GOLDMAN-CECIL MEDICINE - 2 VOLUME SET [25th Edition][2016][UnitedVRG] [PDF] | Online Book Share (2024)

4,177 Pages • 906,424 Words • PDF • 158.6 MB

GOLDMAN-CECIL MEDICINE

1600 John F. Kennedy Blvd. Ste. 1800 Philadelphia, PA 19103-2899

GOLDMAN-CECIL MEDICINE, 25TH EDITION

ISBN: 978-1-4557-5017-7 Volume 1 Part Number: 9996096564 Volume 2 Part Number: 9996096629

International Edition (IE):

ISBN: 978-0-323-28800-2 IE Volume 1 Part Number: 9996118347 IE Volume 2 Part Number: 9996118282

Copyright © 2016, 2012, 2008, 2004, 2000, 1996, 1991, 1988, 1982, 1979, 1975, 1971, 1963, 1959, 1955 by Saunders, an imprint of Elsevier Inc. Copyright 1951, 1947, 1943, 1940, 1937, 1933, 1930, 1927 by Saunders, an imprint of Elsevier Inc. Copyright renewed 1991 by Paul Beeson. Copyright renewed 1979 by Russell L. Cecil and Robert F. Loeb. Copyright renewed 1987, 1975, 1971, 1965, 1961, 1958, 1955 by Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the Publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data Goldman’s Cecil medicine.   Goldman-Cecil medicine / [edited by] Lee Goldman, Andrew I. Schafer.—25th edition.    p. ; cm.   Cecil medicine   Preceded by Goldman’s Cecil medicine / [edited by] Lee Goldman, Andrew I. Schafer. 24th ed. c2012.   Includes bibliographical references.   ISBN 978-1-4557-5017-7 (hardcover, 2 vol set : alk. paper)—ISBN 978-0-323-28800-2 (international edition : alk. paper)—ISBN 978-9996096563 (volume 1 : alk. paper)—ISBN 9996096564 (volume 1 : alk. paper)—ISBN 978-9996096624 (volume 2 : alk. paper)—ISBN 9996096629 (volume 2 : alk. paper)   I.  Goldman, Lee (Physician), editor.  II.  Schafer, Andrew I., editor.  III.  Title.  IV.  Title: Cecil medicine.   [DNLM:  1.  Medicine.  WB 100]   RC46   616—dc23 2014049904 Executive Content Strategist: Kate Dimock Senior Content Development Manager: Maureen Iannuzzi Publishing Services Manager: Anne Altepeter Senior Project Manager: Cindy Thoms Design Specialist: Paula Catalano Printed in the United States of America Last digit is the print number:  9  8  7  6  5  4  3  2  1

ASSOCIATE EDITORS Mary K. Crow, MD

Joseph P. Routh Professor of Rheumatic Diseases in Medicine Weill Cornell Medical College Physician-in-Chief and Benjamin M. Rosen Chair in Immunology and Inflammation Research Hospital for Special Surgery New York, New York

James H. Doroshow, MD Bethesda, Maryland

Jeffrey M. Drazen, MD

Distinguished Parker B. Francis Professor of Medicine Harvard Medical School Senior Physician Brigham and Women’s Hospital Boston, Massachusetts

Robert C. Griggs, MD

Professor of Neurology, Medicine, Pediatrics, and Pathology and Laboratory Medicine University of Rochester School of Medicine and Dentistry Rochester, New York

Donald W. Landry, MD, PhD

Samuel Bard Professor of Medicine Chair, Department of Medicine Physician-in-Chief Columbia University Medical Center New York, New York

Wendy Levinson, MD Professor of Medicine Chair Emeritus Department of Medicine University of Toronto Toronto, Ontario, Canada

Anil K. Rustgi, MD

T. Grier Miller Professor of Medicine and Genetics Chief of Gastroenterology American Cancer Society Professor University of Pennsylvania Perelman School of Medicine Philadelphia, Pennsylvania

W. Michael Scheld, MD

Bayer-Gerald L. Mandell Professor of Infectious Diseases Professor of Medicine Clinical Professor of Neurosurgery Director, Pfizer Initiative in International Health University of Virginia Health System Charlottesville, Virginia

Allen M. Spiegel, MD

Dean Albert Einstein College of Medicine Bronx, New York

PREFACE In the 90 years since the first edition of the Cecil Textbook of Medicine was published, almost everything we know about internal medicine has changed. Progress in medical science is now occurring at an ever-accelerating pace, and it is doing so within the framework of transformational changes in clinical practice and the delivery of health care at individual, social, and global levels. This textbook and its associated electronic products incorporate the latest medical knowledge in multiple formats that should appeal to students and seasoned practitioners regardless of how they prefer to access this rapidly changing information. Even as Cecil’s specific information has changed, however, we have remained true to the tradition of a comprehensive textbook of medicine that carefully explains the why (the underlying pathophysiology of disease) and the how (now expected to be evidence-based from randomized controlled trials and meta-analyses). Descriptions of physiology and pathophysiology include the latest genetic advances in a practical format that strives to be useful to the nonexpert. Medicine has entered an era when the acuity of illness and the limited time available to evaluate a patient have diminished the ability of physicians to satisfy their intellectual curiosity. As a result, the acquisition of information, quite easily achieved in this era, is often confused with knowledge. We have attempted to address this dilemma with a textbook that not only informs but also stimulates new questions and gives a glimpse of the future path to new knowledge. Grade A evidence is specifically highlighted in the text and referenced at the end of each chapter. In addition to the information provided in the textbook, the Cecil website supplies expanded content and functionality. In many cases, the full articles referenced in each chapter can be accessed from the Cecil website. The website is also continuously updated to incorporate subsequent Grade A information, other evidence, and new discoveries. The sections for each organ system begin with a chapter that summarizes an approach to patients with key symptoms, signs, or laboratory abnormalities associated with dysfunction of that organ system. As summarized in E-Table 1-1, the text specifically provides clear, concise information regarding how a physician should approach more than 100 common symptoms, signs, and laboratory abnormalities, usually with a flow diagram, a table, or both for easy reference. In this way, Cecil remains a comprehensive text to guide diagnosis and therapy, not only for patients with suspected or known diseases but also for patients who may have undiagnosed abnormalities that require an initial evaluation. Just as each edition brings new authors, it also reminds us of our gratitude to past editors and authors. Previous editors of Cecil include a short but remarkably distinguished group of leaders of American medicine: Russell Cecil, Paul Beeson, Walsh McDermott, James Wyngaarden, Lloyd H. Smith,

Jr., Fred Plum, J. Claude Bennett, and Dennis Ausiello. As we welcome new associate editors—Mary K. Crow, James H. Doroshow, and Allen M. Spiegel—we also express our appreciation to William P. Arend, James O. Armitage, David R. Clemmons, and other associate editors from the previous editions on whose foundation we have built. Our returning associate editors—Jeffrey M. Drazen, Robert C. Griggs, Donald W. Landry, Wendy Levinson, Anil K. Rustgi, and W. Michael Scheld—continue to make critical contributions to the selection of authors and the review and approval of all manuscripts. The editors, however, are fully responsible for the book as well as the integration among chapters. The tradition of Cecil is that all chapters are written by distinguished experts in each field. We are also most grateful for the editorial assistance in New York of Maribel Lim and Silva Sergenian. These individuals and others in our offices have shown extraordinary dedication and equanimity in working with authors and editors to manage the unending flow of manuscripts, figures, and permissions. We also thank Cassondra Andreychik, Ved Bhushan Arya, Cameron Harrison, Karen Krok, Robert J. Mentz, Gaétane Nocturne, Patrice Savard, Senthil Senniappan, Tejpratap Tiwari, and Sangeetha Venkatarajan, who contributed to various chapters, and we mourn the passing of Morton N. Swartz, MD, co-author of the chapter on “Meningitis: Bacterial, Viral, and Other” and Donald E. Low, MD, author of the chapter “Nonpneumococcal Streptococcal Infections, Rheumatic Fever.” At Elsevier, we are most indebted to Kate Dimock and Maureen Iannuzzi, and also thank Maria Holman, Gabriela Benner, Cindy Thoms, Anne Altepeter, Linda McKinley, Paula Catalano, and Kristin Koehler, who have been critical to the planning and production process under the guidance of Mary Gatsch. Many of the clinical photographs were supplied by Charles D. Forbes and William F. Jackson, authors of Color Atlas and Text of Clinical Medicine, Third Edition, published in 2003 by Elsevier Science Ltd. We thank them for graciously permitting us to include their pictures in our book. We have been exposed to remarkable physicians in our lifetimes and would like to acknowledge the mentorship and support of several of those who exemplify this paradigm— Eugene Braunwald, Lloyd H. Smith, Jr., Frank Gardner, and William Castle. Finally, we would like to thank the Goldman family—Jill, Jeff, Abigail, Mira, Samuel, Daniel, Robyn, Tobin, and Dashel—and the Schafer family— Pauline, Eric, Melissa, Nathaniel, Pam, John, Evan, Samantha, Kate, and Sean, for their understanding of the time and focus required to edit a book that attempts to sustain the tradition of our predecessors and to meet the needs of today’s physician. LEE GOLDMAN, MD ANDREW I. SCHAFER, MD

CONTRIBUTORS Charles S. Abrams, MD Professor of Medicine, Pathology, and Laboratory Medicine, University of Pennsylvania School of Medicine; Director, PENN-Chop Blood Center for Patient Care & Discovery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Thrombocytopenia Frank J. Accurso, MD Professor of Pediatrics, University of Colorado School of Medicine; Attending Physician, Children’s Hospital Colorado, Aurora, Colorado Cystic Fibrosis Ronald S. Adler, MD, PhD Professor of Radiology, New York University School of Medicine; Department of Radiology, NYU Langone Medical Center, New York, New York Imaging Studies in the Rheumatic Diseases Cem Akin, MD, PhD Associate Professor, Harvard Medical School; Attending Physician, Director, Mastocytosis Center, Brigham and Women’s Hospital, Department of Medicine, Division of Rheumatology, Immunology, and Allergy, Boston, Massachusetts Mastocytosis Allen J. Aksamit, Jr., MD Professor of Neurology, Mayo Clinic College of Medicine, Consultant in Neurology, Mayo Clinic, Rochester, Minnesota Acute Viral Encephalitis Qais Al-Awqati, MB ChB Robert F. Loeb Professor of Medicine, Jay I. Meltzer Professor of Nephrology and Hypertension, Professor of Physiology and Cellular Biophysics, Division of Nephrology, Columbia University, College of Physicians and Surgeons, New York, New York Structure and Function of the Kidneys Ban Mishu Allos, MD Associate Professor of Medicine, Division of Infectious Diseases, Associate Professor, Preventive Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee Campylobacter Infections David Altshuler, MD, PhD Professor of Genetics and of Medicine, Harvard Medical School, Massachusetts General Hospital; Professor of Biology (Adjunct), Massachusetts Institute of Technology, Boston and Cambridge, Massachusetts The Inherited Basis of Common Diseases

Larry J. Anderson, MD Professor, Division of Infectious Disease, Department of Pediatrics, Emory University School of Medicine and Children’s Healthcare of Atlanta, Atlanta, Georgia Coronaviruses Aśok C. Antony, MD Chancellor’s Professor of Medicine, Indiana University School of Medicine; Attending Physician, Indiana University Health Affiliated Hospitals and Richard L. Roudebush Veterans Affairs Medical Center, Indianapolis, Indiana Megaloblastic Anemias Gerald B. Appel, MD Professor of Medicine, Division of Nephrology, Department of Medicine, Columbia University College of Physicians and Surgeons, New York, New York Glomerular Disorders and Nephrotic Syndromes Frederick R. Appelbaum, MD Executive Vice President and Deputy Director, Fred Hutchinson Cancer Research Center; President, Seattle Cancer Care Alliance; Professor, Division of Medical Oncology, University of Washington School of Medicine, Seattle Washington The Acute Leukemias Suneel S. Apte, MBBS, DPhil Staff, Cleveland Clinic Lerner College of Medicine at Case Western Reserve University, Cleveland, Ohio Connective Tissue Structure and Function James O. Armitage, MD The Joe Shapiro Professor of Internal Medicine, University of Nebraska Medical Center, Omaha, Nebraska Approach to the Patient with Lymphadenopathy and Splenomegaly; Non-Hodgkin Lymphomas M. Amin Arnaout, MD Professor of Medicine, Departments of Medicine and Developmental and Regenerative Biology, Harvard Medical School; Physician and Chief Emeritus, Division of Nephrology, Massachusetts General Hospital, Boston, Massachusetts Cystic Kidney Diseases Robert M. Arnold, MD Leo H. Criep Professor of Clinical Care, Chief, Section of Palliative Care and Medical Ethics, University of Pittsburgh; Medical Director, UPMC Palliative and Supportive Care Institute, Pittsburgh, Pennsylvania Care of Dying Patients and Their Families

Michael Aminoff, MD, DSc Professor, Department of Neurology, University of California San Francisco, San Francisco, California Approach to the Patient with Neurologic Disease

David Atkins, MD, MPH Director, Health Services Research and Development, Veterans Health Administration, Washington, D.C. The Periodic Health Examination

Jeffrey L. Anderson, MD Professor of Internal Medicine, University of Utah School of Medicine; Vice-Chair for Research, Department of Internal Medicine, Associate Chief of Cardiology and Director of Cardiovascular Research, Intermountain Medical Center, Intermountain Healthcare, Salt Lake City, Utah ST Segment Elevation Acute Myocardial Infarction and Complications of Myocardial Infarction

John P. Atkinson, MD Chief, Division of Rheumatology, Internal Medicine, Washington University School of Medicine in St. Louis, St. Louis, Missouri Complement System in Disease

x

Contributors

Bruce R. Bacon, MD Endowed Chair in Gastroenterology, Professor of Internal Medicine, Co-Director, Saint Louis University Liver Center; Director, Saint Louis University Abdominal Transplant Center, Saint Louis University School of Medicine, St. Louis, Missouri Iron Overload (Hemochromatosis) Larry M. Baddour, MD Professor of Medicine, Chair, Division of Infectious Diseases, Mayo Clinic, Rochester, Minnesota Infective Endocarditis Grover C. Bagby, MD Professor of Medicine and Molecular and Medical Genetics, Knight Cancer Institute at Oregon Health and Science University and Portland VA Medical Center, Portland, Oregon Aplastic Anemia and Related Bone Marrow Failure States Barbara J. Bain, MBBS Professor in Diagnostic Haematology, Imperial College London; Honorary Consultant Haematologist, St. Mary’s Hospital, London, United Kingdom The Peripheral Blood Smear Dean F. Bajorin, MD Attending Physician and Member, Medicine, Memorial Hospital, Memorial Sloan Kettering Cancer Center; Professor of Medicine, Weill Cornell Medical College, New York, New York Tumors of the Kidney, Bladder, Ureters, and Renal Pelvis

Stephen G. Baum, MD Chairman of Medicine, Mount Sinai Beth Israel Hospital; Professor of Medicine and of Microbiology and Immunology, Albert Einstein College of Medicine, New York, New York Mycoplasma Infections Daniel G. Bausch, MD, MPH&TM Associate Professor, Department of Tropical Medicine, Tulane University Health Sciences Center, New Orleans, Louisiana Viral Hemorrhagic Fevers Arnold S. Bayer, MD Professor of Medicine, David Geffen School of Medicine at University of California Los Angeles; LA Biomedical Research Institute; Vice Chair for Academic Affairs, Department of Medicine, Harbor-UCLA Medical Center, Los Angeles, California Infective Endocarditis Hasan Bazari, MD Associate Professor of Medicine, Harvard Medical School, Department of Medicine, Clinical Director, Nephrology, Program Director, Internal Medicine Residency Program, Massachusetts General Hospital, Boston, Massachusetts Approach to the Patient with Renal Disease John H. Beigel, MD National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Antiviral Therapy (Non-HIV)

Robert W. Baloh, MD Professor of Neurology, University of California Los Angeles School of Medicine, Los Angeles, California Neuro-Ophthalmology; Smell and Taste; Hearing and Equilibrium

George A. Beller, MD Professor of Medicine, University of Virginia Health System, Charlottesville, Virginia Noninvasive Cardiac Imaging

Jonathan Barasch, MD, PhD Professor of Medicine and Pathology and Cell Biology, Department of Medicine, Division of Nephrology, Columbia University College of Physicians & Surgeons, New York, New York Structure and Function of the Kidneys

Robert M. Bennett, MD Professor of Medicine, Oregon Health and Science University, Portland, Oregon Fibromyalgia, Chronic Fatigue Syndrome, and Myofascial Pain

Richard L. Barbano, MD, PhD Professor of Neurology, University of Rochester, Rochester, New York Mechanical and Other Lesions of the Spine, Nerve Roots, and Spinal Cord Elizabeth Barrett-Connor, MD Professor of Community and Family Medicine, University of California San Diego, San Diego, California Menopause John R. Bartholomew, MD Section Head, Vascular Medicine, Cardiovascular Medicine, Cleveland Clinic, Professor of Medicine, Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, Ohio Other Peripheral Arterial Diseases Mary Barton, MD, MPP Vice President, Performance Measurement, National Committee for Quality Assurance, Washington, D.C. The Periodic Health Examination Robert C. Basner, MD Professor of Medicine, Columbia University Medical Center; Director, Columbia University Cardiopulmonary Sleep and Ventilatory Disorders Center, Columbia University College of Physicians and Surgeons, New York, New York Obstructive Sleep Apnea

Joseph R. Berger, MD Professor of Neurology, Chief of the Multiple Sclerosis Division, Department of Neurology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania Cytomegalovirus, Epstein-Barr Virus, and Slow Virus Infections of the Central Nervous System; Neurologic Complications of Human Immunodeficiency Virus Infection; Brain Abscess and Parameningeal Infections Paul D. Berk, MD Professor of Medicine, Department of Medicine, Columbia University College of Physicians and Surgeons, New York, New York Approach to the Patient with Jaundice or Abnormal Liver Tests Nancy Berliner, MD Professor of Medicine, Harvard Medical School; Chief, Division of Hematology, Brigham and Women’s Hospital, Boston, Massachusetts Leukocytosis and Leukopenia James L. Bernat, MD Louis and Ruth Frank Professor of Neuroscience, Professor of Neurology and Medicine, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire; Department of Neurology, Dartmouth-Hitchco*ck Medical Center, Lebanon, New Hampshire Coma, Vegetative State, and Brain Death Philip J. Bierman, MD Professor, Department of Internal Medicine, University of Nebraska Medical Center, Omaha, Nebraska Approach to the Patient with Lymphadenopathy and Splenomegaly; Non-Hodgkin Lymphomas

Contributors

xi

Michael R. Bishop, MD Professor of Medicine, Director, Hematopoietic Cellular Therapy Program, Section of Hematology and Oncology, Department of Medicine, University of Chicago, Chicago, Illinois Hematopoietic Stem Cell Transplantation

William E. Boden, MD Professor of Medicine, Albany Medical College; Chief of Medicine, Albany Stratton VA Medical Center; Vice-Chairman, Department of Medicine, Albany Medical Center, Albany, New York Angina Pectoris and Stable Ischemic Heart Disease

Bruce R. Bistrian, MD, PhD, MPH Professor of Medicine, Beth Israel Deaconess Medical Center; Professor of Medicine, Harvard Medical School, Boston, Massachusetts Nutritional Assessment

Jean Bolognia, MD Professor of Dermatology, Yale Medical School; Attending Physician, Yale-New Haven Hospital, New Haven, Connecticut Infections, Hyperpigmentation and Hypopigmentation, Regional Dermatology, and Distinctive Lesions in Black Skin

Joseph J. Biundo, MD Clinical Professor of Medicine, Tulane Medical Center, New Orleans, Louisiana Bursitis, Tendinitis, and Other Periarticular Disorders and Sports Medicine Adrian R. Black, PhD Assistant Professor, Director of Tissue Sciences for the Eppley Institute, The Eppley Institute for Research in Cancer and Allied Diseases, University of Nebraska Medical Center, Omaha, Nebraska Cancer Biology and Genetics Charles D. Blanke, MD Professor of Medicine, Oregon Health and Science University, Portland, Oregon Neoplasms of the Small and Large Intestine Joel N. Blankson, MD, PhD Associate Professor, Johns Hopkins University School of Medicine, Baltimore, Maryland Immunopathogenesis of Human Immunodeficiency Virus Infection Martin J. Blaser, MD Muriel and George Singer Professor of Medicine, Professor of Microbiology, Director, Human Microbiome Program, New York University Langone Medical Center, New York, New York Acid Peptic Disease; Human Microbiome William A. Blattner, MD Professor and Associate Director, Institute of Human Virology, School of Medicine, University of Maryland; Professor of Medicine, School of Medicine, University of Maryland; Professor and Head, Division of Cancer Epidemiology, Department of Epidemiology and Public Health, School of Medicine, University of Maryland, Baltimore, Maryland Retroviruses Other Than Human Immunodeficiency Virus Thomas P. Bleck, MD Professor of Neurological Sciences, Neurosurgery, Internal Medicine, and Anesthesiology, Associate Chief Medical Officer (Critical Care), Rush Medical College, Chicago, Illinois Arboviruses Affecting the Central Nervous System Joel A. Block, MD The Willard L. Wood MD Professor and Director, Division of Rheumatology, Rush University Medical Center, Chicago, Illinois Osteoarthritis Henk Blom, MD Laboratory of Clinical Biochemistry and Metabolism, Department of General Pediatrics, Adolescent Medicine and Neonatology, University Medical Centre Freiburg, Head of Laboratory/Clinical Biochemical Geneticist, Freiburg, Germany hom*ocystinuria and Hyperhom*ocysteinemia Olaf A. Bodamer, MD Medical Genetics, University of Miami Hospital, Miami, Florida Approach to Inborn Errors of Metabolism

Robert A. Bonomo, MD Chief, Medical Service, Louis Stokes Cleveland VA Medical Center; Professor of Medicine, Pharmacology, Biochemistry, Molecular Biology, and Microbiology, Case Western Reserve University School of Medicine, Cleveland, Ohio Diseases Caused by Acinetobacter and Stenotrophom*onas Species Larry Borish, MD Professor of Medicine, Allergy, and Clinical Immunology, University of Virginia Health System, Charlottesville, Virgina Allergic Rhinitis and Chronic Sinusitis Patrick J. Bosque, MD Associate Professor of Neurology, University of Colorado Denver School of Medicine; Neurologist, Denver Health Medical Center, Denver, Colorado Prion Diseases David J. Brenner, PhD, DSc Higgins Professor of Radiation Biophysics, Center for Radiological Research, Columbia University Medical Center, New York, New York Radiation Injury Itzhak Brook, MD, MSc Professor of Pediatrics and Medicine, Georgetown University, Georgetown University Medical Center, Washington, D.C. Diseases Caused by Non–Spore-Forming Anaerobic Bacteria; Actinomycosis Enrico Brunetti, MD Assistant Professor of Infectious Diseases, University of Pavia; Attending Physician, Division of Infectious and Tropical Diseases, IRCCS San Matteo Hospital Foundation; Co-Director, WHO Collaborating Centre for Clinical Management of Cystic Echinococcosis, Pavia, Italy Cestodes David M. Buchner, MD, MPH Shahid and Ann Carlson Khan Professor in Applied Health Sciences, Department of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, Champaign, Illinois Physical Activity Pierre A. Buffet, MD, PhD Research Unit Head, Erythrocyte Parasite Pathogenesis Research Team INSERM–University Paris 6, CIMI–Paris Research Center, University Pierre and Marie Curie; Associate Professor of Parasitology, Faculty of Medicine, University Pierre and Marie Curie, Pitié-Salpêtrière Hospital, Paris, France Leishmaniasis H. Franklin Bunn, MD Professor of Medicine, Harvard Medical School; Physician, Brigham and Women’s Hospital, Boston, Massachusetts Approach to the Anemias David A. Bushinsky, MD John J. Kuiper Distinguished Professor of Medicine, Chief, Nephrology Division, University of Rochester School of Medicine; Associate Chair for Academic Affairs in Medicine, University of Rochester Medical Center, Rochester, New York Nephrolithiasis

xii

Contributors

Vivian P. Bykerk, MD Associate Professor of Medicine, Weill Cornell Medical College; Associate Attending Physician, Hospital for Special Surgery, New York, New York Approach to the Patient with Rheumatic Disease Peter A. Calabresi, MD Professor of Neurology and Director of the Richard T. Johnson Division of Neuroimmunology and Neuroinfectious Diseases, Johns Hopkins University; Director of the Multiple Sclerosis Center, Johns Hopkins Hospital, Baltimore, Maryland Multiple Sclerosis and Demyelinating Conditions of the Central Nervous System David P. Calfee, MD, MS Associate Professor of Medicine and Healthcare Policy and Research, Weill Cornell Medical College; Chief Hospital Epidemiologist, New YorkPresbyterian Hospital/Weill Cornell Medical Center, New York, New York Prevention and Control of Health Care–Associated Infections Douglas Cameron, MD, MBA Professor of Ophthalmology and Visual Neurosciences, University of Minnesota, Minneapolis, Minnesota Diseases of the Visual System Michael Camilleri, MD Atherton and Winifred W. Bean Professor, Professor of Medicine, Pharmacology, and Physiology, College of Medicine, Mayo Clinic, Consultant, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota Disorders of Gastrointestinal Motility Grant W. Cannon, MD Thomas E. and Rebecca D. Jeremy Presidential Endowed Chair for Arthritis Research, Associate Chief of Staff for Academic Affiliations, George E. Wahlen VA Medical Center, Salt Lake City, Utah Immunosuppressing Drugs Including Corticosteroids Maria Domenica Cappellini, MD Professor of Internal Medicine, University of Milan, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, University of Milan, Milan, Italy The Thalassemias Blase A. Carabello, MD Professor of Medicine, Chairman, Department of Cardiology, Mount Sinai Beth Israel Heart Institute, New York, New York Valvular Heart Disease Edgar M. Carvalho, MD Professor of Medicine and Clinical Immunology, Faculdade de Medicina da Bahia, Universidade Federal da Bahia and Escola Bahiana de Medicina e Saúde Pública, Salvador, Bahia, Brazil Schistosomiasis (Bilharziasis) William H. Catherino, MD, PhD Professor and Research Head, Department of Obstetrics and Gynecology, Uniformed Services University of the Health Sciences Division of Reproductive Endocrinology and Infertility; Program in Reproductive and Adult Endocrinology, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland Ovaries and Development; Reproductive Endocrinology and Infertility Jane A. Cauley, DrPH Professor of Epidemiology, University of Pittsburgh Graduate School of Public Health, Vice Chair of the Department of Epidemiology, Pittsburgh, Pennsylvania Epidemiology of Aging: Implications of the Aging of Society

Naga P. Chalasani, MD David W. Crabb Professor and Director, Division of Gastroenterology and Hepatology, Indiana University School of Medicine, Indianapolis, Indiana Alcoholic and Nonalcoholic Steatohepatitis Henry F. Chambers, MD Professor of Medicine, University of California San Francisco School of Medicine; Director, Clinical Research Services, Clinical and Translational Sciences Institute, San Francisco, California Staphylococcal Infections William P. Cheshire, Jr., MD Professor of Neurology, Mayo Clinic, Jacksonville, Florida Autonomic Disorders and Their Management Ilseung Cho, MD, MS Assistant Professor of Medicine, Division of Gastroenterology, Department of Medicine, New York University, New York, New York Human Microbiome Arun Chockalingam, PhD Professor of Epidemiology and Global Health, Director, Office of Global Health Education and Training; Dalla Lana Faculty of Public Health, University of Toronto, Toronto, Ontario, Canada Global Health David C. Christiani, MD Professor of Medicine, Harvard Medical School; Physician, Pulmonary and Critical Care, Massachusetts General Hospital; Elkan Blout Professor of Environmental Genetics, Environmental Health, Harvard School of Public Health, Boston, Massachusetts Physical and Chemical Injuries of the Lung David H. Chu, MD, PhD Director, Contact Dermatitis, Division of Dermatology and Cutaneous Surgery, Scripps Clinic Medical Group, La Jolla, California Structure and Function of the Skin Theodore J. Cieslak, MD Pediatric Infectious Diseases, Clinical Professor of Pediatrics, University of Texas Health Science Center at San Antonio; Department of Pediatrics, Fort Sam Houston, Texas Bioterrorism Carolyn Clancy, MD Interim Under Secretary for Health, Veterans Administration, Washington, D.C. Measuring Health and Health Care David R. Clemmons, MD Kenan Professor of Medicine, University of North Carolina School of Medicine; Attending Physician, Medicine, UNC Hospitals, Chapel Hill, North Carolina Approach to the Patient with Endocrine Disease David Cohen, MD Professor of Medicine, Division of Nephrology; Medical Director, Kidney and Pancreas Transplantation, Columbia University Medical Center, New York, New York Treatment of Irreversible Renal Failure Jeffrey Cohen, MD Chief, Laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Varicella-Zoster Virus (Chickenpox, Shingles)

Contributors

xiii

Myron S. Cohen, MD Associate Vice Chancellor for Global Health, Director, UNC Institute for Global Health and Infectious Diseases, Chief, Division of Infectious Diseases, Yeargan-Bate Eminent Professor of Medicine, Microbiology, and Immunology and Epidemiology, Chapel Hill, North Carolina Approach to the Patient with a Sexually Transmitted Infection; Prevention of Human Immunodeficiency Virus Infection

Mary K. Crow, MD Joseph P. Routh Professor of Rheumatic Diseases in Medicine, Weill Cornell Medical College; Physician in Chief and Benjamin M. Rosen Chair in Immunology and Inflammation Research, Hospital for Special Surgery, New York, New York The Innate Immune Systems; Approach to the Patient with Rheumatic Disease; Systemic Lupus Erythematosus

Steven P. Cohen, MD Professor of Anesthesiology and Critical Care Medicine and Physical Medicine and Rehabilitation, Johns Hopkins School of Medicine, Baltimore, Maryland, and Uniformed Services University of the Health Sciences, Bethesda, Maryland; Director, Pain Research, Walter Reed National Military Medical Center, Bethesda, Maryland Pain

John A. Crump, MB ChB, MD, DTM&H McKinlay Professor of Global Health, Centre for International Health, University of Otago, Dunedin, New Zealand Salmonella Infections (Including Enteric Fever)

Steven L. Cohn, MD Professor of Clinical Medicine, University of Miami Miller School of Medicine; Medical Director, UHealth Preoperative Assessment Center; Director, Medical Consultation Service, University of Miami Hospital, Miami, Florida Preoperative Evaluation Robert Colebunders, MD Emeritus Professor, Institute of Tropical Medicine, Antwerp, Belgium Immune Reconstitution Inflammatory Syndrome in HIV/AIDS Joseph M. Connors, MD Clinical Professor, University of British Columbia; Clinical Director, BC Cancer Agency Centre for Lymphoid Cancer, Vancouver, British Columbia, Canada Hodgkin Lymphoma Deborah J. Cook, MD, MSc Professor of Medicine, Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada Approach to the Patient in a Critical Care Setting Kenneth H. Cowan, MD, PhD Director, Fred & Pamela Buffett Cancer Center; Director, The Eppley Institute for Research in Cancer and Allied Diseases; Professor of Medicine, University of Nebraska Medical Center, Omaha, Nebraska Cancer Biology and Genetics Joseph Craft, MD Paul B. Beeson Professor of Medicine and Immunobiology, Section Chief, Rheumatology, Program Director, Investigative Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut The Adaptive Immune Systems Jill Patricia Crandall, MD Professor of Clinical Medicine, Division of Endocrinology and Diabetes Research Center, Albert Einstein College of Medicine, Bronx, New York Diabetes Mellitus Simon L. Croft, BSc, PhD Professor of Parasitology, Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, United Kingdom Leishmaniasis Kristina Crothers, MD Associate Professor, Department of Medicine, Division of Pulmonary and Critical Care, University of Washington School of Medicine, Seattle, Washington Pulmonary Manifestations of Human Immunodeficiency Virus and Acquired Immunodeficiency Syndrome

Mark R. Cullen, MD Professor of Medicine, Department of Medicine, Stanford University School of Medicine, Stanford, California Principles of Occupational and Environmental Medicine Charlotte Cunningham-Rundles, MD, PhD Professor of Medicine and Pediatrics, Icahn School of Medicine at Mount Sinai, New York, New York Primary Immunodeficiency Diseases Inger K. Damon, MD, PhD Director, Division of High Consequence Pathogens and Pathology, Centers for Disease Control and Prevention, Atlanta, Georgia Smallpox, Monkeypox, and Other Poxvirus Infections Troy E. Daniels, DDS, MS Professor Emeritus of Oral Pathology and Pathology, University of California San Francisco, San Francisco, California Diseases of the Mouth and Salivary Glands Nancy E. Davidson, MD Hillman Professor of Oncology, University of Pittsburgh; Director, University of Pittsburgh Cancer Institute and UPMC CancerCenter, Pittsburgh, Pennsylvania Breast Cancer and Benign Breast Disorders Lisa M. DeAngelis, MD Chair, Department of Neurology, Memorial Sloan-Kettering Cancer Center; Professor of Neurology, Weill Cornell Medical College, New York, New York Tumors of the Central Nervous System Malcolm M. DeCamp, MD Fowler McCormick Professor of Surgery, Feinberg School of Medicine, Northwestern University; Chief, Division of Thoracic Surgery, Northwestern Memorial Hospital, Chicago, Illinois Interventional and Surgical Approaches to Lung Disease Carlos del Rio, MD Hubert Professor and Chair and Professor of Medicine, Hubert Department of Global Health, Rollins School of Public Health and Department of Medicine, Emory University School of Medicine, Atlanta, Georgia Prevention of Human Immunodeficiency Virus Infection Patricia A. Deuster, PhD, MPH Professor and Director, Consortium for Health and Military Performance, Department of Military and Emergency Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland Rhabdomyolysis Robert B. Diasio, MD William J. and Charles H. Mayo Professor, Molecular Pharmacology and Experimental Therapeutics and Oncology, Mayo Clinic, Rochester, Minnesota Principles of Drug Therapy

xiv

Contributors

David J. Diemert, MD Associate Professor, Department of Microbiology, Immunology, and Tropical Medicine, School of Medicine and Health Sciences, The George Washington University, Washington, D.C. Intestinal Nematode Infections; Tissue Nematode Infections Kathleen B. Digre, MD Professor of Neurology, Ophthalmology, Director, Division of Headache and Neuro-Ophthalmology, University of Utah, Salt Lake City, Utah Headaches and Other Head Pain James H. Doroshow, MD Bethesda, Maryland Approach to the Patient with Cancer; Malignant Tumors of Bone, Sarcomas, and Other Soft Tissue Neoplasms John M. Douglas, Jr., MD Executive Director, Tri-County Health Department, Greenwood Village, Colorado Papillomavirus Jeffrey M. Drazen, MD Distinguished Parker B. Francis Professor of Medicine, Harvard Medical School; Senior Physician, Brigham and Women’s Hospital, Boston, Massachusetts Asthma Stephen C. Dreskin, MD, PhD Professor of Medicine and Immunology, Division of Allergy and Clinical Immunology, Department of Medicine, University of Colorado Denver, School of Medicine, Aurora, Colorado Urticaria and Angioedema W. Lawrence Drew, MD, PhD Professor Emeritus, Laboratory Medicine and Medicine, University of California San Francisco, San Francisco, California Cytomegalovirus George L. Drusano, MD Professor and Director, Institute for Therapeutic Innovation, College of Medicine, University of Florida, Lake Nona, Florida Antibacterial Chemotherapy Thomas D. DuBose, Jr., MD Emeritus Professor of Internal Medicine and Nephrology, Wake Forest School of Medicine, Winston-Salem, North Carolina Vascular Disorders of the Kidney F. Daniel Duffy, MD Professor of Internal Medicine and Steve Landgarten Chair in Medical Leadership, School of Community Medicine, University of Oklahoma College of Medicine, Tulsa, Oklahoma Counseling for Behavior Change Herbert L. DuPont, MD, MACP Mary W. Kelsey Chair and Director, Center for Infectious Diseases, University of Texas School of Public Health; H. Irving Schweppe Chair of Internal Medicine and Vice Chairman, Department of Medicine, Baylor College of Medicine; Chief of Internal Medicine, St. Luke’s Hospital System, Houston, Texas Approach to the Patient with Suspected Enteric Infection Madeleine Duvic, MD Professor and Deputy Chairman, Department of Dermatology, The University of Texas MD Anderson Cancer Center, Houston, Texas Urticaria, Drug Hypersensitivity Rashes, Nodules and Tumors, and Atrophic Diseases

Kathryn M. Edwards, MD Sarah H. Sell and Cornelius Vanderbilt Chair in Pediatrics, Vanderbilt University School of Medicine; Director, Vanderbilt Vaccine Research Program, Monroe Carrell Jr. Children’s Hospital at Vanderbilt, Nashville, Tennessee Parainfluenza Viral Disease N. Lawrence Edwards, MD Professor of Medicine, Vice Chairman, Department of Medicine, University of Florida; Chief, Section of Rheumatology, Medical Service, Malcom Randall Veterans Affairs Medical Center, Gainesville, Florida Crystal Deposition Diseases Lawrence H. Einhorn, MD Distinguished Professor, Department of Medicine, Division of Hematology/Oncology, Livestrong Foundation Professor of Oncology, Indiana University School of Medicine, Indianapolis, Indiana Testicular Cancer Ronald J. Elin, MD, PhD A.J. Miller Professor and Chairman, Department of Pathology and Laboratory Medicine, University of Louisville School of Medicine, Louisville, Kentucky Reference Intervals and Laboratory Values George M. Eliopoulos, MD Professor of Medicine, Harvard Medical School; Physician, Division of Infectious Diseases, Beth Israel Deaconess Medical Center, Boston, Massachusetts Principles of Anti-Infective Therapy Perry Elliott, MD Professor in Inherited Cardiovascular Disease, Institute of Cardiovascular Science, University College London, London, United Kingdom Diseases of the Myocardium and Endocardium Jerrold J. Ellner, MD Professor of Medicine, Boston University School of Medicine; Chief, Section of Infectious Diseases, Boston Medical Center, Boston, Massachusetts Tuberculosis Dirk M. Elston, MD Director, Ackerman Academy of Dermatopathology, New York, New York Arthropods and Leeches Ezekiel J. Emanuel, MD, PhD Vice Provost for Global Initiatives, Diane V.S. Levy and Robert M. Levy University Professor, Chair, Department of Medical Ethics and Health Policy, University of Pennsylvania, Philadelphia, Pennsylvania Bioethics in the Practice of Medicine Joel D. Ernst, MD Director, Division of Infectious Diseases and Immunology, Jeffrey Bergstein Professor of Medicine, Professor of Medicine, Pathology, and Microbiology, New York University School of Medicine; Attending Physician, New York University Langone Medical Center, New York, New York Leprosy (Hansen Disease) Gregory T. Everson, MD Professor of Medicine, Director of Hepatology, University of Colorado School of Medicine, Aurora, Colorado Hepatic Failure and Liver Transplantation Amelia Evoli, MD Associate Professor of Neurology, Catholic University, Agostino Gemelli University Hospital, Rome, Italy Disorders of Neuromuscular Transmission

Contributors Douglas O. Faigel, MD Professor of Medicine, Mayo Clinic, Chair, Division of Gastroenterology and Hepatology, Scottsdale, Arizona Neoplasms of the Small and Large Intestine

Manuel A. Franco, MD, PhD Director of Postgraduate Programs, School of Sciences, Pontificia Universidad Javeriana, Bogota, Colombia Rotaviruses, Noroviruses, and Other Gastrointestinal Viruses

Matthew E. Falagas, MD, MSc, DSc Director, Alfa Institute of Biomedical Sciences, Athens, Greece; Adjunct Associate Professor of Medicine, Tufts University School of Medicine, Boston, Massachusetts; Chief, Department of Medicine and Infectious Diseases, Iaso General Hospital, Iaso Group, Athens, Greece Pseudomonas and Related Gram-Negative Bacillary Infections

David O. Freedman, MD Professor of Medicine and Microbiology, University of Alabama at Birmingham; Director, Gorgas Center for Geographic Medicine, Birmingham, Alabama Approach to the Patient before and after Travel

Gary W. Falk, MD, MS Professor of Medicine, Division of Gastroenterology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Diseases of the Esophagus Gene Feder, MBBS, MD Professor, Centre for Academic Primary Care, School of Social and Community Medicine, University of Bristol; General Practitioner, Helios Medical Centre, Bristol, United Kingdom Intimate Partner Violence David J. Feller-Kopman, MD Director, Bronchoscopy and Interventional Pulmonology, Associate Professor of Medicine, The Johns Hopkins University, Baltimore, Maryland Interventional and Surgical Approaches to Lung Disease Gary S. Firestein, MD Dean and Associate Vice Chancellor of Translational Medicine, University of California San Diego School of Medicine, La Jolla, California Mechanisms of Inflammation and Tissue Repair Glenn I. Fishman, MD Director, Leon H. Charney Division of Cardiology, Vice-Chair for Research, Department of Medicine, William Goldring Professor of Medicine, New York University School of Medicine, New York, New York Principles of Electrophysiology Lee A. Fleisher, MD Robert D. Dripps Professor and Chair, Anesthesiology and Critical Care, Professor of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Overview of Anesthesia Paul W. Flint, MD Professor and Chair, Otolaryngology, Head and Neck Surgery, Oregon Health and Science University, Portland, Oregon Throat Disorders Evan L. Fogel, MD, MSc Professor of Clinical Medicine, Indiana University School of Medicine, Indianapolis, Indiana Diseases of the Gallbladder and Bile Ducts Marsha D. Ford, MD Adjunct Professor of Emergency Medicine, School of Medicine, University of North Carolina-Chapel Hill; Director, Carolinas Poison Center, Carolinas HealthCare System, Charlotte, North Carolina Acute Poisoning Chris E. Forsmark, MD Professor of Medicine, Chief, Division of Gastroenterology, Hepatology, and Nutrition, University of Florida, Gainesville, Florida Pancreatitis Vance G. Fowler, Jr., MD, MHS Professor of Medicine, Duke University Medical Center, Durham, North Carolina Infective Endocarditis

xv

Martyn A. French, MD Professor in Clinical Immunology, School of Pathology and Laboratory Medicine, University of Western Australia, Perth, Australia Immune Reconstitution Inflammatory Syndrome in HIV/AIDS Karen Freund, MD, MPH Professor of Medicine, Associate Director, Tufts Clinical and Translational Science Institute, Tufts University School of Medicine, Tufts Medical Center, Boston, Massachusetts Approach to Women’s Health Cem Gabay, MD Professor of Medicine, Head, Division of Rheumatology, University Hospitals of Geneva, Geneva, Switzerland Biologic Agents Kenneth L. Gage, PhD Chief, Entomology and Ecology Activity, Centers for Disease Control and Prevention, Division of Vector-Borne Diseases, Bacterial Diseases Branch, Fort Collins, Colorado Plague and Other Yersinia Infections John N. Galgiani, MD Professor of Medicine, Valley Fever Center for Excellence, University of Arizona, Tucson, Arizona Coccidioidomycosis Patrick G. Gallagher, MD Professor of Pediatrics, Pathology, and Genetics, Yale University School of Medicine; Attending Physician, Yale–New Haven Hospital, New Haven, Connecticut Hemolytic Anemias: Red Blood Cell Membrane and Metabolic Defects Leonard Ganz, MD Director of Cardiac Electrophysiology, Heritage Valley Health System, Beaver, Pennsylvania Electrocardiography Hasan Garan, MD Director, Cardiac Electrophysiology, Dickinson W. Richards, Jr. Professor of Medicine, Columbia University Medical Center, New York, New York Ventricular Arrhythmias Guadalupe Garcia-Tsao, MD Professor of Medicine, Yale University School of Medicine; Chief, Digestive Diseases, VA Connecticut Healthcare System, West Haven, Connecticut Cirrhosis and Its Sequelae William M. Geisler, MD, MPH Professor of Medicine, University of Alabama at Birmingham, Birmingham, Alabama Diseases Caused by Chlamydiae Tony P. George, MD Division of Brain and Therapeutics, Department of Psychiatry, University of Toronto; Schizophrenia Division, The Centre for Addiction and Mental Health, Toronto, Ontario, Canada Nicotine and Tobacco

xvi

Contributors

Lior Gepstein, MD, PhD Edna and Jonathan Sohnis Professor in Medicine and Physiology, Rappaport Faculty of Medicine and Research Institute, Technion–Israel Institute of Technology, Rambam Health Care Campus, Haifa, Israel Gene and Cell Therapy

Larry B. Goldstein, MD Professor of Neurology, Director, Duke Stroke Center, Neurology, Duke University; Staff Neurologist, Durham VA Medical Center, Durham, North Carolina Approach to Cerebrovascular Diseases; Ischemic Cerebrovascular Disease

Susan I. Gerber, MD Team Lead, Respiratory Viruses/Picornaviruses, Division of Viral Diseases/Epidemiology Branch, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia Coronaviruses

Lawrence T. Goodnough, MD Professor of Pathology and Medicine, Stanford University; Director, Transfusion Service, Stanford University Medical Center, Stanford, California Transfusion Medicine

Dale N. Gerding, MD Professor of Medicine, Loyola University Chicago Stritch School of Medicine, Research Physician, Edward Hines, Jr. VA Hospital, Hines, Illinois Clostridial Infections Morie A. Gertz, MD Consultant, Division of Hematology, Mayo Clinic, Rochester, Minnesota; Roland Seidler, Jr. Professor of the Art of Medicine in Honor of Michael D. Brennan, MD, Professor of Medicine, Mayo Clinic, College of Medicine, Rochester, Minnesota Amyloidosis Gordon D. Ginder, MD Professor, Internal Medicine, Director, Massey Cancer Center, Virginia Commonwealth University, Richmond, Virginia Microcytic and Hypochromic Anemias Jeffrey S. Ginsberg, MD Professor of Medicine, McMaster University, Member of Thrombosis and Atherosclerosis Research Institute, St. Joseph’s Healthcare Hamilton, Hamilton, Ontario, Canada Peripheral Venous Disease Geoffrey S. Ginsburg, MD, PhD Director, Duke Center for Applied Genomics and Precision Medicine; Professor of Medicine, Pathology and Biomedical Engineering, Duke University, Durham, North Carolina Applications of Molecular Technologies to Clinical Medicine Michael Glogauer, DDS, PhD Professor, Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada Disorders of Phagocyte Function John W. Gnann, Jr., MD Professor of Medicine, Department of Medicine, Division of Infectious Diseases, Medical University of South Carolina, Charleston, South Carolina Mumps Matthew R. Golden, MD, MPH Professor of Medicine, University of Washington, Director, HIV/STD Program, Public Health–Seattle & King County, Seattle, Washington Neisseria Gonorrhoeae Infections Lee Goldman, MD Harold and Margaret Hatch Professor, Executive Vice President and Dean of the Faculties of Health Sciences and Medicine, Chief Executive, Columbia University Medical Center, Columbia University, New York, New York Approach to Medicine, the Patient, and the Medical Profession: Medicine as a Learned and Humane Profession; Approach to the Patient with Possible Cardiovascular Disease Ellie J.C. Goldstein, MD Clinical Professor of Medicine, David Geffen School of Medicine at University of California Los Angeles, Los Angeles, California; Director, R.M. Alden Research Laboratory, Santa Monica, California Diseases Caused by Non–Spore-Forming Anaerobic Bacteria

Eduardo H. Gotuzzo, MD Professor of Medicine, Director, Alexander von Humboldt Tropical Medicine Institute, Universidad Peruana Cayetano Heredia; Chief Physician, Department of Infectious, Tropical, and Dermatologic Diseases, National Hospital Cayetano Heredia, Lima, Peru Cholera and Other Vibrio Infections; Liver, Intestinal, and Lung Fluke Infections Deborah Grady, MD, MPH Professor of Medicine, University of California San Francisco, San Francisco, California Menopause Leslie C. Grammer, MD Professor of Medicine, Northwestern University Feinberg School of Medicine; Attending Physician, Northwestern Memorial Hospital, Chicago, Illinois Drug Allergy F. Anthony Greco, MD Medical Director, Sarah Cannon Cancer Center, Nashville, Tennessee Cancer of Unknown Primary Origin Harry B. Greenberg, MD Professor, Departments of Medicine and Microbiology and Immunology, Stanford University School of Medicine, Stanford, California Rotaviruses, Noroviruses, and Other Gastrointestinal Viruses Steven A. Greenberg, MD Associate Professor of Neurology, Harvard Medical School; Associate Neurologist, Brigham and Women’s Hospital, Boston, Massachusetts Inflammatory Myopathies Robert C. Griggs, MD Professor of Neurology, Medicine, Pediatrics, and Pathology and Laboratory Medicine, University of Rochester School of Medicine and Dentistry, Rochester, New York Approach to the Patient with Neurologic Disease Lev M. Grinberg, MD, PhD Professor, Chief, Department of Pathology, Ural Medical University; Chief Researcher of the Ural Scientific Research Institute of Phthisiopulmonology, Chief Pathologist of Ekaterinburg, Ekaterinburg, Russia Anthrax Daniel Grossman, MD Vice President for Research, Ibis Reproductive Health, Oakland, California; Assistant Clinical Professor, Bixby Center for Global Reproductive Health, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California San Francisco, San Francisco, California Contraception Lisa M. Guay-Woodford, MD Hudson Professor of Pediatrics, The George Washington University; Director, Center for Translational Science, Director, Clinical and Translational Institute at Children’s National, Children’s National Health System, Washington, D.C. Hereditary Nephropathies and Developmental Abnormalities of the Urinary Tract

Contributors Richard L. Guerrant, MD Thomas H. Hunter Professor of International Medicine, Founding Director, Center for Global Health, Division of Infectious Diseases and International Health, University of Virginia School of Medicine, University of Virginia Health Sciences Center, Charlottesville, Virginia Cryptosporidiosis Roy M. Gulick, MD, MPH Gladys and Roland Harrison Professor of Medicine, Medicine/Infectious Diseases, Weill Cornell Medical College; Attending Physician, New York– Presbyterian Hospital, New York, New York Antiretrovial Therapy of HIV/AIDS Klaus D. Hagspiel, MD Professor of Radiology, Medicine, and Pediatrics, Chief, Noninvasive Cardiovascular Imaging, University of Virginia Health System, Charlottesville, Virginia Noninvasive Cardiac Imaging John D. Hainsworth, MD Chief Scientific Officer, Sarah Cannon Research Institute, Nashville, Tennessee Cancer of Unknown Primary Origin Anders Hamsten, MD, PhD Professor of Cardiovascular Diseases, Center for Molecular Medicine and Department of Cardiology, Karolinska University Hospital, Department of Medicine, Karolinska Institute, Stockholm, Sweden Atherosclerosis, Thrombosis, and Vascular Biology Kenneth R. Hande, MD Professor of Medicine and Pharmacology, Vanderbilt/Ingram Cancer Center, Vanderbilt University School of Medicine, Nashville, Tennessee Neuroendocrine Tumors and the Carcinoid Syndrome H. Hunter Handsfield, MD Professor Emeritus of Medicine, University of Washington Center for AIDS and STD, Seattle, Washington Neisseria Gonorrhoeae Infections Göran K. Hansson, MD, PhD Professor of Cardiovascular Research, Center for Molecular Medicine at Karolinska University Hospital, Department of Medicine, Karolinska Institute, Stockholm, Sweden Atherosclerosis, Thrombosis, and Vascular Biology Raymond C. Harris, MD Professor of Medicine, Ann and Roscoe R. Robinson Chair in Nephrology, Chief, Division of Nephrology, Vanderbilt University School of Medicine, Nashville, Tennessee Diabetes and the Kidney Stephen Crane Hauser, MD Associate Professor of Medicine, Internal Medicine, Division of Gastroenterology and Hepatology, Mayo Clinic College of Medicine, Rochester, Minnesota Vascular Diseases of the Gastrointestinal Tract Frederick G. Hayden, MD Stuart S. Richardson Professor of Clinical Virology and Professor of Medicine, University of Virginia School of Medicine; Staff Physician, University of Virginia Health System, Charlottesville, Virginia Influenza Douglas C. Heimburger, MD, MS Professor of Medicine, Associate Director for Education and Training, Vanderbilt University School of Medicine, Vanderbilt Institute for Global Health, Nashville, Tennessee Nutrition’s Interface with Health and Disease

xvii

Erik L. Hewlett, MD Professor of Medicine and of Microbiology, Immunology, and Cancer Biology, University of Virginia School of Medicine, University of Virginia Health System, Charlottesville, Virginia Whooping Cough and Other Bordetella Infections Richard J. Hift, PhD, MMed School of Clinical Medicine, University of KwaZulu-Natal, Durban, South Africa The Porphyrias David R. Hill, MD, DTM&H Professor of Medical Sciences, Director of Global Public Health, Frank H. Netter MD School of Medicine at Quinnipiac University, Hamden, Connecticut Giardiasis Nicholas S. Hill, MD Professor of Medicine, Tufts University School of Medicine; Chief, Division of Pulmonary, Critical Care, and Sleep Medicine, Tufts Medical Center, Boston, Massachusetts Respiratory Monitoring in Critical Care L. David Hillis, MD Professor and Chair, Department of Medicine, University of Texas Health Science Center at San Antonio, San Antonio, Texas Acute Coronary Syndrome: Unstable Angina and Non-ST Elevation Myocardial Infarction Jack Hirsh, CM, MD, DSc Professor Emeritus, McMaster University, Hamilton, Ontario, Canada Antithrombotic Therapy Steven M. Holland, MD Chief, Laboratory of Clinical Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland The Nontuberculous Mycobacteria Steven M. Hollenberg, MD Professor of Medicine, Cooper Medical School of Rowan University; Director, Coronary Care Unit, Cooper University Hospital, Camden, New Jersey Cardiogenic Shock Edward W. Hook III, MD Professor and Director, Division of Infectious Diseases, University of Alabama at Birmingham, Birmingham, Alabama Granuloma Inguinale (Donovanosis); Syphilis; Nonsyphilitic Treponematoses David J. Hunter, MBBS, MPH, ScD Vincent L. Gregory Professor of Cancer Prevention, Harvard School of Public Health; Professor of Medicine, Harvard Medical School, Brigham and Women’s Hospital, Boston, Massachusetts The Epidemiology of Cancer Khalid Hussain, MBChB, MD, MSc Developmental Endocrinology Research Group, Clinical and Molecular Genetics Unit, Institute of Child Health, University College London, Department of Paediatric Endocrinology, Great Ormond Street Hospital for Children, London, United Kingdom Hypoglycemia/Pancreatic Islet Cell Disorders Steven E. Hyman, MD Director, Stanley Center for Psychiatric Research, Broad Institute, Distinguished Service Professor of Stem Cell and Regenerative Biology, Harvard University, Cambridge, Massachusetts Biology of Addiction

xviii

Contributors

Michael C. Iannuzzi, MD, MBA Chairman, Department of Internal Medicine, State University of New York Upstate Medical University, Syracuse, New York Sarcoidosis

Richard C. Jordan, DDS, PhD Professor of Oral Pathology, Pathology and Radiation Oncology, University of California San Francisco, San Francisco, California Diseases of the Mouth and Salivary Glands

Robert D. Inman, MD Professor of Medicine and Immunology, University of Toronto; Staff Rheumatologist, University Health Network, Toronto, Ontario, Canada The Spondyloarthropathies

Ralph F. Józefowicz, MD Professor, Neurology and Medicine, University of Rochester, Rochester, New York Approach to the Patient with Neurologic Disease

Sharon K. Inouye, MD, MPH Professor of Medicine, Harvard Medical School; Director, Aging Brain Center, Institute for Aging Research, Hebrew SeniorLife, Boston, Massachusetts Neuropsychiatric Aspects of Aging; Delirium or Acute Mental Status Change in the Older Patient

Stephen G. Kaler, MD Senior Investigator and Head, Section on Translational Neuroscience, Molecular Medicine Program, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland Wilson Disease

Geoffrey K. Isbister, MD, BSc Associate Professor, Clinical Toxicologist, Calvary Mater Newcastle, Callaghan, Senior Research Academic, School of Medicine and Public Health, University of Newcastle, New South Wales, Australia Envenomation Michael G. Ison, MD, MS Associate Professor in Medicine-Infectious Diseases and Surgery-Organ Transplantation, Northwestern University Feinberg School of Medicine, Chicago, Illinois Adenovirus Diseases Elias Jabbour, MD Associate Professor, Department of Leukemia, Division of Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas The Chronic Leukemias Michael R. Jaff, DO Professor of Medicine, Harvard Medical School, Chair, Institute for Heart, Vascular, and Stroke Care, Massachusetts General Hospital, Boston, Massachusetts Other Peripheral Arterial Diseases Joanna C. Jen, MD, PhD Professor of Neurology, University of California Los Angeles School of Medicine, Los Angeles, California Neuro-Ophthalmology; Smell and Taste; Hearing and Equilibrium Dennis M. Jensen, MD Professor of Medicine, David Geffen School of Medicine at University of California Los Angeles; Staff Physician, Medicine-GI, VA Greater Los Angeles Healthcare System; Key Investigator, Director, Human Studies Core & GI Hemostasis Research Unit, CURE Digestive Diseases Research Center, Los Angeles, California Gastrointestinal Hemorrhage Michael D. Jensen, MD Professor of Medicine, Endocrine Research Unit, Director, Obesity Treatment Research Program, Mayo Clinic, Rochester, Minnesota Obesity Robert T. Jensen, MD Chief, Cell Biology Section, Digestive Disease Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Clinical Center, Bethesda, Maryland Pancreatic Neuroendocrine Tumors Stuart Johnson, MD Professor of Medicine, Loyola University Chicago Stritch School of Medicine; Associate Chief of Staff for Research, Edward Hines, Jr. VA Hospital, Hines, Illinois Clostridial Infections

Moses R. Kamya, MB ChB, MMed, MPH, PhD Chairman, Department of Medicine, Makerere University College of Health Sciences, Kampala, Uganda Malaria Louise W. Kao, MD Associate Professor of Emergency Medicine, Department of Emergency Medicine, Indiana University School of Medicine, Indianapolis, Indiana Chronic Poisoning: Trace Metals and Others Steven A. Kaplan, MD E. Darracott Vaughan, Jr. Professor of Urology, Chief, Institute for Bladder and Prostate Health, Weill Cornell Medical College; Director, Iris Cantor Men’s Health Center, NewYork–Presbyterian Hospital, New York, New York Benign Prostatic Hyperplasia and Prostatitis Daniel L. Kastner, MD, PhD Scientific Director, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland The Systemic Autoinflammatory Diseases Sekar Kathiresan, MD Associate Professor in Medicine, Harvard Medical School; Director, Preventive Cardiology, Massachusetts General Hospital, Boston, Massachusetts The Inherited Basis of Common Diseases David A. Katzka, MD Professor of and Consultant in Medicine, Gastroenterology, Mayo Clinic, Rochester, Minnesota Diseases of the Esophagus Debra K. Katzman, MD Professor of Pediatrics, Senior Associate Scientist, The Research Institute, The Hospital for Sick Children and University of Toronto, Toronto, Ontario, Canada Adolescent Medicine Carol A. Kauffman, MD Professor of Internal Medicine, University of Michigan Medical School; Chief, Infectious Diseases Section, Veterans Affairs Ann Arbor Healthcare System, Ann Arbor, Michigan Histoplasmosis; Blastomycosis; Paracoccidioidomycosis; Cryptococcosis; Sporotrichosis; Candidiasis Kenneth Kaushansky, MD Senior Vice President for Health Sciences, Dean, School of Medicine, Stony Brook University, Stony Brook, New York Hematopoiesis and Hematopoietic Growth Factors Keith S. Kaye, MD, MPH Professor of Medicine, Division of Infectious Diseases, Wayne State University School of Medicine, Detroit, Michigan Diseases Caused by Acinetobacter and Stenotrophom*onas Species

Contributors Armand Keating, MD Professor of Medicine, Director, Division of Hematology, Epstein Chair in Cell Therapy and Transplantation, Professor, Institute of Biomaterials and Biomedical Engineering, University of Toronto, Toronto, Ontario, Canada Hematopoietic Stem Cell Transplantation Robin K. Kelley, MD Assistant Professor of Medicine, University of California San Francisco, Helen Diller Family Comprehensive Cancer Center, San Francisco, California Liver and Biliary Tract Cancers Morton Kern, MD Chief of Medicine, VA Long Beach Health Care System School of Medicine; Professor of Medicine, Associate Chief, Cardiology, University of California–Irvine, Irvine, California Catheterization and Angiography Gerald T. Keusch, MD Professor of Medicine and International Health and Public Health, Boston University School of Medicine, Boston, Massachusetts Shigellosis Fadlo R. Khuri, MD Professor and Chair, Hematology and Medical Oncology, Deputy Director, Winship Cancer Institute, Emory University, Atlanta, Georgia Lung Cancer and Other Pulmonary Neoplasms David H. Kim, MD Vice Chair of Education, Professor of Radiology, Section of Abdominal Imaging, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin Diagnostic Imaging Procedures in Gastroenterology

xix

Kevin M. Korenblat, MD Associate Professor of Medicine, Department of Medicine, Washington University School of Medicine, St. Louis, Missouri Approach to the Patient with Jaundice or Abnormal Liver Tests Bruce R. Korf, MD, PhD Wayne H. and Sara Crews Finley Chair in Medical Genetics, Professor and Chair, Department of Genetics, University of Alabama at Birmingham, Birmingham, Alabama Principles of Genetics Neil J. Korman, MD, PhD Professor, Dermatology, Case Western Reserve University School of Medicine, University Hospitals Case Medical Center, Cleveland, Ohio Macular, Papular, Vesiculobullous, and Pustular Diseases Mark G. Kortepeter, MD, MPH Associate Dean for Research, Associate Professor of Preventive Medicine and Medicine, Consultant to the Army Surgeon General for Biodefense; Office of the Dean, Edward Hébert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland Bioterrorism Joseph A. Kovacs, MD Senior Investigator and Head, AIDS Section, Critical Care Medicine Department, National Institutes of Health, Bethesda, Maryland Pneumocystis Pneumonia Thomas O. Kovacs, MD Professor of Medicine, Division of Digestive Diseases, David Geffen School of Medicine at University of California Los Angeles, Los Angeles, California Gastrointestinal Hemorrhage

Matthew Kim, MD Instructor of Medicine, Harvard Medical School; Associate Physician, Brigham and Women’s Hospital, Boston, Massachusetts Thyroid

Monica Kraft, MD Professor of Medicine, Duke University School of Medicine; Chief, Division of Pulmonary, Allergy, and Critical Care Medicine, Duke University Medical Center, Durham, North Carolina Approach to the Patient with Respiratory Disease

Louis V. Kirchhoff, MD, MPH Professor, Departments of Internal Medicine (Infectious Diseases) and Epidemiology, University of Iowa Health Care; Staff Physician, Medical Service, Department of Veterans Affairs Medical Center, Iowa City, Iowa Chagas Disease

Christopher M. Kramer, MD Ruth C. Heede Professor of Cardiology, Professor of Radiology, Director, Cardiovascular Imaging Center, University of Virginia Health System, Charlottesville, Virginia Noninvasive Cardiac Imaging

David S. Knopman, MD Professor of Neurology, Mayo Clinic College of Medicine, Rochester, Minnesota Regional Cerebral Dysfunction: Higher Mental Function; Alzheimer Disease and Other Dementias

Donna M. Krasnewich, MD, PhD Program Director, National Institute of General Medical Sciences, National Institutes of Health, Bethesda, Maryland The Lysosomal Storage Diseases

Tamsin A. Knox, MD, MPH Associate Professor of Medicine, Nutrition/Infection Unit, Tufts University School of Medicine, Boston, Massachusetts Gastrointestinal Manifestions of HIV and AIDS D.P. Kontoyiannis, MD, ScD Professor, Department of Infectious Diseases, Infection Control and Employee Health, The University of Texas MD Anderson Cancer Center, Houston, Texas Mucormycosis; Mycetoma Barbara S. Koppel, MD Professor of Clinical Neurology, New York Medical College, Chief of Neurology, Metropolitan Hospital Center, New York City Health and Hospital Corporation, New York, New York Nutritional and Alcohol-Related Neurologic Disorders

Peter J. Krause, MD Senior Research Scientist in Epidemiology, Medicine, and Pediatrics, Yale School of Public Health and Yale School of Medicine, New Haven, Connecticut Babesiosis and Other Protozoan Diseases John F. Kuemmerle, MD Chair, Division of Gastroenterology, Hepatology, and Nutrition, Professor of Medicine, and Physiology and Biophysics, Center for Digestive Health, Virginia Commonwealth University, Richmond, Virginia Inflammatory and Anatomic Diseases of the Intestine, Peritoneum, Mesentery, and Omentum Ernst J. Kuipers, MD, PhD Professor of Medicine, Department of Gastroenterology and Hepatology, Chief Executive Officer, Erasmus MC University Medical Center, Rotterdam, The Netherlands Acid Peptic Disease

xx

Contributors

Paul W. Ladenson, MD Professor of Medicine, Pathology, Oncology, and Radiology and Radiological Sciences, John Eager Howard Professor of Endocrinology and Metabolism, University Distinguished Service Professor, The Johns Hopkins University School of Medicine; Physician and Division Director, The Johns Hopkins Hospital, Baltimore, Maryland Thyroid Daniel Laheru, MD Ian T. MacMillan Professorship in Clinical Pancreatic Research, Medical Oncology, Johns Hopkins University School of Medicine, Baltimore, Maryland Pancreatic Cancer Donald W. Landry, MD, PhD Samuel Bard Professor of Medicine, Chair, Department of Medicine, Physician-in-Chief, NewYork-Presbyterian Hospital/Columbia University Medical Center, New York, New York Approach to the Patient with Renal Disease Anthony E. Lang, MD Director, Division of Neurology, Jack Clark Chair for Research in Parkinson’s Disease, University of Toronto; Director, Morton and Gloria Shulman Movement Disorders Clinic and the Edmond J. Safra Program in Parkinson’s Disease and the Lily Safra Chair in Movement Disorders, Toronto Western Hospital, Toronto, Ontario, Canada Parkinsonism; Other Movement Disorders

Gary R. Lichtenstein, MD Professor of Medicine, Perelman School of Medicine at the University of Pennsylvania, Director, Center for Inflammatory Bowel Disease, University of Pennsylvania, Philadelphia, Pennsylvania Inflammatory Bowel Disease Henry W. Lim, MD Chairman and C.S. Livingood Chair, Department of Dermatology, Henry Ford Hospital; Senior Vice President for Academic Affairs, Henry Ford Health System, Detroit, Michigan Eczemas, Photodermatoses, Papulosquamous (Including Fungal) Diseases, and Figurate Erythemas Aldo A.M. Lima, MD, PhD Professor of Medicine and Pharmacology, School of Medicine, Federal University of Ceará, Fortaleza, Ceará, Brazil Cryptosporidiosis; Amebiasis Geoffrey S.F. Ling, MD, PhD Professor of Neurology, Uniformed Services University of the Health Sciences, Bethesda, Maryland Traumatic Brain Injury and Spinal Cord Injury William C. Little, MD Patrick Lehan Professor of Cardiovascular Medicine, Chair, Department of Medicine, University of Mississippi Medical Center, Jackson, Mississippi Pericardial Diseases

Richard A. Lange, MD, MBA President and Dean, Paul L. Foster School of Medicine, Texas Tech University Health Sciences Center El Paso, El Paso, Texas Acute Coronary Syndrome: Unstable Angina and Non-ST Elevation Myocardial Infarction

Donald M. Lloyd-Jones, MD, ScM Senior Associate Dean, Chair, Department of Preventive Medicine, Eileen M. Foell Professor of Preventive Medicine and Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois Epidemiology of Cardiovascular Disease

Frank A. Lederle, MD Core Investigator, Center for Chronic Disease Outcomes Research, Minneapolis VA Medical Center; Professor of Medicine, University of Minnesota School of Medicine, Minneapolis, Minnesota Diseases of the Aorta

Bennett Lorber, MD Thomas M. Durant Professor of Medicine and Professor of Microbiology and Immunology, Temple University School of Medicine, Philadelphia, Pennsylvania Listeriosis

Thomas H. Lee, MD, MSc Senior Physician, Department of Medicine, Brigham and Women’s Hospital; Chief Medical Officer, Press Ganey, Boston, Massachusetts Using Data for Clinical Decisions

Donald E. Low, MD† Nonpneumococcal Streptococcal Infections, Rheumatic Fever

William M. Lee, MD Professor of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, Texas Toxin- and Drug-Induced Liver Disease James E. Leggett, MD Associate Professor, Department of Medicine, Oregon Health and Science University; Infectious Diseases, Department of Medical Education, Providence Portland Medical Center, Portland, Oregon Approach to Fever or Suspected Infection in the Normal Host Stuart Levin, MD Professor of Medicine, Emeritus Chairman, Department of Medicine, Rush University Medical Center, Chicago, Illinois Zoonoses Stephanie M. Levine, MD Professor of Medicine, Division of Pulmonary Diseases and Critical Care Medicine, The University of Texas Health Science Center San Antonio, South Texas Veterans Health Care System, San Antonio, Texas Alveolar Filling Disorders

Daniel R. Lucey, MD, MPH Adjunct Professor, Microbiology and Immunology, Georgetown University Medical Center, Washington, D.C. Anthrax James R. Lupski, MD, PhD Cullen Professor of Molecular and Human Genetics, Professor of Pediatrics, Baylor College of Medicine and Texas Children’s Hospital, Houston, Texas Gene, Genomic, and Chromosomal Disorders Jeffrey M. Lyness, MD Senior Associate Dean for Academic Affairs, Professor of Psychiatry and Neurology, University of Rochester School of Medicine and Dentistry, Rochester, New York Psychiatric Disorders in Medical Practice Bruce W. Lytle, MD Chair, Heart and Vascular Institute, Professor of Surgery, Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, Ohio Interventional and Surgical Treatment of Coronary Artery Disease

†

Deceased.

Contributors C. Ronald MacKenzie, MD Assistant Attending Physician, Department of Medicine-Rheumatology, C. Ronald MacKenzie Chair in Ethics and Medicine, Hospital for Special Surgery, Associate Professor of Clinical Medicine and Medical Ethics, Weill Cornell Medical College of Cornell University, New York, New York Surgical Treatment of Joint Disease Harriet L. MacMillan, MD, MSc Professor, Departments of Psychiatry and Behavioural Neurosciences, and Pediatrics, Chedoke Health Chair in Child Psychiatry, Offord Centre for Child Studies, McMaster University, Hamilton, Ontario, Canada Intimate Partner Violence Robert D. Madoff, MD Professor of Surgery, Stanley M. Goldberg, MD, Chair, Colon and Rectal Surgery, University of Minnesota, Minneapolis, Minnesota Diseases of the Rectum and Anus Frank Maldarelli, MD, PhD Head, Clinical Retrovirology Section, HIV Drug Resistance Program, National Cancer Institute, National Institutes of Health, Bethesda, Maryland Biology of Human Immunodeficiency Viruses Atul Malhotra, MD Chief of Pulmonary and Critical Care, Kenneth M. Moser Professor of Medicine, Director of Sleep Medicine, University of California San Diego, La Jolla, California Disorders of Ventilatory Control Mark J. Manary, MD Helene B. Roberson Professor of Pediatrics, Washington University School of Medicine; Attending Physician, St. Louis Children’s Hospital, St. Louis, Missouri; Adjunct Professor, Children’s Nutrition Research Center, Baylor College of Medicine, Houston, Texas; Senior Lecturer in Community Health, University of Malawi College of Medicine, Blantyre, Malawi Protein-Energy Malnutrition Donna Mancini, MD Professor of Medicine, Department of Medicine, Division of Cardiology, Columbia University College of Physicians and Surgeons, Center for Advanced Cardiac Care, Columbia University Medical Center, New York, New York Cardiac Transplantation Lionel A. Mandell, MD Professor of Medicine, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada Streptococcus Pneumoniae Infections Peter Manu, MD Professor of Medicine and Psychiatry, Hofstra North Shore–LIJ School of Medicine at Hofstra University, Hempstead, New York; Adjunct Professor of Clinical Medicine, Psychiatry and Behavioral Sciences, Albert Einstein College of Medicine, Bronx, New York; Director of Medical Services, Zucker Hillside Hospital, Glen Oaks, New York Medical Consultation in Psychiatry Ariane Marelli, MD, MPH Professor of Medicine, McGill University, Director, McGill Adult Unit for Congenital Heart Disease, Associate Director, Academic Affairs and Research, Cardiology, McGill University Health Centre, Montreal, Québec, Canada Congenital Heart Disease in Adults Xavier Mariette, MD, PhD Professor, Rheumatology, Université Paris-Sud, AP-HP, Le Kremlin Bicêtre, France Sjögren Syndrome

xxi

Andrew R. Marks, MD Wu Professor and Chair, Department of Physiology and Cellular Biophysics, Founding Director, Helen and Clyde Wu Center for Molecular Cardiology, Columbia University College of Physicians and Surgeons, New York, New York Cardiac Function and Circulatory Control Kieren A. Marr, MD Professor of Medicine and Oncology, The Johns Hopkins University, Director, Transplant and Oncology Infectious Diseases, Baltimore, Maryland Approach to Fever and Suspected Infection in the Compromised Host Thomas J. Marrie, MD Dean, Faculty of Medicine, Dalhousie University; Professor of Medicine, Capital District Health Authority, Halifax, Nova Scotia, Canada Legionella Infections Paul Martin, MD Professor of Medicine and Chief, Division of Hepatology, Miller School of Medicine, University of Miami, Miami, Florida Approach to the Patient with Liver Disease Joel B. Mason, MD Professor of Medicine and Nutrition, Tufts University; Staff Physician, Divisions of Gastroenterology and Clinical Nutrition, Tufts Medical Center, Boston, Massachusetts Vitamins, Trace Minerals, and Other Micronutrients Henry Masur, MD Chief, Critical Care Medicine Department, Clinical Center, National Institutes of Health, Bethesda, Maryland Infectious and Metabolic Complications of HIV and AIDS Eric L. Matteson, MD, MPH Professor of Medicine, Mayo Clinic College of Medicine, Consultant, Divisions of Rheumatology and Epidemiology, Mayo Clinic, Rochester, Minnesota Infections of Bursae, Joints, and Bones Michael A. Matthay, MD Professor, Departments of Medicine and Anesthesia, University of California San Francisco, San Francisco, California Acute Respiratory Failure Toby A. Maurer, MD Professor of Dermatology, University of California San Francisco; Chief of Dermatology, San Francisco General Hospital, San Francisco, California Skin Manifestations in Patients with Human Immunodeficiency Virus Infection Emeran A. Mayer, MD, PhD Professor of Medicine, Physiology, and Psychiatry, Division of Digestive Diseases, Department of Medicine, University of California Los Angeles, Los Angeles, California Functional Gastrointestinal Disorders: Irritable Bowel Syndrome, Dyspepsia, Chest Pain of Presumed Esophageal Origin, and Heartburn Stephan A. Mayer, MD Director, Institute for Critical Care Medicine, Icahn School of Medicine at Mount Sinai, New York, New York Hemorrhagic Cerebrovascular Disease Stephen A. McClave, MD Professor of Medicine, Director of Clinical Nutrition, University of Louisville School of Medicine, Louisville, Kentucky Enteral Nutrition F. Dennis McCool, MD Professor of Medicine, The Warren Alpert Medical School of Brown University; Medical Director of Sleep Center, Memorial Hospital of Rhode Island, Pawtucket, Rhode Island Diseases of the Diaphragm, Chest Wall, Pleura, and Mediastinum

xxii

Contributors

Charles E. McCulloch, PhD Professor of Biostatistics, Department of Epidemiology and Biostatistics, University of California San Francisco, San Francisco, California Statistical Interpretation of Data

Ernest Moy, MD, MPH Medical Officer, Center for Quality Improvement and Patient Safety Agency for Healthcare Research and Quality, Rockville, Maryland Measuring Health and Health Care

William J. McKenna, MD Professor of Cardiology, Institute of Cardiovascular Science, University College London, London, United Kingdom Diseases of the Myocardium and Endocardium

Atis Muehlenbachs, MD, PhD Infectious Diseases Pathology Branch, Centers for Disease Control and Prevention, Atlanta, Georgia Leptospirosis

Vallerie McLaughlin, MD Kim A. Eagle, MD, Endowed Professor of Cardiovascular Medicine, Director, Pulmonary Hypertension Program, University of Michigan, Ann Arbor, Michigan Pulmonary Hypertension

Andrew H. Murr, MD Professor and Chairman, Roger Boles, MD Endowed Chair in Otolaryngology Education, Department of Otolaryngology-Head and Neck Surgery, University of California San Francisco School of Medicine, San Francisco, California Approach to the Patient with Nose, Sinus, and Ear Disorders

John J.V. McMurray, MB, MD Professor of Cardiology, Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, Scotland, United Kingdom Heart Failure: Management and Prognosis Kenneth R. McQuaid, MD Professor of Clinical Medicine, Marvin H. Sleisenger Endowed Chair, Vice Chairman, University of California San Francisco; Chief, Medical Services and Gastroenterology, San Francisco VA Medical Center, San Francisco, California Approach to the Patient with Gastrointestinal Disease Marc Michel, MD Professor of Internal Medicine, Head of the Unit of Internal Medicine at Henri Mondor University Hospital, National Referral Center for Adult’s Immune Cytopenias, Creteil, France Autoimmune and Intravascular Hemolytic Anemias Jonathan W. Mink, MD, PhD Frederick A. Horner, MD Endowed Professor in Pediatric Neurology, Professor of Neurology, Neurobiology & Anatomy, Brain & Cognitive Sciences, and Pediatrics, Chief, Division of Child Neurology, Vice Chair, Department of Neurology, University of Rochester, Rochester, New York Congenital, Developmental, and Neurocutaneous Disorders William E. Mitch, MD Gordon A. Cain Chair in Nephrology, Director of Nephrology, Baylor College of Medicine, Houston, Texas Chronic Kidney Disease Mark E. Molitch, MD Martha Leland Sherwin Professor of Endocrinology, Division of Endocrinology, Metabolism, and Molecular Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois Neuroendocrinology and the Neuroendocrine System; Anterior Pituitary Bruce A. Molitoris, MD Professor of Medicine, and Cellular and Integrative Physiology Director, Indiana Center for Biological Microscopy, Indiana University, Indianapolis, Indiana Acute Kidney Injury Jose G. Montoya, MD Professor of Medicine, Division of Infectious Disease and Geographic Medicine, Stanford University School of Medicine, Stanford, California; Director, Palo Alto Medical Foundation Toxoplasma Serology Laboratory, National Reference Center for the Study and Diagnosis of Toxoplasmosis, Palo Alto, California Toxoplasmosis Alison Morris, MD, MS Associate Professor of Medicine, Clinical Translational Science, and Immunology, Division of Pulmonary, Allergy, and Critical Care Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Pulmonary Manifestations of Human Immunodeficiency Virus and Acquired Immunodeficiency Syndrome

Daniel M. Musher, MD Professor of Medicine, Molecular Virology, and Microbiology, Distinguished Service Professor, Baylor College of Medicine, Infectious Disease Section, Michael E. DeBakey Veterans Affairs Medical Center, Houston, Texas Overview of Pneumonia Robert J. Myerburg, MD Professor of Medicine and Physiology, Division of Cardiology, Department of Medicine, American Heart Association Chair in Cardiovascular Research, University of Miami Miller School of Medicine, Miami, Florida Approach to Cardiac Arrest and Life-Threatening Arrhythmias Sandesh C.S. Nagamani, MD Assistant Professor, Department of Molecular and Human Genetics, Director, Clinic for Metabolic and Genetic Disorders of Bone, Baylor College of Medicine and Texas Children’s Hospital, Houston, Texas Gene, Genomic, and Chromosomal Disorders Stanley J. Naides, MD Medical Director and Interim Scientific Director, Immunology, Quest Diagnostics Nichols Institute, San Juan Capistrano, California Arboviruses Causing Fever and Rash Syndromes Yoshifumi Naka, MD, PhD Professor of Surgery, Department of Surgery, Columbia University College of Physicians and Surgeons, New York, New York Cardiac Transplantation Theodore E. Nash, MD Principal Investigator, Clinical Parasitology Section, Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Giardiasis Avindra Nath, MD Chief, Section of Infections of the Nervous System, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland Cytomegalovirus, Epstein-Barr Virus, and Slow Virus Infections of the Central Nervous System; Neurologic Complications of Human Immunodeficiency Virus Infection; Meningitis: Bacterial, Viral, and Other; Brain Abscess and Parameningeal Infections Eric G. Neilson, MD Vice President for Medical Affairs and Lewis Landsberg Dean, Northwestern University Feinberg School of Medicine, Northwestern Memorial Hospital, Chicago, Illinois Tubulointerstitial Nephritis

Contributors Lawrence S. Neinstein, MD Professor of Pediatrics and Medicine, Keck School of Medicine of USC; Executive Director, Engemann Student Health Center, Division Head of College Health, Assistant Provost, Student Health and Wellness, University of Southern California, Los Angeles, California Adolescent Medicine Lewis S. Nelson, MD Professor of Emergency Medicine, Director, Fellowship in Medical Toxicology, New York University School of Medicine; Attending Physician, New York University Langone Medical Center and Bellevue Hospital Center, New York, New York Acute Poisoning Eric J. Nestler, MD, PhD Nash Family Professor and Chair, Department of Neuroscience, Director, The Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York, New York Biology of Addiction Anne B. Newman, MD, MPH Professor of Epidemiology, The University of Pittsburgh Graduate School of Public Health; Chair, Department of Epidemiology, Director, University of Pittsburgh Center for Aging and Population Health, Pittsburgh, Pennsylvania Epidemiology of Aging: Implications of the Aging of Society Thomas B. Newman, MD, MPH Professor, Epidemiology & Biostatistics and Pediatrics, University of California San Francisco, San Francisco, California Statistical Interpretation of Data William L. Nichols, MD Associate Professor, Medicine and Laboratory Medicine, Mayo Clinic College of Medicine; Staff Physician, Special Coagulation Laboratory, Comprehensive Hemophilia Center, and Coagulation Clinic, Mayo Clinic, Rochester, Minnesota Von Willebrand Disease and Hemorrhagic Abnormalities of Platelet and Vascular Function Lindsay E. Nicolle, MD Professor of Internal Medicine and Medical Microbiology, University of Manitoba, Health Sciences Centre, Winnipeg, Manitoba, Canada Approach to the Patient with Urinary Tract Infection Lynnette K. Nieman, MD Senior Investigator, Program on Reproductive and Adult Endocrinology, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland Approach to the Patient with Endocrine Disease; Adrenal Cortex; Polyglandular Disorders Dennis E. Niewoehner, MD Professor of Medicine, University of Minnesota; Staff Physician, Minneapolis Veterans Affairs Health Care System, Minneapolis, Minnesota Chronic Obstructive Pulmonary Disease S. Ragnar Norrby, MD, PhD Director General, Swedish Institute for Infectious Disease Control, Solna, Sweden Approach to the Patient with Urinary Tract Infection Susan O’Brien, MD Professor, Department of Leukemia, Division of Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas The Chronic Leukemias Christopher M. O’Connor, MD Professor of Medicine and Chief, Division of Cardiology, Director, Duke Heart Center, Durham, North Carolina Heart Failure: Pathophysiology and Diagnosis

xxiii

Francis G. O’Connor, MD, MPH Professor and Chair, Military and Emergency Medicine, Medical Director, Uniformed Services University Consortium for Health and Military Performance, Bethesda, Maryland Disorders Due to Heat and Cold; Rhabdomyolysis Patrick G. O’Connor, MD, MPH Professor and Chief, General Internal Medicine, Yale University School of Medicine, New Haven, Connecticut Alcohol Abuse and Dependence James R. O’Dell, MD Bruce Professor and Vice Chair of Internal Medicine, Chief, Division of Rheumatology, University of Nebraska Medical Center and Omaha VA Nebraska–Western Iowa Health Care System, Omaha, Nebraska Rheumatoid Arthritis Anne E. O’Donnell, MD Professor of Medicine, Chief, Division of Pulmonary, Critical Care, and Sleep Medicine, Georgetown University Medical Center, Washington, D.C. Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders Jae K. Oh, MD Professor of Medicine, Director, Echocardiography Core Laboratory and Pericardial Clinic, Division of Cardiovascular Diseases, Co-Director, Integrated Cardiac Imaging, Division of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota Pericardial Diseases Jeffrey E. Olgin, MD Gallo-Chatterjee Distinguished Professor of Medicine, Chief, Division of Cardiology, Co-Director, Heart and Vascular Center, University of California San Francisco, San Francisco, California Approach to the Patient with Suspected Arrhythmia Walter A. Orenstein, MD Professor of Medicine, Pediatrics, and Global Health, Emory University School of Medicine, Atlanta, Georgia Immunization Douglas R. Osmon, MD, MPH Professor of Medicine, Mayo Clinic College of Medicine; Consultant, Division of Infectious Diseases, Mayo Clinic, Rochester, Minnesota Infections of Bursae, Joints, and Bones Catherine M. Otto, MD J. Ward Kennedy-Hamilton Endowed Chair in Cardiology, Professor of Medicine, University of Washington School of Medicine; Director, Heart Valve Clinic, University of Washington Medical Center, Seattle, Washington Echocardiography Mark Papania, MD, MPH Medical Epidemiologist, Division of Viral Diseases, Measles, Mumps, Rubella, and Herpes Virus Laboratory Branch, Centers for Disease Control and Prevention, Atlanta, Georgia Measles Peter G. Pappas, MD Professor of Medicine, University of Alabama at Birmingham, Birmingham, Alabama Dematiaceous Fungal Infections Pankaj Jay Pasricha, MD Director, The Johns Hopkins Center for Neurogastroenterology; Professor of Medicine and Neurosciences, The Johns Hopkins School of Medicine; Professor of Innovation Management, Johns Hopkins Carey Business School, Baltimore, Maryland Gastrointestinal Endoscopy

xxiv

Contributors

David L. Paterson, MD Professor of Medicine, University of Queensland Centre for Clinical Research, Royal Brisbane and Women’s Hospital Campus, Brisbane, Queensland, Australia Infections Due to Other Members of the Enterobacteriaceae, Including Management of Multidrug Resistant Strains Carlo Patrono, MD Professor and Chair of Pharmacology, Department of Pharmacology, Catholic University School of Medicine, Rome, Italy Prostaglandin, Aspirin, and Related Compounds Jean-Michel Pawlotsky, MD, PhD Professor of Medicine, The University of Paris-Est; Director, National Reference Center for Viral Hepatitis B, C, and Delta and Department of Virology, Henri Mondor University Hospital; Director, Department of Molecular Virology and Immunology, Institut Mondor de Recherche Biomédicale, Créteil, France Acute Viral Hepatitis; Chronic Viral and Autoimmune Hepatitis Richard D. Pearson, MD Professor of Medicine and Pathology, University of Virginia School of Medicine and University of Virginia Health System, Charlottesville, Virginia Antiparasitic Therapy Trish M. Perl, MD, MSc Professor of Medicine and Pathology, The Johns Hopkins School of Medicine; Professor of Epidemiology, Johns Hopkins Bloomberg School of Public Health; Infectious Diseases Specialist and Senior Epidemiologist, The Johns Hopkins Hospital and Health System, Baltimore, Maryland Enterococcal Infections Adam Perlman, MD, MPH Associate Professor, Department of Medicine, Duke University Medical Center; Executive Director, Duke Integrative Medicine, Duke University Health System, Durham, North Carolina Complementary and Alternative Medicine William A. Petri, Jr., MD, PhD Wade Hampton Frost Professor, Departments of Medicine, Pathology, Microbiology, Immunology, and Cancer Biology, School of Medicine, University of Virginia; Chief, Division of Infectious Diseases and International Health, University of Virginia Hospitals, Charlottesville, Virginia Relapsing Fever and Other Borrelia Infections; African Sleeping Sickness; Amebiasis Marc A. Pfeffer, MD, PhD Dzau Professor of Medicine, Harvard Medical School; Senior Physician, Cardiovascular Division, Brigham and Women’s Hospital, Boston, Massachusetts Heart Failure: Management and Prognosis Perry J. Pickhardt, MD Professor of Radiology and Chief, Gastrointestinal Imaging, Section of Abdominal Imaging, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin Diagnostic Imaging Procedures in Gastroenterology David S. Pisetsky, MD, PhD Chief of Rheumatology, Medical Research Service, Durham VA Medical Center; Professor of Medicine and Immunology, Department of Medicine, Duke University Medical Center, Durham, North Carolina Laboratory Testing in the Rheumatic Diseases Marshall R. Posner, MD Professor of Medicine, Director of Head and Neck Medical Oncology, Director of the Office of Cancer Clinical Trials, The Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, New York Head and Neck Cancer

Frank Powell, PhD Professor of Medicine, Chief of Physiology, University of California San Diego, La Jolla, California Disorders of Ventilatory Control Reed E. Pyeritz, MD, PhD William Smilow Professor of Medicine and Genetics and Vice Chair for Academic Affairs, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania Inherited Diseases of Connective Tissue Thomas C. Quinn, MD, MSc Associate Director for International Research, Head, Section of International HIV/AIDS Research, Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health; Professor of Medicine, Pathology, International Health, Molecular Microbiology and Immunology, and Epidemiology, The Johns Hopkins Medical Institutions, Baltimore, Maryland Epidemiology and Diagnosis of Human Immunodeficiency Virus Infection and Acquired Immunodeficiency Syndrome Jai Radhakrishnan, MD, MS Professor of Medicine, Division of Nephrology, Department of Medicine, Columbia University Medical Center; Associate Division Chief for Clinical Affairs, Division of Nephrology, New York-Presbyterian Hospital, New York, New York Glomerular Disorders and Nephrotic Syndromes Petros I. Rafailidis, MD, PhD, MSc Senior Researcher, Alfa Institute of Biomedical Sciences, Attending Physician, Department of Medicine and Hematology, Athens Medical Center, Athens Medical Group, Athens, Greece Pseudomonas and Related Gram-Negative Bacillary Infections Ganesh Raghu, MD Adjunct Professor of Medicine and Laboratory Medicine, University of Washington, Director, CENTER for Interstitial Lung Diseases at the University of Washington; Co-Director, Scleroderma Clinic, University of Washington Medical Center, Seattle, Washington Interstitial Lung Disease Margaret Ragni, MD, MPH Professor of Medicine and Clinical Translational Science, Department of Hematology/Oncology, University of Pittsburgh Medical Center; Director, Hemophilia Center of Western Pennsylvania, Pittsburgh, Pennsylvania Hemorrhagic Disorders: Coagulation Factor Deficiencies Srinivasa N. Raja, MD Professor of Anesthesiology and Neurology, Director, Division of Pain Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland Pain S. Vincent Rajkumar, MD Professor of Medicine, Division of Hematology, Mayo Clinic, Rochester, Minnesota Plasma Cell Disorders Stuart H. Ralston, MB ChB, MD Professor of Rheumatology, Institute of Genetics and Molecular Medicine, Western General Hospital, The University of Edinburgh, Edinburgh, United Kingdom Paget Disease of Bone Didier Raoult, MD, PhD Professor, Aix Marseille Université, Faculté de Médecine; Chief, Hôpital de la Timone, Fédération de Microbiologie Clinique, Marseille, France Bartonella Infections; Rickettsial Infections

Contributors Robert W. Rebar, MD Professor, Department of Obstetrics and Gynecology, Western Michigan University Homer Stryker MD School of Medicine, Kalamazoo, Michigan Ovaries and Development; Reproductive Endocrinology and Infertility Annette C. Reboli, MD Founding Vice Dean, Professor of Medicine, Cooper Medical School of Rowan University, Cooper University Healthcare, Department of Medicine, Division of Infectious Diseases, Camden, New Jersey Erysipelothrix Infections

xxv

Karen Rosene-Montella, MD Professor and Vice Chair of Medicine, Director of Obstetric Medicine, The Warren Alpert Medical School of Brown University; Senior Vice President, Women’s Services and Clinical Integration, Lifespan Health System, Providence, Rhode Island Common Medical Problems in Pregnancy Philip J. Rosenthal, MD Professor, Department of Medicine, University of California San Francisco, San Francisco, California Malaria

K. Rajender Reddy, MD Professor of Medicine, Professor of Medicine in Surgery, Perelman School of Medicine at the University of Pennsylvania; Director of Hepatology, Director, Viral Hepatitis Center, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Bacterial, Parasitic, Fungal, and Granulomatous Liver Diseases

Marc E. Rothenberg, MD, PhD Director, Division of Allergy and Immunology, Director, Cincinnati Center for Eosinophilic Disorders; Professor of Pediatrics, Cincinnati Children’s Hospital Medical Center, University of Cincinnati College of Medicine, Cincinnati, Ohio Eosinophilic Syndromes

Donald A. Redelmeier, MD Professor of Medicine, University of Toronto; Senior Scientist and Staff Physician, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada Postoperative Care and Complications

James A. Russell, MD Professor of Medicine, University of British Columbia; Associate Director, Intensive Care Unit, St. Paul’s Hospital, Vancouver, British Columbia, Canada Shock Syndromes Related to Sepsis

Susan E. Reef, MD Centers for Disease Control and Prevention, Atlanta, Georgia Rubella (German Measles)

Anil K. Rustgi, MD T. Grier Miller Professor of Medicine and Genetics, Chief of Gastroenterology, American Cancer Society; Professor, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania Neoplasms of the Esophagus and Stomach

Neil M. Resnick, MD Thomas P. Detre Endowed Chair in Gerontology and Geriatric Medicine, Professor of Medicine and Division Chief, Geriatrics, Associate Director, University of Pittsburgh Institute on Aging, University of Pittsburgh; Chief, Division of Geriatric Medicine and Gerontology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Incontinence David B. Reuben, MD Director, Multicampus Program in Geriatric Medicine and Gerontology; Chief, Division of Geriatrics, Archstone Professor of Medicine, David Geffen School of Medicine at University of California Los Angeles, Los Angeles, California Geriatric Assessment Emanuel P. Rivers, MD, MPH Professor and Vice Chairman of Emergency Medicine, Wayne State University; Senior Staff Attending, Critical Care and Emergency Medicine, Henry Ford Hospital, Detroit, Michigan Approach to the Patient with Shock Joseph G. Rogers, MD Professor of Medicine, Senior Vice Chief for Clinical Affairs, Division of Cardiology, Durham, North Carolina Heart Failure: Pathophysiology and Diagnosis Jean-Marc Rolain, PharmD, PhD Professor, Institut Hospitalo-Universitaire Méditerranée-Infection, Aix-Marseille Université, Marseille, France Bartonella Infections José R. Romero, MD Professor of Pediatrics, University of Arkansas for Medical Sciences, Horace C. Cabe Professor of Infectious Diseases; Director, Section of Pediatric Infectious Diseases, Arkansas Children’s Hospital, Little Rock, Arkansas Enteroviruses

Daniel E. Rusyniak, MD Professor of Emergency Medicine, Adjunct Professor of Neurology and Pharmacology and Toxicology, Department of Emergency Medicine, Indiana University School of Medicine, Indianapolis, Indiana Chronic Poisoning: Trace Metals and Others Robert A. Salata, MD Professor and Executive Vice Chair, Department of Medicine, Chief, Division of Infectious Diseases and HIV Medicine, Case Western Reserve University, University Hospitals Case Medical Center, Cleveland, Ohio Brucellosis Jane E. Salmon, MD Collette Kean Research Chair, Hospital for Special Surgery, Professor of Medicine, Weill Cornell Medical College, New York, New York Mechanisms of Immune-Mediated Tissue Injury Edsel Maurice T. Salvana, MD, DTM&H Associate Professor of Medicine, Section of Infectious Diseases, Department of Medicine, Philippine General Hospital; Director, Institute of Molecular Biology and Biotechnology, National Institutes of Health, University of the Philippines Manila, Manila, Philippines Brucellosis Renato M. Santos, MD Associate Professor, Cardiology, Wake Forest School of Medicine, Winston-Salem, North Carolina Vascular Disorders of the Kidney Michael N. Sawka, PhD Professor, School of Applied Physiology, Georgia Institute of Technology, Atlanta, Georgia Disorders Due to Heat and Cold Paul D. Scanlon, MD Professor of Medicine, Division of Pulmonary and Critical Care Medicine, Mayo Clinic, Rochester, Minnesota Respiratory Function: Mechanisms and Testing

xxvi

Contributors

Carla Scanzello, MD, PhD Assistant Professor of Medicine, Division of Rheumatology, Perelman School of Medicine at the University of Pennsylvania and Translational Musculoskeletal Research Center, Philadelphia Veterans Affairs Medical Center, Philadelphia, Pennsylvania Osteoarthritis Andrew I. Schafer, MD Professor of Medicine, Director, Richard T. Silver Center for Myeloproliferative Neoplasms, Weill Cornell Medical College, New York, New York Approach to Medicine, the Patient, and the Medical Profession: Medicine as a Learned and Humane Profession; Approach to the Patient with Bleeding and Thrombosis; Hemorrhagic Disorders: Disseminated Intravascular Coagulation, Liver Failure, and Vitamin K Deficiency; Thrombotic Disorders: Hypercoagulable States William Schaffner, MD Professor and Chair, Department of Preventive Medicine, Department of Health Policy; Professor of Medicine (Infectious Diseases), Vanderbilt University School of Medicine, Nashville, Tennessee Tularemia and Other Francisella Infections W. Michael Scheld, MD Bayer-Gerald L. Mandell Professor of Infectious Diseases, Professor of Medicine, Clinical Professor of Neurosurgery, Director, Pfizer Initiative in International Health, University of Virginia Health System, Charlottesville, Virginia Introduction to Microbial Disease: Host-Pathogen Interactions Manuel Schiff, MD Professor, Université Paris 7 Denis Diderot, Sorbonne Paris Cité, Head of Metabolic Unit/Reference Center for Inborn Errors of Metabolism, Robert Debré University Hospital, APHP, Paris, France hom*ocystinuria and Hyperhom*ocysteinemia Michael L. Schilsky, MD Associate Professor, Medicine and Surgery, Yale University School of Medicine, New Haven, Connecticut Wilson Disease Robert T. Schooley, MD Professor and Head, Division of Infectious Diseases, Executive Vice Chair for Academic Affairs, Department of Medicine, University of California San Diego, La Jolla, California Epstein-Barr Virus Infection David L. Schriger, MD, MPH Professor, Department of Emergency Medicine, University of California Los Angeles, Los Angeles, California Approach to the Patient with Abnormal Vital Signs Steven A. Schroeder, MD Distinguished Professor of Health and Healthcare and of Medicine, University of California San Francisco, San Francisco, California Socioeconomic Issues in Medicine Lynn M. Schuchter, MD Professor of Medicine, University of Pennsylvania; Chief, Hematology/ Oncology Division, Program Leader, Melanoma and Cutaneous Malignancies Program, Abramson Cancer Center, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Melanoma and Nonmelanoma Skin Cancers Sam Schulman, MD, PhD Professor, Division of Hematology and Thromboembolism, Director of Clinical Thromboembolism Program, Department of Medicine, McMaster University, Hamilton, Ontario, Canada Antithrombotic Therapy

Lawrence B. Schwartz, MD, PhD Charles and Evelyn Thomas Professor of Medicine, Internal Medicine, Virginia Commonwealth University, Richmond, Virginia Systemic Anaphylaxis, Food Allergy, and Insect Sting Allergy Carlos Seas, MD Associate Professor of Medicine, Vice Director, Alexander von Humboldt Tropical Medicine Institute, Universidad Peruana Cayetano Heredia; Attending Physician, Department of Infectious, Tropical, and Dermatologic Diseases, National Hospital Cayetano Heredia, Lima, Peru Cholera and Other Vibrio Infections Steven A. Seifert, MD Professor of Emergency Medicine, University of New Mexico School of Medicine, Medical Director, New Mexico Poison and Drug Information Center, University of New Mexico Health Sciences Center, Albuquerque, New Mexico Envenomation Julian L. Seifter, MD Associate Professor of Medicine, Harvard Medical School; Senior Physician, Brigham and Women’s Hospital, Boston, Massachusetts Potassium Disorders; Acid-Base Disorders Duygu Selcen, MD Associate Professor of Neurology and Pediatrics, Department of Neurology, Mayo Clinic, Rochester, Minnesota Muscle Diseases Clay F. sem*nkovich, MD Herbert S. Gasser Professor and Chief, Division of Endocrinology, Metabolism and Lipid Research, Washington University School of Medicine, St. Louis, Missouri Disorders of Lipid Metabolism Carol E. Semrad, MD Professor of Medicine, The University of Chicago Medicine, GI Section, Chicago, Illinois Approach to the Patient with Diarrhea and Malabsorption Harry Shamoon, MD Professor of Medicine and Associate Dean for Clinical and Translational Research, Albert Einstein College of Medicine; Director, Harold and Muriel Block Institute for Clinical and Translational Research at Einstein and Montefiore, Bronx, New York Diabetes Mellitus James C. Shaw, MD Associate Professor, Department of Medicine, University of Toronto; Head, Division of Dermatology, Department of Medicine, Women’s College Hospital, Toronto, Ontario, Canada Examination of the Skin and an Approach to Diagnosing Skin Diseases Pamela J. Shaw, DBE, MBBS, MD Professor of Neurology, University of Sheffield, Consultant Neurologist, Royal Hallamshire Hospital, Sheffield, United Kingdom Amyotrophic Lateral Sclerosis and Other Motor Neuron Diseases Robert L. Sheridan, MD Associate Professor of Surgery, Burn Service Medical Director, Boston Shriners Hospital for Children, Massachusetts General Hospital, Division of Burns, Harvard Medical School, Boston, Massachusetts Medical Aspects of Injuries and Burns Stuart Sherman, MD Professor of Medicine and Radiology, Director of ERCP, Indiana University School of Medicine, Indianapolis, Indiana Diseases of the Gallbladder and Bile Ducts

Contributors Michael E. Shy, MD Professor of Neurology, Pediatrics, and Physiology, University of Iowa, Iowa City, Iowa Peripheral Neuropathies

Frederick S. Southwick, MD Professor of Medicine, Division of Infectious Diseases, University of Florida and VF Health, Gainesville, Florida Nocardiosis

Ellen Sidransky, MD Chief, Section on Molecular Neurogenetics, Medical Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland The Lysosomal Storage Diseases

Allen M. Spiegel, MD Dean, Albert Einstein College of Medicine, Bronx, New York Principles of Endocrinology; Polyglandular Disorders

Richard M. Siegel, MD, PhD Clinical Director, National Institute of Arthritis, Musculoskeletal, and Skin Diseases, National Institutes of Health, Bethesda, Maryland The Systemic Autoinflammatory Diseases Robert F. Siliciano, MD, PhD Professor, The Johns Hopkins University School of Medicine, Howard Hughes Medical Institute, Baltimore, Maryland Immunopathogenesis of Human Immunodeficiency Virus Infection Michael S. Simberkoff, MD Chief of Staff, VA New York Harbor Healthcare System; Professor of Medicine, NYU School of Medicine, New York, New York Haemophilus and Moraxella Infections David L. Simel, MD, MHS Professor of Medicine, Duke University; Chief, Medical Service, Durham Veterans Affairs Medical Center, Durham, North Carolina Approach to the Patient: History and Physical Examination Kamaljit Singh, MD Associate Professor of Medicine, Attending Physician, Infectious Diseases, Rush University Medical Center, Chicago, Illinois Zoonoses Karl Skorecki, MD Annie Chutick Professor in Medicine, Rappaport Faculty of Medicine and Research Institute, Technion–Israel Institute of Technology; Director, Medical and Research Development, Rambam Health Care Campus, Haifa, Israel Gene and Cell Therapy; Disorders of Sodium and Water Homeostasis Itzchak Slotki, MD Associate Professor of Medicine, Hebrew University, Hadassah Medical School; Director, Division of Adult Nephrology, Shaare Zedek Medical Center, Jerusalem, Israel Disorders of Sodium and Water Homeostasis Arthur S. slu*tsky, MD Professor of Medicine, Surgery, and Biomedical Engineering, University of Toronto; Vice President (Research), St. Michael’s Hospital, Keenan Research Centre, Li Ka Shing Knowledge Institute, Toronto, Ontario, Canada Acute Respiratory Failure; Mechanical Ventilation Eric J. Small, MD Professor of Medicine and Urology, Deputy Director and Director of Clinical Sciences, Helen Diller Family Comprehensive Cancer Center; Chief, Division of Hematology and Oncology, University of California San Francisco School of Medicine, San Francisco, California Prostate Cancer Gerald W. Smetana, MD Professor of Medicine, Harvard Medical School; Division of General Medicine and Primary Care, Beth Israel Deaconess Medical Center, Boston, Massachusetts Principles of Medical Consultation

xxvii

Robert F. Spiera, MD Professor of Clinical Medicine, Weill Cornell Medical College; Director, Scleroderma, Vasculitis, and Myositis Center, The Hospital for Special Surgery, New York, New York Polymyalgia Rheumatica and Temporal Arteritis Stanley M. Spinola, MD Professor and Chair, Department of Microbiology and Immunology, Professor of Medicine, Microbiology and Immunology, and Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana Chancroid David Spriggs, MD Head, Division of Solid Tumor Oncology, Department of Medicine, Memorial Sloan Kettering Cancer Center; Professor of Medicine, Department of Medicine, Weill Cornell Medical College, New York, New York Gynecologic Cancers Paweł Stankiewicz, MD, PhD Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas Gene, Genomic, and Chromosomal Disorders Paul Stark, MD Professor Emeritus, University of California San Diego; Chief of Cardiothoracic Radiology, VA San Diego Healthcare System, San Diego, California Imaging in Pulmonary Disease David P. Steensma, MD Professor of Medicine, Harvard Medical School, Adult Leukemia Program, Dana-Farber Cancer Institute, Boston, Massachusetts Myelodysplastic Syndrome Martin H. Steinberg, MD Professor of Medicine, Pediatrics, and Pathology and Laboratory Medicine, Boston University School of Medicine; Director, Center of Excellence in Sickle Cell Disease, Boston Medical Center, Boston, Massachusetts Sickle Cell Disease and Other Hemoglobinopathies Theodore S. Steiner, MD Associate Professor, University of British Columbia; Associate Head, Division of Infectious Diseases, Vancouver General Hospital, Vancouver, British Columbia, Canada Escherichia Coli Enteric Infections David S. Stephens, MD Stephen W. Schwarzmann Distinguished Professor of Medicine, Emory University School of Medicine and Woodruff Health Sciences Center, Atlanta, Georgia Neisseria Meningitidis Infections David A. Stevens, MD Professor of Medicine, Stanford University Medical School; President, Principal Investigator, Infectious Diseases Research Laboratory, California Institute for Medical Research, San Jose and Stanford, California Systemic Antifungal Agents

xxviii

Contributors

James K. Stoller, MD, MS Chairman, Education Institute, Jean Wall Bennett Professor of Medicine, Cleveland Clinic Lerner College of Medicine; Staff, Respiratory Institute, Cleveland Clinic, Cleveland, Ohio Respiratory Monitoring in Critical Care John H. Stone, MD, MPH Professor of Medicine, Director, Clinical Rheumatology, Harvard Medical School, Massachusetts General Hospital, Boston, Massachusetts The Systemic Vasculitides Richard M. Stone, MD Professor of Medicine, Harvard Medical School, Clinical Director, Adult Leukemia Program, Dana-Farber Cancer Institute, Boston, Massachusetts Myelodysplastic Syndrome Raymond A. Strikas, MD, MPH Education Team Lead, Immunization Services Division, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia Immunization Edwin P. Su, MD Associate Professor of Clinical Orthopaedics, Orthopaedic Surgery, Weill Cornell University Medical College; Associate Attending Orthopaedic Surgeon, Adult Reconstruction and Joint Replacement, Hospital for Special Surgery, New York, New York Surgical Treatment of Joint Disease Roland W. Sutter, MD, MPH&TM Coordinator, Research, Policy and Product Development, Polio Operations and Research Department, World Health Organization, Geneva, Switzerland Diphtheria and Other Corynebacteria Infections Ronald S. Swerdloff, MD Professor of Medicine, David Geffen School of Medicine at University of California Los Angeles; Chief, Division of Endocrinology, Department of Medicine, Harbor-UCLA Medical Center, Torrance, California The Testis and Male Hypogonadism, Infertility, and Sexual Dysfunction Heidi Swygard, MD, MPH Associate Professor of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina Approach to the Patient with a Sexually Transmitted Infection Megan Sykes, MD Michael J. Friedlander Professor of Medicine, Director, Columbia Center for Translational Immunology, Columbia University Medical Center, New York, New York Transplantation Immunology Marian Tanofsky-Kraff, PhD Associate Professor, Department of Medical and Clinical Psychology, Uniformed Services University of Health Sciences, Bethesda, Maryland Eating Disorders Susan M. Tarlo, MBBS Professor of Medicine, Department of Medicine and Dalla Lana School of Public Health, University of Toronto, Respiratory Physician, University Health Network, Toronto Western Hospital and St. Michael’s Hospital, Toronto, Ontario, Canada Occupational Lung Disease Victoria M. Taylor, MD, MPH Professor of Medicine, University of Washington, Fred Hutchinson Cancer Research Center, Seattle, Washington Cultural Context of Medicine

Ayalew Tefferi, MD Professor of Medicine, Department of Hematology, Mayo Clinic, Rochester, Minnesota Polycythemia Vera, Essential Thrombocythemia, and Primary Myelofibrosis Paul S. Teirstein, MD Chief of Cardiology, Department of Medicine, Scripps Clinic, La Jolla, California Interventional and Surgical Treatment of Coronary Artery Disease Sam R. Telford III, ScD Professor, Tufts University Cummings School of Veterinary Medicine, North Grafton, Massachusetts Babesiosis and Other Protozoan Diseases Rajesh V. Thakker, MD May Professor of Medicine, University of Oxford; Radcliffe Department of Clinical Medicine, OCDEM, Churchill Hospital, Headington, Oxford, United Kingdom The Parathyroid Glands, Hypercalcemia, and Hypocalcemia Antonella Tosti, MD Professor of Clinical Dermatology, Department of Dermatology and Cutaneous Surgery, University of Miami, Miami, Florida Diseases of Hair and Nails Indi Trehan, MD, MPH, DTM&H Assistant Professor of Pediatrics, Washington University School of Medicine; Attending Physician, St. Louis Children’s Hospital, BarnesJewish Hospital, St. Louis, Missouri; Visiting Honorary Lecturer in Paediatrics and Child Health, University of Malawi College of Medicine; Consultant Paediatrician, Queen Elizabeth Central Hospital, Blantyre, Malawi Protein-Energy Malnutrition Ronald B. Turner, MD Professor of Pediatrics, University of Virginia School of Medicine, Charlottesville, Virginia The Common Cold Thomas S. Uldrick, MD Staff Clinician, HIV and AIDS Malignancy Branch, National Cancer Institute, Bethesda, Maryland Hematology and Oncology in Patients with Human Immunodeficiency Virus Infection Anthony M. Valeri, MD Professor of Medicine, Columbia University Medical Center; Director, Hemodialysis, Medical Director, Kidney and Pancreas Transplantation, New York-Presbyterian Hospital (CUMC); Director, Hemodialysis, Columbia University Dialysis Center, New York, New York Treatment of Irreversible Renal Failure John Varga, MD John and Nancy Hughes Professor of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois Systemic Sclerosis (Scleroderma) Bradley V. Vaughn, MD Professor of Neurology, Department of Neurology, University of North Carolina, Chapel Hill, North Carolina Disorders of Sleep Alan P. Venook, MD Professor of Medicine, University of California San Francisco, Helen Diller Family Comprehensive Cancer Center, San Francisco, California Liver and Biliary Tract Cancers Joseph G. Verbalis, MD Professor of Medicine, Georgetown University; Chief, Endocrinology and Metabolism, Georgetown University Hospital, Washington, D.C. Posterior Pituitary

Contributors Ronald G. Victor, MD Professor of Medicine, Burns and Allen Chair in Cardiology Research, Director, Hypertension Center, Associate Director, The Heart Institute, Cedars-Sinai Medical Center, Los Angeles, California Arterial Hypertension Angela Vincent, MBBS Professor of Neuroimmunology, University of Oxford; Honorary Consultant in Immunology, Oxford University Hospital Trust, Oxford, United Kingdom Disorders of Neuromuscular Transmission Robert M. Wachter, MD Professor and Associate Chairman, Department of Medicine, University of California San Francisco, San Francisco, California Quality of Care and Patient Safety Edward H. Wagner, MD, MPH Director Emeritus, MacColl Center for Health Care Innovation, Group Health Research Institute, Seattle, Washington Comprehensive Chronic Disease Management Edward E. Walsh, MD Professor of Medicine, University of Rochester School of Medicine and Dentistry; Head, Infectious Diseases, Rochester General Hospital, Rochester, New York Respiratory Syncytial Virus Thomas J. Walsh, MD Director, Transplantation-Oncology Infectious Diseases Program, Chief, Infectious Diseases Translational Research Laboratory, Professor of Medicine, Pediatrics, and Microbiology and Immunology, Weill Cornell Medical Center; Henry Schueler Foundation Scholar, Sharp Family Foundation Scholar in Pediatric Infectious Diseases, Adjunct Professor of Pathology, The Johns Hopkins University School of Medicine; Adjunct Professor of Medicine, The University of Maryland School of Medicine, Baltimore, Maryland Aspergillosis Jeremy D. Walston, MD Raymond and Anna Lublin Professor of Geriatric Medicine and Gerontology, The Johns Hopkins University School of Medicine, Baltimore, Maryland Common Clinical Sequelae of Aging Christina Wang, MD Professor of Medicine, David Geffen School of Medicine at University of California Los Angeles; Associate Director, UCLA Clinical and Translational Research Institute, Harbor-UCLA Medical Center, Torrance, California The Testis and Male Hypogonadism, Infertility, and Sexual Dysfunction Christine Wanke, MD Professor of Medicine and Public Health, Director, Division of Nutrition and Infection, Associate Chair, Department of Public Health, Tufts University School of Medicine, Boston, Massachusetts Gastrointestinal Manifestions of HIV and AIDS Stephen I. Wasserman, MD Professor of Medicine, University of California San Diego, La Jolla, California Approach to the Patient with Allergic or Immunologic Disease Thomas J. Weber, MD Associate Professor, Medicine/Endocrinology, Duke University, Durham, North Carolina Approach to the Patient with Metabolic Bone Disease; Osteoporosis Heiner Wedemeyer, MD Professor, Department of Gastroenterology, Hepatology, and Endocrinology, Hannover Medical School, Hannover, Germany Acute Viral Hepatitis

xxix

Geoffrey A. Weinberg, MD Professor of Pediatrics, University of Rochester School of Medicine and Dentistry; Director, Pediatric HIV Program, Golisano Children’s Hospital at University of Rochester Medical Center, Rochester, New York Parainfluenza Viral Disease David A. Weinstein, MD, MMSc Professor of Pediatric Endocrinology, Director, Glycogen Storage Disease Program, Division of Pediatric Endocrinology, University of Florida College of Medicine, Gainesville, Florida Glycogen Storage Diseases Robert S. Weinstein, MD Professor of Medicine, Department of Medicine, University of Arkansas for Medical Sciences; Staff Endocrinologist, Department of Medicine, Central Arkansas Veterans Health Care System, Little Rock, Arkansas Osteomalacia and Rickets Roger D. Weiss, MD Professor of Psychiatry, Harvard Medical School, Boston, Massachusetts; Chief, Division of Alcohol and Drug Abuse, McLean Hospital, Belmont, Massachusetts Drug Abuse and Dependence Martin Weisse, MD Chair, Pediatrics, Tripler Army Medical Center, Honolulu, Hawaii; Professor, Pediatrics, Uniformed Services University of the Health Sciences, Bethesda, Maryland Measles Jeffrey I. Weitz, MD Professor of Medicine and Biochemistry, McMaster University; Executive Director, Thrombosis and Atherosclerosis Research Institute, Hamilton, Ontario, Canada Pulmonary Embolism Samuel A. Wells, Jr., MD Medical Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland Medullary Thyroid Carcinoma Richard P. Wenzel, MD, MSc Professor and Former Chairman, Internal Medicine, Virginia Commonwealth University, Richmond, Virginia Acute Bronchitis and Tracheitis Victoria P. Werth, MD Professor of Dermatology and Medicine, Hospital of the University of Pennsylvania and Philadelphia Veterans Administration Medical Center; Chief, Dermatology Division, Philadelphia Veterans Administration Medical Center, Philadelphia, Pennsylvania Principles of Therapy of Skin Diseases Sterling G. West, MD, MACP Professor of Medicine, University of Colorado School of Medicine; Associate Division Head for Clinical and Educational Affairs, University of Colorado Division of Rheumatology, Aurora, Colorado Systemic Diseases in Which Arthritis Is a Feature A. Clinton White, Jr., MD Paul R. Stalnaker Distinguished Professor and Director, Infectious Disease Division, Department of Internal Medicine, University of Texas Medical Branch, Galveston, Texas Cestodes Christopher J. White, MD Professor of Medicine, Ochsner Clinical School, University of Queensland School of Medicine; System Chairman of Cardiovascular Diseases, Ochsner Medical Center, New Orleans, Louisiana Atherosclerotic Peripheral Arterial Disease; Electrophysiologic Interventional Procedures and Surgery

xxx

Contributors

Perrin C. White, MD Professor of Pediatrics, The Audry Newman Rapoport Distinguished Chair in Pediatric Endocrinology, University of Texas Southwestern Medical Center, Chief of Endocrinology, Children’s Medical Center Dallas, Dallas, Texas Disorders of Sexual Development Richard J. Whitley, MD Distinguished Professor of Pediatrics, Loeb Eminent Scholar Chair in Pediatrics, Professor of Pediatrics, Microbiology, Medicine, and Neurosurgery, The University of Alabama at Birmingham, Birmingham, Alabama Herpes Simplex Virus Infections Michael P. Whyte, MD Professor of Medicine, Pediatrics, and Genetics, Division of Bone and Mineral Diseases, Washington University School of Medicine; MedicalScientific Director, Center for Metabolic Bone Disease and Molecular Research, Shriners Hospital for Children, St. Louis, Missouri Osteonecrosis, Osteosclerosis/Hyperostosis, and Other Disorders of Bone Samuel Wiebe, MD, MSc Professor of Clinical Neurosciences, University of Calgary; Co-Director, Calgary Epilepsy Program, Alberta Health Services, Foothills Medical Centre, Calgary, Alberta, Canada The Epilepsies Jeanine P. Wiener-Kronish, MD Henry Isaiah Dorr Professor of Research and Teaching in Anaesthesia and Anesthestist-in-Chief, Department of Anesthesia, Critical Care and Pain Medicine, Massachusetts General Hospital/Harvard Medical School, Boston, Massachusetts Overview of Anesthesia Eelco F.M. Wijdicks, MD, PhD Professor of Neurology, Division of Critical Care Neurology, Department of Neurology, Mayo Clinic, Rochester, Minnesota Coma, Vegetative State, and Brain Death David J. Wilber, MD George M. Eisenberg Professor of Medicine, Loyola Stritch School of Medicine; Director, Division of Cardiology, Director, Clinical Electrophysiology, Loyola University Medical Center, Maywood, Illinois Electrophysiologic Interventional Procedures and Surgery Beverly Winikoff, MD, MPH President, Gynuity Health Projects; Professor of Clinical Population and Family Health, Mailman School of Public Health, Columbia University, New York, New York Contraception Gary P. Wormser, MD Professor of Medicine and Chief, Division of Infectious Diseases, Department of Medicine, New York Medical College, Valhalla, New York Lyme Disease

Myron Yanoff, MD Professor and Chair, Ophthalmology, Drexel University College of Medicine, Philadelphia, Pennsylvania Diseases of the Visual System Robert Yarchoan, MD Branch Chief, HIV and AIDS Malignancy Branch, National Cancer Institute, Bethesda, Maryland Hematology and Oncology in Patients with Human Immunodeficiency Virus Infection Neal S. Young, MD Chief, Hematology Branch, NHLBI and Director, Trans-NIH Center for Human Immunology, Autoimmunity, and Inflammation, National Institutes of Health, Bethesda, Maryland Parvovirus William F. Young, Jr., MD, MSc Professor of Medicine, Mayo Clinic College of Medicine; Chair, Division of Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic, Rochester, Minnesota Adrenal Medulla, Catecholamines, and Pheochromocytoma Alan S.L. Yu, MB, BChir Harry Statland and Solon Summerfield Professor of Medicine, Director, Division of Nephrology and Hypertension and the Kidney Institute, University of Kansas Medical Center, Kansas City, Kansas Disorders of Magnesium and Phosphorus Sherif R. Zaki, MD, PhD Chief, Infectious Diseases Pathology Branch, Centers for Disease Control and Prevention, Atlanta, Georgia Leptospirosis Mark L. Zeidel, MD Herman L. Blumgart Professor of Medicine, Harvard Medical School; Physician-in-Chief and Chairman, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts Obstructive Uropathy Thomas R. Ziegler, MD Professor, Department of Medicine, Division of Endocrinology, Metabolism, and Lipids, Emory University School of Medicine, Atlanta, Georgia Malnutrition, Nutritional Assessment, and Nutritional Support in Adult Hospitalized Patients Peter Zimetbaum, MD Associate Professor of Medicine, Harvard Medical School; Director of Clinical Cardiology, Beth Israel Deaconess Medical Center, Boston, Massachusetts Cardiac Arrhythmias with Supraventricular Origin

VIDEO CONTENTS This icon appears throughout the book to indicate chapters with accompanying video available on ExpertConsult.com. For quick viewing, use your smartphone to scan the QR codes in the front of the book. Aging and Geriatric Medicine Confusion Assessment Method (CAM) Video 28-1 – MARCOS MIALNEZ, JORGE G. RUIZ, AND ROSANNE M. LEIPZIG

Clinical Pharmacology Interlaminar Epidural Steroid Injection Video 30-1 – ALI TURABI

Cardiovascular Disease Standard Echocardiographic Views: Long Axis Image Plane Video 55-1A – CATHERINE M. OTTO Standard Echocardiographic Views: Short Axis Image Plane Video 55-1B – CATHERINE M. OTTO Standard Echocardiographic Views: Short Axis Image Plane Video 55-1C – CATHERINE M. OTTO Standard Echocardiographic Views: Four-Chamber Image Plane Video 55-1D – CATHERINE M. OTTO Dilated Cardiomyopathy: Long Axis View Video 55-2A – CATHERINE M. OTTO Dilated Cardiomyopathy: Short Axis View Video 55-2B – CATHERINE M. OTTO Dilated Cardiomyopathy: Apical Four-Chamber View Video 55-2C – CATHERINE M. OTTO Three-Dimensional Echocardiography Video 55-3 – CATHERINE M. OTTO Stress Echocardiography: Normal Reaction Video 55-4A – CATHERINE M. OTTO Stress Echocardiography: Normal Reaction Video 55-4B – CATHERINE M. OTTO Stress Echocardiography: Proximal Stenosis of the Left Anterior Descending Coronary Artery Video 55-4C – CATHERINE M. OTTO Stress Echocardiography: Proximal Stenosis of the Left Anterior Descending Coronary Artery Video 55-4D – CATHERINE M. OTTO Pericardial Effusion: Parasternal Long Axis Video 55-5A – CATHERINE M. OTTO Pericardial Effusion: Parasternal Short Axis Video 55-5B – CATHERINE M. OTTO Pericardial Effusion: Apical Four-Chamber Views Video 55-5C – CATHERINE M. OTTO Secundum Atrial Septal Defect Video 69-1 – ARIANE J. MARELLI Perimembranous Ventricular Septal Defect Video 69-2 – ARIANE J. MARELLI Coronary Stent Placement Video 74-1 – PAUL S. TEIRSTEIN Guidewire Passage Video 74-2 – PAUL S. TEIRSTEIN Delivering the Stent Video 74-3 – PAUL S. TEIRSTEIN Inflating the Stent Video 74-4 – PAUL S. TEIRSTEIN

Final Result Video 74-5 – PAUL S. TEIRSTEIN Superficial Femoral Artery (SFA) Stent Procedure Video 79-1 – CHRISTOPHER J. WHITE Orthotopic Bicaval Cardiac Transplantation Video 82-1 – Y. JOSEPH WOO

Respiratory Diseases Wheezing Video 87-1 – JEFFREY M. DRAZEN Technique for Use of a Metered-Dose Inhaler Video 87-2 – LESLIE HENDELES and the New England Journal of Medicine VATS Wedge Resection Video 101-1 – MALCOLM M. DeCAMP

Critical Care Medicine Ventilation of an Ex Vivo Rat Lung Video 105-1 – ARTHUR S. slu*tSKY, GEORGE VOLGYESI, AND TOM WHITEHEAD

Renal and Genitourinary Diseases Renal Artery Stent Video 125-1 – RENATO M. SANTOS AND THOMAS D. DUBOSE, JR. Interpretation of a Computed Tomographic Colonography Video 133-1 – DAVID H. KIM Donor Liver Transplantation—Donor and Recipient Video 154-1 – IGAL KAM, THOMAS BAK, AND MICHAEL WACHS

Oncology Snare Polypectomy of a Colon Adenoma Video 193-1 – DOUGLAS O. FAIGEL Laparoscopic-Assisted Double Balloon Enteroscopy with Polypectomy of a Jejunal Adenoma Followed by Surgical Oversew of the Polypectomy Site Video 193-2 – DOUGLAS O. FAIGEL Endoscopic Mucosal Resection Using Saline Lift Polypectomy of a Colon Adenoma Followed by Closure of the Mucosal Defect with Clips Video 193-3 – DOUGLAS O. FAIGEL Endoscopic View of Rectal Cancer Video 193-4 – DOUGLAS O. FAIGEL Endoscopic Ultrasound Video 193-5 – DOUGLAS O. FAIGEL

Nutritional Diseases Laparoscopic Roux-en-Y Gastric Bypass Video 220-1 – JAMES M. SWAIN

Endocrine Diseases Pituitary Surgery Video 224-1 – IVAN CIRIC

Diseases of Allergy and Clinical Immunology Skin Testing Video 251-1 – LARRY BORISH Nasal Endoscopy Video 251-2 – LARRY BORISH

xl

Video Contents

Rheumatic Diseases Hip Arthroscopy Osteochondroplasty Video 276-1 – BRYAN T. KELLY

Neurology Cervical Provocation Video 400-1 – RICHARD L. BARBANO Spurling Maneuver Video 400-2 – RICHARD L. BARBANO Cervical Distraction Test Video 400-3 – RICHARD L. BARBANO Straight Leg Raise Video 400-4 – RICHARD L. BARBANO Contralateral Straight Leg Raise Video 400-5 – RICHARD L. BARBANO Seated Straight Leg Raise Video 400-6 – RICHARD L. BARBANO Discectomy Video 400-7 – JASON H. HUANG Absence Seizure Video 403-1 – SAMUEL WIEBE Left Rolandic Seizure Video 403-2 – SAMUEL WIEBE Left Temporal Complex Partial Seizure Video 403-3 – SAMUEL WIEBE Left Temporal Complex Partial Seizure Postictal Confusion Video 403-4 – SAMUEL WIEBE Left Temporal Complex Partial Seizure Video 403-5 – SAMUEL WIEBE Supplementary Sensory-Motor Seizure Video 403-6 – SAMUEL WIEBE Right Posterior Temporal Seizure-Dramatic Frontal Semiology Video 403-7 – SAMUEL WIEBE Right Mesial Frontal Seizure Video 403-8 – SAMUEL WIEBE Nonconvulsive Status Epilepticus Video 403-9 – SAMUEL WIEBE GTC Seizure Tonic Phase Video 403-10 – SAMUEL WIEBE GTC Seizure Clonic Phase Video 403-11 – SAMUEL WIEBE Myoclonic Facial Seizure Video 403-12 – SAMUEL WIEBE Tonic Seizure Lennox Gastaut Video 403-13 – SAMUEL WIEBE Atonic Seizure Lennox Gastaut Video 403-14 – SAMUEL WIEBE Reflex Auditory Seizure Video 403-15 – SAMUEL WIEBE Four Score Video 404-1 – JAMES L. BERNAT AND EELCO F.M. WIJDICKS Persistent Vegetative State Video 404-2 – JAMES L. BERNAT AND EELCO F.M. WIJDICKS Minimally Conscious State Video 404-3 – JAMES L. BERNAT AND EELCO F.M. WIJDICKS Akinetic Mutism Video 404-4 – JAMES L. BERNAT AND EELCO F.M. WIJDICKS Early Parkinson’s Disease Video 409-1 – ANTHONY E. LANG Freezing of Gait in Parkinson’s Disease Video 409-2 – ANTHONY E. LANG

Gunslinger Gait in Progressive Supranuclear Palsy Video 409-3 – ANTHONY E. LANG Supranuclear Gaze Palsy in Progressive Supranuclear Palsy Video 409-4 – ANTHONY E. LANG Applause Sign in Progressive Supranuclear Palsy Video 409-5 – ANTHONY E. LANG Apraxia of Eyelid Opening in Progressive Supranuclear Palsy Video 409-6 – ANTHONY E. LANG Cranial Dystonia in Multiple System Atrophy Video 409-7 – ANTHONY E. LANG Anterocollis in Multiple System Atrophy Video 409-8 – ANTHONY E. LANG Stridor in Multiple System Atrophy Video 409-9 – ANTHONY E. LANG Alien Limb Phenomenon in Corticobasal Syndrome Video 409-10 – ANTHONY E. LANG Myoclonus in Corticobasal Syndrome Video 409-11 – ANTHONY E. LANG Levodopa-Induced Dyskinesia in Parkinson’s Disease Video 409-12 – ANTHONY E. LANG Essential Tremor Video 410-1 – ANTHONY E. LANG Huntington’s Disease Video 410-2 – ANTHONY E. LANG Hemiballism Video 410-3 – ANTHONY E. LANG Blepharospasm Video 410-4 – ANTHONY E. LANG Oromandibular Dystonia Video 410-5 – ANTHONY E. LANG Cervical Dystonia Video 410-6 – ANTHONY E. LANG Writer’s Cramp Video 410-7 – ANTHONY E. LANG Embouchure Dystonia Video 410-8 – ANTHONY E. LANG Sensory Trick in Cervical Dystonia Video 410-9 – ANTHONY E. LANG Generalized Dystonia Video 410-10 – ANTHONY E. LANG Tics Video 410-11 – ANTHONY E. LANG Tardive Dyskinesia Video 410-12 – ANTHONY E. LANG Hemifacial Spasm Video 410-13 – ANTHONY E. LANG Wernicke Encephalopathy Eye Movements: Before Thiamine Video 416-1 – BARBARA S. KOPPEL Wernicke Encephalopathy Eye Movements: After Thiamine Video 416-2 – BARBARA S. KOPPEL Limb Symptoms and Signs Video 419-1 – PAMELA J. SHAW Bulbar Symptoms and Signs Video 419-2 – PAMELA J. SHAW Normal Swallowing Video 419-3 – PAMELA J. SHAW Charcot-Marie-Tooth Disease Exam and Walk Video 420-1 – MICHAEL E. SHY

QUICK REFERENCE (QR) VIDEO ACCESS The images below are QR codes. Each code corresponds to a video from the Goldman-Cecil Medicine 25 collection. For fast and easy video access, right from your mobile device, follow these instructions. The videos are also available on Expertconsult.com.

What You Need

• A mobile device, such as a smartphone or tablet, equipped with a camera and Internet access • A QR code reader application (If you do not already have a reader installed on your mobile device, look for free versions in your app store.)

How It Works

• Open the QR code reader application on your mobile device. • Point the device’s camera at the code and scan. • Each code opens an individual video player for instant viewing—no log-on required.

Confusion Assessment Method (CAM) Chapter 28, Video 28-1 – Marcos Mialnez, Jorge G. Ruiz, and Rosanne M. Leipzig

Standard Echocardiographic Views: Four-Chamber Image Plane Chapter 55, Video 55-1D – Catherine M. Otto

Interlaminar Epidural Steroid Injection Chapter 30, Video 30-1 – Ali Turabi

Dilated Cardiomyopathy: Long Axis View Chapter 55, Video 55-2A – Catherine M. Otto

Standard Echocardiographic Views: Long Axis Image Plane Chapter 55, Video 55-1A – Catherine M. Otto

Dilated Cardiomyopathy: Short Axis View Chapter 55, Video 55-2B – Catherine M. Otto

Standard Echocardiographic Views: Short Axis Image Plane Chapter 55, Video 55-1B – Catherine M. Otto

Dilated Cardiomyopathy: Apical Four-Chamber View Chapter 55, Video 55-2C – Catherine M. Otto

Standard Echocardiographic Views: Short Axis Image Plane Chapter 55, Video 55-1C – Catherine M. Otto

Three-Dimensional Echocardiography Chapter 55, Video 55-3 – Catherine M. Otto

4

Quick Reference (QR) Video Access

Stress Echocardiography: Normal Reaction Chapter 55, Video 55-4A – Catherine M. Otto

Perimembranous Ventricular Septal Defect Chapter 69, Video 69-2 – Ariane J. Marelli

Stress Echocardiography: Normal Reaction Chapter 55, Video 55-4B – Catherine M. Otto

Coronary Stent Placement Chapter 74, Video 74-1 – Paul S. Teirstein

Stress Echocardiography: Proximal Stenosis of the Left Anterior Descending Coronary Artery Chapter 55, Video 55-4C – Catherine M. Otto

Guidewire Passage Chapter 74, Video 74-2 – Paul S. Teirstein

Stress Echocardiography: Proximal Stenosis of the Left Anterior Descending Coronary Artery Chapter 55, Video 55-4D – Catherine M. Otto

Delivering the Stent Chapter 74, Video 74-3 – Paul S. Teirstein

Pericardial Effusion: Parasternal Long Axis Chapter 55, Video 55-5A – Catherine M. Otto

Inflating the Stent Chapter 74, Video 74-4 – Paul S. Teirstein

Pericardial Effusion: Parasternal Short Axis Chapter 55, Video 55-5B – Catherine M. Otto

Final Result Chapter 74, Video 74-5 – Paul S. Teirstein

Pericardial Effusion: Apical Four-Chamber Views Chapter 55, Video 55-5C – Catherine M. Otto

Superficial Femoral Artery (SFA) Stent Procedure Chapter 79, Video 79-1 – Christopher J. White

Secundum Atrial Septal Defect Chapter 69, Video 69-1 – Ariane J. Marelli

Orthotopic Bicaval Cardiac Transplantation Chapter 82, Video 82-1 – Y. Joseph Woo

Quick Reference (QR) Video Access

Wheezing Chapter 87, Video 87-1 – Jeffrey M. Drazen

Endoscopic Mucosal Resection Using Saline Lift Polypectomy of a Colon Adenoma Followed by Closure of the Mucosal Defect with Clips Chapter 193, Video 193-3 – Douglas O. Faigel

VATS Wedge Resection Chapter 101, Video 101-1 – Malcolm M. DeCamp

Endoscopic View of Rectal Cancer Chapter 193, Video 193-4 – Douglas O. Faigel

Ventilation of an Ex Vivo Rat Lung Chapter 105, Video 105-1 – Arthur S. slu*tsky, George Volgyesi, and Tom Whitehead

Endoscopic Ultrasound Chapter 193, Video 193-5 – Douglas O. Faigel

Renal Artery Stent Chapter 125, Video 125-1 – Renato M. Santos and Thomas D. DuBose, Jr.

Laparoscopic Roux-en-Y Gastric Bypass Chapter 220, Video 220-1 – James M. Swain

Interpretation of a Computed Tomographic Colonography Chapter 133, Video 133-1 – David H. Kim

Pituitary Surgery Chapter 224, Video 224-1 – Ivan Ciric

Donor Liver Transportation–Donor and Recipient Chapter 154, Video 154-1 – Igal Kam, Thomas Bak, and Michael Wachs

Skin Testing Chapter 251, Video 251-1 – Larry Borish

Snare Polypectomy of a Colon Adenoma Chapter 193, Video 193-1 – Douglas O. Faigel

Nasal Endoscopy Chapter 251, Video 251-2 – Larry Borish

Laparascopic-Assisted Double Balloon Enteroscopy with Polypectomy of a Jejunal Adenoma Followed by Surgical Oversew of the Polypectomy Site Chapter 193, Video 193-2 – Douglas O. Faigel

Hip Arthroscopy Osteochondroplasty Chapter 276, Video 276-1 – Bryan T. Kelly

5

6

Quick Reference (QR) Video Access

Cervical Provocation Chapter 400, Video 400-1 – Richard L. Barbano

Left Rolandic Seizure Chapter 403, Video 403-2 – Samuel Wiebe

Spurling Maneuver Chapter 400, Video 400-2 – Richard L. Barbano

Left Temporal Complex Partial Seizure Chapter 403, Video 403-3 – Samuel Wiebe

Cervical Distraction Test Chapter 400, Video 400-3 – Richard L. Barbano

Left Temporal Complex Partial Seizure Postictal Confusion Chapter 403, Video 403-4 – Samuel Wiebe

Straight Leg Raise Chapter 400, Video 400-4 – Richard L. Barbano

Left Temporal Complex Partial Seizure Chapter 403, Video 403-5 – Samuel Wiebe

Contralateral Straight Leg Raise Chapter 400, Video 400-5 – Richard L. Barbano

Supplementary Sensory-Motor Seizure Chapter 403, Video 403-6 – Samuel Wiebe

Seated Straight Leg Raise Chapter 400, Video 400-6 – Richard L. Barbano

Right Posterior Temporal Seizure - Dramatic Frontal Semiology Chapter 403, Video 403-7 – Samuel Wiebe

Discectomy Chapter 400, Video 400-7 – Jason H. Huang

Right Mesial Frontal Seizure Chapter 403, Video 403-8 – Samuel Wiebe

Absence Seizure Chapter 403, Video 403-1 – Samuel Wiebe

Nonconvulsive Status Epilepticus Chapter 403, Video 403-9 – Samuel Wiebe

Quick Reference (QR) Video Access

GTC Seizure Tonic Phase Chapter 403, Video 403-10 – Samuel Wiebe

Minimally Conscious State Chapter 404, Video 404-3 – James L. Bernat and Eelco F. M. Wijdicks

GTC Seizure Clonic Phase Chapter 403, Video 403-11 – Samuel Wiebe

Akinetic Mutism Chapter 404, Video 404-4 – James L. Bernat and Eelco F. M. Wijdicks

Myoclonic Facial Seizure Chapter 403, Video 403-12 – Samuel Wiebe

Early Parkinson’s Disease Chapter 409, Video 409-1 – Anthony E. Lang

Tonic Seizure Lennox Gastaut Chapter 403, Video 403-13 – Samuel Wiebe

Freezing of Gait in Parkinson’s Disease Chapter 409, Video 409-2 – Anthony E. Lang

Atonic Seizure Lennox Gastaut Chapter 403, Video 403-14 – Samuel Wiebe

Gunslinger Gait in Progressive Supranuclear Palsy Chapter 409, Video 409-3 – Anthony E. Lang

Reflex Auditory Seizure Chapter 403, Video 403-15 – Samuel Wiebe

Supranuclear Gaze Palsy in Progressive Supranuclear Palsy Chapter 409, Video 409-4 – Anthony E. Lang

Four Score Chapter 404, Video 404-1 – James L. Bernat and Eelco F. M. Wijdicks

Applause Sign in Progressive Supranuclear Palsy Chapter 409, Video 409-5 – Anthony E. Lang

Persistent Vegetative State Chapter 404, Video 404-2 – James L. Bernat and Eelco F. M. Wijdicks

Apraxia of Eyelid Opening in Progressive Supranuclear Palsy Chapter 409, Video 409-6 – Anthony E. Lang

7

8

Quick Reference (QR) Video Access

Cranial Dystonia in Multiple System Atrophy Chapter 409, Video 409-7 – Anthony E. Lang

Hemiballism Chapter 410, Video 410-3 – Anthony E. Lang

Anterocollis in Multiple System Atrophy Chapter 409, Video 409-8 – Anthony E. Lang

Blepharospasm Chapter 410, Video 410-4 – Anthony E. Lang

Stridor in Multiple System Atrophy Chapter 409, Video 409-9 – Anthony E. Lang

Oromandibular Dystonia Chapter 410, Video 410-5 – Anthony E. Lang

Alien Limb Phenomenon in Corticobasal Syndrome Chapter 409, Video 409-10 – Anthony E. Lang

Cervical Dystonia Chapter 410, Video 410-6 – Anthony E. Lang

Myoclonus in Corticobasal Syndrome Chapter 409, Video 409-11 – Anthony E. Lang

Writer’s Cramp Chapter 410, Video 410-7 – Anthony E. Lang

Levodopa-Induced Dyskinesia in Parkinson’s Disease Chapter 409, Video 409-12 – Anthony E. Lang

Embouchure Dystonia Chapter 410, Video 410-8 – Anthony E. Lang

Essential Tremor Chapter 410, Video 410-1 – Anthony E. Lang

Sensory Trick in Cervical Dystonia Chapter 410, Video 410-9 – Anthony E. Lang

Huntington’s Disease Chapter 410, Video 410-2 – Anthony E. Lang

Generalized Dystonia Chapter 410, Video 410-10 – Anthony E. Lang

Quick Reference (QR) Video Access

Tics Chapter 410, Video 410-11 – Anthony E. Lang

Limb Symptoms and Signs Chapter 419, Video 419-1 – Pamela J. Shaw

Tardive Dyskinesia Chapter 410, Video 410-12 – Anthony E. Lang

Bulbar Symptoms and Signs Chapter 419, Video 419-2 – Pamela J. Shaw

Hemifacial Spasm Chapter 410, Video 410-13 – Anthony E. Lang

Normal Swallowing Chapter 419, Video 419-3 – Pamela J. Shaw

Wernickes Encephalopathy Eye Movements: Before Thiamine Chapter 416, Video 416-1 – Barbara S. Koppel

Charcot-Marie-Tooth Disease Exam and Walk Chapter 420, Video 420-1 – Michael E. Shy

Wernickes Encephalopathy Eye Movements: After Thiamine Chapter 416, Video 416-2 – Barbara S. Koppel

9

GOLDMAN-CECIL MEDICINE 25TH EDITION Volume I

EDITED BY

LEE GOLDMAN, MD

Harold and Margaret Hatch Professor Executive Vice President and Dean of the Faculties of Health Sciences and Medicine Chief Executive, Columbia University Medical Center Columbia University New York, New York

ANDREW I. SCHAFER, MD

Professor of Medicine Director, Richard T. Silver Center for Myeloproliferative Neoplasms Weill Cornell Medical College New York, New York

2

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION: MEDICINE AS A LEARNED AND HUMANE PROFESSION LEE GOLDMAN AND ANDREW I. SCHAFER

  APPROACH TO MEDICINE

Medicine is a profession that incorporates science and the scientific method with the art of being a physician. The art of tending to the sick is as old as humanity itself. Even in modern times, the art of caring and comforting, guided by millennia of common sense as well as a more recent, systematic approach to medical ethics (Chapter 2), remains the cornerstone of medicine. Without these humanistic qualities, the application of the modern science of medicine is suboptimal, ineffective, or even detrimental. The caregivers of ancient times and premodern cultures tried a variety of interventions to help the afflicted. Some of their potions contained what are now known to be active ingredients that form the basis for proven medications (Chapter 29). Others (Chapter 39) have persisted into the present era despite a lack of convincing evidence. Modern medicine should not dismiss the possibility that these unproven approaches may be helpful; instead, it should adopt a guiding principle that all interventions, whether traditional or newly developed, can be tested vigorously, with the expectation that any beneficial effects can be explored further to determine their scientific basis. When compared with its long and generally distinguished history of caring and comforting, the scientific basis of medicine is remarkably recent. Other than an understanding of human anatomy and the later description, albeit widely contested at this time, of the normal physiology of the circulatory system, almost all of modern medicine is based on discoveries made within the past 150 years. Until the late 19th century, the paucity of medical knowledge was perhaps exemplified best by hospitals and hospital care. Although hospitals provided caring that all but well-to-do people might not be able to obtain elsewhere, there is little if any evidence that hospitals improved health outcomes. The term hospitalism referred not to expertise in hospital care but rather to the aggregate of iatrogenic afflictions that were induced by the hospital stay itself. The essential humanistic qualities of caring and comforting can achieve full benefit only if they are coupled with an understanding of how medical science can and should be applied to patients with known or suspected diseases. Without this knowledge, comforting may be inappropriate or misleading, and caring may be ineffective or counterproductive if it inhibits a sick person from obtaining appropriate, scientific medical care. Goldman-Cecil Medicine focuses on the discipline of internal medicine, from which neurology and dermatology, which are also covered in substantial detail in this text, are relatively recent evolutionary branches. The term internal medicine, which is often misunderstood by the lay public, was developed in 19th-century Germany. Inneren medizin was to be distinguished from clinical medicine because it emphasized the physiology and chemistry of disease, not just the patterns or progression of clinical manifestations. Goldman-Cecil Medicine follows this tradition by showing how pathophysiologic abnormalities cause symptoms and signs and by emphasizing how therapies can modify the underlying pathophysiology and improve the patient’s well-being. Modern medicine has moved rapidly past organ physiology to an increasingly detailed understanding of cellular, subcellular, and genetic mechanisms. For example, the understanding of microbial pathogenesis and many inflammatory diseases (Chapter 256) is now guided by a detailed understanding of the human immune system and its response to foreign antigens (Chapters 45 to 49). Advances in our understanding of the human microbiome raise the possibility that our complex interactions with microbes, which outnumber our cells by a factor of 10, will help explain conditions ranging from inflammatory bowel disease (Chapter 141) to obesity (Chapter 220).1 Health, disease, and an individual’s interaction with the environment are also substantially determined by genetics. In addition to many conditions

that may be determined by a single gene (Chapter 41), medical science increasingly understands the complex interactions that underlie multigenic traits (Chapter 42). The decoding of the human genome holds the promise that personalized health care increasingly can be targeted according to an individual’s genetic profile, in terms of screening and presymptomatic disease management, as well as in terms of specific medications and their adjusted dosing schedules.2 Although gene therapy has been approved for only one disease, lipoprotein lipase deficiency (Chapter 206), and only in Europe, it has shown promise in other conditions, such as Leber congenital amaurosis (Chapter 423). Cell therapy is now beginning to provide vehicles for the delivery of genes, gene products, and vaccines. It has also opened the way for “regenerative medicine” by facilitating the regeneration of injured or diseased organs and tissues. Such advances and others, such as nanomedicine, have already led to targeted and personalized therapies for a variety of cancers.3 Knowledge of the structure and physical forms of proteins helps explain abnormalities as diverse as sickle cell anemia (Chapter 163) and prion-related diseases (Chapter 415). Proteomics, which is the normal and abnormal protein expression of genes, also holds extraordinary promise for developing drug targets for more specific and effective therapies. Concurrent with these advances in fundamental human biology has been a dramatic shift in methods for evaluating the application of scientific advances to the individual patient and to populations. The randomized controlled trial, sometimes with thousands of patients at multiple institutions, has replaced anecdote as the preferred method for measuring the benefits and optimal uses of diagnostic and therapeutic interventions (Chapter 10). As studies progress from those that show biologic effect, to those that elucidate dosing schedules and toxicity, and finally to those that assess true clinical benefit, the metrics of measuring outcome has also improved from subjective impressions of physicians or patients to reliable and valid measures of morbidity, quality of life, functional status, and other patient-oriented outcomes (Chapter 11). These marked improvements in the scientific methodology of clinical investigation have expedited extraordinary changes in clinical practice, such as recanalization therapy for acute myocardial infarction (Chapter 73), and have shown that reliance on intermediate outcomes, such as a reduction in asymptomatic ventricular arrhythmias with certain drugs, may unexpectedly increase rather than decrease mortality. Just as physicians in the 21st century must understand advances in fundamental biology, similar understanding of the fundamentals of clinical study design as it applies to diagnostic and therapeutic interventions is needed. An understanding of human genetics will also help stratify and refine the approach to clinical trials by helping researchers select fewer patients with a more hom*ogeneous disease pattern to study the efficacy of an intervention. This explosion in medical knowledge has led to increasing specialization and subspecialization, defined initially by organ system and more recently by locus of principal activity (inpatient vs. outpatient), reliance on manual skills (proceduralist vs. nonproceduralist), or participation in research. Nevertheless, it is becoming increasingly clear that the same fundamental molecular and genetic mechanisms are broadly applicable across all organ systems and that the scientific methodologies of randomized trials and careful clinical observation span all aspects of medicine. The advent of modern approaches to managing data now provides the rationale for the use of health information technology. Computerized health records, oftentimes shared with patients in a portable format, can avoid duplication of tests and assure that care is coordinated among the patient’s various health care providers. Extraordinary advances in the science and practice of medicine, which have continued to accelerate with each recent edition of this textbook, have transformed the global burden of disease.4 Life expectancies for men and women are increasing, a greater proportion of deaths are occurring among people older than age 70 years, and far fewer children are dying before the age of 5 years. Nevertheless, huge regional disparities remain, and disability from conditions such as substance abuse, mental health disorders, injuries, diabetes, musculoskeletal disease, and chronic respiratory disease have become increasingly important issues for all health systems.

  APPROACH TO THE PATIENT

Patients commonly have complaints (symptoms). These symptoms may or may not be accompanied by abnormalities on examination (signs) or on laboratory testing. Conversely, asymptomatic patients may have signs or laboratory abnormalities, and laboratory abnormalities can occur in the absence of symptoms or signs.

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

Symptoms and signs commonly define syndromes, which may be the common final pathway of a wide range of pathophysiologic alterations. The fundamental basis of internal medicine is that diagnosis should elucidate the pathophysiologic explanation for symptoms and signs so that therapy may improve the underlying abnormality, not just attempt to suppress the abnormal symptoms or signs. When patients seek care from physicians, they may have manifestations or exacerbations of known conditions, or they may have symptoms and signs that suggest malfunction of a particular organ system. Sometimes the pattern of symptoms and signs is highly suggestive or even pathognomonic for a particular disease process. In these situations, in which the physician is focusing on a particular disease, Goldman-Cecil Medicine provides scholarly yet practical approaches to the epidemiology, pathobiology, clinical manifestations, diagnosis, treatment, prevention, and prognosis of entities such as acute myocardial infarction (Chapter 73), chronic obstructive lung disease (Chapter 88), obstructive uropathy (Chapter 123), inflammatory bowel disease (Chapter 141), gallstones (Chapter 155), rheumatoid arthritis (Chapter 264), hypothyroidism (Chapter 226), tuberculosis (Chapter 324), and virtually any known medical condition in adults. Many patients, however, have undiagnosed symptoms, signs, or laboratory abnormalities that cannot be immediately ascribed to a particular disease or cause. Whether the initial manifestation is chest pain (Chapter 51), diarrhea (Chapter 140), neck or back pain (Chapter 400), or a variety of more than 100 common symptoms, signs, or laboratory abnormalities, Goldman-Cecil Medicine provides tables, figures, and entire chapters to guide the approach to diagnosis and therapy (see E-Table 1-1 or table on inside back cover). By virtue of this dual approach to known disease as well as to undiagnosed abnormalities, this textbook, similar to the modern practice of medicine, applies directly to patients regardless of their mode of manifestation or degree of previous evaluation. The patient-physician interaction proceeds through many phases of clinical reasoning and decision making. The interaction begins with an elucidation of complaints or concerns, followed by inquiries or evaluations to address these concerns in increasingly precise ways. The process commonly requires a careful history or physical examination, ordering of diagnostic tests, integration of clinical findings with test results, understanding of the risks and benefits of the possible courses of action, and careful consultation with the patient and family to develop future plans. Physicians can increasingly call on a growing literature of evidence-based medicine to guide the process so that benefit is maximized while respecting individual variations in different patients. Throughout Goldman-Cecil Medicine, the best current evidence is highlighted with specific grade A references that can be accessed directly in the electronic version. The increasing availability of evidence from randomized trials to guide the approach to diagnosis and therapy should not be equated with “cookbook” medicine. Evidence and the guidelines that are derived from it emphasize proven approaches for patients with specific characteristics. Substantial clinical judgment is required to determine whether the evidence and guidelines apply to individual patients and to recognize the occasional exceptions. Even more judgment is required in the many situations in which evidence is absent or inconclusive. Evidence must also be tempered by patients’ preferences, although it is a physician’s responsibility to emphasize evidence when presenting alternative options to the patient. The adherence of a patient to a specific regimen is likely to be enhanced if the patient also understands the rationale and evidence behind the recommended option. To care for a patient as an individual, the physician must understand the patient as a person. This fundamental precept of doctoring includes an understanding of the patient’s social situation, family issues, financial concerns, and preferences for different types of care and outcomes, ranging from maximum prolongation of life to the relief of pain and suffering (Chapters 2 and 3). If the physician does not appreciate and address these issues, the science of medicine cannot be applied appropriately, and even the most knowledgeable physician will fail to achieve the desired outcomes. Even as physicians become increasingly aware of new discoveries, patients can obtain their own information from a variety of sources, some of which are of questionable reliability. The increasing use of alternative and complementary therapies (Chapter 39) is an example of patients’ frequent dissatisfaction with prescribed medical therapy. Physicians should keep an open mind regarding unproven options but must advise their patients carefully if such options may carry any degree of potential risk, including the risk that they may be relied on to substitute for proven approaches. It is crucial for the

3

physician to have an open dialogue with the patient and family regarding the full range of options that either may consider. The physician does not exist in a vacuum, but rather as part of a complicated and extensive system of medical care and public health. In premodern times and even today in some developing countries, basic hygiene, clean water, and adequate nutrition have been the most important ways to promote health and reduce disease. In developed countries, adoption of healthy lifestyles, including better diet (Chapter 213) and appropriate exercise (Chapter 16), is the cornerstone to reducing the epidemics of obesity (Chapter 220), coronary disease (Chapter 52), and diabetes (Chapter 229). Public health interventions to provide immunizations (Chapter 18) and to reduce injuries and the use of tobacco (Chapter 32), illicit drugs (Chapter 34), and excess alcohol (Chapter 33) can collectively produce more health benefits than nearly any other imaginable health intervention.

  APPROACH TO THE MEDICAL PROFESSION

In a profession, practitioners put the welfare of clients or patients above their own welfare.5 Professionals have a duty that may be thought of as a contract with society. The American Board of Internal Medicine and the European Federation of Internal Medicine have jointly proposed that medical professionalism should emphasize three fundamental principles: the primacy of patient welfare, patient autonomy, and social justice.6 As modern medicine brings a plethora of diagnostic and therapeutic options, the interactions of the physician with the patient and society become more complex and potentially fraught with ethical dilemmas (Chapter 2). To help provide a moral compass that is not only grounded in tradition but also adaptable to modern times, the primacy of patient welfare emphasizes the fundamental principle of a profession. The physician’s altruism, which begets the patient’s trust, must be impervious to the economic, bureaucratic, and political challenges that are faced by the physician and the patient (Chapter 5). The principle of patient autonomy asserts that physicians make recommendations but patients make the final decisions. The physician is an expert advisor who must inform and empower the patient to base decisions on scientific data and how these data can and should be integrated with a patient’s preferences. The importance of social justice symbolizes that the patient-physician interaction does not exist in a vacuum. The physician has a responsibility to the individual patient and to broader society to promote access and to eliminate disparities in health and health care. To promote these fundamental principles, a series of professional responsibilities has been suggested (Table 1-1). These specific responsibilities represent practical, daily traits that benefit the physician’s own patients and society as a whole. Physicians who use these and other attributes to improve their patients’ satisfaction with care are not only promoting professionalism but also reducing their own risk for liability and malpractice. An interesting new aspect of professionalism is the increasing reliance on team approaches to medical care, as exemplified by physicians whose roles are defined by the location of their practice—historically in the intensive care unit or emergency department and more recently on the inpatient general hospital floor. Quality care requires coordination and effective communication across inpatient and outpatient sites among physicians who themselves now typically work defined hours.7 This transition from reliance on a single, always available physician to a team, ideally with a designated coordinator, places new challenges on physicians, the medical care system, and the medical profession.

TABLE 1-1  PROFESSIONAL RESPONSIBILITIES Commitment to: Professional competence Honesty with patients Patient confidentiality Maintaining appropriate relations with patients Improving the quality of care Improving access to care Just distribution of finite resources Scientific knowledge Maintaining trust by managing conflicts of interest Professional responsibilities From Brennan T, Blank L, Cohen J, et al. Medical professionalism in the new millennium: a physician charter. Ann Intern Med. 2002;1136:243-246.

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

E-TABLE 1-1  GUIDE TO THE APPROACH TO COMMON SYMPTOMS, SIGNS, AND LABORATORY ABNORMALITIES CHAPTER

SPECIFIC TABLES OR FIGURES

SYMPTOMS Constitutional Fever Fatigue Poor appetite Weight loss Obesity Snoring, sleep disturbances

280 274 132 132, 219 220 100, 405

Tables 280-1 to 280-8 E-Table 274-1 Table 132-1 Figure 132-4; Tables 132-4, 219-1, 219-2 Figure 220-1 Table 405-6

Head, Eyes, Ears, Nose, Throat Headache Visual loss, transient Ear pain Hearing loss Ringing in ears (tinnitus) Vertigo Nasal congestion, rhinitis, or sneezing Loss of smell or taste Dry mouth Sore throat Hoarseness

398 423, 424 426 428 428 428 251, 426 427 425 429 429

Tables 398-1, 398-2 Tables 423-2, 424-1 Table 426-3 Figure 428-1 Figure 428-2 Figure 428-3 Figure 251-1; Table 251-2 Table 427-1 Table 425-7 Figure 429-2; Table 429-1

Cardiopulmonary Chest pain Bronchitis Shortness of breath Palpitations Dizziness Syncope Cardiac arrest Cough Hemoptysis

51, 137 96 51, 83 51, 62 51, 62, 428 62 63 83 83

Tables 51-2, 137-5, 137-6

132 132, 138 135, 153 132, 137, 138, 139

Figure 132-5; Table 132-5 Table 132-1 Figure 135-3; Table 135-1 Figures 132-6, 138-2; Tables 137-3, 137-4, 139-1

132, 142 132, 137 137, 140 135 136, 137 145 145

Figures 132-1, 132-2; Tables 132-2, 132-3, 142-1 Figure 132-3; Tables 132-2, 137-1 Figures 137-1, 140-1 to 140-4 Figures 135-3, 135-4, 135-6; Table 135-4 Figures 136-3, 136-5, 137-1; Table 136-2 Figure 145-5

284, 285 284 26 123 126 285 236 236 240 234 234 200 285

Tables 284-3, 284-5, 285-2 Table 284-3 Tables 26-1 to 26-3 Tables 123-1 to 123-3 Figure 126-1

Gastrointestinal Nausea and vomiting Dysphagia, odynophagia Hematemesis Heartburn/dyspepsia Abdominal pain   Acute   Chronic Diarrhea Melena, blood in stool Constipation Fecal incontinence Anal pain Genitourinary Dysuria Frequency Incontinence Urinary obstruction Renal colic vagin*l discharge Menstrual irregularities Female infertility Hot flushes Erectile dysfunction Male infertility Scrotal mass Genital ulcers or warts

Figure 83-3 Figure 62-1; Tables 51-4, 62-5 Figure 62-1; Table 428-1 Figure 62-1; Tables 62-1, 62-2, 62-4 Figures 63-2, 63-3 Figure 83-1; Tables 83-2, 83-3 Tables 83-6, 83-7

Figure 236-3; Tables 236-3, 236-4 Table 236-5 Table 240-1 Figure 234-10 Figures 234-8, 234-9; Table 234-7 Figure 200-1 Table 285-1

3.e1

3.e2

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

E-TABLE 1-1  GUIDE TO THE APPROACH TO COMMON SYMPTOMS, SIGNS, AND LABORATORY ABNORMALITIES—cont’d CHAPTER

SPECIFIC TABLES OR FIGURES

Musculoskeletal Neck or back pain

400

Figures 400-4, 400-5, 400-6; Tables 400-3 to 400-5

Painful joints

256

Figure 256-1; Tables 256-1, 256-3

Swollen feet, ankles, or legs   Bilateral   Unilateral

51 81

Figure 51-8 Figure 81-2; Table 81-2

Claudication

79

Table 79-3

Acute limb ischemia

79

Figure 79-5; Table 79-1

Weakness

396, 420, 421, 422

Tables 396-1, 420-2, 421-2, 421-4

Sensory loss

396, 420

Figure 420-1; Tables 420-1, 420-3 to 420-5

Memory loss

402

Figures 402-1, 402-2; Tables 402-1 to 402-6

Abnormal gait

396

Table 396-2

Seizures

403

Tables 403-1 to 403-6

Abnormal bleeding

171

Table 171-1

Rash

436, 441

Figure 436-1; Tables 436-1 to 436-6, 441-5

Hives

252, 440

Figure 252-2; Tables 252-1, 440-1, 440-2

Abnormal pigmentation

441

Table 441-2

Alopecia and hirsutism

442

Tables 442-1, 442-3

Nail disorders

442

Table 442-4

Fever

280, 281

Figure 281-1; Tables 280-1 to 280-8, 281-2

Hypothermia

8, 109

Table 109-4

Tachycardia/bradycardia

8, 62, 64, 65

Figures 62-2, 62-3; Tables 64-4, 65-2

Hypertension

67

Table 67-5

Hypotension/shock

8, 106

Figures 106-3, 108-1; Tables 106-1, 107-1, 107-2

Altered respiration

8, 86, 104

Tables 86-1, 86-2, 104-2

Eye pain

423

Table 423-3

Red eye

423

Tables 423-4, 423-6

Dilated pupil

424

Figure 424-4

Nystagmus

424

Table 424-5

Papilledema

424

Table 424-2

Strabismus

424

Figure 424-6

Jaundice

147

Figure 147-2; Tables 147-1 to 147-3

Otitis

426

Table 426-3

Sinusitis

251, 426

Tables 251-3, 426-1, 426-2

Oral ulcers and discolorations

425

Tables 425-1 to 425-4

Salivary gland enlargement

425

Table 425-6

Neck mass

190

Figure 190-3

Lymphadenopathy

168

Tables 168-1 to 168-6

Thyroid nodule

226

Figure 226-4

Thyromegaly/goiter

226

Figures 226-1, 226-3

Extremities

Neurologic

Integumentary

SIGNS Vital Signs

Head, Eyes, Ears, Nose, Throat

Neck

Breast Breast mass

198

Lungs Wheezes

83

Table 83-4

Heart murmur or extra sounds

51

Figure 51-6; Tables 51-7, 51-8

Jugular venous distention

51

Table 51-6

Carotid pulse abnormalities

51

Figure 51-5

Cardiac

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

3.e3

E-TABLE 1-1  GUIDE TO THE APPROACH TO COMMON SYMPTOMS, SIGNS, AND LABORATORY ABNORMALITIES—cont’d CHAPTER

SPECIFIC TABLES OR FIGURES

Abdomen Hepatomegaly

146

Figure 146-5

Splenomegaly

168

Tables 168-7, 168-9

Acute abdomen

142, 143

Figure 143-1; Table 142-1

Abdominal swelling/ascites

142, 153

Table 153-3

Rectal bleeding/positive stool

135, 193

Figures 135-3, 135-4, 135-6; Table 135-4

Hemorrhoids

145

Table 145-1

Arthritis

256

Figure 256-1

Edema

51

Figure 51-8

Cyanosis

51

Clubbing

51

Figure 51-10

Delirium

28

Figure 28-1; Tables 28-1, 28-2

Psychiatric disturbances

397

Tables 397-1 to 397-4, 397-6 to 397-8, 397-10, 397-11, 397-13, 397-14

Coma

404

Tables 404-1 to 404-4

Stroke

407, 408

Figure 407-1; Tables 407-2, 407-3, 407-5, 407-6, 408-5, 408-6

Movement disorders

409, 410

Tables 409-4, 410-1 to 410-9

Neuropathy

420

Figure 420-1; Tables 420-1 to 420-5, E-Table 420-1

Suspicious mole

203

Table 203-1

Nail diseases

442

Table 442-4

Anemia

158

Tables 158-2 to 158-6

Polycythemia

166

Figure 166-2; Table 166-4

Leukocytosis

167

Figure 167-4; Table 167-1

Lymphocytosis

167

Table 167-3

Monocytosis

167

Table 167-2

Eosinophilia

170

Figure 170-2; Table 170-1

Neutropenia   With fever

167 281

Figure 167-7; Tables 167-4 and 167-5 Figure 281-1

Thrombocytosis

166

Figure 166-6; Table 166-6

Thrombocytopenia

172

Figure 172-1; Tables 172-1, 172-3

Prolonged PT or PTT

171

Figure 171-4

Urinalysis

114, 120

Tables 114-2, 120-6

Abnormal liver enzymes

147

Figures 147-2 to 147-4

Elevated BUN/creatinine   Acute   Chronic

120 130

Figure 120-1; Tables 120-1 to 120-5 Table 130-1

Hyperglycemia

229

Tables 229-1, 229-2

Hypoglycemia

230

Tables 230-1, 230-2

Electrolyte abnormalities

116, 117

Figure 116-4; Tables 116-6, 116-7, 117-2, 117-3

Acid-base disturbances

118

Figures 118-1, 118-2; Tables 118-1 to 118-6

Hypercalcemia

245

Figure 245-3; Tables 245-2 to 245-4

Hypocalcemia

245

Figure 245-4; Table 245-6

Hypo- and hyperphosphatemia

119

Tables 119-2, 119-3

Magnesium deficiency

119

Table 119-1

Elevated Pco2

86

Figure 86-2

Solitary pulmonary nodule

191

Figure 191-2

Pleural effusion

99

Tables 99-4 to 99-6

ECG abnormalities

54

Tables 54-2 to 54-5

Musculoskeletal/Extremities

Neurologic

Skin and Nails

COMMON LABORATORY ABNORMALITIES Hematology/Urinalysis

Chemistries

Chest Radiograph/ECG

BUN = blood urea nitrogen; ECG = electrocardiogram; PT = prothrombin time; PTT = partial thromboplastin time.

The changing medical care environment is placing increasing emphasis on standards, outcomes, and accountability. As purchasers of insurance become more cognizant of value rather than just cost (Chapter 12), outcomes ranging from rates of screening mammography (Chapter 198) to mortality rates with coronary artery bypass graft surgery (Chapter 74) become metrics by which rational choices can be made. Clinical guidelines and critical pathways derived from randomized controlled trials and evidence-based medicine can potentially lead to more cost-effective care and better outcomes. These major changes in many Western health care systems bring with them many major risks and concerns. If the concept of limited choice among physicians and health care providers is based on objective measures of quality and outcome, channeling of patients to better providers is one reasonable definition of better selection and enlightened competition. If the limiting of options is based overwhelmingly on cost rather than measures of quality, outcomes, and patient satisfaction, it is likely that the historical relationship between the patient and the truly professional physician will be fundamentally compromised. Another risk is that the same genetic information that could lead to more effective, personalized medicine will be used against the very people whom it is supposed to benefit—by creating a stigma, raising health insurance costs, or even making someone uninsurable. The ethical approach to medicine (Chapter 2), genetics (Chapter 40), and genetic counseling provides means to protect against this adverse effect of scientific progress. In this new environment, the physician often has a dual responsibility: to the health care system as an expert who helps create standards, measures of outcome, clinical guidelines, and mechanisms to ensure high-quality, costeffective care; and to individual patients who entrust their well-being to that physician to promote their best interests within the reasonable limits of the system. A health insurance system that emphasizes cost-effective care, that gives physicians and health care providers responsibility for the health of a population and the resources required to achieve these goals, that must exist in a competitive environment in which patients can choose alternatives if they are not satisfied with their care, and that places increasing emphasis on health education and prevention can have many positive effects. In this environment, however, physicians must beware of overt and subtle pressures that could entice them to underserve patients and abrogate their professional responsibilities by putting personal financial reward ahead of their patients’ welfare. The physician’s responsibility to represent the patient’s best interests and avoid financial conflicts by doing too little in the newer systems of capitated care provides different specific challenges but an analogous moral dilemma to the historical American system in which the physician could be rewarded financially for doing too much. In the current health care environment, all physicians and trainees must redouble their commitment to professionalism. At the same time, the challenge to the individual physician to retain and expand the scientific knowledge base and process the vast array of new information is daunting. In this spirit of a profession based on science and caring, Goldman-Cecil Medicine seeks to be a comprehensive approach to modern internal medicine. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

GENERAL REFERENCES 1. Martin R, Miquel S, Langella P, et al. The role of metagenomics in understanding the human microbiome in health and disease. Virulence. 2014;5:413-423. 2. Ganesh SK, Arnett DK, Assimes TL, et al. Genetics and genomics for the prevention and treatment of cardiovascular disease: update: a scientific statement from the American Heart Association. Circulation. 2013;128:2813-2851. 3. Paoletti C, Hayes DF. Molecular testing in breast cancer. Annu Rev Med. 2014;65:95-110.

4.e1

4. The Global Burden of Disease Study 2010. Lancet. 2012-2013;380:2053-2260. 5. Walton M, Kerridge I. Do no harm: is it time to rethink the Hippocratic Oath? Med Educ. 2014;48:17-27. 6. Snyder L, for the American College of Physicians Ethics, Professionalism, and Human Rights Committee. American College of Physicians ethics manual: sixth edition. Ann Intern Med. 2012; 156:73-104. 7. O’Malley AS, Reschovsky JD. Referral and consultation communication between primary care and specialist physicians: finding common ground. Arch Intern Med. 2011;171:56-65.

4

CHAPTER 2  Bioethics in the Practice of Medicine  

2  BIOETHICS IN THE PRACTICE OF MEDICINE EZEKIEL J. EMANUEL It commonly is argued that modern advances in medical technology, antibiotics, dialysis, transplantation, and intensive care units have created the bioethical dilemmas that confront physicians in the 21st century. In reality, however, concerns about ethical issues are as old as the practice of medicine itself. The Hippocratic Oath, composed sometime around 400 bc, attests to the need

of ancient Greek physicians for advice on how to address the many bioethical dilemmas that they confronted. The Oath addresses issues of confidentiality, abortion, euthanasia, sexual relations between physician and patient, divided loyalties, and, at least implicitly, charity care and executions. Other Hippocratic works address issues such as termination of treatments to dying patients and telling the truth. Whether we agree with the advice dispensed or not, the important point is that many bioethical issues are not created by technology but instead are inherent in medical practice. Technology may make these issues more common and may change the context in which they arise, but many, if not most, bioethical issues that regularly confront physicians are timeless and inherent in the practice of medicine. Many physicians have been educated that four main principles can be invoked to address bioethical dilemmas: autonomy, nonmaleficence, beneficence, and justice. Autonomy is the idea that people should have the right and freedom to choose, pursue, and revise their own life plans. Nonmaleficence is the idea that people should not be harmed or injured knowingly; this principle is encapsulated in the frequently repeated phrase that a physician has an obligation to “first do no harm”—primum non nocere. This phrase is not found either in the Hippocratic Oath or in other Hippocratic writing; the only related, but not identical, Hippocratic phrase is “at least, do not harm.” Whereas nonmaleficence is about avoiding harm, beneficence is about the positive actions that the physician should undertake to promote the wellbeing of his or her patients. In clinical practice, this obligation usually arises from the implicit and explicit commitments and promises surrounding the physician-patient relationship. Finally, there is the principle of justice as the fair distribution of benefits and burdens. Although helpful in providing an initial framework, these principles have limited value because they are broad and open to diverse and conflicting interpretations. In addition, as is clear with the principle of justice, they frequently are underdeveloped. In any difficult case, the principles are likely to conflict. Conflicting ethical principles are precisely why there are bioethical dilemmas. The principles themselves do not offer guidance on how they should be balanced or specified to resolve the dilemma. These principles, which are focused on the individual physician-patient context, are not particularly helpful when the bioethical issues are institutional and systemic, such as allocating scarce vaccines or organs for transplantation or balancing the risks and benefits of mammograms for women younger than 50 years. Finally, these four principles are not comprehensive. Other fundamental ethical principles and values, such as communal solidarity, duties to future generations, trust, and professional integrity, are important in bioethics but not encapsulated except by deformation in these four principles. There is no formula or small set of ethical principles that mechanically or magically gives answers to bioethical dilemmas. Instead, medical practitioners should follow an orderly analytic process. First, practitioners need to obtain the facts relevant to the situation. Second, they must delineate the basic bioethical issue. Third, it is important to identify all the crucial principles and values that relate to the case and how they might conflict. Fourth, because many ethical dilemmas have been analyzed previously and subjected frequently to empirical study, practitioners should examine the relevant literature, whether it is commentaries or studies in medical journals, legal cases, or books. With these analyses, the particular dilemma should be reexamined; this process might lead to reformulation of the issue and identification of new values or new understandings of existing values. Fifth, with this information, it is important to distinguish clearly unethical practices from a range of ethically permissible actions. Finally, it is important not only to come to some resolution of the case but also to state clearly the reasons behind the decisions, that is, the interpretation of the principles used and how values were balanced. Although unanimity and consensus may be desirable ideals, reasonable people frequently disagree about how to resolve ethical dilemmas without being unethical or malevolent. A multitude of bioethical dilemmas arise in medical practice, including issues of genetics, reproductive choices, and termination of care. In clinical practice, the most common issues revolve around informed consent, termination of life-sustaining treatments, euthanasia and physician-assisted suicide, and conflicts of interest.

  PHYSICIAN-PATIENT RELATIONSHIP: INFORMED CONSENT

History

It commonly is thought that the requirement for informed consent is a relatively recent phenomenon. Suggestions about the need for a patient’s informed consent can be found as far back as Plato, however. The first

CHAPTER 2  Bioethics in the Practice of Medicine  

recorded legal case involving informed consent is the 1767 English case of Slater v. Baker and Stapleton, in which two surgeons refractured a patient’s leg after it had healed improperly. The patient claimed they had not obtained consent. The court ruled: [I]t appears from the evidence of the surgeon that it was improper to disunite the callous without consent; this is the usage and law of surgeons: then it was ignorance and unskillfulness in that very particular, to do contrary to the rule of the profession, what no surgeon ought to have done. Although there may be some skepticism about the extent of the information disclosed or the precise nature of the consent obtained, the notable fact is that an 18th-century court declared that obtaining prior consent of the patient is not only the usual practice but also the ethical and legal obligation of surgeons. Failure to obtain consent is incompetent and inexcusable. In contemporary times, the 1957 case of Salgo v. Leland Stanford Junior University Board of Trustees constitutes a landmark by stating that physicians have a positive legal obligation to disclose information about risks, benefits, and alternatives to patients; this decision popularized the term informed consent.

Definition and Justification

Informed consent is a person’s autonomous authorization of a physician to undertake diagnostic or therapeutic interventions for himself or herself. In this view, the patient understands that he or she is taking responsibility for the decision while empowering someone else, the physician, to implement it. However, agreement to a course of medical treatment does not necessarily qualify as informed consent. There are four fundamental requirements for valid informed consent: mental capacity, disclosure, understanding, and voluntariness. Informed consent assumes that people have the mental capacity to make decisions; disease, development, or medications can compromise patients’ mental capacity to provide informed consent. Adults are presumed to have the legal competence to make medical decisions, and whether an adult is incompetent to make medical decisions is a legal determination. Practically, physicians usually decide whether patients are competent on the basis of whether patients can understand the information disclosed, appreciate its significance for their own situation, and use logical and consistent thought processes in decision making. Incompetence in medical decision making does not mean a person is incompetent in all types of decision making and vice versa. Crucial information relevant to the decision must be disclosed, usually by the physician, to the patient. The patient should understand the information and its implications for his or her interests and life goals. Finally, the patient must make a voluntary decision (i.e., one without coercion or manipulation by the physician). It is a mistake to view informed consent as an event, such as the signing of a form. Informed consent is viewed more accurately as a process that evolves during the course of diagnosis and treatment. Typically, the patient’s autonomy is the value invoked to justify informed consent. Other values, such as bodily integrity and beneficence, have also been cited, especially in early legal rulings.

Empirical Data

Fairly extensive research has been done on informed consent. In general, studies show that in clinical situations, physicians frequently do not communicate all relevant information for informed decision making. In a study of audiotapes from 1057 outpatient encounters, physicians mentioned alternatives in only 11.3% of cases, provided pros and cons of interventions in only 7.8% of situations, and assessed the patient’s understanding of the information in only 1.5% of decisions. The more complex the medical decisions, the more likely it was that the elements of informed consent would be fulfilled. Importantly, data suggest that disclosure is better in research settings, both in the informed consent documents and in the discussions. For instance, in recorded interactions between researchers and prospective participants, the major elements of research, such as that the treatment was investigational and the risks and benefits of treatment, were disclosed in more than 80% of interactions. Greater disclosure in the research setting may be the consequence of requiring a written informed consent document that has been reviewed by an independent committee, such as an institutional review board or a research ethics committee. Some have suggested that for common medical interventions, such as elective surgery, standardized informed consent documents should include the risks and benefits as quantified in randomized controlled trials, relevant data on the surgeon, the institution’s clinical outcomes for the procedure, and a list of acceptable alternatives.1

5

Patients frequently fail to recall crucial information disclosed, although they usually think they have sufficient information for decision making. Whether patients fail to recall key information because they are overwhelmed by the information or because they do not find much of it salient to their decision is unclear. The issue is what patients understand at the point of decision making, not what they recall later. Studies aimed at improving informed consent in the clinical setting suggest that interactive media, such as videos and interactive computer software, can improve understanding by patients. A1  Conversely, data on shared decision making show that interactive media do not improve participants’ understanding, whereas more personal interaction, whether as an additional telephone call by a research nurse or as an additional face-to-face meeting, does enhance understanding.2 One of the most important results of empirical research on informed consent is the gap between information and decision making. Many studies show that most patients want information, but far fewer prefer decisionmaking authority. One study showed that most patients wanted information, but only about one third desired decision-making authority, and patients’ decision-making preferences were not correlated with their informationseeking preferences. Several investigators found that patients’ preference for decision-making authority increases with higher educational levels and declines with advancing age. Most important, the more serious the illness, the more likely patients are to prefer that physicians make the decisions. Several studies suggest that patients who have less of a desire to make their own decisions generally are more satisfied with how the decisions were made.

Practical Considerations

Implementing informed consent raises concerns about the extent of information to be disclosed and exceptions to the general requirement. A major area of ethical and legal disagreement has been what information to disclose and how to disclose it. As a practical matter, physicians should disclose at least six fundamental elements of information to patients: (1) diagnosis and prognosis; (2) nature of the proposed intervention; (3) alternative interventions, including no treatment; (4) risks associated with each alternative; (5) benefits of each alternative; and (6) likely outcomes of these alternatives (Table 2-1). Because risk is usually the key worry of physicians, it generally is recommended that physicians disclose (1) the nature of the risks, (2) their magnitude, (3) the probability that each risk will occur, and (4) when the consequence might occur.3 Increasingly, these disclosures should include data both from clinical trials as well as the actual data from the institution and physician performing the test and treatments. Some argue that minor risks need not be disclosed. In general, all serious risks, such as death, paralysis, stroke, infections, or chronic pain, even if rare, should be disclosed, as should common risks. The central problem is that the physician should provide this detailed information within reasonable time constraints and yet not overwhelm patients with complex information in technical language. The historical constraint of office time is no longer tenable. Interactive electronic media, which patients can view at home on their own time, can facilitate the transfer of information outside of the physician’s office. Different states have adopted two contrasting legal standards defining how much information should be disclosed. The physician or customary standard, adapted from malpractice law, states that the physician should disclose information “which a reasonable medical practitioner would make under the same or similar circ*mstances.” Conversely, the reasonable person or lay-oriented standard states that physicians should disclose all information that a “reasonable person in the patient’s circ*mstances would find material to” the medical decision. The physician standard is factual and can be determined empirically, but the patientoriented standard, which is meant to engage physicians with patients, is hypothetical. Currently, each standard is used by about half the states.

TABLE 2-1  FUNDAMENTAL ELEMENTS FOR DISCLOSURE TO PATIENTS Diagnosis and prognosis Nature of proposed intervention Reasonable alternative interventions Risks associated with each alternative intervention Benefits associated with each alternative intervention Probable outcomes of each alternative intervention

6

CHAPTER 2  Bioethics in the Practice of Medicine  

There are exceptions to the requirements of informed consent. In emergency situations, consent can be assumed because patients’ interests concentrate on survival and retaining maximal mental and physical functioning; as a result, reasonable persons would want treatment. In some circ*mstances, physicians may believe the process of informed consent could pose a serious psychological threat. In rare cases, the “therapeutic privilege” promoting a patient’s well-being trumps autonomy, but physicians should be wary of invoking this exception too readily. If patients are deemed incompetent, family members—beginning with spouse, children, parents, siblings, then more distant relatives—usually are selected as surrogates or proxies, although there may be concerns about conflicting interests or knowledge of the patient’s wishes. In the relatively rare circ*mstance in which a patient formally designated a proxy, that person has decision-making authority. The substituted judgment standard states that the proxy should choose what the patient would choose if he or she were competent. The best interests standard states that the proxy should choose what is best for the patient. Frequently, it is not clear how the patient would have decided because the situation was not discussed with the patient and he or she left no living will. Similarly, what is best for a patient is controversial because there are usually tradeoffs between quality of life and survival. These problems are exacerbated because a proxy’s predictions about a patient’s quality of life are poor; proxies tend to underestimate patients’ functional status and satisfaction. Similarly, proxy predictions are inaccurate regarding life-sustaining preferences when the patient is mentally incapacitated. Families tend to agree with patients about two thirds of the time in deciding whether to provide life-sustaining treatments if the patient became demented, when chance alone would generate agreement in 50% of the cases. Such confusion about how to decide for incapacitated patients can create conflicts among family members or between the family and medical providers. In such circ*mstances, an ethics consultation may be helpful.

TERMINATION OF MEDICAL INTERVENTIONS

History

Since the start of medicine, it has been viewed as ethical to withhold medical treatments from the terminally ill and “let nature take its course,” while keeping the patient as comfortable as possible.4 Hippocrates argued that physicians should “refuse to treat those [patients] who are overmastered by their disease.” In the 19th century, prominent American physicians advocated withholding of cathartic and emetic “treatments” from the terminally ill and using ether to ease pain at the end of life. John Collins Warren, who wrote Etherization: with Surgical Remarks in 1848, included a chapter on using ether to ease the pain of a cancer patient’s death. The editors of The Lancet, in 1900, argued that physicians should intervene to ease the pain of death and that they did not have an obligation to prolong a clearly terminal life. The contemporary debate on terminating care began in 1976 with the Quinlan case, in which the New Jersey Supreme Court ruled that patients had a right to refuse life-sustaining interventions on the basis of a right of privacy and that the family could exercise the right for a patient in a persistent vegetative state.

Definition and Justification

It generally is agreed that all patients have a right to refuse medical interventions. Ethically, this right is based on the patient’s autonomy and is implied by the doctrine of informed consent. Legally, state courts have cited the right to privacy, right to bodily integrity, or common law to justify the right to refuse medical treatment. In the 1990 Cruzan case and in the subsequent physician-assisted suicide cases, the U.S. Supreme Court affirmed that there is a “constitutionally protected right to refuse lifesaving hydration and nutrition.” The Court stated that “[A] liberty interest [based on the 14th Amendment] in refusing unwanted medical treatment may be inferred from our prior decisions.” All patients have a constitutional and an ethical right to refuse medical interventions. These rulings were the basis of the consistent state and federal court rulings to permit the husband to terminate artificial nutrition and hydration in the Schiavo case.

Empirical Data

Data show that termination of medical treatments is now the norm, and the trend has been to stop medical interventions more frequently based on the preferences of patients and their surrogate decision makers.5 More than 85% of Americans die without cardiopulmonary resuscitation, and more than 90% of decedents in intensive care units do not receive cardiopulmonary resuscitation. Of decedents in intensive care units, more than 85% die after the

withholding or withdrawal of medical treatments, with an average of 2.6 interventions being withheld or withdrawn per decedent. Despite extensive public support for use of advance care directives and the passage of the Patient Self-Determination Act mandating that health care institutions inform patients of their right to complete such documents, less than 30% of Americans have completed one.6 Even among severely or terminally ill patients, less than 50% have an advance directive in their medical record. Data suggest that more than 40% of patients required active decisionmaking about terminating medical treatments in their final days, but more than 70% lacked decision-making capacity, thereby emphasizing the importance of advance directives. Efforts to improve completion of advance care directives have generated mixed results. In La Crosse County, Wisconsin, for example, after health care organizations in the county added an “Advance Directive” section to their electronic medical records, 90% of decedents had some type of advance directive. Unfortunately, even successful pilot efforts like La Crosse County’s have not been adopted or easily scaled. A persistent problem has been that even when patients complete advance care directives, the documents frequently are not available, physicians do not know they exist, or they tend to be too general or vague to guide decisions. The increasing use of electronic health records should make it possible for advance directives to be available whenever and wherever the patient presents to a health care provider. Although electronic health records will help in making existing advance directives available, they will not solve the problem of actually having a conversation between the physician and the patient about advance care planning. Starting that conversation still seems to be a persistent barrier. Just as proxies are poor at predicting patients’ wishes, data show that physicians are probably even worse at determining patients’ preferences for lifesustaining treatments. In many cases, life-sustaining treatments are continued even when patients or their proxies desire them to be stopped. Conversely, many physicians discontinue or never begin interventions unilaterally without the knowledge or consent of patients or their surrogate decision makers. These discrepancies emphasize the importance of engaging patients early in their care about treatment preferences.

Practical Considerations

There are many practical considerations in enacting this right (Table 2-2). First, patients have a right to refuse any and all medical interventions, from blood transfusions and antibiotics to respirators, artificial hydration, and nutrition. Although initiation of cardiopulmonary resuscitation was the focus of the early court cases, this issue is viewed best as addressing just one of the many medical interventions that can be stopped or withheld. The question of what medical interventions can be terminated—or not started—is a recurrent topic of debate among physicians and other health care providers. The fact is that any treatment prescribed by a physician and administered by a health care provider can be stopped. The issue is not whether the treatment is ordinary, extraordinary, or heroic, or whether it is high technology or low technology. Treatments that can be stopped include not only ventilators, artificial nutrition, and hydration but also dialysis, pacemakers, ventricular assist devices, antibiotics, and any medications. Second, there is no ethical or legal difference between withholding an intervention and withdrawing it. If a respirator or other treatment is started because physicians are uncertain whether a patient would have wanted it, they always can stop it later when information clarifies the patient’s wishes. Although physicians and nurses might find stopping a treatment to be more difficult psychologically, withdrawal is ethically and legally permitted—and required—when it is consonant with the patient’s wishes. Third, competent patients have the exclusive right to decide about terminating their own care.7 If there is a conflict between a competent patient and his or her family, the patient’s wishes are to be followed. It is the patient’s right to refuse treatment, not the family’s right. For incompetent patients, the situation is more complex; if the patients left clear indications of their wishes, whether as explicit oral statements or as written advance care directives, these wishes should be followed. Physicians should not be overly concerned about the precise form patients use to express their wishes; because patients have a constitutional right to refuse treatment, the real concern is whether the wishes are clear and relevant to the situation. If an incompetent patient did not leave explicit indications of his or her wishes or designate a proxy decision maker, the physician should identify a surrogate decision maker and rely on the decision maker’s wishes while being cognizant of the potential problems noted. There is a potential problem in terminating life-sustaining care to patients who are permanently incompetent but still conscious. Some state courts have restricted what treatments a proxy decision maker can terminate,

CHAPTER 2  Bioethics in the Practice of Medicine  

TABLE 2-2  PRACTICAL CONSIDERATIONS IN TERMINATION OF MEDICAL TREATMENTS

TABLE 2-3  DEFINITIONS OF ASSISTED SUICIDE AND EUTHANASIA

PRACTICAL QUESTION

TERM

ANSWER

Is there a legal right to refuse medical interventions?

Yes. The U.S. Supreme Court declared that competent people have a constitutionally protected right to refuse unwanted medical treatments based on the 14th Amendment.

What interventions can be legally and ethically terminated?

Any and all interventions (including respirators, antibiotics, pacemakers, ventricular assist devices, intravenous or enteral nutrition and hydration) can be legally and ethically terminated.

Is there a difference between withholding life-sustaining interventions and withdrawing them?

No. The consensus is that there is no important legal or ethical difference between withholding and withdrawing medical interventions. Stopping a treatment once begun is just as ethical as never having started it.

Whose view about terminating The views of a competent adult patient prevail. life-sustaining interventions It is the patient’s body and life. prevails if there is a conflict between the patient and family? Who decides about terminating life-sustaining interventions if the patient is incompetent?

Are advance care directives legally enforceable?

If the patient appointed a proxy or surrogate decision maker when competent, that person is legally empowered to make decisions about terminating care. If no proxy was appointed, there is a legally designated hierarchy, usually (1) spouse, (2) adult children, (3) parents, (4) siblings, and (5) available relatives. Yes. As a clear expression of the patient’s wishes, they are a constitutionally protected method for patients to exercise their right to refuse medical treatments. In almost all states, clear and explicit oral statements are legally and ethically sufficient for decisions about withholding or withdrawing medical interventions.

thereby requiring the incompetent patient to have given very specific instructions about the particular treatments he or she does not want to receive and the conditions under which care should be withheld or withdrawn. This requirement severely limits the authority and power of proxy decision makers in these cases. Fourth, the right to refuse medical treatment does not translate into a right to demand any treatment, especially treatments that have no pathophysiologic rationale, have already failed, or are known to be harmful. Futility has become a justification to permit physicians unilaterally to withhold or withdraw treatments despite the family’s requests for treatment. Some states, such as Texas, have enacted futility laws, which prescribe procedures by which physicians can invoke futility either to transfer a patient or to terminate interventions. However, the principle of futility is not easy to implement in medical practice. Initially, some commentators advocated that an intervention was futile when the probability of success was 1% or lower. Although this threshold seems to be based on empirical data, it is a covert value judgment. Because the declaration of futility is meant to justify unilateral determinations by physicians, it generally has been viewed as an inappropriate assertion that undermines physician-patient communication and violates the principle of shared decision making. Similar to the distinction between ordinary and extraordinary, futility is viewed increasingly as more obfuscating than clarifying, and it is being invoked much less often.

ASSISTED SUICIDE AND EUTHANASIA

History

Since Hippocrates, euthanasia and physician-assisted suicide have been controversial issues. In 1905, a bill was introduced into the Ohio legislature to legalize euthanasia; it was defeated. In the mid-1930s, similar bills were introduced and defeated in the British Parliament and the Nebraska legislature. As of January 2014, physician-assisted suicide is legal in Oregon and Washington State, based on statewide public referenda, and in Vermont, based on legislation passed in May 2013. Both euthanasia and physician-assisted

7

DEFINITION

Voluntary active euthanasia

Intentional administration of medications or other interventions to cause the patient’s death with the patient’s informed consent

Involuntary active euthanasia

Intentional administration of medications or other interventions to cause the patient’s death when the patient was competent to consent but did not consent (e.g., the patient may not have been asked)

Nonvoluntary active Intentional administration of medications or other euthanasia interventions to cause the patient’s death when the patient was incompetent and was mentally incapable of consenting (e.g., the patient might have been in a coma) Passive euthanasia

Withholding or withdrawal of life-sustaining medical treatments from a patient to let him or her die (termination of life-sustaining treatments)—a poor term that should not be used

Indirect euthanasia

Administration of narcotics or other medications to relieve pain with the incidental consequence of causing sufficient respiratory depression to result in the patient’s death

Physician-assisted suicide

A physician provides prescription medications or other interventions to a patient with the understanding that the patient can use them to commit suicide

suicide are legal in the Netherlands, Belgium, and Luxembourg, and physician-assisted suicide is legal in Switzerland. The Montana Supreme Court did not recognize a constitutional right to physician-assisted suicide, but it ruled that the law permitting the termination of life-sustaining treatment protected physicians from prosecution if they helped hasten the death of a consenting, rational, terminally ill patient.

Definition and Justification

The terms euthanasia and physician-assisted suicide8 require careful definition (Table 2-3). So-called passive and indirect euthanasia are misnomers and are not instances of euthanasia, and both are deemed ethical and legal. There are four arguments against permitting euthanasia and physicianassisted suicide. First, Kant and Mill thought that autonomy did not permit the voluntary ending of the conditions necessary for autonomy, and as a result, both philosophers were against voluntary enslavement and suicide. Consequently, the exercise of autonomy cannot include the ending of life because that would mean ending the possibility of exercising autonomy. Second, many dying patients may have pain and suffering because they are not receiving appropriate care, and it is possible that adequate care would relieve much pain and suffering (Chapter 3). Although a few patients still may experience uncontrolled pain and suffering despite optimal end-of-life care, it is unwise to use the condition of these few patients as a justification to permit euthanasia or physician-assisted suicide for any dying patient. Third, there is a clear ethical distinction between intentional ending of a life and termination of life-sustaining treatments. The actual acts are different— injecting a life-ending medication, such as a muscle relaxant, or providing a prescription for one is not the same as removing or refraining from introducing an invasive medical intervention. Finally, adverse consequences of permitting euthanasia and physician-assisted suicide must be considered. There are disturbing reports of involuntary euthanasia in the Netherlands and Belgium, and many worry about coercion of expensive or burdensome patients to accept euthanasia or physician-assisted suicide. Permitting euthanasia and physician-assisted suicide is likely to lead to further intrusions of lawyers, courts, and legislatures into the physician-patient relationship. There are four parallel arguments for permitting euthanasia and physicianassisted suicide. First, it is argued that autonomy justifies euthanasia and physician-assisted suicide. To respect autonomy requires permitting individuals to decide when it is better to end their lives by euthanasia or physicianassisted suicide. Second, beneficence—furthering the well-being of individuals—supports permitting euthanasia and physician-assisted suicide. In some cases, living can create more pain and suffering than death; ending a painful life relieves more suffering and produces more good. Just the reassurance of having the option of euthanasia or physician-assisted suicide, even if people do not use it, can provide “psychological insurance” and be

8

CHAPTER 2  Bioethics in the Practice of Medicine  

beneficial to people. Third, euthanasia and physician-assisted suicide are no different from termination of life-sustaining treatments that are recognized as ethically justified. In both cases, the patient consents to die; in both cases, the physician intends to end the patient’s life and takes some action to end the patient’s life; and in both cases, the final result is the same: the patient’s death. With no difference in the patient’s consent, the physician’s intention, or the final result, there can be no difference in the ethical justification. Fourth, the supposed slippery slope that would result from permitting euthanasia and physician-assisted suicide is not likely. The idea that permitting euthanasia and physician-assisted suicide would undermine the physicianpatient relationship or lead to forced euthanasia is completely speculative and not borne out by the available data. In its 1997 decisions, the U.S. Supreme Court stated that there is no constitutional right to euthanasia and physician-assisted suicide but that there also is no constitutional prohibition against states legalizing these interventions. Consequently, the legalization of physician-assisted suicide in Oregon, Vermont, and Washington State was constitutional.

Empirical Data

Attitudes and practices related to euthanasia and physician-assisted suicide have been studied extensively. First, surveys consistently indicate that between 50 and 80% of the American and British public support legalizing euthanasia and physician-assisted suicide for terminally ill patients who are suffering intractable pain.9 However, public support declines significantly for euthanasia and physician-assisted suicide in other circ*mstances, such as for psychological reasons.10 Physicians tend to be much less supportive of euthanasia and physician-assisted suicide, with oncologists, palliative care physicians, and geriatricians among the least supportive. Among American and British physicians, the majority opposes legalizing either practice. Second, approximately 25% of American physicians have received requests for euthanasia or physician-assisted suicide, including about 50% of oncologists. Third, multiple studies indicate that less than 5% of American physicians have performed euthanasia or physician-assisted suicide. Among oncologists, 4% have performed euthanasia and 11% have performed physician-assisted suicide during their careers. Fourth, in many cases, the safeguards are violated. One study found that in 54% of euthanasia cases, it was the family who made the request; in 39% of euthanasia and 19% of physician-assisted suicide cases, the patient was depressed; in only half of the cases was the request repeated. In the Netherlands and Belgium, where euthanasia and physician-assisted suicide are legal, less than 2% of all deaths are by these measures, with 0.4 to 1.8% of all deaths as the result of euthanasia without the patient’s consent.11 Since the practice of assisted suicide was legalized in Oregon in 1997, a cumulative 0.2% of all deaths are by physician-assisted suicide. Counterintuitively, data indicate that it is not pain that primarily motivates requests for euthanasia or physician-assisted suicide but rather psychological distress, especially depression and hopelessness. Interviews with physicians and with patients with amyotrophic lateral sclerosis, cancer, or infection with human immunodeficiency virus show that pain is not associated with interest in euthanasia or physician-assisted suicide; instead, depression and hopelessness are the strongest predictors of interest. Studies of patients in Australia and the Netherlands confirm the importance of depression in motivating requests for euthanasia. The desire to avoid dependence and loss of dignity are key motivations. Finally, data from the Netherlands and the United States suggest that there are significant problems in performing euthanasia and physician-assisted suicide. Dutch researchers reported that physician-assisted suicide causes complications in 7% of cases, and in 15% of cases, the patients did not die, awoke from coma, or vomited up the medication. Ultimately, in nearly 20% of physician-assisted suicide cases, the physician ended up injecting the patient with life-ending medication, converting physician-assisted suicide to euthanasia. These data raise serious questions about how to address complications of physician-assisted suicide when euthanasia is illegal or unacceptable.

interventions; (3) there should be a waiting period to ensure that the patient’s desire for euthanasia or physician-assisted suicide is stable and sincere; and (4) the physician should obtain a second opinion from an independent physician. Oregon and Washington State require patients to be terminally ill, whereas the Netherlands, Belgium, and Switzerland have no such requirement. Although there have been some prosecutions in the United States, there have been no convictions—except for Dr. Kevorkian—when physicians and others have participated in euthanasia and physician-assisted suicide.

FINANCIAL CONFLICTS OF INTEREST

History

Worrying about how payment and fees affect medical decisions is not new. In 1899, a physician reported that more than 60% of surgeons in Chicago were willing to provide a 50% commission to physicians for referring cases. He subsequently argued that in some cases, this fee splitting led to unnecessary surgical procedures. A 1912 study by the American Medical Association confirmed that fee splitting was a common practice. Selling patent medicines and patenting surgical instruments were other forms of financial conflicts of interest thought to discredit physicians a century ago. In the 1990s, the ethics of capitation for physician services and pharmaceutical prescriptions and payments by pharmaceutical and biotechnology companies to clinical researchers and practitioners raised the issue of financial conflicts of interest.

Definition and Justification

It commonly is argued that physicians have certain primary interests: (1) to promote the well-being of their patients, (2) to advance biomedical research, (3) to educate future physicians, and, more controversially, (4) to promote public health (Table 2-4). Physicians also have other, secondary interests, such as earning income, raising a family, contributing to the profession, and pursuing avocational interests, such as hobbies. These secondary interests are not evil; typically, they are legitimate, even admirable. A conflict of interest occurs when one of these secondary interests compromises pursuit of a primary interest, especially the patient’s well-being. Conflicts of interest are problematic because they can or appear to compromise the integrity of physicians’ judgment, compromising the patient’s well-being or research integrity. Conflict of interest can induce a physician to do something—perform a procedure, fail to order a test, or distort data—that would not be in a patient’s best interest. These conflicts can undermine the trust of patients and the public, not only in an individual physician but also in the entire medical profession. Even the appearance of conflicts of interest can be damaging because it is difficult for patients and the public “to determine what motives have influenced a professional decision.” The focus is on financial conflicts of interest, not because they are worse than other types of conflicts, but rather because they are more pervasive and more easily identified and regulated compared with other conflicts. Since ancient times, the ethical norm on conflicts has been clear: the physician’s primary obligation is to patients’ well-being, and a physician’s personal financial well-being should not compromise this duty.

Empirical Data

Financial conflicts are not rare but are frequently under-reported.12 The increased use of medical services and escalating health care spending, sometimes without clear benefit to patients, have been linked, at least statistically, to ownership of imaging facilities and referral to specialty hospitals owned by physicians. In Florida, it was estimated that nearly 40% of physicians were involved as owners of freestanding facilities to which they referred patients. In one study, 4 to 4.5 times more imaging examinations were ordered by self-referring physicians than by physicians who referred patients to radiologists. Similarly, patients referred to joint-venture physical therapy facilities have an average of 16 visits compared with 11 at non–joint-venture facilities. A recent study of urologists found that those who had integrated radiation

Practical Considerations

There is widespread agreement that if euthanasia and physician-assisted suicide are used, they should be considered only after all reasonable attempts at physical and psychological palliation have failed. A series of safeguards have been developed and embodied in the Oregon and the Dutch procedures, as follows: (1) the patient must be competent and must request euthanasia or physician-assisted suicide repeatedly and voluntarily; (2) the patient must have pain or other suffering that cannot be relieved by optimal palliative

TABLE 2-4  PRIMARY INTERESTS OF PHYSICIANS Promotion of the health and well-being of their patients Advancement of biomedical knowledge through research Education of future physicians and health care providers Promotion of the public health

facilities into their practices increased their use of the radiation by 2.5 times compared with urologists who did not have financial relationships with radiation facilities.13 There are no comparable data on the influence of capitation on physicians’ judgment. Similarly, multiple studies have shown that interaction with pharmaceutical representatives can lead to prescribing of new drugs, nonrational prescribing, and decreased use of generic drugs by physicians. Industry funding for continuing medical education payment for travel to educational symposia increases prescribing of the sponsor’s drug. Regarding researcher conflicts of interest, the available data suggest that corporate funding does not compromise the design and methodology of clinical research; in fact, commercially funded research may be methodologically more rigorous than government- or foundation-supported research. Conversely, data suggest that financial interests do distort researchers’ interpretation of data. The most important impact of financial interests, however, appears to be on dissemination of research studies. Growing evidence suggests the suppression or selective publication of data unfavorable to corporate sponsors but the repeated publication of favorable results.

bioethical issues. Because these tests have serious implications for the patient and others, scrupulous attention to informed consent must occur. The bioethical issues raised by genetic tests for somatic cell changes, such as tests that occur commonly in cancer diagnosis and risk stratification, are no different from the issues raised with the use of any laboratory or radiographic test. In some cases, ethics consultation services may be of assistance in resolving bioethical dilemmas, although current data suggest that consultation services are used mainly for problems that arise in individual cases and are not used for more institutional or policy problems.

Practical Considerations

For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

First, financial conflicts of interest are inherent in any profession when the professional earns income from rendering a service. Second, conflicts come in many different forms, from legitimate payment for services rendered to investments in medical laboratories and facilities, drug company dinners and payment for attendance at meetings, payment for enrolling patients in clinical research trials, and consultation with companies. Third, in considering how to manage conflicts, it is important to note that people are poor judges of their own potential conflicts. Individuals often cannot distinguish the various influences that guide their judgments, do not think of themselves as bad, and do not imagine that payment shapes their judgments. Physicians tend to be defensive about charges of conflicts of interest. In addition, conflicts tend to act insidiously, subtly changing practice patterns so that they then become what appear to be justifiable norms. Fourth, rules—whether laws, regulations, or professional standards—to regulate conflicts of interest are based on two considerations: (1) the likelihood that payment or other secondary interests would create a conflict and (2) the magnitude of the potential harm if there is compromised judgment. Rules tend to be of three types: (1) disclosure of conflicts, (2) management of conflicts, and (3) outright prohibition. Federal law bans certain types of self-referral of physicians in the Medicare program. The American Medical Association and the Pharmaceutical Research and Manufacturers of America have established joint rules that permit physicians to accept gifts of minimal value but “refuse substantial gifts from drug companies, such as the costs of travel, lodging, or other personal expenses . . . for attending conferences or meetings.” Additionally, the Physician Payment Sunshine Act, which was passed in 2010 as part of the Affordable Care Act and went into effect in August 2013, requires that drug and device manufacturers report all payments and transfers of value given to physicians to the Centers for Medicare and Medicaid Services so that information can be published on a searchable public website. Fifth, there is much emphasis on disclosure of conflicts, with the implicit idea being that sunshine is the best disinfectant. Disclosure may be useful in publications, but it is unclear whether this is a suitable safeguard in the clinical setting. Disclosure just may make patients worry more. Patients may have no context in which to place the disclosure or to evaluate the physician’s clinical recommendation, and patients may have few other options in selecting a physician or getting care, especially in an acute situation. Furthermore, self-disclosure often is incomplete, even when required. Finally, some conflicts can be avoided by a physician’s own action. Physicians can refuse to engage in personal investments in medical facilities or to accept gifts from pharmaceutical companies at relatively little personal cost. In other circ*mstances, the conflicts may be institutionalized, and minimizing them can occur only by changing the way organizations structure reimbursem*nt incentives. Capitation encourages physicians to limit medical services, and its potentially adverse effects are likely to be managed by institutional rules rather than by personal decisions.

FUTURE DIRECTIONS

In the near future, as genetics moves from the research to the clinical setting, practicing physicians are likely to encounter issues surrounding genetic testing, counseling, and treatment. The use of genetic tests without the extensive counseling so common in research studies would alter the nature of the

Grade A Reference A1. Stacey D, Légaré F, Bennett CL, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev. 2014;1:CD001431.

GENERAL REFERENCES

CHAPTER 2  Bioethics in the Practice of Medicine  

GENERAL REFERENCES 1. Krumholz HM. Informed consent to promote patient-centered care. JAMA. 2010;303:1190-1191. 2. Flory J, Emanuel E. Interventions to improve research participants’ understanding in informed consent for research: a systematic review. JAMA. 2004;292:1593-1601. 3. Caring Connections. http://www.caringinfo.org/. Accessed February 9, 2015. 4. Education in Palliative and End-of-life Care. http://www.epec.net. Accessed February 9, 2015. 5. Silveira MJ, Kim SY, Langa KM. Advance directives and outcomes of surrogate decision making before death. N Engl J Med. 2010;362:1211-1218. 6. Rao JK, Anderson LA, Lin FC, et al. Completion of advance directives among U.S. consumers. Am J Prev Med. 2014;46:65-70. 7. Loggers ET, Starks H, Shannon-Dudley M, et al. Implementing a Death with Dignity program at a comprehensive cancer center. N Engl J Med. 2013;368:1417-1424.

9.e1

8. Boudreau JD, Somerville MA, Biller-Andorno N. Clinical decisions. Physician-assisted suicide. N Engl J Med. 2013;368:1450-1452. 9. Seale C. Legalisation of euthanasia or physician-assisted suicide: survey of doctors’ attitudes. Palliat Med. 2009;23:205-212. 10. Emanuel EJ. Euthanasia and physician-assisted suicide: a review of the empirical data from the United States. Arch Intern Med. 2002;162:142-152. 11. van der Heide A, Onwuteaka-Philipsen BD, Rurup ML, et al. End-of-life practices in the Netherlands under the Euthanasia Act. N Engl J Med. 2007;356:1957-1965. 12. Okike K, Kocher MS, Wei EX, et al. Accuracy of conflict-of-interest disclosures reported by physicians. N Engl J Med. 2009;361:1466-1474. 13. Mitchell JM. Urologists’ use of intensity-modulated radiation therapy for prostate cancer. N Engl J Med. 2013;369:1629-1637.

9

CHAPTER 3  Care of Dying Patients and Their Families  

3  CARE OF DYING PATIENTS AND THEIR FAMILIES ROBERT ARNOLD By 2030, 20% of the U.S. population will be older than 65 years, and people older than 85 years constitute the fastest growing segment of the population. Owing to successes in public health and medicine, many of these people will live the last years of their lives with chronic medical conditions such as cirrhosis, end-stage kidney disease, heart failure, and dementia. Even human immunodeficiency virus (HIV) and many cancers, once considered terminal, have turned into chronic diseases. The burden associated with these illnesses and their treatments is high. Chronically ill patients report multiple physical and psychological symptoms that lower their quality of life. The economic pressures associated with medical care adversely affect patients’ socioeconomic status and cause family stress, especially among caregivers, who spend 20 or more hours a week helping their loved ones. Palliative care, which was developed to decrease the burden associated with chronic illness, emphasizes patient- and family-centered care that optimizes quality of life by anticipating, preventing, and treating suffering. Palliative care throughout the continuum of illness addresses physical, intellectual, emotional, social, and spiritual needs while facilitating the patient’s autonomy, access to information, and choice. Palliative care and services, which are coordinated by an interdisciplinary team, are available concurrently with or independent of curative or life-prolonging care. Palliative and nonpalliative health care providers should collaborate and communicate about care needs while focusing on peace and dignity throughout the course of illness, during the dying process, and after death. Five points deserve special emphasis. First, palliative care can be delivered at any time during the course of an illness and is often provided concomitantly with disease-focused, life-prolonging therapy. Waiting until a patient is dying to provide palliative care is a serious error. For example, most elderly patients with chronic incurable illnesses, who might benefit from palliative care, are in the last 10 years of their lives but do not consider themselves to be dying. If palliative care is to have an impact on patients’ lives, it should be provided earlier in a patient’s illness, in tandem with other treatments. A1  Second, prediction is an inexact science. Although many cancers have a predictable trajectory in the last 3 to 6 months of life, for most illnesses, doctors rarely can accurately predict whether a patient is in the last 6 months

10

CHAPTER 3  Care of Dying Patients and Their Families  

1

of life (E-Fig. 3-1). Third, palliative care primarily focuses on the illness’s burden rather than treating the illness itself. Because these burdens can be physical, psychological, spiritual, or social, good palliative care requires a multidisciplinary approach. Fourth, palliative care takes the family unit as the central focus of care. Treatment plans must be developed for both the patient and the family. Fifth, palliative care recognizes that medical treatments are not uniformly successful and that patients die. At some point in a patient’s illness, the treatments may cause more burden than benefit. Palliative care recognizes this reality and starts with a discussion of the patient’s goals and the development of an individualized treatment plan. Many people confuse palliative care with hospice—an understandable confusion because hospices epitomize the palliative care philosophy. The two, however, are different. In the United States, hospice provides palliative care, primarily at home, for patients who have a life expectancy of 6 months or less and who are willing to forgo life-prolonging treatments. However, the requirement that patients must have a life expectancy of 6 months or less limits hospice’s availability, as does the requirement that patients give up expensive and potentially life-prolonging treatments. Moreover, because doctors and patients often are unwilling to cease these treatments until very late in the disease course, so are most patients. Palliative care is both a subspeciality and a domain of good internal medicine.2 Given the need for palliative care, every clinician must be able to provide basic palliative care, and subspecialties such as oncology need special expertise.

PALLIATIVE CARE DOMAINS

Palliative care is a holistic discipline with physical, psychological, spiritual, existential, social, and ethical domains. When caring for patients with chronic life-limiting illness, good palliative care requires that the following questions be addressed:

Is the Patient Physically Comfortable?

Across many chronic conditions, patients have a large number of inadequately treated physical symptoms (Table 3-1). The reasons are multifactorial and range from inadequate physician education, to societal beliefs regarding the inevitability of suffering in chronic illness, to public concerns regarding opioids, to the lack of evidence-based treatments in noncancer patients. The first step to improve symptom management is a thorough assessment. Standardized instruments such as the Brief Pain Inventory (Fig. 3-1) measure both the patient’s symptoms and the effect of those symptoms on the patient’s life. Use of standardized instruments assures that physicians will identify overlooked or underreported symptoms and, as a result, will enhance the satisfaction of both the patient and family. The evidence for the treatment of end-stage symptoms continues to improve. The use of nonsteroidal anti-inflammatory agents and opioids A2  can result in effective pain management in more than 75% of patients with cancer. Advances such as intrathecal pumps and neurolytic blocks are helpful in the remaining 25% (Chapter 30). The use of oxygen is not helpful for refractory dyspnea except when hypoxia has been documented A3 , whereas use of medications for depression often can be helpful A4  (Chapter 397).

Is the Patient Psychologically Suffering?

Patients may be physically comfortable but still suffering. Psychological symptoms and syndromes such as depression, delirium, and anxiety are common in patients with life-limiting or chronic illnesses. It may be difficult to determine whether increased morbidity and mortality are caused by the physical effects of the illness or by the psychological effects of depression and anxiety on energy, appetite, or sleep. Screening questions focusing on mood (e.g., “Have you felt down, depressed, and hopeless most of the time for the past 2 weeks?”) and anhedonism (e.g., “Have you found that little brings you pleasure or joy in the past 2 weeks?”) have been shown to help in diagnosing depression in this population. Increasing data show that treatment of depression in chronic illness is possible and improves both morbidity and mortality. A4-A6  For patients and families facing mortality, existential and spiritual concerns are common. Progressive illness often raises questions of love, legacy, loss, and meaning. A physician’s role is not to answer these questions or to provide reassurance, but rather to understand concerns of the patient and family, how they are coping, and what resources might help. Spirituality often is a source of comfort, and physicians can ascertain a patient’s beliefs using a brief instrument such as the FICA Spiritual Assessment Tool (Table 3-2). A single

screening question such as “Are you at peace?” may identify patients who are in spiritual distress and facilitate referrals to chaplains.

Is the Family Suffering?

Families, defined broadly as those individuals who care most for the patient, are an important source of support for most patients. Families provide informal caregiving, often at the expense of their own physical, economic, and psychological health. Good palliative care requires an understanding of how the family is coping and a search for ways to provide family members with the social or clinical resources they need to improve their well-being. Comprehensive and individually targeted interventions can reduce caregivers’ burdens, although the absolute benefits are relatively small. Because patients in palliative care often die, the palliative care team must address bereavement and postdeath family suffering. Good communication and informational brochures in an intensive care unit can decrease family members’ adverse psychological outcomes after death. A7  A letter of condolence or a follow-up phone call to the next of kin after a patient’s death is respectful and offers the opportunity to clarify questions about the patient’s care. Some family members suffer from complicated grief— a recently described syndrome associated with separation and traumatic distress, with symptoms persisting for more than 6 months. Primary care physicians, who have ongoing relationships with the loved one, and hospices, which provide bereavement services for a year after the patient’s death, have the opportunity to assess whether the grief symptoms persist or worsen.

Is the Patient’s Care Consistent with the Patient’s Goals?

The sine qua non for palliative care is ensuring that the treatment plan is consistent with the patient’s values. In one European cohort of elderly patients, most preferred longevity over quality of life, and half wanted resuscitation if necessary.3 However, a large proportion of elderly, seriously ill patients are not focused on living as long as possible. Instead, they want to maintain a sense of control, relieve their symptoms, improve their quality of life, avoid being a burden on their families, and have a closer relationship with their loved ones. Ensuring that treatment is consistent with a patient’s goals requires good communication skills (Table 3-3). The approaches to giving bad news, discussing goals of care, and talking about forgoing life-sustaining treatment have similar structures (Table 3-4). First, the patient needs to understand the basic facts about the diagnosis, possible treatments, and prognosis. The communication skill that helps physicians communicate information is Ask-TellAsk—exploring what the patient knows or wants to know, then explaining or answering questions, and then providing an opportunity for the patient to ask more. In the hospital, where discontinuity of care is common and misunderstandings frequent, it is important to determine what the patient knows before providing information so as to keep everyone well coordinated. When giving bad news, knowing what the patient knows allows the physician to anticipate the patient’s reaction. Finally, information must be titrated based on the patient’s preferences. Although most patients want to hear everything about their disease, a minority do not. There is no foolproof way to ascertain what any patient wants to know other than by asking. When giving patients information, it is important to give small pieces of information, not use jargon, and check the patient’s understanding.4 Giving information is like dosing a medication: one gives information, checks understanding, and then gives more information based on what the patient has heard. After ensuring that the doctor and the patient have a shared understanding of the medical facts, the physician should engage in an open-ended conversation about the patient’s goals as the disease progresses. This strategy requires that the patient be asked about both hopes and fears. One might ask: “What makes life worth living for you?” “If your time is limited, what are the things that are most important to achieve?” “What are your biggest fears or concerns?” “What would you consider to be a fate worse than death?” The clinician can use an understanding of these goals to make recommendations about which treatments to provide and which treatments would not be helpful. As a result, early palliative care can improve quality of life, mood, and even survival. Physicians find talking about prognosis particularly difficult for two reasons: first, it is hard to foretell the future accurately; and second, they fear this information will “take away patients’ hope.” Thus, they often avoid talking to patients about these issues unless specifically asked. Although some patients do not want to hear prognostic information, for many patients, this

CHAPTER 3  Care of Dying Patients and Their Families  

Mostly cancer

Function

High

10.e1

Death Low

Time Short period of evident decline

High Chronic, consistent with usual role

Mostly heart and lung failure

Death

Healthy

Low

Time Long-term limitations with intermittent serious episodes

Chronic, progressive, eventually fatal illness

Mostly frailty and dementia

Function

High

Low

Time Prolonged dwindling

E-FIGURE 3-1.  Different disease trajectories for different illnesses. Permission obtained from RAND Corporation © Lynn J. Perspectives on care at the close of life. Serving patients who may die soon and their families: the role of hospice and other services. JAMA. 2001;285:925-932.

CHAPTER 3  Care of Dying Patients and Their Families  

11

TABLE 3-1  APPROACHES TO THE MANAGEMENT OF PHYSICAL AND PSYCHOLOGICAL SYMPTOMS SYMPTOM

ASSESSMENT

TREATMENT

Pain

How severe is the symptom (as assessed with the use of validated instruments) and how does it interfere with the patient’s life? What is the etiology of the pain? Is the pain assumed to be neuropathic or somatic? What has the patient used in the past (calculate previous days’ equal analgesic dose)?

Prescribe medications to be administered on a standing or regular basis if pain is frequent. For mild pain: use acetaminophen or a nonsteroidal anti-inflammatory agent (see Table 30-3). For moderate pain: titrate short-acting opioids (see Table 30-4). For severe pain: rapidly titrate short-acting opioids until pain is relieved or intolerable side effects develop; start long-acting opiates once pain is controlled. Rescue doses: prescribe immediate-release opioids—10% of the 24-hour total opiate every hour (orally) or every 30 minutes (parenterally) as needed. Concomitant analgesics (e.g., corticosteroids, anticonvulsants, tricyclic antidepressants, and bisphosphonates) should be used when applicable (particularly for neuropathic pain). Consider alternative medicine and interventional treatments for pain.

Constipation

Is the patient taking opioids? Does the patient have a fecal impaction?

Prescribe laxatives for all patients on opiates. If ineffective, add drugs from multiple classes (e.g., stimulant, osmotic laxatives, and enemas). Prescribe methylnaltrexone if still constipated.

Shortness of breath

Ask the patient to assess the severity of the shortness of breath. Does the symptom have reversible causes?

Prescribe oxygen to treat hypoxia-induced dyspnea, but not if the patient is not hypoxic. Opioids relieve breathlessness without measurable reductions in respiratory rate or oxygen saturation; effective doses are often lower than those used to treat pain. Aerosolized opiates do not work. Fans or cool air may work through a branch of the trigeminal nerve. Consider anxiolytics (e.g., low-dose benzodiazepines) and use reassurance, relaxation, distraction, and massage therapy.

Fatigue

Is the patient too tired to do activities of daily living? Is the fatigue secondary to depression? Is a disease process causing the symptom or is it secondary to reversible causes?

Provide cognitive education about conserving energy use. Treat underlying conditions appropriately.

Nausea

Which mechanism is causing the symptom (e.g., stimulation of the chemoreceptor trigger zone, gastric stimulation, delayed gastric emptying or “squashed stomach” syndrome, bowel obstruction, intracranial processes, or vestibular vertigo)? Is the patient constipated?

Prescribe an agent directed at the underlying cause (Chapter 132). If persistent, give antiemetic around the clock. Multiple agents directed at various receptors or mechanisms may be required.

Anorexia and cachexia

Is a disease process causing the symptom, or is it secondary to other symptoms (e.g., nausea and constipation) that can be treated? Is the patient troubled by the symptom or is the family worried about what not eating means?

A nutritionist may help find foods that are more appetizing (Chapter 213). Provide counseling about the prognostic implications of anorexia (Chapter 219).

Delirium

Is the confusion acute, over hours to days? Does consciousness wax and wane? Are there behavioral disturbances, marked by a reduced clarity in the patient’s awareness of the environment, e.g., a problem of attention? Does the patient have disorganized thinking? Does the patient have an altered level of consciousness—either agitated or drowsy? Is there a reversible reason for the delirium? D: Drugs (opioids, anticholinergics, sedatives, benzodiazepines, steroids, chemotherapies and immunotherapies, some antibiotics) E: Eyes and Ears (poor vision and hearing, isolation) L: Low-flow states (hypoxia, myocardial infarction, congestive heart failure, chronic obstructive pulmonary disease, shock) I: Infections R: Retention (urine/stool), Restraints I: Intracranial (central nervous system metastases, seizures, subdural, cerebrovascular accident, hypertensive encephalopathy) U: Underhydration, Undernutrition, Undersleep M: Metabolic disorders (sodium, glucose, thyroid, hepatic, deficiencies of vitamin B12, folate, niacin, and thiamine) and toxic (lead, manganese, mercury, alcohol)

Identify underlying causes and manage symptoms (Chapter 28). Recommend behavioral therapies, including avoidance of excess stimulation, frequent reorientation, and reassurance. Ensure presence of family caregivers and explain delirium to them. Prescribe haloperidol, risperidone, or olanzapine.

Depression

Have you felt down, depressed, or hopeless most of the time during the past 2 weeks? Have you found that little brings you pleasure or joy during the past 2 weeks? (Somatic symptoms are not reliable indicators of depression in this population.)

Recommend supportive psychotherapy, cognitive approaches, behavioral techniques, pharmacologic therapies (see Table 397-5), or a combination of these interventions; prescribe psychostimulants for rapid treatment of symptoms (within days) or selective serotonin reuptake inhibitors, which may require 3 to 4 weeks to take effect; tricyclic antidepressants are relatively contraindicated because of their side effects.

Anxiety (applicable also Does the patient exhibit restlessness, agitation, insomnia, for family members) hyperventilation, tachycardia, or excessive worry? Is the patient depressed? Is there a spiritual or existential concern underlying the anxiety?

Recommend supportive counseling and consider prescribing benzodiazepines.

Spiritual distress

Inquire about spiritual support.

Are you at peace?

Modified from Morrison RS, Meier DE. Palliative care. N Engl J Med. 2004;350:2582-2590.

12

CHAPTER 3  Care of Dying Patients and Their Families  

information helps them plan their lives. Patients who are told that their disease is generally terminal are more likely to spend a longer period of time in hospice and to avoid aggressive technology at the end of life, without adverse psychological consequences. Furthermore, their families usually have fewer postdeath adverse psychological outcomes. Given that one cannot guess how much information to provide, a physician can start these conversations by asking, “Are you the kind of person who wants to hear about what might happen in the future with your illness or

would you rather take it day by day?” If the patient requests the latter, the physician can follow up by asking if there is someone else with whom he or she can talk about the prognosis. Second, before giving prognostic information, it is useful to inquire about the patient’s concerns in order to provide information in the most useful manner. Finally, it is appropriate when discussing prognostic information to acknowledge uncertainty: “The course of this cancer can be quite unpredictable, and physicians don’t have a crystal ball. I think you should be aware of the possibility that your health may

STUDY ID#

HOSPITAL ID# DO NOT WRITE ABOVE THIS LINE

Brief Pain Inventory (Short Form) Time:

Date: Name: Last

First

Middle Initial

1. Throughout our lives, most of us have had pain from time to time (such as minor headaches, sprains, and toothaches). Have you had pain other than these everyday kinds of pain today? 1. Yes

2. No

2. On the diagram, shade in the areas where you feel pain. Put an X on the area that hurts the most.

Right

Left

Left

Right

3. Please rate your pain by circling the one number that best describes your pain at its worst in the last 24 hours.

0 No pain

1

2

3

4

5

6

7

8

9

10 Pain as bad as you can imagine

4. Please rate your pain by circling the one number that best describes your pain at its least in the last 24 hours.

0 No pain

1

2

3

4

5

6

7

8

9

10 Pain as bad as you can imagine

9

10 Pain as bad as you can imagine

9

10 Pain as bad as you can imagine

5. Please rate your pain by circling the one number that best describes your pain on the average.

0 No pain

1

2

3

4

5

6

7

8

6. Please rate your pain by circling the one number that tells how much pain you have right now.

0 No pain

1

2

3

4

5

6

7

8

FIGURE 3-1.  Brief Pain Inventory (short form). (Copyright 1991. Charles S. Cleeland, PhD, Pain Research Group. All rights reserved.)

CHAPTER 3  Care of Dying Patients and Their Families  

13

7. What treatments or medications are you receiving for your pain?

8. In the last 24 hours, how much relief have pain treatments or medications provided? Please circle the one percentage that most shows how much relief you have received. 0% No pain

10%

20%

30%

40%

50%

60%

70%

80%

90%

100% Complete relief

9. Circle one number that describes how, during the past 24 hours, pain has interfered with your: A. General Activity 0 1 Does not interfere

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

B. Mood 0 1 Does not interfere C. Walking Ability 0 1 Does not interfere

D. Normal Work (includes both work outside the home and housework) 0 1 Does not interfere

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

E. Relations with Other People 0 1 Does not interfere F. Sleep 0 1 Does not interfere G. Enjoyment of Life 0 1 Does not interfere FIGURE 3-1, cont’d.

deteriorate quickly, and you should plan accordingly. We probably are dealing with weeks to months, although some patients do better, and some do worse. Over time, the course may become clearer, and if you wish, I may be able to be a little more precise about what we are facing.” The physician must discuss these topics in an empathic way. Palliative care conversations are as much about emotions as facts.5 Talking about disease progression or death may elicit negative emotions such as anxiety, sadness, or frustration. These emotions decrease a patient’s quality of life and interfere with the ability to hear factual information. Empathic responses strengthen the patient-physician relationship, increase the patient’s satisfaction, and make the patient more likely to disclose other concerns. The first step is recognizing when the patient is expressing emotions. Once the physician recognizes the emotion being expressed, he or she can respond empathically.

It is also important for physicians to recognize their own emotional reactions to these conversations. The physician’s emotional reactions color impressions of the patient’s prognosis, thereby making it hard to listen to the patient, and may influence the physician to hedge bad news. The physician should become aware of her or his own emotional reactions to ensure that the conversation focuses on the patient rather than the health care provider’s needs. In addition to good communication skills, palliative care requires a basic knowledge of medical ethics and the law. For example, patients have the moral and legal right to refuse any treatment, even if refusal results in their death. There is no legal difference between withholding and withdrawing life-sustaining treatment. When confronted with areas of ambiguity, the physician should know how to obtain either a palliative care or ethics consultation.

14

CHAPTER 3  Care of Dying Patients and Their Families  

TABLE 3-2  FICA SPIRITUAL ASSESSMENT TOOL

TABLE 3-4  DISCUSSING PALLIATIVE CARE

F—What is your faith/religion? Do you consider yourself a religious or spiritual person? What do you believe in that gives meaning/importance to life? I—Importance and influence of faith. Is your faith/religion important to you? How do your beliefs influence how you take care of yourself? What are your most important hopes? What role do your beliefs play in regaining your health? What makes life most worth living for you? How might your disease affect this? C—Are you part of a religious or spiritual community? Is this of support to you, and how? Is there a person you really love or is very important to you? How is your family handling your illness? What are their reactions/expectations? A—How would you like me to address these issues in your health care? What might be left undone if you were to die today? Given the severity or chronicity of your illness, what is most important for you to achieve? Would you like me to talk to someone about religious/spiritual matters?

GENERAL APPROACH

From Puchalski C, Romer A. Taking a spiritual history. J Palliat Med. 2000;3:129-137.

• Plan what to say. Create the right setting, allow adequate time, and determine who else should be present at the meeting. • Listen carefully. Be prepared for strong emotions, respond empathetically, encourage description of feelings, and allow time for silence and response. ESTABLISHING GOALS OF MEDICAL CARE • Determine what the patient knows. Clarify any uncertainties or misconceptions. • Understand what the patient is hoping to accomplish as well as any fears and worries. • Repeat the goals back to the patient to make sure they are heard. • Suggest treatments to meet these goals and clarify what will not be done because it will not help achieve the goals. Focus on the goals that you think you can achieve. Plan follow-up, review and revise plan as needed. COMMUNICATING BAD NEWS

TABLE 3-3  CORE COMMUNICATION SKILLS RECOMMENDED SKILL

EXAMPLE

A.  IDENTIFYING CONCERNS AND RECOGNIZING CUES Elicit Concerns Open-ended questions “Is there anything you wanted to talk to me about today?” Active listening

Allowing patient to speak without interruption; allowing pauses to encourage patient to speak

Recognize Cues Informational concerns Patient: “I’m not sure about the treatment options” Emotional concerns

Patient: “I’m worried about that”

• • • • • •

Determine what the patient knows, wants to know, and can comprehend. Share information, recognizing that people handle information in different ways. Avoid jargon, pause frequently, check for understanding, and use silence. Recognize and support the patient’s emotional reaction. Assess the patient’s safety. Agree to a plan that enlists potential sources of support.

WITHDRAWING TREATMENT • Discuss the context of the current discussion and what has changed to precipitate it. • Review prior treatment goals and reassess their virtues. • Discuss alternative treatments based on the new goals. • Document a plan for forgoing treatment and share with the patient, the patient’s family, and the health care team. Adapted from Morrison RS, Meier DE. Clinical practice. Palliative care. N Engl J Med. 2004;350:2582-2590.

B.  RESPONDING TO INFORMATIONAL CONCERNS “Ask-tell-ask”

Topic: communicating information about cancer stage

Ask

“Have any of the other doctors talked about what stage this cancer is?”

Tell

“That’s right, this is a stage IV cancer, which is also called metastatic cancer…”

Ask

“Do you have questions about the staging?”

C.  RESPONDING TO EMOTIONAL CONCERNS Nonverbal Empathy: S-O-L-E-R S

Face the patient Squarely

O

Adopt an Open body posture

L

Lean toward the patient

E

Use Eye contact

R

Maintain a Relaxed body posture

Verbal Empathy: N-U-R-S-E N

Name the emotion: “You seem worried”

U

Understand the emotion: “I see why you are concerned about this”

R

Respect the emotion: “You have shown a lot of strength”

S

Support the patient: “I want you to know that I will still be your doctor whether you have chemotherapy or not”

E

Explore the emotion: “Tell me more about what is worrying you”

From Back AL, Arnold RM, Tulsky JA. Discussing Prognosis. Alexandria, VA: American Society of Clinical Oncology; 2008.

During the past 10 years, there has been a societal push to encourage patients to designate health care proxies and to create advance care planning documents, typified by the use of living wills. These documents are meant to protect patients against unwanted treatments and to ensure that as they are dying, their wishes are followed.6 Unfortunately, there are few empirical data showing that these documents actually change practice. Still, discussions of the documents with health professionals and family members generally provoke important conversations about end-of-life care decisions and may help families confronted with difficult situations know they are respecting their loved one’s wishes.

Is the Patient Going to Die in the Location of Choice?

Most patients say that they want to die at home. Unfortunately, most patients die in institutions—either hospitals or nursing homes. Burdensome transitions decrease quality in end-of-life care. Good palliative care requires establishing a regular system of communication to minimize transitional errors. A social worker who knows about community resources is important in the development of a dispositional plan that respects the patient’s goals. Hospice programs are an important way to allow patients to die at home. In the United States, hospice refers to a specific, government-regulated form of end-of-life care, available under Medicare since 1982 but subsequently adopted by Medicaid and many other third-party insurers. Hospice care typically is given at home, a nursing home, or specialized acute care unit. Care is provided by an interdisciplinary team, which usually includes a physician, nurse, social worker, chaplain, volunteers, bereavement coordinator, and home health aides, all of whom collaborate with the primary care physician, patient, and family. Bereavement services are offered to the family for a year after the death. Hospices are paid on a per diem rate and are required to cover all the costs related to the patient’s life-limiting illness. Because of this and the fact that their focus is on comfort rather than life prolongation, many hospices will not cover expensive treatments such as inotropic agents in heart failure or chemotherapy in cancer, even if they have a palliative effect. Many hospices are experimenting with different service models in an attempt to enroll patients earlier in the course of their illness and increase access to their services.

Grade A References A1. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with metastatic non-smallcell lung cancer. N Engl J Med. 2010;363:733-742. A2. Michna E, Cheng WY, Korves C, et al. Systematic literature review and meta-analysis of the efficacy and safety of prescription opioids, including abuse-deterrent formulations, in non-cancer pain management. Pain Med. 2014;15:79-92. A3. Abernethy AP, McDonald CF, Frith PA, et al. Effect of palliative oxygen versus room air in relief of breathlessness in patients with refractory dyspnoea: a double-blind, randomised controlled trial. Lancet. 2010;376:784-793. A4. Laoutidis ZG, Mathiak K. Antidepressants in the treatment of depression/depressive symptoms in cancer patients: a systematic review and meta-analysis. BMC Psychiatry. 2013;13:140. A5. Gallo JJ, Morales KH, Bogner HR, et al. Long term effect of depression care management on mortality in older adults: follow-up of cluster randomized clinical trial in primary care. BMJ. 2013;346:f2570.

A6. Jiang W, Krishnan R, Kuchibhatla M, et al. Characteristics of depression remission and its relation with cardiovascular outcome among patients with chronic heart failure (from the SADHART-CHF Study). Am J Cardiol. 2011;107:545-551. A7. Lautrette A, Darmon M, Megarbane B, et al. A communication strategy and brochure for relatives of patients dying in the ICU. N Engl J Med. 2007;356:469-478.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 3  Care of Dying Patients and Their Families  

GENERAL REFERENCES 1. Gill TM, Gahbauer EA, Han L, et al. Trajectories of disability in the last year of life. N Engl J Med. 2010;362:1173-1180. 2. Quill TE, Abernethy AP. Generalist plus specialist palliative care: creating a more sustainable model. N Engl J Med. 2013;368:1173-1175. 3. Brunner-La Rocca HP, Rickenbacher P, Muzzarelli S, et al. End-of-life preferences of elderly patients with chronic heart failure. Eur Heart J. 2012;33:752-759.

15.e1

4. Center to Advance Palliatrive Care. https://www.capc.org/. Accessed February 9, 2015. 5. National Consensus Project for Quality Palliative Care. Clinical Practice Guidelines for Quality Palliative Care, 2013. http://www.nationalconsensusproject.org. Accessed February 9, 2015. 6. Silveira MJ, Kim SY, Langa KM. Advance directives and outcomes of surrogate decision making before death. N Engl J Med. 2010;362:1211-1218.

15.e2

CHAPTER 3  Care of Dying Patients and Their Families  

REVIEW QUESTIONS 1. A 75-year-old man with lung cancer is admitted to the hospital with severe shortness of breath. Work-up reveals no other cause of his shortness of breath other than lymphogenic spread of his cancer. His oxygen saturation is 94%. Which of the following treatments should be instituted for his dyspnea? A. Morphine B. Benzodiazepines C. Oxygen D. A and C E. All the above Answer: A  In randomized controlled data, opioids have been shown to decrease dyspnea both in lung cancer patients and in patients with COPD. Oxygen is helpful only if the patient has hypoxia. Benzodiazepines have not been shown to decrease breathlessness. 2. Which of the following is NOT required for a patient to be in hospice? A. The patient must be DNR. B. The patient must have a life-limiting illness, which is likely to cause her death in 6 months. C. The patient wishes to focus on quality of life rather than longevity of life. D. If the patient lives at home, she must have a primary caregiver. Answer: A  The patient does not have to be DNR to be in hospice. The others are requirements of hospice.

3. Which of the following is true of depression in life-limiting illnesses? A. It is a normal reaction when people have a life-limiting illness, and it should not be treated. B. It cannot be improved because the treatments take too long to work in patients with serious illness. C. Treatment of depression decreases both morbidity and mortality. D. It requires a psychiatric consult because treatment is very complicated. Answer: C  Data show that the treatment of depression improves both quality of life and mortality. 4. Which of the following is true? A. Telling patients that they have a terminal illness will result in their losing hope. B. Telling patients they have a terminal illness has no impact on their desire for future treatment. C. Telling patients that they have terminal illnesses is associated with their choosing hospice more frequently. D. Patients have clearly stated that they do not want to be told that they have a terminal illness. Answer: C  Data suggest that telling patients that they have a life-limiting illness is associated with a lower likelihood of choosing aggressive care at the end of life and is not associated with poorer psychiatric outcomes.

15

CHAPTER 4  Cultural Context of Medicine  

DISPARITIES IN HEALTH CARE ACCESS AND QUALITY

Components of health care access include the ability to get into the health care system as well as to obtain appropriate care once in the system. The availability of health care providers who meet an individual patient’s needs is another key component of access to care. Quality care is based on scientific evidence (i.e., is effective), avoids injury to the patient (i.e., is safe), minimizes harmful delays (i.e., is timely), is responsive to the individual patient’s needs (i.e., is patient centered), promotes communication among providers (i.e., is coordinated), does not vary because of personal characteristics (i.e., is equitable), and avoids waste (i.e., is efficient).

4  CULTURAL CONTEXT OF MEDICINE VICTORIA M. TAYLOR The 2010 U.S. Census counted about 39 million blacks or African Americans (13% of the population), nearly 15 million Asian Americans (5% of the population), about 3 million American Indians and Alaska Natives, and more than 500,000 Native Hawaiians and other Pacific Islanders. It also counted more than 50 million individuals of Hispanic or Latino origin (16% of the population). Approximately 40 million Americans (13% of the population) were foreign born. One in 2 immigrants to the United States have limited English proficiency (i.e., they do not speak English very well or fluently), and 1 in 10 immigrants do not speak English at all (Fig. 4-1). During the past two decades, a large body of literature has documented substantial disparities in health status. Although some of these disparities are based on socioeconomic status, many are based on race, ethnicity, or other characteristics. Black men have a substantially higher age-adjusted incidence of prostate cancer than do white men (236 per 100,000 versus 147 per 100,000). American Indians/Alaska Natives are more than twice as likely as non-Latino whites of a similar age to have diabetes. More than half of the Americans who are living with chronic hepatitis B infection are Asians or Pacific Islanders. Lesbian, gay, bisexual, and transgender individuals have higher rates of suicidal behavior compared with heterosexual individuals. A major goal of Healthy People 2020 is to eliminate health disparities for preventable and treatable conditions such as cancer, diabetes, and human immunodeficiency virus infection. Culture can be defined as a shared system of values, beliefs, and patterns of behavior, and it is not simply defined by race and ethnicity. Culture can also be shaped by factors such as country and region of origin, acculturation, language, religion, and sexual orientation. For instance, the black population of the northeastern United States includes individuals who moved from southern states decades ago as well as recent immigrants from Ethiopia. As the United States population becomes increasingly diverse and as pronounced differences in health status continue to be documented, consideration of the cultural context of medicine is becoming a national priority.

Access to Health Care

Racial and ethnic minority groups, particularly immigrants, disproportionately have problems accessing health care. Before the implementation of the Affordable Care Act, the proportions of Latinos and Native Americans/ Alaska Natives who lacked health insurance was more than twice the proportion among non-Latino whites, and less than two thirds of Americans with limited English proficiency were insured. About 1 in 3 Korean American and Vietnamese American adults had no regular source of medical care compared with about 1 in 10 non-Latino white adults. Blacks and Latinos are far less likely than are whites and Asians to have access to physicians of their own race and ethnicity. This imbalance is important because racial concordance between physicians and patients can improve the processes of care. For example, patients with race-concordant physicians are more likely to use needed health services, are less likely to postpone or delay seeking care, and are more satisfied with their care than are patients in race-discordant relationships. Whether these differences translate into different health outcomes, however, is less clear.1

Quality of Health Care

National surveys confirm population-level disparities in the quality of preventive care. Recent immigrants have far lower levels of interval screening for breast, cervical, and colorectal cancer than do individuals who were born in the United States (Fig. 4-2). The proportion of Native Hawaiians and other Pacific Islanders whose serum cholesterol levels are measured at least once every 5 years is significantly lower than among whites. In 2011, only 40% of Asians aged 65 years and older had ever received the pneumococcal vaccine compared with 67% of non-Latino whites. Racial and ethnic disparities have been documented for a number of specific clinical situations. For example, Latino women with breast cancer are less likely to receive radiation therapy within a year of breast-conserving surgery than are white women, Native Americans and Alaska Natives are less likely than whites to receive recommended care such as initial antibiotics within 6 hours of hospital arrival, and blacks with end-stage renal disease

Breast

US-born Foreign-born, in US ≥ 10 years

Cervical Latin America

Foreign-born, in US < 10 years

Asia Africa Colorectal Europe Oceania 0%

Total 0%

10%

20%

30%

40%

50%

60%

70%

FIGURE 4-1.  Proportion of immigrants aged 5 years and older with limited English proficiency by region of origin. (From Grieco EM, Acosta YD, de la Cruz P, et al. The foreign-born population in the United States: 2010. Washington DC: U.S. Department of Commerce; 2012.)

20%

40%

60%

80%

100%

FIGURE 4-2.  Adherence to cancer screening guidelines by immigration status. Breast = mammography during last 2 years among women aged 50 to 74 years. Cervical = Papanicolaou test during last 3 years among women aged 21 to 65 years. Colorectal = among individuals aged 50 to 75 years, fecal occult blood test last year; sigmoidoscopy last 5 years and fecal occult blood test last 3 years; or colonoscopy last 10 years. (From Centers for Disease Control and Prevention. Cancer screening—United States, 2010. MMWR Morb Mortal Wkly Rep. 2012; 61:41-45.)

16

CHAPTER 4  Cultural Context of Medicine  

are less likely to be entered to a transplant list than are whites. Moreover, disparities in the quality of care are found even when variations in insurance status, income, and comorbid conditions are taken into account. Disparities in health care quality exist even in systems that are generally believed to provide equal access.2 For example, in the Veterans Affairs Health System, disparities between blacks and whites have been documented for blood pressure control among patients with hypertension, cholesterol control among patients with coronary heart disease, and glucose control among patients with diabetes. Moreover, these disparities persist even after adjusting for location and socioeconomic status. Similar disparities have been documented in Medicare managed care programs between elderly blacks and whites with diabetes and cardiovascular disorders.

CULTURAL COMPETENCE IN HEALTH CARE

Health disparities can be reduced or perhaps even eliminated by maintaining culturally competent health care systems. Cultural competence may be defined as a set of congruent attitudes, behaviors, and policies that come together both among professionals and within systems to enable effective work in cross-cultural situations (Fig. 4-3). Ongoing efforts to improve cultural competence in the health care system target organizational, structural, and clinical barriers. These initiatives aim to close gaps in health status, to decrease differences in the quality of care, to enhance patients’ satisfaction, and to increase patients’ trust.

Organizational Barriers and Interventions

Diversity among health care professionals is associated with better access to care for disadvantaged populations. Black and Latino physicians are more likely than their white colleagues to work in medically underserved communities and to have a better understanding of barriers to health care. Because less than 10% of practicing physicians are black or Latino, and only about 15% of medical school students are from one of these groups, many U.S. medical schools have implemented comprehensive programs to infuse diversity among their students, resident physicians, and faculty. About two thirds of the patients who receive care at federally funded community health centers in medically underserved areas are members of racial and ethnic minority groups. In these health centers, patients are three times more likely to have limited proficiency in English compared with the general population. The community health center model has proved effective not

only in increasing access to care but also in improving continuity of care and health outcomes. For example, medically underserved communities with community health centers have fewer preventable hospitalizations and uninsured emergency department visits than do similar communities without health centers. Compared with national rates, community health centers report minimal racial and ethnic disparities in clinical outcomes such as the control of diabetes and hypertension.3

Structural Barriers and Interventions

Accumulating evidence suggests that trained professional interpreters can improve the clinical care received by individuals with limited English proficiency. A1  However, interpreter services often remain ad hoc, with family members and untrained nonclinical employees acting as interpreters.4 Use of ad hoc services has potentially negative clinical consequences, including breach of the patient’s confidentiality and inaccurate communication. One major obstacle to the implementation of professional interpreter programs is a lack of reimbursem*nt; Medicare and most private insurers do not pay for interpretation and related services, and most states do not pay for interpretation under Medicaid. Assistance with navigation represents a promising model to enable racial and ethnic minority patients to move through the health system effectively and to be actively involved in decision making about their medical care.5 Guides may be nurses, social workers, or volunteers who are familiar with the health care system. They help patients and their families navigate the treatment process, steering them around obstacles that may limit their access to quality care, choice of doctors, and access to treatment options. For example, an American Cancer Society navigation program is effective in reducing the time to diagnostic resolution after abnormal cancer screening tests in medically underserved patients. A2  Another option for closing the gap in health care among various minority populations is community health workers.6 In general, community health workers live locally and share the language and culture of the patients being served. Lay community health workers provide cultural mediation between communities and the health care system; culturally appropriate and accessible health education and information; help in obtaining needed medical services, informal counseling, and social support; and advocacy within the health care system. The effectiveness of community health workers is documented by a study in which Mexican American women randomized to

Health Care Interventions

Organizational

Systematic

Programs to increase the diversity of health care providers Culturally specific health care settings

Programs to recruit and retain staff who reflect cultural diversity of the community Use of interpreter services or bilingual providers Patient navigator and community health worker programs Use of linguistically and culturally appropriate health education materials

Clinical

Cultural competency training for health care providers

Intermediate Outcomes Health care providers reflect diversity of communities served

Increase cultural relevance and acceptability of health information

Less miscommunication due to language differences or cultural understanding of health events

Increase accuracy of diagnosis and use of appropriate interventions

Greater provider knowledge of variation in health beliefs, practices, and conditions

Increase patient understanding of and adherence to treatment recommendations

More provider sensitivity to their own beliefs and behaviors that marginalize ethnic groups

Improve access to quality health care services by diverse populations

Patient Outcomes Increase patient satisfaction with health care system Increase patient confidence in health care system

Health Outcomes Decrease inappropriate differences in the characteristics and quality of care provided Close gaps in health status across diverse populations

FIGURE 4-3.  Analytic framework for evaluating the effectiveness of health care interventions to increase cultural competence. (From Anderson LM, Scrimshaw SC, Fullilove MT, et al., for the Task Force on Community Preventive Services. Culturally competent healthcare systems: a systemic review. Am J Prev Med. 2003; 24[suppl]:68-79.)

receive health worker education were significantly more likely to obtain Papanicolaou tests than were women randomized to receive usual care. A3  The largest formal system of community health workers is the Indian Health Service, which currently has about 1400 community health representatives.

Clinical Barriers and Interventions

Patients who are members of racial and ethnic minority groups often understand health and disease (i.e., explanatory model) differently than the general population. For example, many Vietnamese people believe that disease is caused by an imbalance of the humoral forces of yin and yang. When ill, they commonly use Chinese herbal medicine as well as indigenous folk practices known as Southern medicine in an effort to restore the balance of humoral forces. In addition, Vietnamese patients may think that Western medicine is too strong and will upset the internal balance. Consequently, a hypertensive Vietnamese patient may, for example, use Chinese herbal medicines instead of prescribed antihypertensive medication. Alternatively, the patient may take a lower dose of medication than prescribed by his or her physician. Cultural competency training for health care providers generally includes teaching cross-cultural knowledge and communication skills, while avoiding stereotypes.7 Examples include the effect of prejudice on gays and lesbians and how this prejudice shapes their interactions with the health care system, and common spiritual practices that might interfere with prescribed therapies (such as Ramadan fasting practices, when observed by diabetic Muslim patients). Communication skills that can be addressed in cultural competence training include approaches to eliciting patients’ explanatory models and use of traditional treatments, as well as methods for negotiating different styles of communication and levels of family participation in decisionmaking. Cultural competency training improves the attitudes and skills of health professionals as well as patient satisfaction, but there is less evidence that it improves clinical outcomes. A4 

SUMMARY

Individual clinical practices should regularly assess their current organizational climate, policies, and training related to diversity. Practices can address health disparities by hiring clinical and office staff who are representative of the communities they serve, by routinely using professional interpreters during clinical encounters with patients who have limited proficiency with English, by offering cultural competency education and training to physicians and staff, and by providing educational and informational materials that are culturally and linguistically appropriate for their patient populations.8 National and state efforts to improve cultural competence in health care, whether used alone or in conjunction with socioeconomic initiatives, are likely to play a significant role in reducing health disparities across population subgroups. An important goal of the Affordable Care Act is to reduce health disparities by expanding health insurance coverage, addressing diversity in the health care workforce, increasing the capacity of community health centers, and promoting the use of patient navigators and community health workers.

Grade A References A1. Bagchi AD, Dale S, Verbitsky-Savitz N, et al. Examining effectiveness of medical interpreters in emergency departments for Spanish-speaking patients with limited English proficiency: results of a randomized controlled trial. Ann Emerg Med. 2011;57:248-256. A2. Paskett ED, Katz ML, Post DM, et al. The Ohio Patient Navigation Research Program: does the American Cancer Society patient navigation model improve time to resolution in patients with abnormal screening tests? Cancer Epidemiol Biomarkers Prev. 2012;21:1620-1628. A3. Byrd TL, Wilson KM, Smith JL, et al. AMIGAS: a multicity, multicomponent cervical cancer prevention trial among Mexican American women. Cancer. 2013;119:1365-1372. A4. Sequist TD, Fitzmaurice GM, Marshall R, et al. Cultural competency training and performance reports to improve diabetes care for black patients: a cluster randomized, controlled trial. Ann Intern Med. 2010;152:40-46.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 4  Cultural Context of Medicine  

GENERAL REFERENCES 1. Meghani SH, Brooks JM, Gipson-Jones T, et al. Patient-provider race-concordance: does it matter in improving minority patients’ health outcomes. Eth Health. 2009;14:107-130. 2. Trivedi AN, Grebla RC, Wright SM, et al. Despite improved quality of care in the Veterans Affairs Health Care System, racial disparity persists for some clinical outcomes. Health Aff. 2011;4: 707-715. 3. Lebrun LA, Shi L, Zhu J, et al. Racial/ethnic differences in clinical quality performance among health centers. J Ambul Care Manage. 2013;36:24-34. 4. VanderWielen LM, Enurah AS, Rho HY, et al. Medical interpreters: improvements to address access, equity, and quality of care for limited-English-proficient patients. Acad Med. 2014;89:1324-1327.

17.e1

5. Natale-Pereira A, Enard KR, Nevarez L, et al. The role of patient navigators in eliminating health disparities. Cancer. 2011;117(suppl):3543-3552. 6. Brownstein JN, Hirsch GR, Rosenthal EL, et al. Community health workers 101 for primary care providers and other stakeholders in health care systems. J Ambul Care Manage. 2011;34:210-220. 7. Betancourt JR, Cervantes MC. Cross-cultural medical education in the United States: key principles and experiences. Kaohsiung J Med Sci. 2009;25:472-478. 8. Chin MH, Clarke AR, Nocon RS, et al. A roadmap and best practices for organizations to reduce racial and ethnic disparities in health care. J Gen Intern Med. 2012;27:992-1000.

17.e2

CHAPTER 4  Cultural Context of Medicine  

REVIEW QUESTIONS 1. What percentage of immigrants to the United States have limited English proficiency? A. 20% B. 30% C. 40% D. 50% E. 60% Answer: D  The 2010 Census found that one in two immigrants have limited English proficiency. (Grieco EM, Acosta YD, de la Cruz GP, et al. The foreign born population in the United States: 2010. Washington, DC: U.S. Department of Commerce, 2012.) 2. A 19-year-old Vietnamese man presents for a routine physical examination. He is a recent immigrant, has no symptoms or significant medical history, and has not previously had a physical examination or any blood testing in the United States. Which of the following blood tests should you perform? A. HIV B. Hepatitis B C. Glucose D. Cholesterol E. All of the above Answer: B  More than half of the Americans who have chronic hepatitis B infection are Asians or Pacific Islanders. Therefore, all immigrants from Asia should be tested for hepatitis B. (Pollack H, Wang S, Wyatt Le, et al. A comprehensive screening and treatment model for reducing disparities in hepatitis B. Health Aff. 2011;30:1974-1983.) 3. Which of the following statements about health insurance is incorrect? A. Whites are more likely to have insurance than American Indians/ Alaska Natives. B. Latinos are less likely to have insurance than non-Latino whites. C. Gay/lesbian/hom*osexual individuals are less likely to have insurance than heterosexual individuals. D. Citizens are more likely to have insurance than noncitizens. E. Individuals who are proficient in English are more likely to have insurance than individuals who are not. Answer: C  The California Health Interview Survey provides information about health insurance coverage among population subgroups and shows differences by race/ethnicity, citizenship status, and level of English proficiency but not by sexual orientation. (University of California Los Angeles. Ask CHIS 2009, http://www.chis.ucla.edu. Accessed March 22, 2014.)

4. Which of the following are core community health worker functions? A. Cultural mediation B. Health education C. Informal counseling D. Social support E. All of the above Answer: E  Cultural mediation, health education, informal counseling, and social support are all core community health worker functions. (Brownstein JN, Hirsch GR, Rosenthal EL, Rush CH. Community health workers 101 for primary care providers and other stakeholders in health care systems. J Ambul Care Manage. 2011;34:210-220.) 5. Your practice has recently started seeing a large number of Hispanic immigrant patients with limited English proficiency. You have decided to make some changes to your practice to accommodate the specific needs of these patients. Which of the following approaches are appropriate? A. Hire Hispanic staff B. Ask limited English proficiency patients to bring a family member with them to provide C. Medical interpretation D. Provide cultural competency training to providers E. A and C F. All of the above Answer: D  Clinical practices can address health disparities in multiple ways, including by hiring staff who are representative of the practice population, providing professional interpreter services, offering cultural competency training to providers, and providing linguistically appropriate patient education materials. Family members should not be asked to provide medical interpretation. (Washington DL, Bowles J, Saha S. Transforming clinical practice to eliminate racial-ethnic disparities in healthcare. J Gen Intern Med. 2008;23:685-691.)

17

CHAPTER 5  Socioeconomic Issues in Medicine  

5  SOCIOECONOMIC ISSUES IN MEDICINE STEVEN A. SCHROEDER

All nations—rich and poor—struggle with how to improve the health of the public, obtain the most value from medical services, and restrain rising health care expenditures. Many developed countries also wrestle with the paradox that their citizens have never been so healthy or so unhappy with their medical care. Despite the reality that only about 10% of premature deaths result from inadequate medical care, the bulk of professional and political attention focuses on how to obtain and pay for state-of-the-art medical care. By comparison, 40% of premature deaths stem from unhealthy behaviors— including smoking (about 44%; Chapter 32), excessive or unwise drinking (about 11%; Chapter 33), obesity and insufficient physical activity (about 15% but estimated to rise substantially in the years to come; Chapters 16 and 220), illicit drug use (about 2%; Chapter 34), and imprudent sexual behavior (about 3%; Chapter 285). Genetics (Chapter 40) account for an additional 30%; social factors—discussed next—account for 15%, and environmental factors (Chapter 19) account for 5%. Of the major behavioral causes of premature deaths, tobacco use (Chapter 32) is by far the most important, although recent increases in obesity (Chapter 220) and physical inactivity (Chapter 16) are also alarming. Health is influenced by genetic predisposition, behavioral patterns, environmental exposures, social circ*mstances, and health care.

SOCIAL STATUS INFLUENCES HEALTH

Socioeconomic status, or class, is a composite of many different factors, including income, net wealth, education, occupation, and neighborhood. In general, people in lower classes are less healthy and die earlier than people at higher socioeconomic levels, a pattern that holds true in a stepwise fashion from the poorest to the richest. In the United States, the association between health and class is usually discussed in terms of racial and ethnic disparities; but in fact, race and class are independently associated with health status, and it can be argued that class is the more important factor. For example, U.S. racial disparities in the prevalence of adult smoking are relatively small among whites, blacks, and Hispanic Americans, whereas there are huge differences among smoking rates by educational level (Fig. 5-1).1 U.S. physicians have reduced their smoking prevalence to a record low of only 1%. Although both smoking rates and the numbers of cigarettes smoked by those who continue to smoke are gradually declining (Fig. 5-2), more than 43 million Americans and millions more elsewhere continue to smoke.2 Because people of higher socioeconomic status adopt health-promoting behaviors at a faster rate than people of lower socioeconomic status, overall population health can increase while health disparities also widen (Fig. 5-3).

No high school diploma

25.5%

GED diploma

45.3%

High school graduate

23.8%

Some college

22.3%

Undergraduate degree

9.3%

Graduate degree

5.0% 0%

10%

20%

30%

40%

50%

FIGURE 5-1.  Prevalence of adult smoking, by education, United States, 2011. GED = General Education Development. (From Centers for Disease Control and Prevention. Current cigarette smoking among adults: United States, 2011. MMWR Morb Mortal Wkly Rep. 2012;61:889-894.)

18

CHAPTER 5  Socioeconomic Issues in Medicine  

45 40

Smoking prevalence (%)

35

Average number cigarettes smoked/day per smoker

30 25 20 15 10 5 0

1965 1970 1976 1978 1980 1985 1988 1991 1993 1995 1998 2000 2002 2004 2006 2008 2010 2012

Percent/number of cigarettes smoked daily

In part, the relationship between class and health is mediated by higher rates of unhealthy behaviors among the poor, such as the inverse relationship between educational attainment and cigarette smoking, but unhealthy behaviors do not fully explain the poor health of those in the lower socioeconomic classes. Even when such behaviors are held constant, people in lower socioeconomic classes are much more likely to die prematurely than are people of higher classes. Of interest is that first-generation immigrants to the United States appear to be more protected from the adverse health consequences of low socioeconomic status than are subsequent generations. It is unclear which of the components of class—education, wealth (either absolute wealth or the extent of the gap between rich and poor), occupation, or neighborhood—makes the greatest impact on a person’s health. Most likely, it is a combination of all of them. For example, the constant stress of a lower class existence—lack of control over one’s life circ*mstances, social isolation, and the anxiety derived from the feeling of having low status—is linked to poor health. This stress may trigger a variety of neuroendocrinologic responses that are useful for short-term adaptation but bring long-term adverse health consequences. What can clinicians do with this knowledge? Clearly, it is difficult to write prescriptions for more income, a better education, good neighborhoods, or high-paying jobs. Physicians can, however, encourage healthy behavior. At key times of transition, such as during discharge planning for hospitalized patients, clinicians should be attentive to social circ*mstances. For patients who are likely to be socially isolated, clinicians should encourage or arrange interactions with family, neighbors, religious organizations, or community agencies to improve the likelihood of optimal outcomes. Access points to vital social services, such as child care, disability insurance, and food supplementation, can be provided in clinical settings.3 In addition, physicians should seek to identify and eliminate any aspects of racism in health care institutions (Chapter 4). Finally, in their role as social advocates, physicians

FIGURE 5-2.  Smoking prevalence and average number of cigarettes smoked per day per current smoker. (Data based on Centers for Disease Control and Prevention (CDC). Smoking prevalence, 1965-2010. MMWR Morb Mortal Wkly. 2011;60:109-113; Current cigarette smoking in the United States: current estimate. CDC; http://www.cdc.gov/ tobacco/data_statistics/fact_sheets/adult_data/cig_smoking. Accessed February 10, 2015; National Health Interview Survery. CDC; http://www.cdc.gov/nchs/nhis/quest_ data_related_1997_forward.htm. Accessed February 10, 2015; Jamal A, Agaku IT, O’Connor E, et al. Current cigarette smoking among adults—United States, 2005-2013. MMWR Morb Mortal Wkly. 2014;63:1108-1112.)

Health

Gap

Upper SES Lower SES Time FIGURE 5-3.  Health improves while disparities widen. SES = socioeconomic status.

can promote such goals as safe neighborhoods, improved schools, and access to quality health care.

ECONOMIC ISSUES IN MEDICAL CARE

Medical care today is on a collision course. On the one hand, an everexpanding science base continuously generates new technologies and drugs that promise a longer and healthier life. Add a public eager to obtain the latest breakthroughs touted in the media and over the Internet, plus a well-stocked medical industry eager to meet that demand, and it is easy to understand why expenditures continue to soar. On the other hand, payers for medical care— health insurance companies, government (federal, state, and local), and employers—increasingly bridle at medical care costs. The United States continues to lead the world in health care expenditures.4 In 2011, it spent more than $2.7 trillion, amounting to 17.9% of the gross domestic product. Most policy analysts contend that this rate of increase in medical care expenditures is unsustainable, but this claim has been made for many years. A potent combination of supply and demand factors explains why the United States spends so much.5 On the supply side, the United States far exceeds other countries in the availability and use of expensive diagnostic technologies, such as magnetic resonance imaging and computed tomography. For example, the United States has four times as many magnetic resonance imaging machines per capita as does Canada. Similar patterns exist for therapeutic technologies, whether coronary angioplasty, cancer chemotherapy, or joint prostheses. The differences are especially dramatic in older patients. Other supply factors that drive high medical expenditures in the United States include a fee-for-service payment system that compensates physicians much more when they use expensive technologies than when they do not6; a medical professional work force that earns much higher incomes relative to the population than in other nations and that emphasizes specialist rather than generalist practice; accelerated development of new and costly medications that are directly marketed to consumers; much higher administrative costs; higher rates of fraud and abuse; and a high rate of defensive medicine in response to pervasive fears about medical malpractice suits. Supply factors that do not appear to be unique to the United States are the number of physicians or hospitals. Many other developed countries have a much larger physician work force relative to their population, as well as a much higher ratio of primary care physicians to specialists. The number of hospitals and hospital beds, the frequency of hospitalizations, and the length of hospital stay are relatively low in the United States, although it does have a much greater proportion of intensive care beds. Finally, recent analyses suggest that a principal driver of high expenditures on health care in the United States is the much greater price charged per unit of service compared with other developed countries. Demand factors also drive medical expenditures. The extent to which the media and the medical profession feature medical “breakthroughs” is extensive and one-sided. New promising treatments merit front-page stories and commercial advertisem*nts, whereas subsequent disappointing results are buried or ignored. The cumulative result is to whet patients’ appetite for more and to leave the impression that good health depends only on finding the right treatment. This same quest explains the popularity of alternative medicine, for which patients are willing to spend $34 billion annually out of their own pockets (Chapter 39). The cumulative impact of these supply and demand drivers is that there are incentives to do more at every step of the American medical system.5 It could be argued that rising expenditures for medical care are not a bad thing. What could be more important than ensuring maximal health? There are several rebuttals to that argument. First, it is not clear that money spent on medical care brings appropriate value in the United States, given that its health statistics are worse than those of virtually every other developed country. Second, there are substantial regional differences in the supply and use of medical care, such as a two-fold difference in the supply of acute hospital beds and a four-fold difference in the risk of being hospitalized in an intensive care unit at the end of life. Similar regional differences exist for procedures such as transurethral prostatectomy, hysterectomy, and coronary artery bypass surgery. Yet there is no evidence that “more is better” on a regional basis. Consequently, rising health care expenditures are stressing public programs such as Medicare, Medicaid, the Veterans Administration health system, and municipal hospitals, with budget requests outstripping the tax base to pay for them. Medical debt is by far the most important cause of bankruptcy. Finally, as health care becomes less affordable for businesses and

government, the number of people without health insurance will continue to increase.

Cost-Containment Strategies

Since the mid-1970s, a variety of strategies to contain rising medical expenditures have yielded limited success.5 These attempts have tried to restrict the supply of costly medical technologies as well as the production of physicians, especially specialists; to promote health maintenance organizations that have incentives to spend less on medical care; to ration indirectly by limiting health insurance coverage; to institute prospective payment for hospital care; to use capitation payments or discounted fee schedules for physician reimbursem*nt; to introduce gatekeeper mechanisms to reduce access to costly care; to put patients at more financial risk for their own medical care; to reform malpractice procedures; to reduce administrative costs; and to encourage less aggressive care at the end of life. The most recent suggestions— comparative effectiveness research to curtail the use of unnecessary technology, electronic medical records to avoid duplication of tests, payment for performance, accountable care organizations that change payment incentives—all hold promise to improve quality, but their potential for substantial cost reduction is only theoretical at present. Recently, however, the rate of increase in health care expenditures has slowed relative to the gross domestic product.7 Two basic hypotheses have been offered: the recession that began in 2008, and heightened cost consciousness among hospitals, health insurers, and some physician groups. Payment for medical care varies by country. In the United States, health insurance coverage is an incomplete patchwork, consisting of governmentsponsored programs for elderly people (Medicare), poor people (Medicaid), and veterans, plus employer-based coverage for workers and their families. Medicare covers acute care services in the hospital and in physicians’ offices but has limited coverage for prescription drugs and long-term care. More than half of all Medicare subscribers also buy supplemental insurance. Medicaid covers more services than Medicare does, but Medicaid payments to physicians and hospitals are so low in many states that patients have restricted access to care. At any given time, more than 44 million Americans have lacked health insurance, and 70 million have been without insurance at some point during the year. In addition, millions of immigrant workers are also uninsured. The lack of health insurance contributes to poor health, such as delayed diagnosis and undertreatment of asthma, diabetes, hypertension, and cancer. The 2010 Patient Protection and Affordable Care Act (ACA) contains numerous insurance reform features that took effect in 2010 and 2011, as well as coverage expansions that began in 2014.8 The ACA was originally expected to cover 32 million previously uninsured Americans, with about half enrolling in subsidized private insurance plans and half in expanded state Medicaid programs. However, a 2012 Supreme Court decision gave states the choice of opting out of the Medicaid expansion. As a result, only 27 states plus the District of Columbia accepted that expansion. States that opted out of Medicaid expansion, such as Texas, tend to be those with the highest proportion of uninsured—mainly poor—people. In addition, various coverage components, especially regarding contraception, continue to be litigated. Revenuegenerating provisions of the ACA are split about evenly between spending reductions and cost containment. In contrast to what happened after the passage of Medicare and Medicaid, the ACA continues to be highly controversial politically, and thus subject to potential changes, depending on election results. Because medical care is both so valued and so expensive, physicians everywhere will inevitably become more involved in issues of medical economics. As cost-containment pressures force patients to assume more of their medical expenses, patients will become more aware of costs and more demanding about the price and value of care. In addition, knowledge will continue to accumulate about the real and potential harm from unnecessary or marginally useful medical services. Thus, informed clinical decision making will require that physicians have accurate information about the risks, benefits, and costs of medical care and better ways to communicate what is known and what is not. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 5  Socioeconomic Issues in Medicine  

GENERAL REFERENCES 1. Centers for Disease Control and Prevention. Current cigarette smoking among adults: United States, 2011. MMWR Morb Mortal Wkly Rep. 2012;61:889-894. 2. Jha P, Ramasundarahettige C, Landsman V, et al. 21st Century hazards of smoking and benefits of cessation in the United States. N Engl J Med. 2013;368:341-350. 3. Gottlieb L, Sandel M, Adler NE. Collecting and applying data on social determinants of health in health care settings. JAMA Intern Med. 2013;173:1017-1020. 4. Lorenzoni L, Belloni A, Sassi F. Health-care expenditure and health policy in the USA versus other high-spending OECD countries. Lancet. 2014;384:83-92.

19.e1

5. Schroeder SA. Personal reflections on the high cost of American medical care. Arch Intern Med. 2011;171:722-727. 6. Schroeder SA, Frist W. Phasing out fee-for-service payment. N Engl J Med. 2013;368:2929-2932. 7. Fuchs VR. The gross domestic product and health care spending. N Engl J Med. 2013;369: 107-109. 8. Shaw FE, Asomugha CN, Conway PH, et al. The Patient Protection and Affordable Care Act: opportunities for prevention and public health. Lancet. 2014;384:75-82.

19.e2

CHAPTER 5  Socioeconomic Issues in Medicine  

REVIEW QUESTIONS 1. Preventing premature mortality is a prime goal for all clinicians. Of the following statements regarding premature mortality, which one is incorrect? A. About 10% of premature deaths could be prevented by assuring highquality medical care to all. B. Among the various general causes of premature deaths, behavioral factors such as risky sexual behavior, alcohol and drug abuse, smoking, and obesity and physical inactivity are the most important and also offer opportunities for remediation. C. Among behavioral factors, the most important cause of premature death is smoking cigarettes. D. As smoking prevalence gradually decreases, the remaining smokers are smoking more cigarettes per day. E. Based on current trends, obesity and physical inactivity will likely become more important causes of premature deaths. Answer: D  Statement A is correct. As important as good medical care is, the lack of such services only accounts for about 10% of premature deaths. Statements B, C, and E are correct. As regards statement D, at the same time as overall smoking prevalence has declined, the number of daily cigarettes consumed by those who continue to smoke has also declined. This decline is probably of function of several factors, including rising tobacco taxation that makes smoking more expensive, the spread of clean indoor air laws, and the increasing stigma attached to smoking. 2. Increasing evidence points to the important health impacts of social economic status (SES). Which one of the following statements regarding health and social class is correct? A. Among the various components of SES, racial and ethnic characteristics contribute the most to health disparities. B. Virtually all of the worst health among low SES populations can be explained by personal behaviors such as cigarette smoking and drug and alcohol abuse. C. The relationship between SES and health is concentrated among those with low SES. In other words, for people above 400% of the poverty level, SES does not contribute to health status. D. Because so many of the determinants of low SES (e.g., education, income, housing, net wealth) are outside the purview of clinicians, they should not be concerned about them. E. Co-location of access to important social services—such as food stamps or disability payments—at medical sites can improve the social status of selected patients. Answer: E  Answer A is incorrect. Class, as measured by income, net wealth, educational status, and neighborhood, is the most important factor leading to health disparities. Regarding answer B, although personal behaviors such as smoking and physical activity are important, many other factors associated with low social class contribute to poor health. Regarding answer C, there is a stepwise association between SES and health, even at the higher levels of SES. For example, those in the top decile of SES enjoy better health than those in the ninth decile, even though both deciles have high SES. For answer D, it is true that clinicians would have a hard time improving these factors. Nevertheless, they should be conscious of them because they often influence treatment strategies (e.g., the ability to obtain nutritious food or to buy medications). Finally, answer E is correct. For example, convenient provision of food stamps for those with food insecurity would help stabilize diabetic patients. 3. Regarding per capita medical expenditures, population health, and access to medical care, which statement best expresses the performance of the United States versus other developed nations? A. It leads the world in medical expenditures but trails badly in health outcomes and access to health care services. B. It trails the world in medical expenditures, health outcomes, and access to care. C. It is about in the middle for expenditures, health, and access. D. It leads the world in expenditures, health, and access. E. It leads the world in expenditures and health but lags in access.

Answer: A  Answer A is correct as written. Of the three statements in answer B, only the second is correct. Of the three answers in C, none are correct. Of those in D, only the first is correct. And of those in E, only the first is correct. 4. Regarding the causes of rising medical expenditures in the United States, which answer is correct? A. Fee-for-service payment to physicians is no longer a major driver of cost escalation because it will soon be replaced by bundled or capitated payment. B. The United States is an outlier in that it has more hospital beds per capita and a higher length of stay. C. The United States is an outlier in that it features more physicians per capita. D. There are multiple factors responsible for the patterns of expenditures in the United States, and it is unlikely that any one bears major responsibility for the high expenditure profile. E. The threat and reality of malpractice suits, in a country with the highest number of lawyers per capita, is the major reason for high U.S. medical expenditures. Answer: D  For answer A, although there is much current talk about the imminent demise of fee-for-service payment, it remains the dominant way of paying physicians. Regarding answer B, the United States actually has fewer hospital beds per capita than other developed countries as well as a shorter length of stay. The cost of a day in the hospital, however, is far higher in the United States. Similarly, for answer C, the United States has fewer physicians per capita than most developed nations but has a much higher proportion of specialists. Answer D, the correct answer, reflects the multiple reasons that medical care is more costly in the United States. Regarding answer E, malpractice suits are indeed more of a factor in the United States than other nations. Nevertheless, if all malpractice costs were to vanish, the United States would still have the most expensive health care system by far. 5. Regarding responses to high medical expenditures in the United States, which of the following statements is most correct? A. Because expenditures on medical care are so essential and because wealthy countries characteristically spend more on health care as wealth increases, there is little interest in curtailing rising medical expenditures. B. The cost-containment measures found in the Patient Protection and Affordable Care Act will be sufficient to rein in rising medical costs. C. The combination of pressures on personal and governmental health care spending and crowding out of other social expenditures makes it likely that more intense cost-containment activity will occur. D. The combination of widespread electronic medical record use and the spread of Accountable Care Organizations will be sufficient to curb rising medical expenditures. E. Political obstacles to cost containment, such as the fear of rationing, will not be a problem. Answer: C  Answer A is incorrect. There is emerging consensus that something must be done to curtail rising medical expenditures (as phrased in answer C), but not on how to accomplish that. Although there are some costcontainment features in the Patient Protection and Affordable Care Act (PPACA) (answer B), it is unproved whether they will reduce medical expenditures. Answer C is correct. Regarding answer D, there is no good evidence that either the spread of electronic medical records or Accountable Care Organizations will curb rising medical expenditures, despite the enthusiasm of advocates for those programs. Finally, political obstacles (answer E) remain potent barriers to medical cost containment, such as the assertion during the debate over the PPACA that access to palliative care services would be tantamount to creating death panels.

19

CHAPTER 6  Global Health  

6  GLOBAL HEALTH ARUN CHOCKALINGAM

Health is a human right, but more than 2 billion people live with a daily income of less than $2 and have no access to good health care. Health is determined by the context of people’s lives. Individuals are unable to control many of the social determinants of health (Chapter 5), such as income and social status, education, physical environment, social support network, genetics, health services, and gender.1 In the process of modernization from a less developed to a more developed nation, the epidemiologic transition of modern sanitation, medications, and health care has drastically reduced infant and maternal mortality rates and extended average life expectancy. As a result, the world has progressed from the age of pestilence and famine, with a life expectancy between 20 and 40 years, to the age of receding pandemics, with a life expectancy of 30 to 50 years, and now to the current age of degenerative and man-made diseases, with a life expectancy of 60 years or more. These trends, coupled with subsequent declines in fertility rates, have driven a demographic transition in which the major causes of death change from infectious diseases to chronic and degenerative diseases.2 As many countries around the world have undergone globalization, owing to their internal urbanization, modernization, and economic development, an increased proportion of their burden of morbidity and mortality is now due to chronic noncommunicable diseases, including cardiovascular, cerebrovascular, and renovascular diseases as well as cancer, diabetes, chronic respiratory diseases, and mental disorders (Table 6-1).

WHAT IS GLOBAL HEALTH?

The term global health is sometimes confused with public health, international health, tropical medicine, and population health. Global health, which is defined as the health of populations in a global context, transcends the perspectives and concerns of individual nations and crosses national borders. Global health depends on the public health efforts and institutions of all countries, including their strategies for improving health, both populationwide and for individuals. Global health depends on multiple factors, including social, political, environmental, and economic determinants of health. Although global health often focuses on improving the health of people who live in low- and middle-income countries, it also includes the health of any marginalized population in any country. Global health requires use of a wide range of institutions that collaborate in addressing all health issues. Global health also depends on the constructive use of evidence-based information to provide health and health equity, in part by strengthening primary health care and the health care delivery system.

Millennium Development Goals

In an attempt to address global inequity, the United Nations advanced eight millennium development goals with the objective of achieving these goals between 2000 and 2015. These eight goals incorporate 21 targets (Table 6-2), with a series of measurable health and economic indicators for each target.3 Although many of the targets have not yet been achieved, substantial progress has been made toward all targets. The millennium development goals emphasize that health and development are interconnected. To address global inequity, fundamental issues include reducing poverty, improving education, and empowering people. In addition to specific goals for reducing infant and child mortality, maternal mortality, and mortality due to infectious diseases such as human immunodeficiency virus infection/acquired immunodeficiency syndrome (HIV/ AIDS), malaria, and tuberculosis, the millennium development goals strongly encourage environmental sustainability and global partnership.

GLOBAL BURDEN OF DISEASES

The global burden of disease is measured in terms of total and cause-specific mortality and morbidity as well as the national economic burden for health care. The Global Burden of Diseases, Injuries and Risk Factors Study 20104 shows that an estimated 53 million people died from all causes in 2010, with

20

CHAPTER 6  Global Health  

TABLE 6-1  EPIDEMIOLOGIC TRANSITION IN CARDIOVASCULAR DISEASES

STAGES OF DEVELOPMENT

LIFE EXPECTANCY

BURDEN OF CARDIOVASCULAR DISEASE DEATHS,   % OF TOTAL DEATHS

1. Age of pestilence and famine

20-40 years

5-10

2. Age of receding pandemics

30-50 years

3. Age of degenerative and man-made diseases

50->60 years

3A. Age of delayed degenerative diseases

>60 years

3B. Age of health regression and social upheaval

50-60 years

PREDOMINANT CARDIOVASCULAR DISEASES AND RISK FACTORS

MODERN REGIONAL EXAMPLES

Infections, rheumatic heart disease, and nutritional cardiomyopathies

Rural India, sub-Saharan Africa, South America

10-35

As above plus hypertensive heart disease and hemorrhagic strokes

China

35-65

All forms of strokes; ischemic heart disease at young ages; increasing obesity and diabetes

Aboriginal communities, urban India, former socialist economies

Stroke and ischemic heart disease at old age

Western Europe, North America, Australia, New Zealand

Re-emergence of deaths from rheumatic heart disease, infections, increased alcoholism and violence; increase in ischemic and hypertensive diseases in the young

Russia

6.5) (Chapter 425). Patients who use smokeless tobacco products are at significantly increased risk for premalignant and malignant oral lesions (Chapter 32). Bimanual palpation of the cheeks and floor of the mouth facilitates identification of potentially malignant lesions (Chapter 425).

Eyes

The eye examination begins with simple visual inspection to look for symmetry in the lids, extraocular movements, pupil size and reaction, and the presence of redness (Chapters 423 and 424). Abnormalities in extraocular movements should be grouped into nonparalytic (usually chronic with onset in childhood) or paralytic causes (third, fourth, or sixth cranial nerve palsy). Pupillary abnormalities may be symmetrical or asymmetrical (anisocoria). Red eyes should be categorized by the pattern of ciliary injection, presence of

pain, effect on vision, and papillary abnormalities. When the eye examination is approached systematically, the generalist physician can evaluate the likelihood of conjunctivitis, episcleritis or scleritis, iritis, and acute glaucoma. Routine determination of visual acuity can confirm a patient’s report of diminished vision but does not replace the need for formal ophthalmologic evaluation in patients with visual complaints (Chapter 423). Aging patients often experience acute flashes and floaters, especially with posterior vitreal detachments. If acute flashes and floaters are associated with visual loss, the patient should be urgently referred for an ophthalmologic examination for the evaluation of a possible acute retinal detachment.9 Cataracts can be detected with direct ophthalmoscopy, but the generalist’s proficiency in this evaluation is uncertain. After identifying the optic disc by ophthalmoscopy, the examiner should note the border of the disc for clarity, color, and the size of the central cup in relation to the total diameter (usually less than half the diameter of the disc). A careful observer usually can see spontaneous venous pulsations that indicate normal intracranial pressure, but about 10% of patients with normal intracranial pressure will not have spontaneous pulsations. Abnormalities of the optic disc include optic atrophy (a white disc), papilledema (see Fig. 423-27) (blurry margins with a pink, hyperemic disc), and glaucoma (a large, pale cup with retinal vessels that dive underneath and that may be displaced toward the nasal side). The generalist’s examination inadequately detects early glaucomatous changes, so high-risk patients should undergo routine ophthalmologic examination for glaucoma.10 After inspecting the disc, the upper and lower nasal quadrants should be examined for the appearance of vessels and the presence of any retinal hemorrhages (see Fig. 423-24) or lesions. Proceeding from the nasal quadrants to the temporal quadrants decreases the risk for papillary constriction from the bright light focused on the fovea. Dilating the pupils leads to an improved examination. Patients with diabetes (Chapter 229) should undergo routine examination by eye care experts because the sensitivity of a generalist’s examination is not adequate to exclude diabetic retinopathy or monitor it over time.

Neck Carotid Pulses

The carotid pulses should be palpated for contour and timing in relation to the cardiac impulse. Abnormalities in the carotid pulse contour reflect underlying cardiac abnormalities (e.g., aortic stenosis) but are generally appreciated only after detecting an abnormal cardiac impulse or murmur (Chapter 51). Many physicians listen for bruits over the carotid arteries because asymptomatic carotid bruits are associated with an increased incidence of cerebrovascular and cardiac events in older patients (Chapters 406 and 407). In asymptomatic patients, the presence of a carotid bruit increases the likelihood of a 70 to 90% stenotic lesion (LR 4 to 10), but the absence of a bruit is of uncertain value. Unfortunately, clinical data do not provide adequate data for judging the importance of detecting bruits in asymptomatic patients.

Jugular Veins

The examination of the neck veins is an interesting but often unreliable indicator of central venous pressure or fluid responsiveness in hospitalized sick patients (Chapter 51).11 Inspection of the waveforms may facilitate the interpretation of the cardiac examination for right heart valvular lesions. The waves are seen best by shining a penlight obliquely on the vein while the examiner looks for the dynamic changes of the projected shadow on the bed linen.

Thyroid

The thyroid gland is felt best when standing behind the patient and using both hands to palpate the thyroid gland gently (Chapter 226). Palpation is enhanced when the patient swallows sips of water to allow the thyroid to glide underneath the fingers. When viewed from the side, lateral prominence of the thyroid between the cricoid cartilage and the suprasternal notch indicates thyromegaly. The generalist physician should estimate the size of the thyroid gland as normal or enlarged; the impression of an enlarged thyroid gland by a generalist physician has an LR of almost 4, whereas assessment of normal size makes thyromegaly less likely (LR 0.4). The volume of a normal thyroid gland is no greater than the volume of the patient’s distal thumb phalanx.

Lymphatic System

While palpating the thyroid, the examiner may also identify enlarged cervical lymph nodes (Chapter 168). Lymph nodes can also be palpated in the supraclavicular area, axilla, epitrochlear area, and inguinofemoral region. Simple

CHAPTER 7  Approach to the Patient: History and Physical Examination  

lymph node enlargement confined to one region is common and does not usually represent an important underlying disorder. Unexpected gross lymph node enlargement in a single area or diffuse lymph node enlargement is more important. Patients with febrile illnesses, underlying malignancy, or inflammatory diseases should routinely undergo an examination of each of the aforementioned areas for lymph node enlargement.

Chest

Inspection of the patient’s posture may reveal lateral curves in the back (scoliosis) or kyphosis that may be associated with loss of vertebral height from osteoporosis (Chapter 243). When patients have back pain, the spine and paravertebral muscles should be palpated for spasm and tenderness (Chapter 400). The patient may be placed through maneuvers to assess loss of mobility associated with ankylosing spondylitis (Chapter 265), but a history of loss of lateral mobility may be just as efficient in the early stages of spondylitis.

Lungs

The incremental value of palpation and percussion of the chest to supplement the history, auscultation, and eventual chest radiograph is unknown. Normal vesicular sounds, which approximate a 3:1 inspiratory:expiratory ratio with no pause between phases, are heard throughout most of the normal posterior chest during quiet breathing. Auscultated wheezes are continuous adventitial sounds. Crackles (formerly called rales) are discontinuous sounds heard in conditions that stiffen the lung (heart failure, pulmonary fibrosis, and obstructive lung disease). The best piece of information for increasing the likelihood of chronic obstructive pulmonary disease is a history of more than 40 pack years of smoking (LR 19). The presence of wheezing or downward displacement of the larynx to within 4 cm of the sternum (distance between the top of the thyroid cartilage and the suprasternal notch) increases the likelihood of obstructive pulmonary disease (LR of 4 for either).

Heart

The patient should be examined in the sitting and lying positions (Chapter 51). Palpation of the apical impulse in the left lateral decubitus position helps detect a displaced apical impulse and can reveal a palpable S3 gallop. When the apical impulse is lateral to the midclavicular line, radiographic cardiomegaly (LR 3.5) and an ejection fraction of less than 50% (LR 6) are more likely. Most examiners auscultate in sequence the second right then the second left intercostal spaces, the left sternal border, and then the apex. The examiner should concentrate on the timing, intensity, and splitting of sounds with respiration. The first and second heart sounds are heard best with the diaphragm, as are pericardial rubs. Gallops (S3 and S4) are heard best with the stethoscope bell. High-pitched versus low-pitched murmurs are detected by switching from the diaphragm to the bell. The location, timing, intensity, radiation patterns, and respiratory variation of murmurs should be noted. Special maneuvers during auscultation (e.g., Valsalva, auscultation during sudden squatting or standing) do not usually need to be performed if the results of routine precordial examination are entirely normal. The presence of an S3 gallop is useful for detecting left ventricular systolic dysfunction (LR > 4 for identifying patients with an ejection fraction of 4 cm in diameter). However, palpation misses a substantial proportion of small to medium aneurysms (Chapter 78). The presence of bowel sounds in patients with acute symptoms can be falsely reassuring because the sounds can be present despite an ileus and may be increased early in an obstruction. For patients without gastrointestinal symptoms or abnormalities on palpation, auscultation for bruits is important primarily to detect renal bruits in patients with hypertension (Chapters 67 and 125). The presence of an abdominal bruit in a hypertensive patient, if heard in systole and diastole, strongly suggests renovascular hypertension (LR ≈ 40).

Liver

Detection of liver disease depends mostly on the history and laboratory evaluations (Chapter 146). By the time that signs are present on physical examination, the patient usually has advanced liver disease. The first abnormalities on physical examination associated with liver disease are extrahepatic. The clinician should assess the patient for ascites, peripheral edema, jaundice, or splenomegaly. In patients with an enlarged liver, palpation should begin at the liver edge, but palpation of the edge below the costal margin increases the likelihood of hepatomegaly only slightly (LR 1.7). The upper border of the liver may be detected by percussion, and a span of less than 12 cm reduces the likelihood of hepatomegaly. In the absence of a known diagnosis (e.g., a hepatoma, which may cause a hepatic bruit), auscultation of the liver rarely is helpful.

Spleen

Examination for splenomegaly in patients without findings suggestive of a disorder associated with splenomegaly almost always reveals nothing (Chapter 168). Approximately 3% of healthy teenagers may have a palpable spleen. The examination for an enlarged spleen begins first with percussion in the left upper quadrant to detect dullness. Palpation can be performed by any of the following three approaches (κ ≈ 0.2 to 0.4): palpating with the right hand while providing counterpressure with the left hand behind the spleen, palpating with one hand without counterpressure (with the patient in the right lateral decubitus position for both techniques), or placing the patient supine with the left fist under the left costovertebral angle while the examiner tries to hook the spleen with the hands.

Musculoskeletal System

The musculoskeletal examination in adult patients is almost always driven by symptoms (Chapters 256 and 263). Most patients have back pain at some point during their life (Chapter 400). The patient’s history helps assess the likelihood of an underlying systemic disease (age, history of systemic malignancy, unexplained weight loss, duration of pain, responsiveness to previous therapy, intravenous drug use, urinary infection, or fever). The most important physical examination findings for lumbar disc herniation in patients with sciatica all have excellent reliability, including ipsilateral straight leg raising causing pain, contralateral straight leg raising causing pain, and ankle or great toe dorsiflexion weakness. The generalist physician should evaluate an adult patient with knee discomfort for torn menisci or ligaments. The best maneuver for demonstrating a tear in the anterior cruciate ligament is the anterior drawer or Lachman maneuver, in which the examiner detects the lack of a discrete end point as the tibia is pulled toward the examiner while the femur is stabilized. A variety of maneuvers that assess for pain, popping, or grinding along the joint line between the femur and tibia are used to evaluate for meniscal tears. As with many musculoskeletal disorders, no single finding has the accuracy of the orthopedist’s examination, which factors in the history and a variety of clinical findings. The shoulder examination is directed toward determining range of motion, maneuvers that cause discomfort, and assessment of functional disability.

Hip osteoarthritis is detected by evidence of restriction of internal rotation and abduction of the affected hip. Generalist physicians often rely on radiographs to determine the need for referral to orthopedic physicians, but routine radiographs are not needed early in the course of shoulder or hip disorders. The degree of pain and disability experienced by the patient may prompt confirmation of the diagnosis and referral. The hands and feet may show evidence of osteoarthritis (local or as part of a systemic process) (Chapter 262), rheumatoid arthritis (Chapter 264), gout (Chapter 273), or other connective tissue diseases. In addition to regional musculoskeletal disorders, such as carpal tunnel syndrome, a variety of medical and neurologic conditions should prompt routine examination of the distal ends of the extremities to prevent complications (e.g., diabetes [neuropathy or ulcers] or hereditary sensorimotor neuropathy [claw toe deformity]).

Skin

The skin should be examined under good lighting (Chapter 436). It is best to ask the patient to point out any spots on the skin of concern. Examiner agreement on some of the most important features of melanoma (asymmetry, haphazard color, border irregularity) is fair to moderate (Chapter 203). A lesion that is symmetrical, has regular borders, is only one color, is 6 mm or smaller, or has not enlarged in size is unlikely to represent a melanoma (LR 0.07). However, an increasing number of findings greatly enhance the likelihood of melanoma (LR 2.6 for two or more findings and LR 98 for the presence of all five findings) (Chapter 203). Basal cell carcinoma and squamous cell carcinoma occur more frequently than melanoma (Chapter 203). These lesions can be detected during routine examination by paying careful attention to sun-exposed areas of the nose, face, forearms, and hands.

Neurologic Examination

Full details of the neurologic examination are given in Chapter 396.

Psychiatric Evaluation

During the general examination, much of the psychiatric assessment (including cognition) is accomplished while eliciting the routine history and performing the review of systems (Chapter 397). Observation of the patient’s mannerisms, affect, facial expression, and behavior may suggest underlying psychiatric disturbances. When a screening survey and review of systems are obtained by a questionnaire completed by the patient, the clinician should review the responses carefully to determine whether the patient exhibits symptoms of depression. Specific questioning for symptoms of depression is appropriate for all adult patients. Military veterans should be screened for post-traumatic stress disorder and possible prior traumatic brain injuries that may affect their behaviors. Delirium (Chapter 28) is common in both medical and surgical inpatients and is recognized by fluctuating mental status. Delirium should be suspected when the patient has trouble carrying on a normal conversation during bedside rounds; but the patient’s nurse and visitors may detect delirium before the physician, so their report may contribute to diagnosis.12

Genitalia and Rectum Pelvic Examination

A complete examination includes a description of the external genitalia, appearance of the vagin* and cervix as seen through a speculum, and bimanual palpation of the uterus and ovaries (Chapters 199 and 237). About 10 to 15% of asymptomatic women have some abnormality on examination, and 1.5% have abnormal ovaries. However, screening for ovarian cancer is limited by the low sensitivity of the physical examination for detecting early-stage ovarian carcinoma (Chapter 199). In the emergency setting, all women of reproductive age with vagin*l bleeding and pelvic pain should have a pregnancy test and an ultrasound to evaluate them for a possible ectopic pregnancy.13

Male Genitalia

Examination of the male genitalia should begin with a description of whether the penis is circumcised and whether there are any visible skin lesions (e.g., ulcers or warts). Palpation should confirm the presence of bilateral testes in the scrotum. The epididymis and testes should be palpated for nodules. The low incidence of testicular carcinoma means that most nodules are benign (Chapter 200). The prostate should be examined in all quadrants, with attention focused on surface irregularities or differences in consistency throughout the prostate

(Chapter 201). An estimate of prostate size may be confounded by the size of the examiner’s fingers. It may be best to estimate the size of the prostate in centimeters of width and height.

Rectum

Patients can be examined while lying on their side, although this approach may place the examiner in an awkward stance (Chapters 132 and 145). The rectal examination in women can be performed as part of a bimanual examination, with the index finger in the vagin* and the third finger in the rectum to permit palpation of the rectovagin*l vault. Men may be asked to stand and lean over the examining table; alternatively, they may be examined while on their back with their hips and knees flexed. This latter maneuver is not used often, although it may facilitate examination of the prostate, which falls into the finger in this position. The rectal examination begins with inspection of the perianal area for skin lesions. A well-lubricated, gloved finger is placed on the anus, and while applying gentle pressure, the examiner asks that the patient bear down as though having a bowel movement. This maneuver facilitates entry of the finger into the rectum. A normal rectal response includes tightening of the anal sphincter around the finger. The examiner should palpate circumferentially around the length of the fully inserted finger for masses. On withdrawing the gloved finger, the finger should be wiped on a stool guaiac card for fecal blood testing to assess for acute blood loss. As a screening test for colorectal carcinoma (Chapter 193), digital examination does not replace the need for testing stool samples collected by the patient (or using alternative screening strategies, such as flexible sigmoidoscopy or colonoscopy).

SUMMARIZING THE FINDINGS FOR THE PATIENT

The physician should summarize the pertinent positive and negative findings for the patient and be willing to express uncertainty to the patient, provided that it is accompanied by a plan of action (e.g., “I will reexamine you on your next visit”). The rationale for subsequent laboratory, imaging, or other tests should be explained. A plan should be established for providing further feedback and results to the patient, especially when there is a possibility that bad news may need to be delivered. Some physicians ask the patient if there is “anything else” to be covered. Patients who express additional new concerns at the end of the visit may have been fearful to address them earlier (e.g., “by the way, doctor, I’m getting a lot of chest pain”); when the problems seem non-urgent, it is acceptable to reassure the patient and offer the promise of evaluating the patient in a follow-up phone call or at the next visit.

FUTURE DIRECTIONS

The common assumption that physicians’ diagnostic skills are deteriorating is not supported by evidence. There is considerable evidence that the scientific approach to understanding what is worthwhile and what is not worthwhile during the clinical examination identifies a core set of skills for clinical diagnosticians. Because good patient outcomes at good value are driven primarily by the quality of the information obtained during the clinical examination, continued application of scientific principles to the history and physical examination should improve diagnostic skills. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 7  Approach to the Patient: History and Physical Examination  

GENERAL REFERENCES 1. Krogsboll LT, Jorgensen KJ, Larsen CG, et al. General health checks in adults for reducing morbidity and mortality from disease: Cochrane systematic review and meta-analysis. BMJ. 2012;345:e7191. 2. Verghese A, Brady E, Kapur CC, et al. The bedside evaluation: ritual and reason. Ann Intern Med. 2011;155:550-553. 3. Powers BJ, Trinh JV, Bosworth HB. Can this patient read and understand written health information? JAMA. 2010;304:76-84. 4. Makadon HJ. Ending LGBT invisibility in health care: the first step in ensuring equitable care. Cleve Clin J Med. 2011;78:220-224. 5. Department of Veterans Affairs. Military Health History Pocket Card for Clinicians. http://www. va.gov/oaa/pocketcard/military-health-history-card-for-print.pdf. Accessed February 9, 2015. 6. Murff HJ, Spigel DR, Syngal S. Does this patient have a family history of cancer? An evidence-based analysis of the accuracy of family cancer history. JAMA. 2004;292:1480-1489. 7. Brenner S, Guder G. The patient with dyspnea. Rational diagnostic evaluation. Herz. 2014; 39:8-14.

28.e1

8. Bagai A, Thavendiranathan P, Detsky AS. Does this patient have hearing impairment? JAMA. 2006;295:416-428. 9. Hollands H, Johnson D, Brox AC, et al. Acute-onset floaters and flashes: is this patient at risk for retinal detachment? JAMA. 2009;302:2243-2249. 10. Hollands H, Johnson D, Hollands S, et al. Do findings on routine examination identify patients at risk for primary open-angle glaucoma? The rational clinical examination systematic review. JAMA. 2013;309:2035-2042. 11. Marik PE, Cavallazzi R. Does the central venous pressure predict fluid responsiveness? An updated meta-analysis and a plea for some common sense. Crit Care Med. 2013;41:1774-1781. 12. Wong CL, Holroyd-Leduc J, Simel DL, Straus SE. Does this patient have delirium? Value of bedside instruments. JAMA. 2010;304:779-786. 13. Crochet JR, Bastian LA, Chireau MV. Does this woman have an ectopic pregnancy? The rational clinical examination systematic review. JAMA. 2013;309:1722-1729.

28.e2

CHAPTER 7  Approach to the Patient: History and Physical Examination  

REVIEW QUESTIONS 1. A 24-year-old-woman with right lower quadrant abdominal pain and vagin*l bleeding has a positive home pregnancy. Before a pelvic ultrasound is ordered, the pelvic examination is performed and reveals cervical motion tenderness. Cervical motion tenderness has the following diagnostic characteristics for an ectopic pregnancy: sensitivity 45%, specificity 91%, likelihood ratio positive (LR+) 4.9, LR negative (LR−) 0.62, positive predictive value (PPV) 46%. Which of these values is most helpful for assessing the probability that she has an ectopic pregnancy? A. Sensitivity B. Specificity C. Likelihood ratio positive D. Likelihood ratio negative E. Positive predictive value Answer: C  The sensitivity is the percentage of patients who have cervical motion tenderness among women with an ectopic pregnancy. The specificity is the percentage of patients who do not have cervical motion tenderness among women without an ectopic pregnancy. These individual values are not helpful for this particular patient (Crochet JR, Bastian LA, Chireau MV. does this woman have an ectopic pregnancy? The rational clinical examination systematic review. JAMA. 2013;309:1722-1729) because without knowing whether or not she has an ectopic pregnancy, we do not know which result applies. The PPV describes the probability of ectopic pregnancy when there is cervical motion tenderness. To use the PPV for this patient requires knowing the prevalence of disease from which the PPV was derived. Without knowing that value, you cannot be certain whether the PPV is appropriate for your patient. The LR+ quantifies the increase in odds of disease among those with cervical motion tenderness, and the LR− quantifies the decrease in odds of disease when cervical motion tenderness is absent. Based on your assessment of the prior probability of ectopic pregnancy, which is 10 to 20% among all pregnant women with abdominal pain and/or vagin*l bleeding, the LR+ is the most relevant value because it can be used to determine the increase in likelihood of ectopic pregnancy (Chapter 10). 2. The primary role of completing a review of systems in helping with clinical diagnosis during the clinical evaluation is which of the following? A. To review symptoms associated with the presenting problem B. To pick up the presence of concerns that were uncomfortable to address during the history of the present illness C. To complete a record that enhances patient billing D. To focus on the presence or absence of findings during a constrained time period E. To allow open-ended questioning that enhances the patient’s relationship with the physician Answer: D  A complete medical history should reveal the most important symptoms experienced by the patient that are pertinent to the presenting problems. Open-ended questioning should occur during elicitation of the medical history. Although completing a review of systems may be required for electronic medical records and billing, simply fulfilling that function does not help with diagnosis. However, about 10% of the time, the review of systems might pick up on an important finding that was not addressed during the rest of the evaluation. The main purpose of the review of systems is to use direct questioning to pick up on a limited set of symptoms, during a specified recent time interval (e.g., over the past week, or the past month). This approach helps the patient focus on a well-defined and recent time period that should enhance the reliability of their answers that pertain to their current condition.

3. While conducting your new patient evaluation on a 35-year-old male veteran who served in combat duty in Afghanistan, you note that he seems anxious and that he is questioning you frequently about a variety of difficulties in making his appointment and the attitudes of your office staff. You recognize the need to screen for post-traumatic stress disorder using a four-item screening instrument. You start by asking, “In your life, have you ever had an experience so horrible, frightening, or upsetting that, in the past month you …” All of the items below are part of the four-item screening instrument for post-traumatic stress disorder that follows this introduction except: A. Have little interest or pleasure in doing things? B. Have had nightmares about it or thought about it when you did not want to? C. Tried hard not to think about it or went out of your way to avoid situations that reminded you of it? D. Were constantly on guard, watchful, or easily startled? E. Felt numb or detached from others, activities, or your surroundings? Answer: A  The loss of interest or pleasure in doing things is a symptom that suggests depression. The other symptom that is elicited in a two-item screener for depression asks the patient whether they feel down, depressed, or hopeless (Kroenke K, Spitzer RL, Williams JB. The Patient Health Questionnaire-2: validity of a two-item depression screener. Med Care. 2003;41:1284-1292). Patients who answer “yes” to at least one of the items have an LR+ of 2.7 for depression, whereas answering “no” to both questions makes depression much less likely with an LR of 0.14 (Williams JW Jr, Noël P, Cordes JA, et al. Is this patient clinically depressed? JAMA. 2002;287:11601170). Although it is important to screen for depression among patients with post-traumatic stress disorder (military related or not) (Campbell DG, Felker BL, Liu CF, et al. Prevalence of depression-PTSD comorbidity: implications for clinical practice guidelines and primary care-based interventions. J Gen Intern Med. 2007;22:711-718), screening for depression alone may not detect the associated post-traumatic stress disorder.

28

CHAPTER 8  Approach to the Patient with Abnormal Vital Signs  

8  APPROACH TO THE PATIENT WITH ABNORMAL VITAL SIGNS DAVID L. SCHRIGER Care of the patient is guided by integration of the chief complaint, history, vital signs, and physical examination findings (Chapter 7). Physicians should be keenly aware of a patient’s vital signs but should seldom make them the centerpiece of the evaluation.

CHAPTER 8  Approach to the Patient with Abnormal Vital Signs  

TABLE 8-1  NORMAL AND PANIC RANGES FOR KEY VITAL SIGNS IN ADULTS* NORMAL

PANIC

Temperature

36°-38° C (96.8°-100.4° F)

40° C (104° F)

Pulse

60-100 beats/min

130 beats/min

Respirations

12-20 breaths/min

26 breaths/min

Oxygen saturation

95-100%

115 mm Hg) should stimulate an evaluation for hypertensive urgencies (Chapter 67). Note that hypertension in the absence of signs of acute end-organ damage does not require acute treatment, which can reduce intracranial perfusion pressure and cause stroke. Patients with elevated blood pressure should be offered standard evaluation and treatment for chronic hypertension (Chapter 67). Markedly low pulse or blood pressure in patients receiving cardioactive medications should lead to a confirmation that the patient is truly asymptomatic, an inquiry into the dosing of these medications, and a reconsideration of the regimen. Markedly low pulse in elderly patients who are not receiving rate-controlling drugs should trigger an evaluation of the patient’s cardiac conduction system. Oxygen saturation below 93% in the absence of known pulmonary problems should prompt an evaluation of the patient’s pulmonary status.

measurement of an elevated blood pressure leading to a diagnosis of hypertension is the classic example of the value of vital signs in such patients.

Patients Who Complain of Systemic Illness but Do Not Appear to Be Very Ill

Vital signs serve two additional roles in symptomatic patients who do not appear particularly ill. First, abnormalities in vital signs provide information that may suggest or support a diagnosis. The presence of elevated temperature in a patient with productive cough, shortness of breath, and localized rales and egophony supports a diagnosis of infectious pneumonia. Vital signs may also play a role in defining therapy and triage. For example, guidelines for patients with community-acquired pneumonia (Chapter 97) formally incorporate vital signs. The second role of vital signs in the stable symptomatic patient is to provide warning that the patient is sicker than he or she appears. For example, the presence of hypotension in a well-appearing patient thought to have pyelonephritis may be an indication of sepsis or hypovolemia. For vital signs to be of use, the physician must be aware of them and must incorporate them explicitly into a thought process that considers the dangerous diagnoses associated with the abnormal vital sign. The physician then must decide whether the likelihood of each potentially dangerous diagnosis is high enough to warrant specific evaluation. Unfortunately, no quick or easy rules differentiate spurious abnormalities that can be ignored from those that should trigger additional testing or treatment. What can be said is that the well-trained physician who is aware of abnormal vital signs and is willing to contemplate a change in treatment or disposition in response to them is less likely to make mistakes. A few specific points bear mention. First, for most vital signs, “normal” is relative. Blood pressure must be interpreted in the context of the patient. For example, a blood pressure of 88/64 mm Hg may be reasonable for an otherwise healthy, young 50-kg woman but should cause concern in a 90-kg middle-aged man. Similarly, a blood pressure of 128/80 mm Hg would be fine in a 60-year-old man but worrisome in a 34-week pregnant woman. Second, because vital signs are insensitive measures of disease, normal vital signs should not dissuade the physician from pursuing potentially critical diagnoses. For example, young, well-conditioned adults may maintain normal vital signs well into the course of shock.

Use of Vital Signs in Patients Who Appear to Be Ill

For some patients, abnormal vital signs are expected on the basis of their appearance and their symptoms. For patients in extremis, care should proceed according to established guidelines such as Advanced Cardiac Life Support (Chapter 63), Advanced Trauma Life Support, and algorithms for the treatment of shock (Chapters 107 and 108). For other ill-appearing patients, two processes must occur. In one, the physician, armed with knowledge of the differential diagnosis of each abnormal vital sign and the ability to take a thorough history and to perform an appropriate physical examination, narrows the list of potential diagnoses and decides which are of sufficient probability to warrant evaluation. Simultaneously, the physician considers

the list of treatment options for all diagnoses associated with the abnormal vital sign and, before establishing a diagnosis, initiates those treatments for which the potential benefit of prompt administration exceeds potential harms. For example, antibiotics for febrile patients at risk for bacterial infection, hydrocortisone for hypotensive patients at risk for hypoadrenalism, and thiamine for hypothermic patients at risk for Wernicke encephalopathy may improve outcome and are unlikely to cause harm even if the patient does not have the suspected condition. Although early presumptive treatment can be life-saving in selected patients, it should not be abused; physicians must avoid knee-jerk responses that can cause harm.

Differential Diagnosis and Treatment Options Single Abnormal Vital Signs

Because vital signs can be abnormal in virtually any disease process, no differential diagnosis can be encyclopedic. The physician should focus initially on common diseases and diseases that require specific treatment. The thought process should begin with the chief complaint and history and then incorporate information about the vital signs and the remainder of the physical examination.

Multiple Abnormal Vital Signs

Patients who are acutely ill are likely to have several abnormal vital signs. Although certain patterns of abnormal vital signs predominate in specific conditions (e.g., hypotension, tachycardia, and hypothermia in profound sepsis), no pattern can be considered pathognomonic. The physician’s goal is to work toward a diagnosis while simultaneously providing treatments whose benefits outweigh potential harms. Fever is generally accompanied by tachycardia, with the general rule of thumb that the heart rate will increase by 10 beats per minute for every 1° C increase in temperature. The absence of tachycardia with fever is known as pulse-temperature dissociation and has been reported in typhoid fever (Chapter 308), legionnaires disease (Chapter 314), babesiosis (Chapter 353), Q fever (Chapter 327), infection with Rickettsia spp (Chapter 327), malaria (Chapter 345), leptospirosis (Chapter 323), pneumonia caused by Chlamydia spp (Chapter 318), and viral infections such as dengue fever (Chapter 382), yellow fever (Chapter 381), and other viral hemorrhagic fevers (Chapter 381), although the predictive value of this finding is unknown. Much can be learned by comparing the respiratory rate with pulse oximetry. Hyperventilation in the presence of high oxygen saturation suggests a central nervous system process or metabolic acidosis rather than a cardiopulmonary process. Low respiratory rates in the presence of low levels of oxygen saturation suggest central hypoventilation, which may respond to narcotic antagonists. Hypertension and bradycardia in the obtunded or comatose patient are known as the Cushing reflex, a relatively late sign of elevated intracranial pressure. Physicians should strive to diagnose and treat this condition before the Cushing reflex develops.

Approach to Abnormalities of Specific Vital Signs Elevated Temperature

Normal temperature is often cited as 37° C (98.6° F), but there is considerable diurnal variation and variation among individuals, so 38° C is the most commonly cited threshold for fever. Fever thought to be due to infection should be treated with antipyretics and appropriate antimicrobials (Chapter 280). The importance of early administration of antibiotics to potentially septic patients cannot be overstated (Chapters 280 and 281). Hyperthermia (temperature above 40° C) should be treated with cooling measures such as ice packs, cool misting in front of fans, cold gastric lavage, and, for medicationrelated syndromes, medications such as dantrolene (Chapter 109). Most hospital anesthesia departments will have a designated kit for the treatment of malignant hyperthermia (Chapters 432 and 434).

Low Temperature

The treatment of hypothermia is guided by its cause (Chapter 109). The body’s temperature decreases when heat loss exceeds heat production. Every logically possible mechanism for this phenomenon has been observed. Decreased heat production can result from endocrine hypofunction (e.g., Addison disease [Chapter 227], hypopituitarism [Chapter 224], hypothyroidism [Chapter 226]) and loss of the ability to shiver (e.g., drug-induced or neurologic paralysis or neuromuscular disorders). Malfunction of the hypothalamic regulatory system can be due to hypoglycemia (Chapter 229) and a variety of central nervous system disorders (Wernicke encephalopathy

CHAPTER 8  Approach to the Patient with Abnormal Vital Signs  

[Chapter 416], stroke [Chapter 407], tumor [Chapter 189], and trauma [Chapter 399]). Resetting of the temperature set point can occur with sepsis. Increased heat loss can be due to exposure, behavioral and physical disorders that prevent the patient from sensing or responding to cold, skin disorders that decrease its ability to retain heat, and vasodilators (including ethanol). A careful history and physical examination should illuminate which of these possibilities is most likely. Several considerations are worthy of emphasis. The spine of an obtunded hypothermic patient who is “found down” must be protected and evaluated because paralysis from a fall may have prevented the patient from seeking shelter and may have diminished the ability to produce heat. The physician should not forget to administer antibiotics to patients who may be septic (Chapter 108), thiamine to those who may have Wernicke encephalopathy (Chapter 416), hydrocortisone to those who may be hypoadrenal (Chapter 227), and thyroid hormone to those who may have myxedema coma (Chapter 226). Severely hypothermic patients (Chapter 109) should be treated gently because any stimulation may trigger ventricular dysrhythmias; even in the absence of pulses, cardiopulmonary resuscitation should be used only in patients with ventricular fibrillation or asystole.

Elevated Heart Rate

The rate, rhythm, and electrocardiogram differentiate sinus tachycardia from tachyarrhythmias (Chapters 62 to 65). Tachyarrhythmias can be instigated by conditions that may require specific treatment (e.g., sepsis [Chapter 108], electrolyte disorders [Chapters 116, 117, and 118], endocrine disorders [Chapter 221], and poisonings [Chapters 22 and 110]) before the arrhythmia is likely to resolve. For sinus tachycardia, treatment of the underlying cause is always paramount. Treatments may include antipyretics (for fever); anxiolytics; oral or intravenous fluids (for hypovolemia); nitrates, angiotensinconverting enzyme inhibitors, and diuretics (for heart failure and fluid overload [Chapter 59]); oxygen (for hypoxemia); α-blockers (for stimulant overdose); β-blockers (for acute coronary syndromes [Chapters 72 and 73] or thyroid storm [Chapter 226]); and anticoagulation (for pulmonary embolism [Chapter 98]). Tachycardia is often an appropriate response to a clinical condition and should not be treated routinely unless it is causing or is likely to cause secondary problems.

Low Pulse

Bradycardia can be physiologic (athletes and others with increased vagal tone), due to prescribed cardiac medications (e.g., β-blockers, calciumchannel blockers, digoxin), overdoses (e.g., cholinergics, negative chronotropic agents), disease of the cardiac conducting system, electrolyte abnormalities (severe hyperkalemia), and inferior wall myocardial infarction (Chapters 64 and 73). Asymptomatic patients do not require immediate treatment. The goal of therapy is to produce a heart rate sufficient to perfuse the tissues and alleviate the symptoms (Chapter 63). Overdoses should be treated with specific antidotes (Chapter 110). Endocrine disorders should be treated with replacement therapy. In patients with acute coronary syndrome (Chapter 72), the goal is to restore perfusion and alleviate the ischemia. Patients with profound bradycardia or hypotension may require chronotropic drugs to increase perfusion even if these agents increase myocardial oxygen demand. In normotensive patients with milder bradycardia, chronotropic agents should be used only if symptoms and ischemia cannot be resolved by other means. Atropine is the primary therapy for bradycardia; isoproterenol and cardiac pacing are reserved for those who do not respond (Chapter 63).

Elevated Blood Pressure

Elevated blood pressure does not require acute treatment in the absence of symptoms or signs of end-organ damage (Chapter 67). In patients whose blood pressure is markedly above their baseline, the history and physical examination should assess for the conditions that define “hypertensive emergency”: evidence of encephalopathy, intracranial hemorrhage, ischemic stroke, heart failure, pulmonary edema, acute coronary syndrome, aortic dissection, renal failure, and preeclampsia. In the absence of these conditions, treatment should consist of restarting or adjusting the medications of patients with known hypertension and initiating a program of blood pressure checks and appropriate evaluation for those with no prior history of hypertension (Chapter 67). The patient with a true hypertensive emergency should be treated with agents appropriate for the specific condition. Because rapid decreases in blood pressure can be as deleterious as the hypertensive state itself, intrave-

31

nous agents with short half-lives, such as nitroprusside, labetalol, nitroglycerin, and esmolol, are preferred (Chapter 67).

Low Blood Pressure

Low blood pressure must be evaluated in the context of the patient’s symptoms, general appearance, and physical examination findings. Treatment depends on context. The same blood pressure value may necessitate intravenous inotropic agents in one patient and no treatment in another. In tachycardic hypotensive patients, the physician must rapidly integrate all available evidence to determine the patient’s volume state, cardiac function, vascular capacitance, and primary etiology (Chapter 106). Not all patients with hypotension and tachycardia are in shock, and not all patients in shock will have hypotension and tachycardia. Patients in shock should be treated on the basis of the cause (Chapters 106 to 108). Symptomatic hypotensive patients thought to be intravascularly volume depleted should receive intravenous fluid resuscitation with crystalloid or blood, depending on their hemoglobin level (Chapter 106). In patients with known heart disease, patients who are frail or elderly, and patients whose volume status is uncertain, small boluses of fluid (e.g., 250 mL of normal saline), each followed by reassessment, are preferred so that iatrogenic heart failure may be avoided. Inotropic support should be reserved for patients who do not respond to fluid resuscitation. High-output heart failure should be kept in mind in patients with possible thyroid storm or stimulant overdose.

Increased Respiratory Rate

Tachypnea is a normal response to hypoxemia (see later). Treatment of tachypnea in the absence of hypoxemia is directed at the underlying cause, which often is pain (Chapter 30). Anxiolytics (e.g., diazepam, 5 to 10 mg PO or IV; lorazepam, 1 to 2 mg PO, IM, or IV) or reassurance can calm patients with behavioral causes of hyperventilation. Breathing into a paper bag has been shown to be an ineffective treatment. Pulmonary embolism (Chapter 98) does not necessarily reduce the oxygen saturation or cause a low Po2 and should always be considered in at-risk patients with unexplained tachypnea.

Decreased Respiratory Rate

Any perturbation of the respiratory center in the central nervous system can slow the respiratory drive (Chapter 86). Narcotics and other sedatives and neurologic conditions are common causes of a decreased respiratory rate. The primary treatment of apnea is mechanical ventilation (Chapter 105), but narcotic antagonists can be tried in patients with a history or physical examination findings (miosis, track marks, opiate patch) suggestive of narcotic use or abuse (Chapter 34). In nonapneic patients, mechanical ventilation is indicated for patients who are breathing too slowly to maintain an acceptable oxygen saturation and for patients who are retaining carbon dioxide in quantities sufficient to depress mental function. Patients who are unable to protect their airway should be intubated. Oxygen should be administered to all hypopneic patients who are hypoxemic (see earlier). Patients with chronic hypoventilation (Chapter 86) may have retained HCO3– to compensate for an elevated Pco2 and so may depend on hypoxia to maintain respiratory drive; in these patients, overaggressive administration of oxygen can decrease the respiratory rate, increase the Pco2, and increase obtundation (Chapter 104).

Decreased Oxygen Saturation

In hypopneic patients, initial efforts should try to increase the respiratory rate (see earlier) and tidal volume. Regardless of etiology, oxygen, in amounts adequate to restore adequate oxygen saturation (Po2 > 60 mm Hg, oxygen saturation >90%), is the mainstay of therapy. When oxygen alone fails, noninvasive methods for improving ventilation or tracheal intubation are required (Chapter 104). Oxygen should increase the Po2 in all patients except those who have severe right-to-left shunting (Chapter 69). Treatment of conditions that cause hypoxemia includes antibiotics (pneumonia), bronchodilators (asthma, chronic obstructive pulmonary disease), diuretics and vasodilators (pulmonary edema), anticoagulants (pulmonary embolism), hyperbaric oxygen (carbon monoxide poisoning), methylene blue (methemoglobinemia, sulfhemoglobinemia), and transfusion (anemia). GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 8  Approach to the Patient with Abnormal Vital Signs  

GENERAL REFERENCES 1. Straede M, Brabrand M. External validation of the simple clinical score and the HOTEL score, two scores for predicting short-term mortality after admission to an acute medical unit. PLoS ONE. 2014;9:e105695. 2. Lamantia MA, Stewart PW, Platts-Mills TF, et al. Predictive value of initial triage vital signs for critically ill older adults. West J Emerg Med. 2013;14:453-460.

31.e1

3. Lighthall GK, Markar S, Hsiung R. Abnormal vital signs are associated with an increased risk for critical events in US veteran inpatients. Resuscitation. 2009;80:1264-1269. 4. Gabayan GZ, Sun BC, Asch SM, et al. Qualitative factors in patients who die shortly after emergency department discharge. Acad Emerg Med. 2013;20:778-785. 5. Bleyer AJ, Vidya S, Russell GB, et al. Longitudinal analysis of one million vital signs in patients in an academic medical center. Resuscitation. 2011;82:1387-1392.

31.e2

CHAPTER 8  Approach to the Patient with Abnormal Vital Signs  

REVIEW QUESTIONS 1. A patient presents with malaise, cough, and shortness of breath. Vital signs include temperature 40° C, blood pressure 120/74 mm Hg, respiratory rate 18 breaths per minute, pulse 70 beats per minute, and oxygen saturation 97%. This presentation could be consistent with: A. Streptococcal pneumonia B. Pyelonephritis due to Escherichia coli C. Legionella pneumonia D. Influenza-like illness E. Mycoplasma pneumonia Answer: C  This patient is exhibiting a pulse-temperature dissociation because the pulse (70) is far lower than one would expect given that the patient is febrile to 40° C. This phenomenon is seen in a number of conditions, including typhoid fever and legionella infection. The other conditions would all be expected to produce tachycardia unless the patient could not become tachycardic because of medications (e.g., β-blockers) or cardiac conduction problems. 2. An 88-year-old man presents from a nursing home with slight agitation and vital signs that include temperature 38.7° C, blood pressure 96/64 mm Hg, respiratory rate 22 breaths per minute, pulse 94 beats per minute, and oxygen saturation 96%. Physical examination reveals dry mucous membranes, clear lungs, a soft abdomen, an indwelling Foley catheter, and slightly cool but noncyanotic extremities. The patient should be given: A. Antipyretics (e.g., acetaminophen) B. Intravenous normal saline, 500 mL with additional boluses as tolerated C. Intravenous antibiotics D. All of the above E. Only A and B until urine culture results are available Answer: D  This case is an example of how vital signs can guide treatment in the absence of a firm diagnosis. The patient meets all three of the physical examination criteria for the systemic inflammatory response syndrome (SIRS) and is likely septic. The physician should not wait for his white blood cell count or other laboratory results to initiate antibiotic treatment because evidence suggests that early antibiotics are a crucial step in preventing morbidity and mortality. Although antibiotics should not be overused, the early provision of appropriate broad-spectrum antibiotics before the confirmation of a specific diagnosis is prudent and may be life-saving for this patient. 3. An intern is awakened at 3 am by the ward nurse regarding a patient who is postoperative day 2 from a hip replacement and is newly tachycardic. Vital signs include temperature 36° C, blood pressure 146/82 mm Hg, respiratory rate 18 breaths per minute, pulse 112 beats per minute, and oxygen saturation 97% on room air. The intern drowsily orders a 1000-mL normal saline fluid challenge for dehydration. Later that morning, the patient is acutely intubated for respiratory distress. What most likely went wrong? A. The intern failed to consider pulmonary embolism as a possible cause for the tachycardia. B. The intern failed to consider fat embolism in a patient who had recently undergone hip surgery. C. The intern failed to consider sepsis in the differential diagnosis. D. The intern failed to consider failure of the patient controlled anesthesia (PCA) pump in the differential diagnosis. E. The intern failed to realize that tachycardia can be present in both dehydration and heart failure. Answer: E  First and foremost, the intern’s main mistake was not getting out of bed to evaluate the patient in person. Vital signs alone are not sufficient data on which to base an important clinical decision. Although choices A and B are certainly possible in a postoperative orthopedic patient, heart failure is a more likely diagnosis. There is little clinical support for the other choices.

4. A patient arrives in the emergency department comatose with decreased respiratory rate in the winter. Vital signs are temperature 36° C, blood pressure 128/68 mm Hg, respiratory rate 10 breaths per minute, pulse 100 beats per minute, and oxygen saturation 100% on room air. Pupils are 6 mm and reactive, and lungs are clear. What is the single most important initial treatment? A. High-flow O2 administered by non-rebreather mask B. Intravenous normal saline, 1000 mL with additional boluses as tolerated C. Intravenous antibiotics D. Naloxone, 0.8 mg IV E. Immediate endotracheal intubation Answer: A  This patient may have carbon monoxide poisoning. It is winter, a time when people use heating devices that may have incomplete combustion. The pupillary examination is not suggestive of opiate intoxication (D), and no other diagnosis is apparent. Because oxygen is the best treatment for this condition and is generally harmless in adults, it makes sense to initiate this therapy while efforts (e.g., blood gas analysis with co-oximetry) are made to confirm the diagnosis. There is no basis for thinking this patient is dehydrated (B) or infected (C), and intubation would be premature (E). Remember that pulse oximetry is falsely elevated in carbon monoxide poisoning, so the 100% oxygen saturation means nothing.

32

CHAPTER 9  Statistical Interpretation of Data  

9  STATISTICAL INTERPRETATION OF DATA THOMAS B. NEWMAN AND CHARLES E. MCCULLOCH

ROLE AND LIMITATIONS OF STATISTICS

Much of medicine is inherently probabilistic. Not everyone with hypercholesterolemia who is treated with a statin is prevented from having a myocardial infarction, and not everyone not treated does have one, but statins reduce the probability of a myocardial infarction in such patients. Because so much of medicine is based on probabilities, studies must be performed on groups of people to estimate these probabilities. Three component tasks of statistics are: selecting a sample of subjects for study, describing the data from that sample, and drawing inferences from that sample to a larger population of interest.1

SAMPLING: SELECTING SUBJECTS FOR A STUDY

The goal of research is to produce generalizable knowledge, so that measurements made by researchers on samples of individuals will eventually help draw inferences to a larger group of people than was studied. The ability to draw such inferences depends on how the subjects for the study (the sample) were selected. To understand the process of selection, it is helpful to begin by identifying the group to which the results are to be generalized and then work backward to the sample of subjects to be studied.

Target Population

The target population is the population to which it is hoped the results of the study will be generalizable. For example, to study the efficacy of a new drug to treat obesity, the target population might be all people with a certain level of obesity (e.g., body mass index [BMI] of ≥30 kg/m2) who might be candidates for the drug.

Sampling

The intended sample is the group of people who are eligible to be in the study based on meeting inclusion criteria, which specify the demographic, clinical, and temporal characteristics of the intended subjects, and not meeting exclusion criteria, which specify the characteristics of subjects whom the investigator does not wish to study. For example, for the study of a new obesity drug, the intended sample (inclusion criteria) might be men and women 18 years or older who live in one of four metropolitan areas, who have a BMI of 30 kg/m2 or higher, and who have failed an attempt at weight loss with a standard diet. Exclusion criteria might include an inability to speak English or Spanish, known alcohol abuse, plans to leave the area in the next 6 months, and being pregnant or planning to become pregnant in the next 6 months. In some cases, particularly large population health surveys such as the National Health and Nutrition Examination Survey (NHANES), the intended sample is a random sample of the target population. A simple random sample is a sample in which every member of the target population has an equal chance of being selected. Simple random samples are the easiest to handle statistically but are often impractical. For example, if the target population is the entire population of the United States (as is the case for NHANES), a simple random sample would include subjects from all over the country. Getting subjects from thousands of distinct geographic areas to examination sites would be logistically difficult. An alternative, used in NHANES, is cluster sampling, in which investigators take a random sample of “clusters” (e.g., specific census tracts or geographic areas) and then try to study all or a sample of the subjects in each cluster. Knowledge of the cluster sampling process must then be used during analysis of the study (see later) to draw inferences correctly back to the target population. Regardless of the method used to select the intended sample, the actual sample will almost always differ in important ways because not all intended subjects will be willing to enroll in the study and not all who begin a study will complete it. In a study on treatment of obesity, for example, those who consent to be in the study probably differ in important, but difficult-toquantify ways from those who do not (and may be more likely to do well with treatment). Furthermore, subjects who respond poorly to treatment

may drop out, thus making the group that completes the study even less representative. Statistical methods address only some of the issues involved in making inferences from a sample to a target population. Specifically, most statistical methods address only the effect of random variation on the inference from the intended sample to the target population. Estimating the effects of differences between the intended sample and the actual sample depends on the quantities being estimated and content knowledge about whether factors associated with being in the actual sample are related to those quantities. One rule of thumb about generalizability is that associations between variables are more often generalizable than measurements of single variables. For instance, subjects who consent to be in a study of obesity may be more motivated than average, but this motivation would be expected to have less effect on the difference in weight loss between groups than on the average weight loss in either group.

DESCRIBING THE SAMPLE

Types of Variables

A key use of statistics is to describe sample data. Methods of description depend on the type of variable (E-Table 9-1). Numerical variables include continuous variables (those that have a wide range of possible values), count variables (e.g., the number of times a woman has been pregnant), and timeto-event variables (e.g., the time from initial treatment to recurrence of breast cancer). Whereas numerical variables describe the data with numbers, categorical variables consist of named characteristics. Categorical variables can be further divided into dichotomous variables, which can take on only two possible values (e.g., alive/dead); nominal variables, which can take on more than two values but have no intrinsic ordering (e.g., race); and ordinal variables, which have more than two values and an intrinsic ordering of the values (e.g., tumor stage). Numerical variables are also ordinal by nature and can be made binary by breaking the values into two disjointed categories (e.g., systolic blood pressure >140 mm Hg or not), and thus sometimes methods designed for ordinal or binary data are used with numerical variable types, often for ease of interpretation.

Univariate Statistics for Numerical Variables: Mean, Standard Deviation, Median, and Percentiles

When describing data in a sample, it is a good idea to begin with univariate (one variable at a time) statistics. For numerical variables, univariate statistics typically measure central tendency and variability. The most common measures of central tendency are the mean (or average, i.e., the sum of the observations divided by the number of observations) and the median (the 50th percentile, i.e., the value that has equal numbers of observations above and below it). One of the most commonly used measures of variability is the standard deviation (SD). The SD is defined as the square root of the variance, which is calculated by subtracting each value in the sample from the mean, squaring that difference, totaling all of the squared differences, and dividing by the number of observations minus 1. Although this definition is far from intuitive, the SD has some useful mathematical properties, namely, that if the distribution of the variable is the familiar bell-shaped, normal, or Gaussian distribution, about 68% of the observations will be within 1 SD of the mean, about 95% within 2 SD, and about 99.7% within 3 SD. Even when the distribution is not normal, these rules are often approximately true. For variables that are not normally distributed, including most count and time-to-event variables, the mean and SD are not as useful for summarizing the data. In that case, the median may be a better measure of central tendency because it is not influenced by observations far below or far above the center. Similarly, the range and pairs of percentiles, such as the 25th and 75th percentiles or the 15th and 85th percentiles, will provide a better description of the spread of the data than the SD will. The 15th and 85th percentiles are particularly attractive because they correspond, in the Gaussian distribution, to about −1 and +1 SD from the mean, thus making reporting of the 50th, 15th, and 85th percentiles roughly equivalent to reporting the mean and SD.

Univariate Statistics for Categorical Variables: Proportions, Rates, and Ratios

For categorical variables, the main univariate statistic is the proportion of subjects with each value of the variable. For dichotomous variables, only one proportion is needed (e.g., the proportion female); for nominal variables and ordinal variables with few categories, the proportion in each group can be provided. Ordinal variables with many categories can be summarized by

CHAPTER 9  Statistical Interpretation of Data  

E-TABLE 9-1  TYPES OF VARIABLES AND COMMONLY USED STATISTICAL METHODS ASSOCIATED STATISTICAL METHODS TYPE OF OUTCOME VARIABLE

EXAMPLES

BIVARIATE

MULTIVARIATE

Categorical (dichotomous)

Alive; readmission to 2 × 2 table, the hospital chi-square within 30 days analysis

Logistic regression

Categorical (nominal)

Race; cancer, tumor type

Nominal logistic regression

Categorical (ordinal)

Glasgow Coma Scale Mann-WhitneyWilcoxon, Kruskal-Wallis

Numerical (continuous)

Cholesterol; SF-36 scales*

Chi-square analysis

t Test, analysis of variance

Ordinal logistic regression Linear regression

Numerical (count) Number of times Mann-Whitneypregnant; number Wilcoxon, of mental health Kruskal-Wallis visits in a year

Poisson regression, linear models

Time to event regression

Cox proportional hazards

Time to breast cancer; time to viral rebound in HIV-positive subjects

Log rank

*Numerical scores with many values are often treated as though they were continuous. HIV = human immunodeficiency virus; SF-36 = short-form 36-item health survey.

32.e1

CHAPTER 9  Statistical Interpretation of Data  

33

using proportions or by using medians and percentiles, as with continuous data that are not normally distributed. It is worth distinguishing among proportions, rates, and ratios because these terms are often confused. Proportions are unitless, always between 0 and 1 inclusive, and express what fraction of the subjects have or develop a particular characteristic or outcome. Strictly speaking, rates have units of inverse time; they express the proportion of subjects in whom a particular characteristic or outcome develops over a specific time period. The term is frequently misused, however. For example, the term false-positive rate is widely used for the proportion of subjects without a disease who test positive, even though it is a proportion, not a rate. Ratios are the quotients of two numbers; they can range between zero and infinity. For example, the male-to-female ratio of people with a disease might be 3 : 1. As a rule, if a ratio can be expressed as a proportion instead (e.g., 75% male), it is more concise and easier to understand.

conventionally the risk is for something bad, and the risk in the group of interest is subtracted from the risk in a comparison group, so the ARR will be positive for effective interventions. In this case, the ARR = 0.06% per year, or 6 in 10,000 per year.

Incidence and Prevalence

Two terms commonly used (and misused) in medicine and public health are incidence and prevalence. Incidence describes the number of subjects who contract a disease over time divided by the population at risk. Incidence is usually expressed as a rate (e.g., 7 per 1000 per year), but it may sometimes be a proportion if the time variable is otherwise understood or clear, as in the lifetime incidence of breast cancer or the incidence of diabetes during pregnancy. Prevalence describes the number of subjects who have a disease at one point in time divided by the population at risk; it is always a proportion. At any point in time, the prevalence of disease depends on how many people contract it and how long it lasts: prevalence = incidence × duration.

When the treatment increases the risk for a bad outcome, the difference in risk between treated and untreated patients should still be calculated, but it is usually just called the risk difference rather than an ARR (because the “reduction” would be negative). In that case, the NNT is sometimes called the number needed to harm. This term is a bit of a misnomer. The reciprocal of the risk difference is still a number needed to treat; it is just a number needed to treat per person harmed rather than a number needed to treat per person who benefits. In the WHI, treatment with estrogens was estimated to cause about 12 additional strokes per 10,000 women per year, so the number needed to be treated for 1 year to cause a stroke was about 10,000/12, or 833.

Bivariate Statistics

Odds Ratio

Bivariate statistics summarize the relationship between two variables. In clinical research, it is often desirable to distinguish between predictor and outcome variables. Predictor variables include treatments received, demographic variables, and test results that are thought possibly to predict or cause the outcome variable, which is the disease or (generally bad) event or outcome that the test should predict or treatment prevent. For example, to see whether a bone mineral density measurement (the predictor) predicts time to vertebral fracture (the outcome), the choice of bivariate statistic to assess the association of outcome with predictor depends on the types of predictor and outcome variables being compared.

Dichotomous Predictor and Outcome Variables

A common and straightforward case is when both predictor and outcome variables are dichotomous, and the results can thus be summarized in a 2 × 2 table. Bivariate statistics are also called measures of association (E-Table 9-2).

Relative Risk

The relative risk or risk ratio (RR) is the ratio of the proportion of subjects in one group in whom the outcome develops divided by the proportion in the other group in whom it develops. It is a general (but not universal) convention to have the outcome be something bad and to have the numerator be the risk for those who have a particular factor or were exposed to an intervention. When this convention is followed, an RR greater than 1 means that exposure to the factor was (on average) bad for the study subjects (with respect to the outcome being studied), whereas an RR less than 1 means that it was good. That is, risk factors that cause diseases will have RR values greater than 1, and effective treatments will have an RR less than 1. For example, in the Women’s Health Initiative (WHI) randomized trial, conjugated equine estrogen use was associated with an increased risk for stroke (RR = 1.37) and decreased risk for hip fracture (RR = 0.61).

Relative Risk Reduction

The relative risk reduction (RRR) is 1 − RR. The RRR is generally used only for effective interventions, that is, interventions in which the RR is less than 1, so the RRR is generally greater than 0. In the aforementioned WHI example, estrogen had an RR of 0.61 for hip fracture, so the RRR would be 1 − 0.61 = 0.39, or 39%. The RRR is commonly expressed as a percentage and used only when it is positive.

Absolute Risk Reduction

The risk difference or absolute risk reduction (ARR) is the difference in risk between the groups, defined as earlier. In the WHI, the risk for hip fracture was 0.11% per year with estrogen and 0.17% per year with placebo. Again,

Number Needed to Treat

The number needed to treat (NNT) is 1/ARR. To see why this is the case, consider the WHI placebo group and imagine treating 10,000 patients for a year. All but 17 would not have had a hip fracture anyway because the fracture rate in the placebo group was 0.17% per year, and 11 subjects would sustain a fracture despite treatment because the fracture rate in the estrogen group was 0.11% per year. Thus, with treatment of 10,000 patients for a year, 17 − 11 = 6 fractures prevented, or 1 fracture prevented for each 1667 patients treated for 1 year. This calculation is equivalent to 1/0.06% per year.

Risk Difference

Another commonly used measure of association is the odds ratio (OR). The OR is the ratio of the odds of the outcome in the two groups, where the definition of the odds of an outcome is p/(1 − p), with p being the probability of the outcome. From this definition it is apparent that when p is very small, 1 − p will be close to 1, so p/(1 − p) will be close to p, and the OR will closely approximate the RR. In the WHI, the ORs for stroke (1.37) and fracture (0.61) were virtually identical to the RRs because both stroke and fracture were rare. When p is not small, however, the odds and probability will be quite different, and ORs and RRs will not be interchangeable.

Absolute versus Relative Measures

RRRs are usually more generalizable than ARRs. For example, the use of statin drugs is associated with about a 30% decrease in coronary events in a wide variety of patient populations (Chapter 206). The ARR, however, will usually vary with the baseline risk, that is, the risk for a coronary event in the absence of treatment. For high-risk men who have already had a myocardial infarction, the baseline 5-year risk might be 20%, which could be reduced to 14% with treatment, an ARR of 6%, and an NNT of about 17 for approximately 5 years. Conversely, for a 45-year-old woman with a high low-density lipoprotein cholesterol level but no history of heart disease, in whom the 5-year risk might be closer to 1%, the same RRR would give a 0.7% risk with treatment, a risk difference of 0.3%, and an NNT of 333 for 5 years. The choice of absolute versus relative measures of association depends on the intended use of the measure. As noted earlier, RRs are more useful as summary measures of effect because they are more often generalizable across a wide variety of populations. RRs are also more helpful for understanding causality. However, absolute risks are more important for questions about clinical decision making because they relate directly to the tradeoffs between risks and benefits—specifically, the NNT, as well as the costs and side effects that need to be balanced against potential benefits. RRRs are often used in advertising because they are generally more impressive than ARRs. Unfortunately, the distinction between relative and absolute risks may not be appreciated by clinicians, thereby leading to higher estimates of the potential benefits of treatments when RRs or RRRs are used.

Risk Ratios versus Odds Ratios

The choice between RRs and ORs is easier: RRs are preferred because they are easier to understand. Because ORs that are not equal to 1 are always farther from 1 than the corresponding RR, they may falsely inflate the perceived importance of a factor. ORs are, however, typically used in two circ*mstances. First, in case-control studies (Chapter 11), in which subjects with and without the disease are sampled separately, the RR cannot be

CHAPTER 9  Statistical Interpretation of Data  

E-TABLE 9-2  COMMONLY USED MEASURES OF ASSOCIATION FOR DICHOTOMOUS PREDICTOR AND OUTCOME VARIABLES* OUTCOME PREDICTOR

Yes

No

Total

a

No

c

d

c+d

Total

a+c

b+d

N

Risk ratio or relative risk (RR)

a c ÷ (a + b) (c + d )

Relative risk reduction (RRR)

1 – RR

Risk difference or absolute risk reduction (ARR)

a c − (a + b) (c + d)

Number needed to treat (NNT) Odds ratio (OR)

b

a+b

Yes

1 ARR ad bc

*The numbers of subjects in each of the cells are represented by a, b, c, and d. Case-control studies allow calculation of only the odds ratio.

33.e1

34

CHAPTER 9  Statistical Interpretation of Data  

calculated directly. This situation does not usually cause a problem, however, because case-control studies are generally performed to assess rare outcomes, for which the OR will closely approximate the RR. Second, in observational studies that use a type of multivariate analysis called logistic regression (see later), use of the OR is convenient because it is the parameter that is modeled in the analysis.

Dichotomous Predictor Variable, Continuous Outcome Variable

Many outcome variables are naturally continuous rather than dichotomous. For example, in a study of a new treatment of obesity, the outcome might be change in weight or BMI. For a new diuretic, the outcome might be change in blood pressure. For a palliative treatment, the outcome might be a qualityof-life score calculated from a multi-item questionnaire. Because of the many possible values for the score, it may be analyzed as a continuous variable. In these cases, dichotomizing the outcome leads to loss of information. Instead, the mean difference between the two groups is an appropriate measure of the effect size. When the outcome is itself a difference (e.g., change in blood pressure over time), the effect is measured by the difference in the withingroup differences between the groups. Most measurements have units (e.g., kg, mm Hg), so differences between groups will have the same units and be meaningless without them. If the units of measurement are familiar (e.g., kg or mm Hg), the difference between groups will be meaningful without further manipulation. For measurements in unfamiliar units, such as a score on a new quality-of-life instrument, some benchmark is useful to help judge whether the difference between groups is large or small. In that case, authors typically express the difference in relation to the spread of values in the study by calculating the standardized mean difference (SMD), which is the difference between the two means divided by the SD of the measurement. It is thus expressed as the number of SDs by which the two groups are apart. To help provide a rough feel for this difference, a 1-SD difference between means (SMD = 1) would be a 15-point difference in IQ scores, a 600-g difference in birthweight, or a 40-mg/dL difference in total cholesterol levels.

Continuous Predictor Variable

When predictor variables are continuous, the investigator can either group the values into two or more categories and calculate mean differences or SMDs between the groups as discussed earlier or use a model to summarize the degree to which changes in the predictor variable are associated with changes in the outcome variable. Use of a model may more compactly describe the effects of interest but involves assumptions about the way the predictor and outcome variables are related. Perhaps the simplest model is to assume a linear relationship between the outcome and predictor. For example, one could assume that the relationship between systolic blood pressure (mm Hg) and salt intake (g/day) was linear over the range studied: SBPi = a + (b × SALTi ) + ε i where SBPi is the systolic blood pressure for study subject i, SALTi is that subject’s salt intake, and εi is an error term that the model specifies must average out to zero across all of the subjects in the study. In this model, a is a constant, the intercept, and the strength of the relationship between the outcome and predictor can be summarized by the slope b, which has units equal to the units of SBP divided by the units of SALT, or mm Hg per gram of salt per day in this case. Note that without the units, such a number is meaningless. For example, if salt intake were measured in grams per week instead of grams per day, the slope would only be one seventh as large. Thus, when reading an article in which the association between two variables is summarized, it is critical to note the units of the variables. As discussed earlier, when units are unfamiliar, they are sometimes standardized by dividing by the SDs of one or both variables. It is important to keep in mind that use of a model to summarize a relationship between two variables may not be appropriate if the model does not fit. In the preceding example, the assumption is that salt intake and blood pressure have a linear relationship, with the slope equal to b mm Hg/g salt per day. The value of b is about 1 mm Hg/g salt per day for hypertensive patients. If the range of salt intake of interest is from 1 to 10 g/day, the model predicts that blood pressure will increase 1 mm Hg as a result of a 1-g/day increase in salt intake whether that increase is from 1 to 2 g/day or from 9 to 10 g/day. If the effect of a 1-g/day change in salt intake differed in subjects ingesting low- and high-salt diets, the model would not fit, and misleading conclusions could result.

When the outcome variable is dichotomous, the relationship between the probability of the outcome and a continuous predictor variable is often modeled with a logistic model: 1 Pr{Yi = 1} = 1 + e −(a+bxi ) where the outcome Yi is coded 0 or 1 for study subject i, and xi is that subject’s value of the predictor variable. Once again, a is a constant, in this case related to the probability of the disease when the predictor is equal to zero, and b summarizes the strength of the association; in this case, it is the natural logarithm of the OR rather than the slope. The OR is the OR per unit change in the predictor variable. For example, in a study of lung cancer, an OR of 1.06 for pack years of smoking would indicate that the odds of lung cancer increase by 6% for each pack year increase in smoking. Because the outcome variable is dichotomous, it has no units, and “standardizing” it by dividing by its SD is unnecessary and counterproductive. On the other hand, continuous predictor variables do have units, and the OR for the logistic model will be per unit change in the predictor variable or, if standardized, per SD change in the predictor variable. Re-expressing predictors in standardized or at least more sensible units is often necessary. For example, suppose 10-year mortality risk decreases by 20% (i.e., RR = 0.8) for each increase in gross income of $10,000. The RR associated with an increase in gross income of $1 (which is what a computer program would report if the predictor were entered in dollars) would be 0.99998, apparently no effect at all because a change of $1 in gross income is negligible and associated with a negligible change in risk. To derive the coefficient associated with a $1 change, the coefficient for a $10,000 change is raised to the 110 ,000 power: 0.8(1/10,000) = 0.99998.

Multivariable Statistics

In many cases, researchers are interested in the effects of multiple predictor variables on an outcome. Particularly in observational studies, in which investigators cannot assign values of a predictor variable experimentally, it will be of interest to estimate the effects of a predictor variable of interest independent of the effects of other variables. For example, in studying whether breastfeeding reduces the mother’s risk for subsequent breast cancer, investigators would try to take differences in age, race, family history, and parity into account. Trying to stratify by all these variables would require a massive data set. Instead, models are used because they enable the information about individual predictors to be summarized by using the full data set. In this way, the estimated coefficients from the model are powerful descriptive statistics that allow a sense of the data in situations in which simpler methods fail. These models are similar to those described earlier but include terms for the additional variables.

Multiple Linear Regression

The multiple linear regression model for an outcome variable Y as function or predictor variables x1, x2, and so forth is as follows: Yi = a + (b1 × x 1i ) + (b2 × x 2 i ) + …+ (bk × x ki ) + ε i , where the subscripts 1, 2, …, k are for the first, second, … kth variables of the model, and the i subscripts are for each individual. As before, the relationships between each of these predictor variables and the outcome variable are summarized by coefficients, or slopes, which have units of Y divided by the units of the associated predictor. In addition, the linear combination of predictor variables adds a major simplifying constraint (and assumption) to the model: it specifies that the effects of each variable on the outcome variable are the same regardless of the values of other variables in the model. Thus, for example, if x1 is the variable for salt intake and x2 is a variable for sex (e.g., 0 for females and 1 for males), this model assumes that the average effect of a 1-g increase in daily salt intake on blood pressure is the same in men and women. If such is not believed to be the case, either based on previous information or from examining the data, the model should include interaction terms, or separate models should be used for men and women.

Multiple Logistic Regression

The logistic model expands to include multiple variables in much the same way as the linear model: 1 Pr{Yi = 1} = − ( a+b1i +b2 x2 i +…+bk xki ) 1+ e Again, the additional assumption when more than one predictor is included in the model is that in the absence of included interaction terms,

CHAPTER 9  Statistical Interpretation of Data  

the effect of each variable on the odds of the outcome is the same regardless of the values of other variables in the model. Because the logistic model is multiplicative, however, the effects of different predictors on the odds of the outcome are multiplied, not added. Thus, for example, if male sex is associated with a doubling of the odds for heart disease, this doubling will occur in both smokers and nonsmokers; if smoking triples the odds, this tripling will be true in both men and women, so smoking men would be predicted to have 2 × 3 = 6 times higher odds of heart disease than nonsmoking women.

Recursive Partitioning

Recursive partitioning, or “classification and regression trees,” is a prediction method often used with dichotomous outcomes that avoids the assumptions of linearity. This technique creates prediction rules by repeatedly dividing the sample into subgroups, with each subdivision being formed by further separating the sample on the value of one of the predictor variables. The optimal choice of variables and cut points may depend on the relative costs of falsepositive and false-negative predictions, as set by the investigator. The end result is a set of branching questions that forms a treelike structure in which each final branch provides a yes/no prediction of the outcome. The methods of fitting the tree to data (e.g., cross-validation) help reduce overfitting (inclusion of unnecessary predictor variables), especially in cases with many potential predictors.

Proportional Hazards (Cox) Model

A multivariate model often used in studies in which subjects are monitored over time for development of the outcome is the Cox or proportional hazards model. Like the logistic model, the Cox model is used for continuous or dichotomous predictor variables, but in this case with a time-to-event outcome (e.g., time to a stroke). This approach models the rate at which the outcome occurs over time by taking into account the number of people still at risk at any given time. The coefficients in the Cox model are logarithms of hazard ratios rather than ORs, interpretable (when exponentiated) as the effect of a unit change in predictors on the hazard (risk in the next short time period) of the outcome developing. Like the logistic model, the Cox model is multiplicative; that is, it assumes that changes in risk factors multiply the hazard by a fixed amount regardless of the levels of other risk factors. A key feature of the Cox model and other survival analysis techniques is that they accommodate censored data (when the time to event is known only to exceed a certain value). For example, if the outcome is time to stroke, the study will end with many subjects who have not had a stroke, so their time to stroke is known only to exceed the time to their last follow-up visit.

INFERRING POPULATION VALUES FROM A SAMPLE

The next step after describing the data is drawing inferences from a sample to the population from which the sample was drawn. Statistics mainly quantify random error, which arises by chance because even a sample randomly selected from a population may not be exactly like the population from which it was drawn. Samples that were not randomly selected from populations may be unrepresentative because of bias, and statistics cannot help with this type of systematic (nonrandom) error.

Inferences from Sample Means: Standard Deviation versus Standard Error

The simplest case of inference from a sample to a population involves estimating a population mean from a sample mean. Intuitively, the larger the sample size, N, the more likely it will be that the sample mean will be close to the population mean, that is, close to the mean that would be calculated if every member of the population were studied. The more variability there is in the population (and hence the sample), the less accurate the sample estimate of the population mean is likely to be. Thus, the precision with which a population mean can be estimated is related to both the size of the sample and the SD of the sample. To make inferences about a population mean from a sample mean, the standard error of the mean (SEM), which takes both of these factors into account, is as follows: SD SEM = N To understand the meaning of the SEM, imagine that instead of taking a single sample of N subjects from the population, many such samples were taken. The mean of each sample could be calculated, as could the mean of those sample means and the SD of these means. The SEM is the best estimate from a single sample of what that SD of sample means would be.

35

Confidence Intervals

The SEM expresses variability of sample means in the same way that the SD expresses variability of individual observations. Just as about 95% of observations in a population are expected to be within ±1.96 SD of the mean, 95% of sample means are expected to be within 1.96 SEM of the population mean, thereby providing the 95% confidence interval (CI), which is the range of values for the population mean consistent with what was observed from the sample. CIs can similarly be calculated for other quantities estimated from samples, including proportions, ORs, RRs, regression coefficients, and hazard ratios. In each case, they provide a range of values for the parameter in the target population consistent with what was observed in the study sample.

Significance Testing and P Values

Many papers in the medical literature include P values, but the meaning of P values is widely misunderstood and mistaught. P values start with calculation of a test statistic from the sample that has a known distribution under certain assumptions, most commonly the null hypothesis, which states that there is no association between variables. P values provide the answer to the question, “If the null hypothesis were true, what would be the probability of obtaining, by chance alone, a value of the test statistic this large or larger (suggesting an association between groups of this strength or stronger)?” When the P value is small, there are two possible explanations. First, something with a small possibility of occurring actually happened; or second, the null hypothesis is false, and there is a true association. Values of P less than 0.05 are customarily described as “statistically significant.” There are a number of common pitfalls in interpreting P values. The first is that because P values less than .05 are customarily described as being “statistically significant,” the description of results with P values less than .05 sometimes gets shortened to “significant” when in fact the results may not be clinically significant (i.e., important) at all. A lack of congruence between clinical and statistical significance most commonly arises when studies have a large sample size and the measurement is of a continuous or frequently occurring outcome. A second pitfall is concluding that no association exists simply because the P value is greater than .05. However, it is possible that a real association exists, but that it simply was not found in the study. This problem is particularly likely if the sample size is small because small studies have low power, defined as the probability of obtaining statistically significant results if there really is a given magnitude of difference between groups in the population. One approach to interpreting a study with a nonsignificant P value is to examine the power that the study had to find a difference. A better approach is to look at the 95% CI. If the 95% CI excludes all clinically significant levels of the strength of an association, the study probably had an adequate sample size to find an association if there had been one. If not, a clinically significant effect may have been missed. In “negative” studies, the use of CIs is more helpful than power analyses because CIs incorporate information from the study’s results. Finally, a common misconception about P values is that they indicate the probability that the null hypothesis is true (e.g., that there is no association between variables). Thus, it is not uncommon to hear or read that a P value less than .05 implies at least a 95% probability that the observed association is not due to chance. This statement represents a fundamental misunderstanding of P values. Calculation of P values is based on the assumption that the null hypothesis is true. The probability that an association is real depends not just on the probability of its occurrence under the null hypothesis but also on the probability of another basis for the association (see later)—an assessment that depends on information from outside the study, sometimes called the prior probability of an association (of a certain magnitude) estimated before the study results were known and requiring a different approach to statistical inference. Similarly, CIs do not take into account previous information on the probable range of the parameter being estimated. Bayesian methods, which explicitly combine prior knowledge with new information, are beginning to enter the mainstream medical literature.2 Appropriate test statistics and methods for calculating P values depend on the type of variable, just as with descriptive statistics (see E-Table 9-1). For example, to test the hypothesis that the mean values of a continuous variable are equal in two groups, a t test would be used; to compare the mean values across multiple groups, analysis of variance would be used. Because there are many different ways for the null hypothesis to be false (i.e., many different ways that two variables might be associated) and many test statistics that could be calculated, there are many different ways of calculating a P value for

36

CHAPTER 9  Statistical Interpretation of Data  

the association of the same two variables in a data set, and they may not all give the same answer.

Meta-analysis

Statistical techniques for inferring population values from a sample are not restricted to samples of individuals. Meta-analysis is a statistical method for drawing inferences from a sample of studies to derive a summary estimate and confidence interval for a parameter measured by the included studies, such as a risk ratio for a treatment effect.3 Meta-analysis allows the formal combination of results while estimating and accommodating both the within-study and between-study variations. Meta-analysis is most often done when raw data from the studies are not available, as is typically the case when synthesizing information from multiple published results. For example, the previously cited estimate that a 1-g/day change in salt intake is associated with a 1-mm Hg change in blood pressure was obtained from a meta-analysis of randomized trials of low-salt diets in adults.

INFERRING CAUSALITY

In many cases, a goal of clinical research is not just to identify associations but also to determine whether they are causal, that is, whether the predictor causes the outcome. Thus, if people who take vitamin E live longer than those who do not, it is important to know whether it is because they took the vitamin or for some other reason. Determination of causality is based on considering alternative explanations for an association between two variables and trying to exclude or confirm these alternative explanations. The alternatives to a causal relationship between predictor and outcome variables are chance, bias, effect-cause, and confounding. P values and CIs help assess the likelihood of chance as the basis for an association. Bias occurs when systematic errors in sampling or measurements can lead to distorted estimates of an association. For example, if those making measurements of the outcome variable are not blinded to values of the predictor variable, they may measure the outcome variable differently in subjects with different values of the predictor variable, thereby distorting the association between outcome and predictor. Effect-cause is a particular problem in cross-sectional studies, in which (in contrast to longitudinal studies) all measurements are made at a single point in time, thereby precluding demonstration that the predictor variable preceded the outcome—an important part of demonstrating causality. Sometimes biology provides clear guidance about the direction of causality. For example, in a cross-sectional study relating levels of urinary cotinine (a measure of exposure to tobacco smoke) to decreases in pulmonary function, it is hard to imagine that poor pulmonary function caused people to be exposed to smoke. Conversely, sometimes inferring causality is more difficult: are people overweight because they exercise less, or do they exercise less because they are overweight (or both)?

Confounding

Confounding can occur when one or more extraneous variables are associated with both the predictor of interest and the outcome. For example, observational studies suggested that high doses of vitamin E might decrease the risk for heart disease. However, this association seems to have been largely due to confounding: people who took vitamin E were different in other ways from those who did not, including differences in factors causally related to coronary heart disease. If such factors are known and can be measured accurately, one way to reduce confounding is to stratify or match on these variables. The idea is to assemble groups of people who did and did not take vitamin E but who were similar in other ways. Multivariate analysis can accomplish the same goal—other measured variables are held constant statistically, and the effect of the variable of interest (in this case the use of vitamin E) can be examined. Multivariate analysis has the advantage that it can control simultaneously for more potentially confounding variables than can be considered with stratification or matching, but it has the disadvantage that a model must be created (see earlier), and this model may not fit the data well. A new technique that is less dependent on model fit but still requires accurate measurements of confounding variables is the use of propensity scores. Propensity scores are used to assemble comparable groups in the same way as stratification or matching, but in this case the comparability is achieved on the basis of the propensity to be exposed to or be treated with the predictor variable of primary interest. Although propensity scores can adjust only for known confounders and are more subject to manipulation by investigators, systematic reviews suggest that they usually give answers that are generally similar to those of randomized trials addressing the same question.4

A major limitation of these methods of controlling for confounding is that the confounders must be known to the investigators and accurately measured. In the case of vitamin E, apparent favorable effects persisted after controlling for known confounding variables. It is for this reason that randomized trials provide the strongest evidence for causality. If the predictor variable of interest can be randomly assigned, confounding variables, both known and unknown, should be approximately equally distributed between the subjects who are and are not exposed to the predictor variable, and it is reasonable to infer that any significant differences in outcome that remain in these now comparable groups would be due to differences in the predictor variable of interest. In the case of vitamin E, a recent meta-analysis of randomized trials found no benefit and in fact suggested harm from high doses.

OTHER COMMON STATISTICAL PITFALLS

Missing Data

Research on human subjects is challenging. People drop out of studies, refuse to answer questions, miss study visits, and die of diseases that are not being studied directly in the protocol. Consequently, missing or incomplete data are a fact of medical research. When the particular data that are missing are unrelated to the outcome being studied (which might be true, for example, if the files storing the data got partially corrupted), analyses using only the data present (sometimes called a complete case analysis) are unlikely to be misleading. Unfortunately, such is rarely the case. Subjects refusing to divulge family income probably have atypical values, patients not coming for scheduled visits in a study of depression may be more or less depressed, and patients in an osteoporosis study who die of heart disease probably differ in many ways from those who do not. Whenever a sizable fraction of the data is missing (certainly if it is above 10 or 15%), there is the danger of substantial bias from an analysis that uses only the complete data. This is the gap noted earlier between the intended and actual samples. Any study with substantial missing data should be clear about how many missing data there were and what was done to assess or alleviate the impact; otherwise, the critical consumer of such information should be suspicious. Multiple imputation is a technique of using observations with nonmissing data to estimate missing values; it can produce less biased estimates than simply excluding the observations with missing data.5 In a randomized trial, the general rule is that the primary analysis should include all subjects who were randomized, regardless of whether they followed the study protocol, in an intention-to-treat analysis.

Clustered or Hierarchical Data

Data are often collected in a clustered (also called hierarchical) manner; for example, NHANES used a cluster sample survey, and a study of patient outcomes might be conducted at five hospitals, each with multiple admission teams. The cluster sample or the clustering of patients within teams within hospitals leads to correlated data. Said another way, and other things being equal, data collected on the same patient, by the same admission team, or in the same cluster are likely to be more similar than data from different patients, teams, or clusters. Failure to use statistical methods that accommodate correlated data can seriously misstate standard errors, widths of CIs, and P values, most often leading to overly optimistic estimates, that is, standard errors and P values that are incorrectly too small and CIs that are incorrectly too narrow. Statistical methods for dealing with correlated data include generalized estimating equations and the use of robust standard errors and frailty models (for time-to-event data). Studies with obvious hierarchical structure that fail to use such methods may be in serious error.

Multiple Hypothesis Testing

The “multiple hypothesis testing” or “multiple comparisons” issue refers to the idea that if multiple statistical tests are conducted, each at a significance level of .05, the chance that at least one of them will achieve a P value of less than .05 is considerably larger than .05, even when all the null hypotheses are true. For example, when comparing the mean value of a continuous variable across many different groups, analysis of variance is a time-tested method of performing an overall test of equality and avoiding making a large number of pairwise comparisons. Because most medical studies collect data on a large number of predictor variables, performing a test on the association of each one with the outcome may generate false-positive results. The risk for falsely positive results is especially high with genomic studies, in which a researcher may test a million single-nucleotide polymorphisms for association with a disease.

A typical method for dealing with the problem of multiple testing is the Bonferroni correction, which specifies that the P value at which the null hypothesis will be rejected (e.g., .05) should be divided by the number of hypothesis tests performed. Although simple to use, a problem with this approach is that it is overly conservative. Studies with many listed or apparent outcomes or predictors (or both) are subject to inflation of the error rate to well above the nominal .05. Automated stepwise regression methods for choosing predictors in regression models typically do not alleviate and may exacerbate this problem. If no adjustment or method for dealing with multiple comparisons is used, the high probability of false-positive results should be kept in mind. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 9  Statistical Interpretation of Data  

GENERAL REFERENCES 1. Newman TB, Kohn MA. Evidence-Based Diagnosis. New York: Cambridge University Press; 2009. A practical text for clinicians. 2. Goodman SN. Bayesian methods for evidence evaluation: are we there yet? Circulation. 2013;127: 2367-2369.

37.e1

3. Murad MH, Montori VM, Ioannidis JP, et al. How to read a systematic review and meta-analysis and apply the results to patient care: users’ guides to the medical literature. JAMA. 2014;312:171-179. 4. Lonjon G, Boutron I, Trinquart L, et al. Comparison of treatment effect estimates from prospective nonrandomized studies with propensity score analysis and randomized controlled trials of surgical procedures. Ann Surg. 2014;259:18-25. 5. Cummings P. Missing data and multiple imputation. JAMA Pediatr. 2013;167:656-661.

37.e2

CHAPTER 9  Statistical Interpretation of Data  

REVIEW QUESTIONS 1. A study of 105 vegan Buddhist nuns randomly sampled from monasteries around Ho Chi Minh City found that the average femoral neck bone mineral density was 0.62 g/cm2, with a standard deviation of 0.11 g/cm2 (Ho-Pham LT, Nguyen PL, Le TT, et al. Veganism, bone mineral density, and body composition: a study in Buddhist nuns. Osteoporos Int. 2009;20:2087-2093). Which of the following statements about this result is correct? A. The 95% confidence interval for the mean bone mineral density in these women is about 0.4 to 0.84 g/cm2. B. If bone mineral density is normally distributed, we would expect about 10% of the women in the sample to have bone mineral density outside of the interval: 0.4 to 0.84 g/cm2. C. The 95% confidence interval for the mean bone mineral density in these women is about 0.60 to 0.64 g/cm2. D. Because the women were sampled randomly, there is a 95% chance that a randomly selected woman from the population would have a bone mineral density between 0.60 and 0.64 g/cm2. E. Because the women were sampled randomly, there is a 95% chance that a randomly selected woman from the sample would have a bone mineral density between 0.60 and 0.64 g/cm2. Answer: C  We would expect about 95% of observations to be within 2 standard deviations of the sample mean, leaving 5% out, so choice B is incorrect. The 95% confidence interval for a sample mean is about mean ± 2 standard SD errors of the mean (SEM), where the SEM = . In this case, the SD is N 2 0.11 g/cm , and the N is about 100, so the SEM will be about 0.11/10 = 0.01 g/cm2, and the 95% CI will be about 0.60 to 0.64 g/cm2, as indicated in choice C. Choices D and E are incorrect because the range given is too narrow: it is ± 2 SEM when it should be ± 2 SD. 2. A study of data collected through the Get with the Guidelines-Stroke Program examined the time from onset of stroke symptoms to treatment with tissue-type plasminogen activator (tPA) among 58,353 patients with acute ischemic stroke treated within 4.5 hours of the onset of symptoms (Saver JL, Fonarow GC, Smith EE, et al. Time to treatment with intravenous tissue plasminogen activator and outcome from acute ischemic stroke. JAMA. 2013;309:2480-2488). The authors reported that “faster onset-to treatment time, in 15-minute increments, was associated with … increased achievement of independent ambulation at discharge (OR, 1.04; 95% CI 1.03-1.05; P < 0.001) … .” Which of the following is a correct interpretation of these findings? A. In this study, the effect of a 15-minute reduction in onset-to treatment time was associated with a 4 % (relative) increase in the odds of independent ambulation at discharge. B. Because the odds ratio is very close to 1.0, the results are not statistically significant. C. Because the odds ratio is very close to 1.0, the results, although highly statistically significant, are not clinically significant. D. The 4% increase in odds of independent ambulation translates into a number-needed-to treat (NNT) of 25. E. None of the above is correct. Answer: A  Choice A exactly expresses the meaning of the odds ratio for this study. Choices B and C are incorrect because the proximity of the odds ratio to 1 is in this case based partly on the choice of the authors to express it per 15 minutes onset-to-treatment time. This illustrates the importance of knowing the units of the predictor variable when it is not dichotomous. If the authors had expressed the difference per hour instead of per 15 minutes, the odds ratios would have been taken to the fourth power, that is, the odds ratio for independent ambulation at discharge would have been about 1.044 = 1.17. Choice D is incorrect because estimation of the NNT requires knowing the absolute risk reduction, which was not provided in this case.

3. Assume a study of both smoking and nonsmoking mothers reports that the effect of smoking on birthweight is about a 32-g decrease in birthweight per cigarette smoked per day during pregnancy (similar to what is reported by Juarez SP, Merlo J. Revisiting the effect of maternal smoking during pregnancy on offspring birthweight: a quasi-experimental sibling analysis in Sweden. PLoS One. 2013;8:e61734.). Which of the following statements about this finding is NOT correct? A. The result is based on a model, in which the predicted birthweight is linearly related to the number of cigarettes smoked per day. B. This model predicts that the difference in birthweight between a baby whose mother did not smoke and one who smoked 5 cigarettes per day is the same as the difference between babies of mothers who smoked 20 and 25 cigarettes per day. C. The model predicts the same effect of smoking on birthweight, regardless of the mother’s age and prepregnancy weight. D. This model predicts that cutting cigarette smoking in half will lead to a 64-g expected weight increase in the baby. E. The model could include additional terms that would reflect the effect of mother’s age and prepregnancy weight. Answer: D  Choices A, B and C accurately describe characteristics of a linear model for the effect of smoking on birthweight. Choice D is not consistent with a linear model. Choice E reflects that the effects of other variables can be taken into account, while still maintaining a linear model for the effect of cigarettes smoked on birthweight. 4. A case-control study of the relationship of breast cancer and the use of COX-2 inhibitors found that the odds ratio and confidence interval relating cancer to use of baby aspirin were OR = 0.77 and 95% CI (0.42-1.41) (Harris RE, Beebe-Donk J, Alshafie GA. Reduction in the risk of human breast cancer by selective cyclooxygenase-2 [COX-2] inhibitors. BMC Cancer. 2006;6:27). The authors then stated, “Neither acetaminophen nor baby aspirin had any effect on the relative risk of breast cancer.” This is incorrect because of which of the following? A. The confidence interval crosses 1. B. The authors do not give the P value. C. The confidence interval has a lower limit of 0.42. D. Odds ratios are inappropriate for this study. E. The authors should have used a higher level of confidence. Answer: C  C is correct because the confidence interval allows for a 58% reduction (from [1 to 0.42]*100%) in the chance of breast cancer associated with the use of baby aspirin, a potentially important effect. Choice A is incorrect because it merely indicates a lack of a statistically significant result and does not bear on the size of the effect. Choice B is incorrect because CIs are more useful for ruling out important effects. Odds ratios are especially useful in case-control studies, so D is incorrect. And E is incorrect because a higher level of confidence would make the CI even wider. 5. In a randomized trial of an intervention to reduce hypertension, researchers selected subjects from a single antihypertensive patient club in Shanghai (Xue F, Yao W, Lewin RJ. A randomised trial of a 5 week, manual based, self-management programme for hypertension delivered in a cardiac patient club in Shanghai. BMC Cardiovasc Disord. 2008;8:10). About one third of those approached agreed to be randomized. The results of this study can be safely generalized to which of the following? A. All hypertensives B. All hypertensives in China C. All hypertensives in Shanghai D. All hypertensives belonging to the single club in Shanghai E. None of the above. Answer: E  Because two thirds of the participants refused to participate, the extrapolation of the effect of the intervention might not even apply to the particular club in which the study was conducted, much less a broader population.

CHAPTER 10  Using Data for Clinical Decisions  

37

TABLE 10-1  KEY DEFINITIONS* Probability

A number between 0 and 1 that expresses an estimate of the likelihood of an event

Odds

The ratio of [the probability of an event] to [the probability of the event’s not occurring]

TEST PERFORMANCE CHARACTERISTICS Sensitivity

Percentage of patients with disease who have an abnormal test result

Specificity

Percentage of patients without disease who have a normal test result

Positive predictive value

Percentage of patients with an abnormal test result who have disease

Negative predictive value

Percentage of patients with a normal test result who do not have disease

BAYESIAN ANALYSIS Pretest (or prior) probability

10  USING DATA FOR CLINICAL DECISIONS THOMAS H. LEE Key functions in the professional lives of all physicians are the collection and analysis of clinical data. Decisions must be made on the basis of these data, including which therapeutic strategy is most appropriate for the patient and whether further information should be gathered before the best strategy can be chosen. This decision-making process is a blend of science and art in which the physician must synthesize a variety of concerns, including the patient’s most likely outcome with various management strategies, the patient’s worst possible outcome, and the patient’s preferences among these strategies. Only rarely does the physician enjoy true certainty regarding any of these issues, so a natural inclination for physicians is to seek as much information as possible before making a decision. This approach ignores the dangers inherent in the collection of information. Some of these dangers are immediate, such as the risk of cerebrovascular accident associated with coronary angiography. Some dangers are delayed, such as the risk of a malignant neoplasm due to radiation exposure from diagnostic tests. And some dangers are subtle, such as the risk of unnecessary anguish for patients due to delays, uncertainty, and confusion. An additional concern is the cost of information gathering, including the direct costs of the tests themselves and the indirect costs that flow from decisions made on the basis of the test results. Substantial data demonstrate marked variation in use of tests among physicians located in different regions and even within the same group practice. Standards of medical professionalism endorse the need for physicians to exert their influence to minimize inefficiency, but this challenge grows increasingly complex as medical progress leads to proliferation of alternative testing strategies. For the physician, there are three key questions in this sequence: Should I order a test to improve my assessment of diagnosis or prognosis? Which test is best? Which therapeutic strategy is most appropriate for this patient?

SHOULD I ORDER A TEST?

The decision of whether to order a test depends on the physician’s and the patient’s willingness to pursue a management strategy with the current degree of uncertainty.1 This decision is influenced by several factors, including the patient’s attitudes toward diagnostic and therapeutic interventions (e.g., a patient with claustrophobia might prefer an ultrasound to magnetic resonance imaging) and the information provided by the test itself. The personal tolerance of the patient and physician for uncertainty also frequently influences test-ordering approaches. A decision to watch and wait rather than to obtain a specific test also should be considered an information-gathering alternative because the information obtained while a patient is being observed often reduces uncertainty about the diagnosis and outcome. In other words,

The probability of a disease before the information is acquired

Post-test (or posterior) probability The probability of a disease after new information is acquired Pretest (or prior) odds

(Pretest probability of disease)/(1 − pretest probability of disease)

Likelihood ratio

(Probability of result in diseased persons)/ (Probability of result in nondiseased persons)

*Disease can mean a condition, such as coronary artery disease, or an outcome, such as postoperative cardiac complications.

the “test of time” should be recognized as one of the most useful tests available when this tactic does not seem inappropriately risky. Most tests do not provide a definitive answer about diagnosis or prognosis but instead reduce uncertainty. Accordingly, the impact of information from tests often is expressed as probabilities (Table 10-1). A probability of 1.0 implies that an event is certain to occur, whereas a probability of 0 implies that the event is impossible. When all the possible events for a patient are assigned probabilities, these estimates should sum to 1.0. It is often useful to use odds to quantify uncertainty instead of probability. Odds of 1 : 2 suggest that the likelihood of an event is only half the likelihood that the event will not occur, or a probability of 0.33. The relationship between odds and probability is expressed in the following formula: Odds = P /(1 − P) where P is the probability of an event.

Performance Characteristics

Sensitivity and specificity are key terms for the description of test performance. These parameters describe the test and are in theory true regardless of the population of patients to which the test is applied. Research studies that describe test performance often are based, however, on highly selected populations of patients; test performance may deteriorate when tests are applied in clinical practice. The result of a test for coronary artery disease, such as an electron beam computed tomography scan, rarely may be abnormal if it is evaluated in a low-risk population, such as high-school students. Falsepositive abnormal results secondary to coronary calcification in the absence of obstructive coronary disease are common when the test is performed in middle-aged and elderly people. Another increasingly appreciated factor that can distort the performance of screening tests is the phenomenon of “overdiagnosis,” in which the “disease” that is detected would not have led to clinical harm if it had not been found.2 Although researchers are interested in the performance of tests, the true focus of medical decision making is the patient. Physicians are more interested in the implications of a test result on the probability that a patient has a specific disease or outcome, that is, the predictive values of abnormal or normal test results. These predictive values are extremely sensitive to the population from which they are derived (Table 10-2; see also Table 10-1). An abnormal lung scan result in an asymptomatic patient has a much lower positive predictive value than that same test result in a patient with dyspnea and diminished oxygen saturation. Bayes theorem (see later) provides a

38

CHAPTER 10  Using Data for Clinical Decisions  

1

Question: What is the probability of coronary disease for a patient with a 50% pretest probability of coronary disease who undergoes an exercise test if that patient develops (a) no ST segment changes, (b) 1 mm of ST segment depression, or (c) 2 mm of ST segment depression? Step 1. Calculate the pretest odds of disease: P /(1 − P) = 0.5/(1 − 0.5) = 0.5/0.5 =1

Post-test probability

TABLE 10-2  EXAMPLE OF ODDS RATIO FORM OF BAYES THEOREM

Step 2. Calculate the likelihood ratios for the various test results, using the formula LR = sensitivity/(1 − specificity). (Data from pooled literature.) TEST RESULT No ST segment changes 1-mm ST segment depression 2-mm ST segment depression

SENSITIVITY

SPECIFICITY

LIKELIHOOD RATIO

0.34 0.66 0.33

0.15 0.85 0.97

0.4 4.4 11

Step 3. Calculate the post-test odds of disease and convert those odds to post-test probabilities. TEST RESULT No ST segment changes 1-mm ST segment depression 2-mm ST segment depression

PRETEST ODDS

LIKELIHOOD RATIO

POST-TEST ODDS

POST-TEST PROBABILITY

1

0.4

0.4

0.29

1

4.4

4.4

0.81

1

11

11

0.92

framework for analyzing the interaction between test results and a patient’s pretest probability of a disease. As useful as the performance characteristics may be, they are limited by the fact that few tests truly provide dichotomous (i.e., positive or negative) results. Tests such as exercise tests have several parameters (e.g., ST segment deviation, exercise duration, hemodynamic response) that provide insight into the patient’s condition, and the normal range for many blood tests (e.g., a serum troponin level) varies markedly according one’s willingness to “miss” patients with disease. Tests that require human interpretation (e.g., radiologic studies) are particularly subject to variability in the reported results.

Bayes Theorem

The impact of a test result on a patient’s probability of disease was first quantified by Bayes, an 18th century English clergyman who developed a formula that describes the probability of disease in the presence of an abnormal test result. The classic presentation of Bayes theorem is complex and difficult to use. A more simple form of this theorem is known as the odds ratio form, which describes the impact of a test result on the pretest odds (see Table 10-1) of a diagnosis or outcome for a specific patient. To calculate the post-test odds of disease, the pretest odds are multiplied by the likelihood ratio (LR) for a specific test result. The mathematical presentation of this form of Bayes theorem is as follows: Post-test odds = (Pretest odds) × (LR ) The LR is the probability of a particular test result in patients with the disease divided by the probability of that same test result in patients without disease. In other words, the LR is the test result’s sensitivity divided by the false-positive rate. A test of no value (e.g., flipping a coin and calling “heads” an abnormal result) would have an LR of 1.0 because half of patients with disease would have abnormal test results, as would half of patients without disease. This test would have no impact on a patient’s odds of disease. The further an LR is above 1.0, the more that test result raises a patient’s probability of disease. For LRs less than 1.0, the closer the LR is to 0, the more it lowers a patient’s probability of disease. When it is displayed graphically (Fig. 10-1), a test of no value (dotted line) does not change the pretest probability, whereas an abnormal or normal result from a useful test moves the probability up or down. For a patient with a high pretest probability of disease, an abnormal test result changes the

0 0

Pre-test probability

1

FIGURE 10-1.  Impact of various test results on the patient’s probability of disease. The x-axis depicts a patient’s probability of disease before a test. If the test is of no value, the post-test probability (dotted line) is no different from the pretest probability. An abnormal test result raises the post-test probability of disease, as depicted by the concave downward arc, whereas a normal test result lowers the probability.

patient’s probability only slightly, but a normal test result leads to a marked reduction in the probability of disease. Similarly, for a patient with a low pretest probability of disease, a normal test result has little impact, but an abnormal test result markedly raises the probability of disease. Consider how various exercise test results influence a patient’s probability of coronary disease (see Table 10-2). For a patient whose clinical history, physical examination, and electrocardiographic findings suggest a 50% probability of disease, the pretest odds of disease are 1.0. LRs for various test results are developed by pooling data from published literature. The sensitivity of an exercise test with any amount of ST segment changes is the rate of such test results in patients with coronary disease, and the specificity is the percentage of patients without coronary disease who do not have this test result. The LR for no ST change is less than 1, whereas the LRs for patients with ST changes are greater than 1 (see Table 10-2). Therefore, when the LRs for various test results are multiplied by the pretest odds to calculate post-test odds, the odds decrease for patients without ST segment changes but increase for patients with 1 or 2 mm of ST segment change. Post-test odds can be converted to post-test probabilities according to the following formula: Probability = Odds/(1 + odds) The calculations quantify how the absence of ST segment changes reduces a patient’s probability of disease, whereas ST segment depression raises the probability of disease. This form of Bayes theorem is useful for showing how the post-test probability of disease is influenced by the patient’s pretest probability of disease. If a patient’s clinical data suggest a probability of coronary disease of only 0.1, the pretest odds of disease would be only 0.11. For such a low-risk patient, an exercise test with no ST segment changes would lead to post-test probability of coronary disease of 4%, whereas 1-mm or 2-mm ST segment changes would lead to a post-test probability of disease of 33% or 55%, respectively. Even if clinicians rarely perform the calculations that are described in Bayes theorem, there are important lessons from this theorem that are relevant to principles of test ordering (Table 10-3). The most crucial of these lessons is that the interpretation of test results must incorporate information about the patient. An abnormal test result in a low-risk patient may not be a true indicator of disease. Similarly, a normal test result in a high-risk patient should not be taken as evidence that disease is not present. Figure 10-2 provides an example of the post-test probabilities for positive and negative results for a test with a sensitivity of 85% and a specificity of 90% (e.g., radionuclide scintigraphy for diagnosis of coronary artery disease). In a high-risk population with a 90% prevalence of disease, the positive predictive value of an abnormal result is 0.99 compared with 0.31 for the same test result obtained in a low-risk population with a 5% prevalence of disease. Similarly, the negative predictive value of a normal test result is greater in the low-risk population than in the high-risk population.

CHAPTER 10  Using Data for Clinical Decisions  

TABLE 10-3  PRINCIPLES OF TEST ORDERING AND INTERPRETATION The interpretation of test results depends on what is already known about the patient. No test is perfect; clinicians should be familiar with its diagnostic performance (see Table 10-1) and never believe that a test “forces” them to pursue a specific management strategy. Tests should be ordered if they may provide additional information beyond that already available. Tests should be ordered if there is a reasonable chance that the data will influence the patient’s care. Two tests that provide similar information should not be ordered. In choosing between two tests that provide similar data, use the test that has lower costs or causes less discomfort and inconvenience to the patient. Clinicians should seek all of the information provided by a test, not just an abnormal or normal result. The cost-effectiveness of strategies using noninvasive tests should be considered in a manner similar to that of therapeutic strategies.

1000 Patients

900 with disease

765 truepositive results

A

100 without disease

135 falsenegative results

Test with: Sensitivity = 85% Specificity = 90%

10 falsepositive results

90 truenegative results

pathophysiology, such as ventilation-perfusion scintigraphy and pulmonary angiography. Regardless of whether tests are independent, the performance of multiple tests increases the likelihood that an abnormal test result will be obtained in a patient without disease. If a chemistry battery includes 20 tests and the normal range for each test has been developed to include 95% of healthy individuals, the chance that a healthy patient will have a normal result for any specific test is 0.95. However, the probability that all 20 tests will be normal is (0.95)20, or 0.36. Most healthy people can be expected to have at least one abnormal result. Unless screening test profiles are used thoughtfully, falsepositive results can subject patients to unnecessary tests and procedures.

Threshold Approach to Decision Making

Even if a test provides information, that information may not change management for an individual patient. Lumbar spine radiographs of a patient who is not willing to undergo surgery may reveal the severity of disease but expose the patient to needless radiation. Similarly, a test that merely confirms a diagnosis that already is recognized is a waste of resources (see Table 10-3). Before ordering a test, clinicians should consider whether that test result could change the choice of management strategies. This approach is called the threshold approach to medical decision making, and it requires the physician to be able to estimate the threshold probability at which one strategy will be chosen over another. The management of a clinically stable patient with a high probability of coronary disease might not be changed by any of the posttest probabilities shown in Table 10-2. If that patient had no ST segment changes, the post-test probability of 0.29 still would be too high for a clinician to consider that patient free of disease. An abnormal test result that strengthened the diagnosis of coronary disease might not change management unless it suggested a greater severity of disease that might warrant another management strategy.

Testing for Peace of Mind Positive predictive value: 765/775 = 0.99 Negative predictive value: 90/225 = 0.40

1000 Patients

950 without disease

50 with disease

Physicians frequently order tests even when there is little chance that the outcomes will provide qualitatively new insights into a patient’s diagnosis or prognosis or alter a patient’s management. In such cases, the cited goal for testing may be to improve a patient’s peace of mind. Although a decrease in uncertainty can improve quality of life for many patients, individuals with hypochondriasis and somatization disorders rarely obtain comfort from normal test results; instead, their complaints shift to a new organ system, and their demands focus on other tests. For such patients, management strategies using frequent visits and cognitive tactics are recommended.

WHICH TEST IS BEST?

42.5 truepositive results

B

7.5 falsenegative results

Test with: Sensitivity = 85% Specificity = 90%

95 falsepositive results

855 truenegative results

Positive predictive value: 42.5 /137.5 = 0.31 Negative predictive value: 855/862.5 = 0.99

FIGURE 10-2.  Interpretation of test results in high-risk and low-risk patients. A,

High-risk population (90% prevalence of disease). B, Low-risk population (5% prevalence of disease).

Multiple Testing

39

Clinicians frequently obtain more than one test aimed at addressing the same issue and at times are confronted with conflicting results. If these tests are truly independent (i.e., the tests do not have the same basis in pathophysiology), it may be appropriate to use the post-test probability obtained through performance of one test as the pretest probability for the analysis of the impact of the second test result. If the tests are not independent, this strategy for interpretation of serial test results can be misleading. Suppose a patient with chronic obstructive pulmonary disease and a history vaguely suggestive of pulmonary embolism is found to have an abnormal lung ventilation-perfusion scan. Obtaining that same test result over and over would not raise that patient’s probability of pulmonary embolism further and further. In this extreme case, the tests are identical; serial testing adds no information. More commonly, clinicians are faced with results from tests with related but not identical bases in

If the clinician decides that more information is needed to reduce uncertainty, and if it appears possible that tests might lead to a change in management strategies, the question arises as to which test is most appropriate. Note that just because guideline development committees have concluded that a specific test is “appropriate” in a given clinical context, it does not mean that this test is the most appropriate option. Several factors influence the choice among diagnostic strategies, including the patient’s preferences, the costs and risks associated with the tests, and the diagnostic performance of alternative tests. Diagnostic performance of a test often is summarized in terms of sensitivity and specificity,3 but as shown in the example in Table 10-2, these parameters depend on which threshold (e.g., 1 mm vs. 2 mm of ST segment change) is used. A low threshold for calling a test result abnormal might lead to excellent sensitivity for detecting disease but at the expense of a high false-positive rate. Conversely, a threshold that led to few false-positive results might cause a clinician to miss many cases of true disease. The receiver operating characteristic (ROC) curve is a graphic form of describing this tradeoff and providing a method for comparing test performance (Fig. 10-3). Each point on the ROC curve describes the sensitivity and the false-positive rate for a different threshold for abnormality for a test. A test of no value would lead to an ROC curve with the course of the dotted line, whereas a misleading test would be described by a curve that was concave upward (not shown). The more accurate the test, the closer its ROC curve comes to the upper left corner of the graph, which would indicate a test threshold that has excellent sensitivity and a low false-positive rate. The closer an ROC curve comes to the upper left corner, the greater the area under the curve. The area under ROC curves can be used to compare the information provided by two tests.

40

CHAPTER 10  Using Data for Clinical Decisions  

1

TABLE 10-4  STEPS IN PERFORMANCE OF DECISION ANALYSIS

Sensitivity

Frame the question. Create the decision tree. Identify the alternative strategies. List the possible outcomes for each of the alternative strategies. Describe the sequence of events as a series of decision nodes and chance nodes. Choose a time horizon for the analysis. Determine the probability for each chance outcome. Assign a value to each outcome. Calculate the expected utility for each strategy. Perform sensitivity analysis.

0 0

False-positive rate

1

FIGURE 10-3.  Receiver operating characteristic curve. The points on the curve reflect the sensitivity and false-positive (1 − specificity) rates of a test at various thresholds. As the threshold is changed to yield greater sensitivity for detecting the outcome of interest, the false-positive rate rises. The better the test, the closer the curve comes to the upper left corner. A test of no value (e.g., flipping a coin) would lead to a curve with the course of the dotted line. The area under the curve is used often to compare alternative testing strategies.

Even if one test is superior to another as shown by a greater area under its ROC curve, the question still remains as to what value of that test should be considered abnormal. The choice of threshold depends on the purpose of testing and on the consequences of a false-positive or false-negative diagnosis. If the goal is to screen the population for a disease that is potentially fatal and potentially curable, a threshold with excellent sensitivity is appropriate even if it leads to frequent false-positive results. In contrast, if a test is used to confirm a diagnosis that is likely to be treated with a high-risk invasive procedure, a threshold with high specificity is preferred. Only 1 mm of ST segment depression might be the appropriate threshold when exercise electrocardiography is used to evaluate the possibility of coronary disease in a patient with chest pain. If the question is whether to perform coronary angiography in a patient with stable angina in search of severe coronary disease that might benefit from revascularization, a threshold of 2 mm or more would be more appropriate.

CHOOSING A STRATEGY

Physicians and patients ultimately must use clinical information to make decisions. These choices usually are made after consideration of a variety of factors, including information from the clinical evaluation, patients’ preferences, and expected outcomes with various management strategies. Insight into the impact of these considerations can be improved through the performance of decision analysis (Table 10-4). The first step in a decision analysis is to define the problem clearly; this step often requires writing out a statement of the issue so that it can be scrutinized for any ambiguity. After the problem is defined, the next step is to define the alternative strategies. Consider the question of which test is most appropriate to screen patients for breast cancer: mammography with or without breast magnetic resonance imaging—a technology that is highly sensitive for detecting breast cancer but is more costly and less specific. The expected outcomes for these strategies depend on each test’s sensitivity and specificity for detecting breast cancer, which is influenced in turn by other factors, such as the frequency with which the test is performed. Patients’ outcomes also are influenced by their underlying risk for breast cancer and the likelihood that earlier detection of tumors reduces the risk for death. Each of these variables must be known or estimated for calculations to be made of each strategy’s predicted life expectancy and direct medical costs. These outcomes differ for patients according to age, medical history, family history, and presence or absence of genetic markers such as BRCA mutations. Optimal strategies for an elderly patient with a short life expectancy and low clinical risk of cancer are unlikely to be the same as those for a younger patient with inherited mutations of the BRCA1 or BRCA2 gene, indicating a cumulative lifetime risk of breast cancer of 50 to 85% (Chapter 198).

The credibility of the decision analysis depends on the credibility of these estimates. Published reports often do not provide information on the outcomes of interest for specific subsets of patients, or there may not have been sufficient statistical power within subsets of patients for the findings to be statistically significant. Randomized trial data are relevant to the populations included in the trial; the extension of the findings to other genders, races, and age groups requires assumptions by individuals performing the analysis. For many issues, expert opinion must be used to derive a reasonable estimate of the outcome. For many diseases, the potential outcomes are more complex than perfect health or death. With chronic diseases, patients may live many years in a condition somewhere between these two, and the goal of medical interventions may be to improve quality of life rather than to extend survival. The value of life in imperfect health must be reflected in decision analyses. These values by convention are expressed on a scale of 0 to 100, where 0 indicates the worst outcome and 100 indicates the best outcome. Life-expectancy and quality-of-life estimates are combined in many decision analyses to calculate quality-adjusted life years. A strategy that leads to a 10-year life expectancy with such severe disability that utility of the state of health is only half that of perfect health would have a quality-adjusted life expectancy of 5 years. With such adjustments to life-expectancy data, the impact of interventions that improve quality of life but do not extend life can be compared with interventions that extend life but do not improve its quality or perhaps even worsen it.4 After the value and the probability of the various outcomes have been estimated, the expected utility of each strategy can be calculated. In comparing the different strategies available at a decision node, the analysis generally selects the option with the highest expected utility. At chance nodes, the expected utility is the weighted average of the utility of the various possible branches. After the analysis has been performed with the baseline assumptions, sensitivity analyses should be performed in which these assumptions are varied over a reasonable range. These analyses can reveal which assumptions have the most influence over the conclusions and identify threshold probabilities at which the conclusions would change. For example, the threshold at which breast magnetic resonance imaging should be added to mammography is likely to be influenced by the cost of the magnetic resonance imaging and the accuracy of the radiologists who interpret the images.

Cost-Benefit and Cost-Effectiveness Analyses

For clinicians and health care policymakers, the choices that must be addressed go beyond the choices within any single decision analysis. Because resources available for health care are limited, policymakers may have to choose among many competing options for “investments” in health. Although such decisions frequently are made on the basis of political considerations, cost-benefit and cost-effectiveness analyses can be informative in making the choices. The methodology of these techniques is similar to that of decision analysis except that costs for the various possible outcomes and strategies also are calculated. Discounting is used to adjust the value of future benefits and costs because resources saved or spent currently are worth more than resources saved or expended in the future. In cost-benefit analyses, all benefits are expressed in terms of economic impact. Extensions in life expectancy are translated into dollars by estimating societal worth or economic productivity. Because of the ethical discomfort associated with expressing health benefits in financial terms, cost-effectiveness analyses are used more commonly

TABLE 10-5  ESTIMATED COST-EFFECTIVENESS OF SELECTED HEALTH INTERVENTIONS INTERVENTION

COST PER QUALITY-ADJUSTED LIFE YEAR (QALY) (2010 DOLLARS)

Treating rheumatoid arthritis with drugs that slow disease progression

Saves money and improves health

Using warfarin for 70-year-olds with atrial fibrillation

$3,000 per QALY

Daily dialysis for 60-year old critically ill men with kidney injury

$6,000 per QALY

Using an implantable cardioverterdefibrillator to prevent sudden cardiac death in high-risk patients

$38,000 per QALY

Treating spinal stenosis and leg pain with spine surgery

$90,000 per QALY

Screening 60-year-old heavy smokers with $140,000 per QALY annual CT scans Annual HIV screening for people with a low to moderate risk

Increases costs and makes health worse

Modified from the CEA Registry.6

than cost-benefit analyses. In these analyses, the ratio of costs to health benefits is calculated; one frequently used method for evaluating a strategy is calculation of cost per quality-adjusted life year.5 These estimates can be used to identify strategies that are both cost-saving and health-improving, to compare strategies by which the health care system can “purchase” additional quality-adjusted life years, and even to caution about strategies that increase costs while worsening health (Table 10-5).6 Cost-effectiveness analyses can provide important insights into the relative attractiveness of different management strategies and can help guide policymakers in decisions about which technologies to make available on a routine basis. No medical intervention can have an attractive cost-effectiveness if its effectiveness has not been proved. The cost-effectiveness of an intervention depends heavily on the population of patients in which it is applied. An inexpensive intervention would have a poor cost-effectiveness ratio if it were used in a low-risk population unlikely to benefit from it. In contrast, an expensive technology can have an attractive cost-effectiveness ratio if it is used in patients with a high probability of benefiting from it. Table 10-5 shows cost-effectiveness estimates from published literature for some selected medical interventions. Such estimates should be used only with understanding of the population for which they are relevant. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 10  Using Data for Clinical Decisions  

GENERAL REFERENCES 1. Laine C. High-testing begins with a few simple questions. Ann Intern Med. 2012;156:162-163. 2. Etzioni R, Gulati R, Mallinger L, et al. Influence of study features and methods on overdiagnosis estimates in breast and prostate cancer screening. Ann Intern Med. 2013;158:831-838. 3. Otero HJ, Fang CH, Sekar M, et al. Accuracy, risk and the intrinsic value of diagnostic imaging: a review of the cost-utility literature. Acad Radiol. 2012;19:599-606.

41.e1

4. Heijnsdijk EAM, Wever EM, Auvinen A, et al. Quality-of-life effects of prostate-specific antigen screening. N Engl J Med. 2012;367:595-605. 5. Ryen L, Svensson M. The willingness to pay for a quality adjusted life year: a review of the empirical literature. Health Econ. 2014. [Epub ahead of print]. 6. Cost-Effectiveness Analysis Registry. https://research.tufts-nemc.org/cear4/. Accessed February 10, 2015.

41.e2

CHAPTER 10  Using Data for Clinical Decisions  

REVIEW QUESTIONS 1. A 53-year-old man presents to his primary care physician with a chief complaint of chest pain for the past 2 weeks. The pain is described as intermittent, usually exertional, midline, and aching in nature. His electrocardiogram is completely normal. Which one of the following is the most appropriate next step? A. Watch and wait B. Exercise electrocardiography C. B-type natriuretic peptide D. Exercise nuclear scintigraphy E. Coronary angiography Answer: B The immediate question is whether this patient’s symptoms represent new ischemic heart disease. His clinical presentation suggests a moderate probability of coronary artery disease, so watch-and-wait is probably too risky a strategy. At the same time, he does not appear to have unstable ischemic disease, so there is no need to proceed immediately to coronary angiography in preparation for coronary revascularization. Measurement of B-type natriuretic peptide would not alter management. Of the two noninvasive tests for ischemic heart disease, exercise electrocardiography is the least expensive, is the most convenient, and carries no radiation exposure. Guidelines would thus suggest that answer B is the most appropriate next step. 2. A patient undergoes an exercise test and has 2 mm of ST-segment depression on electrocardiogram. In which one of these patients would this finding be most likely to change management? A. A healthy 19-year-old volunteer in a research study B. A 62-year-old woman who is completely asymptomatic C. A 62-year-old woman with frequent nonexertional aching chest pain D. A 62-year-old woman with recent myocardial infarction and chest pain at rest E. A 62-year-old woman who is completely symptom free 4 months after coronary artery bypass graft surgery

Answer: C Patients A and B have a sufficiently low probability of coronary disease that the exercise test abnormality is highly likely to be a false-positive result. Patient D has an extremely high probability of coronary disease and likely needs coronary angiography and revascularization as next steps; she does not need exercise electrocardiography because the test is unlikely to change this management plan. Patient E’s care is also not likely to be influenced by an exercise test result because she is asymptomatic after major coronary revascularization surgery. Patient C has a low to moderate probability of coronary disease, and this abnormal exercise test result moves her into a mid-range probability. Thus, she is likely to undergo either further testing, initiation of antianginal therapy, or both as a result of this exercise test result. 3. Which one of the following is NOT an important consideration when weighing whether to order a test for a patient? A. Test may influence decision making for patient’s care B. Test is less expensive than alternative strategies C. Test is safer than alternative strategies D. Test reduces patient’s or clinician’s uncertainty E. Test is expected to be abnormal in presence of patient’s already confirmed diagnosis Answer: E Tests should be ordered when they are expected to change care and should be chosen on basis of safety, cost, and impact. They should not be ordered simply because they can confirm an already known diagnosis.

CHAPTER 11  Measuring Health and Health Care  

41

professionals and accredited facilities is sufficient to ensure consistent highquality care. A second major trend is attributable to the successes of biomedical science: the major challenge in health care today is the management of chronic disease for a population with increased life expectancy. For chronic conditions, health benefits are increasingly measured in improvements in functional status or quality of life, rather than simply using mortality rates or life expectancy. A third trend relates directly to how the increasing costs of health care are now threatening public budgets and investments in other social goals, such as education. Although the United States spends more per capita on health care than any other developed nation (Chapter 5), the outcomes achieved lag far behind. Finally, advances in communication and information technologies have inspired more people to play an active role in their health and health care. These innovations have accelerated demands for transparency and shared decision making. As health insurance and health care regulation have expanded, requirements to track and justify health care services have grown. Intensifying urgency to improve the quality of health care, reduce disparities, control costs, and enhance transparency will likely lead patients and insurers to demand more data and to link quality measures to payments for services. Fortunately, modern technology can help to meet the demand for data. Patients can record and submit their health parameters using hand-held devices connected to personal health records. Automated billing programs can track health care services, while electronic health records can assess the quality of physician care. Ultimately, fully integrated health information systems will allow patient information to be retrieved instantly and seamlessly wherever and whenever it is needed. In addition to assessing care quality today, these tools offer enormous promise for learning as a byproduct of care delivery.

HOW ARE HEALTH AND HEALTH CARE MEASURED?

11  MEASURING HEALTH AND HEALTH CARE CAROLYN M. CLANCY AND ERNEST MOY Physicians routinely quantify a variety of health measures, including symptoms, vital signs, and findings on physical examination, to improve diagnosis, treatment, and prognostication. Similarly, the efficacy and quality of health care also can and should be measured for several reasons. First, the quality of care delivered is often suboptimal.1 Persistent variations in practice for patients with the same diagnosis reflect a combination of clinical uncertainty, individualized practice styles, patients’ preferences and characteristics (age, race, ethnicity, education, income), and other factors. Both suboptimal care and varied care for the same condition undermine the historical assumption that a combination of highly trained health

Three types of measures typically assess health and health care. Measures of health quantify the sickness or well-being of a person. Measures of health care quality quantify the extent to which a patient receives needed care and does not receive unnecessary care. Health care quality is assessed using measures of structure (e.g., education and credentialing of clinicians), process (adherence to professional standards and evidence-based recommendations), and outcomes (or end results of care, including how patients experience their care and their self-reported health and function). Measures of health care resources quantify the resources used (e.g., radiographs, surgery, medication, intensive care) to improve the health of a patient.2 All measures can be summed up across populations within a practice or community (Table 11-1). Measures of health and health care often overlap (E-Fig. 11-1). Health measures that can be improved by health care, such as blood pressure or blood glucose levels, are often used as health care quality outcome measures. The delivery of quality health care requires the use of resources and the generation of direct health care costs, which may or may not improve health care at the margins of spending. Impaired health that reduces the ability to do work and earn wages but that could have been prevented by the delivery of health care contributes to the indirect costs of health care. At the intersection of health, health care quality, and health care resources are measures of health care value. These measures compare the health benefits of specific health care services with their costs.2

WHERE DO MEASURES OF HEALTH AND HEALTH CARE COME FROM?

Most researchers, provider groups, insurers, regulators, and credentialing organizations that develop measures consult with physicians to ensure that their metrics are consistent with professional standards. Insurers may create measures to allocate health care resources, to plan for future needs, and to identify efficient physicians for inclusion on panels or to be rewarded with performance bonuses. Regulators may develop measures to establish licensure requirements and to identify physicians who might benefit from remedial instruction. Credentialing organizations may construct measures to demonstrate the superior performance of physicians who meet their high standards. For example, the National Committee for Quality Assurance maintains the Healthcare Effectiveness Data and Information Set that is widely used to accredit health plans, and the American Board of Internal Medicine and other specialty boards include measures of practice performance as well as measures of medical knowledge for the maintenance of certification.

CHAPTER 11  Measuring Health and Health Care  

Health

Health care outcomes

Indirect cost Health care value

Health care quality

Direct cost

Health care resources

E-FIGURE 11-1.  Type of measures of health and health care.

41.e3

42

CHAPTER 11  Measuring Health and Health Care  

TABLE 11-1  MEASURES OF HEALTH AND HEALTH CARE MEASURES OF HEALTH • Mortality: rates of death typically adjusted for age and sex • Morbidity: incidence and prevalence rates of diseases and their sequelae • Functional status: assessments of a patient’s ability to perform various actions such as activities of daily living or instrumental activities of daily living as observed by a provider or reported by the patient • Self-reported health status: a patients’ assessment of their health and well-being MEASURES OF HEALTH CARE QUALITY • Health care outcomes: the end results or health benefits derived from good health care or the health loss attributable to poor health care • Health care processes: assessments of whether the right care was delivered at the right time and in the right way • Health care infrastructure: the availability of resources needed to deliver good health care • Patient perceptions of health care: a patient’s assessment of health care received, usually emphasizing patient-provider communication and shared decision making • Access to health care: the ability of patients to gain entry into health care and navigate to needed resources MEASURES OF HEALTH CARE RESOURCES • Health care utilization: the quantity of health care services that are used • Direct costs: the costs of providers, supplies, and equipment needed to deliver health care • Indirect costs: the costs of lost wages and decreased productivity due to illness or injury that could have been prevented by appropriate health care • Nonmedical costs: the costs of health care not related to the delivery of services, such as administration, advertising, research, and profits earned by health industries TYPE OF MEASURE

EXAMPLE

HEALTH Mortality

Deaths due to colorectal cancer per 100,000 population

Morbidity

New AIDS cases per 100,000 population

Functional status

% of people unable to perform one or more activities of daily living

Self-reported health status

% of people reporting that their overall health is excellent

HEALTH CARE QUALITY Health care outcomes

Death per 1000 hospitalizations with pneumonia

Intermediate outcomes

% of adults with diabetes whose blood pressure is 5 L), whereas glimepiride has a relatively small VD (0.18 L). As discussed later, VD is a useful pharmaco*kinetic tool for calculating the loading dose and appreciating how various changes can affect a drug’s half-life.

Absorption

Elimination

Absorption refers to the transfer of a drug from the site of administration to the systemic circulation. Many drugs cross a membrane barrier by passive diffusion and enter the systemic circulation. Because passive diffusion in this setting depends on the concentration of the solute at the membrane surface, the rate of drug absorption is affected by the concentration of free drug at the absorbing surface. Factors that influence the availability of free drug thus affect drug absorption from the administration site; this effect can be exploited to design medications that release a drug slowly into the circulation by prolonging drug absorption. With certain sustained-released oral prepa­ rations, the rate of dissolution of the drug in the gastrointestinal tract determines the rate at which the drug is absorbed (e.g., timed-release antihistamines). Similarly, a prolonged drug effect can be obtained by the use of transdermal medications (e.g., nitroglycerin) or intramuscular depot preparations (e.g., benzathine penicillin G).

First-Pass Effect

Some drugs that are administered orally are absorbed relatively efficiently into the portal circulation but are metabolized by the liver before they reach the systemic circulation. Because of this “first-pass” or “presystemic” effect, the oral route may be less suitable than other routes of administration for such drugs. A good example is nitroglycerin, which is well absorbed but efficiently metabolized during the first pass through the liver. However, the same drug can achieve adequate systemic levels when it is given sublingually or transdermally.

Bioavailability

The extent of absorption of a drug into the systemic circulation may be incomplete. The bioavailability of a particular drug is the fraction (F) of the total drug dose that ultimately reaches the systemic circulation from the site of administration. This fraction is calculated by dividing the amount of the drug dose that reaches the circulation from the administration site by the amount of the drug dose that would enter the systemic circulation after direct intravenous injection into the circulation (essentially the total dose). Bioavailability, or F, can range from 0, in which no drug reaches the systemic

Drugs are removed from the body by two major mechanisms: hepatic elimination, in which drugs are metabolized in the liver and excreted through the biliary tract; and renal elimination, in which drugs are removed from the circulation by either glomerular filtration or tubular secretion. For most drugs, the rates of hepatic and renal elimination are proportional to the plasma concentration of the drug. This relationship is often described as a “first-order” process. Two measurements, clearance and half-life, are used to evaluate elimination.

Clearance

The efficiency of elimination can be assessed by quantifying how fast the drug is cleared from the circulation. Drug clearance is a measure of the volume of plasma cleared of drug per unit of time. It is similar to the clinical measurement used to assess renal function—creatinine clearance, which is the volume of plasma from which creatinine is removed per minute. Total drug clearance (Cltot) is the rate of elimination by all processes (Eltot) divided by the plasma concentration of the drug (Cp):

Cl tot = El tot /C p

(2)

Drugs may be cleared by several organs, but as noted earlier, renal clearance and hepatic clearance are the two major mechanisms. Total drug clearance (Cltot) can best be described as the sum of clearances by each organ. For most drugs, this is essentially the sum of renal clearance and hepatic clearance:

Cl tot = El Ren + Cl Hep

(3)

Table 29-1 shows the wide variation in clearance values among commonly used medications; some drugs (e.g., phenobarbital) have relatively low clearances (500 mL/minute). Tobramycin is cleared almost entirely by the kidneys, whereas aspirin, carbamazepine, and phenytoin are cleared less than 5% by the kidneys. Drug clearance is affected by several factors, including blood flow through the organ of clearance, protein binding to the drug, and activity of the clearance processes in the organs of elimination (e.g., glomerular filtration rate and

CHAPTER 29  Principles of Drug Therapy  

125

Site of drug administration Absorption Distribution

Bound drug

Distribution Systemic circulation Free drug

Sites of action and/or toxicity

Effect and/or toxicity

Metabolism Metabolites Distribution Excretion

Urine, bile, feces, and other routes Pharmaco*kinetics

Pharmacodynamics

FIGURE 29-1.  Schematic of a drug’s movement through the body, from the site of administration to production of a drug effect. The relationship between pharmaco*kinetics and

pharmacodynamics is shown.

The time needed to eliminate the drug is best described by its half-life (t1/2), which is the time required during the elimination phase (see Fig. 29-2) for the plasma concentration of the drug to be decreased by half. Mathematically, the half-life is equal to the natural logarithm of 2 (representing a reduction of drug concentration to half) divided by Ke. Substituting for Ke from Equation 4 and calculating the natural logarithm of 2, the half-life can be represented by the following equation:

Log concentration of drug X in plasma (mg/L)

100

Distribution phase 10 Cp0

Ca

Elimination phase

t1/2

1

0.1 0

30

60

90

t 1 2 = 0.693 VD/Cl

1/2 C a

120 150 180 210 240 270 300

Time after dosing (min) FIGURE 29-2.  Representative drug concentration versus time plot used in pharmaco*kinetic studies. Concentration of drug is plotted with a logarithmic scale on the ordinate, and time is plotted with a linear scale on the abscissa. The resultant curve has two phases: the distribution phase, which is the initial portion of the plotted line when the concentration of drug decreases rapidly; and the later elimination phase, during which there is an exponential disappearance of drug from the plasma over time. The dotted line extrapolated from the elimination phase back to time zero is used to calculate plasma concentration at time zero (Cp0). During the elimination phase, the half-life (t½) can be calculated as the time it takes to decrease the concentration by half (shown here as the time needed to decrease from concentration Ca to ½ Ca).

(5)

From this equation, one can predict that at a given clearance, as the VD increases, the half-life increases. Similarly, at a given VD, as the clearance increases, the half-life decreases. Clinically, many disease states (see later) can affect VD and clearance. Because disease affects VD and clearance differently, the half-life may increase, decrease, or not change much at all. Therefore, the half-life by itself is not a good indicator of the extent of abnormality in elimination. The half-life is useful to predict how long it takes for a drug to be eliminated from the body. For any drug that has a first-order elimination, one would expect that by the end of the first half-life, the drug would be reduced to 50%; by the end of the second half-life, to 25%; by the end of the third half-life, to 12.5%; by the end of the fourth half-life, to 6.25%; and by the end of the fifth half-life, to 3.125%. In general, a drug can be considered essentially eliminated after three to five half-lives, when less than 10% of the effective concentration remains. Table 29-1 shows the wide variation in half-life for several commonly used drugs.

CLINICAL APPLICATION OF PHARMAco*kINETIC PRINCIPLES

Using a Loading Dose tubular secretion in the kidney, enzyme activity in the liver). Drug clearance is not affected by the distribution of drug throughout the body (VD) because clearance mechanisms act only on drug in the circulation.

Half-Life

The amount of time needed to eliminate a drug from the body depends on the clearance and the VD. The first-order elimination constant (Ke) represents the proportion of the apparent VD that is cleared of drug per unit of time during the drug’s exponential disappearance from the plasma over time (elimination phase):

K e = Cl / VD

(4)

The value of this constant for a particular drug can be determined by plotting drug concentration versus time on a log-linear plot (see Fig. 29-2) and measuring the slope of the straight line obtained during the exponential (elimination) phase.

To attain a desired therapeutic concentration rapidly, a loading dose is often used. In determining the amount of drug to be given, the clinician must consider the “volume” within the body into which the drug will be distributed. This volume is best described by the apparent VD. The loading dose can be calculated by multiplying the desired concentration by the VD:

Loading dose = desired concentration × VD

(6)

Rapid administration of the entire loading dose may produce an initially high peak concentration that results in toxicity. This problem can be avoided either by administering the loading dose as a divided dose or by varying the rate of access to the circulation, such as by administering the drug as an infusion (with an intravenous drug) or by taking advantage of the slower access to the circulation from various other routes (e.g., oral dosing). This approach is illustrated by phenytoin (see Table 29-1), which may need to be administered with a loading dose to achieve a therapeutic level (10 to 20 mg/L) rapidly. Because the VD for phenytoin is approximately 0.6 L/kg, the loading dose calculated from Equation 6 is 420 mg/L to attain a minimally

126

CHAPTER 29  Principles of Drug Therapy  

TABLE 29-1  PHARMAco*kINETIC PARAMETERS FOR SOME COMMONLY USED DRUGS DRUG Amoxicillin

VD (L/kg)

PROTEIN BINDING (%)

TOTAL CLEARANCE (mL/min)

% OF TOTAL CLEARANCE AS RENAL CLEARANCE

HALF-LIFE (hr)

THERAPEUTIC RANGE (mg/L)

86

1.2

2-8

99.5

0.62 ± 0.26

25 to 75%) in daily smoking consumption, there is little if any decrease in cardiovascular disease and lung or other smoking-related cancer risk, further substantiating the merits of quitting versus reducing smoking.

PATHOBIOLOGY

Nicotine is the primary reinforcer in tobacco smoke, with contributions from more than 4000 components to the sensory (non-nicotine) aspects of cigarette smoking. The primary site of action of nicotine is the α4β2 nicotinic acetylcholine receptor (nAChR), and the endogenous neurotransmitter acting on nAChRs is acetylcholine. nAChRs in the central nervous system (CNS) are pentameric ion channel complexes comprising two α- and three β-subunits; the seven α-subunits are designated α2 to α9 and the three β-subunits are designated β2 to β4. This produces considerable diversity in subunit combinations, which may explain the region-specific and functional selectivity of nicotinic effects in the CNS.3 Activation of nAChRs leads to Na+/Ca2+ ion channel fluxes and neuronal membrane depolarization. nAChRs are located presynaptically on several neurotransmitter-secreting neuron types in the CNS, including mesolimbic dopaminergic (DA) neurons that project from the ventral tegmental area (VTA) to the nucleus accumbens (NAc). Activation of nAChRs on mesolimbic DA neurons leads to DA secretion in the nucleus accumbens. At low concentrations of nicotine, α4β2 nAChR stimulation of afferent GABAergic projections onto mesoaccumbal DA neurons predominates, leading to reduced mesolimbic DA neuron firing and DA release. At higher nicotine concentrations, α4β2 nAChRs desensitize, and predominant activation of α7 nAChRs on glutamatergic projections occurs, leading to increased mesolimbic DA neuron firing and release. Within milliseconds of activation by nicotine, nAChRs desensitize. After overnight abstinence, nAChRs resensitize; this may explain why most smokers report that the first cigarette in the morning is the most satisfying. Interestingly, positron emission tomography (PET) neuroimaging studies have shown that smoking 2 or 3 puffs from a cigarette produces saturation of nAChRs in the brain reward system, suggesting that although binding to central nAChRs is an important first step in the effects of nicotine, it is not a complete explanation for continued smoking behaviors.

tar. Besides positive reinforcement, withdrawal, and craving, there are several secondary effects of nicotine and tobacco use that may contribute to both maintenance of smoking and smoking relapse, including mood modulation (e.g., reduction of negative affect), stress reduction, and weight control. In addition, conditioned cues can elicit the urge to smoke even after prolonged periods of abstinence. Specific effects might be most relevant to smokers wishing to lose weight and to those with psychiatric presentations (mood modulation, cognitive enhancement, stress reduction). These secondary effects may present additional targets for pharmacologic intervention in certain subgroups of smokers (e.g., those with schizophrenia or depression, or those concerned about their weight).

DIAGNOSIS

The Diagnostic and Statistical Manual, 5th edition (DSM-5),4 which was released in 2013 by the American Psychiatric Association, has changed the diagnostic terminology for nicotine and tobacco, eliminating the term dependence and instead using the term tobacco use disorder. Tobacco use disorder is established clinically by historical documentation of 2 of the following 11 criteria: 1. Tobacco often taken in larger amounts or over a longer period than was intended 2. Persistent desire or unsuccessful efforts to cut down or control tobacco use 3. A great deal of time spent in activities necessary to obtain or use tobacco 4. Presence of craving, or a strong desire or urge to use tobacco 5. Recurrent tobacco use resulting in failure to fulfill major obligations at work, school, or home 6. Continued tobacco use despite persistent or recurrent social or interpersonal problems caused or exacerbated by the effects of tobacco 7. Important social, occupational, or recreational activities given up or reduced because of tobacco use 8. Recurrent tobacco use in situations in which it is physically hazardous (e.g., smoking in bed) 9. Continued tobacco use despite persistent or recurrent physical or psychological problems that are caused or exacerbated by tobacco use 10. Tolerance, as defined by either a need for markedly increased amounts of tobacco to achieve desired effects, or markedly diminished effects with continued use of the same amount of tobacco 11. Withdrawal, manifested by the presence of the characteristic tobacco abstinence syndrome (e.g., four of the following: irritability, anxiety, difficulty concentrating, increased appetite, restlessness, dysphoric mood, insomnia), or tobacco (or nicotine) taken to relieve or avoid tobacco withdrawal symptoms. For abstinent smokers, remission is classified as early (between 3 and 12 months of abstinence) or sustained (>12 months of abstinence). Moreover, current severity of tobacco use disorder is coded as mild (2 or 3 symptoms), moderate (4 or 5 symptoms) or severe (6 or more symptoms). In addition, most physiologically dependent tobacco smokers state that they smoke their first cigarette of the day within the first 5 minutes of awakening (e.g., time to first cigarette 4 drinks/day Women: >7 drinks/week or >3 drinks/day ALCOHOL USE DISORDER CRITERIA* Tolerance Withdrawal More use than intended Craving Unsuccessful attempts to cut down Excessive time acquiring alcohol Activities given up because of use Use despite negative effects Failure to fulfill major role obligations Recurrent use in hazardous situations Continued use despite social or intrapersonal problems

NADH + H+

Alcohol dehydrogenase

NADH NAD+ CH3COH CH3COO– Acetaldehyde Acetate Aldehyde Microsomal ethanol dehydrogenase oxidizing system

CH3CH2OH Ethanol

NADPH + H+ + 2O2

NADP + H2O

FIGURE 33-1.  Ethanol metabolism. Alcohol dehydrogenase predominates at low to moderate ethanol doses. The microsomal ethanol-oxidizing system is induced at high ethanol levels of chronic exposure and by certain drugs. Aldehyde dehydrogenase inhibition (genetic or drug induced) leads to acetaldehyde accumulation, particularly in the latter group.

*Mild = 2-3 criteria, moderate = 4-5 criteria, severe = 6 or more criteria. (From American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Publishing; 2013.)

PATHOBIOLOGY

At-risk drinking is defined differently for men younger than 65 years than for women of all ages because of generally lower body weights and lower rates of metabolism of alcohol in women; the definition in men older than 65 years is the same as in women because of the age-related increased risk for alcohol problems, in part owing to changes in alcohol metabolism in older individuals. Binge drinking or heavy drinking is the episodic consumption of large amounts of alcohol, usually five or more drinks per occasion for men and four or more drinks per occasion for women. One standard drink contains 12 g of pure alcohol, an amount equivalent to that contained in 5 ounces of wine, 12 ounces of beer, or 1.5 ounces of 80-proof spirits. The recently published Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) replaced the previous terminology of alcohol abuse and alcohol dependence with the term alcohol use disorders (see Table 33-1) in order to more clearly describe the spectrum of symptoms experienced by patients. Patients who meet 2 or 3 criteria are considered to have mild, 4 or 5 criteria moderate, and 6 to 11 criteria severe alcohol use disorder.1

EPIDEMIOLOGY

In national surveys, 52% of American adults reported that they use alcoholic beverages (liquor, wine, or beer), whereas 23% reported binge drinking, and 6.5% reported heavy drinking in the past 30 days.2 Among individuals who use alcohol, many experience problems because of their drinking. It has been estimated that more than $100 billion is spent by American society each year to treat alcohol use disorders and to recover the costs of alcohol-related economic losses. Excessive alcohol consumption ranks as the third leading preventable cause of death in the United States after cigarette smoking and obesity. More than 100,000 deaths per year in the United States are attributed to alcohol use disorders. Population-based epidemiologic studies have shown that alcohol use disorders are among the most prevalent medical, behavioral, or psychiatric disorders in the general population. An epidemiologic survey of the general population in the United States estimated a prevalence of alcohol abuse and dependence (using the older DSM-IV criteria) to be between 7.4 and 9.7%. The lifetime prevalence of abuse and dependence is estimated to be even higher. Despite higher thresholds and tolerance, men are at least twice as likely as women to meet criteria for alcohol abuse and dependence by standard diagnostic survey techniques. Although sociodemographic features, such as young age, low income, and low education level, have been associated with an increased risk for problem drinking, alcohol use disorders are prevalent throughout all sociodemographic groups, and all individuals should be screened carefully. The “skid row” stereotype of the alcohol-dependent patient is much more the exception than the rule. The prevalence of alcohol use disorders is higher in most health care settings than it is in the general population because alcohol problems often result in treatment-seeking behaviors. The prevalence of problem drinking in general outpatient and inpatient medical settings has been estimated between 15 and 40%. These data strongly support the need for physicians to screen all patients for alcohol use disorders.

Beverage alcohol contains ethanol, which acts as a sedative-hypnotic drug. Alcohol is absorbed rapidly into the blood stream from the stomach and intestinal tract. Because women have lower levels of gastric alcohol dehydrogenase, the enzyme primarily responsible for metabolizing alcohol, they experience higher blood alcohol concentrations than do men who consume similar amounts of ethanol per kilogram of body weight. The absorption of alcohol can be affected by other factors, including the presence of food in the stomach and the rate of alcohol consumption. By means of metabolism in the liver, alcohol is converted to acetaldehyde and acetate (Fig. 33-1). Metabolism is proportional to an individual’s body weight, but a variety of other factors can affect how alcohol is metabolized. A genetic variation in a significant proportion of the Asian population alters the structure of an aldehyde hydrogenase isoenzyme, resulting in the development of an alcohol flush reaction, which includes facial flushing, hot sensations, tachycardia, and hypotension. In the brain, alcohol seems to affect a variety of receptors, including γ-aminobutyric acid (GABA), N-methyl-d-aspartate, and opioid receptors. Glycinuric and serotoninergic receptors also are thought to be involved in the interaction between alcohol and the brain. The phenomena of reinforcement and cellular adaptation are thought, at least in part, to influence alcoholdependent behaviors. Alcohol is known to be reinforcing because withdrawal from ethanol and ingestion of ethanol itself are known to promote further alcohol consumption. After chronic exposure to alcohol, some brain neurons seem to adapt to this exposure by adjusting their response to normal stimuli. This adaptation is thought to be responsible for the phenomenon of tolerance, whereby increasing amounts of alcohol are needed over time to achieve desired effects. Although much has been learned about the variety of effects alcohol can have on various brain receptors, no single receptor site has been identified. A variety of neuropsychological disorders are seen in association with chronic ethanol use, including impaired short-term memory, cognitive dysfunction, and perceptual difficulties. Although the brain is the primary target of alcohol’s actions, a variety of other tissues have a major role in how alcohol affects the human body. Direct liver toxicity may be among the most important consequences of acute and chronic alcohol use (Chapter 152). A variety of histologic abnormalities ranging from inflammation to scarring and cirrhosis have been described. The pathophysiologic mechanism of these effects is thought to include the direct release of toxins and the formation of free radicals, which can interact negatively with liver proteins, lipids, and DNA. Alcohol also has substantial negative effects on the heart and cardiovascular system. Direct toxicity to myocardial cells frequently results in heart failure (Chapter 58), and chronic heavy alcohol consumption is considered to be a major contributor to hypertension (Chapter 67). Other organ systems that experience significant direct toxicity from alcohol include the gastrointestinal tract (esophagus, stomach), immune system (bone marrow, immune cell function), and endocrine system (pancreas, gonads).

CLINICAL MANIFESTATIONS

Alcohol has a variety of specific acute and chronic effects. The acute effects seen most commonly are alcohol intoxication and alcohol withdrawal. Chronic clinical effects of alcohol include almost every organ system.

CHAPTER 33  Alcohol Use Disorders  

Acute Effects Alcohol Intoxication

After entering the blood stream, alcohol rapidly passes through the bloodbrain barrier. The clinical manifestations of alcohol intoxication are related directly to the blood level of alcohol. Because of tolerance, individuals chronically exposed to alcohol generally experience less severe effects at a given blood alcohol level than do individuals who are not chronically exposed to alcohol. The symptoms of mild alcohol intoxication in nontolerant individuals typically occur at blood alcohol levels of 20 to 100 mg/dL and include euphoria, mild muscle incoordination, and mild cognitive impairment. At higher blood alcohol levels (100 to 200 mg/dL), more substantial neurologic dysfunction occurs, including more severe mental impairment, ataxia, and prolonged reaction time. Individuals with blood alcohol levels in these ranges can be obviously intoxicated, with slurred speech and lack of coordination. These effects progress as the blood alcohol level rises to higher levels, to the point at which stupor, coma, and death can occur at levels equal to or greater than 300 to 400 mg/dL, especially in individuals who are not tolerant to the effects of alcohol. The usual cause of death in individuals with very high blood levels of alcohol is respiratory depression and hypotension.

Alcohol Withdrawal Syndrome

Alcohol withdrawal can occur when individuals decrease their alcohol use or stop using alcohol altogether. The severity of symptoms can vary greatly. Many individuals experience alcohol withdrawal without seeking medical attention, whereas others require hospitalization for severe illness. Because ethanol is a central nervous system depressant, the body’s natural response to withdrawal of the substance is a hyperexcitable neurologic state. This state is thought to be the result of adaptive neurologic mechanisms being unrestrained by alcohol, with an ensuing release of a variety of neurohumoral substances, including norepinephrine. In addition, chronic exposure to alcohol results in a decrease in the number of GABA receptors and impairs their function. The clinical manifestations of alcohol withdrawal include hyperactivity resulting in tachycardia and diaphoresis. Patients also experience tremulousness, anxiety, and insomnia. More severe alcohol withdrawal can result in nausea and vomiting, which can exacerbate metabolic disturbances. Perceptual abnormalities, including visual and auditory hallucinations and psychom*otor agitation, are common manifestations of more moderate to severe alcohol withdrawal. Grand mal seizures commonly occur during alcohol withdrawal, although they do not generally require treatment beyond the acute withdrawal phase. The time course of the alcohol withdrawal syndrome can vary within an individual and by symptom complex, and the overall duration of symptoms can be a few to several days (Fig. 33-2). Tremor is typically among the earliest symptoms and can occur within 8 hours of the last drink. Symptoms of tremulousness and motor hyperactivity typically peak within 24 to 48 hours. Although mild tremor typically involves the hands, more severe tremors can involve the entire body and greatly impair a variety of basic motor functions. Perceptual abnormalities typically begin within 24 to 36 hours after the last drink and resolve within a few days. When withdrawal seizures occur, they are typically generalized tonic-clonic seizures and most often occur within 12 to 24 hours after reduction of alcohol intake. Seizures can occur, however, at later time periods as well.

Tremor, anxiety, insomnia

The most severe manifestation of the alcohol withdrawal syndrome is delirium tremens. This symptom complex includes disorientation, confusion, hallucination, diaphoresis, fever, and tachycardia. Delirium tremens typically begins after 2 to 4 days of abstinence, and the most severe form can result in death.3

Chronic Effects

Acute manifestations, including intoxication and withdrawal, are generally stereotypical in their appearance and time course, but chronic manifestations tend to be more varied. Many patients with alcohol dependence may be without evidence of any chronic medical manifestations for many years. As time goes on, however, the likelihood that one or more of these manifestations will occur increases considerably. All major organ systems can be affected, but the primary organ systems involved are the nervous system, cardiovascular system, liver, gastrointestinal system, pancreas, hematopoietic system, and endocrine system (Table 33-2). Patients who drink are at risk for a variety of malignant neoplasms, such as head and neck, esophageal, colorectal, breast, and liver cancers (see individual chapters on those cancers). Excessive alcohol use often causes significant psychiatric and social morbidity that can be more common and more severe than the direct medical effects, especially earlier in the course of problem drinking.

Nervous System

In addition to the acute neurologic manifestations of intoxication and withdrawal, alcohol has major chronic neurologic effects. About 10 million Americans have identifiable nervous system impairment from chronic alcohol use. Individual predisposition to these disorders is highly variable and is related to genetics, environment, sociodemographic features, and gender; the relative contribution of these factors is unclear. In the central nervous system, the major effect is cognitive impairment. Patients may present with mild to moderate short-term or long-term memory problems or may have severe dementia resembling Alzheimer disease (Chapter 402). The degree to which the direct toxic effect of alcohol is responsible for these problems or the impact of alcohol-related nutritional

TABLE 33-2  ALCOHOL-RELATED COMPLICATIONS SYSTEM/REALM OF PROBLEM

Delirium tremens 0

1

2

3

4

5

Days since last drink FIGURE 33-2.  Time course of alcohol withdrawal.

6

7

14

COMPLICATIONS

Nervous system

Intoxication Withdrawal Cognitive impairment Cerebellar degeneration Peripheral neuropathy

Cardiovascular system

Cardiac arrhythmias Chronic cardiomyopathy Hypertension

Liver

Fatty liver Alcoholic hepatitis Cirrhosis

Gastrointestinal tract, esophagus

Chronic inflammation Malignant neoplasms Mallory-Weiss tears Esophageal varices

Stomach

Gastritis Peptic ulcer disease

Pancreas

Acute pancreatitis Chronic pancreatitis

Other medical problems

Cancers: mouth, oropharynx, esophagus, colorectal, breast, hepatocellular carcinoma Pneumonia Tuberculosis

Psychiatric

Depression Anxiety Suicide

Behavioral and psychosocial

Injuries Violence Crime Child or partner abuse Tobacco, other drug abuse Unemployment Legal problems

Hallucinations

Seizures

151

152

CHAPTER 33  Alcohol Use Disorders  

deficiencies is uncertain (Chapter 416). The deficiency of vitamins such as thiamine may plays a major role in promoting alcoholic dementia and severe cognitive dysfunction, as is seen in Wernicke encephalopathy and Korsakoff syndrome (Chapter 416). Alcohol also causes a polyneuropathy that can present with paresthesias, numbness, weakness, and chronic pain (Chapters 416 and 420). As with the central nervous system, peripheral nervous system effects are thought to be caused by a combination of the direct toxicity of alcohol and nutritional deficiencies. A small proportion (1%

75,000

Fraction of all human heterozygosity attributable to variants with a frequency of >1%

98%

contains essentially all common sequence variants in the human population (with frequency >1%). At the time of this writing, this public database contains more than 44 million human genetic variants (www.ncbi.nlm.nih.gov:80/ SNP/index.html). Not all these entries represent common variants (some are rare), and a small fraction may represent technical false-positive findings. The major contribution of common variation in human sequence diversity is explained by the unique demographic history of the human population. Despite the global distribution of the current human population, it is now clear that all humans are the descendants of a single population that lived in Africa only 10,000 to 40,000 years ago. The ancestral population was small (with an effective size of perhaps 10,000 individuals), lived a hunter-gatherer existence at low population densities (relative to other humans and later domesticated animals), and had evolved in Africa during millions of years. Most human genetic variation arose in this phase of human history, before the more recent migrations, expansions, and invention of technologies (e.g., farming) that resulted in widespread population of the globe. Most common human genetic variation predates the Diaspora and is shared by all populations on earth. A second factor is the slow rate of change in human DNA. Mutation and recombination occur at very low rates, on the order of 10−8 per base pair per generation; and yet, any pair of human genes traces a lineage back to a shared ancestor who lived on the order of 103 to 104 generations ago (if a generation is 20 years, then 104 generations is 200,000 years). In other words, considering the typical nucleotide in two unrelated humans, it is more likely that they trace back to a shared ancestor without any mutation having occurred than it is that a mutation has arisen in the intervening time. This explains why 99.9% of base pairs are identical when any two copies of the human genome are compared. Another aspect of human variation is explained by these simple mathematical and population genetic relationships: the extent of human DNA sequence diversity attributable to rare and common variants. Each of us inherits from our parents some 3 million common polymorphisms (classically defined as those with frequency of >1%). We inherit genetic variants that are shared by apparently unrelated individuals but are at frequencies less than 1%, and we inherit thousands of variants that are limited only to the individual and the individual’s closest relatives. The shared ancestry of human populations explains another aspect of human genetic variation: the correlations among nearby variants known as linkage disequilibrium. Empirically, individuals who carry a particular common variant at one site in the genome are observed to be more likely than chance to carry a particular set of variants at nearby positions along the chromosome. That is, not all combinations of nearby variants are observed in the population but rather only a small subset of the possible combinations. These correlations reflect the fact that most variants in our genomes arose once in human history (typically long ago) and did so on an arbitrary but unique copy carried by the individual in whom the mutation first arose. This unique ancestral copy can be recognized in the current population by the stretch of

197

particular alleles (known as a haplotype). These ancestral haplotypes, passed down from shared prehistoric ancestors in Africa, offer a practical tool in association studies of human disease because it is not necessary to measure directly each nucleotide to capture much of the information.

THE SEARCH FOR GENES UNDERLYING MONOGENIC DISEASES

The genetic architecture of a disease refers to the number and magnitude of genetic risk factors that exist in each patient and their frequencies and interactions in the population. Diseases can be due to a single gene in each family (monogenic) or to multiple genes (polygenic). It is easiest to identify genetic risk factors when only a single gene is involved and this gene has a large impact on disease in that family. In cases in which a single gene is necessary and sufficient to cause disease, the condition is termed a mendelian disorder because the disease tracks perfectly with a mutation (in the family) that obeys Mendel’s simple laws of inheritance. Some single-gene disorders are caused by the same gene in all affected families; for example, cystic fibrosis is always caused by mutations in CFTR. Although many individuals with cystic fibrosis carry the same founder mutation (δ-508), others carry any pair of a wide variety of different mutations in CFTR. The existence of many different mutations at a given disease gene is known as allelic heterogeneity. A mendelian disorder can be due to a single genetic lesion in any given family but in different families can be due to mutations in a variety of genes. This phenomenon, termed locus heterogeneity, is illustrated by retinitis pigmentosa. Although mutation in a single gene is typically necessary and sufficient to cause retinitis pigmentosa, there are dozens of different genes in which retinitis pigmentosa mutations have been found (Online Mendelian Inheritance in Man #268000). In each family, however, only one such gene is mutated to cause disease. Most single-gene disorders are rare (present in 1%) B. Rare mutations (500 conditions

PharmGKB

http://www.pharmgkb.org

Information on potentially clinically actionable gene-drug associations and genotypephenotype relationships; currently lists 186 well-known pharmacogenomic associations and provides 46 summaries for very important genes

CHAPTER 43  Application Of Molecular Technologies to Clinical Medicine  

REVIEW QUESTIONS 1. Genome sequencing of DNA can be used for all of the following except which one? A. Diagnosis of rare diseases B. Detection of pharmacogenetic variants C. Newborn screening D. Prognosis of cancer E. Diagnosis of infection Answer: D  Various types of genome sequencing have served special clinical functions. Exome sequencing is being increasingly used to diagnose rare genetic diseases and patients with diagnostic dilemmas. Germline genotyping has allowed prediction of adverse events and dosing of many commonly used drugs; although these discoveries have been “actionable,” most have failed to diffuse widely into clinical practice at this time. Screening newborns, prenatal diagnosis, and preconception carrier testing are feasible for identified mendelian disorders. In infectious diseases, diagnosis of causative pathogens can be performed by next-generation sequencing, which may in the future supplant the need to first grow microorganisms in culture. Genome sequencing has not been demonstrated to be useful for prognosis of cancer at this time. 2. Transcriptional (RNA) profiling may be useful for all of the following except which one? A. Susceptibility B. Diagnosis C. Prognosis D. Pharmacogenetics E. Monitoring Answer: A  Next-generation sequencing has allowed sequencing of RNA transcripts in cells. The so-called transcriptome, unlike the static genome, is continually changing. It allows examination, at any given time, of alternatively gene-spliced transcripts, post-transcriptional changes, gene fusion, and changes in gene expression. Transcriptional (RNA) profiling has been applied to diagnosis, prognosis, pharmacogenomics, and monitoring, but not susceptibility analysis.

203.e3

3. Among the great challenges to implementing genomic diagnostics in the clinic is which one of the following? A. The precision of the results B. The evidence to support use C. The ability of the laboratories to perform the test D. Their integration into electronic medical records E. Lack of patient understanding Answer: B  Surmountable challenges to implementing genomic diagnostics at the bedside include increasing the precision of results, expanding the availability of expert laboratories to perform the tests, the ability to integrate the information into electronic medical records, and lack of understanding by patients who are, however, now becoming increasingly savvy about this aspect of their own health care. Evidence to support the utility and costeffectiveness of wide application of genomic diagnostics in clinical practice is only now being taken up by investigators in comparative effectiveness and health services research. 4. Which of the following describes the microbiome? A. A small genome B. A community of microbes colonizing humans C. A tool used to make thinly sliced materials (e.g., paraffin blocks) D. The sequence of a virus or bacterium E. An organelle of the cell Answer: B  The microbiome is the community of microbes that colonize a human host. It is the ecological community of commensal, symbiotic, and pathogenic microorganisms that actually inhabit our body space and outnumber our own native human cells by a ratio of 10 : 1. 5. Which of the following methods may be used for detection of microbial pathogens? A. Transcriptional profiling B. Metabolomics C. Sequencing D. DNA methylation E. Proteomics Answer: C  At this time, sequencing is used to detect and identify a microbial pathogen.

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

203

44  REGENERATIVE MEDICINE, CELL, AND GENE THERAPIES LIOR GEPSTEIN AND KARL SKORECKI

CELL THERAPY

Introduction and Definitions

A remarkable clinical need exists for the development and clinical assessment of various methods to facilitate the regeneration of injured or diseased tissues and organs. This need derives from the unrelenting prevalence of trauma, congenital disorders, ischemia, and degenerative processes, which becomes increasingly urgent as the global population expands and ages. Cancer is tied to this field both directly (e.g., replacement of lost vital organ function as a result of cancer invasion or treatment modalities, cell-based delivery of cancer immune and gene therapies) and indirectly (e.g., role of stem cells in cancer

204

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

pathogenesis, risk for tumorigenesis in stem cell−based therapies). The recent developments in stem cell biology, molecular interventions, biopolymers, and other related biologic and engineering disciplines have paved the way to the emerging research and clinical discipline of regenerative medicine.1

Regenerative Medicine

Regenerative medicine seeks to harness methods for the replacement or repair of dysfunctional cells, tissue, or organs in an attempt to restore normal function. It therefore draws on therapies from the three conventional pillars of medical therapeutics (pharmaceuticals, biologics, and medical devices) as well as from the newest platform technology, namely, cell therapy. The longterm goal of regenerative medicine is to cure disease by replacing the lost functions of tissues and organs, and thus it truly represents a paradigm shift from conventional therapies aiming to alter the natural course of disease or to provide symptomatic control. Consequently, regenerative medicine aims to develop curative strategies for unmet clinical needs such as diabetes, heart failure, and neurodegenerative disorders, among others.

Cell Therapy

Cell therapy involves the application of cells to achieve a therapeutic benefit, regardless of the cell type or clinical indication. Although achieving tissue and organ regeneration through cell replacement represents an important goal of cell therapy technology, its applications may reach far beyond the field of regenerative medicine. Hence, the spectrum of cell therapy approaches may range from permanent cell replacement strategies (attempting to replace lost or dysfunctional cells) to more transient cell therapies aiming to modulate disease progression or to protect tissues at risk, to achieve immunomodulatory effects (e.g., for prevention of graft-versus-host disease), to act as vehicles for the delivery of genes or gene products (cell-based gene therapy strategies), and even to act as cell-based cancer vaccines. This chapter focuses on the use of cell therapy for regenerative medicine and specifically concentrates on the potential role of different stem cell types to meet this challenge.

Stem Cells

Stem cells possess two defining properties: (1) the capacity for self-renewal and (2) the ability to differentiate into cell types with specialized cellular functions (Fig. 44-1). This may occur at the individual stem cell level through the process of asymmetrical cell division or at the cell population level wherein a subset of cells differentiate and the remaining stem cells remain dormant or replicate themselves as stem cells. After asymmetrical cell division, non–stem cell derivatives may either generate a pool of organ system–restricted, transitamplifying cells with enhanced proliferative capacity or continue to differentiate by epigenetic and gene expression profile changes until reaching the terminally differentiated state. This conceptual framework was developed after the discovery of bone marrow cells that were capable of reconstituting the adult hematopoietic system. These hematopoietic stem cells constitute the basis for hematopoietic stem cell transplantation, the only form of stem cell therapy currently routinely well established in clinical practice (Chapter 178).

Stem cell

Stem cell

Specialized cell

(e.g., neuron) FIGURE 44-1.  Asymmetrical cell division. Although this first characteristic was considered a required characteristic for stem cells based on their original description in the adult hematopoietic system, not all cell types currently named as stem cells necessarily display this property. For instance, human embryonic stem cells divide by symmetrical cell division.

The different stem cells types are routinely classified based on the protein or transcription factors they express, but also according to three basic additional attributes. These include replicative capacity (limited vs. unlimited), the scope or potency of differentiation (e.g., pluripotent, multipotent, oligopotent, unipotent), and their place in the life history of the organism (developmental or postdevelopmental). Thus, more recent terminology has broadened use of the term stem cells to cover a wider array of cell types that contribute to organ development or have the capacity to repopulate tissues and organ systems. The term stem cells, together with the formulations noted previously, has also recently been extrapolated to describe certain cellular subpopulations that may be principally responsible for the growth of malignant tumors. However, because cancer stem cells have no role in tissue regeneration, they are considered in Chapter 181. Adult (Postnatal) Stem Cells

After birth, many tissues are thought to contain a subpopulation of cells with the capacity for extended self-renewal, combined with the ability to differentiate into more mature cell types with specialized functions (Fig. 44-2). Adult stem cells, thought to represent less than 0.01% of the total number of cells, are located in specialized supportive niche compartments at various sites within the hematopoietic system and elsewhere, and respond to cues in their local microenvironment. As a result of the success of hematopoietic stem cell transplantation in the treatment of bone marrow failure or in conjunction with myeloablative therapy in malignancy, scientists have been motivated to find adult stem cells in other organs. Adult tissues and organ systems reported to contain putative stem cells include bone marrow (hematopoietic and mesenchymal compartments) and peripheral blood, blood vessel endothelium, dental pulp, epithelia of the skin, adipose tissue, digestive system, cornea, retina, testis, and liver. Similar stem/progenitor cells were also reported in organs historically not thought to contain such cells, such as the central nervous system, the heart, and the kidney. Whether adult stem cells represent remnants of developmental stem cells that persist into adulthood for purposes of organ maintenance and repair or represent a distinct cell type dedicated for this latter purpose is not clear. Importantly, in many organs, despite the presence of such tissue-specific stem cells, their regenerative capacity is still inadequate to deal with massive cell loss such as occurs, for example, after a large myocardial infarction or after ischemic brain injury. Embryonic and Induced Pluripotent Stem Cells

In contrast to adult stem cells that have relatively limited differentiation potency, cells in the developing preimplantation embryo still retain the capacity to differentiate into derivatives of all three germ layers (ectoderm, mesoderm, and endoderm), eventually contributing to all tissues in the body (Fig. 44-3). In normal development, however, such cells do not persist beyond the blastocyst stage. When isolated from unused preimplantation blastocysts generated for in vitro fertilization, the inner cell mass cells isolated can be used to generate human embryonic stem cell (hESC) lines (see Fig. 44-3). The generated hESCs exhibit unlimited self-renewal in cell culture in the undifferentiated state, while retaining the capacity to differentiate into cell derivatives of all three germ layers, essentially giving rise to any cell type in the body. Taking advantage of lessons learned from embryology, scientists were able to utilize the sequential application of different combinations of growth factors to achieve efficient differentiation systems from hESCs, yielding purified populations of different types of neurons, glial cells, cardiomyocytes, vascular endothelial and smooth muscle cells, pancreatic β cells, hepatocytes, different blood cells (platelets, red blood cells), and several other cell lineages. One of the limitations of the hESC technology is the inability to derive such cells from an adult individual, preventing their utilization in a patientspecific manner. These limitations can be overcome with the introduction of induced pluripotent stem cell (iPSC) technology.2 This approach allows adult somatic cells (e.g., fibroblasts) to be reprogrammed into pluripotent stem cells by introduction of a set of transcription factors linked to pluripotency (the originally reported combination of factors included OCT3/4, SOX2, c-MYC, and KLF4). The human iPSCs (hiPSCs) generated in this manner can then be coaxed to differentiate into a variety of cell types, using differentiation protocols similar to those already in place for hESC (Fig. 44-4). Importantly, because the hiPSCs can be generated in a patient-specific manner, this technology can potentially be used to develop autologous cellreplacement strategies that can evade the immune system, to generate patient- and disease-specific models of different genetic disorders, and to establish screens for drug testing and drug discovery.

205

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

Neutrophil

Natural killer (NK) cell T lymphocytes

Basophil

Bone Lymphoid progenitor cell Eosinophil

B lymphocyte

Hematopoietic stem cell

Multipotent stem cell

Monocytemacrophage

Myeloid progenitor cell Red blood cells

Bone matrix

Bone (or cartilage) Stromal cell

Platelets

Hematopoietic supportive stroma

Marrow adipocyte

Osteoblast Stromal stem cell

Lining cell Osteocyte

Blood vessel

Pre-osteoblast Skeletal muscle stem cell? Adipocyte

Hepatocyte muscle stem cell?

Hematopoietic stem cell

Osteoclast FIGURE 44-2.  Adult stem cells. Adult stem cells can be multipotent and have the capacity to differentiate into a limited number of different cell types, often restricted to a given tissue or organ system, as in the case of adult hematopoietic or epidermal stem cells. Two stem cell types have been isolated from adult bone marrow—the hematopoietic stem cell and the mesenchymal stem cell. Adult mesenchymal stem cells of bone marrow origin, although their range of differentiation has been shown to be broader than that of any other adult stem cell type, do not reach pluripotency. It is thought that in some organ systems, such as the gastrointestinal epithelium, a unipotent pool of progenitors exists for repopulating a rapid population turnover of only one type of cell—although it is difficult to be certain whether such progenitors can be distinguished from the overall population of fully differentiated cells in tissues with high cellular turnover.

Cell Therapy Approaches to Regenerative Medicine

Historically, the field of cell therapy can be traced to the transfusion of blood and blood products (Chapter 177), solid organ transplantation (Chapter 49), in vitro fertilization, and bone marrow transplantation (Chapter 178). Nevertheless, beyond the aforementioned therapies, which have become the mainstay treatments in several medical fields, additional cell therapy approaches are considered highly experimental and are still at different stages of preclinical and clinical development. These ongoing efforts can be conceptually grouped into six different approaches (Fig. 44-5).

Delivery of Bone Marrow− and Blood-Derived Stem/ Progenitor Cells

A flurry of studies during the past decade evaluated the ability of bone marrow−derived hematopoietic or mesenchymal stem cells to achieve tissue repair after delivery to a variety of organs. These studies were based initially on the assumption that these types of adult stem cells may display some degree of plasticity, allowing them to transdifferentiate into the relevant cell types (e.g., heart cells, nerve cells, and liver cells) after transplantation into the appropriate tissue environment. Although mounting evidence suggests that such transdifferentiation probably does not occur to a significant extent, many of these studies appeared to result in some degree of functional improvement after stem cell delivery to different organs. This clinical benefit may stem from the secretion of different growth factors by the engrafted cells (“paracrine hypothesis”); these factors in turn are thought to augment endogenous tissue repair mechanisms, improve tissue vascularization, modulate inflammation, and protect tissues at risk.

Delivery or Activation of Tissue-Specific Stem/Progenitor Cells or Induction of Cell Proliferation

In contrast to the conventional dogma, recent evidence suggests that a number of organs previously believed to lack any regenerative capacity (e.g., the brain, pancreas, kidney, and heart) in fact do possess such ability, albeit at a limited capacity. Whether this capability is due to the presence of tissue-specific stem/progenitor cells or due to some replication capa-

bility of terminally differentiated cells is still a matter of debate for each organ. Significant efforts have been made in recent years to isolate such putative tissue-specific stem/progenitor cells based on the expression of general or specific stem cell markers or based on their unique culturing properties. These studies also highlighted the potential of such cells to be cultured in a clonal manner and to give rise to one or more cell types relevant to the organs from which they were isolated. Current efforts to utilize the aforementioned findings for regenerative medicine are focused either on the isolation, ex vivo expansion, and transplantation of such putative stem/progenitor cells back to their respective native organs or on the augmentation of their endogenous reparative potential in vivo. The former strategy can be exemplified in the central nervous system where progenitor cells are harvested, cultivated in culture (as neurospheres), and give rise to different types of neurons and supporting glial cells. Similar efforts have followed in other organs. In the heart, for example, such efforts have already reached early clinical trials, in which autologous cardiac stem cells were harvested from the heart, expanded ex vivo, and then engrafted back to the heart. The latter approach, in contrast, aims to influence putative stem cell niches within damaged organs to enhance the endogenous reparative properties of those stem/progenitor cells. Such an effect may underlie the potential therapeutic benefit of bone marrow−derived stem cells after their delivery to different organs. The final strategy aims to boost endogenous organ repair through the replication of terminally differentiated tissue-specific cells. Such strategies can either augment the inherent physiologic capability of a given organ (e.g., insulin secretagogues for pancreatic β cells) or attempt to induce replication in cells that have already withdrawn from the cell cycle. Caution is warranted with respect to the latter approach because induction of uncontrolled proliferation (e.g., by genetic manipulation) may increase the risk for tumorigenesis.

Engraftment of Fetal Tissue

The most straightforward approach to organ repair would be to replace the missing cells with identical counterparts. Harvesting and expanding adult

206

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

In vitro fertilization Day 0

Totipotent cells Day 3

Blastocyst Day 5 Origin: Derived from preimplantation or peri-implantation embryo

Self-renewal: The cells can divide to make copies of themselves for a prolonged period of time without differentiating.

Stem cell

Pluripotency: Embryonic stem cells can give rise to cells from all three embryonic germ layers even after being grown in culture for a long time.

The three germ layers and one example of a cell type derived from each layer: Ectoderm

Mesoderm

Neuron Ectoderm gives rise to: brain, spinal cord, nerve cells, hair, skin, teeth, sensory cells of eyes, ears, nose, and mouth, and pigment cells.

Blood cells

Mesoderm gives rise to: muscles, blood, blood vessels, connective tissues, and the heart.

Endoderm

Liver cell Endoderm gives rise to: the gut (pancreas, stomach, liver, etc.), lungs, bladder, and germ cells (egg or sperm).

FIGURE 44-3.  Embryonic stem cells. Totipotency refers to the capacity to differentiate into all cell types in an organism, including extraembryonic tissues, placenta, and umbilical cord, a property confined to the fertilized egg itself, including the cells derived from the first few cell divisions after fertilization. Pluripotency refers to the capacity to differentiate into all the specialized cell types derived from the three germ layers (ectoderm, mesoderm, endoderm) of the developing embryo and is a hallmark feature of embryonic stem and germ cells.

human cells for transplantation, however, may not be possible in the case of several organs with limited regenerative capacity. During prenatal human development, cells of fetal origin often show enhanced proliferative capacity as well as the ability to differentiate into more than one type of mature or specialized cell. Moreover, animal studies have demonstrated that transplantation of tissues harvested from developing organs (harvested within a specific time window during embryonic development) may give rise to entire functioning organs such as kidneys, lungs, and pancreas. Nevertheless, to date, the only fetal-derived cells that have been used in human clinical applications are the dopaminergic cells derived from the developing fetal nervous system for the treatment of Parkinson disease (Chapter 409). The broader use of fetal tissues for regenerative medicine may be hampered by the limited

access to such cells for both technical and ethical reasons, the allogeneic nature of such procedures (requiring immune suppression), and the potential for tumor formation as already described in some case reports.

Transplantation of Ex Vivo Differentiation of Pluripotent Stem Cells

Unlike fetal tissues, hESCs are truly pluripotent (can give rise to advanced cell derivatives of all three germ layers). Importantly, hESCs can be propagated in the undifferentiated state and then coaxed to differentiate into a variety of cell types, giving rise to a potentially unlimited number of specialized cell types for transplantation. Consequentially, numerous preclinical studies have demonstrated the ability of hESC derivatives to engraft, survive,

207

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

Reprogramming

cMyc

Sox2

Human iPS cells

Klf4

Oct4

Somatic cells: (Fibroblasts)

Guided differentiation

Cell therapy and tissue engineering

Neurons

Gene correction through gene editing

45

45

40

40

35

35

30

30

25

25

20

20

15

15

10

10

5

5

β-cells

Patient-specific therapies (personalized medicine)

CMs

Disease Modeling

Drug Screening and Discovery

FIGURE 44-4.  Application of the induced pluripotent stem cells (iPSC) technology. Patient-specific human iPSC can be generated by reprogramming of adult somatic cells (fibroblasts) with a set of transcription factors and then coaxed to differentiate into a variety of cell lineages. The patient-specific human iPSCs can then be transplanted back to the patient in an autologous manner for regenerative medicine applications. In a similar manner disease- and patient-specific human iPSC models of inherited disorders could be generated (“disease-in-a-dish models”) and used for better understanding of genetic disorders, for drug development, and for optimizing patient-specific therapies. Gene editing techniques can be used for mutation correction and for transplantation of healthy cells. CM = cardiomyocytes; iPS = induced pluripotent stem cells.

and improve organ performance in a wide spectrum of relevant animal disease models (e.g., heart failure, Parkinson disease and other neurodegenerative disorders, diabetes). Early clinical studies using hESC derivatives are just emerging and have been focused so far on the retina (transplantation of retinal pigment epithelium [RPE] cells) and spinal cord injury (using oligodendrocyte progenitors). Despite the significant achievements made with hESCs, the inability to create patient-specific hESCs from adult individuals, the ethical issues arising from destructive use of human embryos, and the anticipated immune rejection associated with such allogeneic cell transplantation impose important hurdles for their clinical utilization. The hiPSC technology provides a potential solution to these challenges. As noted, the patient’s own somatic cells (fibroblasts, hair follicles, urine epithelial cells, or blood cells) could be reprogrammed by a set of transcription and chemical factors to yield pluripotent stem cells. The patient-specific hiPSCs could then be coaxed to differentiate to a variety of cell lineages, using protocols similar to those already in place for hESCs. In turn, these differentiated derivatives could then be transplanted either in an autologous or allogeneic manner. Clinical trials using hiPSC-derived cell lineages are expected to be initiated in the coming few years, with the initial targets being macular degeneration (RPE cells), Parkinson disease (dopaminergic neurons), blood product transfusion (hiPSCderived platelets and red blood cells), and heart failure (cardiomyocytes). One of the concerns in translating hESCs and hiPSCs into a therapeutic platform is the oncogenic risk. This concern stems from the potential for remaining undifferentiated cells within the cell grafts to form teratomas, from

the use of oncogenic reprogramming factors, from the random integration of the viral vectors used in cellular reprogramming (“insertional oncogenesis”), and from genetic instability, potentially leading to both chromosomal aberrations and mutations. Progress to clinical trials requires definitive clarification of this key concern.

Direct Reprogramming

In contrast to the iPSC approach, which seeks to initially reprogram somatic cells to a pluripotent state followed by differentiation of the generated iPSCs to specific cell lineages, recently described direct reprogramming strategies aim to convert the phenotype of one mature cell type (fibroblasts) directly to another. The prototype for such a strategy was the demonstration that MyoD, a master regulator of skeletal muscle formation, can convert fibroblasts directly to skeletal muscle. Progress to derive other cell types after this report was delayed for many years because, unlike skeletal muscle, a single master developmental regulatory gene does not exist for most cell lineages. Based on the experimental approach used to identify the combination of transcription factors that can reprogram somatic cells into iPSCs, researchers evaluated the ability to achieve analogous transcription factor reprogramming strategies to convert the cell fate of somatic cells directly. Consequentially, using a combination of lineage-specific developmental transcription factors, scientists were able to convert terminally differentiated fibroblasts or other somatic cells directly to neurons, β cells, different hematopoietic cell lineages, and cardiomyocyte-like cells. Recent studies have taken this concept a further step forward by demonstrating that transcription factor−based

208

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

Induction of Endogenous Regeneration/Repair

Cell/Tissue Transplantation Blastocyst Fetal tissue (fetal ventral mesencephalon)

Activation of tissue-specific stem/progenitor cells hESCs, hiPSCs

Oct4, Sox2, Klf4, cMyc Fibroblasts

Neurons Fibroblasts Induction of cell replication

CMs

β-cells

Direct TF reprogramming

Hepatocytes

In vivo TF direct reprogramming

Tissue engineering (biopolymers and cells)

Fibroblasts

Additional mechanisms: • Promotion of angiogenesis • Modulation of inflammation • Cell protection • Trophic effects

Bone marrow Hematopoietic/ mesenchymal stem cells

FIGURE 44-5.  Conceptual framework for regenerative medicine approaches. These strategies can be divided into those attempting to augment endogenous regeneration (left side) and those focusing on transplantation of cells (right side). The former could be achieved through the activation of putative tissue-specific stem/progenitor cells, through induction of cell replication, by in vivo transcription factor (TF)-based direct reprogramming (directly converting one somatic cell [fibroblast] into another), and by several other indirect means (e.g., modulation of inflammation, induction of angiogenesis, trophic effect, protection of tissue at risk). Cell sources that can be used for cell transplantation include fetal tissues (e.g., dopaminergic-rich fetal ventral mesencephalon for Parkinson disease), pluripotent stem cells (human embryonic stem cells [hESCs] and human induced pluripotent stem cells [hIPSCs]), derived cell-lineages, and somatic cells that can be generated ex vivo by direct transcription factor−based reprogramming of fibroblasts. CMs = cardiomyocytes.

transdifferentiation can also be achieved in vivo, suggesting a method whereby resident cells (fibroblasts, hepatic cells, or other cells) could be converted to the appropriate cell types for organ repair. Development of the latter approach for clinical application may be considered more analogous to gene therapy, with the associated advantages, shortcomings, and challenges of this discipline (see later).

Tissue Engineering

Tissue engineering is an interdisciplinary technology combining principles from life sciences and engineering with the goal of developing functional substitutes for damaged tissues and organs.3,4 Rather than simply introducing cells into a diseased area, in tissue engineering, cells are embedded or seeded onto three-dimensional scaffolds (derived from different biomaterials) before

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

transplantation. Regardless of the specific clinical application, tissueengineering strategies usually involve the utilization of combinations of biomaterials, cells, and biologically active factors. The scaffold serves many purposes, including the control of the shape and size of the engrafted tissue, the delivery of biologic signals and adequate biomechanical support to the cells, the induction of vascularization of the graft, and the protection of the cells from physical damage. Scaffolds used in tissue engineering approaches are commonly divided into two general categories: (1) cellular scaffolds that are seeded ex vivo with cells before their in vivo transplantation; and (2) acellular scaffolds that depend on cells in the recipient to repopulate the scaffold with subsequent reconstitution after transplantation. Such tissueengineered efforts have already reached proof-of-concept clinical trials. These efforts have mainly concentrated on the musculoskeletal system (bone and cartilage repair) but have also targeted other organs such as the heart and even complex organ structures such as the esophagus, trachea, and urinary bladder.

Specific Disease Applications in Cell Therapy

Although a growing number of experimental cell therapies have reached various stages of clinical trials, as yet none has become an established or approved treatment, with the aforementioned exception of hematopoietic stem cells and solid organ transplantation. Nonetheless, with the expectation of significant advances on the horizon, examples of some of the current cell therapy efforts being made in the fields of neurodegenerative disorders, heart failure, and diabetes are provided.

Neurodegenerative Disorders

The central nervous system has limited capacity for regenerating lost tissue both in slowly progressive degenerative neurologic conditions such as Parkinson disease, Alzheimer disease, and amyotrophic lateral sclerosis (ALS) and in acute injuries leading to rapid cell loss (ischemic stroke or traumatic spinal cord injury). Stem cell−based therapies are being explored as potential novel therapeutic paradigms for both acute and chronic neurodegenerative disorders.5 Consistent with the spectrum of cell therapy−related mechanistic actions described above, these procedures could potentially act through the following mechanisms: (1) cell replacement, whereby cells (precommitted to specific neuronal or glial lineages) are transplanted to replace the specific subtypes of cells that were lost (i.e., dopaminergic neurons in Parkinson disease, motor neurons in ALS, or a mixture of different neuronal and glial subtypes in other disorders); (2) trophic support, whereby the engrafted cells are used to promote the survival of affected neurons or glia or stimulate endogenous repair of the diseased central nervous system through the secretion of neurotrophic factors; and (3) modulation of the inflammatory process thought to contribute to the pathogenesis of many neurodegenerative processes. Achieving the first mechanistic goal, despite being the most attractive, is probably also the most challenging because one would need not only to derive a clinically relevant number of the specific glial or neuronal subtypes or a combination of these cells but also to deliver them to the appropriate site (either focally or diffusely throughout the brain), as well as to assure cellgraft survival, its continuous and appropriate function, and importantly, its integration with the host neuronal network. Parkinson disease (Chapter 409) involves loss of melanin-containing dopaminergic neurons within the substantia nigra pars compacta of the midbrain, coupled with accompanying depletion of striatal dopamine. This cellular loss is responsible for the major motor features of the disease. In the search for a more definitive therapy than pharmacology, early reports of cell replacement therapy suggested significant improvement in motor function after intrastriatal implantation of mesencephalic dopamine-rich tissue, obtained from aborted human fetuses aged 6 to 9 weeks. Long-term immunosuppressive treatment is essential to allow transplanted dopaminergic neurons to develop into their full functional potential despite the notion of an immunologic sanctuary within the brain. Clinical assessment standards have provided evidence of long-lived graft survival, morphologic and functional integration, and clinical benefit after therapy with cells of fetal origin that have now lasted up to 10 years or longer in some patients. Further progress, however, has been limited by lack of sufficient source tissue to treat a large number of affected patients, prohibitive variability in functional outcome, reports of serious dyskinesias in a subset of treated patients, and ethical considerations. Given the aforementioned limitations of fetal tissue engraftment, stem cell derivatives could offer a viable alternative for the treatment of Parkinson disease by either replacing the dopaminergic neurons or slowing the degeneration process and restoring the integrity of the nigrostriatal pathway

209

through the release of trophic factors. Importantly, dopaminergic neuroblastlike cells have been generated ex vivo from different stem cell sources, including pluripotent stem cells (hESCs and hiPSCs) by direct fibroblast reprogramming, neural stem cells and progenitors from the embryonic ventral mesencephalon, and adult neural stem cells from the subventricular zone. Preclinical engraftment studies demonstrated that such cells could survive in animal models of Parkinson disease and exert beneficial functional effects after cell maturation. Nevertheless, some properties that are fundamental for successful clinical translation have not been fully met in animal transplantation trials employing human stem cell–derived dopaminergic neurons. Additional challenges that should be addressed include development of methods to prevent the disease process from also destroying the grafted neurons (e.g., engineering the cells to secrete neurotrophic factors) and limiting graft-induced dyskinesia (e.g., by minimizing the number of serotonergic neuroblasts in the grafted tissue). Investigations of stem cell–based approaches for the treatment of other neurodegenerative diseases, including ALS, Alzheimer disease, Batten disease, stroke, and brain and spinal cord injury, are now moving from experimental animal model studies to planning of clinical trials. Recent reports have shown a major clinical benefit in animal models of directed differentiation and transplantation of hESCs and hiPSCs toward retinal pigment epithelum. Human studies with these cells were recently initiated for patients suffering from macular degeneration.

Heart Disease

Although recent studies have challenged the dogma of the heart being a completely terminally differentiating organ, the endogenous repair mechanisms of the adult heart are usually inadequate in dealing with an extensive myocardial infarction. The resulting decrease in the contractile mass, which is associated with the loss of approximately 1 billion cardiomyocytes, may lead to the development of clinical heart failure (Chapters 58 and 59). With heart failure being the leading cause of hospitalization and with the paucity of donor organs limiting the number of heart transplantations worldwide, it is not surprising that the heart has become the focus of various regenerative medicine efforts.6 The first cells that reached clinical trials for heart failure were skeletal myoblasts. Such cells could be harvested (satellite cells) in an autologous manner, expanded ex vivo, and transplanted to the heart. However, skeletal myoblasts display different physiologic properties than cardiomyocytes and cannot form electromechanical connections with host cardiac tissue. Consequentially, these clinical efforts have largely been abandoned because of lack of efficacy as well as evidence suggesting increasing arrhythmogenicity in some patients. The largest clinical experience in myocardial cell therapy comes from the use of bone marrow−derived stem cells (primarily hematopoietic stem cells and more recently also mesenchymal stem cells). The effects of delivery of such cells (mainly through the coronary circulation) were studied in thousands of patients, primarily in the setting of acute or recent myocardial infarction. These studies revealed either a neutral effect on myocardial performance or mild functional improvement. Although a recent meta-analysis of 33 randomized controlled trials studying transplantation of adult bone marrow−derived cells revealed a statistically significant improvement in left ventricular ejection fraction, this improvement was not associated with a change in mortality. A1  Bone marrow−derived stem cells are thought to exert their beneficial effects through the secretion of different growth factors rather than transforming to become new heart cells. Consequentially, a cell source that could truly re-muscularize the heart is direly needed. A potential candidate for such a task could be the recently described cardiac progenitor cells. Several reports have described cardiac progenitor cells as multipotent clonogenic cells that could be isolated based on different markers or culturing properties and potentially differentiate into cardiomyocytes and vascular cells. Such cells can be harvested from the heart (during surgery or a percutaneous cardiac catheterization biopsy approach), expanded ex vivo, and then transplanted back to the left ventricle in an autologous manner. In contrast to the aforementioned cell types, human pluripotent stem cell lines (hESCs and hiPSCs) can undoubtedly become cardiomyocytes during ex vivo differentiation. Research efforts in recent years established efficient directed differentiation systems that could give rise to clinically relevant numbers of cardiomyocytes and demonstrated the ability of the generated cells to engraft, functionally integrate with host cardiac tissue, and improve myocardial performance in animal models of myocardial infarction. Nevertheless, issues related to ethics and the allogeneic nature of the graft (hESCs),

210

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

to the inefficient and incomplete reprogramming process (iPSCs), to the heterogeneous and relatively immature properties of the generated cardiomyocytes, to the tumorigenic risk, and to the complex regulatory and financial issues have hindered clinical development of these cells to date. Most recent efforts in the field have focused on attempting to induce mature cardiomyocytes to reenter the cell cycle (directly or after an initial dedifferentiation phase) or to convert the phenotype of nonmyocytes (fibroblasts) into cardiomyocytes. In the latter approach, recent studies have demonstrated the ability to convert the phenotype of murine fibroblasts both in vitro and in vivo into cardiomyocyte-like cells by the expression of a combination of cardiomyocyte-specific transcription factors (GATA-4, MEF-2C, and TBX-5 in one study). Although these and other efforts have the potential to augment the number of cardiomyocytes and consequentially improve contraction of the failing heart, they are still in the early phase of discovery.

Diabetes Mellitus

Successful pancreatic transplantation and improved glucocorticoid-free protocols for transplantation of islets of Langerhans have been shown not only to restore glucose control in patients with diabetes mellitus but also to prevent or even reverse some of the disease’s complications (Chapters 229). However, whole organ or islet-based transplantation approaches are limited both by immunologic rejection and by limitation of an available source of transplantable tissues. This has motivated the search for cell types that can replace (type 1 diabetes mellitus) or augment (type 2 diabetes mellitus) deficient β-cell function.7 The development of hESCs and hiPSCs, coupled with improved understanding of β-cell development, has provided a potentially unique cell source to derive β cells for transplantation therapy. β cells make an especially attractive case for cell replacement strategies because only a single cell type is missing, cell replacement does not necessarily need to be performed in the native environment (pancreas), and, theoretically, such cells could even be engrafted subcutaneously. Harnessing lessons from embryology, efficient protocols were developed to promote differentiation of pluripotent cells in vitro into precursor or early-stage β-cell phenotype. More recent efforts have moved the field even closer to clinical application by tackling the challenge of creating more mature and functional β cells. One of the problems with using β cells for cell replacement therapy is that the autoimmune destruction of endogenous β cells, which underlies the pathogenesis of type 1 diabetes, will probably also result in the destruction of the pluripotent cell−derived β cells, even when derived from an autologous (hiPSCs) source. Consequentially, significant efforts are being made to develop the biotechnologic means (encapsulation technologies) to deliver the cells in an immunoprotective environment that will prevent cell rejection but will retain the capacity of the engrafted β cells to sense glucose and to secrete insulin. Beyond the derivation of new β cells from pluripotent stem cells, progress has also been made in reprogramming closely related cell types to β cells by the overexpressing of master regulatory transcription factors. Early studies focused on the conversion of hepatocytes to β-like cells through the overexpression of PDX1, the transcription factor MAFA, and NeuroD. In vivo transdifferentiation of mouse acinar cells to β cells has been achieved by transient viral overexpression of three transcription factors (PDX1, NGN3, and MAFA), whereas overexpression of a single transcription factor, PAX4, has successfully converted murine α cells to β cells. Regenerative strategy focuses on increasing pancreatic β-cell mass by inducing the replication of existing β cells. This therapeutic approach would probably mainly target type 2 diabetic patients by decreasing the burden on existing overworked β cells but may also be beneficial for some patients with type 1 diabetes who still retain some β-cell mass. Whereas several tissues are regenerated by differentiation of tissue-specific stem cells, new pancreatic β cells are derived from the replication of existing β cells. Promising candidates for augmenting β-cell replication were recently identified and include the use of gluco*kinase activators or betatrophin, a protein secreted by the liver.

Stem Cell–Derived Platforms for Disease Modeling, Personalized Medicine, and Drug Discovery

In addition to the generation of cells for regenerative applications, the ability to grow a wide variety of different specialized cell types of human origin in culture provides unparalleled opportunities for gene and drug discovery and testing. For example, the ability to grow human cardiomyocytes in culture provides a preclinical human cellular-based experimental platform for

screening newly developed drugs in terms of their potential to cause QT-interval prolongation and hence the risk for arrhythmia in the clinical setting. Other examples include the creation of an experimental tissue microenvironment of human origin for studying the stromal response to tumor growth and testing anticancer drugs that target tumorigenic responses such as angiogenesis.8 The hiPSC technology has further revolutionized this field because it allows for the first time the generation of disease/genotype- and patientspecific hiPSC models of a wide array of inherited disorders. Initial studies focused on diseases with monogenic inheritance, but more recent studies have included diseases with more complex inheritance patterns.9 Consequently, different types of patient-specific hiPSC-derived neurons, cardiomyocytes, skeletal muscle, blood cells, hepatocytes, and other cell types were demonstrated to recapitulate in a culture dish the abnormal phenotype of a wide array of genetic disorders, including neurodegenerative disorders (e.g., spinal muscular atrophy, familial dysautonomia, ALS, schizophrenia, and even late-onset diseases such as Parkinson and Alzheimer disease), different cardiomyopathies and arrhythmogenic syndromes, a wide array of blood disorders, and several other genetic disorders. These models have already yielded important insights into the mechanisms underlying these disease states and have established unique experimental platforms that will enable the testing of existing therapies in a patient-specific manner (personalized medicine) to evaluate evolving therapies (“clinical studies in the culture dish”) and to develop new therapeutic strategies.

GENE THERAPY

Gene therapy can be broadly defined as the transfer of genetic material into cells to restore or correct a cellular dysfunction or to provide a new cellular function in an attempt to cure a disease or at least to improve the clinical status of a patient. The use of genes as therapeutic platforms emerged during the mid-20th century, and in the 1990s, the first regulated registered studies were performed in the United States. In the first clinical study, a 4-year-old girl with adenosine deaminase (ADA) deficiency was treated by transfecting the ADA gene into her white blood cells, resulting in improvements in her immune system. Since then, more than 10,000 patients have been involved in more than 1700 gene therapy clinical studies performed throughout the world. The most common patient populations targeted in these studies have been cancer patients (more than 1000 studies), with another important category being monogenic inherited disorders (more than 100 studies). Although gene therapy initially was conceived as a way to treat life-threatening disorders (inborn errors, cancers) refractory to conventional treatment, it is now being explored for non-life-threatening conditions that adversely affect a patient’s quality of life. Although early clinical failures and a number of reported deaths (only two of which were actually attributed directly to gene therapy) and cases of gene therapy−related leukemic transformation led many to dismiss gene therapy as hazardous and premature, recent clinical successes have bolstered new optimism in the promise of this discipline. These include entirely novel initiatives in treating primary immunodeficiency syndromes, the improvement of vision in patients with the retinal disease (e.g., Leber congenital amaurosis), the successful treatment of X-linked adrenoleukodystrophy,10 and the encouragement of experimental results in treating different forms of cancer. Despite these success stories, only a few gene therapy agents are currently approved and available. Fomivirsen (Vitravene) is used for the treatment of cytomegalovirus retinitis in patients with AIDS. In 2012, Glybera became the first gene therapy treatment to be approved for clinical use in either Europe or the United States. Glybera uses a virus injected into a patient to deliver a working copy of a gene for producing lipoprotein lipase (LPL) to treat the rare inherited disorder of LPL deficiency. Finally, the p53 tumor suppressor coding sequence in an adenovirus vector is used for the treatment of head and neck cancer but is registered only in China.

Classifications and Mechanisms of Action

In general, somatic gene therapy applications can be divided into those aiming to treat or correct various genetic disorders and those targeting nongenetic diseases by attempting to alter cell, tissue, and organ function in a favorable manner. According to the World Health Organization, there are more than 10,000 disorders with monogenic inheritance described in humans (http://www.who.int/genomics/public/geneticdiseases/en/index2.html), but only a small fraction of these may be amenable to gene therapy. Traditional gene therapy efforts for inherited disorders have mainly focused on the

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

exogenous expression of genes encoding the missing or abnormal proteins and to a lesser extent also on altering the abnormal gene expression patterns. Future efforts are expected to shift the focus from uncontrolled overexpression of the missing protein to directly correcting the mutation at the DNA level (gene-editing strategies) in affected cells, using the newly emerging technologies known as the TALEN and CRISPR11 approaches described in greater detail later under Gene Editing. With successful widespread use in research studies and proven applications in editing of gene sequence in stem cells, the transition to clinical application is sure to follow. For nongenetic disorders, gene therapy efforts are aimed at overexpressing a specific protein in an attempt to alter cellular function favorably (e.g., to increase contractility in heart failure by overexpression of the sarcoplasmic reticulum calcium ATPase SERCA2a), protect tissue at risk (e.g., in acute kidney injury), exert paracrine effects through local secretion of specific proteins by the engineered cells (e.g., promote angiogenesis in ischemic tissues or induce neurotrophic effects in neurodegenerative disorders), and even secrete proteins systemically (e.g., in gene therapy trials attempting to correct bleeding disorders by secretion of coagulation factors or for systemic delivery of hormones such as erythropoietin for the treatment of anemia). Major efforts in the gene therapy arena to date have been in designing various methods to treat cancer (see later). Progress in the field of gene therapy has developed into two different strategies: ex vivo and in vivo gene therapy. The ex vivo gene therapy approach (combined cell and gene therapy strategy) involves the initial harvesting of cells from a given patient followed by genetic modification of these cells in the laboratory. The genetically modified cells can then be selected, amplified in numbers, and returned to the same patient in an attempt to achieve the desired therapeutic effect. This strategy is particularly attractive for the genetic modification of stem cells that could reconstitute the relevant tissues, organs, and organ systems after transplantation. The most prominent example is using hematopoietic stem cell grafts in gene therapy trials for hematopoietic disorders. The in vivo gene therapy approach, in contrast, involves the delivery of the relevant transgene (through the use of various vectors) directly to the targeted tissue, followed by the stable or transient expression of the transgene in the relevant cells. The expression of the transgene only in the relevant cells/ tissues can be achieved by a combination of localized delivery (injection), a particular tropism of the vector used for the tissue of interest, and the expression of the transgene under the control of a cell/tissue-specific promoter.

Gene Therapy Delivery Methods

Gene therapy agents are often composed of two elements: the genetic material itself (i.e., the DNA expression cassette [the most common therapeutic payload used], short interfering RNA, or an antisense molecule) and the vector delivery system. The latter is usually the more complex and limiting component, and it is important to select the most efficient delivery method for any genetic therapy as well as to be aware of the potential adverse effects of each vector type, thus tailoring the therapy to specific clinical considerations. There are formidable barriers to successful gene transfer, such as crossing the cellular membrane, escaping from the endosome, moving through the nuclear membrane, and integrating into the host genome. Vectors that have been developed to try to overcome these obstacles fall into two broad categories: nonviral and viral vectors. Gene therapy mediated by nonviral vectors is referred to as transfection and consists of the direct delivery of naked DNA by injection, the use of liposomes (cationic lipids mixed with nucleic acids), nanoparticles, and other means. Although nonviral vectors can be produced in relatively large amounts and are likely to present minimal toxic or immunologic problems, their major shortcoming is inefficient gene transfer. In addition, expression of the foreign gene tends to be transient, precluding the application of nonviral vectors to many disease states in which sustained and high-level expression of the transgene is required. The efficiency of nonviral vector delivery could be enhanced by the use of different physical methods that have evolved, such as electroporation (for well-circ*mscribed body compartments or masses such as muscle, skin, and tumors), gene gun (for DNA vaccination), and ultrasound delivery (for cardiovascular and tumor-related applications). Gene therapy mediated by viral vectors is referred to as transduction, and this approach has been the main conduit for transferring genes to human cells in most gene therapy trials. The basic concept of viral vectors is to harness the innate ability of viruses to deliver genetic material into the infected cell. Viruses used in gene therapy have been modified to enhance safety, increase

211

specific uptake, and improve efficiency. However, for each specific virusbased gene therapy vector, there have been major disadvantages that should be balanced against potential therapeutic benefits. For example, in cancer gene therapy, the immune response to the delivery vehicle carrying the anticancer genetic material can be used to advantage by serving as an adjuvant. In contrast, the system for delivery of a gene to be expressed for a prolonged period to replace or supplement a missing gene product in monogenic disease states should preferably be ignored by the immune system. Viral vectors are derived from viruses with either RNA (retroviruses and lentiviruses) or DNA (adenovirus, adeno-associated virus [AAV], herpes simplex virus [HSV], and poxvirus [vaccine virus]) genomes. Viral vectors also fall into one of two main categories: integrating vectors, which insert themselves into the recipient’s genome, and nonintegrating vectors, which often (although not always) form an extrachromosomal genetic element. Integrating vectors, such as γ-retroviral vectors and lentiviral vectors, are generally used to transfect actively dividing cells because they are stably inherited. Integrating vectors, however, may carry the risk for insertional mutagenesis (with clinical oncogenic transformations reported with the use of retroviruses). Nonintegrating vectors, such as adenoviral vectors and AAV vectors, can be used to transfect quiescent or slowly dividing cells, but they are quickly (in the case of adenoviral vectors) lost from cells that divide rapidly. Finally, efficient gene transduction can also be achieved using vectors that are maintained as episomes, especially in nondividing cells. Adenoviral vectors and retroviral vectors based on Moloney murine leukemia virus featured prominently in early gene therapy trials. There has been a movement away from both, however, after the case of a fatality, which was linked to the toxicity of the adenoviral vector (used to introduce the ornithine transcarbamylase gene in that specific study) and the leukemia cases in SCID-X1 patients (which were linked with activation of LMO2, an oncogene on chromosome 11, due to insertional mutagenesis associated with the murine leukemia viral vector). Consequentially, these vectors have been largely been replaced with AAV and lentiviral vectors, respectively, which have become the most common vectors used in clinical trials today. Other viral vectors may have applications in specific settings. For example, in gene therapy applications being developed for pain management, a replicationdefective HSV vector is being used because of its tropism for nerve tissues. Also, different oncolytic viruses with a preferential tropism to cancer cells are being used for gene therapy applications in cancer.

Diseases Treated by Gene Therapy Inherited Immunodeficiency

More than 30 patients reported to date worldwide have undergone treatment with different retroviral vectors for inherited immunodeficiencies (Chapter 250). Patients with one of the following three diseases are included in this group: two types of severe combined immunodeficiency (SCID), both of which are characterized by dysregulation of lymphocyte development, and X-linked chronic granulomatous disease (X-CGD), an inherited immune deficiency with absent phagocyte reduced nicotinamide adenine diphosphate oxidase activity caused by mutations in the gp91 (phox) gene. Individuals with adenosine deaminase (ADA) SCID suffer from premature death of T, B, and natural killer (NK) cells as a result of the accumulation of purine metabolites; patients with this condition have been treated with vectors expressing the ADA gene. In the first patients with ADA SCID, transduced T cells expressing transgenic ADA have been shown to persist for longer than 10 years; however, the therapeutic effect of gene therapy resulted in incomplete correction of the metabolic defect. More recently, an improved gene transfer protocol of bone marrow CD34-positive cells, combined with lowdose busulfan, resulted in multilineage, stable engraftment of transduced progenitors at substantial levels, restoration of immune function, correction of the ADA metabolic defect, and proven clinical benefit.12 Overall, no adverse effect or toxicity has been observed in patients treated with ADA gene transfer in mature lymphocytes or hematopoietic progenitors. The X-linked type (X-SCID group), in which there is defective cytokinedependent survival signaling in T and NK cells, was shown to be corrected by introduction of the wild-type sequence of the common γ-C chain, which is an essential component of five cytokine receptors. In one clinical study, hematologic malignancies developed in four patients. One of the four died of this complication. Ten patients were successfully treated with a different viral transduction protocol, with one reported malignancy in up to 8 years of follow-up. Two adult X-CGD patients who suffered recurrent bacterial infections have been treated with CD34-positive cells transduced with a γ-retroviral vector expressing gp91 phox, with significant clinical improvement in the

212

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

short term. However, in both these patients, there was an expansion of genetransduced cells caused by the transcriptional activation of growth-promoting genes leading to myelodysplasia and gradual loss of efficacy. In summary, of the nearly 30 patients worldwide treated with gene therapy for immunodeficiency disorders, significant clinical improvement has been observed in many. However, severe and even life-endangering adverse consequences have been encountered with certain viral vectors and protocols. Additional clinical information from long-term observation and new clinical studies will be important for a clearer assessment of clinical benefit.13

that deliver a copy of the CFTR gene to the airway of CF patients have been developed. Several placebo-controlled clinical trials of liposome-mediated CFTR gene transfer to the nasal epithelium have confirmed its safety and demonstrated variable degrees of functional correction. In addition, several clinical studies have assessed the potential of retrovectors, adenovectors, and AAV vectors for gene therapy for CF. With both nonviral and viral delivery systems, there were only mild side effects. However, the long-term clinical benefit has been marginal. Improved vectors are being assessed in preclinical studies.

Visual Loss

Cancer

Both cell- and gene-based therapy approaches are leading areas for promising inroads in retinal disease. Although early trials of stem cell−based retinal cell therapy have not yet achieved proof of efficacy, at the level of gene therapy, clinical scientists have used gene augmentation therapy with direct subretinal injection of a recombinant AAV expressing RPE65 complementary DNA in adults and children with Leber congenital amaurosis. This rare inherited eye disease destroys photoreceptors (Chapter 424), and the gene therapy results have shown medical evidence of visual preservation despite continued retinal degeneration.14

Cardiovascular and Pulmonary Conditions

Gene therapy efforts in the cardiovascular field have focused on achieving therapeutic angiogenesis in patients suffering from chronic ischemic heart disease A2  A3  or from critical limb ischemia (CLI) and for improving cardiac function in heart failure patients. The use of genes to revascularize the ischemic myocardium due to coronary artery disease and CLI due to peripheral artery disease has been the focus of two decades of preclinical research with a variety of angiogenic mediators, including vascular endothelial growth factor, fibroblast growth factor, hepatocyte growth factor, and others, encoded by DNA plasmids or adenovirus vectors. Overall, these gene therapy studies in animal experimental models of ischemia were very encouraging, leading eventually to several clinical trials. Despite the established proof of concept and reasonable safety, however, results of the latest clinical trials on therapeutic angiogenesis for myocardial ischemia and CLI have provided inconsistent results, and the definite means of inducing clinically useful therapeutic angiogenesis remain elusive. These less than optimal results may stem from a number of reasons, including the application of a single growth factor that may not be sufficient to meet the multifaceted challenge for developing efficient induction of collateral vessels, the need for more sustained growth factor delivery in order to establish more stable vessels, and the need to target arteriogenesis rather than angiogenesis to achieve a more significant increase in perfusion. Therefore, efforts in the field are moving toward the use of different cell therapies for these ischemic conditions, as well as using combined cell and gene delivery strategies to achieve better outcomes. For example, a recent trial has used combined delivery of endothelial and smooth muscle cells (each cell type modified to secrete a different angiogenic growth factor) in CLI patients. For heart failure, gene therapy trials have focused on restoring the abnormal calcium handling characteristic of failing human cardiomyocytes.15 Because a reduction in levels of the sarcoplasmic reticulum calcium ATPase (SERCA2a), the sarcoplasmic reticulum calcium pump, was found to be a key factor in the alteration of calcium cycling in heart failure, this protein became an attractive clinical target for gene delivery purposes. Overexpression of SERCA2a levels by cardiomyocyte gene delivery has led to the restoration of previously abnormal calcium transients and to improved cardiac contractility, reduction of the frequency of arrhythmias, and improved oxygen utilization in animal models of heart failure. More recently, the clinical benefits of overexpressing SERCA2a have been demonstrated in phase I and II of the Calcium Upregulation by Percutaneous Administration of Gene Therapy in Cardiac Disease (CUPID) trials. A4  These studies demonstrated that AAV delivery of the SERCA2a transgene by intracoronary delivery is feasible and safe, results in persistent expression of the transgene, and is associated with a significant improvement in associated biochemical alterations and clinical symptoms of heart failure in the treated patients. ,

Cystic Fibrosis

Experimental protocols for gene therapy for cystic fibrosis (CF) (Chapter 89) have been implemented since 1990. The cystic fibrosis transmembrane conductance regulator (CFTR) protein is mutated in patients with CF. Transducing the epithelium of the nasal and bronchial tree is potentially feasible through nonsystemic approaches. Nonviral gene therapy methods

One of the most exciting opportunities for gene therapy lies in the cancer arena. Gene therapy strategies targeting cancer can be grouped according to their proposed mechanisms of action and include gene therapies aiming to directly induce cytotoxic effects in cancer cells (through the use of oncolytic viruses or by the delivery of apoptotic inducers and suicide genes), gene therapies aiming to boost the immune response to tumor antigens, and gene therapies targeting the tumor microenvironment.

Direct Cytotoxic Effects

An interesting approach for cancer gene therapy is to harness the action of oncolytic viruses.16 Oncolytic viruses are therapeutically useful anticancer viruses that will selectively infect, amplify, and then damage cancerous tissues without causing harm to normal tissues. Cancer selectivity of the different oncolytic viruses takes advantage of defects commonly found across many tumor types, such as lack of antiviral responses, activation of Ras pathways, loss of tumor suppressors, and defective apoptosis. Oncolytic viruses can kill infected cancer cells in many different ways, ranging from direct virus-mediated cytotoxicity through a variety of cytotoxic immune effector mechanisms. Several viruses such as the Newcastle disease virus (which activates the innate or adaptive immune response), reovirus (which activates host protein kinases to shut down protein production), and mumps virus have an inherent ability to specifically target cancer cells and, upon virus replication, cause significant cell death and tumor regression. Other viruses (HSV, adenovirus, vaccinia virus, vesicular stomatitis virus, and poliovirus) need to be genetically engineered to engender oncolytic activity. Genetically engineered viruses and inherently antitumor-selective viruses are being tested in early and late clinical conditions to determine their effectiveness in specific types of cancer (e.g., metastatic melanoma and different brain tumors). Beyond the direct viral cytopathic effect, viral vectors can be used to deliver genes to cancer cells that will result in tumor cell death. The relevant transgenes encode for cellular proteins that are involved in apoptosis or prevent proliferation. The selectivity for the activation of such genes only in tumor cells is achieved either through the use of the aforementioned oncolytic viruses or by the expression of the transgenes under the control of promoters that are activated only in cancer cells, either as a general property of cancer (e.g., human telomerase or survivin) or in specific types of tumors (probasin in prostate cancer, ceruloplasmin in ovarian cancer, HER2 in breast cancer, and carcinoembryonic antigen in colon cancer). The most clinically advanced gene therapy drug against cancer is the replication-deficient adenovector expressing the human p53 gene. This therapy (Gendicine) is approved in China for the treatment of patients with head and neck squamous cell carcinoma by direct administration into the tumor bed. Another attractive approach is the use of suicide genes. Suicide gene therapy involves delivery of a pro-drug activating enzyme (suicide gene) that converts nontoxic pro-drugs to cytotoxic metabolites. The prototype for such a suicide gene/pro-drug combination is HSV thymidine kinase (TK)/ganciclovir (GCV). The TK gene is selectively expressed only in cancer cells (by one of the methods described previously), and after application of GCV, it converts it to the cytotoxic agent phosphorylated GCV. Interestingly, phosphorylated GCV is only toxic to dividing cells, further increasing the selectivity to the cancer cells. Other cytotoxic strategies are to express secreted pro-apoptotic proteins, such as tumor necrosis factor−related apoptosisinducing ligand (TRAIL) or cytotoxins such as Pseudomonas exotoxin.

Immunomodulatory Cell and Gene Therapy for Cancer and Autoimmune Disease

In recent years, the focus of gene- and cell-based therapy for cancer has shifted away from directly manipulating or targeting the cancer cells toward modulation of the immune system itself. Cytotoxic T-lymphocyte antigen 4 (CTLA-4) and programmed death 1 (PD-1) are two T-lymphocyte

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

proteins that have long been known to attenuate immune destruction of cancer cells. Blocking monoclonal antibodies to circumvent this attenuation have been shown to induce limited remissions in several forms of previously intractable metastatic tumors, including malignant melanoma. These partial successes have now revived still more sophisticated therapeutic approaches, based on the ex vivo personalized genetic engineering of cytotoxic T lymphocytes of cancer patients to enable the immune system to target tumor cells. Chimeric antigen receptor (CAR) therapy releases the encumbrance of major histocompatibility complex restriction in cancer antigen recognition by combining the antigen-binding site of a monoclonal antibody with the signal-activating machinery of the cytotoxic T lymphocytes. This enables combining a high level of target specificity typical of monoclonal antibodies with in vivo expansion and the potential for a durable response, as has been demonstrated in clinical treatment protocols in leukemia and other malignancies.17 It can be anticipated that CAR-modified T lymphocytes might also prove useful as a combined genetic engineering/cell therapy approach to the management of autoimmune disease.

Disrupting Tumor Microenvironment

Targeting the tumor microenvironment is another attractive approach for cancer gene therapy because it consists of normal cells that should not develop resistance to the therapy. The most obvious target is the tumor neovascularization process. The use of antiangiogenic drugs such as bevacizumab (Avastin), an anti−vascular endothelial growth factor monoclonal antibody, has shown success in clinical trials for some cancer cell types, but the effect may be transient or negligible in others. This may be because the angiogenesis process is complex, and inhibiting just one aspect may not be sufficient. Developing alternative strategies such as combination therapies, including targeting multiple angiogenic pathways, might be a better strategy, especially because inhibiting angiogenesis is cytostatic and not cytotoxic. A number of antiangiogenic factors (e.g., angiostatin) have been expressed in viral vectors and have been used in preclinical studies but have not reached the clinic yet.

Other Forms of Molecular Therapies: RNA Interference and Gene Editing RNA Interference

RNA interference (RNAi) regulates gene expression by a highly precise mechanism of sequence-directed gene silencing at the stage of translation by degrading specific messenger RNAs or by blocking their translation into protein. Research on the use of RNAi for therapeutic applications has gained considerable momentum. It has been suggested that many of the novel disease-associated targets that have been identified are amenable to conventional small molecule drug blockade and can potentially be targeted with RNAi. In the coming years, the concept of RNAi will be actively translated into a therapeutic option, with numerous early-phase trials already underway.

Gene Editing

The center of gravity for gene therapy may be shifting from gene restoration (where a whole new gene is pasted into the genome) to genome editing, whereby the pathogenic mutation is corrected in its natural gene location with zinc finger nucleases, transcription activator−like effector nucleases (TALENs), or clustered regulatory interspaced short palindromic repeats

213

(CRISPRs).18,19 These hybrid molecules act as highly specific “molecular scissors,” which are engineered to target a specific location in the genome and introduce a double-strand break in the DNA proximal to the targeted mutation. The cleavage in the DNA is then resolved by hom*ologous recombination between the endogenous genes and an exogenously introduced donor fragment containing the normal sequence. In this fashion, the pathogenic mutation is permanently changed back to the normal sequence. This also preserves the architecture of the genome and maintains gene control under the normal cellular regulatory elements. Consequentially, gene editing represents a paradigm shift in the way gene therapy could be performed. To date, gene editing techniques have been used to correct the disease-causing mutations associated with X-linked SCID, hemophilia B, sickle cell disease,20 and α1-antitrypsin deficiency and to repair Parkinson disease−associated mutations (SNCA gene) in patient-derived hiPSCs or in preclinical mouse models. Targeted gene knockout through similar technologies promises to be a potentially powerful strategy for combating HIV/AIDs. Zinc finger nucleases have been used to confer HIV-1 resistance by disabling the HIV coreceptor C-C chemokine receptor type 5 (CCR5) in primary T cells and hematopoietic stem/progenitor cells. This approach is currently used in clinical trials. Additionally, zinc finger nucleases have been used to improve the performance of T-cell-based immunotherapies by inactivating the expression of endogenous T-cell-receptor genes, thereby enabling the generation of tumor-specific T cells with improved efficacy profiles. Finally, site-specific nucleases may also bring a unique value to the conventional gene-adding approach by enabling insertion of therapeutic transgenes into specific “safe harbor” locations in the human genome, ensuring longterm expression of the transgene as well as reducing the potential for random insertional mutagenesis. It is important to mention that the use of site-specific nuclease technology at its current state requires the presence of proliferating cells, and its utility is therefore still relatively limited for nonproliferating somatic cells and for direct in vivo applications. Continued progress in stem cell research, including the production and manipulation of hiPSCs cells, will ultimately open countless new directions for gene therapy, including treatments based on autologous stem cell transplantation.

Grade A References A1. Clifford DM, Fisher SA, Brunskill SJ, et al. Stem cell treatment for acute myocardial infarction. Cochrane Database Syst Rev. 2012;2:CD006536. A2. Fisher SA, Brunskill SJ, Doree C, et al. Stem cell therapy for chronic ischaemic heart disease and congestive heart failure. Cochrane Database Syst Rev. 2014;4:CD007888. A3. Wang ZX, Li D, Cao JX, et al. Efficacy of autologous bone marrow mononuclear cell therapy in patients with peripheral arterial disease. J Atheroscler Thromb. 2014;21:1183-1196. A4. Zsebo K, Yaroshinsky A, Rudy JJ, et al. Long-term effects of AAV1/SERCA2a gene transfer in patients with severe heart failure: analysis of recurrent cardiovascular events and mortality. Circ Res. 2014;114:101-108.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

GENERAL REFERENCES 1. Daley GQ. The promise and perils of stem cell therapeutics. Cell Stem Cell. 2012;10:740-749. 2. Takahashi K, Yamanaka S. Induced pluripotent stem cells in medicine and biology. Development. 2013;140:2457-2461. 3. Atala A, Kasper FK, Mikos AG. Engineering complex tissues. Sci Transl Med. 2012;4:160rv112. 4. Doulatov S, Daley GQ. Development. A stem cell perspective on cellular engineering. Science. 2013;342:700-702. 5. Lindvall O, Barker RA, Brustle O, et al. Clinical translation of stem cells in neurodegenerative disorders. Cell Stem Cell. 2012;10:151-155. 6. Xin M, Olson EN, Bassel-Duby R. Mending broken hearts: cardiac development as a basis for adult heart regeneration and repair. Nat Rev Mol Cell Biol. 2013;14:529-541. 7. Pagliuca FW, Melton DA. How to make a functional beta-cell. Development. 2013;140:2472-2483. 8. Abelson S, Shamai Y, Berger L, et al. Intratumoral heterogeneity in the self-renewal and tumorigenic differentiation of ovarian cancer. Stem Cells. 2012;30:415-424. 9. Imaizumi Y, Okano H. Modeling human neurological disorders with induced pluripotent stem cells. J Neurochem. 2014;129:388-399. 10. Cartier N, Hacein-Bey-Abina S, Bartholomae CC, et al. Hematopoietic stem cell gene therapy with a lentiviral vector in X-linked adrenoleukodystrophy. Science. 2009;326:818-823. 11. Sheridan C. Gene therapy finds its niche. Nat Biotechnol. 2011;29:121-128.

213.e1

12. Candotti F, Shaw KL, Muul L, et al. Gene therapy for adenosine deaminase-deficient severe combined immune deficiency: clinical comparison of retroviral vectors and treatment plans. Blood. 2012;120:3635-3646. 13. Zhang L, Thrasher AJ, Gaspar HB. Current progress on gene therapy for primary immunodeficiencies. Gene Ther. 2013;20:963-969. 14. Cideciyan AV, Jacobson SG, Beltran WA, et al. Human retinal gene therapy for Leber congenital amaurosis shows advancing retinal degeneration despite enduring visual improvement. Proc Natl Acad Sci U S A. 2013;110:E517-E525. 15. Tilemann L, Ishikawa K, Weber T, et al. Gene therapy for heart failure. Circ Res. 2012;110: 777-793. 16. Russell SJ, Peng KW, Bell JC. Oncolytic virotherapy. Nat Biotechnol. 2012;30:658-670. 17. Porter DL, Levine BL, Kalos M, et al. Chimeric antigen receptor-modified T cells in chronic lymphoid leukemia. N Engl J Med. 2011;365:725-733. 18. Gaj T, Gersbach CA, Barbas CF 3rd. ZFN, TALEN, and CRISPR/Cas-based methods for genome engineering. Trends Biotechnol. 2013;31:397-405. 19. Tebas P, Stein D, Tang WW, et al. Gene editing of CCR5 in autologous CD4 T cells of persons infected with HIV. N Engl J Med. 2014;370:901-910. 20. Romero Z, Urbinati F, Geiger S, et al. beta-globin gene transfer to human bone marrow for sickle cell disease. J Clin Invest. 2013;123:3317-3330.

213.e2

CHAPTER 44  Regenerative Medicine, Cell, and Gene Therapies  

REVIEW QUESTIONS 1. Meniscus repair by constructing an implantable scaffold seeded with mesenchymal stem cells that differentiate into chondrocytes would be an example of which of the following? A. Ex vivo gene therapy B. Autologous tissue meniscus implantation C. Tissue engineering D. Direct stem cell reprogramming E. Somatic gene therapy Answer: C  Rather than simply introducing cells into a diseased area, in tissue engineering, cells are embedded or seeded onto three-dimensional scaffolds (derived from different biomaterials) before transplantation. This can involve either (i) cellular scaffolds that are seeded ex vivo with cells before their in vivo transplantation, or (ii) acellular scaffolds that require the recipient’s cells to repopulate the scaffold to reconstitute it after transplantation. To date, these efforts have mainly concentrated on the musculoskeletal system, as in the example presented. Ex vivo and somatic gene therapies by definition involve the transfer of specific genetic material into cells to correct or restore a cellular defect. Direct stem cell programming is the direct conversion of the phenotype of one cell type (e.g., fibroblasts) to another (e.g., chondrocytes). Autologous tissue meniscus implantation does not involve any type of cell and gene therapy. (See Tissue Engineering.) 2. To date, the only established clinical application of fetal-derived cell therapy has been in patients with which of the following? A. Parkinson disease B. Myocardial infarction C. Heart failure D. Blood product transfusion E. Cartilage repair Answer: A  Cells of fetal origin show enhanced proliferative capacity and enhanced ability to differentiate into mature or specialized cells. Although this form of cell therapy represents a form of regenerative medicine with great potential, the only fetal-derived cells that have been used in clinical applications to date are the dopaminergic cells derived from the developing fetal nervous system for the treatment of Parkinson disease. (See Engraftment of Fetal Tissue.)

3. Which one of the following in not a form of gene therapy? A. Transfection B. Transduction C. RNA interference D. Gene editing E. DNA electroporation Answer: E  Electroporation is a strictly in vitro research method in cell biology that electropermeabilizes cell membranes by an externally applied electrical field to introduce a piece of DNA (or other agents) into a cell. All the other choices are methods of gene therapy. Transfection is the direct delivery of naked DNA by injection, the use of liposomes, nanoparticles, and other means. Transduction is gene therapy that is mediated by viral vectors. RNA interference is a highly precise mechanism of regulating gene expression by sequence-directed silencing (by degrading specific messenger RNAs or by blocking their translation into protein). Gene editing is a way of correcting a pathogenic mutation in its natural gene location (rather than the more conventional gene therapy method of gene restoration, in which a whole gene is inserted into the genome). (See Gene Therapy Delivery Methods and Other Forms of Molecular Therapies: RNA Interference and Gene Editing under the main heading of Gene Therapy.)

216

CHAPTER 45  The Innate Immune System  

45  THE INNATE IMMUNE SYSTEM MARY K. CROW

THE INNATE IMMUNE SYSTEM IN HOST DEFENSE AND DISEASE PATHOGENESIS

The immune system, comprising cells, the molecules they produce, and the organs that organize those components, evolved over millions of years in response to infections with pathogenic microorganisms.1 Its essential role in maintaining health is based on its recognition and elimination or control of those foreign microbes. Central to the success of the protective role of the immune system is its capacity to distinguish foreign and dangerous invaders from self-components.2,3 In addition to its contributions to host defense, the immune system is involved in the prevention of malignancy by surveying and recognizing self-cells that express novel antigens,4 and it also plays a role in resolution and repair of tissue damage. The immune system is generally described as including an innate immune system and an adaptive immune system. The former provides the first and rapid line of defense and cellular response to a foreign stimulus. The latter, dependent on activation by the innate immune response, develops a more specific response targeted to the offending organism and generates memory for that stimulus that can be elicited rapidly should that organism be encountered again on a later occasion. Immune system cells derive from precursor cells of the hematopoietic lineage and populate discrete lymphoid organs, including lymph nodes, spleen, and thymus, as well as skin and intestine. Cells of the innate immune system serve as sentinels at locations that are likely to encounter foreign organisms, and after activation they will often travel to a local lymphoid organ. The induction of the adaptive immune response occurs in the context of structured aggregates of innate and adaptive immune cells in the lymphoid organs. Once activated and differentiated to produce effector molecules, immune system cells can be sampled in blood as they travel to sites of infection or tissue damage. There they can interact directly with target cells to mediate cell death or, alternatively, provide activating signals to expand or regulate a response, or secrete high local levels of immunomodulatory substances called cytokines. Cytokines are small soluble proteins that communicate among cells within the immune system or between immune system cells and cells in other tissues.5 The cells and products of the immune system function as an exquisitely regulated complex system.6 Inherited variations in hundreds of genes have evolved, under pressure of microbial challenge, to ensure adequate defense against pathogenic organisms across the human population.1 However, in any one individual, the composite genetic profile can generate predisposition to infection or, alternatively, autoimmune or inflammatory disease. The innate immune response was traditionally viewed as mediating nonspecific protection through the production of preformed effector molecules. However, important advances in characterization of the cell surface and intracellular pattern recognition receptors (PRR), particularly the toll-like receptor (TLR) family, and signaling pathways used by innate immune cells to implement a defensive response are now understood to have relative specificity for pathogen-associated molecular patterns (PAMPs) that are characteristic of categories of microbes.7,8 In contrast to those receptor systems that initiate an innate immune response, the protein products that implement the response, whether to expand the reaction to additional cells, promote trafficking to the most relevant location, or shape the differentiation programs of adaptive immune system cells, do not show specificity based on the initial triggering stimulus. The products of the innate immune response can be highly effective at ablating or limiting the extent of infection and can generate a tissue repair program that establishes a satisfactory resolution of the episode of infection. However, when sustained or poorly regulated, they can represent an important pathophysiologic mechanism for many autoimmune and inflammatory diseases.

Cells of the Innate Immune System Monocytes and Macrophages

Monocytes circulate in the peripheral blood with a half-life of 1 to 3 days. Macrophages arise from monocytes that have migrated out of the circulation

and have proliferated and differentiated in tissue. Tissue macrophages include alveolar macrophages in the lung, Kupffer cells in the liver, osteoclasts in bone, microglia in the central nervous system, and type A synoviocytes in the synovial membrane. Macrophages secrete myriad products, including hydrolytic enzymes, reactive oxygen species, cytokines, and chemokines. Macrophages engulf microorganisms and foreign particles directly or are activated by protein complexes containing antibodies that bind to cell surface receptors for the Fc portion of immunoglobulin molecules (Fc receptors, or FcRs). These encounters activate intracellular signaling pathways that induce transcription of target genes, primarily those encoding mediators that promote inflammation or enzyme-mediated death of the microbe. Cytokines from other immune system cells, including interferon (IFN)-γ or interleukin (IL)-4, can drive macrophage differentiation toward the production of mediators that are primarily pro-inflammatory or to a wound healing functional profile. Researchers have characterized those functional phenotypes as M1 or M2, although it is recognized that the context of an innate immune response will determine the functional response, with composite profiles common.9 In addition to responding to foreign microbes, macrophages contribute to the elimination of senescent or apoptotic cells in a manner that avoids induction of an inflammatory response. Macrophages also interact with other cell types through complementary cell surface adhesion or costimulatory receptors. After capturing antigen, they can function as antigen-presenting cells for T lymphocytes, and they can interact with non−immune system cells such as endothelial cells or fibroblasts.

Dendritic Cells

Dendritic cells (DCs) comprise a complex family of cells that perform essential functions in the innate immune response and serve as a bridge to activation of an adaptive immune response. Myeloid dendritic cells can incorporate antigens derived from invading microbes, travel to nearby lymph nodes, and present processed antigenic peptides to T lymphocytes (T cells) in the form of peptide−major histocompatibility complex (MHC) molecule complexes. They are the most effective antigen-presenting cells based on expression of cell surface costimulatory molecules, and they produce cytokines, including IL-12 and IL-23, after interaction with PAMPs. They thereby contribute to the shaping of the T-cell differentiation program to generate effector cell functions. Plasmacytoid dendritic cells (pDCs) have been identified as highly effective producers of type I interferon, a key mediator of host defense against viral infections.

Natural Killer Cells

Natural killer (NK) and NK T cells provide early defense against viral infections and other intracellular pathogens while adaptive responses are developing.10 NK cells are sensitized by cytokines, including type I interferons, released from pDCs and macrophages, and secrete abundant IFN-γ, which activates macrophages and other cells. They also are poised to kill virusinfected cells by injecting pore-forming enzymes and granzymes. Activation of NK cells is inhibited by interaction with self-MHC class I molecules on target cells. When those self-histocompatibility antigens are not present, NK cell−mediated killing is implemented. NK cells are important in tumor surveillance because they are able to kill MHC class I–deficient tumor cells that are no longer susceptible to adaptive immune responses. In addition to NK cells, a type of lymphocytes, so-called innate lymphoid cells, which participate early in innate immune responses but do not express rearranged receptors, is a focus of current study.11

Neutrophils

Neutrophils are the most abundant circulating white blood cells. They are recruited rapidly to inflammatory sites and can phagocytose and digest microbes (Chapters 167 and 169). Activation of neutrophils and phagocytosis is facilitated through the triggering of FcRs or complement receptors. Microbe-containing phagosomes fuse with lysosomes, which contain enzymes, proteins, and peptides that inactivate and digest microbes. Beyond their phagocytic capability, neutrophils produce a variety of toxic products. The release of toxic products is known as the respiratory burst because it is accompanied by an increase in oxygen consumption. During the respiratory burst, oxygen radicals are generated by nicotinamide adenine dinucleotide phosphate (NADPH) oxidases. Neutrophils also contribute to host defense through extrusion of DNA and associated proteins in the form of neutrophil extracellular traps, or NETs, to which bacteria can stick, facilitating their clearance. Despite their effective contributions to the innate immune

CHAPTER 45  The Innate Immune System  

response and microbial host defense, neutrophils can generate considerable collateral damage. NETs have the capacity to induce production of cytokines by pDCs and may damage vascular endothelial cells. Secretion of neutrophil granule contents, particularly their enzymes (myeloperoxidase, elastase, collagenase, and lysozyme), causes direct cellular injury and damages macromolecules at inflamed sites.

Eosinophils

In contrast to macrophages and neutrophils, eosinophils are only weakly phagocytic but are potent cytotoxic effector cells against parasites. Their major effector mechanism is the secretion of cationic proteins (major basic protein, eosinophil cationic protein, and eosinophil-derived neurotoxin). These proteins are released into the extracellular space, where they directly destroy the invading microorganism but can also damage host tissue (Chapter 170).

Basophils and Mast Cells

Basophils and tissue mast cells secrete inflammatory mediators such as histamine, prostaglandins, leukotrienes, and some cytokines.12 Release of these substances is triggered when cell surface immunoglobulin E (IgE) receptors encounter monomeric IgE. They play a role in atopic allergies, in which allergens bind immunoglobulin (IgE) and cross-link FcεRs. Mast cells have been observed in rheumatoid arthritis synovial tissue and have been implicated in local inflammatory responses (Chapter 255). Like pDCs and macrophages, mast cells express TLRs and FcRs and produce cytokines after encountering immune complexes composed of TLR ligands.

Recognition Receptors and Triggers of an Innate Immune Response Toll-like Receptors

The innate immune system utilizes both cell surface and intracellular PRRs to recognize conserved structures on microbes (PAMPs). Examples of PAMPs are bacterial lipopolysaccharides, peptidoglycans, mannans, bacterial DNA, double-stranded RNA, and glucans. The discovery and characterization of the TLR family of receptors and their relevant ligands has focused attention on the mechanisms that allow an innate immune response to shape the nature of the resulting inflammatory or repair programs, as well as the T-cell effector cell functions that follow recognition of antigens from the relevant pathogen. The TLRs have in common leucine-rich domains and bind PAMPs common to classes of pathogenic organisms.7,8 For example, TLR-4, a cell surface−expressed PRR, binds lipopolysaccharide of gram-negative bacteria, and TLR-2 recognizes bacterial peptidoglycans and lipoproteins, often based on dimerization with other TLR family members. Important advances in understanding systemic autoimmune diseases have followed the characterization of endosomal TLRs with relative specificity for single-stranded RNA (TLR-7 and TLR-8), demethylated CpG-enriched DNA (TLR-9), and double-stranded RNA (TLR-3, which has both cell surface and endosomal forms). The distribution of particular TLRs among cells of the innate immune system varies, and additional members of the TLR family may still be discovered and characterized. The TLRs play central roles in alerting the immune system that a microbe, typically a bacteria in the case of TLR-2 and TLR-4 or a virus in the case of TLR-3, TLR-7, TLR-,8 and TLR-9, is threatening the host. But in some cases, when an immune complex with self-nucleic acid gains access to an endosomal TLR, a self-directed innate immune response can be initiated or amplified.

Cytoplasmic Nucleic Acid Sensors

Following the description of the TLR family and the capacity of the endosomal TLRs to recognize microbial and self-nucleic acids, a second category of intracellular innate immune system receptors was defined that recognize RNA or DNA from microbes, primarily viruses, that gain access to the cell cytoplasm. The DExD/H-box family of helicases include retinoic acid−inducible gene I (RIG-I) and melanoma differentiation−associated protein 5 (MDA5), described as members of the RIG-I-like receptor (RLR) family that recognizes viral RNAs with particular structural characteristics that distinguish the viral RNA from most host RNAs (Fig. 45-1).13,14 Cytoplasmic DNA receptors have also been defined, with cyclic guanosine monophosphate−adenosine monophosphate synthase (cGAS) recently identified as an important sensor of cytoplasmic DNA that triggers an innate immune response after interacting with the stimulator of interferon genes (STING).15 Whether RNA or DNA triggers these cytoplasmic sensors, the result is transcription and production of interferon-β and other pro-

217

inflammatory cytokines that orchestrate the early phase of an antiviral immune response.

NOD Receptors

Another category of intracellular receptors is proving important in antimicrobial defense as well as contributing to activation of inflammatory states. The nucleotide-binding oligomerization domain (NOD)-like receptor (NLR) family comprises components of an intracellular structure called the inflammasome, a signaling platform that organizes innate immune system activation in response to some stimuli.16 The inflammasome can activate caspase 1, an enzyme important for maturation of the pro-inflammatory cytokines IL-1β and IL-18. The NLRP3-containing inflammasome has been best studied and implicated in the inflammatory response to monosodium urate crystals, the triggers of gout attacks (Chapter 273). Mutations in the NLRP3 gene are the basis of chronic autoinflammatory syndromes that are associated with exaggerated production of IL-1 (reviewed in Chapter 261).

C-Type Lectin Receptors

Members of the C-type lectin receptor family have a carbohydrate recognition domain and a calcium-binding domain that promotes signaling after interaction with carbohydrate-expressing microbes as well as self-molecules. DC-SIGN (DC-specific intracellular adhesion molecule-3 grabbing nonintegrin) is an example of a family member that recognizes high-mannosecontaining structures on foreign antigens and supports DC activation. Mannose receptors on macrophages, dendritic cells, and other cell types, such as renal mesangial cells, participate in clearance of microbes as well as antigen trapping for presentation to adaptive immune system cells. The selectin family of proteins have a lectin domain, bind to carbohydrate ligands, and mediate the first steps of leukocyte migration. L-selectin is present on virtually all leukocytes; P-selectin and E-selectin are expressed on activated endothelial cells, and P-selectin is also stored in platelets. Selectins capture floating leukocytes and initiate their attachment and rolling on activated endothelial cells.

Scavenger Receptors

Scavenger receptors comprise a diverse family of receptors with the common functional role of binding various ligands and transporting or removing nonself or altered-self targets.17 They can participate in clearance of microorganisms and cholesterol transport but can also contribute to disease pathology. For example, among the scavenger receptors is the receptor for oxidized low-density lipoproteins, which can promote generation of lipid-laden macrophages and atherosclerosis when accumulated in excess, and receptors for relatively inert substances such as silicon, which can drive an inflammatory response once taken into phagocytic cells. Scavenger receptors can also participate in activation of the inflammasome, as can occur after binding serum amyloid A protein.

Inhibitory Natural Killer Cell Receptors

The immunoglobulin-like killer inhibitory receptor (KIR) family of receptors participates in distinguishing self-cells from cells of foreign origin or tumor cells expressing modified-self-molecules. NK cells are ready to produce their toxic mediators, but they are held in check by inhibitory receptors that recognize MHC class I or MHC class I–like molecules.10 Recognition of MHC class I molecules provides a negative signal that suppresses cell activity. The observation that NK cells kill target cells lacking MHC class I molecules recognized as self led to the missing-self hypothesis. By screening cell surfaces for the expression of MHC class I molecules, the innate immune system collects information about the intactness of tissues, emphasizing the crucial role of MHC class I molecules as markers of tissue integrity.

Fc and Complement Receptors

Most cells of the innate immune system possess receptors (FcRs) that specifically interact with the constant region (Fc portion) of immunoglobulins and can bind antibodies attached to antigens. The isotype of the antibody determines which cell type is activated in a given response. Triggering of most FcRs transmits activating signals; however, inhibitory FcRs on B lymphocytes (B cells) and macrophages can limit responses. Ligation of an FcγR on macrophages or neutrophils triggers phagocytosis of the antigen, activation of respiratory burst, and induction of cytotoxicity. On NK cells, FcγRs initiate antibody-dependent cell-mediated cytotoxicity. FcRs on pDCs are important for bringing immune complexes into intracellular compartments containing endosomal TLRs. FcRs on mast cells, basophils, and activated

218

CHAPTER 45  The Innate Immune System  

IFNα/β

IFNα/β IFNα/β IFNα/β

IFNα/β receptor

Virus

Cytoplasm

P

Tyk2

Jak1

P

RNA P RIG-I

STAT 1

P STAT 2 P IRF-9 P

MDA5

RIG-I NEMO IPS-1

IKKα IKKβ

TRAF3

IRF-3

IKKε Mitochondria

TBK1 P

IκB NF κB

IRF-3 P P

P

STAT 1 STAT 2

P P

IRF-9 ISGF3

IκB IRF-3 P IRF-3 P

Nucleus P

P CBP/ IRF-3 NF κB p300 IRF-3 P

STAT 1 STAT 2

P

ISGs

P

IRF-9 IFN-β IFN-α FIGURE 45-1.  Induction of antiviral type I interferon response. Cytoplasmic sensors of RNA, including RIG-I and MDA5, trigger a signaling cascade that results in translocation of IRF-3 to the nucleus and transcription of interferons. Those cytokines promote an antiviral immune response after binding to their receptor and activating the JAK-STAT pathway. CBP/p300 = CREB binding protein; NEMO = NF-κB essential modulator; IFN = interferon; IKK = inhibitor of nuclear factor κB kinase subunit; IPS-1 = interferon-β promoter stimulator-1; IRF = interferon response factor; ISG = interferon stimulated gene; ISGF3 = interferon-stimulated gene factor 3; JAK = Janus kinase; MDA5 = melanoma differentiation-associated protein 5; RIG-1 = retinoic acid−inducible gene 1; STAT = signal transducer and activator of transcription; TRAF3 = TNF (tumor necrosis factor) receptor−associated factor; Tyk = tyrosine kinase. (From Wilkins C, Gale M Jr. Recognition of viruses by cytoplasmic sensors. Curr Opin Immunol. 2010;22:41-47.)

eosinophils bind monomeric IgE with extremely high affinity. Cross-linking of the constitutively cell surface–bound IgE induces cell activation and the release of cytoplasmic granules. Some immunoglobulin isotypes fix complement, and complement receptors on monocytes amplify cell activation induced by antigen-antibody-complement immune complexes18 (Chapter 50). Complement receptor 1 (CR1) binds C3b and C4b, initial degradation products of complement activation, and when activated promotes phagocytosis of a complement-bearing immune complex. CR3 and CR4 are β2integrins and bind the degradation product iC3b.

of activated macrophages but also binds to those cells through its specific receptor, expanding an inflammatory response. Innate immune cells also express receptors for IL-6, which induces acute phase reactants and type I interferon, which orchestrates a broad host defense program in response to virus infection (see Fig. 45-1). Chemokine receptors include many family members that are differentially distributed among immune system cells and sense the gradient generated by soluble chemokines, resulting in attraction of cells to sites where they are needed to implement inflammatory or immune functions.

Cytokine and Chemokine Receptors

Signaling Pathways and Effector Mediators of the Innate Immune System

Cells of the innate immune system express receptors for many cytokines, soluble, low-molecular-weight glycoproteins that derive from many cellular sources.5 Binding of IFN-γ, produced by NK or type 1 helper T cells (TH1 cells), by its receptor on monocytes activates a differentiation program that expands an inflammatory response. Receptors for IL-4 on monocytes induce a gene transcription program that is more supportive of a wound healing and repair program. Tumor necrosis factor-α (TNF-α) is a product

Each family of innate immune system receptors utilizes a complex network of molecules to transmit information from the cell surface or its cytoplasm to the nucleus, resulting in induction of a broad gene transcription and protein synthesis program that implements the next phase of the response. The contributions of each of the signal transduction pathways to the overall innate immune response will depend on the proteins produced and will determine whether the resulting cell products focus the overall immune

CHAPTER 45  The Innate Immune System  

function on ablating the damaging effects of virus infection on the host, limiting the inflammation and tissue damage that follow a bacterial or fungal infection, or healing a tissue wound through the production of scar tissue.

Receptor-Mediated Signaling Pathways

Certain common cell signaling systems are utilized by many cells and receptor systems.6,7 Arguably the most important is the nuclear factor κ light-chain enhancer of activated B cells (NF-κB) pathway. NF-κB is a rapid-acting transcription factor because it is preformed in cells of the innate immune system and does not require new protein synthesis to take action. Its activity is induced by ligation of TLRs and many cytokine receptors. Its component transcription factors translocate to the cell nucleus after degradation of an inhibitory component, inhibitor of κB (IκB), and bind to promoter regions of genes encoding mediators of inflammation and cell proliferation. Another important pathway is mediated by the interferon regulatory factor (IRF) family, including transcription factors that are activated by endosomal TLRs in response to ligation by DNA or RNA, or by cytoplasmic nucleic acid sensors, usually from viral sources. IRF-3 is particularly important for promoting transcription of interferon-β, typically produced early in an antivirus innate immune response. IRF-7 is particularly supportive of interferon-α production induced by endosomal TLRs and is constitutively present in pDCs, the most active producers of IFN-α. The Janus kinase ( JAK)-signal transducer and activator of transcription (STAT) pathway is utilized by many cytokine receptors and involves sequential enzymatic reactions by kinases that eventuate in translocation of STAT proteins to the nucleus, where they bind to gene promoters and induce transcription and production of products important in implementing immunoregulation and inflammation. TNF receptor family members activate a complex signaling pathway that involves proteins called TNF receptor−associated death domain (TRADD) proteins and TNF receptor−associated factors (TRAFs), ultimately activating the NF-κB and the mitogen-activated protein (MAP) kinase pathways. The TGF-β receptor is a serine/threonine receptor kinase that phosphorylates cytoplasmic proteins of the SMAD family, which act as transcription factors after receptor engagement by TGF-β. TGF-β signaling can play an important role in terminating an innate immune response and initiating a wound healing or tissue repair program. It is apparent that common intracellular signaling strategies are used by many of the receptor systems that activate and regulate the innate immune system, with ligand-receptor engagement triggering the activation of kinases that phosphorylate downstream pathway proteins, and result in translocation of important transcription factors from cytoplasm to nucleus where new gene transcription takes place.

Soluble Products of the Innate Immune Response

Cells of the innate immune system are the principal producers of many proinflammatory and regulatory cytokines already mentioned, and are also their targets. In addition to the cytokines described, cells of the innate immune system produce chemokines that attract immune system cells to sites of tissue damage or infection, and they produce cell survival and differentiation factors that help to develop an adaptive immune response. Macrophages and dendritic cells produce IL-12 and IL-23 to support development of effector T-cell programs, and they produce B-cell-activating factor (BAFF), a soluble mediator of the TNF family. BAFF supports B-cell survival and can provide costimulatory signals to B cells that have received antigen-specific activation signals through their surface B-cell antigen receptors, promoting differentiation to antibody-producing plasma cells. A particularly important set of products includes components of the complement system, a group of plasma enzymes and regulatory proteins that are converted from inactive pro-enzymes to active enzymes in a controlled and systematic cascade, which is crucial in linking microbial recognition to cellular effector function (Chapter 50). Mannose-binding lectin circulates in the plasma, functioning as an opsonin, and is involved in activation of the complement pathway. C-reactive protein, an acute phase protein, participates in opsonization by binding to bacterial phospholipids. Macrophages and neutrophils are important in the initiation phase of an innate immune response through their production of antimicrobial defensins, cysteine-rich cationic proteins, and cathelicidin peptides, such as LL37.19 Both categories of mediators can assist in killing of microbes in phagosomes. Neutrophils extrude stimulatory DNA in the form of NETs or release mitochondrial DNA, along with DNA-associated proteins like high mobility group box 1 (HMGB1) that amplifies TLR responses in pDCs or macrophages.

219

Role of the Innate Immune System in Localization, Extension, and Resolution of a Host Defense Reaction Localization of Innate Immune System Cells

Most cells of the innate immune response are free agents, moving through blood or lymph in transit from one site to another. Mobility of the cellular constituents of the innate immune system is required for effective initiation of a response to invading microbes. Cells use a multistep process of adherence and activation. Initially, leukocytes roll on activated endothelial cells, activate chemokine receptors, increase adhesiveness, and eventually migrate through the endothelial layer across a chemokine gradient. The selectin family of proteins mediates the first steps of leukocyte migration. P-selectin and E-selectin are expressed on activated endothelial cells, and P-selectin is also stored in platelets. Selectins capture floating leukocytes and initiate their attachment and rolling on activated endothelial cells. To transform attachment and rolling into firm adhesion, the concerted action of chemokines, chemokine receptors, and integrins is necessary. Integrins are heterodimers formed of many different α chains and β chains; different α/β combinations are expressed on different cell subsets. Only after activation can integrins interact with ligands on endothelial cells. Activation involves modification of the cytoplasmic domain of the β chain, which leads to a structural change of the extracellular domains. This process is termed inside-out signaling. The last step of homing is transendothelial migration. Here, the firmly attached leukocytes migrate through the endothelial cell monolayer and the basem*nt membrane of the vessel wall.

Transition to an Adaptive Immune Response

Movement of innate immune system cells is also required to transition a host response from primarily one depending on cells of the innate immune system to one that engages T and B lymphocytes. Dendritic cells resident in the skin and gut serve as sentinels and a first line of defense against invading organisms. When those cells are activated following sensing of PAMPs by PRRs and following uptake of microbial components by those cells, the DCs migrate to local lymph nodes where their contents, by now expressed on their surface in association with MHC class I or II molecules, can be sampled by T cells. As noted, activated macrophages, DCs, and pDCs produce cytokines that shape the differentiation program of T cells. In addition, cell surface costimulatory molecules induced after TLR-mediated activation, such as CD80 and CD86, provide essential accessory activation signals to T cells to ensure their effective activation. Macrophages and DCs also support the development of an adaptive immune response through their production of survival and differentiation factors. Chapter 46 provides a full description of the adaptive immune system and its implementation.

Role of Innate Immune System Cells in Resolution of an Immune Response and Wound Repair

Macrophages are particularly important in resolving an immune response and organizing the repair of damaged tissue. A classic paradigm describing pro-inflammatory/classically activated (M1) and anti-inflammatory/alternatively activated (M2) macrophages (see earlier under Monocytes and Macrophages) is likely to be overly simplistic. Yet it is clear that in the course of a chronic infection, macrophages can shift their functional profile from M1 to M2, in some cases promoted by the T-cell cytokines IL-4 and IL-13, to develop a gene expression program that includes production of TGF-β, supportive of a fibrotic response, and IL-10, a cytokine that inhibits antigenpresenting cell function.9 Although an M1-like profile driven by IFN-γ is highly productive in achieving initial control over a pathogenic invading microbe, and M2-derived mediators promote wound healing, it should be recognized that either macrophage phenotype, and complex in-between profiles, can also be associated with pathologic states (Fig. 45-2). Current research is unraveling the innate immune mechanisms that account for such diverse diseases as atherosclerosis (Chapter 70), viewed as associated with M1 macrophages, and idiopathic pulmonary fibrosis (Chapter 92), possibly involving M2-like macrophages.

Contribution of the Innate Immune Response to Pathogenesis of Autoimmune Disease

Among the most significant insights of the past decade is the essential contribution of the innate immune system to the pathogenesis of autoimmune and inflammatory diseases. As described, the cells of the innate immune system are integral players in the early recognition of invading pathogenic microbes, and when the functions of this complex system are carefully

Chronic phase

Mycobacteria

Viruses

Granuloma

Helminths

Breast cancer

Glomerulonephritis

Steatosis

Atopic dermatitis

Arthritis

Wound healing

Atherosclerosis

M1

Diabetes

M2

Fibrosis

Airway inflammation (asthma)

FIGURE 45-2.  Schematic representation of macrophage plasticity and polarization in pathology. Dynamic changes occur over time with evolution of pathology: for instance, a switch from M1 to M2 macrophage polarization characterizes the transition from early to chronic phases of infection. Moreover, mixed phenotypes or populations with different phenotypes can coexist. (From Sica A, Mantovani A. Macrophage plasticity and polarization: in vivo veritas. J Clin Invest. 2012;122:787-795.)

orchestrated and balanced, the result is efficient ablation, or at least isolation, of the microbe. However, if the microbe is not effectively cleared from the system and persists, a chronic state of infection associated with immune activation and tissue damage is the result. Interestingly, many parallels can be seen between the immune alterations observed in the setting of chronic viral infection and the impaired immunoregulation characteristic of the prototypic autoimmune disease systemic lupus erythematosus. Excessive production of interferon-α is a feature of most patients with that disease, and it is now understood that activation of the endosomal TLRs by nucleic acid−containing immune complexes amplifies the activity of the innate immune response and drives production of interferon-α and other proinflammatory cytokines. Neutrophils are now recognized to contribute to the induction of that response through their production of HMGB1, cathelicidins, and extrusion of stimulatory DNA aggregates. TLR activation is proposed to contribute to many additional autoimmune and inflammatory diseases; as endogenous TLR ligands can act as effective TLR stimuli in the setting of a pro-inflammatory environment associated with oxidative cell damage. The inflammasome and its component proteins, including the NOD-like receptors, are recognized as mediators of inflammatory responses induced by urate crystals that result in gout attacks (Chapter 273), and they are targets of mutations that define dramatic autoinflammatory syndromes (Chapter 261), particularly seen in children.

Conclusion

The cells and products of the innate immune response, for many years viewed as less sophisticated and important than the highly specific T and B lymphocytes of the adaptive immune response, have taken their place as essential defenders against pathogenic microbes. Through the recognition of common molecular patterns characteristic of microbes by members of receptor families, some still being discovered, the cells of the innate immune response orchestrate the effector programs that are fine-tuned to target the vulnerabilities of each pathogen and kill, or at least limit the expansion of, that microbe. Advances in understanding the mechanisms utilized by the innate immune

response and the clinical syndromes that result when components of that system are genetically altered, have elucidated the central role that receptors and products of the innate immune system play in the pathogenesis of autoimmune and inflammatory diseases. These insights are guiding efforts to develop targeted therapies that will leverage the new knowledge to control or even prevent human diseases in which the innate immune system plays an important pathogenic role. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 45  The Innate Immune System  

GENERAL REFERENCES 1. Quintana-Murci L, Clark AG. Population genetic tools for dissecting innate immunity in humans. Nat Rev Immunol. 2013;13:280-293. 2. Iwasaki A, Pillai PS. Innate immunity to influenza virus infection. Nat Rev Immunol. 2014; 14:315-328. 3. Busca A, Kumar A. Innate immune responses in hepatitis B virus (HBV) infection. Virol J. 2014;11:22. 4. Marcus A, Gowen BG, Thompson TW, et al. Recognition of tumors by the innate immune system and natural killer cells. Adv Immunol. 2014;122:91-128. 5. Torrado E, Cooper AM. Cytokines in the balance of protection and pathology during mycobacterial infections. Adv Exp Med Biol. 2013;783:121-140. 6. Zak DE, Tam VC, Aderem A. Systems-level analysis of innate immunity. Annu Rev Immunol. 2014;32:547-577. 7. Broz P, Monack DM. Newly described pattern recognition receptors team up against intracellular pathogens. Nat Rev Immunol. 2013;13:551-565. 8. O’Neill LA, Golenbock D, Bowie AG. The history of toll-like receptors: redefining innate immunity. Nat Rev Immunol. 2013;13:453-460. 9. Sica A, Mantovani A. Macrophage plasticity and polarization: in vivo veritas. J Clin Invest. 2012;122:787-795.

220.e1

10. Terabe M, Berzofsky JA. The immunoregulatory role of type I and type II NKT cells in cancer and other diseases. Cancer Immunol Immunother. 2014;63:199-213. 11. Hazenberg MD, Spits H. Human innate lymphoid cells. Blood. 2014;124:700-709. 12. Cromheecke JL, Nguyen KT, Huston DP. Emerging role of human basophil biology in health and disease. Curr Allergy Asthma Rep. 2014;14:408. 13. Schlee M. Master sensors of pathogenic RNA - RIG-I like receptors. Immunobiology. 2013;218: 1322-1335. 14. Wu J, Chen ZJ. Innate immune sensing and signaling of cytosolic nucleic acids. Annu Rev Immunol. 2014;32:461-488. 15. Cai X, Chiu YH, Chen ZJ. The cGAS-cGAMP-STING pathway of cytosolic DNA sensing and signaling. Mol Cell. 2014;54:289-296. 16. Latz E, Xiao TS, Stutz A. Activation and regulation of the inflammasomes. Nat Rev Immunol. 2013;13:397-411. 17. Canton J, Neculai D, Grinstein S. Scavenger receptors in homeostasis and immunity. Nat Rev Immunol. 2013;13:621-634. 18. Holers VM. Complement and its receptors: new insights into human disease. Annu Rev Immunol. 2014;32:433-459. 19. Silva PM, Gonçalves S, Santos NC. Defensins: antifungal lessons from eukaryotes. Front Microbiol. 2014;5:97.

220.e2

CHAPTER 45  The Innate Immune System  

REVIEW QUESTIONS 1. Which of the following cell types is not considered to be a component of the innate immune system? A. Macrophages B. Neutrophils C. T lymphocytes D. Plasmacytoid dendritic cells E. Eosinophils Answer: C  T lymphocytes are important components of the adaptive immune system. T and B lymphocytes use mechanisms that rearrange DNA to form novel specific antigen-binding receptors. Cells of the innate immune system express pattern recognition receptors but do not express specific antigen-binding receptors. 2. Endosomal toll-like receptors (TLRs) recognize which of the following pathogen-associated molecular patterns (PAMPs)? A. Flagellin B. Lipopolysaccharide C. Antigenic peptides D. Nucleic acids Answer: D  TLRs are important innate immune system receptors that recognize patterns expressed by pathogenic microbes and some endogenous molecules. Cell surface−expressed TLRs recognize molecules that are typically expressed on the surface of microbes. Intracellular endosomal TLRs, such as TLR-3, -7, -8, and -9, recognize RNA or DNA. The sequestering of those nucleic acid−responsive TLRs protects the immune system from inadvertent activation by self-nucleic acids. However, in diseases such as systemic lupus erythematosus, nucleic acid−containing immune complexes can gain access to the endosomal TLRs and induce an innate immune response. 3. Which of the following innate immune system stimuli utilizes the inflammasome to trigger an inflammatory disease? A. Peptide−major histocompatibility class (MHC) class II complex B. Monosodium urate crystals C. Interleukin-6 (IL-6) D. Immunoglobulin E E. Complement Answer: B  Urate crystals access the components of the NOD-like receptors of the inflammasome, activate caspase I, and induce the formation of IL-1, a pro-inflammatory mediator that can amplify an innate immune response. In some patients, this response leads to the acute inflammatory arthritis known as gout. Peptide−MHC class II complexes are stimuli for activation of an adaptive immune response. IL-6 is a broadly active cytokine, and immunogloblulin E is a component of an allergic response.

4. M2 macrophages participate in which of the following? A. Wound healing responses B. Antigen presentation C. Production of IL-12 D. Recognition of oxidized low-density lipoprotein (LDL) E. Complement activation Answer: A  Although the designation of M1 and M2 macrophages is overly simplistic, macrophages do shift their functional program as the course of an immune response progresses toward a more chronic state, with M2 macrophages expanding and producing mediators, such as transforming growth factor-β and IL-10, that contribute to resolution of responses and repair of damaged tissue. Antigen presentation, production of IL-12, and recognition of oxidized LDL are more likely to be features of M1 macrophages. 5. Cells of the innate immune system produce soluble mediators that contribute to activation and expansion of an adaptive immune response. Among those mediators are which of the following? A. BAFF B. IL-23 C. LL37 D. All of the above E. A and B Answer: E  Macrophages and dendritic cells produce mediators that influence both the T and B cell arms of the adaptive immune response. IL-23 can promote generation of T-cell effector programs. BAFF is a B-cell survival and differentiation factor. LL37 is a cathelicidin that is produced by neutrophils and participates in the killing of phagocytosed microbes.

220

CHAPTER 46  The Adaptive Immune System  

46  THE ADAPTIVE IMMUNE SYSTEM JOSEPH CRAFT

PRINCIPLES OF ADAPTIVE IMMUNE SYSTEM ACTIVATION: RECOGNITION OF ANTIGEN

Structure of Antigen-Specific Receptors

The innate immune system recognizes structural patterns that are common in the microbial world, whereas the adaptive immune system is designed to respond to the entire continuum of antigens. This goal is achieved through two principal types of antigen recognition receptors: antibodies and T-cell receptors (TCRs). Antibodies, or immunoglobulins, are expressed as cell surface receptors on B cells or are secreted, both of which have the same

221

CHAPTER 46  The Adaptive Immune System  

specificity for antigen. They recognize conformational structures formed by the tertiary configuration of proteins. In contrast, α/β TCRs, the most abundant class of TCRs, fit specifically to epitopes formed by a small linear peptide embedded into major histocompatibility complex (MHC) molecules on the surface of antigen-presenting cells.

Antibodies

Antibodies consist of two identical heavy chains and two identical light chains, which are covalently linked by disulfide bonds. The amino (N)-terminal domain of each chain is variable and represents the recognition structure that interacts with the antigen. Each antibody has two binding arms of identical specificity. The carboxy (C)-terminal ends of the heavy and light chains form the constant region, which defines the subclass of the antibody (κ or λ for light chains; immunoglobulin M (IgM), IgA, IgD, IgE, or IgG for heavy chains). Additional subclasses can be distinguished for IgG and IgA. The constant region of antibodies includes the Fc region. Fc regions can polymerize (IgA) or pentamerize in the presence of a J (joining) chain (IgM). Fc regions are also the ligand for Fc receptors (FcRs) on cells of the innate immune system.

T-Cell Receptors

TCRs are dimers of α chains and β chains or of γ chains and δ chains, each of which contains three complementary-determining binding sites in the N-terminal domain. These complementary-determining sites define the specificity. α/β TCRs recognize peptide fragments in the context of MHC molecules, although certain ones bind glycolipid antigens, for example from mycobacteria, displayed by molecules with structural similarity to MHC. γ/δ TCRs are more variable and can recognize peptides or certain glycolipid antigens in the context of MHC-like molecules, or even unprocessed antigens, functioning similar to antibodies; the latter is a reflection of their structural similarity.

antigen presentation. The two classes of MHC molecules are used as restriction elements by two different subsets of T cells. CD4+ T cells recognize antigenic peptides embedded into MHC class II molecules, whereas CD8+ T cells bind peptides complexed with MHC class I molecules. Generally, MHC class II molecules are expressed only on specialized, so-called professional, antigen-presenting cells, such as dendritic cells, monocytes, macrophages, and B cells, whereas class I proteins are displayed by virtually all nucleated cells, facilitating recognition by CD8+ T cells of peptides from viruses that often have a broad range of target tissues. Peptides bound to MHC class II molecules typically derive from extracellular antigens that are captured and internalized into endosomes to be digested by proteinases, notably cathepsin. Occasionally, however, intracellular proteins or membrane proteins are also funneled into this pathway. MHC class II molecules are assembled in the endoplasmic reticulum in association with a protein called the invariant chain (Fig. 46-1). The molecules are transported to the endosome, where the invariant chain is removed from the peptide-binding cleft, making the cleft accessible to peptides derived from extracellular proteins. MHC class II molecules, stabilized by peptides of 10 to 30 amino acids in length, are displayed on the cell surface, where they are recognized by CD4+ T cells. MHC class I–associated peptides are produced in the cytosol by the proteasome, a large cytoplasmic multiprotein enzyme complex (see Fig. 46-1). Specialized transporter proteins, called transporter in antigen processing (TAP), facilitate translocation of peptides from the cytosolic proteasome to the endoplasmic reticulum. There, the peptides bind to newly formed MHC class I molecules and are transported to the cell surface, where they are recognized by antigen-specific CD8+ T cells. MHC class I−associated peptides may also originate in the extracellular environment and be presented to T cells through the appropriately named cross-presentation pathway. This enables CD8+ T cells to recognize foreign peptides, for example, from viruses, that

Specificities of Antibodies and T-Cell Receptors

The repertoires, or total number of specificities, of antibodies and TCRs are extremely diverse and have been estimated in the human to account for up to 1011 or higher, and 1018, respectively, combinations. This enormous diversity reflects the anticipatory nature of adaptive immune receptors and must be acquired; it cannot be genetically encoded in contrast to that of innate receptors. Its foundation consists of fewer than 400 genes that are recombined and modified. Immunoglobulin heavy chains are formed from four gene segments—the variable, diversity, joining, and constant region gene segments. Also, TCR β chains and δ chains are assembled by the recombination of variable, diversity, joining, and constant region segments of TCR genes. Immunoglobulin light chains and TCR α chains and γ chains lack the diversity segment and are composed of three gene segments. During antibody or TCR rearrangement, gene segments are cut out by nucleases and recombined at the DNA level to form linear coding units for each receptor gene. Through the combination of several different mechanisms, an enormous diversity of receptors is generated. First, the genome contains multiple forms of gene segments; each receptor or antibody uses a different combination of these gene segments. Second, the splicing process is imprecise, introducing nucleotide variations at the variable-diversity, diversity-joining, and variable-joining junctions. These inaccuracies lead to frame shifts and result in completely different amino acid sequences. Finally, random nucleotides can be inserted at the junctional region by an enzyme, deoxyribonucleotidyl transferase. Once generated, TCR sequences remain unchanged. This rule does not apply to immunoglobulins, which undergo modification. Immunoglobulin modification includes (1) replacement of an entire variable region, or receptor editing, typically occurring in the bone marrow during B-cell development to modify those immunoglobulin receptors that inadvertently bind self-antigens on initial recombination of gene segments; (2) class switching, in which the variable-diversity-joining unit combines with different constant region genes (isotype switching); or (3) somatic hypermutation, in which the antigen-contact areas of the antibody undergo mutations during an immune response to improve the affinity (affinity maturation). The latter two events occur in secondary lymphoid tissues, such as the spleen, lymph nodes, and mucosal lymphoid tissue, where immune responses to antigens are initiated.

Antigen Processing

T cells bearing α/β TCRs recognize peptide fragments that are displayed in the context of MHC class I and class II molecules through a process named

Endoplasmic reticulum MHC I Antigen TAP

MHC II Invariant chain Antigen Golgi complex

Proteosome Endosome

CD8 T cell

CD4 T cell FIGURE 46-1.  Pathways of antigen processing and delivery to major histocompatibility complex (MHC) molecules. Cytosolic proteins are broken down by the proteosome to generate peptide fragments, which are transported into the endoplasmic reticulum by specialized peptide transporters (TAP). After peptides are bound to MHC class I molecules, MHC-peptide complexes are released from the endoplasmic reticulum and travel to the cell surface, where they are ligands for CD8+ T-cell receptors (TCRs). Extracellular foreign antigens are taken into intracellular vesicles, called endosomes. As the pH in the endosomes gradually decreases, proteases are activated that digest antigens into peptide fragments. After fusing with vesicles that contain MHC class II molecules, antigenic peptides are placed in the antigen-binding groove. Loaded MHC class II–peptide complexes are transported to the cell surface, where they are recognized by the TCRs of CD4+ T cells.

222

CHAPTER 46  The Adaptive Immune System  

are derived from infected and dying cells that are ingested by myeloid cells and then presented by MHC class I molecules. The nature of the antigen-processing pathway determines the sequence of events in immune responses. Extracellular antigens, in general, enter the endosomal pool and associate with MHC class II molecules to stimulate CD4+ T cells. Cytosolic antigens, including antigens from intracellular infectious agents, are degraded and displayed in the context of MHC class I molecules to initiate CD8+ T-cell responses.

CELLULAR ELEMENTS OF THE ADAPTIVE IMMUNE SYSTEM

T Cells T-Cell Development

T precursor cells are derived from hematopoietic stem cells that migrate to the thymus, a primary lymphoid tissue, where all the subsequent stages of T-cell maturation occur (Fig. 46-2). Pre-T cells express two enzymes, recombinase and terminal deoxynucleotidyl transferase, enabling them to recombine TCR genes. The β chain of the TCR is rearranged first and is expressed together with a pre-TCR α chain. Signals from the immature TCR complex inhibit rearrangement of the second β-chain allele and induce thymocyte proliferation and expression of both CD4 and CD8 molecules, so-called double positive thymocytes. Subsequently, the TCR α chain is recombined, with formation of a mature TCR. From here, the thymocyte undergoes many

CD4−CD8− pre-T cells

Nurse cell Cortex

differentiation and selection steps modulated by the thymic microenvironment, with the end result being formation of a T cell that is ready to migrate to secondary lymphoid tissues and to be poised to recognize antigenic peptides. Early-stage thymocytes reside in the thymic cortex, where they mostly interact with epithelial cells. They then migrate toward the medulla, encountering dendritic cells and macrophages at the corticomedullary junction. Thymic stromal cells regulate T-cell proliferation by secreting lymphopoietic growth factors, such as interleukin-7 (IL-7). Interactions of the TCR with MHC molecules expressed on epithelial cells and on dendritic cells or macrophages determine the fate of the thymocyte.1 Low-avidity recognition of peptide-MHC complexes on thymic epithelial cells by the TCR results in positive selection.2 This recognition event rescues cells from apoptotic cell death and ensures that only T cells with functional receptors that can recognize MHC molecules, critical for T-cell activation on subsequent residence in the spleen and lymph nodes, survive. Thymocytes that express a receptor not fitting any MHC antigen complex die by neglect. High-affinity interaction between the TCR and peptide-MHC complex induces apoptotic death of the recognizing T cell. This process of negative selection eliminates T cells with specificity for self-antigens and is responsible for central tolerance to many autoantigens. It has been estimated that approximately 1% of thymocytes survive the stringent selection process. While undergoing selection, T cells continue to differentiate, with orderly expression of cell surface molecules. Double-positive thymocytes expressing both CD4 and CD8 molecules downregulate one or the other, developing into single-positive CD4+ helper T cells that have been selected on MHC class II complexes or CD8+ cytotoxic T cells that are restricted to MHC class I complexes. These single-positive cells are now mature T cells that are ready for exit and migration through the circulation to secondary lymphoid organs, including the spleen, lymph nodes, and mucosal lymphoid tissues, following chemokine cues and using adhesion molecules to enter. They exist in these tissues as inactivated, or naïve, cells until receiving the appropriate antigenic signal for activation and subsequent effector function.

T-Cell Stimulation and Accessory Molecules CD4+CD8+ T cells

TCR

MHC I+ self-antigen

CD8

MHC II+ self-antigen Positive selection

Cortical epithelial cells

CD4

Negative selection

CD8

Dendritic cell

High-avidity self-recognition

CD4 Low-avidity self-recognition

Self-tolerant cells MHC class I MHC class II restricted restricted CD8 CD4 Apoptosis Medulla

CD8

CD4

Periphery FIGURE 46-2.  Maturation of T cells in the thymus. Precursors committed to the T-cell lineage arrive in the thymus and begin to rearrange their T-cell receptor (TCR) genes. Immature T cells with receptors binding to self–major histocompatibility complex (MHC) on cortical epithelial cells receive signals for survival (positive selection). At the corticomedullary junction, surviving T cells probe self-antigens presented by dendritic cells and macrophages. T cells reacting strongly to self-antigens are deleted by apoptosis (negative selection). T cells released into the periphery are tolerant toward self and recognize foreign antigens in the context of self-MHC.

T-cell activation is initiated when TCR complexes recognize antigenic peptides in the context of the appropriate MHC molecule on the surface of an antigen-presenting cell in secondary lymphoid organs. The principal antigenpresenting cells for activation of naïve T cells are dendritic cells. MHCpeptide recognition by the TCR, the first signal for T-cell activation, leads to receptor clustering and phosphorylation of the intracellular portion of the CD3 protein complex, the signaling component of the TCR, by receptorassociated tyrosine kinases. These events transmit signals to the nucleus of the T cell and initiate its activation. The coreceptors CD4 and CD8 are also critical for the initial events in T-cell activation, through their interaction with MHC class II and class I molecules, respectively, supporting CD3mediated signals. Yet, this first activation signal delivered by the TCR and coreceptors is not alone sufficient for robust T-cell survival and differentiation. It needs to be complemented by the interaction of accessory molecules on the T cell and their ligands on the antigen-presenting cell. A spectrum of accessory molecules is known, of which the best known is CD28, which are engaged by CD80 and CD86 (also known as B7.1 and B7.2, respectively) on antigen-presenting cells (E-Table 46-1). Engagement of CD28 provides to the T cells a second, or costimulatory, signal to the T cell.3 This second signal, delivered by the antigen-presenting dendritic cell, ensures T-cell survival and expansion. CD28-mediated signals are mandatory for the expression of many activation markers on the responding T cells and, in particular, for the secretion of IL-2. In the absence of such a second signal, T cells are rendered nonresponsive and anergic or undergo apoptosis. Finally, adhesion molecules (integrins) stabilize the interactions between T cells and antigenpresenting cells. Signals from the TCR result in the activation of many genes and entry of the T cell into the cell cycle. The signals are transmitted by a cascade of cytoplasmic events. Cross-linking of the TCR and associated CD3 molecules results in the recruitment and activation of phosphotyrosine kinases and the phosphorylation of molecular constituents of the TCR and various adapter molecules. Signals mediated through the TCR then activate several biochemical pathways, which collectively lead to the activation of transcription factors that regulate gene expression Three major variables determine the outcome of TCR stimulation: the duration and affinity of the TCR-antigen interaction, the maturation stage of the responding T cell, and the nature of the antigen-presenting cell. Antigenpresenting cells are gatekeepers in the initiation of T-cell responses. They can

CHAPTER 46  The Adaptive Immune System  

222.e1

E-TABLE 46-1  CYTOKINES AND CYTOKINE FUNCTION CYTOKINES

MAJOR PRODUCER CELLS

PRINCIPAL ACTION

HEMATOPOIETIN FAMILY IL-2 IL-3 IL-4 IL-5 IL-6 IL-7 IL-9 IL-11 IL-13 G-CSF IL-15 GM-CSF

T cells T cells T cells, mast cells T cells, mast cells Macrophages, endothelial cells Bone marrow, thymic epithelium T cells Stromal fibroblasts T cells Fibroblasts and monocytes Non-T cells Macrophages, T cells

Proliferation of T cells, B cells, and NK cells Early hematopoiesis B-cell activation, IgE switch, inhibition of TH1 cells Eosinophil growth and differentiation T-cell and B-cell growth and differentiation, induction of acute phase proteins Growth of pre-B cells and pre-T cells Stimulation of mast cells and TH2 cells Hematopoiesis B-cell growth and differentiation, inhibition of TH1 cells and macrophages Neutrophil development and differentiation Growth of T cells and NK cells Growth and differentiation of myelomonocytic lineage cells

Leukocytes Fibroblasts T cells, NK cells

Antiviral, increases MHC class I expression Antiviral, increases MHC class I expression Macrophage activation, increases expression of MHC molecules, Ig class switching, inhibition of TH2 cells

INTERFERON FAMILY IFN-α IFN-β IFN-γ

TUMOR NECROSIS FACTOR FAMILY TNF-α TNF-β (LT-α) LT-β

Macrophages, NK cells, T cells T cells, B cells T cells, B cells

Induction of pro-inflammatory cytokines, endothelial cell activation, apoptosis Cell death, endothelial activation, lymphoid organ development Cell death, lymphoid organ development

Monocytes, T cells Macrophages, endothelial cells T cells, macrophages Macrophages, dendritic cells T cells, mast cells, eosinophils CD4 memory cells Macrophages Macrophages, dendritic cells

Anti-inflammatory, inhibits cell growth, induces IgA secretion Acute phase response, fever, macrophage activation, costimulation Suppression of macrophage functions NK cell activation, TH1 cell differentiation Chemoattractant for CD4 T cells, monocytes, and eosinophils Cytokine production by epithelia, endothelial cells, and fibroblasts IFN-γ production by T cells and NK cells TH17 cell differentiation

OTHERS TGF-β IL-1α, IL-1β IL-10 IL-12 IL-16 IL-17 IL-18 IL-23

CD = cluster of differentiation; G-CSF = granulocyte colony-stimulating factor; GM-CSF = granulocyte-macrophage colony-stimulating factor; IFN = interferon; Ig = immunoglobulin; IL = interleukin; LT = lymphotoxin; MHC = major histocompatibility complex; NK = natural killer; TGF = transforming growth factor; TH = helper T lymphocyte; TNF = tumor necrosis factor.

CHAPTER 46  The Adaptive Immune System  

upregulate the expression of accessory molecules that provide costimulatory signals. MHC-peptide complexes are particularly dense on dendritic cells, enabling them to activate naïve T cells. In contrast, memory and effector cells have a lower threshold for activation and can react to antigens presented on peripheral tissue cells.

T-Cell Differentiation and Effector Functions

T-cell activation induces T-cell proliferation, with the goal of clonally selecting and expanding antigen-specific T cells. The extent of clonal proliferation is impressive. Antigen-specific CD8+ T cells expand several thousand−fold; CD4+ T cells expand somewhat less. During the phase of rapid growth, T cells differentiate from naïve T cells that are essentially devoid of effector functions into effector T cells that are needed for clearance of infectious organisms, or pathogens. The transition into effector cells is associated with a fundamental shift in functional profiles. First, effector T cells have a lower activation threshold; they do not require costimulation and can scan tissues that lack professional antigen-presenting cells. Second, they switch the expression of chemokine receptors and adhesion molecules to gain access to peripheral tissues. Finally, they gain effector functions. The principal effector function of CD8+ T cells is to lyse infected, antigenbearing target cells. This commitment to eventual cytotoxic function is made during development in the thymus. Upon emigration from the thymus in the naïve, or inactivated, state, CD8+ T cells circulate through secondary lymphoid tissues, surveying antigen-presenting dendritic cells for the appropriate MHC class I−peptide complex that can engage the TCR and that can supply costimulatory signals. On activation, CD8+ T cells acquire cytotoxic functions and, using a variety of receptors and adhesion molecules, can emigrate from secondary lymphoid organs to peripheral tissue sites seeking cells infected by viruses or intracellular bacteria displaying pathogen-derived peptides on MHC class I molecules. On recognizing the appropriate MHC class I–peptide complex, CD8+ T cells induce apoptosis of target cells. The T cell polarizes toward the area of antigen contact; specialized lytic granules are clustered in the contact area. A pore-forming protein, perforin, is released from the lytic granules and inserted into the target cell membrane. Proteases (granzymes) are injected into the target cells to initiate the apoptotic process by activating enzyme cascades. Mechanisms deployed by CD8+ T cells are essentially identical to those of natural killer (NK) cells. CD4+ T cells can also induce apoptosis but by a different mechanism than CD8+ T cells. On activation, they express cell surface molecules such as Fas ligand (CD178) and TRAIL, which initiate the apoptotic cascade selectively in cells expressing the respective ligands Fas (CD95) or the death receptors DR4 and DR5. Compared with CD8+ T cells, the spectrum of options for CD4+ T cells is larger. They are generally characterized as helper T, or TH cells, because they produce cytokines and express cell surface molecules that promote the effector function of other lymphocytes and phagocytes. Like CD8+ T cells, they are initially activated in secondary lymphoid tissues on contact with dendritic cells displaying the MHC−peptide complex (MHC class II, compared with MHC class I for CD8+ T-cell activation) bound by a specific TCR along with the proper costimulatory signals. On activation, different subsets of CD4+ effector T cells can be distinguished based on the preferential production of certain cytokines (see E-Table 46-1). TH1 T cells predominantly produce interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) and are involved in cell-mediated immunity, such as delayed-type hypersensitivity reactions. These cytokines, among other actions, promote macrophage activation that is critical for protective responses against intracellular pathogens such as mycobacteria and listeria. TH2 T cells preferentially produce IL-4, IL-5, and IL-13, cytokines that promote eosinophil maintenance, expansion, and tissue accumulation, as well as macrophage function; these are all important for host protection following infection with helminths, such as schistosomes and other worms. TH17 T cells produce IL-17, critical for neutrophil expansion and function, with killing of extracellular bacteria, such as streptococci, and pathogenic fungi.4 These cells may also produce IL-22 that promotes hostprotective function at barrier surfaces, such as the skin and gut. Follicular helper T (TFH) cells home to lymphoid follicles, where B cells congregate, where they express CD40 ligand (CD154) and other surface proteins along with cytokines, including IL-21, IL-4, and IFN-γ, that are critical for B-cell maturation to plasma cells and memory B cells. The decision as to which differentiation pathway to take is made during the early stages of naïve T-cell activation by antigen-presenting cells in secondary lymphoid organs. Pathway differentiation depends on several factors, including (1) the cytokines produced by the activating antigen-presenting cell and other innate cells in the

223

microenvironment, (2) the nature of costimulatory signals, and (3) the avidity of the TCR–MHC antigen interaction. CD4+ T-cell subset, or lineage, development is generally correlated with the expression of specific transcription factors (T-bet for TH1, GATA3 for TH2, RORγt for TH17, and Bcl6 for TFH cells). However, lineage commitment among differentiated CD4+ T cells is not absolute and is not terminal, and transition between different effector types is possible.

Regulatory T Cells

Depending on their cytokine profile, CD4+ T cells have the ability to crossregulate each other, influence T-cell differentiation, and suppress T-cell effector activity. Classic examples of T cells with regulatory activity generated during the normal immune response are IL-10– and transforming growth factor-β (TGF-β)–producing cells. In addition, specialized subsets of regulatory T (Treg) cells are characterized by expression of the transcription factor forkhead box P3 (Foxp3). Naturally occurring Foxp3+ Treg cells are generated during T-cell development in the thymus and recognize self-antigens. Foxp3+ Treg cells can also arise from conventional CD4+ T cells in the periphery. Natural and inducible Treg cells are in many ways indistinguishable, particularly because their development and function depend on Foxp3, and they are able to suppress T-cell expansion and constitutively express several cell surface markers, albeit markers that are not necessarily specific for Treg because activated T cells can also express them. Treg cells are important in peripheral tolerance, controlling the expansion of autoreactive T cells. They also play a role in immune responses to pathogens by virtue of their ability to suppress T-cell effector function and consequently downmodulate the inflammatory response incited by the former, a natural consequence of pathogen elimination. A principal difference between natural and induced Treg cells is that the latter largely survey mucosal and other environmentally exposed surfaces. Despite extensive studies in various models, the mechanism by which Treg cells function in vivo remains incompletely understood, although it is certainly a consequence of secretion of regulatory cytokines, like IL-10 and TGF-β, that can dampen inflammatory responses. Tregs may also express the T-cell molecule cytotoxic T-lymphocyte antigen (CTLA)-4 (CD152) that, like CD28, engages CD80 and CD86 on antigen-presenting cells. In contrast to CD28, which receives a positive signal from CD80 and CD86, leading to robust T-cell activation, engagement of these molecules on antigen-presenting cells by CTLA-4 on Tregs suppresses the ability of antigen-presenting cells to activate naïve T cells.

T-Cell Homeostasis

Effective immunity depends on the ability of the immune system to generate large numbers of antigen-specific T cells rapidly, yet the space in the T-cell compartment is limited. To avoid competition for space and resources and to prevent perturbation of T-cell diversity by lifelong exposure to antigens, the adaptive immune system employs several counterbalancing mechanisms. In the later stages of the activation process, a strong negative signal derives from interaction of CTLA-4 with CD80/CD86 on antigen-presenting cells. In addition, T cells undergo activation-induced cell death. Activated CD4+ T cells begin to secrete Fas ligand and acquire sensitivity to Fas-mediated death, inducing apoptotic suicide and fratricide in neighboring T cells. These mechanisms impose constraints in the early stages of the T-cell antigen response. Other mechanisms control the rapid decline of expanded antigen-specific T cells when elimination of the antigen has been achieved. Removal of the driving antigen causes a deprivation of cytokines and costimulatory molecules, and growth factor–deprived T cells die from apoptosis. It has been estimated that only 5% of the antigen-expanded population survives after antigen clearance, becoming memory cells that are poised to respond if the host is again challenged by the same offending pathogen.

B Lymphocytes B-Cell Development

B cells are generated in the bone marrow, like the thymus, a primary lymphoid organ. Lymphoid stem cells differentiate into distinctive B-lineage cells in the marrow, supported by a specialized microenvironment of nonlymphoid stromal cells supplying necessary chemokines, including stromal cell– derived factor 1 and cytokines (IL-7). Precursor B cells enter a process of tightly controlled sequential rearrangements of heavy chain and light chain immunoglobulin genes. On pre-B cells, the membrane µ chain is associated with a surrogate light chain to form a pre-B-cell receptor (BCR). Signals provided through this receptor induce proliferation of progeny that subsequently rearrange different light chain gene segments.

224

CHAPTER 46  The Adaptive Immune System  

Bone marrow

Lymph node

IgM

Pre-B cell

IgM

Stem cell

IgD

Bone marrow stromal cell

NaÏve mature B cell

Germinal center

Mantle zone Follicular dendritic cell

Memory cells Plasma cells Isotype switching

T-cell zone

B-cell proliferation Somatic hypermutation

Positive selection

FIGURE 46-3.  B-cell development and differentiation. The early stages of B-cell development occur in the bone marrow, with cells progressing through a developmental program determined by the rearrangement and expression of immunoglobulin (Ig) genes. Immature B cells with receptors for multivalent self-antigens die in the bone marrow. Surviving B cells coexpress IgD and IgM surface receptors. They are seeded into peripheral lymphoid organs, where they home to selected locations and receive signals to survive and become longer-lived naïve B cells. Antigen-binding B cells and antigen-presenting B cells that receive help from antigen-specific T cells are activated through membrane-bound and secreted molecules. Activated B cells migrate into the follicles, leading to the formation of germinal centers. B cells in germinal centers undergo somatic hypermutation of immunoglobulin genes; cells with high affinity for antigens presented on the surface of follicular dendritic cells are selected to differentiate into either memory B cells or plasma cells.

It is estimated that only 10% of B cells generated in the bone marrow reach the circulating pool. Losses are mostly due to negative selection and clonal deletion of immature B cells that express receptors directed against selfantigens. Cross-linking of surface IgM by multivalent self-antigens causes immature B cells to die. Such self-reactive B cells can be rescued from death by replacing the light chain with a newly rearranged light chain that is no longer self-reactive, a process named receptor editing.5 On maturation, B cells begin to express surface IgD. B cells positive for IgD and IgM are exported from the bone marrow and migrate to peripheral lymphoid tissues following a chemokine gradient, in a process analogous to the migration of naïve T cells from the thymus to the same tissues (Fig. 46-3). There, colocalization of both types of lymphocytes facilitates their interaction following pathogen challenge. This enables B cells to receive T-cell help for the former’s activation and subsequent function, including memory development and antibody secretion, required for responses to protein antigens.

B-Cell Stimulation

Mature, but naïve, B cells in secondary lymphoid organs are activated by soluble and cell-bound antigens to develop into antibody-secreting effector cells. B cells respond to a large variety of antigens, including proteins, polysaccharides, and lipids. Binding of antigen to cell surface IgM molecules induces BCR clustering, the initial step in B-cell activation. In addition to the antigenbinding immunoglobulin, the BCR comprises two proteins, Ig-α and Ig-β. The Ig-α/Ig-β heterodimer functions to transduce a signal and initiates the intracellular signaling cascade, analogous to the CD3 molecule of the TCR. Thus, the composition of the BCR, with ligand-binding and signal-transducing units, and the signaling events that lead to gene induction, are similar to those of the TCR. BCR triggering is enhanced by coreceptors, as for the TCR. The BCR-coreceptor complex is composed of CD81, CD19, and CD21, analogous to the TCR coreceptors CD4 and CD8. CD21 binds to complement fragments on opsonized antigens that are bound by the BCR, resulting in phosphorylation of the intracellular tail of CD19 by tyrosine kinases and augmentation of the BCR-mediated signal. Like naïve T cells, naïve B cells require accessory signals in addition to triggering of their antigen-binding receptor. They receive second signals either from follicular helper T cells or from microbial components. Microbial

constituents, such as bacterial polysaccharides, can induce antibody production in the absence of helper T cells, comprising thymus-independent, or T-independent, antigens.6 In contrast, in the case of protein antigens, which are thymus- or T-dependent, the initial BCR stimulation prepares the cell for subsequent interaction with follicular helper T cells. These activated B cells start to enter the cell cycle; upregulate cell surface molecules, such as CD80 and CD86, that provide costimulatory signals to T cells; and upregulate certain cytokine receptors. As such, these B cells are prepared to activate helper T cells and to respond to cytokines secreted by those T cells, but they cannot differentiate into antibody-producing cells in the absence of T-cell help. Survival and differentiation factors produced by myeloid cells, such as B-cell-activating factor (BAFF), also stimulate B cells and help to maintain the B-cell pool.7

B-Cell Differentiation

Differentiation of B cells activated by protein antigens depends on interaction with helper T cells. B cells use their antigen receptor not only to recognize antigens but also to internalize them. After processing endocytosed antigens, MHC class II–peptide complexes appear on the cell surface, where antigenspecific CD4+ T cells detect them. Also, B cells express costimulatory molecules and provide optimal conditions for T-cell activation. On activation, CD4+ T cells express CD154, also known as CD40 ligand, on their surface and are able to stimulate the CD40 molecule on their B-cell partner. CD40CD154 interaction is essential for subsequent B-cell proliferation and differentiation. Cytokines secreted by the helper T cells act in concert with CD154 to amplify B-cell differentiation and to determine the antibody type by controlling isotype switching. Isotypes greatly influence the versatility of antibodies as effector molecules, and cytokines drive isotype switching by stimulating the transcriptional activation of heavy chain constant region genes and enabling switching from transcription of the IgM heavy chain gene to that of IgG, IgA, or IgE. T-cell-dependent B-cell differentiation and maturation take place in germinal centers, specialized areas in secondary lymphoid tissues where B cells rapidly proliferate, with mutation of the variable, or antigen-binding portion, of their immunoglobulin surface receptors (BCRs) (see Fig. 46-3). Those B cells bearing receptors with the highest affinity for antigen are selected for

CHAPTER 46  The Adaptive Immune System  

survival with the help of specific signals delivered by follicular helper T cells, whereas those with lesser affinity die by apoptosis. This process enables affinity maturation of B cells that most efficiently bind antigen and thereby facilitate its removal. As somatic hypermutation and affinity maturation proceed in the germinal center, isotype class switching of the immunoglobulin receptors is also occurring.8

Lymphocytes and Lymphoid Tissue

The initiation of adaptive immune responses depends on rare antigen-specific T cells and B cells meeting antigen-presenting cells and their relevant antigen. The recognition of a specific antigen in the tissue by uncommon T cells has a low probability, and it is unlikely that sufficient numbers of antigenpresenting cells and lymphocytes can be brought together to provide crucial momentum. The immune system uses specialized lymphoid microstructures to bring antigens to the site of lymphocyte traffic and accumulation. Secondary lymphoid organs include the spleen for blood-borne antigens, the lymph nodes for antigens encountered in peripheral tissues, and the mucosaassociated, bronchial-associated, and gut-associated lymphoid tissues, where antigens from epithelial surfaces are collected. Lymphocytes circulate through secondary lymphoid organs, constantly searching for their antigen. Their homing to lymph nodes is facilitated by specialized microvessels, called high endothelial venules, which provide the proper structure for them to leave the circulation and enter the tissue. Secondary lymphoid tissues have developed several strategies to sequester the relevant antigen. Antigens in peripheral tissue are encountered first by dendritic cells that, after activation, are mobilized to transport antigens into the local lymph nodes by the draining lymph. These antigen-bearing dendritic cells enter the lymph nodes through the afferent lymphatic vessel and settle in the T-cell-rich zones to present processed antigens to T cells. The net result of this process is an accumulation and concentration of the antigen in an environment that can be readily screened by infrequent antigen-specific T cells. B cells are segregated from T cells in the lymph nodes and are localized in follicles. If, on antigen engagement, B cells find their cooperating (cognate) T cells at the borders of the T-cell-rich areas and the follicle, they receive cues to enter germinal centers along with their cognate follicular helper T cells. Germinal centers contain a network of follicular dendritic cells that capture particulate antigen or immune complexes on the cell surface. This unprocessed antigen is taken up by antigen-specific B cells, processed and presented, and recognized by antigen-specific TFH cells. These T cells provide cytokines and cell-cell contact signals to support the germinal center reaction, a process that includes somatic hypermutation, affinity selection, and isotype switching (see Fig. 46-3). Germinal centers are essential for generating long-lived antibody-secreting plasma cells and memory B cells. Lymphoid organ development is highly dependent on environmental cues. The symbiotic relationship between the host immune system and microorganisms is best exemplified in the gastrointestinal tract. Development of gutassociated lymphoid tissue is absolutely dependent on bacterial colonization. Increasing evidence suggests that host-symbiont interactions regulate adaptive immune functions throughout life. Disturbances in the bacterial microbiota and failure to maintain intestinal homeostasis are important in diverse diseases, including inflammatory bowel disease (Chapter 143) and HIVassociated immune defects.

Memory

An important consequence of adaptive immunity is the generation of immunologic memory, the basis for long-lived protection after a primary infection. Memory induction by vaccination is one of the landmark successes in medicine. Immunologic memory is defined as the ability to respond more rapidly and effectively to pathogens that have been encountered previously. The bases of immunologic memory are qualitative and quantitative changes in antigen-specific T cells and B cells. As a direct result of clonal expansion and selection in antigen-driven responses, the frequencies of antigen-specific memory B cells and memory T cells are increased 10-fold to 1000-fold compared with the naïve repertoires. The mechanisms through which memory T cells and B cells escape clonal downsizing in the terminal stages of the primary immune response are consequences of upregulation of a selected group of transcription factors that ensure survival. The enrichment of antigen-specific B cells and T cells enhances the sensitivity of the system to renewed challenges and provides a head start of 4 to 10 cell divisions. In addition to increased frequencies, memory T cells and B cells are functionally different from their naïve counterparts. Memory cells are long-lived and survive in the presence of certain cytokines without the need for continuous antigenic

225

stimulation, guaranteeing immunologic memory for the life expectancy of the individual cell. Memory B cells produce predominantly IgG and IgA antibodies with evidence of somatic hypermutation and high affinity for the antigen. Cell surface expression of high-affinity antibodies allows more efficient antigen uptake, which enhances the crucial interaction with T cells. On antigen encounter, memory B cells change to antibody-secreting plasma cells, or re-enter the germinal center, where the high affinity of their immunoglobulin receptor gives them a competitive advantage over naïve B cells in antigen binding, leading to progressive affinity maturation of somatically mutated antibody molecules. Because the TCR does not undergo isotype switching or affinity maturation, memory T cells are more difficult to distinguish from naïve or effector T cells. In contrast to effector cells, memory T cells lack activation markers and need antigen stimulation to resume effector functions. In contrast to naïve T cells, memory T cells have a lower activation threshold and are less dependent on costimulatory signals. In essence, their requirements for antigen stimulation are fewer, and their clonal size is larger, permitting fast, efficient responses to secondary antigen encounters. Also, memory T cells resume effector functions without having to undergo cell divisions.

Immunologic Tolerance and Autoimmunity

Unresponsiveness to self is a fundamental property of the immune system and is a condition, sine qua non, to maintain tissue integrity of the host. Self/ nonself distinction is relatively straightforward for the innate immune system, in which receptors to nonself molecules are genetically encoded and evolutionarily selected. Self/nonself discrimination is much more complex for the adaptive immune system, in which antigen-specific receptors are generated randomly and the entire spectrum of antigens can be recognized. Thus, the adaptive immune system must acquire the ability to distinguish between self and nonself. Several different mechanisms are used, collectively called tolerance. Tolerance is antigen specific; its induction requires the recognition of antigen by lymphocytes in a defined setting. Failure of self-tolerance results in immune responses against self-antigens. Such reactions are called autoimmunity and may give rise to chronic inflammatory autoimmune disease. Central and peripheral tolerance mechanisms can be distinguished. In central tolerance, self-reactive lymphocytes are deleted during development. This process of negative selection is particularly important for T cells. During thymic development, T cells that recognize antigen with high affinity, in particular antigens that are constitutively expressed on antigen-presenting cells, are deleted. Central tolerance for B cells follows the same principles. Recognition of antigen by developing B cells in the bone marrow induces apoptosis, or receptor editing that replaces the self-reactive receptor with one containing the product of a newly rearranged light chain gene. Negative selection is particularly important for B cells that recognize multivalent antigens because they do not depend on T-cell help and cannot be controlled peripherally. Not all self-reactive T cells are centrally purged from the repertoire; certain antigens are not encountered at sufficient densities in the thymus. Also, all T cells have some degree of self-reactivity, which is necessary for positive selection in the thymus and for peripheral survival. Mechanisms of peripheral T-cell tolerance include (1) anergy, (2) peripheral deletion, (3) clonal ignorance, and (4) suppression of immune responses by regulatory T cells. T-cell anergy is transient and is actively maintained. It is induced if CD4+ T cells recognize antigens presented by MHC class II molecules without receiving costimulatory signals. In general, costimulatory molecules such as CD80 and CD86 are restricted to antigen-presenting cells, and their expression is dependent on microbial recognition, leading to activation of the antigenpresenting cells. MHC-peptide presentation to T cells by immature or inactivated, resting antigen-presenting cells or on any cell other than peripheral antigen-presenting cells results in anergy because these cells typically lack expression of costimulatory molecules. Tissue-residing immature dendritic cells need to be activated by cytokines or recognition of pathogen-associated molecular patterns (PAMPs) to stimulate and not to anergize T cells. A second tolerance mechanism, peripheral deletion, is induced as a consequence of hyperstimulation. Hyperstimulation of T cells (e.g., by high doses of antigen and high concentrations of IL-2) preferentially activates proapoptotic pathways and causes elimination of the responding T-cell specificity. This mechanism may be responsible for the elimination of T cells specific for plentiful peripheral self-antigens and for foreign antigens abundantly present during infection. Whereas induction of anergy and activationinduced cell death are active consequences of antigen recognition, the third tolerance mechanism, clonal ignorance, is less well understood. Clonal

ignorance is defined as the presence of self-reactive lymphocytes that fail to recognize or to respond to peripheral antigens. These cells remain responsive to antigenic challenge if given in the right setting. An example of clonal ignorance is nonresponsiveness to sequestered antigens that are not accessible to the immune system. Other mechanisms must exist, however, because clonal ignorance has also been shown for accessible antigens. Fourth, Treg cells play a pivotal role in maintaining peripheral tolerance. During an immune response, T cells can acquire the ability to produce regulatory cytokines, such as TGF-β, IL-10, or IL-4, that dampen or suppress immune responses. A dedicated subset of Treg cells, Foxp3 CD4+ T cells, has been identified and characterized. Harnessing the frequencies and function of these cells may offer a promising approach to restoring peripheral tolerance in treating autoimmune diseases or facilitating transplantation tolerance; their elimination or functional suppression may potentiate cancer immunotherapy. A critically important mechanism of peripheral tolerance of B cells is maintained through the absence of T-cell help. B cells require signals from T cells to differentiate into effector cells. B lymphocytes that recognize self-antigens in the periphery in the absence of T-cell help are rendered anergic or are unable to enter lymphoid follicles, where they could receive T-cell help, effectively excluding them from immune responses. Generation and maintenance of self-tolerance can fail, in which case autoimmune responses are generated. Overall, chronic inflammatory diseases induced by tolerance failure occur in about 5% of the general population. Given the complexity of regulation, it is surprising that autoimmune diseases are not more frequent. It is thought that most autoimmune diseases result from dysfunction of the adaptive immune system, although activation of the innate immune system can set the stage for a self-reactive adaptive immune response. Many models of autoimmunity rely on the hypothesis that peripheral anergy is broken. Aberrant expression of costimulatory molecules on nonprofessional antigen-presenting cells or inappropriate activation of tissue-residing dendritic cells sets the stage for the induction of “forbidden” T-cell responses. Also, autoreactive B cells that recognize self-antigen complexed with foreign antigen may engulf this complex and receive help from T cells specific for the foreign antigen. Autoimmunity also may emerge if antigen ignorance is broken. This could happen if tissue barriers break down and antigens that are usually sequestered from the immune system, such as antigens from the central nervous system or the eye, become accessible. Tolerance mechanisms of anergy or clonal ignorance can also fail if a foreign antigen is sufficiently different from a self-antigen to initiate an immune response but sufficiently similar for activated T cells to elicit T-cell and B-cell effector functions (molecular mimicry). GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 46  The Adaptive Immune System  

GENERAL REFERENCES 1. Klein L, Kyewski B, Allen PM, et al. Positive and negative selection of the T cell repertoire: what thymocytes see (and don’t see). Nat Rev Immunol. 2014;14:377-391. 2. Fu G, Rybakin V, Brzostek J, et al. Fine-tuning T cell receptor signaling to control T cell development. Trends Immunol. 2014;35:311-318. 3. Hubo M, Trinschek B, Kryczanowsky F, et al. Costimulatory molecules on immunogenic versus tolerogenic human dendritic cells. Front Immunol. 2013;4:82. 4. Peters A, Yosef N. Understanding Th17 cells through systematic genomic analyses. Curr Opin Immunol. 2014;28:42-48.

226.e1

5. Luning Prak ET, Monestier M, Eisenberg RA. B cell receptor editing in tolerance and autoimmunity. Ann N Y Acad Sci. 2011;1217:96-121. 6. Bortnick A, Allman D. What is and what should always have been: long-lived plasma cells induced by T cell-independent antigens. J Immunol. 2013;190:5913-5918. 7. Rickert RC, Jellusova J, Miletic AV. Signaling by the tumor necrosis factor receptor superfamily in B-cell biology and disease. Immunol Rev. 2011;244:115-133. 8. Matthews AJ, Zheng S, DiMenna LJ, et al. Regulation of immunoglobulin class-switch recombination: choreography of noncoding transcription, targeted DNA deamination, and long-range DNA repair. Adv Immunol. 2014;122:1-57.

226.e2

CHAPTER 46  The Adaptive Immune System  

REVIEW QUESTIONS 1. IgG antibodies are characterized by all of the following except which one? A. Two heavy chains and two light chains B. A common γ chain C. Disulfide bonds D. Subclasses Answer: B  IgG molecules consist of two identical heavy and two identical light chains, linked by disulfide bonds. IgG and IgA molecules include distinct subclasses encoded by distinct genomic DNA sequences. A common γ chain is a signaling component of certain cytokine and Fc receptors but is not a component of IgG antibodies. 2. CD4+ T cells recognize antigenic peptides presented on which of the following? A. Epithelial cells B. Major histocompatibility complex (MHC) class I molecules C. MHC class II molecules D. Costimulatory molecules E. CD80 Answer: C  Antigenic peptides processed by specialized antigen-presenting cells associate with the peptide-binding cleft of MHC class II molecules and are recognized by the T-cell receptor expressed on CD4+ T cells. CD8+ T cells recognized antigenic peptides associated with MHC class I molecules. CD80 is an example of a costimulatory molecule expressed on antigen-presenting cells that supports the activation of T cells that have interacted with peptideMHC complexes. Epithelial cells do not typically activate CD4+ T cells. 3. Which of the following functions is not provided by differentiated T cells? A. Production of interleukin-17 (IL-17) B. Lysis of virus-infected target cells C. Help for B-cell activation and differentiation D. Production of BAFF E. Production of interferon-γ Answer: D  The cytokine milieu of T cells contributes to differentiation of those cells to provide various effector functions. These include support for B-cell activation and differentiation (through CD154-CD40 interactions and production of IL-21), secretion of cytokines (interferon-γ) that support macrophage activation and delayed-type hypersensitivity reactions, induction of inflammatory responses (IL-17), and mediating the killing of virus-infected cells. Myeloid cells, rather than T cells, are the major producers of BAFF.

4. B cells minimize self-reactivity through a process called which of the following? A. Isotype switching B. Costimulation C. Somatic hypermutation D. Recombination E. Receptor editing Answer: E  Isotype switching, somatic hypermutation, and recombination are processes that promote the diversity of the B-cell repertoire. Costimulation of T-cell activation supports amplification of an adaptive immune response. Receptor editing modifies the sequence of the B-cell receptor to avoid production of autoreactive B cells. 5. Immunologic tolerance is mediated by all of the following except which of the following? A. Thymic deletion of T cells that recognize antigen with high affinity B. Regulatory T cells C. Toll-like receptor activation D. T-cell activation in the absence of costimulation E. It is mediated by all of the above. Answer: C  Central and peripheral tolerance mechanisms include thymic deletion of self-reactive T cells, T-cell activation without costimulation, and control of self-reactive cells by regulatory T cells. Toll-like receptor activation is an important feature of innate immune system activation.

226

CHAPTER 47  Mechanisms of Immune-Mediated Tissue Injury  

Effector mechanisms that eliminate pathogens in adaptive immune responses are essentially identical to those of innate immunity. The specific antigen recognition feature of the adaptive immune response seems to have been appended to the preexisting innate defense system. As a result, the inflammatory cells and molecules of the innate immune system are essential for the effector functions of B and T lymphocytes. In addition to initiating protective responses, they mediate tissue injury in allergy, hypersensitivity, and autoimmunity.

Effector Mechanisms

47  MECHANISMS OF IMMUNE-MEDIATED TISSUE INJURY JANE E. SALMON

THE ADAPTIVE IMMUNE RESPONSE

Definition

The adaptive immune response is a crucial component of host defense against infection. Its distinguishing and unique feature is the ability to recognize pathogens specifically, based on clonal selection of lymphocytes bearing antigen-specific receptors. Antigens unassociated with infectious agents also may elicit adaptive immune responses. Many clinically important diseases are characterized by normal immune responses directed against an inappropriate antigen, typically in the absence of infection. Immune responses directed at noninfectious antigens occur in allergy, in which the antigen is an innocuous foreign substance, and in autoimmunity, in which the response is to a self-antigen.

Effector actions of antibodies depend on recruiting cells and molecules of the innate immune system. Antibodies are adapters that bind antigens to nonspecific inflammatory cells and direct their destructive effector responses. Antibodies also activate the complement system, which enhances opsonization of antigens, recruits phagocytic cells, and amplifies (or “complements”) antibody-triggered damage. The isotype or class of antibodies produced determines which effector mechanisms are engaged. Cell-bound receptors for immunoglobulin (Ig) constitute the link between humoral and cellular aspects of the immune cascade and play an integral part in the process by which foreign and endogenous opsonized material is identified and destroyed. These cell-based binding sites for antibodies, termed Fc receptors, interact with the constant region (Fc portion) of the immunoglobulin heavy chain of a particular antibody class regardless of its antigen specificity. Accessory cells that lack intrinsic specificity, such as neutrophils, macrophages, and mast cells, are recruited to participate in inflammatory responses through the interaction of their Fc receptors with antigen-specific antibodies. Distinct receptors for different immunoglobulin isotypes are expressed on different effector cells. Receptors for IgG (FcγRs) are a diverse group of receptors expressed as hematopoietic cell surface molecules on phagocytes (macrophages, monocytes, neutrophils), platelets, mast cells, eosinophils, and natural killer (NK) cells. FcγRs often are expressed as stimulatory and inhibitory pairs.1 Triggering of stimulatory FcγRs initiates a series of events, including phagocytosis; antibody-dependent, cell-mediated cytotoxicity; secretion of granules; and release of inflammatory mediators, such as cytokines, reactive oxidants, and proteases. Extensive structural diversity among FcγR family members leads to differences in binding capacity, signal transduction pathways, and cell type–specific expression patterns. This diversity allows IgG complexes to activate a broad program of cell functions relevant to inflammation, host defense, and autoimmunity. Phagocyte activation is triggered by stimulatory FcγRs, facilitating the recognition, uptake, and destruction of antibodycoated targets, whereas multivalent IgG binding to FcγRs on platelets leads to platelet aggregation and thrombosis, and binding to FcγRs on NK cells mediates cytotoxicity of antibody-coated targets. IgE binds to high-affinity FcεRs on mast cells, basophils, and activated eosinophils.2 In contrast to FcγRs, which are low affinity and bind to multivalent IgG rather than circulating individual IgG molecules, FcεRs can bind monomeric IgE. A single mast cell may be armed with IgE molecules specific for different antigens, all bound to surface FcεRs. Mast cells, localized beneath the mucosa of the gastrointestinal and respiratory tracts and the dermis of the skin, await exposure to multivalent antigens, which cross-link surface IgE bound to FcεRs and cause release of histamine-containing granules and generation of cytokines and other inflammatory mediators. IgEmediated activation of eosinophils, cells normally present in the connective tissue of underlying respiratory, urogenital, and gut epithelium, leads to the release of highly toxic granule proteins, free radicals, and chemical mediators such as prostaglandins, cytokines, and chemokines. These amplify local inflammatory responses by activating endothelial cells and recruiting and activating more eosinophils and leukocytes. Prepackaged granules and highaffinity FcεRs that bind to free monomeric IgE enable an immediate response to pathogens or allergens at the first site of entry, a location where FcεRbearing cells reside. Inhibitory FcγRs, which modulate activation thresholds and terminate stimulating signals, are key elements in the regulation of effector function. Given that inhibitory and stimulatory Fc receptors are often coexpressed on the same cells, the effector response to a specific stimulus in a particular cell represents the balance between stimulatory and inhibitory signals. Inhibitory FcγRs can dampen responses triggered by FcεRs on mast cells and FcγRmediated inflammation at sites of immune complex deposition. Effector activities targeted by IgG and IgM also may be mediated by components of the complement system (Chapter 50). Antigen-bound multimeric immunoglobulin can initiate activation of the classic pathway of

227

CHAPTER 47  Mechanisms of Immune-Mediated Tissue Injury  

complement, causing enhanced phagocytosis of antigen-antibody complexes, increased local vascular permeability, and recruitment and activation of inflammatory cells. The target of injury is specified by the antibody, and the extent of damage is determined by the synergistic activities of immunoglobulin and complement. Antigen-specific effector T cells also may initiate tissue injury. On exposure to an appropriate antigen, memory T cells are stimulated to release cytokines and chemokines that activate local endothelial cells and recruit and activate macrophages and other inflammatory cells. The effector cells directed by T-cell-derived cytokines, or cytolytic T cells themselves, mediate tissue damage. T helper 1 (TH1) cells produce interferon-γ (IFN-γ) and activate macrophages to cause injury, whereas TH2 cells produce interleukin-4 (IL-4), IL-5, and eotaxin (an eosinophil-specific chemokine) and trigger inflammatory responses in which eosinophils predominate. TH17 cells secrete several effector molecules, including IL-17, which act on both immune and nonimmune cells to trigger differentiation; release of antimicrobial molecules, cytokines, and chemokines; and recruitment to sites of inflammation.3 New TH effector subsets have recently been identified, including follicular T helper cells (TFH), which provide help to B cells in germinal centers and thus are key regulators of humoral responses and antibody production.

HYPERSENSITIVITY REACTIONS

In predisposed individuals, innocuous environmental antigens may stimulate an adaptive immune response, immunologic memory, and, on subsequent

exposure to the antigen, inflammation. These “overreactions” of the immune system to harmless environmental antigens (allergens), called hypersensitivity or allergic reactions, produce tissue injury and can cause serious disease. Hypersensitivity reactions are grouped into four types according to the effector mechanisms by which they are produced (Table 47-1). The effectors for types I, II, and III hypersensitivity reactions are antibody molecules, whereas type IV reactions are mediated by antigen-specific effector T cells.4 Autoimmune disease is characterized by the presence of antibodies and T cells specific for self-antigens expressed on target tissues. The mechanisms of antigen recognition and effector function that lead to tissue damage in autoimmune disease are similar to the mechanisms elicited in response to pathogens and environmental antigens. These mechanisms resemble certain hypersensitivity reactions and may be classified accordingly (Table 47-2). Autoimmune disease caused by antibodies directed against cell surface or extracellular matrix antigens corresponds to type II hypersensitivity reactions; disease caused by formation of soluble immune complexes that subsequently are deposited in tissue corresponds to type III hypersensitivity; and disease caused by effector T cells corresponds to type IV hypersensitivity. Typically, several of these pathogenic mechanisms are operative in autoimmune disease. However, IgE responses are not associated with damage in autoimmunity.

Type I Hypersensitivity Reactions

Type I hypersensitivity reactions (Fig. 47-1) are triggered by the interaction of antigen with antigen-specific IgE bound to FcεRs on mast cells, which

TABLE 47-1  FOUR MAJOR TYPES OF IMMUNOLOGICALLY MEDIATED HYPERSENSITIVITY REACTIONS* IMMUNOLOGIC SPECIFICITY

TYPE I (IgE ANTIBODY)

TYPE II (IgG ANTIBODY)

TYPE III (IgG ANTIBODY) Soluble antigen

TYPE IV (T CELLS) TH1 Cells

TH2 Cells

TH17 Cells

T Cells

Antigen

Soluble antigen allergen

Cell- or matrixassociated antigen

Soluble antigen

Soluble antigen

Soluble antigen

Cell-associated antigen

Effector mechanism

FcεRI- or FcγRIIIdependent mast cell activation, with release of mediators/ cytokines

FcγR+ cells FcγR+ cells, (phagocytes, NK complement cells), complement

Macrophage activation

Eosinophil activation

Macrophage activation Neutrophil activation

Direct cytotoxicity

Examples

Systemic anaphylaxis, Certain drug Arthus reaction and Contact dermatitis, Chronic allergic Contact dermatitis, Contact dermatitis asthma, allergic reactions and other immune tuberculin inflammation atopic dermatitis, (e.g., poison ivy), rhinitis, urticaria, reactions to complex–mediated reaction (e.g., chronic asthma, reactions to angioedema incompatible reactions (e.g., asthma, rheumatoid certain blood transfusions serum sickness, chronic allergic arthritis virus-infected subacute bacterial rhinitis) cells, some endocarditis) instances of graft rejection

*Hypersensitivity reactions were classified into four types by Coombs and Gell (1963) and modified by Janeway and colleagues (2001). FcγR = Fc receptor for immunoglobulin G; FcεR = Fc receptor for immunoglobulin E; NK = natural killer. From Coombs RRA, Gell PGH: Classification of allergic reactions responsible for clinical hypersensitivity and disease. In: Gell PGH, Coombs RA, eds. Clinical Aspects of Immunology. Oxford, UK: Blackwell; 1963; and Janeway C, Travers P, Walport M, Shlomchick M: Immunobiology: The Immune System in Health and Disease. 5th ed. New York: Garland Publishing; 2001.

TABLE 47-2  CLASSIFICATION OF AUTOIMMUNE DISEASES ACCORDING TO MECHANISM OF TISSUE INJURY HYPERSENSITIVITY REACTION

AUTOIMMUNE DISEASE

AUTOANTIGEN

TYPE II Antibody against cell surface antigens Antibody against receptors Antibody against matrix antigens

Autoimmune hemolytic anemia Autoimmune thrombocytopenic purpura Graves disease Myasthenia gravis Goodpasture syndrome

Rh blood group antigens, I antigen Platelet integrin glycoprotein IIb/IIIa Thyroid-stimulating hormone receptor (agonistic antibodies) Acetylcholine receptor (antagonistic antibodies) Basem*nt membrane collagen (α3-chain of type IV collagen)

Pemphigus vulgaris

Epidermal cadherin (desmoglein)

Mixed essential cryoglobulinemia Systemic lupus erythematosus

Rheumatoid factor IgG complexes (with or without hepatitis C antigens) DNA, histones, ribosomes, binuclear proteins

Insulin-dependent diabetes mellitus Rheumatoid arthritis Multiple sclerosis

Pancreatic B-cell antigen Unknown synovial joint antigen Myelin basic protein, proteolipid protein

TYPE III Immune complex diseases TYPE IV T-cell-mediated diseases

228

CHAPTER 47  Mechanisms of Immune-Mediated Tissue Injury  

Cell-associated antigen

IgE FcεRI Soluble antigen

Mast cell Erythrocyte Histamine Proteolytic enzymes Cytokines (IL-4, IL-5, TNF-α) Leukotrienes Chemokines

IgG FcγR

MΦ Tissue damage FIGURE 47-1.  Type I hypersensitivity. Type I responses are mediated by immunoglobulin E (IgE), which induces mast cell activation. Cross-linking of the Fc receptor for IgE (FcεR) on mast cells, triggered by the interaction of multivalent antigen with antigenspecific IgE bound to FcεR, causes the release of preformed granules containing histamine and proteases. Cytokines, chemokines, and lipid mediators are synthesized after cell activation. IL = interleukin; TNF = tumor necrosis factor.

causes mast cell activation. Proteolytic enzymes and toxic mediators, such as histamine, are released immediately from preformed granules, and chemokines, cytokines, and leukotrienes are synthesized after activation. Together, these mediators increase vascular permeability, break down tissue matrix proteins, promote eosinophil production and activation (IL-3, IL-5, and granulocyte-macrophage colony-stimulating factor [GM-CSF]), and cause influx of effector leukocytes (tumor necrosis factor-α [TNF-α], plateletactivating factor, and macrophage inflammatory protein [MIP-1]), constriction of smooth muscle, stimulation of mucus secretion, and amplification of TH2 cell responses (IL-4 and IL-13). Eosinophils and basophils, activated through cell surface FcεRs, rapidly release highly toxic granular proteins (major basic protein, eosinophil peroxidase, and collagenase) and, over a longer period, produce cytokines (IL-3, IL-5, and GM-CSF), chemokines (IL-8), prostaglandins, and leukotrienes that activate epithelial cells, leukocytes, and eosinophils to augment local inflammation and tissue damage. FcεR-bearing effectors act in a coordinated fashion. The immediate allergic inflammatory reaction initiated by mast cell products is followed by a latephase response that involves recruitment and activation of eosinophils, basophils, and TH2 lymphocytes.5 The manifestations of IgE-mediated reactions depend on the site of mast cell activation. Mast cells reside in vascular and epithelial tissue throughout the body. In a sensitized host (an individual with IgE responses to antigens), re-exposure to antigen leads to type I hypersensitivity responses only in the mast cells exposed to the antigen. Inhalation of antigens produces bronchoconstriction and increased mucus secretion (asthma and allergic rhinitis); ingestion of antigens causes increased peristalsis and secretion (diarrhea and vomiting); and the presence of subcutaneous antigens initiates increased vascular permeability and swelling (urticaria and angioedema). Blood-borne antigens cause systemic mast cell activation, increased capillary permeability, hypotension, tissue swelling, and smooth muscle contraction—the characteristics of systemic anaphylaxis.

Type II Hypersensitivity Reactions

Type II hypersensitivity reactions (Fig. 47-2) are caused by chemical modification of cell surface or matrix-associated antigens that generates “foreign” epitopes to which the immune system is not tolerant. B cells respond to this antigenic challenge by producing IgG, which binds to these modified cells and renders them susceptible to destruction through complement activation, phagocytosis, and antibody-dependent cytotoxicity. This phenomenon is seen clinically when drugs interact with blood constituents and alter their cellular antigens. Hemolytic anemia caused by immune-mediated destruction of erythrocytes (Chapter 160) and thrombocytopenia caused by destruction of platelets (Chapter 172), both type II hypersensitivity reactions, are adverse effects of certain drugs. Chemically reactive drug molecules bind covalently to the surface of red cells or platelets creating new epitopes that in a small subset of individuals are recognized as foreign antigens by the immune system and stimulate production of IgM and IgG antibodies reactive with the conjugate of drug and cell surface protein. Penicillin-specific IgG binds to penicillin-modified proteins on red blood cells and triggers activation of the complement cascade. Activation of complement components C1 through C3 results in covalent binding of C3b to

FIGURE 47-2.  Type II hypersensitivity. Type II responses are mediated by immunoglobulin G (IgG) directed against cell surface or matrix antigens, which initiates effector responses through the Fc receptor for IgG (FcγR) and complement. The relative contributions of these pathways vary with the IgG subclass and the nature of the antigen. Only FcγR-mediated phagocytosis by macrophages (MΦ) is depicted in this figure. Activation of complement components would result in binding of C3b to the red blood cell membrane, rendering red blood cells susceptible to phagocytosis and leading to formation of the membrane attack complex and cell lysis.

the red cell membrane and renders circulating red cells susceptible to phagocytosis by FcγR and complement receptor–bearing macrophages in the spleen or liver. Activation of complement components C1 through C9 and formation of the membrane attack complex cause intravascular lysis of red cells. The factors that predispose only some people to drug-induced type II hypersensitivity reactions are unknown. Penicillin, quinidine, and methyldopa have been associated with hemolytic anemia and thrombocytopenia through this mechanism. Another example is heparin-induced thrombocytopenia or thrombosis, a severe, life-threatening complication that occurs in 1 to 3% of patients exposed to heparin (Chapter 172). Interactions among heparin, human platelet factor 4, antibodies to the human platelet factor 4– heparin complex, platelet FcγRIIA, and splenic FcγRs (which remove opsonized platelets) are involved in the pathogenesis of this disease. Autoantibodies directed at antigens on the cell surface or extracellular matrix cause tissue damage by mechanisms similar to type II hypersensitivity reactions. IgG or IgM antibodies against erythrocytes lead to cell destruction in autoimmune hemolytic anemia because opsonized cells (coated with IgG or IgM and complement) are removed from the circulation by phagocytes in the liver and spleen or are lysed by formation of the membrane attack complex. Platelet destruction in autoimmune thrombocytopenic purpura occurs through a similar process. Because nucleated cells express membrane-bound complement regulatory proteins, they are less sensitive to lysis through the membrane attack complex, but when coated with antibody, they become targets for phagocytosis or antibody-dependent cytotoxicity. This mechanism is responsible for autoimmune and alloimmune neutropenia (Chapter 167). IgM and IgG antibodies recognizing antigens within tissue or binding to extracellular antigens cause local inflammatory damage through FcγR and complement mechanisms. Pemphigus vulgaris (Chapter 439) is a serious blistering disease that results from a loss of adhesion between keratinocytes caused by autoantibodies against the extracellular portions of desmoglein 3, an intercellular adhesion structure of epidermal keratinocytes. Another example of a type II hypersensitivity reaction is Goodpasture disease (Chapter 121), in which antibodies against the α3-chain of type IV collagen (the collagen in basem*nt membranes) are deposited in glomerular and lung basem*nt membrane. Tissue-bound autoantibodies activate monocytes, neutrophils, and basophils through FcγRs, initiating release of proteases, reactive oxidants, cytokines, and prostaglandins. Local activation of complement, particularly C5a, recruits and activates inflammatory cells and amplifies tissue injury. Neighboring cells are lysed by assembly of the membrane attack complex or by FcγR-initiated, antibody-dependent cytotoxicity. Autoantibodies against cell surface receptors produce disease by stimulating or blocking receptor function. In myasthenia gravis (Chapter 422), autoantibodies against the acetylcholine receptors on skeletal muscle cells bind the receptor and induce its internalization and degradation in lysosomes, reducing the efficiency of neuromuscular transmission and causing pro­ gressive muscle weakness. In contrast, Graves disease (Chapter 226) is

229

CHAPTER 47  Mechanisms of Immune-Mediated Tissue Injury  

Proteolytic enzymes Reactive oxidants Cytokines Chemokines Leukotrienes

Ag Immune complex

Complement activation C3a C5a Chemotaxis

FcγR activation Monocyte

FcγR

Eosinophil activation Ag

Platelets

Mediator release Oxidants Proteases Chemokines Cytokines

Tissue damage

IFN-γ

PMN

Complement receptors

T H1

Mφ

Mast cell

Mφ

Proteolytic enzymes Cytokines Chemokines Leukotrienes

TH2 Toxic proteins

IL-4 IL-5 Eotaxin

FIGURE 47-3.  Type III hypersensitivity. Type III responses are mediated by immunoglobulin G (IgG) directed against soluble antigens. Localized deposition of immune complexes activates mast cells, monocytes, neutrophils, and platelets bearing the Fc receptor for IgG (FcγR), and initiates the complement cascade, all effectors of tissue damage. Generation of complement components C3a and C5a recruits and stimulates inflammatory cells and amplifies effector functions. PMN = polymorphonuclear leukocyte (also called neutrophil).

CTL

Tissue damage

Direct cytotoxicity

Cell-associated antigen

characterized by autoantibodies that act as agonists. Autoantibodies to thyroid-stimulating hormone receptors bind the receptor, mimicking the natural ligand, inducing thyroid hormone overproduction, disrupting feedback regulation, and causing hyperthyroidism.

Type III Hypersensitivity Reactions

Type III hypersensitivity reactions (Fig. 47-3) are caused by tissue deposition of small soluble immune complexes that contain antigens and high-affinity IgG antibodies directed at these antigens. Localized deposition of immune complexes activates FcγR-bearing mast cells and phagocytes and initiates the complement cascade, all effectors of tissue damage.6 Immune complexes are generated in all antibody responses. The formation and the fate of immune complexes depend on the biophysical and immunologic properties of the antigen and the antibody. These properties include the size, net charge, and valence of the antigen; the class and subclass of the antibody; the affinity of the antibody-antigen interaction; the net charge and concentration of antibody; the molar ratio of available antigen and antibody; and the ability of the immune complex to interact with the proteins of the complement system. The lattice size of the immune complex is influenced strongly by the physical size and valence of the antigen, the association constant of antibody for that antigen, the molar ratio of antigen and antibody, and the absolute concentrations of the reactants. Larger aggregates fix complement more efficiently, present a broader multivalent array of ligands for complement and FcγRs to bind, and are taken up more readily by mononuclear phagocytes in the liver and spleen and thereby removed from the circulation. Smaller immune complexes, which form in antigen excess—as occurs early in an immune response—circulate in the blood and are deposited in blood vessels, where they initiate inflammatory reactions and tissue damage through interactions with FcγRs and complement receptors. Serum sickness is a systemic type III hypersensitivity reaction, historically described in patients injected with therapeutic horse antiserum for the treatment of bacterial infections. In general, serum sickness occurs after the injection of large quantities of a soluble antigen. Clinical features include chills, fever, rash, urticaria, arthritis, and glomerulonephritis. Disease manifestations become evident 7 to 10 days after exposure to the antigen, when antibodies are generated against the foreign protein and form immune complexes with these circulating antigens. Immune complexes are deposited in blood vessels, where they activate phagocytes and complement, producing widespread tissue injury and clinical symptoms. The effects are transient, however, and resolve after the antigen is cleared. A syndrome similar to serum sickness occurs in chronic infections in which pathogens persist in the face of continued immune response. In subacute bacterial endocarditis (Chapter 76), antibody production continues

FIGURE 47-4.  Type IV hypersensitivity. Type IV responses are mediated by T cells through three different pathways. In the first, type 1 helper T (TH1) cells recognize soluble antigens (Ag) and release interferon-γ (IFN-γ) to activate effector cells, in this case macrophages (MΦ), and cause tissue injury. In TH2-mediated responses, eosinophils predominate. TH2 cells produce cytokines to recruit and activate eosinophils, leading to their degranulation and tissue injury. In the third pathway, damage is caused directly by cytolytic T lymphocytes (CTL). IL = interleukin.

but fails to eliminate the infecting microbes. As the pathogens multiply, generating new antigens, immune complexes form in the circulation and are deposited in small blood vessels, where they lead to inflammatory damage of skin, kidney, and nerve. Hepatitis B virus infection (Chapters 148 and 149) may be associated with immune complex deposition early in its course, during a period of antigen excess, because antibody production in response to hepatitis B surface antigen is as yet relatively insufficient; some anicteric patients may present with acute arthritis. Mixed essential cryoglobulinemia, which may be associated with hepatitis C viral infection, is an immune complex–mediated vasculitis in which deposition of complexes containing IgG, IgM, and hepatitis C antigens causes inflammation in peripheral nerves, kidneys, and skin. Serum sickness also can develop in transplant recipients who are treated with mouse monoclonal antibodies specific for human T cells to prevent rejection, and in patients with myocardial infarction who are treated with the bacterial enzyme streptokinase to effect thrombolysis. Systemic lupus erythematosus (Chapter 266), the prototypical immune complex–mediated autoimmune disease, is characterized by circulating IgG directed against common cellular constituents, typically DNA and DNAbinding proteins. Small immune complexes are deposited in skin, joints, and glomeruli and initiate local tissue damage.

Type IV Hypersensitivity Reactions

Type IV hypersensitivity reactions (Fig. 47-4), also known as delayed-type hypersensitivity reactions, are mediated by antigen-specific effector T cells. They are distinguished from other hypersensitivity reactions by the lag time from exposure to the antigen until the response is evident (1 to 3 days). Antigen is taken up, processed, and presented by macrophages or dendritic cells. TH1 effector cells that recognize the specific antigen (these are scarce and take time to arrive) are stimulated to release chemokines, which recruit macrophages to the site, release cytokines that mediate tissue injury and growth factors that stimulate monocyte production. IFN-γ activates macrophages and enhances their release of inflammatory mediators, whereas TNF-α and TNF-β activate endothelial cells, enhance vascular permeability, and damage local tissue. The prototypical type IV hypersensitivity reaction

is the tuberculin test, but similar reactions can occur after contact with sensitizing antigens (e.g., poison ivy, certain metals) and lead to epidermal reactions characterized by erythema, cellular infiltration, and vesicles. CD8+ T cells also may mediate damage by direct toxicity. In contrast to TH1-mediated hypersensitivity reactions, in which the effectors are macrophages, eosinophils predominate in TH2-mediated responses. TH2 effector T cells are associated with tissue damage in chronic asthma (Chapter 87). TH2 cells produce cytokines to recruit and activate eosinophils (IL-5 and eotaxin), leading to degranulation, further tissue injury, and chronic, irreversible airway damage. Additional TH effector cells, such as TH17 cells, mediate tissue damage. TH17 cells produce IL-17 family cytokines, as well as IL-21, IL-22, and GM-CSF, that regulate innate effectors and orchestrate local inflammation by inducing release of proinflammatory cytokines and chemokines, proliferation and activation of effector cells and other target cells, recruitment of neutrophils, and enhanced TH2-mediated inflammation, all of which amplify allergic and autoimmune responses.3,7 TH17 cells have been implicated in allergic disorders (atopic dermatitis, asthma) and autoimmune and inflammatory diseases (psoriasis, inflammatory bowel disease, rheumatoid arthritis, systemic lupus erythematosus, multiple sclerosis). In some autoimmune diseases, effector T cells specifically recognize selfantigens to cause tissue damage, either by direct cytotoxicity or by inflammatory responses mediated by activated macrophages. In type 1 insulin-dependent diabetes mellitus, T cells mediate destruction of β cells of the pancreatic islets. IFN-γ-producing T cells specific for myelin basic proteins have been implicated in multiple sclerosis. Rheumatoid arthritis is another autoimmune disease caused, at least in part, by activated TH1 cells. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 47  Mechanisms of Immune-Mediated Tissue Injury  

GENERAL REFERENCES 1. Hogarth PM, Anania JC, Wines BD. The FcγR of humans and non-human primates and their interaction with IgG: implications for induction of inflammation, resistance to infection and the use of therapeutic monoclonal antibodies. Curr Top Microbiol Immunol. 2014;382:321-352. 2. Salazar F, Ghaemmaghami AM. Allergen recognition by innate immune cells: critical role of dendritic and epithelial cells. Front Immunol. 2013;4:356. 3. Singh RP, Hasan S, Sharma S, et al. Th17 cells in inflammation and autoimmunity. Autoimmun Rev. 2014;13:1174-1181.

230.e1

4. Shah A. The pathologic and clinical intersection of atopic and autoimmune disease. Curr Allergy Asthma Rep. 2012;12:520-529. 5. Voehringer D. Protective and pathological roles of mast cells and basophils. Nat Rev Immunol. 2013;13:362-375. 6. Karsten CM, Kohl J. The immunoglobulin, IgG Fc receptor and complement triangle in autoimmune diseases. Immunobiology. 2012;217:1067-1079. 7. Maddur MS, Miossec P, Kaveri SV, et al. Th17 cells: biology, pathogenesis of autoimmune and inflammatory diseases, and therapeutic strategies. Am J Pathol. 2012;181:8-18.

230.e2

CHAPTER 47  Mechanisms of Immune-Mediated Tissue Injury  

REVIEW QUESTIONS 1. Type III hypersensitivity reactions are caused by tissue deposition of immune complexes which activate local effectors and lead to injury. Which of the following mediators are NOT characteristic of this pathway of inflammation? A. Fc receptors for IgE B. Complement C. Neutrophils D. Mononuclear phagocytes Answer: A  Type III responses are mediated by IgG directed against soluble antigens. Localized deposition of immune complexes activates mast cells, monocytes, neutrophils, and platelets bearing the Fc receptor for IgG (FcγR) and initiates the complement cascade, all effectors of tissue damage. Generation of complement components C3a and C5a recruits and stimulates inflammatory cells and amplifies effector functions. IgE does not participate in the inflammatory response. Of note, mast cells may be activated in type III responses through receptors for IgG or complement. 2. Type IV hypersensitivity reactions, as exemplified by the tuberculin test, does NOT require which of the following elements? A. IgG B. Macrophages C. TH1 effector cells D. IFN-γ Answer: A  The prototypic type IV hypersensitivity reaction is the tuberculin test, in which antigen is taken up, processed, and presented by macrophages. Type 1 helper T (TH1) effector cells that recognize the specific antigen are stimulated to release chemokines, which recruit macrophages to the site, and release cytokines including IFN-γ, which activate macrophages and enhance their release of inflammatory mediators. IgG does not participate in type IV hypersensitivity reactions. 3. In which of the following clinical situations are autoantibodies NOT critical triggers of tissue damage? A. Asthma B. Pemphigus vulgaris C. Graves disease D. Systemic lupus erythematosus Answer: A  IgE antibodies against innocuous environmental antigens initiate asthma, not autoantibodies. Autoantibodies against TSH receptors act as agonists in Graves disease. Autoantibodies against desmoglein-3 effect adhesion between epidermal keratinocytes and cause blister formation. In systemic lupus erythematosus, inflammation is initiated by deposition of immune complexes containing autoantibodies directed against common cellular constituents (e.g., DNA and ribonucleolar proteins).

4. A syndrome similar to serum sickness occurs in which of the following infections? A. Hepatitis B B. Tuberculosis C. Pneumococcal pneumonia D. Influenza Answer: A  A syndrome similar to serum sickness occurs in chronic infections in which pathogens persist in the face of continued immune response. Hepatitis B virus infection may be associated with immune complex deposition early in its course, during a period of antigen excess, because antibody production in response to hepatitis B surface antigen is as yet relatively insufficient. Efficiency of clearance of immune complexes depends on their size and valence. Smaller immune complexes, which form in antigen excess—as occurs early in an immune response—circulate in the blood and are deposited in blood vessels, where they initiate inflammatory reactions and tissue damage through interactions with FcγRs and complement receptors. In influenza and pneumococcal pneumonia, there is no clinical evidence of systemic immune complex deposition. Tuberculosis generates T-cell responses. 5. The effector mechanisms that lead to tissue damage in autoimmune diseases are similar to those elicited in response to environmental antigens that result in allergy. Which hypersensitivity reaction is correctly matched to an autoimmune condition that occurs through a similar mechanism? A. Systemic lupus erythematosus and serum sickness B. Idiopathic thrombocytopenia purpura and asthma C. Multiple sclerosis and heparin-induced thrombocytopenia D. Myasthenia gravis and atopic dermatitis Answer: A  Systemic lupus erythematosus, the prototypical immune complex–mediated autoimmune disease, is characterized by circulating IgG directed against common cellular constituents, typically DNA and DNAbinding proteins. Like in the case of serum sickness, small immune complexes are deposited in skin, joints, and glomeruli and initiate local tissue damage. Idiopathic thrombocytopenia purpura, like type II hypersensitivity reactions, is mediated by autoantibodies that are directed to platelet surface antigens, whereas asthma is a triggered by IgE against innocuous environmental antigens (type I hypersensitivity reaction). T cells are key effectors in multiple sclerosis, whereas heparin-induced thrombocytopenia is caused by IgG autoantibodies. Myasthenia gravis is caused by IgG autoantibodies that recognize acetylcholine receptors and impair neuromuscular signaling, whereas atopic dermatitis is mediated by T cells.

230

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

complex somatic mutations and gene rearrangements. This provides defense tailored for each member of the species; its complexity and beauty permit specificity but also provide opportunities for error such as responses against self-antigens in autoimmunity.

Pathogen-Associated Molecular Pattern Recognition

48  MECHANISMS OF INFLAMMATION AND TISSUE REPAIR GARY S. FIRESTEIN Host defense mechanisms have evolved to recognize pathogens rapidly, render them harmless, and repair the damaged tissue. This complex and highly regulated sequence of events can also be triggered by environmental stimuli such as noxious mechanical and chemical agents. Under normal circ*mstances, tightly controlled responses protect against further injury and clear damaged tissue. In disease states, however, pathologic inflammation can lead to marked destruction of the extracellular matrix (ECM) and organ dysfunction.

INITIATION OF THE INFLAMMATORY RESPONSE

When normal tissue encounters a pathogen, resident cells are stimulated by engagement of pattern recognition receptors that activate an ancient arm of host defense known as innate immunity. In contrast to adaptive immunity, which provides exquisite antigen specificity, innate immune responses recognize common motifs on pathogens (Chapter 45). Additional cytoplasmic receptors can sense “danger” signals from a toxic environment or cellular stress, such as urate or adenosine triphosphate (ATP). Innate mechanisms are designed for rapid responses (minutes to hours) compared with the more leisurely adaptive system that can take days to weeks to develop. In addition to orchestrating early events that are critical to host defense, cells of the innate system like dendritic cells orchestrate the subsequent adaptive cascade through the generation of chemokines that organize lymphoid tissue and presentation of antigens to lymphocytes. Innate immunity provides intergenerational continuity in that the receptors are encoded in the germline and are passed unchanged to progeny to protect the species. In contrast, each individual must generate his or her own adaptive immune system through

The toll-like receptor (TLR) family of proteins binds common patterns of molecular structures on microbial pathogens that normally are not found in mammalian cells. The TLRs are critical members of the innate immune system and serve as sentinels that initiate a rapid response.1 Some are expressed on the cell surface, such as TLR2, which is activated primarily by bacterial peptidoglycan and lipoproteins, and TLR4, which is activated by lipopolysaccharide (LPS, or endotoxin). Others are expressed mainly on the inner leaflet of cytoplasmic vesicles, like TLR9, which is activated by unmethylated bacterial sequences that are enriched for CpG motifs (regions of DNA where cytosine and guanine nucleotides in the linear sequence of bases along its length are separated by one phosphate), or TLR3 and TLR7, which are important for antiviral defense because they bind double-stranded and single-stranded viral RNA, respectively. In addition to exogenous molecules, some endogenous structures can bind to TLRs, including heat shock proteins and oxidized low-density lipoproteins (oxLDLs). The latter might be especially important in the pathogenesis of atherosclerosis, in which LDL activates TLR4 within vascular plaques. Local endothelial cell– and macrophage-derived chemotactic factors can then recruit activated T cells into the atheroma. Signaling by TLR2 and TLR4 progresses through adaptor proteins and often converges on a kinase known as MyD88, which orchestrates several downstream cascades. By directing the phosphorylation of IκB kinase-β (IKKβ), MyD88 activates nuclear factor-κB (NF-κB), a master switch for inflammatory genes.2 Translocation of NF-κB to the cell nucleus stimulates the production of cytokines (e.g., interleukin-6 [IL-6], IL-8, and tumor necrosis factor [TNF]), the machinery for prostaglandin release (e.g., cyclooxygenase 2 [COX2]), and genes that regulate the ECM (e.g., metalloproteinases). This rapid response is normally transient, although it can persist in pathogenic states. MyD88-independent pathways that stimulate innate immunity also exist. For instance, TLR3 stimulation by RNA viruses uses a separate pathway involving IKKε and interferon regulating factor-3 (IRF-3). IRF-3, in combination with several other transcription factors, induces the expression of genes such as interferon-β (IFN-β) to establish an antiviral state. These genes primarily offer protection against pathogens by initiating key defense mechanisms. However, these same pathways can create a hazardous milieu that is toxic to normal cells through the production of oxygen radicals, nitric oxide, and other reactive intermediaries. These molecules can damage DNA and harm bystander cells, or even lead to neoplasia (E-Table 48-1). For instance, long-standing inflammation in the colon, as in ulcerative colitis, is associated with adenocarcinoma. Increased COX2 expression as a result of NF-κB translocation is another mechanism that contributes to the development of tumors at inflammatory sites. An unanticipated finding is that NF-κB itself can also directly augment carcinogenesis by serving as a survival signal for damaged cells that would normally be deleted by apoptosis. The TLR signal transduction mechanisms integrate the environmental stimuli and generate a broadly antipathogen response. Fine-tuning of host defenses against unique pathogen structures to provide long-lived immunity requires the slower, more precise adaptive immune system. Although it is more cumbersome and primitive, innate immunity provides signals that activate adaptive responses. For instance, TLRs can direct dendritic cells (Chapter 45), which have internalized and processed antigen, to migrate from peripheral tissues to central lymphoid organs. The dendritic cells can also produce cytokines and, after maturation, present antigens to T cells in the context of class II major histocompatibility molecules and surface costimulatory proteins. The activated T cells can then migrate to the tissue to enhance and amplify the host response. T cells also provide help to B cells, thereby stimulating antibody production and activating other components of innate immunity (e.g., the complement system, Chapter 50). Other non-TLR cytoplasmic sensors also serve a similar purpose in the environment. For instance, retinoic acid−inducible gene 1 (RIG-1) and melanoma differentiation−associated gene 5 (MDA5) can detect RNA viruses and initiate an inflammatory response. These are, in some cases, partially redundant with TLR3 and TLR7 and can activate similar signaling mechanisms, such as NF-κB through the IKKß and IRFs through a distinct pathway involving IKKε and TBK1.

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

E-TABLE 48-1  EXAMPLES OF INFLAMMATION PATHWAYS IN DISEASE DISEASE

ACTIVATED PATHWAYS

Atherosclerosis

Toll-like receptor activation (e.g., oxLDL) Chemokine-mediated leukocyte recruitment (e.g., MCP-1)

Cancer

Reactive oxygen and nitrogen intermediate-induced mutations Cyclooxygenase 2–mediated neoplasia (e.g., colon, breast) NF-κB prolonging survival of damaged cells

Asthma

IgE-mediated mast cell activation TH2 cytokine-mediated leukocyte activation Leukotriene-induced bronchospasm Protease-induced airway remodeling

Rheumatoid arthritis

Toll-like receptor activation (e.g., peptidoglycan) Macrophage/fibroblast cytokine production, including IL-1, TNF, and IL-6 Cyclooxygenase 2 induction Protease-mediated cartilage destruction Synovial complement activation

Systemic lupus erythematosus

Complement activation in multiple organs α-Interferon production and interferon signature

Autoinflammatory diseases, including psoriasis

Inflammasome activation, including production of IL-1, IL-18, and IL-33 TH17 cell activation IL-17A-, IL-12-, and IL-23-mediated inflammation

IgE = immunoglobulin E; MCP-1 = monocyte chemoattractant protein 1; NF-κB = nuclear factor-κB; oxLDL = oxidized low-density lipoprotein; TH17 = helper T lymphocyte type 17; TH2 = helper T lymphocyte type 2; TNF = tumor necrosis factor.

230.e3

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

Environmental Stress and Danger-Associated Molecular Patterns Danger-associated molecular pattern molecules serve as a mechanism to detect and respond to damage to the microenvironment. Tissue injury due to direct trauma or a noxious stimulus initiates an inflammatory response and is associated with microvascular damage, extravasation of leukocytes through vascular walls, and leakage of plasma and proteins into the tissue. Endogenous proteins, including ATP receptors, S100, heat shock proteins, and high mobility-group box 1 (HMGB1), mediate release of molecules that reflect cellular toxicity and induce a cellular response. Acid-sensitive ion channels (ASICs) on the cell surface can also detect the environmental stress caused by a decrease in tissue pH. ASICs can mediate a variety of cellular functions, including cell death through apoptosis or pain responses that can lead to adaptive pain behaviors that limit further exposure to noxious stimuli.

Proteases, Coagulation, and Inflammation Although the coagulation system’s primary function is to maintain vascular integrity (Chapter 171), the proteases that regulate its functions also play an important role in the early responses to tissue damage and inflammation. For example, plasminogen is a circulating proenzyme that can be cleaved to plasmin by enzymes in the coagulation pathway, including factors XIa and XIIa. Tissue plasminogen activating factor and kallikrein also have this capacity. When activated, the serine protease plasmin can digest fibrin, fibronectin, thrombospondin, and laminin as well as activating pro-matrix metalloproteinases like collagenase (MMP1). By remodeling the extracellular matrix, this system can ultimately regulate cell recruitment and tissue damage. Thrombus formation at the site of vascular damage can begin the inflammatory cascade through the release of vasoactive amines (e.g., serotonin), release of lysosomal proteases, and formation of eicosanoid products. The platelets can also later regulate healing with release of growth factors such as platelet-derived growth factor (PDGF) and transforming growth factor-β (TGFβ).

Inflammasome The inflammasome3 is among the best characterized mechanisms for sensing danger and includes the 22-member human Nod-like receptor (NLR) family of cytoplasmic proteins. The activated NLR proteins recruit additional proteins to form a complex with caspase-1 and adaptor molecule apoptosisassociated specklike protein (ASC). Activation of caspase-1 is a key function of inflammasomes, with resultant cleavage and activation of IL-1, IL-18, and IL-33. The latter molecule is also known as an “alarmin” because of its rapid release in the presence of tissue damage or a pathogen. Alarmins are often preformed in cells, such as mast cells, and can be either released directly into the microenvironment or quickly processed and secreted. Other alarmins include products of cell destruction, such as ATP or uric acid. Disorders of the inflammasome are associated with a group of conditions known as autoinflammatory diseases (Chapter 261). The prototypic syndromes known as familial cold autoinflammatory disease, Muckle-Wells disease, and neonatal-onset multisystem inflammatory disease (NOMID) are due to nonconserved mutations in the NLR gene that encodes cryopyrin (also known as NALP3). These rare diseases are characterized by abnormal inflammasome activation with aberrant release of processed IL-1β. The clinical manifestations, including fever, rash, hearing impairment, and arthritis, depend on the specific amino acid substitution as well as other less welldefined genetic influences. The critical role of IL-1 has been proved by studies using treatment with IL-1 inhibitors, which prevent flares and can reverse end-organ damage. The inflammasome also participates in some common diseases, such as gout (Chapter 273), in which urate crystals can activate the inflammasome.

Immune Complexes and Complement The complement system (Chapter 50) is another ancient defense mechanism that links innate immunity and the humoral arm of adaptive immunity. Both the classical complement pathway, activated by immunoglobulin G (IgG)and IgM-containing immune complexes, and the alternative pathway, activated by bacterial products, converge at the third component of complement, C3, with proteolytic release of fragments that amplify the inflammatory response and mediate tissue injury. C3a and C5a directly increase vascular permeability and contraction of smooth muscle. C5a induces mast cell release of histamine, thereby indirectly mediating increased vascular permeability. C5a also activates leukocytes and enhances their chemotaxis,

231

adhesion, and degranulation, with release of proteases and toxic metabolites. C5b attaches to the surface of cells and microorganisms and is the first component in the assembly of the C5b-9 membrane attack complex. Individuals with abnormalities of the early complement components, especially C1q, C2, and C4, usually have a minimally increased incidence of infection but demonstrate an enhanced risk for developing autoimmune diseases such as systemic lupus erythematosus (SLE) (Chapter 266). The mechanism of increased disease susceptibility is probably related to inefficient clearance of immune complexes. Enhanced activation and consumption of complement proteins can also occur in SLE accompanied by low plasma C3 and C4 levels, especially in association with disease exacerbations. C3 or C5 deficiency increases susceptibility to bacterial infections, whereas defects in the late components that form the membrane attack complex result in an increased incidence of Neisseria sp bacteremia (Chapter 298).

SECOND WAVE OF THE INFLAMMATORY RESPONSE

Activation of innate immunity quickly leads to the robust influx of inflammatory cells. Resident cells, such as vascular endothelial cells, mast cells, dendritic cells, and interstitial fibroblasts, respond by releasing soluble mediators, including eicosanoids and pro-inflammatory cytokines (E-Table 48-2). These mediators amplify the inflammatory response and recruit additional leukocytes. Locally stimulated cells, along with the newly arrived inflammatory cells, release toxic reactive intermediates of nitrogen and oxygen as well as a myriad of proteases, principally matrix metalloproteinases (MMPs), serine proteases, and cysteine proteases. These molecules help destroy infectious agents and remove damaged cells, thus clearing the injured site for tissue repair. In most situations, the normal physiologic response is an exquisitely coordinated program that uses proteolytic enzymes to remodel the ECM and promote a supportive environment for wound healing rather than tissue damage.

Cellular Response Inflammatory cell infiltration at the site of initial tissue damage typically begins with release of chemokines and soluble mediators from resident cells, including interstitial fibroblasts, mast cells, and vascular endothelial cells. Signaling from these events alters the local adhesion molecule profile and creates a chemotactic gradient that recruits cells from the blood stream. Mast cells, in particular, act as sentinels that degranulate within seconds after ligation of immunoreceptors and activation of the signaling molecule spleen tyrosine kinase (Syk) to release vasoactive amines. In most acute responses, polymorphonuclear leukocytes (PMNs) are the first inflammatory cells to arrive at the site of injury, followed later by mononuclear cells. Most tissue fibroblasts and vascular endothelial cells are generally quiescent before migration of PMNs into the tissue. However, these resident cells can be triggered to proliferate and migrate toward the site of injury as well as to synthesize cytokines, proteases, and ECM components. Growth factors are released, such as basic fibroblast growth factor (bFGF) and vascular endothelial growth factor (VEGF), stimulating new blood vessel formation. Together with granulocyte-macrophage colony-stimulating factor (GMCSF), these locally released growth factors contribute to cellular proliferation and amplification of the inflammatory response and also induce maturation of dendritic cells that process antigens. In addition, fibroblasts and endothelial cells secrete new ECM proteins, MMPs, and other ECM-digesting enzymes. Initially, the response favors proteolytic activity to clear damaged infrastructure. This is followed by a shift to increased production of new ECM to allow tissue repair and wound healing. Increased vascular permeability, caused by disruption of endothelial cell tight junctions, allows blood-borne proteins such as fibrinogen, fibronectin, and vitronectin to extravasate into the perivascular ECM. Interaction with preexisting ECM allows the assembly of new ligands for a subset of adhesion molecules (e.g., integrins α5β1 and αvβ3). This increased vascular permeability and change in the profiles of adhesion molecules and ligands, in conjunction with release of chemoattractant molecules, leads to the recruitment of leukocytes to sites of inflammation. Some of the chemokines involved are IL-8 (for neutrophils), macrophage chemoattractant protein-1 (MCP-1) for monocytes, RANTES (regulated on activation, T-cell expressed and secreted) for monocytes and eosinophils, and IL-16 (for CD4+ T cells). Chemokines have the capacity to recruit specific subsets of cells by binding to G protein−coupled chemokine receptors. Directly targeting chemokines, either with biologics or with small molecules, has met with limited success in clinical trials, perhaps because the system is quite complex and highly

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

E-TABLE 48-2  SIGNALS FOR INDUCTION AND REPAIR OF INFLAMMATION INFLAMMATION

RESOLUTION AND TISSUE REPAIR

CYTOKINES AND GROWTH FACTORS TNF

TGF-β

IL-1 family (IL-1, IL-18, IL-33)

IL-10

IL-6 family (IL-6, IL-11, LIF, osteopontin)

FGF

IL-4, IL-13

Osteoprotegerin

IL-15

IL-1RII

IL-17 family (IL-17A-F)

IL-1Ra

IL-12 family (IL-12, IL-23, IL-27)

Soluble TNF-R

VEGF

IL-18 binding protein

Chemokines HMBG1 PROTEASES Matrix metalloproteinases   Collagenases   Gelatinases   Stromelysins   Matrilysins

TIMPs

Serine proteases   Trypsin   Chymotrypsin   Kallikrein   Plasmin

SERPINs, α2-macroglobulin

Cysteine proteases ADAMTS family   Aggrecanases SMALL MOLECULE MEDIATORS Prostaglandins (especially PGE2) Leukotrienes (especially LTB4) C3a and C5a

Lipoxins Cyclopentenone Antioxidants

Histamine Bradykinin Reactive oxygen Reactive nitrogen APOPTOSIS REGULATORS Soluble Fas ligand

Fas TRAIL Reactive oxygen Reactive nitrogen

ADAMTS = a disintegrin and metalloproteinase family; FGF = fibroblast growth factor; IL = interleukin; LIF = leukemia inhibitory factor; R = receptor; Ra = receptor antagonist; SERPINs = serine protease inhibitors; TGF = transforming growth factor; TIMPs = tissue inhibitors of metalloproteinase; TNF = tumor necrosis factor; TRAIL, TNF-related apoptosis-inducing ligand; VEGF = vascular endothelial growth factor; HMBG1= high mobility-group box 1.

231.e1

232

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

redundant. An alternative approach might be to target intracellular mechanisms distal to receptor ligation. Chemokine receptors generally signal through the phosphoinositide-3 kinase (PI3K) system, especially the gamma isoform. PI3Kγ is mainly expressed in bone marrow−derived cells and is the convergence point for multiple chemotactic factors. Preclinical studies suggest that blocking this pathway decreases inflammatory cell recruitment in models of lupus and rheumatoid arthritis. The precise combination of chemokines and vascular adhesion molecules present in an inflammatory lesion determines the timing for recruitment of individual inflammatory cell types. Ligation of integrins on leukocytes also prolongs cell survival after they have moved into the tissue, by preventing apoptosis. The central role of certain specific adhesion molecule−ligand pairs has been confirmed in human diseases. For instance, α4β1 plays a key role in the recruitment of lymphocytes to the central nervous system in multiple sclerosis, and blocking this interaction suppresses disease activity (Chapter 411). Eosinophils use the same adhesion receptors to migrate into the lung in allergen-induced asthma (Chapter 87). Increased expression of intracellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1), as well as increased chemokine expression, is evident in other cell types, such as the airway epithelium after allergen challenge in asthma. Rapid transient influx of neutrophils occurs in allergic airway disease, along with activation of the local T cells and mast cells. These neutrophils produce lipid mediators, reactive oxygen intermediates, and proteases such as elastase, which may contribute to airflow obstruction, epithelial damage, and remodeling. Neutrophil elastase, together with chemokines released by both recruited and allergen-activated T cells and mast cells, serves to recruit eosinophils.

Soluble Mediators

PRO-INFLAMMATORY CYTOKINES

Pro-inflammatory cytokines, often derived from macrophages and fibroblasts, are mediators that activate the immune system. The pro-inflammatory members of the IL-1 family (e.g., IL-α, IL-1β, IL-18, and IL-33) and TNF have pleiotropic activities and can enhance adhesion molecule expression on endothelial cells, induce proliferation of endogenous cells, and stimulate antigen presentation. IL-1 and TNF also increase expression of matrixdegrading enzymes, such as collagenase and stromelysin. In addition, they stimulate synthesis of other inflammatory mediators such as prostaglandins from fibroblasts. TNF inhibitors (Chapter 36) are effective in inflammatory diseases such as psoriasis, rheumatoid arthritis, and inflammatory bowel disease, and IL-1 inhibitors (Chapter 36) are beneficial in genetic diseases such as Muckle-Wells syndrome and familial cold autoinflammatory syndrome. IL-1 and TNF comprise only a small fraction of the acute cytokine response. Many other factors also participate, including IL-6 and its related cytokines (IL-11, osteopontin, and leukemia inhibitory factor), which can both induce acute phase reactants and bias an immune response toward a helper T type 1 (TH1) or TH2 phenotype (Chapter 47). GM-CSF can regulate dendritic cell maturation, increase expression of human leukocyte antigen (HLA-DR) on these cells, and enhance antigen presentation. The TH1 lymphokine IFN-γ, although often considered part of the secondary wave that ensues after T-cell activation, can also induce expression of HLA-DR, increase expression of endothelial cell adhesion molecules, and inhibit collagen production. IL-1, IL-6, and IL-23 can coordinate differentiation toward TH17 cells, a phenotype that is thought to play a major role in inflammation and autoimmunity owing to the production of IL-17 family members (IL-17A through F). Of these, IL-17A and perhaps IL-17F are especially important because they can synergize with IL-1 and TNF. The growth factor TGF-β biases cells toward the regulatory T cell (Treg) phenotype, which can suppress antigen-specific responses of other T cells (see later). The benefit of individual cytokine inhibitors varies depending on the disease. For instance, IL-6 blockade is effective in rheumatoid arthritis, whereas IL-12/23 and IL-17A inhibition suppresses skin inflammation in psoriasis. Clinical trials now clearly show that IL-17A antibodies are effective in psoriasis. A1  Many cytokines activate cells by ligating their receptors and engaging the Janus kinase ( JAK) family of signaling molecules, including JAK1, JAK2, JAK3, and Tyk2. These kinases, in turn, phosphorylate the signal transducer and activator of transcription (STAT) proteins. The STATs serve as transcription factors that initiate expression of many other cytokines and mediators of the inflammation and amplify the response. JAK inhibition represents an alternative approach to abrogating the inflammatory response.

Cytokines play a key role in the establishment and perpetuation immunemediated diseases. As noted earlier, autocrine and paracrine cytokine networks play a critical role in the perpetuation of inflammation in rheumatoid arthritis4 (Chapter 264). MCP-1 recruits and activates macrophages into atheromas containing oxLDLs and foam cells. In allergic asthma (Chapter 87), IL-13 is emerging as a central inflammatory cytokine. IL-13 functions through binding to cell surface IL-4 receptors, and IL-4R–deficient mice are relatively resistant to the development of asthma.

EICOSANOIDS

In addition to cytokines and immune complexes, local inflammatory responses lead to the release of eicosanoids, which are lipid-derived molecules. Because lipids are present in the cell membrane, they are readily available substrates for the synthesis of mediators. These molecules are produced adjacent to sites of injury, and their half-lives range from seconds to minutes. Eicosanoids are not stored but are produced de novo from membrane lipids when cell activation by mechanical trauma, cytokines, growth factors, or other stimuli leads to release of arachidonic acid. Cytosolic phospholipase A2 (cPLA2) is the key enzyme in eicosanoid production. Cell-specific and agonist-dependent events coordinate the translocation of cPLA2 to the nuclear envelope, endoplasmic reticulum, and Golgi apparatus, where interaction with COX (in the case of prostaglandin synthesis) or 5-lipoxygenase (in the case of leukotriene synthesis) can occur.

PROSTAGLANDINS

Prostanoids5 are produced when arachidonic acid is released from the plasma membrane of injured cells by phospholipases and metabolized by cyclooxygenases and specific isomerases (Chapter 37). These molecules act both at peripheral sensory neurons and at central sites within the spinal cord and brain to evoke pain and hyperalgesia. Their production is increased in most acute inflammatory conditions, including arthritis and inflammatory bowel disease. In response to exogenous and endogenous pyrogens, prostaglandin E2 (PGE2) derived from COX2 mediates a central febrile response. In addition, prostaglandins synergize with bradykinin and histamine to enhance vascular permeability and edema. The levels of prostaglandins are usually very low in normal tissues and increase rapidly with acute inflammation, well before leukocyte recruitment. COX2 induction with inflammatory stimuli most likely accounts for the high levels of prostanoids in chronic inflammation. COX2 also plays a key role in platelet−endothelial cell interactions by increasing the production of prostacyclin (PGI2) in endothelial cells (Chapter 37). Increased risk for myocardial infarction associated with the use of selective COX2 inhibitors may be related to unopposed production of thromboxane A2 by COX1 in platelets. Prostacyclin also protects against atherosclerosis in mice, and COX2 blockade abrogates this beneficial effect. Thus, COX inhibitors can potentially increase thrombotic events.

LEUKOTRIENES

A distinct set of enzymes direct arachidonic acid metabolites toward the synthesis of leukotrienes (Chapter 87). Their relative importance depends on the specific target organ of an inflammatory response. For instance, leukotriene receptor antagonists are effective in asthma, whereas similar approaches have been less impressive in rheumatoid arthritis. Unlike prostaglandins, leukotrienes are primarily produced by inflammatory cells such as neutrophils, macrophages, and mast cells. 5-Lipoxygenase is the key enzyme in this cascade, transforming released arachidonic acid to the epoxide leukotriene A4 (LTA4) in concert with 5-lipoxygenase-activating protein (FLAP). LTA4 can be hydrolyzed by cytosolic LTA4 hydrolase to LTB4, a potent neutrophil chemoattractant and stimulator of leukocyte adhesion to endothelial cells. LTA4 can also conjugate with glutathione to form LTC4 by LTC4 synthase at the nuclear envelope. LTC4 can be metabolized extracellularly to LTD4 and LTE4. These three cysteinyl leukotrienes promote plasma leakage from postcapillary venules, upregulation of expression of cell surface adhesion molecules, and bronchoconstriction.

HISTAMINE

Histamine is a vasoactive amine produced by basophils and mast cells that markedly increases capillary leakage. In basophils, histamine is released in response to bacterial formylmethionyl-leucyl-phenylalanine (f-MLP) sequences, complement fragments C3a and C5a, and IgE. The resultant edema can be readily observed clinically in urticaria (Chapters 252 and 440) and allergic rhinitis (Chapter 251). The stimulus for release of histamine from

233

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

mast cell granules is the same as in basophils, except for the absence of f-MLP receptors in this cell type. Histamine can also synergize with locally produced LTB4 and LTC4. In addition, histamine enhances leukocyte rolling and firm adhesion, and induces gaps in the endothelial cell lining, enhancing leukocyte extravasation. Despite the production of histamine in asthma and in acute synovitis, currently available histamine blockers have minimal therapeutic effect in these conditions. Targeting the more recently described histamine type 4 receptor (HR4), which has a variety of immunomodulatory effects on bone marrow−derived cells, suggests that more precise inhibition of this novel histamine pathway might have greater success.6

KININS

Kinins induce vasodilation, edema, and smooth muscle contraction, as well as pain and hyperalgesia, through stimulation of C fibers. They are formed from high- and low-molecular-weight kininogens by the action of serine protease kallikreins in plasma and peripheral tissues. The primary products of kininogen digestion are bradykinin and lysyl-bradykinin. These products have high affinity for the B2 receptor, which is widely expressed and is responsible for the most common effects of kinins. The peptides desArg-BK and Lys-desArg-BK are generated by carboxypeptidases and bind the kinin B1 receptor subtype, which is not expressed in normal tissues but is rapidly upregulated by TLR ligands and cytokines. The kinin B2 receptor is internalized rapidly and desensitized, whereas the B1 receptor remains highly responsive. Kinin actions are associated with the secondary production of other mediators of inflammation, including nitric oxide, mast cell–derived products, and the pro-inflammatory cytokines IL-6 and IL-8. In addition, kinins can increase IL-1 production through initial stimulation of TNF and can increase prostanoid production through activation of phospholipase A2 and release of arachidonic acid.

NEURAL NETWORKS

Neural outflow also can rapidly activate inflammatory mechanisms and alter vascular permeability at sites of tissue damage. Pain receptors can activate type δ fibers and carry information to the spinal cord about noxious stimuli where cytokines like IL-1 or TNF are produced. Spinal cytokines lead to phosphorylation of signal molecules in the central nervous system like mitogen activated protein kinases (MAPKs). Reflex neural loops, including sympathetic and parasympathetic nerves, release mediators like substance P, acetylcholine, epinephrine or norepinephrine into the immediate location as well as surrounding tissue. Vascular permeability and activation of resident cells like macrophages can help recruit additional cells to the affected region.

MECHANISMS OF TISSUE DAMAGE IN INFLAMMATION

Nitric oxide synthases (NOS) convert l-arginine and molecular oxygen to l-citrulline and nitric oxide (NO). There are three known isoforms of NOS: neuronal NOS (ncNOS or NOS1) and endothelial cell NOS (ecNOS or NOS3) are both constitutively expressed, whereas macrophage NOS (macNOS, iNOS, or NOS2) is induced by inflammatory cytokines such as TNF and IFN-γ, as well as by products of viruses, bacteria, protozoa, and fungi and by low oxygen tension and low environmental pH. Together with prostaglandins, the production of NO by NOS2 and ROIs by NADPH oxidase is a key mechanism by which macrophages paradoxically impair T-cell proliferation. This might control inflammatory processes or delete autoreactive T cells and partially accounts for the immunosuppression observed in certain infections and malignancies.

Proteases and Matrix Damage Production of enzymes that degrade the ECM regulates tissue turnover in inflammation. Reconfiguring of the matrix remodels damaged tissue, releases matrix-bound growth factors and cytokines, prepares the tissue for the ingrowth of new blood vessels, and alters the local milieu to permit adherence and retention of newly recruited cells. The MMPs are a family of more than 20 extracellular endopeptidases that participate in degradation and remodeling of the ECM matrix (Table 48-1). They are produced as pro-enzymes and require limited proteolysis or partial denaturation to expose the catalytic site. Their name is derived from their dependence on metal ions (zinc/metzincin superfamily) for activity and from their potent ability to degrade structural ECM proteins. MMPs can also cleave cell surface molecules and other pericellular nonmatrix proteins, thereby regulating cell behavior. For instance, MMPs can alter cell growth by digesting matrix proteins associated with growth factors. FGF and TGF-β have high affinities for matrix molecules that serve as depots for storage of these cytokines. Matrix proteolysis releases some growth factors and can make them available to cell surface receptors. In addition, MMPs can directly cleave and activate growth factors. MMPs affect cell migration by altering cell-matrix or cell-cell receptor sites. The adhesion molecule β4 integrin is

TABLE 48-1  COMMON MATRIX METALLOPROTEINASES AND THEIR SUBSTRATES MMP FAMILY

OTHER SUBSTRATES

Collagen I, II, III, VII, and X Pro-MMP-1, -2, -8, -9, and -13 Aggrecan Pro-TNF

Entactin

α1-Proteinase inhibitors Gelatin Tenascin

Gelatinases

Aggrecan Denatured collagen Elastin Fibronectin Laminin Vitronectin

Pro-MMP-1, -2, and -13 Pro-TNF Pro-IL-1β Latent TGF-β

Matrilysins

Proteoglycans Denatured collagens Entactin Fibrin, fibrinogen Fibronectin Gelatin Laminin Tenascin Vitronectin

Pro-MMP-2 and -7 Pro-TNF Membrane-bound Fas ligand (FasL) Plasminogen β4 Integrins

Stromelysins

Proteoglycans Aggrecan Collagen III, IV, V, IX, X, and XI Pro-IL-1β Entactin Fibrin, fibrinogen Fibronectin Gelatin Laminin Tenascin Vitronectin

Pro-MMP-1, -3, -7, -8, -9, -10, and -13 Pro-TNF Plasminogen α1-Proteinase inhibitors

Reactive Oxygen and Nitrogen

Macrophages, neutrophils, and other phagocytic cells can generate large amounts of toxic reactive oxygen intermediates (ROIs) and reactive nitrogen intermediates (RNIs) that can directly kill pathogens. ROIs and RNIs also serve as critical signal transduction molecules that regulate expression of inflammatory genes. These molecules can also have deleterious effects on normal tissue by damaging DNA, oxidizing membrane lipids, and nitrosylating proteins. Release of reactive intermediates can be initiated by microbial products such as LPS and lipoproteins, by cytokines such as IFN-γ and IL-8, and by engagement of Fc receptors by IgG. These events cause translocation of several cytosolic proteins, including Rac2 and Rho-family guanosine triphosphatase (GTPase) to the membrane-bound complex carrying cytochrome c, with subsequent activation of reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase. The reaction catalyzed by NADPH oxidase leads to superoxide production, which, in turn, increases hydrogen peroxide, hydroxyl radicals and anions, hypochlorous acid, and chloramines. In some cases, ROIs can contribute directly to the initiation of chronic disease. Lipid oxidation produces aldehydes that substitute lysine residues in apolipoprotein B-100. This altered moiety either binds to TLR2 to induce cytokine production or is internalized by macrophages, leading to the production of foam cells and fatty streaks, the primary lesions of atherosclerosis (Chapter 70). Subsequently, altered epitopes in damaged host proteins can be presented to T cells to initiate an adaptive immune response that amplifies the inflammatory vascular lesion.

MATRIX SUBSTRATES

Collagenases

IL = interleukin; MMP = matrix metalloproteinase; TGF = transforming growth factor; TNF = tumor necrosis factor.

234

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

cleaved by MMP-7. MMP-3 and MMP-7 digest E-cadherin and not only disrupt endothelial cell junctions but also stimulate cell migration. Degradation of the ECM is usually initiated by collagenases, which cleave native collagen. Denatured collagen is then recognized and further degraded by gelatinases and stromelysins. Unlike the collagenases, stromelysins demonstrate broad substrate specificity and act on many ECM proteins, such as proteoglycan, fibronectin, laminin, and many cartilage proteins. Stromelysins can also amplify the remodeling process by activating collagenase through limited proteolysis. MMP gene expression can be induced by many pro-inflammatory cytokines, including TNF, IL-1, IL-17A, and IL-18. One common element in MMP promoters that regulates transcription is activator protein-1 (AP-1). AP-1 is a dimer that includes members of the Jun and Fos families. Cytokines can regulate the MMP gene by activating MAPKs, especially c-Jun amino terminal kinase ( JNK), which, in turn, phosphorylates c-Jun and markedly enhances MMP production. NF-κB and NF-κB-like binding sites also can contribute to protease transcription. Several other classes of proteases remodel the matrix, including serine proteases and cysteine proteases. High levels of active serine proteases, such as trypsin, chymotrypsin, and elastase, are released by infiltrating PMNs at sites of inflammation and can directly digest the ECM or activate the proenzyme forms of secreted MMPs. The ADAM (a disintegrin and metalloproteinase) family can cleave the extracellular domain of cytokine receptors. These ECM proteases include two members of the aggrecanase family. One of the aggrecanases (aggrecanase 2, or ADAMTS5) has been implicated in osteoarthritis because mice deficient in this enzyme have decreased cartilage destruction in models of osteoarthritis (Chapter 262).

TISSUE REPAIR AND RESOLUTION OF INFLAMMATION

Inflammation is a normal physiologic response but can cause serious host injury if allowed to persist. Additional mechanisms are required to reestablish homeostasis after this response is initiated. Suppression of acute inflammation by removal or deactivation of mediators and effector cells permits the host to repair damaged tissues through elaboration of appropriate growth factors and cytokines (Fig. 48-1). As in the initial generation of an inflammatory response, components of resolution include a cellular response (apoptosis and necrosis), formation of soluble mediators (such as anti-inflammatory cytokines and antioxidants), and production of direct effectors (such as protease inhibitors).

Deletion of Inflammatory Cells Cells can be removed from an inflammatory site by several mechanisms. First, the influx of cells can be decreased by suppressing chemotactic factor produc-

tion and vascular adhesion molecule expression. Second, cells, especially lymphocytes, can be released from the tissue and return to the circulation through lymphatics. Third, stressed cells can undergo necrosis with the release of their contents into the local environment. A fourth mechanism, known as autophagy,7 can lead to digestion of internal organelles and ultimately to cell death. Perhaps the most critical and effective method for clearing cells from an inflammatory site is programmed cell death, or apoptosis. Apoptosis is a highly regulated process in eukaryotic cells that leads to cell death and marks the surface membrane for rapid removal by phagocytes. This clearance process does not elicit an inflammatory response, in contrast to cell death by necrosis. PMN phagocytes have a very short half-life in the tissue, and the persistence or release of their contents into the microenvironment after death can be deleterious. In some pathologic conditions, such as leukocytoclastic vasculitis (Chapter 270), abundant neutrophil apoptosis is readily apparent on histopathologic examination. Other cells, including T lymphocytes, undergo postactivation apoptosis to prevent an overwhelming persistent host response. Defective apoptosis or even persistence of apoptotic cells that escape clearance may contribute to chronic inflammatory and autoimmune diseases. For instance, loss of tolerance to self-antigens might participate in autoimmune responses in SLE. Commitment of a cell to apoptosis can be initiated by a number of factors, including the ROIs in the cellular microenvironment as well as signaling through several death receptor pathways (e.g., FasL/Fas and TNF-related apoptosis-inducing ligand [TRAIL]). The former can damage DNA, which is a common byproduct of the genotoxic environment created by inflammation. If DNA damage is excessive, repair by tightly regulated mismatch repair mechanisms is terminated, and programmed cell death can be initiated by genes such as the p53 tumor suppressor. The burden of mutations induced by ROIs or RNIs in chronic inflammation can potentially accumulate over time and eventually lead to amino acid substitutions in key regulatory proteins. Ultimately, as has been observed in ulcerative colitis, neoplastic disease can ensue. Removal of apoptotic bodies, or the remnants of packaged apoptotic cells, is rapid and can be accomplished by macrophages, fibroblasts, epithelial and endothelial cells, muscle cells, and dendritic cells. The surface receptors used in recognition and engulfment of apoptotic cells include integrins (e.g., αvβ3), lectins, scavenger receptors, ATP-binding cassette transporter 1, LPS receptor, CD14, and complement receptors CR3 and CR4. However, some of these membrane molecules can be used in both pro-inflammatory and apoptotic pathways, the divergence of which may be based on differing ligands and accessory molecules. Apoptotic cells display a series of membraneassociated molecular patterns that interact with receptors on phagocytes. A general feature of apoptotic cells is loss of phospholipid asymmetry, with external presentation of phosphatidylserine. Externalized phosphatidylserine

Inflammation

Resolution

Apoptosis

Deletion of innate and adaptive immune response cells

Antioxidants

Cytokines

Anti-inflammatory cytokines, IL-10, TGF-β

TIMPs Collagen α2-Macroglobulin SERPINs Metalloproteinases

Fibrosis

Cytokine antagonists

Soluble receptors IL-1, TNF-α

Intracellular signals

Phosphatases Signaling inhibitors

Binding proteins IL-18

Natural antagonists IL-1Ra

Cell deactivation

Decreased adaptive immune response

FIGURE 48-1.  Anti-inflammatory mechanisms that resolve inflammation and lead to repair of the extracellular matrix. IL = interleukin; SERPINs = serine protease inhibitors; TGF = transforming growth factor; TIMPs = tissue inhibitor of metalloproteinases; TNF = tumor necrosis factor.

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

may be sufficient to trigger phagocytosis, but other apoptotic cell surface structures exist. Although some inflammatory and immune cells are being deleted, other cell lineages expand during the resolution phase. Mesenchymal cells, especially fibroblasts, proliferate and produce new matrix that can contract to form a fibrotic scar. Locally produced growth factors such as PDGF induce DNA synthesis of these stromal cells through activation of PI3Ks. TGF-β8 also stimulates fibroblast proliferation and converts cell phenotype to matrix formation rather than matrix destruction by increasing collagen production and suppressing MMP expression. In addition, mesenchymal stem cells that either reside in the tissue or migrate from the peripheral blood can differentiate into the appropriate organ-specific lineage. The pluripotential cells, in the presence of the appropriate milieu, can become adipocytes, chondrocytes, bone cells, or other terminally differentiated stromal cells.

Soluble Mediators

ANTI-INFLAMMATORY CYTOKINES

A variety of anti-inflammatory cytokines are released by resident and infiltrating cells. TGF-β and IL-10 are examples that are produced by macrophages, interstitial fibroblasts, or T cells. Some T-cell cytokines, including IL-4, IL-10, and IL-13, suppress the expression of MMP by cells stimulated by IL-1 or TNF. In addition to increasing fibroblast proliferation, TGF-β suppresses collagenase production, increases collagen deposition, and decreases MMP activity by inducing production of the tissue inhibitors of metalloproteinases (TIMPs). The repair phase is abnormal in diseases in which tissue fibrosis represents a major pathologic manifestation. For example, scleroderma (Chapter 267) is marked by diffuse fibrosis and is accompanied by high levels of TGF-β and increased production of ECM. Cytokine decoy receptors can also downregulate the inflammatory response. Receptors can also be shed from the cell surface after proteolytic cleavage and can absorb cytokines, thereby preventing them from ligating functional receptors on cell membranes. These cytokine inhibitors can be released as a coordinated attempt to prevent unregulated inflammation, as in septic shock (Chapter 108), in which endotoxin induces production of soluble receptors after initial massive production of TNF and IL-1. Other types of cytokine-binding proteins are also produced as counter-regulatory mechanisms, including IL-18-binding protein (IL-18BP), which is an Ig superfamily−related receptor that captures IL-18. In bone remodeling (Chapter 243), interactions of receptor activator of NF-κB (RANK) with RANK ligand are required for osteoclast-mediated resorption. The competitive antagonist osteoprotegerin is a member of the TNF receptor family that binds to RANK ligand and inhibits osteoclast activation. At least two distinct mechanisms contribute to natural IL-1 inhibition. An IL-1 decoy receptor (type II IL-1R) has both cell membrane and soluble forms that neutralize IL-1 activity. In addition, a natural IL-1 antagonist, IL-1Ra, can bind to functional IL-1 receptors and compete with IL-1α or IL-1β. However, IL-1Ra does not transduce a signal to the cell and blocks the biologic functions of ambient IL-1. The balance of IL-1 and IL-1Ra production depends on many influences. For instance, monocytes produce more IL-1, whereas mature macrophages produce IL-1Ra.

DEACTIVATION OF SIGNALING PATHWAYS

The signaling pathways described previously that initiate an inflammatory response have intracellular mechanisms to ensure that the process is selflimited. Many kinases, such as the MAPKs, require post-translational modification through phosphorylation to increase enzyme activity. A system of phosphatases that remove these phosphates can return the kinase to its resting form. For example, dual specificity phosphatase 1 (DUSP1) is an enzyme that dephosphorylates p38 MAPK as well as other MAPKs. DUSP1 expression is increased by p38 MAPK; thus, the very process of activating the cell through p38 is responsible for its own counter-regulatory mechanism. NF-κB activation is typically initiated by phosphorylation of the inhibitor of kB (IkB), which targets it for proteolysis. IkB expression later increases dramatically and stops the signaling through this pathway. JAK-STAT signaling is inhibited by the suppressor of cytokine stimulation (SOCS) proteins. Thus, cellular defense mechanisms have evolved to prevent persistent cell activation.

235

none prostaglandins (CyPG). The prostanoids can serve as ligands for peroxisome proliferator-activated receptors (PPARs) (Chapter 206). There are three main classes of PPAR receptors—PPARα, PPARβ/δ, and PPARγ— all of which bind to DNA as heterodimers in association with the retinoid X receptor. Activation of PPARγ by CyPG is associated with the suppression of AP-1 and STAT transcriptional pathways in macrophages. A variety of natural and synthetic PPAR agonists have demonstrated efficacy in models of ischemia-reperfusion injury, arthritis, and inflammatory airway disease.

Inhibitors of Direct Effectors

ANTIOXIDANTS

Antioxidant enzymes that can inactivate the toxic intermediates and protect normal tissues include catalase and superoxide dismutase. Catalase is a peroxisomal enzyme that catalyzes the conversion of hydrogen peroxide to water and oxygen. Superoxide dismutases (SODs) catalyze the dismutation of superoxide to hydrogen peroxide, which is then removed by catalase or glutathione peroxidase. Glutathione peroxidases and glutathione reductase are additional mechanisms for maintaining redox balance and removal of toxic metabolites. Insufficient production of intracellular antioxidants such as glutathione can suppress lymphocyte responses and could account for defective T-cell receptor signaling and blunted immunity in T cells derived from rheumatoid arthritis synovium (Chapter 264). Interactions of free radicals with surrounding molecules can generate secondary radical species in a self-propagating chain reaction. Chain-breaking antioxidants are small molecules that can receive or donate an electron and thereby form a stable byproduct with a radical. These antioxidant molecules are categorized as either aqueous phase (vitamin C, albumin, reduced glutathione) or lipid phase (vitamin E, ubiquinol-10, carotenoids, and flavonoids). In addition, transition metal-binding proteins (ceruloplasmin, ferritin, transferrin, and lactoferrin) can serve as antioxidants by sequestering cationic iron and copper and thereby inhibiting the propagation of hydroxyl radicals.

PROTEASE INHIBITORS

Protease inhibitors regulate the function of endogenous proteases and reduce the likelihood of collateral damage to tissues. These proteins form two functional classes, active site inhibitors and α2-macroglobulin (α2M). The latter class of protease inhibitors acts by covalently linking the protease to the α2M chain and thereby blocking access to substrates. α2M binds to all classes of proteases and, after forming a covalent bond, conveys them to cells through receptor-mediated endocytosis with subsequent enzymatic inactivation. The family of inhibitors of serine proteases (SERPINs) are the most abundant members of the former class of protease inhibitors and play a major role in regulation of blood clot resolution and inflammation, as indicated by many of their names: antithrombin III, plasminogen activator inhibitors 1 and 2, α2-antiplasmin, α1-antitrypsin, and kallistatin. The TIMP family blocks the function of most MMPs. The TIMPs bind to activated MMPs and irreversibly block their catalytic sites. Examples of disease states with an unfavorable balance between TIMPs and MMPs include loss of cartilage in arthritis and regulation of tumor metastasis. TIMP-MMP imbalance in destructive forms of arthritis appears be caused by the limited production capacity for protease inhibitors, which is overwhelmed by the prodigious expression of MMPs. Whereas IL-1 and TNF induce MMPs, IL-6 and TGF-β suppress production of MMPs and increase levels of TIMPs. Therefore, the cytokine profile has a profound influence on the status of remodeling. When pro-inflammatory cytokines predominate, the balance favors matrix destruction; in the presence of pro-inflammatory cytokine inhibitors and growth factors, matrix protein production increases, and MMPs are inhibited by TIMPs.

Grade A Reference A1. Langley RG, Elewski BE, Lebwohl M, et al. Secukinumab in plaque psoriasis—results of two phase 3 trials. N Engl J Med. 2014;371:326-338.

ANTI-INFLAMMATORY PROSTANOIDS AND CYCLOOXYGENASE

COX2 induced by pro-inflammatory mediators appears early and can contribute to inflammatory responses. However, COX2 expression late in the process has led to speculation that it also functions in the resolution of inflammation. This regulation might occur through formation of the cyclopente-

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 48  Mechanisms of Inflammation and Tissue Repair  

GENERAL REFERENCES 1. Bryant CE, Symmons M, Gay NJ. Toll-like receptor signalling through macromolecular protein complexes. Mol Immunol. 2015;63:162-165. 2. Pal S, Bhattacharjee A, Ali A, et al. Chronic inflammation and cancer: potential chemoprevention through nuclear factor kappa B and p53 mutual antagonism. J Inflamm (Lond). 2014;11:23. 3. Strowig T, Henao-Mejia J, Elinav E, Flavell R. Inflammasomes in health and disease. Nature. 2012;481:278-286. 4. Arend W, Firestein GS. Pre-rheumatoid arthritis: predisposition and transition to chronic synovitis. Nature Rev Rheumatol. 2012;8:573-586.

235.e1

5. Aoki T, Narumiya S. Prostaglandins and chronic inflammation. Trends Pharmacol Sci. 2012;33:304-311. 6. Yamaura K, Shigemori A, Suwa E, et al. Expression of the histamine H4 receptor in dermal and articular tissues. Life Sci. 2013;92:108-113. 7. Choi AM, Ryter SW, Levine B. Autophagy in human health and disease. N Engl J Med. 2013;368:651-662. 8. Samarakoon R, Overstreet JM, Higgins PJ. TGF-β signaling in tissue fibrosis: redox controls, target genes and therapeutic opportunities. Cell Signal. 2013;25:264-268.

236

CHAPTER 49  Transplantation Immunology  

49  TRANSPLANTATION IMMUNOLOGY MEGAN SYKES

DEFINITION

Clinical transplantation encompasses transplantation of organs and islets of Langerhans containing insulin-producing β cells, in which it is necessary to overcome the host-versus-graft (HVG) immune response to avoid rejection, as well as hematopoietic cell transplantation (HCT) (Chapter 178), in which it is necessary to contend with not only the HVG but also the graft-versushost (GVH) immune response. Because preparations of bone marrow or mobilized peripheral blood stem cells (mPBSCs) contain mature T cells, their administration to conditioned and consequently immunoincompetent recipients is associated with the risk for GVH disease. Organs transplanted include corneas, kidneys, livers, hearts, lungs, small intestines, pancreases, and composite tissue allografts such as hands and faces. The list of transplanted allogeneic cells is likely to expand in the future to include other cell types, such as hepatocytes, myoblasts, and stem cell–derived replacement cells. Transplants originating from a member of the same species are referred to as allotransplants. However, transplants from other species, termed xenografts, are believed by many to be a promising solution to the severely inadequate supply of allogeneic organs and tissues, and such grafts may be used in the future. Transplants of tissues or cells originating from the recipient, either by processing of cells from the recipient’s own organ (e.g., islets of Langerhans following pancreatectomy for chronic pancreatitis) or cell populations (e.g., CD34+ hematopoietic progenitor and stem cells collected from leukapheresis products following mobilization from the bone marrow before high-dose radiation or chemotherapy to treat cancer) are referred to as autologous. In the future, these transplants may include stem cell−derived autologous cells used for therapeutic purposes.

ANTIGENS IN TRANSPLANTATION

The major antigens recognized during graft rejection and the cell types targeting them are summarized in Table 49-1.

Major Histocompatibility Antigens

The major histocompatibility complex (MHC; human leukocyte antigens [HLAs] in the human) controls adaptive and some innate immune responses and is of central importance in many immune-mediated diseases. The MHC also presents the strongest immunologic obstacle to all types of allografts. The HLA molecules include two major isoforms, termed class I and class II, and are all encoded in the MHC complex of chromosome 6. Although all HLA molecules have a similar general structure, class I and class II molecules show different expression patterns, with class I MHC expressed on most cells of the body, whereas class II antigens are expressed mainly on antigen-

presenting cell (APC) populations, such as dendritic cells, macrophages, and B cells, as well as thymic epithelial cells involved in T-cell selection. Class II MHC can also be expressed on vascular endothelial cells and activated T cells of some species, including humans. The specialized function of both classes of MHC molecules is the presentation of peptide antigens to T-cell receptors (TCRs), allowing adaptive immune responses to occur. In general terms, the peptides presented by class I molecules are 8- to 9-amino acid peptides derived from cytosolic proteins (e.g., viral proteins) that are transported into the endoplasmic reticulum, where they are processed and loaded onto class I molecules during their synthesis. CD8 molecules interact with the α3 domain of the class I heavy chain, thereby strengthening the interaction of CD8+ T cells that recognize class I−peptide complexes. Peptides presented by class II MHC molecules, on the other hand, are mostly 10 to 20mers derived from exogenous proteins (e.g., phagocytosed bacteria) that are processed through the endosomal processing pathway, and these are recognized by TCRs of T cells whose CD4 molecules strengthen the overall T cell−APC interaction. The class I presentation pathway is of particular importance in allowing destruction of virally infected cells, consistent with the expression of class I MHC on almost all cell types in the body. However, there are exceptions to this paradigm that account for the phenomenon of cross-priming and cross-presentation, wherein peptides from exogenous proteins are presented by class I molecules, a phenomenon that may have significance for transplantation. Class II MHC presentation of exogenous antigens takes place primarily on professional APCs and B cells, consistent with the role of CD4+ T cells in initiating immune responses by activating APCs, providing direct and indirect (through activated APCs that also present peptides on class I molecules) “help” for CD8+ cytotoxic T lymphocytes (CTLs), and providing help for antibodyproducing B cells. B cells are able to focus the antigens recognized by their specific surface immunoglobulin receptors by binding and internalizing these antigens, which thereby predominate in the endosomal antigen-processing pathway and become presented by a high proportion of class II molecules on that B cell. This ability of B cells to preferentially present peptides derived from their cognate antigens to CD4 T cells recognizing those alloantigens is very important in driving alloantibody production. A number of MHC molecules have been crystallized, both alone and with TCRs that recognize them. The TCR binding structure of class I and II MHC molecules is similar overall and includes both the peptide binding cleft formed by a β-pleated sheet and two α-helices forming the sides of the cleft (E-Fig. 49-1). However, class I and II MHC molecules also have significant structural differences, as summarized in E-Table 49-1. Although class I molecules are formed by the combination of a highly variable heavy (45-kD) chain (α chain) noncovalently linked to a nonpolymorphic, smaller (12-kD) light chain (β2-microglobulin), class II molecules are heterodimers of two polymorphic chains, a 32-kD α chain and a noncovalently bound 28-kD β chain. TCRs interact physically with both the α-helices of the MHC molecules and side chains of the peptide that is bound in the groove, representing a trimolecular MHC-peptide-TCR interaction (see E-Fig. 49-1). It is the most variable (“hypervariable”) portion of the TCR, produced by V-D-J somatic rearrangements and N insertions in the TCR α and β chains, known

TABLE 49-1  LYMPHOCYTES INVOLVED IN GRAFT REJECTION CELL TYPE

ANTIGENS RECOGNIZED

FUNCTION

RELEVANCE

CD4+ T cells

Allogeneic class II MHC (+ peptide) Self class II MHC + donor peptide

Antigen-presenting cell activation Help (cytokines and costimulation) Proinflammatory cytokine production Cytotoxicity Regulatory function

Organ allografts Cellular allografts Xenografts GVHD

CD8+ T cells

Allogeneic class I MCH (+ peptide) Self class I MHC + donor peptide

Cytotoxicity Cytokine production Regulatory function

Organ allografts Cellular allografts Xenografts GVHD

NK cells

Class I MHC (activates or inhibits NK cell function) Other activating ligands

Cytotoxicity Cytokine production

? Organ allografts Cellular allografts Xenografts

B cells

Class I and class II MHC blood group antigens Xenogeneic carbohydrates

Antibody-mediated rejection (hyperacute, acute humoral, and chronic rejection)

Organ allografts Cellular allografts Xenografts

CTL = cytotoxic T lymphocyte; GVHD = graft-versus-host disease; MHC = major histocompatibility complex; NK = natural killer.

CHAPTER 49  Transplantation Immunology  

Peptidebinding cleft

α1

236.e1

α2

N N Peptide-binding cleft

N C

C β2m

α3

A

B

E-FIGURE 49-1.  Two views of an HLA class I molecule. A, Ribbon diagram showing the x-ray crystallographic structure of an HLA class I molecule (side view). The β-strand structures are indicated by thick green arrows (oriented in an amino to carboxy direction), whereas connecting loops are indicated as thin lines. The α-helices are shown flanking a peptide-binding cleft at the top (membrane distal portion) of the molecule. The base (membrane proximal portion) of the molecule is formed by the noncovalent association between the α3 domain of the class I α chain and β2-microglobulin (β2m). B, View from the top of the molecule emphasizing that the base of the peptide-binding cleft consists of β-pleated sheets flanked by α-helical structures. C = C terminal; N = N terminal. (Adapted from Bjorkman PJ, Saper MA, Samraoi B, et al. Structure of the class I histocompatibility antigen HLA-A2. Nature. 1987;329:506-512.)

E-TABLE 49-1  COMPARISON OF STRUCTURAL AND FUNCTIONAL FEATURES OF HLA CLASS I AND CLASS II ISOTYPES FEATURE

HLA CLASS I

HLA CLASS II

Chain structure of heterodimer

45-kD α chain 12-kD β2-microglobulin

34-kD α chain 28-kD β chain

Tissue distribution

All nucleated cells

Antigen-presenting cells (monocytes, B cells, dendritic cells, Langerhans cells), thymic epithelium, and some T cells; inducible on other cell types by interferon-γ

Size of bound peptides

8-9 amino acids

10-20 amino acids

Source peptides

Cytosolic

Endosomal

Functions

Presentation of antigenic peptides to CD8+ T cells; ligands for natural killer cell receptors

Presentation of antigenic peptides to CD4+ T cells

CHAPTER 49  Transplantation Immunology  

as complementarity-determining region 3 (CDR3), that recognizes specific MHC-peptide complexes. The HLA molecules are all encoded within a 3.6-million base-pair region that encodes more than 200 genes, including complement and tumor necrosis factor (TNF) genes and many others in addition to MHC that have immunologic functions. The organization of the HLA region is illustrated in E-Figure 49-2, which shows that the heavy chains for “classic” class I HLA-A, B, and C and “nonclassic” class I molecules are encoded in a region that is telomeric to the “central MHC” region that includes complement and TNF genes among others, and lies between the class II and class I HLA regions. The class II region contains two α- and β-chain genes, only one of which is functional, for each of HLA-DQ and DP. However, the DR locus contains different numbers of β chains for different HLA alleles. Some of these DR β chains are pseudogenes, but various HLA-DR alleles contain either one or two functional β-chain genes. One of the striking features of HLA molecules (and the MHC of most mammalian species) is their extensive polymorphism. There are thousands of defined HLA alleles in the class I and class II regions. Because the primary function of antigen presentation to T cells is to permit responsiveness to and clearance of pathogenic microorganisms, this polymorphism may have evolved to maintain the diversity of immune responsiveness to various pathogens within a population, avoiding annihilation of that population by a single microorganism that might not be presented well by a particular MHC. HLA alleles were originally distinguished by panels of highly sensitized human sera containing multiple alloantibodies. Although it effectively identified structurally related HLA alleles, this method failed to distinguish many allelic differences that are of functional importance for antigen binding and T-cell recognition. It was only with the development of molecular methods to distinguish alleles at the genomic level, eventually through specific genomic sequences, that the full extent of the polymorphism in this region was revealed. In association with this knowledge, it has been necessary to continually revise and refine the system of nomenclature defining these alleles. According to the most recently accepted nomenclature,1 HLA alleles are identified by the locus (e.g., HLA-A), followed by an asterisk, and then a unique number with up to four sets of digits separated by colons. The first set describes the allele group (e.g., HLA-A*02), which usually corresponds to a serologically defined antigen, and the second set indicates the specific allele (e.g., HLA-A*02:101). The third and fourth set of digits are of less practical importance because they identify silent nucleotide substitutions in different alleles and variations in the nontranslated regions of the gene, respectively. Within certain populations, however, the level of diversity within allele groups may be quite limited because of the common genetic origin of the allele. For example, for the originally serologically defined HLA-DR3 allele group, there is little diversity among Northern Europeans, such that most carry the DRB1*0301 allele. Thus, for this population, it is reasonable to refer to the serologic HLA-DR3 type as defining this allele. There are certain alleles that predominate within racial groups. For example, as few as five DRB1 alleles predominate among Northern Europeans, with each allele represented in 10 to 30% of this population. E-Table 49-2 summarizes the major DRB1 allelic groups defined initially at the serologic level and later at the level of genomic sequencing. Most organ transplantations are performed across HLA disparities, and the strong immunosuppressive regimens used in transplant recipients are designed to prevent rejection by this exceptionally strong immune response. In contrast to T-cell responses to peptide antigens derived from foreign proteins, which are recognized by a very small fraction of naïve T cells (in the range of 1 in 105), a very high proportion, estimated at 1 to 10% of the T-cell repertoire, recognizes MHC alloantigens. The strong immunogenicity of allogeneic MHC molecules relates to the manner in which T cells are selected in the thymus; developing thymocytes do not survive unless they can weakly recognize a self MHC/peptide complex on a thymic stromal cell. This process is termed positive selection. Thymocytes whose receptors have high affinity for self/MHC complexes are deleted, however, so strongly autoreactive T cells rarely make it into the peripheral T-cell pool. Allogeneic antigens are not part of this negative selection process. The net result of these two selection steps is that the human T-cell “repertoire” is strongly biased to have cross-reactivity to allogeneic MHC molecules, providing a barrier to organ and hematopoietic cell transplantation. In the case of organ transplantation, in which long-term pharmacotherapy with powerful immunosuppressive drugs is used in an effort to prevent graft rejection, this can translate into improved results with matched organs in some situations. However, for unrelated, deceased donor transplantation, the benefits of HLA matching may be counterbalanced by

237

the disadvantages associated with prolonged graft ischemia when attempts are made to transport organs to the most closely matched recipient.2 For hematopoietic cell transplantation (Chapter 178), the risks for GVH disease and marrow graft failure are so greatly amplified in the presence of extensive HLA mismatches that such transplantations have been avoided whenever possible; if a sufficiently matched, related donor cannot be found, a search is conducted through large registries containing millions of volunteer unrelated donors. Because of its extensive polymorphism, truly MHCidentical, unrelated donors can be difficult to find in the human population at large. For individuals with common HLA genotypes, the likelihood of finding a matched unrelated donor is markedly greater than that for individuals with rare genotypes. This situation relates in part to the phenomenon of linkage disequilibrium, wherein alleles at nearby loci are found together on the same chromosomal segment, or haplotype, more frequently than would be predicted by chance. The pattern of linkage disequilibrium is different in different racial groups, so the chance of finding a truly genotypically identical haplotype is greatest within the same population. For example, among whites, the DRB1*0301 allele is in linkage disequilibrium with DQB1*0201, which is located several hundred thousand base pairs away on chromosome 6; this complex, in addition to the DR4 alleles that are in linkage disequilibrium with DQB1*0302, confers the greatest genetic component of risk for the development of type 1 diabetes. Many autoimmune diseases demonstrate similarly strong HLA associations. Although there are data to indicate that HLAspecific autoantigen presentation plays a major role in determining disease susceptibility, non-HLA genes in linkage disequilibrium likely account for a significant component of these genetic risk factors. The use of alternative donors has also increased the availability of hematopoietic cell transplantation (HCT) in individuals for whom an HLAidentical related or unrelated donor cannot be identified (Chapter 178). The use of cord blood transplantation, which has reduced GVH disease−inducing activity compared with adult stem cell products, as well as advances in avoiding GVH disease in haploidentical related donor HCT, has recently increased the safety and use of HLA-mismatched HCT.3

Minor Histocompatibility Antigens

“Minor” histocompatibility antigens are peptides derived from polymorphic peptides presented by an MHC molecule. Even genotypically HLA-identical siblings have different minor histocompatibility antigens. These are sufficient to induce graft rejection if immunosuppressive pharmacotherapy is not used. In the case of HCT, significant GVH disease frequently (about 30 to 50% of the time) complicates transplantation between HLA-identical siblings, even with the use of pharmacologic immunoprophylaxis.

Other Antigens

The major blood group (ABO) antigens can be the targets of a dramatic “hyperacute” rejection process that occurs when mismatched vascularized grafts are transplanted. Recognition of blood group antigens on the endothelial surface of the graft vessels by recipient “natural” antibodies (antibodies that are present without known sensitization to the antigens) activates the complement and coagulation cascades, resulting in rapid graft thrombosis and ischemia. A similar outcome can occur after transplantation to an individual with preformed anti–donor HLA antibodies resulting from presensitization by prior transplantations, transfusions, or pregnancies. Antibodies against other polymorphic antigens, such as MHC class I−related chain A (MICA), have been associated with graft rejection. In the past, transplantation could not be successfully performed in the presence of a positive antidonor crossmatch. However, considerable success has been achieved in the transplantation of ABO-mismatched kidneys, livers, and hearts (the latter in the neonatal period only), and in transplantation of kidneys to highly presensitized patients.4,5 In the case of kidney and liver transplantation, initial removal of the antibody and sometimes depletion of B cells, as well as the infusion of intravenous immunoglobulin (IVIG), has led to these successes. ABO-mismatched neonatal heart transplantation has succeeded because the transplantations are performed before the recipient has developed high levels of anti–blood group antigen antibodies, and the B cells seem to be rendered tolerant to the donor blood group antigen by the grafting process. Recognition of blood group antigens can also be of significance in HCT, in which ABO barriers are routinely crossed in both directions. This can cause hemolysis of recipient erythrocytes if the mismatch is in the GVH direction, but this complication can be avoided by washing the cellular product before infusion. Mismatches in the HVG direction can cause more persistent problems due to ongoing destruction of donor erythropoietic cells, resulting in

CHAPTER 49  Transplantation Immunology  

HLA class II region DP

Chromosome 6p21

"Central" MHC

HLA class I region

DQ DR Telomere MICB B C E AG MICA TNF-α / β

C4A C4B DPB2 DPB1

DPA2

237.e1

DQB2

DPA1

HFE

fB C2

DQB1

DQA2 DQA1 DRB1

DRA

DRB subregion gene organization DRB1 ψDRB DR1,10 DRB1 ψDRB

DRB5

DR15,16 DRB1 ψDRB DRB3 DR3,5 DRB1 ψDRB ψDRB DRB4 DR4,7,9 DRB1 DR8 E-FIGURE 49-2.  Map of the human major histocompatibility complex (MHC) spanning approximately 3.5 million base pairs on the short arm of chromosome 6. The HLA class I and class II molecules are encoded in distinct regions of the MHC. The HLA class II region contains three subregions: DR, DQ, and DP. Each of these subregions contains a variable number of α- and β-chain genes. HLA class II loci with known functional protein products are labeled in bold. In the case of DR, different numbers of DRB genes are present in different haplotypes, some of which are nonfunctional pseudogenes (ψ). A summary of the most common of these is shown in the box. The DQ and DP subregions each contain one pair of functional α- and β-chain genes. The HLA class I region contains the three “classic” class I genes—HLA-A, HLA-B, and HLA-C—as well as other related “nonclassic” class I molecules such as MICA, MICB, HLA-E, and HLA-G. The gene for familial hemochromatosis (HFE) is found just telomeric to the HLA class I region, about 3 million base pairs distant from HLA-A. The “central” MHC also contains a number of genes related to immune function, including the complement components (C4A, C4B, C2, and factor B), as well as tumor necrosis factor (TNF)-α and -β. Not shown in the figure are more than 100 additional genes, many of which are located in the central MHC. A complete listing of MHC-encoded genes can be found in Horton R, Wilming L, Rand V, et al. Gene map of the extended human MHC. Nat Rev Genet. 2004;5:889-899.

E-TABLE 49-2  SUMMARY OF MAJOR ALLELIC GROUPS AT THE DRB1 LOCUS AND THEIR RELATIONSHIP TO COMMON DRB1 ALLELES DEFINED AT THE SEQUENCE LEVEL ALLELIC GROUPS (SEROLOGIC TYPING) Major Groups

Serologic “Splits”

DR1 DR2

EXAMPLES OF COMMON ALLELES (NORTHERN EUROPEAN WHITE INDIVIDUALS) DEFINED BY SEQUENCE* DRB1*0101, 0102, 0103

DR15 DR16

DR3

DRB1*1501, 1502 DRB1*1601 DRB1*0301

DR4

DRB1*0401, 0402, 0403, 0404, 0405, 0406, 0407, 0408

DR5

DR11 DR12

DRB1*1101, 1102, 1103, 1104 DRB1*1201

DR6

DR13 DR14

DRB1*1301, 1302, 1303 DRB1*1401

DR7

DRB1*0701

DR8

DRB1*0801, 0802, 0803, 0804, 0806

DR9

DRB1*0901

DR10

DRB1*1001

*Alleles in bold are found in at least 10% of individuals in the population. From Williams F, Meenagh A, Single R, et al. High resolution HLA-DRB1 identification of a Caucasian population. Hum Immunol. 2004;65:66-77.

238

CHAPTER 49  Transplantation Immunology  

pure red cell aplasia. More often, however, donor erythropoiesis is successfully established, and antidonor isohemagglutinins disappear from the circulation. A and B blood group antigens are the consequence of the presence or absence of specific glycosylation enzymes in different individuals. Likewise, an antigenic specificity of the utmost importance in xenotransplantation is a carbohydrate epitope, Galα1–3Galβ1–4GlcNAc (αGal), which is produced by a specific galactosyl transferase. Humans and Old World monkeys lack a functional αGal transferase and produce high levels of natural antibodies against the ubiquitous αGal epitope. Because animals of interest as xenograft sources (e.g., pigs) express αGal at high levels on their vascular endothelium, transplantation of vascularized organs from pigs results in hyperacute rejection unless something is done to absorb the antibodies or inactivate complement. The development of αGal-knockout pigs, therefore, was an important milestone, and encouraging results have been obtained in pig-to-primate transplantation in initial studies. In another type of transplant reaction, recognition as foreign results not from the presence of an antigen, but paradoxically from the absence of a self MHC molecule. Natural killer (NK) cells express a series of surface inhibitory and activating receptors that, collectively, determine whether the NK cell does or does not kill a potential target cell. The ligands for the inhibitory receptors are MHC class I molecules, and the receptors recognize specific groups of alleles. An NK cell may kill an allogeneic target that lacks a self MHC inhibitory ligand. This phenomenon has been shown in animal models to result in rapid bone marrow rejection when the donor marrow cells are not given in excess numbers or when a fraction of them are destroyed by an incompletely suppressed T-cell response. A similar phenomenon has not been clearly demonstrated in clinical HCT. The possibility that NK cells play a role in organ allograft rejection has long been an area of controversy. NK cells may be of particular importance in xenotransplantation, where they appear early in infiltrates of organ xenografts undergoing acute vascular rejection. NK cells clearly play a strong role in rejection of xenogeneic bone marrow, an observation that is relevant in one approach to inducing tolerance (see later discussion).

MECHANISMS OF REJECTION AND GRAFT-VERSUS-HOST DISEASE

Cellular Mediators

Many different cell types participate in rejection responses, and there is considerable redundancy. T cells are key players in most forms of rejection, with the exception of rejection that can be induced by antibodies in the absence of T-cell help. These include hyperacute and acute vascular rejection processes that may be induced by natural antibodies, as described earlier, or by antibodies that are present due to presensitization. The possible role of NK cells has already been discussed.

Direct and Indirect Allorecognition

T-cell responses are induced by APCs that present alloantigens. There are two forms of alloantigen recognition, termed direct and indirect (Fig. 49-1). Direct allorecognition denotes recognition of donor antigens on donor APCs provided by the graft. The extraordinarily high frequency of T cells with alloreactivity is caused by direct recognition of allogeneic MHC. Indirect recognition is the recognition of donor antigens that are picked up and presented on recipient MHC molecules on recipient APCs. The indirect response is more similar to “normal” T-cell responses, in which professional APCs present peptide antigens to T cells that are present at relatively low frequency in the naïve repertoire. In organ transplantation, direct alloreactivity is particularly important in the early post-transplantation period, when APCs within the transplanted organ are still present; many of these cells migrate to the lymphoid tissues, where they initiate the alloresponse. However, the APC supply that comes with the donor graft is not renewable, so if the direct response is not maintained by recognition of donor antigens on endothelial cells or other cells in the graft, it may recede in importance. The indirect response, on the other hand, can be maintained by the constantly renewed pool of recipient APCs. The indirect response is of particular importance in inducing antibody responses.

Effector Mechanisms of Rejection

T cells can promote graft rejection through several effector mechanisms. One is the antibody-dependent processes that have already been discussed, which can be induced by CD4+ helper T cells that promote differentiation and

Direct Allorecognition T-cell receptor

Donor Class II MHC

Recipient T cell Donor APC

CD4

Indirect Allorecognition T-cell receptor Recipient T cell

Host Class II MHC

Donor peptide

Recipient APC

CD4 FIGURE 49-1.  Direct and indirect allorecognition. Direct allorecognition involves the recognition by a T-cell receptor of major histocompatibility complex (MHC) molecules (with or without a peptide) on a donor antigen-presenting cell (APC). Indirect allorecognition involves recognition by the T-cell receptor of a donor peptide presented on a recipient APC that has picked up and processed donor antigens.

immunoglobulin (Ig) class switching of B cells that recognize other specificities on the same alloantigens. T cells provide cognate help to B cells when the TCRs recognize complexes of self MHC with donor MHC−derived peptide antigens (produced by B cells whose surface Ig receptors recognize and pick up the donor MHC antigen). If antidonor antibody is not present before transplantation but is induced afterward, the response can lead to the pathologic picture of acute humoral rejection. Antibodies may also participate in a slower, poorly understood process of chronic rejection, which, in the case of kidney and heart, is characterized by unique vascular lesions with intimal thickening and loss of the vessel space, and in the case of lung transplantation, by obliterative bronchiolitis. The mechanisms underlying these chronic rejection lesions are not well understood, and several different immune processes may in fact lead to similar lesions. Another major effector pathway leading to graft rejection involves CTLs, which are predominantly members of the CD8+ T-cell subset but also include CD4+ T cells. Several effector mechanisms lead to killing of target cells by CTLs, and these include the granzyme/perforin-mediated pathway and the pathways involving Fas/Fas ligand (FasL) and other members of the TNF receptor family and their ligands (Chapter 47). Because CD8+ cells recognize class I MHC molecules, which are widely expressed, it is not difficult to envision graft destruction by CD8+ CTLs. CD8+ CTLs may be activated through an APC that is stimulated initially through contact with an alloreactive CD4+ cell. This is one form of CD4 “help” for CD8+ cells. In addition, CD8+ cells may be dependent on cytokines such as interleukin-2 (IL-2) from CD4+ cells for their expansion and cytotoxic differentiation. However, there are also many examples of CD8+ cell–mediated rejection that is independent of “help” from CD4+ cells. Class II MHC, which is recognized by CD4+ T cells, is less widely expressed on graft tissues than is class I MHC, although it may be induced on endothelial cells and graft parenchymal cells in the presence of inflammatory cytokines such as interferon-γ (IFN-γ). In addition to cytotoxic mechanisms resulting from direct allorecognition, CD4+ and CD8+ T cells with indirect specificity seem also to be capable of causing graft destruction under some circ*mstances. Cytokines such as IFN-γ have been implicated in some instances, but in general, the pathways of indirect graft destruction are not well understood. A CD8+ cell–mediated form of skin graft rejection that is dependent on donor antigens crosspresented on recipient MHC molecules (a form of indirect allorecognition for CD8+ cells) has been described in an animal model. This form of graft rejection may be directed at antigen presented on endothelial cells of recipient vessels that revascularize the graft. This mechanism would not apply to primarily vascularized organ allografts.

The Role of T-Cell Trafficking

All the rejection processes described require trafficking of T cells into the graft. This process is made possible after the initial activation of naïve T cells

CHAPTER 49  Transplantation Immunology  

in the lymphoid tissues. Naïve T cells can migrate into lymph nodes because of their expression of the CCR7 chemokine receptor and the adhesion molecule L-selectin. These T cells are activated by migratory graft APCs that also enter the lymph nodes. T-cell activation is associated with loss of CCR7 and L-selectin expression and acquisition of a new set of chemokine receptors and adhesion molecules that allow rolling and adhesion on the graft endothelium and entry into the graft parenchyma (Chapter 47). Inflammation in the graft, such as that induced by ischemia-reperfusion injury and the transplantation procedure, as well as that induced by initially responding T cells, is associated with upregulation of chemokines and adhesion ligands that promote entry of lymphocytes into the graft. Nevertheless, well healed-in grafts can be slowly rejected by adoptively transferred memory T cells, demonstrating that acute graft injury and inflammation are not essential for rejection in the presence of an established memory T-cell response. Rejection of hematopoietic cell grafts may involve many of the same mechanisms as those discussed for solid organs, although less detailed work has been done in this area.

Mechanisms of Graft-versus-Host Disease

Initiation of GVH disease (Chapter 178) requires that donor T cells recognize host alloantigens. The disease involves attacks on a variety of recipient epithelial tissues, namely skin, the intestine, and liver. Animal models have demonstrated clear roles for both CD4+ and CD8+ cells in initiating GVH disease, and each subset is able to do so independently of the other. The mechanisms of GVH disease include activation of alloreactive donor T cells by recipient APCs, leading to the differentiation of effector cells with direct cytotoxic activity and cytokine production in response to host antigens. A prominent role is played by TNF-α, whose production is induced in part by the translocation of bacteria across the intestinal wall, promoting innate immune system activation through toll-like receptors (Chapter 45). An intensely pro-inflammatory environment is produced by the combination of conditioning-induced tissue injury and disruption of mucosal barriers, bacterial activation of the innate immune system, and the GVH alloresponse. An important role is now appreciated for the inflamed microenvironment in target tissues in promoting the trafficking of GVH-reactive T cells into these tissues.6

STRATEGIES TO PREVENT GRAFT-VERSUS-HOST DISEASE

In view of the critical role of donor T cells in inducing GVH disease, an obvious strategy for preventing this complication is to remove mature T cells from the marrow graft. This approach has indeed been shown in both animal models and clinical studies to prevent GVH disease effectively. However, there are several disadvantages to this approach. One is that adult humans, particularly those who have undergone prior chemotherapy and radiotherapy, have little remaining thymic tissue and therefore demonstrate sluggish T-cell recovery, leading to serious opportunistic infections. The second disadvantage applies to the most common indication for allogeneic HCT, namely the treatment of hematologic malignancies (Chapter 178). In this setting, T-cell depletion is often associated with an increased relapse rate due to loss of a graft-versus-tumor (GVT) effect, which is in large part mediated by GVH alloreactivity. Separation of GVH disease from GVT effects is a major goal of research in HCT, and some promising strategies are being explored (E-Table 49-3). These include control of T-cell trafficking so that the GVH alloresponse is confined to the lymphohematopoietic tissues where the tumor resides and a number of other approaches.6,7 The third disadvantage of donor T-cell depletion in HCT is that it increases the rate of engraftment failure. GVH alloreactivity and a “veto” effect of donor T cells help to overcome host resistance to donor engraftment. A veto cell, which may be a T cell or an NK cell, kills a CTL that attacks it. Although the phenomenon has been well established in animal models, its mechanisms are not clearly established, and its potential role in humans is uncertain. NK-cell recognition in the GVH direction resulting from the absence in the recipient of a class I MHC ligand (E-Fig. 49-3) that can trigger a donor NK-cell inhibitory receptor (KIR) may promote donor marrow engraftment and antitumor effects against acute myeloid leukemias in the setting of T-celldepleted, HLA-mismatched HCT. Clinically, pharmacologic immunosuppressive prophylaxis is usually used in at least the first 6 months after HCT to minimize the complication of GVH disease. Additionally, HLA-matched or closely matched donors are chosen whenever possible because GVH disease increases in frequency and severity as increased HLA barriers are transgressed. These measures, nevertheless, are

239

insufficient, and GVH disease remains a major complication of HCT. Therefore, many of the new strategies being explored in organ transplantation and other fields are also being examined for the prevention of GVH disease in experimental models. It should be borne in mind, however, that tolerance of donor T cells to recipient alloantigens (see later discussion) might not be entirely beneficial in the HCT setting for the treatment of malignant disease because loss of GVH alloreactivity is likely to come with loss of antitumor effects.

STRATEGIES TO PREVENT ALLOGRAFT REJECTION

Nonspecific Immunosuppression

Immunosuppressive drugs are the mainstay of clinical organ transplantation, and improvements in these drugs following the discovery of cyclosporine have extended organ transplantation to include hearts, lungs, pancreases, livers, and other organs and tissues in the past 30 years. The mechanisms of action of these agents are discussed in Chapter 35. However, it is noteworthy that, despite these improvements and their enormous impact on early graft survival, these agents have been less effective in attenuating late graft loss. Because chronic immunologic rejection processes and side effects of the immunosuppressive drugs themselves are responsible for much of this late graft loss, improved immunosuppressive agents and induction of immune tolerance (see later discussion) are major research goals in transplantation.

Costimulatory Blockade

As understanding of immune responses has increased, recent years have seen the exploration of numerous biologic agents, including antibodies and small molecules targeting receptors of the immune system as well as cell-based therapies, in efforts to improve allograft survival. Because of the central role played by T cells in the immune response, considerable attention has been focused on blockers of T-cell costimulation. When a naïve T cell recognizes antigen through its unique TCR, additional “costimulatory” signals are required to allow full activation, expansion, and differentiation to occur. These signals are often provided by APCs in the form of ligands (e.g., B7-1, B7-2) for costimulatory receptors (e.g., CD28) on the T cell. Cross-talk between the T cell and the APC (e.g., due to CD40 activation by CD154 upregulation on the activated T cell) further amplifies the costimulatory activity of the APC, allowing it to effectively activate other T cells as well. The CD154 (T cell)–CD40 (B cell) interaction also promotes Ig class switching and functioning of B cells as APCs. Blockade of these processes (e.g., by CTLA4Ig and anti-CD154 monoclonal antibodies [mAbs]) has led to marked prolongation of allograft survival in stringent rodent and largeanimal models. Robust, systemic tolerance to donor antigens has been achieved in rodents receiving bone marrow transplantation with costimulatory blockade and little or no additional conditioning. Some of these agents have joined the armamentarium of immunosuppressive agents in clinical trials in transplantation and autoimmune diseases.8 Although anti-CD154 antibodies have been associated with thromboembolic complications, precluding further evaluation in transplantation trials, recently developed anti-CD40 antibodies have shown promise in animal studies. Numerous additional costimulatory and inhibitory pathways that affect T-cell responses have been described, and these all are potential targets for further manipulation of the alloresponse.

Immune Tolerance

Immune tolerance denotes a state in which the immune system is specifically unreactive to the donor graft (or recipient in the case of GVH reactivity) while remaining normally responsive to other antigens.9-11 Tolerance is distinct from the state produced by nonspecific immunosuppressive agents, which increase risks for infection and malignancy. Numerous approaches to tolerance induction have been described in rodent models, largely owing to the strong tolerogenicity of primarily vascularized heart, liver, and kidney grafts in these animals. Because such grafts are less tolerogenic in humans, none of these strategies has been effectively applied clinically to date. Therefore, tolerance strategies that are appropriate for clinical evaluation must first be tested in “stringent” models, including relatively nontolerogenic grafts such as MHC-mismatched skin in rodents and vascularized organ graft models in large animals. In most of the models, only a superficial understanding of the mechanisms leading to tolerance is currently available. The three major mechanisms of T-cell tolerance are deletion, anergy, and suppression (often referred to as “regulation”). Deletion denotes the

CHAPTER 49  Transplantation Immunology  

239.e1

E-TABLE 49-3  EXPERIMENTAL STRATEGIES TO PREVENT GRAFT-VERSUS-HOST DISEASE STRATEGY

ADVANTAGES

LIMITATIONS

Donor T-cell TH2 polarization (e.g., conditioning with ATG and TLI; in vitro stimulation with cytokine exposure)

May preserve GVL

May limit GVL; TH2 can contribute to acute and chronic GVHD

Tolerance induction of donor T cells (e.g., costimulatory blockade; regulatory cells)

Some strategies may selectively tolerize GVH-reactive T cells (e.g., in vitro antigen exposure with costimulatory blockade)

Global immunosuppression may limit GVL and antiinfectious immunity; tolerance (i.e., GVH protection) may be incomplete

Donor T-cell depletion plus NK-cell infusion with class I mismatched transplantation

NK cells do not cause GVHD but may mediate antitumor effects; donor NK cells may eliminate host APCs that trigger GVHD

Antitumor effect against only certain types of malignancies; requires appropriate MHC disparity and expression of polymorphic NK-cell receptors; insufficient T-cell immunity to infection

Donor T-cell depletion followed by delayed donor lymphocyte infusion (DLI)

Preserves high level of GVL due to GVH reactivity. GVHD does not occur if host inflammation from conditioning has subsided and initial HCT was devoid of donor T cells

Antitumor effect delayed until time of DLI; most applicable for indolent lymphohematopoietic tumors. GVHD more difficult to control in humans than animal models, probably owing to occult or overt infection resulting from T-cell deficiency before DLI

Depletion of donor T cells recognizing host alloantigens by in vitro or in vivo activation/ depletion (i.e., “allodepletion”)

Preserves anti-infectious immunity and tumor antigen−specific responses while limiting GVHD

Loss of GVH reactivity will limit GVL and engraftment; highly efficient allodepletion methods not yet available. Residual T cells may cause GVHD

Donor T-cell depletion with infusion of expanded infection-specific T cells (e.g., CMV or EBV specific)

Reduces GVHD potential while protecting against significant infectious organisms

Lack of GVL effect; lack of broad anti-infectious immunity; expense and inefficiency of in vitro T-cell expansion; loss of survival/homing potential of cultured T cells

Donor T-cell depletion with infusion of expanded tumor antigen-specific T cells (expanded from natural repertoire or transduced with a T-cell receptor or chimeric antigen receptor)

GVL without GVHD

Lack of anti-infectious immunity; expense and inefficiency of in vitro expansion of tumor-specific T cells; loss of survival/homing potential of cultured T cells

Insertion of suicide gene (e.g., thymidine kinase) into donor T cells

Drug targeting inserted gene (e.g., ganciclovir) kills donor T cells to treat GVHD after GVL initiated.

Expense and inefficiency of in vitro transduction of T cells; loss of function/survival/homing potential of cultured T cells; risk for GVHD if transduction incomplete; curtailment of GVL when donor T cells killed in vivo

Block T-cell trafficking to epithelial GVHD target tissues (e.g., blockade of adhesion molecules or chemokines, sphingosine 1 phosphate agonists)

Permits lymphohematopoietic GVH reactions to occur, with associated GVL effects

Redundancy of trafficking pathways in inflammatory environment may limit efficacy; tumors outside of lymphohematopoietic system not targeted

Block injury/promote repair in epithelial target tissues (e.g., keratinocyte growth factor)

Permits lymphohematopoietic GVH reactions to occur, with associated GVL effects

Efficacy may be limited

APC = antigen-presenting cell; ATG = antithymocyte globulin; CMV = cytomegalovirus; DLI = donor lymphocyte infusion; EBV = Epstein-Barr virus; GVHD = graft-versus-host disease; GVL = graft-versus-leukemia effects; HCT = hematopoietic cell transplantation; MHC = major histocompatibility complex; NK = natural killer; TH2 = helper T lymphocytes type 2; TLI = total lymphoid irradiation.

Receptor 1 A

HLA group 1

C

Receptor 2 B Receptor 3

D

HLA group 2 HLA group 3 HLA group 4

Receptor 4 Autologous cell A C B

HLA group 1 HLA group 2

D Receptor 4

No inhibitory HLA ligand—cytotoxicity

Allogeneic cell E-FIGURE 49-3.  Killing of allogeneic targets by natural killer (NK) cells due to “missing self.” NK cells express clonally distributed inhibitory receptors (KIRs) with specificity for different groups of major histocompatibility complex (MHC) class I alleles, referred to in the figure as human leukocyte antigen (HLA) groups 1, 2, 3, and 4. Four different NK cells (A, B, C, and D) are shown, each with a different set of KIRs (referred to as receptors 1, 2, 3, and 4). Examples of HLA allele groups in the human are the HLA-Cw4, HLA-Cw3, and HLA-Bw4 groups; examples of KIRs are the ligands for these allele groups—namely, KIR2DL1, KIR2DL2/3, and KIR3DL1, respectively. Each functional NK cell has one or more inhibitory receptors that recognize a “self” (autologous) HLA molecule. Although some of the NK cells (e.g., cells A and B in the figure) will also find an HLA ligand to which their receptors bind on allogeneic cells, others (e.g., cells C and D) will not. The latter cells therefore will not receive inhibitory signals from the allogeneic cells and will kill them due to recognition by other (activating) receptors.

destruction of T cells with receptors that recognize donor antigens; it can be achieved during T-cell development in the thymus, for example, by induction of mixed chimerism in T-cell-depleted hosts. Deletion can also be applied to mature T cells in the periphery, for example, by transplantation of a tolerogenic organ or marrow graft in combination with blockade of costimulatory molecules. Anergy denotes the inability of T cells to respond fully to antigens they recognize, and it can be induced by antigen presentation without costimulation. Suppression has attracted considerable interest since the discovery that constitutively CD25+ T cells of the CD4+ subset have suppressive activity that is dependent on expression of the transcription factor Forkhead Box Protein 3 (FoxP3). These and other types of suppressive T cells (e.g., NKT cells, regulatory CD8+ cells and B cells, myeloid-derived suppressor cells) have been implicated in rodent transplantation tolerance models and in prevention of autoimmunity. The use of expanded regulatory cells has recently entered clinical trials, and both the ultimate practicality of the approach and the relative advantages of antigen-specific versus nonspecific regulatory cell therapy remain to be determined. There is also interest in strategies for activating or expanding regulatory T cells in vivo, thereby favoring the suppressive immune response over destructive alloimmunity.12 The developments in animal models and understanding of immune mechanisms described here have provided impetus for efforts to achieve immune tolerance in clinical transplantation. Every transplantation center has anecdotal cases of patients who have removed themselves from chronic immunosuppression without experiencing graft rejection. However, for every such patient, there are dozens more who have experienced rejection on dose reduction or removal of immunosuppressive drugs. Although trials of minimization and slow withdrawal of nonspecific immunosuppressive therapy are underway in organ transplant recipients, a major current limitation is the absence of good predictors of success. It remains to be seen whether recently identified molecular “tolerance signatures” will provide markers with sufficient predictive value to allow such withdrawal to be safely undertaken. One approach developed in animal models has been successfully applied to the induction of immune tolerance in a small group of patients receiving renal allografts. This approach, involving bone marrow transplantation after nonmyeloablative conditioning, which is much less toxic than standard HCT conditioning, was shown to be effective in the most stringent rodent and large-animal models before being evaluated clinically. Initial success using combined kidney and bone marrow transplantation in patients with renal failure due to multiple myeloma led to pilot studies in patients with renal failure without malignant disease, with encouraging preliminary results. This approach and others that have emerged from ongoing investigations provide hope that, in the future, transplantation might be routinely performed without the need for chronic immunosuppressive therapy, with its attendant

complications and limited ability to control chronic rejection.11 Because autoimmune diseases are major contributors to end-stage renal disease, diabetes, and other types of organ failure, the potential for tolerance strategies to reverse autoimmunity while inducing allograft tolerance is also a source of hope. All these approaches must, however, be undertaken with the caution that successful regimens could also lead to immune tolerance to active infectious organisms. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 49  Transplantation Immunology  

GENERAL REFERENCES 1. Marsh SG, Albert ED, Bodmer WF, et al. Nomenclature for factors of the HLA system, 2010. Tissue Antigens. 2010;75:291-455. 2. Susal C, Opelz G. Current role of human leukocyte antigen matching in kidney transplantation. Curr Opin Organ Transplant. 2013;18:438-444. 3. Brunstein CG, Fuchs EJ, Carter SL, et al. Alternative donor transplantation after reduced intensity conditioning: results of parallel phase 2 trials using partially HLA-mismatched related bone marrow or unrelated double umbilical cord blood grafts. Blood. 2011;118:282-288. 4. Montgomery JR, Berger JC, Warren DS, et al. Outcomes of ABO-incompatible kidney transplantation in the United States. Transplantation. 2012;93:603-609. 5. Montgomery RA, Lonze BE, King KE, et al. Desensitization in HLA-incompatible kidney recipients and survival. N Engl J Med. 2011;365:318-326.

240.e1

6. Li HW, Sykes M. Emerging concepts in haematopoietic cell transplantation. Nat Rev Immunol. 2012;12:403-416. 7. Blazar BR, Murphy WJ, Abedi M. Advances in graft-versus-host disease biology and therapy. Nat Rev Immunol. 2012;12:443-458. 8. Pilat N, Schwarz C, Wekerle T. Modulating T-cell costimulation as new immunosuppressive concept in organ transplantation. Curr Opin Organ Transplant. 2012;17:368-375. 9. Ferrer IR, Hester J, Bushell Wood KJ. Induction of transplantation tolerance through regulatory cells: from mice to men. Immunol Rev. 2014;258:102-116. 10. Fuchs EJ. Transplantation tolerance: from theory to clinic. Immunol Rev. 2014;258:64-79. 11. Zachary AA, Leffell MS. Desensitization for solid organ and hematopoietic stem cell transplantation. Immunol Rev. 2014;258:183-207. 12. Issa F, Robb RJ, Wood KJ. The where and when of T cell regulation in transplantation. Trends Immunol. 2013;34:107-113.

240

CHAPTER 50  Complement System in Disease  

50  COMPLEMENT SYSTEM IN DISEASE JOHN P. ATKINSON The complement system consists of plasma and membrane proteins that participate in host defense against infections and in clearance of cellular and extracellular debris, as well as in a wide variety of autoimmune and inflammatory states (Fig. 50-1).1,2 Complement is essential in innate immunity and a potent effector arm of adaptive (humoral) immunity. It is a first responder, especially in blood, to bacterial and viral invasion (Table 50-1). It helps to maintain sterility (“guardian of the intravascular space”) by depositing within seconds its opsonic and membrane-perturbing fragments on a pathogen’s surface. A second major activity of complement is to promote the inflammatory response via the release of soluble fragments (anaphylatoxins). They bind to their receptors, leading to cellular activation, including chemokinesis and chemotaxis by phagocytic cells, and thereby enhance protection against infections. Furthermore, the deposition of complement fragments on immune complexes keeps them from precipitating and promotes their adherence to red blood cells (RBCs) for a hand-off to monocytes and dendritic cells in the liver and spleen. Through these interactions, complement also instructs the adaptive immune response. Antigens decorated by complement proteins are taken up Complement

Membrane modification

Inflammation (anaphylatoxins)

Pathogen

Lysis

Opsonization

Degranulation of mast cell

Induction of neutrophil chemotaxis

C C C

C

Nature’s adjuvant FIGURE 50-1.  Function of the complement system. The most important function of the complement system is to alter the membrane of the pathogen by coating its surface with clusters of activation fragments. In one case, they facilitate the key process of opsonization in which C4b and C3b interact with complement receptors. In the other case, as with certain gram-negative bacteria and viruses, the membrane attack complex lyses the organism. The second critical function of complement is to activate cells and thus promote inflammatory and immune responses. The complement fragments C3a and C5a (known as anaphylatoxins) stimulate many cell types such as mast cells to release their contents and stimulate phagocytic cells to migrate to sites of inflammation (chemotaxis). Through these phenomena of opsonization and cell activation, complement serves as nature’s adjuvant to prepare, facilitate, and instruct the host’s adaptive immune response. Because complement activation occurs in a few seconds, this innate immune system initially engages most pathogens, especially those that try to enter the vascular space. As will be illustrated, these basic functions are also required to handle immune complexes and prevent autoimmunity. (Modified from Arthritis Foundation. Primer on the Rheumatic Diseases. 12th ed. Arthritis Foundation; Atlanta, Ga 2001.)

CHAPTER 50  Complement System in Disease  

241

TABLE 50-1  COMPLEMENT SYSTEM IN HOST DEFENSE AGAINST BACTERIA AND VIRUSES

TABLE 50-2  SALIENT FEATURES OF THE COMPLEMENT SYSTEM

THE ACTIVITY

Ancient innate system of immunity predominantly found in blood (the “guardian of the intravascular space”) Capable of rapidly opsonizing and lysing bacteria and viruses (millions of active fragments can be deposited on a target) Works in seconds! Most proteins are synthesized by the liver Constantly turning over (AP protein C3 “ticks over” at a rate of 1% to 2% per hr) The AP also features a feedback or amplification loop, which requires tight control Effector arm of the humoral immune system (IgM and IgG) Critical for clearance of self-debris (garbage removal) After immunoglobulins and albumin, complement proteins are among the most abundant in blood Nature’s adjuvant (almost all foreign antigens are coated with complement fragments); instructs the adaptive immune response A deficiency of an activator leads to bacterial infections or autoimmunity (SLE) A deficiency of a regulator leads to undesirable cellular and tissue damage at sites of injury or degeneration (excessive activation)

THE PLAYERS

Opsonization

(C3b > C4b, C1q, MBL)*

Membrane perturbation including lysis (the membrane attack complex)

(C5b-C9)

Proinflammatory via cellular activation (the anaphylatoxins and their receptors)

(C3a, C5a)

*C3b is the major opsonin of the complement system. C1q and MBL (mannose- or mannanbinding lectin) both participate in classical and lectin pathway activation, respectively, but also bind to their specific receptors upon attachment to a target.

by monocytes, follicular-dendritic cells, B lymphocytes, and other antigenpresenting cells, resulting in an adaptive immune response. (The complement system is often called “nature’s adjuvant.”) Thus, complement activation is required for an optimal antibody response to most foreign antigens. Individuals lacking a functional complement system are predisposed to bacterial infections, predominantly by encapsulated organisms, including streptococcus, staphylococcus, Haemophilus spp., and Neisseria spp.3 Surprisingly, a complete deficiency in an early component of the classical complement pathway predisposes to autoimmune diseases, particularly systemic lupus erythematosus (SLE).4 This association suggests that complement is required not only for host defense against foreign agents but also to identify and safely clear self-materials (debris removal), particularly RNA and DNA species. A remarkable feature of the complement system is that it reacts within seconds (Table 50-2). In less than 2 minutes, it can coat an encapsulated gram-positive bacterium with several million C3b opsonic fragments and lyse gram-negative bacteria by insertion of its terminal components (the membrane attack complex [MAC]). It works even more efficiently if driven by IgM or IgG binding to an antigen on a microbial membrane to activate the cascade. Antibodies and lectins direct the activation process to the pathogen’s surface. Overall, the complement cascade is designed to become engaged on the surface of a pathogen, particularly bacteria. Plasma and membrane regulators of complement activation inhibit formation on normal “self ” cells. Much of the complement-mediated pathology revolves around the alternative pathway’s (AP’s) amplification loop. This feedback amplification loop is key in triggering activation early in an immune response; however, it must be rigorously regulated to prevent activation on normal self and excessive activation on injured self.5 Approximately half of the proteins associated with the complement system are dedicated to the control of its activation and effector functions, especially to maintain homeostasis of the AP’s amplification loop. In clinical medicine (Table 50-3), the complement system participates in three pathologic processes (Table 50-4): (1) an inherited decrease in functional activity leading to increased susceptibility to bacterial infections and to autoimmunity, (2) mediating undesirable tissue damage upon activation by autoantibodies and immune complexes, and (3) excessive activation at sites of tissue injury in individuals carrying genetic variants in regulators. Knowledge of how complement is activated and how it can be controlled points to opportunities for the development of therapeutic agents such as anti-C5 monoclonal antibody (mAb) therapy, which has been recently approved to treat several complement-dependent hemolytic disorders.

ACTIVATION OF COMPLEMENT

Classical Pathway

The binding of IgM or IgG to a target antigen activates this exceptionally powerful and quick acting pathway to destroy microbes (Figs. 50-2 and 50-3). The classical pathway (CP) reaction cascade is designed to opsonize and perturb the surface membrane of microorganisms. Of course, autoantibodies also trigger this highly efficient CP. Complement action mediated by immune complexes may then lead to cellular and tissue damage. Instructive examples of autoantibodies and complement-mediated diseases are immune hemolytic anemias, myasthenia gravis, and bullous pemphigoid. The basic problem or pathologic defect in this type of human disease is, of course, the formation of the autoantibody. A misidentification of self that has occurred because of a breaking of tolerance. In this pathologic situation, the complement system is working at the behest of the autoantibody.

AP = alternative pathway; SLE = systemic lupus erythematosus.

TABLE 50-3  PARTICIPATION OF THE COMPLEMENT SYSTEM IN HUMAN DISEASE Activation by autoantibody (formation of immune complexes) Engagement with modified self (clearance of debris or garbage) • Degenerative processes (diseases of aging such as age-related macular degeneration) • Cell and tissue damage (ischemia-reperfusion injury; atypical hemolytic uremic syndrome)

TABLE 50-4  PATHOLOGIC CONDITIONS ASSOCIATED WITH COMPLEMENT ACTIVATION Examples of diseases in which complement activation contributes to the immunopathology:   Atypical hemolytic uremic syndrome*†   Paroxysmal nocturnal hemoglobinuria†   Age-related macular degeneration*†   Membranoproliferative glomerulonephritis (types 1, 2 and 3)†‡   Myasthenia gravis‡   Bullous pemphigoid‡   Systemic lupus erythematosus/antiphospholipid syndrome‡   Rheumatoid arthritis‡   Immune hemolytic anemias‡   Immune vasculitis (the ANCA-positive syndromes)‡   Ischemia reperfusion injury*†   Allotransplantation‡   Serum sickness‡   Exposures to foreign materials (e.g., membranes, nanoparticles)* *Injury, ischemia, trauma, degeneration, or foreign body is the trigger (innate immune activation). † Lack of adequate regulation contributes to disease pathogenesis. ‡ Antibody dependent activation of the complement system (adaptive humoral immune activation).

The CP is also activated by means other than the formation of IgM- and IgG-bearing immune complexes. β-Amyloid in the neuritic plaques of patients with Alzheimer disease directly engages the CP via an interaction with C1q. Likewise, C-reactive protein (CRP) and serum amyloid protein (SAP) bind to chromatin and other ribonucleoprotein complexes released from apoptotic cells, and these types of complexes activate the CP. As noted, the CP plays a key role in the opsonization and removal of nuclear debris. Approximately 80% of patients with hereditary absence of C1q or C4 develop SLE. Deposits of CRP and activated C1 have been demonstrated in ischemic tissue such as infarcted human myocardium. These observations indicate that CP activation via these antibody-independent means is critical in protecting against autoimmune responses by facilitating debris clearance. Regulation of the CP activation occurs at two levels. First, the serine protease inhibitor (serpin) known as the C1-inhibitor (C1-INH) blocks the activity of many proteases, including factor XIIa, kallikrein, and factor XIa of

242

CHAPTER 50  Complement System in Disease  

Complement Activation CLASSICAL PATHWAY

LECTIN PATHWAY

ALTERNATIVE PATHWAY

Antibody binds to specific antigen on pathogen surface

Mannose-binding lectin binds to pathogen surface

Pathogen surface creates local environment conducive to complement activation

FIGURE 50-2. The three pathways of complement activation.

TABLE 50-5  TISSUE INJURY OR DEGENERATION AND COMPLEMENT ACTIVATION* Age-related macular degeneration Osteoarthritis (degenerative joint disease) Ischemic stroke Myocardial infarction Traumatic brain injury (e.g., liver, kidney, gut) Ischemia-reperfusion injury Burns Acute respiratory distress syndrome Septic shock Multiorgan failure syndromes Alzheimer disease

Alternative pathway Continuous and spontaneous tickover of C3 in blood and on cells C3a C3 (H2O) or C3b + C3b + FB C3bB + FD Ba Feedback loop

*In these conditions, complement activation leads to deposition of fragments at the site of injury; however, how much of the tissue injury is attributable to complement system is unknown. In many cases, animal models support a pathologic role for the complement system. Only in age-related macular degeneration do we also have powerful genetic evidence in humans to indicate a key role for the complement system.

the clotting system as well as C1r, C1s, and MASP2 of the complement system. The importance of C1-INH is exemplified by its role in hereditary angioedema (Table 50-6). In this dominantly inherited disease, a deficiency of C1-INH allows uncontrolled proteolysis of C4 and C2 and generation of bradykinin, leading to recurrent swelling episodes. This serpin prevents chronic activation of the CP cascade and, after a few minutes, helps to shut down the system. CP activation is also regulated by multiple inhibitors at the key step of C3 activation. These plasma and membrane proteins inhibit C3 convertase formation on healthy self. Membrane regulators are highly expressed on most cell types, where they prevent activation on normal self and overexuberant activation on altered and nonself.

Lectin Pathway

The protein mannose-binding lectin (MBL) is a member of the collectin family that also includes pulmonary surfactants A and D.6 MBL has a structure similar to C1q in that it consists of several subunits; namely, a globular recognition head domain for carbohydrates and a collagen-like tail that interacts with serine proteases. In the case of MBL, the globular domain is a lectin (protein) that binds to repeating mannose and N-acetylglucosamine residues on the surface of pathogens (see Figs. 50-2 and 50-3). Many microorganisms are recognized by MBL, including gram-positive and gram-negative bacteria, mycobacteria, fungi, parasites, and viruses (including human immunodeficiency virus 1 [HIV-1]). In general, as would be expected, mammalian glycoproteins and glycolipids are not readily recognized by MBL and the related lectins (ficolins and collectins) that activate the lectin pathway. Three serine proteases, MASP-1, MASP-2, and MASP-3, associate with MBL (and the ficolins and collectins) through their collagen-like domain. This is analogous to the association of C1r and C1s with C1q. Activation of MASP-2, with some help from MASP-1, results in cleavage of C2 and C4, leading to formation of the classical/lectin pathway C3 convertase (C4b2a). Genetic variations in the structural and regulatory portions of the MBL gene lead to wide differences in serum levels. A low level of MBL is associated with recurrent infections in children and adults and is a risk factor for the development of SLE. More striking is the association of low levels of MBL with infections in the setting of the treatment of SLE. For example, heterozygous MBL deficiency has been associated with a fourfold increase in the risk of bacterial pneumonia and hom*ozygous deficiency with a more than 100-fold increase.

C3bBb + P (properdin) C3bBbP +

Classical pathway

C3b on targets Lectin pathway

Enzymes Proconvertase

C3 convertase

Stabilized C3 convertase

C3

C3b

(C3b)2 BbP +

C5 convertase

C5 C5a C5b + C6, C7, C8, C9 C5b-C9 (MAC)

FIGURE 50-3.  Complement activation pathways. In the reaction cascade shown, C3b or C3 (H2O) binds the proenzyme factor B (FB), and the C36B complex then cleaved by the protease factor D (FD). The addition of properdin (P) to the enzyme complex increases the half-life of the enzyme complex approximately 10-fold. Although the source of the C3b can be from spontaneous turnover or via lectin pathway (LP) and classical pathway (CP) activation, the alternate pathway (AP) feedback loop commonly takes over to generate most of the C3b that binds to a target. The alternate pathway is continuously turning over. If activated C3b or C3 (H2O) remains in the fluid phase, it is rapidly inhibited by the plasma regulator factor H. if activated C3 binds to normal or healthy self, it is prevented from forming a convertase by the ubiquitously expresses membrane cofactor protein (MCP [CD46]) and decay-accelerating factor (DAF [CD55]). DAF “kicks out” the catalytic Bb domain (a temporary stop), but MCP is a permanent stop because, upon its binding, the C3b is proteolytically cleaved to inactive C3b (iC3b) by a serine protease known as factor I. The feedback loop is a powerful amplification system. A single Escherichia coli organism in blood can be coated with several million C3bs in a couple of minutes!

Alternative Pathway

The AP takes advantage of the fact that C3 undergoes spontaneous, chronic, low-grade activation (Figs. 50-2 to 50-4). This C3b may covalently attach to any cell; however, on normal self, amplification of the cascade is blocked by inhibitors. In contrast, deposition on polysaccharides of bacterial membranes and to other targets, such as endotoxin and virally infected cells, leads to a rapid engagement of this pathway. These sites, similar to immune complexes and almost any type of biomaterial (cardiopulmonary bypass and hemodialysis membranes, nanoparticles, and so on), lack regulators, so rapid, massive activation may occur. During spontaneous activation, called tickover, small amounts of activated C3 are continuously generated (C3 turns over in blood at 1% to 2%/hr). It can initiate a feedback loop and cleave more C3 to C3b. Also, the initial C3b

243

CHAPTER 50  Complement System in Disease  

TABLE 50-6  SOLUBLE AND MEMBRANE FACTORS REGULATING COMPLEMENT SOLUBLE FACTORS REGULATING COMPLEMENT NAME

LIGAND OR BINDING FACTOR

C1-INH

C1r, C1s, MASP-2

Binds to and displaces C1r and C1s from C1q and MASP-2 from MBL

FUNCTIONAL ACTIVITY HAE

PATHOLOGY, IF DEFICIENT

C4bp* (C4 binding protein)

C4b, GAGs

Displaces C2a (DAA); cofactor for C4b cleavage by factor I (CA)

No clinical syndrome clearly defined

CPN-1 (carboxypeptidase-N)

C3a, C5a

Inactivates C3a and C5a

Urticaria and angioedema

Factor H*†

C3b, C3d, GAGs

Displaces Bb from AP C3 and C5 convertases (DAA) and is a cofactor for factor I to cleave C3b (CA)

AMD, aHUS, C3 glomerulopathies; bacterial infections secondary to low C3

Factor I†

C3b, C4b

Serine protease; cleaves C3b and C4b, requires a cofactor protein (CA)

AMD, aHUS; bacterial infections secondary to low C3

Protein S (vitronectin)

C5b67

Inhibits membrane attachment by C5b67

None defined

NAME

LIGAND OR BINDING FACTOR

DAF (CD55)

C3 and C5 convertases

Displaces Bb from AP convertase and C2a from CP or LP convertases, respectively

PNH

Membrane cofactor protein (MCP, CD46)

C3b, C4b

Cofactor for factor I (CA)

aHUS

Protectin (CD59)

C8, C9

Inhibits MAC formation or insertion

PNH

CR1 (CD35) (immune adherence or C4b/C3b receptor)

C3b, C4b, C3, and C5 convertases

Cofactor for factor I to cleave C4b and C3b (CA); displaces Bb from C3b and C2a from C4b to inhibit convertases (DAA)

No complete deficiency described; decreased levels in immune complex– mediated diseases such as lupus

CRIg

C3b, iC3b, C3c

Inhibits activation of AP

None defined

MEMBRANE-BOUND FACTORS REGULATING COMPLEMENT FUNCTIONAL ACTIVITY

DISEASE, IF DEFICIENT

*Factor H and C4bp also bind to surfaces, particularly at sites of tissue and cellular injury, where they also carry out regulatory activity. † If heterozygous deficient, individual is predisposed to AMD and aHUS. If hom*ozygous deficient, the AP turns over excessively, resulting in kidney disease (C3 glomerulopathies) and bacterial infections (secondary to the very low C3). aHUS, atypical hemolytic uremic syndrome; AMD, age-related macular degeneration; AP, alternative pathway; (C3b)2 Bb, alternative pathway C5 convertase; C3bC4bC2a, classical and lectin pathway C5 convertase; C4bC2a, classical and lectin pathway C3 convertase; CA, cofactor activity; CR1, complement receptor type 1; CRIg, complement receptor of the Ig superfamily; DAA, decay-accelerating activity; GAG, glycosaminoglycan; HAE, hereditary angioedema; MAC, membrane attack complex; MASP, mannan-binding lectin-associated serine protease; MBL, mannan or mannose binding lectin; PNH, paroxysmal nocturnal hemoglobinuria.

Vitronectin, clusterin, protectin CP/LP C5 con

C5

C6

C4b2a3b

C8

C5b

C3bBb3b

MAC C7

C5a

C9

AP C5 con

A

Increasing lipophilicity Topography of C5b-9 assembly

C6 β α'

β C7 α'

β

β

n

β α' C6

α' C6

γ α C8

C6 C7

C7 α β

γ

C9

β α' C6

C9 site

C7 α

γ C9 C9C9

β C9

C5b

B

C5b-6 C5b-7

C5b-8

C5b-9n

FIGURE 50-4.  Activation of C5 and the membrane attack complex (MAC). A, The C5 convertases (“con”) are the same as C3 convertases except a C3b has been attached to C4bC2a or a second C3b in the case of AP C5 convertase. B, Schematic representation of the assembly of the assembly of the MAC on a cell membrane. C5b (composed of two chains) binds C6 and then C7. The C5b-7 complex can insert into a membrane and then bind C8 (composed of three chains) and multiple C9s to form a pore or channel in the membrane. (Modified from Liszewski MK, et al. The Human Complement System in Health and Disease. Marcel Dekker; New York, NY 1998.)

may be derived from either the classical or lectin pathway. Thus, activation of complement by any one of the three pathways has the potential to be rapidly magnified.7 The AP C3 convertase is negatively controlled (to maintain homeostasis) both in the fluid phase and on host cells by two abundant plasma proteins and two widely expressed membrane proteins.8

The central role of the AP as an amplifier of complement activation is borne out by its association with a number of clinicopathologic states in the setting of deficient regulation (Tables 50-6 and 50-7). For example, multiple forms of membranoproliferative glomerulonephritis are associated with excessive C3 fragment deposition in the kidney because of either the

244

CHAPTER 50  Complement System in Disease  

TABLE 50-7  COMPLEMENT SYSTEM MEDIATES THE DISEASE PROCESS AND ITS INHIBITION TREATS THE CONDITION DISEASE

PATHOPHYSIOLOGY

ETIOLOGY

TREATMENT

FDA APPROVED

PNH

Lyse RBCs

Acquired hemopoietic somatic stem cell mutation in gene required for synthesis of GPI anchor

mAb to C5

Yes

aHUS

Damage to endothelial cells

Inherited loss of function variants in AP regulators or gain of function variants in AP activators

mAb to C5

Yes

HAE

Bradykinin generation

Autosomal dominant variants in the C1-inhibitor gene

C1 inhibitor replacement Bradykinin receptor blockage Kallikrein inhibitor

Yes Yes Yes

AMD

Degeneration of the retina

Inherited variants in a regulator (FH or FI) or gain of function in an alternative pathway component (C3 or FB)

Clinical trials in progress

No

aHUS= atypical hemolytic uremic syndrome; AMD = age-related macular degeneration; FDA = Food and Drug Administration; HAE= hereditary angioedema; GPI = glycosyl phosphatidylinositol; mAb = monoclonal antibody; PNH= paroxysmal nocturnal hemoglobinuria; RBC = red blood cell.

presence of autoantibodies (C3 or C4 nephritic factors) that stabilize C3 convertases or a genetic deficiency in complement regulatory protein (factor H or factor I).9 Likewise, atypical hemolytic uremic syndrome (i.e., not associated with a preceding enteropathic infection featuring a Shiga-like toxin) occurs in individuals who harbor heterozygous missense mutations in factor H or I or have gain-of-function mutations in factor B or C3.10 Genome-wide association and targeted deep sequencing studies have also linked age-related macular degeneration (AMD) to functional coding mutations in factor H and factor I and more uncommonly in factor B and C3.11 Finally, rodent models of rheumatoid arthritis, SLE, and ANCA-positive vasculitic syndromes are ameliorated if the AP is disrupted.

C3 and C5 Convertases

The three activation pathways converge at C3. A remarkable feature of C3 is the presence of a thioester bond. Buried within the three-dimensional structure of the C3 protein lies a γ-carboxy group of a reactive glutamic acid residue linked to a cysteine in an “internal thioester.” Upon its cleavage, for a few microseconds, a covalent attachment can occur via an ester or amide linkage to any nearby hydroxyl or amino group. Most of the cleaved thioester bonds are hydrolyzed by water to produce a form of C3 (known as C3 [H2O]); however, a substantial percentage forms an amide or ester bond to amino groups or carbohydrates thereby covalently attaching C3b to a target’s surface. Also, the addition of a C3b to C4b2a (classical/lectin pathway C3 convertase) or to C3bBb (AP C3 convertase) then forms a convertase for C5 (C3bBbC3b for the AP and C4bC2aC3b for the CP/LP).

Regulators of Complement Activation at the C3 and C5 steps

The regulators of complement activation (RCA) (see Table 50-2) limit the production of C3b, primarily by the AP C3 convertases. Because the addition of C3b to a C3 convertase makes it a C5 convertase, regulation of the two enzyme complexes is linked. Modulation of their activity on host cells limits tissue destruction and the production of inflammatory mediators. The RCA proteins control complement activation by two processes. Decayaccelerating activity refers to when the inhibitor transiently binds to C3b or C4b in the convertase and thereby dissociates the other members of the complex, rendering it enzymatically inactive (as the component released is the catalytic domain of the protease). The second is cofactor activity, which requires recognition of C3b or C4b by a plasma cofactor protein. Upon this interaction, the protease, factor I, cleaves C3b or C4b. Cleavage of C3b by factor I renders the convertase irreversibly inactive (generates iC3b which cannot participate in convertase formation).

Membrane Attack Complex

The cleavage of C5 generates C5a, the most potent of the complement anaphylatoxins, and C5b. C5b associates with C6 and C7 to create a lipophilic trimer as the initial part of the MAC (Fig. 50-4). The C5b67 trimer inserts into the lipid bilayer and serves as a binding site for C8 and C9. C9 selfpolymerizes, leading to 12 to 18 C9 molecules that form a ring structure (completing the MAC). The MAC resembles a doughnut with a 10-nm pore running through the center. This pore allows water and ions to enter the cells, ultimately leading to osmotic lysis. Many pathogens such as gram-positive bacteria possess a capsule that makes them resistant to lysis.12 Opsonization leading to phagocytosis is thus the major means of eliminating such organisms.

TABLE 50-8  DISTRIBUTION OF ANAPHYLATOXIN RECEPTORS AND THEIR CELLULAR RESPONSES CELL TYPE

RESPONSES

C5aR (CD88) Neutrophils Eosinophils Basophils Mast cells Monocytes Hepatocytes Pulmonary epithelium Neuronal cells Endothelial cells Renal epithelial/mesangial cells

Chemotaxis Enzyme release Generation of reactive oxygen species Upregulation of adhesion molecules Increased synthesis of IL-1, IL-6, and IL-8 Prostaglandin and leukotriene synthesis Increased synthesis of acute phase reactants Increased IL-8 Cellular activation Increased expression of P-selectin Proliferation Synthesis of growth factors

C3aR Eosinophils Mast cells Platelets Epithelial, endothelial, etc.

Chemotaxis Enzyme release Generation of reactive oxygen species Upregulation of adhesion molecules Cellular activation

CNS = central nervous system; IL = interleukin.

The MAC appears to be essential only for elimination of Neisseria spp. Individuals completely deficient in C5, C6, C7, C8, or C9 are at an increased risk only for meningococcal and gonococcal infections. C9 deficiency is a common immunodeficiency in Japan, with a heterozygote frequency of 3% to 5%. Thus, heterozygous deficiency seems to not be deleterious to the population in general but may have a selective advantage. Extensive complement activation during an inflammatory response can result in sufficient MAC deposition to produce host cell lysis. Host cells, however, have mechanisms in place to resist the osmotic changes caused by the MAC and to block assembly of the MAC as it is formed (the protein is known as protectin or CD59). Rather, the nonlethal effects of sublytic MAC deposition are more likely to contribute to pathology. In most cells, this occurs through a general activation of multiple cell signaling pathways. The response to MAC deposition at sites of complement activation depends on the cell type (Table 50-8). In phagocytic cells, such as neutrophils or macrophages, sublytic MAC insertion leads to the production of reactive oxygen species (e.g., superoxide, hydrogen peroxide) as well as release of prostaglandins and leukotrienes. Platelets undergoing a “MAC attack” incorporate phosphatidylserine on their outer membrane, facilitating formation of blood coagulation enzyme complexes with a potentially procoagulant effect. On endothelial cells, MAC deposition induces the synthesis of interleukin-1α (IL-1α), which leads to further autocrine and paracrine endothelial cell activation. It stimulates a procoagulant state by (1) altering the phospholipid composition of the endothelial membrane; (2) inducing the synthesis of tissue factor and upregulating the synthesis of plasminogen activator inhibitor; (3) upregulating the expression of adhesion molecules, including intercellular adhesion molecule 1 (ICAM-1) and E-selectin; and (4) stimulating endothelial cells to proliferate through growth factor production. In summary,

CHAPTER 50  Complement System in Disease  

LP AP

245

CP

TABLE 50-9  RESPONSES TO SUBLYTIC MEMBRANE ATTACK COMPLEX ACTIVATION

Eculizumab C3

CELL TYPE

C5

C3b

C5a

C3b Bb

C3b

C3b B

Bb

C3b

C5b

MAC

C3 con

C5 con FIGURE 50-5.  Complement activation and the mechanism of action of eculizumab (monoclonal antibody [mAb] to C5). The alternate pathway (AP) constantly undergoes “tickover” but can also be primed by the classical pathway (CP) and lectin pathway (LP). The C3b that is formed interacts with factor B (B), which is then cleaved by factor D to form a C3 convertase (C3Bb). As more C3b is generated, some binds to the C3 convertase to form a C5 convertase. mAb to C5 (eculizumab) prevents the cleavage of C5 by the C5 convertase. Not shown is properdin that binds to both the C3 and C5 convertases to increase their half-lives approximately five- to 10-fold (from ≈30 seconds to several minutes).7 (Modified from Wong EK, Goodship TH, Kavanagh D. Complement therapy in atypical haemolytic uraemic syndrome (aHUS). Mol Immunol. 2013;56(3):199-212.)

although cell death does not usually occur, deposition of the sublytic levels of MAC leads to a potentially dangerous situation, with increased inflammation, a procoagulant state, and cellular proliferation. Of course, part of this response is necessary at sites of injury to eliminate pathogens and debris and to facilitate wound repair. The short duration of complement activation and the presence of inhibitors help to maintain homeostasis. Regulation of MAC formation is important clinically (Fig. 50-5). Two plasma proteins, clusterin and S-protein (vitronectin), bind the C5b-7 complex and prevent its association with the lipid membrane. C8 and multiple C9 molecules adhere to this soluble complex, termed soluble C5b-9, which is lytically inactive. CD59 (protectin) is a membrane-bound inhibitor of MAC formation. This small glycoprotein is attached to the cell membrane through a glycosyl phosphatidylinositol tail (GPI anchor). It binds to C5b-8, inserted in the cell membrane, to prevent binding to and polymerization of C9. The expression of CD59 is defective in patients with paroxysmal nocturnal hemoglobinuria (PNH), owing to the failure to synthesize the GPI anchor used by this and many other membrane proteins (including decay-accelerating factor [DAF]) to insert on the cell. The clinical features of PNH are primarily chronic hemolysis and intermittent thrombosis. Hemolysis is caused by complement activation on RBCs because of a lack of DAF and particularly CD59. Thrombosis is likely secondary to intravascular complement activation, leading to endothelial cell activation. The primary defect is an acquired hematopoietic stem cell mutation of a gene on the X chromosome responsible for encoding the first enzyme in the pathway to synthesize a GPI-anchor.

Anaphylatoxins

The anaphylatoxins serve a key early role in initiating a local inflammatory response as they trigger pathways to prepare a cell to face a pathogen or injury (Table 50-9). Similar to the MAC, anaphylatoxins are another major source of potential pathologic damage to self that results from complement activation. These peptides, C3a and C5a, are cleaved from their respective proteins during complement activation. They were named in 1910 to describe their toxic effects, including shock after the transfer of complement-activated serum into laboratory animals. They are 77 (C3a) or 74 (C5a) amino acids long and contain a key carboxy (C)-terminal arginine. They interact with the anaphylatoxin receptors. In plasma, the C-terminal arginine is removed by carboxypeptidase-N from anaphylatoxins not bound to their receptors. Depending on the response studied, this removal totally inactivates the anaphylatoxin or reduces its potency by about 1000-fold. The C5a receptor (C5aR [CD88]) is a seven-transmembrane-spanning protein that couples ligand binding to G-protein signaling. Expressed on myeloid cells, particularly neutrophils and eosinophils, it mediates the potent

EFFECTS

Most cells

Increased intracellular calcium flux Activation of G proteins Activation of protein kinases Activation of transcription factors Proliferation

Neutrophils and macrophages

Release of reactive oxygen species Activation of phospholipase A2 Release of prostaglandins, thromboxane, and leukotrienes

Platelets

Release of ATP Increased P-selectin expression Procoagulant membrane changes

Endothelial cells

Increased synthesis of IL-1α Increased release of tissue factor Increased release of von Willebrand factor Increased synthesis of basic fibroblast and plateletderived growth factors

Synoviocytes

Increased synthesis of prostaglandin Increased synthesis of IL-6 Increased production of matrix metalloproteinase

Glomerular epithelium

Activation of phospholipase A2 Synthesis of prostaglandin Increased synthesis of collagen and fibronectin

Oligodendrocytes

Increased synthesis of myelin basic protein and proteolipids Increased proliferation

ATP = adenosine triphosphate; IL = interleukin.

chemoattractant property of C5a for both of these cell types. Signaling through CD88 leads to rapid secretion of all granule contents. These include lipases and proteases as well as lactoferrin from neutrophils, and peroxidase, major basic protein, and cationic protein from eosinophils. C5a also induces the release of cytokines, such as tumor necrosis factor (TNF), IL-1, IL-6, IL-8, and adhesion molecules, promoting the inflammatory response. The C5aR is expressed by numerous other tissues, including hepatocytes, bronchial and alveolar epithelium, vascular endothelium, renal mesangial and tubular epithelial cells, and brain neuronal cells. These cells are activated by receptor engagement, leading to production and release of cytokines, chemokines, and prostaglandins and to cellular proliferation. The C3a receptor is also a seven-transmembrane-domain protein. It is expressed on almost all myeloid cells, including mast cells, where it mediates the release of allergic mediators. The C3aR also has been detected on many tissues, including in the brain and lung. The anaphylatoxins have multiple biologic effects. In general, they cause smooth muscle contraction and recruitment of granulocytes, monocytes, and mast cells. In theory, they can contribute to the pathophysiology of any inflammatory condition. In disease models, C3a and C5a have been shown to play a role in diseases such as acute respiratory distress syndrome (ARDS), multisystem organ failure, septic shock, myocardial ischemia-reperfusion injury, asthma, rheumatoid arthritis, SLE, and inflammatory bowel disease. The anaphylatoxin peptides also are responsible for the “postpump” syndrome seen in patients undergoing cardiopulmonary bypass or hemodialysis. Exposure of blood to dialysis or perfusion membranes leads to complement activation. Within minutes of starting bypass, there is a sharp increase in the levels of C3a and C5a in the extracorporeal circuit being returned to the patient. This increase can be associated with respiratory distress, pulmonary hypertension, and pulmonary edema. It has been shown that the length of time that patients stay on the ventilator after bypass surgery correlates with the level of C3a generated during reperfusion. C3a and C5a have been implicated in the initiation and prolongation of ARDS and multisystem organ failure. After severe trauma, levels of C3a have been measured that suggest activation of the entire circulating C3 pool. This activation leads to bronchoconstriction, increased vascular permeability, hypotension, and vascular plugging with leukocytes. The activation of white blood cells continues the cycle of tissue damage with further complement activation. Continued elevation of C3a in shock or ARDS is a poor prognostic sign. C3a and C5a also appear to play a major role in the pathogenesis of asthma.

246

CHAPTER 50  Complement System in Disease  

C4b

C3b

C3b

CR1 CR2

C3d

C4bp

x7 MCP

Other concerns about complement inhibition include whether it is shortor long term and whether it is systemic or local. Long-term inhibition of complement, particularly at one of the early steps, is likely to predispose to infection and possibly autoimmunity. Short-term (hours to days) inhibition at any step is unlikely to cause problems. Given that inflammation is usually a local phenomenon, several mechanisms are being tested to target complement inhibitors to these sites. In this way, higher levels of inhibition can be achieved when needed and with lower doses of inhibitor.

Natural Complement Inhibitors C3b Factor H

DAF

FIGURE 50-6.  Model of the complement regulators required to inhibit complement activation on self at the steps of C3 and C5 cleavage. Circles represent individual complement control repeats (~60 amino acids each), and shading indicates higher organizational units composed of several repeats. The approximate locations for binding of C3b and C4b fragments are indicated. Factor H and C4bp are plasma proteins. C4bp = C4-binding protein; CR = complement receptor; DAF = decay-accelerating factor; MCP = membrane cofactor protein.

Complement Receptors

Opsonization of target by C4b and C3b is effective in preventing infections because these two complement fragments (and the further cleavage products in the case of C3b) are ligands for complement receptors (Fig. 50-6). After covalent attachment of C4b and C3b, immune adherence occurs between the opsonized microbe and immune cells, predominantly neutrophils, monocytes, and macrophages. Complement opsonins are highly effective mediators of immune adherence. On phagocytic cells, this is the prelude to the ingestion and destruction of the target antigen. On RBCs, immune adherence is followed by transfer of the C4b/C3b-coated cargo to monocytes and macrophages in the liver and spleen. CR1 is particularly efficient at immune adherence. Proteolytic modification of C3b leads to iC3b, which is a ligand for the highly phagocytic CR3 and CR4. A further degradation of iC3b to C3d leads to an interaction with CR2 to lower the threshold of B cell activation. Overall, the process is designed with two goals in mind: first is destruction by phagocytosis of the microbe and second is to coat microbial antigens for an adaptive immune response. For example, follicular dendritic cells and B lymphocytes express CR1, CR2, CR3, and CR4 that facilitate complementcoated antigens to be bound, internalized, and presented to other immune cells. CR3 and CR4 facilitate phagocytosis, and CR2 on follicular dendritic cells facilitates immunologic memory generation.

COMPLEMENT INHIBITORS

Given the many disease states in which complement is one of the central mediators of pathology, it is no surprise that complement inhibitors are in preclinical or clinical development for treatment of human diseases (see Table 50-2 and Fig. 50-6). These inhibitors take several different forms. Whereas some are variations of physiologic inhibitors, others are the products of molecular biologic searches for novel compounds. It is important to consider where in the complement pathway to design an inhibitor to act. Inhibition of the activation pathways limits the production of biologically active peptides. Inhibiting the activation of C3 not only prevents the generation of the C3a anaphylatoxin but also may leave the patient susceptible to infection by limiting the deposition of C3b on targets as an opsonin. Inhibition of C3b deposition would decrease the patient’s ability to clear immune complexes, potentially resulting in renal, pulmonary, and vascular damage. It also might promote the development of antibodies to self-antigens. Inhibition of the C5 convertases is an attractive goal because it would prevent the generation of the C5a anaphylatoxin and the MAC (see Fig. 50-5). This strategy would inhibit complement activation without limiting C3b deposition. Inhibitors based on this concept have been successful; the mAb to C5 is approved by the Food and Drug Administration (FDA) to treat PNH and atypical hemolytic-uremic syndrome (aHUS).

Naturally occurring compounds that control complement activation include products or extracts of plants, fungi, insects, bacteria, viruses, and venoms.12 The mechanism of complement inhibition by some of these natural products is known and is of clinical and experimental importance. In particular, to protect themselves from the host’s complement system, poxviruses, herpesviruses, and flaviviruses produce either mimics of the human regulators that they at one time hijacked from their hosts or proteins that bind the hosts’ regulators such as the plasma protein factor H. Bacteria also express a wide variety of inhibitors of the human complement system. Staphylococcus aureus, for example, synthesizes up to 10 distinct proteins that inhibit at almost every key step of the complement cascade. Cobra venom factor (CVF) is a modified form of cobra C3b secreted by venom glands in the oral cavity. It is an 144,000-dalton glycoprotein that forms an AP convertase in association with host factor B. Upon injection, CVF leads to massive activation of the AP, leading to shock and pulmonary microvascular injury in experimental animals. It is resistant to the host’s inhibitors because the site for that interaction is altered on CVF. Perhaps the most widely used natural inhibitor of complement activation is heparin. It decreases activation of the CP and AP. In clinical practice, the anticomplementary effect of heparin has been used to prevent complement activation during cardiopulmonary bypass. Measurement of complement activation products such as C3a or soluble C5b-9 after bypass showed decreases of 35% to 70% for adult and pediatric patients when heparin-coated extracorporeal circuits were used. Although numerous studies have looked at the decrease in complement activation by heparin-coated bypass circuits, there have been few attempts to correlate this with clinical outcome.

Anti-C5

The complement inhibitor that has achieved the widest attention as a therapeutic agent to stop complement activation is a mAb to C5 that prevents its cleavage to C5a (potent anaphylatoxin) and C5b (initiator of the MAC) (see Fig. 50-5). The generation of the C3b and C4b still occurs, allowing opsonization of pathogens and formation of immune complexes. Because activation of early complement components is also important for the maintenance of tolerance to self-antigens, inhibition of C5 activation is less worrisome than inhibition of C3 activation. The one consequence of C5 deficiency in humans is an increased risk of Neisseria infections that can largely be mitigated through vaccination. The anti-C5 mAb eculizumab has been approved for use in patients with PNH and aHUS. The complement system is undergoing a renaissance. There are several reasons but probably the foremost is the discovery of mutations in complement regulators leading to aHUS and AMD. Second is the therapeutic success of a mAb to C5 in the treatment of aHUS and PNH. Third is the introduction of purified C1-Inhibitor, kallikrein inhibitors, and a bradykinin receptor antagonist to prevent and treat swelling attacks in hereditary angioedema. Last, intriguing recent data implicate the complement system in the pathophysiology of multiple disorders, including AMD, ischemia/reperfusion injury, organ regeneration, brain development (pruning of undesirable synapses), obesity, asthma, T-cell activation phenomena associated with allergic and rheumatic diseases,13 and more. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 50  Complement System in Disease  

GENERAL REFERENCES 1. Ricklin D, Hajishengallis G, Yank K, et al. Complement: a key system for immune survelliance and homeostasis. Nat Immunol. 2010;11:785-797. 2. Frank MM. Complement disorders and hereditary angioedema. J Allergy Clin Immunol. 2010; 125:5262-5271. 3. Clarke EV, Tenner AJ. Complement modulation of T cell immune responses during homeostasis and disease. J Leukoc Biol. 2014;96:745-756. 4. Cozzani E, Drosera M, Gasparini G, et al. Serology of lupus erythematosus: Correlation between immunopathological features and clinical aspects. Autoimmune Dis. 2014;2014:321359. 5. Sethi S, Fervenza FC. Pathology of renal diseases associated with dysfunction of the alternative pathway of complement: C3 glomerulopathy and atypical hemolytic uremic syndrome (aHUS). Semin Thromb Hemost. 2014;40:416-421. 6. Genster N, Takahashi M, Sekine H, et al. Lessons learned from mice deficient in lectin complement pathway molecules. Mol Immunol. 2014;61:59-68.

246.e1

7. Kemper C, Atkinson JP, Hourcade DE. Properdin: emerging roles of a pattern-recognition molecule. Annu Rev Immunol. 2010;28:131-155. 8. Nesargikar PN, Spiller B, Chavez R. The complement system: history, pathways, cascade and inhibitors. Eur J Microbiol Immunol (Bp). 2012;2:103-111. 9. Sethi S, Fervenza FC. Membranoproliferative glomerulonephritis—a new look at an old entity. N Engl J Med. 2012;366:1119-1131. 10. Joseph C, Gattineni J. Complement disorders and hemolytic uremic syndrome. Curr Opin Pediatr. 2013;25:209-215. 11. Seddon JM, Yu Y, Miller EC, et al. Rare variants in CFI, C3 and C9 are associated with high risk of advanced age-related macular degeneration. Nat Genet. 2013;45:1366-1370. 12. Okumura CY, Nizet V. Subterfuge and sabotage: evasion of host innate defenses by invasive grampositive bacterial pathogens. Annu Rev Microbiol. 2014;68:439-458. 13. Liszewski MK, Kolev M, Le Friec G, et al. Intracellular complement activation sustains T cell homeostasis and mediates effector differentiation. Immunity. 2013;39:1143-1157.

246.e2

CHAPTER 50  Complement System in Disease  

REVIEW QUESTIONS 1. What statement is false about the alternate pathway of complement activation? A. It serves as a feedback or amplification loop for each pathway. B. It mediates lysis and cell damage in several types of hemolytic disorders. C. It is involved in debris removal and wound repair. D. It requires lectins and antibodies to be initiated. Answer: D  The alternate pathway is the original complement system. For example, it is present in insects and echinoderms and functions in hemolymph similar to its role in blood vertebrates. It is often called the “guardian of the intravascular space,” being particular designed to prevent invasion by bacteria. It is continuously “turning over” at a low rate. Antibody and lectins are more evolutionary recent means to specifically target complement activation. 2. Which statement is not true of the classical pathway? A. In conjunction with IgG and IgM, the classical pathway mediates tissue damage in autoimmune diseases. B. A complete deficiency of an early component predisposes to SLE. C. A total complement titer (CH50 or THC) measures the quantity of hemolytic activity in this pathway. D. Proteins such as CRP and serum amyloid protein (SAP) also can activate the classical pathway. E. All are true. Answer: E  It requires only a single IgM or two IgGs in close proximity to activate the classical complement pathway. IgA and IgE do not activate the classical complement pathway. 3. A deficiency of a complement inhibitor at the steps of C3 and C5 activation leads to all of the following except: A. excessive production of the C3a and C5a anaphylatoxins. B. excessive production of opsonins and the membrane attack complex. C. systemic lupus erythematosus and closely related rheumatic diseases. D. renal disease (atypical hemolytic uremic syndrome, membrano­ proliferative glomerulonephritis, C3 glomerulonephritis, and C3 glomerulopathies. Answer: C  Excessive activation at sties of injury or degeneration is a hallmark of inadequately regulated C3 and C5 convertases. The diseases for which this type of pathogenesis has been most studies are atypical hemolytic uremic syndrome and age-related macular degeneration. The other commonly observed phenotype is renal disease as described in Answer D.

4. Which statement is not true? A. CP is activated by the Fc portion of IgG or IgM engaging C1q, which is part of C1 complex. B. Complement receptor one (CR1, CD35) on RBCs serves as a taxi or ferry to take C4b and C3b bearing immune complexes to liver and spleen for disposal. C. Only a limited number of specific foreign antigens can be coated with complement fragments. D. Complement proteins in blood are synthesized by the liver, but most cell types also express complement proteins locally. Answer: C  Almost all foreign materials become coated with complement C3 fragments. Of note, a role in host defense and response to injury for local synthesis of complement components remains to be definitely established. 5. The C3a and C5a anaphylatoxins accomplish all but which one of the following? A. Engage their specific receptors to activate many cell types. B. If liberated in substantial amounts, they can lead to shock and death. C. They are responsible for opsonic activity of the complement system. D. Both are produced by all complement pathways. Answer: C  C3a and C5a are liberated in substantial amounts and this occurs in seconds. It is an early warning system to the host cell to develop a proinflammatory environment. The major opsonin of the complement system is C3b and its subsequent limited proteolytic degradation fragments. C4b is also an opsonin if the classical pathway or lectin pathway is activated.

248

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

51  APPROACH TO THE PATIENT WITH POSSIBLE CARDIOVASCULAR DISEASE LEE GOLDMAN Patients with cardiovascular disease may present with a wide range of symptoms and signs, each of which may be caused by noncardiovascular conditions. Conversely, patients with substantial cardiovascular disease may be asymptomatic. Because cardiovascular disease is a leading cause of death in the United States and other developed countries, it is crucial that patients be evaluated carefully to detect early cardiovascular disease, that symptoms or signs of cardiovascular disease be evaluated in detail, and that appropriate therapy be instituted. Improvements in diagnosis, therapy, and prevention have contributed to a 70% or so decline in age-adjusted cardiovascular death rates in the United States since the 1960s. Furthermore, among people age 65 years and older, regular visits to a primary care physician are associated with a 25 to 30% reduction in overall mortality. However, the absolute number of deaths from cardiovascular disease in the United States has not declined proportionately because of the increase in the population older than 40 years as well as the aging of the population in general. In evaluating a patient with known or suspected heart disease, the phy­ sician must determine quickly whether a potentially life-threatening condition exists. In these situations, the evaluation must focus on the specific issue at hand and be accompanied by the rapid performance of appropriately directed additional tests. Examples of potentially life-threatening conditions include acute myocardial infarction (MI) (Chapter 73), unstable angina (Chapter 72), suspected aortic dissection (Chapter 78), pulmonary edema (Chapter 59), and pulmonary embolism (Chapter 98).

USING THE HISTORY TO DETECT CARDIOVASCULAR SYMPTOMS

Patients may complain spontaneously of a variety of cardiovascular symptoms (Table 51-1), but sometimes these symptoms are elicited only by obtaining a careful, complete medical history. In patients with known or suspected cardiovascular disease, questions about cardiovascular symptoms are key components of the history of present illness; in other patients, these issues are a fundamental part of the review of systems.

Chest Pain Chest discomfort or pain is the cardinal manifestation of myocardial ischemia resulting from coronary artery disease or any condition that causes myocardial ischemia by an imbalance of myocardial oxygen demand compared with myocardial oxygen supply (Chapter 71). New, acute, often ongoing pain may indicate an acute MI, unstable angina, or aortic dissection; a pulmonary cause, such as acute pulmonary embolism or pleural irritation; a musculoskeletal condition of the chest wall, thorax, or shoulder; or a gastrointestinal abnormality, such as esophageal reflux or spasm, peptic ulcer disease, or cholecystitis (Table 51-2). The chest discomfort of MI commonly occurs without an immediate or obvious precipitating clinical cause and builds in intensity for at least several minutes; the sensation can range from annoying discomfort to severe pain (Chapter 73). Although a variety of adjectives may be used by patients to describe the sensation, physicians must be suspicious of any discomfort, especially if it radiates to the neck, shoulder, or arms. The probability of an acute MI can be estimated by integrating information from the history, physical examination, and electrocardiogram (Fig. 51-1). The chest discomfort of unstable angina is clinically indistinguishable from that of MI except that the former may be precipitated more clearly by activity and may be more rapidly responsive to antianginal therapy (Chapter 72). Aortic dissection (Chapter 78) classically presents with the sudden onset of severe pain in the chest and radiating to the back; the location of the pain often provides clues to the location of the dissection. Ascending aortic dissections commonly present with chest discomfort radiating to the back, whereas dissections of the descending aorta commonly present with back pain radiating to the abdomen. The presence of back pain or a history of hypertension or other predisposing factors, such as Marfan syndrome, should prompt a careful assessment of peripheral pulses to determine whether the great vessels are affected by the dissection and of the chest radiograph to

evaluate the size of the aorta. If this initial evaluation is suggestive, further testing with transesophageal echocardiography, computed tomography (CT), or magnetic resonance imaging (MRI) is indicated. The pain of pericarditis (Chapter 77) may simulate that of an acute MI, may be primarily pleuritic, or may be continuous; a key physical finding is a pericardial rub. The pain of pulmonary embolism (Chapter 98) is commonly pleuritic in nature and is associated with dyspnea; hemoptysis also may be present. Pulmonary hypertension (Chapter 68) of any cause may be associated with chest discomfort with exertion; it commonly is associated with severe dyspnea and often is associated with cyanosis. Recurrent, episodic chest discomfort may be noted with angina pectoris and with many cardiac and noncardiac causes (Chapter 71). A variety of stress tests (Table 51-3) can be used to provoke reversible myocardial ischemia in susceptible individuals and to help determine whether ischemia is the pathophysiologic explanation for the chest discomfort (Chapter 71).

Dyspnea Dyspnea, which is an uncomfortable awareness of breathing, is commonly caused by cardiovascular or pulmonary disease. A systematic approach (see Fig. 83-3) with selected tests nearly always reveals the cause. Acute dyspnea can be caused by myocardial ischemia, heart failure, severe hypertension, pericardial tamponade, pulmonary embolism, pneumothorax, upper airway obstruction, acute bronchitis or pneumonia, or some drug overdoses (e.g., salicylates). Subacute or chronic dyspnea is also a common presenting or accompanying symptom in patients with pulmonary disease (Chapter 83). Dyspnea also can be caused by severe anemia (Chapter 158) and can be confused with the fatigue that often is noted in patients with systemic and neurologic diseases (Chapters 256 and 396). In heart failure, dyspnea typically is noted as a hunger for air and a need or an urge to breathe. The feeling that breathing requires increased work or effort is more typical of airway obstruction or neuromuscular disease. A feeling of chest tightness or constriction during breathing is typical of bronchoconstriction, which is commonly caused by obstructive airway disease (Chapters 87 and 88) but also may be seen in pulmonary edema. A feeling of heavy breathing, a feeling of rapid breathing, or a need to breathe more is classically associated with deconditioning. In cardiovascular conditions, chronic dyspnea usually is caused by increases in pulmonary venous pressure as a result of left ventricular failure (Chapters 58 and 59) or valvular heart disease (Chapter 75). Orthopnea, which is an exacerbation of dyspnea when the patient is recumbent, is caused by increased work of breathing because of either increased venous return to the pulmonary vasculature or loss of gravitational assistance in diaphragmatic effort. Paroxysmal nocturnal dyspnea is severe dyspnea that awakens a patient at night and forces the assumption of a sitting or standing position to achieve gravitational redistribution of fluid.

Palpitations Palpitations (Chapter 62) describe a subjective sensation of an irregular or abnormal heartbeat. Palpitations may be caused by any arrhythmia (Chapters 64 and 65) with or without important underlying structural heart disease. Palpitations should be defined in terms of the duration and frequency of the episodes; the precipitating and related factors; and any associated symptoms of chest pain, dyspnea, lightheadedness, or syncope. It is crucial to use the history to determine whether the palpitations are caused by an irregular or a regular heartbeat. The feeling associated with a premature atrial or ventricular contraction, often described as a “skipped beat” or a “flip-flopping of the heart,” must be distinguished from the irregularly irregular rhythm of atrial fibrillation and the rapid but regular rhythm of supraventricular tachycardia. Associated symptoms of chest pain, dyspnea, lightheadedness, dizziness, or diaphoresis suggest an important effect on cardiac output and mandate further evaluation. In general, evaluation begins with ambulatory electrocardiography (ECG) (Table 51-4), which is indicated in patients who have palpitations in the presence of structural heart disease or substantial accompanying symptoms. Depending on the series, 9 to 43% of patients have important underlying heart disease. In such patients, more detailed evaluation is warranted (see Fig. 62-1). Lightheadedness or syncope (Chapter 62) can be caused by any condition that decreases cardiac output (e.g., bradyarrhythmia, tachyarrhythmia, obstruction of the left ventricular or right ventricular inflow or outflow, cardiac tamponade, aortic dissection, or severe pump failure), by reflexmediated vasomotor instability (e.g., vasovagal, situational, or carotid sinus syncope), or by orthostatic hypotension (see Table 62-1). Neurologic

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

TABLE 51-1  CARDINAL SYMPTOMS OF CARDIOVASCULAR DISEASE Chest pain or discomfort Dyspnea, orthopnea, paroxysmal nocturnal dyspnea, wheezing Palpitations, dizziness, syncope Cough, hemoptysis Fatigue, weakness Pain in extremities with exertion (claudication)

249

diseases (e.g., migraine headaches, transient ischemic attacks, or seizures) also can cause transient loss of consciousness. The history, physical examination, and ECG are often diagnostic of the cause of syncope (see Table 62-2). Syncope caused by a cardiac arrhythmia usually occurs with little warning. Syncope with exertion or just after conclusion of exertion is typical of aortic stenosis and hypertrophic obstructive cardiomyopathy. In many patients, additional testing is required to document central nervous system disease, the cause of reduced cardiac output, or carotid sinus syncope. When the history, physical examination, and ECG do not provide helpful diagnostic information that points toward a specific cause of syncope, it is imperative that patients with heart disease or an abnormal ECG be tested with continuous ambulatory ECG monitoring to diagnose a possible arrhythmia (see Fig. 62-1); in selected patients, formal electrophysiologic testing may be indicated

TABLE 51-2  CAUSES OF CHEST PAIN CONDITION

LOCATION

QUALITY

DURATION

AGGRAVATING OR RELIEVING FACTORS

ASSOCIATED SYMPTOMS OR SIGNS

CARDIOVASCULAR CAUSES Angina

Retrosternal region; radiates to or occasionally isolated to neck, jaw, epigastrium, shoulder, or arms (left common)

Pressure, burning, squeezing, heaviness, indigestion

1 flight in normal conditions

Patient can perform to completion any activity requiring ≥2 metabolic equivalents but cannot and does not perform to completion any activities requiring ≥5 metabolic equivalents, e.g., shower without stopping, strip and make bed, clean windows, walk 2.5 mph, bowl, play golf, dress without stopping

IV

Patients with cardiac disease resulting in inability to carry on any physical activity without discomfort Symptoms of cardiac insufficiency or of the anginal syndrome may be present even at rest. If any physical activity is undertaken, discomfort is increased.

Inability to carry on any physical activity without discomfort—anginal syndrome may be present at rest

Patient cannot or does not perform to completion activities requiring ≥2 metabolic equivalents; cannot carry out activities listed above (Specific Activity Scale, class III)

From Goldman L, Hashimoto B, Cook EF, et al. Comparative reproducibility and validity of systems for assessing cardiovascular functional class: advantages of a new specific activity scale. Circulation. 1981;64:1227-1234. Reproduced by permission of the American Heart Association.

(Chapter 62). In patients with no evident heart disease, tilt testing (Chapter 62) can help detect reflex-mediated vasomotor instability.

Other Symptoms Nonproductive cough (Chapter 83), especially a persistent cough (see Fig. 83-1), can be an early manifestation of elevated pulmonary venous pressure and otherwise unsuspected heart failure. Fatigue and weakness are common accompaniments of advanced cardiac disease and reflect an inability to perform normal activities. A variety of approaches have been used to classify the severity of cardiac limitations, ranging from class I (little or no limitation)

to class IV (severe limitation) (Table 51-5). Hemoptysis (Chapter 83) is a classic presenting finding in patients with pulmonary embolism, but it is also common in patients with mitral stenosis, pulmonary edema, pulmonary infections, and malignant neoplasms (see Table 83-6). Claudication, which is pain in the extremities with exertion, should alert the physician to possible peripheral arterial disease (Chapters 79 and 80).

Complete Medical History The complete medical history should include a thorough review of systems, family history, social history, and past medical history (Chapter 15). The

252

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

review of systems may reveal other symptoms that suggest a systemic disease as the cause of any cardiovascular problems. The family history should focus on premature atherosclerosis or evidence of familial abnormalities, such as may be found with various causes of the long QT syndrome (Chapter 65) or hypertrophic cardiomyopathy (Chapter 60). The social history should include specific questioning about cigarette smoking, alcohol intake, and use of illicit drugs. The past medical history may reveal prior conditions or medications that suggest systemic diseases, ranging from chronic obstructive pulmonary disease, which may explain a complaint of dyspnea, to hemochromatosis, which may be a cause of restrictive cardiomyopathy. A careful history to inquire about recent dental work or other procedures is crucial if bacterial endocarditis is part of the differential diagnosis.

PHYSICAL EXAMINATION FOR DETECTION OF SIGNS OF CARDIOVASCULAR DISEASE

Jugular venous distention Sternal angle

45°

FIGURE 51-2.  Jugular venous distention is defined by engorgement of the internal jugular vein more than 5 cm above the sternal angle at 45 degrees. The central venous pressure is the observed venous distention above the sternal angle plus 5 cm.

The cardiovascular physical examination, which is a subset of the complete physical examination, provides important clues to the diagnosis of asymptomatic and symptomatic cardiac disease and may reveal cardiovascular manifestations of noncardiovascular diseases. The cardiovascular physical examination begins with careful measurement of the pulse and blood pressure (Chapter 8). If aortic dissection (Chapter 78) is a consideration, blood pressure should be measured in both arms and, preferably, in at least one leg. When coarctation of the aorta is suspected (Chapter 69), blood pressure must be measured in at least one leg and in the arms. Discrepancies in blood pressure between the two arms also can be caused by atherosclerotic disease of the great vessels. Pulsus paradoxus, which is more than the usual 10-mm Hg drop in systolic blood pressure during inspiration, is typical of pericardial tamponade (Chapter 77).

General Appearance The respiratory rate may be increased in patients with heart failure. Patients with pulmonary edema are usually markedly tachypneic and may have labored breathing. Patients with advanced heart failure may have CheyneStokes respirations. Systemic diseases, such as hyperthyroidism (Chapter 226), hypothyroidism (Chapter 226), rheumatoid arthritis (Chapter 264), scleroderma (Chapter 267), and hemochromatosis (Chapter 212), may be suspected from the patient’s general appearance. Marfan syndrome (Chapter 260), Turner syndrome (Chapter 235), Down syndrome (Chapter 41), and a variety of congenital anomalies also may be readily apparent.

FIGURE 51-3. Typical distention of the internal jugular vein. (From http:// courses.cvcc.vccs.edu/WisemanD/jugular_vein_distention.htm.)

A ECG S1

Ophthalmologic Examination Examination of the fundi may show diabetic (see Fig. 423-24) or hypertensive retinopathy (see Fig. 67-8) or Roth spots (see Fig. 423-28) typical of infectious endocarditis. Beading of the retinal arteries is typical of severe hypercholesterolemia. Osteogenesis imperfecta, which is associated with blue sclerae, also is associated with aortic dilation and mitral valve prolapse. Retinal artery occlusion (see Fig. 423-29) may be caused by an embolus from clot in the left atrium or left ventricle, a left atrial myxoma, or atherosclerotic debris from the great vessels. Hyperthyroidism may present with exophthalmos and typical stare (see Fig. 423-6), whereas myotonic dystrophy, which is associated with atrioventricular block and arrhythmia, often is associated with ptosis and an expressionless face (see Fig. 421-2).

S2

Phono LSB

A

C

V

Jugular Veins The external jugular veins help in assessment of mean right atrial pressure, which normally varies between 5 and 10 cm H2O; the height (in centimeters) of the central venous pressure is measured by adding 5 cm to the height of the observed jugular venous distention above the sternal angle of Louis (Fig. 51-2). The normal jugular venous pulse, best seen in the internal jugular vein (and not seen in the external jugular vein unless insufficiency of the jugular venous valves is present), includes an a wave, caused by right atrial contraction; a c wave, reflecting carotid artery pulsation; an x descent; a v wave, which corresponds to isovolumetric right ventricular contraction and is more marked in the presence of tricuspid insufficiency; and a y descent, which occurs as the tricuspid valve opens and ventricular filling begins (Fig. 51-3). Abnormalities of the jugular venous pressure (Fig. 51-4) are useful in detecting heart failure, and they correlate well with brain natriuretic peptide levels (Chapter 58) and echocardiographic evidence of an elevated pulmonary artery pressure (Chapter 55).1 The jugular venous pressure also helps in the diagnosis of pericardial disease, tricuspid valve disease, and pulmonary hypertension (Table 51-6).

Y JUG X

0.1 sec

FIGURE 51-4.  Normal jugular venous pulse. ECG = electrocardiogram; JUG = jugular vein; LSB = left sternal border; phono = phonocardiogram; S1 = first heart sound; S2 = second heart sound.

Carotid Pulse The carotid pulse should be examined in terms of its volume and contour. The carotid pulse (Fig. 51-5) may be increased in frequency and may be more intense than normal in patients with a higher stroke volume secondary to aortic regurgitation, arteriovenous fistula, hyperthyroidism, fever, or anemia. In aortic regurgitation or arteriovenous fistula, the pulse may have a bisferious quality. The carotid upstroke is delayed in patients with valvular aortic

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

stenosis (Chapter 75) and has a normal contour but diminished amplitude in any cause of reduced stroke volume.

Cardiac Inspection and Palpation Inspection of the precordium may reveal the hyperinflation of obstructive lung disease or unilateral asymmetry of the left side of the chest because of right ventricular hypertrophy before puberty. Palpation may be performed with the patient either supine or in the left lateral decubitus position; the latter position moves the left ventricular apex closer to the chest wall and increases the ability to palpate the point of maximal impulse and other

Positive hepatojugular reflux

Suspect heart failure, particularly left ventricular systolic dysfunction (echocardiography recommended)

Elevated systemic venous pressure without obvious x or y descent, quiet precordium, and pulsus paradoxus

Suspect cardiac tamponade (echocardiography recommended)

Elevated systemic venous pressure with sharp y descent, Kussmaul sign, and quiet precordium

Suspect constrictive pericarditis (cardiac catheterization and MRI or CT recommended)

Elevated systemic venous pressure with a sharp brief y descent, Kussmaul sign, and evidence of pulmonary hypertension and tricuspid regurgitation

Suspect restrictive cardiomyopathy (cardiac catheterization and MRI or CT recommended)

A prominent a wave with or without elevation of mean systemic venous pressure

Exclude tricuspid stenosis, right ventricular hypertrophy caused by pulmonary stenosis, and pulmonary hypertension (echo-Doppler study recommended)

A prominent v wave with a sharp y descent

Suspect tricuspid regurgitation (echo-Doppler or cardiac catheterization to determine etiology)

S4

S1

S4

P A2 2

S1

S1

S4

P A2 2

B

S4

S1

P A2 2

Dicrotic notch

C

S4

P A2 2

S1

P A2 2

Dicrotic notch

Dicrotic notch

D

The first heart sound (Fig. 51-6), which is largely produced by closure of the mitral and—to a lesser extent—the tricuspid valves, may be louder in patients with mitral valve stenosis and intact valve leaflet movement and less audible in patients with poor closure caused by mitral regurgitation (Chapter 75). The second heart sound is caused primarily by closure of the aortic valve, but closure of the pulmonic valve is also commonly audible. In normal individuals, the louder aortic closure sound occurs first, followed by pulmonic closure. With expiration, the two sounds are virtually superimposed. With inspiration, by comparison, the increased stroke volume of the right ventricle commonly leads to a discernible splitting of the second sound. This splitting may be fixed in patients with an atrial septal defect (Chapter 69) or a right bundle branch block. The split may be paradoxical in patients with left bundle branch block or other causes of delayed left ventricular emptying. The aortic component of the second sound is increased in intensity in the presence of systemic hypertension and decreased in intensity in patients with aortic stenosis. The pulmonic second sound is increased in the presence of pulmonary hypertension. Early systolic ejection sounds are related to forceful opening of the aortic or pulmonic valve. These sounds are common in congenital aortic stenosis, with a mobile valve; in hypertension, with forceful opening of the aortic valve; and in healthy young individuals, especially when cardiac output is increased. Midsystolic or late systolic clicks are caused most commonly by mitral valve prolapse (Chapter 75). Clicks are relatively high-frequency sounds that are heard best with the diaphragm of the stethoscope. An S3 corresponds to rapid ventricular filling during early diastole. It may occur in normal children and young adults, especially if stroke volume is

Dicrotic notch

Dicrotic notch

A

phenomena. Low-frequency phenomena, such as systolic heaves or lifts from the left ventricle (at the cardiac apex) or right ventricle (parasternal in the third or fourth intercostal space), are felt best with the heel of the palm. With the patient in the left lateral decubitus position, this technique also may allow palpation of an S3 gallop in cases of advanced heart failure or an S4 gallop in cases of poor left ventricular distensibility during diastole. The left ventricular apex is more diffuse and sometimes may be frankly dyskinetic in patients with advanced heart disease. The distal palm is best for feeling thrills, which are the tactile equivalent of cardiac murmurs. By definition, a thrill denotes a murmur of grade 4/6 or louder. Higher-frequency events may be felt best with the fingertips; examples include the opening snap of mitral stenosis or the loud pulmonic second sound of pulmonary hypertension.

Auscultation

TABLE 51-6  ABNORMALITIES OF VENOUS PRESSURE AND PULSE AND THEIR CLINICAL SIGNIFICANCE

CT = computed tomography; MRI = magnetic resonance imaging. From Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular Medicine. 5th ed. Philadelphia: WB Saunders; 1997.

253

E

FIGURE 51-5.  Schematic diagrams of the configurational changes in the carotid pulse and their differential diagnosis. Heart sounds also are illustrated. A, Normal. B, Anacrotic pulse with slow initial upstroke. The peak is close to the second heart sound. These features suggest fixed left ventricular outflow obstruction, such as valvular aortic stenosis.  C, Pulsus bisferiens, with percussion and tidal waves occurring during systole. This type of carotid pulse contour is observed most frequently in patients with hemodynamically significant aortic regurgitation or combined aortic stenosis and regurgitation with dominant regurgitation. It rarely is observed in patients with mitral valve prolapse or in normal individuals. D, Pulsus bisferiens in hypertrophic obstructive cardiomyopathy. This finding rarely is appreciated at the bedside by palpation. E, Dicrotic pulse results from an accentuated dicrotic wave and tends to occur in sepsis, severe heart failure, hypovolemic shock, and cardiac tamponade and after aortic valve replacement. A2 = aortic component of the second heart sound; P2 = pulmonary component of the second heart sound); S1 = first heart sound; S4 = atrial sounds. (From Chatterjee K. Bedside evaluation of the heart: the physical examination. In: Chatterjee K, Chetlin MD, Karliner J, et al, eds. Cardiology: An Illustrated Text/Reference. Philadelphia: JB Lippincott; 1991:3.11-3.51.)

254

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

S4

Atrial or presystolic gallop (S4)

A M1 T1 Split first heart sound

B EC Aortic or pulmonary systolic ejection click (EC)

C A2 P2 Split second heart sound

D OS Opening snap of mitral stenosis (OS)

E S3 Third heart sound (S3)

supravalvular, or infravalvular aortic stenosis and pulmonic stenosis. The murmur of hypertrophic obstructive cardiomyopathy has a similar ejection quality, although its peak may be later in systole when dynamic obstruction is maximal (Chapter 60). Pansystolic murmurs are characteristic of mitral or tricuspid regurgitation or with a left-to-right shunt from conditions such as a ventricular septal defect (left ventricle to right ventricle). A late systolic murmur is characteristic of mitral valve prolapse (Chapter 75) or ischemic papillary muscle dysfunction. Ejection quality murmurs also may be heard in patients with normal valves but increased flow, such as occurs with marked anemia, fever, or bradycardia secondary to congenital complete heart block; they also may be heard across a valve that is downstream from increased flow because of an intracardiac shunt. Maneuvers such as inspiration, expiration, standing, squatting, and hand gripping can be especially useful in the differential diagnosis of a murmur; however, echocardiography commonly is required to make a definitive diagnosis of cause and severity (Table 51-8). High-frequency, early diastolic murmurs are typical of aortic regurgitation and pulmonic regurgitation from a variety of causes. The murmurs of mitral and tricuspid stenosis begin in early to mid diastole and tend to diminish in intensity later in diastole in the absence of effective atrial contraction, but they tend to increase in intensity in later diastole if effective atrial contraction is present. Continuous murmurs may be caused by any abnormality that is associated with a pressure gradient in systole and diastole. Examples include a patent ductus arteriosus, ruptured sinus of Valsalva aneurysm, arteriovenous fistula (of the coronary artery, pulmonary artery, or thoracic artery), and a mammary soufflé. In some situations, murmurs of two coexistent conditions (e.g., aortic stenosis and regurgitation, atrial septal defect with a large shunt and resulting flow murmurs of relative mitral and pulmonic stenosis) may mimic a continuous murmur. Unfortunately, the physical examination is limited for detecting meaningful valvular heart disease.2 As a result, echocardiography (Chapter 55) is critical to the evaluation of patients with suspected structural heart disease.

Abdomen

F SC S1

S2 Midsystolic click (SC)

G FIGURE 51-6. Timing of the different heart sounds and added sounds. (Modified from Wood P. Diseases of the Heart and Circulation. 3rd ed. Philadelphia: JB Lippincott; 1968.)

increased. After about 40 years of age, however, an S3 should be considered abnormal; it is caused by conditions that increase the volume of ventricular filling during early diastole (e.g., mitral regurgitation) or that increase pressure in early diastole (e.g., advanced heart failure). A left ventricular S3 gallop is heard best at the apex, whereas the right ventricular S3 gallop is heard best at the fourth intercostal space at the left parasternal border; both are heard best with the bell of the stethoscope. An S4 is heard rarely in young individuals but is common in adults older than 40 or 50 years because of reduced ventricular compliance during atrial contraction; it is a nearly ubiquitous finding in patients with hypertension, heart failure, or ischemic heart disease. The opening snap of mitral and, less commonly, tricuspid stenosis (Chapter 75) occurs at the beginning of mechanical diastole, before the onset of the rapid phase of ventricular filling. An opening snap is high pitched and is heard best with the diaphragm; this differential frequency should help distinguish an opening snap from an S3 on physical examination. An opening snap commonly can be distinguished from a loud pulmonic component of the second heart sound by the differential location (mitral opening snap at the apex, tricuspid opening snap at the left third or fourth intercostal space, pulmonic second sound at the left second intercostal space) and by the longer interval between S2 and the opening snap. Heart murmurs may be classified as systolic, diastolic, or continuous (Table 51-7). Murmurs are graded by intensity on a scale of 1 to 6. Grade 1 is faint and appreciated only by careful auscultation; grade 2, readily audible; grade 3, moderately loud; grade 4, loud and associated with a palpable thrill; grade 5, loud and audible with the stethoscope only partially placed on the chest; and grade 6, loud enough to be heard without the stethoscope on the chest. Systolic ejection murmurs usually peak in early to mid systole when left ventricular ejection is maximal; examples include fixed valvular,

The most common cause of hepatomegaly in patients with heart disease is hepatic engorgement from elevated right-sided pressures associated with right ventricular failure of any cause. Hepatojugular reflux is elicited by pressing on the liver and showing an increase in the jugular venous pressure; it indicates advanced right ventricular failure or obstruction to right ventricular filling. Evaluation of the abdomen also may reveal an enlarged liver caused by a systemic disease, such as hemochromatosis (Chapter 212) or sarcoidosis (Chapter 95), which also may affect the heart. In more severe cases, splenomegaly and ascites also may be noted. Large, palpable, polycystic kidneys (Chapter 127) commonly are associated with hypertension. A systolic bruit suggestive of renal artery stenosis (Chapter 125) or an enlarged abdominal aorta (Chapter 78) is a clue of atherosclerosis.

Extremities Extremities should be evaluated for peripheral pulses, edema, cyanosis, and clubbing. Diminished peripheral pulses suggest peripheral arterial disease (Chapters 79 and 80). Delayed pulses in the legs are consistent with coarctation of the aorta and are seen after aortic dissection. Edema (Fig. 51-7) is a cardinal manifestation of right-sided heart failure.3 When it is caused by heart failure, pericardial disease, or pulmonary hypertension, the edema is usually symmetrical and progresses upward from the ankles; each of these causes of cardiac edema commonly is associated with jugular venous distention and often with hepatic congestion. Unilateral edema suggests thrombophlebitis or proximal venous or lymphatic obstruction (Fig. 51-8). Edema in the absence of evidence of right-sided or left-sided heart failure suggests renal disease, hypoalbuminemia, myxedema, or other noncardiac causes. Among unselected patients with bilateral edema, about 40% have an underlying cardiac disease, about 40% have an elevated pulmonary blood pressure, about 20% have bilateral venous disease, about 20% have renal disease, and about 25% have idiopathic edema. Cyanosis (Fig. 51-9) is a bluish discoloration caused by reduced hemoglobin exceeding about 5 g/dL in the capillary bed. Central cyanosis is seen in patients with poor oxygen saturation resulting from a reduced inspired oxygen concentration or inability to oxygenate the blood in the lungs (e.g., as a result of advanced pulmonary disease, pulmonary edema, pulmonary arteriovenous fistula, or right-to-left shunting); it also may be seen in patients with marked erythrocytosis. Methemoglobinemia (Chapter 158) also can present with cyanosis. Peripheral cyanosis may be caused by reduced blood flow to the extremities secondary to vasoconstriction, heart failure, or shock.

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

255

TABLE 51-7  SOME COMMON CAUSES OF HEART MURMURS* USUAL LOCATION

COMMON ASSOCIATED FINDINGS

SYSTOLIC Holosystolic   Mitral regurgitation   Tricuspid regurgitation   Ventricular septal defect

Apex → axilla LLSB LLSB → RLSB

Early to mid systolic   Aortic valvular stenosis    Fixed supravalvular or subvalvular    Dynamic infravalvular

↑ with handgrip; S3 if marked mitral regurgitation; left ventricular dilation common ↑ with inspiration; right ventricular dilation common Often with thrill

RUSB RUSB

  Pulmonic valvular stenosis    Infravalvular (infundibular)    Supravalvular   “Flow murmurs”

LUSB LUSB LUSB LUSB

Ejection click if mobile valve; soft or absent A2 if valve immobile; later peak associated with more severe stenosis Hypertrophic obstructive cardiomyopathy; murmur louder if left ventricular volume lower or contractility increased, softer if left ventricular volume increased†; can be later in systole if obstruction delayed ↑ with inspiration ↑ with inspiration ↑ with inspiration Anemia, fever, increased flow of any cause‡

Mid to late systolic   Mitral valve prolapse   Papillary muscle dysfunction

LLSB or apex → axilla Apex → axilla

Preceded by click; murmur lengthens with maneuvers that decrease left ventricular volume† Ischemic heart disease

Early diastolic   Aortic regurgitation

RUSB, LUSB

High-pitched, blowing quality; endocarditis, diseases of the aorta, associated aortic valvular stenosis; signs of low peripheral vascular resistance Pulmonary hypertension as a causative factor

LLSB → apex + axilla

DIASTOLIC

  Pulmonic valve regurgitation

LUSB

Mid to late diastolic   Mitral stenosis, tricuspid stenosis

Apex, LLSB

  Atrial myxomas

Apex (L), LLSB (R)

Continuous   Venous hum   Patent ductus arteriosus   Arteriovenous fistula    Coronary    Pulmonary, bronchial, chest wall   Ruptured sinus of Valsalva aneurysm

Low pitched; in rheumatic heart disease, opening snap commonly precedes murmur; can be caused by increased flow across normal valve‡ “Tumor plop”

Over jugular or hepatic vein or breast LUSB LUSB Over fistula RUSB

Sudden onset

*See also Chapters 69 and 75. † Left ventricular volume is decreased by standing or during prolonged, forced expiration against a closed glottis (Valsalva maneuver); it is increased by squatting or by elevation of the legs; contractility is increased by adrenergic stimulation or in the beat after an extrasystolic beat. ‡ Including a left-to-right shunt through an atrial septal defect for tricuspid or pulmonic flow murmurs, and a ventricular septal defect for pulmonic or mitral flow murmurs. LLSB = left lower sternal border (fourth intercostal space); LUSB = left upper sternal border (second and third intercostal spaces); RLSB = right lower sternal border (fourth intercostal space); RUSB = right upper sternal border (second and third intercostal spaces).

TABLE 51-8  SENSITIVITY AND SPECIFICITY OF BEDSIDE MANEUVERS IN THE IDENTIFICATION OF SYSTOLIC MURMURS MANEUVER

RESPONSE

MURMUR

SENSITIVITY (%)

SPECIFICITY (%)

Inspiration

↑

RS

100

88

Expiration

↓

RS

100

88

Valsalva maneuver

↑

HC

65

96

Squat to stand

↑

HC

95

84

Stand to squat

↓

HC

95

85

Leg elevation

↓

HC

85

91

Handgrip

↓

HC

85

75

Handgrip

↑

MR and VSD

68

92

Transient arterial occlusion

↑

MR and VSD

78

100

HC = hypertrophic cardiomyopathy; MR = mitral regurgitation; RS = right sided; VSD = ventricular septal defect. Modified with permission from Lembo NJ, Dell’Italia IJ, Crawford MH, et al. Bedside diagnosis of systolic murmurs. N Engl J Med. 1988;318:1572-1578. Copyright 1988 Massachusetts Medical Society. All rights reserved.

Clubbing (Fig. 51-10), which is loss of the normal concave configuration of the nail as it emerges from the distal phalanx, is seen in patients with pulmonary abnormalities such as lung cancer (Chapter 191) and in patients with cyanotic congenital heart disease (Chapter 69).4

Examination of the Skin Examination of the skin may reveal bronze pigmentation typical of hemochromatosis (Chapter 212); jaundice (see Fig. 146-1) characteristic of severe right-sided heart failure or hemochromatosis; or capillary hemangiomas typical of Osler-Weber-Rendu disease (see Fig. 173-1), which also is associated with pulmonary arteriovenous fistulas and cyanosis. Infectious endocarditis may be associated with Osler nodes (see Fig. 76-2), Janeway lesions, or splinter hemorrhages (Fig. 51-11) (Chapter 76). Xanthomas (Fig. 51-12) are subcutaneous deposits of cholesterol seen on the extensor surfaces of the extremities or on the palms and digital creases; they are found in patients with severe hypercholesterolemia.

Laboratory Studies All patients with known or suspected cardiac disease should have an ECG and chest radiograph. The ECG (Chapter 54) helps identify rate, rhythm, conduction abnormalities, and possible myocardial ischemia. The chest radiograph (Chapter 56) yields important information on chamber enlargement, pulmonary vasculature, and the great vessels. Blood testing in patients with known or suspected cardiac disease should be targeted to the conditions in question. In general, a complete blood cell count, thyroid indices, and lipid levels are part of the standard evaluation. Point-of-care biomarker measurements in the emergency department can

256

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

decrease unnecessary admissions and reduce median length-of-stay. For example, among patients who are being evaluated for an acute MI, an undetectable high-sensitivity troponin level at presentation reduces the probability of acute MI to less than 1%.5 A protocol in which the ECG and troponin level are repeated in 2 hours is as good as longer observation periods for

evaluating patients with acute chest pain and suspected MI. A1  However, the advent of high-sensitivity troponin assays has also greatly increased the risk for a false-positive diagnosis of MI,5 especially because of chronic troponin elevations in many cardiac conditions and in elderly patients (Chapter 72).6 Echocardiography (Chapter 55) is the most useful test to analyze valvular and ventricular function. By use of Doppler flow methods, stenotic and regurgitant lesions can be quantified. Hand-held ultrasonography performed by generalists can improve the assessment of left ventricular function, cardiomegaly, and pericardial effusion. Transesophageal echocardiography is the preferred method to evaluate possible aortic dissection and to identify clot in the cardiac chambers. Radionuclide studies (Chapter 56) can measure left ventricular function, assess myocardial ischemia, and determine whether ischemic myocardium is viable. CT can detect coronary calcium, which is a risk factor for symptomatic coronary disease (Chapter 56). In the setting

B

A

FIGURE 51-7.  Pitting edema in a patient with cardiac failure. A depression (“pit”) remains in the edema for some minutes after firm fingertip pressure is applied. (From Forbes CD, Jackson WD. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.)

FIGURE 51-9.  Arterial embolism causing acute ischemia and cyanosis of the leg. Initial pallor of the leg and foot was followed by cyanosis. (From Forbes CD, Jackson WD. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.)

Unilateral or bilateral? Bilateral

Unilateral

Detailed history Physical exam

R/O DVT

Yes

No

Anticoagulation

Pain?

Urine dipstick

Yes

No

Fever or increased WBC?

Postphlebitic syndrome?

−

+

Obvious findings of CHF?

R/O concurrent cardiac and hepatic disease

Yes

No

Yes

No

Yes

No

Cellulitis or other infection?

Characteristic physical signs of popliteal cyst or gastrocnemius rupture

Continue anticoagulation

R/O malignancy Detailed history Pelvic exam Rectal exam

Initiate appropriate therapy

Creatinine Electrolytes Albumin Cholesterol Prothrombin time Liver enzymes TSH Chest x-ray Cardiac echo

Antibiotic treatment

Yes

No

Initiate symptomatic therapy

Consider MRI

Pursue further diagnostic work-up as appropriate

Renal disease

Occult CHF

Cirrhosis

Consider renal biopsy

Initiate appropriate therapy

Hypothyroidism

Other or idiopathic

Follow-up abnormalities Initiate appropriate therapy FIGURE 51-8.  Diagnostic approach to patients with edema. CHF = congestive heart failure; DVT = deep vein thrombosis; MRI = magnetic resonance imaging; R/O = rule out; TSH = thyroid-stimulating hormone; WBC = white blood cell count. (From Chertow G. Approach to the patient with edema. In: Braunwald E, Goldman L, eds. Primary Cardiology. 2nd ed. Philadelphia: WB Saunders; 2003.)

FIGURE 51-10.  Severe finger clubbing in a patient with cyanotic congenital heart disease. (From Forbes CD, Jackson WD. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.)

on echocardiography.7,8 These tests are often crucial in diagnosis of possible myocardial ischemia (Chapter 71) and in establishment of prognosis in patients with known ischemic heart disease. However, they are not recommended for the screening of asymptomatic individuals9 or prior to participation in sports.10 Cardiac catheterization (Chapter 57) can measure precise gradients across stenotic cardiac valves, judge the severity of intracardiac shunts, and determine intracardiac pressures. Coronary angiography provides a definitive diagnosis of coronary disease and is a necessary prelude to coronary revascularization with a percutaneous coronary intervention or coronary artery bypass graft surgery (Chapter 74). Continuous ambulatory ECG monitoring can help diagnose arrhythmias. A variety of newer technologies allow longer-term monitoring in patients with important but infrequently occurring symptoms (Chapter 62). Formal invasive electrophysiologic testing can be useful in the diagnosis of ventricular or supraventricular wide-complex tachycardia, and it is crucial for guiding a wide array of new invasive electrophysiologic therapies (Chapter 66).

SUMMARY

The history, physical examination, and laboratory evaluation should help the physician establish the cause of any cardiovascular problem; identify and quantify any anatomic abnormalities; determine the physiologic status of the valves, myocardium, and conduction system; determine functional capacity; estimate prognosis; and provide primary or secondary prevention. Key preventive strategies, including diet modification, recognition and treatment of hyperlipidemia, cessation of cigarette smoking, and adequate physical exercise, should be part of the approach to every patient, with or without heart disease.

Grade A References

FIGURE 51-11.  Splinter hemorrhage (solid arrow) and Janeway lesions (open arrow). These findings should stimulate a work-up for endocarditis. (Courtesy of Daniel L.  Stulberg, MD.)

A1. Than M, Aldous S, Lord SJ, et al. A 2-hour diagnostic protocol for possible cardiac chest pain in the emergency department: a randomized clinical trial. JAMA Intern Med. 2014;174:51-58. A2. Goodacre SW, Bradburn M, Cross E, et al. The randomised Assessment of Treatment using Panel Assay of Cardiac Markers (RATPAC) trial: a randomised controlled trial of point-of-care cardiac markers in the emergency department. Heart. 2011;97:190-196. A3. Litt HI, Gatsonis C, Snyder B, et al. CT angiography for safe discharge of patients with possible acute coronary syndromes. N Engl J Med. 2012;366:1393-1403. A4. Hoffmann U, Truong QA, Schoenfeld DA, et al. Coronary CT angiography versus standard evaluation in acute chest pain. N Engl J Med. 2012;367:299-308.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

FIGURE 51-12.  Eruptive xanthomas of the extensor surfaces of the lower extremities. This patient had marked hypertriglyceridemia. (From Massengale WT, Nesbitt LT Jr. Xanthomas. In: Bolognia JL, Jorizzo JL, Rapini RP, eds. Dermatology. Philadelphia: Mosby; 2003:1449.)

of acute chest pain, multislice CT is effective in diagnosing coronary disease. A2  In a randomized trial of emergency department patients at low to intermediate risk for a possible acute coronary syndrome, coronary CT angiography resulted in a higher rate of discharge from the emergency department (50% vs. 23), a shorter length of stay (median, 18 vs. 24.8 hours), and a higher rate of detection of coronary disease (9% vs. 3.5%) without any change in the rate of serious adverse events. A3  However, in a subsequent randomized trial of emergency department patients with symptoms suggestive of acute coronary syndromes but without ischemic ECG changes or an initially positive troponin test, incorporating coronary CT angiography into the triage strategy did not decrease overall costs of care. A4  Stress testing by exercise or pharmacologic stress is useful to precipitate myocardial ischemia that may be detected by ECG abnormalities, perfusion abnormalities on radionuclide studies, or transient wall motion abnormalities

CHAPTER 51  Approach to the Patient with Possible Cardiovascular Disease  

GENERAL REFERENCES 1. Pellicori P, Kallvikbacka-Bennett A, Zhang J, et al. Revisiting a classical clinical sign: jugular venous ultrasound. Int J Cardiol. 2014;170:364-370. 2. Roberts KV, Brown AD, Maguire GP, et al. Utility of auscultatory screening for detecting rheumatic heart disease in high-risk children in Australia’s Northern Territory. Med J Aust. 2013;199: 196-199. 3. Clark AL, Cleland JG. Causes and treatment of oedema in patients with heart failure. Nat Rev Cardiol. 2013;10:156-170. 4. Rutherford JD. Digital clubbing. Circulation. 2013;127:1997-1999. 5. Storrow AB, Christenson RH, Nowak RM, et al. Diagnostic performance of cardiac troponin I for early rule-in and rule-out of acute myocardial infarction: Results of a prospective multicenter trial. Clin Biochem. 2014; [Epub ahead of print].

257.e1

6. Korley FK, Jaffe AS. Preparing the United States for high-sensitivity cardiac troponin assays. J Am Coll Cardiol. 2013;61:1753-1758. 7. Mancini GB, Gosselin G, Chow B, et al. Canadian Cardiovascular Society guidelines for the diagnosis and management of stable ischemic heart disease. Can J Cardiol. 2014;30:837-849. 8. Mieres JH, Gulati M, Bairey Merz N, et al. Role of noninvasive testing in the clinical evaluation of women with suspected ischemic heart disease: a consensus statement from the American Heart Association. Circulation. 2014;130:350-379. 9. Chou R, Arora B, Dana T, et al. Screening asymptomatic adults with resting or exercise electrocardiography: a review of the evidence for the U.S. Preventive services task force. Ann Intern Med. 2011;155:375-385. 10. Sharma S, Estes NA 3rd, Vetter VL, et al. Clinical decisions: cardiac screening before participation in sports. N Engl J Med. 2013;369:2049-2053.

CHAPTER 52  Epidemiology of Cardiovascular Disease  

257

52  EPIDEMIOLOGY OF CARDIOVASCULAR DISEASE DONALD M. LLOYD-JONES Cardiovascular diseases are the leading cause of death, disability, and medical costs in the world, and they are expected to remain so for the foreseeable future. Cardiovascular disease manifests in a number of different ways, including congenital heart and vascular malformations (Chapter 69); coronary heart disease (Chapters 70, 71, 72, 73, and 74); heart failure (Chapter 59); cardiomyopathies (Chapter 60); valvular heart disease (Chapter 75); dysrhythmias (Chapters 62, 63, 64, and 65); pericardial diseases (Chapter 77); aortic (Chapter 78), peripheral (Chapter 79), and cerebrovascular diseases (Chapter 406); systemic hypertension (Chapter 67); vasculitides (Chapter 270); venous thromboembolic disease (Chapter 81); and pulmonary vascular hypertension (Chapter 68). Of these, coronary heart disease, stroke, and heart failure, which share many common underlying risk factors, have by far the largest impact on the population in terms of incidence, prevalence, quality of life, and medical costs.

258

CHAPTER 52  Epidemiology of Cardiovascular Disease  

BURDEN IN THE UNITED STATES

Cardiovascular diseases have been the leading cause of death in the United States in every year of the 20th and 21st century except for 1918, when the influenza epidemic surpassed them. Cardiovascular diseases account for 1 in 3 deaths in America annually, or about 790,000 deaths, including about 400,000 in women and about 390,000 in men.1 The overall rate of death due to cardiovascular disease in the United States is about 230 per 100,000 persons, with higher rates in men than in women and in blacks than in whites. Because of secular trends over the past 40 to 50 years, coronary heart disease alone may soon fall below all cancers combined, but all cardiovascular diseases combined are expected to remain the leading causes of death in the United States and globally for the foreseeable future. Cardiovascular diseases also are the leading cause of hospitalizations and medical costs in the United States. Each year, about 5.8 million Americans are hospitalized for cardiovascular disease, more than 1.3 million cases of which are due to coronary heart disease and another 1 million or more due to heart failure. The United States currently spends more than $300 billion annually on direct and indirect costs for cardiovascular diseases, and these total costs are projected roughly to triple to more than $1 trillion annually by 2030. In the United States, about 15.4 million adults have coronary heart disease, roughly half of whom have had a myocardial infarction. Each year, Americans suffer more than 900,000 new and recurrent myocardial infarctions, with about 380,000 deaths due to coronary heart disease, a large percentage of which are sudden cardiac deaths. There are about 6.8 million stroke survivors in the United States, with 800,000 new or recurrent strokes occurring every year. Strokes are especially prominent in the so-called “stroke belt” in the southeastern United States, where many African Americans live. With aging, the risks for stroke and heart failure tend to increase earlier in women and African Americans than in white men, whose coronary risk increases earlier. At present, more than 5 million Americans suffer from chronic heart failure, with approximately equal numbers of men and women affected. However, the prevalence of heart failure is about twice as high in blacks as in whites.

GLOBAL BURDEN

Cardiovascular diseases, including coronary heart disease and stroke, became the leading cause of death and disability globally in the early 21st century.2 About 80% of cardiovascular deaths and events now occur in low- and middle-income countries, and the onset of cardiovascular disease tends to be at an earlier age in these countries. For example, about 50% of coronary deaths occur before age 70 years in India, whereas only 25% occur by that age in high-income countries. Unfavorable global trends in eating patterns, high rates of smoking, and increasing burdens of obesity, diabetes, and hypertension are driving the burden of cardiovascular disease.3 Whereas stroke was the dominant cause of death and disability in East Asian countries for decades owing to high sodium intake and resulting hypertension, recent changes in diet, activity levels, and smoking have made coronary heart disease an equivalent or greater health burden in this area of the world.

RISK FACTORS FOR CARDIOVASCULAR DISEASE

Established Risk Factors

A number of factors have been established for cardiovascular disease based on their strength and consistency of associations, specificity, temporality, and biologic plausibility.4,5 Furthermore, these established risk factors explain the vast majority of risk for incident myocardial infarction. Longitudinal cohort studies demonstrate that 90% of individuals who suffer a myocardial infarction have at least one established clinical risk factor before their first event, and adverse levels of nine risk factors and behaviors collectively account for 90% or more of the risk for myocardial infarction in men and women, in older and younger individuals, and in all regions of the world. These nine risk factors and behaviors include smoking (Chapter 32), elevated apolipoprotein B−to−apolipoprotein A1 ratio (Chapter 206), hypertension (Chapter 67), diabetes (Chapter 229), abdominal obesity (Chapter 220), psychosocial factors, lower consumption of fruits and vegetables (Chapter 213), alcohol intake (Chapter 33), and physical inactivity (Chapter 16). Many of the established risk factors tend to cluster in a metabolic syndrome, which is characterized by abdominal obesity, insulin resistance, hyperglycemia, elevated blood pressure, elevated triglyceride levels, and lower high-density lipoprotein (HDL) cholesterol levels. Age is the most powerful risk factor for the development of most cardiovascular diseases, especially stroke (Chapter 407), heart failure (Chapters 58 and 59), and atrial fibrillation (Chapter 64). Chronologic age represents a

person’s aggregate exposure to multiple physiologic and environmental effects on the cardiovascular system. The incidence of cardiovascular disease at least doubles with each additional decade of age in adulthood until the oldest ages, when the heavy burden of competing causes of mortality (Chapter 23) limits further progression. The impact of a person’s sex on cardiovascular disease is important. More women than men die of cardiovascular diseases annually. However, women tend to develop risk factors later in life than do men, and women’s incidence rates lag men’s by approximately 10 years. The precise contributions of sex hormones to these age trends are uncertain, but many women develop worsening risk factor levels, particularly with regard to lipids, blood pressure, weight, and insulin resistance, during and after the menopausal transition (Chapter 240). Race per se is not thought to be an independent risk factor for cardiovascular disease, and the established causal risk factors have broadly similar effects in all race and ethnic groups. Nevertheless, hypertension tends to be more prevalent in individuals of African ancestry, especially in environments with higher sodium intake, and to have a somewhat stronger association with cardiovascular events, especially heart failure and stroke. Compared with whites, individuals of East Asian and South Asian descent have a greater risk for developing the metabolic syndrome, insulin resistance, and diabetes at a lower overall body mass index. However, some of the cardiovascular risk differences observed across race and ethnic groups can be attributed to differences in socioeconomic status, rather than race or ethnicity. Blood lipid levels (Chapter 206), including the total serum cholesterol level and its subfractions, particularly low-density lipoprotein (LDL) cholesterol, have significant, continuous, and graded associations with the risk for coronary heart disease and peripheral arterial atherothrombotic disease. By comparison, independent associations of blood lipids with stroke and heart failure events are much weaker, indicating a potentially lesser role in the pathogenesis of these diseases when they occur independently of their relationship to coexisting coronary heart disease. Apolipoprotein B−containing particles make up the subpopulation of circulating cholesterol-containing particles that represent the atherogenic lipoprotein fractions. These particles are considered to be the central actors in the initiation and promotion of atherogenesis on the basis of a substantial body of epidemiologic, clinical, and basic science evidence. Among U.S. adults aged 20 years and older, 43% (or nearly 100 million) have total cholesterol levels above the desirable range of less than 200 mg/dL, and 14% (31 million) have elevated levels of 240 mg/ dL or higher. Mean total cholesterol levels have been falling sharply in recent decades, mostly because of changes in dietary composition but also because of more widespread use of lipid-lowering medications. In the 1970s, mean total cholesterol concentrations were approximately 220 mg/dL, whereas currently they are just under 200 mg/dL. These improvements have been a major contributor to the decline in coronary death rates over the same time period. Randomized clinical trials have unequivocally established LDL cholesterol as a causal agent for coronary heart disease, and statins are effective at reducing rates of both coronary heart disease and stroke, significantly and substantially. A1  By comparison, niacin is of no apparent added value A2  and other medications are being actively investigated (Chapter 206). Blood pressure (Chapter 67) has a continuous, graded association with incident coronary heart disease, stroke, and heart failure events. In worldwide studies of nearly 1 million individuals, the risk at every age for all types of cardiovascular disease death doubled with each 20-mm Hg higher systolic blood pressure and each 10-mm Hg higher diastolic blood pressure, beginning at a blood pressure of 115/75 mm Hg.6 Although the relationship with outcomes is linear, hypertension is typically defined by blood pressures of 140 mm Hg or higher systolic or 90 mm Hg diastolic (Chapter 67). Using this definition, hypertension is the most prevalent modifiable cardiovascular risk factor worldwide. Among people who are normotensive at age 55 years, the remaining lifetime risk for development of hypertension is 90%. Approximately one third of all American adults currently have hypertension, and its prevalence has been increasing owing to the obesity epidemic. Hypertension has stronger relative associations with stroke and heart failure than with coronary heart disease, in part because of its effects on myocardial and cerebrovascular remodeling. In the United States, rates of treatment and control for hypertension have been gradually increasing. The effective treatment of hypertension reduces the risk for stroke, heart failure, and coronary heart disease events. A3  Cigarette smoking (Chapter 32) is one of the strongest risk factors for cardiovascular disease events. After adjustment for other risk factors, smoking confers two- to three-fold higher risk for all manifestations of cardiovascular

CHAPTER 52  Epidemiology of Cardiovascular Disease  

disease, especially coronary heart disease and peripheral arterial disease. Fortunately, consistent public health efforts have reduced the prevalence of smoking in the United States from about 45% in the 1960s to just under 20% currently. The prevalence of smoking remains higher in many European and Asian countries, and its continued increase in some parts of the world drives unfavorable trends in cardiovascular morbidity and mortality. A large body of evidence indicates that environmental exposure to tobacco smoke in nonsmokers (“second-hand” or “passive” smoking) also increases risk for cardiovascular events substantially (Chapter 32) and contributes to the population burden of disease. Substantial data also support the benefits of smoking cessation for reducing the risks for a subsequent coronary event and death.7 Overweight and obesity have been increasing in the United States and worldwide. Before 1985, fewer than 10% of Americans were obese, defined as having a body mass index of 30 kg/m2 or higher. Now, however, about 35% of Americans are obese, and another 35% are overweight (Chapter 220). Major societal changes in the availability of food and in dietary content, coupled with reductions in physical activity, have produced this unprecedented epidemic. Although overweight and obesity themselves tend to be weak independent predictors of cardiovascular events in the short term, they are major drivers of elevated blood pressure, elevated blood glucose levels, and adverse lipid profiles that are themselves major contributors to the incidence of cardiovascular disease.8 Blood glucose and its surrogate marker, hemoglobin A1c, have a continuous and graded association with cardiovascular events. People with diabetes (Chapter 229), whether diagnosed or undiagnosed, have two- to three-fold higher adjusted risk for cardiovascular events compared with persons without diabetes, and they also have substantially higher risks for developing chronic renal disease (Chapter 130). Whereas diabetes was relatively uncommon before the 1980s, the obesity epidemic has led to a dramatic increase in the prevalence of type 2 diabetes and of impaired fasting glucose levels, termed pre-diabetes. At present in the United States, nearly 20 million people, representing more than 8% of all adults, have diagnosed diabetes, and another 8 million (about 3.5% of adults) have undiagnosed diabetes. Fully 87 million more adults, or about 38% of the adult U.S. population, currently have prediabetes. If current trends continue, an estimated 77% of men and 53% of women in the U.S. could have pre-diabetes by 2020. Diabetes affects nonwhite racial and ethnic groups, such as American Indians, African Americans, South Asians, East Asians, and Latinos, who appear to have greater sensitivity to insulin resistance at lower body mass index, in much greater proportions than whites. Unfortunately, tight control of glucose levels in persons with diabetes has not been associated with significant reductions in risk for macrovascular cardiovascular disease. A4  A5  Adverse diet (Chapter 213) is a major contributor to obesity, diabetes, hypertension, and hyperlipidemia. Healthy eating patterns emphasize a lower caloric intake and focus on fruits and vegetables, healthy fats from nuts and olive oil, lean sources of protein such as fish, whole grains, a reduced sodium intake, and limiting the intake of processed foods, unhealthy fats, and simple sugars. This eating pattern is typical of the “Mediterranean diet,” which has been shown to be associated with a lower incidence of cardiovascular disease. A6  By comparison, no vitamin or mineral supplement has been shown conclusively to reduce cardiovascular risk.9 Alcohol (Chapter 33) has a complex association with cardiovascular events. Moderate intake of one serving of alcohol per day is associated with a modestly lower risk for cardiovascular disease. At higher levels of intake, however, risks for total mortality, hypertension, stroke, and heart failure tend to increase. Physical inactivity (Chapter 16) and a sedentary lifestyle are also significant risk factors for cardiovascular disease. Individuals who participate in no physical activity are at highest risk for events. The risk is significantly lower for people who participate in even minimal physical activity, and risks decrease further with greater activity levels, particularly to the extent that they contribute to improvement in objective physical fitness. The biology and risks of sedentary time may be more than just the absence of physical activity because sedentary lifestyle, measured best by the hours of time spent in front of a television or computer screen, seems to have an adverse effect independent of time spent doing physical activity. Family history is clearly an important cardiovascular risk factor, independent of other measurable risk factors. However, ideal levels of cardiovascular health do not appear to be genetically programmed nor inexorably compromised as a consequence of aging. Data indicate that the heritability of ideal cardiovascular health is less than 20%, thereby suggesting strong environmental and behavioral influences on this trait. ,

259

Novel Risk Markers

Blood markers of inflammation, thrombosis, and target organ damage also appear to characterize the atherosclerotic process (Chapter 70). Serum biomarkers such as C-reactive protein, fibrinogen, plasminogen activator inhibitor-1, interleukin-6, and lipoprotein-associated phospholipase A2 have significant associations with incident cardiovascular events that are independent of established risk factors.10 However, because of their lack of specificity and their relatively weak independent associations with incident disease, none of these markers has yet proved useful for routine screening or for incorporation into risk assessment algorithms in primary or secondary prevention. To date, none has provided meaningful reclassification of risk in individuals after quantitative assessment using traditional established risk factors. Newer biomarkers that indicate the presence of existing target organ damage, such as high-sensitivity troponin or natriuretic peptide levels, hold promise for screening and targeting of prevention efforts in older, asymptomatic individuals (Chapter 23). Noninvasive cardiac testing and imaging holds the potential for detecting preclinical disease and potentially guiding early intervention. For example, electrocardiographic evidence of left ventricular hypertrophy confers significant excess risk for coronary heart disease over and above the presence of hypertension and other risk factors. High levels of coronary calcification on computed tomography (CT) imaging of the heart (Chapter 56) or greater carotid intima-media thickness measured by B-mode ultrasound of the carotid arteries portends a higher risk for future cardiovascular events. Because these imaging markers detect evidence of the actual underlying diseases of interest (i.e., left ventricular hypertrophy or atherosclerosis), rather than nonspecific risk factors, they are more effective at identifying individuals at high risk for incident clinical events, such as heart failure, stroke, and myocardial infarction. Of the available modalities, CT screening for coronary artery calcification appears to be the best widely available means for detecting individuals at near-term risk. For example, in the Multi-Ethnic Study of Atherosclerosis, asymptomatic individuals with coronary artery calcium scores of more than 100 Agatston units had relative hazards for a coronary event that were 7- to 10-fold higher than in individuals without any coronary calcification, even after adjustment for major established risk factors.11 Coronary calcium scoring also has been shown to be the most effective and reliable means for reclassifying risk after a quantitative risk assessment using established risk factors, with the ability to identify otherwise low-risk individuals who nonetheless will have a cardiovascular event. Although noninvasive screening for cardiovascular disease holds much promise for the future, its precise role remains uncertain at the present time (Chapter 56).

Assessment of Risk for Cardiovascular Disease Estimation of Short-Term Risk

Adverse levels of any single risk factor or risk marker are associated with elevated risk for incident cardiovascular events. However, combinations of adverse risk factors are additive and sometimes synergistic for increasing risk. To improve the prediction of cardiovascular events and provide quantitative risk assessment, a number of multivariable risk equations or scores, such as the Framingham equations (E-Tables 52-1 and 52-2), have been developed. The vast majority of risk scores available have focused on predicting 10-year absolute risk, and essentially all include age, sex, smoking status, cholesterol, and blood pressure, with some also including diabetes, family history, body mass index, socioeconomic status, or novel biomarkers. The end points of interest for diverse risk equations have varied widely, from the prediction of cardiovascular death alone to the prediction of fatal and nonfatal major coronary events, major atherosclerotic events (coronary disease and stroke), and a broader range of cardiovascular events (including heart failure, coronary revascularization, angina, or claudication). For example, the 10-year risk for incident atherosclerotic cardiovascular disease can be predicted in 50-yearold men and women according to sex, race, and different levels of risk factors (Fig. 52-1), and the risks are dramatically higher with a greater risk factor burden.

Lifetime Risk Estimation

Despite the widespread use of 10-year risk estimates to guide prevention strategies, this approach has important limitations. For example, one consequence of the substantial weighting of age in 10-year risk equations is that younger men and women, even those with substantial risk factor burden, do not tend to have a high short-term risk. When treatment thresholds are applied to quantitative risk estimates for clinical guidelines, men younger

259.e1

CHAPTER 52  Epidemiology of Cardiovascular Disease  

E-TABLE 52-1  FRAMINGHAM RISK SCORE FOR CARDIOVASCULAR DISEASE PREDICTION ACCORDING TO TRADITIONAL RISK FACTORS POINTS

AGE (yr)

HDL-C (mg/dL)

TOTAL CHOLESTEROL (mg/dL)

SBP NOT TREATED

SBP TREATED

SMOKER

DIABETES

No

No

WOMEN 65 Hypertension Diabetes mellitus Congestive heart failure Prior stroke or TIA Yes

Drug dosages Dabigatran 150 mg bid, 75 mg bid if creatinine clearance < 30 Rivaroxaban 20 mg qd, 15 mg qd if creatinine clearance 15-50 Apixaban 5 mg bid, 2.5 mg bid if 2 or more (age > 80, body weight ≤ 60 kg, creatinine ≥ 1.5 mg/dL)

365

as the advent of reliable ablative therapies for AF may necessitate a reevaluation of this question. If a strategy of rate control is chosen, it is important to confirm a heart rate of 80 to 110 beats per minute at rest and less than 140 beats per minute with exercise, preferably by monitoring the heart rate during exercise on an exercise treadmill test or with an ambulatory monitor. More strict rate control is not beneficial. A5  Failure to confirm rate control can result in the development of tachycardia-induced cardiomyopathy. First-line therapy for rate control includes β-blockers or calcium-channel blockers; digoxin can also be used but is generally less effective. Patients commonly require a combination of mediations to achieve goal heart rates.7 If a strategy of rhythm control is chosen, many patients will first require cardioversion, either pharmacologic or electrical (Chapter 66). The risk for clot formation must be mitigated before cardioversion in all patients with AF of more than 48 hours’ duration. The first step generally is to perform transesophageal echocardiography (TEE) (Chapter 55). If TEE shows no evidence  of a left atrial clot, cardioversion can be undertaken without systemic anticoagulation; if the patient has risk factors for stroke in association with AF, however, most clinicians administer anticoagulation during the cardioversion and for the subsequent 4 weeks. If the TEE shows evidence of clot, 4 con­ secutive weeks of warfarin anticoagulation with an INR of at least 2, or equivalent anticoagulation with therapeutic doses of dabigatran, rivaroxaban, or apixaban is required; anticoagulation must be maintained for at least 3 to 4 weeks after cardioversion. Electrical cardioversion, which should be performed with a minimum of 200 joules, is successful in more than 90% of cases. Pharmacologic cardioversion can be performed with intravenous drugs such as ibutilide, which is more successful for atrial flutter (60% efficacy) than AF (50%). Oral medications can also be used as a “pill in the pocket” strategy. Patients can take a single dose of propafenone (600 mg) or flecainide (300 mg) with a conversion rate for recent-onset AF ( 48 hr duration

Warfarin, dabigatran, apixaban, or rivaroxaban for 3-4 weeks

FIGURE 64-23.  Management of recent-onset atrial fibrillation (AF). CV = cardioversion; INR = international normalized ratio; TEE = transesophageal echocardiography; TIA, transient ischemic attack.

366

CHAPTER 64  Cardiac Arrhythmias with Supraventricular Origin  

weeks) can also be used for cardioversion and is successful in approximately 50% of patients with both recent and more prolonged AF.

Long-Term Management

AVNRT, AVRT, Atrial Tachycardias, and Atrial Flutter

Chronic therapy for AVNRT is guided by the frequency and severity of symptoms.8 Many patients are able to live with this rhythm with infrequent recurrences, which terminate spontaneously or with adenosine. If chronic therapy is required, β-blockers or calcium-channel blockers and less commonly digoxin are used. Ablation of AVNRT (Chapter 66) is highly effective and should be considered before using sodium- or potassium-channel blocking drugs. Most patients with symptomatic AVRT are treated with catheter ablation (Chapter 66). Ablation of an accessory pathway located near the AV node or His bundle carries a 1% risk for complete heart block, whereas ablation of accessory pathways on the left side of the heart and distant from the AV node and His bundle region is not associated with a risk for heart block but carries a small risk for stroke. At present, it is not standard of care to ablate accessory pathways in patients without symptomatic arrhythmias.9,10 The long-term management of atrial tachycardia depends on symptoms. If the rhythm is highly symptomatic, it is generally managed with a β-blocker or calcium-channel blocker. If these medications are unsuccessful or not tolerated, ablation is frequently recommended, but antiarrhythmic medications are an alternative. In patients with atrial flutter, ventricular rate control is possible by achieving AV nodal block with β-blockers, calcium-channel blockers, and digitalis. However, radiofrequency ablation, which is curative, is now the preferred choice for most patients with atrial flutter (Chapter 66), especially recurrent atrial flutter. Because atrial flutter carries a 3% per year risk for thromboembolism, patients with atrial flutter should also receive long-term anticoagulation similar to what is recommended for AF (see later). If atrial flutter is successful, the risk for recurrence is very small, and long-term anticoagulation is not necessary.

Atrial Fibrillation

Therapies for the chronic maintenance of sinus rhythm in patients with AF include pharmacologic and procedural approaches. The procedural approaches include catheter-based ablation inside the left atrium with the goal of electrically isolating the pulmonary veins from the left atrium. Similarly, a minimally invasive surgical approach can electrically isolate the pulmonary veins from the external surface of the heart with the additional resection of the left atrial appendage. Both these procedures have become standard options for AF, especially in patients who have recurrent AF despite at least one antiarrhythmic drug. A6  The catheter approach carries a small risk for cardiac perforation, including pericardial tamponade and atrioesophageal fistula formation, and a 1% risk for stroke. There is also a small risk for pulmonary vein stenosis, which has been reduced by newer technologies. A second procedure typically is offered to patients who have recurrent AF following a first catheter-based procedure. In randomized trials of patients with paroxysmal AF, the cumulative burden of AF over a period of 2 years appeared to be slightly lower with initial radiofrequency catheter ablation therapy compared with antiarrhythmic medications, but at the expense of procedural risks and without any differences  in patient-reported quality of life. A7  A8  Catheter ablation may, however, be preferred as initial therapy in patients with persistent AF and symptomatic heart failure. A9  For patients with long-standing persistent AF, 5-year success rates are 20% for a single ablation procedure and 45% for multiple ablation procedures. The surgical approach carries a higher risk for cardiac bleeding, particularly during the resection of the left atrial appendage, and is associated with a significantly longer recovery time than the percutaneous approach. However, there should be no stroke risk associated with the surgical procedure because it is performed completely from the epicardial surface of the heart. A more extensive surgical operation, called the maze procedure, requires a full thoracotomy and is most often performed concomitantly as part of open coronary artery bypass surgery or an open valve operation. In this procedure, electrical lines of block are created in the left atrium to interrupt the perpetuation of AF, the pulmonary veins are isolated, and the left atrial appendage is resected. Success rates for this procedure, which should be reserved for refractory, symptomatic AF, exceed 80%. The pharmacologic options for the treatment of AF work by blocking sodium, potassium, or a combination of cardiac channels. Blockade of these channels results in slowing of cardiac conduction (sodium channels) and  prolongation in cardiac repolarization (potassium channels) as well as additional effects from modulation of the autonomic nervous system. The choice of antiarrhythmic drug is based on the patient’s underlying clinical condition (Table 64-6). Amiodarone is the most widely used medication for AF with an efficacy of 60 to 70% at 1 year. It is associated with a number of drug interactions, most notably with warfarin and digoxin. Its associated risk for thyroid, liver, and lung toxicities, related in part to the iodine moieties on this compound, necessitate ,

TABLE 64-6  SELECTION OF ANTIARRHYTHMIC DRUGS PATIENT CHARACTERISTICS

ANTIARRHYTHMIC DRUG CHOICES

No structural heart disease

First line: flecainide, propafenone, dronedarone, sotalol Second line: amiodarone, dofetilide

Depressed left ventricular ejection fraction with heart failure

First line: amiodarone, dofetilide Avoid: dronedarone, flecainide, propafenone

Coronary artery disease without congestive heart failure

First line: sotalol, dronedarone, dofetilide, amiodarone Avoid: flecainide, propafenone

Hypertrophic cardiomyopathy

First line: amiodarone, sotalol, dronedarone Second line: disopyramide

TABLE 64-7  CURRENT RECOMMENDATIONS FOR THROMBOEMBOLIC PROPHYLAXIS FOR PATIENTS WITH ATRIAL FIBRILLATION BASED ON RISK FACTORS FOR STROKE RISK FACTORS* Heart failure (1 point) Hypertension (1 point) Age ≥65 (1 point), ≥75 (2 points) Diabetes (1 point) Stroke/TIA (2 points) Vascular disease (1 point) Female gender (1 point)

RECOMMENDATIONS 2 or more points: anticoagulation with warfarin or a new oral anticoagulant 1 point: anticoagulation or no therapy depending on the preference of the patient and treating physician 0 points: no therapy

*Based on the CHA2DS2-VASc risk stratification scoring system. TIA = transient ischemic attack. Data from January CT, Wann S, Alpert JS, et al. 2014 AHA/ACC/HRS Guideline for the Management of Patients with Atrial Fibrillation: A Report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014;64:e1-e76.

careful follow-up. For prevention of recurrent AF, oral amiodarone is significantly more effective than propafenone, flecainide, dofetilide, or sotalol, which are the recommended alternatives. Dronedarone (400 mg twice daily), which is related to amiodarone but has no iodine and a 24-hour half-life, is well tolerated in terms of noncardiovascular side effects but has been associated with an increased risk for heart failure, stroke, and death in patients with permanent AF. As a result, it should be discontinued in patients in whom sinus rhythm is not well maintained. Quinidine, procainamide, and disopyramide are predominantly sodiumchannel blocking drugs that also block potassium channels at slow heart rates. Each of these drugs is moderately successful in AF, with about 50% of treated patients in sinus rhythm at 1 year, but each also has idiosyncratic noncardiovascular toxicities that can significantly limit their utility (see Table 64-5). Propafenone and flecainide are also sodium-channel blockers that are widely used for the maintenance of sinus rhythm. These drugs are moderately effective, with a 50% rate of sinus rhythm at 1 year, and are generally well tolerated but must be avoided in patients with structural heart disease, particularly with a history of prior myocardial infarction and impaired left ventricular function, because of a risk for drug-induced ventricular arrhythmia. Dofetilide is a potassium-channel blocking medication that is moderately effective for suppressing AF but carries a dose-dependent risk for QT prolongation and torsades de pointes.

Anticoagulation

The presence or absence of associated conditions helps determine which patients with AF require chronic anticoagulation with warfarin or other systematic coagulants (Table 64-7). Long-term anticoagulation therapy with warfarin, dabigatran, rivaroxaban, or apixaban is generally recommended in all patients who have persistent or paroxysmal AF, who are older than 65 years, and who have no contraindications to anticoagulation.4 Anticoagulation also should be maintained for 6 months after both catheter and surgical procedures in patients without clinical risk factors for stroke and chronically in patients with risk factors. Catheter-based procedures directed at excluding the left atrial appendage from the systemic blood stream may become options in patients who have a high risk for stroke and who cannot tolerate systemic anticoagulation owing to an excessive risk for bleeding. A10  Warfarin alone is superior to aspirin or the combination of clopidogrel  and aspirin, with meta-analysis showing that adjusted-dose warfarin and 

antiplatelet agents reduce stroke by approximately 60% and 20%, respectively. A11  Although there is some protective effect at an INR as low as 1.8, the target INR for chronic anticoagulation with warfarin should be 2 to 3 to avoid INRs less than 1.8. Guidelines no longer recommend aspirin or other antiplatelet agents in patients without an indication for warfarin or the newer anticoagulants. New oral anticoagulant medications have the potential to replace warfarin as more effective and safer (except for gastrointestinal bleeding) A12  primary therapy to prevent systemic emboli in patients with AF. In a randomized trial of patients with nonvalvular atrial fibrillation, rivaroxaban (an oral factor Xa inhibitor at 20 mg per day) was better than warfarin at preventing stroke  or systemic embolization, with significantly less intracranial and fatal bleeding. A13  In another randomized trial of patients with atrial fibrillation, apixaban (an oral factor Xa inhibitor at 5 mg twice daily) prevented more strokes and systemic emboli than warfarin, with less bleeding from all causes and fewer deaths. A14  Dabigatran, a direct thrombin inhibitor (150 mg twice daily), is superior to warfarin for preventing thromboembolism, with a lower risk for intracranial bleeding but a slightly higher risk for extracranial bleeding. A15  All three drugs are eliminated by the kidney (apixaban 25%, rivaroxaban 65%, and dabigatran 85%), so they are not recommended in patients with substantial renal dysfunction, and the doses should be reduced in patients with moderate renal dysfunction (Chapter 38). A reasonable approach is to use apixaban in patients at the highest risk for bleeding, to use rivaroxaban in patients who prefer once-daily dosing, and to avoid dabigatran in patients older than 80 years because of increased bleeding risk. The addition of aspirin to moderateintensity warfarin (INR 2 to 3) or to dabigatran, rivaroxaban, or apixaban can decrease vascular events and is recommended, despite its increased risk of causing bleeding, in some AF patients with concomitant risk factors, such as coronary artery disease or a prior stroke that is attributed to vascular disease rather than to AF.

Grade A References A1. Curtis AB, Worley SJ, Adamson PB, et al. Biventricular pacing for atrioventricular block and systolic dysfunction. N Engl J Med. 2013;368:1585-1593. A2. Cappato R, Castelvecchio S, Ricci C, et al. Clinical efficacy of ivabradine in patients with inappropriate sinus tachycardia: a prospective, randomized, placebo-controlled, double-blind, crossover evaluation. J Am Coll Cardiol. 2012;60:1323-1329. A3. Al-Khatib SM, Allen LaPointe NM, Chatterjee R, et al. Rate- and rhythm-control therapies in patients with atrial fibrillation: a systematic review. Ann Intern Med. 2014;160:760-773. A4. Roy D, Talajic M, Nattel S, et al. Rhythm control versus rate control for atrial fibrillation and heart failure. N Engl J Med. 2008;358:2667-2677. A5. Van Gelder I, Groenveld H, Crijns H. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med. 2010;362:1363-1373. A6. Wilber DJ, Pappone C, Neuzil P, et al. Comparison of antiarrhythmic drug therapy and radiofrequency catheter ablation in patients with paroxysmal atrial fibrillation: a randomized controlled trial. JAMA. 2010;303:333-340. A7. Morillo CA, Verma A, Connolly SJ, et al. Radiofrequency ablation vs antiarrhythmic drugs as first-line treatment of paroxysmal atrial fibrillation (RAAFT-2): a randomized trial. JAMA. 2014;311:692-700. A8. Cosedis Nielsen J, Johannessen A, Raatikainen P, et al. Radiofrequency ablation as initial therapy in paroxysmal atrial fibrillation. N Engl J Med. 2012;367:1587-1595. A9. Jones DG, Haldar SK, Hussain W, et al. A randomized trial to assess catheter ablation versus rate control in the management of persistent atrial fibrillation in heart failure. J Am Coll Cardiol. 2013;61:1894-1903. A10. Reddy VY, Möbius-Winkler S, Miller MA, et al. Left atrial appendage closure with the Watchman device in patients with a contraindication for oral anticoagulation: the ASAP study (ASA Plavix Feasibility Study With Watchman Left Atrial Appendage Closure Technology). J Am Coll Cardiol. 2013;61:2551-2556. A11. Hart RG, Pearce LA, Aguilar MI, et al. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med. 2007;146:857-867. A12. Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet. 2014;383:955-962. A13. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883-891. A14. Granger CB, Alexander JH, McMurray JJ, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981-992. A15. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361:1139-1151.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 64  Cardiac Arrhythmias with Supraventricular Origin  

GENERAL REFERENCES 1. Brignole M, Auricchio A, Baron-Esquivias G, et al. 2013 ESC guidelines on cardiac pacing and cardiac resynchronization therapy: the task force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association (EHRA). Europace. 2013;15:1070-1118. 2. Epstein AE, DiMarco JP, Ellenbogen KA, et al. 2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2013;61: e6-e75. 3. Zimetbaum P. Antiarrhythmic drug therapy for atrial fibrillation. Circulation. 2012;125:381-389. 4. Wann LS, Curtis AB, Ellenbogen KA, et al. Management of patients with atrial fibrillation (compilation of 2006 ACCF/AHA/ESC and 2011 ACCF/AHA/HRS recommendations): a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. Circulation. 2013;127:1916-1926. 5. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: Executive Summary: A Report of the American College of

6. 7. 8. 9.

10.

367.e1

Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2014;64:e1-e76. Heidbuchel H, Verhamme P, Alings M, et al. EHRA practical guide on the use of new oral anticoagulants in patients with non-valvular atrial fibrillation: executive summary. Eur Heart J. 2013;34: 2094-2106. Verma A, Cairns JA, Mitchell LB, et al. 2014 focused update of the Canadian Cardiovascular Society guidelines for the management of atrial fibrillation. Can J Cardiol. 2014;30:1114-1130. Link MS. Evaluation and initial treatment of supraventricular tachycardia. N Engl J Med. 2012;367: 1438-1448. Cohen MI, Triedman JK, Cannon BC, et al. PACES/HRS expert consensus statement on the management of the asymptomatic young patient with a Wolff-Parkinson-White (WPW, ventricular preexcitation) electrocardiographic pattern: developed in partnership between the Pediatric and Congenital Electrophysiology Society (PACES) and the Heart Rhythm Society (HRS). Endorsed by the governing bodies of PACES, HRS, the American College of Cardiology Foundation (ACCF), the American Heart Association (AHA), the American Academy of Pediatrics (AAP), and the Canadian Heart Rhythm Society (CHRS). Heart Rhythm. 2012;9:1006-1024. Pappone C, Vicedomini G, Manguso F, et al. Wolff-Parkinson-White syndrome in the era of catheter ablation: insights from a registry study of 2169 patients. Circulation. 2014;130:811-819.

367.e2

CHAPTER 64  Cardiac Arrhythmias with Supraventricular Origin  

REVIEW QUESTIONS 1. Supraventricular tachyarrhythmias. A 25-year-old woman presents to the emergency department with palpitations and the attached electrocardiogram. Which of the following statements is false? I

aVR

V1

II

aVL

V2

V5

III

aVF

V3

V6

V4

II

A. The rhythm can be treated with intravenous diltiazem. B. The rhythm is best treated with adenosine. C. The rhythm is curable with radiofrequency ablation. D. The rhythm is most consistent with atrial flutter. E. The rhythm is most consistent with atrioventricular (AV) nodal re-entrant tachycardia. Answer: D  The rhythm is AV nodal re-entrant tachycardia given the absence of evident P waves. This rhythm is typical in women in this age group. It can be treated acutely with adenosine or diltiazem, and long-term treatment can include a curative ablation procedure. The absence of flutter waves makes atrial flutter unlikely. 2. Supraventricular tachyarrhythmias. A 54-year-old man presents to the emergency department with the new onset of palpitations and shortness of breath. The accompanying electrocardiogram is obtained. What is the diagnosis? aVR

V1

V4

I

aVL

V2

V5

II

aVF

V3

V6

I

II

II

II

A. AV nodal re-entrant tachycardia B. AV re-entrant tachycardia C. Atrial flutter D. Atrial fibrillation E. Atrial tachycardia Answer: C  The electrocardiogram demonstrates negative (“sawtooth”) waves in leads 2, 3, and aVF, positive flutter waves in V1 and negative in V6. There are two flutter waves for every QRS complex, consistent with typical atrial flutter with 2 : 1 conduction.

CHAPTER 64  Cardiac Arrhythmias with Supraventricular Origin  

367.e3

3. Atrioventricular conduction disturbances. An 18-year-old man presents to the hospital with fatigue for the prior week. It is late August and he has spent the summer on Nantucket as a life guard. His pulse is 40 beats per minute, and the accompanying electrocardiogram is obtained. What is the most likely diagnosis? I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

A. Sinus bradycardia B. AV Wenckebach C. Complete heart block D. 2 : 1 AV block E. Junctional rhythm Answer: D  The rhythm shows 2 : 1 AV block with a long PR interval associated with the conducted P wave. The clinical history is consistent AV block secondary to Lyme disease. It should be treated with antibiotics, and in most cases there will be resolution of conduction disease. 4. Supraventricular tachycardia. An 18-year-old man presents with a syncopal episode while playing basketball. He is brought to the emergency department where the following electrocardiogram is recorded. His blood pressure is stable. What is the diagnosis? I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

I

A. Ventricular tachycardia B. AV nodal re-entrant tachycardia with aberration C. Atrioventricular tachycardia with aberration D. Atrioventricular tachycardia without aberration E. Atrial fibrillation with conduction over an accessory pathway (bypass tract) Answer: E  The rhythm is irregularly irregular, consistent with atrial fibrillation, and the variable QRS durations are consistent with varying amounts of conduction over an accessory pathway. This classic fast, broad, and irregular pattern is consistent with atrial fibrillation conducted over an accessory pathway.

367.e4

CHAPTER 64  Cardiac Arrhythmias with Supraventricular Origin  

5. Bradyarrhythmias. A 45-year-old man complains of daytime somnolence. Evaluation is notable for morbid obesity and hypertension. He takes 20 mg of lisinopril for hypertension. The following telemetry strip occurred during sleep and was not associated with symptoms. During daytime hours, his heart rate never dropped below 70 beats per minute. Which statement is true? N II 1 mV

N

N

N

N

PLETA 2 : 04 : 10

N

N

N

5/2/07 2 : 04 : 16

N

N

22 : 04 : 21

***BRADY 39 < 40

M

2 : 04 : 22

A. The telemetry strip shows complete heart block. B. The telemetry strip shows sinus node dysfunction, which requires a pacemaker. C. The telemetry strip shows sinus bradycardia progressing to sinus arrest, consistent with a vagal mechanism. D. Sinus bradycardia is due to metoprolol. E. The telemetry strip shows sinus arrhythmia. Answer: C  The telemetry strip shows sinus bradycardia progressing to a sinus pause. This finding is most consistent with a vagal mechanism, probably associated with obstructive sleep apnea. The best approach is to treat the sleep apnea. In the absence of symptoms such as syncope, a pacemaker is not indicated.

CHAPTER 65  Ventricular Arrhythmias  

367

65  VENTRICULAR ARRHYTHMIAS HASAN GARAN

DEFINITIONS

Ventricular arrhythmias are cardiac rhythms that originate in the ventricular myocardium or in the His-Purkinje tissue. They include a wide spectrum of arrhythmias, from the most innocuous isolated premature ventricular contraction (PVC) to the most malignant and life-threatening ventricular arrhythmia (Fig. 65-1). Two consecutive PVCs are termed a couplet, whereas ventricular tachycardia (VT) is arbitrarily defined as three or more ventricular contractions in a row at a rate faster than 100 beats per minute. The definition of sustained VT—a continuous ventricular rhythm, at a rate faster than 100 beats per minute, with no interruption for 30 seconds or longer—is equally arbitrary. However, most if not all sustained VTs are much faster than 100 beats per minute, persist for more than 30 seconds, and cause a substantial decrease in ventricular function and cardiac output, especially in patients with underlying organic heart disease. These abrupt physiologic changes may result in acute heart failure, hypotension, syncope, or even circulatory collapse within several seconds to minutes after the onset of VT. Monomorphic VT is electrocardiographically defined as a wide-complex tachycardia with no change in QRS configuration, frontal axis, or horizontal axis from one beat to the next (see Fig. 65-1C). Monomorphic ventricular tachycardia (Fig. 65-2A) at a very rapid (>250 beats per minute) rate is sometimes called ventricular flutter, but there is no consensus for a definite rate cutoff, and it is not possible to separate the QRS clearly from the T waves when the rate exceeds 250 beats per minute. Polymorphic VT is characterized by beat-to-beat changes in the QRS morphology and axis, and very fast polymorphic VT may be difficult to distinguish from ventricular fibrillation (VF) (Fig. 65-2B). VF is a grossly irregular ventricular rhythm, usually at a rate faster than 300 beats per minute and with markedly variable low amplitude in the QRS morphology, during which there is no cardiac output. Torsades de pointes and bidirectional polymorphic VT are two distinct subtypes of polymorphic VT. To avoid confusion, the term pleomorphic VT should be used rather than the term polymorphic VT to describe the phenomenon of multiple clinical monomorphic VTs, each with distinct QRS configurations and axis observed at different times in the same patient.

EPIDEMIOLOGY

The prevalence of PVCs is a function of sampling method and duration, and PVCs may be seen in 50% of apparently healthy individuals if the monitoring time is 24 hours or longer. Nonsustained VT may be recorded in up to 3% of apparently healthy individuals with no identifiable heart disease. The prevalence of PVCs and nonsustained VT increases with age, but also with the presence and severity of an underlying heart disease. Therefore, the finding of nonsustained VT often leads to a cardiac evaluation to exclude organic heart disease, even if it is incidentally discovered in an asymptomatic patient. The prevalence of nonsustained VT rises to 7 to 12% in the late phase of myocardial infarction (MI) and may be as high as 80% in patients with heart failure owing to dilated cardiomyopathy (Chapter 60). Approximately 10% of patients with documented sustained VT have no identifiable heart disease, in which case idiopathic VT is diagnosed. Idiopathic VF is exceedingly rare. Sudden cardiac death (Chapter 63) owing to ventricular arrhythmias accounts for an estimated 50% of all annual cardiovascular deaths in the United States.1 The nature of the underlying heart disease in patients dying of VT or VF is age dependent. Before 30 years of age, the organic heart disease most commonly associated with VT and VF is genetic cardiomyopathy (Chapter 60), whereas acute MI and chronic ischemic cardiomyopathy are the most common underlying heart diseases in individuals older than 40 years. In about one third of cases of sudden cardiac death without obvious underlying organic heart disease at autopsy, post-mortem genetic analysis may identify a deleterious mutation in an ion channel—a so-called channelopathy that predisposes to VT and VF.

368

CHAPTER 65  Ventricular Arrhythmias  

A

B

C FIGURE 65-1.  Ventricular arrhythmias. A, Multifocal premature ventricular beats. B, Nonsustained monomorphic ventricular tachycardia. Note dissociated P waves indicated by arrows. C, Sustained monomorphic ventricular tachycardia. Dissociated P waves are indicated by arrows.

I

II

III

aVR

V1

aVL

V2

aVF

V3

V4

V5

V6

A

B FIGURE 65-2.  A, Monomorphic ventricular tachycardia (VT) in a patient with chronic myocardial infarction. The arrows identify P waves in lead V1, showing atrioventricular dissociation. No R wave is recorded in any of the precordial leads V1 to V6 during VT. B, Polymorphic VT in a patient with chronic ischemic cardiomyopathy and marked first-degree atrioventricular block. There is no QT prolongation before the onset of the polymorphic VT.

PATHOBIOLOGY

Based on their underlying mechanisms, ventricular arrhythmias are classified as re-entrant, triggered, or automatic (Chapter 61). Re-entry, which results from activation in pathways sharing a common isthmus, is initiated by the simultaneous presence of conduction block in one limb and abnormally slow conduction in an adjacent limb, thereby allowing recovery of excitability in the former (E-Fig. 65-1A). One type of triggered activity results from early afterdepolarizations, which are oscillatory depolarizations occurring during the late phase of the action potential (E-Fig. 65-1B). Another type of triggered activity results from delayed afterdepolarizations, which are transient

depolarizations that occur immediately after the termination of the action potential and may reach activation threshold. Automatic arrhythmias arise from accelerated pacemaker activity (E-Fig. 65-1C). Sustained re-entrant activation in the myocardium, which is the most common cause of monomorphic VT, usually arises from subendocardial scarring, which is the result of prior ischemic injury and which creates an electrophysiologically abnormal substrate that results in re-entry. Other pathologic conditions capable of creating a substrate for re-entry include inflammation, granuloma (e.g., cardiac sarcoidosis), fibrofatty infiltration (e.g. arrhythmogenic right ventricular cardiomyopathy [ARVC]), genetically caused sarcomeric disarray (e.g., hypertrophic cardiomyopathy), and iatrogenic scar or

CHAPTER 65  Ventricular Arrhythmias  

368.e1

LAD Base

Apex Left lateral

A

EAD Shorter cyle length, abnormal automaticity

0 mV

Em DAD

–80

Faster phase 4 depolarization

B C

E-FIGURE 65-1.  A, Re-entry within the myocardial infarction zone in an experimental canine model of ventricular tachycardia (VT). A central region of slow abnormal conduction, commonly referred to as the isthmus, is characterized by narrow crowded isochrones flanked by arcs of bidirectional conduction block depicted by dark lines, isolating the isthmus and enabling the maintenance of re-entry. The arrows indicate the spread of the wave of depolarization outside the central isthmus, in the shape of a figure of 8, with the red zone as the early breakthrough of activation and the dark blue area as the late activation in the VT cycle, which is also the point of re-entry into the isthmus. LAD = left anterior descending artery. B, Schematic depiction of the cardiac action potential with early afterdepolarizations (EAD) during phase 3 of a prolonged action potential (dotted lines) and delayed afterdepolarizations (DAD) reaching threshold and resulting in a premature action potential at the end of the phase 3 and the very start of the phase 4. Em = membrane potential. C, Schematic depiction of cardiac action potential with an increased slope of depolarization toward the threshold during phase 4, at a site of automatic tachycardia.

CHAPTER 65  Ventricular Arrhythmias  

patch (e.g., surgical repair of tetralogy of Fallot). These substrates may also result in polymorphic VT and VF by more than one mechanism. The mechanism of ventricular arrhythmias in Brugada syndrome is not completely understood. One proposed mechanism is based on intraventricular phase 2 re-entry owing to an exaggerated endocardial-to-epicardial gradient in membrane potential due to differences in transient outward current. Other evidence suggests abnormal conduction in the epicardium of the right ventricular outflow tract. Triggered activity, which results from adenosine-sensitive delayed afterdepolarizations rather than re-entry, is thought to be the underlying mechanism for idiopathic monomorphic VT of outflow tract origin. Idiopathic VT from re-entry in the fascicles of the left bundle branch has a relatively narrow QRS complex that always manifests right bundle branch block mimicry, most commonly with left, but rarely with right, frontal axis deviation. Torsades de pointes is caused by early afterdepolarizations that arise during an abnormally prolonged action potential owing to a delayed repolarization process in the setting of genetic long QT syndromes or acquired long QT during therapy with QT-prolonging drugs. The cause may be either diminished outflowing potassium currents or enhanced inflowing sodium or calcium currents. Although many episodes terminate spontaneously, the rates are usually very fast, and a torsade episode, if long enough, can transform into VF. Bundle branch re-entry, which results from re-entrant activation incorporating the right and the left bundle branches distally joined by the slowly conducting septal myocardium, may cause one or two nonsustained ventricular beats in a normal heart. However, sustained bundle branch re-entry occurs when myocardial disease causes chamber enlargement and bundle branch elongation and/or disease in the conduction system causes abnormal slow conduction, thereby creating the scenario for sustained bundle branch re-entry. The common type of bundle branch re-entry has anterograde activation over the right bundle and uses the left bundle retrogradely, thereby resulting in a left bunch branch block (LBBB) pattern on surface electrocardiogram (ECG), but the reverse direction with right bundle branch block (RBBB) may also occur rarely. Accelerated pacemaker activity in an ectopic location, with rates exceeding the underlying sinus rhythm rate, may arise in settings such as transient inflammation, excess digoxin levels, intracellular calcium loading, electrolyte imbalance, and coronary reperfusion following thrombotic occlusion. Bidirectional VT is thought to result from calcium overload of the myocytes owing to congenitally acquired abnormal calcium release from the ryanodine receptor or digitalis toxicity. Finally, there is no consensus regarding the mechanisms underlying VF. Theoretically, VF may be initiated when early or delayed afterdepolar­ izations fall in the vulnerable period of the action potential, thereby pre­ cipitating a re-entrant wave that breaks into sister wavelets and results in high-frequency electrical activity. In fact, VF may be regarded as an end stage for a variety of severe electrophysiologic abnormalities that result in chaotic activation.

CLINICAL MANIFESTATIONS

Ventricular arrhythmias can present in a variety of clinical settings (Table 65-1). Often, ventricular arrhythmias are asymptomatic and are detected by an irregular pulse on a physical examination, on a routine ECG, on an exercise test, or on routine inpatient monitoring. In other patients, symptomatic ventricular arrhythmias can present as palpitations, dizziness, syncope (Chapters 51 and 62), shortness of breath, or sudden cardiac arrest (Chapter 63). The diagnosis usually can be confirmed on an ECG, but ambulatory monitoring (Chapter 62) is often needed because the arrhythmia may be intermittent. Ambulatory monitoring can also help correlate arrhythmias with any potentially related symptoms. In some patients, exercise testing can be helpful, especially in patients with exercise-induced symptoms. On the ECG, the QRS complex duration will typically be more than 0.12 seconds. In monomorphic VT (Fig. 65-3), the QRS complexes are the same from beat to beat, whereas polymorphic VT has multiple and changing QRS morphologies (Fig. 65-4). In VF, the ECG shows continuous irregular activation without any discrete QRS complexes (Fig. 65-5). Although underlying structural heart disease is usually present, these arrhythmias do not require a fixed structural substrate.

Acute Myocardial Infarction

VT and VF may arise as early as minutes to hours after the onset of symptoms during acute myocardial infarction (MI), and prehospital VT and VF during

369

TABLE 65-1  VENTRICULAR TACHYCARDIA AND CARDIAC DIAGNOSIS STRUCTURAL HEART DISEASE Acquired heart disease Acute myocardial infarction Chronic myocardial infarction, ischemic heart disease Nonischemic dilated cardiomyopathy Hypertensive heart disease Valvar heart disease Cardiac sarcoidosis Cardiac amyloidosis Other infiltrative diseases (e.g., Chagas disease) Cardiac tumors Congenital heart disease Arrhythmogenic right ventricular cardiomyopathy Hypertrophic cardiomyopathy Genetic dilated cardiomyopathies Iatrogenic Surgically repaired congenital heart disease Left ventricular assist devices NO STRUCTURAL HEART DISEASE Idiopathic ventricular tachycardia Right and left ventricular outflow tract tachycardias Left intrafascicular re-entry Papillary muscle tachycardias Idiopathic ventricular fibrillation Ion channel mutations Long QT syndromes Catecholaminergic polymorphic ventricular tachycardia Short QT syndrome Mixed etiology Brugada syndrome

acute MI are responsible for a large proportion of out-of-hospital sudden cardiac deaths (Chapter 63). The incidence of peri-infarction VF has declined over the past two decades, presumably related to the widespread practice of coronary revascularization (Chapter 74) during acute MI. Among patients with ST elevation MI who now reach the hospital, about 3 to 4% develop VT, mostly during the acute phase. The incidence of VT in patients with non-ST elevation MI (Chapter 72) is lower, about 1%. Accelerated idioventricular rhythm (AIVR) is an automatic ventricular rhythm that is faster than the sinus rate but usually less than 120 beats per minute. It may occur in the setting of acute MI and is commonly observed immediately after coronary reperfusion. AIVR rates are slower than those of the fast and malignant VT and VF of acute MI, and this arrhythmia typically terminates spontaneously without causing hemodynamic instability.

DIAGNOSIS

Not every wide-complex tachycardia is VT. The diagnosis is straightforward from the His bundle electrogram recorded at the time of a wide-complex tachycardia during a cardiac electrophysiology study, but diagnosis on a standard 12-lead ECG may be challenging (Table 65-2). The differential diagnosis of a sustained regular-rate wide-complex tachycardia includes any type of supraventricular tachycardia with aberrant conduction (Chapter 64), supraventricular tachycardia with ventricular preexcitation, bundle branch re-entry (which is a specific type of VT), and myocardial VT. The clinical setting and the patient’s background (e.g., history of previous MI or cardiomyopathy) play a major role in making an accurate diagnosis. New-onset wide-complex tachycardia in a young and otherwise healthy individual with no structural heart disease is most likely supraventricular tachycardia (SVT) with aberration, an SVT with preexcitation, or idiopathic VT. The most reliable observation in favor of VT is evidence of AV dissociation, that is, absence of any relationship between the atrial and ventricular rate, with the ventricular rate faster than the atrial (see Fig. 65-2A), or a regular wide-complex tachycardia with the atria fibrillating. However, the absence of atrioventricular (AV) dissociation does not exclude VT because ventriculoatrial conduction is present in about 25% of VTs. Fusion beats (which occur when an occasional sinus beat conducts through the AV node and reaches the His-Purkinje system at the same time as the VT source activates the myocardium, thereby resulting in a beat with a morphology that is the hybrid of a conducted QRS complex and the VT complex) confirm AV

370

CHAPTER 65  Ventricular Arrhythmias  

I

II

aVR

aVL

V1

V4

V2

V5 220 ms

III aVF

V3

V6

FIGURE 65-3.  Monomorphic ventricular tachycardia in a patient with chronic ischemic cardiomyopathy. In lead V2, the duration from the onset of the R wave to the nadir of the S wave is more than 200 msec. See text for further explanation.

V1

V5 FIGURE 65-4. Torsades de pointes (TdP) in a patient with a markedly prolonged QT interval. A premature ventricular beat just after the peak of the T wave initiates TdP. As the tachycardia progresses, the rotation or the “twist” in the QRS axis is clearly observed in lead V1, with the polarity of the signal changing gradually from negative to positive.

V6 FIGURE 65-5. This electrocardiogram in a patient with idiopathic ventricular fibrillation (VF) shows recurrent closely coupled premature ventricular contractions (PVCs) and the initiation of VF by one of these closely coupled PVCs.

TABLE 65-2  DISTINGUISHING VENTRICULAR TACHYCARDIA FROM SUPRAVENTRICULAR TACHYCARDIA WITH ABERRANT CONDUCTION VENTRICULAR TACHYCARDIA AV dissociation aVR: initial R > S or initial R or Q > 40 msec Absence of any R wave in V1 to V6 V1 to V6: onset of R to S > 100 msec in any lead QRS duration >160 msec Initial R wave in aVR

SUPRAVENTRICULAR TACHYCARDIA Same QRS morphology as preexisting bundle branch block in sinus rhythm V1: rsR′

AV = atrioventricular.

dissociation but are observed only when VT rates are relatively slow. Other findings that favor VT include a QRS duration longer than 160 msec, or longer than 140 msec with an RBBB pattern. One approach, based on the QRS configuration on the ECG, uses the absence of RS complex in all precordial leads or an interval of more than 100 msec from the onset of R to the nadir of S wave as observations strongly favoring VT (see Figs. 65-2A and 65-3). The absence of any R waves in the QRS complexes recorded from all six precordial leads, described as negative concordance, strongly suggests VT, but unfortunately is not a common finding. Prominent R waves observed in all six precordial ECG leads, termed positive concordance, may be seen in SVT with left ventricular preexcitation but otherwise also suggests VT with a basal site of origin. In the absence of preexcitation, a slow rate of rise in the voltage

during the first 40 to 60 msec of the QRS onset suggests VT, as does the presence of initial R wave in lead aVR. A wide-complex tachycardia with a QRS morphology identical to that of aberrantly conducted beats manifesting bundle branch block (BBB) on a previously recorded ECG in the same patient should raise suspicion of bundle branch re-entry VT if AV dissociation is present. If AV dissociation is not present, the differential diagnosis includes SVT with aberrant conduction, but the rare condition of preexcitation with an atriofascicular accessory pathway should also be considered in a patient with LBBB aberration. A monomorphic wide-complex tachycardia with an irregular rate, manifested by more than 60-msec difference in cycle length from one beat to the next, is likely to be atrial fibrillation (AF) or atrial flutter, with variable AV block and aberrant conduction or with preexcitation. It is important to emphasize that electrolyte imbalances or the use of antiarrhythmic drugs diminishes the predictive accuracy of all of these diagnostic clues. Sustained polymorphic wide-complex tachycardia with marked beat-tobeat changes in the QRS morphology is always ventricular and either terminates spontaneously or transforms into VF. Torsades de pointes, a specific type of polymorphic VT, derives its name from the “twisting” or rotating of the QRS axis as the tachycardia progresses. It occurs in genetic or acquired long QT syndrome and is frequently pause dependent—typically starting when a premature beat falls on the prolonged T wave of the beat following a long RR interval (see Fig. 65-4). Finally, bidirectional VT manifesting a unique feature of beat-by-beat axis alternans may occur with digitalis toxicity or in the congenital catecholaminergic polymorphic ventricular tachycardia syndrome. Several different algorithms based on the configurations of the QRS complexes have high sensitivity, high specificity, and acceptable predictive accuracy for distinguishing epicardial VT from endocardial VT (Table 65-3).2 All

CHAPTER 65  Ventricular Arrhythmias  

I

aVR

V1

II

aVL

V2

III

aVF

V3

371

V4

V5

V6

FIGURE 65-6.  Monomorphic epicardial ventricular tachycardia in a patient with nonischemic dilated cardiomyopathy. The positive polarity pseudo-delta wave is prominent in the right precordial leads and the negative polarity pseudo-delta wave is prominent in the inferior limb leads.

TABLE 65-3  ELECTROCARDIOGRAPHIC PARAMETERS USED TO PREDICT AN EPICARDIAL ORIGIN OF VENTRICULAR TACHYCARDIA PARAMETER

CRITERIA

Pseudo-delta wave

>75 msec favors epicardial site

Intrinsicoid deflection time

>85 msec favors epicardial site

Onset of R to nadir of S in precordial leads

>120 msec favors epicardial site

QRS duration

Epicardial longer

Q waves during VT in lead I

Favors epicardial site

Q waves during VT in II-III-aVF

Favors endocardial site

aVR/aVL amplitude ratio

Epicardial higher

VT = ventricular tachycardia.

are based on ECG criteria for whether the initial activation likely starts at an epicardial site. If so, the rapidly conducting His-Purkinje system is not available immediately, and the intramyocardial conduction delay produces a slurred initial component of the QRS complex, often called a pseudo-delta wave, which is manifested as a slow rate of rise of voltage before it reaches the intrinsicoid deflection (Fig. 65-6). Early recognition of ECG findings suggesting an epicardial origin of VT is important in planning and preparing a patient before a catheter ablation procedure (Chapter 66) because the epicardial approach requires a special technique in the cardiac electrophysiology laboratory. Cardiac electrophysiology testing (Chapter 62) may be indicated in patients who have organic heart disease and recurrent syncope but in whom the history, physical examination, ECG, echocardiogram, and ambulatory cardiac rhythm monitoring fail to clarify the cause, especially if the patient has a history of myocardial infarction or cardiomyopathy, either of which increases the probability that VT may be the cause of syncope. A second diagnostic indication is to identify the mechanism underlying a documented wide-complex tachycardia before the consideration of catheter ablation therapy (Chapter 66).

Identifying the Underlying Cause of Ventricular Arrhythmias

In patients with a diagnosed ventricular arrhythmia, the next step is to conduct a careful evaluation to exclude any underlying structural heart disease. This evaluation must include a comprehensive history and physical examination (Chapter 51), echocardiography (Chapter 55), and stress testing (Chapter 71). The family history may provide clues to guide genetic testing for an inherited cardiomyopathy (Chapter 60). Cardiac magnetic resonance imaging (Chapter 56) is indicated in selected patients to exclude conditions such as sarcoidosis and ARVC. Despite a comprehensive evaluation, about 10 to 15% of patients will have PVCs or VT with no identifiable structural or genetically identifiable cause. Most of the idiopathic monomorphic VTs are in one of two categories,

defined by ECG morphology. VTs that arise in the right or left ventricular outflow tract typically manifest an inferiorly directed frontal axis and are markedly positive in inferior leads (E-Fig. 65-2); the QRS configuration observed in the right precordial leads may further discriminate the sites of origin as the right or the left ventricular outflow tract or one of the sinuses of Valsalva. By comparison, idiopathic left ventricular tachycardia usually manifests RBBB mimicry and left axis deviation, but there may also be right axis deviation. The QRS complexes typically are not very wide because the involved region is His-Purkinje tissue adjacent to the interventricular septum. The differential diagnosis includes idiopathic VT arising in one of the left ventricular papillary muscles (E-Fig. 65-3). When either of these typical patterns is observed in a patient with no structural heart disease, the physician should suspect idiopathic VT. Conversely, sustained VT that does not fall into either of these two broad categories should always raise a high index of suspicion that organic heart disease may be present.

Chronic Ischemic Heart Disease and Post−Myocardial Infarction Ventricular Tachycardia

In survivors of ST elevation MI, the prevalence of sustained VT by 6 weeks is about 1%, and VT may occur as late as 15 to 20 years after the acute MI without any intervening event. VT commonly, but not invariably, reflects poor left ventricular function, especially a dyskinetic left ventricular wall segment. The electrophysiologic substrate is the surviving but electrophysiologically abnormal tissue embedded in the infarcted zone, which creates the conditions for re-entry. The areas that harbor pathways underlying re-entry can be identified by low-amplitude fractionated local electrograms recorded from the endocardium. Up to 16% of the patients have VT of epicardial origin. The same substrate may cause polymorphic VT and VF, which do not depend on a long QT interval and are different than torsades de pointes seen with repolarization abnormalities.

Nonischemic Dilated Cardiomyopathy

The most common cause of sustained monomorphic VT in nonischemic cardiomyopathy (Chapter 60) is also re-entry within the myocardium, but it differs from the post-infarction VT of chronic ischemic heart disease. The pathologic substrate, such as fibrosis, may be hard to identify. The abnormal, low-voltage, fractionated local electrograms tend to be located in basal, lateral, and often perivalvar left ventricular areas, which may correlate with the location of intramyocardial or subepicardial scarring identified by cardiac magnetic resonance imaging. The proportion of monomorphic VTs due to bundle branch re-entry is higher in nonischemic dilated cardiomyopathy compared with chronic ischemic heart disease, and VT with a focal rather than re-entrant mechanism rarely may be observed. Also, VT of nonischemic dilated cardiomyopathy is more likely to have an epicardial origin—as high as 22 to 35% in many series—and reaching 70% in Chagas disease.3 Ventricular tachycardia resulting from bundle branch re-entry also is more common in nonischemic dilated cardiomyopathy.

Heart Failure

The failing heart from any underlying cause (Chapter 58) is highly vulnerable to ventricular arrhythmias, and 40 to 60% of the deaths in patients with

CHAPTER 65  Ventricular Arrhythmias  

I

V1

V4

V2

V5

aVR

II

aVL

III

aVF

V3

371.e1

V6

E-FIGURE 65-2. Electrocardiogram recorded during idiopathic ventricular tachycardia originating in the right ventricular outflow tract and manifesting a deeply inferior frontal axis and left bundle branch block mimicry in the precordial leads.

I

II

III

aVR

aVL

aVF

V1

V2

V3

V4

V5

V6

VI

II

V5 E-FIGURE 65-3.  Ventricular tachycardia (VT) originating in the anterolateral papillary muscle. Note the right bundle branch mimicry of the QRS and the right axis deviation, similar to the electrocardiographic configuration of an interfascicular re-entrant VT.

372

CHAPTER 65  Ventricular Arrhythmias  

heart failure are sudden and commonly from VT and VF. Re-entrant VT is common especially in patients whose heart failure is due to advanced ischemic heart disease, but triggered activity resulting from derangements of calcium homeostasis may also play a prominent role. In addition, hormonal factors, electrolyte abnormalities, and changes in autonomic nervous system activity also increase the vulnerability of the failing heart to ventricular arrhythmias.

Inflammatory and Infiltrative Disease

Among patients with sarcoidosis (Chapter 95), about 40 to 50% have cardiac involvement, which may first manifest as progressive AV block and VT. Although the true prevalence of VT in sarcoidosis is not known, in the selected patients who have received implantable cardiac defibrillators (ICDs) for cardiac sarcoidosis diagnosed by endomyocardial biopsy, cardiac magnetic resonance imaging, or cardiac positron emission tomographic scans, about 15% per year have appropriate ICD discharges for sustained VT.4 Patients with other infiltrative heart diseases such as amyloidosis (Chapter 188) also have an elevated risk for VT and life-threatening ventricular arrhythmias.5

Adult Congenital Heart Disease

VT may occur in the setting of any adult congenital heart disease when there is a ventricular surgical scar or patch, as is seen after repair of tetralogy of Fallot or a ventricular septal defect closure, or a failing ventricle such as after a Mustard or Senning procedure to palliate transposition of great arteries (Chapter 69). In patients with surgically repaired tetralogy of Fallot, the prevalence of VT is about 5%, and about 2% have sudden cardiac death.

Genetically Inherited Cardiomyopathies

Hypertrophic cardiomyopathy (Chapter 60) is responsible for more than one third of sudden cardiac deaths in patients younger than age 25 years (Chapter 63), and mortality in young hypertrophic cardiomyopathy patients is almost exclusively due to VT and VF. Neither genetic testing nor a cardiac EP study can definitively identify patients at high risk for VT and VF, and the risk is determined based on findings such as a history of syncope, documented nonsustained VT especially in a young patient, a markedly thickened (>3 cm) interventricular septum, and a paradoxical decrease in blood pressure during exercise.6 ARVC is a congenital cardiomyopathy (Chapter 60), usually with an autosomal dominant inheritance. The fibrofatty infiltration of the right ventricular myocardium, which may also involve the interventricular septum and the left ventricle, results in progressive histologic change and marked electrophysiologic abnormalities, which may be manifest on the surface ECG as an epsilon wave (Fig. 65-7). The markedly altered conduction characteristics are conducive to re-entry. The incidence of VT in ARVC is related to the severity of the pathologic myocardial changes and ranges from 25 to 100%, depending on the penetrance and the expressivity of the disease. VT typically is initiated by exercise and demonstrates LBBB mimicry in the precordial ECG leads. However, unlike idiopathic right ventricular outflow tract VT, the frontal axis may be variable and not always inferiorly directed, and the site of origin may be epicardial in about 40% of cases.

Genetically Inherited “Channelopathies”

Several genetically acquired syndromes, including the long QT syndromes, Brugada syndrome, and catecholaminergic polymorphic VT increase the risk for sudden cardiac death due to ventricular tachyarrhythmias. Despite the remarkable heterogeneity of the long QT syndrome, most of the cases (LQT1, LQT2, LQT3) result from mutations in the genes coding for one of the potassium channels or the sodium channel.7 The other genetic mutations are extremely rare. The VT of long QT syndrome is torsades de pointes, and both bradycardia and pauses increase its probability in patients who are predisposed. The incidence of torsades de pointes is influenced by multiple factors, including age, gender, the particular genetic mutation, and the magnitude of QT prolongation (Fig. 65-8A). The electrocardiographic hallmark of Brugada syndrome, which also predisposes to ventricular tachyarrhythmias and sudden cardiac death, is the coved ST segment elevations in the right precordial leads (Fig. 65-8B). In some cases, this pattern may not be present except when the patient is febrile. The inheritance is autosomal dominant, a sodium channel mutation is present in 20 to 30% of the cases, but the genetics are heterogenous. Catecholaminergic polymorphic VT is a rare genetic condition resulting from abnormal calcium homeostasis. It is characterized by exercise-induced, wide-complex tachycardia manifesting alternating ECG axes from one beat to the next. This condition also predisposes the patient to exercise-induced VF. In addition, a chromosomal haplotype causing overexpression of dipeptidyl peptidase-like protein-6 has been described in one type of familial idiopathic VF, a rare but challenging subset of inheritable arrhythmia syndromes causing sudden cardiac death.

Iatrogenic Ventricular Tachycardia and Ventricular Fibrillation

QT-prolonging drugs, including class III antiarrhythmic drugs (see www.sads.org.uk), may precipitate torsades de pointes and VF in genetically predisposed individuals even if the baseline QTc is normal or borderline. Class IC antiarrhythmic drugs may cause life-threatening VT in patients with ischemic or any other organic heart disease and in patients with Brugada syndrome. Ventricular scarring owing to aneurysmectomy, tetralogy of Fallot repair, ventricular septal defect repair, alcohol ablation of the interventricular septum to relieve dynamic outflow tract obstruction in hypertrophic cardiomyopathy, or implantation of a left ventricular assist device may create a substrate for re-entry and VT.

TREATMENT  Premature Ventricular Contractions and Nonsustained Ventricular Tachycardia

In the absence of structural heart disease, there is no convincing evidence that ventricular ectopic activity influences survival. Therefore PVCs do not need treatment in asymptomatic patients. If ventricular ectopy results in symptoms that substantially decrease quality of life, a cardioselective β-blocker (e.g., metoprolol 50 mg twice daily or atenolol 50 mg once daily) is safe but

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

FIGURE 65-7. This electrocardiogram was recorded in a patient with arrhythmogenic right ventricular cardiomyopathy, marked first-degree heart block, and recurrent ventricular tachycardia. Epsilon waves, marked by the arrow, are visible in the right precordial leads.

CHAPTER 65  Ventricular Arrhythmias  

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

373

A

I

aVR

II

aVL

III

aVF

V1

V4

V2

V5

V3

V6

B FIGURE 65-8.  A, Electrocardiogram showing a QT interval of 640 msec in a woman with LQT1 syndrome, with the terminal portion of the T wave merging with the P wave. B, Electrocardiogram of a man with Brugada syndrome, showing the typical “coved” ST elevation in lead V1.

not a very effective first-choice therapy to eradicate PVCs. PVCs may be more effectively suppressed using class IC antiarrhythmic agents such as flecainide (50 to 100 mg twice daily) or propafenone (150 to 225 three times daily), which are safe in the absence of organic heart disease but are contraindicated in organic heart disease, especially ischemic heart disease.8 In some patients, the frequency of PVCs and nonsustained VT reaches a critical level that results in decreased systolic ventricular function. These patients should be treated aggressively, including catheter mapping and ablation (Chapter 66) to avoid the potential adverse effects of antiarrhythmic drugs.

Acute Management of Ventricular Tachycardia and Ventricular Fibrillation

The management of VT with hemodynamic instability and VF should conform to the guidelines for advanced cardiac life support (Chapter 63), with an emphasis on defibrillation. For patients who have sustained VT with modest hypotension and normal mental status, intravenous drug therapy with lidocaine (given as a 50-mg bolus) or amiodarone (150 mg infused intravenously over 10 minutes) may be tried. Lidocaine, which works best at rapid heart rates, is an effective drug to terminate VT, which invariably occurs at a high rate, but not to prevent recurrences, except in the setting of acute ischemia. By comparison, amiodarone is more effective at slower heart rates and therefore is better for preventing recurrent VT after sinus rhythm is restored. Intravenous calcium channel blocker therapy should not be given unless the mechanism is known with certainty to be verapamil-sensitive idiopathic left ventricular tachycardia. The most important factor in preventing early recurrence is the prompt identification and reversal of any precipitating causes. Examples include hypokalemia and other electrolyte imbalances, low oxygen saturation, intravenous β-agonist agents, and acute heart failure (Chapter 59) or myocardial ischemia (Chapter 72 and 73). Heart failure should be rigorously treated (Chapter 59). VF suggests the presence of residual ischemia; the feasibility of coronary revascularization (Chapter 74) should be addressed, but even then recurrences are common.

Treatment of Electrical Storm

Electrical storm is a term used to describe frequently recurrent VT or VF requiring repeated defibrillations. Electrical storm rarely occurs in nonischemic cardiomyopathy or in genetically acquired ventricular arrhythmias. When this condition is encountered in the early phase of acute MI, relief of ischemia is of paramount importance. If the electrical storm continues even after coronary reperfusion, insertion of an intra-aortic balloon pump and use of an intravenous β-blocker therapy, preferably with a short half-life drug (e.g., esmolol 50

to 300 µg/kg per minute by intravenous infusion) should be considered. Intravenous lidocaine (2 to 4 mg per minute) or intravenous amiodarone (0.5 to 1.0 mg per minute) may also be used if esmolol is ineffective.

Idiopathic Ventricular Tachycardia and Ventricular Fibrillation

Although cardiac arrest resulting from transformation of idiopathic VT to VF is exceedingly rare, sustained VT at a rate faster than 200 beats per minute commonly causes cardiopulmonary symptoms and even syncope (Chapter 62). Idiopathic VTs of outflow tract origin may respond to β-blocker therapy (e.g., metoprolol 50 mg every 12 hours, or atenolol 50 mg daily), and a few may respond to an empirical trial of calcium-channel blockers (e.g., sustainedrelease diltiazem 120 to 240 mg daily, or sustained-release verapamil 120 to 240 mg daily). The so-called idiopathic left ventricular tachycardia resulting from left fascicular re-entry frequently responds to verapamil (e.g., sustainedrelease 180 to 360 mg daily), but papillary muscle VT may not. Catheter ablation therapy (Chapter 66), which may be curative for idiopathic VTs because of their focal origin in sites such as the right or left ventricular outflow tracts, epicardium, or papillary muscle, should be considered as a preferred alternative to long-term antiarrhythmic drug therapy.9 The longterm success rate of catheter ablation for these focal sites can be about 85% or even higher. Studies with smaller groups of patients have reported even higher rates of success for idiopathic VT arising in the sinuses of Valsalva. By comparison to the relatively benign prognosis of idiopathic sustained VT, idiopathic VF accounts for 5 to 10% of all cases of sudden cardiac death (Chapter 63). The appropriate treatment for survivors of idiopathic VF is no different than for any other survivor of VF (i.e., ICD therapy). The ECG may show recurrent PVCs with short coupling intervals. If these PVCs are monomorphic, they may be amenable to catheter ablation. Catheter ablation may decrease the risk for recurrence but still does not obviate the need for ICD protection.

Ventricular Tachycardia and Ventricular Fibrillation with Structural Heart Disease

Sustained VT and VF in patients with organic heart disease have become a common indication for ICD therapy. Three randomized trials comparing ICD therapy to antiarrhythmic drug therapy all showed significant survival benefit with ICD over drug therapy. After an acute MI, an ICD reduces mortality in patients who have survived more than 40 days and have a left ventricular ejection fraction of 30% or less or who have symptomatic heart failure and an ejection fraction of less than 0.35%; and patients more than 5 days after MI who have a reduced ejection fraction, nonsustained VT, and inducible sustained VT or VF on electrophysiologic testing. A1  By comparison, ICDs do not

reduce mortality when routinely implanted soon after MI or in patients after recent coronary artery revascularization. A2  In patients with a chronically depressed ejection fraction of less than 30%, insertion of an ICD reduces the mortality rate by 20%, from 36% to 29%, over the next 5 years. A3  By comparison, amiodarone suppresses ventricular ectopy and reduces sudden death but does not appear to improve survival. A3  A4  However, there are no randomized placebo-controlled trials of antiarrhythmic drug therapy or catheter ablation for the secondary prevention of recurrent VT in patients with organic heart disease, probably because treatment of a potentially lethal arrhythmia with placebo has been considered unacceptable. As a result, antiarrhythmic drugs and catheter ablation currently serve  as palliative treatments to modify the course of VT in patients who receive  too many ICD shocks because of frequently recurrent or nearly incessant  VT. If medications are chosen, amiodarone (100 to 400 mg daily) is more successful than β-blockers or sotalol for palliative therapy of VT in patients with ICD. A5  However, sustained VT can be reproducibly initiated by focal electrophysiologic stimulation in 65% of the patients with chronic ischemic heart disease, and the identified site is usually amenable to treatment with catheter ablation. Among such patients who have an ICD for ischemic VT but experience recurrent VT, an average of about 50% of patients treated with catheter ablation will be free of VT for 2 years, A6  although some studies report even higher success rates. A7  These procedures should be undertaken only at institutions with the highest level of expertise and experience. In patients who are clinically unstable because of incessant VT despite an ICD, reversible precipitating factors should be sought and, if present, corrected promptly. Intravenous β-blockers (e.g., esmolol 50 to 300 µg/kg per minute) and amiodarone (0.5 to 1.0 mg per minute) constitute the first line of therapy. Intravenous lidocaine (e.g., 2 to 4 mg per minute) can be added, especially if myocardial ischemia is present, and emergent catheter ablation may be considered. ,

Torsades de Pointes

The first line of therapy for torsades de pointes in patients with long QT syndrome is β-blocker therapy (e.g., metoprolol 50 to 100 mg daily, atenolol 50 mg daily, or nadolol 40 mg daily, and titrated as tolerated), but its success is influenced by gender and the magnitude of QT prolongation, as well as by the specific genotype.10 In an acute setting with frequently recurrent torsades de pointes, magnesium sulfate (1 to 2 g intravenous infusion over 10 to 30 minutes) may be effective. An ICD is recommended if patients who are on β-blocker therapy develop recurrent syncope or documented torsades de pointes. Whether nonpharmacologic therapy, such as sympathetic ganglionic denervation, will prove effective in some cases is uncertain. β-Blockers are also the drugs of choice for catecholaminergic polymorphic VT, with ICD therapy recommended for patients with recurrent syncope or documented VT while on β-blocker therapy. Whether oral flecainide can obviate the need for ICD protection in such patients is uncertain.

Treatment of Genetically Acquired Ventricular Tachycardia and Ventricular Fibrillation

Most of the VTs observed in patients with ARVC may be reproducibly induced by programmed electrophysiologic stimulation and are amenable to catheter ablation. Recent clinical studies suggest that nearly half of these VTs have an epicardial site of origin, and simultaneous endocardial and epicardial catheter mapping and ablation may be the most effective method of treatment.11 However, catheter ablation cannot substitute for ICD therapy. In patients with hypertrophic cardiomyopathy, ICD therapy is routinely recommended in high-risk patients (Chapter 60). Case reports suggest a benefit of catheter ablation in highly selected patients, and amiodarone may sometimes be useful. For Brugada syndrome, catheter ablation of the electrophysiologically abnormal substrate in the right ventricular outflow tract epicardium can lead to eventual disappearance of the pathognomonic ST elevation on the ECG.12 Quinidine (600 to 900 mg daily in three or four divided doses) is the only antiarrhythmic drug that appears to be useful to treat this syndrome, but whether catheter ablation or quinidine will be a substitute for an ICD in the highest risk patients, who have unprovoked Brugada pattern in their resting ECG and have a history of syncope, is unproved.

Iatrogenic Ventricular Tachycardia

For sustained monomorphic VT in patients with surgically repaired congenital heart disease, catheter mapping and ablation are recommended. The techniques are similar to those used for catheter ablation of ischemic VT. Although sustained VT or VF during the course of LVAD therapy is usually well tolerated acutely, recurrent intractable VT or VF can result in right heart failure and in frequent ICD shocks, both of which may carry significant morbidity. Amiodarone (200 to 400 mg daily) and β-blocker therapy (e.g., metoprolol 100 to 200 mg daily) may be effective in at least rendering VT no longer incessant, and catheter ablation therapy has occasionally been tried in refractory VT with modest success.

PROGNOSIS

VT or VF in the early minutes or hours of acute MI has not been shown to affect the long-term prognosis in patients who survive to hospital discharge. Sustained monomorphic VT occurring after the hyperacute phase but within the next few days of an anterior wall MI portends poor prognosis, with about a seven-fold increase in subsequent mortality, because it usually occurs after the necrosis of a large amount of myocardium. The modern natural course of untreated VT and VF perhaps can best be assessed from prospective studies of patients receiving ICD therapy. Among patients who have had documented sustained VT, appropriate ICD therapy for VT or VF occurs in about 70% of patients within 2 years and 85% of patients within 3 years. For patients who have survived cardiac arrest, the rate is about 70% at 3 years. These figures underscore the high rate of recurrence of VT and VF in patients with organic heart disease presenting with sustained ventricular arrhythmias. The prognosis of patients with PVCs or nonsustained VT is less well known but is critically dependent on the underlying heart disease. No study to date has shown any survival benefit of treating PVCs or nonsustained VT in patients who do not have underlying organic heart disease and who do not develop a cardiomyopathy as a result of their ventricular ectopic activity. In very high-risk patients with ischemic heart disease, an ejection fraction of less than 40%, and electrically induced sustained VT, the 2-year rate of cardiac arrest or death from ventricular arrhythmia is more than 30%.

Grade A References A1. Goldenberg I, Gillespie J, Moss AJ, et al. Long-term benefit of primary prevention with an implantable cardioverter-defibrillator: an extended 8-year follow-up study of the Multicenter Automatic Defibrillator Implantation Trial II. Circulation. 2010;122:1265-1271. A2. Steinbeck G, Andresen D, Seidl K, et al. Defibrillator implantation early after myocardial infarction. N Engl J Med. 2009;361:1427-1436. A3. Piccini JP, Berger JS, O’Connor CM. Amiodarone for the prevention of sudden cardiac death: a meta-analysis of randomized controlled trials. Eur Heart J. 2009;30:1245-1253. A4. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med. 2005;352:225-237. A5. Connolly SJ, Dorian P, Roberts RS, et al. Comparison of beta-blockers, amiodarone plus betablockers, or sotalol for prevention of shocks from implantable cardioverter defibrillators: the OPTIC Study: a randomized trial. JAMA. 2006;295:165-171. A6. Kuck KH, Schaumann A, Eckardt L, et al. Catheter ablation of stable ventricular tachycardia before defibrillator implantation in patients with coronary heart disease (VTACH): a multicentre randomised controlled trial. Lancet. 2010;375:31-40. A7. Reddy VY, Reynolds MR, Neuzil P, et al. Prophylactic catheter ablation for the prevention of defibrillator therapy. N Engl J Med. 2007;357:2657-2665.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 65  Ventricular Arrhythmias  

GENERAL REFERENCES 1. Goldberger JJ, Basu A, Boineau R, et al. Risk stratification for sudden cardiac death: a plan for the future. Circulation. 2014;129:516-526. 2. Vallès E, Bazan V, Marchlinski FE. ECG criteria to identify epicardial ventricular tachycardia in nonischemic cardiomyopathy. Circ Arrhythm Electrophysiol. 2010;3:63-71. 3. Della Bella P, Brugada J, Zeppenfeld K, et al. Epicardial ablation for ventricular tachycardia: a European multicenter study. Circ Arrhythm Electrophysiol. 2011;4:653-659. 4. Betensky BP, Tschabrunn CM, Zado ES, et al. Long-term follow-up of patients with cardiac sarcoidosis and implantable cardioverter-defibrillators. Heart Rhythm. 2012;9:884-891. 5. Lin G, Dispenzieri A, Kyle R, et al. Implantable cardioverter defibrillators in patients with cardiac amyloidosis. J Cardiovasc Electrophysiol. 2013;24:793-798. 6. O’Mahony C, Jichi F, Pavlou M, et al. A novel clinical risk prediction model for sudden cardiac death in hypertrophic cardiomyopathy (HCM Risk-SCD). Eur Heart J. 2014;35:2010-2020.

374.e1

7. Abrams DJ, Macrae CA. Long QT syndrome. Circulation. 2014;129:1524-1529. 8. Pedersen CT, Kay GN, Kalman J, et al. EHRA/HRS/APHRS expert consensus on ventricular arrhythmias. Europace. 2014;16:1257-1283. 9. Stevenson WG. Current treatment of ventricular arrhythmias: state of the art. Heart Rhythm. 2013;10:1919-1926. 10. Napolitano C, Bloise R, Monteforte N, et al. Sudden cardiac death and genetic ion channelopathies: long QT, Brugada, short QT, catecholaminergic polymorphic ventricular tachycardia, and idiopathic ventricular fibrillation. Circulation. 2012;125:2027-2034. 11. Bai R, Di Biase L, Shivkumar K, et al. Ablation of ventricular arrhythmias in arrhythmogenic right ventricular dysplasia/cardiomyopathy: arrhythmia-free survival after endo-epicardial substrate based mapping and ablation. Circ Arrhythm Electrophysiol. 2011;4:478-485. 12. Brugada P, Brugada J, Roy D. Brugada syndrome 1992-2012: 20 years of scientific excitement, and more. Eur Heart J. 2013;34:3610-3615.

374

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

66  ELECTROPHYSIOLOGIC INTERVENTIONAL PROCEDURES AND SURGERY DAVID J. WILBER

PACEMAKERS

Temporary Pacemaking

In emergencies such as asystolic cardiac arrest (Chapter 63), transcutaneous pacing with electrode pads applied to the chest wall occasionally can be lifesaving. Usually, however, time allows for temporary pacemaker leads to be inserted percutaneously, through an internal jugular or subclavian vein, and to be positioned and gently embedded in the right ventricular apex under fluoroscopic guidance. The lead is then attached to an external generator. A temporary pacemaker is often required as urgent therapy in a patient who has an indication for a permanent pacemaker and is awaiting that definitive procedure. Another indication for temporary pacing is the treatment of a

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

375

TABLE 66-1  CLASS I INDICATIONS* FOR IMPLANTATION OF A PERMANENT PACEMAKER SINUS NODE DYSFUNCTION Symptomatic sinus bradycardia Symptomatic chronotropic incompetence Symptomatic sinus bradycardia resulting from required drug therapy ATRIOVENTRICULAR (AV) BLOCK Third-degree and advanced second-degree AV block associated with symptomatic bradycardia Third-degree and advanced second-degree AV block in an awake patient with asystole of >3 seconds or an escape rate of 3 seconds in duration *Class I indications are conditions for which a pacemaker is indicated. Adapted from Gillis AM, Russo AM, Ellenbogen KA, et al. HRS/ACCF expert consensus statement on pacemaker device and mode selection. J Am Coll Cardiol. 2012;60:682-703.

Permanent Pacing

Permanent pacemaker leads can be inserted during cardiac surgery, but they much more frequently are inserted percutaneously through the subclavian vein or by cutdown through a cephalic vein. Ventricular leads are typically positioned in the right ventricular apex or, alternatively, higher on the right ventricular septum or outflow tract, and then are secured in place with a screw mechanism. Atrial leads are usually placed in the right atrial appendage (Fig. 66-1). Lithium iodide pacemaker batteries, which have a 7- to 8-year lifespan and weigh less than 30 g, typically are implanted subcutaneously in the infraclavicular region (E-Fig. 66-1). The programmability of many different parameters has become standard, as has the ability of the pacemaker to provide diagnostic and telemetric data.

Indications for Permanent Pacemaking

Pacemakers are implanted either to alleviate symptoms caused by bradycardia or to prevent severe symptoms in patients in whom symptomatic bradycardia is likely to develop (Tables 66-1 and 66-2).1-3 The most common bradycardia-induced symptoms are dizziness or lightheadedness, syncope or near-syncope (Chapters 51 and 62), exercise intolerance, and heart failure. Because these symptoms are nonspecific, documentation of an association between symptoms and bradycardia should be obtained before a pacemaker is recommended. If the bradycardia is persistent, such as in a patient with a complete AV block, a simple electrocardiogram may be sufficient to document the need for a pacemaker. If the bradycardia is intermittent, other diagnostic testing, such as 24-hour ambulatory monitoring, a continuous loop recorder, an implantable event monitor, or an electrophysiology test (Chapter 62), may be needed to document a relationship between symptoms and bradycardia. Even after a symptomatic bradycardia has been documented, however, a correctable cause for the bradycardia (Chapter 64) should be excluded before a pacemaker is implanted. Correctable causes of symptomatic bradycardia include hypothyroidism (Chapter 226), an overdose with drugs such as

TABLE 66-2  CLASS IIA INDICATIONS* FOR IMPLANTATION OF A PERMANENT PACEMAKER SINUS NODE DYSFUNCTION Heart rate of 40 beats per minute in an asymptomatic adult without cardiomegaly Asymptomatic second-degree infranodal AV block First-degree or second-degree AV block associated with symptoms similar to pacemaker syndrome Asymptomatic type II second-degree AV block with a narrow QRS CHRONIC BIFASCICULAR BLOCK Syncope, when other potential causes of syncope have been excluded An HV interval of ≥100 msec Pathologic pacing-induced infranodal AV block during electrophysiologic testing CAROTID SINUS SYNDROME Syncope without clear provocative events and with asystole of >3 seconds during carotid sinus pressure HIV = His-ventricle. *Class IIA indications are conditions for which a pacemaker is reasonable. Adapted from Tracy CM, Epstein AE, Darbar D, et al. 2012 ACCF/AHA/HRS focused update of the 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2012;60:1297-1313.

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

E-FIGURE 66-1.  Site of implantation of a permanent pacemaker battery and generator. (From Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.)

375.e1

376

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

digitalis, electrolyte disturbances, and medications such as β-adrenergic blocking agents (administered either orally or in the form of eye-drops for glaucoma), calcium-channel blocking agents, and antiarrhythmic medications (Chapter 64). At times, a pacemaker is necessary to allow continued treatment with a medication that is responsible for the bradycardia, such as in a patient in whom symptomatic sinus bradycardia develops after initiation of therapy with a β-adrenergic blocking agent for paroxysmal atrial fibrillation (AF) associated with a rapid ventricular response.

Pacing Modes

Pacing modes are described by a simple code. The first letter represents the chamber being paced (A for atrium, V for ventricle, D for dual chamber). The second letter identifies the chamber whose depolarizations are being sensed by the pacemaker (A, V, D, or O for no sensing). The third letter indicates whether the pacemaker is functioning in an inhibited (I) mode, a tracking (T) mode, in both modes (D), or asynchronously (O). The fourth letter designates whether the pacemaker can modulate the heart rate on its own, independent of the patient’s intrinsic atrial activity. An additional fifth letter may be used to define the pacemaker’s ability to provide antitachycardia pacing (P), to deliver shocks (S), or both (D). The most appropriate pacing mode must be determined for each individual. By far the most common permanent pacing modes now used in the United States are DDD (pacing and sensing of the atrium and ventricle in both inhibited and tracking fashion) and DDDR (with the additional ability to adjust the atrial rate independently in patients with a poor intrinsic heart rate response to exercise). In patients with sinus node dysfunction, atrial or dual-chamber pacing significantly reduces the risk for AF A1  and improves quality of life A2  compared with ventricular pacing. Although atrial pacing alone is an option for younger active patients with normal AV conduction, the high risk that these patients will develop symptomatic AV block makes initial dual-chamber pacing attractive. In patients with AV block, dual-chamber pacing improves quality of life, reduces the risk for developing AF, and avoids the 25% risk for developing the pacemaker syndrome, which consists of symptoms of weakness, lightheadedness, exercise intolerance, or palpitations owing to the absence of AV synchrony during ventricular pacing. This syndrome is treated by restoring AV synchrony with atrial-based pacing modes, which would require an additional procedure to implant an atrial lead if a dual-chamber pacemaker were not originally placed. For these reasons, current consensus guidelines recommend dual-chamber pacing for most patients with sinus node dysfunction or AV block. In patients who have paroxysmal AF and dual-chamber pacemakers, the ventricular rate will attempt to track the rapid atrial rates during the arrhythmia. Mode-switching pacemakers can pace in the DDD mode during sinus rhythm and automatically switch to rate-responsive ventricular pacing during AF or other supraventricular arrhythmias (Fig. 66-2). In patients who have long-standing persistent AF and in whom further attempts to restore sinus rhythm are not planned, there is no indication for atrial pacing or for the placement of an atrial lead. An exception to the recommendation for dual-chamber pacing is in patients who have chronic AF with occasional symptomatic pauses. VVIR pacing is recommended to protect against the pauses and to provide a normal rate response to exercise if needed.

Another option for patients with AV block is biventricular pacing. In patients with AV block, class I to III heart failure, and left ventricular systolic dysfunction, biventricular pacing is superior to conventional right ventricular pacing with an insignificant 17% reduction in death but a larger reduction in severe heart failure. A3 

Complications of Pacemakers

About 1 to 2% of patients develop complications from the implantation procedure itself, including pocket hematoma, pneumothorax, perforation of the atrium or ventricle, lead dislodgement, subclavian vein thrombosis, and infection. A strategy of continued warfarin treatment at the time of implantation of a pacemaker or an implantable cardioverter-defibrillator (ICD) markedly reduces the incidence of clinically significant device-pocket hematoma compared with bridging therapy with heparin. A4  The subcutaneous pocket may develop a hematoma or local infection.4 Pacemaker infections typically involve primarily the subcutaneous pacemaker pocket, but long-term resolution of infection generally requires removal of both the pulse generator and leads as well as long-term antibiotic therapy (Chapter 76). If the device becomes infected, nearly 40% of patients have coexisting valve involvement, predominantly tricuspid valve infection (Chapter 76), with mortality rates as high as 15% in the hospital and 20 to 25% at 1 year.4 During long-term follow-up after pacemaker implantation, potential problems include failure to pace, failure to capture, and changes in the pacing rate. These problems may be a manifestation of suboptimal programming, fracture of a lead or a break in its insulation, generator malfunction, or battery depletion. Ventricular pacing, particularly from the right ventricular apex, is associated with a delayed and abnormal activation sequence, and interventricular and intraventricular mechanical dyssynchrony. When more than 40% of heart beats are the result of ventricular pacing, even in dual-chamber pacing modes, patients can develop adverse ventricular remodeling with ventricular dilation, systolic dysfunction, altered myocardial metabolism, and functional mitral regurgitation. Clinically, such patients are at greater risk for developing AF and heart failure. To avoid those complications, every attempt should be made to minimize the amount of ventricular pacing, including programming longer AV delays (220 to 250 msec) or implanting pacemakers with algorithms that minimize the cumulative percentage of ventricular pacing. In patients with bradycardia and a left ventricular ejection fraction of 35% or less, cardiac resynchronization therapy should be considered if significant right ventricular pacing (>40% of heart beats) is anticipated.

TRANSTHORACIC CARDIOVERSION AND DEFIBRILLATION

Techniques

Defibrillators generate and then discharge an electrical current across two paddle electrodes. The resulting shock simultaneously depolarizes large portions of the atria or ventricles, thereby terminating re-entrant circuits and extinguishing re-entrant arrhythmias that rely on such circuitry (Chapters 61, 64, and 65). Synchronization with the QRS complex (cardioversion) is always advised in patients with either a supraventricular tachycardia (SVT) or ventricular tachycardia (VT) because a nonsynchronized shock coincident with the T wave may precipitate ventricular fibrillation (VF). If a shock is needed

A V

A V

* B 1 sec FIGURE 66-2.  Rhythm strips from a Holter monitor in a patient with complete atrioventricular block, sinus bradycardia, paroxysmal atrial fibrillation, and a rate-responsive dual-chamber pacemaker with mode-switching capability. A, When the patient is in sinus rhythm, the pacemaker functions in a DDDR mode, with synchronized atrial and ventricular pacing at 105 beats per minute while the patient is walking. B, At the onset of an episode of atrial fibrillation, there is tracking of the atrium that results in ventricular pacing at 140 beats per minute, which is the upper rate limit of the pacemaker. Within 2 seconds (asterisk), the mode-switch feature results in VVIR pacing, and the ventricular pacing rate gradually falls to 70 beats per minute, which is the lowest rate limit of the pacemaker. A = atrial stimulus; V = ventricular stimulus. (Courtesy of Dr. Fred Morady.)

377

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

to terminate VF, however, this defibrillation does not require synchronization to the QRS complex. The success of cardioversion or defibrillation is affected by the shock waveform and shock strength. Biphasic shock waveforms are recommended because they are significantly more effective than monophasic waveforms at equivalent energies. Other technique-dependent variables that maximize the energy delivered to the heart include increasing paddle pressure, delivery of the shock during expiration, and repetitive shocks. Patient-related factors that may decrease the probability of successful cardioversion and defibrillation include metabolic disturbances, a longer duration of arrhythmia, and higher body weight.

Indications and Technique

The most common arrhythmias treated by cardioversion and defibrillation are VF, VT, AF, and atrial flutter (Chapters 63, 64, and 65). Treatment of VF is always an emergency: a 200-J defibrillation shock should be delivered emergently, followed by one or more 360-J shocks if necessary. Cardioversion of VT may be a life-saving emergency procedure, similar to defibrillation for VF, or an urgent but controlled procedure with an initial shock strength of 50 to 100 J followed by higher energy shocks if needed. For AF, cardioversion is usually an elective procedure, with an initial shock of 200 J in adults, followed by shocks of 300 to 360 J if necessary. For cardioversion of atrial flutter, an initial shock of 50 to 100 J is appropriate. Regardless of the underlying arrhythmia, the energy required is a probability function and not a discrete value, so subsequent shocks may be effective for successful cardioversion or defibrillation even if the first 360-J shock is not effective. Elective cardioversion requires fasting for at least 8 hours, a reliable catheter in a peripheral vein, oxygen, suction, and equipment for potential emergency airway management. Patients are premedicated (Chapter 432), usually with propofol. In the anteroposterior configuration, which may be more effective for initial cardioversion of AF, one electrode is positioned to the left of the sternum at the fourth intercostal space, with the second electrode placed posteriorly, to the left of the spine, at the same level as the anterior electrode. In the anteroapical configuration, one electrode is placed to the right of the sternum at the level of the second intercostal space, and the second electrode is placed at the mid-axillary line, lateral to the apical impulse.

Precautions and Complications

Cardioversion of AF (Chapter 64) may be complicated by thromboembolism. If no atrial thrombi are seen on a transesophageal echocardiogram, preprocedure anticoagulation is not necessary. Otherwise, anticoagulation is necessary for 3 weeks before elective cardioversion. All patients should be anticoagulated for 1 month after cardioversion if AF has been present for 48 hours or longer.5 VF may rarely occur even when shocks are synchronized to the QRS complex. The risk for post-shock ventricular arrhythmias is increased in patients with electrolyte disturbances and digitalis toxicity, so elective cardioversion should be delayed in such patients. Many patients develop elevations of serum troponin levels, sometimes with transient ST segment elevation, after cardioversion, especially if higher energies were delivered in a short period of time, but clinical myocardial dysfunction is rare. Post-shock bradycardia or asystole, which may occur because of vagal discharge or an underlying sick sinus syndrome, sometimes can require atropine or emergency transcutaneous pacing. If a patient has a pacemaker or ICD, the shocking electrodes should be placed as far away from the generator as possible, and both the generator and pacing threshold should be checked after the procedure.

OTHER IMPLANTABLE DEVICES: CARDIOVERTER-DEFIBRILLATORS AND CARDIAC RESYNCHRONIZATION THERAPY

Implantable Cardioverter-Defibrillators ICD Pulse Generators and Leads

The procedures for implanting ICDs are analogous to those used for permanent pacemakers. The 60-g pulse generators are similarly implanted subcutaneously in the infraclavicular area. ICDs deliver biphasic shocks at strengths of less than 1 to 42 J while recording the electrogram during the arrhythmia and its treatment. They also can provide antitachycardia overdrive pacing as well as dual-chamber antibradycardia pacing. A developing alternative is the totally subcutaneous, implantable ICD.6 With this device, defibrillation is achieved by current flowing between a pulse

Pulse generator

FIGURE 66-3.  Schematic of pulse generator and lead position for the subcutaneous

defibrillator.

generator implanted in the axilla and a subcutaneous coil implanted parallel and just lateral to the sternum (Fig. 66-3). The pulse generator must deliver higher stored energy (80 J) compared with transvenous defibrillation to ensure an adequate margin of safety for defibrillation, but initial data suggest that the device’s effectiveness for sensing and terminating VF is comparable to transvenous systems. The device is not suitable for patients with concomitant bradycardia, in those with indications for cardiac resynchronization therapy, or in patients in whom frequent antitachycardia pacing is needed or anticipated. However, it is likely to become an attractive option for patients who have had prior complications from transvenous systems (e.g., vein thrombosis, infection) and for primary prevention in patients in whom frequent shocks are not anticipated.

Indications

ICD therapy initially evolved as secondary prevention to treat patients who had survived an episode of ventricular fibrillation or hemodynamically unstable ventricular tachycardia and were at high risk for subsequent death owing to recurrent ventricular arrhythmias.6 Based on multiple randomized trials and careful risk-to-benefit analyses, ICDs now are also implanted as primary prevention in individuals at high risk for a first cardiac arrest (Table 66-3), including patients who have idiopathic dilated cardiomyopathy (Chapter 60) and unexplained episodes of syncope; patients who have dilated ischemic or nonischemic cardiomyopathy with an ejection fraction of 35% or less and class II or III heart failure; patients who have coronary artery disease, an ejection fraction or 35% or less, spontaneous episodes of nonsustained VT, and inducible sustained VT in the electrophysiology laboratory (Chapter 65); patients who have had a previous myocardial infarction and now have an ejection fraction of less than 30% (Chapter 73); and selected patients with conditions such as hypertrophic cardiomyopathy, Brugada syndrome, and long-QT syndrome (Chapters 60 and 65).7,8 Further refinements in these criteria are to be expected over the coming years as data from ongoing and planned trials become available.

Programming of ICDs

A fundamental goal of ICD implantation and lead configuration is to deliver sufficient energy to the ventricular myocardium to ensure reliable defibrillation. The most common pathway for this energy is from the implanted pectoral pulse generator to a coil electrode on the distal portion of the transvenous lead that is positioned in the right ventricle. The energy required for successful defibrillation is probabilistic, and the relationship between successful defibrillation and energy of the shock is a sigmoidal curve, with an intermediate zone in which the success rate of any single defibrillation attempt is variable. Most transvenous ICD pulse generators can store and deliver 30 to 40 J, which exceeds the typical defibrillation energy threshold of 10 to 15 J, thereby allowing a substantial margin of safety for defibrillation. If the

378

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

standard lead configuration is insufficient for reliable defibrillation, alternative configurations can be tested at the time of implantation. ICDs can perform a variety of functions beyond defibrillation. A substantial proportion of ICD recipients have sustained monomorphic VT that can be painlessly terminated by appropriately timed, overdrive antitachycardia pacing. ICDs can be programmed to deliver different combinations of antitachycardia pacing rates and shocks depending on the rate of the spontaneous arrhythmia. ICDs also incorporate a variety of programmable detection algorithms designed to withhold unnecessary therapies for brief episodes of rapid supraventricular arrhythmias, especially sinus tachycardia or atrial fibrillation

TABLE 66-3  INDICATIONS FOR IMPLANTABLE CARDIOVERTER-DEFIBRILLATOR IMPLANTATION CLASS I*: SECONDARY PREVENTION Cardiac arrest survivor (VF or unstable sustained VT not associated with a completely reversible cause) Spontaneous sustained VT (irrespective of stability) and structural heart disease Syncope of unknown origin associated with clinically relevant and hemodynamically significant sustained VT or VT induced at electrophysiologic testing CLASS I*: PRIMARY PREVENTION Prior myocardial infarction (≥40 days), LVEF < 35%, NYHA class II or III Prior myocardial infarction (≥40 days), LVEF < 30%, NYHA class I Nonischemic dilated cardiomyopathy, LVEF ≤ 35%, NYHA class II or III Prior myocardial infarction, LVEF < 40%, spontaneous NSVT, inducible sustained VT or VF at electrophysiologic testing CLASS IIA† Nonischemic dilated cardiomyopathy, significant LV dysfunction, and syncope Sustained VT and normal or near-normal ventricular function Hypertrophic cardiomyopathy and one or more risk factors for sudden death Arrhythmogenic right ventricular cardiomyopathy and one or more risk factors for sudden death Long QT syndrome with syncope or VT despite β-blocker therapy Nonhospitalized patients awaiting cardiac transplantation Brugada syndrome and syncope or VT Catecholaminergic polymorphic VT and syncope or VT on β-blocker therapy Cardiac sarcoidosis, giant cell myocarditis, or Chagas disease *Class I indications are conditions for which an implantable cardioverter-defibrillator is indicated. † Class IIA indications are conditions for which an implantable cardioverter-defibrillator is reasonable. LVEF = left ventricular ejection fraction; NSVT = nonsustained ventricular tachycardia, VF = ventricular fibrillation; VT = ventricular tachycardia. Adapted from Tracy CM, Epstein AE, Darbar D, et al. 2012 ACCF/AHA/HRS focused update of the 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2012;60:1297-1313.

(Fig. 66-4). In addition, many ventricular arrhythmias, particularly at lower rates, will self-terminate within seconds, so algorithms designed to delay cardioversion for a few seconds can reduce unnecessary shocks. Optimal programming reduces the patient’s discomfort, maximizes the life of the pulse generator battery, and reduces the potential adverse effects of unneeded shocks on left ventricular function. For example, optimized programming (incorporating higher rate cutoffs, detection delay, rhythm discriminators, and antitachycardia overdrive pacing can reduce inappropriate device shocks by 50% and mortality by 30% compared with standard ICD programming, without increasing the risk for syncope. A5  A6  Many athletes with ICDs can engage in vigorous and competitive sports without physical injury or failure to terminate the arrhythmia.9 ,

Cardiac Resynchronization Therapy

Cardiac resynchronization therapy (CRT) requires the transvenous placement of electrodes into the right ventricle and into the coronary venous system for synchronous pacing of both ventricles (Fig. 66-5).

Indications for CRT

Abnormal and prolonged ventricular activation, as indicated by QRS prolongation of more than 120 msec on the electrocardiogram (ECG), contributes to poorly coordinated dyschronous activation of the left ventricle, thereby resulting in spontaneously reduced systolic function as would occur iatrogenically with right ventricular pacing. Over time, these abnormal electrical patterns can lead to progressive ventricular remodeling with dilation, further impairment of systolic function, and new or worsening heart failure (Chapter 58). Implantation of an additional ventricular lead, typically into a posterolateral coronary vein, permits pacing of the region of latest left ventricular activation. Synchronizing the timing between pacing of the right ventricular septum and the left ventricular free wall promotes more synchronous ventricular contraction and results in “reverse remodeling” with a reduction in ventricular volumes, improved systolic function, and clinical improvement in heart failure (Chapter 59).10 In clinical trials of patients with reduced systolic function, prolonged QRS duration, and class III or ambulatory class IV heart failure, the addition of CRT to guideline-directed medical therapy is associated with an approximately 30% reduction in hospitalizations and a 24 to 36% reduction in total mortality. A7  Functional improvement in heart failure symptoms and quality of life is seen in a majority of patients, but 30 to 40% of patients may not improve symptomatically. Superior outcomes are associated with longer QRS delays and a left bundle branch block QRS configuration.2 Although a prolonged QRS duration is typically associated with mechanical dyssynchrony as demonstrated by cardiac imaging, there is no evidence that imaging-identified mechanical dyssynchrony alone, in the absence of QRS prolongation, identifies patients who will improve with CRT. Clinical trials have also established that the benefit of CRT extends to patients with less severe heart failure symptoms (class I or II), in whom CRT

A

B

C 1 sec FIGURE 66-4.  Examples of stored electrograms obtained several hours after three different patients had experienced a flurry of shocks from an implantable cardioverterdefibrillator and showing the rhythm recorded by the device immediately before a shock was delivered. A, In this patient, the stored electrogram demonstrates ventricular tachycardia at a rate of 300 beats per minute, thus indicating that the shock was appropriate. He was treated with amiodarone to reduce the frequency of episodes of ventricular tachycardia. B, This patient received shocks because of paroxysmal supraventricular tachycardia at a rate of 206 beats per minute, which exceeded the programmed rate cutoff of 170 beats per minute. He underwent radio frequency ablation of the paroxysmal supraventricular tachycardia and received no further inappropriate shocks. C, The stored electrograms in this patient indicate that the patient received inappropriate shocks that were triggered by atrial fibrillation at a rate of 180 beats per minute. The rate cutoff of the device in this patient was 150 beats per minute. This patient was treated with a β-blocker to keep the ventricular rate less than 150 beats per minute during atrial fibrillation. (Courtesy of Dr. Fred Morady.)

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

379

FIGURE 66-5.  Typical position of the left ventricular pacing lead in a posterolateral branch of the coronary sinus in a patient with a cardiac resynchronization device. Note the presence of a defibrillator coil on the distal right ventricular lead, indicating that the device is also capable of defibrillation (CRT-D).

appears to delay the onset of symptomatic heart failure and significantly reduce heart failure events over the next 1 to 7 years if the ejection fraction is 30% or less and the QRS duration is more than 130 msec, especially in patients with left bundle branch block. A8  Current indications for CRT therapy (Table 66-4) likely will continue to evolve as additional evidence accrues. In patients who meet criteria for both CRT and ICD implantation, some evidence suggests that implantation of devices with both functions (CRT-D) may provide additional mortality benefit.3 However, CRT alone may be appropriate for some patients with more advanced heart failure, extensive comorbidity, and limited life expectancy, in whom the primary objective is symptomatic improvement.

Complications

Many of the complications related to ICD and CRT implantation procedures are somewhat more frequent but are similar to those associated with pacemakers. Major procedure-related complications include pneumothorax, myocardial perforation, and infection, all of which should have an incidence of less than 1%. The approach to infection is similar as for permanent pacemakers. Overall, major complications occur in 2 to 3% of new ICD implants and are more common during pulse generator replacement (5 to 6%). Longterm complications are primarily related to infection and lead failure. Given their larger size and complexity, ICD leads are more likely to fail (1 to 4% annually) than are pacemaker leads (40%) ventricular pacing *Class I indications are conditions for which CRT is indicated. † Class IIA indications are conditions for which CRT is reasonable. CRT = cardiac resychronization therapy; LBBB = left bundle branch block; LVEF = left ventricular ejection fraction; NYHA=New York Heart Association (functional class). Adapted from Brignole M, Auricchio A, Baron-Esquivias G, et al. 2013 ESC Guidelines on cardiac pacing and cardiac resynchronization therapy: the Task Force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association (EHRA). Eur Heart J. 2013;34:22812329; and Tracy CM, Epstein AE, Darbar D, et al. 2012 ACCF/AHA/HRS focused update of the 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2012;60:1297-1313.

of the clinical arrhythmia during diagnostic electrophysiologic testing (Chapter 62) is critical and can be facilitated by careful selection of pacing sites, judicious use of sedation, and intravenous infusion of catecholamines. Alternatively, the target site can sometimes be selected based on specific anatomic landmarks or tissue characteristics identified during sinus rhythm.

Tissue Effects of Applied Energy

Radio frequency current, typically in the range of 300 to 750 kHz, is the most common form of energy used in catheter ablation. The energy is applied in a unipolar fashion between a small electrode in contact with the targeted myocardium and a large dispersive cutaneous patch electrode placed on the back. The small electrode area at the myocardial interface results in a high-density current and rapid resistive heating in the myocardium that is immediately subjacent to the electrode, with slower conductive heating of deeper myocardial layers. A tissue temperature greater than 60° C is required for irreversible myocyte injury. Saline irrigation of the electrode reduces heating at the tissue-electrode interface, moves the zone of maximal heating deeper into the tissue, and results in larger, deeper lesions. Among other energy sources,

380

CHAPTER 66  Electrophysiologic Interventional Procedures and Surgery  

cryoablation is the most widely used, whereas microwave, laser, and ultrasound have limited applications at present.

Radio Frequency Ablation of Supraventricular Tachycardias

Radio frequency ablation is the recommended first-line treatment for paroxysmal SVT, Wolff-Parkinson-White (WPW) syndrome, or type 1 (typical) atrial flutter that is symptomatic enough to warrant therapy (Chapter 64). For atrial flutter other than type 1 and for inappropriate sinus tachycardia, an ablation procedure is recommended only in patients who have significant symptoms and recurrences despite antiarrhythmic medications. AV nodal re-entrant tachycardia (Chapter 64), which is the most common type of paroxysmal SVT, is successfully eliminated in 98% of cases (with a 198 mg/dL) ESTABLISHED CARDIOVASCULAR OR RENAL DISEASE • • • • • •

Stroke or TIA CAD: myocardial infarction, angina, myocardial revascularization Heart failure (with decreased or preserved ejection fraction) Intermittent claudication (symptomatic peripheral artery disease) Chronic kidney disease with eGFR < 30 mL/min/1.73m2 Advanced retinopathy: hemorrhages or exudates, papilledema

BMI = body mass index; CAD = coronary artery disease; ECG = electrocardiogram; eGFR = estimated glomerular filtration rate; TIA = transient ischemic attack. Adapted from Mancia G, fa*gard R, Narkiewicz K, et al. 2013 ESH/ESC Guidelines for the management of arterial hypertension: the Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens. 2013;31:1281-1357.

385

In patients with mild (stage 2: GFR of 60 to 90 mL/minute per 1.73 m2) or moderate (stage 3: GFR of 30 to 60 mL/minute per 1.73 m2) proteinuric chronic kidney disease, stringent blood pressure control is important both to slow the progression to end-stage renal disease and to reduce the excessive cardiovascular risk. In patients with severe chronic kidney disease, hypertension often becomes difficult to treat and may require either (1) intensive medical treatment with loop diuretics, potent vasodilators (e.g., minoxidil), high-dose β-adrenergic blockers, and central sympatholytics; or (2) initiation of chronic hemodialysis as the only effective way to reduce plasma volume. In chronic hemodialysis patients, the challenge is to control interdialytic hypertension without exacerbating dialysis-induced hypotension. The annual mortality rate in the hemodialysis population is 25%; half of this excessive mortality is caused by cardiovascular events that are related, at least in part, to hypertension.

RENOVASCULAR HYPERTENSION PATHOBIOLOGY AND CLINICAL MANIFESTATIONS

The two main causes of renal artery stenosis (Chapter 125) are atherosclerosis (85% of cases), typically in older persons with other clinical manifestations of systemic atherosclerosis, and fibromuscular dysplasia (15% of cases), typically in young women who are otherwise healthy. Although renal artery stenosis and hypertension frequently coexist, the presence of a renal artery stenosis proves neither that the patient’s hypertension is renovascular in origin nor that revascularization will improve renal perfusion and blood pressure. Unilateral renal artery stenosis can lead to underperfusion of the juxtaglomerular cells, thereby causing renin-dependent hypertension even though the contralateral kidney is able to maintain normal blood volume. In contrast, bilateral renal artery stenosis (or unilateral stenosis with a solitary kidney) constitutes a potentially reversible cause of progressive renal failure and volume-dependent hypertension. The following clinical clues increase the suspicion of renovascular hypertension: any hospitalization for urgent or emergent hypertension; recurrent “flash” pulmonary edema; recent worsening of long-standing, previously well-controlled hypertension; severe hypertension in a young adult or after the age of 50 years; precipitous and progressive worsening of renal function in response to angiotensin-converting enzyme (ACE) inhibition or angiotensin II receptor blockade; unilateral small kidney by any radiographic study; extensive peripheral arteriosclerosis; and a flank bruit.

DIAGNOSIS

Contrast-enhanced computed tomography (CT) and magnetic resonance angiography are the preferred screening tests for renal artery stenosis, but gadolinium-enhanced magnetic resonance imaging (MRI) is contraindicated in patients with advanced chronic kidney disease to avoid potentially fatal gadolinium-induced nephrogenic systemic fibrosis (Chapter 267). Fibromuscular dysplasia classically causes a “string of beads” lesion in the midportion of a renal artery (Fig. 67-5A), whereas atherosclerotic renal artery lesions

TABLE 67-5  GUIDE TO EVALUATION OF IDENTIFIABLE CAUSES OF HYPERTENSION SUSPECTED DIAGNOSIS

CLINICAL CLUES

DIAGNOSTIC TESTING

Chronic kidney disease

Estimated GFR < 60 mL/min/1.73 m2 Urine albumin-to-creatinine ratio ≥ 30 mg/g

Renal sonography

Renovascular disease

New elevation in serum creatinine, marked elevation in serum creatinine with ACE inhibitor or ARB, drug-resistant hypertension, flash pulmonary edema, abdominal or flank bruit

Renal sonography (atrophic kidney), CT or MR angiography, invasive angiography

Coarctation of the aorta

Arm pulses > leg pulses, arm BP > leg BP, chest bruits, rib notching on chest radiography

MR angiography, TEE, invasive angiography

Primary aldosteronism

Hypokalemia, drug-resistant hypertension

Plasma renin and aldosterone, 24-hour urine aldosterone and potassium after oral salt loading, adrenal vein sampling

Cushing syndrome

Truncal obesity, wide and blanching purple striae, muscle weakness

1 mg dexamethasone-suppression test, urinary cortisol after dexamethasone, adrenal CT

Pheochromocytoma

Paroxysms of hypertension, palpitations, perspiration, and pallor; diabetes

Plasma metanephrines, 24-hour urinary metanephrines and catecholamines, abdominal CT or MR imaging

Obstructive sleep apnea

Loud snoring, large neck, obesity, somnolence

Polysonography

ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; BP = blood pressure; CT = computed tomography; GFR = glomerular filtration rate; MR, magnetic resonance; TEE = transesophageal echocardiography.

386

CHAPTER 67  Arterial Hypertension  

Fibromuscular Dysplasia

Atherosclerosis

“String of beads” Proximal stenosis

A

B

FIGURE 67-5.  Computed tomographic angiogram with three-dimensional reconstruction. A, The classic “string of beads” lesion of fibromuscular dysplasia (bilateral in this patient). B, A severe proximal atherosclerotic stenosis of the right renal artery and mild stenosis of the left renal artery. (Images courtesy of Bart Domatch, MD, Radiology Department, University of Texas Southwestern Medical Center, Dallas, Texas.)

are proximal and discrete (Fig. 67-5B). Invasive renal angiography is the gold standard for confirming the diagnosis of renal artery stenosis.

TREATMENT  Balloon angioplasty is the treatment of choice for fibromuscular dysplasia, with overall favorable outcomes and at least one third of patients no longer needing any antihypertensive medications.7 In contrast, most atherosclerotic renal artery lesions do not cause hypertension or progressive renal failure, and most patients will not benefit from revascularization (balloon angioplasty or stenting), which carries substantial risks for serious complications. A1  A2  As a result, medical management of hypertension (with a regimen that includes a renin-angiotensin system inhibitor) and the associated atherosclerotic risk factors is first-line treatment for atherosclerotic renal artery stenosis, except that patients with truly drug-resistant hypertension, a progressive decline in renal function (ischemic nephropathy), or recurrent acute (“flash”) pulmonary edema may benefit from stent-based revascularization (Chapter 125). ,

MINERALOCORTICOID-INDUCED HYPERTENSION DUE TO PRIMARY ALDOSTERONISM PATHOBIOLOGY

The most common causes of primary aldosteronism (Chapter 227) are a unilateral aldosterone-producing adenoma and bilateral adrenal hyperplasia. Because aldosterone is the principal ligand for the mineralocorticoid receptor in the distal nephron, excessive aldosterone production causes excessive renal Na+-K+ exchange, often resulting in hypokalemia.

CLINICAL MANIFESTATIONS AND DIAGNOSIS

The diagnosis should always be suspected when hypertension is accompanied by either unprovoked hypokalemia (serum potassium concentration below 3.5 mmol/L in the absence of diuretic therapy) or a tendency to develop excessive hypokalemia during diuretic therapy (serum potassium concentration below 3.0 mmol/L). However, more than one third of patients do not have hypokalemia on initial presentation, and the diagnosis should also be considered in any patient with resistant hypertension. Screening for hyperaldosteronism should be restricted to the small fraction of hypertensive patients with hypokalemia or severe drug-resistant hypertension. If such patients have a positive screening test—a high serum

aldosterone level and a suppressed plasma renin activity level—and want to consider laparoscopic adrenalectomy, the patient should be referred to an experienced center for further evaluation: salt-loading to test for nonsuppressible aldosteronism and, if present, adrenal vein sampling to test for lateralization.

TREATMENT  Both CT and MRI have too many false-positive and false-negative results to be used as a noninvasive alternative to invasive adrenal vein sampling. Laparoscopic adrenalectomy and mineralocorticoid receptor blockade with eplerenone (50 to 100 mg per day) constitute highly effective therapeutic options that target the disease-causing mechanism with a favorable risk-to-benefit ratio.

MENDELIAN FORMS OF MINERALOCORTICOIDINDUCED HYPERTENSION

Almost all the rare mendelian forms of hypertension are mineralocorticoid induced and involve excessive activation of the epithelial Na+ channel (ENaC), the final common pathway for reabsorption of sodium from the distal nephron (E-Fig. 67-2). Thus, salt-dependent hypertension can be caused both by gain-of-function mutations of ENaC or the mineralocorticoid receptor and by increased production or decreased clearance of mineralocorticoid receptor ligands, which are aldosterone, deoxycorticosterone, and cortisol.

PHEOCHROMOCYTOMAS AND PARAGANGLIOMAS

Pheochromocytomas are rare catecholamine-producing tumors of the adrenal chromaffin cells, whereas paragangliomas are even rarer tumors of the extra-adrenal chromaffin cells (Chapter 228). The diagnosis should be suspected when hypertension is drug resistant or paroxysmal, particularly when accompanied by paroxysms of headache, palpitations, pallor, or diaphoresis. In some patients, pheochromocytoma is misdiagnosed as panic disorder. A family history of early-onset hypertension may suggest pheochromocytoma as part of the multiple endocrine neoplasia syndromes (Chapter 231). An increasing number of pheochromocytomas are being detected incidentally on abdominal imaging studies for nonadrenal indications. If the diagnosis is missed, outpouring of catecholamines from the tumor can cause unsuspected hypertensive crisis during unrelated surgical procedures, in which case mortality rates exceed 80%.

CHAPTER 67  Arterial Hypertension  

Distal collecting duct

Angiotensinogen Ang I

Cl– NCCT

386.e1

Renin

High BP

Ang II

WNK1 + 4 PHA2

Na+

GRA

17 α HD, 11βHD Aldosterone DOC deficiencies

Progesterone MR

Na+ ENaC

HEP

Cortisol

Liddle Syndrome

Na+ K+

11β -HSD2 AME Cortisone

ROMK Urine K+

WNK1 + 4 PHA2

Blood

Cortical collecting duct

E-FIGURE 67-2.  Mendelian forms of hypertension that cause mineralocorticoid-induced hypertension. AME = apparent mineralocorticoid excess; Ang = angiotensin; BP = blood

pressure; GRA = glucocorticoid-remediable aldosteronism; 17αHD and 11βHD = 17α- and 11β-hydroxylase deficiency; 11β-HSD2 = 11β-hydroxysteroid dehydrogenase type 2; DOC = deoxycorticosterone; ENaC = epithelial Na+ channel; HEP = hypertension exacerbated by pregnancy; MR = mineralocorticoid receptor; NCCT = sodium-chloride cotransporter; PHA2 = pseudohypoaldosteronism type 2; ROMK = rectifying outer medullary K+ channel; WNK = with no lysine kinases. See text for explanation. (Modified from Lifton RP, Gharavi AG, Geller DS. Molecular mechanisms of human hypertension. Cell. 2001;104:545-556.)

CHAPTER 67  Arterial Hypertension  

OTHER NEUROGENIC CAUSES

Other causes of neurogenic hypertension that can be confused with pheochromocytoma include sympathomimetic agents (cocaine, methamphetamine; Chapter 34), baroreflex failure, and obstructive sleep apnea (Chapter 100). A history of surgery and radiation therapy for head and neck tumors (Chapter 190) raises suspicion of baroreceptor damage. Snoring and somnolence suggest sleep apnea, but continuous positive airway pressure to treat the sleep apnea rarely improves blood pressure substantially (Chapter 100).

OTHER CAUSES OF SECONDARY HYPERTENSION

Coarctation of the aorta typically occurs just distal to the origin of the left subclavian artery, so the blood pressure is lower in the legs than in the arms (opposite of the normal situation) (see Fig. 69-7). The clue is that the pulses are weaker in the lower than in the upper extremities, indicating the need to measure blood pressure in the legs as well as in both arms. Intercostal collaterals can produce bruits on examination and rib notching on the chest radiograph. Coarctations can be cured with surgery or angioplasty. Hyperthyroidism tends to cause systolic hypertension with a wide pulse pressure, whereas hypothyroidism tends to cause mainly diastolic hypertension. Treatment is for the underlying disease. Hyperparathyroidism (Chapter 245) also has been associated with hypertension. Cyclosporine and tacrolimus are important causes of secondary hypertension in transplant recipients, apparently by inhibition of calcineurin, the calcium-dependent phosphatase that is expressed not only in lymphoid tissue but also in neural, vascular, and renal tissue. In the absence of outcomes data, nondihydropyridine calciumchannel blockers (CCBs) have become the drugs of first choice, but they increase cyclosporine blood levels. Combination therapy with diuretics, CCBs, and central sympatholytics often is required.

PREVENTION AND TREATMENT OF HYPERTENSION  At the population level, the primary prevention of hypertension requires large-scale societal changes, including further efforts to influence the food industry to reduce salt in processed foods, efforts to increase exercise, and the availability of fresh fruits and vegetables. After a person’s blood pressure rises to hypertensive or even prehypertensive levels, lifestyle modifications alone are almost never enough to return blood pressure to normal, and recidivism is typical. Although short-term pharmacologic therapy with a low-dose angiotensin receptor blocker (ARB) may prevent the conversion from prehypertension to full-blown hypertension, blood pressure quickly rises again if the ARB is discontinued. Thus, lifelong prescription medication is the cornerstone of effective therapy for primary hypertension, with lifestyle modification serving as a very important adjunct but not as an alternative. The objective is to reduce the blood pressure and associated metabolic abnormalities sufficiently to reduce the risk for cardiovascular events and end-stage renal disease without compromising the patient’s quality of life. Randomized trials have proved beyond any doubt that antihypertensive drug therapy reduces cardiovascular risk, with benefits that are proportional to the reduction in blood pressure achieved (E-Fig. 67-3). A3  However, in practice, most treated patients do not achieve the same low risk levels of truly normotensive persons because their blood pressures remain higher than optimal owing to the threshold levels of guidelines, hesitation of practicing physicians to start and intensify drug treatment, costs of medications, and medication noncompliance despite the declining costs of generic medications. This residual risk also may be attributable to the cardiovascular damage sustained before starting drug therapy. Multidrug regimens with two, three, or even more medications of different drug classes are almost always required to achieve currently recommended blood pressure goals. Low-dose drug combinations exert synergistic effects on blood pressure while minimizing dose-dependent side effects. For most patients with hypertension, lipid-lowering therapy (Chapter 206) is indicated as part of a comprehensive cardiovascular risk-reduction strategy (Chapter 52).

Lifestyle Modification

Lifestyle modification (Table 67-6) should be part of every antihypertensive regimen. A4  However, dietary and exercise interventions are difficult to sustain long term. For example, short-term trials have proved that individuals with prehypertension or stage 1 hypertension can lower their blood pressures on average by 6/3 mm Hg even without restricting calorie or sodium intake if they adhere to a diet rich in fresh fruits and vegetables and low-fat dairy products (www.nhlbi.nih.gov/files/docs/public/heart/dash_brief.pdf ). Modest dietary sodium restriction produces a further reduction in blood pressure and decreases cardiovascular disease risk.8 Sodium reduction is particularly effec-

387

TABLE 67-6  LIFESTYLE RECOMMENDATIONS TO LOWER BLOOD PRESSURE IN ADULTS WITH HYPERTENSION OR PRE-HYPERTENSION DIET 1. Adopt a diet that is: • High in vegetables, nuts, fruits, grains, low-fat dairy products, fish, poultry, etc. • Low in sweets, sugar-sweetened beverages, and red meats Adapt this dietary pattern to calorie requirements, personal/cultural food preferences, and medical conditions such as diabetes. 2. Lower sodium intake PHYSICAL ACTIVITY 1. Engage in three to four 40-minute sessions of moderate-to-intense aerobic physical activity per week. Adapted from Eckel RH, Jakicic JM, Ard JD, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63:2960-2984.

tive in black hypertensive patients. Most dietary sodium comes from processed foods, and patients should be taught to read food labels (6 g of NaCl = 2.4 g of sodium = 100 mmol of sodium). Moderately intense aerobic exercise programs can lower blood pressure by 2 to 5/1 to 2 mm Hg. The larger reductions in blood pressure are seen immediately after a bout of aerobic exercise (Chapter 16), but smaller reductions can persist for several hours. Relaxation and stress management techniques (e.g., meditation, biofeedback, breathing exercises) can decrease blood pressure transiently but generally produce little if any demonstrable effect on ambulatory blood pressure (Chapter 39). However, some individuals with overwhelming home or job strain or anger can benefit from cognitive behavior therapy and anxiolytics (Chapter 397). Blood pressure increases transiently by 10 to 15 mm Hg after each cigarette, so smokers of more than 20 cigarettes per day often have higher blood pressures out of the office than in the smoke-free medical office. Smokers should be counseled to quit tobacco (Chapter 32) not only because it raises blood pressure but also because it is such a potent risk factor for coronary heart disease, stroke, and the progression of hypertensive kidney disease. Blood pressure increases by up to 10 to 15 mm Hg with the first morning cup of coffee, but the pressor response to caffeine often habituates throughout the day. Thus, caffeine consumption need not be totally eliminated but may need to be reduced. Moderate alcohol (Chapter 33) consumption (one or two drinks per day) does not seem to increase the risk for hypertension in Western populations; but in Japanese populations, hypertension is more common in men who are moderate drinkers than in men who cannot drink because of a loss-of-function mutation in the alcohol dehydrogenase gene. In all populations, heavy drinking (three or more standard-sized drinks per day) and especially binge drinking activate the sympathetic nervous system the next day during withdrawal and are associated with an increased incidence and severity of hypertension, which is reversible if alcohol consumption decreases.

Antihypertensive Drugs

Although every hypertensive patient should adopt sensible lifestyle modifications, almost all will require medication to optimize outcomes. Lowering blood pressure with medication reduces but does not eliminate the risks for cardiovascular events, renal failure, and death.

Classes of Oral Antihypertensive Drugs

Multiple classes of oral antihypertensive drugs are approved by the U.S. Food and Drug Administration, although all have specific contraindications (Tables 67-7 and 67-8).

First-Line Drugs for Hypertension

Multiple practice guidelines1-4 recommend initiating drug treatment with one or more of three classes of first-line drugs (Fig. 67-6), which have additive or synergistic effects when used in combination: (1) CCBs, (2) reninangiotensin system blockers—either ACE inhibitors or ARBs, and (3) thiazidelike diuretics.

Calcium-Channel Blockers Mechanism of Action. The CCBs block the opening of voltage-gated

(L-type) Ca2+ channels in cardiac myocytes and vascular smooth muscle cells. The resultant decrease in the cytosolic Ca2+ signal decreases heart rate and ventricular contractility and relaxes vascular smooth muscle. Blood pressure lowering is related mainly to peripheral arterial vasodilation, with the rank

387.e1

CHAPTER 67  Arterial Hypertension 4.0

4.0

2.0

2.0

1.0 0.5

0.25

A

Risk ratio

Risk ratio

diltiazem ≫ verapamil. In contrast, for negative chronotropic and inotropic effects, the rank order of potency is verapamil ≫ diltiazem > dihydropyridines. Therapeutic Principles. The most recommended CCBs are amlodipine followed by diltiazem. Amlodipine’s long half-life permits once-daily dosing, and its costs are low since it became generic. Amlodipine is equivalent to a potent diuretic or lisinopril in protecting against nonfatal coronary events, stroke, and death, but it provides less protection against heart failure. A5  Unlike diuretics, ARBs, and ACE inhibitors, a high-salt diet or concurrent nonsteroidal anti-inflammatory drug (NSAID) therapy does not compromise the effectiveness of dihydropyridine CCBs. The CCBs have some diuretic action because

they dilate the afferent renal arteriole and may reduce requirements for additional diuretic therapy in mild hypertension. Amlodipine and other dihydropyridine CCBs are less renoprotective than ACE inhibitors or ARBs in patients with proteinuric chronic kidney disease. Such patients should not receive amlodipine as first-line therapy, but it may be useful as adjunctive therapy after initiation of appropriate first-line therapy with either an ACE inhibitor or ARB, as well as a diuretic. Diltiazem is a usually well-tolerated alternative in patients who cannot  tolerate amlodipine or would benefit from its other effects. Verapamil is not recommended because it is a weak antihypertensive medication and causes constipation.

390

CHAPTER 67  Arterial Hypertension  

Blood pressure 140/90 in adults aged >18 years (For age ≤80 years, pressure ≤150/90 or ≤140/90 if high risk [diabetes, kidney disease]) Start lifestyle changes (lose weight, reduce dietary salt and alcohol, stop smoking) Drug therapy (consider a delay in uncomplicated stage 1 patients)* Stage 1 140–159/90–99 Black patients

If needed, add… ACE-I or ARB or Combine CCB + thiazide If needed…

Stage 2 ≥160/100

Nonblack patients

CCB or thiazide

Start drug therapy (in all patients)

Age ≥60 years

Age 2-3) Old ischemic stroke (>3 months ago); intracerebral disease other than above Recent (10 minutes) cardiopulmonary resuscitation or internal bleeding Active peptic ulcer Recent noncompressible vascular punctures Pregnancy For streptokinase/anistreplase: prior exposure (especially if >5 days ago) or allergic reaction Modified from Kushner FG, Hand M, Smith SC Jr, et al. 2009 Focused updates: ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction and the ACC/ AHA/SCAI guidelines on percutaneous coronary intervention. Circulation. 2009;120:2271-2306.

mobilize and to treat patients quickly (3 hours), and for patients with greater risk of intracranial hemorrhage (age >70 years, female gender, therapy with hypertensive agents). In non–PCI-capable hospitals and in other situations in which PCI is not feasible or would be significantly delayed (e.g., by a long transfer time to a PCI-capable facility) to more than 120 minutes after first medical contact, fibrinolytic therapy should be given to patients with STEMI within 12 hours of the onset of symptoms unless it is contraindicated. Fibrinolytic therapy is reasonable within 12 to 24 hours of the onset of symptoms in the setting of a large MI or hemodynamic instability. Prehospital fibrinolysis followed by a routine emergent (i.e., within 1 to 2 hours) invasive strategy on hospital arrival causes a higher rate of in-hospital mortality, cardiac ischemic events, and strokes compared with primary PCI alone or by a more delayed invasive approach after fibrinolysis in stabilized patients and cannot be recommended.

448

CHAPTER 73  ST SEGMENT ELEVATION ACUTE MYOCARDIAL INFARCTION  

However, immediate transfer to a PCI-capable hospital for coronary angiography is recommended if such patients develop cardiogenic shock or severe heart failure and is reasonable in patients with suspected failed reperfusion or reocclusion after fibrinolytic therapy and even in hemodynamically stable patients with apparently successful reperfusion, provided angiography is delayed for at least 2 to 3 hours after fibrinolytic therapy.

Ancillary and Other Therapies Initial Medical Management

Aspirin (162 to 325 mg) should be given on presentation to all patients unless it is contraindicated (Fig. 73-4). A loading dose of an adenosine diphosphate receptor (P2Y12) inhibitor (i.e., clopidogrel, 600 mg, or prasugrel, 60 mg, or ticagrelor, 180 mg) also should be given as early as possible or at the time of primary PCI to STEMI patients for whom an invasive approach is planned. In addition, it is reasonable to start treatment with a GPIIb-IIIa receptor antagonist—abciximab (IV bolus of 0.25 mg/kg, then 0.125 µg/kg/minute [maximum, 10 µg/minute] for up to 12 hours), tirofiban (IV bolus of 25 µg/kg,

Recommendations for antiplatelet and anticoagulant therapy for STEMI Aspirin

Select Management Strategy

Invasive

Medical (Fibrinolysis)

Anticoagulant therapy: Pre-angiography: UFH At angiography: bivalirudin (or UFH)

Anticoagulant therapy: Enoxaparin or fondaparinux (UFH if renal dysfunction)

Antiplatelet therapy: Pre-angiography: prasugrel or ticagrelor (or clopidogrel) At angiography: may add GPI (selectively)

Antiplatelet therapy: Clopidogrel

FIGURE 73-4.  Recommendations for antiplatelet and anticoagulant therapy for ST segment elevation myocardial infarction (STEMI). See text for doses. GPI = glycoprotein IIb/IIIa inhibitor; UFH = unfractionated heparin.

then 0.15 µg/kg/minute for up to 12 to 18 hours; reduce infusion rate by 50% for estimated creatinine clearance < 30 mL/minute), or eptifibatide (IV bolus of 180 µg/kg, second bolus after 10 minutes, then 2.0 µg/kg/minute for up to 18 hours; reduce infusion by 50% for estimated creatinine clearance < 50 mL/ minute)—at the time of primary PCI for STEMI in selected patients, such as those with a large burden of thrombus or those who have not received an adequate loading dose of a P2Y12 inhibitor. The value of starting a GPIIb-IIIa receptor antagonist before arrival in the catheterization laboratory is less certain. Anticoagulant therapy should be initiated on presentation. Options include intravenous heparin (initial bolus of 60 IU/kg [maximum, 4000 IU], then 12 IU/ kg/hour [maximum, 1000 IU/hour] for patients >70 kg, adjusted to maintain activated partial thromboplastin time 1.5 to 2 times the control value), lowmolecular-weight heparin (LMWH; e.g., enoxaparin, IV bolus of 30 mg, then 1 mg/kg subcutaneously twice daily for patients 2.5

≤12

None required

Hyperdynamic

Normal or high

Anxious

>3

25

Avoid hypotensive agents; place intra-aortic balloon pump; urgent revascularization if possible

RV infarct

Very low

↑ JVP with clear lungs

90%) are caused by Enterococcus faecalis. The HACEK group of gram-negative organisms (Haemophilus spp., Aggregatibacter spp. [formerly Actinobacillus actinomycetemcomitans], Cardiobacterium hominis, Eikenella corrodens, and Kingella spp.) accounts for about 5% of endocarditis cases. Because these fastidious organisms usually grow in blood cultures within 7 days using current methods, prolonged incubation is no longer required to isolate HACEK strains. Many other gram-negative bacilli have been reported to cause infective endocarditis but are even more unusual. Traditionally, injection drug use has been regarded as the primary risk factor for enteric gram-negative bacterial endocarditis. However, recent experience from large multinational studies shows that health care contact, not injection drug use, is the most common risk factor for enteric gramnegative endocarditis. Fungal endocarditis is often difficult to diagnose and treat; it is most commonly found in patients with a history of injection drug use, recent cardiac valve surgery, or prolonged use of indwelling vascular catheters, especially those used for total parenteral nutrition. The most common fungi found in infective endocarditis are Aspergillus and Candida spp. Aspergillus (Chapter 339) rarely grows in blood cultures and must usually be cultured from a pathologic specimen (either an embolic site or vegetation); by contrast, Candida spp. (Chapter 338) frequently grows in blood cultures. Mortality is very high, and valve replacement surgery is usually necessary for fungal endocarditis.

CHAPTER 76  Infective Endocarditis  

Prosthetic valve endocarditis can be classified into one of two groups based on the time between valve surgery and disease onset: early (2 months) (Table 76-4). Staphylococci, particularly S. aureus, predominate during the early period, when most episodes of infective endocarditis are thought to be related to perioperative infection. In the late period, the spectrum of organisms becomes more akin to that of communityacquired native valve disease, in which S. aureus and viridans group streptococci predominate. Of note, among the coagulase-negative staphylococci, oxacillin resistance can be seen in these late cases. Staphylococcal species account for the large majority (≥70%) of implantable cardiac device infections. The prevalence of oxacillin resistance among S. aureus strains varies from study to study but is generally in the 30% to 50% range.

Endocarditis with Negative Blood Cultures

In most patients with infective endocarditis who have not received previous antibiotic therapy, every blood culture is positive because the bacteremia of endocarditis is continuous. Blood cultures are truly negative in fewer than 5% of cases of endocarditis; however, prior antibiotic administration may decrease the yield of blood cultures by up to 35%. Accordingly, most “culturenegative” cases of endocarditis occur in patients who have recently received antimicrobial agents. These cases are probably caused by the same organisms responsible for most native valve endocarditis; viridans group streptococci and the HACEK organisms are the most likely suspects because they are

TABLE 76-4  CAUSES OF PROSTHETIC VALVE ENDOCARDITIS* EARLY (2 mo POSTOPERATIVELY) Coagulase-negative staphylococci Staphylococcus aureus Viridans group streptococci Enterococci

*Listed in order of relative frequency. Adapted from Wang A, Athan E, Pappas PA, et al. Contemporary clinical profile and outcome of prosthetic valve endocarditis. JAMA. 2007; 297:1354-1361.

477

much more fastidious than staphylococci and enterococci and are therefore more likely to be affected by previous antibiotic administration. Ultimately, however, when blood cultures are negative and endocarditis is suspected, especially when a history of recent antimicrobial treatment is lacking, consideration should be given to fastidious organisms, fungi, and noncultivatable organisms (Table 76-5), particularly when the patient’s history suggests exposure to farm animals or unpasteurized milk (Coxiella burnetii, Brucella spp.), cats (Bartonella henselae), body lice (Bartonella quintana), or birds (Chlamydia psittaci). It is important to notify the microbiology laboratory that endocarditis is suspected because special culture techniques can increase the yield for the HACEK species, nutritionally variant streptococci (Abiotrophia and Granulicatella spp.), Brucella spp., Legionella spp., and some fungi. The traditional practice of holding blood cultures for 2 to 4 weeks is no longer required routinely. Specific serologic tests can diagnose endocarditis related to C. burnetii (the agent of Q fever), Brucella spp., Bartonella spp., and C. psittaci. Tropheryma whippelii, the etiologic agent in Whipple disease, and multiple other organisms may be diagnosed by polymerase chain reaction. Histopathologic features of resected tissue also can provide clues in the etiologic diagnosis of culture-negative endocarditis. If the search for a causative organism is fruitless, noninfectious causes such as marantic or Libman-Sacks endocarditis and atrial myxoma (Chapter 60) should be considered.

Laboratory Findings

Initial laboratory tests should include a complete blood count with differential, serum electrolytes, measurement of renal function, and urinalysis. Most patients with subacute infective endocarditis have the serum iron profile of anemia of chronic disease (Chapter 159). The white blood cell count is frequently elevated in acute infective endocarditis, particularly if S. aureus is the causative organism, but may not be elevated in more subacute forms. Microscopic hematuria is common, as is proteinuria. The chest radiograph is abnormal—demonstrating consolidation, atelectasis, pleural effusion, or clear septic emboli—in the overwhelming majority of patients with right-sided endocarditis. In others, it may show evidence of heart failure. The electrocardiogram (ECG) should be carefully examined for evidence of atrioventricular conduction blocks, especially a prolonged PR interval (see Figs. 64-5 through 64-9 in Chapter 64), suggestive of an aortic ring abscess or frank myocardial infarction (see Figs. 73-1 and 73-2 in Chapter 73). Rheumatoid factor, which is an ancillary test that has been included in the modified Duke criteria as a “minor criterion” in the category of “immunologic phenomenon,” may be positive in subacute or chronic endocarditis. Other ancillary tests, such as the erythrocyte sedimentation rate, the C-reactive

TABLE 76-5  ORGANISMS CAUSING “CULTURE-NEGATIVE” ENDOCARDITIS* ORGANISM

EPIDEMIOLOGY

DIAGNOSTIC TESTS

HACEK spp.

Mostly oral flora, so often history of periodontal disease

May require up to 7 days to grow

Nutritionally variant streptococci

Slow and indolent course

Supplemented culture media or growth as satellite colonies around Staphylococcus aureus streak

Coxiella burnetii (Q fever)

Worldwide; exposure to raw milk, farm environment, or rural areas

Serologic tests (high titers of antibody to both phase 1 and phase 2 antigens); also PCR on blood or valve tissue

Brucella spp.

Ingestion of contaminated milk or milk products; close contact with infected livestock

Bulky vegetations usually seen on echocardiography; blood cultures positive in 80% of cases with incubation time of 4-6 wk; lysis-centrifugation technique may expedite growth; serologic tests are available

Bartonella spp.

Bartonella henselae: transmitted by cat scratch or bite or by cat fleas Bartonella quintana: transmitted by human body louse; predisposing factors include homelessness and alcohol abuse

Serologic testing (may cross-react with Chlamydia spp.); PCR of valve or emboli is best test; lysis-centrifugation technique may be useful

Chlamydia psittaci

Exposure to birds

Serologic tests available, but must exclude Bartonella spp. because of cross-reactivity; monoclonal antibody direct stains on tissue may be useful; PCR now available

Tropheryma whippelii (Whipple disease)

Systemic symptoms include arthralgias, diarrhea, abdominal pain, lymphadenopathy, weight loss, CNS involvement; however, endocarditis may be present without systemic symptoms

Histologic examination of valve with PAS stain; valve cultures may be done using fibroblast cell lines; PCR on vegetation material

Legionella spp.

Contaminated water distribution systems; often nosocomial outbreaks; usually prosthetic valves

Lysis-centrifugation technique; also periodic subcultures onto buffered charcoal yeast extract medium; serologic tests and PCR available

Aspergillus and other noncandidal fungi

Prosthetic valve

Lysis-centrifugation technique; also culture and direct examination of any emboli

*Listed in approximate order of relative frequency. CNS = central nervous system; HACEK = Haemophilus spp., Aggregatibacter spp., Cardiobacterium hominis, Eikenella corrodens, and Kingella spp.; PAS = periodic acid–Schiff; PCR = polymerase chain reaction.

478

CHAPTER 76  Infective Endocarditis  

Infective endocarditis suspected

No or low risk from underlying cardiac condition, low clinical suspicion

Underlying cardiac risk factors or other high-risk features,* moderate to high clinical suspicion, or difficult imaging candidate

Initial TTE

Initial TEE

Negative

Positive

Negative

Positive

Therapy Increased suspicion during clinical course

Low suspicion persists

High-risk echo features†

Therapy

No high-risk echo features†

Look for other source of symptoms

High suspicion persists Negative

TEE

Negative Look for other source

TEE for detection of complications

Alternative diagnosis

Repeat TEE

No TEE unless clinical status deteriorates

Positive

Positive

Negative

Therapy

Therapy

Look for other source

*Initial high-risk features include prosthetic heart valves, many congenital heart diseases, previous endocarditis, new murmur, heart failure, or other stigmata of IE.

Positive

Follow-up TTE or TEE to reassess vegetations, complications, or therapeutic response as clinically indicated

†High-risk echocardiographic features include large or mobile vegetations, valvular insufficiency,

suggestions of perivalvular extension, or secondary ventricular dysfunction.

FIGURE 76-3. Algorithm for the diagnostic use of echocardiography (echo) in suspected cases of infective endocarditis (IE). TEE = transesophageal echocardiography; TTE = transthoracic echocardiography. (Adapted from Bayer AS, Bolger AF, Taubert KA, et al. Diagnosis and management of infective endocarditis and its complications. Circulation. 1998; 98:2936-2948.)

protein level, and the procalcitonin level, are generally not helpful in establishing an endocarditis diagnosis.

Echocardiography

Both transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE) (Chapter 55) are highly specific tests (≈98%) when used as part of the diagnostic evaluation of suspected endocarditis. By contrast, TEE has a much higher sensitivity (90%-95%) in this setting than TTE (48%63%). In most cases in which endocarditis is a serious diagnostic consideration, the evaluation should begin with TEE because negative TTE findings are not sensitive enough to exclude endocarditis (Fig. 76-3). Because TEE is the only relatively noninvasive means of detecting perivalvular extension of infection, any patient with a new conduction system abnormality or persistent fever—clinical predictors of perivalvular extension—should be evaluated with TEE. Likewise, TEE is strongly preferred in the evaluation of suspected prosthetic valve– or device-related endocarditis, although bland clots can occur on the leads of 5% to 10% of patients with intracardiac devices, and the finding of a “vegetation” on a lead is not specific for infection. The high sensitivity of TEE in detecting valvular vegetations on native valves also may be used in combination with clinical parameters (e.g., prompt resolution of bacteremia and defervescence) to support the clinical decision to abbreviate therapy in patients with vascular catheter–associated S. aureus bacteremia. A negative TEE result has a negative predictive value of about 95%. Nevertheless, when clinical suspicion of endocarditis is high and the initial TEE is negative, repeat TEE in 7 to 10 days may reveal the diagnosis. If TEE is unavailable, technically impossible, or considered too invasive by the patient, it is reasonable to begin with TTE.

TREATMENT  Definitive antibiotic treatment of infective endocarditis (Table 76-6) is guided by antimicrobial susceptibility testing of the responsible pathogen isolated from clinical cultures. Although it is often advisable to begin empirical treatment before definitive culture results are available, not all patients who are admitted because of possible endocarditis necessarily need to be treated empirically. Patients who are clinically stable, with a subacute presentation syndrome, and without evidence of heart failure or other end-organ complications, can be closely observed without antibiotics so that serial blood cultures can be obtained. Likewise, such stable patients who were started on empiric antibiotics before hospitalization and before blood was drawn for cultures can discontinue antibiotics so that blood cultures can be obtained, preferably as long as possible after stopping the antibiotics. By contrast, acutely ill patients, patients with evidence of complications of endocarditis, and patients who are at high risk for endocarditis (e.g., prosthetic valve recipients) should be treated empirically with antibiotics pending culture results. In most cases of infective endocarditis, an infectious diseases specialist can assist in guiding the diagnostic evaluation and designing an appropriate antibiotic regimen. Either of two regimens provides appropriate empirical coverage for patients with suspected native valve endocarditis: nafcillin (or oxacillin)–penicillin– gentamicin or vancomycin–gentamicin (Table 76-7). Nafcillin–penicillin–  gentamicin is suitable in most cases of suspected native valve endocarditis because it provides optimal coverage for viridans group streptococci, methicillin-sensitive staphylococci, enterococci, and HACEK organisms. Some experts recommend a regimen of nafcillin–ceftriaxone–penicillin–gentamicin to cover for HACEK isolates that produce β-lactamase. If methicillin-resistant S. aureus (MRSA) is an important consideration, as in injection drug users and patients with health care contact, empirical therapy should consist  of vancomycin–ceftriaxone–gentamicin. This regimen is also acceptable for

CHAPTER 76  Infective Endocarditis  

479

TABLE 76-6  DEFINITIVE THERAPY OF BACTERIAL ENDOCARDITIS ORGANISM AND REGIMEN*

COMMENTS

PCN-SUSCEPTIBLE VIRIDANS STREPTOCOCCI (MIC ≤0.1 µg/mL) AND STREPTOCOCCUS GALLOLYTICUS (formerly S. bovis) 1. PCN 2-3 million units IV q4h × 4 wk

1. Also effective for other PCN-susceptible nonviridans streptococci

2. Ceftriaxone 2 g IV qd × 4 wk

2. Uncomplicated infection with viridans streptococci in a candidate for outpatient therapy; also for those with PCN allergy

3. PCN 2-3 million units IV q4h × 2 wk plus gentamicin 1 mg/kg IV q8h × 2 wk

3. Uncomplicated infection with none of the following features: renal insufficiency, eighth cranial nerve deficit, prosthetic valve infection, CNS complications, severe heart failure, age >65 yr; also not acceptable for nutritionally variant streptococci

4. PCN 2-4 million units IV q4h × 4 wk plus gentamicin 1 mg/kg IV q8h for at least 2 wk with ID input

4. Nutritionally variant strain; for prosthetic valve, give 6 wk of PCN

5. Vancomycin 15-20 mg/kg IV q8-12h × 4 wk

5. For PCN allergy; goal trough level of 15-20 mg/L

RELATIVELY PCN-RESISTANT VIRIDANS STREPTOCOCCI (MIC 0.12-0.5 µg/mL) †

1. PCN‡ 18-30 million units IV per day in divided doses × 4-6 wk or ampicillin 12 g/24 hr IV in 6 equally divided doses plus gentamicin 1 mg/kg IV q8h × 4-6 wk

1. Increase duration of both drugs to 6 wk for prosthetic valve infection or symptoms >3 mo in enterococcal infection

2. Vancomycin 15-20 mg/kg IV q8-12h × 6 wk plus gentamicin 1 mg/kg q8h × 6 wk§

2. For PCN allergy; PCN desensitization is also an option; high risk of nephrotoxicity with this regimen

3. Ampicillin 12 g/24 h IV in 6 equally divided doses plus ceftriaxone 2 g IV q12h

3. PCN-susceptible, aminoglycoside-resistant enterococci or patients who have significant underlying renal disease

STAPHYLOCOCCUS AUREUS 1. Nafcillin 2 g IV q4h × 4-6 wk

1. Methicillin-susceptible strain; omit gentamicin if significant renal insufficiency

2. Vancomycin 15-20 mg/kg IV q8-12h × 6 wk

2. PCN allergy (immediate hypersensitivity or anaphylaxis) or MRSA

3. Nafcillin 2 g IV q4h × 2 wk plus gentamicin 1 mg/kg IV q8h × 2 wk

3. Methicillin-susceptible strain; 2-wk regimen only for use in IV drug abusers with only tricuspid valve infection, no renal insufficiency, and no extrapulmonary infection

4. Nafcillin 2 g IV q4h × >6 wk plus gentamicin 1 mg/kg IV q8h × 2 wk plus rifampin 300 mg PO/IV q8h × ≥6 wk

4. Prosthetic valve infection with methicillin-susceptible strain; use vancomycin instead of nafcillin for MRSA

5. Cefazolin 2 g IV q8h × 4-6 wk

5. PCN allergy other than immediate hypersensitivity

6. Daptomycin 6 mg/kg IV qd × 14-42 days

Daptomycin is FDA-approved for treatment of right-sided S. aureus infective endocarditis; for adults, some experts recommend 8-10 mg/kg IV

COAGULASE-NEGATIVE STAPHYLOCOCCI, PROSTHETIC VALVE INFECTION Vancomycin 15-20 mg/kg IV q8-12h × >6 wk plus gentamicin 1 mg/kg IV q8h × 2 wk plus rifampin 300 mg PO/IV q8h × >6 wk

Can substitute nafcillin in above doses for vancomycin if isolate is methicillin sensitive

HACEK STRAINS 1. Ceftriaxone 2 g IV qd × 4 wk; 6 wk for prosthetic valves

—

2. Ampicillin–sulbactam 3 g IV q6h × 4 wk; 6 wk for prosthetic valves

2. HACEK strains increasingly may produce β-lactamase

NON-HACEK GRAM-NEGATIVE BACILLI Enterobacteriaceae Extended-spectrum PCN or cephalosporin plus aminoglycosides for susceptible strains

Treat for a minimum of 6-8 wk; some species exhibit inducible resistance to third-generation cephalosporins; valve surgery is required for most patients with left-sided endocarditis caused by gram-negative bacilli; consultation with a specialist in infectious diseases is recommended

Pseudomonas aeruginosa High-dose tobramycin (8 mg/kg/day IV or IM in once-daily doses) with maintenance of peak and trough concentrations of 15 to 20 µg/mL and ≤2 µg/ mL, respectively, in combination with an extended-spectrum PCN (e.g., ticarcillin, piperacillin, azlocillin); ceftazidime, cefepime, or imipenem in full doses; or imipenem

Treat for a minimum of 6-8 wk; early valve surgery usually required for left-sided Pseudomonas endocarditis; consultation with a specialist in infectious diseases is recommended

Fungi Treatment with a parenteral antifungal agent (usually a lipid-containing amphotericin B product, 3-5 mg/kg/day IV for at least 6 weeks) and valve replacement; Fluconazole, 400 mg daily PO is an alternative for susceptible yeasts; other azoles, such as voriconazole, may be required for resistant yeasts or molds.

Long-term or lifelong suppressive therapy with PO antifungal agents often required; consultation with a specialist in infectious diseases is recommended

*Dosages are for patients with normal renal function; for those with renal insufficiency, adjustments must be made for all drugs except nafcillin, rifampin, and ceftriaxone. Gentamicin doses should be adjusted to achieve a peak serum concentration of approximately 3 µg/mL 30 min after dosing and a trough gentamicin level of 1 week after the institution of appropriate antibiotics) should prompt repeat blood cultures. If such cultures are negative, several possibilities should be considered: myocardial abscess, extracardiac infection (e.g., mycotic aneurysm, psoas or splenic abscess, vertebral osteomyelitis, septic arthritis), immune complex–mediated tissue damage, or a complication of hospitalization and therapy (e.g., drug fever, nosocomial superinfection, pulmonary embolism). Appropriate studies might include TEE, computed tomography (CT) scan of the abdomen, bone

481

CHAPTER 76  Infective Endocarditis  

scan, and urinalysis with microscopy (to elicit evidence of interstitial nephritis). IV line sites should be carefully examined for evidence of infection, and indwelling central lines should be changed according to published guidelines. Anticoagulation in individuals with infective endocarditis is controversial. Although new anticoagulation in the setting of native valve endocarditis does not appear to provide a benefit, continuing ongoing anticoagulation may  be advisable. Some authorities recommend continuing anticoagulation in patients with mechanical prosthetic valve endocarditis. However, discontinuation of all anticoagulation for at least the first 2 weeks of antibiotic therapy is generally advised in patients with S. aureus prosthetic valve endocarditis who have experienced a recent CNS embolic event; this approach allows the thrombus to organize and potentially prevents the acute hemorrhagic transformation of embolic lesions. Reintroduction of anticoagulation in these patients must be cautious, and the international normalized ratio must be monitored carefully. The best option for patients with other indications for anticoagulation, such as deep vein thrombosis, major vessel embolization, or atrial fibrillation, is less clear and should be decided in a multidisciplinary fashion that balances the risks and benefits for each individual patient. High-dose (325 mg/day) aspirin does not prevent embolic events and  tends to increase the incidence of bleeding in patients with infective endocarditis. A4  Whether a patient should remain on chronic, low-dose (81 mg) aspirin if they develop subsequent infective endocarditis is uncertain.

Some complications of infective endocarditis result when bacteremic seeding causes metastatic infection at a distant site. Patients may present with or develop osteomyelitis, septic arthritis, or epidural abscess. Purulent meningitis (Chapter 412) is a rare complication except in pneumococcal endocarditis, although many patients with S. aureus infective endocarditis who undergo lumbar puncture have a pleocytosis. Intracranial abscesses are uncommon in bacterial endocarditis but frequent in Aspergillus endocarditis; such a finding in the setting of culture-negative endocarditis should prompt the consideration of Aspergillus as an etiologic agent. Importantly, the finding of one metastatic complication of infective endocarditis does not exclude the possibility of additional sites of hematogenous infection, particularly in S. aureus endocarditis. Thus, the need for additional diagnostic evaluations should be guided by the patient’s clinical course. The immunologic phenomena of infective endocarditis are often directly related to high levels of circulating immune complexes. Renal biopsy results nearly always are abnormal in the setting of active infective endocarditis, which classically causes a hypocomplementemic glomerulonephritis (Chapter 121). Histopathologically, the glomerular changes may be focal, diffuse, or membranoproliferative, or they may be akin to the immune complex disease found in systemic lupus erythematosus. In addition, many of the musculoskeletal conditions associated with infective endocarditis, including monoarticular and oligoarticular arthritides, are probably immune mediated. These immunologic phenomena usually abate with successful antimicrobial therapy.

Complications

Surgery

The complications of infective endocarditis can be divided into four groups for ease of classification: direct valvular damage and consequences of local invasion, embolic complications, metastatic infections from bacteremia, and immunologic phenomena. Local damage to the endocardium or myocardium may directly erode through the involved cardiac valve or adjacent myocardial wall, resulting in hemodynamically significant valvular perforations or intra- or extracardiac fistulae. Such local complications typically present clinically with the acute onset of heart failure and carry a poor prognosis, even with prompt cardiac surgery. Valve ring abscesses also require surgical intervention and are more frequent in patients with prosthetic valves. Although a conduction defect on ECG may suggest the diagnosis, TEE is the diagnostic technique of choice for detecting paravalvular abscess, valve perforation, or intracardiac fistulae. Frank myocardial abscesses are found in up to 20% of cases on autopsy, and Aspergillus endocarditis invades the myocardium in more than 50% of cases. Pericarditis is rare and is usually associated with myocardial abscess. Myocardial infarction (MI), thought to be caused by embolism of vegetative material into the coronary arteries, is seen in 40% to 60% of cases on autopsy, although most cases are clinically silent and lack characteristic ECG changes. However, up to 15% of elderly patients may present with clinical evidence of acute MI, with potentially disastrous complications if the MI is thought to be the primary event and the patient is given thrombolytic therapy. Heart failure is the leading cause of death in infective endocarditis, usually related to direct valvular damage. Embolic events are less common now than in the preantibiotic era, but about 35% of patients have at least one clinically evident embolic event. In fungal endocarditis, the majority of patients have at least one embolic event, frequently with a large embolus. The presence of large (>10 mm), mobile vegetations on the echocardiogram, particularly when the anterior mitral valve leaflet is involved, predicts a high risk of embolic complications. In addition, patients may have frank infarction of cutaneous tissue from emboli. In addition to the skin, systemic emboli most commonly lodge in the kidneys, spleen, large blood vessels, or CNS. Vegetations of right-sided endocarditis usually embolize to the lungs and cause abnormalities on the chest radiograph, although occasionally such emboli reach the left-sided circulation via a patent foramen ovale. Renal abscesses are rare in infective endocarditis; however, bland renal infarction is a frequent asymptomatic finding on abdominal CT scanning, seen in more than 50% of cases at autopsy. Similarly, splenic infarction occurs in up to 44% of autopsy-confirmed cases. Such emboli may be asymptomatic but also can cause left upper quadrant pain radiating to the left shoulder, sometimes as the presenting symptom of infective endocarditis. A splenic infarction that progresses to form an abscess can cause persistent fever or bacteremia, so such patients should undergo abdominal CT to search for this complication. Mycotic vascular aneurysms, which frequently occur at bifurcation points, may be clinically silent until they rupture (which may be months to years after apparently successful antibiotic treatment of infective endocarditis) and have been found in 10% to 15% of cases at autopsy. Whereas peripheral mycotic aneurysms require surgical resection, intracerebral aneurysms should be resected or managed with intravascular techniques (e.g., coils) if they bleed or if they are causing a mass effect. Many patients may have evidence of cerebrovascular emboli, which have a predilection for the middle cerebral artery distribution and may be devastating. Most emboli to the CNS occur early in the course of the disease and are evident at the time of presentation or shortly thereafter. Embolic strokes may undergo hemorrhagic transformation, with a sudden worsening of the patient’s neurologic status. Many patients with fungal endocarditis present with an embolic stroke or large emboli that occlude major vessels.

Some patients with infective endocarditis require surgical treatment, either to cure the infection or to avoid its complications7,8 (Table 76-8). Most patients with evidence of direct extension of infection to myocardial structures, 

TABLE 76-8  INDICATIONS FOR SURGERY IN ENDOCARDITIS INDICATION

CLASS*

NATIVE VALVE ENDOCARDITIS Acute aortic insufficiency or mitral regurgitation with heart failure

I

Acute aortic insufficiency with tachycardia and early closure of the mitral valve on echocardiogram

I

Fungal endocarditis

I

Evidence of annular or aortic abscess, sinus or aortic true or false aneurysm, valvular dehiscence, rupture, perforation, or fistula

I

Evidence of valve dysfunction and persistent infection after a prolonged period (7-10 days) of appropriate therapy, provided there are no noncardiac causes of infection

I

Recurrent emboli after appropriate antibiotic therapy

I

Infection with enteric gram-negative organisms or organisms with a poor response to antibiotics in patients with evidence of valve dysfunction

I

Anterior mitral leaflet vegetation (especially with size >10 mm) or persistent vegetation after systemic embolization

IIa

Increase in vegetation size despite appropriate antimicrobial therapy

IIb

Early infections of the mitral valve that can probably be repaired, especially in the presence of large vegetations and/or recurrent emboli

III

Persistent fever and leukocytosis with negative blood cultures

III

PROSTHETIC VALVE ENDOCARDITIS Early prosthetic valve endocarditis (10 mm), early surgery did not significantly reduce all-cause mortality at 6 months but markedly decreased the risk of systemic embolism, including stroke and MI. A5  Surgical management should also be considered for patients with recurrent (two or more) embolic events or those with large vegetations (>10 mm) on echocardiography and one embolic event, although the data in these situations are less convincing. The presence of S. aureus endocarditis involving the anterior mitral valve leaflet and large vegetations (>10 mm) may be a special circ*mstance calling for early surgical intervention to reduce the high risk of CNS emboli, especially when mitral valve repair, rather than valve replacement, can be accomplished. Unfortunately, only about 15% to 20% of these latter patients end up being good candidates for valve repair. Delaying surgery in patients with deteriorating cardiac function in an attempt to sterilize the affected valve is ill advised because the risk of progressive heart failure or further complications usually outweighs the relatively small risk of recurrent infective endocarditis after prosthetic valve implantation. Relative contraindications to valve replacement include recent large CNS emboli (>2 cm) or bleed (because of the risk of bleeding in the perioperative period, when systemic anticoagulation is required), multiple prior valve replacements (because of the difficulty of sewing a new valve into tissue already weakened from previous surgeries), and ongoing injection drug use. On occasion, patients have both a compelling indication for valve replacement (e.g., acute heart failure) and a recent CNS embolic event. The risk of hemorrhagic transformation of such lesions during cardiac bypass–associated anticoagulation is controversial. However, it appears that the greatest risk of such transformation events is in larger (>2 cm) emboli, especially those that have exhibited a hemorrhagic component. In these latter scenarios, it is prudent to try to delay surgery for at least 2 to 4 weeks to allow organization and resolution of such emboli. However, there appears to be no survival benefit in delaying indicated valve replacement surgery (>7 days) after an ischemic stroke. After definitive surgical treatment, most patients should receive further antibiotic therapy unless a full course of antibiotics was administered before surgery and there is no evidence of ongoing infection. If the patient received antibiotics for less than 1 week before surgery or the culture from the operative site is positive, the patient should receive the equivalent of a full initial course of antibiotics appropriate for the organism. If the patient received antibiotics for 2 weeks or more and the culture result from the operative site is negative (regardless of whether valve histopathology shows inflammation or a positive Gram stain result), the patient should receive whatever remains of the originally planned course of appropriate antibiotic therapy. In patients with infective endocarditis related to implanted cardiovascular devices, complete device removal is mandatory, regardless of the pathogen, if the goal is to cure the infection. If a replacement device needs to be implanted, the optimal timing for such a procedure is unclear. However, blood culture results should be negative, and any concomitant local or pocket site infection should be completely resolved. The duration of antimicrobial therapy after device extraction depends on the device and the infection.9 For lead-related infective endocarditis, which is usually associated with bloodstream infection, 2 weeks of therapy is recommended if there are no infection complications. For infection caused by S. aureus, therapy should be extended for up to 4 weeks. In patients with valve infection, 4 to 6 weeks of therapy is recommended.

PREVENTION

Despite a lack of definitive data for dental procedures,10 prophylactic antibiotics are recommended to prevent infective endocarditis (Table 76-9) when patients with the highest risk of adverse outcomes from endocarditis undergo dental procedures that involve manipulation of gingival tissue or the periapical region of teeth or perforation of the oral mucosa; an invasive procedure of the respiratory tract, with incision or biopsy of the respiratory mucosa, such as tonsillectomy and adenoidectomy; or invasive procedures involving infected skin, skin structures, or musculoskeletal tissue (Table 76-10).11 Other consensus guidelines have also narrowed the indications for antimicrobial prophylaxis. In the United Kingdom, for example, no prophylaxis is advised for any dental patient, regardless of underlying cardiac valvular conditions. In contrast, French and other European guidelines are largely consistent with current AHA guidelines. Since the recent publications of these more limited recommendations from the AHA, France, and

TABLE 76-9  HIGH-RISK CARDIAC CONDITIONS FOR WHICH ENDOCARDITIS PROPHYLAXIS WITH DENTAL PROCEDURES IS REASONABLE Prosthetic cardiac valve or prosthetic material used for cardiac valve repair Previous endocarditis Complex congenital heart disease involving unrepaired cyanotic congenital heart disease (including palliative shunts and conduits), completely repaired congenital heart disease with prosthetic material within 6 mo of the procedure, or repaired congenital heart disease with residual defects at the site or adjacent to the site of prosthetic material Cardiac transplantation recipients who develop cardiac valvuloplasty Adapted from Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.

TABLE 76-10  RECOMMENDATIONS FOR ENDOCARDITIS PROPHYLAXIS PROPHYLAXIS IS RECOMMENDED* Dental: all dental procedures involving manipulation of gingival tissue or the periapical region of teeth or perforation of the oral mucosa Respiratory: procedures involving incision or biopsy of the respiratory mucosa, such as tonsillectomy and adenoidectomy Other: procedures involving infected skin, skin structures, or musculoskeletal tissue prior to incision and drainage PROPHYLAXIS IS NOT RECOMMENDED Dental: routine anesthetic injections through noninfected tissue, dental radiographs, placement of removable prosthodontic or orthodontic appliances, adjustment of orthodontic appliances, placement of orthodontic brackets, shedding of deciduous teeth, bleeding from trauma to the lips or oral mucosa Respiratory: procedures not involving incision or biopsy of the respiratory mucosa, including bronchoscopy (unless the procedure involves incision of the respiratory tract mucosa) Genitourinary: antibiotic prophylaxis solely to prevent infective endocarditis is not recommended Gastrointestinal: antibiotic prophylaxis solely to prevent infective endocarditis is not recommended *Only in patients with underlying cardiac conditions associated with the highest risk for adverse outcome from endocarditis (listed in Table 76-9). Adapted from Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.

the United Kingdom, follow-up surveys in these countries have shown no appreciable increase in the incidence of viridans group streptococcal infective endocarditis.12,13 The antibiotics chosen for preprocedure prophylaxis should be active against the organisms most likely to be released into the blood stream by the procedure of interest (Table 76-11). Thus, antibiotics that cover primarily oral flora are recommended for dental and upper respiratory procedures. For patients with the conditions listed in Table 76-9 who undergo a procedure for infected skin, skin structure, or musculoskeletal tissue, the therapeutic regimen should contain an agent active against staphylococci and β-hemolytic streptococci. Patients with implanted cardiac devices do not require antibiotic prophylaxis for dental or other invasive procedures. However, such patients require surgical site prophylaxis at the time of device placement.2 The recommended regimens generally include a β-lactam (commonly cefazolin, 1 g IV 1 hour before device placement), regardless of whether a new device is being placed or a device is being revised.

PROGNOSIS

Untreated infective endocarditis is uniformly fatal. Aggressive medical and surgical management dramatically improves the outcome. The overall mortality rate from both native and prosthetic valve endocarditis remains fairly high, ranging from 17% to 36%. Whereas certain subgroups, such as patients with viridans group streptococcal endocarditis, have a lower risk of death,

TABLE 76-11  SUGGESTED ANTIBIOTICS FOR ENDOCARDITIS PROPHYLAXIS FOR DENTAL OR RESPIRATORY TRACT PROCEDURES* IN PATIENTS WITH HIGH-RISK CARDIAC CONDITIONS† PATIENT CHARACTERISTICS

REGIMEN‡

Able to take oral medications

Amoxicillin 2 g PO

Unable to take oral medications

Ampicillin 2 g IV or IM; or cefazolin or ceftriaxone 1 g IM or IV

Allergic to penicillin or ampicillin and able to take oral medications

Cephalexin 2 g PO (or other first- or second-generation oral cephalosporin in equivalent adult doses); clindamycin 600 mg PO; azithromycin 500 mg PO; or clarithromycin 500 mg PO Cephalosporins should not be used in an individual with a history of anaphylaxis, angioedema, or urticaria with penicillin or ampicillin

Allergic to penicillin or ampicillin and unable to take oral medications

Cefazolin or ceftriaxone 1 g IM or IV; or clindamycin 600 mg IM or IV

*For the applicable procedures, see Table 76-10. † For the applicable conditions, see Table 76-9. ‡ All regimens consist of a single dose 30-60 min before the procedure. IM = intramuscular; IV = intravenous; PO = oral. Adapted from Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.

patients with S. aureus, fungal, and zoonotic endocarditis have higher mortality rates. Heart failure and CNS events are the most frequent causes of death. Endocarditis recurs in about 12% to 16% of patients and is more common in injection drug users, elderly people, and patients with prosthetic valves. The rate of relapse also varies depending on the causative organism. Easily treated infections, such as those with viridans group streptococci, have a low rate of relapse (5%), but more difficult-to-eradicate organisms may have significantly higher rates.

FUTURE DIRECTIONS

As cardiac imaging technology continues to improve, the duration of treatment of endocarditis may be dictated in part by the characteristics of visualized vegetations. In addition, now that large vegetations have been demonstrated to cause more embolic events, earlier interventions to remove vegetations from functionally competent valves or to introduce agents that prevent the formation or promote the dissolution of vegetations may be feasible. Moreover, advances in imaging may allow routine screening of patients for subclinical infections. Novel therapeutic and prophylactic approaches, such as antibacterial antibodies, targeted bacterial vaccines, and cell wall– specific enzymes that can act as adjuncts to antibiotics in facilitating bacteriologic clearance, are currently in development.

Grade A References A1. Cosgrove SE, Vigliani GA, Fowler VG Jr, et al. Initial low-dose gentamicin for Staphylococcus aureus bacteremia and endocarditis is nephrotoxic. Clin Infect Dis. 2009;48:713-721. A2. Sexton DJ, Tenenbaum MJ, Wilson WR, et al. Ceftriaxone once daily for four weeks compared with ceftriaxone plus gentamicin once daily for two weeks for treatment of endocarditis due to penicillinsusceptible streptococci. Endocarditis Treatment Consortium Group. Clin Infect Dis. 1998;27: 1470-1474. A3. Fowler VG Jr, Boucher HW, Corey GR, et al. Daptomycin versus standard therapy for bacteremia and endocarditis caused by Staphylococcus aureus. N Engl J Med. 2006;355:653-665. A4. Chan KL, Dumesnil JG, Cujec B, et al. A randomized trial of aspirin on the risk of embolic events in patients with infective endocarditis. J Am Coll Cardiol. 2003;42:775-780. A5. Kang DH, Kim YJ, Kim SH, et al. Early surgery versus conventional treatment for infective endocarditis. N Engl J Med. 2012;366:2466-2473.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 76  Infective Endocarditis  

GENERAL REFERENCES 1. Baddour LM, Cha YM, Wilson WR. Clinical practice. Infections of cardiovascular implantable electronic devices. N Engl J Med. 2012;367:842-849. 2. Baddour LM, Epstein AE, Erickson CC, et al. Update on cardiovascular implantable electronic device infections and their management: a scientific statement from the American Heart Association. Circulation. 2010;121:458-477. 3. Hoen B, Duval X. Clinical practice. Infective endocarditis. N Engl J Med. 2013;368:1425-1433. 4. Thuny F, Grisoli D, Cautela J, et al. Infective endocarditis: prevention, diagnosis, and management. Can J Cardiol. 2014;30:1046-1057. 5. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis: diagnosis, antimicrobial therapy, and management of complications: a statement for healthcare professionals from the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Cardiovascular Disease in the Young, and the Councils on Clinical Cardiology, Stroke, and Cardiovascular Surgery and Anesthesia, American Heart Association: endorsed by the Infectious Diseases Society of America. Circulation. 2005;111:e394-e434. 6. Tattevin P1, Boutoille D2, Vitrat V, et al. Salvage treatment of methicillin-resistant staphylococcal endocarditis with ceftaroline: a multicentre observational study. J Antimicrob Chemother. 2014;69: 2010-2013. 7. Barsic B, Dickerman S, Krajinovic V, et al. Influence of the timing of cardiac surgery on the outcome of patients with infective endocarditis and stroke. Clin Infect Dis. 2013;56:209-217. 8. Lalani T, Cabell CH, Benjamin DK, et al. Analysis of the impact of early surgery on in-hospital mortality of native valve endocarditis: use of propensity score and instrumental variable methods to adjust for treatment selection bias. Circulation. 2010;121:1005-1013.

483.e1

9. Sandoe JA, Barlow G, Chambers JB, et al. Guidelines for the diagnosis, prevention and management of implantable cardiac electronic device infection. Report of a joint Working Party project on behalf of the British Society for Antimicrobial Chemotherapy (BSAC, host organization), British Heart Rhythm Society (BHRS), British Cardiovascular Society (BCS), British Heart Valve Society (BHVS) and British Society for Echocardiography (BSE). J Antimicrob Chemother. 2015;70: 325-359. 10. Glenny AM, Oliver R, Roberts GJ, et al. Antibiotics for the prophylaxis of bacterial endocarditis in dentistry. Cochrane Database Syst Rev. 2013;10:CD003813. 11. Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007; 116:1736-1754. 12. Desimone DC, Tleyjeh IM, Correa de Sa DD, et al. Incidence of infective endocarditis caused by viridans group streptococci before and after publication of the 2007 American Heart Association’s endocarditis prevention guidelines. Circulation. 2012;126:60-64. 13. Duval X, Delahaye F, Alla F, et al. Temporal trends in infective endocarditis in the context of prophylaxis guideline modifications: three successive population-based surveys. J Am Coll Cardiol. 2012;59:1968-1976.

483.e2

CHAPTER 76  Infective Endocarditis  

REVIEW QUESTIONS 1. A 76-year-old man with known rheumatic valvular heart disease underwent elective mitral valve replacement with a Saint Jude prosthetic valve. His dentist calls you for advice regarding choice of antibiotic prophylaxis before dental extraction because the patient had developed bronchospasm and a diffuse urticarial rash after amoxicillin administration before a dental cleaning approximately 6 months ago. Which antibiotic should be administered? A. Clindamycin 600 mg orally 1 hour before the procedure B. Cefuroxime axetil 500 mg orally before the procedure C. Amoxicillin 2 g intravenously 1 hour before the procedure with corticosteroid and antihistamine coverage D. Nafcillin sodium 2 g IV 1 hour before the procedure E. Gentamicin sulfate 1 mg/kg IV 1 hour before the procedure Answer: A  The current (2007) AHA guidelines recommend clindamycin in patients who have a history of an immediate type hypersensitivity reaction to β-lactam antibiotics, which this patient demonstrated. Therefore, amoxicillin, cephalosporins, and nafcillin should be avoided. Levofloxacin and gentamicin are not recommended in these guidelines. (Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.) 2. A 68-year-old man with diabetes on chronic hemodialysis developed the acute onset of fever, chills, and left-sided abdominal pain. He had an ICD implanted 3 years ago. Blood cultures grew Staphylococcus aureus, and transesophageal echocardiography demonstrated vegetations on the mitral valve. Splenic infarctions were seen on computed tomographic scanning. Which one of the following is true regarding this presentation? A. Health care–associated infection is accounting for an increasing number of infective endocarditis cases in this country. B. Escherichia coli is a common cause of infective endocarditis in the hemodialysis population. C. Pending susceptibility testing results, gentamicin should be administered. D. To reduce health care costs, transthoracic echocardiography should have been performed instead of transesophageal echocardiography. E. For chronic hemodialysis, a tunneled catheter has a lower risk of blood stream infection compared with an arteriovenous fistula. Answer: A  Health care exposure accounts for an increasing number of cases of infective endocarditis in developed countries. Staphylococcus aureus, including methicillin-resistant strains, is a common cause of these infections. Empiric vancomycin should be administered until susceptibility results are known. (Athan E, Chu VH, Tattevin P, et al. Clinical characteristics and outcome of infective endocarditis involving implantable cardiac devices. JAMA. 2012;307:1727-1735.)

3. A 55-year-old veterinarian presents with several months of low-grade fever and night sweats. On an echocardiograph, he has evidence of endocarditis. Despite no recent antibiotic therapy, three sets of blood cultures remain negative for 7 days. Which one of the following is the most likely pathogen? A. Coagulase-negative staphylococcus B. Coxiella burnetii C. Enterococcus faecium D. Escherichia coli E. Orf virus Answer: B  Exposure to animals, particularly sheep and goats, is a risk factor for Coxiella infection and a well-known cause of culture-negative endocarditis. Orf virus does not cause endocarditis. The other choices should result in positive blood culture results in a patient who has infective endocarditis and who has not received antibiotic therapy recently. (Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis: diagnosis, antimicrobial therapy, and management of complications: a statement for healthcare professionals from the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Cardiovascular Disease in the Young, and the Councils on Clinical Cardiology, Stroke, and Cardiovascular Surgery and Anesthesia, American Heart Association: endorsed by the Infectious Diseases Society of America. Circulation. 2005;111:e394-434.) 4. A 78-year-old woman presents with left upper chest pain, swelling, and purulent drainage at the site where a permanent pacemaker generator was implanted. She has had no fever or chills, and her white blood cell count is normal. She underwent generator exchange 4 months ago and underwent a dental cleaning without prophylaxis 3 months ago. Which one of the following is true regarding cardiac implantable electronic device (CIED) infections? A. Surgical site prophylaxis has not been shown to reduce the risk of a CIED site infection. B. Prophylactic antibiotic should have been given before her dental cleaning. C. Device manipulation is a risk factor for device infection. D. The most likely cause of this infection is a HACEK organism. E. Antibiotic therapy for 4 weeks will likely cure the device infection without the device being removed. Answer: C  Manipulation of an implantable cardiac device is associated with the development of acute infection. Antibiotic prophylaxis before manipulation of the surgical site is beneficial, but dental prophylaxis is not. The most likely cause is Staphylococcus spp., and removal of the device is required for cure of the infection. (Baddour LM, Cha YM, Wilson WR. Clinical practice. Infections of cardiovascular implantable electronic devices. N Engl J Med. 2012;367:842-849.) 5. A 25-year-old morbidly obese man who injects heroin and cocaine presents with fever and blood cultures that grow Staphylococcus aureus. He had a past history of prior S. aureus blood stream infection 2 years ago, when he had an allergic reaction to vancomycin. At that time, he had evidence of tricuspid valve endocarditis. Which one of the following is true? A. A transthoracic echocardiography should be obtained. B. Initial empiric therapy should include daptomycin until susceptibility results are known. C. His mortality risk is high (>50%). D. Two weeks of antibiotic therapy should be curative. E. Rifampin should be administered. Answer: B  Daptomycin should be administered in case the blood culture isolate is methicillin-resistant Staphylococcus aureus. Transesophageal, rather than transthoracic, echocardiography should be obtained to evaluate for both right-sided and left-sided endocarditis. Cure rates with active antibiotic therapy for 2 weeks are high, provided there is no evidence of left-sided endocarditis or of metastatic foci of infection. Rifampin is not routinely recommended in native valve infections caused by staphylococci. (Fowler VG Jr, Boucher HW, Corey GR, et al. Daptomycin versus standard therapy for bacteremia and endocarditis caused by Staphylococcus aureus. N Engl J Med. 2006;355:653-665.)

CHAPTER 77  Pericardial Diseases  

483

77  PERICARDIAL DISEASES WILLIAM C. LITTLE AND JAE K. OH

The pericardium, which is a relatively avascular fibrous sac that surrounds the heart, has two layers, the visceral and parietal pericardia. The potential space between the two layers normally contains only 10 to 50 mL of fluid, which is an ultrafiltrate of plasma. The pericardium is well innervated, so pericardial inflammation may produce severe pain and trigger vagally mediated reflexes. As a result of its relatively inelastic physical properties, the pericardium limits acute cardiac dilation and enhances mechanical interactions of the cardiac chambers. In response to long-standing stress, the pericardium dilates to allow a slowly accumulating pericardial effusion to become quite large without compressing the cardiac chambers and to allow left ventricular remodeling to occur without pericardial constriction. Conversely, a scarred or thickened pericardium can limit the filling of the heart, resulting in pericardial constriction. Despite the important functions of the normal pericardium, congenital absence or surgical resection of the pericardium does not appear to have any major untoward effects.

ACUTE PERICARDITIS EPIDEMIOLOGY AND PATHOBIOLOGY

Acute inflammation of the pericardium, with or without an associated pericardial effusion, can occur as an isolated clinical problem or as a manifestation of systemic disease. Although about 85% of isolated cases of acute pericarditis are idiopathic or viral, the list of other potential causes is quite extensive (Table 77-1). Patients with fever greater than 38° C or a subacute course or who fail to respond promptly to therapy are most likely to have pericarditis caused by a systemic autoimmune disease, malignancy, or viral or bacterial infection. Pericarditis can occur after an acute myocardial infarction (MI). It occurs 1 to 3 days after a transmural MI, presumably owing to the interaction between the healing necrotic epicardium and the overlying pericardium. Dressler syndrome, which is another form of pericarditis associated with MI, typically occurs weeks to months after MI. It is similar to the pericarditis that can occur days to months after traumatic pericardial injury, surgical manipulation of the pericardium, or pulmonary infarction. This syndrome is presumed to be mediated by an autoimmune mechanism and is associated with signs of systemic inflammation, including fever and polyserositis.

CLINICAL MANIFESTATIONS

Most patients with acute pericarditis experience sharp retrosternal chest pain (see Table 51-2 in Chapter 51), which can be quite severe and debilitating. In some cases, however, pericarditis is asymptomatic, such as when it accompanies rheumatoid arthritis. Pericardial pain is usually worse with inspiration and when supine, and it is generally relieved by sitting and leaning forward. Typically, pericardial pain is referred to the scapular ridge, presumably owing to irritation of the phrenic nerves, which pass adjacent to the pericardium. The chest pain of acute pericarditis must be differentiated from pulmonary embolism and myocardial ischemia or infarction (Table 77-2). The pericardial friction rub is the classic finding in patients with acute pericarditis. A friction rub is a high-pitched, scratchy sound that can have one, two, or three components occurring when the cardiac volumes are most rapidly changing: during ventricular ejection, during rapid ventricular filling in early diastole, and during atrial systole. A pericardial rub, which is differentiated from a murmur by its scratchy quality, is sometimes localized to a small area on the chest wall and may come and go spontaneously or with changes in position. To hear a rub, it may be necessary to auscultate the heart with the patient in multiple positions, especially using the diaphragm with the patient learning forward and not breathing after full expiration. The pericardial friction rub must be differentiated from a pleural rub, which is absent during suspended respiration, but the pericardial rub is unaffected.

DIAGNOSIS

Early in the course of acute pericarditis, the electrocardiogram (ECG) typically displays diffuse ST elevation in association with PR depression (Fig. 77-1). The ST elevation is usually present in all leads except for aVR,

484

CHAPTER 77  Pericardial Diseases  

TABLE 77-1  CAUSES OF PERICARDITIS: INFECTIOUS AND NONINFECTIOUS INFECTIOUS PERICARDITIS (2 3 OF CASES) Viral (most common): echovirus and coxsackievirus (usual), influenza, EBV, CMV, adenovirus, varicella, rubella, mumps, HBV, HCV, HIV, parvovirus B19, human herpesvirus 6 (increasing reports) Bacterial: tuberculosis (4%-5%)* and Coxiella burnetii (most common); other bacterial causes (rare) include pneumococcosis, meningococcosis, gonococcosis, Haemophilus, staphylococci, Chlamydia, Mycoplasma, Legionella, Leptospira, Listeria Fungal (rare): histoplasmosis more likely in immunocompetent patients; aspergillosis, blastomycosis, candidiasis more likely in immunosuppressed patients Parasitic (very rare): Echinococcus, Toxoplasma NONINFECTIOUS PERICARDITIS ( 1 3 OF CASES) Autoimmune Pericarditis (5000/mm3 (autoreactive lymphocytic) or the presence of antibodies against heart muscle tissue (antisarcolemmal) in the pericardial fluid (autoreactive antibody mediated); (2) signs of myocarditis on epicardial or endomyocardial biopsies by ≥14 cells/mm2; and (3) exclusion of infections, neoplasia, and systemic and metabolic disorders. CMV = cytomegalovirus; EBV = Epstein-Barr virus; HBV = hepatitis B virus; HCV = hepatitis C virus; HIV = human immunodeficiency virus. From Imazio M, Spodick DH, Brucato A, et al. Controversial issues in the management of pericardial diseases. Circulation. 2010;121:916-928.

TABLE 77-2  DIFFERENTIATION OF PERICARDITIS FROM MYOCARDIAL ISCHEMIA OR INFARCTION AND PULMONARY EMBOLISM MYOCARDIAL ISCHEMIA OR INFARCTION

FINDINGS

PERICARDITIS

PULMONARY EMBOLISM

CHEST PAIN Character

Pressure-like heavy, squeezing

Sharp, stabbing, occasionally dull

Sharp, stabbing

Change with respiration

No

Worsened with inspiration

In phase with respiration (absent when the patient is apneic)

Change with position

No

Worse when supine; improved when sitting up or leaning forward

No

Duration

Minutes (ischemia); hours (infarction)

Hours to days

Hours to days

Response to nitroglycerin

Improved

No change

No change

Absent (unless pericarditis is present)

Present in most patients

Pleural friction rub may occur

ST segment elevation

Localized convex

Widespread concave

Limited to leads III, aVF, and V1

PR segment depression

Rare

Frequent

None

PHYSICAL EXAMINATION Friction rub ELECTROCARDIOGRAM

Modified from Little WC, Freeman GL. Pericardial disease. Circulation. 2006;113:1622-1632.

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

FIGURE 77-1.  Electrocardiogram demonstrating typical features of acute pericarditis on presentation. There are diffuse ST elevation and PR depression except in aVR, where there is ST depression and PR elevation.

485

CHAPTER 77  Pericardial Diseases  

but the changes may be more localized in post-MI pericarditis. Classically, the ECG changes of acute pericarditis evolve over several days; resolution of the ST elevation is followed by widespread T-wave inversion that subsequently normalizes. Uremic pericarditis usually occurs without the typical ECG abnormalities. Patients with acute pericarditis usually have evidence of systemic inflammation, including leukocytosis, an elevated erythrocyte sedimentation rate (ESR), and increased C-reactive protein (CRP) level. A low-grade fever is common, but a temperature greater than 38° C is unusual and suggests the possibility of bacterial pericarditis. About 85% of cases of acute pericarditis are idiopathic or viral. Viral causes include echoviruses and group B coxsackieviruses, but obtaining specific viral titers does not alter patient management. About 6% of cases are neoplastic in origin, about 4% are caused by tuberculosis, about 3% are caused by other bacterial or fungal infections, and about 2% are caused by collagen vascular disease. A targeted evaluation (Table 77-3) can help identify the various causes (Table 77-4). Troponin levels typically are minimally elevated in acute pericarditis owing to some involvement of the epicardium by the inflammatory process. An elevated troponin level in acute pericarditis usually returns to normal within 1 to 2 weeks and is not associated with a worse prognosis. Although the elevated troponin level may lead to the misdiagnosis of an ST elevation MI (Chapter 73), most patients with elevated troponin levels and acute pericarditis have normal coronary angiograms. An echocardiogram (Chapter 55) can help avoid a misdiagnosis of MI. Interestingly, patients with myopericarditis and elevated troponin levels tend to have a lower recurrence rate than do patients with pure pericarditis and normal troponin levels.1 Echocardiography may demonstrate a small pericardial effusion in the presence of acute pericarditis, but normal echocardiogram results do not exclude the diagnosis of acute pericarditis. An echocardiogram is critical, however, in excluding the diagnosis of cardiac tamponade (see later). When the diagnosis of acute pericarditis is unclear, cardiac magnetic resonance imaging (MRI) can demonstrate pericardial inflammation as delayed enhancement of the pericardium (Fig. 77-2). Diagnostic pericardiocentesis is indicated in suspected purulent tuberculosis or malignant pericarditis or if the patient has cardiac tamponade.

TABLE 77-3  SELECTED DIAGNOSTIC TESTS IN ACUTE PERICARDITIS IN ALL PATIENTS Tuberculin skin test (plus control skin test to exclude anergy) BUN and creatinine to exclude uremia Erythrocyte sedimentation rate Electrocardiogram Chest radiograph Echocardiogram IN SELECTED PATIENTS Cardiac magnetic resonance imaging ANA and rheumatoid factor to exclude SLE or rheumatoid arthritis in patients with acute arthritis or pleural effusion TSH and T4 to exclude hypothyroidism in patients with clinical findings suggestive of hypothyroidism and in asymptomatic patients with unexplained pericardial effusion HIV test to exclude AIDS in patients with risk factors for HIV disease or a compatible clinical syndrome Blood cultures in febrile patients to exclude infective endocarditis and bacteremia Fungal serologic tests in patients from endemic areas or in immunocompromised patients ASO titer in children or teenagers with suspected rheumatic fever Heterophil antibody test to exclude mononucleosis in young or middle-aged patients with a compatible clinical syndrome or acute fever, weakness, and lymphadenopathy AIDS = acquired immunodeficiency virus; ANA = antinuclear antibody; ASO = antistreptolysin O; BUN = blood urea nitrogen; HIV = human immunodeficiency virus; SLE = systemic lupus erythematosus; T4 = thyroxine; TSH = thyroid-stimulating hormone. Modified from Nishimura RA, Kidd KR. Recognition and management of patients with pericardial disease. In: Braunwald E, Goldman L, eds. Primary Cardiology. 2nd ed. Philadelphia: WB Saunders; 2003:625.

TABLE 77-4  PRESENTATION AND TREATMENT OF THE MOST COMMON CAUSES OF PERICARDITIS TYPE

PATHOGENESIS OR ETIOLOGY

DIAGNOSIS

TREATMENT

COMPLICATIONS

COMMENTS

Viral

Coxsackievirus B Echovirus type 8 Epstein-Barr virus

Leukocytosis Elevated ESR Mild cardiac biomarker elevation

Symptomatic relief, NSAIDs, colchicine

Tamponade Relapsing pericarditis

Peaks in spring and fall

Tuberculous

Mycobacterium tuberculosis

Isolation of organism from biopsy fluid Granulomas not specific

Triple-drug antituberculosis regimen Pericardial drainage followed by early (4-6 wk) pericardiectomy if signs of tamponade or constriction develop

Tamponade Constrictive pericarditis

1%-8% of patients with tuberculosis pneumonia; rule out HIV infection

Bacterial

Group A streptococcus Staphylococcus aureus Streptococcus pneumoniae

Leukocytosis with marked left shift Purulent pericardial fluid

Pericardial drainage by catheter or surgery Systemic antibiotics Pericardiectomy if constrictive physiology develops

Tamponade in one third of patients

Very high mortality rate if not recognized early

Post–myocardial infarction

12 hr–10 days after infarction

Fever Pericardial friction rub Echo: effusion

Aspirin Prednisone

Tamponade rare

More frequent in large Q wave infarctions Anterior > inferior

Uremic

Untreated renal failure: 50% Chronic dialysis: 20%

Pericardial rub: 90%

Intensive dialysis Indomethacin: probably ineffective Catheter drainage Surgical drainage

Tamponade Hemodynamic instability on dialysis

Avoid NSAIDs ≈50% respond to intensive dialysis

Neoplastic

In order of frequency: lung cancer, breast cancer, leukemia and lymphoma, others

Chest pain, dyspnea Echo: effusion CT, MRI: tumor metastases to pericardium Cytologic examination of fluid positive in 85%

Catheter drainage Subxiphoid pericardiectomy Chemotherapy directed at underlying malignant neoplasm

Tamponade Constriction

CT = computed tomography; ESR = erythrocyte sedimentation rate; HIV = human immunodeficiency virus; MRI = magnetic resonance imaging; NSAIDs = nonsteroidal anti-inflammatory drugs. Modified from Malik F, Foster E. Pericardial disease. In: Wachter RM, Goldman L, Hollander H, eds. Hospital Medicine. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2005:449.

486

CHAPTER 77  Pericardial Diseases  

Cardiac tamponade? Yes

Drain effusion

No

Evaluate per Table 77-3

Evidence of high risk? • Tamponade • Moderate or large effusion • Fever >38° C • Suspicion of systemic illness Yes FIGURE 77-2.  Cardiac magnetic resonance image of a patient with acute pericarditis shows late gadolinium hyperenhancement of the pericardium and epicardium.

Admit to hospital NSAIDs ± colchicine Search for etiology

TREATMENT 

Yes

Acutely ill patients with fever should be hospitalized, as should patients with suspected acute MI (Chapter 73), large effusions, evidence of impending hemodynamic compromise, or a cause other than viral or idiopathic pericarditis because of the risk of a rapidly accumulating effusion with potential tamponade. Patients without effusions can usually be followed as outpatients (Fig. 77-3). If acute pericarditis is a manifestation of an underlying disease, it often responds to treatment of the primary condition. Most cases of acute idiopathic or viral pericarditis are self-limited and respond to treatment with aspirin (650 mg every 6 hours) or another nonsteroidal antiinflammatory drug (NSAID) such as ibuprofen (300 to 800 mg every 6 to 8 hours). The dose of NSAID should be tapered after symptoms and any pericardial effusion have resolved, but the medication should be taken for at least 3 to 4 weeks to minimize the risk of recurrent pericarditis. In addition, colchicine (0.6 to 1.2 mg/day for 3 months) should be started in all patients with acute pericarditis to reduce the rate of persistent symptoms at 72 hours, reduce the likelihood of recurrent pericarditis from 55% to 24% at 18 months, and reduce the rate of subsequent hospitalization. A1  A2  The major side effect of colchicine is diarrhea. The lower dose of colchicine should be used in patients who weigh less than 70 kg or who have side effects with the higher dose. Colchicine should be avoided in patients with abnormal renal or hepatic function and in patients being treated with macrolide antibiotics, which alter its metabolism. A proton pump inhibitor, such as omeprazole (20 mg/day), should be considered to improve the gastric tolerability of NSAIDs. Warfarin and heparin should be avoided to minimize the risk of hemopericardium, but anticoagulation may be required if the patient is in atrial fibrillation or has a prosthetic heart valve. It is prudent to avoid exercise until after the chest pain completely resolves. If pericarditis recurs, the patient can be reloaded with colchicine and intravenous ketorolac (20 mg) and then continued on an oral NSAID and colchicine for at least 3 months. Although acute pericarditis usually responds dramatically to systemic corticosteroids, observational studies strongly suggest that the use of steroids increases the probability of relapse in patients treated with colchicine. Except when needed to treat an underlying inflammatory disease, every effort should be made to avoid the use of steroids, reserving low-dose steroids for patients who cannot tolerate aspirin and other NSAIDs or whose recurrence is not responsive to colchicine and intravenous NSAIDs. If steroids are used, low-dose prednisone (0.2 to 0.5 mg/kg) appears to be as effective as higher doses and is less likely to be associated with recurrence. Steroids should be continued for at least 1 month before slow tapering, which can be guided by return of the CRP level to the normal range. Pericardiocentesis is not recommended unless purulent or tuberculous pericarditis is clinically suspected or the patient fails to respond to 2 to 3 weeks of NSAID therapy. ,

PROGNOSIS

The course of viral and idiopathic pericarditis is usually self-limited, and most patients recover completely.2 About 25% of patients, however, have recurrent pericarditis weeks to months later, probably caused by an immune response, and some patients may have multiple debilitating episodes. In patients whose acute pericarditis is accompanied by myocarditis, as evidenced by elevation of serum troponin levels, the recurrence rate is closer to 10%.1 Recurrent pericarditis is more common in patients treated with steroids for the acute episode, especially during a rapid steroid taper. In these patients, prolonged

Outpatient follow-up Taper NSAIDs after 3-4 wk

No

Trial of NSAIDs Pain relieved in 2 cm) effusions in patients who are hemodynamically stable and in whom tamponade is not suspected, a follow-up echocardiogram should be performed in 7 days and then every month until the effusion is minimal.4 If bacterial or malignant pericarditis is

Moderate-large pericardial effusion Cardiac tamponade or suspicion of infection? Yes

No

Drain effusion

Large effusion (>20 mm)? No

Yes

Treat pericarditis

Present for 2 mm) that can be imaged by echocardiography, CT, and MRI (Fig. 77-10). It is important to recognize, however, that pericardial constriction can be present without pericardial calcification and, in about 20% of patients, without any obvious pericardial thickening.

490

CHAPTER 77  Pericardial Diseases  

TABLE 77-6  DIFFERENTIATION OF PERICARDIAL CONSTRICTION FROM RESTRICTIVE CARDIOMYOPATHY FINDINGS

PERICARDIAL CONSTRICTION

RESTRICTIVE CARDIOMYOPATHY

PHYSICAL EXAMINATION Pulmonary congestion

Usually absent

Usually present

Early diastolic sound

Pericardial knock

S3 (low pitched)

Respiratory variation in E wave (%)

>25

7

2 mm (but 95%) D-dimer.

calf veins. If the repeated ultrasound examination 1 week later also is normal, further investigation and therapy can be safely withheld. In centers with highly skilled operators, one normal ultrasound of the proximal veins and the calf veins near the popliteal vein at presentation is sufficiently accurate to exclude clinically important DVT and eliminate the need for follow-up testing. In patients with either a normal D-dimer test result or a low clinical pretest probability, normal two-point compression ultrasonography excludes DVT.

Magnetic Resonance Venography

Magnetic resonance venography (MRV), which uses the difference in magnetic resonance signals between flowing blood and stationary clot, has a high sensitivity and specificity for proximal DVT. Recent interest has focused on magnetic resonance for direct imaging of the thrombus because a thrombus produces a positive image without the use of contrast material, owing to its methemoglobin content. Although MRV is accurate in diagnosing and excluding DVT, it is expensive and not readily available in most centers outside of the United States.

Contrast Venography

FIGURE 81-3.  Abnormal venogram demonstrates a persistent (two or more different views) intraluminal filling defect in the popliteal vein.

ultrasonography is diagnostic of DVT in symptomatic patients and is an indication for treatment. Of patients with symptoms suggestive of DVT but with normal findings on initial ultrasound examination of the proximal veins, approximately 15% will have undetected isolated calf DVT; progression into the proximal veins occurs in a minority of patients, usually within a week of presentation. Isolated calf DVT that does not extend into the proximal veins is rarely if ever associated with clinically important pulmonary embolus. The sensitivity of ultrasonography for calf DVT is well below 90%, with a wide range of accuracies reported for different populations of patients. Imaging of the calf veins is time-consuming and potentially inaccurate. Rather, two-point (common femoral and popliteal) or three-point (twopoint plus the calf “trifurcation”) compression ultrasonography should be performed. If two-point compression is normal, the test should be repeated about 1 week after the initial examination. This approach will identify the 20 to 25% of patients who have had proximal extension of distal clot in the

Ascending contrast venography remains the gold standard for diagnosis, but because of its expense, discomfort to the patient, and potential for adverse experiences, venography is currently indicated in symptomatic patients only when diagnostic uncertainty persists after noninvasive testing or if noninvasive testing is unavailable. A constant intraluminal filling defect is diagnostic of acute thrombosis (Fig. 81-4), and DVT can be excluded in patients who have a normal, adequately performed venogram. Minor side effects of local pain, nausea, and vomiting are not uncommon, whereas more serious adverse reactions, such as anaphylaxis or other allergic manifestations, are rare. Venography also can induce DVT.

Laboratory Findings D-Dimer

D-Dimer is a plasma protein specifically produced after lysis of cross-linked fibrin by plasmin. Levels are almost invariably elevated in the presence of acute VTE, so measurement of D-dimer levels is a sensitive test for recent DVT and pulmonary embolism. Unfortunately, numerous nonthrombotic conditions, including sepsis, pregnancy, surgery, and cardiac or renal failure, also can cause elevated levels. As a result of this nonspecificity, the role of D-dimer assays is limited to helping exclude VTE when levels are not raised. Laboratory tests for D-dimer use enzyme-linked immunosorbent assay or agglutination techniques, both involving specific monoclonal antibodies. Sensitivity and cut points vary among assays, so results cannot be generalized. Highly sensitive tests, consisting of new rapid ELISA or immunoturbidimetric assays, have sensitivities of 95 to 100% for acute VTE but in general have low specificities (20 to 50%). Highly sensitive D-dimer assays can be

514

CHAPTER 81  Peripheral Venous Disease  

FIGURE 81-4.  Compression venous ultrasonography demonstrates thrombosis of the popliteal vein. The sonograms in the top row demonstrate examination without (left side) and with (right side) gentle probe compression of the skin overlying the popliteal vein. The lack of compressibility is diagnostic of deep vein thrombosis. The bottom row shows analogous views of the femoral vein, which shows partial compressibility.

employed as stand-alone tests for exclusion of DVT, but clinicians must be aware of the accuracy of the assay in their institution before using the D-dimer assay to make management decisions. D-Dimer measured after a 3-month (or longer) initial treatment with warfarin also appears to be predictive of recurrent DVT. In addition, an elevated D-dimer level 1 month after stopping warfarin predicts a clinically and statistically significant higher recurrence rate than is seen in patients in whom the D-dimer levels were normal or low.

Algorithms for Diagnosis of Deep Venous Thrombosis and Their Risk for Recurrence

A number of diagnostic algorithms have been tested in prospective management trials (see Fig. 81-2).

Clinical Assessment and Venous Ultrasonography

It is safe to perform only a single ultrasound examination in patients with a low pretest probability by a validated clinical prediction rule (Table 81-2). Other patients require serial ultrasonographic testing if only clinical assessment and ultrasonography are used. Venography should be considered in patients with a high pretest probability and normal compression ultrasonography because the probability of DVT is still approximately 20% in such patients.

Clinical Assessment, D-Dimer Testing, and Venous Ultrasonography

Diagnostic imaging and treatment can be safely withheld in patients who have (1) a low pretest probability based on a validated clinical prediction rule and a negative value on a moderately sensitive D-dimer assay or (2) a low or intermediate pretest probability and a negative value on a highly sensitive D-dimer assay. Patients with a high pretest probability require ultrasonography regardless of the D-dimer result. A normal D-dimer result with use of either a moderately or highly sensitive assay can safely obviate the need for repeated imaging in patients with normal findings on the initial ultrasound examination. Algorithms for predicting the recurrence of an initially unprovoked DVT after the cessation of anticoagulant therapy are undergoing validation in prospective trials.

TABLE 81-2  ALTERNATIVE DIAGNOSES IN 87 CONSECUTIVE PATIENTS WITH CLINICALLY SUSPECTED VENOUS THROMBOSIS AND NORMAL VENOGRAMS* DIAGNOSIS

PATIENTS (%)

Muscle strain

24

Direct twisting injury to the leg

10

Leg swelling in paralyzed limb

9

Lymphangitis, lymphatic obstruction

7

Venous reflux

7

Muscle tear

6

Baker cyst

5

Cellulitis

3

Internal abnormality of the knee

2

Unknown

26

*The diagnosis was made once venous thrombosis was excluded by venography.

Differential Diagnosis

A number of conditions can mimic DVT (see Table 81-2), but DVT often can be excluded only by accurate diagnostic testing. In some patients, however, the cause of pain, tenderness, and swelling remains uncertain.

Suspected Recurrent Deep Venous Thrombosis

Approximately 10% of patients with unprovoked VTE will experience recurrent thromboembolism in the first year after ceasing anticoagulant therapy. In addition, many patients will have positional leg swelling and pain early during treatment as a result of venous outflow obstruction or later (≥6 months after diagnosis) because of the post-thrombotic syndrome after endogenous thombolysis has maximized removal of the thrombus and venous valvular incompetence manifests. These and other nonthrombotic

CHAPTER 81  Peripheral Venous Disease  

disorders can produce symptoms that are similar to those of acute recurrent DVT, so accurate diagnostic testing to confirm recurrence is mandatory. However, residual venous abnormalities are common after an initial event; persistent abnormalities are seen on compression ultrasonography in approximately 80% of patients at 3 months and 50% of patients at 1 year after a documented proximal DVT. Therefore comparison with previous ultrasound images is required in patients with suspected recurrence. Although an increase in diameter of 4 mm or more in the compressed vein strongly suggests recurrent DVT, a new noncompressible proximal venous segment is the most reliable criterion for the diagnosis of recurrence. When compression ultrasonography is inconclusive, venography should be considered; a new intraluminal filling defect is diagnostic of acute DVT, and the absence of a filling defect excludes the diagnosis. Nonfilling of venous segments may mask recurrent DVT and is considered a nondiagnostic finding. A normal D-dimer test result is useful in excluding recurrent DVT.

Pregnancy

Symptoms of leg pain or swelling, shortness of breath, and atypical chest pain are common during pregnancy, so objective testing is needed to diagnose VTE. As in nonpregnant patients, compression ultrasonography is the initial test of choice. A normal D-dimer test is also reassuring in excluding DVT. Because isolated iliac and iliofemoral DVT is more common in pregnancy and has the potential to be missed by ultrasonography, efforts should be made to image the iliac veins to detect such thrombi. MRV, which is sensitive for pelvic DVT, may be useful when the clinical suspicion is high or if Doppler imaging of the iliac vein is inconclusive.

TREATMENT  The large majority of patients with acute DVT can now be treated on an outpatient basis, regardless of their treatment regimen (Fig. 81-5).3 The principal indications for admission are clinical instability, the inability to adhere to

515

outpatient therapy, or the need to use intravenous heparin for extensive iliofemoral thrombosis.

Initial Treatment

Low-molecular-weight heparin (LMWH) preparations (Chapter 38) are administered subcutaneously using weight-based dosing to provide reliable outpatient management of DVT without the need for routine laboratory monitoring. Dosage regimens differ for the various LMWH formulations (Table 81-3), but once-daily administration of LMWH is thought to be as safe and effective as twice-daily administration.4 Anti–factor Xa monitoring should be considered for three populations of patients: (1) patients with renal insufficiency (calculated creatinine clearance of less than 30 mL/minute); (2) obese patients, in whom the volume of distribution of LMWH might be different, so weight-adjusted dosing might not be appropriate; and (3) pregnant women, in whom it is unclear whether the dose should be adjusted according to the woman’s weight change. Levels are usually determined on blood samples drawn 4 hours after subcutaneous injection; therapeutic ranges of 0.6 to 1.0 U/mL for twice-daily administration and 1.0 to 2.0 U/mL for once-daily treatment have been proposed. Fixed-dose subcutaneous injection of LMWH is at least as effective and safe as adjusted-dose intravenous administration of unfractionated heparin for the treatment of acute DVT, with a trend toward a significant difference in mortality benefit favoring LMWH, probably because of improved survival in patients with malignant disease. A1  However, patients with extensive iliofemoral DVT have often been excluded from trials of LMWH, and extended-duration (i.e., >5 days) intravenous unfractionated heparin therapy is often administered to such patients. Unfractionated heparin is usually administered by continuous intravenous infusion (Table 81-4), with either fixed initial dosing or dosing according to a patient’s weight, results in more rapid achievement of therapeutic activated partial thromboplastin time (aPTT) levels. The initial aPTT level should be measured 6 hours after therapy is commenced. Up to 25% of patients with acute VTE have resistance to heparin, defined as a requirement for greater than expected doses of unfractionated heparin to achieve a “therapeutic” aPTT. If it is available, anti–factor Xa monitoring is recommended in patients with heparin resistance.

DVT diagnosed Contraindication to anticoagulants?

Yes

No

Transient contraindication

Permanent contraindication

Retrievable IVC filter

Permanent IVC filter

Phlegmasia or extensive iliofemoral DVT

No

Yes

Suitable for outpatient therapy

Yes

Yes

No

LMWH or fondaparinux in treatment doses

IV UFH or LMWH or fondaparinux in hospital

Contraindication to thrombolytic therapy No Catheterdirected thrombolysis available?

Yes

No

Consider catheterdirected thrombolysis

Consider IV systemic thrombolytic therapy

FIGURE 81-5.  Guidelines for treatment of deep vein thrombosis (DVT). IVC = inferior vena cava; IV = intravenous; LMWH = low-molecular-weight heparin; UFH = unfractionated

heparin.

516

CHAPTER 81  Peripheral Venous Disease  

TABLE 81-3  GUIDELINES FOR ANTICOAGULATION WITH LOW-MOLECULAR-WEIGHT HEPARIN AND FONDAPARINUX INDICATIONS

GUIDELINES

VTE suspected

Obtain baseline aPTT, PT, CBC Check for contraindication to heparin therapy Order imaging study; consider giving IV unfractionated heparin (5000 IU) or LMWH

VTE confirmed

Give LMWH (dalteparin,* enoxaparin,† nadroparin,‡ tinzaparin,§ fondaparinux¶) Start warfarin therapy on day 1 at 5 mg and adjust the subsequent daily dose according to INR Check platelet count between days 3 and 5 Stop LMWH therapy after at least 4 or 5 days of combined therapy when the INR is > 2 Anticoagulate with warfarin for at least 3 months at an INR of 2.5, range of 2-3 (See text for alternatives to warfarin: "Oral Direct Thrombin and Factor Xa Inhibitors")

Modified from Hyers TM, Agnelli G, Hull RD, et al. Antithrombotic therapy for venous thromboembolic disease. Chest. 2001;119:176S-193S. *Dalteparin sodium, 200 anti-Xa IU/kg/day SC. A single dose should not exceed 18,000 IU (approved in Canada). † Enoxaparin sodium, 1 mg/kg q12h SC, or enoxaparin sodium, 1.5 mg/kg/day SC. A single daily dose should not exceed 180 mg (approved in both the United States and Canada). ‡ Nadroparin calcium, 86 anti-Xa IU/kg two times daily SC for 10 days (approved in Canada), or nadroparin calcium, 171 anti-Xa IU/kg SC daily. A single dose should not exceed 17,100 anti-Xa IU. § Tinzaparin sodium, 175 anti-Xa IU kg/day SC daily (approved in Canada and the United States). ¶ Fondaparinux according to weight: 100 kg, 10 mg SC. aPTT = activated partial thromboplastin time; CBC = complete blood count; INR = international normalized ratio; LMWH = low-molecular-weight heparin; PT = prothrombin time; VTE = venous thromboembolism.

TABLE 81-4  WEIGHT-BASED NOMOGRAM FOR INITIAL INTRAVENOUS HEPARIN THERAPY aPTT

DOSE (IU/kg)

Initial dose

80 bolus, then 18/hr

3×)

Hold infusion 1 hr, then decrease infusion rate by 3/hr

Modified from Raschke RA, Reilly BM, Guidry JR, et al. The weight-based heparin dosing nomogram compared with a “standard care” nomogram: a randomized controlled trial. Ann Intern Med. 1993;119:874-881. *Figures in parentheses show comparison with control. aPTT = activated partial thromboplastin time. In general, with contemporary aPTT reagents, the target therapeutic range is more than 1.2 to 2.3 times control.

Fondaparinux

Fondaparinux is a synthetic analogue of the critical pentasaccharide sequence required for binding of heparin molecules to antithrombin (Chapter 38). Given subcutaneously, fondaparinux demonstrates 100% bioavailability, with peak plasma concentrations occurring 1.7 hours after dosing. Once-daily subcutaneous administration of fondaparinux (5 mg/day if weight is < 50 kg; 7.5 mg/day if weight is 50 to 100 kg; 10 mg/day if weight is > 100 kg) is an effective and safe alternative to LMWH for the initial 5 to 10 days of treatment of DVT. A2  Clearance is predominantly renal, with approximately 70% of the initial dose recovered in the urine in an unchanged form. Patients with reduced creatinine clearance, such as elderly patients, have higher peak drug levels and longer drug half-life, so their dose may need to be adjusted downward.

Transition to Oral Treatment: Coumarin Derivatives (Warfarin)

Warfarin is a vitamin K antagonist that inhibits the production of clotting factors II (prothrombin), VII, IX, and X, as well as the naturally occurring anticoagulants protein C and protein S. In patients with DVT, the drug should be started within 24 to 48 hours of initiation of heparin with a goal of achieving international normalized ratio (INR) results between 2.0 and 3.0 (Chapter 38). A higher target INR of 3.0 to 4.0 is associated with more bleeding but no better efficacy, even in patients with the antiphospholipid antibody syndrome

(Chapter 176), and lower intensity warfarin therapy (target INR, 1.5 to 1.9) is significantly less effective at preventing recurrent VTE, despite similar rates of major bleeding. A3  The dose is empirical, but a starting dose of 5 to 10 mg is suitable for most patients. Warfarin doses are adjusted according to the prothrombin time, expressed as the INR, performed daily or every other day until the results are in the therapeutic range for at least 24 hours. After initial dosing, warfarin can be monitored two or three times per week for 1 to 2 weeks and then less frequently, depending on the stability of INR results, up to intervals as long as 4 to 6 weeks. If dose adjustment is needed, such as when medications that can interact with warfarin are introduced, the cycle of more frequent monitoring is repeated until a stable dose response is again achieved. It is now clear that pharmacogenetics have a large impact on the relatively wide range of warfarin dose requirements among different populations and the variability of warfarin requirements over time in any individual patient.5 Polymorphisms in the gene encoding cytochrome P-450 2C9 enzyme, the enzyme that primarily clears the S-enantiomer of warfarin, contribute to variable responses to warfarin. Vitamin K epoxide reductase (VKORC1) recycles vitamin K epoxide to the reduced form of vitamin K and is the target of warfarin. Genotyping for CYP2C9*2, CYP2C9*3, VKORC1 can help guide warfarin dosing and increase the amount of time patients are in the the therapeutic INR range. A4  Polymorphisms are associated with a need for lower doses of warfarin during long-term therapy. Routine pharmacogenetic testing may ultimately be recommended in candidates for long-term (>3 months) warfarin therapy to identify individuals who are likely to require higher or lower warfarin doses.

Long-Term Treatment

The preferred long-term treatment of DVT for most patients is warfarin or another coumarin derivative (e.g., acenocoumarol), continued until the benefits of treatment for reducing recurrent VTE no longer outweigh its risks for major bleeding. The decision to prolong or to stop anticoagulation should be individualized, and a patient’s preferences should be considered.6 Patients with symptomatic proximal DVT or pulmonary emboli should be treated for at least 3 months, even if the VTE was associated with a transient risk factor, A5  but the optimal duration of treatment for patients whose VTE is not associated with a transient risk factor is controversial. Three months of treatment is associated with a 10 to 27% risk for a recurrence during the 12 months after anticoagulant therapy is stopped, whereas 6 months of anticoagulant therapy reduces the risk for recurrence in the first year after stopping to approximately 10%. In patients whose VTE developed in association with minor risk factors (e.g., air travel, pregnancy, within 6 weeks of estrogen therapy, after leg injury or immobilization), the risk for recurrence is probably lower than 10%. Continuation of treatment beyond 6 months reduces the risk for recurrent VTE during the course of therapy, but the benefit is lost after warfarin is discontinued. Current guidelines recommend 3 months of therapy for a first proximal DVT, pulmonary embolism, or both that is provoked by surgery or by a nonsurgical risk factor. For unprovoked VTE, the recommendation is also 3 months if the bleeding risk is high but extended therapy if the bleeding risk is low or moderate. For patients VTE associated with active cancer, extended therapy is recommended using LMWH (see later) rather than warfarin. A6  The most convincing association of thrombophilia with the risk for recurrent VTE is the antiphospholipid antibody (lupus anticoagulant or anticardiolipin antibody [Chapter 176]), which is associated with a two-fold increase in the risk for recurrence. hom*ozygous factor V Leiden, and deficiencies of antithrombin, protein C, and protein S also have been associated with an increased risk for recurrence in some reports, but other data suggest that testing for heritable thrombophilia does not predict recurrent VTE in the first 2 years after anticoagulant therapy is stopped. In the absence of randomized trials to assess different durations of anticoagulation in patients with VTE and thrombophilia, routine testing for thrombophilias need not be performed but should be considered in young ( 5 mm Hg or pulmonary vascular resistance > 6 Wood units not reduced with vasodilators, parenteral inotropic agents, phosphodiesterase type V inhibitors, endothelin receptor antagonists, or a mechanical assist device. II. Relative contraindications 1. Age > 72 yr 2. Any active infection (with exception of device-related infection in ventricular assist device recipients) 3. Active peptic ulcer disease 4. Diabetes mellitus with moderate end-organ involvement (neuropathy, nephropathy, or retinopathy) 5. Severe peripheral vascular or cerebrovascular disease 6. Morbid obesity (BMI > 35) or cachexia (BMI < 18) 7. Significant chronic renal impairment with creatinine > 2.5 mg/dL or creatinine clearance < 25 mL/min* 8. Significant hepatic impairment with bilirubin > 2.5 mg/dL, serum transaminase levels > 3 times normal, INR > 1.5 off warfarin 9. Severe pulmonary dysfunction with FEV1 < 40% normal 10. Recent pulmonary infarction within 6-8 wk 11. Irreversible neurologic or neuromuscular disorder 12. Active mental illness or psychosocial instability 13. Drug, tobacco, or alcohol abuse within 6 mo 14. Significant coagulopathies *May be suitable for cardiac transplantation if inotropic support and hemodynamic management produce a creatinine < 2 mg/dL and creatinine clearance > 50 mL/min. Transplantation may also be advisable as combined heart-kidney transplant. BMI = body mass index; FEV1 = forced expiratory volume in one second; INR = international normalized ratio.

which increases the risk for immediate postoperative right ventricular failure and the 30-day mortality rate. In most patients with advanced heart failure, however, pulmonary hypertension is reversible with vasodilators (Chapter 68) or after implantation of a left ventricular assist device. Although diabetes mellitus with evidence of significant end-organ damage (e.g., neuropathy or nephropathy) is a relative contraindication to heart transplantation, carefully selected patients with diabetes can undergo successful transplantation with morbidity and mortality similar to that in patients

without diabetes. In patients with diabetes who have renal dysfunction (Chapter 124), combined heart and kidney transplantation (Chapter 131) can be considered, with a survival rate comparable to that of heart transplantation alone. Patients with an active or recent malignancy may be offered mechanical support either before or after cancer treatment as a way to bridge them to transplant. However, any patient with a history of malignancy has an increased risk for developing a second malignancy owing to immunosuppression after the transplantation (Chapter 49). Combined heart and stem cell transplantation (Chapter 178) is an option in patients with primary amyloid light-chain amyloidosis (Chapter 188), but survival rates are lower than in other transplant patients because of the frequent recurrence of amyloidosis in the transplanted heart. In contrast, survival of patients with familial amyloidosis caused by a mutant form of the protein transthyretin is comparable to that in other transplant recipients. Repeat transplantation now accounts for 3% of U.S. heart transplants, usually in patients who have developed chronic allograft dysfunction because of severe transplant coronary artery disease, often with a left ventricular ejection fraction less than 45% or with restrictive cardiomyopathy, but without any other significant comorbid conditions. However, repeat transplantation is associated with greater risk for infection and malignancies, owing to the heightened immunosuppression, and a poorer long-term survival. Currently, 3% of adults undergoing cardiac transplantation have complex congenital heart disease (Chapter 69) as the cause of their heart failure. As more patients with complex congenital heart disease survive into adulthood, however, an estimated 10 to 20% of such patients will become candidates for heart or combined heart-lung transplantation at some time during their lives. In such patients, the short-term post-transplant survival is significantly lower compared with patients who have ischemic or dilated cardiomyopathies owing to their higher rate of intraoperative and post-operative bleeding. If a patient with congenital heart disease survives the surgery, however, 10-year survival post-transplant is excellent. Active bacterial infection is a temporary absolute contraindication to heart transplantation, except in the setting of mechanical device infection, in which transplant is felt to be curative. Patients who are positive for human immunodeficiency virus (HIV) and have end-stage cardiomyopathy can be considered for transplant, with good short-term outcome in carefully screened patients who have low or undetectable viral loads and no recent significant bacterial infections. Patients with chronic hepatitis B or C (Chapter 149) have an increased incidence of postoperative liver disease, but their posttransplant survival is not reduced.

Organ allocation Donor Criteria

Donors and recipients are matched for ABO blood compatibility and size. Weight matching is generally within 25% of recipient body weight, though

CHAPTER 82  Cardiac Transplantation  

donors of equal size or larger are preferred for recipients with high PVR. Height mismatches greater than 6 inches are currently not recommended. Males under age 40 and females under age 45 are suitable donors, provided echocardiography shows no evidence of preexisting heart disease or impaired myocardial function. Older individuals also may be suitable donors if coronary atherosclerotic lesions can be excluded, optimally by cardiac catheterization. Donors with serologic findings positive for HIV, hepatitis B and C, and nonprimary brain malignancies are generally not accepted. Organs procured with ischemic times in excess of 4 hours are associated with a higher rate of primary graft failure.

Matching Donors and Recipients

Approximately 10% of transplant candidates have human leukocyte antigen (HLA) antibodies that could lead to a positive crossmatch. The sensitized candidate is typically a multiparous woman, a patient who has received multiple prior transfusions, or patients supported with a mechanical assist device. Patients with high antibody levels require a donor-specific T-cell crossmatch before transplantation to exclude the presence of lymphocytotoxic immunoglobulin G antibodies against donor HLA class I antigens, a situation that can cause hyperacute rejection. Sensitized patients are also at risk for acute humoral rejection and an earlier onset of accelerated coronary artery disease. A donor-specific T-cell crossmatch is a contraindication to transplantation; thus, sensitized candidates have longer wait times before receiving a cardiac allograft. With technologies using solid-phase assays and flow cytometric techniques that can rapidly identify class I and II antibodies, longdistance donors can be screened for unacceptable antigens—a “virtual crossmatch” that enlarges the potential donor pool for sensitized patients.

Surgical Technique

Orthotopic cardiac transplantation can be performed by using a biatrial anastomosis and reconnecting the pulmonary artery and aorta above the semilunar valves. Increasingly, however, atrial function is preserved by performing a bicaval anastomosis, which results in improved atrial geometry, better right ventricular function, less frequent atrioventricular valve regurgitation, and less sinus node dysfunction.

Immunosuppression

Most transplant centers use a triple drug therapy regimen including a calcineurin inhibitor (cyclosporine or tacrolimus), an antiproliferative agent (usually mycophenolate mofetil), and steroids (Chapter 49). Some centers, however, use only a single agent such as tacrolimus. A1  By the third month after surgery, most patients receive only prednisone 5 mg/day and approximately 25 to 60% of patients can tolerate total withdrawal of steroids by the end of the first year. Acute rejection occurs most frequently in the first 3 months after transplant. Some centers use selective induction agents, such as basiliximab, that target the activated interleukin-2 (IL-2) receptor on T cells, or potent nonselective immunosuppressive agents, such as thymoglobulin, in the perioperative period to decrease early allograft rejection. However, this intensification of immunosuppression can predispose patients to more frequent opportunistic infections. Currently, approximately 50% of transplant centers use induction therapy. For calcineurin inhibitors, prospective randomized trials have demonstrated a decreased incidence of allograft rejection, less hypertension, lower lipid levels, less hirsuitism, and less gingival hyperplasia using tacrolimus compared with cyclosporine, although most studies demonstrating no difference in survival. A2-A5  Mycophenolate mofetil, a selective de novo purine inhibitor, is the preferred antiproliferative agent to reduce rejection and the development of transplant vasculopathy. A6  Although everolimus may be better than mycophenolate mofetil in preventing early transplant vasculopathy, it also is associated with more side effects, A7  so mycophenolate mofetil remains the current agent of choice.

Rejection

Allograft rejection, which can be antibody-mediated or cell-mediated, occurs most frequently in the first 6 months after the transplant. Antibody-mediated, or humoral, rejection generally occurs very early after the transplant, particularly in a previously sensitized recipient. It is characterized histologically by immunoglobulin and complement deposition in the absence of cellular rejection, and often it is associated with hemodynamic compromise. T-cell mediated rejection, triggered by the recognition of foreign antigens on the surface

521

TABLE 82-3  HISTOLOGIC GRADING OF CELLULAR REJECTION* Grade 0R

No rejection

Grade 1R (mild)

Interstitial, perivasicular, or both infiltrate with up to 1 focus of myocyte damage

Grade 2R (moderate)

≥2 foci of infiltrate with myocyte damage

Grade 3R (severe)

Diffuse infiltrate with multifocal myocyte damage, edema, hemorrhage, vasculitis

*Modified from Stewart S, Winters GL, Fishbein MC, et al. Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection. J Heart Lung Transplant. 2005;24:1710-1720.

TABLE 82-4  2013 INTERNATIONAL SOCIETY FOR HEART & LUNG TRANSPLANTATION CLASSIFICATION FOR DIAGNOSIS OF CARDIAC ANTIBODY-MEDIATED REJECTION (AMR)* GRADE

DEFINITION

pAMR 0

No rejection

pAMR 1 (H+)

Histopathologic changes are present without immunopathologic findings.

pAMR 1 (I+)

Immunopathologic findings are positive without histologic findings.

pAMR 2

Both immunologic and histologic findings are present.

pAMR 3

Severe pathologic antibody-mediated rejection with interstitial hemorrhage, capillary fragmentation, mixed inflammatory infiltrates, endothelial cell pyknosis, karyorrhexis, or a combination of these, and marked edema and immunopathologic findings are present. These cases may be associated with profound hemodynamic dysfunction and poor clinical outcomes.

*Modified from Berry GJ, Burke MM, Andersen C, et al. The 2013 International Society for Heart and Lung Transplantation Working Formulation for the standardization of nomenclature in the pathologic diagnosis of antibody-mediated rejection in heart transplantation. J Heart Lung Transplant. 2013;32:1147-1162.

of engrafted cells, accounts for more than 90% of rejection episodes, usually within the first 6 months. To diagnose allograft rejection, endomyocardial biopsies are usually performed by the transjugular approach weekly for the first month, then every other week for 2 months, then every 1 to 2 months for the first year. Biopsy grading of cellular rejection is based on the severity of lymphocyte infiltration and myocyte necrosis on hematoxylin and eosin staining (Table 82-3), whereas antibody-mediated rejection also includes immunologic staining (Table 82-4). Although the presence of donor-specific antibodies is not a requirement for the diagnosis of antibody-mediated rejection, patients suspected of having antibody-mediated rejection are frequently tested and serially monitored for donor-specific antibody. In asymptomatic patients on low doses of steroids, gene expression profiling of peripheral blood samples can reduce the number of biopsies while providing equivalent clinical outcomes. A8 

Treatment of Rejection

The treatment of allograft rejection is determined by the presence of symptoms, the degree of left ventricular dysfunction, the time since transplant, and the pathologic grade of the biopsy. Most episodes of cellular rejection are easily treated with high-dose oral or intravenous steroids. Patients with hemodynamically significant rejection will require rescue therapy with thymoglobulin. Humoral rejection therapy includes modalities that both clear and reduce the production of the antibody, such as plasmapheresis, intravenous immunoglobulin, thymoglobulin, high-dose steroids, and B cell–specific monoclonal antibodies such as rituximab8 (Chapter 49).

Transplant Vasculopathy

Transplant vasculopathy, which is primarily a form of chronic rejection, occurs at an annual incidence rate of 5 to 10% and remains one of the main causes of late death after cardiac transplantation. Risk factors for transplant

522

CHAPTER 82  Cardiac Transplantation  

vasculopathy include an increased number of HLA mismatches, increased number of acute rejection episodes, older donor age, prior cytomegalovirus (CMV) infection, ischemia-reperfusion injury, and the classic risk factors for atherosclerotic disease—age, smoking, obesity, diabetes, dyslipidemia, and hypertension. Histologic examination shows subendothelial accumulation of primarily T cells, myointimal proliferation of smooth muscle cells, lipidladen foam cells, and perivascular fibrosis. Patients rarely experience exertional angina and are more likely to present with severe fatigue, heart failure, myocardial infarction, ventricular arrhythmia, or sudden death. Patients should be screened annually either by angiography, with or without intravascular ultrasound, or by dobutamine stress echocardiography. An increase in intimal thickness of at least 0.5 mm, as detected by intravascular ultrasound, is a reliable indicator of both cardiac allograft vasculopathy and 5-year mortality. In contrast to routine coronary artery disease, cardiac allograft vasculopathy is usually manifested by concentric narrowing, owing to neointimal proliferation of vascular smooth muscle cells throughout the length of the vessel and angiographic evidence of rapid tapering, pruning, and obliteration of vessels. Some patients may have focal coronary lesions that are amenable to stent placement, but generally the disease is diffuse and not amenable to percutaneous coronary interventions or bypass grafting. Treatment has been predominantly aimed at prevention. Use of high-dose or low-dose statins (see Table 206-6 in Chapter 206) can reduce the development of cardiac vasculopathy. A9  Sirolimus and everolimus also can reduce the incidence of cardiac vasculopathy and slow its progression. However, the only definitive treatment of transplant vasculopathy is repeat transplantation, and the survival of patients undergoing repeat transplantation for severe vasculopathy is comparable to that of de novo heart transplant recipients.

Patients with positive tuberculosis skin tests should be treated with isoniazid and pyridoxine (Chapter 324), and patients with latent syphilis should be treated with penicillin (Chapter 319). Infections early after transplant are predominantly bacterial from hospitalacquired organisms (Chapter 282), catheters, the surgical site, a prior left ventricular assist device, or occasionally from donor-transmitted disease. Any infection early after transplantation increases the risk for a subsequent fatal CMV infection (Chapter 376). Any CMV-positive patient or CMV-negative patient receiving a CMV-positive organ should receive prophylactic ganciclovir followed by valganciclovir. Fungal and viral infections are usually more frequent starting a month or so after transplant. Antibiotic prophylaxis includes perioperative antibacterial agents, such as cefazolin, and initiation of prophylaxis against CMV infection, Pneumocystis jiroveci pneumonia, herpes simplex virus infection, and oral candidiasis (Chapter 338). The prophylactic use of one single-strength trimethoprim-sulfamethoxazole tablet daily, typically for the first year after transplantation, has virtually eliminated P. jiroveci (Chapter 341) and also prevents nocardial infections (Chapter 330) and toxoplasmosis (Chapter 349). Aspergillosis (Chapter 339) and candidiasis (Chapter 338) are the most common fungal infections after heart transplantation; oral nystatin solution or clotrimazole troches are routinely used in the first 3 to 6 months (Chapter 331).

Comorbid Conditions

Within 5 years, hypertension (Chapter 67) occurs in over 92% of transplant recipients, primarily owing to the side effects of calcineurin inhibitors. Diabetes (Chapter 229) is observed in approximately 40% of patients owing to treatment with corticosteroids or tacrolimus. Hyperlipidemia is found in approximately 90% of patients and is treated with statins, with the same approach as for patients with known coronary disease (Chapter 206). Osteoporosis (Chapter 243) is also common because of chronic steroid treatment as well as the use of calcineurin inhibitors. Significant renal insufficiency (serum creatinine > 2.5 mg/dL) occurs in 20% of patients by 10 years post-transplant.

Malignancy

With the improved survival of heart transplant recipients and longer exposure to immunosuppressive drugs, malignancies, especially lymphomas (Chapter 185), are now almost equal to transplant vasculopathy as the leading cause of long-term mortality. Post-transplant lymphoproliferative disease (Chapters 49 and 185) includes a spectrum of predominantly B-cell lymphomas (~90%) frequently associated with Epstein-Barr virus (Chapter 377). T-cell lymphomas are much less frequent (10%) and often more difficult to treat (Chapter 185). Skin cancers (Chapter 203) are also more frequent in post-transplant patients, but the incidence of solid organ malignancies is not much different from that in the general population.

PROGNOSIS

During the first year after transplantation, early causes of death are graft failure, infection, multiorgan failure, and allograft rejection. Overall survival, which is approximately 85% at 1 year and 75% at 5 years, has been improving over time (Fig. 82-2). After 5 years, cardiac allograft vasculopathy (14%), late graft failure (18%), malignant disease (25%), and non-CMV infection (10%) are the most prominent causes of death. Of the patients, 50% will survive for more than 11 years and many survive for 20 to 30 years, often with good ventricular function.9 Functional capacity is usually excellent, with more than 90% of 1-year survivors reporting no functional limitations. Exercise capacity improves but remains reduced compared with that of age- and gender-matched controls, owing to denervation of the heart, side effects of immunosuppressive drugs,

Infection

Infections (Chapter 281) account for approximately 20% of deaths within the first year after transplant surgery and continue to be a common cause of morbidity and mortality throughout the recipient’s life. In the waiting period, careful attention should be given to updating immunizations against pneumococcal pneumonia, hepatitis B, and herpes zoster (Chapter 18). Young women should receive vaccination against the human papilloma virus. 100

All pair-wise comparisons were significant at p < 0.0001 except 2002–2005 vs. 2006–6/2011 (p = 0.9749)

Survival (%)

80

60 1982–1991 (N = 21,342) 1992–2001 (N = 38,966) 2002–2005 (N = 13,496) 2006–6/2011 (N = 18,896)

40

20

0 0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

Years FIGURE 82-2.  Survival by transplant era. Adult heart transplants: Kaplan-Meier survival by era, January 1982-June 2011. (From Lund LH, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: thirtieth official adult heart transplant report—2013; focus theme: age. J Heart Lung Transplant. 2013;32:951-964.

CHAPTER 82  Cardiac Transplantation  

and donor-recipient size mismatch. Many patients return to full-time employment, travel extensively, and participate in vigorous sports such as skiing, running, and hiking. Some young patients may bear children of their own, though genetic counseling is recommended in patients with familial cardiomyopathies (Chapter 60). In female recipients, pregnancy, which requires modification of immunosuppression and close monitoring for allograft rejection, is recommended only in 1-year survivors with normal graft function.

FUTURE DIRECTIONS

With the continuing scarcity of donor organs, physicians must use this scarce resource wisely by selecting candidates who would benefit the greatest over time. In the future, warm preservation techniques may extend harvest time as well and enable the resuscitation of donor organs whose function may initially appear marginal. The field of mechanical circulatory support continues to evolve rapidly, and devices (Chapter 59) rather than transplantation will probably offer a greater chance for long-term survival for most patients with advanced heart failure.

Grade A References A1. Baran DA, Zucker MJ, Arroyo LH, et al. A prospective, randomized trial of single-drug versus dualdrug immunosuppression in heart transplantation: the tacrolimus in combination, tacrolimus alone compared (TICTAC) trial. Circ Heart Fail. 2011;4:129-137.

523

A2. Ye F, Ying-Bin X, Yu-Guo W, et al. Tacrolimus versus cyclosporine microemulsion for heart transplant recipients: a meta-analysis. J Heart Lung Transplant. 2009;28:58-66. A3. Penninga L, Moller CH, Gustafsson F, et al. Tacrolimus versus cyclosporine as primary immunosuppression after heart transplantation: systematic review with meta-analyses and trial sequential analyses of randomised trials. Eur J Clin Pharmacol. 2010;66:1177-1187. A4. Sanchez-Lazaro IJ, Almenar L, Martinez-Dolz L, et al. A prospective randomized study comparing cyclosporine versus tacrolimus combined with daclizumab, mycophenolate mofetil, and steroids in heart transplantation. Clin Transplant. 2011;25:606-613. A5. Guethoff S, Meiser BM, Groetzner J, et al. Ten-year results of a randomized trial comparing tacrolimus versus cyclosporine a in combination with mycophenolate mofetil after heart transplantation. Transplantation. 2013;95:629-634. A6. Eisen HJ, Kobashigawa J, Keogh A, et al. Three-year results of a randomized, double-blind, controlled trial of mycophenolate mofetil versus azathioprine in cardiac transplant recipients. J Heart Lung Transplant. 2005;24:517-525. A7. Eisen HJ, Kobashigawa J, Starling RC, et al. Everolimus versus mycophenolate mofetil in heart transplantation: a randomized, multicenter trial. Am J Transplant. 2013;13:1203-1216. A8. Pham MX, Teuteberg JJ, Kfoury AG, et al. Gene-expression profiling for rejection surveillance after cardiac transplantation. N Engl J Med. 2010;362:1890-1900. A9. Som R, Morris PJ, Knight SR. Graft vessel disease following heart transplantation: a systemic review of the role of statin therapy. World J Surg. 2014;38:2324-2334.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 82  Cardiac Transplantation  

GENERAL REFERENCES 1. Lund LH, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: thirtieth official adult heart transplant report—2013; focus theme: age. J Heart Lung Transplant. 2013;32:951-964. 2. Kirklin JK, Naftel DC, Kormos RL, et al. Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant. 2013;32: 141-156. 3. Colvin-Adams M, Smithy JM, Heubner BM, et al. OPTN/SRTR 2012 Annual Data Report: heart. Am J Transplant. 2014;14(suppl 1):113-138. 4. Yancy CW, Jessup M, Bozkurt B, et al. 2013 ACCF/AHA guideline for the management of heart failure: executive summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation. 2013;128:1810-1852.

523.e1

5. Mancini D, Lietz K. Selection of cardiac transplantation candidates in 2010. Circulation. 2010; 122:173-183. 6. Colvin-Adams M, Valapour M, Hertz M, et al. Lung and heart allocation in the United States. Am J Transplant. 2012;12:3213-3234. 7. Schulze PC, Kitada S, Clerkin K, et al. Regional differences in recipient waitlist time and pre- and post-transplant mortality after the 2006 United Network for Organ Sharing policy changes in the donor heart allocation algorithm. JACC Heart Fail. 2014;2:166-177. 8. Kobashigawa J, Crespo-Leiro MG, Ensminger SM, et al. Report from a consensus conference on antibody-mediated rejection in heart transplantation. J Heart Lung Transplant. 2011;30:252-269. 9. Galeone A, Kirsch M, Barreda E, et al. Clinical outcome and quality of life of patients surviving 20 years or longer after heart transplantation. Transpl Int. 2014;26:576-582.

526

CHAPTER 83  Approach to the Patient with Respiratory Disease  

83  APPROACH TO THE PATIENT WITH RESPIRATORY DISEASE MONICA KRAFT Respiratory symptoms, which are among the most common reasons why patients seek medical care, are responsible for approximately 20% of office visits to a primary care physician. In addition to a careful history, a systematic physical examination is critical for accurate diagnosis. A careful pulmonary examination complements the cardiac physical examination (Chapter 51). Inspection may reveal an elevated jugular pressure, indicative of right heart failure owing to cor pulmonale (Chapter 68). Cervical or supraclavicular adenopathy (Chapter 168) may be the first clue to suggest a thoracic malignancy (Chapter 191) or mycobacterial infection (Chapter 324). Unilateral arm swelling can be caused by venous thrombosis (Chapter 81), whereas venous engorgement of the head and neck can be caused by a tumor that results in superior vena cava syndrome (see Fig. 99-8 in Chapter 99). On the cardiac examination, a loud pulmonic second heart sound is suggestive of pulmonary hypertension, which also can result in a murmur of tricuspid (see Table 51-7 in Chapter 51) or pulmonic valve insufficiency. Inspection of the chest may show hyperinflation and reduced diaphragmatic excursion, typical of chronic obstructive pulmonary disease (COPD; Chapter 88), chest wall abnormalities such as kyphoscoliosis (Chapter 99), or diaphragmatic muscle wall weakness as in many hypoventilation syndromes (Chapter 86). Percussion may reveal dullness in patients with pleural effusions or with lung that has been consolidated by pneumonia. Auscultation of the lungs1 includes listening at both apices and over both upper and lower lobes, anteriorly and posteriorly, and during inspiration and respiration. Normal lung sounds are heard during inspiration and early expiration as soft and non-musical sounds (Table 83-1). Bronchial breath sounds, which sound similar to but often somewhat harsher than normal lung sounds, are heard throughout expiration as well as inspiration, similar to what would be heard by placing a stethoscope over the trachea. The term rales is no longer used and has been replaced by the term crackles. Fine crackles are non-musical and heard typically in late inspiration; they are most commonly a sign of heart failure (Chapter 58) or interstitial lung disease (Chapter 92). By comparison, coarse crackles, which unlike fine crackles tend to be transmitted through the mouth and cleared by coughing, are typical of bronchitis (Chapter 96) and COPD (Chapter 88). Wheezes are high-pitched, musical sounds heard during expiration and sometimes inspiration, most commonly in asthma (Chapter 87) and sometimes in COPD (Chapter 88). When these diseases are severe, however, the degree of airflow may be insufficient to produce wheezes. A rhonchus is a musical, low-pitched sound typically heard in expiration and sometimes during inspiration; it often resolves with coughing. Like coarse crackles, rhonchi are common in bronchitis (Chapter 96) and COPD (Chapter 88). A pleural friction rub, which classically occurs during inspiration but sometimes also during expiration, is heard in patients with inflammatory diseases or malignancies

TABLE 83-1  DIAGNOSTIC UTILITY OF LUNG AUSCULTATION AUSCULTATORY FINDING

CLINICAL CORRELATION

Bronchial breathing

Pneumonia or interstitial lung disease

Fine crackle

Heart failure, interstitial lung disease, alveolar filling disorders

involving the pleura (Chapters 99 and 191). Stridor is a musical, high-pitched sound that may be audible without a stethoscope and that indicates upper airway obstruction, such as found with acute inflammatory or chronic degenerative diseases of the larynx (Chapter 429) or obstruction of the trachea, as may be caused by intrathoracic malignant diseases (Chapter 191). An absence of breath sounds would be noted if the lung is not ventilated because of a complete bronchial obstruction or if it is displaced by a pleural effusion. Tactile fremitus, which is a vibratory sensation noted during breathing, is increased in patients who have consolidated lung from pneumonia, because the vibratory sensation conducts better through such lung tissue and is diminished in patients with pleural effusion. Egophony, by which a patient’s recitation of the long E sound is heard on auscultation as a long A sound, is another indication of consolidation typical of pneumonia. Evaluation of the abdomen may show a readily palpable liver, sometimes mistaken for hepatomegaly, in patients with COPD and low diaphragm. Examination of the extremities may reveal cyanosis in patients who are hypoxemic, usually with a partial pressure of oxygen less than 55 mm Hg, although it also may be observed in patients with methemoglobinemia (Chapter 158). Clubbing (Chapter 51) is indicative of chronic hypoxemia, as seen in patients with chronic right-to-left-shunting from congenital heart disease (Chapter 69) or other causes of long-standing hypoxemia (Chapters 88 and 92), but it also may be indicative of pleural-based diseases (Chapter 99) as part of the syndrome of hypertrophic pulmonary osteoarthropathy (Chapters 179 and 275). In patients with suspected hypoxemia, careful analyses of arterial blood gases can help determine its severity and guide therapy (Chapter 103). In patients in whom it is difficult to distinguish heart failure from a pulmonary cause of hypoxemia, an elevated brain natriuretic peptide level may point to a cardiac cause (Chapter 58). Chest imaging (Chapter 84) is a crucial part of the evaluation of many potential pulmonary complaints, and pulmonary function testing (Chapter 85) can be extremely helpful in distinguishing among causes of acute and chronic lung disease. Among the most common respiratory complaints are cough, wheezing, dyspnea, and hemoptysis. Each can and should be approached in a systematic way.

  APPROACH TO THE PATIENT WITH COUGH

Cough is the single most common respiratory complaint for which patients seek care. Referrals of patients with persistently troublesome chronic cough of unknown cause account for 10 to 38% of outpatient visits to respiratory specialists. For acute cough, defined as coughing that has been present for less than 8 weeks, a careful medical history and physical examination will usually reveal the diagnosis (Table 83-2). Although most acute coughs are of minor

TABLE 83-2  SPECTRUM OF CAUSES AND FREQUENCIES OF COUGH IN IMMUNOCOMPETENT ADULTS COMMON

LESS COMMON

ACUTE COUGH Common cold

Asthma

Acute bacterial sinusitis

Pneumonia

Pertussis

Heart failure

Exacerbations of COPD

Aspiration syndromes

Allergic rhinitis

Pulmonary embolism

Environmental irritant rhinitis

Exacerbation of bronchiectasis

CHRONIC COUGH Rhinosinus conditions/UACS

Bronchogenic carcinoma

Asthma

Chronic interstitial pneumonia

Gastroesophageal reflux

Sarcoidosis

Chronic bronchitis

Left heart failure

Coarse crackle

Bronchitis

Wheeze

Asthma, COPD

Rhonchus

Bronchitis, COPD

Eosinophilic bronchitis

Obstructive sleep apnea

Stridor

Upper-airway obstruction from laryngeal or tracheal inflammation, mass lesions, or external compression

Bronchiectasis

Chronic tonsillar enlargement

ACE inhibitors

Pleural friction rub

Pleural inflammation or tumors

COPD = chronic obstructive pulmonary disease.

Postinfection ACE = angiotensin-converting enzyme; COPD = chronic obstructive pulmonary disease; UACS = upper airway cough syndrome.

527

CHAPTER 83  Approach to the Patient with Respiratory Disease  

consequence, cough can occasionally be a sign of a potentially life-threatening illness, such as pulmonary embolism (Chapter 98), pneumonia (Chapter 97), or heart failure (Chapter 58). Up to 98% of all cases of chronic cough, defined as a cough that persists for more than 8 weeks, in immunocompetent adults are caused by eight common conditions: postnasal drip syndrome from a variety of rhinosinus conditions (Chapter 251), asthma (Chapter 87), gastroesophageal reflux disease (GERD) (Chapter 138), chronic bronchitis (Chapter 88), eosinophilic bronchitis, bronchiectasis (Chapter 90), use of angiotensin-converting enzyme (ACE) inhibitors, and postinfectious cough. Postinfectious cough is usually nonproductive and lasts for 3 to 8 weeks after an upper respiratory tract infection; patients have a normal chest radiograph. Uncommon causes of chronic cough include bronchogenic carcinoma (Chapter 191), chronic interstitial pneumonia (Chapter 92), sarcoidosis (Chapter 95), left ventricular failure (Chapter 58), and aspiration (Chapter 94).

DIAGNOSIS

In chronic cough (Fig. 83-1), the character and timing are not of diagnostic help. A chest radiograph should be obtained in all patients, but other tests should not be ordered in current smokers or patients taking ACE inhibitors until the response to smoking cessation or discontinuation of the drug for at least 4 weeks can be assessed. Sinus radiographs, barium esophagography, methacholine challenge, esophageal pH, and bronchoscopy can be ordered as part of the initial evaluation, depending on the history and physical examination (Table 83-3; see Fig. 83-1). If a test points toward a possible diagnosis, a trial of treatment for that condition is needed to confirm the diagnosis.2

TREATMENT  The specific cause of cough can be diagnosed and treated successfully 84 to 98% of the time, so nonspecific therapy3 aimed to suppress the cough per se is rarely indicated. There is no strong evidence that nonspecific therapies such as antitussives, mucolytics, decongestants, or antihistamine-decongestant

TABLE 83-3  TESTING CHARACTERISTICS OF DIAGNOSTIC PROTOCOL FOR EVALUATION OF CHRONIC COUGH TESTS

DIAGNOSIS

POSITIVE PREDICTIVE VALUE, %

NEGATIVE PREDICTIVE VALUE, %

Sinus radiograph

Sinusitis

57-81

95-100

Methacholine inhalation challenge

Asthma

60-82

100

Modified barium esophagography

GERD, esophageal stricture

38-63

63-93

Esophageal pH*

GERD

89-100

Bronchoscopy

Endobronchial mass/lesion

50-89

100

*24-Hour esophageal pH monitoring. GERD = gastroesophageal reflux disease.

Chronic Cough

Investigate and treat

Inadequate response to optimal Rx

A cause of cough is suggested

History, physical examination, chest radiograph

Smoking ACE-inhibitor

Discontinue

No response Post-nasal drip/rhinitis/sinusitis Empiric treatment (Chapter 426) with anti-histamine/decongestant, nasal saline irrigation Asthma (Chapter 87) Evaluate via spirometry, bronchodilator reversibility, methacholine challenge; then treat with inhaled corticosteroids, beta-adrenergic inhalers, leukotriene receptor antagonists (Chapter 87); Empiric treatment as a second option Gastroesophageal Reflux Disease (GERD) Empiric treatment (Chapter 138) with protein pump inhibitor, diet/lifestyle

Inadequate response to optimal Rx Further Investigations to consider if empiric treatments partially effective or ineffective (see Table 83-3 for testing regarding specific diagnoses): • 24h esophageal pH monitoring • endoscopic or videofluoroscopic swallow evaluation • barium esophagram • sinus imaging • HRCT • bronchoscopy • echocardiogram • environmental assessment • polysomnogram

Important General Considerations Optimize therapy for each diagnosis Check adherence with medications Due to the possibility of multiple causes, maintain all partially effective treatments

FIGURE 83-1.  Algorithm for the management of chronic cough lasting longer than 8 weeks. ACE = angiotensin-converting enzyme; HRCT = high-resolution computed tomography; Rx = prescription.

528

CHAPTER 83  Approach to the Patient with Respiratory Disease  

combinations are efficacious for acute cough in the setting of an upper respiratory tract infection. A1  For nonspecific persistent cough, effective treatment of chronic gastroesophageal reflux disease with a proton pump inhibitor (Chapter 138) provides no more than modest benefit, with approximately one in five patients improving. A2  Inhaled corticosteroids can reduce cough but should be used only after evaluation by chest radiography and often spirometry. A3  Dextromethorphan and codeine-containing cough suppressants can reduce chronic cough by approximately 40%. In adults with refractory chronic cough without active respiratory disease or infection, gabapentin (up to a maximum daily dose of 1800 mg) significantly improves cough-specific quality of life compared with placebo. A4  For chronic refractory cough despite comprehensive evaluation and opioid therapy, a combination of education, breathing exercises, cough suppression techniques, and counseling can significantly reduce cough and its negative impact on quality of life. A5  Coughing can also be reduced by training patients to focus externally rather than internally. A6 

  APPROACH TO THE PATIENT WITH WHEEZING

Wheeze is a continuous musical sound that lasts longer than 80 to 100 msec, likely generated by flow through critically narrowed collapsible bronchi. Although expiratory wheezing is a common physical finding in asthma (Chapter 87), the many causes of wheezing (Table 83-4) (e.g., COPD [Chapter 88], pulmonary edema [Chapter 58], bronchiolitis [Chapter 92], bronchiectasis [Chapter 90], and less common entities such as carcinoid [Chapter 232] and parasitic infections) often can be distinguished based

on the history, physical examination, and pulmonary function testing (Chapter 85).4

DIAGNOSIS

On pulmonary function testing, the shape of inspiratory and expiratory flowvolume loops provide key information about the presence of airway obstruction and whether the obstruction is extrathoracic or intrathoracic (Fig. 83-2). An important cause of extrathoracic obstruction is vocal cord lesions (Chapter 190). Variable intrathoracic obstruction can be caused by tracheomalacia, whereas fixed upper airway obstruction can be caused by a proximal tracheal tumor.

TREATMENT  Treatment of the specific cause will usually lead to complete or at least partial resolution of wheezing. However, treatment of associated asymptomatic or minimally symptomatic gastroesophageal reflux disease is not beneficial. A7 

  APPROACH TO THE PATIENT WITH DYSPNEA

Dyspnea is the sensation of difficult, labored, or unpleasant breathing. The word unpleasant is very important to this definition because the labored or difficult breathing encountered by healthy individuals while exercising does

TABLE 83-4  DIAGNOSIS OF SELECTED WHEEZING ILLNESSES OTHER THAN ASTHMA DISEASES

DISTINGUISHING FEATURES

UPPER AIRWAY DISEASES Postnasal drip syndrome

History of postnasal drip, throat clearing, nasal discharge; physical examination shows oropharyngeal secretions or cobblestone appearance to mucosa.

Epiglottis

History of sore throat out of proportion to pharyngitis. Evidence of supraglottitis on endoscopy or lateral neck radiographs.

Vocal cord dysfunction syndrome

Lack of symptomatic response to bronchodilators, presence of stridor plus wheeze in absence of increased P(A-a)o2; extrathoracic variable obstruction on flow-volume loops; paradoxical inspiratory, and/or early expiratory adduction of vocal cords on laryngoscopy during wheezing. This syndrome can masquerade as asthma, be provoked by exercise, and often coexists with asthma.

Retropharyngeal abscess

History of stiff neck, sore throat, fever, trauma to posterior pharynx; swelling noted by lateral neck or CT radiographs.

Laryngotracheal injury due to tracheal cannulation

History of cannulation of trachea by endotracheal or tracheostomy tube; evidence of intrathoracic or extrathoracic variable obstruction on flow-volume loops, neck and chest radiographs, laryngoscopy, or bronchoscopy.

Neoplasms

Bronchogenic carcinoma, adenoma, or carcinoid tumor is suspected when there is hemoptysis, unilateral wheeze, or evidence of lobar collapse on chest radiograph or combinations of these; diagnosis is confirmed by bronchoscopy.

Anaphylaxis

Abrupt onset of wheezing with urticaria, angioedema, nausea, diarrhea, and hypotension, especially after insect bite, in association with other signs of anaphylaxis such as hypotension or hives, or administration of drug or IV contrast, or family history.

LOWER AIRWAY DISEASES COPD

History of dyspnea on exertion and productive cough in cigarette smoker. Because productive cough is nonspecific, it should only be ascribed to COPD when other cough-phlegm syndromes have been excluded, forced expiratory time to empty more than 80% of vital capacity >4 sec, and there is decreased breath sound intensity, unforced wheezing during auscultation, and irreversible, expiratory airflow obstruction on spirometry.

Pulmonary edema

History and physical examination consistent with passive congestion of the lungs, ARDS, impaired lung lymphatics; abnormal chest radiograph, echocardiogram, radionuclide ventriculography, cardiac catheterization, or combinations of these.

Aspiration

History of risk for pharyngeal dysfunction or gastroesophageal reflux disease; abnormal modified barium swallow, 24-hr esophageal pH monitoring, or both.

Pulmonary embolism

History of risk for thromboembolic disease, positive confirmatory tests.

Bronchiolitis

History of respiratory infection, connective tissue disease, transplantation, ulcerative colitis, development of chronic airway obstruction over months to a few years rather than over many years in a nonsmoker; mixed obstructive and restrictive pattern on PFTs and hyperinflation; may be accompanied by fine nodular infiltrates on chest radiograph.

Cystic fibrosis

Combination of productive cough, digital clubbing, bronchiectasis, progressive COPD with Pseudomonas sp colonization and infection, obstructive azoospermia, family history, pancreatic insufficiency, and two sweat chloride determinations of > 60 mEq/L; some patients are not diagnosed until adulthood, in one instance as late as age 69 yr; when sweat test is occasionally normal, definitive diagnosis may require nasal transepithelial voltage measurements and genotyping.

Carcinoid syndrome

History of episodes of flushing and watery diarrhea; elevated 5-hydroxyindoleactic acid level in 24-hr urine specimen.

Bronchiectasis

History of episodes of productive cough, fever, or recurrent pneumonias; suggestive chest radiographs or typical chest CT findings; ABPA should be considered when bronchiectasis is central.

Lymphangitic carcinomatosis

History of dyspnea or prior malignancy; reticulonodular infiltrates with or without pleural effusions; suggestive high-resolution chest CT scan; confirmed by bronchoscopy with biopsies.

Parasitic infections

Consider in a nonasthmatic patient who has traveled to an endemic area and complains of fatigue, weight loss, fever; peripheral blood eosinophilia; infiltrates on chest radiograph; stools for ova and parasites for nonfilarial causes; blood serologic studies for filarial causes.

ABPA = allergic bronchopulmonary aspergillosis; ARDS = acute respiratory distress syndrome; COPD = chronic obstructive pulmonary disease; CT = computed tomography; IV = intravenous; P(A-a)o2 = alveolar-arterial oxygen tension gradient; PFTs = pulmonary function tests.

CHAPTER 83  Approach to the Patient with Respiratory Disease

•

V (L/S) Inspiration Expiration

A

7 6 5 4 3 2 1 0 1 2 3 4 5 6

B

C

D

529

E

TABLE 83-5  DISEASES THAT CAUSE DYSPNEA GROUPED BY PHYSIOLOGIC MECHANISMS OF ACTION* INCREASED RESPIRATORY DRIVE Stimulation of Chemoreceptors

100

0100

0100 0100 Vital capacity (%)

0 100

FIGURE 83-2.  Schematic flow-volume loop configurations in a spectrum of airway lesions. A is normal; B is variable extrathoracic upper airway obstruction; C is variable intrathoracic upper airway lesion; D is fixed upper airway obstruction; and E is small airway obstruction. L/S = liters per second; = ventilation.

not qualify as dyspnea because it is at the level expected for the degree of exertion. The sensation of dyspnea is often poorly or vaguely described by the patient. The physiology of dyspnea remains unclear, but multiple neural pathways can be involved in processes that lead to dyspnea. In acute dyspnea, or shortness of breath of sudden onset, the history, physical examination, and laboratory testing must first focus on potential lifethreatening conditions, including pulmonary embolism (Chapter 98), pulmonary edema (Chapters 58 and 59), acute airway obstruction from anaphylaxis or foreign bodies, pneumothorax (Chapter 99), or pneumonia (Chapter 97). For chronic dyspnea, specific conditions to consider include COPD (Chapter 88), asthma (Chapter 87), interstitial lung disease (Chapter 92), heart failure (Chapter 58), cardiomyopathy (Chapter 60), GERD (Chapter 138), other respiratory diseases, or hyperventilation syndrome (Table 83-5).

DIAGNOSIS

A chest radiograph, electrocardiogram (ECG), pulmonary function testing, and an exercise test with electrocardiographic monitoring and pulse oximetry at rest and during exercise are key tests to assess patients with unexplained dyspnea (Fig. 83-3).5 For acute dyspnea, B-type natriuretic peptide testing can be extremely helpful in distinguishing heart failure from other causes. A8  The utility of more detailed pulmonary testing with maximal inspiratory and expiratory pressures, flow-volume loops, with or without methacholine challenge, computed tomographic screening of the chest, and echocardiography depends on history and physical examination and the results of these tests. When GERD is a suspected cause of dyspnea, a modified barium esophagogram or 24-hour esophageal pH monitoring, or both, should be considered (Chapter 138). Other more invasive tests such as cardiac catheterization or lung biopsy may be indicated when the results of less invasive tests have not been conclusive.

TREATMENT  Whenever possible, the final determination of the cause of dyspnea is made by observing which specific therapy eliminates it. Because dyspnea may be simultaneously the result of more than one condition, it may be necessary to treat more than one condition.

  APPROACH TO THE PATIENT WITH HEMOPTYSIS Hemoptysis is the expectoration of blood from the lung parenchyma or airways.6 Hemoptysis may be scant, with just the appearance of streaks of bright red blood in the sputum, or massive, with the expectoration of a large volume of blood. Massive hemoptysis, which is defined as the expectoration of at least 600 mL of blood in 24 to 48 hours, may occur in 3 to 10% of patients with hemoptysis. Dark red clots also may be expectorated when the blood has been present in the lungs for days. Pseudohemoptysis, which is the expectoration of blood from a source other than the lower respiratory tract, may cause diagnostic confusion when patients cannot clearly describe the source of the bleeding. Pseudohemoptysis can occur when blood from the oral cavity, nares, pharynx, or tongue clings to the back of the throat and initiates the cough reflex, or when patients

Conditions leading to acute hypoxemia Impaired gas exchanger (e.g., asthma, pulmonary embolism, pneumonia, congestive heart failure†) Environmental hypoxia (e.g., altitude, contained space with fire) Conditions leading to increased dead space, acute hypercapnia Impaired gas exchanger (e.g., acute, severe asthma; exacerbation of COPD; severe pulmonary edema) Impaired ventilator pump (e.g., muscle weakness, airflow obstruction) Metabolic acidosis Renal disease (e.g., renal failure, renal tubular acidosis) Decreased oxygen carrying capacity (e.g., anemia) Decreased release of oxygen to tissues (e.g., hemoglobinopathy) Decreased cardiac output Stimulation of Pulmonary Receptors (irritant, mechanical, vascular)‡ Interstitial lung disease Pleural effusion (compression atelectasis) Pulmonary vascular disease (e.g., thromboembolism, idiopathic pulmonary hypertension) Heart failure Mild asthma Behavioral Factors Hyperventilation syndrome, anxiety disorders, panic attacks VENTILATORY PUMP: INCREASED EFFORT OR WORK OF BREATHING Muscle Weakness Myasthenia gravis, Guillain-Barré syndrome, spinal cord injury, myopathy, postpoliomyelitis syndrome Decreased compliance of the chest wall Severe kyphoscoliosis, obesity, pleural effusion Airflow Obstruction (including increased resistive load from narrowing of the airways and increased elastic load from hyperinflation) Asthma, COPD, laryngospasm, aspiration of foreign body, bronchitis *Some diseases appear in more than one category, because they act via several physiologic mechanisms. † Heart failure includes both systolic and diastolic dysfunction. Systolic dysfunction may produce dyspnea at rest and with activity. Diastolic dysfunction typically leads to symptoms primarily with exercise. In addition to the mechanisms noted above, systolic heart failure may also produce dyspnea via metaboreceptors, which are postulated to exist in muscles and be stimulated by changes in the metabolic milieu when oxygen delivery does not meet oxygen demand. ‡ These conditions probably produce dyspnea by a combination of increased ventilator drive and primary sensory input from the receptors. COPD = chronic obstructive pulmonary disease.

who have hematemesis aspirate into the lower respiratory tract. When the oropharynx is colonized with Serratia marcescens, a red-pigment–producing aerobic gram-negative rod, the sputum can also be red and be confused with hemoptysis. Hemoptysis can be caused by a wide variety of disorders. Virtually all causes of hemoptysis (Table 83-6) may result in massive hemoptysis, but massive hemoptysis is most frequently caused by infection (e.g., tuberculosis [Chapter 324], bronchiectasis and lung abscess [Chapter 90], and cancer [Chapter 191]). Infections with aspergilloma (Chapter 339) and in patients with cystic fibrosis (Chapter 89) also are associated with massive hemoptysis. Iatrogenic causes of massive hemoptysis include rupture of a pulmonary artery after less than 0.2% of cases of balloon-guided flotation catheterization and tracheal artery fistula as a complication of tracheostomy. In nonmassive hemoptysis, the cause is bronchitis in more than one third of cases (Chapter 96), bronchogenic carcinoma (Chapter 191) in one fifth of cases, tuberculosis (Chapter 324) in 7%, pneumonia (Chapter 97) in 5%, and bronchiectasis in 1% (Chapter 90). Using a systematic diagnostic approach (see later), the cause of hemoptysis can be found in 68 to 98% of cases. The remaining 2 to 32% have idiopathic or central hemoptysis, which occurs most commonly in men between 30 and 50 years of age. Prolonged follow-up of idiopathic hemoptysis almost always fails to reveal the source of bleeding, even though 10% continue to have occasional episodes of hemoptysis.

DIAGNOSIS

The diagnostic evaluation for hemoptysis begins with a detailed medical history and a complete physical examination. Information on the amount of

530

CHAPTER 83  Approach to the Patient with Respiratory Disease  

Evaluation of Patients with Chronic Dyspnea Patient with suspected chronic dyspnea

Conduct detailed history and physical examination. Conduct appropriate level 1 testing as needed to confirm diagnosis. Is the diagnosis evident? Yes

No

Level 1: Complete blood count Metabolic profile Chest radiograph Electrocardiogram Spirometry Pulse Oximetry

Conduct appropriate Level 2 testing

Possible diagnoses: Asthma Chronic obstructive pulmonary disease Heart failure Pleural effusion Anemia Kyphoscoliosis

Is the diagnosis evident?

Yes

Level 2: Echocardiogram Brain natriuretic peptide Pulmonary function testing Arterial blood gas High-resolution computed tomography Holter monitor Radionuclide study Ventilation-perfusion (V/Q) scan Exercise treadmill testing

No

Possible diagnoses: Chronic pulmonary embolism Restrictive lung disease Interstitial lung disease Pericardial disease Heart failure Valvular heart disease Coronary artery disease Cardiac dysrhythmia

Conduct appropriate level 3 testing (specialty consultation for these tests) Is the diagnosis evident?

Yes

Possible diagnoses: Gastroesophageal reflux disease Primary pulmonary hypertension Coronary artery disease Deconditioning

No Consider: Psychogenic dyspnea Specialty consultation

Level 3: Bronchoscopy Esophageal pH probe testing Lung biopsy Cardiac catheterization Cardiopulmonary exercise testing Bronchoscopy Esophageal pH probe testing Lung biopsy

FIGURE 83-3.  Algorithm outlining the approach to chronic dyspnea. (Modified from Karnani NG, Reisfield GM, Wilson GR. Evaluation of chronic dyspnea. Am Fam Phys. 2005;71:1529-1537.)

TABLE 83-6  COMMON CAUSES OF MASSIVE HEMOPTYSIS Cardiovascular Arterial bronchial fistula Heart failure, especially from mitral stenosis Pulmonary arteriovenous fistula Diffuse intrapulmonary hemorrhage Diffuse parenchymal disease Iatrogenic Malposition of chest tube Pulmonary artery rupture following pulmonary arterial catheterization Tracheoarterial fistula Infections Aspergilloma Bronchiectasis Bronchitis Cystic fibrosis Lung abscess Sporotrichosis Tuberculosis Malignancies Bronchogenic carcinoma Leukemia Metastatic cancer Trauma Drugs and toxins Penicillamine Solvents Crack cocaine Trimelletic anhydride Bevacizumab

bleeding should be obtained, as well as details about the frequency, timing, and duration of hemoptysis. For example, repeated episodes of hemoptysis occurring over a period of months to years suggest a bronchial adenoma or bronchiectasis as the cause, whereas small amounts of hemoptysis occurring every day for weeks are more likely to be caused by bronchogenic carcinoma. A travel history can suggest coccidioidomycosis (Chapter 333) and histoplasmosis (Chapter 332) in the United States, paragonimiasis and ascariasis (Chapter 358) in the Far East, and schistosomiasis (Chapter 355) in South America. Orthopnea and paroxysmal nocturnal dyspnea suggest heart failure (Chapter 58), especially from mitral stenosis (Chapter 75). In patients who have occupational exposure to trimellitic anhydride, which occurs when heated metal surfaces are sprayed with a corrosion-resistant epoxy resin, hemoptysis can be part of the postexposure syndrome. In a patient with the triad of upper airway disease, lower airway disease, and renal disease, granulomatosis with polyangiitis (Chapter 270) should be suspected. Pulmonary hemorrhage also may be a presenting manifestation of systemic lupus erythematosus (Chapter 266). Goodpasture syndrome, which typically occurs in young men, is also associated with renal disease (Chapter 121). Diffuse alveolar hemorrhage occurs in 20% of cases during autologous bone marrow transplantation (Chapter 178) and should be suspected in patients who have undergone recent bone marrow transplantation when they present with cough, dyspnea, hypoxemia, and diffuse pulmonary infiltrates. On physical examination, inspection of the skin and mucous membranes may show telangiectasias suggesting hereditary hemorrhagic telangiectasia (Chapter 173) or ecchymoses and petechiae, suggesting a hematologic abnormality (Chapter 172). Pulsations transmitted to a tracheostomy cannula should heighten suspicion of a tracheal artery fistula. Inspection of the thorax should show evidence of recent or old chest trauma, and unilateral

wheeze or crackles may herald localized disease such as a bronchial adenoma or carcinoma. Although pulmonary embolism (Chapter 98) cannot be definitively diagnosed on physical examination, tachypnea, phlebitis, and pleural friction rub suggest this disorder. If crackles are heard on the chest examination, heart failure as well as other diseases causing diffuse pulmonary hemorrhage (see earlier) or idiopathic pulmonary hemosiderosis (Chapter 92) should be considered. Careful cardiovascular examination may help diagnose mitral stenosis (Chapter 75), pulmonary artery fistulas, or pulmonary hypertension (Chapter 68). Routine laboratory studies should include a complete blood count, urinalysis, and coagulation studies. The complete blood count may suggest an infection, hematologic disorder, or chronic blood loss. Urinalysis may reveal hematuria and suggest the presence of a systemic disease (e.g., Wegener granulomatosis, Goodpasture syndrome, systemic lupus erythematosus) associated with renal disease. Coagulation studies may uncover a hematologic disorder that is primarily responsible for hemoptysis or that contributes to excessive bleeding from another disease. The ECG may help suggest the presence of a cardiovascular disorder. Although as many as 30% of patients with hemoptysis have a normal chest radiograph, routine chest radiographs may be diagnostically valuable. Bronchoscopy can localize the bleeding site in up to 93% of patients by fiberoptic bronchoscopy and in up to 86% with rigid bronchoscopy.7 It may establish sites of bleeding different from those suggested by the chest radiograph. The best results are obtained when bronchoscopy is performed during or within 24 hours of active bleeding, and rates of diagnosis fall to approximately 50% by 48 hours after bleeding. When there is no active bleeding, bronchoscopy with bronchoalveolar lavage can be helpful in patients thought to have diffuse intrapulmonary hemorrhage. Typical findings include bright red or blood-tinged lavage fluid from multiple lobes in both lungs or a substantial number of hemosiderin-laden macrophages (i.e., at least 20% of the total number of alveolar macrophages). Depending on the results of the initial evaluation and the likely categories of hemoptysis, additional diagnostic tests can be helpful (Table 83-7). Bronchoscopy may not be needed in patients who have stable chronic bronchitis (Chapter 88) with one episode of blood streaking or who have acute tracheobronchitis (Chapter 88). Bronchoscopy also may not be needed with obvious cardiovascular causes of hemoptysis, such as heart failure and pulmonary embolism.

TABLE 83-7  EXAMPLES OF SPECIAL EVALUATIONS FOR HEMOPTYSIS ACCORDING TO CATEGORY OF DISEASE* TRACHEOBRONCHIAL DISORDERS Expectorated sputum for TB, parasites, fungi, and cytology Bronchoscopy (if not done) High-resolution chest CT scan LOCALIZED PARENCHYMAL DISEASES Expectorated sputum for TB, parasites, fungi, and cytology Chest CT scan Lung biopsy with special stains DIFFUSE PARENCHYMAL DISEASES Expectorated sputum for cytology Blood for BUN, creatinine, ANA, RF, complement, cryoglobulins, ANCA, anti-GBM antibody Lung or kidney biopsy with special stains CARDIOVASCULAR DISORDERS Echocardiogram Arterial blood gas on 21% and 100% oxygen Ventilation-perfusion scans Pulmonary arteriogram Aortogram, contrast-enhanced CT scan HEMATOLOGIC DISORDERS Coagulation studies Bone marrow *This table is not meant to be all inclusive. ANA = antinuclear antibody; ANCA = antineutrophil cytoplasmic antibody; BUN = blood urea nitrogen; CT = computed tomography; GBM = glomerular basem*nt membrane; RF = rheumatoid factor; TB = tuberculosis.

TREATMENT  Treatment is targeted toward the cause of hemoptysis. Bronchoscopic approaches (Chapter 101) are increasingly used for endobronchial lesions.

Grade A References A1. Smith SM, Schroeder K, Fahey T. Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database Syst Rev. 2014;11:CD001831. A2. Shaheen NJ, Crockett SD, Bright SD, et al. Randomised clinical trial: high-dose acid suppression for chronic cough—a double-blind, placebo-controlled study. Aliment Pharmacol Ther. 2011;33:225-234. A3. Johnstone KJ, Chang AB, Fong KM, et al. Inhaled corticosteroids for subacute and chronic cough in adults. Cochrane Database Syst Rev. 2013;3:CD009305. A4. Ryan NM, Birring SS, Gibson PG. Gabapentin for refractory chronic cough: a randomised, doubleblind, placebo-controlled trial. Lancet. 2012;380:1583-1589. A5. Chamberlain S, Birring SS, Garrod R. Nonpharmacological interventions for refractory chronic cough patients: systematic review. Lung. 2014;192:75-85. A6. Janssens T, Silva M, Davenport PW, et al. Attentional modulation of reflex cough. Chest. 2014;146:135-141. A7. Mastronarde JG, Anthonisen NR, Castro M, et al. Efficacy of esomeprazole for treatment of poorly controlled asthma. N Engl J Med. 2009;360:1487-1499. A8. Lam LL, Cameron PA, Schneider HG, et al. Meta-analysis: effect of B-type natriuretic peptide testing on clinical outcomes in patients with acute dyspnea in the emergency setting. Ann Intern Med. 2010;153:728-735.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 83  Approach to the Patient with Respiratory Disease  

GENERAL REFERENCES 1. Bohadana A, Izbicki G, Kraman SS. Fundamentals of lung auscultation. N Engl J Med. 2014;370:744-751. 2. Goldsobel AB, Kelkar PS. The adult with chronic cough. J Allergy Clin Immunol. 2012;130:825825.e6. 3. Dicpinigaitis PV, Morice AH, Birring SS, et al. Antitussive drugs: past, present, and future. Pharmacol Rev. 2014;66:468-512.

531.e1

4. Busse WW. What is the best pulmonary diagnostic approach for wheezing patients with normal spirometry? Respir Care. 2012;57:39-46. 5. Parshall MB, Schwartzstein RM, Adams L, et al. An official American Thoracic Society statement: update on the mechanisms, assessment, and management of dyspnea. Am J Respir Crit Care Med. 2012;185:435-452. 6. Hurt K, Bilton D. Haemoptysis: diagnosis and treatment. Acute Med. 2012;11:39-45. 7. Sakr L, Dutau H. Massive hemoptysis: an update on the role of bronchoscopy in diagnosis and management. Respiration. 2010;80:38-58.

531.e2

CHAPTER 83  Approach to the Patient with Respiratory Disease  

REVIEW QUESTIONS 1. Which of the following is true about cough variant asthma? A. The diagnosis requires an improvement in the cough with traditional asthma medications. B. It does not reverse with a bronchodilator. C. Sputum eosinophils are absent. D. The cough is not responsive to inhaled corticosteroids. E. Hemoptysis has been present on two occasions. Answer: A  Cough variant asthma responds to asthma medications, in contrast to other causes of cough, which may be either partially or not affected by traditional asthma medications. 2. A 53-year-old woman returns for reevaluation of cough of 1 year duration. The cough is occasionally productive, occurs during the day and night, and is triggered by talking, laughing, and cold air. Her pulmonary function tests are normal. The cough has not responded to appropriate therapy for asthma with inhaled corticosteroids and short-acting β-agonists. She denies any history of lung disease, cigarette smoking, atopy, rhinitis, gastroesophageal reflux disease, or obstructive sleep apnea. The physical examination is unrevealing. Further evaluation, including esophageal impedance and manometry, were unremarkable. What is the appropriate next step in this patient’s management? A. High-resolution computed tomography of the chest B. Computed tomography of the sinuses C. Bronchoscopy D. Echocardiogram E. Video laryngeal swallowing study Answer: A  With a cough that is productive, even occasionally, an assessment of airway structure via computed tomography should be performed first before bronchoscopy to assess for bronchiectasis, sarcoidosis, or occult interstitial lung disease. The patient has no history of atopy or rhinitis, so sinus computed tomography is not indicated. A cough due to cardiac causes is less likely to be productive, so an echocardiogram is not indicated. The likely diagnosis is mild bronchiectasis causing cough. 3. A 55-year-old man presents with dyspnea on exertion beginning approximately 2 months ago. He denies chest pain, cough, wheezing, chest tightness, or a history of cardiovascular disease, lung disease, diabetes, or cigarette smoking. He takes medication daily for hypertension and gastroesophageal reflux disease. He does not exercise regularly. His BMI is 36 kg/m2. On physical examination his blood pressure is 135/70, and his other vital signs and examination are normal. Pulmonary function tests, chest radiograph, complete blood count, and brain natiuretic peptide levels are normal. What is the next appropriate step in this patient’s evaluation? A. Echocardiogram B. Pulse oximetry with ambulation (6-minute walk test) C. Exercise treadmill test D. High-resolution chest computed tomography E. 24-hour Holter monitoring Answer: C  This patient likely has coronary artery disease manifesting as dyspnea on exertion. His risk factors include hypertension, obesity, and lack of exercise. With a normal chest radiograph and pulmonary function tests, primary lung diseases are less likely. A normal brain natiuretic peptide level makes reduced cardiac function less likely, and anemia is excluded by a normal complete blood count. Cardiovascular disease is the most likely diagnosis, and additional testing likely will be required if the exercise treadmill test is unrevealing.

4. A 40-year-old woman with a history of atopy, rhinitis, and gastroesophageal reflux disease (GERD) presents with wheezing of 3 months duration. The wheezing occurs primarily during the day, is triggered by cold air and exercise, and has not responded to a 6-week course of an inhaled corticosteroid and short-acting β-agonist. She denies chest tightness, chest pain, or a history of lung or cardiovascular disease. Her history is notable for allergies to dust, multiple weeds, grasses, and animal dander. She has persistent rhinitis for which she performs nasal saline rinses and uses intranasal corticosteroids daily. She also takes a proton pump inhibitor once daily for treatment of GERD, and these symptoms are well controlled. Her physical examination reveals normal vital signs and erythema of the upper airway without nasal polyps. The remainder of her physical examination is normal without wheezing. Pulmonary function tests, chest radiograph, brain natriuretic peptide, and methacholine challenge testing are normal. What is the next step in this patient’s evaluation? A. Sinus computed tomography B. High-resolution chest computed tomography C. Direct laryngoscopy after exercise D. Echocardiogram E. Bronchoscopy with biopsy Answer: C  This presentation suggests vocal cord dysfunction presenting as wheezing. Wheezing in the setting of rhinitis and GERD suggests an airway process as the most likely diagnosis. Because the patient has not responded to therapy for asthma and a methacholine challenge is negative, asthma is highly unlikely. A normal BNP makes cardiac dysfunction less likely. In the setting of rhinitis and GERD, evaluation of the upper airway to exclude vocal cord dysfunction during an episode of wheezing is the next appropriate step. An echocardiogram should be obtained if direct laryngoscopy is negative.

531

CHAPTER 84  Imaging in Pulmonary Disease  

84  IMAGING IN PULMONARY DISEASE PAUL STARK

  IMAGING OF THE LUNGS, MEDIASTINUM, AND CHEST WALL EPIDEMIOLOGY

Worldwide, chest radiography is the most commonly performed imaging procedure; more than 75 million chest radiographs are performed every year in the United States alone. Chest radiographs provide useful information about the patient’s anatomy and disease at a minimal monetary cost and with radiation exposure that most experts agree is negligible (0.05 to 0.1 mSv) (Chapter 20). Although many novel imaging techniques are available, the conventional chest radiograph remains invaluable in the initial assessment of disorders of the lung, pleura, mediastinum, and chest wall.

Imaging Techniques

The standard chest radiograph is performed at 2 m from the x-ray tube focal spot to the image detector, in frontal and lateral projections. If possible, the radiographs should be obtained with the patient inhaling to total lung capacity. These images, which provide views of the lungs, mediastinum, and chest wall simultaneously, are typically acquired, stored, and distributed digitally.

Bedside Radiography

Although bedside radiography accounts for a large number of chest radiographs, especially in the intensive care unit (ICU), the images obtained are generally of lower technical quality, cost more, and are more difficult to interpret. Lung volumes are low, thereby leading to crowding of vascular structures, and the low kilovoltage technique required for the mobile equipment yields radiographs with overexposed lungs and an underpenetrated mediastinum. The anteroposterior projection and the slightly lordotic angulation of the x-ray beam combine to distort the basal lung structures and magnify

532

CHAPTER 84  Imaging in Pulmonary Disease  

the cardiac silhouette. Recumbent studies also make recognition of pleural effusions or pneumothoraces more difficult. In the ICU, chest radiography can be ordered selectively rather than as a daily routine, without compromising care.1

Computed Tomography

Computed tomography (CT) has multiple advantages over conventional radiography. It displays cross-sectional anatomy free of superimposition, with a 10-fold higher contrast resolution. Multislice CT scanners acquire a continuous, volumetric, near-isotropic data set with possibilities for high-quality two-dimensional or three-dimensional reformatting (volume rendering) in any plane. High-resolution CT of the lung parenchyma is an important application; narrow collimation of the beam combined with an edge-enhancing high spatial frequency algorithm results in exquisite detail of normal and abnormal lungs, and correlation with pathologic anatomy is high.

Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) depends on the magnetic properties of hydrogen atoms. Magnetic coils and radio frequency coils lead to induction, excitation, and eventual readout of magnetized protons. The molecular environment of hydrogen atoms will affect the rate at which they release energy; this energy yields a spatial distribution of signals that is converted into an image by computer algorithms, similar to CT. Because of its soft tissue specificity, MRI has applications in the assessment of chest wall invasion, mediastinal infiltration, and diaphragmatic involvement by lung cancer or malignant mesothelioma.

Positron Emission Tomography

Fluorodeoxyglucose positron emission tomography (FDG-PET) uses labeled fluorodeoxyglucose to image the glycolytic pathway of tumor cells or other metabolically active tissues with affinity for glucose. This technique has proved helpful in studying intrathoracic tumors and has facilitated the work-up of solitary pulmonary nodules. Integrated PET-CT scans have improved the diagnosis and staging of intrathoracic tumors.2

Ultrasonography

Outside the heart, ultrasonography plays only a limited role in thoracic imaging. Its primary use is to localize pleural effusions and guide their drainage (Chapter 99). In the intensive care setting, ultrasound also may help with the diagnosis of pneumothorax and diffuse alveolar damage.

Evaluation of Chest Images

Images of the chest are best evaluated by examining regions of the lung for specific findings and relating these findings to known diagnostic groups. A number of critical radiographic features should be considered, with an appreciation for the known causes of these changes.

Diffuse Lung Disease

Diffuse lung disease is an overall term for a number of related abnormal parenchymal radiographic patterns. Although radiologists have attempted to separate alveolar from interstitial lung disease radiographically, this distinction is no longer recommended because the correlation between the radiographic localization to a compartment and the actual histopathologic findings is relatively poor.3 For example, nodular patterns can be produced by either interstitial or alveolar disease. Conversely, so-called alveolar disease processes can induce an interstitial reaction. Ground-glass opacities can be induced by either alveolar or interstitial disease. Air bronchograms, the presumed paradigm of air space disease, can be identified in a small percentage of patients with predominantly interstitial lung disease, such as sarcoidosis, pulmonary lymphoma, and pulmonary calcinosis. Because of such limitations, a graphically descriptive approach that combines analysis of predominant opacities, assessment of lung expansion, and distribution and profusion of disease yields a differential diagnosis. The term infiltrate should be avoided; instead, the term pulmonary opacities should be used, with opacities further classified as large (i.e., >1 cm in largest dimension) or small (i.e., 10% difference). The usual cause is the presence of more than one restrictive process, such as a parenchymal restrictive disorder plus obesity, respiratory muscle weakness, or heart failure. Some patients have a mixed disorder with evidence of both obstruction and restriction. Common causes include cystic fibrosis (Chapter 89), sarcoidosis (Chapter 95), and heart failure (Chapters 58 and 59) as well as cases in which the cause of the obstructive disorder and the restrictive disorder are unrelated. Disorders of the central airways can cause characteristic patterns of abnormality. In a “fixed airway obstruction” such as tracheal stenosis (Fig. 85-1H), flow is typically reduced on both inspiration and expiration. In contrast, in a variable extrathoracic upper airway obstruction (Fig. 85-1F), inspiration is disproportionately reduced; however, expiration is often abnormal, merely less so. Likewise, in variable intrathoracic obstruction (e.g., relapsing polychondritis, tracheomalacia, or a dynamic intrathoracic tracheal tumor), the expiratory flow-volume curve is reduced but in a pattern unlike that seen in asthma or COPD (Fig. 85-1G). These central airway obstructive patterns may signify a locally treatable cause of obstruction.

  PROVOCATIVE TESTING

Assessing Airway Responsiveness

Hyperresponsiveness of airways to the smooth muscle–contracting effect of pharmacologic agents such as methacholine, as well as to cold air, dry air, and other physical stimuli, is characteristic of asthma (Chapter 87). It is also observed in COPD and other obstructive airway diseases. Bronchoprovocation studies, in which graded doses of a stimulus are used to elicit airway constriction, are performed to measure airway responsiveness. A responsive airway, that is, one in which a small stimulus leads to a fall in FEV1, may be used to confirm the diagnosis of asthma (Chapter 87). Exhaled nitric oxide is a marker of eosinophilic airway inflammation and can be used to predict the likelihood that airway obstruction will improve with corticosteroid treatment. However, the utility of exhaled nitric oxide levels for asthma management is controversial.

  CARDIOPULMONARY EXERCISE TESTS

Some patients have dyspnea (Chapter 83) or exercise limitation that is not adequately explained by the clinical examination, standard pulmonary

function testing, and chest imaging. For such patients, laboratory testing of physiologic performance during exercise can be enlightening. Cardiopulmonary exercise testing, which is usually performed on a cycle ergometer or treadmill, includes monitoring of the heart rate, electrocardiography, and pulse oximetry as well as breath-by-breath measurement of tidal volume, breathing rate, oxygen consumption, and carbon dioxide production. Optional measurements include arterial blood gases and noninvasive cardiac output. Outcomes include maximal oxygen uptake ( V O2 max ), maximal workload, maximal heart rate, ventilation parameters during exercise, and measurements of gas exchange. Results are analyzed to determine if anaerobic metabolism occurs when the study subject reaches maximal effort and to determine what limits the ability of a patient to exercise—a gas exchange abnormality, ventilatory limitation, cardiac limitation, or deconditioning. Simple tests of exercise performance, such as the 6-minute walk test, can quantify and serially assess exercise performance.8

  BRONCHOALVEOLAR LAVAGE

Bronchoalveolar lavage can be useful for evaluation of opportunistic infections in immunocompromised hosts (Chapter 281), but its utility in the evaluation of interstitial lung disease is more controversial.9 The procedure is generally safe, although provision must be made for the transient deterioration in gas exchange after the procedure. Oxygen supplementation is usually necessary, and intubation and mechanical ventilation are sometimes needed. A normal bronchoalveolar lavage specimen includes 85% macrophages or more, 10 to 15% lymphocytes, 3% neutrophils or less, 1% eosinophils or less, 1% mast cells or less, and less than 5% squamous epithelial cells (which are an indicator of contamination from the upper airway). Smokers may have higher cell counts and a higher percentage of neutrophils. Increased lymphocyte counts are seen in sarcoidosis (Chapter 95), hypersensitivity pneumonitis (Chapter 94), nonspecific interstitial pneumonitis (Chapter 92), collagen vascular diseases (Chapter 92), radiation pneumonitis (Chapter 94), cryptogenic organizing pneumonia (Chapter 92), and lymphoproliferative disorders. Increased neutrophil counts are seen in idiopathic pulmonary fibrosis (Chapter 92), collagen vascular diseases (Chapter 92), infectious pneumonia (Chapter 97), aspiration pneumonia (Chapter 97), acute respiratory distress syndrome (Chapter 104), diffuse alveolar damage (Chapter 91), acute interstitial pneumonia (Chapter 92), and asbestosis (Chapter 93). Increased eosinophils can be seen in asthma (Chapter 87), bronchitis (Chapter 96), allergic bronchopulmonary aspergillosis (Chapter 339), Churg-Strauss vasculitis (Chapter 270), Hodgkin lymphoma (Chapter 186), and drug-induced lung disease (Chapter 94). If eosinophils are more than 25%, eosinophilic pneumonia is likely (Chapter 170). If lymphocytes are increased and the clinical differential diagnosis includes sarcoidosis or hypersensitivity pneumonitis, analysis of T-cell populations may be helpful; the CD4:CD8 ratio is typically increased in sarcoidosis but reduced in hypersensitivity pneumonitis. If more than 20% of macrophages stain positive for hemosiderin, diffuse alveolar hemorrhage is considered likely (Chapter 91), particularly if lavage fluid is progressively bloody in successive aliquots of lavage fluid. Cellular constituents of bronchoalveolar lavage are usually stained for cytologic analysis for malignant cells and viral inclusions. If Langerhans cell histiocytosis (Chapter 92) is considered possible, 5% or more CD1a–positive cells supports the diagnosis. If chronic beryllium disease or beryllium sensitization is possible, a lymphocyte proliferation test in response to exposure to beryllium salts can be helpful (Chapter 93). Staining of solid material from the bronchoalveolar lavage with periodic acid–Schiff (PAS) stain for the presence of PAS-positive material is essential to the diagnosis of pulmonary alveolar proteinosis (Chapter 91). A diagnosis of lipoid pneumonia (Chapter 94), caused by the aspiration of oil, can be confirmed by an excess of lipidladen macrophages from bronchoalveolar lavage. The presence of asbestos bodies or silica is not diagnostic of lung disease related to these substances (Chapter 93) but does indicate significant exposure.

  PULMONARY FUNCTION IN OBESITY

The epidemic of obesity is manifested in many organ systems. Dyspnea, exercise limitation, and respiratory failure are more common in obese persons than in the nonobese. Asthma is more common and more severe in obese patients.10 The effects of obesity on lung function are usually relatively modest among ambulatory patients with a body mass index (BMI) of less than 40. The most commonly observed effect of obesity on lung function is a reduction in expiratory reserve volume (the amount of air exhaled between FRC and residual volume), which is substantially reduced even in persons who are

overweight (BMI 25-30) or mildly obese (BMI 30-35). Vital capacity is reduced in obesity, but the effect is modest and highly variable. In large studies, on average, for each unit increase in BMI above 25, vital capacity or FVC is reduced by 0.5 to 0.8%. Effects of obesity on total lung capacity and FEV1 are somewhat smaller. The FEV1/FVC ratio and Dlco actually increase slightly with increase in BMI. In exercise studies, the effects of obesity among ambulatory outpatients are likewise modest. Such patients have an increased work of breathing, but maximal oxygen uptake is often normal.

  FUTURE DIRECTIONS

New test methods are likely to evolve, such as using “electronic nose” devices to identify volatile compounds in exhaled gases. Such efforts are encouraged by reports of dogs that can be trained to identify persons with malignant neoplasms and other conditions. For example, cancer-specific volatile carbonyl aldehydes and ketones can be identified in the exhaled breath condensate of patients with lung cancer, and these concentrations may return to normal after surgery. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 85  Respiratory Function: Mechanisms and Testing  

GENERAL REFERENCES 1. Burney PG, Hooper R. Forced vital capacity, airway obstruction and survival in a general population sample from the USA. Thorax. 2011;66:49-54. 2. Garcia-Rio F, Calle M, Burgos F, et al. Spirometry. Arch Bronconeumol. 2013;49:388-401. 3. Qaseem A, Wilt TJ, Weinberger SE, et al. Diagnosis and management of stable chronic obstructive pulmonary disease: a clinical practice guideline update from the American College of Physicians, American College of Chest Physicians, American Thoracic Society, and European Respiratory Society. Ann Intern Med. 2011;155:179-191. 4. Redlich CA, Tarlo SM, Hankinson JL, et al. Official American Thoracic Society Technical Standards: spirometry in the occupational setting. Am J Respir Crit Care Med. 2014;189:983-993. 5. Iyer VN, Schroeder DR, Parker KO, et al. The nonspecific pulmonary function test: longitudinal follow-up and outcomes. Chest. 2011;139:878-886.

545.e1

6. Ford ES, Mannino DM, Wheaton AG, et al. Trends in the prevalence of obstructive and restrictive lung function among adults in the United States: findings from the National Health and Nutrition Examination surveys from 1988-1994 to 2007-2010. Chest. 2013;143:1395-1406. 7. Schmidt SL, Tayob N, Han MK, et al. Predicting pulmonary fibrosis disease course from past trends in pulmonary function. Chest. 2014;145:579-585. 8. Casanova C, Celli BR, Barria P, et al. The 6-min walk distance in healthy subjects: reference standards from seven countries. Eur Respir J. 2011;37:150-156. 9. Meyer KC, Raghu G, Baughman RP, et al. An official American Thoracic Society clinical practice guideline: the clinical utility of bronchoalveolar lavage cellular analysis in interstitial lung disease. Am J Respir Crit Care Med. 2012;185:1004-1014. 10. Gibeon D, Batuwita K, Osmond M, et al. Obesity-associated severe asthma represents a distinct clinical phenotype: analysis of the British Thoracic Society Difficult Asthma Registry Patient cohort according to BMI. Chest. 2013;143:406-414.

545.e2

CHAPTER 85  Respiratory Function: Mechanisms and Testing  

REVIEW QUESTIONS 1. A 58-year-old man with exercise-related cough and a body mass index of 42 has a hemoglobin level of 14 g/dL and a diffusing capacity (Dlco) that is 144% of the reference value. The most appropriate next test is A. Bronchoalveolar lavage for hemosiderin-laden macrophages B. Echocardiography to exclude intracardiac shunt C. Quantitative assay for JAK2 mutation D. Methacholine challenge E. Measurement of hemoglobin P50 Answer: D  Obesity and asthma are the most likely causes of an increased Dlco, and a search for rare causes of an increased Dlco usually is not indicated. Bronchoalveolar lavage can be useful if clinical information suggests pulmonary hemorrhage. An echocardiogram may demonstrate a left to right shunt, which is a rare cause of an increased Dlco. JAK2 mutations are associated with polycythemia vera, but the patient is not polycythemic. 2. A 61-year-old male former smoker (40 pack-years) complains of dyspnea and cough. Pulmonary function testing shows normal spirometry and lung volumes; there is an isolated reduction in diffusing capacity (Dlco). The most useful next test is A. Echocardiography B. Right-sided heart catheterization C. High-resolution computed tomography of the chest D. Maximal respiratory pressures E. Bronchoalveolar lavage for hemosiderin-laden macrophages Answer: C  An isolated reduction in Dlco, which is most often associated with emphysema or fibrosis or both, is seen best on computed tomography. An isolated reduction in Dlco is less often due to pulmonary vascular disorders such as pulmonary hypertension, so echocardiography and right-sided heart catheterization would have lower yields. Muscle weakness can reduce the Dlco, but it also reduces lung volumes. More than 20% hemosiderin-laden macrophages on bronchoalveolar lavage is suggestive of diffuse alveolar hemorrhage, which is a rare cause of an increased Dlco. 3. A 53-year-old never-smoker with a saddle nose deformity has severe dyspnea and dry cough. His pulmonary function test results are as follows: 10 Control Bronchodilator Bronchodilator Insp

8 6 4 FVC FEV1 FEV1/FVC FEF max FIFmax MVV

4.51 1.52 33.7 2.7 4.5 55

93% 40% 31% 36%

2 0 1

2

3

4

5

−2 −4 −6

3604414

He reports episodes of ear pain and erythema, refractory to antibiotics but responsive to steroids. What is the next most appropriate test? A. Methacholine challenge B. Maximal respiratory pressures C. Airway resistance D. Imaging of the central airways (computed tomography or bronchoscopy) E. Measurement of exhaled nitric oxide Answer: D  He has relapsing polychondritis. His main respiratory issue is central airway collapse due to chondromalacia of the tracheal and bronchial cartilage. The flow-volume curve shows characteristic flattening, as opposed to the “scooped out” pattern of asthma and COPD. He does not have a disorder of airway reactivity, so methacholine challenge adds little useful information and may not be safe with this degree of obstruction. Maximal respiratory pressures are not likely to be abnormal. Airway resistance will be abnormal but will add nothing diagnostically. Exhaled nitric oxide is abnormal in patients with eosinophilic airway inflammation and would not be expected to be abnormal in this case.

545.e3

CHAPTER 85  Respiratory Function: Mechanisms and Testing  

4. A patient with mild obstruction on spirometry has a maximal voluntary ventilation that is reduced out of proportion to the FEV1. Which of the following is least likely to be helpful? A. Maximal respiratory pressures B. Inspiratory flow-volume curve C. Cardiopulmonary exercise challenge D. Airway resistance measurement E. Careful scrutiny of test for repeatability of measures and technician comments on patient performance Answer: C  A disproportionate reduction in maximal voluntary ventilation may be due to inspiratory obstruction, muscle weakness, or poor performance. Cardiopulmonary exercise testing is likely to be abnormal regardless of the cause of the abnormality. The other four options would yield more specific diagnostic information. 5. A 43-year-old woman is being evaluated for dyspnea and lack of energy. Results are as follows: TLC, 79% predicted; FVC, 46%; FEV1, 54%; FEV1/FVC, 0.99; Dlco, 81%. The expiratory flow-volume curve is as shown: 10 Control

TLC RV RV/TLC FVC FEV1 FEV1/FVC FEF max DLCO SPO2

3.56 1.92 53.9 1.47 1.45 98.6% 3.7 18 94%

79% 145% 183% 46% 54% 64% 81% 91%

8 6 4 2 0 0

1

2

3

4

5 4396224

What test is likely to be most helpful? A. Maximal respiratory pressures B. Airway resistance C. Methacholine challenge D. Cardiopulmonary exercise test E. Arterial blood gases Answer: A  The convex shape of the flow-volume curve in an adult suggests muscle weakness or poor performance. In a patient with restriction, the disproportionate reduction in FVC compared with TLC suggests muscle weakness. The patient probably has a myopathy. Alternative possibilities include chest wall limitation, poor performance, and occult airflow obstruction. The most helpful measurements on this patient will be maximal respiratory pressures, which will likely result in referral to a neurologist. Airway resistance is unlikely to be abnormal. There is little to suggest asthma, and an exercise study is likely to be abnormal but may not reveal the cause of the abnormality. Arterial blood gases are usually normal in neuromuscular disorders until the FEV1 and FVC are severely reduced, after which hypercapnia develops as a manifestation of end-stage respiratory failure.

545.e4

CHAPTER 85  Respiratory Function: Mechanisms and Testing  

6. Case FX-1 50 yo M Ht = 178 cm Wt = 79 kg BMI = 25 Dx: Tracheal stenosis, never-smoker CONTROL

%PRED

TLC

5.73

83

RV

2.32

124

RV/TLC

0.41

148

FVC

3.40

68

FEV1

1.51

38

FEV1/FVC

44.5

MVV

11

FEF50/FIF50

56 7

1.1

110

Dlco (hb adj)

18.58

61

Spo2

98

%PRED = percentage of predicted value.

Exp flow (L/sec)

6

4 2 0

Insp flow

1

2

3

4

Expired volume (L) 5 inspired

2 4

6

Interpretation: Abnormal.  Severe fixed airway obstruction is indicated by the reduced FEV1 and MVV and shape of the inspiratory and expiratory flowvolume curves. There is no immediate response to bronchodilator. Dlco is mildly reduced, consistent with a pulmonary parenchymal or vascular process. Lung volumes and oxygen saturations are normal.

545

CHAPTER 86  Disorders of Ventilatory Control  

kyphoscoliosis (Chapter 99). The epidemiology of these hypoventilation syndromes is poorly studied, but about 15% of patients with severe COPD or morbid obesity have an elevated Paco2. Regardless of the cause, patients with hypoventilation frequently have further worsening of their ventilation at the onset of sleep due to loss of the wakefulness stimulus, which is the normal drive to breathe while awake, and some degree of upper airway collapse after the onset of sleep (Chapter 100). Patients with central sleep apnea (Chapter 100), which is a group of conditions in which cessation of airflow occurs because of a lack of respiratory effort, are classified into those with inadequate ventilatory drive and those with excessive drive.1 The apparent paradox of how excessive drive leads to central apnea is explained by the concept of loop gain. A negative feedback control system with a high loop gain is prone to instability that leads to periods of excessive breathing followed by periods of apnea (Table 86-2). The prototype of a condition with high loop gain is periodic breathing or CheyneStokes breathing (Fig. 86-1).

Cheyne-Stokes Breathing Cheyne-Stokes breathing is a waxing and waning pattern of breathing, which is classically described as crescendo-decrescendo and often includes periods of central apnea. Cheyne-Stokes is seen most commonly during sleep in patients with heart failure.

EPIDEMIOLOGY

86  DISORDERS OF VENTILATORY CONTROL ATUL MALHOTRA AND FRANK POWELL

Cheyne-Stokes breathing is a form of ventilatory instability that occurs in 30 to 40% of patients with left ventricular systolic dysfunction.2 Male sex, advanced age, low baseline Paco2, and atrial fibrillation are risk factors for Cheyne-Stokes breathing among patients with heart failure. Controversy remains regarding whether this breathing pattern itself is deleterious or whether it is simply a marker of the underlying severity of cardiac disease. Cheyne-Stokes breathing represents about 5 to 10% of all cases of sleep apnea (Chapter 100) and is uncommon among patients who do not have heart failure.

PATHOBIOLOGY

DEFINITIONS AND PATHOGENESIS

Ventilatory Control

Ventilation is controlled by complex interactions between central chemoreceptors, which predominantly are responsive to carbon dioxide tensions in arterial blood, and peripheral chemoreceptors, which primarily respond to carbon dioxide and oxygen tensions (Table 86-1). Disorders of ventilatory control are caused by derangements in these control systems.

  HYPOVENTILATION SYNDROMES

Hypoventilation syndromes are defined by a lack of adequate alveolar ventilation to maintain a normal arterial carbon dioxide tension of 40 mm Hg. The two most common clinical settings that result in chronic hypoventilation are severe chronic obstructive pulmonary disease (COPD; Chapter 88) and morbid obesity (Chapters 100 and 220); less common causes are chronic opiate therapy, neuromuscular weakness (Chapters 421 and 422), and severe

Individuals with Cheyne-Stokes breathing have robust chemosensitivity as evidenced by marked increases in ventilation with small increases in Paco2. The drive to breathe may be further increased by neural reflexes that are triggered by extravascular lung fluid and an elevated left atrial pressure. Intermittent hypoxemia and catecholamine surges, which are frequent in these patients, contribute to oxidative stress and neuroendocrine activation, both of which are thought to contribute to worsening of the underlying heart failure.

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Patients with Cheyne-Stokes breathing can sometimes be diagnosed at the bedside by careful observation of their breathing pattern. During sleep or exercise, breathing becomes dependent primarily on metabolic stimuli. Patients may complain of fatigue or sleepiness because arousals from sleep tend to occur during the hyperpneic phase. Paroxysmal nocturnal dyspnea, a classic symptom of heart failure (Chapter 58), most commonly reflects

TABLE 86-1  CLASSIFICATION OF CENTRAL SLEEP APNEA CENTRAL SLEEP APNEA SYNDROME

MECHANISM

THERAPY

Sleep transition apneas

Carbon dioxide fluctuations during transitions from sleep to wake to sleep

Reassurance, occasionally hypnotics or oxygen

Chronic narcotic therapy

Lack of central drive

Reduce narcotic dose Consider positive-pressure device

Cheyne-Stokes breathing

High loop gain from robust chemosensitivity and ventilatory drive

Optimize medical therapy for heart failure, consider PAP devices

Idiopathic central apnea

Unknown

Supportive, bilevel PAP; consider ventilatory stimulants

Treatment of emergent central apnea or “complex apnea”

Lowering upper airway resistance at CPAP initiation improves efficiency of carbon dioxide excretion

Reassurance, generally resolves spontaneously

Sleep hypoventilation syndromes

Fall in drive with loss of wakefulness stimulus, loss of accessory muscle activity during REM sleep

Noninvasive ventilation

CPAP = continuous positive airway pressure; PAP = positive airway pressure; REM = rapid eye movement.

546

CHAPTER 86  Disorders of Ventilatory Control  

TABLE 86-2  CLASSIFICATION OF HYPERCAPNIC DISEASES HYPERCAPNIC DISEASE

MECHANISM

DIAGNOSIS

TREATMENT

Narcotic overdose

Reduced central drive

History, narcotized pupils, toxicology

Supportive care, naloxone

Acute severe asthma

Severe airflow obstruction, high dead space

Typical history, wheezing on examination, low FEV1/FVC

Bronchodilators, anti-inflammatories, mechanical ventilation (usually invasive)

Acute exacerbation of COPD

Airflow obstruction, high dead space

History, cigarette smoking, low FEV1/ FVC, infectious etiology

Bronchodilators, anti-inflammatories, noninvasive ventilation

Obesity-hypoventilation syndrome

Low respiratory system compliance, high upper airway resistance, low central drive

High BMI, lack of other diagnoses; blunted carbon dioxide response

Weight loss, nocturnal bilevel positive airway pressure

Central congenital hypoventilation syndrome

PHOX2B mutation, lack of central drive

Genetic testing

Supportive care, mechanical ventilation (usually noninvasive)

Neuromuscular disease (e.g., myasthenia gravis, ALS, polymyositis, GBS/AIDP)

Lack of respiratory muscle force

Immediate orthopnea, low VC, low MIPs/ MEPs

Underlying cause; nocturnal noninvasive ventilation; supportive care

Severe parenchymal lung disease, e.g., COPD

Lack of alveolar surface area; high pulmonary dead space and work of breathing

Typical history, smoking, low FEV1 and FEV1/FVC

Bronchodilator, anti-inflammatory therapy, possible nocturnal noninvasive ventilation, smoking cessation

Kyphoscoliosis

Low respiratory system compliance

Physical examination

Supportive care, noninvasive ventilation

AIDP = acute inflammatory demyelinating polyneuropathy; ALS = amyotrophic lateral sclerosis; BMI = body mass index; COPD = chronic obstructive pulmonary disease; FEV1 = forced expiratory volume in 1 second; FVC = forced vital capacity; GBS = Guillain-Barré syndrome; MEPs = maximal expiratory pressures; MIPs = maximal inspiratory pressures; VC = vital capacity.

Cheyne-Stokes Respiration

FIGURE 86-1.  Cheyne-Stokes breathing with crescendo-decrescendo pattern of breathing. The thermistor detects air temperature changes at the mouth and nose. Note absences in airflow without respiratory effort seen in the abdominal belts. This breathing pattern leads to intermittent desaturations, arousals from sleep, and bursts of tachycardia. The loop gain concept can be understood by considering the thermostat analogy in which a control system is working to regulate a stable room temperature (e.g., 20° C). By analogy, the respiratory control system is working primarily to maintain a stable PaCO2 of 40 mm Hg and stable pH. Situations in which marked fluctuations in room temperature might occur would include one in which the thermostat is excessively sensitive (i.e., furnace turns on if room temperature falls to 19.999° C); if the furnace is too powerful, a marked overshoot in room temperature will be followed by a prolonged period when the furnace does not run. In the analogy to Cheyne-Stokes breathing, carbon dioxide is equated to room temperature and would be predicted to be unstable if chemosensitivity (i.e., the thermostat) were excessively robust (i.e., a marked increase in ventilation for a small change in carbon dioxide) or if the efficiency of carbon dioxide excretion were high (i.e., marked fall in PaCO2 with increased ventilation). Situations that increase the propensity for carbon dioxide fluctuations lead to elevated loop gain and thus increase the risk for Cheyne-Stokes breathing.

underlying Cheyne-Stokes breathing. Patients often are diagnosed in the sleep laboratory while undergoing investigation for possible obstructive sleep apnea. The diagnosis of Cheyne-Stokes breathing, if it is not readily apparent, can be made during overnight polysomnography, when the typical oscillatory pattern of tidal volume is seen in the absence of ventilatory efforts during the

apneic periods. In evaluating such recordings, and in contrast to obstructive apnea, it is important to note that Cheyne-Stokes breathing usually resolves during rapid eye movement (REM) sleep, that arousals on the electroencephalogram typically occur during the hyperpneic phase, and that CheyneStokes breathing generally does not resolve immediately when nasal continuous positive airway pressure (CPAP) is applied.

CHAPTER 86  Disorders of Ventilatory Control  

TREATMENT  Medical management of Cheyne-Stokes breathing most often is treatment of the underlying heart failure (Chapter 59). After optimization of medical management, the Cheyne-Stokes breathing pattern frequently resolves. CPAP can improve breathing indices but is no better than standard medical therapy from the standpoint of mortality. A1  Newer approaches to non-invasive ventilation show promise but require further evaluation.3

547

mechanical or neuromuscular abnormalities have a larger work of breathing compared with normal individuals; the most common underlying conditions are severe COPD (Chapter 88) and morbid obesity (Chapter 220) with the obesity-hypoventilation syndrome. Such individuals have diminished but not absent chemoresponsiveness. Another cause of inadequate gas exchange is neuromuscular disease; common causes include disorders of neuromuscular transmission (Chapter 422), severe muscle weakness (Chapter 421), the residua from poliovirus infection (Chapter 379), Guillain-Barré syndrome (Chapter 420), and acute poisoning (Chapter 110).

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Central Congenital Hypoventilation Syndrome

DEFINITION AND EPIDEMIOLOGY

Central congenital hypoventilation syndrome is a rare congenital condition, previously referred to as Ondine curse, characterized by a diminished ventilatory response to carbon dioxide.4 The central congenital hypoventilation syndrome was traditionally diagnosed in neonates, but more subtle forms of disease are increasingly noted in older children and adults.

PATHOBIOLOGY

The syndrome is now defined by a mutation in the PHOX2B gene, located on chromosome 4p12.5 The PHOX2B gene is a highly conserved homeobox gene that is expressed mainly in the afferent and efferent pathways of respiratory, cardiovascular, and digestive reflexes. Deletion of the gene in mice causes irregular breathing, a reduced hypercapnic ventilatory response, and death from central apnea. These mice have neuronal loss in the retrotrapezoid nucleus and parafacial region of the brain stem, thereby suggesting the importance of this medullary region in normal breathing. Abnormalities in PHOX2B genes have also been associated with Hirschsprung disease (Chapter 136), neural crest tumors, cardiac asystole (Chapter 63), and other abnormalities of the autonomic nervous system (Chapter 418). Because most parents of affected children with the central congenital hypoventilation syndrome do not carry a PHOX2B mutation, the mutations are de novo. About 90% of patients are heterozygous for a polyalanine repeat expansion mutation, in which the affected allele has 24 to 33 alanines rather than the normal 20 alanines. The remaining 10% of central congenital hypoventilation syndrome patients have missense, nonsense, or frameshift mutations in the PHOX2B gene.

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Neonates can present with cyanosis at birth, recurrent central apneas, or both. Adults can present with idiopathic central sleep apnea, unexplained hypercapnia, or autonomic abnormalities (Chapter 418). Confirmation of the diagnosis requires the demonstration of an abnormality in the PHOX2B gene.

TREATMENT  There are currently no specific therapies for central congenital hypoventilation syndrome beyond supportive care. Genetic counseling is required for afflicted individuals and their families, given the autosomal dominant pattern of inheritance. Patients must be cautioned against the use of sedatives, which could precipitate respiratory failure. Mechanical ventilation during sleep either invasively (through tracheostomy) or noninvasively (through bilevel positive airway pressure support [Chapter 100]) is required in most patients. Some patients remain fully ventilator dependent. Alternative treatments, such as ventilatory stimulants and diaphragmatic pacing, are generally ineffective.

Acquired Hypoventilation Syndromes

DEFINITION AND EPIDEMIOLOGY

Patients with hypoventilation syndromes cannot maintain adequate minute ventilation to keep their Paco2 at 40 mm Hg. Patients can be classified into those who lack central ventilatory drive and those who have a pulmonary mechanical or neuromuscular abnormality that prevents adequate gas exchange. The case frequency is unknown, but hypercapnic respiratory failure is one of the more common admission diagnoses in intensive care units.

PATHOBIOLOGY

Patients with conditions characterized by the lack of central drive have reasonably normal lungs and respiratory muscle function but lack adequate response to carbon dioxide and hypoxia. In contrast, most patients with

Patients with hypoventilation have myriad presentations ranging from asymptomatic abnormalities in laboratory testing (e.g., elevated Paco2, unexplained low Sao2, or elevated serum bicarbonate level) to respiratory failure in the intensive care unit (e.g., respiratory infection with laboratory evidence of chronic abnormalities, such as acute-on-chronic respiratory acidosis). Patients who acutely overdose on sedative-hypnotic or narcotic agents may present with acute respiratory acidosis and loss of consciousness. Patients who take chronic narcotics may present with central sleep apnea-hypopnea or otherwise unexplained oxygen desaturation at night. Once it is suspected, the diagnosis of hypoventilation is confirmed by the finding of Paco2 higher than 42 mm Hg on analysis of an arterial blood sample. If the increase in Paco2 is of short duration so that renal compensation has not yet occurred (Chapter 118), the serum bicarbonate level is increased by 1 mEq/L for every rise of 10 mm Hg in Paco2. By comparison, if the respiratory acidosis is of sufficient duration for renal compensation to occur, the serum bicarbonate level will be increased by 4 mEq for every rise of 10 mm Hg in Paco2 (Fig. 86-2). Once an elevated Paco2 is established, it is appropriate to distinguish patients who “can’t breathe” from those who “won’t breathe.” “Can’t breathe” implies that a respiratory mechanical problem or neuromuscular weakness is causing the elevation in Paco2. Abnormalities in pulmonary function testing (e.g., a very low vital capacity) suggest a parenchymal or chest wall disorder. Ultrasound can identify phrenic neuropathy causing diaphragmatic dysfunction.6 Patients who “won’t breathe” have central nervous system abnormalities that affect central drive, chemosensitivity, or both.

TREATMENT AND PROGNOSIS  The treatment of hypoventilation should focus on the underlying cause. Acute poisonings can be managed supportively or, in some cases, with specific antidotes (Chapter 110). Chronic conditions can be treated by addressing the underlying cause, such as weight loss in obesity-hypoventilation syndrome or cholinesterase inhibitors in myasthenia gravis (Chapter 422). For parenchymal lung disease, treatment is directed at the underlying cause, if possible (Chapters 88 and 92). Sedative medications should be used cautiously because they can occasionally precipitate acute respiratory failure. Although profound hypoxemia can clearly be deleterious, oxygen occasionally can precipitate severe acute respiratory acidosis, particularly in patients with acute exacerbations of COPD (Chapter 88). As a result, hypoventilating patients with COPD require cautious management including the careful administration of supplemental oxygen, which should be titrated to an arterial oxygen saturation of 90% or an arterial oxygen tension of 60 mm Hg. Severe hypoventilation requires mechanical ventilation (Chapter 105), such as noninvasive ventilation for an acute exacerbation of COPD. For other presentations in which the Paco2 is believed to be acutely elevated, endotracheal intubation and mechanical ventilation are frequently used, especially in patients with impaired consciousness. For chronic hypoventilation in hypercapnic COPD, noninvasive bilevel positive airway pressure through a face mask during sleep can maintain alveolar ventilation, but there is no definitive evidence that noninvasive positive-pressure ventilation can prolong life or reduce hospitalizations in patients with COPD and chronic respiratory failure. A2  In addition, the considerable difficulty of adhering to nocturnal bilevel therapy in COPD emphasizes the need for discussions with patients and families regarding its risks and benefits. Other chronic hypoventilation syndromes are also commonly treated with bilevel positive airway pressure, although data are not compelling. In some chronic conditions, such as motor neuron disease (Chapter 419), tracheostomy should be discussed, although the impact of such interventions on quality of life should be carefully considered. Regardless of the underlying cause, an elevation in the Paco2 level is considered a poor prognostic sign. End-of-life discussions are also important in such cases because the prognosis of patients with chronic respiratory failure is generally poor.

PaCO2

pH > 7.4 Consider metabolic alkalosis if a 10-mEq rise in HCO3 yields 7-mm Hg rise in PaCO2 Treat underlying cause

pH < 7.4

Acute if 10-mm Hg rise in PaCO2 yields 1-mEq rise in HCO3 Abrupt presentation See Chapter 118

Chronic if 10-mm Hg rise in PaCO2 yields 4-mEq rise in HCO3

Consider cause Careful history Can’t breathe

Neuromuscular • Immediate orthopnea • Diaphragmatic percussion MIPs/ MEPs

Parenchymal √ Breath sounds √ PFTs Consider severe COPD

Won’t breathe: Check duration Low P0.1 Chest wall • Kyphoscoliosis • Obesity-hypoventilation

Symptoms since birth: √ PHOX2B gene

Acquired: Consider brain stem lesions, drugs, etc.

FIGURE 86-2.  A flow chart of a systematic approach to hypercapnia and various causes of hypoventilation. The change in pH can help determine the cause and chronicity. A careful history and physical examination, coupled with pulmonary function testing, can help classify patients into those who “can’t breathe” because of neuromuscular or mechanical abnormalities of the respiratory system compared with those who “won’t breathe” because of central nervous system disease. COPD = chronic obstructive pulmonary disease; MEPs = maximal expiratory pressures; MIPs = maximal inspiratory pressures; P0.1 = the negative mouth pressure generated during the first 100 msec of an occluded inspiration; PFTs = pulmonary function tests.

Grade A References A1. Bradley TD, Logan AG, Kimoff RJ, et al. Continuous positive airway pressure for central sleep apnea and heart failure. N Engl J Med. 2005;353:2025-2033. A2. Kohnlein T, Windisch W, Kohler D, et al. Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial. Lancet Respir Med. 2014;2:698-705.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 86  Disorders of Ventilatory Control  

GENERAL REFERENCES 1. Pack AI. Central sleep apnea. Handb Clin Neurol. 2011;98:411-419. 2. McGee S. Cheyne-stokes breathing and reduced ejection fraction. Am J Med. 2013;126:536-540. 3. Combs D, Shetty S, Parthasarathy S. Advances in positive airway pressure treatment modalities for hypoventilation syndromes. Sleep Med Clin. 2014;9:315-325. 4. Weese-Mayer DE, Berry-Kravis EM, Ceccherini I, et al. An official ATS clinical policy statement. Congenital central hypoventilation syndrome: genetic basis, diagnosis, and management. Am J Respir Crit Care Med. 2010;181:626-644.

548.e1

5. Rand CM, Yu M, Jennings LJ, et al. Germline mosaicism of PHOX2B mutation accounts for familial recurrence of congenital central hypoventilation syndrome (CCHS). Am J Med Genet A. 2012;158A: 2297-2301. 6. Boon AJ, Sekiguchi H, Harper CJ, et al. Sensitivity and specificity of diagnostic ultrasound in the diagnosis of phrenic neuropathy. Neurology. 2014;83:1264-1270.

548.e2

CHAPTER 86  Disorders of Ventilatory Control  

REVIEW QUESTIONS 1. Which of the following is currently the treatment of choice for CheyneStokes breathing in congestive heart failure? A. Optimize medical therapy B. Nasal continuous positive airway pressure C. Nasal bilevel positive airway pressure D. Oral appliance therapy E. Uvulopalatopharyngoplasty Answer: A  The treatment of choice for Cheyne-Stokes breathing is currently optimization of medical therapy for the underlying heart failure. Trials of nasal continuous positive airway pressure have failed to improve outcome compared with usual care. Bilevel therapy has not been rigorously studied but may make the situation worse. Although the upper airway can sometimes narrow or collapse in central apnea, there is no role for oral appliance therapy or uvulopalatopharyngoplasty in the absence of obstructive sleep apnea. 2. Which of the following best defines the measurement of loop gain in control of breathing? A. A measure of cardiac output B. A measure of extravascular lung fluid C. A measure of hemoglobin concentration D. A measure of the tendency toward instability in the ventilatory control system E. A measure of partial pressure of oxygen in the arterial circulation Answer: D  Loop gain refers to the overall instability in a negative feedback control system, such as the ventilatory control system, whose role is to maintain Paco2 levels. The other factors listed can all influence control of breathing in some way but are not defining loop gain. 3. Which of the following is the gene associated with the central congenital hypoventilation syndrome? A. CFTR B. PHOX2B C. A1AT D. RET E. BMP Answer: B  The PHOX2B gene is the hallmark for the diagnosis of central congenital hypoventilation syndrome. The other genes have been associated with various respiratory conditions but not with the central congenital hypoventilation syndrome.

4. Which of the following is true regarding obesity-hypoventilation syndrome? A. Serum bicarbonate is a useful screening test. B. Leptin deficiency is generally seen in afflicted humans. C. Obstructive sleep apnea is uncommon in afflicted patients. D. It is present in roughly 50% of patients with obstructive sleep apnea. E. Hypercapnia usually persists despite major weight loss. Answer: A  Serum bicarbonate is a useful screening test because it is elevated in the majority of patients with chronic hypoventilation. Leptin is deficient in some animal models but rare in human obesity. Obstructive sleep apnea is common in the obesity-hypoventilation syndrome, and obesityhypoventilation syndrome is seen in roughly 10% of obstructive sleep apnea. Weight loss usually improves gas exchange in these patients. 5. Which of the following is true regarding sleep stages in sleep apnea? A. Obstructive sleep apnea is generally worst during slow wave sleep. B. Cheyne-Stokes breathing is generally worst during REM sleep. C. Periodic breathing at altitude usually resolves during REM sleep. D. Obstructive sleep apnea is rare during REM sleep. E. Arousal in obstructive sleep apnea has no major effect on breathing. Answer: C  Both Cheyne-Stokes and periodic breathing at altitude are improved in REM compared with NREM sleep. Slow wave sleep is associated with improvement in obstructive sleep apnea. Arousal in obstructive sleep apnea can serve to restore pharyngeal airway patency and allow the resumption of effective tidal breathing.

548

CHAPTER 87  Asthma  

airway obstruction is largely reversible, it is currently thought that changes in the asthmatic airway may be irreversible in some settings.

EPIDEMIOLOGY

87  ASTHMA JEFFREY M. DRAZEN

DEFINITION

Asthma is a clinical syndrome of unknown etiology characterized by three distinct components: (1) recurrent episodes of airway obstruction that resolve spontaneously or as a result of treatment; (2) exaggerated bronchoconstrictor responses to stimuli that have little or no effect in nonasthmatic subjects, a phenomenon known as airway hyperresponsiveness; and (3) inflammation of the airways as defined by a variety of criteria. Although

Asthma is an extremely common disorder affecting boys more commonly than girls and, after puberty, women slightly more commonly than men; approximately 8% of the adult population of the United States has signs and symptoms consistent with a diagnosis of asthma. Although most cases begin before the age of 25 years, new-onset asthma may develop at any time throughout life. The worldwide prevalence of asthma has increased more than 45% since the late 1970s. In the last decade alone, the prevalence of wheezing in children has increased by about 0.1% per year. During the past four decades, the greatest increases in the prevalence of asthma have occurred in countries that adopted an “industrialized” lifestyle. For example, epidemiologic data suggest that being raised in a farming environment is associated with a much lower risk of asthma, independent of genetic factors, and this difference may be attributable to exposure to a greater diversity of environmental microbes early in life.1 Asthma is among the most common reasons to seek medical treatment. In the United States, it is responsible for about 15 million annual outpatient visits to physicians and for nearly 2 million annual inpatient hospital days of treatment. The estimated yearly direct and indirect costs of asthma care in the United States are more than $55 billion.

PATHOBIOLOGY

Genetics

In twin studies, asthma has about 60% heritability, indicating that both genetic and environmental factors are important in its etiology. A region on chromosome 17q21, at or near the locus for ORMDL3, a member of a gene family that encodes endoplasmic reticulum transmembrane proteins, has been repeatedly associated with childhood-onset asthma. Although the exact

CHAPTER 87  Asthma  

functional variant in this region has not been identified, the isolation of a locus for childhood-onset asthma supports the clinical observation that adult- and childhood-onset asthma appear to be distinct disorders. Genetic variants that influence the response to treatment also have been identified and widely replicated.

Pathology

The pathology of mild asthma, as delineated by bronchoscopic and biopsy studies, is characterized by edema and hyperemia of the mucosa and by infiltration of the mucosa with mast cells, eosinophils, and lymphocytes bearing the TH2 phenotype. Controlled trials using antibodies against interleukin-5 or the interleukin-4 receptor α chain in patients with persistent asthma symptoms and eosinophilia, despite treatment with corticosteroids, provide solid evidence for the pathobiologic role of these inflammatory cytokines in asthma. As a result of these inflammatory stimuli coupled with the mechanical deformation of the epithelium from airway, smooth muscle constriction,2 the airway wall is thickened by the deposition of type III and type V collagen below the true basem*nt membrane. In addition, in severe chronic asthma, there is hypertrophy and hyperplasia of airway glands and of both surface and glandular secretory cells as well as hyperplasia of airway smooth muscle. Morphometric studies of airways from asthmatic subjects have demonstrated airway wall thickening of sufficient magnitude to increase airflow resistance and to enhance airway responsiveness. During a severe asthmatic event, the airway wall is thickened markedly; in addition, patchy airway occlusion occurs by a mixture of hyperviscous mucus and clusters of shed airway epithelial cells. The episodic airway obstruction that constitutes an asthma attack results from narrowing of the airway lumen to airflow. Although it is now well established that asthma is associated with infiltration of the airway by inflammatory cells, the links between the presence of these cells and the pathobiologic processes that account for asthmatic airway obstruction are just beginning to be delineated. Three possible but not mutually exclusive links have been postulated: the constriction of airway smooth muscle, the thickening of airway epithelium, and the presence of liquids within the confines of the airway lumen. Among these mechanisms, the constriction of airway smooth muscle due to the local release of bioactive mediators or neurotransmitters is the most widely accepted explanation for the acute reversible airway obstruction in asthma attacks. Several bronchoactive mediators are thought to be the agents that initiate the airway obstruction characteristic of asthma. Moreover, the chronic airway narrowing, termed airway wall remodeling, that occurs in many patients with asthma likely results from the actions of inflammatory cells in the asthmatic airway.

549

antihistamines indicate only a minor role for histamine as a mediator of airway obstruction in asthma.

Leukotrienes and Lipoxins

The cysteinyl leukotrienes, namely, LTC4, LTD4, and LTE4, as well as the dihydroxy leukotriene LTB4 are derived by the lipoxygenation of arachidonic acid released from target cell membrane phospholipids during cellular activation. 5-Lipoxygenase, the 5-lipoxygenase–activating protein, and LTC4 synthase make up the cellular protein and enzyme content needed to produce the cysteinyl leukotrienes. The production of LTB4 requires 5-lipoxygenase, the 5-lipoxygenase–activating protein, and LTA4 epoxide hydrolase. Mast cells, eosinophils, and alveolar macrophages have the enzymatic capability to produce cysteinyl leukotrienes from their membrane phospholipids, whereas polymorphonuclear leukocytes produce exclusively LTB4, which is predominantly a chemoattractant molecule; LTC4 and LTD4 are among the most potent contractile agonists ever identified for human airway smooth muscle. The efficacy of a leukotriene receptor antagonist (i.e., pranlukast, zafirlukast, and montelukast) or a synthesis inhibitor (i.e., zileuton) in the treatment of chronic persistent asthma has led to the conclusion that the leukotrienes are important but not exclusive mediators of the asthmatic response. Lipoxins, which are double lipoxygenase products of arachidonic acid metabolism, have been shown to be endogenous downregulators of the inflammatory response. The amounts of lipoxins are decreased in the airways of patients with severe asthma.

Nitric Oxide

Nitric oxide (NO•) is produced enzymatically by airway epithelial cells and by inflammatory cells found in the asthmatic lung. Free NO• has a half-life on the order of seconds in the airway and is stabilized by conjugation to thiols to form RS-NO, where R designates any one of a number of molecular entities that can support this chemical linkage. Both NO• and RS-NO have bronchodilator actions and may play a homeostatic role in the airway. Paradoxically, high levels of NO•, when it is coavailable with superoxide anion, may form toxic oxidation products, such as peroxynitrite (OONO−), which could damage the airway. Patients with asthma have higher than normal levels of NO• in their expired air, and these levels decrease consistently after treatment with corticosteroids.

Physiological Changes in Asthma

Histamine, or β-imidazolylethylamine, was identified as a potent endogenous bronchoactive agent more than 100 years ago. Mast cells, which are prominent in airway tissues obtained from patients with asthma, constitute the major pulmonary source of histamine. Clinical trials with novel potent

An increased resistance to airflow is the consequence of the airway obstruction induced by smooth muscle constriction, thickening of the airway epithelium, or free liquid within the airway lumen. Obstruction to airflow is manifested by increased airway resistance and decreased flow rates throughout the vital capacity. At the onset of an asthma attack, obstruction occurs at all airway levels; as the attack resolves, these changes are reversed—first in the large airways (i.e., mainstem, lobar, segmental, and subsegmental bronchi) and then in the more peripheral airways. This anatomic sequence of onset and reversal is reflected in the physiological changes observed during resolution of an asthmatic episode. Specifically, as an asthma attack resolves, flow rates first normalize at volumes high in the vital capacity and only later at volumes low in the vital capacity. Because asthma is an airway disease, not an air space disease, no primary changes occur in the static pressure-volume curve of the lungs. However, during an acute attack of asthma, airway narrowing may be so severe as to result in airway closure, with individual lung units closing at a volume that is near their maximal volume. This closure results in a change of the pressure-volume curve such that for a given

Acute-Severe

Late-Resolution

Mediators of the Acute Asthmatic Response Acetylcholine

Acetylcholine released from intrapulmonary motor nerves causes constriction of airway smooth muscle through direct stimulation of muscarinic receptors of the M3 subtype. The potential role for acetylcholine in the bronchoconstriction of asthma primarily derives from the observation that tiotropium bromide, a muscarinic antagonist, can reduce bronchoconstriction.

Histamine

Early-Resolution

.

.

.

VE

VE

VE

TLC TLC

RV VL

RV

VL

VL

FIGURE 87-1.  Schematic flow-volume curves in various stages of asthma. In each figure, the dashed line depicts the normal flow-volume curve. Predicted and observed total lung capacity (TLC) and residual volume (RV) are shown at the extremes of each curve. V E = expiratory flow rate; VL = lung volume.

550

CHAPTER 87  Asthma  

contained gas volume within the thorax, elastic recoil is decreased, which in turn further depresses expiratory flow rates. Additional factors influence the mechanical behavior of the lungs during an acute attack of asthma. During inspiration in an asthma attack, the maximal inspiratory pleural pressure becomes more negative than the subatmospheric pressure of 4 to 6 cm H2O usually required for tidal airflow. The expiratory phase of respiration also becomes active as the patient tries to force air from the lungs. As a consequence, peak pleural pressures during expiration, which normally are, at most, only a few centimeters of water above atmospheric pressure, may be as high as 20 to 30 cm H2O above atmospheric pressure. The low pleural pressures during inspiration tend to dilate airways, whereas the high pleural pressures during expiration tend to narrow airways. During an asthma attack, the wide pressure swings, coupled with alterations in the mechanical properties of the airway wall, lead to a much higher resistance to expiratory airflow than to inspiratory airflow. The respiratory rate is usually rapid during an acute asthmatic attack. This tachypnea is driven not by abnormalities in arterial blood gas composition but rather by stimulation of intrapulmonary receptors with subsequent effects on central respiratory centers. One consequence of the combination of airway narrowing and rapid airflow rates is a heightened mechanical load on the ventilatory pump. During a severe attack, the load can increase the work of breathing by a factor of 10 or more and can predispose to fatigue of the ventilatory muscles. With respect to gas exchange, the patchy nature of asthmatic airway narrowing results in a maldistribution of ventilation (V) relative to pulmonary perfusion (Q). A shift occurs from the normal preponderance of V/Q units, with a ratio of near unity, to a distribution with a large number of alveolar-capillary units, with a V/Q ratio of less than unity. The net effect is to induce arterial hypoxemia. In addition, the hyperpnea of asthma is reflected as hyperventilation with a low arterial Pco2.

CLINICAL MANIFESTATIONS

History

During an acute asthma attack, patients seek medical attention for shortness of breath accompanied by cough, wheezing, and anxiety. The degree of breathlessness experienced by the patient is not closely related to the degree of airflow obstruction but is often influenced by the acuteness of the attack. Dyspnea may occur only with exercise (exercise-induced asthma),3 after aspirin ingestion (aspirin-exacerbated respiratory disease),4 after exposure to a specific known allergen (extrinsic asthma), or for no identifiable reason (intrinsic asthma). Variants of asthma exist in which cough, hoarseness, or inability to sleep through the night is the only symptom. Identification of a provoking stimulus through careful questioning helps establish the diagnosis of asthma and may be therapeutically useful if the stimulus can be avoided. Most patients with asthma complain of shortness of breath when they are exposed to rapid changes in the temperature and humidity of inspired air. For example, during the winter months in less temperate climates, patients commonly become short of breath on leaving a heated house; in warm humid climates, patients may complain of shortness of breath on entering a cold dry room, such as an air-conditioned theater. An important factor to consider in taking a history from a patient with asthma is the potential for occupational exposures in asthma (Chapter 93). Asthma that is brought on by occupational exposures is termed occupational asthma; preexisting asthma that is exacerbated by workplace exposures is termed workplace-exacerbated asthma. In reactive airway dysfunction syndrome, a single large exposure leads to a persistent asthma-like phenotype in a previously normal individual.5

Physical Examination Vital Signs

Common features noted during an acute attack of asthma include a rapid respiratory rate (often 25 to 40 breaths per minute), tachycardia, and pulsus paradoxus (an exaggerated inspiratory decrease in the systolic pressure). The magnitude of the pulsus is related to the severity of the attack; a value greater than 15 mm Hg indicates an attack of moderate severity. Pulse oximetry, with the patient respiring ambient air, commonly reveals an oxygen saturation near 90%.

Thoracic Examination

Inspection may reveal that patients experiencing acute attacks of asthma are using their accessory muscles of ventilation; if so, the skin over the thorax may be retracted into the intercostal spaces during inspiration. The chest is

usually hyperinflated, and the expiratory phase is prolonged relative to the inspiratory phase. Percussion of the thorax demonstrates hyperresonance, with loss of the normal variation in dullness due to diaphragmatic movement; tactile fremitus is diminished. Auscultation reveals wheezing, which is the cardinal physical finding in asthma but does not establish the diagnosis (Chapter 83). Wheezing, commonly louder during expiration but heard during inspiration as well, is characterized as polyphonic in that more than one pitch may be heard simultaneously (Video 87-1). Accompanying adventitious sounds may include rhonchi, which are suggestive of free secretions in the airway lumen, or rales, which should raise the suspicion of an alternative diagnosis and are indicative of localized infection or heart failure. The loss of intensity or the absence of breath sounds in a patient with asthma is an indication of severe airflow obstruction.

DIAGNOSIS

Laboratory Findings Pulmonary Function Findings

A decrease in airflow rates throughout the vital capacity is the cardinal pulmonary function abnormality during an asthmatic episode.6,7 The peak expiratory flow rate (PEFR), the forced expiratory volume in the first second (FEV1), and the maximal mid-expiratory flow rate (MMEFR) are all decreased in asthma (Chapter 85). In severe asthma, dyspnea may be so severe as to prevent the patient from performing a complete spirogram. In this case, if 2 seconds of forced expiration can be recorded, useful values for PEFR and FEV1 can be obtained. Gradation of attack severity (Table 87-1) must be assessed by objective measures of airflow; no other methods yield accurate and reproducible results. As the attack resolves, the PEFR and the FEV1 increase toward normal together while the MMEFR remains substantially depressed; as the attack resolves further, the FEV1 and the PEFR may normalize while the MMEFR remains depressed (see Fig. 87-1). Even when the attack has fully resolved clinically, residual depression of the MMEFR is not uncommon; this depression may resolve during a prolonged course of treatment. If the patient is able to cooperate such that more complete measurements of lung function can be made, lung volume measurements made during an attack demonstrate an increase in both total lung capacity and residual volume; the changes in total lung capacity and residual volume resolve with treatment. Because of the extra cooperation needed for this testing, it is not advised during an acute asthmatic event but is indicated before discharge in a patient hospitalized for the treatment of asthma or between episodes of asthma. Pulmonary function testing obtained when the patient is relatively stable usually demonstrates airway obstruction, as indicated by low FEV1 (as a percentage of the patient’s predicted value), low forced vital capacity, and slightly elevated total lung capacity and residual volume values. These results may fully normalize after administration of a bronchodilator, but a “bronchodilator response” is canonically defined as a 12% increase in the FEV1, provided it is at least 200 mL (E-Fig. 87-1).

TABLE 87-1  RELATIVE SEVERITY OF AN ASTHMATIC ATTACK AS INDICATED BY PEFR, FEV1, AND MMEFR TEST

PREDICTED VALUE (%)

SEVERITY OF ASTHMA

PEFR FEV1 MMEFR

>80 >80 >80

No spirometric abnormalities

PEFR FEV1 MMEFR

>80 >70 55-75

Mild asthma

PEFR FEV1 MMEFR

>60 45-70 30-50

Moderate asthma

PEFR FEV1 MMEFR

800-1600

Ciclesonide (Alvesco)†

80-160

160-320

>320-1280

Flunisolide (AeroBid/AeroBid-M)†

500-1000

1000-2000

Fluticasone (Flovent)†

100-250

250-500

>500-1000

Mometasone furoate (Asmanex)†

200-400

400-800

>800-1200

Triamcinolone acetonide (Azmacort)†

400-1000

1000-2000

DRUG

>2000

>2000

*Once-a-day dosing is acceptable for low daily dose. † Trade name in the United States. Note: Some doses may be outside package labeling. Metered-dose inhaler doses are expressed as the amount of drug leaving the valve, not all of which is available to the patient. Dry powder inhaler doses are expressed as the amount of drug in the inhaler after activation. Modified from 2012 Global Initiative for Asthma guidelines. www.ginasthma.com.

daily) or the action of leukotrienes at the CysLT1 receptor (montelukast [Singulair], 10 mg once a day; pranlukast [Onon, Ultair], 225 mg twice a day, not available in the United States; and zafirlukast [Accolate], 20 mg twice a day) are effective oral controller medications for patients with mild or moderate persistent asthma.11 In patients treated with zileuton, alanine aminotransferase levels should be monitored for the first 3 to 6 months of treatment; if levels rise to more than three times the upper limit of normal, the drug should be stopped. Theophylline metabolism is slowed by zileuton, so monitoring of levels is indicated if both are prescribed. These treatments can be used on their own for mild persistent asthma or in combination with inhaled steroids for more severe asthma.

Long-Acting β-Agonists

In contrast to short- to medium-acting β-agonists, long-acting β-agonists currently available in the United States (salmeterol [Serevent, 42 µg per puff; the same dose is labeled 50 µg per puff outside of the United States; one or two puffs should be delivered every 12 hours], formoterol [Foradil, 12 µg through a proprietary dry powder inhaler every 12 hours], and bambuterol [Bambec and Oxeol, 10 to 20 mg orally each evening]) have a duration of action of nearly 12 hours and are considered controller agents rather than acute bronchodilator agents. Indacaterol (trade names: Arcapta in the United States and Onbrez in Europe), which is an ultralong-lasting β-agonist delivered by a dry powder inhaler, is used only once a day; its label indications are for chronic obstructive lung disease, not asthma, but it may be used in patients whose asthma is also being treated concomitantly with an inhaled corticosteroid. Randomized controlled trials demonstrate that long-acting β-agonists should not be used as a sole controller agent. Other trials have shown that there are excess asthma deaths (about one for every 650 patient-years of treatment) when long-acting β-agonists are used. Therefore, long-acting β-agonists should be used in patients with asthma only when they are given in concert with inhaled corticosteroids. A number of combination products contain both inhaled steroids and longacting β-agonists in the same aerosol device. These products prevent patients with asthma from using inhaled long-acting β-agonists without inhaled corticosteroids. When prescribing a combination inhaler, the physician should determine the inhaled dose of corticosteroids (fluticasone, budesonide, beclomethasone, mometasone) that the patient requires and then choose a combination product that will deliver a dose of long-acting β-agonist with the inhaled corticosteroid when it is given as two puffs twice per day. The dose of long-acting β-agonist varies with brand and type of inhaler used.

Theophylline

Theophylline and its more water-soluble congener aminophylline are moderately potent bronchodilators that are useful in both inpatient and outpatient management of asthma. Treatment with theophylline is recommended only for patients who have moderate or severe persistent asthma and who are receiving controller medications, such as inhaled steroids or antileukotrienes, but whose asthma is not adequately controlled despite these treatments.

The mechanism by which theophylline exerts its effects has not been established with certainty but is probably related to the inhibition of certain forms of phosphodiesterase. Theophylline is not widely used because of its toxicity and the wide variations in the rate of its metabolism, both in a single individual over time and among individuals in a population. Because blood levels need to be monitored for optimal dosing, most physicians have reserved theophylline for third- or fourth-line therapy. For most preparations, the starting dose should be about 300 mg/day; the frequency will depend on the preparation used. Acceptable plasma levels for therapeutic effects are between 10 and 20 µg/ mL; higher levels are associated with gastrointestinal, cardiac, and central nervous system toxicity, including anxiety, headache, nausea, vomiting, diarrhea, cardiac arrhythmias, and seizures. These last catastrophic complications may occur without antecedent mild side effects when plasma levels exceed 20 µg/mL. Because of these potentially life-threatening complications of treatment, plasma levels need to be measured with great frequency in hospitalized patients receiving intravenous aminophylline and less frequently in stable outpatients receiving one of the long-acting theophylline preparations. Most asthma care providers use dosing amounts and intervals to achieve steadystate theophylline levels of 10 to 14 µg/mL, thereby avoiding the toxicity associated with decrements in metabolism.

Systemic Corticosteroids

Systemic corticosteroids are effective for the treatment of moderate to severe persistent asthma as well as for occasional severe exacerbations of asthma in a patient with otherwise mild asthma, but the mechanism of their therapeutic effect has not been established. No consensus has been reached on the specific type, dose, or duration of corticosteroid to be used in the treatment of asthma. In nonhospitalized patients with asthma refractory to standard therapy, a steroid “pulse” with initial doses of prednisone on the order of 40 to 60 mg/day, tapered to zero during 7 to 14 days, is recommended. For patients who cannot stop taking steroids without having recurrent uncontrolled bronchospasm despite the addition of multiple other controller treatments, alternate-day administration of oral steroids is preferable to daily treatment. For patients whose asthma requires in-hospital treatment but is not considered life-threatening, an initial intravenous bolus of 2 mg/kg of hydrocortisone, followed by continuous infusion of 0.5 mg/kg/hour, has been shown to be beneficial within 12 hours. In attacks of asthma that are considered life-threatening, the use of intravenous methylprednisolone (125 mg every 6 hours) has been advocated. In each case, as the patient improves, oral steroids are substituted for intravenous steroids, and the oral dose is tapered during 1 to 3 weeks; addition of inhaled steroids to the regimen is strongly recommended when oral steroids are started.

Monoclonal Antibody Treatment Omalizumab

Subcutaneous administration of omalizumab, a humanized murine monoclonal antibody that binds circulating IgE, is associated with decreased serum free (not total) IgE levels. In patients who have moderate to severe allergic asthma with elevated levels of serum IgE and who are receiving inhaled corticosteroids, omalizumab treatment improves asthma control even as doses of inhaled steroids are decreased. Dosing is guided by weight and by pretreatment IgE levels: a monthly subcutaneous dose of 0.016 mg × body weight (kg) × IgE level (IU/mL). For example, in a patient weighing 70 kg with a pretreatment total IgE level of 300 IU/mL, 336 mg of omalizumab would be administered monthly by subcutaneous injection. Dosing calculators can be found online (e.g., http://www.xolairhcp.com/hcp/determining-the-dose.html). AntiIgE antibodies can reduce exacerbations and improve quality of life in patients with severe allergic asthma, but their place in treatment schema has not been established. Because of the potential for anaphylaxis, all patients need to be monitored after injection; the duration of the monitoring period is not specified by the U.S. Food and Drug Administration (FDA), but most physicians monitor for 30 to 60 minutes.

Other Monoclonal Antibodies

In a randomized trial, lebrikizumab (a monoclonal antibody against interleukin-13 at 250 mg subcutaneously once per month for 6 months) enhanced lung function in adults with asthma, especially in patients with low pretreatment serum periostin levels. A5  Mepolizumab is a monoclonal antibody directed against interleukin-5. In randomized trials using 75 to 100 mg daily, it reduced asthma exacerbations by about 50% among relatively rare patients with moderately severe asthma who still had sputum eosinophilia despite treatment with oral and inhaled corticosteroids. A6  A7  Among patients with more conventional asthma, however, mepolizumab treatment did not have a salutary effect. A trial of a monoclonal antibody against the α subunit of the shared interleukin-4 and interleukin-13 receptor, dupilumab, in patients whose asthma and eosinophilia was not controlled with conventional doses of inhaled corticosteroids and long-acting β-agonists showed that both the inhaled long-acting β-agonists and inhaled corticosteroids could be withdrawn without losing asthma control when the monoclonal antibody was administered. A8  None of these drugs is currently approved by the U.S. FDA. ,

554

CHAPTER 87  Asthma  

Other Controller Drugs

Cromolyn sodium (two to four times a day by nebulizer using 20-mg nebules) is a nonsteroid inhaled treatment used in the management of mild to moderate persistent asthma. It appears to be most useful in pediatric populations or when an identifiable stimulus (such as exercise or allergen exposure) elicits an asthmatic response. The use of systemic gold (as in rheumatoid arthritis), methotrexate, or cyclosporine has been suggested as adjunctive treatment of patients with severe chronic asthma who cannot otherwise discontinue high-dose corticosteroid treatment. However, these agents are experimental, and their routine use is not advocated. Despite initial encouraging trials, agents that inhibit the action of tumor necrosis factor-α do not benefit patients with asthma and should not be used. Based on the concern that asthma could be caused by silent gastroesophageal reflux disease, treatment with a proton pump inhibitor has been advocated in patients with mild to moderate asthma even in the absence of gastrointestinal symptoms. Adequately powered clinical studies suggest that this approach provides no benefit for asthma control.

Asthma in the Emergency Department

A proprietary system to ablate airway smooth muscle by delivery of radio frequency energy through a bronchoscopically placed probe has reduced asthma exacerbations in sham-controlled trials among patients whose asthma remained out of control despite the use of multiple controller medications. Although a device for such treatment has been approved by the FDA, the long-term impacts of this treatment on airway or lung function are not known.

When a patient with asthma presents for acute emergency care, objective measures of the severity of the attack, including quantification of pulsus paradoxus and measurement of airflow rates (PEFR or FEV1), should be evaluated in addition to the usual vital signs. If the attack has been prolonged and failed to respond to treatment with bronchodilators (e.g., albuterol by metered-dose inhaler, two puffs every 2 to 3 hours) and high-dose inhaled steroids (e.g., more than 2000 µg/day of beclomethasone or half that amount of fluticasone) before arrival at the emergency department, intravenous steroids (40 to 60 mg of methylprednisolone or its equivalent) should be administered. If the patient has not been receiving treatment with a leukotriene receptor antagonist, such agents should be administered (10 mg of montelukast or 20 mg of zafirlukast) as soon as possible. Treatment with inhaled β-agonists (either nebulized albuterol, 0.5 mL of a 0.5% solution repeated at 20- to 30-minute intervals, or albuterol by metered-dose inhaler, two puffs every 30 minutes) should be used until the PEFR or FEV1 increases to greater than 40% of the predicted values. If this point is not reached within 2 hours, admission to the hospital for further treatment is strongly advocated. When patients have PEFR and FEV1 values that are greater than 60% of their predicted value on arrival in the emergency department, treatment with inhaled β2-agonists alone, albuterol (0.5 mL of an albuterol 0.083% solution) or equivalent, is likely to result in an objective improvement in airflow rates. If significant improvement takes place in the emergency department, such patients can usually be treated as outpatients with inhaled β2-agonists and a controller agent (see Fig. 87-2). A good strategy is to add inhaled corticosteroids if the patient has not been receiving this treatment or has been using a single controller therapy. For patients whose PEFR and FEV1 values are between 40% and 60% of the values predicted at the time of initial evaluation in the emergency care setting, a plan of treatment varying in intensity between these two plans is indicated. Failure to respond to treatment by objective criteria (PEFR or FEV1) within 2 hours of arrival at the emergency department is an indication for the use of systemic corticosteroids.

Control-Driven Asthma Therapy

Status Asthmaticus

Vaccination for Seasonal Influenza and Pneumococcal Disease

Vaccination of patients for seasonal influenza is safe and not associated with enhanced asthma exacerbations. Vaccination against seasonal influenza and pneumococcal disease is recommended in patients with asthma.

Radio Frequency Ablation of Airway Smooth Muscle

Because all current asthma treatment is symptomatic (i.e., no current treatment changes the disease history), the approach to the management of asthma is to titrate treatment to achieve an adequate level of control. If a patient’s asthma is well controlled, treatment can be continued or stepped down (see Fig. 87-2). A9  If a patient’s asthma is poorly controlled, treatment intensity should be stepped up. At the mild end of the spectrum, a patient who has rare limitations in activities of daily life, has nearly normal lung function, and sleeps without interruption from asthma can be prescribed nothing more than inhaled rescue treatment on an as-needed basis. In general, if a patient can control his or her asthma with the use of a single metered-dose inhaler of rescue treatment dispensed every 7 to 8 weeks or less frequently, there is no need for background controller treatment. If a patient has a requirement for more rescue treatment, has symptoms that interfere with sleeping through the night, or has moderately deranged lung function, controller therapy should be added. Single-agent controller therapy should consist of an inhaled corticosteroid or an antileukotriene. If control is not achieved with one of these agents, the patient can be switched to the other or have a second agent added. The best studied two-agent combination is inhaled corticosteroids and a long-acting inhaled β2-adrenergic agonist, available in a single inhaler under the trade names of Symbicort, Advair, and Dulera in the United States; trade names vary in other parts of the world. These combinations provide excellent disease control and often allow a reduction in the dose of inhaled corticosteroids. Data indicate that another combination, an antileukotriene and inhaled steroid, is more effective than either treatment alone, but this regimen does not have as substantial an evidence base as the combination of inhaled corticosteroids and a long-acting β-agonist.

Specific Treatment Scenarios Concurrent Pulmonary Infection

In some patients, acute exacerbations of asthma may be due to concurrent infection, which requires targeted therapy (Chapters 88, 90, and 97).

Aspirin-Exacerbated Respiratory Disease (Previously Termed Aspirin-Induced Asthma)

Approximately 5% of patients with moderate to severe persistent asthma develop asthma when they ingest agents that inhibit cyclooxygenase, such as aspirin and other nonsteroidal anti-inflammatory drugs (Chapter 37). Inhibitors of cyclooxygenase 2 are less likely to cause these reactions, but aspirintype reactions have been reported in sensitive patients treated with selective cyclooxygenase 2 inhibitors. Although the physiologic manifestations of laboratory-based aspirin challenge can be blocked by leukotriene pathway inhibitors, these agents do not prevent clinical aspirin-exacerbated respiratory disease. Thus, patients with this form of asthma must avoid aspirin and other nonsteroidal anti-inflammatory drugs.

The asthmatic subject whose PEFR or FEV1 does not increase to greater than 40% of the predicted value with treatment, whose Paco2 increases without improvement of indices of airflow obstruction, or who develops major complications such as pneumothorax or pneumomediastinum should be admitted to the hospital for close monitoring. Frequent treatments with inhaled β-agonists (0.5 mL of an albuterol 0.083% solution every 2 hours), intravenous aminophylline (at doses to yield maximal acceptable plasma levels, that is, 15 to 20 µg/mL; 500- to 1000-mg loading dose given during an hour followed by an infusion of 30 to 60 mg/hour), and high-dose intravenous steroids (methylprednisolone, 40 to 60 mg every 4 to 6 hours) are indicated. Oxygen should be administered by face mask or nasal cannula in amounts sufficient to achieve Sao2 values between 92% and 94%; a higher Fio2 promotes absorption atelectasis and provides no therapeutic benefit. If objective evidence of an infection is present, appropriate treatment should be given for that infection. If no improvement is seen with treatment and if respiratory failure appears imminent, bronchodilator treatment should be intensified to the maximum tolerated by the patient as indicated by the maximum tolerated heart rate, usually 130 to 140 beats per minute. If indicated, intubation of the trachea and mechanical ventilation can be instituted; in this case, the goal should be to provide a level of ventilation just adequate to sustain life but not sufficient to normalize arterial blood gases. For example, a Paco2 of 60 to 70 mm Hg, or even higher, is acceptable for a patient in status asthmaticus.

Asthma in Pregnancy

Asthma may be exacerbated, remain unchanged, or remit during pregnancy (Chapter 239). There need not be substantial departures from the ordinary management of asthma during pregnancy, although one randomized trial suggests that unlike in other settings, measurement of the fraction of exhaled nitric oxide can improve the management of asthma during pregnancy. However, no unnecessary medications should be administered; systemic steroids should be used sparingly to avert fetal complications, and certain drugs should be avoided, including tetracycline (as a treatment of intercurrent infection), ipratropium bromide (which may cause fetal tachycardia), terbutaline (which is contraindicated during active labor because of its tocolytic effects), and iodine-containing mucolytics (such as saturated solution of potassium iodide). Moreover, use of prostaglandin F2α as an abortifacient should be avoided in asthmatic patients.

PROGNOSIS

Asthma is a chronic relapsing disorder. Most patients have recurrent attacks without a major loss in lung function for many years. A minority of patients experience a significant irreversible loss in lung function over and above the normal pulmonary senescence. Methods to distinguish these various clinical phenotypes have not been developed.

Grade A References A1. Petsky HL, Cates CJ, Lasserson TJ, et al. A systematic review and meta-analysis: tailoring asthma treatment on eosinophilic markers (exhaled nitric oxide or sputum eosinophils). Thorax. 2012;67: 199-208. A2. Peters SP, Kunselman SJ, Icitovic N, et al. Tiotropium bromide step-up therapy for adults with uncontrolled asthma. N Engl J Med. 2010;363:1715-1726. A3. Busse WW, Pedersen S, Pauwels RA, et al. START Investigators Group. The Inhaled Steroid Treatment As Regular Therapy in Early Asthma (START) study 5-year follow-up: effectiveness of early intervention with budesonide in mild persistent asthma. J Allergy Clin Immunol. 2008;121: 1167-1174. A4. Calhoun WJ, Ameredes BT, King TS, et al. Asthma Clinical Research Network of the National Heart, Lung, and Blood Institute. Comparison of physician-, biomarker- and symptom-based strategies for adjustment of inhaled corticosteroid therapy in adults with asthma: the BASALT randomized controlled trial. JAMA. 2012;308:987-997. A5. Corren J, Lemanske RF, Hanania NA, et al. Lebrikizumab treatment in adults with asthma. N Engl J Med. 2011;365:1088-1098. A6. Bel EH, Wenzel SE, Thompson PJ, et al. Oral glucocorticoid-sparing effect of mepolizumab in eosinophilic asthma. N Engl J Med. 2014;371:1189-1197. A7. Ortega HG, Liu MC, Pavord ID, et al. Mepolizumab treatment in patients with severe eosinophilic asthma. N Engl J Med. 2014;371:1198-1207. A8. Wenzel S, Ford L, Pearlman D, et al. Dupilumab in persistent asthma with elevated eosinophil levels. N Engl J Med. 2013;368:2455-2466. A9. Peters SP, Anthonisen N, Castro M, et al. American Lung Association Asthma Clinical Research Centers. Randomized comparison of strategies for reducing treatment in mild persistent asthma. N Engl J Med. 2007;356:2027-2039.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 87  Asthma  

GENERAL REFERENCES 1. Ege MJ, Mayer M, Normand AC, et al. Exposure to environmental microorganisms and childhood asthma. N Engl J Med. 2011;364:701-709. 2. Grainge CL, Lau LC, Ward JA, et al. Effect of bronchoconstriction on airway remodeling in asthma. N Engl J Med. 2011;364:2006-2015. 3. Anderson SD, Kippelen P. Assessment and prevention of exercise-induced bronchoconstriction. Br J Sports Med. 2012;46:391-396. 4. Kowalski ML, Makowska JS, Blanca M, et al. Hypersensitivity to nonsteroidal anti-inflammatory drugs (NSAIDs)—classification, diagnosis and management: review of the EAACI/ENDA and GA2LEN/HANNA. Allergy. 2011;66:818-829.

555.e1

Tarlo SM, Lemiere C. Occupational asthma. N Engl J Med. 2014;370:640-649. Bel EH. Clinical practice. Mild asthma. N Engl J Med. 2013;369:549-557. Martinez FD, Vercelli D. Asthma. Lancet. 2013;382:1360-1372. Global Initiative for Asthma. http://www.ginasthma.org. Accessed January 7, 2015. Cazzola M, Page CP, Rogliani P, et al. β2-Agonist therapy in lung disease. Am J Respir Crit Care Med. 2013;187:690-696. 10. Rogers L, Hanania NA. Role of anticholinergics in asthma management: recent evidence and future needs. Curr Opin Pulm Med. 2015;21:103-108. 11. Price D, Musgrave SD, Shepstone L, et al. Leukotriene antagonists as first-line or add-on asthmacontroller therapy. N Engl J Med. 2011;364:1695-1707. 5. 6. 7. 8. 9.

555.e2

CHAPTER 87  Asthma  

REVIEW QUESTIONS 1. A 25-year-old woman who works in an office complains of shortness of breath during the recovery period from her usual aerobic exercise routine. She had a history of asthma as a child, but it went into remission when she was in junior high school. What single test would be the best to order to easily confirm that she is now having a return of her asthma? A. Blood eosinophils B. Measurement of forced vital capacity before and after bronchodilator C. Measurement of the forced expiratory volume in the first second (FEV1) before and after bronchodilator D. Testing for airway hyperresponsiveness with inhalation of cold air E. Measurement of the diffusion capacity for carbon monoxide Answer: C  If the FEV1, measured before and after bronchodilator treatment, shows an improvement of 12% or more (assuming an absolute increase in the FEV1 of at least 200 mL), this finding would be diagnostic of asthma in this setting. The same is not true of the forced vital capacity. Measurement of airway hyperresponsiveness would make the diagnosis but is complicated and time-consuming. Measurements of blood eosinophils or diffusion capacity are not indicated in this setting. 2. For the same patient as in question 1, you take a medical history to determine if her asthma is in control. Which of the following questions would you not need to ask? A. Has she been able to sleep through the night without asthma symptoms? B. How often has she needed to use her rescue inhaler to control her asthma symptoms? C. Has she needed to seek unscheduled medical care for her asthma? D. Can she fulfill the activities of her daily life without asthma symptoms? E. Is she able to scuba dive without difficulty? Answer: E  Answers A through D are part of the information that should be sought from all asthma patients. Patients with asthma should not scuba dive.

3. A 35-year-old woman with asthma comes for her yearly checkup. For the past year, she has been treated with an albuterol inhaler that she uses on an as-needed basis. During this time, she has used two inhalers (each with 200 puffs of treatment) and no other asthma medications. She is able to sleep through the night without asthmatic symptoms. She is able to participate in all her desired activities, including a regular exercise routine, without asthmatic symptoms, except during the colder months of the year, when she pretreats herself with albuterol before running. She has not had any unscheduled visits for asthma care. Her lung function test results are essentially normal. At this point you should A. Add an inhaled corticosteroid at low dose to her asthma regimen B. Add a leukotriene modifier to her asthma regimen C. Add a combination inhaler (long-acting β-agonist and inhaled corticosteroid) D. Leave her regimen as it is E. Remove her albuterol before she becomes addicted to it Answer: D  She is well controlled while she is treated with albuterol alone. There is no need to add or to subtract other therapies. 4. A 25-year-old woman whose asthma is in good control comes to your office indicating that she and her husband are trying to have a child. Her only treatment has been inhaled albuterol on an as-needed basis, and she uses about four inhalers (each contains 200 puffs of medication) a year. What is the single statement most likely to be true about her asthma treatment? A. There is no need to change her treatment while she is trying to conceive or becomes pregnant. B. She should stop the albuterol inhaler and simply “suffer through” any asthma events. C. She should continue the albuterol inhaler and start an inhaled corticosteroid. D. She should consult an asthma specialist now before she attempts to become pregnant. E. She should begin to use her albuterol inhaler on a regularly scheduled basis, two puffs four times a day. Answer: A  There is no need to change her asthma therapy at this time because albuterol is safe to use during pregnancy. The other answers are, by elimination, incorrect.

555

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

be anticipated in Asia and other regions of the world because of rapidly increasing cigarette consumption. More than 10% of the population older than 45 years in the United States has airflow obstruction of at least moderate severity as judged by spirometric criteria. COPD is the third leading cause of death in the United States, and mortality from COPD has increased during the past 30 years in both men and women.1 Worldwide, COPD also is the third leading cause of death globally and the fifth leading cause of years lived with disability.2,3 Medical costs and lost productivity attributable to COPD exceed $40 billion annually in the United States. Direct medical costs rise precipitously as COPD becomes more severe, with hospitalization for exacerbations accounting for more than half of the total.

Cigarette Smoking

88  CHRONIC OBSTRUCTIVE PULMONARY DISEASE DENNIS E. NIEWOEHNER

Cigarette smoking is the principal cause of COPD, but the relationship is complex and COPD may develop without a smoking history.4 Airflow obstruction is the sentinel physiologic disturbance in COPD, and the forced expiratory volume in the first second (FEV1) is the single best indicator of severity. Cigarette smoking causes declines in lung function that exceed those expected from aging alone, and the magnitude of loss is dependent on both the intensity and duration of exposure to cigarette smoke. Thus, the cumulative effects of smoking largely account for the increasing prevalence of COPD with advancing age. Individual losses of lung function vary widely, even after adjustment for smoking intensity. After the age of 30 years, everyone loses lung function on a yearly basis, but smoking further affects the rate of lung function loss. The mean annual reduction in the FEV1 (Chapter 85) in normal nonsmoking white men is about 25 mL per year, but the loss increases to an average of about 40 mL per year among smokers (Fig. 88-1). A small minority of smokers, “susceptible smokers,” suffer annual FEV1 losses of 100 mL or more and may develop clinically significant airflow obstruction in the fourth and fifth decades of life. Factors that distinguish the susceptible smoker from the average smoker remain largely unknown. Adverse effects of cigarette smoke on lung function may extend as far back as fetal development. Maternal smoking during pregnancy, secondhand cigarette smoke exposure during early childhood, and active smoking during adolescence impair lung growth. As a consequence, the lower lung function in early adulthood constitutes a significant risk factor for COPD later in life.

Other Environmental Exposures

EPIDEMIOLOGY

COPD represents a growing global public health problem, although prevalence estimates vary widely according to the definition used. Cigarette smoking (Chapter 32) is the principal risk factor for COPD, so prevalence tends to reflect societal smoking habits with a lag phase of 20 to 30 years. Cigarette consumption has leveled off or decreased in large segments of North America and Europe, but the prevalence of COPD may continue to increase as exposed populations age. A greater future burden of COPD may

100 FEV1 (% predicted at age 25 yr)

DEFINITIONS

Chronic obstructive pulmonary disease (COPD) is now the preferred term for a condition that is characterized by progressive, largely irreversible airflow obstruction, usually with clinical onset in middle-aged or elderly persons with a history of cigarette smoking, and that cannot be attributed to another specific disease, such as bronchiectasis (Chapter 90) or asthma (Chapter 87). Commonly used terms for this condition in the past included chronic bronchitis and emphysema. That terminology is outdated because nearly all patients with a clinical diagnosis of COPD have both air space destruction (i.e., emphysema) and pathologic changes of the conducting airways consistent with chronic bronchitis. Emphysema is defined pathologically by abnormal enlargement of the air spaces due to destruction and deformation of alveolar walls. The severity of emphysema may vary widely in COPD patients with similar degrees of airflow obstruction. Chronic bronchitis is defined clinically as persistent cough and sputum production and pathologically as abnormal enlargement of the mucous glands within the central cartilaginous airways. Chronic bronchitis was once thought to be a key element in the pathogenesis of chronic airflow obstruction, but it is now known that increased airflow resistance in COPD can be attributed principally to a variety of pathologic changes within the distal airways of the lung (“small airways disease”).

Workers exposed to dust in certain workplace environments, such as mines, cotton mills, and grain-handling facilities, commonly develop symptoms of cough and sputum and may suffer permanent loss of lung function (Chapter 93). In some regions of the world, repeated exposure to biomass combustion in confined living quarters causes airflow obstruction. Current urban air

Normal

75

50

Average smoker

Disability Impaired lung growth Death

25

Susceptible smoker

0 0

20

40

60

80

Age (yr) FIGURE 88-1.  Lung growth occurs during childhood and adolescence, with the forced expiratory volume in 1 second (FEV1) reaching a maximum at about 25 years of age. Thereafter, the FEV1 steadily declines owing to normal aging effects. Lung function declines more rapidly in smokers, but the average effect is so small that clinically significant airflow obstruction would never develop. However, a proportion of “susceptible smokers” lose lung function much faster than the average, so they develop disabling chronic obstructive pulmonary disorder (COPD). If lung growth is impaired, lung function reserve is less as a young adult, and a susceptible smoker will develop disabling COPD at an earlier age.

556

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

pollution in economically advanced countries appears to have little effect on the prevalence of airflow obstruction, but this factor may be more important in heavily polluted urban centers in industrializing countries.

Respiratory Infections

Recurrent respiratory infections were once thought to be a major factor in the development of airflow obstruction, but longitudinal cohort studies have yielded inconclusive findings. An effect, if present, appears weak relative to cigarette smoking. Whether childhood respiratory infections leave residual effects on adult lung function is similarly unclear.

Airway Responsiveness

Acute bronchoconstriction after inhalation of dilute concentrations of methacholine or histamine, termed bronchial hyperresponsiveness (Chapter 87), is a defining feature of asthma but is also present in many COPD patients. Bronchial hyperresponsiveness independently predicts accelerated loss of lung function in persons with mild to moderate COPD, especially among persons who continue to smoke.

Genetic Factors

A severe deficiency of α1-antitrypsin is the only genetic risk factor proven to have a major impact on the development of COPD.5 This deficiency is found in about 1 to 2% of patients with an established diagnosis of COPD. α1-Antitrypsin, which is a serine protease inhibitor that is secreted into the circulation from the liver, is thought to protect lung tissue against digestion by neutrophil elastase and related serine proteinases that have been implicated in the pathogenesis of human emphysema. The most common allele at the α1-antitrypsin genetic locus is M, and MM hom*ozygotes have what are considered normal levels of α1-antitrypsin (100 to 300 mg/dL). Numerous variant alleles have been identified, but severe deficiency is most commonly found in persons who are hom*ozygous for the Z allele, in whom serum levels are generally less than 20 to 30% of the lower range of normal. Affected persons are very susceptible to cigarette smoke–induced damage and may develop severe COPD at a relatively early age. The risk for clinically important emphysema appears to be much less if patients with the risk alleles do not smoke. Emphysema associated with severe α1-antitrypsin deficiency is characteristically of the panacinar type with a predominant basal distribution. About 2 to 3% of northern European populations possess the MZ heterozygote serum and have α1-antitrypsin levels about half of normal. These individuals may be at greater risk for development of chronic airflow obstruction, but the magnitude of the effect, if present, appears to be quite small. Studies of family aggregations and of molecular genetics suggest some heritable risks beyond those associated with α1-antitrypsin deficiency. Women with severe COPD appear to have relatively more airway disease and less emphysema compared with men with similar airflow obstruction.

PATHOBIOLOGY

Pathology Emphysema

Emphysema is characterized by abnormal enlargement of the air spaces distal to the terminal bronchiole, with destruction of the alveolar walls but without obvious fibrosis. The terminal bronchiole, which is the most distal nonalveolated airway within the bronchial tree, supplies ventilation to a lung unit that is termed the acinus. Distal to the terminal bronchiole are two or three generations of partially alveolated respiratory bronchioles, and then the alveolar zone, where most gas exchange occurs. Air spaces may enlarge throughout the alveolated zone owing to destruction or rearrangement of their walls. Human emphysema consists of two major subtypes. Centriacinar emphysema localizes to the respiratory bronchioles just distal to the terminal bronchiole, whereas the remainder of the acinus is largely spared. Individual lesions, which may be up to 10 mm in diameter, tend to be more prominent in the upper lobe. Severe centriacinar emphysema is almost always related to cigarette smoking, but mild centriacinar emphysema can occur from other environmental exposures. Focal areas of inflammation, fibrosis, and carbonaceous pigment are commonly present in adjacent alveolar and bronchiolar walls. In panacinar emphysema, alveolar ducts are diffusely enlarged; adjacent alveoli may become effaced to the extent that individual units can no longer be identified. With progression of the disease, individual lesions can coalesce to form large bullae. Panacinar emphysema, which is typical of severe

α1-antitrypsin deficiency, also commonly occurs in patients in whom the major risk factor for COPD is cigarette smoking. Most patients with severe COPD appear to have mixed elements of centriacinar and panacinar emphysema, and individual subtypes cannot be reliably distinguished in advanced disease.

Chronic Bronchitis and Bronchiolitis

Mucous glands, located between the epithelial basem*nt membrane and the cartilage plates within the central bronchial tree, and goblet cells in the airway epithelium secrete mucus into the bronchial lumen to aid in host defenses. Enlargement of the bronchial mucous glands and expansion of the epithelial goblet cell population, which occur commonly in COPD, are correlated with clinical symptoms of cough and excess sputum production but not with airflow obstruction. A low-grade inflammatory response, consisting of neutrophils, macrophages, and CD8+ T lymphocytes, may also be seen in the cartilaginous airways of COPD patients. The principal sites of increased airflow resistance in COPD are the small distal airways that have an internal diameter near the lung’s functional residual capacity of less than 2 mm. The earliest pathologic changes identified in young cigarette smokers consist of focal collections of brown-pigmented macrophages in the respiratory bronchiole and a sparse infiltrate of neutrophils and lymphocytes in the walls of the terminal bronchiole. In older patients with established COPD, the inflammatory response is more intense, but still with a similar mix of neutrophils, macrophages, and lymphocytes. Other pathologic changes in the distal airways include fibrosis, goblet cell and squamous cell metaplasia of the lining epithelium, smooth muscle enlargement within the airway walls, and scattered regions of mucous plugging. Compared with normal subjects, distal airways in patients with COPD have thicker airway walls and smaller lumens.

Pulmonary Vasculature

Hypoxemia causes vasoconstriction in small pulmonary arteries and a consequent increase in pulmonary vascular resistance. Vascular remodeling in response to chronic hypoxemia results in irreversible pulmonary hypertension (Chapter 68). Medial smooth muscle enlargement and intimal fibrosis in small pulmonary arteries are the most important vascular changes. In addition, a substantial portion of the capillary bed may be destroyed by severe emphysema.

Pathogenesis

Emphysema appears to be caused by an elastase-antielastase imbalance in the lung due to either elastase excess or antielastase deficiency. Human lungs contain a rich network of elastin-containing fibers and other matrix proteins that confer structural integrity and elasticity to alveolar walls. Intratracheal instillation of proteinases, particularly those capable of hydrolyzing native elastin, induces lesions with morphologic and functional features of human emphysema in experimental animals. Chronic inflammation induced from cigarette smoke increases the burden of inflammatory cell–derived proteinases within lung parenchyma. Severe deficiency of α1-antitrypsin, a potent inhibitor of neutrophil elastase and other serine proteinases, is associated with development of severe panacinar emphysema in humans. In addition to neutrophil elastase, other neutrophil-derived serine proteinases, such as proteinase 3 and cathepsin G, and matrix metalloproteinases degrade elastin and other matrix components, including collagen, proteoglycans, and fibronectin. A macrophage-derived metalloproteinase, MMP-12, is essential to the development of cigarette smoke–induced emphysema in an animal model, and genetic studies show that a single-nucleotide polymorphism in the promoter region of the MMP-12 gene is associated with reduced risk of COPD in adult smokers. Relatively less is understood about the pathogenesis of distal airways disease. Particulate matter and toxic gases from inhaled cigarette smoke initiate an inflammatory response composed primarily of macrophages and neutrophils. This early inflammatory response may be mediated by the innate defense system as a response to cell injury. In more advanced disease, inflammation persists even after the patient has stopped smoking. At this stage, humoral and cellular components of the adaptive immune system may predominate, possibly in response to infection or specific antigens from other sources. Infiltration of airway walls with CD4+, CD8+, and B lymphocytes is a prominent feature of more advanced COPD. Repair from either type of immune response might cause airway remodeling by stimulating connective tissue matrix synthesis and smooth muscle formation and by increasing the proportion of mucus-secreting goblet cells within the epithelial layer.

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

Lung and Heart Mechanics

Elastic recoil refers to the lung’s intrinsic tendency to deflate after inflation. A dense labyrinth of elastic fibers and other matrix elements within the lung parenchyma, along with surface tension at the alveolar air-liquid interface, confers this important mechanical property. Elastic recoil maintains the patency of small airways through radial alveolar attachments, similar to the way a tent is held up by its guy ropes, and provides a portion of the driving pressure during expiration. Age-related loss of lung elasticity largely explains the normal decline in FEV1 with advancing age. In emphysema, loss of lung elastic recoil results from damage to elastic fibers and loss of alveolar surface area, with consequent airflow obstruction. An increase in bronchial airflow resistance is another sentinel feature of lung mechanics in COPD. The increased resistance in COPD is due primarily to narrowing and loss in airways of less than 2 mm in diameter, known as small airways, even before emphysematous destruction occurs. As a result, peripheral airflow resistance of COPD is higher than in normal lungs by an order of magnitude or more. In contrast, airflow resistance in the central airways of lungs from COPD patients differs little from that of normal lungs. One of the key physiologic aspects of COPD is limitation of expiratory airflow (Fig. 88-2) due to loss of lung elastic recoil and increased viscous resistance to airflow in the small airways (Chapter 85). The severity of emphysema and airflow obstruction is directly related to impaired left ventricular filling, reduced stroke volume, and lower cardiac output without reducing the ejection fraction.

Expiration

Maximal effort Tidal volume with exercise Tidal volume at rest

Inspiration

Flow rate

COPD Forced vital capacity 100

75

Residual volume 50

25

% Predicted total lung capacity

Gas Exchange

Mild hypoxemia may be detected in the early stages of COPD, and hypoxemia often becomes more prominent as airflow obstruction worsens. Hypercapnia usually appears only with severe COPD but is sometimes absent even in late-stage disease. Ventilation-perfusion mismatching, due to changes in both the airways and pulmonary vessels, is largely responsible for hypoxemia, with uneven ventilation being the primary event. Gas exchange is most efficient when the ratio of ventilation to perfusion is uniform in all lung regions. In COPD, there is “wasted ventilation” because some lung regions have inadequate pulmonary blood flow for the ventilation. The calculated A-a gradient for oxygen is larger than anticipated for the patient’s age (Chapters 85 and 103). Thus, in most cases of COPD, modest increases in the fraction of inspired oxygen result in a resolution of clinical hypoxemia.

CLINICAL MANIFESTATIONS

History and Physical Examination

COPD should be suspected in all adults who complain of chronic respiratory symptoms, particularly dyspnea (Chapter 83) that limits activities of daily living.6,7 Clinical features that increase the likelihood of COPD include older age, current or past cigarette use, insidious onset of dyspnea with slow progression, history of acute bronchitis for which medical care is sought, and symptoms of chronic cough, sputum production, or wheezing. Symptoms of cough and sputum may antedate dyspnea by many years. Some patients date the onset of dyspnea to a respiratory infection, but careful questioning usually elicits some history of impaired exercise tolerance before that event. Absence of cigarette smoking does not preclude a diagnosis of COPD because a few persons develop severe irreversible airflow obstruction without smoking history and even without known genetic predispositions. Some nonsmokers may relate a history of occupational dust or noxious gas exposure (Chapters 93 and 94), but in others no putative cause can be discerned. The physical examination findings are usually normal in patients with mild to moderate disease, and characteristic physical signs may be absent even in severe disease. Physical examination findings commonly present in severe COPD include the appearance of a barrel-shaped chest, low diaphragm detected by percussion, prolonged expiratory phase, and use of accessory muscles of respiration. Heart sounds are usually distant, and auscultation of the chest may reveal diminished breath sounds or a variety of rhonchi, wheezes, and rattles. Auscultatory wheezes may be prominent, particularly during exacerbations, but this physical sign does not reliably differentiate COPD from asthma. With severe hypoxemia, cyanosis may be clinically evident. Clubbing is not associated with COPD, and its presence should suggest another diagnosis. Pedal edema, distended jugular veins, and hepatic congestion are signs of pulmonary hypertension and cor pulmonale (Chapter 68). Patients with advanced COPD may be cachectic, with loss of muscle mass and subcutaneous fat.

Expiration

Clinical Phenotypes

Inspiration

Flow rate

Normal Forced vital capacity

Residual volume

FIGURE 88-2.  Inspiratory and expiratory flow-volume loops at rest, with exercise, and with maximal effort in a normal subject are compared with those in a patient with chronic obstructive pulmonary disorder (COPD). The normal subject can easily increase both tidal volume and breathing frequency to match the metabolic requirements of vigorous exercise. In contrast, the COPD patient exhibits maximal expiratory flow limitation even at rest and must breathe at larger lung volumes to optimize expiratory airflow. Lung hyperinflation requires greater respiratory work because the lung and chest wall become stiffer at larger volumes. This effect is accentuated during exercise, which causes end-expiratory lung volume to increase further. This phenomenon is described as dynamic hyperinflation and is an important mechanism in limiting exercise and causing dyspnea.

557

One of the more enduring efforts to categorize COPD into subtypes is the description of the patients as either “pink puffers” or “blue bloaters.”8 The pink puffer is described as a cachectic individual with unrelenting dyspnea, clinical and radiographic signs of severe lung hyperinflation, and normal or near-normal arterial blood gases at rest. Salient features of the blue bloater are a stout body habitus, chronic cough and sputum, less troubling dyspnea, and severe hypoxemia and hypercapnia resulting in polycythemia and signs of cor pulmonale. In the original description of these phenotypes, the pink puffer phenotype was equated with severe emphysema, whereas the blue bloater was thought to have predominant chronic bronchitis. Selected COPD patients do fit one or the other of these clinical subtypes, but most cannot be simply categorized. Limited information from clinical and pathologic correlative studies fails to show a consistent association of either clinical subtype with distinguishing pathologic features in lung parenchyma or airways. The blue bloater phenotype may now be less common, possibly because hypoxemia is recognized and treated earlier or because some COPD patients once described as blue bloaters may have had coexisting obstructive sleep apnea (Chapter 100). COPD patients with similar degrees of airflow obstruction vary greatly with respect to severity of dyspnea, impairment of exercise tolerance, frequency of exacerbations, body habitus, and severity of arterial blood gas disturbances. There is limited understanding about the mechanisms underlying these clinical characteristics.

558

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

DIAGNOSIS

Oximetry

Pulmonary Function Tests

Airflow obstruction can be determined best by spirometry. If the ratio of FEV1 to forced vital capacity (FEV1/FVC) is less than 0.70 (Chapter 85) after administration of an inhaled bronchodilator, an obstructive defect is present (Table 88-1). A large improvement of perhaps 30 to 40% in FEV1 after treatment with an inhaled bronchodilator may help identify a patient with predominant asthma, but the test otherwise has little clinical utility and cannot reliably identify patients who will benefit from any particular form of therapy. Lung volume measurements may help distinguish obstructive and restrictive lung diseases in selected patients, but they are unnecessary in most COPD patients. The diffusing capacity for carbon monoxide (Dlco; Chapter 85) measures the uptake of carbon monoxide between inspired air and the blood stream. Decreases in the Dlco reflect the loss of alveolar surface area that is available for gas transfer and roughly correspond to the severity of emphysema. However, the test provides information of no practical value in the customary management of COPD. After the diagnosis is established, follow-up spirometry may help determine whether worsening breathlessness is due to COPD, as indicated by a decrease in FEV1, or to another cause, such as heart failure (Chapter 58). However, repeated spirometry should not be used as a guide to drug therapy because the background variability of the measurement is large relative to treatment effects.

TABLE 88-1  SEVERITY OF AIRFLOW OBSTRUCTION IN COPD ACCORDING TO POSTBRONCHODILATOR SPIROMETRY STAGE AND SEVERITY

DEFINITION

I: Mild

FEV1/FVC < 0.70, FEV1 ≥ 80% of predicted

II: Moderate

FEV1/FVC < 0.70, 50% ≤ FEV1 < 80% of predicted

III: Severe

FEV1/FVC < 0.70, 30% ≤ FEV1 < 50% of predicted

IV: Very severe

FEV1/FVC < 0.70, FEV1 < 30% of predicted or FEV1 < 50% of predicted plus chronic respiratory failure

COPD = chronic obstructive pulmonary disease; FEV1 = forced expiratory volume in the first second; FVC = forced vital capacity. Data from Global Initiative for Chronic Obstructive Lung Disease. http://www.goldcopd.com.

A

Hypoxemia and hypercapnia become increasingly common as COPD worsens. Because treatment with supplemental oxygen improves mortality, patients with severe COPD should be tested for hypoxemia at regular intervals. Hypoxemia can be detected and quantified by oximetry or arterial blood gases. Oximetry is generally preferred because it is simpler, cheaper, and causes no discomfort. The added information from a set of arterial blood gases (Chapter 103) is most helpful in COPD patients with severe exacerbations.

Radiographic Studies

Common signs of severe COPD on a chest radiograph include hyperinflated lungs, flattened diaphragms, and increased retrosternal clear space (Fig. 88-3). The walls of large emphysematous bullae may be visualized as thin curvilinear lines, and severe emphysema may appear as regions of relative hyperlucency. Chest radiographs are usually normal in mild to moderate COPD and sometimes in severe COPD. Hence, a chest radiograph is not an adequate diagnostic test for COPD, and it is used mostly to exclude other pulmonary diseases. Chest computed tomography (CT), which is a superior imaging modality to assess the magnitude and distribution of emphysema (Fig. 88-4), is not helpful in the usual management of COPD.

Other Studies

Measurement of the serum level of α1-antitrypsin deficiency may be considered, particularly if the patient has a strong family history of COPD or if the onset of airflow obstruction occurs at an early age. If the α1-antitrypsin level is less than 20 to 30% of normal, further testing with specialized phenotyping and genotyping studies is required to confirm the diagnosis.

Differential Diagnosis

COPD is most commonly confused with asthma (Chapter 87), particularly in older patients. Clinical features that favor asthma over COPD include onset of disease at an early age, presence of atopy, lack of a smoking history, substantial variability of symptoms over time, and largely reversible airflow obstruction. However, new onset of asthma may occur in elderly people, some asthmatics smoke, an atopic history is not a requisite for development of asthma, and airflow obstruction may become fixed in patients with severe, long-standing asthma. Because treatment is much the same, distinguishing asthma from COPD may not be so important. Bronchiectasis (Chapter 90) is characterized by chronic inflammation and abnormal dilation of airways associated with chronic cough and expectora-

B

FIGURE 88-3.  Posteroanterior (A) and lateral (B) radiographs of the thorax in a patient with emphysema. The most obvious abnormalities are those associated with increased lung volume. The lungs appear dark because of their increased air relative to tissue. The diaphragms are caudal to their normal position and appear flatter than normal. The heart is oriented more vertically than normal because of caudal displacement of the diaphragm, and the transverse diameter of the rib cage is increased; as a result, the width of the heart relative to the rib cage on the posteroanterior view is decreased. The space between the sternum and heart and great vessels is increased on the lateral view.

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

FIGURE 88-4.  High-resolution axial computed tomography scan of a 1-mm section of the thorax of a patient with emphysema at the level of the tracheal carina. The right lung is on the left. Multiple large bullae—black holes—are evident. Many smaller areas of similar tissue destruction are also present in both lungs. The right upper lobe bronchus is seen entering the lung; its walls are thickened, suggesting chronic inflammation. (Courtesy Dr. Bruce Maycher.)

tion of purulent sputum. It can be distinguished from COPD with a predominant bronchitis component by chest CT imaging. Bronchiolitis obliterans is characterized by cicatricial narrowing of the distal airways with severe irreversible airflow obstruction. The condition may occur in association with collagen vascular diseases (Chapters 264 and 266) and is commonly seen after lung transplantation (Chapter 101). A similar disorder has been described with certain industrial inhalants, such as diacetyl, a butter-like flavoring manufactured for use with microwavable popcorn (Chapter 93). In nonsmokers, the diagnosis of bronchiolitis obliterans can be reliably inferred from the history, the presence of irreversible airflow obstruction, and the absence of emphysema or other explanatory conditions on chest CT images. Attribution of cause in smokers is more difficult because the airway disease with diacetyl exposure is similar to that found with cigarette smoke.

TREATMENT  Stable Disease

Smoking Cessation

Smoking cessation (Chapter 32) reduces symptoms of cough and sputum production in many patients with COPD, but it improves lung function to only a small extent. Most important, about a decade after smoking cessation, the rate of decline of FEV1 in patients with mild to moderate disease reverts to that seen in lifelong nonsmokers, thereby making it unlikely that these former smokers will ever develop severe COPD. Smoking cessation in patients with mild to moderate COPD also improves long-term mortality by reducing both respiratory and cardiovascular deaths. Smoking cessation probably also slows the decline of lung function in patients with more severe COPD. Limited information indicates that counseling and pharmacotherapy achieve the same low success rates for smoking cessation in COPD patients as in the general population (Chapter 32). COPD patients tend to quit smoking as the disease progresses, possibly because they have greater awareness of their disease or because cigarette smoke makes their respiratory symptoms worse. Sharing information about abnormal spirometry has not been shown to motivate patients to quit smoking.

Bronchodilators

Both β2-adrenergic agonists and anticholinergics are widely used to treat COPD (Table 88-2). Short-acting β2-adrenergic agonists, such as albuterol and the short-acting anticholinergic ipratropium bromide, can be administered

559

either by oral inhaler devices or by nebulization, with little objective superiority of one delivery device over the other if a spacer is used with oral inhaler devices. Longer-acting bronchodilators have largely replaced shorter-acting drugs, but a short-acting bronchodilator, such as albuterol, is still recommended for “rescue” or “as-needed” use in patients who experience bothersome dyspnea (Table 88-3). Inhaled long-acting β2-adrenergic agonist bronchodilators widely used for COPD include salmeterol and formoterol, administered by one inhalation twice daily, and indacaterol, administered by one inhalation once daily (Table 88-2). Inhaled anticholinergics include tiotropium (administered by one inhalation once daily) and aclidinium (administered by one inhalation twice daily). Tiotropium (18 µg once daily) appears to be superior to salmeterol (50 µg twice daily) for reducing exacerbations, A1  and combining a long-acting β2adrenergic agonist inhaler with a separate long-acting anticholinergic inhaler appears to be more effective than either agent alone. A2  Once-daily combination inhalers likely will soon be commercially available. Compared with placebo, each class of long-acting bronchodilators reduces exacerbation rates by about 15 to 20% in relative terms. Because the average patient with severe COPD has about one serious exacerbation per year, the number of patients that need to be treated to prevent one exacerbation is about six. Adverse symptomatic events of both classes of long-acting bronchodilators in COPD patients are generally minor. Theophylline is a poor bronchodilator that largely has been replaced with inhaled drugs, but its effect is additive when it is given along with inhaled bronchodilators. Theophylline may also reduce exacerbations. To be used effectively and safely, it should be started with an oral daily dose of between 150 and 300 mg and titrated to achieve serum levels of 8 to 12 µg/mL. Higher levels are poorly tolerated, especially in older patients. Theophylline interacts with numerous other drugs (e.g., allopurinol, diazepam, cimetidine, ciprofloxacin), and conditions such as heart failure and liver disease may reduce its elimination rates. Patients’ drug levels must be monitored on a regular basis, and toxic levels can develop even in a patient receiving a stable dose. Oral roflumilast, a phosphodiesterase 4 inhibitor at 500 µg once daily, can increase FEV1 by 50 mL and reduce moderate to severe exacerbations in patients with COPD and chronic bronchitis, even in patients already treated with tiotropium.

Corticosteroids

Inhaled corticosteroids produce marginal improvements in lung function and respiratory health status in COPD patients, and they reduce COPD exacerbation rates by about 15 to 20% in relative terms. A3  Inhaled corticosteroids combined with an inhaled long-acting β2-agonist provide added benefit over that seen with either monotherapy, but added benefit appears quite small when added to both a long-acting β2-agonist and a long-acting anticholinergic. A4  (Table 88-3). Multiple large trials have found little effect of inhaled corticosteroids in reducing FEV1 loss during periods of several years. The most common adverse effects of inhaled corticosteroids are dysphonia and upper airway thrush. Less commonly, they may predispose patients to pneumonia. Observational studies suggest that long-term inhaled corticosteroids may cause osteoporosis (Chapter 243) and cataracts (Chapter 423). A few COPD patients are prescribed systemic corticosteroids on a regular basis, usually in doses of 10 to 15 mg/day of prednisone or its equivalent. These patients are sometimes considered “prednisone dependent” because it is frequently difficult to wean them completely off drug. There are no proven benefits of chronic, low-dose prednisone in COPD, and adverse effects involving bone, eyes, and other organs are well documented (Chapter 35). Consequently, efforts should be made to reduce or to discontinue chronic systemic corticosteroids while optimizing other treatment.

Oxygen

Chronic hypoxemia in patients with COPD can induce irreversible pulmonary hypertension and cor pulmonale (Chapter 68). Long-term oxygen therapy extends life in patients who are persistently hypoxemic. The principal qualifying criteria are an arterial Pao2 of less than 56 mm Hg and an arterial oxygen saturation of less than 89%, both while breathing ambient air at rest in a stable clinical state. Patients should also be considered for home oxygen if their Pao2 is less than 60 mm Hg in the presence of right-sided heart failure or polycythemia. Treatment should consist of home oxygen to be used for at least 18 hours daily, to include sleep time. To determine an appropriate prescription, the oxygen flow rate should be adjusted in 1-L/minute increments at 15-minute intervals until the resting oxygen saturation remains above 90%. In qualifying patients, long-term oxygen may also decrease polycythemia and pulmonary hypertension and improve neuropsychiatric function. Oxygen does not improve mortality in patients with similarly severe airflow obstruction but milder hypoxemia. Many patients with severe COPD may be normoxemic at rest while breathing ambient air but exhibit oxygen desaturation with exercise. Physicians commonly prescribe ambulatory oxygen in this setting, with the expectation that it will improve exercise tolerance and increase daily activity. Ambulatory oxygen modestly improves exercise endurance for such patients in a laboratory setting, but efforts to show benefit during activities of daily living have

560

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

TABLE 88-2  COMMONLY USED MEDICATIONS FOR STABLE CHRONIC OBSTRUCTIVE PULMONARY DISEASE (COPD) MODE OF DELIVERY

DOSE AND FREQUENCY

POSSIBLE ADVERSE REACTIONS

SHORT-ACTING INHALED BRONCHODILATORS Albuterol (β2-adrenergic agonist)

Inhaler

Ipratropium (anticholinergic)

Inhaler

Albuterol/ipratropium

Inhaler device

Nebulizer

Nebulizer

Nebulizer

100 µg per inhalation; 1-2 inhalations every 4-6 hours, as needed 2.5 mg; every 4-6 hours, as needed

Palpitations, tachycardia, tremor, hypersensitivity reaction

17 µg per inhalation; 2 inhalations 4 times daily, up to 12 inhalations a day 0.5 mg; every 6-8 hours

Dry mouth, cough, blurred vision, hypersensitivity reaction

90 µg/18 µg per inhalation; 2 inhalations 4 times daily, up to 12 inhalations per day 2.5 mg/0.5 mg; 4 times daily, up to 2 additional doses daily

All those occurring with either albuterol or ipratropium

LONG-ACTING INHALED BRONCHODILATORS Formoterol (β2-adrenergic agonist)

Inhaler Nebulizer

12 µg; 1 inhalation twice daily 20 µg; twice daily

Dizziness, tremor, throat irritation, hypersensitivity reaction

Salmeterol (β2-adrenergic agonist)

Inhaler

50 µg; 1 inhalation twice daily

Headache, tremor, throat irritation, hypersensitivity reaction

Indacaterol (β2-adrenergic agonist)

Inhaler

75 µg; 1 inhalation daily

Cough, oropharyngeal pain, nasopharyngitis, headache, nausea, hypersensitivity reaction

Tiotropium (anticholinergic)

Inhaler

18 µg; 1 inhalation each morning

Dry mouth, urinary retention, symptoms of narrow-angle glaucoma, hypersensitivity reaction

Aclidinium (anticholinergic)

Inhaler

400 µg; 1 inhalation twice daily

Same as for tiotropium

Fluticasone powder

Inhaler

250 µg; 1-2 inhalations twice daily

Sore throat, dysphonia, headache, hypersensitivity reaction

Budesonide

Inhaler

160 µg; 1-2 inhalations twice daily

Nasopharyngitis, thrush, hypersensitivity reactions

Fluticasone/salmeterol

Inhaler

250 µg/50 µg; 1 inhalation twice daily

All those occurring with either fluticasone or salmeterol

Budesonide/formoterol

Inhaler

160 µg/4.5 µg; 2 inhalations twice daily

All those occurring with either budesonide or formoterol

Theophylline (24-hour sustained release)

Pill

200-800 mg, once daily; start with daily dose of 150-300 mg and titrate to blood level of 8-12 µg/mL

Nausea and vomiting, seizures, tremor, insomnia, multifocal atrial tachyarrhythmia, hypersensitivity reaction

Roflumilast

Pill

500 µg, once daily

Depression, suicidal thought, insomnia, loss of appetite, weight loss, diarrhea

INHALED CORTICOSTEROIDS

COMBINATION INHALERS

ORAL DRUGS

TABLE 88-3  GUIDELINE RECOMMENDATIONS FOR DIAGNOSIS AND MANAGEMENT OF STABLE COPD Spirometry should be obtained to diagnose airflow obstruction in patients with respiratory symptoms. Spirometry should not be used to screen for airflow obstruction in individuals without respiratory symptoms. For stable COPD patients with respiratory symptoms and FEV1 between 60% and 80% of predicted, treatment with inhaled bronchodilators may be used. Stable COPD patients with respiratory symptoms and FEV1 < 60% should be treated with inhaled bronchodilators. Clinicians should prescribe monotherapy with either long-acting inhaled anticholinergics or long-acting inhaled β-agonists for symptomatic patients with COPD and FEV1 < 60% predicted. Clinicians should base the choice of specific monotherapy on the patient’s preference, the cost, and the adverse effect profile. Clinicians may administer combination inhaled therapies (long-acting inhaled anticholinergics, long-acting inhaled β-agonists, or inhaled corticosteroids) for symptomatic patients with stable COPD and FEV1 < 60% predicted. Clinicians should prescribe pulmonary rehabilitation for symptomatic patients with an FEV1 < 50% predicted. Clinicians may consider pulmonary rehabilitation for symptomatic or exercise-limited patients with an FEV1 > 50% predicted. Clinicians should prescribe continuous oxygen therapy in patients with COPD who have severe resting hypoxemia (arterial oxygen partial pressure ≤ 55 mm Hg or arterial oxygen saturation ≤ 88%) COPD = chronic obstructive pulmonary disease; FEV1 = forced expiratory volume in 1 second. Modified from Qaseem A, Wilt TJ, Weinberger SE, et al. Diagnosis and management of stable chronic obstructive pulmonary disease: a clinical practice guideline update from the American College of Physicians, American College of Chest Physicians, American Thoracic Society, and European Respiratory Society. Ann Intern Med. 2011;155:179-191.

been mostly unsuccessful. A5  Even normoxemic COPD patients with isolated nocturnal hypoxemia have not been shown to benefit from oxygen therapy.

Immunizations and Prophylactic Antibiotics

An annual influenza vaccination (Chapter 18) is recommended for all patients with COPD, although few trials have targeted this population of patients. Observational studies suggest that influenza vaccination substantially reduces hospitalization and mortality rates in COPD patients. Polysaccharide pneumococcal vaccination (Chapter 18) is also recommended, although supporting evidence is weak. Chronic prophylactic macrolide use (e.g., azithromycin 250 mg daily) in addition to regular treatment can decrease exacerbations and improve quality of life in patients with COPD, A6  but such treatment may cause hearing loss and may not be safe if patients have a prolonged QTc interval or if they are taking other drugs known to prolong the QTc interval.

Pulmonary Rehabilitation

COPD patients become increasingly sedentary as their disease progresses. Lack of physical activity causes muscle and cardiovascular deconditioning, which further complicates the ability to perform routine tasks. The principal goal of pulmonary rehabilitation is to reverse this process with a program of exercise endurance training. Educational and behavior modification elements are usually included in an effort to improve coping skills and psychological functioning. Most programs are hospital based and consist of 3- to 4-hour sessions, three times a week, during a 6- to 12-week period. Patients who become breathless with minimal activity or who have exercise-limiting comorbidities are not suitable candidates. Numerous randomized, controlled trials have shown that pulmonary  rehabilitation confers substantial improvements in respiratory health status and in walking distance and possibly in reduction of health care use. A7  Unfortunately, the benefits of pulmonary rehabilitation erode rapidly in the 

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

absence of a continuation plan after completion of the initial program. Pulmonary rehabilitation is also not accessible to most patients for a variety of reasons.

Surgical Options

In lung volume reduction surgery (Chapter 101), severely emphysematous tissue is resected from the upper lobes of both lungs to permit less diseased portions of unresected lung to expand and to function more normally. Patients who are severely disabled from COPD and who have no other major comorbid conditions may be candidates for this procedure if CT imaging shows that severe emphysema is mostly localized to the upper lobes. Compared with controls receiving no surgical treatment, lung volume reduction surgery improves lung function, exercise capacity, and respiratory health status in COPD patients with severe emphysema, but there is no mortality benefit from this procedure, with a possible exception in the subset of patients with both predominant upper lobe emphysema and low exercise capacity. A8  Procedures that deflate severely emphysematous regions of the lung by endoscopic placement of one-way bronchial valves or of transbronchial stents or that completely ablate bronchi that subtend regions of severe emphysema, with either thermal injury or biologic sealants, have not yielded clear clinical benefits and are not recommended (Chapter 101). Lung transplantation is an option for patients who are severely incapacitated from COPD and have no major comorbid conditions (Chapter 101). Median survival after lung transplantation is only about 5 years, primarily because of the development of bronchiolitis obliterans, a form of chronic graft rejection causing severe airflow obstruction in the peripheral airways. It is unclear whether lung transplantation extends survival in patients with COPD, but patients who are fortunate enough to avoid complications are able to resume normal daily activities.

Exacerbations

Exacerbations represent an important element in the natural history of COPD.9 An exacerbation is defined as some combination of dyspnea, cough, and productive sputum, each of which has worsened from the stable state or has newly appeared. Exacerbations may also be associated with symptoms of rhinorrhea, sore throat, fever, and chest congestion. A symptom-based clinical event, as described previously, coupled with administration of an antibiotic or a systemic corticosteroid or admission to a hospital, is a definition that has been widely used in clinical trials. Patients with severe COPD experience an average of about one such exacerbation per year along with additional milder exacerbations that meet the symptomatic definition but do not require a medical intervention. Exacerbations are acute in onset, but recovery may require several weeks. Severe exacerbations have a major adverse impact on health status and may cause

561

permanent loss of lung function. Hospitalization for exacerbations consumes more than half of total medical costs for COPD. For poorly understood reasons, some patients suffer frequent exacerbations, whereas others have very few, despite similar degrees of airflow obstruction. Independent risk factors include low lung function, older age, history of frequent exacerbations, elevated blood levels of inflammatory biomarkers,10 and prior hospitalizations as well as the presence of a productive cough, gastroesophageal reflux, and cardiovascular comorbidities. Respiratory infections are thought to cause most exacerbations, although many of these implicated microorganisms may be recovered from sputum during periods of stable disease. Bacteria commonly implicated include Haemophilus influenzae (Chapter 300), Streptococcus pneumoniae (Chapter 289), and Moraxella catarrhalis (Chapter 300). Pseudomonas aeruginosa (Chapter 306) and enteric gram-negative bacilli (Chapters 304 and 305) are less common but are seen in patients with very severe COPD who were recently hospitalized or intubated. Putative viral pathogens include rhinoviruses (Chapter 361), influenza (Chapter 364), parainfluenza (Chapter 363), and respiratory syncytial virus (Chapter 362). Periods of increased airborne pollution with diesel particulates, sulfur dioxide, ozone, and nitrogen dioxide are associated with more COPD hospitalizations, but no cause can be assigned to many exacerbations. Evaluation and management of a patient with a suspected exacerbation vary according to severity. Mild exacerbations encountered in an office setting can be diagnosed and treated on the basis of a brief history and physical examination. Patients seen in emergency department or hospital settings generally are sicker and require a more extensive evaluation (Table 88-4). A chest radiograph should be obtained to look for signs of pneumonia (Chapter 97), pneumothorax (Chapter 99), and heart failure (Chapter 58). If pulmonary embolism (Chapter 98) is suspected, spiral CT of the chest is the test of choice. Arterial blood gases should be measured if there is any suspicion of hypercapnia because this information influences subsequent therapy. Sputum cultures need not be done routinely because they are unproven guides to antibiotic therapy. During seasonal outbreaks of influenza (Chapter 364), type A and B viruses can be identified with rapid commercially available polymerase chain reaction assays having a sensitivity of greater than 90%. These tests should not be relied on to withhold antiviral therapy if the patient is severely ill or if there is a strong clinical suspicion of influenza. Cardiac disease is a common comorbidity in COPD patients, and distinguishing a COPD exacerbation from left ventricular failure (Chapter 58) by history and physical examination alone is often problematic. Dyspnea (Chapter 83) is common to both conditions. Peripheral edema (Chapter 51) and elevated jugular venous pressure may occur with either left ventricular failure or cor pulmonale secondary to COPD. Echocardiography (Chapter 55) and serum brain natriuretic peptide (BNP) levels (Chapter 58) are useful in this clinical setting, although echocardiography is more difficult to perform in patients

TABLE 88-4  GUIDELINE RECOMMENDATIONS FOR HOSPITAL MANAGEMENT OF COPD EXACERBATIONS GLOBAL INITIATIVE FOR CHRONIC OBSTRUCTIVE LUNG DISEASE*

NATIONAL INSTITUTE FOR CLINICAL EXCELLENCE†

Date of statement

2013

2010

Diagnostic testing

Chest radiograph, oximetry, ABGs, and ECG Other testing as warranted by clinical indication

Chest radiograph, ABGs, ECG, complete blood count, sputum smear and culture, blood cultures if febrile

Bronchodilator therapy

Inhaled short-acting β2-agonist is recommended Consider ipratropium if inadequate clinical response Consider theophylline or aminophylline as second-line intravenous therapy

Administer inhaled drugs by nebulizer or hand-held inhaler with spacer device Specific agents and dosing regimens not specified Consider theophylline if inadequate response to inhaled bronchodilators

Antibiotics (see text for dosing)

Recommended if (1) increases in dyspnea, sputum volume, and sputum purulence all are present; (2) increase in sputum purulence along with increase in either dyspnea or sputum volume; or (3) need for assisted ventilation Initial empirical therapy with aminopenicillin with or without clavulanic acid, macrolide, or tetracycline, based on local bacterial resistance patterns Subsequent therapy based on sputum and blood cultures

Administer only if the patient has a history of purulent sputum Initiate with an aminopenicillin, a macrolide, or a tetracycline, taking into account local bacterial resistance patterns Adjust therapy according to sputum and blood cultures

Systemic corticosteroids

Daily prednisolone 30-40 mg (or its equivalent) for 10-14 days

Daily prednisolone 30 mg (or its equivalent) orally for 7-14 days

Supplemental oxygen

Maintain oxygen saturation 88-92% Monitor ABGs for hypercapnia and acidosis

Maintain oxygen saturation within the individualized target range Monitor ABGs

Assisted ventilation

Indications for NPPV include respiratory acidemia (arterial pH ≤ 7.35) or severe dyspnea with clinical signs of respiratory muscle fatigue or increased work of breathing

NPPV is the treatment of choice for persistent hypercapnic respiratory failure Consider functional status, body mass index, home oxygen, comorbidities, prior ICU admissions, age, and FEV1 when assessing suitability for intubation and ventilation

*Data from http://www.goldcopd.com. † Data from http://www.nice.org.uk. ABGs = arterial blood gases; COPD = chronic obstructive pulmonary disease; ECG = electrocardiogram; ICU = intensive care unit; NPPV = noninvasive positive-pressure ventilation.

with severe COPD. BNP levels may be modestly elevated in both stable and exacerbated COPD in the absence of left ventricular dysfunction. A normal BNP level excludes a diagnosis of left-sided heart failure with a high level of confidence, but an elevated level does not confirm its presence unless it is markedly elevated. Decisions about the need for hospitalization rely mostly on clinical judgment because there are no well-validated guidelines. Clinical assessment should consider intensity of dyspnea, use of accessory muscles of respiration, arterial blood gas disturbances, hemodynamic stability, and mental alertness. Guideline recommendations (see Table 88-4) for treatment of patients  hospitalized for COPD exacerbations emphasize that antibiotics hasten recovery. A9  Antibiotics are most effective when cough and purulent sputum are present, but there are no well-validated methods for determining which patients should be treated. If patients are sufficiently ill to seek medical attention for an exacerbation, most should probably receive an antibiotic. Most randomized placebo-controlled trials evaluated first-generation antibiotics, such as amoxicillin, trimethoprim-sulfamethoxazole, and tetracyclines, and it is unclear whether newer classes of antibiotics, such as macrolides and fluoroquinolones, are more effective. Choice of an antibiotic should be made with considerations to cost, safety, and local patterns of antibiotic resistance among the bacterial species commonly isolated from sputa during exacerbations. Doxycycline, 100 mg twice daily for 7 to 10 days, or trimethoprimsulfamethoxazole, 160/800 mg twice daily for 7 to 10 days, would be reasonable choices for initial therapy in many locales. Systemic corticosteroids improve lung function, shorten the recovery period, and prevent relapse when given to patients who are hospitalized  or present to an emergency department with a COPD exacerbation. Severely symptomatic patients seen in other clinical settings are also likely to benefit. Prednisone, 40 mg once daily for 5 days, is appropriate for most patients. A10  Longer courses of systemic corticosteroid therapy are strongly discouraged because they are no more effective and they increase the  likelihood of adverse effects. Parenteral corticosteroids should be given  only if gastrointestinal absorption is thought to be impaired. The major adverse effect of systemic corticosteroids is transient hyperglycemia, which may require treatment, particularly in patients with known diabetes mellitus (Chapter 229). Patients should be encouraged to increase their use of short-acting bronchodilators during outpatient treatment of an exacerbation. For hospitalized patients, a short-acting bronchodilator should be administered on a regular schedule, every 4 to 6 hours and more frequently as needed. Anticholinergic and β2-agonist agents are similarly effective, and a few small trials found no significant additive effect during exacerbations. Some patients express a preference for a nebulizer delivery system, although equivalent objective results can be achieved when inhalers are used with a spacer. Sufficient oxygen should be provided to maintain arterial oxygen saturations just above 90%, usually with oxygen flow rates of 2 to 3 L/minute delivered through a nasal cannula. Even at low flow rates, oxygen therapy can be expected to increase Paco2 by an average of about 5 to 10 mm Hg in patients with chronic hypercapnia. It is prudent to use the lowest flow of oxygen that achieves the desired result. If oxygen is prescribed for hypoxemia during an exacerbation, it is important to retest the patient several weeks later after recovery to determine when long-term oxygen is needed. The introduction of noninvasive positive-pressure ventilation (NIPPV) has significantly improved the care of patients with severe COPD exacerbations who have respiratory failure. A11  With NIPPV, the patient wears a tightly fitting nasal or full facial mask that is attached to a positive-pressure ventilator, avoiding the need for an endotracheal tube or a tracheostomy (Chapter 105). Compared with usual care, treatment with NIPPV is associated with fewer intubations, a shorter hospital stay, and improved all-cause mortality.

PROGNOSIS

About two-thirds of patients have progressive disease.11 Severe COPD is associated with excess mortality, and lung function, usually expressed as the percentage of predicted FEV1, is the single strongest predictor of death. Patients with COPD have variable rates of decline in FEV1, with more rapid average rates in smokers than in former smokers, but spirometry repeated at intervals of 1 year or more provides only limited information about prognosis. Only about half of patients with an FEV1 that is about 40% of predicted will survive 5 years. The severity of emphysema by CT or carbon monoxide diffusion is independently associated with a rapid annual decline in FEV1. Additional risk factors include the severity of dyspnea, weight loss, limited walking distance, hospitalization for exacerbation, hypoxemia, hypercapnia, and impaired quality of life. The development of bronchiectasis is independently associated with an increased risk of all-cause mortality in patients with moderate to severe COPD.12 The only interventions shown to reduce mortality are smoking cessation in patients with mild to moderate COPD, long-term

oxygen therapy for the subset of patients with chronic hypoxemia, and NIPPV in selected patients who are hospitalized for respiratory failure.

Grade A References A1. Vogelmeier C, Hederer B, Glaab T, et al. Tiotropium versus salmeterol for the prevention of exacerbations of COPD. N Engl J Med. 2011;364:1093-1103. A2. Wedzicha JA, Decramer M, Ficker JH, et al. Analysis of chronic obstructive pulmonary disease exacerbations with the dual bronchodilator QVA149 compared with glycopyrronium and tiotropium (SPARK): a randomised, double-blind, parallel-group study. Lancet Respir Med. 2013;1:199-209. A3. Yang IA, Clarke MS, Sim EHA, et al. Inhaled corticosteroids for stable chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2012;7:CD002991. A4. Magnussen H, Disse B, Rodriguez-Roisin R, et al. Withdrawal of inhaled glucocorticoids and exacerbations of COPD. N Engl J Med. 2014;371:1285-1294. A5. Abernethy AP, McDonald CF, Frith PA, et al. Effect of palliative oxygen versus room air in relief of breathlessness in patients with refractory dyspnoea: a double-blind, randomised controlled trial. Lancet. 2010;376:784-793. A6. Herath SC, Poole P. Prophylactic antibiotic therapy in chronic obstructive pulmonary disease. JAMA. 2014;311:2225-2226. A7. COPD Working Group. Pulmonary rehabilitation for patients with chronic pulmonary disease (COPD): an evidence based analysis. Ont Health Technol Assess Ser. 2012;12:1-75. A8. Shah PL, Slebos DJ, Cardoso PF, et al. Bronchoscopic lung-volume reduction with Exhale airway stents for emphysema (EASE trial): randomised, sham-controlled, multicentre trial. Lancet. 2011;378:997-1005. A9. Vollenweider DJ, Jarrett H, Steurer-Stey CA, et al. Antibiotics for exacerbations of chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2012;12:CD010257. A10. Leuppi JD, Schuetz P, Bingisser R, et al. Short-term vs conventional glucocorticoid therapy in acute exacerbations of chronic obstructive pulmonary disease. The REDUCE randomized clinical trial. JAMA. 2013;309:2223-2231. A11. McCurdy BR. Noninvasive positive pressure ventilation for acute respiratory failure patients with chronic obstructive pulmonary disease (COPD): an evidence-based analysis. Ont Health Technol Assess Ser. 2012;12:1-102.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

GENERAL REFERENCES 1. Thun MJ, Carter BD, Feskanich D, et al. 50-year trends in smoking-related mortality in the United States. N Engl J Med. 2013;368:351-364. 2. Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380:2095-2128. 3. Vos T, Flaxman AD, Naghavi M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380:2163-2196. 4. Gan WQ, FitzGerald JM, Carlsten C, et al. Associations of ambient air pollution with chronic obstructive pulmonary disease hospitalization and mortality. Am J Respir Crit Care Med. 2013;187: 721-727. 5. Stoller JK, Aboussouan LS. A review of α1-antitrypsin deficiency. Am J Respir Crit Care Med. 2012; 185:246-259. 6. Niewoehner DE. Clinical practice: outpatient management of severe COPD. N Engl J Med. 2010; 362:1407-1416.

562.e1

7. Vestbo J, Hurd SS, Agusti AG, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med. 2013;187:347-365. 8. Kim V, Criner GJ. Chronic bronchitis and chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2013;187:228-237. 9. Criner GJ, Bourbeau J, Diekemper RL, et al. Prevention of acute exacerbations of chronic obstructive pulmonary disease: American College of Chest Physicians and Canadian Thoracic Society Guideline. Chest. 2014; [Epub ahead of print]. 10. Thomsen M, Ingebrigtsen TS, Marott JL, et al. Inflammatory biomarkers and exacerbations in chronic obstructive pulmonary disease. JAMA. 2013;309:2353-2363. 11. Vestbo J, Agusti A, Wouters EF, et al. Should we view chronic obstructive pulmonary disease differently after ECLIPSE? A clinical perspective from the study team. Am J Respir Crit Care Med. 2014;189:1022-1030. 12. Martinez-Garcia MA, de la Rosa Carrillo D, Soler-Cataluna JJ, et al. Prognostic value of bronchiectasis in patients with moderate-to-severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2013;187:823-831.

562.e2

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

REVIEW QUESTIONS 1. A 54-year-old man seeks medical attention because of increasing exercise intolerance. He works as a construction worker and finds it increasingly difficult to complete the assigned work. He started smoking at the age of 18 years and currently smokes about one pack per day. In addition to dyspnea, he also has daily cough and sputum and admits to frequent “chest colds.” A diagnosis of COPD is strongly suspected. Which of the following would constitute the most appropriate initial assessment? A. History and physical examination; spirometry; chest computed tomography; oximetry B. History and physical examination; spirometry; electrocardiogram; oximetry C. History and physical examination; spirometry; diffusing capacity for carbon monoxide; arterial blood gases D. History and physical examination; spirometry; chest radiograph; oximetry E. History and physical examination; lung volumes; chest radiograph; arterial blood gases Answer: D  On the basis of the history, a diagnosis of COPD is likely in this patient. Clinical suspicion must always be confirmed by spirometry to demonstrate airflow obstruction and, if it is present, to obtain information about severity. Measurements of lung volumes or the diffusing capacity for carbon monoxide are more complex than spirometry and provide little additional clinically useful information. Chest computed tomography is superior to a chest radiograph in defining the presence and severity of emphysema, but this information presently has no practical implications for managing most COPD patients. Assessment of need for supplemental oxygen can be done with oximetry or arterial blood gases. For patients with stable disease, oximetry is preferred because it is simple and noninvasive. 2. A 63-year-old woman with severe COPD confirmed by spirometry (FEV1/FVC, 0.53; FEV1, 37% of predicted) was admitted to the hospital with an exacerbation. At time of discharge, her arterial oxygen saturation was 81% while breathing ambient air at rest. She was given a prescription for home oxygen. At a follow-up visit 3 months later, the patient stated that she had fully recovered from the exacerbation and was able to resume all of her usual activities. Her arterial oxygen saturation at this time was 86%, again while breathing ambient air at rest. Which of the following is the most appropriate recommendation? A. Discontinue home oxygen as the patient has fully recovered clinically B. Continue oxygen long term, titrate oxygen flow rate to ensure an arterial oxygen saturation of 90% or higher, and instruct the patient to use oxygen at least 18 hours daily C. Continue home oxygen with instructions that it be used only if the patient feels more breathless D. Use oxygen only during sleep and with exercise E. Prescribe oxygen at a flow rate of 2 L/min with instructions to use it at least 12 hours daily Answer: B  This patient has severe COPD with persistent hypoxemia after recovery from a severe exacerbation. Long-term oxygen should be prescribed because it improves all-cause mortality in such patients, and 18 hours of daily use is better than 12 hours. Long-term oxygen may confer other benefits, such as improved neurocognitive function and reduced risk of pulmonary hypertension. When long-term oxygen is first prescribed, the flow rate should be titrated to achieve an arterial oxygen saturation of at least 90%. In patients who do not qualify for long-term oxygen but experience modest desaturation with exercise, oxygen therapy has not been shown to decrease breathlessness or to increase daily activity.

3. A 68-year-old man with long-standing, severe COPD presents to an emergency department with a several-day history of increasing breathlessness, coupled with worsening cough and purulent sputum. After a brief evaluation, it is determined that hospitalization is required. The admitting physician obtains a history and physical examination, which reveals a severely breathless patient with audible expiratory wheezing. A chest radiograph shows hyperinflation and a focal pneumonic infiltrate. An electrocardiogram is unremarkable except for a sinus tachycardia. Routine laboratory test results are normal, and arterial blood gas analysis on ambient air shows a Po2 of 48 mm Hg, a Pco2 of 64 mm Hg, and a pH of 7.18. The patient is capable of swallowing oral medications. Which of the following would constitute the most appropriate treatment? A. Respiratory antibiotic appropriate to the locale; prednisone given orally (40 mg daily for 5 days); short-acting bronchodilator by nebulizer every 4 to 6 hours; supplemental oxygen sufficient to raise saturation to 90%; noninvasive positive-pressure ventilation B. Respiratory antibiotic appropriate to the locale; prednisone given orally (40 mg daily for 5 days); short-acting bronchodilator by nebulizer every 4 to 6 hours; supplemental oxygen sufficient to raise saturation to 90%; intubation and mechanical ventilation C. Respiratory antibiotic appropriate to the locale; prednisone given orally (40 mg daily for 5 days); short-acting bronchodilator by nebulizer every 4 to 6 hours; intravenous theophylline; supplemental oxygen sufficient to raise saturation to 90%; noninvasive positivepressure ventilation D. Respiratory antibiotic pending results of sputum culture; prednisone given orally (40 mg daily for 5 days); short-acting bronchodilator by nebulizer every 4 to 6 hours; supplemental oxygen sufficient to raise saturation to 90%; noninvasive positive-pressure ventilation E. Respiratory antibiotic appropriate to the locale; methylprednisolone given intravenously (30 mg four times daily for 10 days); short-acting bronchodilator by nebulizer every 4 to 6 hours; supplemental oxygen sufficient to raise saturation to 90%; intubation and mechanical ventilation Answer: A  Although never tested in rigorous randomized clinical trials, there is a consensus that regularly scheduled short-acting bronchodilators and controlled supplemental oxygen are beneficial for hospitalized patients with COPD. Antibiotics appear to be most beneficial for COPD exacerbations that result in hospitalization. No single antibiotic is proven to be superior to any other, and the choice should be based primarily on local bacterial resistance patterns. There is very good evidence that systemic corticosteroids improve clinical outcomes in this setting as well as good evidence that short courses of oral prednisone are as effective as longer courses. Several small, randomized clinical trials failed to show any benefit from theophylline in hospitalized COPD patients, and gastrointestinal side effects can be troublesome. Strong evidence supports the use of noninvasive positive-pressure ventilation, rather than intubation and mechanical ventilation, for COPD patients with mild to moderate respiratory failure.

CHAPTER 88  Chronic Obstructive Pulmonary Disease  

4. A 53-year-old woman with severe COPD confirmed by spirometry is seen in the office because of progressive dyspnea with exertion to the point that she could walk only one block on the level before having to stop and catch her breath. She has regularly smoked cigarettes continuously since she was 18 years old and still smokes half a pack per day. She made two visits to emergency departments in the previous year for “bronchitis,” and on each visit she had been given an antibiotic but no other respiratory medications. She has no known history of cardiac disease. Examination of the chest reveals signs of lung hyperinflation with decreased breath sounds. The chest radiograph shows signs of hyperinflation. Spirometry results include an FEV1/FVC of 0.53 and an FEV1 of 37% of predicted, thus confirming the diagnosis of severe COPD. Oxygen saturation at rest on ambient air is 90%. Which of the following would constitute the most appropriate management plan for this woman with newly diagnosed severe, exacerbation-prone COPD? A. Smoking cessation; short-acting bronchodilator; long-acting β2adrenergic bronchodilator or long-acting anticholinergic bronchodilator B. Smoking cessation; short-acting bronchodilator; long-acting β2adrenergic or long-acting anticholinergic bronchodilator; theophylline; pulmonary rehabilitation C. Smoking cessation; short-acting bronchodilator; combination inhaled therapy with a long-acting β2-adrenergic bronchodilator, a long-acting anticholinergic bronchodilator, or a glucocorticosteroid; low-dose prednisone (10-15 mg daily) D. Smoking cessation; short-acting bronchodilator; combination inhaled therapy with at least two of the following: long-acting β2-adrenergic bronchodilator, long-acting anticholinergic bronchodilator, inhaled glucocorticosteroid; pulmonary rehabilitation E. Smoking cessation; short-acting bronchodilator; long-acting β2adrenergic or anticholinergic bronchodilator; daily azithromycin; ambulatory oxygen Answer: D  This woman with newly diagnosed COPD has severe airflow obstruction by spirometric criteria, a pronounced limitation of exercise, and a history of two exacerbations in the past year. Smoking cessation is the most important element in her long-term care. The severity of her disease would justify combination therapy including at least two of these three classes of inhaled drugs: long-acting β2-adrenergic bronchodilators, long-acting anticholinergic bronchodilators, or inhaled glucocorticosteroids. Theophylline and daily azithromycin are not considered first-line treatment of COPD. Chronic prednisone is not known to be clinically effective in patients with stable COPD but is known to cause harm. Assuming that the patient is motivated and that a program is available, pulmonary rehabilitation is recommended for all patients who have exercise-limiting COPD.

562.e3

5. A 58-year-old woman with severe COPD has become increasingly incapacitated from her lung disease to the point that she is able to walk only about a half-block. She was once very physically active but has had to curtail most of these activities. She is compliant with her medications (including an inhaled short-acting bronchodilator, an inhaled long-acting bronchodilator, an inhaled corticosteroid, and theophylline), but she finds that these medications have had only modest effect on her exercise tolerance. She quit smoking 10 years ago and has no comorbid conditions that should significantly limit her physical activities. Her arterial oxygen saturation is 92% while at rest and breathing ambient air, but it decreases to 86% after a brisk walk down the hallway. Which of the following is the most appropriate intervention at this stage of the disease? A. Ambulatory oxygen B. Lung volume reduction surgery C. Lung transplantation D. A pulmonary vasodilating drug E. Pulmonary rehabilitation Answer: E  Assuming that the patient is motivated, has no other severe disabling comorbid conditions, and has access to a program, pulmonary rehabilitation has been shown to increase walking distance and respiratory health status substantially. Most rehabilitation programs are only 6 to 12 weeks in length, and the clinical benefits that accrue during that period erode during the ensuing year unless a longer-term program is put in place. Lung volume reduction surgery or a lung transplant may be considered at a later stage of her disease if all other interventions fail. The best available evidence raises serious doubts that ambulatory oxygen confers significant clinical benefits when it is prescribed for isolated exercise-induced hypoxemia. Studies of limited size have failed to show that pulmonary vasodilators are of any clinical benefit in COPD.

562

CHAPTER 89  Cystic Fibrosis  

89  CYSTIC FIBROSIS FRANK J. ACCURSO

DEFINITION

Cystic fibrosis is an autosomal recessive disease caused by mutations in the gene that encodes the cystic fibrosis transmembrane conductance regulator (CFTR) protein, which is a membrane protein that regulates ion flux at epithelial surfaces. Cystic fibrosis affects the lungs, pancreas, intestines, liver, sweat glands, sinuses, and vas deferens, thereby resulting in substantial morbidity and premature mortality. Progressive lung disease is the cause of death in 80% of patients.

EPIDEMIOLOGY

The incidence of cystic fibrosis in the United States, Europe, and Australia is one in 3000 to 5000 births. Cystic fibrosis is most common in the nonHispanic white population but also occurs in significant numbers in Hispanics (one in 7000), African Americans (one in 12,000), and some Native American populations. It also occurs rarely in individuals of Asian origin. Approximately 30,000 persons in the United States have cystic fibrosis, for an estimated prevalence of approximately one in 10,000. Worldwide, an estimated 100,000 individuals are affected. Intensive daily care and exacerbations, particularly those that require hospitalization, are associated with enormous social and monetary costs.

PATHOBIOLOGY

Lung and Sinus

The pathobiology of cystic fibrosis is based on the ion transport activities of the CFTR, which is a membrane glycoprotein that functions as a chloride channel but is also involved in the regulation of transepithelial sodium and bicarbonate transport. In the airway, CFTR dysfunction reduces chloride

CHAPTER 89  Cystic Fibrosis  

563

secretion from the epithelial lining cell into the airway lumen. In addition, sodium absorption from the lumen into the cell is markedly increased. The net effect is a thinning of the airway surface’s liquid lining layer, thereby crucially impairing mucociliary clearance. The subsequent chronic infection leads to an intense neutrophil-dominated inflammatory response. Neutrophil products, including proteolytic enzymes and oxidants, are thought to mediate the pathologic changes in the airway, including bronchiectasis, bronchiolectasis, bronchial stenosis, and fibrosis. Mucus plugging of airways, likely owing to chronic infection and inflammation as well as to CFTR dysfunction in mucus glands, is another prominent feature of airway disease (Fig. 89-1). The origin of sinus disease is believed to be similar to that in the lung. Impaired mucociliary clearance leads to chronic infection and inflammation. Nasal and sinus polyps are common, but their cause is poorly understood.

Intestine and Liver

Pancreas

In the sweat gland, CFTR dysfunction leads to a failure of chloride absorption from the lumen into the sweat ductal lining cell. In contrast, the abnormality in the lung involves chloride secretion. The failure to absorb chloride and, by electroneutrality, sodium, results in marked elevations in the chloride and sodium content of sweat. This abnormality is not accompanied by tissue destruction.

Pathologic studies of the pancreas in infants demonstrate ductal obstruction and dilation as well as acinar dilation. The CFTR is expressed in ductal tissue, suggesting that impairment of chloride and bicarbonate secretion into the lumen of the ducts leads to the viscous secretions that obstruct the ducts and cause acinar dilation. The exposure of pancreatic tissue to proteolytic enzymes of acinar origin leads to a cystic and fibrotic pancreas in the first few years of life. Unlike the lung, injury to the exocrine pancreas does not involve infection. Almost complete exocrine pancreatic insufficiency is seen in 85% of patients and is related to genotype.

CFTR is expressed throughout the intestine. In approximately 15% of cases, cystic fibrosis is accompanied by meconium ileus as a manifestation of severe intestinal obstruction at birth. The incidence of jejunal and ileal stenoses and atresias is greatly increased compared with normal individuals. It is unclear how these severe abnormalities arise, but mucus obstruction, which is frequently seen in intestinal crypts at birth, suggests that abnormalities in CFTR lead to viscous meconium that interferes with normal intestinal development. In the liver, bile duct obstruction is the first pathologic change noted. Focal areas of sclerosis ensue, probably owing to obstructed bile ducts. Infection is not involved in hepatic injury.

Sweat Gland

Male Reproductive Tract

The vas deferens appears to be the organ that is most sensitive to CFTR dysfunction. It often becomes obstructed in fetuses or infants. Resorption of the vas deferens occurs very early in life, and the vas is ultimately not identifiable in most males.

Other Organ Involvement

The primary abnormalities in cystic fibrosis result in secondary involvement of a number of other systems. Diabetes (Chapter 229), which is increasingly common in adolescents and adults, has historically been attributed to the extension of scarring from the exocrine pancreas into the islets of Langerhans. Recent evidence, however, suggests that functional β-cell abnormalities related to an abnormal CFTR. Osteopenia and osteoporosis (Chapter 243), which are common in adults, result from a combination of malnutrition and chronic infection.1 Delayed puberty (Chapters 234 and 235) is also common. Patients can experience recurrent vasculitis or arthralgias that are believed to be caused by the host response to chronic infection. Exocrine pancreatic insufficiency leads to impaired growth and to a multitude of potential nutritional complications, including deficiencies in fat-soluble vitamins and trace elements (Chapter 218).2

Genetics

FIGURE 89-1. Section through the right lung from a 13-year-old young woman with cystic fibrosis demonstrating the gross appearance of cavity formation, bronchiectasis, and purulent mucus plugging.

The gene that encodes the CFTR spans more than 250,000 base pairs on the long arm of chromosome 7. The CFTR (ABCC7), which is a protein of 1480 amino acids, belongs to the adenosine triphosphate–binding cassette transporter family. More than 1500 mutations of five different classes have been described (Table 89-1).3 In the United States, only five mutations are present in more than 1% of cases. The F508δ mutation is by far the most

TABLE 89-1  CLASSES OF CFTR MUTATIONS CLASS I

MECHANISM Defective protein production

GENETIC AND MOLECULAR ABNORMALITIES

REPRESENTATIVE GENOTYPE

Unstable mRNA Truncated protein Premature stop mutations Frameshift Splicing variants

W1282X Del394TT 1717-1G to A

II

Defective protein processing

Trafficking abnormality Protein degraded in proteasome Deletion

F508del

III

Defective channel regulation

Protein at membrane Failure of gating Amino acid substitution

G551D

IV

Defective channel conductance

Protein at membrane Decreased gating Amino acid substitution

R117H

V

Decreased active CFTR

CFTR has normal activity at membrane but is decreased in amount Splice variant Substitution

3849+10kb C to T A455E

CFTR = cystic fibrosis transmembrane conductance regulator.

564

CHAPTER 89  Cystic Fibrosis  

common and is present in approximately 90% of patients in the United States. The next most common mutation, G542X, is present in only 5% of patients. Class 1 mutations are nonsense mutations that result in essentially no expression of the CFTR protein. Class II mutations lead to defective protein processing; in the case of 508δF, protein trafficking to the cell membrane is disrupted because the protein is recognized as defective by cellular quality control mechanisms, which direct it to the proteasome for degradation. In class III mutations, a protein is produced and processed correctly, but the channel remains closed in response to physiologic stimuli. In class IV mutations, the channel is present in the membrane but opens only partially in response to stimuli. In class V mutations, normal CFTR is produced but in reduced amounts because of defective splicing. Different mutations lead to differing levels of CFTR dysfunction.4 Whereas severe mutations (classes 1-3) may reduce CFTR activity to 1% to 3% of normal, mild mutations (classes 4 and 5) may be associated with CFTR activity that is 10% to 20% of normal. An important clinical correlation of CFTR activity is in the exocrine pancreas: patients with severe mutations almost always have pancreatic insufficiency, but some patients with milder mutations may retain pancreatic sufficiency. Patients with mild mutations tend, on average, to have less severe lung disease as well. The clinical course of cystic fibrosis is variable even after controlling for the type of mutation in CFTR, suggesting additional heritable and environmental influences. Genes that code for transforming growth factor-β, mannose-binding lectin, and interferon-related developmental regulator 1 are among the identified modifiers of the severity and course of cystic fibrosis. Most modifiers have to do with the host response to infection or the development of fibrosis rather than the ion transport function of CFTR.

FIGURE 89-2.  A computed tomography image of a 13-year-old young woman with cystic fibrosis demonstrating bronchiectasis in several different regions of the lung, right middle lobe collapse, partial lingular collapse, patchy tree-in-bud opacities, and mild hypoattenuation.

Without specific supportive care, most patients succumb in infancy or early childhood because of malnutrition or lung disease. With the use of pancreatic enzyme replacement therapy, better pulmonary care, and the establishment of specialized centers of expertise, most patients live into their fourth or fifth decade.

sputum production, and change in sputum color that may last days to weeks.5 Frequently, crackles are increased on physical examination, and both the resting oxygen saturation and lung function may decline. Increasing evidence suggests that the permanent loss of lung function is accelerated during periods of exacerbation. Pulmonary complications can also include pneumothorax (Chapter 99), hemoptysis (Chapter 83), and pulmonary hypertension (Chapter 68).6 Some patients with more advanced disease can develop acute ventilatory failure (Chapter 104) with their exacerbations.

Lung Disease

Gastrointestinal Disease

CLINICAL MANIFESTATIONS

Cough, often persistent after viral infections, is the most prominent early feature of the disease. Viral infection may require more frequent hospitalizations in children with cystic fibrosis than in normal children. Although lung disease begins in infancy, pulmonary function is often preserved until adolescence, when a steep decline frequently begins; at this time, pulmonary exacerbations become common. Most patients with cystic fibrosis have a daily productive cough by late adolescence or young adulthood. Cystic fibrosis causes obstructive lung disease, initially with decreased flows at low lung volumes. Forced expiratory volume in 1 second (FEV1) (Chapter 85) is the best correlate of outcome and starts to differ markedly from normal during late adolescence. The rate of decline in FEV1 often predicts the clinical course. Early in the disease, the chest radiograph demonstrates hyperinflation and peribronchial thickening. Computed tomography (Fig. 89-2) can demonstrate bronchiectasis (Chapter 90) early in the course of the disease, even before pulmonary function abnormalities are notable. Airway infection, which is the key clinical manifestation, can be detected by culture of sputum or bronchoalveolar lavage fluid. Pseudomonas aeruginosa (Chapter 306) is the primary pathogen, although its prevalence is decreasing in the United States, likely owing to improved treatment. Staphylococcus aureus (Chapter 288), which is another prominent pathogen, can be methicillin resistant and exist in a small-colony variant form that makes antibiotic treatment difficult. Most infections remain endobronchial and rarely cause invasive disease. An exception is Burkholderia infection, which can result in sepsis that leads to death. Burkholderia infection can also lead to an accelerated decline in lung function and result in death over months to years. Nontuberculous mycobacterial infection can cause granulomatous disease in the airway. Aspergillus (Chapter 339) and other fungal species, which are often identified in sputum samples, can cause allergic bronchopulmonary mycoses, but whether they contribute to endobronchitis apart from allergy is unknown. The polymicrobial nature of airway disease is increasingly appreciated. Stenotrophom*onas maltophilia, Achromobacter xylosoxidans, and Inquilinus limosus are frequently identified serially in airway cultures. Anaerobic infection may also be important. Individuals with cystic fibrosis are subject to acute exacerbations characterized by cough, dyspnea, decreased exercise tolerance, fatigue, increased

Exocrine pancreatic insufficiency, which is apparent in the first year of life in most patients, results in impaired growth and lifelong difficulty in maintaining a normal weight. Patients at all ages may exhibit signs of malabsorption, including bulky, foul-smelling stools and flatulence. Fat-soluble vitamin and trace element deficiencies are common and are difficult to diagnose without regular laboratory monitoring. About 15% of patients retain exocrine pancreatic sufficiency, most of whom have mild mutations associated with 10% to 20% of CFTR function. About one sixth of these patients are subject to recurrent episodes of pancreatitis (Chapter 144) that can lead to pancreatic pseudocysts or ultimately result in exocrine pancreatic insufficiency. Intestinal obstruction can occur at any age. Frequently, the blockage is at the ileocecal valve, but generalized chronic constipation (Chapter 136) is even more common. Intussusception of the appendix can also occur. Inflammatory bowel disease (Chapter 141) and gastrointestinal malignancies (Chapters 192 and 193) appear to be more common than in the general population. Chronic abdominal pain can occur at any time of life and is often difficult to treat. Most patients who develop liver disease do so in childhood or adolescence. Liver abnormalities are often first appreciated when physical examination reveals splenomegaly or a palpable, firm liver. Occasionally, hematemesis leads to the identification of esophageal or gastric varices that are indicative of portal hypertension. Splenic sequestration can lead to neutropenia or thrombocytopenia. Decreased hepatic production of clotting factors can also contribute to bleeding. Occasionally, jaundice is a presenting sign of hepatobiliary disease. Except for γ-glutamyl transpeptidase (GGT) levels, liver enzymes are frequently normal, even in patients with advanced disease. Gallstones (Chapter 155) are common and may or may not lead to symptoms. The hepatopulmonary syndrome (Chapter 153) can occur.

Other Organ Involvement

Although most patients have radiographic evidence of sinus changes, acute or chronic sinusitis occurs in only a minority of individuals. Sinusitis can be accompanied by debilitating headache and anosmia. Nasal or sinus polyposis can lead to obstructed breathing during sleep. Hypoelectrolytemia from sweat losses can occur at any age. Symptoms range from nausea, vomiting, and decreased appetite to seizures and

CHAPTER 89  Cystic Fibrosis  

circulatory collapse with fatal consequences. Almost all men are sterile because of the changes in the vas deferens. Spermatogenesis is normal, however. Cystic fibrosis–related diabetes (Chapter 229) increases in frequency with age.7 By 30 years of age, approximately one third of patients have diabetes. Although patients rarely develop ketoacidosis, the microvascular and macrovascular complications of diabetes can occur. In addition, patients with diabetes appear to have an accelerated decline in lung function. Osteoporosis (Chapter 243), osteopenia, and increased fractures also increase in frequency with age. Vasculitis accompanied by rash or arthralgia can occur at any time of life. Chronic pain and depression are other important complications that increase with age.

DIAGNOSIS

Newborn Screening and Diagnosis

In the United States, all 50 states require newborn screening for cystic fibrosis to allow early diagnosis and immediate treatment. All newborn screening programs currently measure immunoreactive trypsinogen, a marker of pancreatic injury, from a dried blood spot taken during the first few days of life as the first step in the screening process. This biochemical screen identifies a large number of infants with abnormalities, only a fraction of whom have cystic fibrosis. Most programs perform genetic mutation analysis as the next step. Sweat testing is required to establish the diagnosis if suspected patients carry only one identifiable mutation, but most programs perform confirmatory sweat testing even if two mutations are present. Sweat testing measures the chloride concentration in sweat that is stimulated by pilocarpine iontophoresis. The result is considered abnormal in adults and children when the concentration of chloride in the sweat is greater than 60 mmol/L; in infants, a concentration greater than 40 mmol/L is considered diagnostic. Patients with milder mutations may have normal sweat chloride values. A family history of cystic fibrosis also provides supportive evidence.

Diagnosis in Adulthood

Five percent of patients are diagnosed after 18 years of age, mostly on the basis of recurrent pancreatitis, nasal polyposis, chronic sinusitis, bronchiectasis, male infertility, allergic bronchopulmonary mycoses, and nontuberculous mycobacterial infection (Table 89-2). If the predominant symptoms are respiratory, the differential diagnosis includes primary ciliary dyskinesia, immune deficiency, or postinfectious bronchiectasis (Chapter 90). If the predominant symptom is recurrent pancreatitis (Chapter 144), the differential diagnosis includes hereditary pancreatitis with abnormalities in the SPINK gene. Transepithelial potential differences are altered in cystic fibrosis because of abnormal transport of sodium and chloride. The measurement of nasal potential difference, therefore, can sometimes be used as a diagnostic tool, particularly in adults. It is increasingly recognized that some patients appear to have cystic fibrosis on clinical grounds but do not meet the criteria for diagnosis, usually because their sweat test results are in the normal range or two genetic mutations cannot be identified. These patients are sometimes diagnosed as having atypical cystic fibrosis, nonclassical cystic fibrosis, or variant cystic fibrosis.

TABLE 89-2  APPROACH TO DIAGNOSIS OF CYSTIC FIBROSIS IN ADULT PATIENTS CONDITIONS SUGGESTING THE DIAGNOSIS OF CYSTIC FIBROSIS IN ADULTS Recurrent pancreatitis Male infertility Chronic sinusitis Nasal polyposis Nontuberculous mycobacterial infection Allergic bronchopulmonary mycosis Bronchiectasis RECOMMENDED DIAGNOSTIC STUDIES Sweat electrolyte determination Extended CFTR mutation analysis Nasal potential difference High-resolution CT scan to identify bronchiectasis CT scan of sinuses for polyposis Sputum induction or bronchoalveolar lavage to identify bacterial and fungal pathogens CFTR = cystic fibrosis transmembrane conductance regulator; CT = computed tomography.

565

Full analysis of the CFTR coding and flanking regions may be helpful in making the diagnosis. Such patients should be followed at a cystic fibrosis center so that their lung disease can be treated and they can be monitored for other complications of cystic fibrosis.

TREATMENT  The general consensus is that treatment is best conducted at specialized centers that use a team approach. Much of their success is based on the education of patients and families regarding symptoms, complications, the need  for daily treatment, the importance of close monitoring of pulmonary function, and the potential benefits of rapid intervention for any detected abnormalities.8 Much research has addressed the possibility of treating this genetic disease with gene therapy, but such approaches have not yet yielded positive  results. A1  As a result, interest has shifted to improving CFTR function by using druglike molecules.9,10 In a randomized trial, ivacaftor, a CFTR modulator (150 mg twice daily for 48 weeks), significantly improved lung function, weight, and sweat chloride in cystic fibrosis patients with the G551D mutation. A2  This drug was rapidly approved for clinical use by the U.S. Food and Drug Administration.

Pulmonary Infections

Pulmonary infections can be treated with oral, inhaled, or intravenous antibiotics. An increase in cough or other respiratory symptoms should be addressed with the introduction of antibiotics or a change in antibiotics within a few days. Nebulized antibiotics (4 weeks of either aztreonam 75 mg two or three times a day or tobramycin 300 mg twice daily), alone or in combination with oral antibiotics, improve lung function and decrease exacerbations in patients with chronic Pseudomonas infection. A3  A4  These same antibiotic strategies have also been increasingly successful for eradicating Pseudomonas infection. Chronic oral macrolide treatment (e.g., azithromycin 5-15 mg/kg/ day, 500 mg three times per week) can reduce exacerbations for up to 6 months. A5  It is not yet clear whether the chronic use of antibiotics in this setting leads to the development of more resistant organisms. More severe changes in symptoms or an acute fall in lung function requires intravenous antibiotics aimed at the cultured pathogen (Chapter 97). Nontuberculous mycobacterial infections are treated for 6 months or longer using multiple antibiotic agents (Chapter 325). Allergic bronchopulmonary mycoses are treated with corticosteroids and antifungal agents (Chapter 331). Several agents known to decrease the viscosity of mucus have proven to be of clinical benefit in cystic fibrosis. Daily use of inhaled rhDNase (2.5 mg) is associated with improvement in lung function and fewer exacerbations. Inhaled hypertonic (7%) saline can increase pulmonary function and reduce exacerbations, A6  and adding inhaled mannitol (400 mg twice daily) to standard therapy can produce a sustained improvement in pulmonary function for 26 to 52 weeks. A7  Many patients have hyperreactive airways and may benefit from inhaled bronchodilators (Chapter 87). Inhaled corticosteroids are controversial and do not have proven benefit. Oral corticosteroid “bursts” (e.g., 5 days of prednisone, 1 mg/kg twice a day in children and 60 mg a day in adults) are often useful, but chronic administration of oral corticosteroids can result in severe complications, including diabetes and stunted growth. Most patients perform physical means of airway secretion clearance one or more times a day. Even passive smoke exposure is deleterious. Oxygen therapy is often required to maintain saturation and prevent the development of pulmonary hypertension. Noninvasive ventilation is used mainly in patients with more advanced disease. Pneumothorax almost always requires pleurodesis. Persistent or recurrent hemoptysis is treated with bronchial artery embolization. Occasionally, lobectomy is required. Patients in acute ventilatory failure should receive mechanical ventilation unless they have decided against such treatment. In patients who have advanced disease, the possible need for ventilation should be addressed before the need actually arises. Lung transplantation (Chapter 101) is an option for many patients. Individuals with cystic fibrosis have survival rates after transplantation comparable to or better than those of other patients. ,

Gastrointestinal Diseases

Pancreatic enzyme replacement (Chapter 144) is the mainstay of treatment for exocrine pancreatic insufficiency. Because gastric acid decreases enzyme activity, H2-blockers (e.g., ranitidine 150 mg twice daily in children weighing >30 kg and in adults) or proton pump inhibitors (e.g., lansoprazole 30 mg orally once daily in children weighing >30 kg and in adults) are often used. Children and adolescents frequently use multiple nutritional supplements every day to maintain weight. Fat-soluble vitamin replacement therapy is necessary in most patients. Between 10% and 20% of patients may require gastrostomy feeding to aid growth or maintain weight. To prevent intestinal obstruction, dietary fiber should be increased, and polyethylene glycol at varying doses (e.g., 17 g orally with 8 oz of water one to three times per day) is frequently used on a daily basis. Acute obstructions can be treated with more intensive use of polyethylene glycol or Gastrografin

enema. Occasionally, refractory constipation (Chapter 136) requires surgical approaches that can result in loss of intestine.

Other Organ Systems

A combination of nasal rinses and topically applied corticosteroids and antibiotics is used to treat sinus disease (Chapter 426). Surgery is often required, however, especially for polyps. Many pediatric patients receive daily salt supplementation. Adults should be counseled on the symptoms of salt depletion and encouraged to increase the amount of salt in their diets if there are no medical contraindications to doing so. Regular screening for the onset of impaired glucose homeostasis or frank diabetes is required in all patients older than 10 years. Diabetes is treated with insulin (Chapter 229) because the safety and efficacy of oral antihyperglycemic agents have not been demonstrated in those with cystic fibrosis. Bone health is addressed through vitamin D supplementation, calcium supplementation, and oral bisphosphonate therapy (Chapter 243). Delayed puberty and short stature require consultation with endocrinologists and sometimes hormonal administration. Most clinicians believe that both aerobic exercise and strength training can have beneficial effects, although the implementation of exercise programs has been difficult in clinical practice. Men with cystic fibrosis can father children through the use of epididymal aspiration to retrieve sperm followed by in vitro fertilization.

General Care

Given all the pulmonary, nutritional, and other therapies prescribed for individuals with cystic fibrosis, their care amounts to several hours a day. The transition from pediatric care to adult care can be challenging and requires diligent planning and execution.11 This burden has a major influence on the quality of life in patients and their families and may contribute to the increasing incidence of depression observed in this population. End-of-life care encompasses many complex issues. Patients are often depressed and experience chronic pain. They are asked to perform increasingly intense therapeutic regimens. They may have changed locations to await transplantation. Family, medical, and professional relationships are disrupted. Excellent communication with caregivers about advance directives and other planning is necessary.

PREVENTION

Prenatal carrier screening, which is offered in many countries, can decrease the incidence of cystic fibrosis by approximately 25%. Newborn screening programs may also decrease the incidence by influencing the future reproductive decisions of parents of an affected child.

PROGNOSIS

The median expected survival time for cystic fibrosis patients at birth in the United States is 37 years. However, the peak age at death is 26 years, demonstrating that some patients are particularly vulnerable to devastating lung disease. Late adolescence and early adulthood are high-risk times for pulmonary insufficiency. Patients who survive to their 30s and beyond are often more stable, have milder CFTR mutations, and have a very slow decline in lung function. The success of ivacaftor in treating patients with the G551D mutation has spurred research for small molecules aimed at improving CFTR function for other mutations.

Grade A References A1. Lee TW, Southern KW. Topical cystic fibrosis transmembrane conductance regulator gene replacement for cystic fibrosis-related lung disease. Cochrane Database Syst Rev. 2013;11:CD005599. A2. Ramsey BW, Davies J, McElvaney NG, et al. A CFTR potentiator in patients with cystic fibrosis and the G551D mutation. N Engl J Med. 2011;365:1663-1672. A3. Ramsey BW, Pepe MS, Quan JM, et al. Intermittent administration of inhaled tobramycin in patients with cystic fibrosis. N Engl J Med. 1999;340:23-30. A4. McCoy KS, Quittner Al, Oermann CM, et al. Inhaled aztreonam lysine for chronic airway Pseudomonas aeruginosa in cystic fibrosis. Am J Respir Crit Care Med. 2008;178:921-928. A5. Saiman L, Marshall BC, Mayer-Hamblett N, et al. Azithromycin in patients with cystic fibrosis chronically infected with Pseudomonas aeruginosa: a randomized controlled trial. JAMA. 2003; 290:1749-1756. A6. Elkins MR, Robinson M, Rose BR, et al. A controlled trial of long-term inhaled hypertonic saline in patients with cystic fibrosis. N Engl J Med. 2006;354:229-240. A7. Aitken ML, Bellon G, De Boeck K, et al. Long-term inhaled dry powder mannitol in cystic fibrosis: an international randomized study. Am J Respir Crit Care Med. 2012;185:645-652.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 89  Cystic Fibrosis  

GENERAL REFERENCES 1. Sermet-Gaudelus I, Bianchi ML, Garabedian M, et al. European cystic fibrosis bone mineralisation guidelines. J Cyst Fibros. 2011;10(suppl 2):S16-S23. 2. Culhane S, George C, Pearo B, et al. Malnutrition in cystic fibrosis: a review. Nutr Clin Pract. 2013;28:676-683. 3. Tsui LC, Dorfman R. The cystic fibrosis gene: a molecular genetic perspective. Cold Spring Harb Perspect Med. 2013;3:a009472. 4. Sosnay PR, Siklosi KR, Van Goor F, et al. Defining the disease liability of variants in the cystic fibrosis transmembrane conductance regulator gene. Nat Genet. 2013;45:1160-1167. 5. Mogayzel PJ Jr, Naureckas ET, Robinson KA, et al. Cystic fibrosis pulmonary guidelines. Chronic medications for maintenance of lung health. Am J Respir Crit Care Med. 2013;187:680-689.

566.e1

6. Flume PA, Mogayzel PJ Jr, Robinson KA, et al. Cystic fibrosis pulmonary guidelines: pulmonary complications: hemoptysis and pneumothorax. Am J Respir Crit Care Med. 2010;182:298-306. 7. Kelly A, Moran A. Update on cystic fibrosis-related diabetes. J Cyst Fibros. 2013;12:318-331. 8. Smyth AR, Bell SC, Bojcin S, et al. European Cystic Fibrosis Society Standards of Care: best practice guidelines. J Cyst Fibros. 2014;13(suppl 1):S23-S42. 9. Boyle MP, Bell SC, Konstan MW, et al. A CFTR corrector (lumacaftor) and a CFTR potentiator (ivacaftor) for treatment of patients with cystic fibrosis who have a phe508del CFTR mutation: a phase 2 randomised controlled trial. Lancet Respir Med. 2014;2:527-538. 10. Van Goor F, Yu H, Burton B, et al. Effect of ivacaftor on CFTR forms with missense mutations associated with defects in protein processing or function. J Cyst Fibros. 2014;13:29-36. 11. Nazareth D, Walshaw M. Coming of age in cystic fibrosis—transition from paediatric to adult care. Clin Med. 2013;13:482-486.

566.e2

CHAPTER 89  Cystic Fibrosis  

REVIEW QUESTIONS 1. A 27-year-old male with a nagging, productive cough is known to be infertile. Which diagnostic approach is NOT likely to be helpful in determining whether he has cystic fibrosis? A. Ultrasonography of the kidneys B. Sweat test C. Extended CFTR genotype analysis D. Sputum culture E. Exploration of family history looking for deaths in childhood Answer: A  The kidneys are not usually affected in cystic fibrosis, at least to the extent that can be appreciated on ultrasonography. The other tests can all reveal supportive evidence of the diagnosis of cystic fibrosis. 2. Considering the same 27-year-old man as in question 1, the sweat test result comes back abnormal at 78 mmol/L. Which of the following statements regarding this patient’s care is INCORRECT? A. You tell him that it is unlikely that he will need follow-up at a cystic fibrosis center because he was diagnosed at such a late age. B. You should order a mutation analysis to define his genotype. C. You should order lung function testing. D. You should order a high-resolution computed tomography scan of his chest looking for bronchiectasis. E. You should order a sputum culture. Answer: A  All patients with a diagnosis of cystic fibrosis should have follow-up at a cystic fibrosis care center.

3. Considering the same 27-year-old man as in questions 1 and 2, his lung function testing reveals a forced expiratory volume in 1 second of 65% predicted with no reversibility after a bronchodilator. His computed tomography scan shows definite bronchiectasis. His sputum culture grows Pseudomonas aeruginosa. His genotype is G551D/A455E. Which of the following treatments is NOT indicated? A. Inhaled tobramycin or inhaled aztreonam daily, one month on and one month off B. Inhaled dornase alfa daily C. Ivacaftor 150 mg PO twice daily D. Oral corticosteroid treatment at 30 mg every other day in the morning. E. Daily airway secretion clearance Answer: D  Corticosteroids are not part of the first-time care of cystic fibrosis, and they also can cause complications, especially by making patients more susceptible to infection.

566

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

90  BRONCHIECTASIS, ATELECTASIS, CYSTS, AND LOCALIZED LUNG DISORDERS ANNE E. O’DONNELL

  BRONCHIECTASIS DEFINITION

Bronchiectasis is an abnormal permanent dilation of the bronchi and bronchioles caused by repeated cycles of airway infection and inflammation. The distal airways become thickened; the mucosal surfaces develop edema, inflammation, and suppuration; an ultimately, there is neovascularization of the adjacent bronchial arterioles. Bronchiectasis, which can be focal or diffuse, is triggered by a variety of genetic, anatomic, and systemic processes. Abnormalities of cilia, mucus clearance, mucus rheology, airway drainage, and host defenses can result in bronchiectasis. Regardless of the cause, patients with bronchiectasis develop chronic infections, which may lead to progressive lung destruction.

EPIDEMIOLOGY

Based on insurance claims reviews, an estimated 110,000 or more patients in the United States are receiving treatment for bronchiectasis that is not related to cystic fibrosis (Chapter 89),1 and these numbers appear to be increasing.2 The prevalence in the United States has been reported as 4.2 per 100,000 persons age 18 to 34 years and 272 per 100,000 among those older than 75 years. In the older age category, women are disproportionally represented. Other epidemiologic surveys suggest that there is increased risk for the development of bronchiectasis in individuals with reduced access to health care and higher rates of pulmonary infection in childhood.

PATHOBIOLOGY

In up to one third of cases, the cause of bronchiectasis is not identified. Other cases are related to pulmonary infections, genetic causes, anatomic abnormalities, and immune and autoimmune diseases.3

Pulmonary Infections

Approximately one third of patients with bronchiectasis have an infectious trigger, usually years before the onset of the disease. Childhood viral infections, such as pertussis (Chapter 313) and bacterial infection, can cause permanent damage to the airways, leading to bronchiectasis years after the initial infection. Mycobacterial tuberculosis with its resultant granulomatous inflammation of the airway, lung parenchyma, and lymph nodes can cause subsequent bronchiectasis (Chapter 324), and nontuberculous mycobacterial infections have been recognized as an increasing cause and complication of bronchiectasis, particularly in white women older than 55 years (Chapter 325). Nontuberculous mycobacterial-related bronchiectasis typically involves the right middle lobe and lingula and can be associated with the “tree-in-bud” pattern of bronchiolar infection as well.

Genetics

Cystic fibrosis (Chapter 89) is characterized by bilateral diffuse bronchiectasis. Although many cystic fibrosis patients are diagnosed in childhood with multisystem disease, older patients may present with only pulmonary or pulmonary and sinus manifestations. Some patients with bronchiectasis may have subtle defects in the cystic fibrosis transmembrane conductance regulator channel without a clear-cut diagnosis of cystic fibrosis.4 In primary ciliary dyskinesia, abnormalities in the dynein arms prevent normal ciliary beating. Patients with primary ciliary dyskinesia generally have significant sinopulmonary disease and infertility, and approximately half of these patients have Kartagener syndrome with situs inversus (Chapter 69). Patients with α1-antitrypsin deficiency also may develop bronchiectasis.

Anatomic Causes

Patients with chronic abnormalities of their swallowing mechanism or with esophageal dysfunction may develop focal or diffuse bronchiectasis with lower lobe predominance (Chapter 138). Direct lung injury caused by acid

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

A

567

B

FIGURE 90-1.  A and B, High-resolution computed tomographic images of bilateral bronchiectasis in a patient with primary ciliary dyskinesia.

or particulate matter aspiration or recurrent pneumonia may lead to bronchiectasis. Chronic obstructive pulmonary disease (COPD) is sometimes complicated by bronchiectasis (Chapter 88). Patients with chronic lower airway bacterial colonization and increased airway inflammation may develop areas of bronchiectasis. Rarely, patients with asthma (Chapter 87) have been found to have bronchiectasis. Allergic bronchopulmonary aspergillosis (Chapter 339) can cause a distinct “finger-in-glove” central bronchiectasis owing to chronic inflammation and mucous plugging. Airway abnormalities such as endobronchial tumors (Chapter 191), extrinsic compression by lymph nodes (right middle lobe syndrome), and foreign bodies are also rare causes of focal bronchiectasis. Tracheobronchomegaly (Mounier-Kuhn syndrome) is associated with distal bronchiectasis.

Immune and Autoimmune Diseases

Primary hypogammaglobulinemia (Chapter 250) leads to recurrent pulmonary infections that may result in bronchiectasis. Patients with immunoglobulin G subclass deficiencies may develop bronchiectasis if the deficiency leads to reduction in antibody production. Defects of neutrophil adhesion and chemotaxis (Chapter 169) have been found to cause bronchiectasis. Patients with human immunodeficiency virus infection (Chapter 391) have a higher prevalence of bronchiectasis than individuals with a normally functioning immune system. Bronchiectasis is an increasingly recognized complication of collagen vascular diseases, particularly rheumatoid arthritis (Chapter 264) and Sjögren syndrome (Chapter 268). The airway injury is likely attributable to chronic inflammation or esophageal dysfunction. Inflammatory bowel disease (Chapter 141) also causes bronchiectasis by undetermined mechanisms.

CLINICAL MANIFESTATIONS

Patients present with chronic cough and usually have mucopurulent or purulent sputum production. Occasionally, a dry nonproductive cough is the primary manifestation. Other symptoms include dyspnea, intermittent hemoptysis, and pleuritic chest pain. Weight loss, malaise, and fatigue sometimes develop. When patients have infectious exacerbations, they may develop fever as well as an increase in their baseline symptoms. Physical findings in patients with bronchiectasis are nonspecific and include an abnormal chest examination with wheezing, crackles, or both. Clubbing of the digits is rare. The clinical course of patients with bronchiectasis is variable. Some patients have few to no symptoms, others have daily cough with sputum production, and some patients have occasional to frequent exacerbations. A slow decline in pulmonary function is seen with bronchiectasis; the decline is more rapid in patients infected with Pseudomonas aeruginosa (Chapter 306) and in patients who have more frequent exacerbations.

DIAGNOSIS

Imaging Studies

Although the diagnosis may be suspected by plain chest radiography, highresolution computed tomography (HRCT) is the current “gold standard” for confirming bronchiectasis. The characteristic computed tomography (CT)

FIGURE 90-2. High-resolution computed tomography image of nodular bronchiectasis caused by a nontuberculous mycobacterium infection.

findings are lack of bronchial tapering, bronchi visible in the peripheral 1 cm of the lungs, and an internal bronchial diameter greater than the diameter of the accompanying bronchial artery. Other associated HRCT findings are cysts off the end of a bronchus, tree-in-bud irregular branching lines (E-Fig. 90-E1) indicating mucus impaction (E-Figs. 90-E2 and 90-E3), volume loss (E-Fig. 90-E4), and occasionally associated consolidation (Fig. 90-1). The location of the bronchiectatic airways may suggest the cause: upper lobe predominance is seen in cystic fibrosis and lower lobe predominance in aspiration syndromes (E-Fig. 90-E5). Whereas right middle lobe and lingula involvement suggests the presence of nontuberculous mycobacterial infection (Fig. 90-2 and E-Figs. 90-E6A and 90-E6B), central bronchiectasis is seen with allergic bronchopulmonary aspergillosis (Fig. 90-3). Pulmonary function testing, which should be performed on all patients with suspected bronchiectasis, usually shows airflow obstruction as measured by the ratio between the forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) (Chapter 85). The severity of the airflow obstruction and the rate of decline correlate with radiographic extent of disease and frequency of exacerbation. Bronchoscopy will detect airway abnormalities, including tumors, structural deformities, and foreign bodies, and hence should be considered in the evaluation of localized bronchiectasis. Cultures of sputum and of bronchoalveolar lavage when expectorated sputum is not available have an important role in assessing the infectious complications of bronchiectasis. Molecular techniques have recently demonstrated diverse polymicrobial communities in the lungs of patients with bronchiectasis, both when they are clinically stable and also during exacerbations.5

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

567.e1

E-FIGURE 90-1. A thin-slice high-resolution computed tomography image showing bronchiectasis and volume loss on the left and areas of mild cylindrical bronchiectasis and “tree-in-bud” bronchiolitis in the right lower lobe.

E-FIGURE 90-2. A thin-slice high-resolution computed tomography image through the middle chest; there are mild bronchiectasis in the right middle lobe adjacent to the heart and mucous plugging in areas of bronchiectasis at the right base.

E-FIGURE 90-3.  Bronchiectatic changes and mucous plugging superimposed on diffuse emphysematous changes in the lungs.

E-FIGURE 90-4. Extensive bronchiectasis in lower cuts through the right lung with associated volume loss.

567.e2

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

E-FIGURE 90-5. Computed tomography slice showing a “signet ring” abnormality consistent with a localized area of bronchiectasis in the right lower lobe.

A

B

E-FIGURE 90-6.  A, A pneumonic infiltrate superimposed on an area of bronchiectasis in the right middle lobe. Bilateral breast implants are also visible. The patient had a history of chronic mycobacterial infection but was acutely ill with a superimposed bacterial pneumonia when this study was performed. B, Follow-up imaging after treatment of the acute infection. A residual minor area of right middle lobe bronchiectasis is seen.

568

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

A

B

FIGURE 90-3.  A and B, High-resolution computed tomography images of finger-in-glove central bronchiectasis caused by allergic bronchopulmonary aspergillosis.

The dominant organisms are P. aeruginosa and Haemophilus influenzae, but anaerobic organisms may also be detected. The presence of P. aeruginosa portends a worse prognosis and more frequent exacerbations. Patients with no identifiable pathogens have the mildest disease. Staphylococcus aureus in the airway may suggest cystic fibrosis as the cause of the bronchiectasis. Nontuberculous mycobacteria are found with increasing frequency in the airways of patients with bronchiectasis, usually as a complication of preexisting bronchiectasis but occasionally as its primary cause. The laboratory evaluation of patients with bronchiectasis should be individualized. All patients should have sputum cultures for bacterial and mycobacterial testing. Other tests that should be considered include measurement of serum immunoglobulin levels and screening for genetic diseases, particularly in patients with diffuse bronchiectasis. Cystic fibrosis (Chapter 89) is diagnosed by elevated sweat chloride levels and by genetic testing. Primary ciliary dyskinesia can be evaluated by measurement of nasal nitric oxide levels, ciliary beat frequency and pattern testing, and electron microscopy studies.6 α1-Antitrypsin deficiency is diagnosed by measuring levels and performing phenotyping (Chapter 88). Screening for rheumatoid arthritis (Chapter 264) or Sjögren syndrome (Chapter 268) also may be reasonable in patients with diffuse bronchiectasis.

TREATMENT  The goals of treatment are to reduce the frequency of exacerbations and potentially to improve quality of life, reduce symptoms, and alter the natural history of the disease (Table 90-1). Multimodality maintenance treatment7 for patients with more advanced disease or three or more exacerbations may include airway clearance and anti-inflammatory therapies, as well as short and long-term antibiotic therapy, which reduces markers of airways and systemic inflammation. Exacerbations are treated based on clinical acuity. Because patients are heterogeneous and therapeutic trials are few, therapy is commonly individualized, especially because no therapies are currently approved by the U.S. Food and Drug Administration for non–cystic fibrosis bronchiectasis and because the proven treatments for cystic fibrosis are often not effective.

Preventing Exacerbations

The 23-valent pneumococcal vaccination (Chapter 18) is recommended for patients with bronchiectasis. Routine seasonal influenza vaccination is also standard. At present, no vaccines are available for prevention of the other infectious complications of bronchiectasis.

Treatment of the Underlying Etiology

For treatable conditions, such as immunoglobulin deficiency, replacement therapy (Chapter 250) should be considered even though there are few data on whether that alters the natural history of the lung disease. Patients with allergic bronchopulmonary aspergillosis (Chapter 339) should be treated with steroids to mitigate the inflammatory process that leads to the bronchiectasis.

TABLE 90-1  POTENTIAL THERAPIES FOR BRONCHIECTASIS Treat underlying condition, if possible Mobilization of secretions Pharmacologic Mechanical Anti-inflammatory therapy Inhaled steroids Macrolides Antimicrobial therapy Pathogen specific Surgery Localized or refractory disease Transplantation End-stage disease Adapted from O’Donnell A. Bronchiectasis. Chest. 2008;134:815-823.

Airway Clearance

Chest physiotherapy and the use of devices to aid mucociliary clearance appear to be beneficial in non–cystic fibrosis bronchiectasis. In a randomized trial, for example, twice-daily use of an oscillatory positive expiratory pressure device (Acapella) improved sputum volume and quality of life end points compared with no routine physiotherapy. A1  Other techniques that may also have a role for airway clearance include traditional chest physical therapy with postural drainage and the use of chest wall oscillator vests.8 Formal pulmonary rehabilitation and exercise also likely provide benefit to patients with bronchiectasis. Inhaled therapy with nebulized hypertonic saline (7%) may enhance airway clearance, decrease exacerbations, and improve lung function as well as quality of life. A2  Chronic inhalation of dry powder mannitol improves sputum clearance but does not reduce the frequency of exacerbations. Although recom­ binant human DNase is efficacious in cystic fibrosis bronchiectasis, a large clinical trial showed it had deleterious effects when given a maintenance therapy in patients with non–cystic fibrosis bronchiectasis, so it should not be used. Other mucolytic agents are of unproven benefit. A3  No randomized trials support the use of routine short-acting β-agonist or anticholinergic bronchodilators in bronchiectasis. However, a subset of patients with airway reactivity likely benefits from use of these agents (Chapter 87).

Reduction of Airway Inflammation

One clinical trial demonstrated that inhaled medium-dose budesonide, when combined with formoterol, is safe and more effective than high-dose budesonide in treating patients with non–cystic fibrosis bronchiectasis. A4  Oral steroids, although occasionally used in patients with bronchiectasis, have never been evaluated in a clinical trial.

Antimicrobial Therapy

At present, there is no firm evidence to support the use of routine maintenance antibiotics, although such therapy may be considered in patients with frequent exacerbations and progressive lung destruction. When

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

mycobacterial species are cultured from patients with bronchiectasis, decisions regarding whether to treat and which antimicrobial agents to use are based on published guidelines (Chapters 324 and 325). Chronic low-dose oral macrolide therapy (e.g., azithromycin 500 mg three times per week or 250 mg/day or erythromycin ethylsuccinate 400 mg twice daily) can reduce exacerbations in patients with non–cystic fibrosis bronchiectasis. A5  A6  Whether patients will develop resistant organisms is of concern, and macrolide therapy alone should not be used in patients co-infected with nontuberculous mycobacteria. Targeted inhaled antimicrobial therapies are also an option, particularly in patients infected with Pseudomonas spp. A7  For example, nebulized gentamicin (80 mg twice daily) for 12 months can provide sustained bacteriologic and clinical benefit. Clinical trials have demonstrated microbiologic benefits with inhaled tobramycin, 300 mg twice per day as a 4-week trial for one cycle and a 2-week-on, 2-week-off trial for three cycles, but clinical benefit was not firmly established, and some patients experienced unacceptable respiratory side effects. Antimicrobial resistance is also a concern. Inhaled colistin, 1 million IU twice daily delivered by nebulizer, also was recently shown to be safe and possibly effective in adherent patients with bronchiectasis and pseudomonas aeruginosa infection. A8  Additional inhaled antibiotics currently being evaluated in clinical trials include dry powder ciprofloxacin, nebulized liposomal ciprofloxacin, and dry powder tobramycin. Other off-label antibiotic strategies include prolonged intravenous antibiotics targeted at the cultured pathogens. ,

Surgery and Transplantation

569

or postoperative patient. Patchy atelectasis is caused by alveolar filling processes, such as hemorrhage and edema (Chapter 91). Passive, relaxation, or compression atelectasis occurs when the lung recoils to a smaller volume because of fluid or air in the adjacent pleural space. Obstructive or resorptive atelectasis is caused by bronchial block to the entry of air, with resultant retractile consolidation. Intrinsic airway obstruction may be caused by mucous plugs, foreign bodies, or tumors in the airway. Extrinsic airway obstruction results from compression of the airway owing to peribronchial lymph node enlargement or other masses impinging on the airway. Rounded atelectasis is caused by pleural thickening that invagin*tes and traps adjacent lung. Any chronic pleural disease can cause rounded atelectasis, particularly asbestos-related pleural disease.

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Atelectasis is typically asymptomatic and diagnosed on chest imaging, but it may cause dyspnea and tachypnea and result in hypoxemia. In postoperative patients, atelectasis may be a cause of low-grade fever. Plain chest radiography shows loss of lung volume and the displacement of the lobar fissure, mediastinum, or diaphragm toward the involved lung unit (Figs. 90-4 and 90-5). Platelike or discoid atelectasis manifests as horizontal or curvilinear lines on plain chest radiography. Rounded atelectasis is an ovoid masslike density abutting the pleura. The type and cause of atelectasis can sometimes be

Resectional surgery may have a role for patients who have focal disease or for patients who have hemoptysis that cannot be controlled by embolization of the bleeding vessels (Chapter 101). Surgical resection can also benefit some patients who have diffuse bronchiectasis unresponsive to conventional therapy and some patients infected with nontuberculous mycobacteria. Double-lung transplantation (Chapter 101) has been successfully performed in patients with end-stage lung disease caused by non–cystic fibrosis bronchiectasis, and the clinical outcomes parallel those seen with transplantation for other end-stage lung diseases.

Treatment of Acute Exacerbations of Bronchiectasis

When a patient with bronchiectasis experiences an acute exacerbation, antimicrobial treatments should be aimed at the known infecting organisms. Mild to moderate exacerbations can be treated with oral antibiotics, targeted to the results of the sputum culture, for 2 to 3 weeks. More severe exacerbations or exacerbations caused by resistant organisms generally require intravenous antibiotics administered in hospital or at home. No benefit has yet been demonstrated by adding an inhaled antibiotic to systemic therapy for an acute exacerbation. Patients experiencing an acute exacerbation likely benefit from airway clearance modalities and the other nonantibiotic therapies discussed previously.

PROGNOSIS

Non–cystic fibrosis bronchiectasis is a heterogeneous disease with a widely variable prognosis. Patients with more severe obstructive and restrictive findings on pulmonary function tests, poor gas transfer, and chronic pseudomonal infection have the worst prognosis.9 Independent predictors of future hospitalization include prior hospital admissions, advanced dyspnea, FEV1 less than 30% predicted, P. aeruginosa colonization, colonization with other pathogenic organisms, and three or more lobes involved on HRCT.10 Bronchiectasis in patients with moderate to severe COPD is an independent risk factor for all-cause mortality. Radiographic extent of disease, hypoxemia, hypercapnia, and evidence of right heart failure are also predictors of outcome. Bronchiectasis patients who are admitted to an intensive care unit for respiratory failure have been reported to have a 60% 4-year survival rate.

FIGURE 90-4.  Plain chest radiograph demonstrating right upper lobe atelectasis (caused by an endobronchial tumor).

  ATELECTASIS DEFINITION

Atelectasis, or collapse, is caused by hypoventilation of lung units. Atelectasis may involve an entire lung or a lobe, segment, or subsegment. Atelectasis can be caused by intrinsic obstruction of an airway or external compression from lymph nodes, parenchymal masses, or other entities. When lung units are atelectatic, ventilation-perfusion mismatch leads to hypoxemia. Infection may result from sustained atelectasis.

EPIDEMIOLOGY AND PATHOBIOLOGY

The lung bases and posterior segments are vulnerable to dependent atelectasis, which is caused by inadequate ventilation, particularly in an immobilized

FIGURE 90-5. Computed tomography image of rounded atelectasis.

570

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

A

B

FIGURE 90-6.  Pulmonary sequestration. A, Computed tomography image of pulmonary sequestration in right lower lobe. B, Feeding vessel visible arising from the aorta.

elucidated by CT or ultrasonography. Bronchoscopy is required to confirm intrinsic versus extrinsic compression in obstructive-resorptive atelectasis and to determine the exact pathology of the obstruction. An oxygen saturation concentration can help assess the severity of the atelectasis and overall lung dysfunction.

PREVENTION AND TREATMENT  Incentive spirometry is commonly prescribed to prevent or treat atelectasis in patients with limited mobility because of recent surgery, neuromuscular weakness, or any prolonged immobilization, no randomized controlled trials have proven its effectiveness. Preoperative inspiratory muscle training may reduce atelectasis in patients undergoing upper abdominal surgery,11 and prophylactic use of noninvasive ventilation may reduce pulmonary dysfunction after lung resection surgery. Other modalities such as positive expiratory pressure devices and high-frequency chest wall oscillation airway clearance are of uncertain benefit. Patchy atelectasis is treated by addressing the underlying disease process in the lung parenchyma. Compression atelectasis is treated by alleviating the pleural space process. Obstructive or resorptive atelectasis often requires bronchoscopy for diagnosis and treatment. In patients with obstruction owing to retained secretions, multiple bronchoscopies are sometimes required, but the mucus often rapidly reaccumulates and will resolve only when the patient’s overall status improves. Rounded atelectasis does not require treatment. CT is helpful in distinguishing rounded atelectasis from parenchymal tumor.

  CONGENITAL CYSTIC DISEASES OF THE THORAX

Thoracic cysts, which are exceedingly rare, develop because of abnormal development or branching of the foregut. Cysts may develop in the mediastinum at an early stage of gestation or in the lung parenchyma at a later stage. Abnormalities include bronchogenic cysts (mediastinal and parenchymal), congenital pulmonary airway malformation, and pulmonary sequestrations. The cysts are lined with airway and alveolar epithelium but do not communicate in a normal fashion with the airways or lung tissue. Most patients with thoracic cysts present in childhood, but the cysts can remain asymptomatic and unnoticed until adulthood. In the absence of symptoms, these cystic lesions sometimes present as an incidental finding on chest imaging performed for another indication. Congenital cystic diseases can cause recurrent pneumonia, hemoptysis, or compression of normal structures. Computed tomography scanning with CT angiography can usually detect congenital cystic lesions of the thorax, but pulmonary or bronchial angiography is sometimes necessary to define the blood flow to the lesion. Bronchogenic cysts are usually found in the right paratracheal or subcarinal areas of the mediastinum but are occasionally seen in the lung parenchyma.12 These cysts are often asymptomatic, but they can cause wheezing, dyspnea, and cough when they compress adjacent structures. Secondary infection may develop in the cysts, and there are a few case reports of malignant transformation. Complete surgical resection is generally recommended, but partial excision with de-epithelization of the cysts has also been performed. Observation is also an option when the cysts are asymptomatic.

Congenital pulmonary airway malformation, previously called congenital cystic adenomatoid malformation of the lung, is an exceedingly rare abnormality with reported incidence of one in every 25,000 to 35,000 pregnancies. The abnormality is caused by arrested development of the bronchial tree. Most patients are diagnosed prenatally by ultrasonography, but a few adults have first presented with complications, including pneumothorax and air embolism. Treatment is anatomic surgical resection. Pulmonary sequestrations are areas of nonfunctioning pulmonary parenchyma with no communication to the tracheobronchial tree and abnormal arterial supply and venous drainage (Fig. 90-6). Intralobar sequestration, which accounts for about 75% of cases, does not have visceral pleura and is generally found in a lower lobe, the left more frequently than the right. Extralobar sequestrations have their own visceral pleura, are separate from the normal lobes, and may even be found below the diaphragms. Sequestrations usually have a feeding vessel that arises from the aorta. Patients with sequestrations may be asymptomatic but sometimes develop recurrent infections and or hemoptysis. Surgical excision with special care for the management of the feeding vessel is curative. Embolization of the feeding vessel is sometimes a successful treatment option. Hyperlucent lungs are diagnosed by a paucity of vascular and interstitial markings noted on chest imaging. Lung parenchymal air collections can be caused by congenital parenchymal cysts, congenital lobar emphysema (almost exclusively diagnosed in infancy), giant bullous emphysema (vanishing lung syndrome), or Swyer-James syndrome. Lung parenchymal cysts may be a bullous alveolar type or may contain bronchial wall elements such as cartilage, smooth muscle, and glands. They may become infected and may rupture to cause pneumothorax. Surgical resection is generally recommended unless the lesions are small. Congenital lobar emphysema, otherwise known as congenital large hyperlucent lobe, may cause severe respiratory distress in infants owing to compression of surrounding lung tissue. Giant bullous emphysema is a rare condition that usually affects the upper lobes of young male smokers. Compression of normal lung parenchyma from these overdistended lobes may require surgical resection. Swyer-James-Macleod syndrome, which is characterized by unilateral lucency of an entire lung, is caused by childhood bronchiolitis obliterans owing to viral or bacterial infection or toxic inhalation. CT shows air trapping and hyperlucency of the affected lung, with a normal contralateral lung. No therapy is required.

Grade A References A1. Patterson JE, Hewitt O, Kent L, et al. Acapella versus “usual airway clearance” during acute exacerbation in bronchiectasis: a randomized crossover trial. Chron Respir Dis. 2007;4:67-74. A2. Kellett F, Robert NM. Nebulised 7% hypertonic saline improves lung function and quality of life in bronchiectasis. Respir Med. 2011;105:1831-1835. A3. Wilkinson M, Sugumar K, Milan SJ, et al. Mucolytics for bronchiectasis. Cochrane Database Syst Rev. 2014;CD001289. A4. Martinez-Garcia MA, Soler-Cataluna JJ, Catalan-Serra P, et al. Clinical efficacy and safety of budesonide-formoterol in non-cystic fibrosis bronchiectasis. Chest. 2012;141:461-468. A5. Altenburg J, de Graaff CS, Stienstra Y, et al. Effect of azithromycin maintenance treatment on infectious exacerbations among patients with non-cystic fibrosis bronchiectasis: the BAT randomized controlled trial. JAMA. 2013;309:1251-1259.

A6. Serisier DJ, Martin ML, McGuckin MA, et al. Effect of long-term, low-dose erythromycin on pulmonary exacerbations among patients with non-cystic fibrosis bronchiectasis: the BLESS randomized controlled trial. JAMA. 2013;309:1260-1267. A7. Brodt AM, Stovold E, Zhang L. Inhaled antibiotics for stable non-cystic fibrosis bronchiectasis: a systematic review. Eur Respir J. 2014;44:382-393. A8. Haworth CS, Foweraker JE, Wilkinson P, et al. Inhaled colistin in patients with bronchiectasis and chronic Pseudomonas aeruginosa infection. Am J Respir Crit Care Med. 2014;189:975-982.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

GENERAL REFERENCES 1. McShane PJ, Naureckas ET, Tino G, et al. Non-cystic fibrosis bronchiectasis. Am J Respir Crit Care Med. 2013;188:647-656. 2. Seitz AE, Olivier KN, Adjemian J, et al. Trends in bronchiectasis among Medicare beneficiaries in the United States, 2000 to 2007. Chest. 2012;142:432-439. 3. Moulton BC, Barker AF. Pathogenesis of bronchiectasis. Clin Chest Med. 2012;33:211-217. 4. Bienvenu T, Sermet-Gaudelus I, Burgel PR, et al. Cystic fibrosis transmembrane conductance regulator channel dysfunction in non-cystic fibrosis bronchiectasis. Am J Respir Crit Care Med. 2010; 181:1078-1084. 5. Tunney MM, Einarsson GG, Wei L, et al. Lung microbiota and bacterial abundance in patients with bronchiectasis when clinically stable and during exacerbation. Am J Respir Crit Care Med. 2013; 187:1118-1126. 6. Knowles MR, Daniels LA, Davis SD, et al. Primary ciliary dyskinesia. Recent advances in diagnostics, genetics, and characterization of clinical disease. Am J Respir Crit Care Med. 2013;188: 913-922.

571.e1

7. Pasteur MC, Bilton D, Hill AT. British Thoracic Society guideline for non-CF bronchiectasis. Thorax. 2010;65(suppl 1):i1-i58. 8. Flude LJ, Agent P, Bilton D. Chest physiotherapy techniques in bronchiectasis. Clin Chest Med. 2012;33:351-361. 9. Loebinger MR, Wells AU, Hansell DM, et al. Mortality in bronchiectasis: a long-term study assessing the factors influencing survival. Eur Respir J. 2009;34:843-849. 10. Chalmers JD, Goeminne P, Aliberti S, et al. The bronchiectasis severity index. An international derivation and validation study. Am J Respir Crit Care Med. 2014;189:576-585. 11. Restrepo RD, Wettstein R, Wittnebel L, et al. Incentive spirometry: 2011. Respir Care. 2011; 56:1600-1604. 12. Cilleruelo Ramos A, Ovelar Arribas Y, Garcia Yuste M. Cervical bronchogenic cyst in adults. case report and literature review. Arch Bronconeumol. 2015;51:95-96.

571.e2

CHAPTER 90  Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders  

REVIEW QUESTIONS 1. Of the following, which organism is associated with the worst prognosis in non–cystic fibrosis bronchiectasis? A. Staphylococcus aureus B. Escherichia coli C. Streptococcus pneumoniae D. Pseudomonas aeruginosa E. Moraxella catarrhalis Answer: D  Pseudomonas aeruginosa is the correct answer. Molecular techniques have demonstrated that diverse polymicrobial communities are present in the lungs of patients with bronchiectasis, both when they are clinically stable as well as during exacerbations. The dominant organisms are P. aeruginosa and Haemophilus influenzae, but anaerobic organisms may also be detected. 2. Of the following, which organism found in your patient with bronchiectasis would raise the suspicion for undiagnosed cystic fibrosis? A. Pseudomonas aeruginosa B. Staphylococcus aureus C. Mycobacterium avium complex D. Streptococcus pneumoniae E. Acinetobacter baumannii Answer: B  The presence of Staphylococcus aureus raises the suspicion for cystic fibrosis, and further testing to establish or exclude this diagnosis is indicated.

3. Which imaging technique is used to confirm the diagnosis of bronchiectasis? A. Routine chest radiography B. Noncontrast high-resolution CT scan C. CT of the chest with contrast D. MRI of thorax E. Bronchography Answer: B  High-resolution computed tomography of the chest without the need for contrast is the imaging test of choice for confirming bronchiectasis 4. Which of the following is most predictive of poor prognosis in patients with bronchiectasis? A. Chronic infection with Mycobacterium avium complex B. Total lung capacity less than 70% predicted C. Admission to the ICU for respiratory failure D. Chronic infection with Staphylococcus aureus E. Dependence on supplemental oxygen Answer: C  When a patient with bronchiectasis has developed respiratory failure, the long-term prognosis is poor. More than 60% of patients admitted to a medical intensive care unit for bronchiectasis will die during that hospitalization.

571

CHAPTER 91  Alveolar Filling Disorders  

usually absent. If the initial chest radiograph or chest computed tomography (CT) is consistent with a possible alveolar filling process (Chapter 84) and acute pneumonia and pulmonary edema are excluded, bronchoscopy with bronchoalveolar lavage (BAL) (Chapter 85) and transbronchial biopsy should be performed, particularly if pulmonary alveolar proteinosis, invasive mucinous adenocarcinoma, lepidic predominant nonmucinous adenocarcinoma (Chapter 191), or alveolar hemorrhage is suspected. When these tests are nondiagnostic, and in many cases of suspected acute interstitial pneumonia, a surgical lung biopsy obtained by thoracoscopy or an open surgical procedure may be indicated.

  PULMONARY ALVEOLAR PROTEINOSIS

91  ALVEOLAR FILLING DISORDERS

EPIDEMIOLOGY

Pulmonary alveolar proteinosis is a rare alveolar filling disease caused by the accumulation of phospholipoproteinaceous surfactant material in the alveoli. The incidence is estimated to be 3.7 cases per million people. Pulmonary

STEPHANIE M. LEVINE TABLE 91-1  ALVEOLAR FILLING DISORDERS PATHOPHYSIOLOGY

RADIOGRAPHIC FINDINGS

Pulmonary alveolar proteinosis

Impaired processing of surfactant by alveolar macrophages caused by defects in GM-CSF signaling

Bilateral alveolar opacities with “crazy paving” and diffuse areas of ground-glass attenuation on CT scan

Acute interstitial pneumonia

Diffuse alveolar damage with temporal uniformity

Diffuse alveolar filling process similar to the acute respiratory distress syndrome

Diffuse alveolar hemorrhage

Bleeding from the pulmonary microcirculation, usually from the capillaries

Acute development of bilateral alveolar opacities

Invasive mucinous adenocarcinoma and lepidic predominant nonmucinous adenocarcinoma (formerly called bronchioloalveolar cell carcinoma)

Cancer cells growing along the alveolar septa

Pneumonic opacities, consolidation with air bronchograms, ground-glass opacities (either solitary or multiple)

DISEASES

DEFINITION

Alveolar filling disorders (Table 91-1) are characterized by chest radiographic findings of alveolar involvement ranging from a ground-glass appearance to consolidation; the pathologic process shows primary involvement of the alveolar air spaces distal to the terminal bronchioles. For example, in pulmonary alveolar proteinosis, the alveoli are filled by proteinaceous fluid. By comparison, the alveolar walls are lined by adenocarcinoma cells in invasive mucinous adenocarcinoma and lepidic predominant nonmucinous adenocarcinoma, formerly called bronchioloalveolar cell cancer. In acute interstitial pneumonia, exudative organizing fibroproliferative infiltrates fill the alveolar space; in the alveolar hemorrhage disorders, blood fills the alveolar space. Alveolar spaces filled with acute inflammatory cells, as in bacterial pneumonia (Chapter 97); water, as in cardiogenic or hydrostatic pulmonary edema (Chapter 58); or high-protein fluid, as in noncardiogenic or increased permeability pulmonary edema (Chapter 104), are also part of the radiographic differential diagnosis of alveolar filling disorders and must be excluded. A general approach to these suspected alveolar filling diseases (Fig. 91-1) can be stratified by the time elapsed since the onset of symptoms. The typical patient may present with the onset of cough (usually dry) and dyspnea of variable duration, depending on the disease process. Hemoptysis is a frequent presenting symptom in the alveolar hemorrhagic disorders. With the exception of acute interstitial pneumonia, symptoms suggesting an acute infectious process such as fever, leukocytosis, and productive cough are

CT = computed tomography; GM-CSF = granulocyte-macrophage colony-stimulating factor.

Cough, dyspnea, alveoar infiltrates Exclude pulmonary edema* Exclude infectious pneumonia†

Bronchoscopy

BAL without infectious etiology

BAL with progressively more bloody retum

BAL and TBBX with malignant adenocarcinoma cells

BAL with return of milky fluid BAL and TBBX show PASpositive material

Possible acute interstitial pneumonia: Symptom duration: days to weeks; features of ARDS with no obvious cause unless fever is present

Diffuse alveolar hemorrhage: Symptom duration: hours to days; hemoptysis in 70%; anemia

Invasive mucinous adenocarcinoma and lepidic-predominant nonmucinous adenocarcinoma (bronchioloalveolar cell cancer): Symptom duration: weeks to months, bronchorrhea may be present

Pulmonary alveolar proteinosis: Symptom duration: weeks to months

Confirm with surgical lung biopsy

FIGURE 91-1.  A general approach to the alveolar filling disorders. *See Chapter 58. †See Chapter 97. ARDS = acute respiratory distress syndrome; BAL = bronchoalveolar lavage; PAS = periodic acid–Schiff; TBBX = transbronchial biopsy.

572

CHAPTER 91  Alveolar Filling Disorders  

alveolar proteinosis is a primary acquired, autoimmune disorder in more than 90% of cases, but similar histopathologic features may be found with identifiable secondary causes, such as acute silicosis (silicoproteinosis; Chapter 93), aluminum dust exposure (Chapter 93), indium dust exposure, immunodeficiency disorders (e.g., immunoglobulin G monoclonal gammopathy and severe combined immunodeficiency syndrome), hematologic malignant neoplasms (particularly myeloid leukemias; Chapters 183 and 184), and certain infections (e.g., Pneumocystis jiroveci pneumonia). Pulmonary alveolar proteinosis has also been described after bone marrow transplantation (Chapter 178). A congenital form also usually presents in infancy.

PATHOBIOLOGY

The pathogenesis of pulmonary alveolar proteinosis is related to impaired processing of surfactant by alveolar macrophages caused by defects in granulocyte-macrophage colony-stimulating factor (GM-CSF) signaling. This impairment may be caused by autoantibodies against GM-CSF or GM-CSF receptor gene mutations, which are found in 90% of cases. The secondary forms are caused by a relative GM-CSF deficiency, leading to macrophage dysfunction and reduced surfactant clearance. The autosomal recessive congenital form of pulmonary alveolar proteinosis, caused by a mutation in the genes encoding surfactant protein B or C, or the GM-CSF receptor results in abnormal surfactant function and severe respiratory distress in hom*ozygous infants. The result of this impairment is accumulation of surfactant-rich material and progressive dysfunction in phagocytosis caused by excessive production or diminished clearance of surfactant by alveolar macrophages. Histologic examination in pulmonary alveolar proteinosis reveals alveoli filled with lipoproteinaceous material that stains pink (positive reaction) with periodic acid–Schiff stain. Classically, there is no destruction of alveolar architecture. Electron microscopy reveals lamellar (phospholipid-containing) myelin bodies.

CLINICAL MANIFESTATIONS

Pulmonary alveolar proteinosis presents in patients in the third to fourth decade of life with a 2 : 1 male predominance. Most patients (72%) are smokers. Patients present with the insidious onset of dyspnea and cough, which may be dry or occasionally productive of grayish material. The duration of symptoms before diagnosis is typically 6 weeks to 6 to 8 months. Low-grade fevers, malaise, and weight loss may also be present. Hemoptysis is unusual. On physical examination, rales are present in 50% of cases. Clubbing is an unusual finding until later stages of disease.

E-Fig. 91-E1).1 This radiographic pattern is not specific for this disorder and can be seen with acute respiratory distress syndrome (ARDS; Chapter 104), P. jiroveci pneumonia (Chapter 341), adenocarcinomas formerly described as bronchioloalveolar carcinoma (Chapter 191), lipoid pneumonia (Chapter 94), sarcoidosis (Chapter 95), organizing pneumonia, drug reactions, and pulmonary hemorrhage as well as with cardiogenic pulmonary edema (Chapter 59) and acute interstitial pneumonias. Pulmonary function tests often, but not always, show a restrictive pattern with a reduced diffusing capacity. Arterial blood gas analyses reveal hypoxemia. Bronchoscopy should be the initial procedure when pulmonary alveolar proteinosis is suspected. The diagnosis of pulmonary alveolar proteinosis can be established in most cases by the recovery of milky white to sandy-colored or light brown fluid on BAL. When it is subjected to cytologic analysis, the BAL fluid has a positive reaction on periodic acid–Schiff staining and reveals alveolar macrophages filled with positive staining material. Transbronchial biopsy or thoracoscopic biopsy can confirm the diagnosis by providing tissue that has similar staining characteristics. Serologic analysis for GM-CSF antibodies is now often performed to support the diagnosis.

TREATMENT  About 8% to 30% of cases of pulmonary alveolar proteinosis resolve spontaneously, and smoking cessation may contribute to spontaneous resolution. A second group of patients will progress to respiratory failure. The remainder will have stable disease. Superinfection with Nocardia spp., atypical mycobacteria, fungi, and other opportunistic organisms can occur in more than 15% of patients as a result of macrocyte phagocytic dysfunction.2 Therapy depends on the severity of symptoms.3 Treatment of severely symptomatic patients with dyspnea and hypoxemia begins with multistage or sequential whole-lung lavage performed under general anesthesia with a double-lumen endotracheal tube. The recovered fluid initially has an opaque, sandy-colored appearance (E-Fig. 91-E2). This procedure may have to be repeated at variable intervals. Asymptomatic patients should be observed and followed closely. Mildly symptomatic patients should receive supportive therapy with supplemental oxygen as needed. Subcutaneous or inhaled GM-CSF can improve quality of life, oxygenation, pulmonary function, and exercise capacity in about half of treated patients.4,5 Treatment with GM-CSF is usually reserved for patients who cannot undergo whole-lung lavage or who have failed standard treatment with whole-lung lavage. Isolated case series have reported variable responses to CD-20 monoclonal antibody, rituximab, or plasmapheresis. Corticosteroids are not routinely used. Lung transplantation can be performed, but recurrent pulmonary alveolar proteinosis has been reported. Survival rates at 5 years approach 75%.

DIAGNOSIS

Mildly elevated leukocyte counts and mildly to moderately elevated lactate dehydrogenase (LDH) levels may be found in more than 80% of patients; LDH levels may correlate with the severity of disease. Polycythemia and hypergammaglobulinemia may also be present. The chest radiograph (Fig. 91-2) and chest CT scans demonstrate a diffuse symmetrical alveolar filling process with predominance in the lower two thirds of the lung fields; the radiographic appearance may mimic pulmonary edema. The characteristic CT pattern is often described as “crazy paving,” which is attributable to scattered or diffuse areas of ground-glass attenuation with thickening of intralobular structures and interlobular septa in polygonal shapes (Fig. 91-3 and

FIGURE 91-2.  A chest radiograph showing bilateral alveolar opacities in a patient with pulmonary alveolar proteinosis.

  ACUTE INTERSTITIAL PNEUMONIA DEFINITION

Acute interstitial pneumonia, also referred to as the Hamman-Rich syndrome, is a rare and often fatal disease that mimics ARDS (Chapter 104). The etiology is unknown, and acute interstitial pneumonia is sometimes defined as

FIGURE 91-3.  A chest computed tomography scan showing the “crazy paving” pattern characteristic of pulmonary alveolar proteinosis.

CHAPTER 91  Alveolar Filling Disorders  

572.e1

E-FIGURE 91-1.  A chest computed tomography scan showing the “crazy paving” pattern characteristic of pulmonary alveolar proteinosis. E-FIGURE 91-2.  Whole-lung lavage from a patient with pulmonary alveolar proteinosis revealing sandy-colored fluid.

CHAPTER 91  Alveolar Filling Disorders  

the development of ARDS in the absence of known triggers.6 A similar acute presentation may be seen in patients with an acute exacerbation of idiopathic pulmonary fibrosis (Chapter 92), but most investigators believe that acute interstitial pneumonia is a separate disease process with no histologic evidence of underlying usual interstitial pneumonia.

PATHOBIOLOGY

The pathogenesis of acute interstitial pneumonia is damage to the epithelium of the alveolar membranes by a neutrophil-mediated mechanism; the result is pouring of exudate into the air space in the initial exudative phase of disease. Histologic examination reveals diffuse alveolar damage with intraalveolar hyaline membrane formation, interstitial and intraalveolar edema, acute inflammation, and epithelial cell necrosis with a nonspecific distribution and temporal uniformity. This process progresses to the organizing phase, characterized by alveolar septal thickening, type II pneumocyte hyperplasia, and fibroblast proliferation along the interstitium and alveolar spaces. In situ thrombi of small pulmonary arteries may be present. Finally, a fibrotic phase occurs with alveolar septal thickening from organizing fibrosis. One of the key pathologic findings in acute interstitial pneumonia is the temporal uniformity of the diffuse alveolar damage and of organizing and proliferating connective tissue. This uniformity supports a single acute injury at a particular point in time. Long-standing fibrosis is not a typical pathologic finding in acute interstitial pneumonia.

573

but without a clear etiology. The differential diagnosis histologically and clinically includes other causes of ARDS (Chapter 104), such as severe infection, trauma, and sepsis, as well as other causes of acute lung injury (Chapter 94), such as drug toxicity, inhalation injury, and collagen vascular diseases. The presentation is clinically and radiographically similar to that of diffuse alveolar hemorrhage, acute hypersensitivity pneumonitis, acute exacerbation of pulmonary fibrosis, acute eosinophilic pneumonia, and cryptogenic organizing pneumonia. Bronchoscopy with BAL is often performed to exclude alveolar hemorrhage, eosinophilic pneumonias, and infectious causes of lung injury. In a small number of cases, transbronchial biopsy may yield the diagnosis, but definitive diagnosis in most cases of acute interstitial pneumonia requires a surgical lung biopsy revealing diffuse alveolar damage.

TREATMENT  Treatment includes supportive intensive care unit management. In small case series, corticosteroids at doses of 1 to 2 g of methylprednisolone in divided doses intravenously per day for 3 consecutive days followed by prednisone or equivalent at 1 mg/kg/day with a taper during several weeks to months, with or without cyclophosphamide, may be of benefit, but the mortality rate remains higher than 70%. Patients also can have recurrences in months to years. Some cases of acute interstitial pneumonia may resolve without sequelae, but in some series, more than 50% of survivors may be left with residual fibrosis.

CLINICAL MANIFESTATIONS

Acute interstitial pneumonia manifests with equal frequency in men and women, typically in previously healthy individuals in the 50- to 55-year age range. It develops acutely to subacutely during a few days to a few weeks. The mean duration of symptoms is 15 days. Dry cough, shortness of breath, malaise, and fever (in 50% of patients) are typical clinical findings. A viruslike prodrome period has been described. Pulmonary rales are heard on physical examination, and hypoxemia is characteristic. Clubbing is rare. Acute interstitial pneumonia often progresses to hypoxemic ventilatory failure, and intensive care unit admission with mechanical ventilation is usually required. Early mortality rates are high. Radiographic features of acute interstitial pneumonia are diffuse alveolar opacities and air space consolidation similar to the appearance of ARDS (Fig. 91-4); CT scans reveal bilateral air space consolidation with areas of ground-glass opacities with little honeycombing. Septal thickening and a subpleural distribution of the opacities may also be present.

DIAGNOSIS

The diagnosis of acute interstitial pneumonia is made in the appropriate clinical setting in a patient who has a clinical presentation compatible with ARDS

  DIFFUSE ALVEOLAR HEMORRHAGE DEFINITION

The alveolar hemorrhage syndromes cause alveolar filling disease, usually with an acute onset and often with life-threatening severity. They can be associated with ANCA- (antineutrophil cytoplasmic antibody–) associated vasculitides,7 such as microscopic polyangiitis (Chapter 270) and granulomatosis with polyangiitis (Chapter 270); immunologic diseases, such as Goodpasture syndrome (anti–glomerular basem*nt membrane antibody disease; Chapter 121); collagen vascular diseases, such as systemic lupus erythematosus (Chapter 266); cocaine inhalation (Chapter 34); drugs (including penicillamine, mitomycin C, trimellitic anhydride, all-trans retinoic acid, propylthiouracil, and isocyanates); bone marrow transplantation (Chapter 178); coagulopathy (Chapter 174); and mitral stenosis (Chapter 75). A small percentage of idiopathic and recurrent cases are termed idiopathic pulmonary hemosiderosis. In Goodpasture syndrome, there is a strong association with tobacco use and a male predominance, with young men most frequently affected. A viral syndrome and exposure to hydrocarbons may simulate Goodpasture disease. Idiopathic pulmonary hemosiderosis most often occurs in children and young adults.

PATHOBIOLOGY

Alveolar hemorrhage is caused by bleeding from the pulmonary microcirculation, including the capillaries, arterioles, and venules. It may be associated with injury or neutrophilic inflammation of the alveolar walls and adjacent interstitial capillaries or with a capillaritis, usually when it is associated with collagen vascular or vasculitic processes. In Goodpasture syndrome, for example, the circulating anti–glomerular basem*nt membrane antibodies are directed against the α3 chain of type IV collagen in the glomerular basem*nt membrane, where they cause glomerulonephritis; these core antibodies can cross-react with the alveolar capillary basem*nt membranes, resulting in alveolar hemorrhage. Alternatively, alveolar hemorrhage may be associated with relatively bland pathologic changes with red blood cells in the alveolar spaces. Idiopathic pulmonary hemosiderosis is an example of bland hemorrhage.

CLINICAL MANIFESTATIONS

FIGURE 91-4.  A chest radiograph showing bilateral alveolar opacities in a patient with acute interstitial pneumonia.

Patients present acutely (usually in hours to a week) with dyspnea, shortness of breath, hemoptysis (which may not be present in all patients), and cough. Some patients also have low-grade fever. Lung examination reveals rales. Laboratory examination may reveal anemia, and arterial blood gases reveal hypoxemia. In Goodpasture syndrome and the ANCA-associated vasculitides, hematuria and renal insufficiency caused by glomerulonephritis are typically present. Radiographic features include the acute development of bilateral alveolar filling disease similar to pulmonary edema but without cardiomegaly or

574

CHAPTER 91  Alveolar Filling Disorders  

mide (2 mg/kg/day orally or 0.5 g/m2-0.75 g/m2 intravenously for one dose) and then if needed, every 2 weeks or change to oral therapy. Plasmapheresis (3-14 exchanges) to remove the offending circulating antibody is a mainstay therapy for Goodpasture syndrome and may also be used in some cases of alveolar hemorrhage from ANCA-associated vasculitis and systemic lupus erythematosus. In patients with ANCA-associated vasculitis, rituximab (375 mg/ m2 once a week for 4 weeks) is at least as good as cyclophosphamide followed by azathioprine for 12 to 15 months for inducing and maintaining remission (≈50% at 12 months and about 40% at 18 months), A1  and rituximab (at 500 mg on days 0 and 14 and at 6, 12, and 18 months) is better than azathioprine for maintaining remission at 28 months. A2 

PROGNOSIS

Alveolar hemorrhage can result in acute respiratory failure and death. Recurrent alveolar hemorrhage from any cause, such as idiopathic pulmonary hemosiderosis, can be associated with the development of pulmonary fibrosis. Alveolar hemorrhage related to collagen vascular disease, vasculitides, and idiopathic pulmonary hemosiderosis can have mortality rates ranging from 25% to 50%. With Goodpasture syndrome, renal failure is common, and the degree of renal impairment may correlate with outcome.

  INVASIVE MUCINOUS ADENOCARCINOMA AND LEPIDIC PREDOMINANT NONMUCINOUS ADENOCARCINOMA (FORMERLY CALLED BRONCHIOLOALVEOLAR CELL CARCINOMA) FIGURE 91-5.  A chest computed tomography scan showing alveolar opacities in a patient with diffuse alveolar hemorrhage.

pleural effusions (E-Fig. 91-E3). Rapid remission and recurrences are seen with repeated episodes of bleeding, which also may result in chronic interstitial changes on the chest radiograph. Pulmonary function testing may reveal an increase in the diffusion capacity for carbon monoxide because of the presence of hemoglobin in the alveolar spaces.

DIAGNOSIS

The diagnosis of alveolar hemorrhage is usually made in the appropriate clinical setting by the triad of diffuse alveolar opacities (Fig. 91-5),8 hemoptysis (in two thirds of patients), and anemia. BAL typically demonstrates the return of progressively more bloody aliquots of fluid (E-Fig. 91-E4), and cytologic analysis reveals that more than 20% of the macrophages are hemosiderin laden. Goodpasture syndrome is diagnosed by circulating anti– glomerular basem*nt membrane antibodies, which are present in more than 90% of patients, or by the demonstration of linear deposition of immunoglobulin G antibodies along the alveolar or renal capillary basem*nt membrane tissue when it is viewed by direct immunofluorescence. c-ANCA(cytoplasmic antineutrophil cytoplasmic antibody–) associated vasculitis causes a focal, segmental, necrotizing glomerulonephritis and is associated with the presence of proteinase 3 antineutrophilic cytoplasmic antibodies in 90% of active cases of granulomatosis with polyangiitis (Chapter 270). Necrotizing granulomatous inflammation is often found in the upper airway in addition to the lungs and kidneys. A perinuclear myeloperoxidase antineutrophilic antibody is often present in association with microscopic polyarteritis (Chapter 270). Patients with systemic lupus erythematosus usually have antinuclear antibodies (Chapter 266). Patients using illicit drugs such as cocaine may have a positive drug screen. Idiopathic pulmonary hemosiderosis is a diagnosis of exclusion after other causes of diffuse alveolar hemorrhage have been eliminated.

TREATMENT  Treatment of alveolar hemorrhage varies according to its underlying cause. Massive hemoptysis (Chapter 83) from any cause of alveolar hemorrhage should be managed as needed. In the case of anticoagulation-, drug-, or toxinrelated alveolar hemorrhage, the offending agent should be withdrawn, and supportive care is indicated. In Goodpasture syndrome, the ANCA-associated vasculitides, and other vasculitides (Chapter 270), treatment typically includes immunosuppressant agents such as corticosteroids (methylprednisolone, 5002000 mg/day in divided doses for 3-5 days followed by a prednisone taper beginning at 1 mg/kg/day during the next 6 to 9 months) and cyclophospha-

DEFINITION

In 2011, a multidisciplinary committee eliminated the term bronchioloalveolar cell carcinoma and divided pulmonary adenocarcinomas into five types: adenocarcinoma in situ, minimally invasive adenocarcinoma, lepidic predominant nonmucinous adenocarcinoma, invasive mucinous adenocarcinoma, and invasive adenocarcinoma and its subtypes.9 The types that would most likely be correlated with an alveolar filling appearance pathologically and on chest imaging are invasive mucinous adenocarcinoma, in which consolidation and air-bronchograms may be present, and lepidic-predominant nonmucinous adenocarcinoma, in which a ground-glass appearance is characteristic. Both of these types of adenocarcinoma are among those formally characterized as bronchioloalveolar cell carcinoma. In general, these tumors are characterized by malignant cells lining the alveolar cell wall (Chapter 191). Among bronchogenic carcinomas, these subtypes are the least associated with tobacco use, and patients with these cancers are more likely to be nonsmokers. Unlike other non–small cell lung cancers, the sex ratio approaches 1 : 1 or may be slightly female predominant, and younger patients may be affected.

PATHOBIOLOGY

These types of adenocarcinoma usually arise in the periphery of the lung and may be characterized by lepidic growth, which means contiguous growth along the intact alveolar septa, with varying degrees of stromal, pleural, vascular, or lymphatic invasion and without a known primary adenocarcinoma elsewhere. The mucinous type is thought to derive from respiratory goblet cells and columnar cells, and the nonmucinous type from type II pneumocytes or Clara cells.

CLINICAL MANIFESTATIONS

Patients present with a gradual onset of shortness of breath and cough. The duration of symptoms is usually several months. Constitutional symptoms such as malaise and weight loss may be present. Hemoptysis may occur. An unusual but unique clinical finding is bronchorrhea, with patients reporting the production of copious amounts of clear sputum daily. This finding is more common in the invasive mucinous form of the disease.

DIAGNOSIS

Radiographic patterns vary10 and can include localized disease with peripheral solitary or multiple nodules or masses in 60% of cases or a persistent pneumonic pattern in 40% of cases (E-Fig. 91-E5). The radiographic findings of consolidation with air bronchograms are often initially thought to be consistent with acute pneumonia, but the typical clinical presentation is that of a nonresolving peripheral density on chest radiograph.11 In addition, CT may show areas of ground-glass attenuation. Positron emission tomography may be normal because of the low glucose uptake of these tumors. The diagnosis

CHAPTER 91  Alveolar Filling Disorders  

E-FIGURE 91-3.  A chest radiograph showing bilateral alveolar opacities in a patient

with diffuse alveolar hemorrhage.

E-FIGURE 91-5.  A chest computed tomography scan in a patient with invasive mucinous adenocarcinoma.

574.e1

E-FIGURE 91-4.  Bronchoalveolar lavage fluid revealing a progressively increasingly bloody lavage consistent with diffuse alveolar hemorrhage

of invasive mucinous adenocarcinoma and lepidic predominant nonmucinous adenocarcinoma is most often made by bronchoscopy with transbronchial biopsy.

TREATMENT  For staging and treatment, these types of adenocarcinoma are approached like other types of non–small cell lung cancers (Chapter 191). Testing for epidermal growth factor receptor (EGFR) mutations should be performed, and chemotherapy planned accordingly. In general, the invasive mucinous adenocarcinomas are KRAS positive, and EGFR negative. The lepidic predominant nonmucinous type tends to be EGFR positive. Bilateral lung transplantation has been performed, but recurrence in the transplanted lungs has been reported.

PROGNOSIS

Prognosis correlates with disease stage and with the histologic and radiographic patterns. Patients who undergo surgical resection for adenocarcinoma in situ or minimally invasive adenocarcinoma with a single focus of disease have a better prognosis than patients with other adenocarcinomas of like stage, with the 5-year survival rate approaching 100%. More advanced forms likely have a prognosis similar to that of other adenocarcinomas.

Grade A References A1. Specks U, Merkel PA, Seo P, et al. Efficacy of remission-induction regimens for ANCA-associated vasculitis. N Engl J Med. 2013;369:417-427. A2. Guillevin L, Pagnoux C, Karras A, et al. Rituximab versus azathioprine for maintenance in ANCAassociated vasculitis. N Engl J Med. 2014;371:1771-1780.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 91  Alveolar Filling Disorders  

GENERAL REFERENCES 1. Ben-Dov I, Segel MJ. Autoimmune pulmonary alveolar proteinosis: clinical course and diagnostic criteria. Autoimmun Rev. 2014;13:513-517. 2. Punatar AD, Kusne S, Blair JE, et al. Opportunistic infections in patients with pulmonary alveolar proteinosis. J Infect. 2012;65:173-179. 3. Leth S, Bendstrup E, Vestergaard H, et al. Autoimmune pulmonary alveolar proteinosis: treatment options in year 2013. Respirology. 2013;18:82-91. 4. Khan A, Agarwal R, Aggarwal AN. Effectiveness of granulocyte-macrophage colony-stimulating factor therapy in autoimmune pulmonary alveolar proteinosis: a meta-analysis of observational studies. Chest. 2012;141:1273-1283. 5. Tazawa R, Inoue Y, Arai T, et al. Duration of benefit in patients with autoimmune pulmonary alveolar proteinosis after inhaled granulocyte-macrophage colony-stimulating factor therapy. Chest. 2014;145:729-737. 6. Mukhopadhyay S, Parambil JG. Acute interstitial pneumonia (AIP): relationship to Hamman-Rich syndrome, diffuse alveolar damage (DAD), and acute respiratory distress syndrome (ARDS). Semin Respir Crit Care Med. 2012;33:476-485.

575.e1

7. West S, Arulkumaran N, Ind PW, et al. Diffuse alveolar haemorrhage in ANCA-associated vasculitis. Intern Med. 2013;52:5-13. 8. Lichtenberger JP 3rd, Digumarthy SR, Abbott GF, et al. Diffuse pulmonary hemorrhage: clues to the diagnosis. Curr Probl Diagn Radiol. 2014;43:128-139. 9. Travis WD, Brambilla E, Noguchi M, et al. International Association for the Study of Lung Cancer/ American Thoracic Society/European Respiratory Society: international multidisciplinary classification of lung adenocarcinoma: executive summary. Proc Am Thorac Soc. 2011;8:381-385. 10. Austin JH, Garg K, Aberle D, et al. Radiologic implications of the 2011 classification of adenocarcinoma of the lung. Radiology. 2013;266:62-71. 11. Lee HJ, Lee CH, Jeong YJ, et al. IASLC/ATS/ERS International Multidisciplinary Classification of Lung Adenocarcinoma: novel concepts and radiologic implications. J Thorac Imaging. 2012;27: 340-353.

575.e2

CHAPTER 91  Alveolar Filling Disorders  

REVIEW QUESTIONS 1. The correct treatment for patient with pulmonary alveolar proteinosis and hypoxemia is A. corticosteroids. B. lung transplantation. C. plasmapheresis. D. whole-lung lavage. E. azathioprine. Answer: D  Patients with pulmonary alveolar proteinosis and significant clinical findings, such as dyspnea and hypoxemia, should be treated with whole-lung lavage, which is the current standard therapy. 2. Bronchorrhea is a finding sometimes associated with A. diffuse alveolar hemorrhage. B. pulmonary alveolar proteinosis. C. invasive mucinous adenocarcinoma. D. cardiogenic pulmonary edema. E. ARDS. Answer: C  Bronchorrhea, which is the production of copious amounts of sputum that is usually clear and sometimes salty tasting, is seen in some cases of invasive mucinous adenocarcinoma. 3. Which of the following is best characterized radiographically as an alveolar filling disorder? A. Idiopathic pulmonary fibrosis B. Lymphangioleiomyomatosis C. Acute interstitial pneumonitis D. Lymphocytic interstitial pneumonitis E. Langerhans cell histiocytosis Answer: C  Acute interstitial pneumonia is characterized radiographically as an alveolar filling process. The other options have an interstitial pattern on chest imaging.

4. The most common type of pulmonary alveolar proteinosis is A. acquired or autoimmune. B. congenital C. associated with hematologic disease. D. associated with metal dust exposure. E. associated with respiratory infection. Answer: A  Autoimmune processes account for 90% of adult cases of pulmonary alveolar proteinosis, usually caused by GM-CSF antibodies. It also can be congenital or secondary to dust exposure or hematologic malignancy and is the most common form overall. 5. Plasmapheresis is part of the standard treatment for diffuse alveolar hemorrhage caused by which of the following diseases? A. Idiopathic pulmonary hemosiderosis B. Goodpasture syndrome C. Cocaine inhalation D. Autologous bone marrow transplantation E. Pneumococcal pneumonia Answer: B  Plasmapheresis along with corticosteroids and cyclophosphamide is part of the standard treatment of diffuse alveolar hemorrhage related to Goodpasture syndrome.

CHAPTER 92  Interstitial Lung Disease  

575

TABLE 92-1  CLINICAL CLASSIFICATION OF INTERSTITIAL LUNG DISEASE IDIOPATHIC INTERSTITIAL PNEUMONIAS Chronic Fibrosing Interstitial Pneumonias Idiopathic pulmonary fibrosis Nonspecific interstitial pneumonia Smoking-Related Interstitial Pneumonias Respiratory bronchiolitis–interstitial lung disease Desquamative interstitial pneumonia Acute or Subacute Idiopathic Interstitial Pneumonias Cryptogenic organizing pneumonia Acute interstitial pneumonia Lymphoid and lymphocytic interstitial pneumonia Rare interstitial Pneumonias Histologic pattern of acute fibrinous and organizing pneumonia Histological pattern of interstitial pneumonias with a bronchiolocentric distribution Pleuroparenchymal fibroelastosis INTERSTITIAL LUNG DISEASE ASSOCIATED WITH CONNECTIVE TISSUE DISEASE Progressive systemic sclerosis Rheumatoid arthritis Systemic lupus erythematosus Dermatomyositis and polymyositis Sjögren syndrome Mixed connective tissue disease Ankylosing spondylitis HYPERSENSITIVITY PNEUMONITIS Occupational and environmental factors (e.g., farmer’s lung; bird fancier’s lung) Iatrogenic DRUG-INDUCED AND IATROGENIC INTERSTITIAL LUNG DISEASE See Table 92-2 ALVEOLAR FILLING DISORDERS (Chapter 91)

92  INTERSTITIAL LUNG DISEASE GANESH RAGHU

Goodpasture syndrome Pulmonary alveolar proteinosis Pulmonary hemosiderosis Alveolar hemorrhage syndromes Chronic eosinophilic pneumonia INTERSTITIAL LUNG DISEASE ASSOCIATED WITH PULMONARY VASCULITIS Pulmonary capillaritis Granulomatosis with polyangiitis (formerly known as Wegener granulomatosis) Churg-Strauss syndrome OTHER SPECIFIC FORMS OF INTERSTITIAL LUNG DISEASE

DEFINITION

In an apparently immunocompetent host, interstitial lung disease (ILD) is a clinical term for a heterogenous group of acute and chronic lower respiratory tract disorders with many potential causes. However, clinical and physiological features common to all ILDs include exertional dyspnea, a restrictive pattern on pulmonary function testing (Chapter 85), coexistent airflow obstruction, decreased diffusing capacity (Dlco), increased alveolar-arterial oxygen difference (Pao2-Pao2) (Chapter 103) at rest or during exertion, and absence of pulmonary infection or neoplasm. ILDs comprise several acute and chronic lung disorders with variable degrees of pulmonary fibrosis (Table 92-1). The term interstitial is a misnomer because the pathologic processes are not restricted to the interstitium, which is the microscopic space bounded by the basem*nt membranes of epithelial and endothelial cells. Rather, all of the several cellular and soluble constituents that make up the gas exchange units (alveolar wall, capillaries, alveolar space, and acini) and the bronchiolar lumen, terminal bronchioles, and pulmonary parenchyma beyond the gas exchange units (as well as the pleura and lymphatics and sometimes the lymph nodes) are involved in the pathogenesis and manifestations of ILD.

EPIDEMIOLOGY

Among persons 18 years or older, the prevalence of all ILDs in the United States is about 81 per 100,000 men and 67 per 100,000 women. The overall incidence is also higher in men (31.5 per 100,000 per year) than in women (26.1 per 100,000 per year). Moreover, the prevalence of undiagnosed or early ILD is estimated to be 10 times that of clinically recognized disease; as

Sarcoidosis Langerhans cell histiocytosis (histiocytosis X) Lymphangioleiomyomatosis INHERITED FORMS OF INTERSTITIAL LUNG DISEASE Familial idiopathic pulmonary fibrosis Familial pulmonary fibrosis or interstitial pneumonia Tuberous sclerosis Neurofibromatosis Gaucher disease Niemann-Pick disease Hermansky-Pudlak syndrome

physicians’ awareness of these entities increases, it is expected that the frequency of the diagnosis of ILD will rise. Among the ILDs, the most common is idiopathic pulmonary fibrosis, which represents at least 30% of incident cases. In the United States, the annual incidence of idiopathic pulmonary fibrosis is 94 per 100,000 in the Medicare population, with a mean age of onset of 79 years.1 Recent data suggest that more than 5000 new cases are diagnosed each year in the United Kingdom.

PATHOBIOLOGY

Interstitial lung diseases are thought to result from an unknown tissue injury and attempted repair in the lung of a genetically predisposed person. Genetic variants within the hTERT or hTR components of the telomerase gene and surfactant protein gene have been associated in a subset of familial pulmonary fibrosis and in some sporadic cases.2 An MUC5B promotor polymorphism is

576

CHAPTER 92  Interstitial Lung Disease  

associated with familial interstitial pneumonia and idiopathic pulmonary fibrosis.3 In idiopathic pulmonary fibrosis, varying degrees of acute, subacute, and chronic fibroproliferation are present in the lungs at the time of diagnosis. Ultimately, progressive fibrosis results in honeycombing, an end-stage finding that is often associated with increased pulmonary vascular resistance and secondary pulmonary hypertension. As a reflection of these dynamic processes, histopathologic examination of lung tissue often reveals highly heterogeneous findings; for example, a single biopsy specimen may show normal alveoli adjacent to abnormal areas of inflammation and fibrosis, with or without granulomas, vasculitis, or secondary vascular changes within the pulmonary parenchyma.

CLINICAL MANIFESTATIONS

Interstitial lung diseases are typically characterized by progressive dyspnea. Nonproductive cough and fatigue are also common complaints. Pleuritic chest pain may occur with certain connective tissue or drug-induced ILDs, and acute pleuritic chest pain with dyspnea may represent a spontaneous pneumothorax (Chapter 99) in association with lymphangioleiomyomatosis, tuberous sclerosis (Chapter 417), neurofibromatosis, or Langerhans cell histiocytosis. Hemoptysis suggests a diffuse alveolar hemorrhagic syndrome, systemic lupus erythematosus (SLE) (Chapter 266), lymphangioleiomyomatosis, granulomatosis with polyangiitis (Chapter 270), or Goodpasture syndrome (Chapter 121); it is rare in other ILDs. In patients with existing ILD, new hemoptysis should prompt consideration of a superimposed malignancy, pulmonary embolus, or infection such as aspergillosis. In some patients, the first and the only clue to the presence of an ILD may be the finding of coarse rales (crackles) on auscultation of the lungs. These coarse crackles must be distinguished from the finer rales typical of heart failure (Chapter 58) or noncardiogenic pulmonary edema (Chapter 104). Unlike patients with obstructive lung disease, wheezes are not common. A history of wheezing suggests the coexistence of occult hyperactive airways and airflow obstruction and raises the possibility of allergic bronchopulmonary aspergillosis (Chapter 339), Churg-Strauss syndrome (Chapter 270), chronic eosinophilic pneumonia (see later), or parasitic infection (Chapter 344). In some patients, the initial presentation may be with peripheral cyanosis, clubbing, or the signs and symptoms of an underlying systemic disease (see later).

DIAGNOSIS

The first key in patients with an ILD is to establish the syndromic diagnosis and then pursue the differential diagnosis of its specific cause (Fig. 92-1). However, a conclusive cause often may not be identified despite an exhaustive medical history and invasive diagnostic interventions, including bronchoalveolar lavage (BAL)4 and sufficiently large and multiple lung biopsy specimens. Thus, the cause of several of the ILDs, even when diagnosed as specific entities, remains unknown.

History

The patient’s age, sex, and cigarette smoking history may provide useful clues to the diagnosis. Idiopathic pulmonary fibrosis is an adult disorder that usually occurs in patients older than 50 years. Pulmonary sarcoidosis (Chapter 95), in contrast, is more common in young adults and middle-aged persons. Whereas pulmonary Langerhans cell histiocytosis (previously known as pulmonary histiocytosis X or eosinophilic granuloma) characteristically occurs in young cigarette-smoking men, lymphangioleiomyomatosis occurs exclusively in women of childbearing age. Respiratory bronchiolitis– associated ILD is seen almost exclusively in cigarette smokers but occurs in both men and women of all ages. The medical history also should focus on environmental factors, especially changes in environmental exposures (including domestic, recreational, hot tub, whirlpool baths, indoor swimming pool, ventilation system at home, automobiles, and workplaces), occupational exposure, medications, and drug use (Chapters 93 and 94). A family medical history should address possible familial ILD. Environmental risk factors that may suggest the diagnosis of hypersensitivity pneumonitis include farming or exposure to overt or occult avian antigens at or in the close vicinity of home, bird droppings, feather duvets (“bird fancier’s lung” or “pigeon breeder’s lung”), visible molds, unkempt dusty homes, unchanged filters in furnaces, unkempt ventilatory systems, water leaks, or humidifiers in the domestic environment (hypersensitivity to thermophilic actinomycetes, Aureobasidium pullulans). At-risk occupations include mining (pneumoconioses), machine tool

Interstitial Lung Disease (immunocompetent host)

History (domestic and occupational environmental exposures, drugs, systemic/connective tissue disease, family medical history, etc.) and PE. Routine laboratory studies, pulmonary function tests, CXR, HRCT chest

Not idiopathic interstitial pneumonia (IIP) (e.g., obvious or known connective tissue disease, occupational exposure, drugs)

Diagnostic HRCT patterns (e.g., lymphangioleiomyomatosis, pulmonary Langerhans cell histiocytosis) Potential environmental cause/iatrogenic? No

Possible IIP

Yes

Typical HRCT pattern of UIP and features of idiopathic pulmonary fibrosis (IPF)? Yes

No When appropriate: • Muscle, kidney, fat, sinus biopsy • Specific serologies or biopsy of nonlung tissue to diagnose connective tissue or other systemic disease Bronchoalveolar lavage (BAL) Transbronchial lung biopsy (TLB)

No

IPF diagnosis

Specific systemic disease? Yes

Specific diagnosis

Yes

No further diagnostic workup

No Surgical lung biopsy

Histologic pattern usual interstitial pneumonia (UIP)? No IIP other than IPF (NSIP, RB-ILD, DIP, DAD, OP, LIP)

Yes IPF diagnosis

FIGURE 92-1.  An approach to interstitial lung disease. DAD = diffuse alveolar damage; DIP = desquamative interstitial pneumonia; HRCT = high-resolution computed tomography; IIP = idiopathic interstitial pneumonia; IPF = idiopathic pulmonary fibrosis; LIP = lymphoid interstitial pneumonia; NSIP = nonspecific interstitial pneumonia; OP = organizing pneumonia; PE = physical examination; PFT = pulmonary function test; RB-ILD = respiratory bronchiolitis–associated interstitial lung disease; TLB = transbronchial lung biopsy; UIP = usual interstitial pneumonia. (Adapted from American Thoracic Society/European Respiratory Society: International multidisciplinary consensus classification of idiopathic interstitial pneumonias. Am J Respir Crit Care Med. 2002;165:277304 and Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183:788-824.)

grinding, sandblasting and working with granite (silicosis), welding and working in a shipyard (asbestosis), and working in the aerospace or electronic industries (berylliosis) (Chapters 93 and 94). Because of the long interval between the exposure and the onset of symptoms in many occupations associated with ILD, it is important to take a lifelong occupational history

CHAPTER 92  Interstitial Lung Disease  

TABLE 92-2  DRUG-INDUCED AND IATROGENIC INTERSTITIAL LUNG DISEASE* ANTIMICROBIAL AGENTS Cephalosporins Isoniazid Nitrofurantoin Penicillins Sulfonamides ANTI-INFLAMMATORY AGENTS Aspirin Gold Methotrexate Nonsteroidal anti-inflammatory agents Penicillamine Phenylbutazone Zafirlukast CARDIOVASCULAR DRUGS Amiodarone Angiotensin-converting enzyme inhibitors β-Blockers Hydralazine Hydrochlorothiazide Procainamide Protamine sulfate Tocainide ANTINEOPLASTIC AND CHEMOTHERAPEUTIC AGENTS Bleomycin Busulfan Chlorambucil Cyclophosphamide Erlotinib Gefitinib Gemcitabine Imatinib Melphalan Mercaptopurine Methotrexate Mitomycin Nitrosoureas Procarbazine CENTRAL NERVOUS SYSTEM DRUGS Carbamazepine Chlorpromazine Imipramine Phenytoin ORAL HYPOGLYCEMIC AGENTS Chlorpropamide Tolazamide Tolbutamide ILLICIT DRUGS Cocaine Heroin Methadone Propoxyphene OTHER AGENTS Antithymocyte globulin All-trans-retinoic acid Colony-stimulating factors Interferon-α and -β Irradiation Mycophenolate mofetil Tumor necrosis factor-α modulating agents High fraction of inspired oxygen (Fio2) with mechanical ventilation *This list contains examples only and is not meant to be exhaustive.

(Chapter 19) as well as to establish the interval between exposure and the onset of symptoms. Because the list of medications known to cause ILD is long and continues to grow (Table 92-2), a careful history regarding recent use of prescription and over-the-counter products is essential. Risk factors for immunosuppression, including infection with human immunodeficiency virus, raise the possibility of opportunistic lung infections (Chapter 391), neoplasm (Chapter 191), and transplant-related pulmonary complications.

577

Particular attention should be paid to the onset and duration of symptoms; the rate of disease progression; and association with hemoptysis, fever, or extrathoracic symptoms. Symptoms that persist 4 weeks or less and the presence of fever suggest cryptogenic organizing pneumonia, drug-induced pulmonary injury, or acute hypersensitivity pneumonitis, BUT idiopathic pulmonary fibrosis, ILD associated with connective tissue diseases, and Langerhans cell histiocytosis tend to have a more subacute onset. Extrapulmonary symptoms suggest that the ILD may be associated with systemic disorders (e.g., sarcoidosis; Chapter 95), and symptoms such as dysphagia, dry eyes or mouth, skin rashes, or arthritis may suggest a connective tissue disorder (Chapters 266 to 270). Proximal muscle aches or weakness suggests the possibility of polymyositis or dermatomyositis (Chapter 269), and recurrent sinusitis suggests granulomatosis with polyangiitis (Chapter 270). Extrathoracic manifestations present in tuberous sclerosis (Chapter 417) include hematuria, epilepsy, and mental retardation.

Physical Examination

Physical examination of the respiratory system is rarely helpful in the diagnostic evaluation of ILD because findings such rhonchi and rales on auscultation and digital clubbing are nonspecific. Findings on cardiac examination, such as an accentuated P2, a right ventricular heave, or tricuspid insufficiency, are suggestive of pulmonary hypertension (Chapter 68) and cor pulmonale in patients with advanced lung disease. However, extrathoracic findings such as skin abnormalities, peripheral lymphadenopathy, and hepatosplenomegaly may be more specifically associated with underlying sarcoidosis (Chapter 95); muscle tenderness and proximal muscle weakness may point to coexisting polymyositis (Chapter 269); and signs of arthritis may indicate connective tissue disease (Chapters 264, 266, and 270) or sarcoidosis (Chapter 95). Characteristic rashes occur in several connective tissue diseases, disseminated Langerhans cell histiocytosis, tuberous sclerosis, and neurofibromatosis. Ophthalmologic findings (Chapter 423) such as iridocycl*tis, uveitis, or conjunctivitis, may be a clue to the diagnosis of sarcoidosis or a connective tissue disease, AND central nervous system abnormalities may be present in sarcoidosis, SLE, Langerhans cell histiocytosis, or tuberous sclerosis.

Laboratory Testing

Routine laboratory testing should include a complete blood count, leukocyte differential, erythrocyte sedimentation rate, chemistry panel (calcium, liver enzymes, electrolytes, creatinine), and urinalysis. Although these data rarely yield a specific diagnosis, they may provide helpful clues. Routine serology for occult connective tissue diseases (Chapter 257) may reveal typical findings of SLE (e.g., antinuclear antibodies), rheumatoid arthritis (rheumatoid factor, anticitrullinated peptide antibody), scleroderma (ScL 70), dermatomyositis or polymyositis (creatine kinase, aldolase, and anti–Jo-1 antibody), granulomatosis with polyangiitis (antineutrophil cytoplasmic antibodies), and Goodpasture syndrome (anti–basem*nt membrane antibodies). Mild hypoxemia is typically present on arterial blood gas analysis because of abnormal ventilation-perfusion ratios, especially in moderate to severe cases of ILD. However, carbon dioxide retention is rare and suggests possible coexisting emphysema (Chapter 88) or a hypoventilatory disorder (Chapter 86).

Noninvasive Evaluation Chest Radiograph

The distribution and appearance of radiographic abnormalities (Chapter 84) may prove useful in differentiating the clinicopathologic syndromes in patients with ILD (Table 92-3). Comparison of previous chest radiographs with the current one is important in establishing the rate of progression of the patient’s disease. A diffuse ground-glass pattern is often observed early in the course of ILD followed by progression to reticular (linear) infiltrates with nodules (reticulonodular infiltrates) or, in the case of alveolar filling disorders, ill-defined nodules (acinar rosettes) with air bronchograms. Most ILDs cause infiltrates in the lower lung zones, but upper lobe predominance is typically present in sarcoidosis, berylliosis, Langerhans cell histiocytosis, silicosis, chronic hypersensitivity pneumonitis, cystic fibrosis, and ankylosing spondylitis. The middle and lower lung zones show the most prominent abnormalities in lymphangitic carcinomatosis, idiopathic pulmonary fibrosis, subacute eosinophilic pneumonia, asbestosis, and pulmonary fibrosis caused by rheumatoid arthritis or progressive systemic sclerosis. Hilar adenopathy and mediastinal adenopathy are not common in ILDs; their presence should suggest sarcoidosis, berylliosis, silicosis, lymphocytic interstitial pneumonia (LIP), amyloidosis, or Gaucher disease. A pattern of peripherally located

578

CHAPTER 92  Interstitial Lung Disease  

TABLE 92-3  CHARACTERISTIC CHEST RADIOGRAPHIC PATTERNS IN PATIENTS WITH INTERSTITIAL LUNG DISEASE

TABLE 92-4  RADIOGRAPHIC FEATURES OF IDIOPATHIC INTERSTITIAL PNEUMONIAS

PATTERN

CLINICAL DIAGNOSIS

SUGGESTED DIAGNOSES*

Decreased lung volumes

Idiopathic pulmonary fibrosis, nonspecific interstitial pneumonia, desquamative interstitial pneumonia, connective tissue disease, chronic eosinophilic pneumonia, asbestosis, chronic hypersensitivity pneumonitis, or drug-induced interstitial lung disease (ILD)

Increased or preserved lung volumes

Idiopathic pulmonary fibrosis with emphysema, respiratory bronchiolitis–associated ILD, cryptogenic organizing pneumonia, hypersensitivity pneumonitis, lymphangioleiomyomatosis, Langerhans cell histiocytosis, sarcoidosis, neurofibromatosis, tuberous sclerosis

Micronodules

Infection, hypersensitivity pneumonitis, sarcoidosis, respiratory bronchiolitis–associated ILD

Septal thickening

Malignancy, infection, chronic congestive heart failure, pulmonary veno-occlusive disease

Honeycombing

Idiopathic pulmonary fibrosis, fibrotic nonspecific interstitial pneumonia, connective tissue disease, asbestosis, chronic hypersensitivity pneumonitis, sarcoidosis

Recurrent infiltrates

Cryptogenic organizing pneumonia, chronic eosinophilic pneumonia, drug- or radiation-induced ILD

Migratory or fleeting infiltrates

Cryptogenic organizing pneumonia, hypersensitivity pneumonitis, Churg-Strauss syndrome, Löffler syndrome, allergic bronchopulmonary aspergillosis

Pleural disease

Connective tissue disease, asbestosis, malignancy, radiation-induced ILD, amyloidosis, sarcoidosis, lymphangioleiomyomatosis, nitrofurantoin-induced ILD

Pneumothorax

Langerhans cell histiocytosis, lymphangioleiomyomatosis, tuberous sclerosis, neurofibromatosis

Mediastinal or hilar adenopathy

Lymphocytic interstitial pneumonia, connective tissue disease, silicosis, chronic berylliosis, malignancy, infection, sarcoidosis, amyloidosis, Gaucher disease

Normal (rare)

Cellular nonspecific interstitial pneumonia, respiratory bronchiolitis–associated interstitial lung disease, connective tissue disease, hypersensitivity pneumonitis, sarcoidosis

LOCATION OF RADIOGRAPHIC ABNORMALITY Mid to upper lung zone

SUGGESTED DIAGNOSES* Hypersensitivity pneumonitis, chronic berylliosis, ankylosing spondylitis, silicosis, Langerhans cell histiocytosis, sarcoidosis, pleuroparenchymal fibroelastosis, cystic fibrosis

Lower lung zone

Idiopathic pulmonary fibrosis, nonspecific interstitial pneumonia (fibrotic), connective tissue disease, asbestosis, chronic hypersensitivity pneumonitis

Peripheral

Idiopathic pulmonary fibrosis, nonspecific interstitial pneumonia (fibrotic), cryptogenic organizing pneumonia, chronic eosinophilic pneumonia

*This list is not intended to be comprehensive. Adapted from Raghu G, Brown K. Clinical issues: patient evaluation. In: Baughman RP, du Bois RM, eds. Diffuse Lung Disease: A Practical Approach. New York: Oxford University Press; 2004.

pulmonary infiltrates in the upper and middle lung zones with relatively clear perihilar and central zones is a clue to chronic eosinophilic pneumonia. Recurrent infiltrates raise the possibility of cryptogenic organizing pneumonia, chronic eosinophilic pneumonia, or drug- or radiation-induced pneumonitis, and fleeting or migratory infiltrates may occur in Churg-Strauss syndrome (allergic angiitis), allergic bronchopulmonary aspergillosis, tropical eosinophilic pneumonia, or Löffler syndrome. Whereas localized pleural plaques may indicate asbestosis, diffuse pleural thickening can result from asbestosis, rheumatoid arthritis, progressive systemic sclerosis, radiation pneumonitis, nitrofurantoin, or malignancy. In the absence of left ventricular

USUAL RADIOGRAPHIC FEATURES

TYPICAL FINDINGS ON HRCT

Idiopathic pulmonary fibrosis

Basal-predominant reticulation abnormality with volume loss

Pattern of usual interstitial pneumonia; peripheral, basal, subpleural reticulation; honeycombing, traction bronchiectasis

Nonspecific interstitial pneumonia

Ground-glass and reticular opacification

Peripheral, basal, subpleural, symmetrical ground-glass attenuation with irregular lines and consolidation; subpleural sparing

Cryptogenic organizing pneumonia

Patchy bilateral consolidation

Subpleural or peribronchial patchy consolidation or nodules

Acute interstitial pneumonia

Diffuse ground-glass density or consolidation

Diffuse consolidation and ground-glass opacification, often with lobular sparing and late traction bronchiectasis

Desquamative interstitial pneumonia

Ground-glass opacity

Peripheral, lower lung zone ground-glass attenuation with reticulation and/or small cysts

Respiratory bronchiolitis– associated interstitial lung disease

Bronchial wall thickening, groundglass opacification

Diffuse bronchial wall thickening with poorly defined centrilobular nodules and patchy ground-glass opacification

Lymphocytic interstitial Reticular opacities and pneumonia nodules

Diffuse centrilobular nodules, ground-glass attenuation, septal and bronchovascular wall thickening, and thin-walled cysts

HRCT = high-resolution computed tomography. Adapted from American Thoracic Society/European Respiratory Society. International multidisciplinary revised classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med. 2002;165:277-304; Travis WD, Costabel U, Hansell DM, et al. An official American Thoracic Society/European Respiratory Society statement: update of the international multidisciplinary classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med. 2013;188:733-748; and Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183:788-824.

failure, the presence of a pleural effusion (Chapter 99) raises the possibility of rheumatoid arthritis, SLE, acute hypersensitivity pneumonitis, sarcoidosis, asbestosis, amyloidosis, lymphangioleiomyomatosis, or lymphangitic carcinomatosis. A reduction of lung volumes is typical in most ILDs; the presence of preserved lung volumes or hyperinflation should raise suspicion for chronic hypersensitivity pneumonitis, Langerhans cell histiocytosis, lymphangioleiomyomatosis, neurofibromatosis, sarcoidosis, or tuberous sclerosis. However, plain chest radiographs may be normal in about 10% of patients with ILD.

High-Resolution Computed Tomography

Because of its increased sensitivity and ability to distinguish ground-glass changes, which are generally considered to be reversible areas of lung disease, from irreversible fibrotic and honeycomb changes, high-resolution computed tomography (HRCT) is essential in both the diagnosis and staging of ILD. Although microscopic ILD cannot be excluded by a normal HRCT result, HRCT allows recognition of abnormalities not apparent in plain chest radiographs and may lead to an earlier diagnosis, help narrow the differential diagnosis patterns (Table 92-4), aid in selecting the site or sites for BAL and lung biopsy, and assist in choosing among therapeutic options and in estimating the response to treatment. Whereas normal HRCT excludes the diagnosis of pulmonary fibrosis, the presence of patchy subpleural reticular and basilar septal fibrosis, traction bronchiectasis, and honeycombing increases the level

579

CHAPTER 92  Interstitial Lung Disease  

of diagnostic confidence for the pattern of usual interstitial pattern, which is characteristic of idiopathic pulmonary fibrosis. The finding of bilateral cysts, including their size, configuration, distribution, and appearance, helps differentiate among lymphangioleiomyomatosis, tuberous sclerosis, and pulmonary Langerhans cell histiocytosis. HRCT can detect ILD despite normal chest radiographs in patients with asbestosis, silicosis, sarcoidosis, and scleroderma. Patients with respiratory bronchiolitis–associated ILD typically have patchy ground-glass attenuation on HRCT in concert with bilateral interstitial prominence, fine nodular radiographic infiltrates, and normal lung volumes. Images obtained in the supine and prone positions and on deep inspiration and exhalation sometimes help to differentiate fibrosis from atelectasis.

Pulmonary Function Tests

The most characteristic physiologic abnormalities in patients with ILD, regardless of etiology, are a restrictive lung defect and decreased Dlco (see Table 85-2 in Chapter 85). Forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) are decreased proportionally such that the ratio of the two remains normal or may even be increased. Both total lung capacity (TLC) and lung volumes measured by body plethysmography are reduced. Pulmonary function tests (PFTs) may be useful in monitoring the progression of disease and prognosis; significant changes in FVC, Dlco (corrected to hemoglobin), and physiological measurements (FVC, Dlco) at 1 year portend a worse survival in patients with idiopathic pulmonary fibrosis. Certain PFT findings may also aid in the differential diagnosis. A mixed obstructive–restrictive pattern occurs in patients with Churg-Strauss syndrome, allergic bronchopulmonary aspergillosis, endobronchial sarcoidosis, hypersensitivity pneumonitis, cryptogenic organizing pneumonia, tropical pulmonary interstitial eosinophilia, coexisting chronic obstructive pulmonary disease or asthma, or secondary bronchiectasis. Diseases associated with respiratory muscle weakness, such as polymyositis, progressive systemic sclerosis, and SLE, may exhibit a decrease in maximal voluntary ventilation and increased residual volume out of proportion to the decrease in FEV1.

Exercise Testing

The magnitude of the increase in Pao2-Pao2 on exercise correlates well with the severity of disease and the degree of pulmonary fibrosis in patients with idiopathic pulmonary fibrosis. Other exercise-induced physiologic abnormalities in ILD include a decrease in work rate and maximal oxygen consumption, abnormally high minute ventilation at submaximal work rates, decreased peak minute ventilation, and failure of tidal volumes to increase at submaximal levels of work while the respiratory rate increases disproportionately. The 6-minute walk test, performed on a flat surface, can provide quantitative data on exercise capacity and on oxygen desaturation with exercise and can justify use of supplemental oxygen based on clinical and physiological needs.

Invasive Evaluation

A collegial interaction and multidisciplinary discussions among the pulmonary clinician, chest radiologist, thoracic surgeon, and pathologist can help determine the best diagnostic approach for an individual patient (see Fig. 92-1). Findings on BAL can be diagnostic in some patients with ILD and can narrow the differential diagnosis in others (Chapter 85). For example, a lymphocyte-predominant cellular pattern raises the possibility of sarcoidosis or hypersensitivity pneumonitis in the appropriate clinical setting. Eosinophils are seen in pulmonary Langerhans cell granulomatosis, an asbestos body count greater than 1 fiber per milliliter of BAL fluid is seen in asbestosis, and specially staining surfactant material is seen in pulmonary alveolar proteinosis. A transbronchial lung biopsy may reveal noncaseating granulomas in sarcoidosis, “loose” noncaseating granulomas in hypersensitivity pneumonitis, giant cell granulomas in hard metal pneumoconiosis, or smooth muscle proliferation in lymphangioleiomyomatosis. However, failure to establish a diagnosis on BAL and transbronchial lung biopsy does not exclude these entities. Video-assisted thoracoscopic biopsy (Chapter 101) or open lung biopsy may be required to obtain an adequate sample for histologic evaluation of a patient with unexplained signs and symptoms when other studies have failed to establish a diagnosis, but most patients with idiopathic pulmonary fibrosis do not need to have a biopsy to confirm the diagnosis. The mortality rate for the procedure is generally less than 1%, and the morbidity rate is less than 3%.

TABLE 92-5  INTERSTITIAL LUNG DISEASE: CLINICAL RESPONSE TO SYSTEMIC CORTICOSTEROIDS ALONE* GENERALLY RESPONSIVE Sarcoidosis Acute hypersensitivity pneumonitis Drug induced Environmental causes (some) Idiopathic interstitial pneumonia Cryptogenic organizing pneumonia Nonspecific interstitial pneumonia (cellular) Respiratory bronchiolitis–associated ILD Lymphocytic interstitial pneumonia Desquamative interstitial pneumonia (subset) Acute interstitial pneumonia (?) Acute pulmonary capillaritis Eosinophilic pneumonia (acute and chronic) Acute radiation pneumonitis‡ Organizing pneumonia associated with connective tissue diseases

UNRESPONSIVE† Idiopathic interstitial pneumonia Idiopathic pulmonary fibrosis (usual interstitial pneumonia) Desquamative interstitial pneumonia (subset) Chronic secondary and advanced pulmonary fibrosis Chronic hypersensitivity pneumonitis (subset) Chronic radiation fibrosis Cryptogenic organizing pneumonia (subset) Acute interstitial pneumonia (?) Chronic pulmonary hemorrhage syndromes Pulmonary veno-occlusive disease Environmental (e.g., asbestosis, pneumoconiosis) End-stage ILDs, pulmonary fibrosis coexisting or associated with pulmonary hypertension Pulmonary Langerhans cell granulomatosis Lymphangioleiomyomatosis ILD in inherited disorders (?)

*The dosage plus duration of corticosteroids used is variable and based on anecdotal experience, individual expert opinion, clinical judgment, and response as judged by objective measurements (clinical, radiologic, or physiologic). Oral prednisone or prednisolone is the most common corticosteroid used. Most patients who respond during the first few weeks of 20 to 60 mg of prednisone per day require maintenance low-dose oral prednisone at 5 to 10 mg/day beyond 6 months. Some patients who require maintenance of oral prednisone doses higher than 20 mg/day beyond 4 to 6 months may tolerate lower doses of prednisone if other immune-modulating agents (e.g., azathioprine, mycophenolate) are used in combination. There is no evidence to recommend a specific regimen. Patients should be monitored carefully and regularly for known side effects of corticosteroid use (e.g., osteoporosis, glucose intolerance), and preventive and therapeutic measures must be undertaken appropriately. † Some patients unresponsive to oral corticosteroids alone may respond to combined treatment with corticosteroids and other immune-modulating drugs (e.g., azathioprine, mycophenolate). ‡ Although most patients respond to modest doses of oral prednisone (initially, 40-60 mg/day), it is important to taper the prednisone very slowly to reach a maintenance dose of 5 to 10 mg/day beyond 6 months; rapid taper of oral prednisone has been associated with “rebound,” which is an exaggerated lung injury beyond the irradiated segment of the lung and in the contralateral lung. ILD = interstitial lung disease.

TREATMENT  When the cause of the ILD is clearly known (e.g., acute or subacute hypersensitivity pneumonitis, occupational ILD, iatrogenic), further avoidance of  the inciting agent or agents is essential (Chapter 93). Although systemic corticosteroids are generally indicated and are associated with a favorable response in some ILDs, the dosage and duration are unclear and essentially based on anecdotal experience (Table 92-5). Supportive oxygen supplementation is dictated by clinical needs. For selected patients with end-stage ILDs, such as those associated with significant pulmonary fibrosis and pulmonary hypertension, lung transplantation (Chapter 101) may be a feasible and viable option. Treatments for pulmonary hypertension associated with ILDs (Chapter 68) are indicated in patients with connective tissue diseases, but their clinical benefit for patients with other ILDs have been disappointing. A1 

  SPECIFIC TYPES OF INTERSTITIAL LUNG DISEASE

Idiopathic Interstitial Pneumonias

Idiopathic interstitial pneumonias, which are a subset of acute or chronic ILDs of unknown etiology, are characterized by the presence of varying degrees of interstitial and alveolar inflammation and fibrosis. Distinct clinicopathologic forms of idiopathic interstitial pneumonia include chronic

580

CHAPTER 92  Interstitial Lung Disease  

1 2 3

A

B

FIGURE 92-2.  Diagnosis of idiopathic pulmonary fibrosis. A, The usual interstitial pneumonia pattern of idiopathic pulmonary fibrosis in the lower lobes on high-resolution computed tomography consists of (1) subpleural fibrotic changes with (2) traction bronchiectasis and (3) honeycomb cysts in the lower lobes. B, Usual interstitial pneumonia pattern of idiopathic pulmonary fibrosis. Note the presence of (1) subpleural fibrosis with (2) traction emphysema, (3) fibroblastic foci, and temporal heterogeneity of microscopic abnormalities at low magnification. (Courtesy of Dr. Kevin Leslie.)

fibrosing interstitial pneumonias (idiopathic pulmonary fibrosis and nonspecific interstitial pneumonia), smoking-related interstitial pneumonias (respiratory bronchiolitis–ILD and desquamative interstitial pneumonia), and acute or subacute idiopathic interstitial pneumonias (cryptogenic organizing pneumonia and acute interstitial pneumonia). Rare histologic patterns of acute fibrinous and organizing pneumonia and interstitial pneumonias with a bronchiolocentric distribution and pleuroparenchymal fibroelastosis have been recently recognized.5 In some patients, mixed histopathologic features are evident in different segments of the same lung. When the distinct pathological forms are not evident, the diagnosis of unclassifiable interstitial pneumonia has been recently recognized. The accuracy of the diagnosis of idiopathic interstitial pneumonias is increased by multidisciplinary discussions among expert pulmonologists, radiologists, and pathologists familiar with interstitial lung diseases and idiopathic interstitial pneumonias. Although the clinical severity may vary, the idiopathic interstitial pneumonias tend to manifest as an insidious onset of exertional dyspnea and a nonproductive cough. Chest pain and systemic symptoms such as weight loss and fatigue may be present. Bibasilar end-inspiratory crackles are often heard on auscultation. Clubbing, although not specific, is found in 25% to 50% of patients with idiopathic pulmonary fibrosis. Findings on the chest radiograph are most often nonspecific, and the presence of normal lung markings on the chest radiograph does not exclude ILD. On HRCT, many pathologic entities have characteristic image patterns that have greatly aided diagnosis (see Table 92-4). The clinical course of idiopathic pulmonary fibrosis is heterogeneous.

CHRONIC FIBROSING INTERSTITIAL PNEUMONIA

Idiopathic Pulmonary Fibrosis

Idiopathic pulmonary fibrosis accounts for 50% to 60% of all idiopathic interstitial pneumonias. Idiopathic pulmonary fibrosis occurs in adult men and women with a mean age at onset of 62 years. Some patients have familial disease, likely as an autosomal dominant with variable penetrance. The best validated genetic risk factor is a polymorphism in the promoter of the gene encoding mucin-5B (MUC5B), which is associated with both familial and sporadic forms.6 Variants in the gene encoding surfactant protein C have been strongly associated with familial idiopathic pulmonary fibrosis, and mutations in the gene encoding surfactant protein A2 have been associated with familial pulmonary fibrosis and lung cancer. Telomere shortening caused by genetic variants within the human telomerase RNA or human telomerase reverse transcriptase has been associated with both familial and sporadic idiopathic pulmonary fibrosis. Idiopathic pulmonary fibrosis is limited to the lungs in adults, usually older than 60 years, and it generally occurs in men with a history of cigarette smoking. Most often patients have otherwise been in good health and have no known connective tissue disease or exposure to drugs or environmental factors known to cause pulmonary fibrosis, although patients with significant cigarette smoking history may have coexisting emphysema. Typical clinical

FIGURE 92-3.  Computed tomography scan showing traction bronchiectasis (arrows).

manifestations include a gradual onset and progression of exertional dyspnea, restrictive abnormalities on PFTs (Chapter 85), and a distinct pattern of bilateral pulmonary fibrosis on HRCT.

DIAGNOSIS

Chest radiographs typically show basal-predominant reticular abnormalities with low lung volumes.7 The diagnostic features on HRCT are peripheral, predominantly basilar patchy intralobular reticulation, often with subpleural honeycomb cysts, traction bronchiectasis, and traction bronchiolectasis as the disease becomes more advanced (Fig. 92-2, A). Reticulation may progress to honeycombing, although neither alveolar consolidation nor parenchymal nodules are present. Compared with the other idiopathic interstitial pneumonias, the HRCT appearance of idiopathic pulmonary fibrosis is distinguished by the presence of fibrotic abnormalities, predominantly in the bases of the lower lobes (E-Fig. 92-E1), by subpleural reticulations (E-Fig. 92-E2), and by its hallmark honeycombing (E-Fig. 92-E3) and traction bronchiectasis (Fig. 92-3). There is a notable absence of extensive ground-glass opacification, micronodules, cysts, consolidation, significant air trapping in multiple lobes, pleural plaques, pleural effusion, and extensive mediastinal adenopathy, all of which are inconsistent with the radiographic pattern of usual interstitial pneumonia. Pulmonary function tests usually show a progressive restrictive pattern. However, patients with milder disease may have normal lung volumes and a small decrease in Dlco; rarely, PFT results may be normal.

CHAPTER 92  Interstitial Lung Disease  

580.e1

E-FIGURE 92-2.  Computed tomography scan showing subpleural reticulations (arrows) in a patient with idiopathic pulmonary fibrosis (A) The cysts are interspersed within areas of ground glass attenuation (B and C).

Basal predominate distribution E-FIGURE 92-1.  Computed tomography scan demonstrating lower lobe predominant fibrosis in a patient with idiopathic pulmonary fibrosis.

E-FIGURE 92-3.  Computed tomography scan showing honeycombing (arrows) in a patient with idiopathic pulmonary fibrosis.

CHAPTER 92  Interstitial Lung Disease  

TABLE 92-6  DIAGNOSIS CRITERIA FOR IDIOPATHIC PULMONARY FIBROSIS The diagnosis of idiopathic pulmonary fibrosis requires the presence of usual interstitial pneumonia (UIP) in the absence of other causes of interstitial lung disease (e.g., domestic, occupational, and environmental exposures, connective tissue disease, and drug toxicity) AND a. The presence of a UIP pattern on chest HRCT in the absence of a lung biopsy or b. Specific combinations* of chest HRCT patterns (UIP, possible UIP, inconsistent with UIP) and histopathologic features (UIP, probable UIP, possible UIP, not UIP) on surgical lung biopsy HRCT FEATURES OF UIP • Subpleural, basal predominance • Reticular abnormality • Honeycombing with or without traction bronchiectasis • Absence of peribronchovascular predominance, extensive ground-glass abnormality, diffuse micronodules, discrete cysts, diffuse mosaic attenuation, or consolidation

HISTOPATHOLOGIC FEATURES OF UIP • Marked fibrosis/architectural distortion, +/– honeycombing in a predominantly subpleural/ paraseptal distribution • Patchy parenchymal lung fibrosis • Fibroblast foci • No features suggesting an alternate diagnosis*

*Based on data from Raghu G, Collard HR, Egan JJ, et al. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183:788-824. HRCT = high-resolution computed tomography.

The cellular pattern in BAL fluid, which is nonspecific, is marked by an excess of neutrophils in proportion to the extent of reticular change on HRCT; the percentage of eosinophils may be mildly increased. The histopathologic pattern of usual interstitial pneumonia consists of patchy interstitial changes alternating with zones of honeycombing, fibrosis, minimal inflammatory cells, collagen deposition, and normal lung (Fig. 92-2, B). Subepithelial fibroblastic foci, small aggregates of myofibroblasts, and fibroblasts within myxoid matrix are invariably present and represent areas of active fibrosis. The presence of temporal heterogeneity, or areas at different stages of fibrosis transitioning with normal areas and honeycomb cysts, along with fibrotic foci within the lung, is an essential feature of usual interstitial pneumonia that distinguishes it from other processes such as nonspecific interstitial pneumonia. Interstitial cellular inflammation is minimal in usual interstitial pneumonia. Although usual interstitial pneumonia characterizes the microscopic abnormality in idiopathic pulmonary fibrosis, the same histologic and radiologic pattern can also be seen in patients with rheumatologic lung diseases, chronic hypersensitivity pneumonitis, and asbestosis (Chapter 93). In the appropriate clinical setting (and after definitive exclusion of other known clinical conditions associated with ILD) (see later), a definitive diagnosis of idiopathic pulmonary fibrosis is based on the presence of a pattern of usual interstitial pneumonia on HRCT or surgical lung biopsy (Table 92-6).

TREATMENT  Pirfenidone (1800 mg/day for 1 year) decreases the rate of decline in forced vital capacity in clinical trials of patients who have idiopathic pulmonary fibrosis and mild to moderate impairment in pulmonary function, A2-A4  and pooled data suggest an improvement in survival. A5  Pirfenidone is approved for the treatment of idiopathic pulmonary fibrosis in the U.S., Japan, and Europe.8 Treatment with nintedanib (a tyrosine kinase inhibitor at 150 mg orally twice daily) also decreases the rate of disease progression as measured by FVC over 52 weeks in patients with idiopathic pulmonary fibrosis and mild to moderate impairment in pulmonary function; A6  it is now approved in the U.S. and Europe. Sildenafil (a phosphodiesterase inhibitor at 20 mg orally three times a day) has shown small benefits in terms of dyspnea, oxygenation, and quality of life but not exercise capacity in patients with idiopathic pulmonary fibrosis and severe impairment in pulmonary function. A7  Abnormal acid gastroesophageal reflux (Chapter 138) is very common in patients with idiopathic pulmonary fibrosis, and treatment with standard doses of proton pump inhibitors, H2 receptor antagonists, or both as used for gastroesophageal reflux disease (Chapter 138) can slow the rate of progression of idiopathic pulmonary fibrosis. A8  By comparison, N-acetylcysteine alone A9  or as part of triple-therapy combined with prednisone plus azathioprine A10  is not beneficial. Warfarin increases

581

respiratory hospitalizations and death in patients with idiopathic pulmonary fibrosis. A11  Interferon-γ1b, cyclophosphamide, colchicine, d-penicillamine, dual and selective endothelin receptor antagonists, and oral corticosteroids as monotherapy or in combination with immunosuppressive agents are not beneficial. Despite the absence of data, patients who require hospitalization and intensive care for an acute exacerbation with loss of respiratory function in the absence of infection or other complications are usually treated with empirical intravenous corticosteroids (e.g., methylprednisolone 1.0 g intravenously as a pulse dose once a day for 3 days and followed by hydrocortisone, 125 mg every 6 hours for another 3 to 5 days), with further dosing dependent on the clinical response. Ancillary treatment measures, including supplemental oxygen (based on clinical and physiologic needs); prompt detection and treatment of respiratory tract infections and pulmonary embolism (Chapter 98); pulmonary rehabilitation; and immunization for influenza, herpes zoster, and pneumococcus, are all appropriate. Pulmonary hypertension, if present, may be treated (Chapter 68), but there is no evidence that such treatment will be beneficial. A12  Lung transplantation (Chapter 101) is indicated in selected patients, but about two thirds of patients with idiopathic pulmonary fibrosis are older than 60 to 65 years, which is a relative contraindication to lung transplantation. It is important to initiate discussion of palliative care measures before patients reach the terminal stages of the disease.

PROGNOSIS

The natural course of idiopathic pulmonary fibrosis is heterogeneous. Most patients exhibit a slow and steady decline, with a mortality rate of about 7% at 1 year and 14% at 2 years after diagnosis.9 A small subset of patients declines at a rapid rate over several months, but another subset of patients remains stable over several years before declining. Progressive impairment of lung function and gas exchange ultimately is fatal unless the patient undergoes lung transplantation. Patients who survive longer generally have less fibrosis on HRCT, less functional impairment, no evidence of pulmonary hypertension, and no significant oxygen desaturation during a modified version of the 6-minute walk test. Patients with coexisting emphysema, pulmonary hypertension, or episodes of acute exacerbation have even shorter survival times. By comparison, patients who have a polymorphism in the gene encoding MUC5B may have better survival times.

Nonspecific Interstitial Pneumonia

Nonspecific interstitial pneumonia is often associated with connective tissue diseases, but idiopathic nonspecific interstitial pneumonia is also recognized as a distinct clinical entity. It typically occurs in middle-aged, nonsmoking women with an average age at diagnosis of about 50 years. The prevalence of nonspecific interstitial pneumonia has been estimated at one to nine per 100,000. Two subgroups have been described, cellular and fibrotic. Because the average age at onset is about 10 years earlier in nonspecific interstitial pneumonia than in idiopathic pulmonary fibrosis and because the clinical features of idiopathic fibrotic nonspecific interstitial pneumonia are very similar to early cases of idiopathic pulmonary fibrosis, questions persist as to whether idiopathic fibrotic nonspecific interstitial pneumonia is a separate clinical entity or represents an early form of idiopathic pulmonary fibrosis.

DIAGNOSIS

Chest radiographs show bilateral patchy pulmonary infiltrates with a lower lung zone predominance in all forms of nonspecific interstitial pneumonia. HRCT reveals a predominant ground-glass pattern of attenuation, usually bilateral and often associated with subpleural reticulation (Fig. 92-4), and loss of volume in the lower lobe. In cellular nonspecific interstitial pneumonia, HRCT shows ground-glass opacification, consolidation, or both, but the biopsy shows mild to moderate lymphoplasmacytic interstitial chronic inflammation. The major differential diagnosis to consider as an alternative to cellular nonspecific interstitial pneumonia is acute or subacute hypersensitivity pneumonitis, so a thorough history regarding environmental exposures is crucial. In contrast, fibrotic nonspecific interstitial pneumonia has a bilateral lower lobe distribution with architectural derangement on HRCT; histopathologically, it has uniformly dense interstitial fibrosis and may sometimes be difficult to distinguish from idiopathic pulmonary fibrosis and usual interstitial pneumonia in the early clinical stages. In these circ*mstances, the diagnosis of fibrotic nonspecific interstitial pneumonia can be ascertained only by the histologic features in a surgical lung biopsy specimen.

582

CHAPTER 92  Interstitial Lung Disease  

TREATMENT AND PROGNOSIS  Patients with cellular nonspecific interstitial pneumonia usually respond to treatment with corticosteroids (see Table 92-5), and their prognosis is generally better than that of patients with idiopathic pulmonary fibrosis. Nonetheless, some patients progress over several years, and some manifest acute exacerbations similar to patients with idiopathic pulmonary fibrosis. Immunemodulating drugs, including prednisone, azathioprine, and mycophenolate, have been used empirically, with their doses based on clinical response as assessed by clinicians and not on evidence with randomized clinical trials.

SMOKING-RELATED INTERSTITIAL PNEUMONIAS

Respiratory Bronchiolitis–Associated Interstitial Lung Disease

This ILD is almost invariably associated with chronic and current cigarette smoking, and it usually manifests clinically during the fourth or fifth decade of life. However, it may also be detected incidentally on radiographs

in relatively younger and asymptomatic persons with a previous history of cigarette smoking or in people passively exposed to chronic cigarette smoke Respiratory bronchiolitis–associated ILD is always associated with chronic exposure to cigarette smoke.

DIAGNOSIS

Pulmonary function tests show varying degrees of airway obstruction, mildly decreased or preserved TLC, and decreased Dlco. The chest radiograph typically reveals bronchial wall thickening and areas of ground-glass attenuation. HRCT reveals centrilobular nodules with an upper lobe predominance, patchy ground-glass attenuation, and peribronchial alveolar septal thickening (Fig. 92-5, A). Areas of hypoattenuation (mosaic attenuation) represent air trapping as a result of small airways disease. The characteristic finding on BAL is numerous brown-pigmented alveolar macrophages, often with a modest increase in neutrophils. Lung biopsy is rarely needed, but its hallmark histopathologic feature is the accumulation of pigmented alveolar macrophages with glassy eosinophilic cytoplasm and granular pigmentation within respiratory bronchioles, typically with a chronic inflammatory cell infiltrate in the bronchioles and surrounding alveolar walls (Fig. 92-5, B). Fibroblastic foci and honeycomb change are not present, but centrilobular emphysema is frequent.

TREATMENT AND PROGNOSIS  Progression to honeycomb lung and end-stage fibrosis seldom occurs, and the prognosis is good with cessation of smoking. Discontinuation of cigarette smoking is essential, and patients may benefit from low-dose corticosteroids (e.g., prednisone, 10 to 20 mg/day) for a few months.

Desquamative Interstitial Pneumonia

Desquamative interstitial pneumonia is a rare entity ( 65 yr Upper lobe predominant disease Chronic medical conditions Hepatitis B and/or C HIV infection Renal insufficiency Cirrhosis Neuropathy Poorly controlled diabetes Osteoporosis Severe GERD Poor esophageal motility Malignancy Unable to maintain long-term follow-up Psychiatric issues limiting compliance Insufficient social support

FEV1 ≤ 20% predicted Dlco ≤ 20% predicted hom*ogeneous or lower lobe distribution of disease TLC < 100% predicted RV < 150% predicted Paco2 > 60 mm Hg Pao2 < 45 mm Hg 6 MWD < 140 m or 65 yr Poor nutritional status ( 30 kg/m) Symptomatic osteoporosis Colonization with highly virulent and/or highly resistant fungi, mycobacteria, or bacteria Requirement for invasive ventilation and/or circulatory support Uncontrolled chronic medical conditions (e.g., diabetes, hypertension, GERD) Severely limited functional status with poor rehabilitation potential Psychosocial problems likely to affect the outcome adversely High-dose (>20 mg of prednisone daily) corticosteroid use *Other includes lymphangioleiomatosis, non-retransplantation-related obliterative bronchiolitis, and miscellaneous indications. BMI = body mass index; CAD = coronary artery disease; CF = cystic fibrosis; COPD = chronic obstructive pulmonary disease; GERD = gastroesophageal reflux disease; LV = left ventricle; PCI = percutaneous coronary intervention; PPH = primary pulmonary hypertension. Adapted from Yusen RD, Christie JH, Edwards LB, et al. Twenty-sixth official adult lung and heart-lung transplant report—2013. J Heart Lung Transplant. 2013;32:965-978; and Orens JB, Estenne M, Arcasoy S, et al. International guidelines for the selection of lung transplant candidates: 2006 update—a consensus report from the Pulmonary Scientific Council of the International Society for Heart and Lung Transplantation. J Heart Lung Transplant. 2006;25:745-755.

performed in certain patients with emphysema, primary pulmonary hyper­ tension, and other diseases (see Table 101-3). Bilateral transplantation is preferred for nearly all indications because a double-lung recipient can expect a half-life of 6.9 years compared with 4.6 years for a single-lung recipient. As a result, about 75% of the world’s reported lung transplants are now bilateral. Heart-lung transplantation is now performed in only about 75 cases per year. It is an en bloc procedure with right atrial, aortic, and distal tracheal anastomoses. It is performed in patients with advanced lung disease and coexistent irreparable cardiac disease, usually associated with fixed pulmonary hypertension, and in those with Eisenmenger syndrome (Chapter 69). Living donor lobar transplantation involves the removal of a lower lobe from each of two living donors. One is implanted into each hemithorax of the recipient in a manner similar to bilateral lung transplantation.11

Evaluation of Potential Transplant Recipients

The ideal candidate for lung transplantation has lung disease unresponsive to medical therapy but is in otherwise good health. Patients who experience

646

CHAPTER 101  Interventional and Surgical Approaches to Lung Disease  

critical illness as a result of lung disease often have poor nutritional status, coexistent major organ dysfunction, refractory infection, or other contrain­ dications to transplantation. The specific evidence-based recommendations for referral for transplant evaluation vary with the underlying disease. In the United States, the lung allocation system is based on expected disease-specific and patient-specific survival during the waiting period and after engraftment, thereby reflecting net transplant benefit. Early evaluations of the system, which was introduced in 2005, indicate shorter waiting times, an increase in the total number of transplantations performed, a decreased waitlist mortality, and an unchanged overall survival after transplantation.

Post-transplantation Issues

Most of the medical issues that patients and physicians face after lung trans­ plantation are the consequence of the transplantation and post-transplantation medication rather than the underlying disease for which the transplantation was performed. Examples include immunosuppression, infections and their prophylaxis, acute allograft rejection, chronic allograft rejection, and nonpul­ monary complications of transplantation.

Immunosuppression

The standard chemotherapeutic regimen for immunosuppression after lung transplantation consists of a calcineurin inhibitor such as cyclosporine or tacrolimus, azathioprine or mycophenolate mofetil, and corticosteroids. More than 50% of centers add an antilymphocyte antibody preparation in the first days after transplantation, and this practice has led to a small but statistically significant improvement in long-term survival.

Infections and Prophylaxis after Lung Transplantation

Lung transplant recipients are at high risk for bacterial, viral, fungal, and protozoal infections; infections are the leading causes of death during the early post-transplantation period. In the first 3 months after transplantation, bacterial infections are responsible for most deaths. In approximately one third of patients, pneumonia is diagnosed in the first weeks after transplanta­ tion, with gram-negative organisms as the cause in 75% of cases. Colonization and recurrent infections, usually with Pseudomonas species, often develop in patients with chronic rejection. Among potential viral pathogens, cytomegalovirus (CMV; Chapter 376) is the most important in lung transplant recipients. Seronegative patients who receive an allograft from a seropositive donor are at particularly high risk for the development of a clinically significant CMV infection. Seronegative patients who have a seronegative donor are at low risk for infection if they are treated with seronegative blood products. Epstein-Barr virus (EBV) has been associated with the development of post-transplantation lymphoprolif­ erative disorder. Aspergillus species are the most common cause of invasive fungal infection (Chapter 339). Colonized patients and those deemed at risk may receive prophylactic inhaled amphotericin B. Because of the nature of the immunosuppressive chemotherapeutic regimen used, patients are at high risk for infection by the protozoan Pneumocystis jirovecii (Chapter 341). The use of trimethoprim-sulfamethoxazole prophylaxis (typically 1 double-strength tablet three times weekly indefi­ nitely) has virtually eliminated Pneumocystis pneumonia.

Acute Rejection

Histologically, the initial manifestation of acute rejection is a lymphocytepredominant inflammatory response, usually centered on blood vessels, airways, or both. By convention, acute rejection is graded histologically from 0 (normal) to 4 (severe), with subclasses defined by the presence or absence of airway inflammation. The risk for acute allograft rejection is highest in the early months after transplantation and declines with time. Multiple episodes of acute rejection are the major risk factor for the subsequent development of chronic rejection. Clinically, patients may have fever, cough, and exertional dyspnea. Evalu­ ation may demonstrate rales or rhonchi on chest examination, a decline in pulmonary function by spirometry, leukocytosis, opacities on chest radiog­ raphy, and exertional desaturation. The clinical manifestation is often indis­ tinguishable from infectious pneumonia, and the clinical impression is accurate in only 50% of cases. Treatment of acute rejection most often consists of high-dose corticoste­ roids (typically, 1 g/day of methylprednisolone administered intravenously for 3 days).

Chronic Rejection

PATHOBIOLOGY

The bronchiolitis obliterans syndrome is thought to be a manifestation of chronic rejection. Risk factors for development of the syndrome include the number of acute rejection episodes and, in some series, previous symptom­ atic CMV infection. Pathologically, “early” lesions demonstrate inflammation and disruption of the epithelium of small airways, followed by growth of granulation tissue into the airway lumen and subsequent complete or partial obstruction. The granulation tissue then organizes in a stereotypical pattern with resultant fibrosis that obliterates the lumen of the airway.

CLINICAL MANIFESTATIONS

Clinically, bronchiolitis obliterans is accompanied by nonspecific symp­ toms.12 Progressive exertional breathlessness typically develops, and pulmo­ nary function testing usually demonstrates evidence of progressive airflow obstruction (Chapter 85). Bronchiolitis obliterans is classified according to the FEV1: 0 (no significant abnormality) if FEV1 is greater than 80% of base­ line; 1 (mild) if FEV1 is 65 to 80% of baseline; 2 (moderate) if FEV1 is 50 to 65% of baseline; and 3 (severe) if FEV1 is 50% or less of baseline. In early stages, chest radiography is notable only for hyperinflation, but it may show bronchiectasis as the syndrome progresses. Later stages of bronchiolitis oblit­ erans may include a syndrome of bronchiectasis with chronic productive cough and airway colonization with Pseudomonas species.

DIAGNOSIS

The diagnosis of bronchiolitis obliterans is made on both clinical and patho­ logic grounds. Transbronchial biopsy has a low yield for demonstrating his­ tologic evidence of bronchiolitis obliterans, but when such evidence is seen, it is diagnostic. In patients with a compatible clinical syndrome, exclusion of anastomotic stenosis and occult pulmonary infection is sufficient to establish the diagnosis.

TREATMENT  A variety of therapies have been tried for chronic rejection, including pulse corticosteroids, antilymphocyte antibodies, total lymphoid irradiation, photopheresis, and nebulized cyclosporine, but none has been clearly established as effective. Most patients with bronchiolitis obliterans experience a progressive decline in pulmonary function despite immunosuppression.

PROGNOSIS

Bronchiolitis obliterans is the leading cause of late mortality after lung trans­ plantation. Half of lung transplant recipients surviving to 5 years will have either biopsy-proven bronchiolitis obliterans or the clinical diagnosis of bronchiolitis obliterans syndrome.

Nonpulmonary Medical Complications   of Lung Transplantation

Most of the nonpulmonary medical complications that arise in patients after lung transplantation are the result of immunosuppressive therapy. One or more of these complications develop in virtually all lung transplant recipients. Osteoporosis (Chapter 243) is common because of the long-term use of corticosteroids and cyclosporine. Bone density should be monitored periodi­ cally, and pharmacologic therapy should be instituted if excessive bone loss is identified. Chronic renal insufficiency (Chapter 130) is common and is the result of therapy with the calcineurin inhibitors cyclosporine or tacrolimus, both of which affect afferent vascular tone in the kidneys and result in an average 50% drop in the glomerular filtration rate in the first 12 months after lung trans­ plantation. Systemic arterial hypertension is also common and is caused by corticosteroids and cyclosporine. Calcium-channel blockers, which are often used to treat hypertension, raise serum cyclosporine levels; appropriate mon­ itoring and dose adjustment are needed when starting such therapy. Both corticosteroids and tacrolimus contribute to the development of diabetes mellitus and hyperlipidemia. Solid organ transplantation is associated with an increased incidence of malignancy, thought to be due to pharmacologic immunosuppression and

CHAPTER 101  Interventional and Surgical Approaches to Lung Disease  

Adult Lung Transplants Kaplan-Meier Survival by Procedure Type (Transplants: January 1994 – June 2011) 100 Median survival (years): Double lung: 6.9 years; Conditional = 9.6 years Single lung: 4.6 years; Conditional = 6.5 years All lungs: 5.6 years; Conditional = 7.9 years

Survival (%)

75

p < 0.0001 50

25

Bilateral/double lung (N = 22,181) Single lung (N = 14,225) All lungs (N = 36,406)

0 0

1

2

3

4

5

6

7

8 9 10 11 12 13 14 15 16 17 Years

FIGURE 101-4.  Kaplan-Meier survival estimates for all adult lung transplantations reported to the International Registry for Heart and Lung Transplantation from 1994 to 2011. Note the highly statistically significant survival advantage conferred by double lung grafts. Because the decline in survival is greatest during the first year after transplantation, the conditional survival (i.e., when 50% of the recipients who survive to at least 1 year have died) provides a more realistic expectation of survival time for recipients who survive the early post-transplant period. (Adapted from Yusen RD, Christie JH, Edwards LB, et al. Twenty-sixth official adult lung and heart-lung transplant report— 2013. J Heart Lung Transplant. 2013;32:965-978.)

alteration in immune surveillance. Patients are at increased risk for lympho­ proliferative malignancies and other types of cancers. Post-transplantation lymphoproliferative disorders occur in about 4% of patients after organ trans­ plantation; most are associated with EBV. These syndromes can be polyclonal or monoclonal. Reduction in immunosuppression is sometimes therapeutic

647

in those with polyclonal disease. The prognosis in patients with monoclonal disease is poor, with little response to modification of immunosuppression or antineoplastic chemotherapy. Patients are also at increased risk for skin, bladder, lung, cervical, and hepatobiliary malignancy after solid organ transplantation.

Outcomes after Lung Transplantation

Currently, the annual mortality rate following lung transplantation is 8 to 10% per year, largely owing to bronchiolitis obliterans syndrome. The median survival after lung transplantation is about 5.5 years (Fig. 101-4).

Grade A References A1. Annema JT, van Meerbeeck JP, Rintoul RC, et al. Mediastinoscopy vs endosonography for medias­ tinal nodal staging of lung cancer: a randomized trial. JAMA. 2010;304:2245-2252. A2. Wechsler ME, Laviolette M, Rubin AS, et al. Bronchial thermoplasty: Long-term safety and effec­ tiveness in patients with severe persistent asthma. J Allergy Clin Immunol. 2013;132:1295-1302. A3. Chen JS, Chan WK, Tsai KT, et al. Simple aspiration and drainage and intrapleural minocycline pleurodesis versus simple aspiration and drainage for the initial treatment of primary spontaneous pneumothorax: an open-label, parallel-group, prospective, randomised, controlled trial. Lancet. 2013;381:1277-1282. A4. Naunheim KS, Wood DE, Mohsenifar Z, et al. Long-term follow-up of patients receiving lungvolume-reduction surgery versus medical therapy for severe emphysema by the National Emphy­ sema Treatment Trial Research Group. Ann Thorac Surg. 2006;82:431-443. A5. Fishman A, Martinez F, Naunheim K, et al. A randomized trial comparing lung-volume-reduction surgery with medical therapy for severe emphysema. N Engl J Med. 2003;348:2059-2073. A6. Sciurba FC, Ernst A, Herth FJF, et al. A randomized study of endobronchial valves for advanced emphysema. N Engl J Med. 2010;363:1233-1244. A7. Shah PL, Slebos DJ, Cardoso PF, et al. Bronchoscopic lung-volume reduction with Exhale airway stents for emphysema (EASE trial): randomised, sham-controlled, multicentre trial. Lancet. 2011; 378:997-1005.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 101  Interventional and Surgical Approaches to Lung Disease  

GENERAL REFERENCES 1. Detterbeck FC, Lewis SZ, Diekemper R, et al. Executive summary: diagnosis and management of lung cancer, 3rd ed. American College of Chest Physicians evidence-based clinical practice guide­ lines. Chest. 2013;143:7s-37s. 2. Silvestri GA, Gonzalez AV, Jantz MA, et al. Methods for staging non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed. American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143:e211S-250S. 3. Bacon JL, Patterson CM, Madden BP. Indications and interventional options for non-resectable tracheal stenosis. J Thorac Dis. 2014;6:258-270. 4. Ost DE, Gould MK. Decision making in patients with pulmonary nodules. Am J Respir Crit Care Med. 2012;185:363-372. 5. Paul S, Altorki NK, Sheng S, et al. Thoracoscopic lobectomy is associated with lower morbidity than open lobectomy: a propensity-matched analysis from the STS database. J Thorac Cardiovasc Surg. 2010;139:366-378. 6. Berger RL, DeCamp MM, Criner GJ, et al. Lung volume reduction therapies for advanced emphy­ sema: an update. Chest. 2010;138:407-417.

647.e1

7. Agzarian J, Miller JD, Kosa SD, et al. Long-term survival analysis of the Canadian Lung Volume Reduction Surgery trial. Ann Thorac Surg. 2013;96:1217-1222. 8. Cohen E. Bronchoscopic treatment of end-stage chronic obstructive pulmonary disease. Curr Opin Anaesthesiol. 2014;27:36-43. 9. Yusen RD, Edwards LB, Kucheryavaya AY, et al. The registry of the International Society for Heart and Lung Transplantation: thirty-first adult lung and heart-lung transplant report—2014; focus theme: retransplantation. J Heart Lung Transplant. 2014;33:1009-1024. 10. Cypel M, Yeung JC, Liu M, et al. Normothermic ex vivo lung perfusion in clinical lung transplanta­ tion. N Engl J Med. 2011;364:1431-1440. 11. Date H, Sato M, Aoyama A, et al. Living-donor lobar lung transplantation provides similar survival to cadaveric lung transplantation even for very ill patients. Eur J Cardiothorac Surg. 2014;[Epub ahead of print]. 12. Barker AF, Bergeron A, Rom WN, et al. Obliterative bronchiolitis. N Engl J Med. 2014;370: 1820-1828.

647.e2

CHAPTER 101  Interventional and Surgical Approaches to Lung Disease  

REVIEW QUESTIONS 1. Which of the following statements is correct concerning endobronchial ultrasound (EBUS)? A. The use of EBUS obviates the need for surgical lymph node sampling. B. EBUS can provide adequate material for the analysis of molecular markers in patients with non−small cell lung cancer. C. Radial-probe EBUS allows for real-time sampling of hilar lymph nodes. D. The combined use of RP-EBUS and navigation bronchoscopy has a yield equivalent to transthoracic needle biopsy for the diagnosis of a 2-cm pleural-based nodule. E. All of the above Answer: B  EBUS can obtain adequate material for the evaluation of tumor molecular markers in more than 94% of cases. Although EBUS has very high sensitivity and specificity, specimens showing lymphocytes but no tumor may need to be confirmed as true negatives by surgical lymph node dissec­ tion. Radial-probe EBUS is primarily used to identify parenchymal nodules. Transthoracic needle biopsy is associated with a higher diagnostic yield for peripheral nodules but is more likely to be complicated by the development of pneumothorax compared with radio-probe EBUS and navigation bronchoscopy. 2. Which of the following statements is true concerning patients with severe emphysema? A. Bronchoscopic lung volume reduction can reduce emergency depart­ ment visits for exacerbations of chronic bronchitis. B. The use of oxygen and surgical lung volume reduction has been shown to decrease mortality. C. Surgical lung volume reduction is better than medical therapy for patients with hom*ogenous emphysema. D. Surgical lung volume reduction improves mortality in the subset of patients with high exercise tolerance. E. All of the above Answer: B  The use of oxygen and surgical lung volume reduction improves mortality in appropriately selected patients with advanced emphysema: patients with upper lobe−predominant disease, an FEV1 of less than 20%, and low exercise capacity. For patients with good exercise tolerance, surgical lung volume reduction increases mortality compared with standard medical therapy. There are no data suggesting that bronchoscopic lung volume reduc­ tion reduces emergency room visits for exacerbations of chronic obstructive pulmonary disease.

3. Which of the following statements is correct concerning lung transplantation? A. The outcomes of unilateral lung transplantation are better than for bilateral lung transplantation. B. Lung transplantation improves survival in patients with advanced emphysema. C. Infection with Aspergillus species is a common cause of mortality at 5 years. D. The bronchiolitis obliterans syndrome is the leading cause of late mortality. E. All of the above Answer: D  Bilateral lung transplantation has improved outcomes compared with unilateral transplantation. Transplantation confers no overall survival advantage in patients with advanced emphysema, and obliterative bronchiol­ itis is the leading cause of mortality at 5 years. 4. Which of the following are true concerning central airway obstruction? A. The treatment of malignant central airway obstruction does not affect mortality. B. Laser photoresection is better than argon plasma coagulation to treat malignant central airway obstruction. C. Airway stenting is the procedure of choice for extrinsic airway obstruction. D. Airway stents are well tolerated and associated with minimal complications. E. All of the above Answer: D  Airway stents are the only endoscopic treatment that can palliate extrinsic airway obstruction. They are, however, associated with significant complications. After successful treatment of malignant airway obstruction, patients have mortality rates similar to those of patients with malignancy but no central airway obstruction.

650

CHAPTER 102  Approach to The Patient in a Critical Care Setting  

102  APPROACH TO THE PATIENT IN A CRITICAL CARE SETTING DEBORAH J. COOK

THE INTENSIVIST-LED MULTIDISCIPLINARY TEAM

Patients with critical illness in the intensive care unit (ICU) usually require advanced life support, such as mechanical ventilation, vasopressors, inotropic agents, or renal replacement therapy. Morbidity associated with critical illness includes complications of both acute and chronic diseases, nosocomial and iatrogenic consequences, and impaired quality of life among survivors. Critically ill patients are at a higher risk of death than any other hospital population. Accordingly, the goals of critical care are to reduce morbidity and mortality, to maintain organ function, and to restore health. Unlike many other specialties, critical care medicine is not limited to a particular population, disease, diagnosis, or organ system. Staffing of ICUs with critical care physicians, often referred to as intensivists, who provide mandatory consultations or principal ongoing care is associated with a significantly reduced ICU and hospital mortality and reduced ICU and hospital lengths of stay. The addition of nighttime intensivist staffing appears to reduce mortality by 38% in ICUs with low-intensity daytime staffing but not in centers with high-intensity daytime staffing, such as academic ICUs. A1  These findings emphasize the value of the on-site availability of trained physicians who are dedicated to appropriate triaging, diagnosis, monitoring, treatment, and palliation of critically ill patients. Daily rounds by an ICU physician who leads the coordinated work of nurses, pharmacists, respiratory therapists, physiotherapists, dietitians, chaplains, and other physicians appear to improve outcomes. Observational studies suggest that a standardized, goal-oriented approach to care delivered by multidisciplinary clinicians, with explicitly defined roles and best practices checklists, can help improve the quality of ICU rounds.1 The critical care process can be optimized by interprofessional leadership, communication, and a positive organizational culture.

FLUID RESUSCITATION

Intravenous fluids to maintain or to restore intravascular volume are an important component of ICU therapy. Both crystalloid and colloid solutions are in widespread use. Crystalloids are readily available and inexpensive, whereas colloids generally require less volume to achieve a specific physiologic goal. Fluid replacement with either normal saline or 4% albumin results in similar rates of death, organ failure, and other clinical outcomes, A2  but crystalloids may lower mortality for patients with traumatic brain injury (Chapter 399). Fluid management with hydroxyethyl starch increases the need for renal replacement therapy and increases mortality compared with crystalloid infusions. A3  On the basis of these data, either crystalloid- or albumin-based colloid fluid resuscitation is recommended for most critically ill patients, crystalloids are recommended for head-injured patients, and starches are not recommended.

SEDATION, ANALGESIA, AND SPONTANEOUS BREATHING TRIALS

Endotracheal intubation, central venous catheterization, postoperative pain management, and other ICU procedures require that most patients receive sedation, analgesia, or both. Sedatives and analgesics are used to ensure ongoing tolerance to mechanical ventilation, particularly in patients with shock or severe acute respiratory distress syndrome (ARDS). As long as pain and anxiety are well treated, bolus injections are preferred to continuous infusions because of emerging concerns about drug-induced delirium and delayed weaning from the ventilator. If patients are receiving drug infusions, daily interruption of sedatives and analgesics, by protocols that provide an opportunity for the patient to be observed safely in a less sedated state, are associated with a shorter duration of mechanical ventilation and ICU length of stay than continuous infusions. A second key component of managing sedation and analgesia is to use a drug titration protocol and nurse-led

sedation scale; in these situations, daily interruption of sedation infusions may confer no additional benefit. A4  Discontinuation of ventilation is affected by sedation and analgesic infusions, and vice versa. A daily sedation vacation followed by a spontaneous breathing test increases the days of breathing without assistance and shortens ICU stay and hospital stay compared with usual sedation management plus a daily spontaneous breathing test. A5  In the year after enrollment, patients who were treated with a “wake up and breathe” protocol, which linked daily sedation vacation periods with daily spontaneous breathing trials, had a 32% better survival rate. On the basis of these data, a nurse-implemented sedation and analgesic management scale with daily drug interruption and daily spontaneous breathing trials are recommended for mechanically ventilated critically ill patients.

LONG-TERM OUTCOMES FOR SURVIVORS

Biomarkers of inflammation, residual organ dysfunction, and functional disabilities persist in most ICU survivors even after transfer out of the ICU. Treatments administered in the ICU also have serious sequelae. For example, neuromuscular blockers and corticosteroids may contribute to critical illness polyneuropathy. These problems have particularly serious adverse consequences for elderly critically ill patients who are deconditioned before hospitalization. In addition, anxiety, post-traumatic stress, and major mood disorders are common among patients and their caregivers during recovery. Therefore, although ICU discharge and hospital discharge are milestones in a patient’s trajectory, sequelae of critical illness have rarely resolved completely when patients are on the regular hospital unit. For example, residual muscle weakness is common,2 even 5 years after ICU discharge. The legacy of critical care and the resulting residual functional impairment increase postdischarge morbidity and costs, thereby encouraging rehabilitation interventions to improve long-term outcomes. In a randomized trial of patients who received mechanical ventilation for 72 hours or less, the addition of graduated, individualized, early physical therapy and occupational therapy during daily sedation vacation periods improved functional capacity at hospital discharge, reduced the duration of delirium, and reduced the number of ventilator days during the 28-day follow-up. A6  Discontinuation of physiotherapy as a result of patient instability, usually patient-ventilator asynchrony, occurred in only 4% of all sessions. This trial highlights how the recovery of critically ill patients potentially can be improved by coordinated multidisciplinary care.

APPLYING EVIDENCE TO PREVENT COMPLICATIONS OF CRITICAL ILLNESS

Considerable evidence of effective preventive and therapeutic ICU interventions has emerged in randomized trials during the past decade. For example, evidence-based initial management of a patient with urosepsis and ARDS includes low tidal volume ventilation, A7  avoidance of early high-frequency oscillation, A8  high positive end-expiratory pressure, A9  inotrope or vasopressor infusion, low-dose corticosteroids, early enteral small bowel nutrition, avoidance of antioxidants, A10  head of bed elevation, oral antisepsis with chlorhexidine, stress ulcer prophylaxis,3 thromboprophylaxis with lowmolecular-weight heparin, A11  and insulin therapy aimed at avoiding marked hyperglycemia but not achieving normoglycemia A12  (Chapters 104 and 105). In mechanically ventilated adults, chest radiographs on demand provide clinical outcomes equivalent to those of routine radiographs, despite about one-third fewer radiographs. A13  Later during the stabilization and recovery phase of critical illness, evidence-based management includes targeted protocol-driven sedation, daily interruption of sedation infusions, daily spontaneous breathing trials, and early mobilization. Potential barriers to applying evidence in fast-paced ICUs include a perceived lack of responsibility, unclear decisional authority, and errors of omission. Passive dissemination of information, whether written or verbal, is generally ineffective in modifying physicians’ behavior. More effective strategies to encourage the implementation of evidence-based recommendations are interactive education, audit and feedback, reminders (written or computerized), involvement of local opinion leaders, and multifaceted approaches. In the high-acuity ICU setting, preprinted physician orders may help guide (but not dictate) management (Table 102-1). For example, a statewide intervention coached local safety teams to lead multidisciplinary education about central venous catheter management strategies known to decrease infection risk, including a procedural checklist that incorporated handwashing, full barrier precautions for catheter insertion, chlorhexidine skin cleansing, avoidance of the femoral site, and removal of unnecessary catheters. This multimethod approach, which included periodic site-specific feedback,

651

CHAPTER 102  Approach to The Patient in a Critical Care Setting  

TABLE 102-1  ICU ADMISSION ORDERS: EXAMPLE FOR A PATIENT WITH UROSEPSIS AND ARDS MANAGEMENT STRATEGY ACUTE PHASE Mechanical ventilation Maintenance fluid Norepinephrine Corticosteroids Sedation Analgesia Antibiotics Head of bed Oral antisepsis Small bowel enteral nutrition Stress ulcer prophylaxis Thromboprophylaxis Intensive insulin therapy if glucose >180 mg/dL Glucometer calibration Tests Monitoring

ORDERS

REEVALUATE

Target TV 5-7 mL/kg of ideal body weight, PC 16 cm, rate 12, Fio2 0.7, PEEP 16 cm, plateau pressure 65 mm Hg Hydrocortisone 50 mg IV q6h while vasopressor dependent Midazolam 2-8 mg/hr IV, bolus 2-4 mg PRN Morphine 1-4 mg IV PRN Ampicillin 2 g IV q6h 45-degree elevation from horizontal Chlorhexidine 15 mL q6h 10 mL/hr of a commercial balanced feed containing about 1 kcal/mL; increase by 20 mL q4h to 70 mL/hr Pantoprazole 40 mg IV daily Dalteparin 5000 U SC daily 50 U insulin in 50 mL NS; start at 0.5 U/hr, repeat glucose q1h for 4 hr, and reassess; target 110-150 mg/dL

Calibrate glucose from glucometer and central laboratory every morning Glucose q4h when stable, ABG with each ventilator change, other tests as per ICU team Arterial catheter for systolic blood pressure, central venous catheter for central venous pressure and mixed venous oxygen saturation, ECG, oximetry, ABGs, sedation scale, Foley catheter, others as per ICU monitoring protocols STABILIZATION AND RECOVERY PHASES Sedation vacation Daily interruption of sedation from 0700 h until 0900 h; restart at half prior infusion rate at 0900 h if necessary; aim to discontinue infusion as soon as possible Spontaneous breathing trials Spontaneous breathing trial when weaning readiness criteria met Early mobility Titrated physiotherapy and occupational therapy when able

PRN PRN PRN Daily PRN PRN Daily PRN Daily Daily Daily Daily Daily Daily PRN PRN

Daily Daily Daily

ABG = arterial blood gas; ARDS = acute respiratory distress syndrome; ECG = electrocardiogram; Fio2 = fraction of inspired oxygen; ICU = intensive care unit; IV = intravenous; NS = normal saline; PC = pressure control; PEEP = positive end-expiratory pressure; PRN = as needed; SC = subcutaneous; TV = tidal volume.

decreased catheter-related blood stream infections from 7.7 per 1000 catheter-days at baseline to 1.4 at 18 months’ follow-up.4 In a provincial cluster randomized trial addressing six evidence-based critical care practices in community ICUs, a multimethod approach including video conferencing, education, provision of algorithms, audit, and feedback resulted in a threefold increased adoption of the six management strategies. A14 

PREDICTIONS, PREFERENCES, AND PALLIATIVE CARE

The prognosis of many critically ill patients improves once they are in the ICU. For others, treatment responsiveness is delayed or not realized, organ dysfunction evolves but does not resolve, and complications arise. Despite best efforts of the multidisciplinary ICU team, critical illness proves fatal to between 5 and 40% of adults. Approximately 2% of ICU patients discharged to the ward are readmitted within 48 hours and about 4% within 120 hours.5 When a therapeutic trial of critical care is started, and particularly when it is failing, it is crucial to discuss prognosis openly with families (Chapter 3). Among medical ICU patients older than 80 years at one tertiary care university hospital, ICU mortality was 46%, hospital mortality was 55%, and mortality among hospital survivors was 53% at 2 years.6 About 15% of patients who are admitted to an ICU have clinical courses that probably should generate discussion about palliative care.7 Families bring key information about the patient’s prior function and preferences. In the shared decision-making model dominant in many settings today, these exchanges often result in plans to withhold or to withdraw basic or advanced life support.8 Mechanical ventilation is the most frequent life support administered to and withdrawn from critically ill patients. Ventilator withdrawal very often precedes death in the ICU. Patients undergoing ventilator withdrawal or who die while mechanically ventilated have a shorter ICU stay than patients successfully weaned from the ventilator. When life support modalities are withdrawn because their further use would be futile,9 each can be discontinued or weaned, with attendant considerations and cautions (Table 102-2). Withdrawal may be guided by the severity of the illness and other physiologic characteristics, but it is more heavily influenced by the contemporary life support model that is attentive to a patient’s values and the physician’s predictions about future quality of life. This complexity underscores the need for ICU teams to be expert communicators, sensitive in eliciting patients’ preferences, timely in relieving suffering, and compassionate in

TABLE 102-2  CONSIDERATIONS AND CAUTIONS IN THE WITHDRAWAL OF LIFE SUPPORT ISSUE RISKS Weaning from No risk of physical inotropes or distress vasopressors Discontinuation No risk of physical of inotropes or distress vasopressors

OTHER CONSIDERATIONS May prolong the dying process, particularly if patient requires low doses and this is the only life support withdrawn Death may not occur quickly if the patient requires low doses, particularly if mechanical ventilation is ongoing Death may occur quickly if the patient requires high doses, with or without withdrawal of mechanical ventilation May prolong the dying process, particularly if the patient requires low pressure settings or low oxygen levels and this is the only life support withdrawn Death may not occur quickly if the patient requires low pressure settings or low oxygen levels Death may occur quickly if the patient requires high pressure settings or high oxygen levels Preemptive sedation is typically needed to blunt air hunger due to rapid changes in mechanical ventilation Avoids discomfort and suctioning of endotracheal tube Can facilitate oral communication Informing families about possible physical signs after extubation can prepare and reassure them Allows for the most natural appearance Not advised if the patient has hemoptysis

Weaning from mechanical ventilation

Low risk of dyspnea

Discontinuation of mechanical ventilation

Risk of dyspnea

Extubation

Risk of dyspnea Risk of stridor (steroids) Risk of airway obstruction (jaw thrust) Risk of noisy breathing (glycopyrrolate) Low risk of Death may take several days if this is the physical distress only advanced life support withdrawn

Discontinuation of renal replacement therapy

Reprinted with permission from Cook D, Rocker G. Dying with dignity in the intensive care unit. N Engl J Med. 2014;370:2506-2514. Copyright © 2014 Massachusetts Medical Society.

providing dignity to the dying while administering culturally competent, family-centered end-of-life care. A death with dignity in the ICU infers that whereas some treatments may be foregone, care can be enhanced as death ensues. Fundamental to maintaining dignity is the need to understand a patient’s unique perspectives on what gives life meaning in a setting replete with depersonalizing devices. The goal is caring for patients in a manner consistent with their values at a time of incomparable vulnerability, when they cannot speak for themselves.10

Grade A References A1. Kerlin MP, Small DS, Cooney E, et al. A randomized trial of nighttime physician staffing in an intensive care unit. N Engl J Med. 2013;368:2201-2209. A2. Finfer S, Bellomo R, Boyce N, et al. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004;350:2247-2256. A3. Rochwerg B, Alhazzani W, Sindi A, et al. Fluid resuscitation in sepsis: a systematic review and network meta-analysis. Ann Intern Med. 2014;161:347-355. A4. Mehta S, Burry L, Cook D, et al. Daily sedation interruption in mechanically ventilated critically ill patients cared for with a sedation protocol: a randomized controlled trial. JAMA. 2012;308:1985-1992. A5. Girard T, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet. 2008;371:126-134. A6. Schweickert WD, Pohlman MC, Pohlman AS, et al. Early physical and occupational therapy in mechanically ventilated, critically ill patients: a randomised controlled trial. Lancet. 2009;373:1874-1882. A7. Burns KE, Adhikari NK, slu*tsky AS, et al. Pressure and volume limited ventilation for the ventilatory management of patients with acute lung injury: a systematic review and meta-analysis. PLoS ONE. 2011;6:e14623. A8. Ferguson ND, Cook DJ, Guyatt GH, et al. High-frequency oscillation in early acute respiratory distress syndrome. N Engl J Med. 2013;368:795-805. A9. Briel M, Meade M, Zhou Q, et al. Higher versus lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and individual patient data meta-analysis. JAMA. 2010;303:865-873. A10. Heyland D, Muscedere J, Wischmeyer PE, et al. A randomized trial of glutamine and antioxidants in critically ill patients. N Engl J Med. 2013;368:1489-1497. A11. Cook D, Meade M, Guyatt G, et al. Dalteparin versus unfractionated heparin in critically ill patients. N Engl J Med. 2011;364:1305-1314. A12. Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283-1297. A13. Hejblum G, Chalumeau-Lemoine L, Ioos V, et al. Comparison of routine and on-demand prescription of chest radiographs in mechanically ventilated adults: a multicentre, cluster-randomized, two-period crossover study. Lancet. 2009;374:1687-1693. A14. Scales DC, Dainty K, Hales B, et al. A multifaceted intervention for quality improvement in a network of intensive care units. JAMA. 2011;305:363-372.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 102  Approach to The Patient in a Critical Care Setting  

GENERAL REFERENCES 1. Lane D, Ferri M, Lemaire J, et al. A systematic review of evidence-informed practices for patient care rounds in the ICU. Crit Care Med. 2013;41:2015-2029. 2. Kress JP, Hall JB. ICU-acquired weakness and recovery from critical illness. N Engl J Med. 2014;370:1626-1635. 3. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med. 2013;39:165-228. 4. Pronovost PJ, Goeschel CA, Colantuoni E, et al. Sustaining reductions in catheter related bloodstream infections in Michigan intensive care units: observational study. BMJ. 2010;340:c309. 5. Brown SE, Ratcliffe SJ, Kahn JM, et al. The epidemiology of intensive care unit readmissions in the United States. Am J Respir Crit Care Med. 2012;185:955-964.

652.e1

6. Roch A, Wiramus S, Pauly V, et al. Long-term outcome in medical patients aged 80 or over following admission to an intensive care unit. Crit Care. 2011;15:R36. 7. Hua MS, Li G, Blinderman CD, et al. Estimates of the need for palliative care consultation across United States intensive care units using a trigger-based model. Am J Respir Crit Care Med. 2014;189:428-436. 8. Curtis JR, Vincent JL. Ethics and end-of-life care for adults in the intensive care unit. Lancet. 2010;376:1347-1353. 9. Huynh TN, Kleerup EC, Wiley JF, et al. The frequency and cost of treatment perceived to be futile in critical care. JAMA Intern Med. 2013;173:1887-1894. 10. Cook D, Rocker G. Dying with dignity in the intensive care unit. N Engl J Med. 2014;370: 2506-2514.

652.e2

CHAPTER 102  Approach to The Patient in a Critical Care Setting  

REVIEW QUESTIONS 1. Optimal models for the care of critically ill patients include all except which of the following? A. Critical care delivered by a dedicated intensivist or obligatory intensivist consultation on all intensive care unit (ICU) patients B. Nighttime in-house intensivist coverage C. Intensivist-led multidisciplinary rounds D. A multidisciplinary team with explicitly defined roles E. Use of a “best practices” checklist or equivalent tool on rounds Answer: B  Strategies considered to optimize the process of critical care include a closed ICU model with intensivists delivering the care or consulting on each patient, intensivist-led multidisciplinary rounds in which clinicians have explicitly defined roles, and rounds that incorporate a best practices tool. Randomized trial evidence does not show an impact of 24/7 in-house intensivist coverage in fully staffed academic ICUs. 2. Compared with crystalloids for fluid maintenance or fluid resuscitation, which statement is true regarding starch solutions? A. Starch decreases the risk of renal failure. B. Starch increases the probability of needing renal replacement therapy. C. Starch increases the probability of needing renal replacement therapy, but only in patients with severe sepsis. D. Starch increases the risk of death in patients with severe sepsis. E. Both B and D. Answer: E  Starches do not decrease the risk of renal failure. Large randomized trials show that in both general ICU patients and in patients with severe sepsis, starches increase the probability of needing renal replacement therapy; furthermore, in patients with severe sepsis, resuscitation with starches increases the risk of death. 3. For mechanically ventilated patients receiving infusions of sedation or analgesia, which of the following statements is true? A. Concerns prevail that oversedation may induce delirium, prolonged weakness, and delayed liberation from the ventilator. B. Nurse-led targeted sedation protocols represent an effective method to minimize the chance of oversedation while maintaining the patient’s comfort. C. Compared with continuing infusions, daily interruption of sedation infusions decreases the duration of ventilation and ICU stay. D. Daily interruption of sedation infusions appears to have no additional impact on the duration of mechanical ventilation or safety of the patient if a nurse-led targeted sedation protocol is in place. E. All of the above. Answer: E  Concern is emerging about the adverse effects of prolonged sedation and analgesia infusions. Targeted sedation by a nurse-led protocol represents an optimal approach to minimize unnecessary sedation. Daily interruption of sedation infusions is also effective but may not further improve outcomes if individualized, targeted sedation levels are already managed with a nurse-led protocol.

4. Which of the following statements is not true regarding the legacy of critical illness? A. Disability among survivors of critical illness is pervasive, affecting many domains of quality of life. B. Disability among ICU survivors increases the overall length of hospital stay. C. Disability among family caregivers of ICU survivors is rare. D. Some aspects of disability among ICU survivors may be mitigated by earlier rehabilitation in the ICU. E. Combined early physiotherapy, sedation interruption, and spontaneous breathing tests used in combination have been shown to improve survival in critically ill patients. Answer: C  Depression, anxiety, and post-traumatic stress disorder are common in family caregivers of ICU survivors, and ICU survivors themselves can have pervasive, long-lasting multidimensional disabilities. Some disabilities are targets for early physical rehabilitation; others may benefit from counseling and other supports. Coordinated multidisciplinary approaches to help patients along the trajectory to recovery include early physiotherapy, daily sedation interruptions, and spontaneous breathing tests. 5. Which of the following statements about end-of-life care in the ICU is true? A. Palliative care is exclusively patient centered. B. Optimal end-of-life care is focused primarily on complete, timely management of symptoms in the final days. C. Decisions to withdraw life support are primarily determined by acute and chronic physiologic trends during critical illness. D. Dignified end-of-life care involves caring for patients in a manner consistent with their values when they cannot speak for themselves. E. In contemporary ICUs, death is more commonly preceded by withdrawal of inotropes and vasopressors than by withdrawal of mechanical ventilation. Answer: D  Palliative care in the ICU ensures the patient’s dignity, is holistic, is both patient and family centered, and addresses broad dimensions of health. Mechanical ventilation is the most common life support administered and withdrawn in the ICU. Decisions to withdraw mechanical ventilation are based less on pathophysiology than on the patient’s values, which are typically expressed by the family, and physicians’ predictions about future quality of life.

652

CHAPTER 103  Respiratory Monitoring in Critical Care  

narcotics, slowing the respiratory rate). Conversely, sustained tachypnea (e.g., >35 breaths/minute in an adult) can indicate ongoing increased work of breathing, impending respiratory failure, and the need for mechanical assistance, such as noninvasive ventilation or intubation and mechanical ventilation, depending on the etiology of the respiratory failure. Contraction of the sternocleidomastoid muscles or scalene muscles, often with a seated, bent posture, is called the tripod sign (E-Fig. 103-1). This response indicates inadequate diaphragmatic function, most commonly in the setting of emphysema with associated diaphragmatic flattening, which causes a mechanical disadvantage of diaphragmatic contraction. In this circ*mstance, patients may demonstrate Hoover sign, which is inspiratory retraction of the rib cage at the level of the zone of apposition, where the diaphragm inserts on the chest wall. The physical examination of the nail beds and lips may also reveal cyanosis, which suggests hypoxemia. Cyanosis occurs when saturation falls, but it requires the presence of 5 g of desaturated hemoglobin. As such, polycythemic patients may show cyanosis with relatively high oxyhemoglobin saturation values, whereas patients with profound anemia may not demonstrate cyanosis even in the face of low values of oxyhemoglobin saturation.

SYSTEMIC ARTERIAL BLOOD GAS ANALYSIS

Sampling of arterial blood, either through a percutaneous arterial puncture or by withdrawal of blood from an indwelling arterial catheter, provides important information about the patient’s oxygenation and ventilation status as well as the acuity of and compensation for derangements. The partial pressure of carbon dioxide (Paco2) reflects ventilation, the elimination of carbon dioxide. In many but not all cases, Paco2 is close to the mixed alveolar Paco2. The Paco2 in the arterial blood is closely related to the ratio of metabolic carbon dioxide production to alveolar ventilation:

PaCO2 = (K )(CO2 production rate)/(alveolar ventilation [ VA ]) (1)

The partial pressure of oxygen (Pao2) reflects the level of oxygenation. Normal levels of oxygenation are defined by the alveolar-arterial oxygen gradient, P(A-a)o2, which is calculated as P(A-a)O2 = FIO2 (PB − PH 2O at standard pressure and body temperature) − (PaO2 + PaCO2 /respiratory quotient) (2) where the respiratory quotient equals the number of moles of carbon dioxide produced for each mole of oxygen consumed (generally ~0.8 under normal metabolic conditions at rest but variable with dietary intake and metabolic rate). The normal value of the alveolar-arterial oxygen gradient varies with age and position and can be approximated by the simple equation

103  RESPIRATORY MONITORING IN CRITICAL CARE JAMES K. STOLLER AND NICHOLAS S. HILL Monitoring of the respiratory system involves a broad array of assessment techniques ranging from low-technology approaches like a careful physical examination to sophisticated technologies to monitor oxygenation and ventilation.

PHYSICAL EXAMINATION

The physical examination can provide important information about the patient’s ventilation and oxygenation. Ventilation can be assessed by recording the respiratory rate (normally 12 to 20 breaths/minute in adults) as well as by closely inspecting the pattern of chest wall movement during inspiration and by noting the use of accessory inspiratory muscles (e.g., the scalene, trapezius, and sternocleidomastoid muscles). Hypopnea (shallow or slow breathing) or a slowed respiratory rate (bradypnea) can indicate decreased ventilation. Shallow breathing may relate to muscle weakness (Chapter 421) or increased lung stiffness, which is commonly accompanied by a compensatory increase in the ratio of the respiratory rate to maintain ventilation. Bradypnea may relate to a suppressed respiratory drive (e.g., excessive use of

P(A-a)O2 = (age/4) + 4

(3)

Normal age-related values of Pao2 in the sitting position can be determined by the equation

PaO2 sitting = 104.2 − (0.27 × age in years)

(4)

Normal values of PaO2 are generally in the range of 70 to 95 mm Hg, depending on the patient’s age. The Paco2 helps assess the adequacy of the patient’s ventilation. At sea level, normal values of Paco2 range from 35 to 45 mm Hg. Values of Paco2 below 35 mm Hg indicate hyperventilation, either as a primary respiratory event (e.g., with anxiety) or in response to another insult (e.g., hypoxemia, sepsis, liver disease). Similarly, values of Paco2 exceeding 45 mm Hg indicate hypoventilation, hypercapnia, and respiratory acidosis, which may result either from suppression of the ventilatory drive (Chapter 86) (e.g., excess narcotics; Chapter 34) or from respiratory insufficiency (e.g., respiratory muscle weakness; Chapter 421). Assessment of the patient’s bicarbonate level (HCO3−) helps define the chronicity of changes in the patient’s Paco2, where the value of bicarbonate is defined by the Henderson-Hasselbalch equation:

pH = 6.1 + log10 [HCO3 − ]/0.003 PaCO2

(5)

Acute increases in Paco2 drive the normal kidney to retain bicarbonate (Chapter 118), whereas acute decreases in Paco2, as in hyperventilation from anxiety or liver disease, would be expected to cause the normal kidney to waste bicarbonate to preserve the body’s pH (normally 7.35 to 7.45). The clinician can also assess whether the patient’s ventilatory response to metabolic acidosis is appropriate or inadequate by the Winter equation, which predicts the expected Paco2 in the face of a decreased bicarbonate from a

CHAPTER 103  Respiratory Monitoring in Critical Care  

E-FIGURE 103-1.  This patient with chronic obstructive pulmonary disease (COPD) demonstrates the posture referred to as the tripod sign. The patient is sitting forward with his hands on his knees to provide a mechanical advantage to the accessory muscles of respiration, such as the sternocleidomastoid and trapezius. Diaphragmatic flattening accompanying COPD lessens the diaphragm’s ability to generate pressure, thereby resulting in increased dependence on the accessory muscles of respiration and leading to this posture.

652.e3

CHAPTER 103  Respiratory Monitoring in Critical Care  

metabolic acidosis (Equation 6). Specifically, a measured Paco2 above the expected value indicates an inadequate ventilatory response, whereas a value of Paco2 that falls within the expected range indicates an expected, appropriate ventilatory response to the metabolic derangement (i.e., the acidosis). PaCO2 = (1.5[HCO3 − ] + 8) ± 2

(6)

When the patient is hypercapnic and hypoxemic, a useful step is to calculate the ambient air P(A-a)o2 and to determine whether it is normal or increased for the patient’s age. Of the six mechanisms of hypoxemia, only two (hypoventilation and breathing decreased ambient oxygen, as at altitude or from a hypoxic gas mixture) are associated with a preserved P(A-a)o2 (Table 103-1). Under clinical circ*mstances at sea level, hypoxemia in the face of a normal P(A-a)o2 indicates that the patient’s hypoxemia is caused by hypoventilation and should prompt the clinician to consider the various causes of suppressed respiratory drive (Chapter 86) or respiratory insufficiency that interferes with a normal ventilatory response (e.g., respiratory muscle weakness; Chapter 421).

653

component to isolate the arterial component. The device can estimate the percentage of oxygenated hemoglobin over the range of 100% to about 75%. Most clinicians regard the output of pulse oximeters to be inaccurate for percentage saturation values of less than 70%, although the probability of a low saturation should not be discounted (Fig. 103-1). Pulse oximetry measurements may help identify significant drops in Pao2 below 60 to 65 mm Hg but are relatively insensitive to changes in Pao2 from 90 to 65 mm Hg. The true value of pulse oximetry for decision-making in the emergency department setting remains uncertain. A1 

CARBON DIOXIDE MONITORING: CAPNOMETRY AND TRANSCUTANEOUS CARBON DIOXIDE MEASUREMENT

Ventilation-perfusion mismatch

Pneumonia

Increased

The fraction of carbon dioxide in exhaled air can be measured in real time by infrared capnometry.2 Partial pressures can then be calculated on the basis of knowledge of atmospheric pressure. The expiratory capnogram (Fig. 103-1) represents a continuous plot of exhaled Pco2 versus time or exhaled volume and reflects the sequential appearance of gas from various compartments (e.g., the endotracheal tube, central airways, and finally the alveoli, where the Pco2 is in equilibrium with end-capillary blood). The shape of the capnogram provides clues to the presence of chronic obstructive pulmonary disease, in that emptying of areas of lung with increased dead space (see later) can cause the capnogram to have a rising contour (Fig. 103-1A), whereas the attainment of a so-called alveolar plateau on the normal capnogram (Fig. 103-1B) indicates that alveolar gas is composed of a mix with a relatively small contribution from areas of increased dead space. The value of Peco2 measured at the end of expiration on the capnometer (i.e., the highest value recorded) represents the end-tidal Petco2. Notably, the value of Petco2 is always below the Paco2 because there is a normal component of dead space ventilation (Vd/Vt) related to the anatomic dead space of the conducting airways (i.e., the trachea and airways to the level of gas-exchanging alveolar ducts and alveoli). The numerical difference between the Paco2 and the mixed exhaled carbon dioxide tension (Peco2, defined as the partial pressure of carbon dioxide that would be measured in a balloon in which the entire exhaled volume is gathered) is related to the magnitude of dead space ventilation (i.e., areas of the lung that are ventilated without accompanying blood flow, normally ~0.3 to 0.4) as defined by the Bohr equation:

Diffusion impairment

Interstitial lung disease

Increased

Anatomic right-to-left shunt

Pulmonary arteriovenous malformation

Increased

Hypoventilation

Neuromuscular weakness

Normal

Breathing decreased ambient oxygen (from either hypobaric conditions [e.g., altitude] or breathing a gas mixture with decreased inspired oxygen fraction)

Altitude exposure

Normal

Diffusion-perfusion impairment

Hepatopulmonary syndrome

Increased

The difference between Paco2 and Peco2 may be as low as several millimeters of mercury, but changing conditions of ventilation-perfusion matching (e.g., with pulmonary embolism [Chapter 98], atelectasis [Chapter 90]) may change the gradient over time. Measurement of the Petco2 can be clinically useful to assess trends, to help detect esophageal intubation, to detect disconnection from the ventilator, and to detect perfusion during cardiopulmonary resuscitation, but it is not a reliable surrogate for Paco2. Furthermore, measurement of the dead space fraction has prognostic value in patients with early acute respiratory distress syndrome (Chapter 104), in whom rising dead space is linearly related to increased mortality risk. Measurement of transcutaneous Pco2 by heated probes applied to the skin represents an alternative noninvasive method for estimating Pco2. This

PULSE OXIMETRY

Pulse oximetry is a noninvasive method to assess arterial blood oxygenation.1 The percentage of hemoglobin that is oxygenated is measured by passing light of two different wavelengths (660 nm [for deoxyhemoglobin] and 940 nm [for oxyhemoglobin]) through a blood-carrying tissue (e.g., finger, earlobe, forehead), identifying the pulsatile component (which contains arterial blood and background tissue elements), and subtracting the nonpulsatile

TABLE 103-1  PHYSIOLOGIC MECHANISMS OF HYPOXEMIA AND ACCOMPANYING VALUES OF THE ALVEOLAR-ARTERIAL OXYGEN GRADIENT   ON BREATHING OF ROOM AIR MECHANISM/ PHYSIOLOGIC PROCESS

ALVEOLAR-ARTERIAL OXYGEN GRADIENT ON ROOM AIR

EXAMPLE

VD/VT = (PaCO2 − PECO2 )/PaCO2

(7)

Lack of alveolar plateau Alveolar plateau

PCO2

PCO2

End-tidal PCO2

Expiration

A

Time

End-tidal PCO2

Expiration

B

Time

FIGURE 103-1.  Abnormal and normal end-tidal capnograms. A, Illustration of a capnogram from a patient with chronic obstructive pulmonary disease in which the end-tidal PCO2 rises throughout expiration as carbon dioxide excretion varies from different parts of the lung. B, Illustration of a normal capnogram in which the end-tidal PCO2 reaches a plateau with more uniform carbon dioxide excretion. The end-tidal PCO2 is the highest point of the alveolar plateau.

654

CHAPTER 103  Respiratory Monitoring in Critical Care  

approach is less widely used clinically, at least in adults, because of technical requirements, such as site rotation for the probes and repetitive calibration, and its generally lower accuracy in estimating Pco2.

Pressure (cm H2O)

30

ARTERIAL OXYGEN CONTENT AND SYSTEMIC OXYGEN DELIVERY

Arterial (Cao2) and venous oxygen content (Cvo2) are used to calculate cardiac output by the Fick equation (Equation 8), which is an alternative to determining cardiac output by the thermodilution method with a flowdirected pulmonary artery (Swan-Ganz) catheter (Chapter 57). The Fick equation is

Minute ventilation (Ve), which is the amount of gas exhaled from the airway per minute, is the product of the respiratory rate times the exhaled tidal volume, measured at body temperature and standardized to barometric pressure at sea level, saturated with water vapor (BTPS). The BTPS is a standard condition under which many measurements for most pulmonary function equipment and mechanical ventilators are made. These devices use an airflow meter to measure exhaled airflow and integrate the signal to derive tidal volume. An alternative way to measure tidal volume in an intensive care setting is respiratory impedance plethysmography, which uses calibrated magnetic coils in belts strapped around the chest and abdomen to monitor respiratory frequency and changes in thoracic volume. Alveolar ventilation is the rate of gas delivery in liters per minute to gasexchanging areas of the lung (i.e., the alveoli and alveolar ducts). The portion of minute ventilation that fails to undergo gas exchange is dead space ventilation (Vd) and is determined by Equation 7. Minute, alveolar (Va), and dead space ventilation are related as follows:

VE = VA + VD

(11)

It follows that conditions such as acute lung injury and acute respiratory distress syndrome (ARDS; Chapter 104) that are associated with very high dead space ratios require high Ve to achieve a sufficient Va. Conversely, conditions that cause neuromuscular weakness (Chapter 421) are associated with small tidal volumes and have a high Vd/Vt ratio because the anatomic dead space is fixed and constitutes a higher fraction of the diminished tidal volume.

MEASURING CARBON DIOXIDE PRODUCTION

Measurement of carbon dioxide production is sometimes referred to as indirect calorimetry because it provides an index of metabolic rate and permits estimation of calorie requirements. Metabolic “carts” that simultaneously measure not only carbon dioxide production but also oxygen consumption and respiratory quotient are commonly used clinically to estimate metabolic needs to prescribe nutritional repletion (Chapter 216). The normal baseline carbon dioxide production is in the range of 200 mL/minute but is subject to wide variation because of hypermetabolic states commonly encountered in critically ill patients, such as sepsis and the systemic inflammatory response syndrome. The respiratory quotient also gives insight into the composition of feedings because carbohydrates yield a respiratory quotient of 1, whereas fatty acids yield a ratio of 0.8 and amino acids a ratio of 0.7. Thus, balanced nutrition should yield a respiratory quotient of approximately 0.85. A respiratory quotient of 1 in combination with a high carbon dioxide production suggests that the dietary proportion of carbohydrates is excessive.

5

1

2

3

4

Time (seconds) FIGURE 103-2.  Illustration of inspiratory hold maneuver to determine plateau pressure (Pplateau). Airway pressure during volume-targeted mechanical ventilation rises as the tidal volume is delivered and reaches a peak. An inspiratory hold is initiated at  peak pressure that prevents exhalation, so pressure falls to a “plateau” of about  20 cm H2O. The drop in pressure reflects the pressure needed to overcome airway resistance. After slightly more than 1 second, the inspiratory hold is released, and airway pressure falls to positive end-expiratory pressure (PEEP). The difference between Pplateau and PEEP is used to calculate static compliance by dividing the difference into the tidal volume.

Under normal conditions (with, for example, an arterial percentage saturation of 95% and a hemoglobin level of 15 g/100 mL and an oxygen consumption of 250 mL/minute), arterial oxygen content is about 20 mL/100 mL, and because mixed venous oxygen saturation is about 75%, central venous oxygen content is about 15 mL/100 mL, making the normal arteriovenous oxygen content difference with a normal cardiac output about 5 mL/100 mL. Systemic oxygen transport defines the amount of oxygen delivered to the tissues and multiplies the arterial oxygen content by the cardiac output:

MEASURING VENTILATION: MINUTE VENTILATION AND ALVEOLAR VENTILATION

10

Oxygen content = 1.34 (hemoglobin)(% saturation) + 0.0031(PaO2 )   (9)

where the normal value is about 1000 mL/minute.

PPlateau

15

PEEP

where oxygen content has the units of milliliters of oxygen per 100 mL of blood and is calculated as

Systemic oxygen transport (mL/ min) = cardiac output × CaO2 (10)

20

Oxygen consumption (mL O2 / min) = cardiac output × (CaO2 − CVO2 ) (8)

PPeak

25

MEASURING RESPIRATORY COMPLIANCE Respiratory compliance is the change in respiratory system volume induced by a change in applied pressure (i.e., inspiratory pressure) and is the mathematical inverse of elastance. Compliance diminishes in conditions like lung injury and ARDS (Chapter 104) or pulmonary fibrosis (Chapter 92), in which diffuse inflammation and scarring alter lung structure and contribute to increased lung “stiffness.” Static respiratory compliance is measured in patients receiving volume-limited mechanical ventilation by imposing a brief inspiratory hold at end inspiration. Assuming the patient has no spontaneous breathing effort, the airway pressure measured when airflow ceases is referred to as the plateau pressure (Pplateau). The difference between this pressure and the positive end-expiratory pressure (PEEP) is taken as the driving pressure required to deliver the tidal volume (Fig. 103-2). Static respiratory system compliance (Crs) is then calculated as

CRS = ∆V(exhaled tidal volume)/∆P (Pplateau − PEEP)

(12)

This compliance not only reflects the status of the lung but also includes contributions of the chest wall and abdomen. Thus, patients with chest wall deformities or morbid obesity have lower values of respiratory compliance even in the absence of lung abnormalities (Chapter 99). The normal respiratory compliance is in the range of 50 to 70 mL/cm H2O, and patients with ARDS usually have values of Crs of less than 30 cm H2O. If respiratory compliance is below 20 to 25 cm H2O, weaning from mechanical ventilation (Chapter 105) is difficult or impossible because of the high work of breathing requirements (see later).

MEASURING RESPIRATORY DRIVE

The respiratory center, located in the pons and medulla, regulates respiratory drive. Hypercapnia is a strong stimulus to ventilation (Chapter 86). This response may be blunted by chronic carbon dioxide retention or by drugs like narcotics. Hypoxemia is a weaker ventilatory stimulus that is potentiated by hypercapnia and blunted by hypocapnia. Thus, respiratory drive can be assessed as the response to carbon dioxide in the blood in the hypercapnic ventilatory response. In one technique to measure respiratory drive, the patient rebreathes his or her exhaled air while minute ventilation and Petco2 are monitored; a graph relating Petco2 with minute ventilation is used to measure respiratory drive. However, this technique is impractical in an intensive care unit (ICU) setting. Another technique is to measure the negative swing in airway pressure during the first 100 msec of inspiration (P100). This technique avoids the problem of diminished ventilatory response due to airway obstruction, but it is still subject to blunting by some drugs and still underestimates drive in patients with respiratory muscle weakness, a common problem in the ICU. In patients who are failing to be weaned from mechanical ventilation, a practical way to assess the integrity of respiratory drive is to determine whether the respiratory rate increases, usually into the range of 30 to 40 breaths/minute, as Paco2 rises after the patient is removed from ventilatory support.

GENERAL REFERENCES

Volume

For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

c

d

a

Pressure:

b

Normal

Restrictive lung disease

Obstructive lung disease

FIGURE 103-3.  Pressure-volume curves illustrating components of work in a normal subject and in patients with restrictive or obstructive disease. The line between a and c represents elastic work as the lung expands, but this work is a net zero because static forces return the lung to its neutral position. The restrictive curve is flatter than normal because the lung is stiffer and volume changes less for a given unit change in pressure. The obstructive curve has a greater slope because (e.g., in emphysema) the lung is more compliant and starts inhalation from a higher volume. The abc curve represents resistive work during inspiration, and cda represents resistive work during exhalation. Resistive work during exhalation is greater in patients with obstructive lung disease.

MEASURING RESPIRATORY MUSCLE STRENGTH

Respiratory muscle weakness has long been recognized as a contributor to respiratory failure and failure to be weaned from mechanical ventilation in the ICU (Chapter 105). This recognition has intensified in recent years with the increased awareness of ICU-acquired weakness after critical illness. However, measurement of respiratory muscle strength remains challenging because of the need to differentiate between actual weakness and reduced muscle performance due to inability to cooperate or to exert a full inspiratory effort. The most commonly used measures of respiratory muscle strength are the maximal inspiratory and expiratory pressures (Pimax or MIP and Pemax or MEP). These values are obtained by measuring the pressure change with a manometer when the patient inhales with maximal force from residual volume and exhales with maximal force from total lung capacity. Normal MIP is usually more negative than −75 cm H2O, and normal MEP is usually more positive than 125 cm H2O. When the value for MIP is less negative than −20 or −30 cm H2O, weaning from mechanical ventilation may be difficult, and values less positive than 60 cm H2O suggest cough insufficiency.3 However, these values have poor predictive value in mechanically ventilated patients because many of these patients are unable to cooperate. This problem may be addressed by attaching a one-way valve to the end of an endotracheal tube that permits exhalation but not inhalation and then measuring the inspiratory pressure efforts for 20 to 25 seconds.

MEASURING WORK OF BREATHING

Work of breathing is the product of pressure and volume for each breath (Fig. 103-3). The components include work needed to overcome elastic recoil of the lung and to displace the chest wall and abdomen as well as work needed to overcome airway resistance and lung viscosity and work needed to overcome inertia. With restrictive lung diseases, the inspiratory work of breathing is increased because of the decreased lung elasticity. With obstructive diseases, the work of breathing is increased because of increased airway resistance. In clinical settings, a more practical way to assess the inspiratory work of breathing is to calculate the pressure-time product (in cm H2O-seconds). The pressure-time product can be calculated by the decrease in airway pressure during inspiration, esophageal pressure (measured with an esophageal balloon manometer), or transdiaphragmatic pressure (measured with esophageal and gastric balloon manometers) as an index of diaphragmatic work. The work can be calculated as work of breathing per breath or as work of breathing per minute by multiplying the work per breath by the respiratory frequency. Commercially available devices using esophageal manometry automatically calculate the inspiratory work of breathing, which may be of some value in assessing the likelihood of weaning from mechanical ventilation. If the drop in inspiratory pressure necessary to achieve an adequate tidal volume is too large, the calculated work of breathing will be high, and the likelihood of successful weaning will be reduced.

Grade A Reference A1. Schuh S, Freedman S, Coates A, et al. Effect of oximetry on hospitalization in bronchiolitis: a randomized clinical trial. JAMA. 2014;312:712-718.

CHAPTER 103  Respiratory Monitoring in Critical Care  

GENERAL REFERENCES 1. Pretto JJ, Roebuck T, Beckert L, et al. Clinical use of pulse oximetry: official guidelines from the Thoracic Society of Australia and New Zealand. Respirology. 2014;19:38-46. 2. Ortega R, Connor C, Kim S, et al. Monitoring ventilation with capnography. N Engl J Med. 2012;367:e27. 3. McConville JF, Kress JP. Weaning patients from the ventilator. N Engl J Med. 2013;368:1068-1069.

655.e1

655.e2

CHAPTER 103  Respiratory Monitoring in Critical Care  

REVIEW QUESTIONS 1. Paco2 (arterial carbon dioxide tension) is determined by A. CO2 production rate and minute ventilation B. Carbonic anhydrase level and alveolar ventilation C. CO2 production rate and alveolar ventilation D. Diffusing capacity and CO2 production rate E. Minute ventilation and dead space (Vd/Vt) Answer: C  The Paco2 is determined by the rate of CO2 production and the alveolar ventilation. 2. A patient’s ambient air arterial blood gas shows a Pao2 of 60 torr and a Paco2 of 60 torr (at sea level). What is this patient’s alveolar-arterial oxygen gradient? A. 4 torr B. 14 torr C. 24 torr D. 34 torr E. 37 torr Answer: B  The alveolar-arterial oxygen gradient at sea level (barometric pressure = 760 mm Hg) is calculated as 149 − ([1.25] Paco2 + Pao2). The alveolar-arterial oxygen gradients with these arterial blood gas values = 14 torr (or mm Hg).

3. Variables that affect the ratio of dead space to tidal volume (Vd/Vt) include A. End-tidal CO2 and Paco2 B. Exhaled CO2 tension (Peco2) and Pao2 C. Pao2 and CO2 production rate D. Exhaled CO2 tension (Peco2) and Paco2 E. Exhaled CO2 tension (Peco2) and end-tidal CO2 tension Answer: D  The Vd/Vt is calculated by the Bohr equation, which is (Paco2 − Peco2)/Paco2.

CHAPTER 104  Acute Respiratory Failure  

655

104  ACUTE RESPIRATORY FAILURE MICHAEL A. MATTHAY AND ARTHUR S. slu*tSKY

DEFINITION

Acute respiratory failure occurs when dysfunction of the respiratory system results in abnormal gas exchange that is potentially life-threatening. Each element of this definition is important to understand. The term acute implies a relatively sudden onset (from hours to days) and a substantial change from the patient’s baseline condition. Dysfunction indicates that the abnormal gas exchange may be caused by abnormalities in any element of the respiratory system (e.g., a central nervous system abnormality affecting the regulation of breathing or a musculoskeletal thoracic abnormality affecting ventilation [Chapter 83]) in addition to abnormalities of the lung itself. The term respiration refers, in a broad sense, to the delivery of oxygen (O2) to metabolically active tissues for energy use and the removal of carbon dioxide (CO2) from these tissues (Table 104-1). Respiratory failure is a failure of the process of delivery of O2 to the tissues or removal of CO2 from the tissues. Abnormalities in the periphery (e.g., cyanide poisoning, circulatory shock, pathologic distribution of organ blood flow in sepsis) can lead to tissue hypoxia; although these conditions represent forms of respiratory failure in the broadest terms, this chapter focuses on respiratory failure resulting from dysfunction of the lungs, chest wall, and control of respiration.

PATHOBIOLOGY

Abnormal gas exchange is the physiologic hallmark of acute respiratory failure, which can be classified in several ways (Table 104-2). Although gas exchange can be abnormal for either oxygenation or CO2 removal, significant hypoxemia is nearly always present when patients with acute respiratory failure breathe ambient air. If CO2 is retained at a potentially life-threatening level under these conditions, it must be accompanied by significant hypoxemia (see later). The life-threatening aspect of the condition places the degree of abnormal gas exchange in a clinical context and calls for urgent treatment. The diagnosis of acute respiratory failure requires a significant change in arterial blood gases from baseline. Many patients with chronic respiratory problems can function with blood gas tensions that would be alarming in a physiologically normal individual. Over time, patients with so-called chronic respiratory failure or chronic respiratory insufficiency develop mechanisms to compensate for inadequate gas exchange. Conversely, this chronic condition makes patients vulnerable to insults that could be easily tolerated by a previously healthy individual. In acute respiratory failure, the O2 content in the blood (available for tissue use) is reduced to a level at which the possibility of end-organ dysfunction increases markedly. The value of the partial pressure of O2 in the arterial blood (Pao2) that demarcates this vulnerable zone is often considered to be the point of the oxyhemoglobin dissociation relationship at which any further decrease in the Pao2 results in sharp decreases in the amount of hemoglobin saturated with O2 (Sao2) and in the arterial blood O2 content (Cao2). Thus, acute respiratory failure is often defined in practice as occurring when the Pao2 is less than about 55 mm Hg (Fig. 104-1). The oxyhemoglobin dissociation curve of venous blood, which is the partial pressure at which O2 is being unloaded to the tissues, is a critical determinant of how much O2 is available for the cells and their mitochondria. Other than under conditions of an extremely hypoxic environment (e.g., in utero or on the summit of Mt. Everest), the enhanced ability to unload O2 at the tissue level more than compensates for small decreases in the amount of O2 picked up in the lungs when the oxyhemoglobin dissociation curve is shifted rightward. With a leftward shift in the curve, O2 is bound more tightly to hemoglobin, so less O2 is available for tissue delivery. These clinical considerations imply that any definition of acute respiratory failure based on an absolute level of Pao2 is arbitrary. A healthy, young, conditioned individual climbing at high altitude may have a Pao2 of less than

656

CHAPTER 104  Acute Respiratory Failure  

TABLE 104-1  ABBREVIATIONS COMMONLY USED IN ACUTE RESPIRATORY FUNCTION

Acute respiratory distress syndrome

ARF

Acute respiratory failure

cm H2O

Centimeters of water

Cao2

Content of oxygen in arterial blood

Cco2

Content of oxygen in end-capillary blood

CO2

Carbon dioxide

COPD

Chronic obstructive pulmonary disease

CPAP

Continuous positive airway pressure (used when positive pressure during exhalation is applied with spontaneous ventilation)

Cvo2

Content of oxygen in mixed venous blood

Fio2

Fraction of inspired oxygen

g/dL

Grams per deciliter

HbO2

Saturation of hemoglobin by oxygen

L/min

Liters per minute

mL/kg

Milliliters per kilogram

mL/min

Milliliters per minute

mm Hg

Millimeters of mercury

NIPPV

Noninvasive positive-pressure ventilation

O2

Oxygen

P(A-a)o2

Difference of partial pressure of oxygen between mean alveolar gas and arterial blood (alveolar-to-arterial oxygen difference)

Paco2

Partial pressure of carbon dioxide in alveolar gas

Paco2

Partial pressure of carbon dioxide in arterial blood

Pao2

Partial pressure of oxygen in alveolar gas

Pao2

Partial pressure of oxygen in arterial blood

Pao2/Fio2

Ratio of partial pressure of oxygen in arterial blood to fraction of inspired oxygen

PBW

Predicted body weight

Pc CO2

Partial pressure of carbon dioxide in end-capillary blood

PCO2

Partial pressure of carbon dioxide

Pc O2

Partial pressure of oxygen in end-capillary blood

PEEP

Positive end-expiratory pressure (used when positive pressure during exhalation is applied with mechanical ventilation)

P/F

Pao2/Fio2 ratio

Pio2

Partial pressure of oxygen in inspired gas

Po2

Partial pressure of oxygen

Pvco2

Partial pressure of carbon dioxide in mixed venous blood

Pvo2 Q

Partial pressure of oxygen in mixed venous blood

RR

Respiratory rate

Sao2

Percentage of saturation of hemoglobin by oxygen in arterial blood

V

Ventilation

V/Q

Ventilation-perfusion ratio

Vt

Tidal volume

Blood flow or perfusion

50 mm Hg because of the reduction in inspired O2 pressure.1 This individual is not in acute respiratory failure, even though the Pao2 may be in the low 40s. A patient who has chronic obstructive pulmonary disease (COPD) and whose usual range of Pao2 is 50 to 55 mm Hg would not be considered to be in acute respiratory failure if the Pao2 was 50 mm Hg. However, if a patient’s usual Pao2 is 80 mm Hg, a sudden drop to a Pao2 of 50 mm Hg could be associated with a substantial risk for a further life-threatening reduction in oxygenation; this patient should be considered to have acute respiratory failure. Traditionally, the level of arterial CO2 partial pressure (Paco2) that defines acute respiratory failure has been 50 mm Hg or greater, if it is accompanied

80 15 60

40

10

CaO2 (mL/dL)

Acute lung injury

ARDS

Normal arterial

ALI

20

Mixed venous

Arterial blood gas or arterial blood gas analysis

100

SaO2 (%)

ABG

CaO2 SaO2 Dissolved

5

20

0 0

20

40

60

80

100

600

PaO2 (mm Hg) FIGURE 104-1.  Oxyhemoglobin association-dissociation curve. The axis for oxygen saturation in the arterial blood (SaO2) is on the left, and the axis for arterial content of oxygen (CaO2) is on the right. CaO2 is the sum of the oxygen dissolved in plasma (denoted as “Dissolved” in the figure) plus the oxygen bound to hemoglobin. With a normal hemoglobin, most of the oxygen is carried in combination with hemoglobin, with only a relatively small amount of oxygen dissolved in plasma. When the value of the arterial partial pressure of oxygen (PaO2) is on the “flat” portion of the curve (PaO2 ≥ 60 to 65 mm Hg, normal partial pressure of arterial carbon dioxide [PaCO2], and normal pH), raising the PaO2 further has relatively little effect on total oxygen content. Increases in temperature, PCO2, hydrogen ion concentration, or 2,3-diphosphoglycerate cause a rightward shift in the oxyhemoglobin association-dissociation curve.

by arterial acidosis with a pH of less than about 7.30. The Paco2 is linked to pH in this definition because of the general belief that acidosis is what leads to tissue dysfunction and symptoms. Patients with severe COPD may have chronic CO2 retention, but renal compensation for the respiratory acidosis protects them against abnormalities related to the elevation in CO2. A further acute rise in Paco2 can precipitate symptoms and other organ dysfunction; however, even severe respiratory acidosis (pH 7.1) seems to be better tolerated than metabolic acidosis of the same pH in most previously healthy individuals if arterial and tissue oxygenation is adequate.

Pathophysiology

Five mechanisms can lead to a reduction in Pao2: (1) decreased inspired partial pressure of O2 (Pio2) (e.g., at high altitude or when breathing a reduced percentage O2 mixture); (2) hypoventilation; (3) ventilation ) mismatch; (4) shunting of blood from the pulmonary to perfusion ( V/Q systemic circulation, bypassing the alveoli anatomically or functionally; and (5) any barrier for diffusion of O2 from the alveoli into the capillary blood. mismatch in which blood perfuses In essence, a shunt is an extreme V/Q alveoli with no ventilation; it is differentiated clinically from other V/Q mismatching by the response to breathing of supplemental O2 (see later). For clinical purposes, diffusion abnormalities are not usually important causes of hypoxemia at sea level because there is sufficient time for adequate diffusion of O2 during the transit of a red blood cell through the pulmonary capillary bed, even in the presence of severe lung disease. When diffusion mismatch abnormalities are present and contribute to hypoxemia, V/Q nearly always coexists with the shunting, and this mismatch is an important cause of hypoxemia. Except at high altitude or when a subject is breathing a mismatch, and shunting are the gas mixture low in O2, hypoventilation, V/Q dominant causes of hypoxemia. If only hypoventilation is present, the resulting hypoxemia is associated with a normal difference between the calculated alveolar and the measured arterial oxygenation levels [P(A-a)o2]. In this setting, an elevated Paco2 suggests disease processes that affect nonpulmonary respiratory function (e.g., central respiratory depression resulting from drug overdose, neuromuscular diseases such as Guillain-Barré syndrome, or chest wall disease such as flail mismatch and shunting are associated chest; Chapter 86). In contrast, V/Q with an elevated P(A-a)o2, which may or may not coexist with hypoventilation. The normal value for P(A-a)o2 varies as a function of the fraction of inspired O2 (Fio2), increasing as Fio2 increases.

CHAPTER 104  Acute Respiratory Failure  

657

TABLE 104-2  SYSTEMS TO CLASSIFY ACUTE RESPIRATORY FAILURE HYPOXIC VERSUS HYPERCAPNIC-HYPOXEMIC Causes of Hypoxemic Acute Respiratory Failure Acute lung injury/ARDS Pneumonia Pulmonary thromboembolism Acute lobar atelectasis Cardiogenic pulmonary edema Lung contusion Acute collagen vascular disease (Goodpasture syndrome, systemic lupus erythematosus) Causes of Hypercapnic-Hypoxemic Acute Respiratory Failure Pulmonary disease   COPD   Asthma: advanced, acute, severe asthma Drugs causing respiratory depression Neuromuscular   Guillain-Barré syndrome   Acute myasthenia gravis Spinal cord tumors Metabolic derangements causing weakness (including hypophosphatemia, hypomagnesemia) Musculoskeletal   Kyphoscoliosis   Ankylosing spondylitis Obesity hypoventilation syndrome (often with additional acute, superimposed abnormality as cause of acute respiratory failure) ETIOLOGIC MECHANISMS OF HYPOXEMIA Normal P(A-a)O2* ↓Pio2   High altitude; inadvertent administration of low Fio2 gas mixture   Hypoventilation   See causes of hypercapnic-hypoxic acute respiratory failure above Increased P(A-a)O2* ) mismatch Ventilation-perfusion ( V/Q   Airway disease   Vascular disease, including pulmonary thromboembolism Shunt   Acute lung injury/ARDS   Pneumonia   Parenchymal lung disease   Cardiogenic pulmonary edema   Pulmonary infarction Diffusion limitation†

ACUTE RESPIRATORY FAILURE WITH AND WITHOUT CHRONIC LUNG DISEASE With Chronic Lung Disease COPD Asthma Parenchymal lung diseases Restrictive lung/chest wall diseases Without Chronic Lung Disease‡ Acute lung injury/ARDS Pneumonia Pulmonary thromboembolism ACUTE RESPIRATORY FAILURE BY ORGAN SYSTEM INVOLVED Respiratory (Lungs and Thorax) Airway/airflow obstruction   COPD   Asthma Pulmonary parenchyma   Pneumonia   ARDS   Acute flare of chronic collagen vascular disease (e.g., Goodpasture syndrome, systemic lupus erythematosus) Central Nervous System Respiratory depression   Increased sedatives, tranquilizers with respiratory effect; opiates; alcohol Brain stem and spinal cord involvement   Tumors, trauma, vascular accidents Neuromuscular Guillain-Barré syndrome Myasthenia gravis Cardiovascular Cardiogenic pulmonary edema Pulmonary thromboembolism Renal/Endocrine Volume overload Metabolic abnormalities

*Calculated by the alveolar-air equation; see text for description. † See text for discussion. ‡ These can also be superimposed on chronic disease. ARDS = acute respiratory distress syndrome; COPD = chronic obstructive pulmonary disease; Fio2 = fraction of inspired oxygen; P(A-a)o2 = alveolar-to-arterial oxygen difference; Pio2 = partial pressure of = ventilation-perfusion ratio. inspired oxygen; V/Q

mismatch or shunting is the cause of hypoxemia, some When V/Q alveolar regions have increased levels of Pco2 and associated reduced levels of Po2; the blood in the vessels perfusing these alveoli reflects these abnormal gas tensions. The resulting increased arterial Pco2 (Paco2) usually can be reversed by increasing overall ventilation, but this increased ventilation usually does not correct the decreased arterial Po2 (Pao2). mismatch is distinguished from shunting by assessing the Pao2 V/Q misresponse to enhanced O2 administration. Hypoxemia caused by V/Q match can be corrected to a nearly complete O2 saturation of the hemoglobin in most patients by a relatively small increase in Fio2, such as from 0.24 to 0.28 by face mask or 1 to 2 L/minute O2 by nasal prongs, in patients with acute exacerbations of COPD. If the airways to poorly ventilated alveoli remain open and the enriched O2 mixture is administered for an adequate length of time (ranging from a few minutes to about 20 minutes, depending inequality), the increased Pio2 is reflected by an on the degree of V/Q increased Pao2 and an increased Pao2. When a shunt is present (no ventilation but continued perfusion), a relatively small increase in the Fio2 has little or no effect on the Pao2, and even large increases in Fio2 up to 1.0 result in only modest increases in Pao2 (Fig. 104-2).

CLINICAL MANIFESTATIONS

The hallmark of acute respiratory failure is the inability to maintain adequate oxygenation or the inability to maintain an appropriate Paco2. Patients are

typically dyspneic and tachypneic, unless progressive respiratory failure causes fatigue—sometimes leading to respiratory arrest—or a drug overdose or neuromuscular condition prevents an appropriate respiratory response to hypoxemia or hypercapnic acidosis. Neurologic function may deteriorate, and myocardial ischemia or even infarction may be precipitated by hypoxemia. In addition, each cause has its own specific manifestations (see later).

DIAGNOSIS

As part of the diagnosis of acute respiratory failure, the physician has three objectives: (1) to confirm the clinical suspicion that acute respiratory failure is present, (2) to classify the type of acute respiratory failure (e.g., hypoxemia caused by mismatch or shunting), and (3) hypoventilation vs. hypoxemia caused by V/Q to determine the specific cause (e.g., the acute respiratory distress syndrome [ARDS]) secondary to pulmonary or nonpulmonary sepsis or decompensated COPD because of acute bronchitis. Defining the type of acute respiratory failure and determining the specific cause are prerequisites to optimal management. The initial approach to diagnosis consists of considering information from four sources: (1) clinical history and physical examination; (2) physiologic abnormalities, particularly arterial blood gas derangements, which help establish the mechanisms of hypoxemia; (3) chest radiographic findings; and (4) other tests aimed at elucidating specific causes. In many cases, the clinical picture from the history is so clear that the presumptive type of acute respiratory failure (and sometimes the cause) is obvious, so treatment can be started

658

CHAPTER 104  Acute Respiratory Failure  

•

•

Valv = 6.1 PaO2 = 41 PaCO 2 = 46

Valv = 5.6

PaO2 = 117 PaCO 2 = 32

PaO2 = 105 PaCO 2 = 37

Cv O 2 = 11.9 Pv O 2 = 32 Pv CO 2 = 46 C cO 2 = 14.8 P cO 2 = 41 P cCO 2 = 46

Cv O 2 = 11.9 Pv O 2 = 31 Pv CO 2 = 46 C cO 2 = 20.1 P cO 2 = 117 P cCO 2 = 32

C cO 2 = 11.9 P cO 2 = 32 P cCO 2 = 46

Ca O 2 = 16.9 Pa O 2 = 50 PaCO 2 = 40

FIO2 = 0.21

A

C cO 2 = 19.9 P cO 2 = 105 P cCO 2 = 37 Ca O 2 = 16.9 PaO2 = 50 PaCO 2 = 40 •

•

Valv = 5.8

Valv = 6.3 PaO2 = 665 PaCO 2 = 48

PaO2 = 682 PaCO 2 = 31

PaO2 = 677 PaCO 2 = 36

Cv O 2 = 17.0 Pv O 2 = 53 Pv CO 2 = 48 C cO 2 = 22 P cO 2 = 665 P cCO 2 = 48

FIO2 = 1.0

Cv O 2 = 14.1 Pv O 2 = 38 Pv CO 2 = 46 C cO 2 = 22 P cO 2 = 682 P cCO 2 = 31

C cO 2 = 14.1 P cO 2 = 38 P cCO 2 = 46

Ca O 2 = 22 PaO2 = 672 PaCO 2 = 40

C cO 2 = 22 P cO 2 = 667 P cCO 2 = 36 Ca O 2 = 19.1 PaO2 = 76 PaCO 2 = 40

B FIGURE 104-2.  Arterial oxygenation. Comparison of the effect on arterial oxygenation of increasing the fraction of inspired oxygen (FIO2) from breathing of ambient air (FIO2 = (left) and a shunt (right), using a two-compartment lung model. Shunting and 0.21) (A) and breathing of 100% oxygen (FIO2 = 1.0) (B) with a low ventilation-perfusion ratio ( V/Q) can lead to identical arterial blood gases (partial pressure of oxygen in arterial blood [PaO2] = 50 mm Hg; partial pressure of carbon dioxide in arterial blood [PaCO2] = decreased V/Q 40 mm Hg). The response to supplemental oxygen administration is markedly different. Hypoxemia is only partially corrected by breathing of 100% oxygen when a shunt is present because arterial oxygenation represents an average of the end-capillary oxygen content (CcO2) from various parts of the lung, not an average of the partial pressures of oxygen (partial pressure of carbon dioxide in the end-capillary blood [PcCO2]). When the CcO2 values are mixed, the PaO2 is determined from the resultant content of oxygen in the arterial blood (as is often the case in patients with chronic obstructive pulmonary disease), an (CaO2) by the oxyhemoglobin association-dissociation relationship (see Fig. 104-1). With low V/Q increase in FIO2 increases the alveolar partial pressure of oxygen (PO2) of the low V/Qunit and leads to a marked increase in arterial PO2. The values in this figure were generated from modeling to result in the same PaCO2 (40 mm Hg) for all four situations shown; this is the reason for slight changes in alveolar ventilation ( Valv) for some of the conditions. Several assumptions are made: no diffusion limitation is present; oxygen consumption = 300 mL/minute, and CO2 production = 240 mL/minute; cardiac output = 6.0 L/minute; the low V/Q regions in the left panels represent 60% of the cardiac output perfusing alveoli with a V/Q 25% of normal; and the shunts in the right panels represent a 37% shunt (i.e., 37% of the cardiac output is perfusing alveoli with no ventilation).

while confirmatory laboratory studies are ordered. In other cases, a clinician may be asked to see a patient because of an abnormal chest radiograph or abnormal arterial blood gases ordered by someone else and may elicit the pertinent history based on these clues. When the degree of hypoxemia is life-threatening, therapeutic decisions must be made quickly, even if data are limited. The clinician must obtain updated information continually and should view most therapeutic decisions as therapeutic trials, with careful monitoring to assess desired benefits and possible detrimental effects.

HYPOXEMIA

HYPERCAPNIA

Tachycardia

Somnolence

Tachypnea

Lethargy

Anxiety

Restlessness

Clinical Evaluation

Diaphoresis

Tremor

Altered mental status

Slurred speech

Confusion

Headache

Cyanosis

Asterixis

Hypertension

Papilledema

Hypotension

Coma

Bradycardia

Diaphoresis

The presentation often reflects one of three clinical scenarios: (1) the effects of hypoxemia or respiratory acidosis, (2) the effects of primary (e.g., pneumonia) or secondary (e.g., heart failure) diseases affecting the lungs, and (3) the nonpulmonary effects of the underlying disease process. The clinical effects of hypoxemia and respiratory acidosis are manifested mainly in the central nervous system (e.g., irritability, agitation, changes in personality, depressed level of consciousness, coma) and the cardiovascular system (e.g., arrhythmias, hypotension, hypertension) (Table 104-3). In patients with underlying COPD (Chapter 88) with a gradual onset of acute respiratory failure, central nervous system abnormalities may be the major presenting findings. Cyanosis, which requires at least 5 g/dL of unsaturated hemoglobin to be detectable, may not be seen before serious tissue hypoxia develops, especially in patients with underlying anemia.

TABLE 104-3  CLINICAL MANIFESTATIONS OF HYPOXEMIA AND HYPERCAPNIA

Seizures Coma Lactic acidosis* *Usually requires additional reduction in oxygen delivery because of inadequate cardiac output, severe anemia, or redistribution of blood flow.

CHAPTER 104  Acute Respiratory Failure  

Pulmonary symptoms and signs often reflect the respiratory disease causing the acute respiratory failure. Examples include cough and sputum with pneumonia (Chapter 97) or chest pain from pulmonary thromboembolism with infarction (Chapter 98). Dyspnea and respiratory distress are nonspecific reflections of the respiratory system’s difficulty in meeting the increased demands from pulmonary and nonpulmonary diseases. Physical findings may be associated with a particular pathologic lung process, such as pneumonia (Chapter 97), which often results in bronchial breathing and crackles on auscultation, or the crackles (rales) of cardiogenic pulmonary edema (Chapter 58). Abnormal findings may be minimal or absent in patients with ARDS or pulmonary thromboembolism (Chapter 98). In some patients, the clinical picture is dominated by the underlying disease process, particularly with diseases that cause ARDS, such as sepsis (Chapter 108), severe pneumonia (Chapter 97), aspiration of gastric contents (Chapter 94), and trauma. In these conditions, the physical examination findings are often nonspecific, with no obvious clues except, for example, fever with sepsis or pneumonia and hypotension with septic shock.

659

A

B

Assessment of Physiologic Abnormalities

The clinical suspicion of acute respiratory failure must be addressed by arterial blood gas analysis to answer several questions. Is hypoxemia present? The answer is based largely on the value of the Pao2 or Sao2. The degree of the hypoxemia not only confirms the diagnosis of acute respiratory failure but also helps define its severity. Is hypoventilation present? If the Paco2 is elevated, alveolar hypoventilation is present. Does the degree of hypoventilation fully explain the hypoxemia? If the P(A-a)o2 is normal, hypoventilation fully explains the presence and degree of hypoxemia. When this is the case, the most likely causes of acute respiratory failure are central nervous system abnormalities or a chest wall abnormality. If the P(A-a)o2 is increased but hypoventilation does not fully explain the hypoxemia, another condition must be present; common diagnoses include COPD (Chapter 88), severe asthma (Chapter 87), pneumonia (Chapter 97), and early stages of ARDS. If hypoxemia exists without hypoventilation, an elevated P(A-a)o2 should be confirmed, and the response to breathing of an enhanced O2 mixture would answer this question: Is the increase in P(A-a)O2 the result of a V/Q abnormality or of shunting? If hypoxemia is primarily the result of a V/Q abnormality, the likely cause is an airway disease, either COPD or acute severe asthma, or a vascular disease, such as pulmonary thromboembolism. If shunting is the major explanation for the hypoxemia, processes that fill the air spaces (e.g., cardiogenic pulmonary edema, noncardiogenic pulmonary edema or ARDS, or purulent pulmonary secretions in acute pneumonia) or, less commonly, an intracardiac or anatomic intrapulmonary shunt is the likely cause. Conditions that fill air spaces should be confirmed by abnormal findings on a chest radiograph; if the radiograph is normal, the possibility of intracardiac shunt (Chapter 69) or thromboembolism (Chapter 98) should be evaluated.

Chest Radiography

The chest radiograph in acute respiratory failure is likely to show one of three patterns (Fig. 104-3): (1) normal (or relatively normal), (2) localized alveolar filling opacities, or (3) diffuse alveolar filling opacities. Diffuse interstitial opacities are also possible, but diseases that cause this pattern usually have a more gradual onset and are associated with chronic respiratory failure. If the chest radiograph is normal (i.e., it is clear or relatively clear), airway diseases, such as COPD and asthma, or pulmonary vascular diseases, such as thromboembolism, are more likely. If a localized alveolar filling abnormality is present, pneumonia is the major consideration, but pulmonary embolism and infarction should also be considered. When diffuse (bilateral) alveolar filling abnormalities are present, cardiogenic pulmonary edema and ARDS (e.g., as seen after sepsis, trauma, pneumonia, or aspiration of gastric contents) are the major considerations. The combination of the chest radiograph and the arterial blood gas interpretation can be helpful. The finding of a significant shunt may suggest ARDS in a patient in whom this diagnosis was not clinically obvious; the chest radiograph should help confirm that possibility.

Other Evaluations

All patients with acute respiratory failure should have a complete blood count including a platelet count, routine blood chemistry tests, prothrombin time, and urinalysis to screen for possible underlying causes and comorbid conditions. Other blood tests should be guided by the clinical picture. Examples

C FIGURE 104-3.  Chest radiographs (left) and computed tomography scans (right) of the three most common findings in diseases causing acute respiratory failure. A, Relatively clear chest, consistent with an acute exacerbation of airway disease (e.g., asthma, chronic obstructive pulmonary disease) or a central nervous system or neuromuscular disease as the cause of acute respiratory failure. B, Localized alveolar filling opacity, most commonly seen with acute pneumonia. C, Diffuse bilateral alveolar filling opacities consistent with acute lung injury and acute respiratory distress syndrome. The computed tomography scan in C shows a small left pneumothorax and cavities or cysts that are not apparent on the anteroposterior chest radiograph.

include a serum amylase level if pancreatitis is a possible cause of ARDS and thyroid indices if severe hypothyroidism is a possible cause of hypoventilation. Blood cultures are recommended when an infectious cause such as sepsis is suspected. Any abnormal fluid collections, especially pleural effusion (Chapter 99), should be aspirated for diagnostic purposes. Sputum Gram stain and culture are indicated when pneumonia is suspected. Other specific tests should be directed by the history, physical examinations, arterial blood gas levels, and chest radiograph. An abdominal computed tomography (CT) scan may be indicated to search for the source of infection in a patient with sepsis and ARDS. A chest CT scan may help define pulmonary disease if the chest radiograph is not definitive. CT arteriography of the pulmonary circulation may diagnose pulmonary thromboembolism (Chapter 98). A head CT scan may be indicated if a stroke involving the respiratory center is suspected. Routine blood chemistry studies can detect diabetic ketoacidosis or renal failure as contributing causes.

TREATMENT  General Measures

The management of acute respiratory failure depends on its cause, its clinical manifestations, and the patient’s underlying status. Certain goals apply to all patients: improvement of the hypoxemia to eliminate or markedly reduce the acute threat to life; improvement of the acidosis if it is considered lifethreatening; maintenance of cardiac output or improvement if cardiac output is compromised; treatment of the underlying disease process; and avoidance of predictable complications. The precise methods for improving hypoxemia depend on the cause of the acute respiratory failure. However, an increase in the inspired O2 concentration is a cornerstone of treatment for nearly all patients, even though it may not produce a marked increase in Pao2 in patients whose underlying pathophysiologic process involves a significant amount of lung with low ventilationperfusion ratios or true shunting. The level of acidosis that requires treatment other than for the underlying disease process is a matter of debate. Although normalization of the arterial pH was suggested in the past, respiratory acidosis is apparently well tolerated in many patients with severe ARDS, so a patient with a pH of 7.15 or higher may not require bicarbonate therapy. If the acidemia coexists with clinical complications, such as cardiac arrhythmias or a decreased level of consciousness, that have no other obvious cause, treatments to increase pH should be

660

CHAPTER 104  Acute Respiratory Failure  

considered. The therapeutic goal is alleviation or reduction of the accompanying complications by improving the level of acidosis; normalization of the pH usually is not indicated (Chapter 118). The maintenance of cardiac output is crucial for O2 delivery in acute respiratory failure, especially because mechanical ventilation and positive endexpiratory pressure (PEEP) may compromise cardiac output. Placement of a pulmonary artery catheter allows measurement of cardiac output and filling pressures, but most patients who have these catheters do no better than similar patients managed without them. A1  Nevertheless, selective use of diagnostic pulmonary artery catheterization can help determine the cause of the pulmonary edema (cardiogenic vs. noncardiogenic) and the physiologic basis for shock (sepsis, hypovolemia, or decreased cardiac output from impaired cardiac function) in selected patients in whom either is not clear.2 Many therapeutic interventions that improve short-term physiologic variables may worsen long-term, clinically important outcomes. For example, transfusing all patients to maintain a hemoglobin greater than 10 g/dL increases mortality in critically ill patients who have not had an acute myocardial infarction and do not have unstable angina, even though the O2-carrying capacity of the blood is acutely increased. Use of a relatively large tidal volume (e.g., 12 mL/kg predicted body weight, which is equivalent to approximately 10 to 10.5 mL/kg measured body weight in patients who are somewhat overweight) increases mortality in patients with ARDS compared with a lower tidal volume (6 mg/kg predicted body weight), even though it raises Pao2 more in the short term than does a lower tidal volume. Conservative use of fluids when vasopressors are no longer required to support the systemic blood pressure improves lung function and shortens the duration of mechanical ventilation and intensive care. A2  Improvements in oxygenation, acid-base status, and cardiac output are of no more than temporary benefit unless the underlying disease processes are diagnosed and treated properly. In patients with ARDS, sepsis may worsen injury to the lung and other organs despite optimal supportive care. Similarly, if the precipitating cause of acute respiratory failure in a patient with COPD is not identified and treated, supportive care is likely to be futile. Complications may arise from the physiologic effects of the gas exchange abnormality, from the disease processes causing the acute respiratory failure, from being critically ill and its associated incursions on homeostasis (e.g., sleep deprivation), or from iatrogenic complications of therapy.

Mechanical Therapy to Improve Oxygenation

A Pao2 greater than 60 mm Hg is usually adequate to produce an Sao2 in the low to middle 90s. The Pao2 can be increased by the administration of supplemental O2, by pharmacologic manipulations, by continuous positive airway pressure (CPAP), by mechanical ventilation with or without maneuvers such as PEEP, and by the prone position. PEEP, pharmacologic manipulations, and positioning are used primarily in patients with ARDS (see later). The initial choice of the concentration and amount of supplemental O2 is based on the severity of the hypoxemia, the clinical diagnosis, the likely mechanism causing the hypoxemia, and the O2 delivery systems available. For the tracheal Fio2 to be the same as the delivered Fio2, the O2 delivery system must deliver a flow that matches the patient’s peak inspiratory flow rate with gas of a known Fio2. High-flow O2 blenders can achieve this goal by delivering gas at 80 L/minute or more to a nonintubated patient. These systems require a large flow of O2 (from a wall unit or tank), however, and are not universally available. Other systems for nonintubated patients (including nasal prongs, simple face masks, and non-rebreather and partial rebreather masks) use a simple regulator that mixes room air with O2 from a wall unit or tank, with resulting flows that are frequently unable to match the patient’s peak inspiratory flow rate. The patient entrains more air from the environment, and the resulting tracheal Fio2 or partial pressure of oxygen in inspired gas (Pao2) is unknown. The amount of air entrained depends on the patient’s inspiratory pattern and minute ventilation. Although the resulting Fio2 is unknown, these systems are satisfactory if the delivery is constant and if they result in adequate arterial O2 saturation, as monitored by arterial blood gases or oximetry. Nasal prongs can deliver a tracheal Fio2 of approximately 0.50, and non-rebreather masks can deliver 50 to 100% O2; in both cases, this depends on the inspiratory pattern mismatch is present, only a small and flow rate. If only hypoventilation or V/Q increment in Fio2 (e.g., an Fio2 of 0.24 or 0.28 delivered by a Venturi principle face mask or by mechanical ventilation; or 1 to 2 L/minute O2 delivered by nasal prongs) is likely to be required. By comparison, if marked shunting or are the cause of hypoxemia, a many lung units with low but not zero V/Q considerably higher Fio2 (e.g., >0.7) may be required, and even this high Fio2 may not reverse the hypoxemia. A common practice when a significant shunt is suspected is to give an Fio2 of 1.0, then adjust the Fio2 downward as guided by the resulting Pao2 or Sao2. The O2 concentration that is toxic to the lungs in critically ill patients is not known, but prior injury may provide tolerance to O2 toxicity, whereas other conditioning agents, such as bleomycin, may enhance oxidative injury. An Fio2

of 0.7 or higher is generally considered injurious to the normal human lung. Because it is unknown what lower concentration is safe, however, patients should be given the lowest Fio2 that provides an adequate Sao2 (≥90%). If an Fio2 equal to or greater than 0.5 to 0.7 is required for adequate oxygenation, other measures, especially PEEP or CPAP, should be considered. Even a lower Fio2 of about 0.5 may be associated with impaired ciliary action in the airways and impaired bacterial killing by alveolar macrophages, but the clinical importance of these effects is not known. A low concentration of supplemental O2 can be administered by nasal prongs or nasal cannula, which most patients find comfortable and allows them to cough, speak, eat, and drink while receiving O2. When the nasal passages are open, the Pio2 does not depend too much on whether the patient breathes through the nose or the mouth because O2 is entrained from the posterior nasal pharynx during a breath taken through the mouth. The level of O2 can be adjusted by the flow rate to the nasal prongs. In patients with COPD, flows as low as 0.5 to 2 L/minute are usually adequate unless an intrapulmonary shunt is contributing to the hypoxemia, as usually occurs in  acute pneumonia. At flows greater than approximately 6 L/minute, only a small further augmentation in the Pio2 can be achieved. Because gas flow through the nose has a drying and irritating effect, a face mask should be considered at high flow rates. O2 face masks using the Venturi principle allow the regulation of Fio2 and can be particularly useful when COPD is suspected, and it is important to avoid the CO2 retention that can be associated with the unregulated administration of O2. A higher Fio2 of 0.5 to nearly 1.0 can be administered through a non-rebreathing face mask with an O2 reservoir. If an Fio2 equal to or greater than 0.70 is required for more than several hours, particularly in an unstable patient, endotracheal intubation should be considered so O2 can be administered by a closed system with reliable maintenance of the patient’s Sao2. Indications for placement of an artificial airway in a patient with acute respiratory failure are to protect the airway against aspiration of gastric contents, to deliver an increased Fio2, to facilitate prolonged mechanical ventilation, and possibly to aid in the control of respiratory secretions (Chapter 105). Ventilatory maneuvers that may increase arterial oxygenation include mechanical ventilation itself and the administration of PEEP or CPAP, all of which allow ventilation of areas of the lung that were previously poorly ventilated or unventilated. Although large tidal volumes with mechanical ventilation may open areas of atelectasis and may improve oxygenation initially, these higher tidal volumes can cause lung injury, particularly if the lung is already injured (Chapter 105).2 CPAP refers to the maintenance of positive pressure during the respiratory cycle while breathing spontaneously. PEEP refers to the maintenance of positive pressure throughout the expiratory cycle when it is applied together with mechanical ventilation (Chapter 105). CPAP and PEEP can result in recruitment of microatelectatic regions of the lung that are perfused but were not previously ventilated, thus contributing substantially to hypoxemia. CPAP and PEEP have the theoretical advantage of keeping some of these regions open during exhalation, thus preventing cyclic closure and reopening of lung units, which may result in alveolar wall stress and injury.

Supportive Measures

Every patient with acute respiratory failure is at risk for deep venous thrombosis, pulmonary thromboembolism, and gastric stress ulceration. Prophylactic anticoagulation is recommended in patients who are not at high risk for bleeding complications; sequential leg compression therapy may be preferred for high-risk patients (Chapter 81). Nutrition is important to maintain strength needed for weaning. In patients with ARDS, limited enteral feeding for up to 6 days is as good as full enteral feeding in terms of ventilator-free days, 60-day mortality, and infectious complications, and limited feedings induce less gastrointestinal intolerance. A3  The best means of preventing gastric stress ulceration is not known, but current evidence indicates that the use of an H2-receptor blocker is superior to the gastric administration of sucralfate on the basis of a large randomized, controlled trial that found a higher incidence of significant bleeding in patients receiving sucralfate than in those receiving ranitidine. Evidence also indicates that proton pump inhibitors may be useful in the acute care setting (Chapter 217). Current evidence supports maintaining the head of the bed at a 45-degree angle to reduce aspiration in critically ill patients. Attempts should be made to ensure a normal day-night sleep pattern, including minimizing activity and reducing direct lighting at night. The patient should change position frequently, including sitting in a chair and walking short distances if possible, even while receiving mechanical ventilatory support. Mobilization can enhance the removal of secretions, help maintain musculoskeletal function, reduce the risk of deep venous thrombosis, and provide psychological benefits.

CHAPTER 104  Acute Respiratory Failure  

  SPECIFIC ACUTE RESPIRATORY FAILURE SYNDROMES

Chronic Obstructive Pulmonary Disease

EPIDEMIOLOGY AND PATHOBIOLOGY

The epidemiology and pathobiology of COPD are discussed in Chapter 88.

CLINICAL MANIFESTATIONS

When patients with COPD develop acute respiratory failure, they commonly have a history of increasing dyspnea and sputum production. Acute respiratory failure may be manifested in more cryptic ways, however, such as changes in mental status, arrhythmias, or other cardiovascular abnormalities. Acute respiratory failure must be considered whenever patients with COPD have significant nonspecific clinical changes.

DIAGNOSIS

The diagnosis can be confirmed or excluded by arterial blood gas analysis. The pH is helpful in assessing whether the hypoventilation is partly or exclusively acute. The pH declines by approximately 0.08 for each rise of 10 mm Hg in the Paco2 in acute respiratory acidosis without renal compensation. By comparison, in chronic respiratory acidosis with normal renal compensation, the pH drops only about 0.03 for each rise of 10 mm Hg in the Paco2.

TREATMENT  General Care

As soon as acute respiratory failure is confirmed in a patient with COPD, attention must focus on detecting potential precipitating events (Table 104-4), including decreased ventilatory drive, commonly because of oversedation; decreased muscle strength or function, often related to electrolyte abnormalities, including hypophosphatemia and hypomagnesemia; decreased chest wall elasticity, possibly related to rib fracture, pleural effusion, ileus, or ascites; atelectasis, pneumonia, or pulmonary edema; increased airway resistance, caused by bronchospasm or increased secretions; or increased metabolic O2 requirements, such as may occur with systemic infection. Many of these abnormalities can impair the cough mechanism, diminish the clearance of airway secretions, and precipitate acute respiratory failure.

Infection

The most common specific precipitating event is airway infection, especially acute bronchitis. The role played by viral agents, Mycoplasma pneumoniae, chronic contaminants of the lower airway such as Haemophilus influenzae and Streptococcus pneumoniae, and other acute pathogens is difficult to determine on a clinical or even microbiologic basis. Acute exacerbations of COPD commonly result from new infections rather than from reemergence of an infection by preexisting colonization. Antibiotics modestly shorten the duration of the exacerbation, with no significant increase in toxicity, compared with placebo; the impact of antibiotics on the subsequent emergence of resistant organisms is not known. It is standard practice to use antibiotics to treat a patient with COPD who has an exacerbation severe enough to cause acute respiratory failure and who has evidence consistent with acute tracheobronchitis (Chapters 88 and 96). Pneumonia may account for 20% of cases of acute respiratory failure in patients with COPD. Compared with the physiologically normal population, patients with COPD who have community-acquired pneumonia are more likely to have gram-negative enteric bacteria or Legionella infections and are more likely to have antibioticresistant organisms.

Other Precipitating Causes

Other common precipitating causes of acute respiratory failure include heart failure and worsening of the underlying COPD, often related to noncompliance with medications. Less common and often difficult to diagnose in this setting is pulmonary thromboembolism.

Site of Care

Many patients with COPD and acute respiratory failure can be managed on a general medical hospital unit rather than in an intensive care unit if the precipitating cause of acute respiratory failure has been diagnosed and is potentially responsive to appropriate therapy, provided blood gas abnormalities respond to O2 therapy, the patient can cooperate with the treatment, and appropriate nursing and respiratory care is available (Chapter 88). An unstable patient who requires closer observation and monitoring should be admitted to an intensive care unit.

Mechanical Therapy

The decision to institute mechanical ventilation in patients with COPD and acute respiratory failure must be made on clinical grounds and is not dictated

661

TABLE 104-4  KEY PRINCIPLES IN THE MANAGEMENT OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE PATIENTS WITH ACUTE RESPIRATORY FAILURE 1. Monitor and treat life-threatening hypoxemia (these measures should be performed virtually simultaneously). a. Assess the patient clinically, and measure oxygenation by arterial blood gases and/or oximetry. (1) If the patient is hypoxemic, initiate supplemental oxygen therapy with nasal prongs (low flows [0.5-2. L/min] are usually sufficient) or by Venturi face mask (24 or 28% oxygen delivered). (2) If the patient needs ventilatory support, consider noninvasive ventilation. (3) Determine whether the patient needs to be intubated; this is almost always a clinical decision. Immediate action is required if the patient is comatose or severely obtunded. b. A reasonable goal in most patients is Pao2 of 55-60 mm Hg or Sao2 of 88-90%. c. After changes in Fio2, check blood gases and check regularly for signs of carbon dioxide retention. 2. Start to correct life-threatening acidosis. a. The most effective approach is to correct the underlying cause of acute respiratory failure (e.g., bronchospasm, infection, heart failure). b. Consider ventilatory support, based largely on clinical considerations. c. With severe acidosis, the use of bicarbonate can be considered, but it is often ineffective, and there is little evidence of a clinical benefit. 3. If ventilatory support is required, consider noninvasive mechanical ventilation. a. The patient must have intact upper airway reflexes and be alert, cooperative, and hemodynamically stable. b. Careful monitoring is required; if the patient does not tolerate the mask, becomes hemodynamically unstable, or has a deteriorating mental status, consider intubation. 4. Treat airway obstruction and the underlying disease process that triggered the episode of acute respiratory failure. a. Treat airway obstruction with pharmacologic agents: systemic corticosteroids and bronchodilators (ipratropium and/or β-adrenergic agents). b. Improve secretion clearance: encourage the patient to cough, administer chest physical therapy if cough is impaired and a trial appears effective. c. Treat the underlying disease process (e.g., antibiotics, diuretics). 5. Prevent complications of the disease process and minimize iatrogenic complications. a. Pulmonary thromboembolism prophylaxis: use subcutaneous heparin if no contraindications exist. b. Gastrointestinal complications: administer prophylaxis for gastrointestinal bleeding. c. Hemodynamics: if the patient is ventilated, monitor and minimize auto-PEEP. (1) Treat the underlying obstruction. (2) Minimize minute ventilation; use controlled hypoventilation. (3) Use small tidal volumes; increase the inspiratory flow rate to decrease the inspiratory time and lengthen the expiratory time. d. Cardiac arrhythmias: maintain oxygenation and normalize electrolytes. Fio2 = fraction of inspired oxygen; Pao2 = partial pressure of oxygen in arterial blood; PEEP = positive end-expiratory pressure; Sao2 = oxygen saturation.

by any particular arterial blood gas values. In general, if the patient is alert and is able to cooperate with treatment, mechanical ventilation often is not necessary. If ventilatory support is required (Chapter 105), the decision is whether to use noninvasive positive-pressure ventilation therapy (without endotracheal intubation) or endotracheal intubation with positive-pressure ventilation. A number of studies have demonstrated that noninvasive positive-pressure ventilation is preferred for patients with COPD and can decrease mortality if it is applied in appropriate patients with no factors that are likely to lead to complications. A4 

PROGNOSIS

Acute respiratory failure in patients with severe COPD is associated with an in-hospital mortality of 6 to 20%. The severity of the underlying disease and the severity of the acute precipitating illness are important determinants of hospital survival. Hospital mortality is higher if the respiratory failure is associated with a pH lower than 7.25 and if the patient requires invasive mechanical ventilation. However, the pH, the Paco2, and other clinical characteristics are not reliable in predicting a particular patient’s chances of survival.

Acute Lung Injury/Acute Respiratory Distress Syndrome

DEFINITION

ARDS is the abrupt onset of diffuse lung injury characterized by severe hypoxemia (shunting) and generalized pulmonary infiltrates on the chest

662

CHAPTER 104  Acute Respiratory Failure  

radiograph in the absence of left-sided cardiac failure.3,4 The term acute lung injury has been used to include “traditional” ARDS as well as less severe forms of lung injury. Acute lung injury and its more severe manifestation, ARDS, are diagnosed by bilateral pulmonary infiltrates compatible with pulmonary edema in the absence of clinical heart failure (usually determined by the lack of elevated left atrial pressures). A recent Berlin consensus conference recommended that the term acute lung injury should no longer be used, with ARDS diagnosed on the basis of a Pao2/Fio2 of less than 300 mm Hg. With this new definition, the severity of ARDS is defined on the basis of the Pao2 divided by the Fio2 (Pao2/Fio2, also called the P/F ratio) as mild ARDS (200 < P/F ≤ 300 mm Hg), moderate ARDS (100 < P/F ≤ 200 mm Hg), or severe ARDS (P/F ≤ 100 mm Hg).5

EPIDEMIOLOGY

ARDS is a clinical syndrome triggered by some other cause (Table 104-5), with an annual incidence of about 80 cases per 100,000 adult population. This underlying precipitating factor may affect and injure the lungs directly, such as in diffuse pneumonia or aspiration of gastric contents, or it may affect the lungs indirectly, as in severe sepsis from a nonpulmonary or a pulmonary source (Chapter 108) or severe nonthoracic trauma associated with shock (Chapter 111).6 Severe sepsis is the most common precipitating cause of ARDS worldwide. The organisms vary widely, ranging from gram-negative and gram-positive bacteria and viruses (e.g., H1N1 influenza in 2009) to leptospiral infections or malaria. It can be difficult to determine whether pneumonia is diffuse, with endobronchial spread involving most of the lungs, or whether localized pneumonia has precipitated a sepsis syndrome, with secondary injury to other parts of the lung.

PATHOBIOLOGY

Pathology

Despite the variety of underlying disease processes leading to ARDS, the response to these insults in the lung is monotonously characteristic, with similar clinical findings, physiologic changes, and morphologic abnormalities. The pathologic abnormalities in ARDS are nonspecific and are described as diffuse alveolar damage by pathologists. The initial process is inflammatory, with neutrophils usually predominating in the alveolar fluid. Hyaline membranes are present in some but not all patients,7 similar to those seen in premature infants with infant respiratory distress syndrome, presumably related to the presence of large-molecular-weight proteins that have leaked into the alveolar space. Alveolar flooding leads to impairment of surfactant, which is abnormal in quantity and quality. The result is microatelectasis, which may be associated with impaired immune function. Cytokines and other inflammatory mediators are usually markedly elevated, although with different patterns over time in the bronchoalveolar lavage fluid and the systemic blood. The resolution of ARDS depends in part on restoration of a functional alveolar epithelial barrier, capable of removing alveolar edema

TABLE 104-5  DISORDERS ASSOCIATED WITH THE ACUTE RESPIRATORY DISTRESS SYNDROME COMMON Sepsis (gram-positive or gram-negative bacterial, viral, fungal, or parasitic infection) Diffuse pneumonia (bacterial, viral, or fungal) Aspiration of gastric contents Trauma (usually severe) LESS COMMON Near-drowning (fresh or salt water) Drug overdose   Acetylsalicylic acid   Heroin and other narcotic drugs Massive blood transfusion (likely a marker of severe trauma, but also seen with severe gastrointestinal bleeding, especially in patients with severe liver disease) Leukoagglutination reactions Inhalation of smoke or corrosive gases (usually requires high concentrations) Pancreatitis Fat embolism UNCOMMON Miliary tuberculosis Paraquat poisoning Central nervous system injury or anoxia (neurogenic pulmonary edema) Cardiopulmonary bypass

fluid by sodium-dependent vectorial fluid transport.8 Lung repair is also disturbed; early evidence of profibrotic processes includes the appearance of breakdown products of procollagen in the bronchoalveolar lavage fluid, followed by fibrosis in some patients. Lung function improves over time in survivors of ARDS; however, the fibrosis is often reversible.

Pathophysiology

The physiologic abnormalities are dominated by severe hypoxemia with shunting, decreased lung compliance, decreased functional residual capacity, increased pulmonary dead space, and increased work of breathing. Initially, the Paco2 is low or normal, usually associated with increased minute ventilation. The initial abnormalities in oxygenation are thought to be related to alveolar flooding and collapse. As the disease progresses, especially in patients who require ventilatory support, fibroproliferation may develop; the lungs (including alveoli, blood vessels, and small airways) remodel and scar, with a loss of microvasculature. In some patients, these changes may lead to pulmonary hypertension and increased pulmonary dead space; marked elevations in minute ventilation are required to achieve a normal Paco2, even as oxygenation abnormalities are improving.

CLINICAL MANIFESTATIONS

In most cases of ARDS, the onset either coincides with or occurs within 72 hours of the onset of the underlying disease process; the mean time from onset of the underlying cause to onset of acute lung injury is 12 to 24 hours. The presenting picture is dominated by respiratory distress and the accompanying laboratory findings of severe hypoxemia and generalized infiltrates or opacities on the chest radiograph. Alternatively, it may be dominated by manifestations of the underlying disease process, such as severe sepsis with hypotension and other manifestations of systemic infection.

DIAGNOSIS

The key to diagnosis is to distinguish ARDS from cardiogenic pulmonary edema (Table 104-6). No specific biochemical test exists to define ARDS. Certain blood or bronchoalveolar lavage (Chapter 85) abnormalities are frequent but are not sufficiently specific to be useful clinically.

TREATMENT  Treatment of ARDS consists predominantly of respiratory support and treatment of the underlying disease (Fig. 104-4). Sepsis, which is a common predisposing condition for the development of ARDS, must be treated aggressively (Chapter 108). Current recommendations for lung-protective mechanical ventilation by endotracheal intubation (Table 104-7) emphasize lower tidal volumes based on the patient’s predicted body weight (Chapter 105). A5  A6  This approach also includes achieving a plateau airway pressure less than 30 cm H2O. PEEP is a mainstay in the ventilatory strategy for ARDS; although the method for determining the optimal level of PEEP has not been established, higher PEEP levels may have some benefit for patients with moderate to severe ARDS. A6-A8  PEEP may allow a lower Fio2 to provide adequate oxygenation, thereby reducing the risk of O2 toxicity. It also may prevent the cyclic collapse and reopening of lung units, a process that is thought to be a major cause of ventilator-induced lung injury, even when adequate oxygenation can be obtained at relatively low levels of Fio2.9 On the basis of one clinical trial, the early use of cisatracurium besylate (15 mg rapid infusion followed by 37.5 mg/hour for 48 hours), a neuromuscular blocker, can reduce ARDS mortality rates by about 25% in patients with moderately severe ARDS with a P/F below 150 mm Hg. A9  In patients with severe ARDS who do not respond to standard therapy but otherwise have a reasonable life expectancy and do not have multiorgan failure, extracorporeal membrane oxygenation is an acceptable albeit not fully proven rescue therapy.10, A10  Data also indicate that prone positioning reduces mortality in patients with moderate to severe ARDS (initial P/F < 150 mm Hg). A11  ,

PROGNOSIS

Case-fatality rates are 30 to 50% and are highly dependent on disease severity and the underlying predisposing condition. Based on the degree of hypoxemia (mild, 200 mm Hg < Pao2/Fio2 < 300 mm Hg; moderate, 100 mm Hg < Pao2/Fio2 < 200 mm Hg; and severe, Pao2/Fio2 < 100 mm Hg), inpatient mortality rates are about 27%, 32%, and 45%, respectively, not including patients with severe underlying conditions, such as end-stage cancer.5 Memory, verbal fluency, and executive function are impaired in about 13%, 16%, and 49% of long-term survivors after ARDS.11,12 Lower Pao2 during hospitalization is associated with cognitive and psychiatric impairment.

CHAPTER 104  Acute Respiratory Failure  

663

Acute Lung Injury/Acute Respiratory Distress Syndrome

Place on non-rebreathing mask with 100% O2 Attach pulse oximeter for SaO2 and measure ABG

Begin management of precipitating events or associated underlying diseases and MSOF Consider right heart catheterization if hypotension present and diagnosis uncertain

Patient alert and hemodynamically stable: RR < 35, PaCO2 < 35 mm Hg; SaO2 > 88% Yes

No Intubate: volume cycled ventilation: VT 6 mL/kg PBW; FIO2 1.0; PEEP 5 cm H2O assist control mode; conscious sedation and maintain comfort Code status discussed; NPO; H2 blocker; DVT prophylaxis; semi-recumbent (45°) position

Adjust FIO2 to yield SaO2 88-95% Consider NIPPV to relieve dyspnea

SaO2 < 88%

SaO2 > 95% Increase PEEP in 3-5 cm H2O increments (or consider ARDSNet PEEP/FIO2 ladder)

FIO2 < 0.6

Monitor ABG, blood pressure, urine output, capillary refill time, and (if available) cardiac index Inadequate perfusion Adequate perfusion Give volume

Measure plateau pressure

Consider right heart catheterization ≤ 30 cm H2O

Reduce FIO2 until SaO2 < 96% FIO2 ≤ 0.6

Maintain urine output ≅ fluid intake 24-48 h > 30 cm H2O

Continue to increase PEEP by 3-5 cm H2O increments; repeat above assessment until SaO2 > 88% and FIO2 < 0.6

Decrease VT by 1 mL/kg PBW decrements (to minimum of 4) until Pplat < 30 cm H2O; allow PaCO2 to rise slowly Consider other modes of ventilatory support Note: If chest wall compliance is markedly decreased (e.g., massive ascites), then it may not be necessary to decrease VT Repeat ABG

SaO2 > 95% or PaO2 > 80 mm Hg

SaO2 88-95% or PaO2 55-80 mm Hg

Decrease FIO2 by 0.1 decrements to 0.4 and/or decrease PEEP by 3-5 cm H2O to 8 cm H2O (or consider ARDSNet PEEP/FIO2 ladder) Wean from ventilator as tolerated

Maintain ventilator settings Continue pulse oximetry; repeat ABG in 4-8 h or as clinically indicated

SaO2 < 88% or PaO2 < 55 mm Hg Increase PEEP by 3-5 cm H2O increments to maximum of 25 cm H2O and/or increase FIO2 by 0.1 increments to 1.0 Repeat assessment of plateau pressure and ABG SaO2 remains < 88% PaO2 remains < 55 mm Hg Consider prone position; increase sedation and/or paralysis Accept PaCO2 rise; accept pH decrease to 7.15 or lower; accept SaO2 ≈ 85%

FIGURE 104-4.  Algorithm for the initial management of acute respiratory distress syndrome. ABG = arterial blood gas analysis; CO2 = carbon dioxide; DVT = deep venous thrombosis; FIO2 = inspired oxygen concentration; MSOF = multisystem organ failure; NIPPV = noninvasive intermittent positive-pressure ventilation; O2 = oxygen; PaCO2 = arterial partial pressure of carbon dioxide; PaO2 = arterial partial pressure of oxygen; PBW = predicted body weight; PEEP = positive end-expiratory pressure; Pplat = plateau pressure; RR = respiratory rate; SaO2 = arterial oxygen saturation; VT = tidal volume.

TABLE 104-6  FEATURES ASSOCIATED WITH NONCARDIOGENIC AND CARDIOGENIC PULMONARY EDEMA* NONCARDIOGENIC EDEMA (ARDS)

CARDIOGENIC EDEMA/VOLUME OVERLOAD

PRIOR HISTORY No history of heart disease

Prior history of heart disease

Appropriate fluid balance (difficult to assess after resuscitation from shock or trauma)

Hypertension, chest pain, new-onset palpitations; positive fluid balance

PHYSICAL EXAMINATION Flat neck veins

Elevated neck veins

Hyperdynamic pulses

Left ventricular enlargement, lift, heave, dyskinesis

Physiologic gallop

S3 and S4; murmurs

Absence of edema

Edema: flank, presacral, legs

ELECTROCARDIOGRAM Sinus tachycardia, nonspecific ST-T wave changes

Evidence of prior or ongoing ischemia, supraventricular tachycardia, left ventricular hypertrophy

CHEST RADIOGRAPH Normal heart size

Cardiomegaly

Peripheral distribution of infiltrates

Central or basilar infiltrates; peribronchial and vascular congestion

Air bronchograms common (80%)

Septal lines (Kerley lines), air bronchograms (25%), pleural effusion

HEMODYNAMIC MEASUREMENTS Pulmonary artery wedge pressure 3.5 L/ min/m2

Pulmonary capillary wedge pressure >18 mm Hg, cardiac index 3.5 L/min/m2 with volume overload

*These features are neither highly sensitive nor specific. Although the findings are more commonly associated with the type of pulmonary edema as listed, they do not have high positive or negative predictive value. ARDS = acute respiratory distress syndrome.

patient with neuromuscular disease has Guillain-Barré syndrome (Chapter 420). The treatment for both types of patients is supportive. In the case of a patient with a sedative overdose, the threshold for intubation with mechanical ventilatory support should be low because this temporary condition is quickly reversible when the responsible drug is eliminated. Such a patient may require intubation for airway protection against aspiration of gastric contents. Patients with Guillain-Barré syndrome or other forms of progressive neuromuscular disease should be monitored with serial measurements of vital capacity. In general, when the vital capacity decreases to less than 10 to 15 mL/kg body weight, intubation and mechanical ventilatory support should be considered without regard to the patient’s Paco2.

Grade A References A1. Wheeler AP, Bernard GR, Thompson BT, et al. Pulmonary-artery versus central venous catheter to guide treatment of acute lung injury. N Engl J Med. 2006;354:2213-2224. A2. Wiedemann HP, Wheeler AP, Bernard GR, et al. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006;354:2564-2575. A3. Rice TW, Wheeler AP, Thompson BT, et al. Initial trophic vs full enteral feeding in patients with acute lung injury: the EDEN randomized trial. JAMA. 2012;307:795-803. A4. Williams JW, Cox CE, Hargett CW, et al. Noninvasive Positive-Pressure Ventilation (NPPV) for Acute Respiratory Failure. AHRQ Comparative Effectiveness Reviews, No. 68. Report No. 12-EHC089-EF. Rockville, MD: Agency for Healthcare Research and Quality; 2012. A5. Putensen C, Theuerkauf N, Zinserling J, et al. Meta-analysis: ventilation strategies and outcomes of the acute respiratory distress syndrome and acute lung injury. Ann Intern Med. 2009;151:566-576. A6. Briel M, Meade M, Mercat A, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and metaanalysis. JAMA. 2010;303:865-873. A7. Meade MO, Cook DJ, Guyatt GH, et al. Ventilation strategy using low tidal volumes, recruitment maneuvers, and high positive end-expiratory pressure for acute lung injury and acute respiratory distress syndrome: a randomized controlled trial. JAMA. 2008;299:637-645. A8. Mercat A, Richard JC, Vielle B, et al. Positive end-expiratory pressure setting in adults with acute lung injury and acute respiratory distress syndrome: a randomized controlled trial. JAMA. 2008;299:646-655. A9. Papazian L, Forel JM, Gacouin A, et al. Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med. 2010;363:1107-1116. A10. Peek GJ, Mugford M, Tiruvoipati R, et al. Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet. 2009;374:1351-1363. A11. Hu SL, He HL, Pan C, et al. The effect of prone positioning on mortality in patients with acute respiratory distress syndrome: a meta-analysis of randomized controlled trials. Crit Care. 2014;18:R109.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

TABLE 104-7  ARDS NETWORK VENTILATORY MANAGEMENT PROTOCOL FOR TIDAL VOLUME AND PLATEAU AIRWAY PRESSURE Calculate PBW:   Male PBW: 50 + 2.3 (height in inches – 60) or 50 + 0.91 (height in centimeters – 152.4)   Female PBW: 45.5 + 2.3 (height in inches – 60) or 45.5 + 0.91 (height in centimeters – 152.4) Select assist control mode Set initial Vt at 8 mL/kg PBW Reduce Vt by 1 mL/kg at intervals < 2 hr until Vt = 6 mL/kg PBW Set initial RR to approximate baseline minute ventilation (maximum RR = 35/min) Set inspiratory flow rate higher than patient’s demand (usually > 80 L/min) Adjust Vt and RR further to achieve Pplat and pH goals   If Pplat > 30 cm H2O: decrease Vt by 1 mL/kg PBW (minimum = 4 mL/kg PBW)   If pH ≤ 7.30, increase RR (maximum = 35)   If pH < 7.15, increase RR to 35; consider sodium bicarbonate administration or increase Vt ARDS = acute respiratory distress syndrome; PBW = predicted body weight; Pplat = plateau pressure (airway pressure at the end of delivery of a tidal volume breath during a condition of no airflow); RR = respiratory rate; Vt = tidal volume. See the ARDSNet website (http://www.ardsnet.org) for further details about the protocol, including the approach for setting positive end-expiratory pressure and fraction of inspired oxygen.

Acute Respiratory Failure without Lung Disease Acute respiratory failure without pulmonary abnormalities (see Table 104-2) develops in patients with depressed ventilatory drive secondary to central nervous system dysfunction and in patients with severe neuromuscular disease. The prototypical patient with suppressed ventilatory drive has taken an overdose of a sedative or tranquilizing medication (Chapter 110). The prototypical

CHAPTER 104  Acute Respiratory Failure  

GENERAL REFERENCES 1. Grocott MP, Martin DS, Levett DZ, et al. Arterial blood gases and oxygen content in climbers on Mount Everest. N Engl J Med. 2009;360:140-149. 2. Wilson JG, Matthay MA. Mechanical ventilation in acute hypoxemic respiratory failure: a review of new strategies for the practicing hospitalist. J Hosp Med. 2014;9:469-475. 3. Del Sorbo L, slu*tsky AS. Acute respiratory distress syndrome and multiple organ failure. Curr Opin Crit Care. 2011;17:1-6. 4. Matthay MA, Ware L, Zimmerman GA. The acute respiratory distress syndrome. J Clin Invest. 2012;122:2731-2740. 5. Force ADT, Ranieri VM, Rubenfeld GD, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307:2526-2533. 6. Villar J, Sulemanji D, Kacmarek RM. The acute respiratory distress syndrome: incidence and mortality, has it changed? Curr Opin Crit Care. 2014;20:3-9.

664.e1

7. Thille AW, Esteban A, Fernandez-Segoviano P, et al. Comparison of the Berlin definition for acute respiratory distress syndrome with autopsy. Am J Respir Crit Care Med. 2013;187:761-767. 8. Matthay MA. Resolution of pulmonary edema. Thirty years of progress. Am J Respir Crit Care Med. 2014;189:1301-1308. 9. slu*tsky AS, Ranieri VM. Ventilator-induced lung injury. N Engl J Med. 2013;369:2126-2136. 10. Ventetuolo CE, Muratore CS. Extracorporeal life support in critically ill adults. Am J Respir Crit Care Med. 2014;190:497-508. 11. Herridge MS, Tansey CM, Matte A, et al. Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med. 2011;364:1293-1304. 12. Mikkelsen ME, Christie JD, Lanken PN, et al. The adult respiratory distress syndrome cognitive outcomes study: long-term neuropsychological function in survivors of acute lung injury. Am J Respir Crit Care Med. 2012;185:1307-1315.

664.e2

CHAPTER 104  Acute Respiratory Failure  

REVIEW QUESTIONS 1. Which of the following mechanisms can lead to a reduction in the arterial oxygen tension (Pao2)? A. Ventilation-perfusion mismatch B. Intrapulmonary right-to-left shunt C. Decreased inspired partial pressure of oxygen D. Alveolar hypoventilation E. All of the above Answer: E  It is important for clinicians to understand the physiologic mechanisms of arterial hypoxemia. Once the physiologic basis of hypoxemia is known, the differential diagnosis as to specific cause (i.e., the type of disease at hand leading to this physiologic abnormality) can be made, and a specific treatment often can be initiated. 2. If a patient has the clinical presentation of bilateral pneumonia with no evidence of cardiac disease and has a chest radiograph that shows bilateral pulmonary infiltrates, which of the following Pao2/Fio2 ratios (where Pao2 is in mm Hg) is consistent with the diagnosis of acute respiratory distress syndrome (ARDS)? A. Pao2 = 80; Fio2 = 0.25 B. Pao2 = 200; Fio2 = 0.3 C. Pao2 = 200; Fio2 = 0.8 D. Pao2 = 190; Fio2 = 0.3 E. Pao2 = 490; Fio2 = 0.5 Answer: C  It is the only one with a Pao2/Fio2 ratio of less than 300 mm Hg. It is important to be able to calculate the Pao2/Fio2 ratio to make the diagnosis of ARDS in a patient who has bilateral pulmonary infiltrates consistent with pulmonary edema but no evidence of cardiogenic pulmonary edema as the primary cause. 3. Which of the following treatments has been shown to reduce mortality in patients with ARDS on the basis of multicenter randomized trials. A. High-frequency ventilation B. Lung-protective ventilation with a tidal volume of 6 mL/kg predicted body weight and a plateau airway pressure of less than 30 cm H2O C. Intermittent mandatory ventilation with weaning to pressure support ventilation D. Pressure-control ventilation E. Liquid ventilation with perfluorocarbons Answer: B  on the basis of a National Heart, Lung, and Blood Institute– sponsored ARDS Network clinical trial in 2000. All intensivists need to know that lung-protective ventilation with this strategy markedly reduces mortality in patients with ARDS.

4. Why do most patients with an exacerbation of chronic obstructive pulmonary disease (COPD) require relatively low-flow nasal oxygen to achieve adequate oxygenation? A. Because intrapulmonary right-to-left shunt is an important cause of the hypoxemia B. Because there are many lung units with low ventilation-perfusion ratios, and these units are the primary cause of arterial hypoxemia in these patients C. Because diffusion impairment is an important cause of hypoxemia and COPD D. Because the inspired oxygen tension is low in these patients E. Because such patients have decreased cardiac output and therefore low perfusion Answer: B  because ventilation-perfusion abnormalities predominate. In most patients with an exacerbation of COPD, most gas exchange takes place in units with low ventilation to perfusion values. In most patients, the hypoxemia can be corrected, at least in part, with low-flow oxygen. Clinicians should understand the physiologic basis of treating an exacerbation of COPD, which normally requires low-flow oxygen unless another process, such as pneumonia, pulmonary edema, or atelectasis, could be causing an intrapulmonary shunt. 5. Which of the following clinical conditions have been associated with the development of ARDS? A. Nonpulmonary sepsis B. Pneumonia C. Drug overdose D. Massive blood transfusions E. All of the above Answer: E  Clinicians need to know clinical disorders that may be associated with ARDS.

664

CHAPTER 105  Mechanical Ventilation  

105  MECHANICAL VENTILATION ARTHUR S. slu*tSKY Mechanical ventilation is a life-sustaining therapy in which a ventilator provides partial or full support for patients with respiratory failure (Chapter 104). In setting the ventilator, the clinician can use a variety of modes of ventilation and can also alter the inspired oxygen tension, the pressure at the airway opening at the end of a breath, and other facets of the volume or pressure time pattern imposed on the patient. The main goals of ventilatory support are to maintain adequate gas exchange, to rest the respiratory muscles, and to decrease the oxygen cost of breathing. Modern ventilation strategies focus on minimizing its iatrogenic consequences, such as iatrogenic hyperinflation (from endogenously derived positive pressure at the end of a breath, i.e., auto-PEEP) and ventilatorinduced lung injury. In some patients, the physician should be willing to accept arterial blood gases that are not in the normal range to avoid these complications by using lower levels of minute ventilation or relatively smaller tidal volumes.

CHAPTER 105  Mechanical Ventilation  

TYPES OF MECHANICAL VENTILATORS

Negative-Pressure Ventilators

Delivery of gas to the lungs requires a hydrostatic pressure gradient between the airway opening and the alveoli. During spontaneous breathing, this pressure gradient is generated by developing negative pleural pressure due to respiratory muscle contraction. Some ventilators operate by generating negative pressure around the chest wall (e.g., cuirass) or around the entire body below the neck (e.g., iron lung). The cuirass has the major advantage of minimizing detrimental hemodynamic consequences, but it is difficult to apply because the device must have an adequate seal to the body so that the negative pressure is not dissipated to the room—a task that is not always easy to accomplish in a way that is comfortable for the patient. The iron lung makes nursing care difficult because it encircles the patient’s entire body. Although iron lungs were widely used during the polio epidemic of the mid-1950s, they are rarely used today.

Positive-Pressure Ventilators

The most widely used approach to mechanical ventilation is to deliver gas to the lung with positive-pressure ventilation (PPV) applied through an endotracheal tube, a tracheostomy, or a tight-fitting mask. The approach with a mask is considered noninvasive ventilation (NIV) and is considered separately. The most basic mode of PPV is controlled ventilation, in which a preset tidal volume at a predetermined rate is delivered, regardless of the patient’s requirements or efforts. This form of ventilation is usually used in patients who cannot initiate spontaneous breaths (e.g., heavily sedated or paralyzed patients) or in those who need full ventilatory support because of extremely severe pulmonary or cardiovascular disease (e.g., severe shock). This ventilator mode may be beneficial when it is used for relatively short periods (~48 hours) in patients with the acute respiratory distress syndrome (ARDS) early in their clinical course, when they may be treated with neuromuscular blocking agents (see later). A1  However, a paralyzed patient without any ability to make breathing efforts is at risk of asphyxia in the event of an inadvertent disconnection from the ventilator. If the patient is not making any respiratory efforts, controlled ventilation can rapidly lead to respiratory muscle atrophy. For these reasons, clinicians usually try to limit the time that a patient is paralyzed and receiving controlled ventilation. Assisted ventilation is the term used when the patient’s spontaneous ventilatory efforts trigger the ventilator to deliver breaths, rather than having the breaths delivered by the ventilator at a fixed rate without regard to the patient’s efforts. Mechanical ventilation can be applied by either volume-controlled or pressure-controlled modes. In volume-controlled ventilation, the desired tidal volume and respiratory rate are set by the user, and the airway pressure is the dependent variable. The airway pressure profile depends on the mechanical properties of the patient’s respiratory system and on the ventilator’s flow settings. In pressure-controlled ventilation, the pressure imposed at the airway opening along with the respiratory rate is set by the user, and the tidal volume becomes the dependent variable.

advantage is that the delivered tidal volume is maintained even if lung mechanics change, thereby ensuring a more constant partial pressure of arterial carbon dioxide (Paco2). The potential disadvantage is that if lung mechanics deteriorate, higher pressures may be required to achieve the tidal volume goal, and regions of overinflation may result in regional lung injury. Although controlled ventilation as described earlier can be either volume limited (preset tidal volume) or pressure limited (preset peak airway pressure), clinicians usually use the term controlled mechanical ventilation to refer to volume-limited ventilation with a set ventilatory rate. In volume-controlled ventilation, an upper limit to applied airway pressure is commonly used for safety reasons. The most common form of volume-controlled ventilation is one in which the patient assists the ventilator, thus triggering at least some of the breaths. The term assisted mechanical ventilation can refer to either volumelimited ventilation or pressure-limited ventilation when the patient triggers some or all of the breaths, but in either case the ventilator should be set to deliver breaths if apnea occurs. This mode is also referred to as assist/control (A/C).

Intermittent Mandatory Ventilation

Intermittent mandatory ventilation (IMV) refers to a mode in which the patient is allowed to breathe spontaneously through an endotracheal tube or tracheostomy but also receives some preset (and thus mandatory) volumelimited breaths from the ventilator. In current ventilators, the mandatory breaths are triggered by the patient and are synchronized (synchronized IMV); however, if the patient ceases spontaneous ventilatory efforts, breaths at the rate set on the ventilator will still be delivered. Synchronized IMV is a form of partial ventilatory support because some breaths are spontaneous, in contrast to full ventilatory support, in which all breaths are delivered by the ventilator. This mode allows the patient to do a variable amount of the respiratory work but with the security of a set minimal backup rate should spontaneous ventilatory efforts stop.

Pressure-Controlled Ventilation

Pressure-controlled ventilation is a type of ventilation in which the ventilator delivers pressure-limited breaths to the patient; delivered volume becomes a dependent variable. The initiation of each breath may be triggered by the patient (assisted breaths) or may be initiated by the ventilator (controlled breaths). In the assist mode, a backup control rate protects any patients who cease to make inspiratory efforts on their own. The delivered tidal volume depends on the preset pressure, the ventilatory rate, the inspiratory-toexpiratory ratio, and the patient’s respiratory mechanics (resistance, compliance, and auto-PEEP). At a fixed preset pressure and inspiratory-to-expiratory ratio, tidal volume decreases as respiratory frequency increases. In patients with COPD, the tidal volume at low frequencies is relatively high but decreases substantially as the respiratory rate is increased; whereas in patients with stiff respiratory systems (e.g., ARDS), the tidal volume does not change much with respiratory frequency because the lung fills with gas quickly.

Positive End-Expiratory Pressure

Pressure-Support Ventilation

Volume-Controlled Ventilation

High-Frequency Ventilation

A key characteristic that can be combined with most ventilatory modes is the level of the end-expiratory pressure. Positive end-expiratory pressure (PEEP) is used in patients with diffuse pulmonary diseases (e.g., pulmonary edema or ARDS) to recruit collapsed alveolar regions and to maintain them in a recruited state, to reopen collapsed airways, to redistribute fluid in the lung, to increase functional residual capacity, and to redistribute ventilation to dependent regions. All these changes can improve the matching of ventilation to perfusion, thereby leading to improved oxygenation and allowing the fractional inspiratory concentration of oxygen (Fio2) to be reduced. PEEP does not usually improve alveolar ventilation and, in fact, may increase dead space by overdistending alveoli, with a concomitant decrease in alveolar capillary blood flow in certain regions of the lung. PEEP can also be administered to spontaneously breathing subjects by a technique termed continuous positive airway pressure (CPAP). In patients with exacerbations of chronic obstructive pulmonary disease (COPD), PEEP and CPAP can overcome some of the mechanical consequences of auto-PEEP (see later) to minimize the work of breathing, provided the magnitude of the PEEP is low enough that it does not cause additional hyperinflation. Volume-controlled ventilation (or volume-limited ventilation) refers to mechanical ventilation in which the tidal volume is preset. The major

665

Pressure-support ventilation is a pressure-limited, patient-triggered ventilatory mode. Once the patient triggers the ventilator by creating either a small negative pressure or a low inspiratory flow at the airway opening, the ventilator switches to inspiratory mode and provides the airflow needed to maintain a preset level of pressure. In contrast to pressure-controlled ventilation, inspiration terminates when the inspiratory airflow decreases to a threshold level (the specific algorithm varies from ventilator to ventilator). This mode provides flexibility for the patient with respect to tidal volume, inspiratory flow, and ratio of time allowed for inspiration compared with expiration. Tidal volume depends on patient-related factors (effort), respiratory system mechanics, and level of pressure set for support. During pressure-support ventilation, the size of each breath is determined partially by the patient’s muscle effort and partially by the ventilator. This mode can compensate for the added work of breathing imposed by the resistance of the endotracheal tube. Pressure-support ventilation has been used to wean patients from ventilatory support because it provides a simple way to reduce the magnitude of mechanical support while the patient assumes a larger fraction of the ventilatory work. High-frequency ventilation refers to modes that have the common feature of providing ventilation at frequencies that are substantially greater than those

666

CHAPTER 105  Mechanical Ventilation  

used during normal breathing. During high-frequency ventilation, tidal volumes may be less than the dead space, so adequate gas transport takes place by various convective and diffusive mechanisms. Interest in these modes of ventilation has waned because it appears to be no better or even worse than conventional ventilation for adults with ARDS. A2  A3  ,

Proportional Assist Ventilation and Neurally Adjusted Ventilatory Assist

One of the difficulties in providing assisted ventilation is ensuring that there is adequate synchrony between the patient’s respiratory drive and the delivery of the ventilator’s breaths. This issue is a particular problem for patients with severe obstructive airways disease, especially if they have significant auto-PEEP. Two newer modes of ventilation, proportional assist ventilation (PAV) and neurally adjusted ventilatory assist (NAVA), have been developed and implemented on some ventilators in part to address this concern. Both these modes deliver ventilation in proportion to the instantaneous effort of the patient, but the underlying principles are different. Although both of these modes improve patient-ventilator synchrony, data are insufficient to know whether either will improve clinically important outcomes. PAV is based on the mathematical relationships between airway pressure and airflow; these state that the pressure applied by the respiratory muscles is used to overcome the elastic losses (i.e., compliance) and the resistive losses of the respiratory system. With PAV, the pressure that is applied during inspiration varies on the basis of the patient’s inspiratory effort and respiratory system mechanics. This form of ventilation is not in widespread use, in part because of its complexity and the need to estimate the patient’s compliance and resistance on a regular basis. This latter issue has been addressed in new versions of the technique in which measurements of compliance and resistance are automatically measured repeatedly. NAVA makes use of the electrical activity of the diaphragm (Edi) as measured by an array of electrodes attached to a nasogastric tube inserted into the esophagus. Pressure is then delivered by the ventilator in direct proportion to the (virtually) instantaneous Edi. Because the initiation and delivery of the breath by the ventilator are not dependent on measurement of pressures in the lung, patient-ventilator synchrony is improved in patients with auto-PEEP. Once the array of electrodes has been inserted, the mode is relatively easy to use; the only parameter to set is the proportionality factor linking the Edi and the pressure delivered by the ventilator.

Noninvasive Positive-Pressure Ventilation

PPV can be provided through a mask rather than through an endotracheal tube. This method, which has been termed noninvasive because the patient is not intubated, is conceptually simple but requires appropriate implementation and monitoring for its successful application.1 Of particular importance are patient selection and appropriate training of hospital personnel. Patients must be alert, cooperative, and hemodynamically stable. Patients must also have intact upper airway reflexes to prevent aspiration of material from the upper airway into the lung, and they must not have any facial trauma that would preclude the use of a mask. Once patients are started on NIV, they should be carefully monitored, and NIV should be discontinued if the patient’s clinical condition deteriorates, if the patient develops cardiovascular instability, or if it appears that the patient is likely to aspirate. NIV can also be delivered through a “helmet” that avoids some of the problems associated with the use of face masks. NIV has potential advantages compared with invasive ventilation. It is relatively easy to apply and can be used for short intervals because it can be started and stopped easily. The major advantages are that it avoids the complications associated with intubation, it is usually more comfortable for the patient, and it reduces the need for sedation. Patients receiving NIV are able to communicate verbally with medical staff and family members, are probably able to sleep better, and are able to eat if they are sufficiently stable to remove the mask for short periods. However, NIV has several disadvantages. Implementation of NIV takes more time from caregivers at the bedside initially, and the time course of correction of blood gases is slower than usually occurs in patients who are intubated and ventilated. Gastric distention is an unusual occurrence; medical staff should be aware of this complication and should watch for signs of abdominal distention. Data strongly support the use of NIV for patients with COPD (see later), A4  and it is preferred in cardiogenic pulmonary edema, A5  but whether it provides better outcomes in other forms of respiratory failure is uncertain.

COMPLICATIONS OF MECHANICAL VENTILATION

Intubation

Endotracheal intubation can be used to secure a patient’s airway, to act as a conduit to deliver gas from the ventilator to the patient, to prevent aspiration, and to help with pulmonary toilet when secretions are increased. However, intubation can be associated with complications including the risk of aspiration during insertion of the endotracheal tube, difficulty in swallowing and communicating, disruption of normal host defense mechanisms, and upper airway trauma. Pressure from the cuff of the tube that provides a pneumatic seal between the tube and trachea can lead to regions of tracheal ischemia and may eventually cause tracheal stenosis. The endotracheal tube increases airway resistance because its diameter is smaller than the airway into which it is inserted. The magnitude of the increase depends on the length, diameter, and shape of the tube as well as on the buildup of secretions and mucus that narrow the tube’s diameter. Furthermore, the upper airway is normally an effective means of heating and humidifying inspiratory gases. This natural system is bypassed by an endotracheal tube; inadequately humidified inspiratory gases can reduce mucociliary clearance and can lead to inspissation of tracheal secretions. Intubation affects a number of factors that increase the likelihood of nosocomial pneumonia (Chapters 97 and 282). Normally, cough involves an increase in airway pressure as respiratory muscles are contracted against a closed glottis. When the glottis opens, expiratory flow sharply increases, resulting in dynamic compression of major airways. The presence of an endotracheal tube limits the buildup of airway pressure and alters the dynamics of expiratory flow, thereby greatly impairing the efficacy of the patient’s cough. A cuffed endotracheal tube helps prevent gross aspiration, but pharyngeal secretions that pool at the top of the cuff often seep into the lungs. Endotracheal tubes also can often become colonized with the microorganisms that cause ventilator-associated pneumonia (Chapter 97). Silver-coated tubes can reduce this risk but are considerably more expensive than conventional endotracheal tubes and are unlikely to be used routinely for initial intubation. Endotracheal tubes with a port that allows suctioning of secretions above the cuff may also reduce the incidence of ventilator-associated pneumonia, although results of studies have been mixed. In addition, endotracheal intubation is often not well tolerated in awake patients, and there is always the danger that the tube will inadvertently be dislodged—a complication that can have tragic consequences. For these reasons and to improve oral care and feeding, a tracheostomy can be performed. However, tracheostomy is associated with its own set of complications, and performing a tracheostomy in the first week is no better than waiting until the patient has been ventilated for about 10 days, A6  A7  in part because clinicians are not accurate in predicting which patients will require prolonged ventilatory support. Early tracheostomy should be avoided in patients in whom uncertainty exists as to how long invasive ventilation will be needed. ,

Hemodynamic Compromise

The major mechanical determinants of cardiovascular hemodynamics during mechanical ventilation are intrathoracic pressure, changes in lung volume, and the patient’s circulatory volume status. An increase in lung volume can cause a beneficial decrease in pulmonary vascular resistance, if lung units that had been closed are opened as a result of mechanical ventilation, or it can lead to a detrimental increase in pulmonary vascular resistance related to overdistention of the lung with concomitant compression and lengthening of alveolar vessels. PPV can affect cardiovascular hemodynamics through its effect on pleural pressure, an effect that is directly related to changes in lung volume and not necessarily directly reflected in measurements of airway pressure; the relation between alveolar pressure and lung volume depends on respiratory system mechanics. For example, in a patient with stiff lungs (e.g., ARDS), a given increase in airway pressure will lead to much less of an increase in lung volume than in a patient with COPD, so the increase in pleural pressure will be much less in the patient with ARDS. As a result, patients with ARDS tolerate relatively high PEEP levels, whereas similarly high levels in patients with normal lungs (e.g., in a drug overdose) or in patients with COPD would markedly reduce cardiac output. At very high lung volumes, a direct effect of the pressure of the lung on the heart can increase pericardial pressure and can thereby decrease cardiac filling.

CHAPTER 105  Mechanical Ventilation  

Auto-PEEP and Dynamic Hyperinflation

A key factor that affects cardiovascular hemodynamics and other physiologic variables during mechanical ventilation is the development of auto-PEEP, which is defined as the difference between alveolar pressure and airway pressure at end expiration. Auto-PEEP is associated with dynamic hyperinflation, which is an increase in the end-expiratory lung volume above the value that would be obtained if there was complete exhalation to the static functional residual capacity. This phenomenon occurs whenever there is insufficient time for a complete exhalation to occur; the respiratory system is thus prevented from reaching its static end-expiratory volume. The major determinants of auto-PEEP and hence dynamic hyperinflation are increased expiratory airway resistance, high minute ventilation, increased respiratory system compliance, and decreased expiratory time. Auto-PEEP may not be detected by routine measurements of pressure at the airway opening because most of the pressure drop occurs across the airways. Moreover, measurements of auto-PEEP are difficult to make in spontaneously breathing patients. When patients are not making spontaneous breathing efforts, auto-PEEP can be assessed as the difference in pressure between the set PEEP and the pressure obtained when the airway opening is occluded at the end of expiration (Fig. 105-1). It can also be assessed by the change in plateau pressure after a prolonged pause during volume cycle ventilation. If it is considered safe for the patient, a rapid estimate of the effect of auto-PEEP on cardiovascular hemodynamics can be obtained by transiently disconnecting the ventilator and allowing the auto-PEEP to approach zero during a long expiration. If the auto-PEEP is less than 5 cm H2O, it is unlikely to cause clinically important changes in the measured intravascular pressures. If auto-PEEP is not considered in the interpretation of respiratory mechanics, measurements of respiratory system compliance will be falsely low. Dynamic hyperinflation can be measured as the volume of gas that is released when the expiratory time of a given breath is lengthened by 20 to 30 seconds. The techniques for measuring auto-PEEP are based on the assumptions that no respiratory efforts are made and that the alveoli communicate with the airway opening, thereby allowing equilibration of pressures or exhalation of trapped gas. However, this assumption is not necessarily correct in patients with severe airways obstruction (e.g., status asthmaticus) because some airways may be completely closed. Endotracheal Alveoli tube

Expiratory port open to atmosphere

No flow

Ventilator

Central airway

A

Manometer 0

Airway obstruction

B

0 No flow

15

C

Auto-PEEP should be suspected whenever flow at end expiration is detectable or when a patient fails to trigger the ventilator consistently with inspiratory efforts. This failure to trigger the ventilator occurs because the patient must generate sufficient pressure to overcome the level of auto-PEEP before a negative deflection of pressure or generation of inspiratory flow (either of which may be used by the ventilator to detect the onset of inspiration) is sensed at the airway opening. Auto-PEEP and the attendant dynamic hyperinflation have numerous detrimental consequences. In a patient who is not breathing spontaneously, dynamic hyperinflation increases pleural pressure and right atrial pressure, thereby leading to a decrease in the driving pressure for venous return, with a concomitant decrease in cardiac output. This effect can be magnified in patients with airway obstruction immediately after intubation and initiation of mechanical ventilation because compensatory mechanisms to enhance venous return are impaired by pharmacologic agents that are often used to prepare the patient for endotracheal tube insertion and that also reduce venous and arterial tone. In such patients, auto-PEEP can also lead to gross misinterpretation of vascular pressures. For example, the absolute value of capillary wedge pressure will be directly affected by the increase in intrathoracic pressure during auto-PEEP. The clinician may interpret this high (absolute) capillary wedge pressure as indicating adequate ventricular filling when, in fact, transmural capillary wedge pressure is low because intrathoracic pressure is also high. This misinterpretation, coupled with the decreased cardiac output related to the high intrathoracic pressure, may suggest the diagnosis of cardiogenic shock rather than the correct diagnosis of auto-PEEP. In a spontaneously breathing patient, dynamic hyperinflation can markedly increase the oxygen cost of breathing for two reasons. First, because the respiratory system is stiffer at higher lung volumes, more energy is required to complete each ventilatory cycle. Second, to initiate flow into the lung, the patient must generate a pressure in the alveolar zone that is lower than atmospheric pressure. However, if dynamic hyperinflation is present, the patient first has to generate an inspiratory effort sufficient to overcome the (positive) end-expiratory alveolar pressure before he or she begins to lower alveolar pressure to less than atmospheric pressure to initiate airflow. The increase in lung volume associated with dynamic hyperinflation also has an impact on the effectiveness of the ventilatory muscles; at high lung volumes, the diaphragm is relatively flat, so it is at a mechanical disadvantage in producing changes in pleural pressure. Auto-PEEP is more likely to occur in patients with airway obstruction. Avoidance of high levels of auto-PEEP by approaches such as controlled hypoventilation—during which minute ventilation is minimized, with the attendant hypercapnia—is a fundamental approach to consider in ventilating patients who have severe airway obstruction. Treatment of the detrimental hemodynamic consequences of auto-PEEP include infusion of fluids and, most important, decreasing the level of auto-PEEP, which can usually be accomplished by increasing expiratory time, decreasing airway resistance (e.g., bronchodilators, when appropriate), or decreasing minute ventilation. The last approach is usually the most effective ventilatory maneuver, but it results in an increase in the Paco2.

Ventilator-Induced Lung Injury

15

15 0

FIGURE 105-1.  The relationships among alveolar, central airway, and ventilator circuit pressure at the end of exhalation under the following conditions: A, Normal conditions (no auto–positive end-expiratory pressure [auto-PEEP]). B, Severe dynamic airway obstruction with the expiratory port open. C, Severe dynamic airway obstruction with the expiratory port occluded at the end of exhalation. The auto-PEEP level is identified by creating an end-expiratory hold, thereby allowing the alveolar, central airway, and ventilator circuit pressures to equilibrate because there is no flow in the circuit. During equilibration, the level of auto-PEEP can be read on the manometer in the ventilator circuit. (Modified from Pepe PE, Marini JJ. Occult positive end-expiratory pressure in mechanically ventilated patients with airflow obstruction: the auto-PEEP effect. Am Rev Respir Dis. 1982;126:166-170.)

667

Mechanical ventilation itself can lead to numerous types of lung injury2,3 (Fig. 105-2) in addition to oxygen toxicity4 when high levels of inspired oxygen concentrations are administered. Barotrauma refers to pulmonary air leaks, such as pneumothorax and pneumomediastinum. However, a much more subtle injury—diffuse alveolar damage presenting as pulmonary edema—can also occur. For both types of injury, the critical factor is the degree of overdistention of the lung, best assessed by the transpulmonary pressure (Ptp), the airway opening minus pleural pressure (Ppl). The esophageal pressure, measured with an esophageal balloon, estimates Ppl, although this measurement is not routinely performed in clinical practice. The usual pressures measured during mechanical ventilation are airway pressures referenced to atmospheric pressure. The peak inspiratory pressure (PIP) is easy to measure, but its interpretation is not always simple. PIP represents the sum of the pressure needed to overcome the resistance to flow plus the pressure required to inflate the lungs. Thus, a high PIP does not necessarily indicate an increased propensity to overdistending the lung with subsequent ventilator-induced lung injury. For example, for a given inspiratory flow, use of a smaller endotracheal tube will increase PIP, but the danger of pulmonary overdistention is no greater than would be present if the patient was ventilated with a larger-bore tube and a lower PIP. The plateau pressure (Pplat) is the airway pressure at the end of an end-inspiratory pause (usually

668

CHAPTER 105  Mechanical Ventilation  

>0.5 second) and is relatively easy to measure at the bedside if the patient is passive (e.g., receiving a paralytic agent). Depending on Ppl, it has some relationship with the development of overdistention. Although Ppl can vary greatly and no single value of Pplat can be defined as “dangerous” from a lung injury perspective, a reasonable maximal value of Pplat in patients with ARDS is 30 cm H2O. Certain caveats should be noted in interpreting Pplat and PIP, related to associated changes in Ppl. If the patient is breathing spontaneously, Ppl will be negative, and overdistention may occur even with a Pplat much lower than 30 cm H2O. Conversely, in a patient who is either paralyzed or not making ventilatory efforts and who has a stiff chest wall (e.g., due to ascites, obesity, pregnancy), as airway pressure increases, most of the pressure drop will be dissipated across the chest wall, thus leading to values of Ppl that are positive. In this setting, a high Pplat may not be indicative of a high Ptp and hence may not indicate increased lung distention. Thus, the physician caring for a patient receiving mechanical ventilatory support must interpret the measured airway pressures within the clinical context. Measurement of Ppl, as noted earlier, may help resolve these difficulties. During mechanical ventilation, some areas of the lung may undergo cyclic recruitment and de-recruitment. This process, which is of particular

The initiation of mechanical ventilation involves several steps in clinical decision making (Table 105-1). Despite the utility of such guidelines, each patient must be evaluated for specific factors that could modify the recommendation or mandate an alternative.

Acute Respiratory Distress Syndrome

Volume

Increased volume (lung stretch) • Volutrauma • Gross barotrauma • Diffuse alveolar damage Pressure

Epithelial injury

SPECIFIC COMMON TREATMENT SCENARIOS

Initiation of Mechanical Ventilation

• Atelectrauma injury-free zone

Decreased lung volumes • Effects on surfactant • Recruitment/de-recruitment

importance in patients who have ARDS, has been termed atelectrauma and can cause significant lung injury. The precise mechanisms of injury are not entirely clear but are thought to result from shear stress due to opening and closing of lung units, regional hypoxia in atelectatic lung units, and effects on surfactant. Prevention of this type of injury provides part of the rationale for the use of PEEP to maintain recruitment of lung units during tidal ventilation (Video 105-1). Finally, evidence suggests that mechanical ventilation strategies that promote overdistention and atelectrauma can lead to an inflammatory response in the lung, a mechanism of injury termed biotrauma, with the release of proinflammatory cytokines and chemokines. To the extent that these mediators can translocate from the lung into the systemic circulation, they could potentially lead to dysfunction of other organs (Fig. 105-3). This concept suggests that optimal ventilatory strategies are important not only to maintain lung function but also to prevent the development of multipleorgan dysfunction (Chapter 104), a condition that is reasonably frequent in very sick, ventilated patients. This hypothesis may explain the decreased mortality recently observed with a strategy designed to avoid overdistention in a large randomized trial of mechanical ventilation in patients with ARDS.

Biotrauma

Pulmonary edema

FIGURE 105-2.  Schematic representation of the pressure-volume curve of a lung with diffuse alveolar edema. Mechanical ventilation can induce or worsen lung injury by numerous mechanisms when ventilation occurs at high lung volumes or when ventilation occurs at low lung volumes. Lung-protective strategies during ventilation of patients with acute respiratory distress syndrome should try to keep the ventilatory pattern in the injury-free zone. Data in patients confirm the benefit of ensuring that overdistention does not occur.

Patients with ARDS (Chapter 104) have noncardiogenic pulmonary edema, with a reduced functional residual capacity and a mortality rate that commonly exceeds 25%. Although therapy may be available for the underlying disease process that led to the development of ARDS (e.g., antibiotics for a predisposing pneumonia), no effective therapy directly reverses diffuse alveolar damage. These patients require mechanical ventilation as supportive therapy to improve oxygenation and to decrease the oxygen cost of breathing until their lungs recover from the primary insult that led to the alveolar damage. The major goal in treating these patients is to provide adequate gas exchange while ensuring that damaged lungs are not further injured by whatever ventilatory strategy is required to provide sufficient oxygenation. The balanced approaches to minimize lung injury are termed lung-protective ventilation or lung-protective strategies.

Lung-Protective Ventilation Strategies

The lungs of a patient with ARDS are stiff and are characterized on computed tomographic scans by patchy, heterogeneous infiltrates that consist of airless atelectatic or consolidated regions. Many patients have a dependent zone that

Mechanical Ventilation

Biochemical injury (biotrauma) Epithelium/ interstitium

Cytokines, complement, PGs, LTs, ROS, proteases

mφ

Bacteria

Biophysical injury • Shear • Overdistention • Cyclic stretch • Intrathoracic pressure

Alveolar-capillary permeability Cardiac output Organ perfusion

Neutrophils Distal Organ Dysfunction FIGURE 105-3.  Mechanisms by which mechanical ventilation may lead to distal organ dysfunction. LTs = leukotrienes; mϕ = macrophages; PGs = prostaglandins; ROS = reactive oxygen species. (Modified from slu*tsky AS, Tremblay LN. Multiple system organ failure: is mechanical ventilation a contributing factor? Am J Respir Crit Care Med. 1998;157:1721-1725.)

CHAPTER 105  Mechanical Ventilation  

TABLE 105-1  STEPS AND GUIDELINES FOR INITIATION OF MECHANICAL VENTILATION* 1. Ventilatory mode Unintubated patients • NIV for patients with COPD and acute hypercapnic respiratory failure if alert, cooperative, and hemodynamically stable • NIV not routinely recommended for acute hypoxemic respiratory failure Intubated patients • Assist/control with volume-limited ventilation as initial mode • Consider specific indications for PCV or HFOV (see text) in acute lung injury • SIMV: consider if some respiratory effort, dyssynchrony • PSV: consider if patient’s effort good, ventilatory needs moderate to low, and patient more comfortable during PSV trial 2. Oxygenation • If infiltrates on chest radiograph, then FIO2: begin with 0.8-1.0, reduce according to SpO2 PEEP: begin with 5 cm H2O, increase according to PaO2 or SpO2, FIO2 requirements, and hemodynamic effects; consider PEEP/FIO2 “ladder” (see Fig. 105-4); goal of SpO2 >90%, FIO2 ≤ 0.6 • No infiltrates on chest radiograph (COPD, asthma, PTE) FIO2: start at 0.4 and adjust according to SpO2 (consider starting higher if pulmonary embolism is strongly suspected) 3. Ventilation • Tidal volume: begin with 8 mL/kg PBW (see Fig. 105-4 for formulas); decrease to 6 mL/kg PBW over a few hours if acute lung injury present (see Fig. 105-4) • Rate: begin with 10-20 breaths/min (10-15 if not acidotic; 15-20 if acidotic); adjust for pH; goal pH > 7.3 with maximal rate of 35; may accept lower goal if minute ventilation high 4. Secondary modifications • Triggering: in spontaneous modes, adjustment of sensitivity levels to minimize effort • Inspiratory flow rate of 40-80 L/min; higher if tachypneic with respiratory distress or if auto-PEEP present, lower if high pressure in ventilator circuit leads to a high-pressure alarm • Assessment of auto-PEEP, especially in patients with increased airways obstruction (e.g., asthma, COPD) • I/E ratio: 1 : 2, either set or as function of flow rate; higher (1 : 3 or more) if auto-PEEP present • Flow pattern: decelerating ramp reduces peak pressure 5. Monitoring • Clinical: blood pressure, ECG, observation of ventilatory pattern including assessment of dyssynchrony, effort or work by the patient; assessment of airflow throughout expiratory cycle • Ventilator: tidal volume, minute ventilation, airway pressures (including auto-PEEP), total compliance • Arterial blood gases, pulse oximetry *Decisions within this algorithm will be influenced by the specific conditions of the individual patient. COPD = chronic obstructive pulmonary disease; ECG = electrocardiogram; Fio2 = fraction of inspired oxygen; HFOV = high-frequency oscillatory ventilation; I/E ratio = inspiratory-to-expiratory ratio; NIV = noninvasive ventilation; Pao2 = partial pressure of oxygen in arterial blood; PBW = predicted body weight; PCV = pressure-controlled ventilation; PEEP = positive end-expiratory pressure; PSV = pressure-support ventilation; PTE = pulmonary thromboembolism; SIMV = synchronized intermittent mandatory ventilation; Spo2 = arterial oxygen saturation by pulse oximetry.

is consolidated, atelectatic, or fluid filled; a nondependent zone that looks relatively normal; and a middle zone that has some collapsed regions that can be recruited to resemble the nondependent regions if sufficiently increased levels of airway pressure are transiently used (these approaches are called recruitment maneuvers). Arterial oxygen saturation can often be increased by high tidal volumes but at the expense of regional overdistention of those lung units that were not affected by the disease process itself—a treatment strategy that can lead, over time, to worse lung injury and poorer clinical outcomes. The injury caused by mechanical ventilation can be reduced by ventilatory strategies that avoid or minimize regional lung overdistention: limiting inspiratory pressure to some “safe” level or using smaller tidal volumes to limit end-inspiratory stretch, or both. However, in some patients, this lower “dose” of ventilation results in higher levels of Paco2 (so-called permissive hypercapnia) and a lower pH. Higher tidal volumes (12 mL/kg predicted body weight) yielded more normal blood gases, but lower tidal volumes (6 mL/kg predicted body weight) decreased mortality by 22% (from an absolute value of 40 to 31%) in a large clinical trial (Fig. 105-4). A8  Data also suggest that limiting tidal volumes in ventilated patients who are intubated for reasons other than ARDS prevents injury later in the course of

669

their intensive care unit stay.5 A lung-protective strategy with limitation of tidal volume should be considered in ventilated patients who are at high risk for development of acute lung injury or ARDS.

Positive End-Expiratory Pressure

PEEP traditionally has been used to improve oxygenation while at the same time allowing reduction in Fio2 to relatively nontoxic levels. Within the context of the current paradigm of trying to minimize iatrogenic complications of mechanical ventilation, PEEP is a therapy that can potentially minimize the injury caused by ventilation at low lung volumes by recruiting lung units and keeping them open. The critical issues are how to assess the level of PEEP in an individual patient and how to determine whether the procedures to recruit the lung units and keep them open are less harmful than allowing the lung units to remain de-recruited. One experimental option is chest computed tomography to assess whether areas of the lung are recruited, but this technique is not practical for routine assessment. Data are inconclusive regarding the benefits of higher (≈13 cm H2O) compared with lower (≈8 cm H2O) PEEP levels, and PEEP levels often are individualized on the basis of a PEEP/Fio2 table (Fig. 105-4). Higher PEEP levels appear to be associated with decreased mortality in ARDS patients with Pao2/Fio2 of less than 200 mm Hg but not in patients with higher Pao2/ Fio2 ratios. A9  PEEP guided by esophageal pressure measured by an intraesophageal balloon6 can significantly increase Po2 levels and respiratory compliance compared with treatment guided by a standard protocol. A10 

Adjunctive Approaches for Ventilating ARDS Patients

The neuromuscular blocking agent cisatracurium (15 mg intravenous bolus followed by 37.5 mg/hour infusion) can decrease mortality in ARDS patients with Pao2/Fio2 ratios below 150 mm Hg when it is given for 48 hours in patients with early ARDS. A11  The putative mechanism is a decrease in ventilator-induced lung injury. The use of prone position in patients with ARDS can improve oxygenation compared with the supine position by permitting a more even distribution of pleural pressure, thereby reducing ventilator-induced lung injury and decreasing Fio2. Use of the prone position has decreased mortality by an absolute 9% in patients who have Pao2/Fio2 below 100 mm Hg A12  and by an absolute 16% in patients with PaO2/FIO2 below 150 mm Hg. A13  A critical factor in the use of the prone position is proper training of medical personnel in how to place patients safely in the prone position.

Obstructive Airways Diseases

The major physiologic abnormality in patients with obstructive airways diseases (e.g., COPD, asthma) is an increase in airway resistance leading to expiratory airflow limitation; patients may also have a concomitant increase in minute ventilation. These factors may lead to dynamic hyperinflation, which is associated with numerous complications (described earlier), including respiratory muscle compromise, increased oxygen cost of breathing, and hemodynamic compromise. Thus, the main goals in the ventilatory support of patients with obstructive airway diseases are to minimize auto-PEEP, to rest the respiratory muscles, to maintain adequate gas exchange, and to decrease the oxygen cost of breathing while simultaneously minimizing the iatrogenic complications of mechanical ventilation. These strategies allow time for the diagnosis and treatment of the primary cause of the exacerbation (Chapters 87 and 88).

Noninvasive Ventilation

For patients who have acute respiratory failure resulting from an exacerbation of COPD and who require ventilatory support, the preferred approach is NIV if the patient is hemodynamically stable, alert, and cooperative and does not need to be intubated to protect the airway.7 It is important to choose a comfortable mask and to reassure the patient because some patients find the mask difficult to tolerate. This strategy may be applied with several ventilation modes, including pressure support and bilevel positive airway pressure. The ventilation settings are adjusted to improve gas exchange and to ensure the patient’s comfort. Despite this approach, some patients with COPD require intubation and ventilation because of cardiac or respiratory arrest, agitation, increased sputum, worsening respiratory failure, or other concomitant severe disorders.

Intubation and Ventilation

The key goal after intubation is to minimize the detrimental effects of dynamic hyperinflation. The most effective way to do this is to decrease the minute

670

CHAPTER 105  Mechanical Ventilation  

Ventilatory Strategy for Patients with ARDS* Goal 1: Low Vt /Pplat Initiation: Calculate PBW —Male: 50 + 2. 3 (height [inches] – 60) —Female: 45.5 + 2.3 (height [inches] – 60) Initiate volume assist control —start with 8 mL/kg, and to 6 mL/kg over a few hours

Keep Pplat (based on 0.5-sec pause) < 35 cm H2O If Pplat > 30 cm H2O, Vt by 1 mL/kg to 5 or 4 mL/kg If Pplat < 25 AND Vt < 6 mL/kg, Vt by 1 mL/kg until Pplat > 25 cm H2O OR Vt = 6 mL/kg If patient severely distressed and/or breath stacking, consider Vt to 7 or 8 mL/kg, as long as Pplat ≤ 30 cm H2O †

Goal 2: Adequate Oxygenation

Goal 3: Arterial pH Goal: pH: 7.30–7.45 Acidosis algorithm If pH 7.15–7.30 • set rate until pH > 7.30 or PaCO2 < 25 mm Hg (max RR = 35) • if RR = 35 & pH < 7.30 NaHCO3 may be given If pH < 7.15 • set RR to 35 • if set RR = 35 & pH < 7.15, Vt may be in 1 mL/kg steps until pH > 7.15 (Pplat target may be exceeded) Alkalosis algorithm If pH > 7.45 • set RR until patient RR > set RR (minimum set RR = 6/min)

Specific goal: PaO2 55-80 mm Hg or SpO2 88-95% Use only FIO2 /PEEP combinations shown below to achieve this target • if oxygenation is low, choose FIO2 /PEEP combination (from FIO2 /PEEP table) to the right • if oxygenation is high, choose FIO2 /PEEP combination to the left

FIO2/PEEP Table FIO2 PEEP

0.3

0.4

0.4

0.5

0.5

0.6

0.7

0.7

0.7

0.8

0.9

0.9

0.9

5

5

8

8

10

10

10

12

14

14

14

16

18

1.0 18–24

*Based on ARDS Network Algorithm † If compliance of the chest wall is markedly decreased (e.g., massive ascites), it may be reasonable or necessary (if the patient is very hypoxemic) to allow a Pplat >30 cm H2O.

FIGURE 105-4.  Ventilatory strategy for patients with the acute respiratory distress syndrome (ARDS). Several caveats should be considered in using the low tidal volume strategy. (1) Tidal volume (Vt) is based on predicted body weight (PBW), not actual body weight; PBW tends to be about 20% lower than actual body weight. (2) The protocol mandates decreases in the Vt lower than 6 mL/kg of PBW if the plateau pressure (Pplat) is greater than 30 cm H2O and allows small increases in Vt if the patient is severely distressed or if there is breath stacking, as long as Pplat remains at 30 cm H2O or lower. (3) Because arterial carbon dioxide (CO2) levels will rise, pH will fall; acidosis is treated with increasingly aggressive strategies dependent on the arterial pH. (4) The protocol has no specific provisions for the patient with a stiff chest wall, which in this context refers to the rib cage and abdomen; in such patients, it seems reasonable to allow Pplat to increase to more than 30 cm H2O, even though it is not mandated by the protocol; in such cases, the limit on Pplat may be modified on the basis of analysis of abdominal pressure, which can be estimated by measuring bladder pressure. RR = respiratory rate; SpO2 = oxygen saturation based on pulse oximeter.

ventilation, even if this means an increase in Paco2—a strategy known as permissive hypercapnia or controlled hypoventilation. Judicious use of sedation may decrease carbon dioxide production and improve patient-ventilator synchrony, although the avoidance of sedation can reduce the duration of ventilation and hospitalization. A14  In a randomized study, no difference was found between dexmedetomidine and midazolam in time at targeted sedation level, but dexmedetomidine resulted in less time on mechanical ventilation, less delirium, and less hypertension and with less tachycardia but more bradycardia. A15  Care must be taken in the use of paralytic agents, especially when patients with asthma are also receiving corticosteroids, because they may lead to prolonged muscle weakness and resulting difficulty in extubation and post–intensive care unit recovery. Increasing expiratory time by use of a higher peak inspiratory flow may be somewhat helpful, but it is not nearly as effective as decreasing minute ventilation. What level of Paco2 (and pH) should be tolerated is not known with certainty, but maintaining pH higher than approximately 7.20 is a reasonable target if the patient is not having side effects (e.g., arrhythmias, increasing right-sided heart failure), although much lower values have been reported in clinical studies. In patients with COPD who are spontaneously breathing, the addition of external (set) PEEP at a level that is just less than what is necessary to overcome the auto-PEEP fully may not increase Pplat and may decrease the inspiratory effort that the patient needs to generate to initiate inspiratory airflow. This strategy does not appear to be as effective in patients with status asthmaticus, in whom it may cause an increase in Pplat. Measurements of autoPEEP by airway occlusion may be inaccurate in some patients with status asthmaticus, probably because of gas trapping at the end of expiration with closed-off lung regions that do not communicate with the central airways.

DISCONTINUATION OF MECHANICAL VENTILATION

To minimize the iatrogenic consequences of intubation and mechanical ventilation, discontinuation of ventilatory support and extubation should occur as expeditiously as possible. However, if discontinuation is attempted too early, patients may deteriorate and require urgent reintubation.

From the moment that mechanical ventilation is instituted, it is important that the clinician start planning for eventual discontinuation of ventilatory support. A key aspect of this approach is serial evaluation, with aggressive treatment of the factors contributing to the patient’s ventilatory dependence, including respiratory systems factors (e.g., respiratory muscles), cardio­ vascular factors (e.g., myocardial ischemia), neurologic factors (e.g., respiratory muscle weakness), and metabolic factors (e.g., increased oxygen consumption). Two major types of weaning strategies have been used historically: a ventilatory mode thought to hasten the weaning process; and daily monitoring of the patient for criteria suggesting the likelihood of successful weaning, with a trial of spontaneous breathing for patients deemed likely to succeed. Studies of ventilatory modes of weaning have included trials in which patients are allowed to breathe spontaneously from a fresh gas supply delivered to the endotracheal tube (a so-called T-tube), trials of IMV, and trials of pressuresupport ventilation. With all approaches, the level of support is gradually decreased until extubation is tolerated by the patient. These methods have been compared in randomized controlled trials, with mixed results, although weaning with IMV appeared less favorable in most trials. Likewise, use of ventilatory criteria to predict weaning success has been disappointing, mainly because some patients who fail to meet the criteria will be successfully weaned if they are given the opportunity to breathe spontaneously. An easily measured variable is the so-called rapid shallow breathing index, in which the respiratory rate is divided by tidal volume (in liters), with a value of less than 105 suggesting the ability to be weaned; however, false-negative and falsepositive test results commonly occur. More recently, the approach to weaning has been based on the concept that a patient is ready to be removed from ventilatory support when the underlying disease process that led to the intubation has resolved or improved substantially. Rather than applying rigorous ventilatory criteria, the only requirements are that the patient be clinically stable (i.e., has shown improvement in the underlying process), be hemodynamically stable, and have oxygen requirements that can be met by face mask once the patient is extubated.8 If the patient meets these general criteria, a spontaneous breathing trial is recommended (Fig. 105-5); if the patient passes the trial, the patient

CHAPTER 105  Mechanical Ventilation  

671

Approach to Discontinuing Ventilation/Extubation 1. Daily assessment: Is patient ready for a spontaneous breathing trial? • General: resolving process, patient alert, no continuous sedation • Gas exchange: P/F > 200; FIO2 ≤ 50% • Hemodynamics: no vasopressors • Respiratory: PEEP ≤ 5-7 cm H2O

Evaluate and treat reversible causes of failure • Sedation, fluid status, myocardial ischemia, pain control, bronchodilator need, etc.

Yes 2. Initiate screening for spontaneous breathing trial (SBT): • Monitor patient with ECG, oximetry • Patient breathes spontaneously on T-piece, or on PSV of 5-7 cm H2O • Monitor physiologic variables (RR, gas exchange, hemodynamics, subjective comfort) If patient physiologically stable, continue

If patient physiologically unstable

Reinstitute ventilation • Stable, nonfatiguing, comfortable

Failed criterion

3. Continue SBT for 30-120 minutes: Discontinue if any of the following occurs: • General: anxiety or sweating • Gas exchange: SpO2 < 88%; PaCO2 by >10 mm Hg • Hemodynamics: sustained HR changes of >± 20% OR HR > 140/min; SBP < 90 OR > 180 mm Hg • Respiratory: RR > 35/min for > 5 min; signs of WOB (paradoxical breathing, accessory muscles…)

Reintubate

Failure No failure criterion met

4. Extubate

5. Monitor

FIGURE 105-5.  Algorithm for assessing whether a patient is ready to be liberated from mechanical ventilation and extubated. ECG = electrocardiogram; HR = heart rate; P/F = PaO2/FIO2 ratio; PSV = pressure support ventilation; RR = respiratory rate; SBP = systolic blood pressure; SpO2 = oxygen saturation based on pulse oximeter; WOB = work of breathing.

can be extubated. A corollary is that gradual weaning is not necessary; instead, patients should be assessed on a daily basis regarding their suitability for removal from ventilatory support, and if they are not ready, a comfortable, nonfatiguing form of mechanical ventilation should be used between the assessments. Assisted modes of ventilation are preferred between the spontaneous breathing trials. After extubation, evidence suggests that noninvasive ventilation may be beneficial in hypercapnic or high-risk patients.9 An important recommendation in relation to weaning or discontinuation of mechanical ventilation relates to evidence that intensive care units should develop weaning or discontinuation protocols that can be implemented by health care professionals other than physicians. Three large randomized trials demonstrated that protocols implemented by health care professionals other than physicians improved care and were associated with substantial savings in costs compared with standard management approaches, even though the specifics of the protocols were different. A strategy that paired spontaneous awakening, based on the interruption of sedatives, with spontaneous breathing trials improved extubation rates, reduced intensive care length of stay, and decreased mortality by 32%. A16  A major issue to assess before extubation is the patency of the patient’s airway and whether the patient will be able to clear secretions after extubation. Assessment of the likely patency of the upper airway can be achieved by use of the cuff-leak volume, which is the difference between the inspiratory and expiratory tidal volume when the cuff of the endotracheal tube is deflated. If this volume is more than 110 mL, it is usually an indication that major upper airway obstruction will not occur after extubation. Although this test is not required before extubation, a low cuff-leak volume warrants added precautions, such as the availability of equipment and personnel for managing a difficult intubation, when the patient is extubated. In patients who have been ventilated for more than 36 hours, methylprednisolone (20 mg intravenously) started 12 hours before a planned extubation and repeated every 4 hours until tube removal substantially reduces postextubation laryngeal edema and reduces the need for reintubation by 50%. A17  Despite the use of all these techniques, approximately 5 to 25% of patients will have to be reintubated and have mechanical ventilation reinstituted.

Once a patient is reintubated, it is again necessary to reevaluate the respiratory and nonrespiratory reasons for the failure. The choice of the specific weaning protocol should be left to the individual institution and should be individualized to the specific group of patients considered. In instituting such protocols, several key issues should be recognized. First, protocols are guides that should not replace clinical judgment. If a clinician does not follow some aspect of the protocol, there should be a mechanism in place for keeping track of what recommendations were not accepted, with an explanation of the rationale; these data should be collated and used to reassess the protocol. Second, protocols should be viewed as dynamic structures that are open to change and should be reevaluated on a regular basis. Third, implementation of a protocol requires adequate resources, and an institution must make a commitment not only to develop protocols but also to implement and regularly assess them.

Grade A References A1. Papazian L, Forel JM, Gacouin A, et al. Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med. 2010;363:1107-1116. A2. Ferguson ND, Cook DJ, Guyatt GH, et al. High-frequency oscillation in early acute respiratory distress syndrome. N Engl J Med. 2013;368:795-805. A3. Young D, Lamb S, Shah S, et al. High-frequency oscillation for acute respiratory distress syndrome. N Engl J Med. 2013;368:806-813. A4. Williams JW, Cox CE, Hargett CW, et al. Noninvasive Positive-Pressure Ventilation (NPPV) for Acute Respiratory Failure. AHRQ Comparative Effectiveness Reviews, No. 68. Report No. 12-EHC089-EF. Rockville, MD: Agency for Healthcare Research and Quality; 2012. A5. Gray A, Goodacre S, Newby DE, et al. Noninvasive ventilation in acute cardiogenic pulmonary edema. N Engl J Med. 2008;359:142-151. A6. Terragni PP, Antonelli M, Fumagalli R, et al. Early vs late tracheotomy for prevention of pneumonia in mechanically ventilated adult ICU patients: a randomized controlled trial. JAMA. 2010;303:1483-1489. A7. Young D, Harrison DA, Cuthbertson BH, et al. Effect of early vs late tracheostomy placement on survival in patients receiving mechanical ventilation: the TracMan randomized trial. JAMA. 2013;309:2121-2129. A8. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342:1301-1308.

A9. Briel M, Meade M, Mercat A, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and metaanalysis. JAMA. 2010;303:865-873. A10. Talmor D, Sarge T, Malhotra A, et al. Mechanical ventilation guided by esophageal pressure in acute lung injury. N Engl J Med. 2008;359:2095-2104. A11. Alhazzani W, Alshahrani M, Jaeschke R, et al. Neuromuscular blocking agents in acute respiratory distress syndrome: a systematic review and meta-analysis of randomized controlled trials. Crit Care. 2013;17:R43. A12. Hu SL, He HL, Pan C, et al. The effect of prone positioning on mortality in patients with acute respiratory distress syndrome: a meta-analysis of randomized controlled trials. Crit Care. 2014;18: R109. A13. Guerin C, Reignier J, Richard JC, et al. Prone positioning in severe acute respiratory distress syndrome. N Engl J Med. 2013;368:2159-2168. A14. Strom T, Martinussen T, Toft P. A protocol of no sedation for critically ill patients receiving mechanical ventilation: a randomised trial. Lancet. 2010;375:475-480. A15. Riker RR, Shehabi Y, Bokesch PM, et al. Dexmedetomidine vs midazolam for sedation of critically ill patients: a randomized trial. JAMA. 2009;301:489-499. A16. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet. 2008;371:126-134. A17. Francois B, Bellissant E, Gissot V, et al. 12-h pretreatment with methylprednisolone versus placebo for prevention of postextubation laryngeal oedema: a randomised double-blind trial. Lancet. 2007;369:1083-1089.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 105  Mechanical Ventilation  

GENERAL REFERENCES 1. Bello G, De Pascale G, Antonelli M. Noninvasive ventilation: practical advice. Curr Opin Crit Care. 2013;19:1-8. 2. slu*tsky AS, Ranieri VM. Ventilator-induced lung injury. N Engl J Med. 2013;369:2126-2136. 3. Saddy F, Sutherasan Y, Rocco PR, et al. Ventilator-associated lung injury during assisted mechanical ventilation. Semin Respir Crit Care Med. 2014;35:409-417. 4. Kallet RH, Matthay MA. Hyperoxic acute lung injury. Respir Care. 2013;58:123-141. 5. Wilson JG, Matthay MA. Mechanical ventilation in acute hypoxemic respiratory failure: a review of new strategies for the practicing hospitalist. J Hosp Med. 2014;9:469-475.

672.e1

6. Akoumianaki E, Maggiore SM, Valenza F, et al. The application of esophageal pressure measurement in patients with respiratory failure. Am J Respir Crit Care Med. 2014;189:520-531. 7. Ramsay M, Hart N. Current opinions on non-invasive ventilation as a treatment for chronic obstructive pulmonary disease. Curr Opin Pulm Med. 2013;19:626-630. 8. Macintyre NR. Evidence-based assessments in the ventilator discontinuation process. Respir Care. 2012;57:1611-1618. 9. Thille AW, Richard JC, Brochard L. The decision to extubate in the intensive care unit. Am J Respir Crit Care Med. 2013;187:1294-1302.

672.e2

CHAPTER 105  Mechanical Ventilation  

REVIEW QUESTIONS 1. Which of the following is not usually a goal when positive end-expiratory pressure (PEEP) is used in patients with the acute respiratory distress syndrome (ARDS)? A. Decrease the negative physiologic consequences of auto-PEEP B. Recruit alveolar regions C. Decrease ventilator-induced lung injury D. Increase oxygenation E. Increase functional residual capacity Answer: A  Although external PEEP may be useful in some patients with auto-PEEP, especially when it is caused by airway obstruction, this problem usually is not an issue in patients with ARDS. 2. Which of the following is not an exclusion criterion for use of noninvasive ventilation with COPD? A. Facial trauma B. Apnea C. Hypercapnia D. Loss of consciousness E. Hemodynamic instability Answer: C  Hypercapnia is not a contraindication to the use of noninvasive ventilation in patients with COPD. In fact, COPD is one of the indications for its use. Noninvasive ventilation should not be used in most patients with severe facial trauma, because of the problems of fitting the mask, or in patients who are unconscious or apneic, because a patient has to be able to breathe spontaneously to trigger the ventilator. In addition, patients with hemodynamic compromise commonly need full respiratory support to minimize the oxygen cost of breathing. 3. Which of the following is not first-line therapy for a patient who has ARDS and a P/F ratio below 100 mm Hg? A. Lung-protective mechanical ventilation B. Use of the prone position C. Neuromuscular blocking agents D. High-frequency ventilation E. Use of PEEP Answer: D  Two recent studies have shown that high-frequency ventilation does not have a role in the treatment of adult patients with ARDS early in their course.

4. Which of the following is an indication not to try a spontaneous breathing trial? A. A Pao2 = 88 mm Hg while breathing an Fio2 = 1.0 B. PEEP = 5 mm Hg C. Patient is not receiving vasopressors D. Patient is alert E. Patient is receiving minimal sedation Answer: A  In general, it is best not to attempt a spontaneous breathing trial if the patient has a P/F ratio below 200 mm Hg. 5. Which of the following statements about lung-protective strategy is incorrect? A. Ventilator-induced lung injury can be associated with release of a number of mediators that may have systemic consequences. B. Regional lung distention is an important factor in causing ventilatorinduced lung injury. C. Use of a lung-protective strategy is important even in patients who do not have ARDS. D. An increased plateau pressure (>~30 cm H2O) always indicates that the patient is at increased risk of ventilator-induced lung injury. E. The use of a lung-protective strategy in an ARDS patient has been shown to decrease mortality. Answer: D  The key variable in terms of overdistention for ventilator-induced lung injury is overdistention of regional lung units. A patient may have an increased plateau pressure because of conditions that affect chest wall mechanics (e.g., ascites), and if so, the high plateau pressure may not indicate that the lung is being overstretched.

672

CHAPTER 106  Approach to the Patient with Shock  

106  APPROACH TO THE PATIENT WITH SHOCK EMANUEL P. RIVERS

Compensatory Responses

DEFINITION

The key feature of shock is tissue hypoperfusion, not a specific level of systemic arterial blood pressure. The clinical picture may be cryptic or obvious.

EPIDEMIOLOGY

More than 1.2 million patients present in shock or develop shock in U.S. hospitals each year, at an annual cost of more than $100 billion. Shock can be categorized as hypovolemic, cardiogenic (Chapter 107), extracardiac/ obstructive, distributive, or dissociative.1

PATHOBIOLOGY

The delivery and utilization of oxygen are essential for cellular viability, and the failure to deliver or to use oxygen is central to the concept of shock and its pathogenesis (Fig. 106-1). Systemic oxygen delivery (i.e., the amount of oxygen delivered to tissues by the arterial blood) depends on the concentration of hemoglobin in the blood, the fractional saturation of the hemoglobin with oxygen (Sao2), the amount of oxygen dissolved in the blood (Pao2), and cardiac output. Cardiac output is a product of stroke volume and heart rate. Stroke volume is determined by ventricular preload and afterload as well as by contractility of the right or left side of the heart. Systemic vascular resistance (SVR), the force resisting cardiac contraction, can be calculated by Equation 1:

a patient presents with a decrease in cardiac output, but a compensatory increase in SVR maintains a nearly normal MAP. Despite the nearly normal blood pressure, however, the patient is in “cryptic shock” because of tissue hypoperfusion. Compensatory mechanisms are organ specific. Blood flow to organs such as the heart and brain is carefully regulated and maintained over a wide range of blood pressures. In other organs, however, such as the intestine or liver, autoregulation is not as tightly maintained. Systemic oxygen consumption, which is the amount of oxygen consumed by the body per minute, is calculated as the systemic oxygen delivery multiplied by the systemic oxygen extraction ratio. Oxygen demand is the amount of oxygen required by the tissues to avoid anaerobic metabolism. Normally, systemic oxygen delivery is sufficient so that systemic oxygen consumption is not altered by or dependent on changes in delivery. However, if systemic oxygen delivery drops below a critical value, a compensatory increase in the oxygen extraction ratio maintains systemic oxygen consumption at adequate levels to meet systemic oxygen demands. When this compensatory response in the oxygen extraction ratio is inadequate to meet systemic oxygen demands, a switch occurs from aerobic metabolism to the less efficient anaerobic metabolism. The result is depletion of adenosine triphosphate (ATP) and intracellular energy reserves. Intracellular acidosis occurs, and anaerobic glycolysis leads to the production of lactate. Below this critical value of systemic oxygen delivery, systemic oxygen consumption is dependent on systemic oxygen delivery, a relationship termed physiologic oxygen supply dependency. This critical value of systemic oxygen delivery varies substantially because comorbid or preexisting conditions affect the rate of systemic oxygen utilization. A pathologic systemic oxygen delivery dependency exists in patients with sepsis, trauma, and acute respiratory distress syndrome (ARDS) and after resuscitation from prolonged cardiac arrest. These patients have systemic oxygen delivery in the normal or elevated range but an impairment of oxygen utilization. This condition of cytopathic tissue hypoxia is a result of maldistribution of blood flow or a defect in the utilization of substrate at the microcirculatory or subcellular level. This pathologic supply dependency is accompanied by very high mixed venous oxygen saturation levels or venous hyperoxia as well as by elevated lactate levels. This process is believed to be an important mechanism of cellular damage in various forms of shock.

SVR = (MAP − CVP)*80/CO

(1)

MAP = diastolic blood pressure + (systolic − diastolic blood pressure)/3 (2) MAP denotes the mean systemic arterial blood pressure, CVP denotes central venous pressure, and CO denotes cardiac output. SVR is determined primarily by the degree of vasomotor tone in the precapillary smooth muscle sphincters. The systemic circulation is normally autoregulated, so that when systemic arterial pressure increases, vessel diameter decreases to maintain flow at a steady level. The clinical significance of these relationships is apparent when

Minor decreases in arterial blood pressure and systemic oxygen delivery activate the baroreceptor reflex through stretch receptors or sensing mechanisms located in the carotid sinus, splanchnic vasculature, aortic arch, right atrium, and juxtaglomerular apparatus of the kidney as well as through chemoreceptors sensitive to concentrations of carbon dioxide or oxygen located in the central nervous system, mostly in the medulla. These compensatory responses mediated by activation of the sympathetic nervous system include the following: release of cortisol, aldosterone, and epinephrine; activation of the renin-angiotensin system; release of arginine vasopressin from the posterior pituitary; augmentation of myocardial contractility and heart rate; constriction of arterial and venous capacitance vessels, particularly in the splanchnic bed, thereby augmenting venous return; redistribution of blood flow away from skeletal muscle beds and the splanchnic viscera; and creation of a local tissue environment to enhance the unloading of oxygen to tissues and to improve its extraction because of acidosis, pyrexia, and increased red blood cell 2,3-diphosphoglycerate.

Noncompensatory Responses

Noncompensatory responses develop when physiologic adjustments are exaggerated or lead to pathologic results. Vasodilatory shock results from many sources, including unregulated nitric oxide synthesis, inadequate ATP synthesis in vascular smooth muscle cells, activation of the enzyme poly(ADPribose) polymerase 1, lipid mediators, and opening of ATP-sensitive potassium channels in vascular smooth muscle cells. This multifaceted insult leads to interstitial fluid and cellular edema, which impairs oxygen diffusion from capillary to cell, causing a failure of energy-dependent ion transport, the production of lactate, and the inability to maintain normal transmembrane gradients of potassium, chloride, and calcium. Cells lose their ability to use available oxygen as a result of mitochondrial dysfunction, abnormal carbohydrate metabolism, and failure of many energy-dependent enzyme reactions. Acidosis commonly accompanies shock. When a molecule of ATP is hydrolyzed to adenosine diphosphate (ADP) and inorganic phosphate, the reaction also generates a proton. The net yield of protons is positive when ATP is hydrolyzed in the cell and then regenerated only by the anaerobic

CHAPTER 106  Approach to the Patient with Shock  

Systemic oxygen delivery (DO2)

×

Systemic oxygen extraction OER (%) = (100 – SvO2)

Hemoglobin Cardiac output Heart rate × stroke volume

Heart rate

Contractility

Arterial oxygen content (CaO2)

Pulmonary gas exchange (PaO2, SaO2)

Stroke volume (SV) Cardiac output/ heart rate

Preload (CVP or PAOP)

=

673

Systemic oxygen consumption (VO2)

Systemic oxygen demands: Stress Pain Hyperthermia Shivering Work of breathing

Microcirculation

Systemic vascular resistance (SVR) MAP−CVP or PAOP × 80 CO

FIGURE 106-1.  The hemodynamic, oxygen transport, and oxygen utilization components of shock management. Systemic oxygen delivery (DO2) is affected by cardiac output (CO) and arterial oxygen content. The cardiac, pulmonary, and blood determinants of DO2 are shown. CVP = central venous pressure; MAP = mean arterial pressure; OER = oxygen extraction ratio; PAOP = pulmonary artery occlusion pressure.

breakdown of glucose. Thus, during anaerobic glycolysis, the use of ATP to power cellular processes, coupled with the anaerobic production of ATP by substrate-level phosphorylation reactions, results in the development of acidosis. Cells in organs such as the kidneys, liver, and brain can convert lactate into glucose through gluconeogenesis or oxidize lactate to pyruvate and then, ultimately, to carbon dioxide and water. Lactate levels are a reflection of tissue hypoxia, clearance, and alternative sources of production. When the splanchnic circulation is compromised in shock, hepatic lactate clearance is impaired, contributing to the buildup of lactate levels in the circulation. In sepsis, however, the rate of glycolysis increases even in the absence of tissue hypoxia. This phenomenon, which has been termed accelerated aerobic glycolysis, may reflect a change in the ratio of the active to the inactive form of pyruvate dehydrogenase, which is the rate-limiting step for the entry of substrate into the mitochondrial tricarboxylic acid cycle. When systemic oxygen delivery continues to fail to meet systemic oxygen demands, the oxygen debt accumulates. Three stages of shock can ensue. The first stage, which is called early, reversible, or compensated shock, is characterized by compensatory responses to minimize tissue injury. This stage of shock can be self-limited, with full recovery and minimal residual morbidity, if the cause is recognized and treated early. If substantial oxygen debt persists without timely repayment or resolution, inflammation and cellular and microvascular injury define the second stage of shock, which is associated with a prolonged recovery and is typically complicated by organ failure, such as acute lung injury and acute kidney injury. The third stage is late, irreversible, or decompensated shock. In this situation, the oxygen debt is large, and repayment is slow or nonexistent. When shock reaches this point, cellular and tissue injury is extensive and largely irreversible. Progression to multisystem organ failure or death is inevitable, regardless of therapy.

CLINICAL MANIFESTATIONS

The five general types of shock are cardiogenic, distributive, hypovolemic, obstructive, and dissociative. The distinction among these five shock syndromes can be made by combining the history, clinical picture, and hemodynamic measurements. Cardiogenic shock (Chapter 107) and shock syndromes related to sepsis (Chapter 108) are covered in detail elsewhere. The clinical manifestation of shock is variable and depends on the initiating cause and the response of multiple organs. Shock typically is manifested as absolute or relative systemic arterial hypotension and evidence of endorgan dysfunction (Table 106-1). Even a one-time hypotensive episode on hospital admission is associated with increased in-hospital mortality. The extremities are cool and pale if shock is associated with peripheral vasocon-

striction, which is typical of hypovolemic, cardiogenic, and obstructive shock, but they are typically warm and pink with the peripheral vasodilation of distributive shock and dissociative shock (cyanide poisoning). Skin mottling is a physical finding that correlates with hemodynamic compromise, organ failure, and mortality.2 The most frequent neurologic finding in shock is alteration in the level of consciousness, ranging from confusion to coma. Many of the clinically apparent manifestations of cardiac involvement in shock result from sympathoadrenal stimulation, with tachycardia being the most sensitive indicator that shock is present. Acute lung injury, which can be immediate or delayed, results in impaired gas exchange; the work of breathing is increased, and respiratory muscle fatigue and ventilatory failure require mechanical ventilation. Hypovolemia with or without acute tubular necrosis results in oliguria, although polyuria may be seen in early shock. Typical clinical manifestations of gut involvement during shock include abdominal pain, ileus, erosive gastritis, pancreatitis, acalculous cholecystitis, and submucosal hemorrhage. If the integrity of the gut barrier is compromised, bacteria and their toxins are translocated into the blood stream. The most common manifestation of liver involvement in shock is a mild increase in serum levels of aminotransferases and lactate dehydrogenase. With severe hypoperfusion, shock liver may be manifested with massive aminotransferase elevations and extensive hepatocellular damage. Thrombocytopenia may result from dilution during volume repletion or from immunologic platelet destruction, which is especially common during septic shock. Activation of the coagulation cascade can lead to disseminated intravascular coagulation (Chapter 175), which results in thrombocytopenia, decreased fibrinogen, elevated fibrin split products, and microangiopathic hemolytic anemia. The finding of nucleated red blood cells on a peripheral blood smear is associated with increased in-hospital mortality.3

Hypovolemic Shock

Hemorrhagic shock, whether from internal or external bleeding, is the most common cause of hypovolemic shock (Table 106-2). Nonhemorrhagic hypovolemic shock can be caused by severe dehydration due to massive urinary or gastrointestinal fluid losses. Such losses are common in conditions such as diabetic ketoacidosis (Chapter 229) and diarrhea from some infectious diseases, such as cholera (Chapter 302). Massive insensible losses of water or perspiration can precipitate shock in patients with major burn injuries (Chapter 111) or heatstroke (Chapter 109). Sequestration of fluid in the extravascular compartment, commonly referred to as third spacing, can cause shock in patients as a result of surgery, bowel obstruction, hepatic failure (Chapter 154), systemic inflammation, acute pancreatitis (Chapter 144), or

674

CHAPTER 106  Approach to the Patient with Shock  

TABLE 106-1  PHYSICAL EXAMINATION AND SELECTED LABORATORY SIGNS IN SHOCK

TABLE 106-2  CLASSIFICATION OF HEMORRHAGIC SHOCK*

Central nervous system

Blood loss (mL)

Up to 750

750-1500

1500-2000

>2000

% Volume

Up to 15

15-30

30-40

>40

Pulse rate (per minute)

100

>120

>140

Blood pressure

Normal

Normal

Decreased

Decreased

Pulse pressure

Normal or increased

Decreased

Decreased

Decreased

Respiratory rate (per minute)

14-20

20-30

30-40

>35

Urine output (mL/hr)

>30

20-30

5-15

Negligible

Mental status

Slightly anxious

Mildly anxious

Anxious, confused

Confused, lethargic

Fluid replacement

Crystalloid

Crystalloid

Crystalloid and blood

Crystalloid and blood

Acute delirium, restlessness, disorientation, confusion, and coma, which may be secondary to decreased cerebral perfusion pressure (mean arterial pressure minus intracranial pressure). Patients with chronic hypertension or increased intracranial pressure may be symptomatic at normal blood pressures. Cheyne-Stokes respirations may be seen with severe decompensated heart failure. Blindness can be a presenting complaint or complication.

Temperature

Hyperthermia results in excess tissue respiration and greater systemic oxygen delivery requirements. Hypothermia can occur when decreased systemic oxygen delivery or impaired cellular respiration decreases heat generation.

Skin

Cool distal extremities (combined low serum bicarbonate and high arterial lactate levels) aid in identifying patients with hypoperfusion. Pallor, cyanosis, sweating, and decreased capillary refill and pale, dusky, clammy or mottled extremities indicate systemic hypoperfusion. Dry mucous membranes and decreased skin turgor indicate low vascular volume. Low toe temperature correlates with the severity of shock.

General cardiovascular

Neck vein distention (e.g., heart failure, pulmonary embolus, pericardial tamponade) or flattening (e.g., hypovolemia), tachycardia, and arrhythmias Decreased coronary perfusion pressures can lead to ischemia, decreased ventricular compliance, and increased left ventricular diastolic pressure. A “mill wheel” heart murmur may be heard with an air embolus.

Heart rate

Usually elevated. However, paradoxical bradycardia can be seen in patients with preexisting cardiac disease and severe hemorrhage. Heart rate variability is associated with poor outcomes.

Systolic blood pressure

May actually increase slightly when cardiac contractility increases in early shock and then fall as shock advances A single episode of undifferentiated hypotension with a systolic blood pressure 10 mm Hg with inspiration) seen in asthma, cardiac tamponade, and air embolus

Mean arterial blood pressure

Diastolic blood pressure + [pulse pressure/3]

Shock index

Heart rate/systolic blood pressure. Normal = 0.5 to 0.7. A persistent elevation of the shock index (>1.0) indicates impaired left ventricular function (as a result of blood loss or cardiac depression) and is associated with increased mortality.

Respiratory

Tachypnea, increased minute ventilation, increased dead space, bronchospasm, hypocapnia with progression to respiratory failure, acute lung injury, and adult respiratory distress syndrome

Abdomen

Low-flow states may result in abdominal pain, ileus, gastrointestinal bleeding, pancreatitis, acalculous cholecystitis, mesenteric ischemia, and shock liver.

Renal

Because the kidney receives 20% of cardiac output, low cardiac output reduces the glomerular filtration rate and redistributes renal blood flow from the renal cortex toward the renal medulla, thereby leading to oliguria. Paradoxical polyuria in early sepsis may be confused with adequate hydration.

Metabolic

Respiratory alkalosis is the first acid-base abnormality, but metabolic acidosis occurs as shock progresses. Hyperglycemia, hypoglycemia, and hyperkalemia may develop.

CLASS I

CLASS II

CLASS III

CLASS IV

*Estimates based on a 70-kg patient. From Committee on Trauma of the American College of Surgeons. Advanced Trauma Life Support for Doctors. Chicago: American College of Surgeons; 1997:108.

thermal injuries (Chapter 111).4 Regardless of whether hypovolemic shock is due to hemorrhage or fluid losses, the rate of loss is a critical component of the presentation. If volume is lost at a slow rate, compensatory mechanisms are usually effective, and any given amount of volume depletion is often better tolerated than if the same volume were lost acutely. In addition, underlying diseases, especially those that limit cardiac reserve, can substantially influence the clinical severity of a hypovolemic insult. As the importance of the microcirculation continues to be elucidated, it may become a target of future management strategies.

Distributive Shock

The most important and prevalent cause of distributive shock is septic shock (Chapter 108), but anaphylaxis (Chapter 253), drug overdose (Chapter 34), neurogenic insults, and addisonian crisis (Chapter 227) can also produce vasodilatory shock. Sepsis can be a combination of hypovolemia, vasodilation, myocardial suppression, and impaired tissue oxygen use (dissociative shock). In approximately 10 to 15% of septic shock patients, myocardial dysfunction results in a low cardiac output form of shock. Early interventions (Chapter 108) can improve outcomes substantially.

Cardiogenic Shock

Cardiogenic shock (Chapter 107) is defined by a decrease in systemic oxygen delivery caused by an acute or chronic deterioration of cardiac function due to myocardial, valvular, structural, toxic, or infectious causes. The clinical picture of cardiogenic shock is variable, depending on which structural component of the ventricle is impaired.

Extracardiac Obstructive Shock

This form of shock results from acute obstruction to flow in the circulation. Examples include impaired diastolic filling of the right ventricle (e.g., superior vena cava syndrome; Chapter 179), obstruction of right ventricular output (e.g., massive pulmonary embolism; Chapter 98), and an air embolus from cardiopulmonary bypass or central line placement (Chapter 98). Systemic arterial hypertension (Chapter 67) severe enough to impair left ventricular function or acute pericardial tamponade or constrictive pericarditis (Chapter 77) can also produce an obstructive shock pattern.

Dissociative Shock

Dissociative shock results from microvascular abnormalities, with maldistribution or shunting of blood flow, or cytopathic tissue hypoxia. Dissociative shock includes disorders that inhibit oxygen utilization, such as cyanide poisoning (Chapter 110), sodium nitroprusside use, and sepsis.

Mixed Shock States

Shock may arise from multiple causes. For example, a patient with pneumonia and a history of ischemic cardiomyopathy may present in a hypodynamic

675

CHAPTER 106  Approach to the Patient with Shock  

rather than in a hyperdynamic state when combined with sepsis. Thus, a mixture of hypovolemic, distributive, cardiogenic, obstructive, and dissociative shock can potentially be seen in the same patient.

DIAGNOSIS

A key element in the approach to shock is a problem-directed history and physical examination. Some patients present with few symptoms other than generalized weakness, lethargy, or altered mental status. A discussion with the patient and family members should specifically address symptoms that suggest volume depletion, including bleeding, vomiting, diarrhea, excessive urination, insensible losses due to fever, and orthostatic lightheadedness. The history should also inquire about prior or current evidence of cardiovascular disease, especially episodes of chest pain (Chapter 51) or symptoms of heart failure (Chapter 58). Prior neurologic diseases can render patients more susceptible to complications from hypovolemia. Medication use, both prescribed and nonprescribed, must be ascertained. Some medications cause volume depletion (e.g., diuretics), whereas others depress myocardial contractility (e.g., β-blockers, calcium-channel blockers). The possibility of an anaphylactic reaction to a new medication or cardiovascular depression due to drug toxicity should be considered. A recent or remote history of steroid use may suggest adrenal insufficiency (Chapters 35 and 227). The physical examination can provide critical information to aid in the diagnosis (see Table 106-1). Traditionally, shock is defined by a systolic blood pressure less than 90 mm Hg or 40 mm Hg less than the baseline systolic blood pressure if the patient has a history of hypertension. Ultrasound can be incorporated into a formal protocol to evaluate cardiac function, cardiac chamber filling, and certain aspects of the peripheral vasculature (Table 106-3).5

Acidosis

A common theme in shock is that tissue hypoxia leads to acidosis (Chapter 118), which develops as a consequence of anaerobic metabolism and generally parallels the severity of shock. Laboratory manifestations may include a base deficit, low arterial and venous pH levels, and an elevated serum lactate level. Base deficit is the absolute decrease in the serum concentration of bicarbonate (normal minus the patient’s bicarbonate). A mild base deficit is −2 to −5, moderate is −6 to −14, and severe is −15 mmol/L or greater. When patients are resuscitated with large volumes of normal saline, the large fluid load can cause a dilutional acidosis, and the large chloride load can induce metabolic acidosis even in the absence of tissue hypoxia and anaerobic metabolism. Base deficit can also be caused by cocaine, alcohol (Chapter 33), and diabetic ketoacidosis (Chapter 229). Despite its limitations, base deficit provides the clinician with a quick indicator to assess the severity of tissue hypoperfusion and the adequacy of resuscitation in relieving anaerobic metabolism and oxygen debt. The low blood pH of metabolic acidosis can result from different acids. Acidosis caused by lactate and unidentified anions produces an anion gap. The blood lactate concentration rises with increased anaerobic metabolism, as is seen in shock but also in diabetic ketoacidosis (Chapter 229), total parenteral nutrition (Chapter 217), seizures (Chapter 403), thiamine deficiency (Chapter 218), treatment of HIV infection with protease inhibitors (Chapter 388), and administration of metformin, salicylate, isoniazid, propofol, and cyanide (Chapter 110). A lactate concentration greater than 4 mmol/L is unusual in normal and non–critically ill hospitalized patients

and warrants concern. A lactate concentration greater than 4 mmol/L is associated with an in-hospital mortality exceeding 25%, regardless of the cause, and failure to decrease lactate levels during the first 6 hours of shock is associated with an increased inflammatory response, the development of organ failure, and mortality. The arterial-venous difference in carbon dioxide content is inversely related to cardiac output. Whether the samples are taken from the pulmonary artery or the central venous circulation, the relationship and clinical interpretation are the same.

Urine Output

The kidneys normally receive 20% of the systemic oxygen delivery, and because of this large amount of blood flow per gram of tissue, they are highly sensitive to changes in renal blood flow. Urine output is a valuable indicator of renal perfusion and vital organ blood flow. Although a significant drop in urine output indicates reduced renal blood flow, an adequate urine output does not always indicate successful resuscitation. In fact, paradoxical polyuria may be present. Other factors that may affect urine output include the use of mannitol or diuretics. Preexisting conditions, such as renal failure, may also limit the ability of this measure to reflect the adequacy of resuscitation.

TREATMENT  The goal of initial management is to restore global and microvascular perfusion to levels that sustain aerobic cellular respiration. Multiple randomized trials have shown significant and consistent reductions in mortality when shock is reversed aggressively before organ failure develops. Once this initial management is accomplished, the definitive diagnosis leads to more specific therapy based on the cause of shock. Markers of shock serve not only as diagnostic tools for risk stratification but also as targets or end points for the early restoration of adequate tissue perfusion. Clinical monitoring of tissue oxygenation and organ function commonly involves measuring traditional end points of resuscitation, such as heart rate, blood pressure, mentation, urine output, and skin perfusion. Many clinicians continue to use these parameters as indicators that systemic oxygenation imbalances have been corrected. However, there is increasing evidence that clinical parameters may be poor indicators of the ongoing tissue hypoxia and microcirculatory dysfunction that are associated with increased mortality.

Initial Management

The initial management of shock requires immediate diagnostic and therapeutic interventions, including attention to airway, breathing, and circulation and definitive diagnosis and treatment (Chapter 63). The first step to optimize systemic oxygen delivery is to provide supplemental oxygen to increase arterial oxygen content. If any doubt exists about the patency of the airway or the adequacy of ventilation, endotracheal intubation should be performed and mechanical ventilation initiated (Chapter 105). Mechanical ventilation helps provide adequate oxygenation and carbon dioxide elimination and decreases oxygen utilization by the respiratory muscles, which may be responsible for up to 40% of systemic oxygen consumption and lactate production. Although endotracheal intubation and mechanical ventilation may be critical for patients in shock, the sudden increase in airway pressure can lead to a series of deleterious hemodynamic complications, especially in patients who are hypovolemic or vasodilated or have compromised cardiac function. In such patients, the resulting decreased venous return, increased pulmonary

TABLE 106-3  RAPID ULTRASOUND IN SHOCK (RUSH) PROTOCOL* RUSH EVALUATION

HYPOVOLEMIC SHOCK

Heart

Hypercontractile LV Small LV chamber size

Hypocontractile or dilated LV

CARDIOGENIC SHOCK

OBSTRUCTIVE SHOCK Hypercontractile LV Pericardial effusion Cardiac tamponade RV strain Cardiac thrombus

Hypercontractile LV in early sepsis, hypocontractile LV in late sepsis

DISTRIBUTIVE SHOCK

Fluid status

Flat IVC Flat jugular veins Peritoneal fluid (fluid loss) Pleural fluid (fluid loss)

Distended IVC Distended jugular veins Pulmonary edema Pleural or peritoneal fluid

Distended IVC Distended jugular veins Pneumothorax

Normal or small IVC in early sepsis Peritoneal or pleural fluid

Extracardiac circulatory system

Abdominal aneurysm Aortic dissection

Normal

DVT

Normal

*Modified from Perera P, Mailhot T, Riley D, et al. The RUSH exam: Rapid Ultrasound in SHock in the evaluation of the critically ill. Emerg Med Clin North Am. 2010;28:29-56. DVT = deep venous thrombosis; IVC = inferior vena cava; LV = left ventricle; RV = right ventricle.

676

CHAPTER 106  Approach to the Patient with Shock  

TABLE 106-4  VASOPRESSOR AGENTS PERIPHERAL VASCULATURE AGENT

DOSE RANGE

Dopamine

1-4 µg/kg/min 5-10 µg/kg/min

Vasopressin Phenylephrine

20-200 µg/min

CARDIAC EFFECTS

Vasoconstriction Vasodilation Heart Rate Contractility Dysrhythmias

TYPICAL USE

0 1-2+

1+ 1+

1+ 2+

1+ 2+

1+ 2+

11-20 µg/kg/min

2-3+

1+

2+

2+

3+

0.04-0.1 units/min

3-4+

1+

Septic shock, post–cardiopulmonary bypass shock state, no outcome benefit in sepsis

4+

1+

Vasodilatory shock; best for supraventricular tachycardia

Norepinephrine 1-20 µg/min

4+

2+

2+

2+

First-line vasopressor for septic shock, vasodilatory shock

Epinephrine

1-20 µg/min

4+

4+

4+

4+

Refractory shock, shock with bradycardia, anaphylactic shock

Dobutamine

1-20 µg/kg/min

1+

2+

1-2+

3+

3+

Cardiogenic shock, septic shock

Milrinone

37.5-75 µg/kg bolus followed by 0.375-0.75 µg/min

2+

1+

3+

2+

Cardiogenic shock, right-sided heart failure, dilates pulmonary artery; caution in renal failure

vascular resistance, and decreased ventricular compliance may lead to hypotension and cardiovascular collapse.6,7 Early sedation and muscle relaxation with mechanical ventilation have been shown to have outcome benefit, especially with acute lung injury. A1  However, early use of sedatives, anxiolytics, or induction agents during and after intubation can decrease catecholamine levels, peripheral vascular resistance, and cortisol levels and may result in hypotension. If possible, preparations should be made to monitor physiologic variables, to ensure adequate fluid administration, and to provide rapid access to vasopressors should systemic arterial pressure fall to dangerously low levels. On initial presentation, it is good practice to place one or two large-bore (≥16 gauge) peripheral intravenous catheters and to administer a crystalloid solution (normal saline or lactated Ringer solution). If MAP is less than 60 to 65 mm Hg, systolic blood pressure is less than 90 mm Hg, or evidence of tissue hypoperfusion is present, an intravenous fluid challenge (20 to 40 mL/kg crystalloid or colloid) should be given rapidly. A bolus of 500 mL every 30 minutes titrated to MAP or measurement of preload is recommended. In an 80-kg person, the average intravascular volume is 5 L. In shock states, such as septic shock, in which intravascular hypovolemia is a predominant feature, 5 to 6 L of fluid during the first 6 hours is considered an average volume resuscitation. If hemorrhage is the likely cause of shock, blood should be used to replace volume. Fluids should not be withheld, even in patients with end-stage renal disease. Central venous access and arterial blood pressure monitoring should be established to administer vasopressors and to monitor hemodynamics and venous and arterial blood gases, respectively. Electrocardiographic monitoring and continuous measurement of oxygen saturation by pulse oximetry are useful adjuncts. Because the Trendelenburg position may impair gas exchange and promote aspiration, an alternative is to raise the patient’s legs above the level of the heart with the patient in the supine position. If the patient remains hypotensive, vasopressors such as norepinephrine, dopamine, or phenylephrine (Table 106-4) should be administered to restore adequate systemic arterial pressure while the diagnostic evaluation is ongoing. Treatment with vasopressors should not be postponed while trying to achieve euvolemia by using fluid boluses because patients with cerebrovascular and coronary artery disease may be intolerant of the hypotensive interval. However, vasopressors may also mask hypovolemia when they increase blood pressure. If the volume status remains undefined or the hemodynamic condition requires repeated fluid challenges or vasopressor treatment, a central venous catheter should be placed to determine central venous oxygen saturation, ventricular filling pressures, and intravascular volume status while echocardiography is performed. On the basis of these data, patients can usually be classified and managed according to their hemodynamic and oxygen transport patterns (Figs. 106-2 and 106-3).

Fluid Replacement

Rapid and appropriate restoration of vascular volume decreases the need for vasopressor therapy, adrenal replacement therapy, and invasive monitoring; in addition, it modulates the inflammation that arises when a patient progresses to severe shock. The goal of fluid resuscitation is not merely to

“Renal dose” does not improve renal function; may be used with bradycardia and hypotension Vasopressor range

achieve a predetermined volume or pressure but rather to titrate fluid to optimize systemic oxygen delivery and to meet tissue oxygen demands.8 To assess the adequacy of cardiac preload during the resuscitation of a patient with shock, decisions based on monitoring of CVP lead to the same outcomes as those based on measuring equivalence to wedged pulmonary artery occlusion pressure. A2  However, the CVP does not correlate well with left ventricular end-diastolic volume; although a low CVP indicates hypovolemia, a “normal” CVP does not exclude inadequate preload as a cause of shock. A fluid challenge in a volume-responsive patient increases cardiac output by about 20% for each change of 2 cm H2O in CVP; by comparison, cardiac output does not change if the CVP is raised in a patient with adequate left ventricular volume. When intrathoracic pressure increases during the application of positive airway pressure in a mechanically ventilated patient, venous return decreases, and as a consequence, left ventricular stroke volume also decreases. The variation in pulse pressure or stroke volume during a positive-pressure breath can also predict the responsiveness of cardiac output to changes in preload. Pulse pressure variation is defined by Equation 3:

100 × (PPmax − PPmin)/[(PPmax + PPmin)/ 2]

(3)

PPmax and PPmin are the maximal and minimal pulse pressures, respectively, in a respiratory cycle; these measurements must be made when the patient is not making any respiratory efforts on his or her own. Pulse pressure variation values of 13 to 15% suggest that hypovolemia is present and that the cardiac index will increase by at least 15% after the rapid infusion of 500 mL of crystalloid. Pulse pressure variation is a reasonable predictor of volume status and response to fluids, although atrial arrhythmias can interfere with the usefulness of this technique.

Types of Fluids

The two most commonly used crystalloid solutions are 0.9% sodium chloride solution (normal saline) and lactated Ringer solution (Table 106-5). Although these two solutions have been regarded as essentially interchangeable, accumulating data suggest that large volumes of normal saline, but not of lactated Ringer solution, promote the development of hyperchloremic metabolic acidosis and coagulopathy. Hypertonic saline is not recommended for routine use in trauma patients. A3  Colloids are higher-molecular-weight solutions that increase plasma oncotic pressure; they are classified as natural (albumin) or artificial (starches, hetastarch, pentastarch, dextrans, and gelatins). Colloids are dissolved in either normal saline or a balanced salt solution. Colloids stay in the intravascular space significantly longer than crystalloids do, with an intravascular halflife of 16 hours for albumin versus 30 to 60 minutes for normal saline and lactated Ringer solution. When they are titrated to the same volume status, colloids and crystalloids restore tissue perfusion to the same magnitude, but a two to four times greater volume of crystalloids is required to achieve the same end point. The outcomes of patients with hypovolemic shock are equivalent for crystalloids and albumin (see Table 106-5). A4  A5  By comparison, hetastarch increases renal failure and mortality in intensive care unit (ICU) patients with shock. A6  ,

677

CHAPTER 106  Approach to the Patient with Shock  

Patient in Suspected Shock

Diagnostic History Physical exam Laboratory Hemoglobin, WBC, platelets PT, aPTT, INR, D-dimer, fibrinogen Arterial blood gases Electrolytes, Mg, Ca, PO4 BUN, creatinine, glucose, lactate, base deficit, pH Troponin, BNP, and ECG Chest radiograph Cultures (blood, urine)

Assure Adequacy of Airway

Initial Steps

Breathing

Supplemental oxygen

Therapeutic Admit to intensive care unit (ICU) Venous access (1 or 2 large bore) Central venous catheter ECG monitoring Pulse oximetry Hemodynamic support Fluid challenge Vasopressors, unresponsive to fluids Blood Inotropes

Definitive Diagnosis

Circulation

Hemodynamic and oxygen transport data: Hypovolemic Distributive Cardiogenic Obstructive Dissociative

Mechanical ventilation Proactive lung strategies (plateau pressure 92%

Urine output 0.5 mL/kg/hr CVP = 8-12 mm Hg PAOP = 15-8 mm Hg PPV or SVV < 13-15% Fluid challenge

MAP of > 60 and < 90 mm Hg SVR = 1200-1400

Cardiac index > 2.2 L/min/m2 or until ScvO2 > 70% or SvO2 > 65%

Hemoglobin > 10 gm/dL

VO2

Demands

Metabolic End Points SvO2 > 65% ScvO2 > 70% Lactate < 2 mM/L Base deficit < 5 mEq/L pH > 7.35 (a-v)CO2 < 5 mm Hg pHi > 7.35-4

System oxygen demands: Stress Pain Hyperthermia Shivering Work of breathing

Reversal of Organ Dysfunction Encephalopathy Liver function tests Renal function

Heart rate < 100 bpm Shock index (HR/SBP) < 0.9 Stroke volume > 60 mL/beat

FIGURE 106-2.  General hemodynamic management. aPTT = activated partial thromboplastin time; BNP = brain natriuretic peptide; bpm = beats per minute; BUN, blood urea nitrogen; CVP = central venous pressure; DO2 = (systemic) oxygen delivery; ECG = electrocardiogram; HR = heart rate; INR = international normalized ratio; MAP = mean arterial pressure; PAOP = pulmonary artery occlusion pressure; pHi = intestinal mucosal pH; PPV = pulse pressure variation; PT = prothrombin time; SBP = systolic blood pressure; SVR = systemic vascular resistance; SVV = stroke volume variation; VO2 = (systemic) oxygen consumption; WBC = white blood cell.

Fluid Management Strategies

After initial aggressive fluid resuscitation within 6 hours of presentation, controversy exists regarding fluid management strategies in the next 2 days or so, especially in patients with ARDS. Beginning an average of about 48 hours after admission to the ICU and 24 hours after the establishment of ARDS, conservative fluid therapy to maintain euvolemia provides significantly better lung and central nervous system function as well as a decreased need for sedation, mechanical ventilation, and ICU care compared with more aggressive fluid therapy. A7  However, patients who receive conservative volume replacement transiently may need more vasopressor support. One of the negative attributes of an open-ended and oftentimes aggressive fluid resuscitation strategy is that patients may develop an intra-abdominal compartment

syndrome in which elevated abdominal pressures impair gas exchange, decrease renal perfusion, decrease visceral organ perfusion, impair venous return, and thereby decrease cardiac output and systemic oxygen delivery.9

Hemoglobin

In hemorrhagic shock, the rapid administration of packed red blood cells and, if indicated, platelets and thawed fresh-frozen plasma can be life-saving. Whenever possible, fully crossmatched packed red blood cells are preferable, but type-specific blood can often be given safely when immediate therapy is warranted (Chapter 177). In dire emergencies, type O Rh-negative blood can be administered to women of childbearing potential, and type O Rh-negative or Rh-positive blood can be given to men or postmenopausal women.

Increased vascular capacitance (venodilation): Sepsis Anaphylaxis Toxins/drugs

Interstitial fluid redistribution: Thermal injury Trauma Anaphylaxis

Fluid depletion (nonhemorrhagic): External fluid loss Dehydration Vomiting Diarrhea Polyuria

Arrhythmias: Bradycardia Tachycardia

Mechanical: Valvular failure (stenotic or regurgitant) Hypertrophic cardiomyopathy Ventricular septal defect

Impaired systolic contraction (increased ventricular afterload): Right ventricle Pulmonary embolus (massive) Air embolus Acute pulmonary hypertension Left ventricle occlusion or aortic dissection

Decreased cardiac compliance: Constrictive pericarditis Cardiac tamponade

Increased intrathoracic pressure: Tension pneumothorax Mechanical ventilation (with excessive pressure or volume depletion) Asthma

Toxic (e.g., nitroprusside, bretylium, cyanide) End-stage sepsis Catecholamine toxicity Septic (bacterial, fungal, viral, rickettsial) Toxic shock syndrome Anaphylactic anaphylactoid Neurogenic (spinal shock) Endocrinologic Adrenal crisis Thyroid storm

Impaired diastolic filling (decreased ventricular preload): Direct venous obstruction (vena cava) Intrathoracic obstructive tumors

Myopathic: Myocardial infarction Left ventricle Right ventricle

Hemorrhagic: Trauma Gastrointestinal Retroperitoneal Postoperative surgical procedures Myocardial contusion (trauma): Myocarditis Cardiomyopathy Postischemic myocardial stunning Septic myocardial depression Pharmacologic Anthracycline cardiotoxicity Calcium-channel blockers

Dissociative Impairment of oxygen utilization or impairment of the cellular machinery.

Distributive Arterial and venous dilation leads to a decrease in preload, with decreased, normal, or elevated cardiac output, depending on the presence of myocardial depression. Inflammatory causes produce microcirculatory dysfunction, hypotension, and multiple organ system dysfunction without a decrease in cardiac output.

Extracardiac/ Obstructive Obstruction to flow in the cardiovascular circuit that leads to inadequate diastolic filling or decreased systolic function because of increased afterload.

Cardiogenic Severe reduction in cardiac function resulting from direct myocardial damage or a mechanical abnormality of the heart.

Hypovolemic Blood or fluid loss, both leading to a decreased circulating blood volume, diastolic filling pressure, and volume.

FIGURE 106-3.  Definitions, etiologies, and therapies of various shock states. CI = cardiac index; CO = cardiac output; CT = computed tomography; CVP = central venous pressure; DO2 = systemic oxygen delivery; ECG = electrocardiogram; LV = left ventricular; MAP = mean arterial pressure; MRI = magnetic resonance imaging; PA = pulmonary artery; PAOP = pulmonary artery occlusion pressure; RA = right atrial; RV = right ventricular; SVR = systemic vascular resistance; US = ultrasonography; VSD = ventricular septal defect.

A

Etiologies

Definition

Shock

678

CHAPTER 106  Approach to the Patient with Shock  

FIGURE 106-3, cont’d.

B

Therapy

Hemodynamic Characterization

SvO2/ScvO2 Comments: Equalization of intracardiac diastolic pressure is consistent with pericardial tamponade. A rapid x and blunted y descent on CVP wave form indicates the inability of the heart to fill.

SVR SvO2/ScvO2 Comments: If there is a VSD, there may be ScvO2/SvO2 or a step up from the right atrium to right ventricle to the pulmonary artery. Large V wave is seen on PA catheter with mitral regurgitation.

LV infarction: intra-aortic balloon pump (IABP) coronary angiography Recanalization: thrombolytic therapy, angioplasty, coronary bypass surgery RV infarction: fluids and inotropes with PA catheter Mechanical abnormality: echocardiography, cardiac catheterization Corrective surgery: mechanical assist devices, transplantation

SVR SvO2/ScvO2 Comments: Preexisting heart disease may alter the filling pressures

Rapid replacement of blood, colloid or crystalloid Identify source of blood/fluid loss Replace deficient coagulation factors Consider factor VIIa Endoscopy/colonoscopy Angiography CT/MRI/US or other

Pericardial tamponade: pericardiocentesis, surgical drainage (if needed) Pneumothorax: chest tube Pulmonary embolism: heparin ventilation-perfusion lung scan pulmonary angiography consider thrombolytic therapy, embolectomy Severe hypertension: afterload reduction

SVR

early normal later early normal later early later early later early later

Identify site of infection Antimicrobial agents Early goal-directed therapy fluids vasopressors inotropic agents Goals: ScO2 > 70% CVP > 8 mm Hg MAP > 65 mm Hg Improving organ function Decreasing lactate levels Consider activated protein C Requiring vasopressors, consider corticosteroids

Comments: In the early stage, filling pressures are low, leading to low to normal cardiac output and variable ScvO2/SvO2. If myocardial depression accompanies, then a similar profile of cardiogenic shock is seen. After resuscitation a hyperdynamic circulation is seen.

SvO2/ScvO2

SVR

CO/CI/DO2

CO/CI/DO2

CO/CI/DO2

PAOP

CVP/RA

PAOP

CO/CI/DO2

early later

PAOP

PAOP

CVP/RA

CVP/RA

CVP/RA

early normal to later early normal to later early later early later early later

Toxin antidote or remove agent, hyperbaric oxygen therapy Recombinant activated protein C Nitroglycerin (speculative) Prostacyclin (speculative)

Comments: Filling pressures can be variable, depending on stage of the disease. Cardiac output is generally increased. SVR is decreased and ScvO2/SvO2 and lactate are increased.

SvO2/ScvO2

SVR

CO/CI/DO2

PAOP

CVP/RA

CHAPTER 106  Approach to the Patient with Shock  

679

680

CHAPTER 106  Approach to the Patient with Shock  

TABLE 106-5  FLUID THERAPY Normal saline

Normal saline is a slightly hyperosmolar solution containing 154 mEq/L of both sodium and chloride. Because of the relatively high chloride concentration and low pH, normal saline carries a risk of inducing hyperchloremic metabolic acidosis when it is given in large amounts.

Lactated Ringer solution (LR)

Lactate is metabolized to carbon dioxide (CO2) and water by the liver, leading to the release of CO2 in the lungs and excretion of water by the kidneys. LR is preferred to normal saline and buffers acidemia. Because LR contains a very small amount of potassium, there is a small risk of inducing hyperkalemia in patients with renal insufficiency or renal failure. LR may be incompletely metabolized in severe hepatic failure.

Albumin

Albumin is a protein derived from human plasma and is available in varying concentrations from 4 to 25%. A study comparing fluid resuscitation with albumin versus saline found similar 28-day mortalities and secondary outcomes in each arm. However, a post hoc subset analysis of patients with sepsis and acute lung injury resuscitated with albumin showed a trend toward a decrease in mortality. There was a significant increase in mortality in trauma patients, particularly those with head injury.

Hydroxyethyl starch (HES)

HES, which is a synthetic colloid derived from hydrolyzed amylopectin, causes renal impairment at recommended doses and impaired long-term survival at high doses. HES can also cause coagulopathy and bleeding complications from reduced factor VIII and von Willebrand factor levels as well as impaired platelet function. HES increases the risk of acute renal failure and reduces the probability of survival in patients with sepsis.

Dextrans

Dextrans are artificial colloids synthesized by Leuconostoc mesenteroides bacteria grown in sucrose media. Dextrans are used more frequently to lower blood viscosity than for rapid plasma expansion. They can cause renal dysfunction as well as anaphylactoid reactions.

Gelatins

Gelatins are produced from bovine collagen. Because they have a small molecular weight, they are not very effective at expanding plasma volume, but they cost less than other options. They have been reported to cause renal impairment as well as allergic reactions ranging from pruritus to anaphylaxis. Gelatins are not currently available in North America.

significant two-fold increase in arrhythmic events with dopamine compared with norepinephrine (24.1 vs. 12.4%). Dopamine’s effects result from transduction at dopaminergic receptors in the renal, mesenteric, coronary, and systemic circulations. The positive chronotropic and inotropic effects of dopamine can lead to tachycardia and tachyarrhythmias; this effect frequently limits its dosing because the increased myocardial oxygen requirements promote the development of myocardial ischemia, especially in the presence of coronary artery disease. Phenylephrine is a synthetic catecholamine that is a selective α-adrenergic agonist and is ideal in patients with tachycardia. However, the resulting increase in myocardial oxygen consumption, decrease in splanchnic blood flow, and decrease in cardiac output can be detrimental for patients with septic shock. Epinephrine, which is a potent α-, β1-, and β2-adrenergic agonist, increases peripheral arteriolar tone as well as cardiac contractility. It is the first-line agent for the treatment of anaphylactic shock and is used to support myocardial contractility after cardiac surgery. Epinephrine increases the white blood cell count and the blood lactate concentration because of accelerated aerobic glycogenolysis or maldistribution of blood flow. Vasopressin deficiency accompanies vasodilatory shock, and administration of low doses of vasopressin (0.03–0.04 units/minute) increases arterial blood pressure in septic patients with intractable hypotension. In patients with septic shock, the addition of low-dose vasopressin to norepinephrine does not have any overall benefit, A11  but it may benefit patients who have less severe forms of shock and who also receive glucocorticoids.10

Adrenal Dysfunction

Beyond their metabolic functions, glucocorticoids are required to maintain responsiveness to vasopressors, intravascular volume, vascular permeability, and myocardial contractility. If the hypothalamic-pituitary-adrenocortical axis is depressed in shock, clinical findings can include unexplained fever, hypoglycemia, hyponatremia, hyperkalemia, metabolic acidosis, hypotension refractory to fluid resuscitation, and eosinophilia. Cortisol levels and the results of cosyntropin stimulation testing may not be clinically helpful.11 If adrenal insufficiency is strongly suspected, or if patients have refractory hypotension despite vasopressors and hemodynamic optimization, stress doses of intravenous hydrocortisone (e.g., 50 mg every 6 hours) are recommended. A12 

Mechanical Support

Mechanical hemodynamic support with an intra-aortic balloon may be indicated in cardiogenic shock, but it does not improve survival in patients with acute myocardial infarction. A13  Extracorporeal membrane oxygenation is another temporary option for cardiopulmonary support in patients with ARDS until more definitive long-term interventions can be performed.12

PROGNOSIS

The appropriate hemoglobin level in shock remains controversial, but a transfusion threshold value of ≥7 g/dL is as good as a transfusion threshold of 9 g/dL in septic shock. A8  A hemoglobin value of 7 to 10 g/dL is appropriate when the patient is in the acute but stable phase of upper gastrointestinal bleeding. A9 

Vasopressor Therapy

To optimize end-organ perfusion, the second phase of intervention after adequate fluid therapy is to maintain perfusion pressure. A specific MAP goal has not been established for all shock states, but a MAP of at least 60 to 65 mm Hg is a reasonable target. The most common vasopressors are agonists at various adrenergic receptors. Receptors include peripheral α-adrenergic receptors that lead to vasoconstriction; cardiac β1 receptors with both chronotropic and inotropic effects; β2 receptors located in the circulation and airways that mediate vasodilation and bronchodilation; and dopaminergic receptors located throughout the cardiovascular, mesenteric, and renal circulations. On the basis of these mechanisms, therapy can be tailored to a specific circ*mstance. For example, a patient with severe tachycardia would be best served by an agent with more α-selective activity and less β activity to avoid tachycardia and increased myocardial oxygen consumption (see Table 106-4). Norepinephrine, which is a vasoconstrictor and an inotrope, provides better splanchnic oxygen utilization compared with dopamine. It is generally considered the first-line vasopressor for treating persistent hypotension in septic patients despite adequate resuscitation, and it may be superior to dopamine in treating cardiogenic shock. For example, a randomized trial comparing dopamine with norepinephrine in patients with shock showed no significant difference in mortality for hypovolemic and septic shock but a significant benefit of norepinephrine in cardiogenic shock. A10  In addition, there was a

Clinical characteristics associated with a poor outcome include the severity of shock; its temporal duration, underlying cause, and reversibility; and preexisting vital organ dysfunction. Decreased systemic oxygen consumption, persistently elevated lactate levels, size of the base deficit, and severity of the anion gap are associated with increased organ failure and are prognostic in trauma, in septic shock, and after cardiac arrest. Regional measurements of pH are highly predictive of outcome; for example, if the gastric mucosal pH remains below 7.3 for 24 hours, the hospital mortality rate is about 50%. Although many of these poor prognostic signs are suggestive of microcirculatory failure, no targeted therapies yet exist to reverse this disorder. The mortality for an undiagnosed patient who is sent to a general medical ward and develops shock is three times higher than for a patient who is admitted directly to the ICU.

Grade A References A1. Papazian L, Forel JM, Gacouin A, et al. Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med. 2010;363:1107-1116. A2. Wheeler AP, Bernard GR, Thompson BT, et al. Pulmonary-artery versus central venous catheter to guide treatment of acute lung injury. N Engl J Med. 2006;354:2213-2224. A3. Wang JW, Li JP, Song YL, et al. Hypertonic saline in the traumatic hypovolemic shock: metaanalysis. J Surg Res. 2014;191:448-454. A4. Rochwerg B, Alhazzani W, Sindi A, et al. Fluid resuscitation in sepsis: a systematic review and network meta-analysis. Ann Intern Med. 2014;161:347-355. A5. Annane D, Siami S, Jaber S, et al. Effects of fluid resuscitation with colloids vs crystalloids on mortality in critically ill patients presenting with hypovolemic shock: the CRISTAL randomized trial. JAMA. 2013;310:1809-1817.

A6. Zarychanski R, Abou-Setta AM, Turgeon AF, et al. Association of hydroxyethyl starch administration with mortality and acute kidney injury in critically ill patients requiring volume resuscitation: a systematic review and meta-analysis. JAMA. 2013;309:678-688. A7. Wiedemann HP, Wheeler AP, Bernard GR, et al. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006;354:2564-2575. A8. Holst LB, Haase N, Wetterslev J, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med. 2014;371:1381-1391. A9. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013;368:11-21. A10. De Backer D, Biston P, Devriendt J, et al. Comparison of dopamine and norepinephrine in the treatment of shock. N Engl J Med. 2010;362:779-789. A11. Russell JA, Walley KR, Singer J, et al. Vasopressin versus norepinephrine infusion in patients with septic shock. N Engl J Med. 2008;358:877-887. A12. Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358:111-124. A13. Thiele H, Zeymer U, Neumann FJ, et al. Intra-aortic balloon counterpulsation in acute myocardial infarction complicated by cardiogenic shock (IABP-SHOCK II): final 12 month results of a randomised, open-label trial. Lancet. 2013;382:1638-1645.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 106  Approach to the Patient with Shock  

GENERAL REFERENCES 1. Vincent JL, De Backer D. Circulatory shock. N Engl J Med. 2013;369:1726-1734. 2. Ait-Oufella H, Lemoinne S, Boelle PY, et al. Mottling score predicts survival in septic shock. Intensive Care Med. 2011;37:801-807. 3. Desai S, Jones SL, Turner KL, et al. Nucleated red blood cells are associated with a higher mortality rate in patients with surgical sepsis. Surg Infect (Larchmt). 2012;13:360-365. 4. Druey KM, Greipp PR. Narrative review: the systemic capillary leak syndrome. Ann Intern Med. 2010;153:90-98. 5. Kanji HD, McCallum J, Sirounis D, et al. Limited echocardiography-guided therapy in subacute shock is associated with change in management and improved outcomes. J Crit Care. 2014;29: 700-705. 6. Funk DJ, Jacobsohn E, Kumar A. The role of venous return in critical illness and shock—part I: physiology. Crit Care Med. 2013;41:255-262.

681.e1

7. Funk DJ, Jacobsohn E, Kumar A. Role of the venous return in critical illness and shock: part II— shock and mechanical ventilation. Crit Care Med. 2013;41:573-579. 8. Myburgh JA, Mythen MG. Resuscitation fluids. N Engl J Med. 2013;369:1243-1251. 9. Kirkpatrick AW, Roberts DJ, De Waele J, et al. Intra-abdominal hypertension and the abdominal compartment syndrome: updated consensus definitions and clinical practice guidelines from the World Society of the Abdominal Compartment Syndrome. Intensive Care Med. 2013;39: 1190-1206. 10. Russell JA, Walley KR, Gordon AC, et al. Interaction of vasopressin infusion, corticosteroid treatment, and mortality of septic shock. Crit Care Med. 2009;37:811-818. 11. Boonen E, Vervenne H, Meersseman P, et al. Reduced cortisol metabolism during critical illness. N Engl J Med. 2013;368:1477-1488. 12. Abrams D, Combes A, Brodie D. Extracorporeal membrane oxygenation in cardiopulmonary disease in adults. J Am Coll Cardiol. 2014;63:2769-2778.

681.e2

CHAPTER 106  Approach to the Patient with Shock  

REVIEW QUESTIONS 1. Hydroxyethyl starch fluid resuscitation is associated with A. Less renal failure B. Better outcomes C. Higher mortality D. Lowers costs of fluid replacement E. Lowered plasma oncotic pressure Answer: C  Hydroxyethyl starch has increasingly been shown to increase morbidity and mortality in shock. 2. Which vasoactive agent is the least likely to increase heart rate? A. Dopamine B. Epinephrine C. Phenylephrine D. Dobutamine E. Nitroprusside Answer: C  Phenylephrine is a pure α agonist and does not have β-adrenergic activity that precipitates tachycardia. 3. A decreased central venous or mixed venous oxygen saturation is seen in all except which of the following? A. Cardiogenic shock B. Severe cyanide poisoning C. Hemorrhagic shock D. During cardiopulmonary resuscitation E. Saddle pulmonary embolus Answer: B  Cyanide poisoning is the correct answer. This type of poisoning uncouples oxidative metabolism, so less oxygen is extracted from blood, thereby leading to an increased mixed venous oxygen content.

4. Obstructive shock can result from each of the following except which one? A. Pericardial tamponade B. Tension pneumothorax C. Pulmonary embolus D. Increased arterial vascular resistance E. Ventricular septal defect Answer: E  A ventricular septal defect does not obstruct the flow of blood, whereas all other choices represent physiologic conditions in which obstruction can occur. 5. Shock can be accompanied by A. Hypovolemia B. Myocardial suppression C. Vasodilation D. Normal or increased lactate level E. All of the above Answer: E  All of the noted items are associated with shock.

681

CHAPTER 107  Cardiogenic Shock  

TABLE 107-1  DIAGNOSIS OF CARDIOGENIC SHOCK CLINICAL SIGNS Hypotension Oliguria Clouded sensorium Cool and mottled extremities HEMODYNAMIC CRITERIA Systolic blood pressure < 90 mm Hg or > 30 mm Hg decrease from baseline for > 30 minutes Cardiac index < 2.2 L/min/m2 Pulmonary capillary wedge pressure > 18 mm Hg OTHER Documented myocardial dysfunction Exclusion of hypovolemia, hypoxia, and acidosis

Rupture/tamponade, 1.4% Acute MR, 6.9% VSD, 3.9%

107 

RV shock, 2.8% Other, 6.5%

CARDIOGENIC SHOCK STEVEN M. HOLLENBERG

LV failure, 78.5%

DEFINITION

Cardiogenic shock is the syndrome that ensues when the heart is unable to deliver enough blood to maintain adequate tissue perfusion.1 The hemodynamic picture includes sustained systemic hypotension, pulmonary capillary wedge pressure (PCWP) greater than 18 mm Hg, and cardiac index less than 2.2 L/minute/m2 (Table 107-1). Although systolic blood pressure less than 90 mm Hg is a commonly accepted threshold for shock, a decrease of 30 mm Hg from baseline is also used. The diagnosis of cardiogenic shock is often made on clinical grounds—hypotension combined with signs of poor tissue perfusion, including oliguria, clouded sensorium, and cool extremities, all in the setting of myocardial dysfunction. To make the diagnosis, it is important to document myocardial dysfunction and to exclude or to correct factors such as hypovolemia, hypoxemia, and acidosis.

EPIDEMIOLOGY

The predominant cause of cardiogenic shock (Fig. 107-1) is left ventricular failure secondary to acute myocardial infarction (MI)—an extensive first acute MI, the cumulative loss of myocardial function in a patient with previous MI or cardiomyopathy, or a mechanical complication of MI (Chapter 73). However, any cause of severe left ventricular (LV) or right ventricular (RV) dysfunction can lead to cardiogenic shock, including end-stage cardiomyopathy (Chapter 60), prolonged cardiopulmonary bypass, valvular disease (Chapter 75), myocardial contusion (Chapter 111), sepsis with unusually profound myocardial depression (Chapter 108), and fulminant myocarditis (Chapter 60) (Table 107-2). Stress-induced (takotsubo) cardiomyopathy (Chapter 60), a syndrome of acute apical LV dysfunction that occurs after emotional distress, may also be manifested with cardiogenic shock. Acute valvular regurgitation, most often caused by endocarditis (Chapter 76) or chordal rupture (Chapter 75), can lead to shock, as can physiologic stress in the setting of severe valvular stenosis. Cardiac tamponade (Chapter 77) and massive pulmonary embolism (Chapter 98) with acute RV failure can cause shock without pulmonary edema. The incidence and mortality associated with cardiogenic shock appear to be declining.2 In the past 30 years, the incidence has fallen from about 8% to 6% of MIs, largely because of the benefit of early perfusion strategies (Chapter 73). In parallel, mortality from cardiogenic shock has decreased from 70 to 80% to 50% or less, suggesting that increasingly effective early treatment and more widespread adoption of early revascularization have improved the outcomes of patients in whom shock has already developed.

FIGURE 107-1.  Causes of cardiogenic shock in patients with myocardial infarction in the SHOCK trial registry. LV = left ventricular; MR = mitral regurgitation; RV = right ventricular; VSD = ventricular septal defect. (Modified from Hochman JS, Boland J, Sleeper LA, et al. Current spectrum of cardiogenic shock and effect of early revascularization on mortality. Results of an International Registry. SHOCK Registry Investigators. Circulation. 1995;91:873-881).

Risk factors for the development of cardiogenic shock in MI parallel those for LV dysfunction and the severity of coronary artery disease (CAD). Characteristics of patients include older age, anterior MI, diabetes, hypertension, multivessel CAD, previous MI, and peripheral vascular and cerebrovascular disease. Clinical risk factors include decreased ejection fractions, larger infarctions, and lack of compensatory hyperkinesis in myocardial territories remote from the infarction. Clinical harbingers of impending shock include the degree of hypotension and tachycardia at hospital presentation. The factors that predict mortality reflect the severity of the acute insult as well as comorbid conditions. Coronary angiography most often demonstrates multivessel CAD. About 30% of patients have a left main coronary artery occlusion, about 60% have three-vessel coronary disease, and only about 20% have single-vessel disease. Multivessel CAD helps explain the failure to develop compensatory hyperkinesis in remote myocardial segments because of either previous infarction or high-grade coronary stenoses. Only one fourth of patients who develop cardiogenic shock are in shock when they initially present to the hospital; in the others, shock usually evolves during several hours, suggesting that early treatment may prevent shock. Comparison of the clinical characteristics of patients with early and late shock shows similar demographic, historical, clinical, and hemodynamic characteristics, but shock tends to develop earlier in patients with single-vessel CAD than in those with triple-vessel disease. This finding suggests that early shock in the setting of acute MI may be more amenable to revascularization of the culprit vessel by thrombolysis or angioplasty (Chapter 73), whereas shock developing later may require more complete revascularization with multivessel percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) surgery (Chapter 74).

682

CHAPTER 107  Cardiogenic Shock  

TABLE 107-2  CAUSES OF CARDIOGENIC SHOCK ACUTE MYOCARDIAL INFARCTION Pump failure   Large infarction   Smaller infarction with preexisting left ventricular dysfunction   Infarct extension   Reinfarction   Infarct expansion Mechanical complications   Acute mitral regurgitation due to papillary muscle rupture   Ventricular septal defect   Free wall rupture   Pericardial tamponade Right ventricular infarction CARDIOMYOPATHY Myocarditis Peripartum cardiomyopathy End-stage low-output heart failure Hypertrophic cardiomyopathy with outflow tract obstruction Stress cardiomyopathy VALVULAR HEART DISEASE Acute mitral regurgitation (chordal rupture) Acute aortic regurgitation Aortic or mitral stenosis with tachyarrhythmia or other comorbid condition causing decompensation Prosthetic valve dysfunction TACHYARRHYTHMIA OTHER CONDITIONS Prolonged cardiopulmonary bypass Septic shock with severe myocardial depression Penetrating or blunt cardiac trauma Orthotopic transplant rejection Massive pulmonary embolism Pericardial tamponade

PATHOBIOLOGY

Cardiogenic shock is characterized by a downward cascade in which myocardial dysfunction reduces stroke volume, cardiac output, and blood pressure; these changes compromise myocardial perfusion, exacerbate ischemia, and further depress myocardial function, cardiac output, and systemic perfusion. Concomitant diastolic dysfunction increases left atrial pressure, which leads to pulmonary congestion and hypoxemia that can exacerbate myocardial ischemia and impair ventricular performance. Compensatory mechanisms include sympathetic stimulation, which increases heart rate and contractility, and renal fluid retention, which increases preload. Increases in heart rate and contractility raise output but also increase myocardial oxygen demand. Another compensatory mechanism, vasoconstriction to maintain blood pressure, increases myocardial afterload, further impairing cardiac performance and increasing myocardial oxygen demand. In the face of inadequate perfusion, this increased demand can worsen ischemia and perpetuate a vicious circle that, if unbroken, may culminate in death. Interruption of this circle of myocardial dysfunction and ischemia is the basis for therapeutic regimens for cardiogenic shock. In cardiogenic shock, LV dysfunction is not always severe. In one large study, the mean LV ejection fraction was 30%, indicating that mechanisms other than primary pump failure were operative. Furthermore, systemic vascular resistance is not always elevated, suggesting that compensatory vasoconstriction is not universal. Inflammatory responses may contribute to the vasodilation and myocardial dysfunction in cardiogenic shock. Patients in cardiogenic shock may have areas of nonfunctional but viable myocardium due to stunning or hibernation. Myocardial stunning represents postischemic dysfunction that persists despite restoration of normal blood flow. Hibernating myocardial segments have persistently impaired function at rest because of severely reduced coronary blood flow. Although hibernation is conceptually different from stunning, the two conditions may not differ much clinically. Repetitive episodes of myocardial stunning can occur in areas of viable myocardium subtended by a critical coronary

TABLE 107-3  CLINICAL SIGNS OF VOLUME STATUS AND PERFUSION SIGNS AND SYMPTOMS OF CONGESTION Orthopnea, paroxysmal nocturnal dyspnea Jugular venous distention Abdominojugular reflux Rales Hepatomegaly Edema Right upper quadrant tenderness POSSIBLE EVIDENCE OF LOW PERFUSION Narrow pulse pressure Obtundation Cool extremities Cachexia, muscle loss Decreased exercise tolerance Renal/hepatic dysfunction Hypotension with angiotensin-converting enzyme inhibition

stenosis. Such episodes can recapitulate the hibernation phenotype, blurring the distinction between myocardial stunning and hibernation. Regardless of the degree of overlap, their therapeutic implications differ in cardiogenic shock. The contractile function of hibernating myocardium improves with revascularization, whereas stunned myocardium retains inotropic reserve and can respond to inotropic stimulation. In addition, the severity of the antecedent ischemic insult determines the intensity of stunning, providing a rationale for reestablishing the patency of occluded coronary arteries in patients with cardiogenic shock. Finally, the notion that some myocardial tissue may recover function emphasizes the importance of measures to support the patient hemodynamically and to minimize myocardial necrosis in patients with shock.

CLINICAL MANIFESTATIONS

The physical examination should be geared toward evaluating congestion and systemic perfusion to characterize the patient’s hemodynamic profile (Table 107-3). An assessment of whether the patient is “wet” or “dry” and “cold” or “warm” is integral to management. Signs of congestion (Chapter 58) include jugular venous distention (see Fig. 51-3) and pulmonary rales and may include peripheral edema and ascites. Whether the patient is cold or warm is an indication of systemic perfusion. The majority of the cardiogenic shock patients present wet and cold. Patients with shock are usually ashen or cyanotic, and they have cool skin and mottled extremities. Cerebral hypoperfusion may cloud the sensorium. Pulses, which are rapid and faint, may be irregular in the presence of arrhythmias. Jugular venous distention and pulmonary rales are usually present, although their absence does not exclude the diagnosis. A precordial heave resulting from LV dyskinesis may be palpable. The heart sounds may be distant, and third and fourth heart sounds are usually present. A systolic murmur of mitral regurgitation or a ventricular septal defect may be heard, but either complication can occur without an audible murmur (Chapter 73).

DIAGNOSIS

After recognizing the clinical manifestations of apparent cardiogenic shock, the clinician must confirm its presence and assess its cause while simultaneously initiating supportive therapy before irreversible damage to vital organs ensues. The clinician must balance overzealous pursuit of an etiologic diagnosis before achieving stabilization with overzealous empirical treatment without establishing the underlying pathophysiologic process. An electrocardiogram (ECG) should be performed immediately. In cardiogenic shock caused by acute MI, the ECG most commonly shows ST elevation, but ST depression or nonspecific changes are found in 25% of cases. If RV infarction is suspected, ST elevation in modified right-sided leads may be diagnostic (Chapter 73). The ECG may also provide information on previous MIs and rhythm abnormalities. A relatively normal ECG or one showing only diffuse, nonspecific changes in a patient with clinical cardiogenic shock should suggest myocarditis (Chapter 60), especially if the patient has arrhythmias. In end-stage heart failure, the ECG may show Q waves or bundle branch block, indicative of extensive disease.

CHAPTER 107  Cardiogenic Shock  

Other initial diagnostic tests include a chest radiograph, complete blood count, and measurement of arterial blood gases, electrolytes, and cardiac biomarkers. A high-quality chest film can assess signs of pulmonary edema and is helpful when signs suggest an alternative diagnosis, such as a widened mediastinum indicative of aortic dissection (Chapter 78).

Echocardiography

Echocardiography should be performed as early as possible, preferably with color flow Doppler, to provide an expeditious assessment of cardiac chamber size, LV and RV function, valvular structure and motion, atrial size, and the pericardium (Chapter 55). Echocardiography can also assess or diagnose overall and regional systolic function, diastolic function, papillary muscle rupture, acute ventricular septal defect, free wall rupture, degree of mitral regurgitation, presence of RV infarction, cardiac tamponade, and valvular stenosis.

Right-Sided Heart Catheterization

If the history, physical examination, chest radiograph, and echocardiogram demonstrate systemic hypoperfusion, low cardiac output, and elevation of venous pressures, right-sided heart catheterization may not be necessary for diagnosis. However, therapy with vasopressors and inotropic agents is best optimized with hemodynamic measurements. Right-sided heart catheterization can exclude other causes of shock, such as volume depletion and sepsis. A step-up in oxygen saturation between the right atrium and pulmonary artery can indicate a ventricular septal defect (Chapter 69), and large v waves in the PCWP waveform can reflect acute severe mitral regurgitation. RV infarction should be suspected when the PCWP is normal but right-sided filling pressures are notably elevated. Right-sided heart catheterization is most useful, however, to optimize therapy in unstable patients. In such patients, clinical estimates of filling pressures can be unreliable, and changes in myocardial performance or therapeutic interventions can change cardiac output and filling pressures precipitously. Although patients with a low cardiac index (38° C) or hypothermia (90 beats/minute); (3) tachypnea (>20 breaths/ minute), hypocapnia (partial pressure of carbon dioxide 12,000 cells/μL), leukopenia (0% immature band cells) in the circulating white blood cell differential and suspected or proven infection. Bacteremia is defined as the growth of bacteria in blood cultures, but infection does not have to be proved to diagnose sepsis at the onset. Severe sepsis is sepsis in addition to dysfunction of one or more organ systems (e.g., hypoxemia, oliguria, lactic acidosis, thrombocytopenia, decreased Glasgow Coma Scale score). Septic shock is defined as severe sepsis in addition to hypotension (systolic blood pressure 40 mm Hg decrease from baseline) despite adequate fluid resuscitation.1

EPIDEMIOLOGY

Approximately 750,000 cases of severe sepsis or septic shock occur every year in the United States. Sepsis causes as many deaths as acute myocardial infarction, and septic shock and its complications are the most common causes of death in noncoronary intensive care units (ICUs). The medical care costs associated with sepsis are approximately $16.7 billion a year in the United States alone. The frequency of septic shock is increasing as physicians perform more aggressive surgery, as more resistant organisms are present in the environment, and as the prevalence of immune compromise resulting from disease and immunosuppressive drugs increases. Studies suggest that African Americans have a higher incidence of severe sepsis than whites do (6.0 vs. 3.6 per 1000 population) and a higher mortality in ICUs (32.1 vs. 29.3%; P < .0001), even after adjustment for poverty levels. The mechanisms of this apparent difference in risk for and mortality from sepsis are not known. Gram-positive or gram-negative bacteria, fungi, and, very rarely, protozoa or rickettsiae can cause septic shock. Increasingly common causes of septic shock are gram-positive bacteria, especially methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, penicillin-resistant Streptococcus pneumoniae, and resistant gram-negative bacilli. The common infections causing septic shock are pneumonia, peritonitis, pyelonephritis, abscess (especially intra-abdominal), primary bacteremia, cholangitis, cellulitis, necrotizing fasciitis, and meningitis. Nosocomial pneumonia is the most common cause of death from nosocomial infection.

PATHOBIOLOGY

At onset, septic shock activates inflammation, leading to enhanced coagulation, activated platelets, increased neutrophils and mononuclear cells, and

686

CHAPTER 108  Shock Syndromes Related to Sepsis  

diminished fibrinolysis. After several days, a compensatory anti-inflammatory response with immunosuppression may contribute to death. Several pathways amplify one another: inflammation triggers coagulation, and coagulation triggers inflammation, resulting in a positive feedback loop that is proinflammatory and procoagulant. Tissue hypoxia in septic shock also amplifies inflammation and coagulation. Many mediators that are critical for the homeostatic control of infection may be injurious to the host (e.g., tumor necrosis factor-α [TNF-α]), so therapies that fully neutralize such mediators are largely ineffective. Widespread endothelial injury is an important feature of septic shock; an injured endothelium is more permeable, so the flux of protein-rich edema fluid into tissues such as the lung increases. Injured endothelial cells release nitric oxide, a potent vasodilator that is a key mediator of septic shock. Septic shock also injures epithelial cells of the lung and intestine. Intestinal epithelial injury increases intestinal permeability; this leads to epithelial translocation of intestinal bacteria and endotoxin, which further augments the inflammatory phenotype of septic shock.

Early Infection, the Innate Immune Response, Inflammation, and the Endothelium

Host defense is organized into innate and adaptive immune responses. The innate immune system responds by using pattern recognition receptors (e.g., toll-like receptors [TLRs]) to pathogen-associated molecular patterns, which are extremely well conserved molecules of microorganisms. Surface molecules of gram-positive and gram-negative bacteria (peptidoglycan and lipopolysaccharide, respectively) bind to TLR-2 and TLR-4, respectively (E-Fig. 108-1). TLR-2 and TLR-4 binding initiates an intracellular signaling cascade that culminates in nuclear transport of the transcription factor nuclear factor κB (NF-κB), which triggers transcription of cytokines such as TNF-α and interleukin (IL)–6. Cytokines upregulate adhesion molecules of neutrophils and endothelial cells, and neutrophil activation leads to bacterial killing. However, cytokines also directly injure host endothelial cells, as do activated neutrophils, monocytes, and platelets. Inhibition of early cytokine mediators of sepsis, such as TNF-α and IL-1β, has not proved successful, probably because TNF-α and IL-1β peak and then decline quickly, before these antagonist therapies can be applied clinically. After the early cytokine inflammatory response, immune cells, including macrophages and neutrophils, release later mediators, such as high-mobility group box 1 (HMGB-1). HMGB-1 activates neutrophils, monocytes, and endothelium. Unlike TNF-α antagonists, inhibitors of HMGB-1 decrease mortality even when they are given 24 hours after the induction of experimental peritonitis. Another adverse effect of sepsis is widespread endothelial injury that leads to increased endothelial permeability with loss of protein and fluids to the interstitial space. This endothelial permeability is a final common pathway for widespread tissue injury. Cytokines and other inflammatory mediators induce intercellular endothelial cell gaps by disrupting intercellular junctions, by changing cytoskeletal structure, or by direct damage to endothelial cells. Several pathways of altered endothelial permeability have been implicated in sepsis, including protease-activated receptor 1 (PAR-1) and disruption of the intercellular VE-cadherin, β-catenin, and p120-catenin complex. PAR-1 binding by activated protein C and low-dose thrombin is cytoprotective, whereas PAR-1 stimulation by high-dose thrombin increases endothelial permeability. Binding of Slit to Robo4 maintains the integrity of the intercellular VE-cadherin, β-catenin, and p120-catenin complexes and thus maintains healthy endothelial permeability.

Adaptive Immunity Adds Specificity and Amplifies the Immune Response

Microorganisms stimulate specific humoral and cell-mediated adaptive immune responses that amplify innate immunity. B cells release immunoglobulins that bind to microorganisms and thereby facilitate delivery of microorganisms to natural killer cells and neutrophils. In sepsis, type 1 helper T (TH1) cells generally secrete proinflammatory cytokines (TNF-α, IL-1β), and type 2 helper T (TH2) cells secrete anti-inflammatory cytokines (IL-4, IL-10).

Coagulation Response to Infection

Septic shock activates the coagulation system (E-Fig. 108-2) and ultimately converts fibrinogen to fibrin, which is bound to platelets to form microvascular thrombi. Microvascular thrombi further amplify endothelial injury by the release of mediators and by tissue hypoxia because of obstruction to blood flow.

Normally, natural anticoagulants (protein C, protein S, antithrombin, and tissue factor pathway inhibitor) dampen coagulation, enhance fibrinolysis, and remove microthrombi. Thrombin-α binds to thrombomodulin, which activates protein C when protein C is bound to the endothelial protein C receptor (EPCR). Activated protein C dampens the procoagulant phenotype because it inactivates factors Va and VIIIa and inhibits the synthesis of plasminogen activator inhibitor 1 (PAI-1). Activated protein C also decreases apoptosis, leukocyte activation and adhesion, and production of cytokines. Septic shock decreases the levels of the natural anticoagulants protein C, protein S, antithrombin, and tissue factor pathway inhibitor. Furthermore, lipopolysaccharide and TNF-α decrease thrombomodulin and EPCR, thereby limiting the activation of protein C. Lipopolysaccharide and TNF-α also increase levels of PAI-1, inhibiting fibrinolysis.

Tissue Hypoxia in Septic Shock

Tissue hypoxia independently activates inflammation (by activation of NF-κB and cytokines, synthesis of nitric oxide, and activation of HMGB-1), induces coagulation (through tissue factor and PAI-1), and activates neutrophils, monocytes, and platelets. Hypoxia induces hypoxia-inducible factor1α (HIF-1α), which upregulates erythropoietin, and vascular endothelial growth factor (VEGF). Erythropoietin is protective to brain and other tissues. VEGF inhibits fibrinolysis and increases inducible nitric oxide synthase, which augments nitric oxide–induced vasodilation. Nitric oxide has a further injurious effect: excessive nitric oxide inhibits the beneficial actions of HIF-1α (e.g., upregulating synthesis of erythropoietin) during hypoxia.

Late Septic Shock, Immunosuppression, and Apoptosis of Immune and Epithelial Cells

After about 1 week of septic shock, death can result from immunosuppression, which is suggested by anergy, lymphopenia, hypothermia, and nosocomial infection (E-Fig. 108-3). Multiple organ dysfunction may be an anti-inflammatory phenotype because of the apoptosis of immune, epithelial, and endothelial cells. Activated CD4+ T cells evolve into either a TH1 proinflammatory (TNF-α, IL-1β) or a TH2 anti-inflammatory (IL-4, IL-10) phenotype. Sepsis leads to migration from a TH1 to a TH2 phenotype; for example, persistent elevation of IL-10 is associated with an increased risk of death. Immunosuppression also develops because of apoptosis of lymphocytes. Proinflammatory cytokines, activated B and T cells, and glucocorticoids induce lymphocyte apoptosis, whereas TNF-α and endotoxin induce apoptosis of lung and intestinal epithelial cells. The fact that glucocorticoids also stimulate apoptosis could be the biologic explanation for the observation that patients with septic shock who are treated with hydrocortisone have more superinfections than do patients treated with placebo. Death from infectious disease appears to be highly heritable. Sepsis is a prime example of a polygenic disease related to the interaction of multiple genes and an environmental insult (infection). Single-nucleotide polymorphisms of cytokines (TNF-α, IL-6, IL-10), coagulation factors (protein C, fibrinogen-β), the catecholamine pathway (β-adrenergic receptor), and innate immunity genes (CD14, TLR-1, TLR-2) have been variably associated with an increased risk of death from sepsis.

Cardiovascular Dysfunction

Inadequate tissue perfusion and tissue hypoxia are the cardinal features of all types of shock. Early in septic shock, most patients have sinus tachycardia and, by definition, decreased blood pressure (65 mm Hg) Central venous pressure (CVP): normal, 6-12 mm Hg Pulmonary artery pressure (PAP): normal, 25/15 mm Hg Pulmonary vascular resistance (PVR): normal, 150-250 dynes/sec/cm

(

≡

PAP − PAOP × 80 CO

)

Pulmonary artery occlusion pressure (PAOP) or pulmonary artery wedge pressure (PAWP): normal, 8-15 mm Hg Systemic vascular resistance (SVR): normal, 900-1400 dynes/sec/cm

(

≡

MAP − CVP × 80 CO

)

Cardiac output (CO): normal, 5 L/min Left ventricular stroke work index (LVSWI): normal, (60-100 grams × meters/ beats) = (SV × [MAP – PAWP] × 0.0136) Oxygen delivery (Do2): normal, 1 L/min (= CO × [Hg × 1.38 × Sao2] + [0.003 × Po2]) Oxygen consumption (Vo2): normal, 250 mL/min (= CO × Hg × 1.38 × [Sao2 – Svo2] + [0.003 × (Pao2 – Pvo2)]) Oxygen extraction ratio: normal, 0.23-0.32 (= Vo2/Do2) Hemodynamic variables are often normalized to account for different body mass by dividing by body surface area (BSA)    Pulmonary vascular resistance index (PVRI): normal (= PVR/BSA)    Systemic vascular resistance index (SVRI): normal (= SVR/BSA)    Cardiac index (CI): normal, 2.5-4.2 L/min/m2 (= CO/BSA)    Left ventricular stroke work index (LVSWI): normal (= LVSW/BSA)    Oxygen delivery index (Do2I): normal, 460-650 mL/min/m2 (= Do2/BSA)    Oxygen consumption index (Vo2I): normal, 95-170 mL/min/m2 (= Vo2/BSA)

diarrhea if gastrointestinal disease is present. Second, fluid loss from the intravascular to the interstitial space (capillary leak) is caused by mediators that induce widespread endothelial injury, which increases capillary permeability. Increased capillary permeability leads to loss of protein-rich edema fluid into the interstitial space. In the lung, increased permeability is a key component of acute lung injury. A third reason that ventricular preload is decreased in septic shock is venodilation induced by mediators such as nitric oxide. Venodilation increases venous capacitance, leading to relative volume depletion, which compounds the absolute volume depletion. Ventricular afterload is decreased because of excessive release of potent vasodilators such as nitric oxide, prostaglandin I2, adenosine diphosphate, and other vasodilators. In addition to abnormal vasodilation, patients have concurrent microvascular vasoconstriction. Microvascular vasoconstriction may not be apparent clinically or hemodynamically, but it can lead to tissue hypoxia, detected by increased arterial lactate concentrations. Microvascular vasoconstriction is caused by increased norepinephrine, thromboxanes, and other local vasoconstrictors. Microvascular vasoconstriction causes focal hypoxia, which is exacerbated by microvascular obstruction by platelets and leukocytes. The abnormal mismatch of oxygen delivery to oxygen demand can disturb the global relationship of oxygen delivery to oxygen consumption. Normally, oxygen consumption is independent of oxygen delivery over a wide range. When oxygen delivery decreases to less than the critical oxygen delivery level, oxygen consumption decreases and leads to a state in which oxygen consumption depends on oxygen delivery. At levels lower than the critical oxygen delivery level, arterial lactate increases as a result of tissue hypoxia. The clinical implication is that oxygen delivery should be increased (e.g., by increasing cardiac output by volume resuscitation, infusion of dobutamine, or transfusion of erythrocytes) to more than the critical level. Cardiovascular function is further compromised in septic shock because of decreased ventricular contractility.3 Decreased ventricular contractility may be difficult to detect clinically and may be diagnosed only by hemodynamic or echocardiographic assessment. Numerous circulating mediators of sepsis, including endotoxin, cytokines (e.g., IL-6, TNF-α), and nitric oxide (locally released into the coronary circulation), decrease contractility. Endotoxin signals through TLRs to upregulate the expression of proteins such as S110A8 and S100A9 to cause a receptor for advanced glycation end products (RAGE)–dependent decrease in calcium flux, which decreases the ejection fraction. Coronary ischemia resulting from microvascular obstruction by leukocytes and oxygen free radicals, which are released by neutrophils adherent

687

to the coronary capillary endothelium, is another mechanism of decreased contractility. Early in septic shock, patients who survive have increased left ventricular end-diastolic volume, which likely allows them to maintain cardiac output despite decreased contractility. In contrast, nonsurvivors do not have increased left ventricular end-diastolic volume, so their cardiac output is compromised. In some patients with septic shock, concurrent acute lung injury and secondary pulmonary hypertension increase right ventricular afterload, with a secondary shift of the interventricular septum from right to left. This septal shift decreases left ventricular end-diastolic volume and can also limit cardiac output.

CLINICAL MANIFESTATIONS

Cardiovascular dysfunction in septic shock is characterized by decreased preload (because of decreased intake, fluid losses, third spacing resulting from increased permeability, and venodilation), decreased afterload, and often decreased ventricular contractility. Decreased ventricular volume is detected clinically by low jugular venous pressure and hemodynamically by decreased central venous pressure. Left ventricular resistance, or afterload, is also commonly decreased and is detected clinically by warm, flushed skin and hemodynamically by decreased systemic vascular resistance.

DIAGNOSIS

Even as the diagnostic evaluation is beginning, the initial assessment of a critically ill patient must focus immediately on the airway (need for intubation), breathing (respiratory rate, respiratory distress, pulse oximetry), circulation (heart rate, blood pressure, jugular venous pressure, skin perfusion), and rapid initiation of resuscitation (Fig. 108-1). Vital signs and the leukocyte count quickly establish whether the patient has SIRS (two of the four criteria). Arterial blood gases and lactate levels are useful complementary tests. A secondary survey is designed to determine the likely source of infection and the status of organ function. Pneumonia (Chapter 97) is suggested by cough, sputum, and respiratory distress; empyema (Chapter 99) is suggested by pleuritic chest pain. Signs of peritonitis, an abdominal mass, and right upper quadrant tenderness suggest abdominal sepsis. Pyelonephritis (Chapter 284) is likely in patients with dysuria and costovertebral angle tenderness. Integumentary assessment for erythema (cellulitis), line site erythema (line sepsis), tenderness (necrotizing fasciitis), crepitus (anaerobic myonecrosis), and petechiae and purpura (meningococcemia) can be illuminating. Headache, stiff neck, and signs of meningismus raise the suspicion of meningitis (Chapter 412). Focal neurologic signs suggest brain abscess (Chapter 413). Laboratory investigations that are helpful to identify the source of infection include appropriate cultures and Gram stains (blood, sputum, urine, fluids, and cerebrospinal fluid). Blood cultures are positive in 40 to 60% of patients who have septic shock. The chest radiograph aids in the diagnosis of pneumonia, empyema, and acute lung injury. Abdominal ultrasound and computed tomography are indicated if abdominal sepsis is suspected. Hemodynamic assessment of the patient includes diagnostic central venous or pulmonary artery catheterization. In early septic shock, central venous pressure is usually low and increases in response to volume resuscitation. Central venous oxygen saturation, cardiac output, and ventricular filling pressures may be determined continuously. Pulmonary artery pressure is usually normal but may be increased because septic shock can cause pulmonary hypertension. Pulmonary artery occlusion (or wedge) pressure is usually low before resuscitation, but it may be normal or increased if the patient has underlying preexisting heart disease (e.g., heart failure or coronary artery disease with prior myocardial infarction) or if left ventricular contractility is decreased by sepsis. Cardiac output may be low or normal before fluid resuscitation and typically increases to higher than normal after fluid resuscitation. If fluid resuscitation increases central venous pressure and pulmonary artery occlusion pressure but cardiac output does not increase, left ventricular dysfunction is presumably present. Echocardiographic features of decreased ventricular contractility include decreased right and left ventricular ejection fractions and increased enddiastolic and end-systolic volumes. Early in septic shock, the left ventricular ejection fraction is decreased, and it remains low in nonsurvivors. In survivors, the left ventricular ejection fraction usually returns to normal during 5 to 10 days. Bedside echocardiography can also be used to assess intravascular volume status, which can be diagnosed on the basis of collapse of the inferior vena cava, and valvular dysfunction. Renal, hepatic, and coagulation function tests are helpful to evaluate organ function. After determination of the source of sepsis, it is crucial to address

688

CHAPTER 108  Shock Syndromes Related to Sepsis  

Clinical Evaluation Airway Breathing • RR • Distress • Pulse oximetry Circulation • HR, BP • Skin • JVP

• • • •

SIRS (2 of 4) HR (>90 min–1) RR (>20 min–1) or PaCO2 < 32 mm Hg or mechanical ventilation T (>38°C) or T (12,000 mm–3) or WBC ( 65 mm Hg • CVP 8-12 mm Hg • Hct > 30% • ScvO2 > 70% (optional) Consider pulmonary artery catheter or echocardiogram especially if known CV disease

• ABGs • Arterial lactate

+ SIRS • CBC • WBC differential

+ D. Drugs Antibiotics: broad spectrum Consider hydrocortisone Consider vasopressin

+

Source of Infection • Respiratory (pneumonia, empyema) • Abdominal (peritonitis, abscess, cholangitis) • Skin (cellulitis, fasciitis) • Pyelonephritis • CNS (meningitis, brain abscess)

Source of Infection • C&S, Gram stain: blood, sputum, urine, +/– fluids, +/– CSF • CXR • U/S, CT scan

Organ Function CNS • LOC, focal signs Renal function • Urine output

Organ Function • Renal function: electrolytes, BUN, creatinine • Hepatic function: bilirubin, AST, AlkPhos • Coagulation: INR, PTT, platelets, D-dimer

+

+ E. Evaluate source of sepsis

+ F. Fix source of sepsis • Abscess, empyema • Cholecystitis, cholangitis • Urinary obstruction • Peritonitis, bowel infarct • Necrotizing fasciitis • Gas gangrene

FIGURE 108-1.  Algorithm for the clinical and laboratory evaluation and management of septic shock. ABGs = arterial blood gases; AlkPhos = alkaline phosphatase; AST = aspartate aminotransferase; BP = blood pressure; BUN = blood urea nitrogen; C&S = culture and sensitivity; CBC = complete blood count; CNS = central nervous system; CSF = cerebrospinal fluid; CT = computed tomography; CV = cardiovascular; CVP = central venous pressure; CXR = chest radiograph; Hct = hematocrit; HR = heart rate; IBW = ideal body weight; INR = international normalized ratio; JVP = jugular venous pressure; LOC = level of consciousness; MAP = mean arterial pressure; PaCO2 = partial pressure of carbon dioxide; PTT = partial thromboplastin time; RR = respiratory rate; ScvO2 = central venous oxygen saturation; SIRS = systemic inflammatory response syndrome; T = temperature; U/S = ultrasound; WBC = white blood cell count.

that source by draining abscesses and empyemas; radiologically or surgically correcting urinary tract obstruction; and surgically managing peritonitis, bowel infarction, cholecystitis, cholangitis, necrotizing fasciitis, and gas gangrene.

without infection. Obstructive shock (from pulmonary thromboembolism, cardiac tamponade, pneumothorax) is manifested similarly to cardiogenic shock.

Differential Diagnosis

Measures to prevent sepsis include handwashing, elevation of the head of the bed, scrupulous sterile techniques for the insertion of catheters, and possibly the use of antibiotic-impregnated catheters. New catheter insertion sites for catheter changes, isolation of patients who have resistant organisms, and isolation of significantly immunocompromised patients may also prevent infection. Preventing the progression from sepsis to septic shock requires early diagnosis and aggressive resuscitation.4 Early fluid resuscitation, lung-protective ventilation, and antibiotics are critical therapies in early septic shock (Table 108-2).

The major differential diagnoses of classic septic shock are other nonseptic causes of SIRS, such as acute pancreatitis (Chapter 144), acute respiratory distress syndrome (Chapter 104), aspiration pneumonitis (Chapter 94), multiple trauma (Chapter 111), and recent major surgery without infection (Chapter 433). Other causes of distributive shock are anaphylactic shock (suggested by angioedema and hives; Chapter 440), spinal shock (recent trauma and paraplegia; Chapter 399), acute adrenal insufficiency (“tanned skin,” hyperkalemia, metabolic alkalosis; Chapter 227), and acute or acuteon-chronic hepatic failure (jaundice, ascites, encephalopathy; Chapter 153). The differential diagnosis of septic shock must include the other causes of shock: hypovolemic, cardiogenic, and obstructive shock (Chapters 106 and 107). Patients with hypovolemic shock (from internal or external fluid losses, hemorrhage) present with a suggestive history and signs of hypovolemia (low jugular venous pressure) and skin hypoperfusion (cool, clammy, cyanotic extremities). Cardiogenic shock (resulting from myocardial infarction or acute-on-chronic congestive heart failure or occurring after cardiovascular surgery) is suggested by the history, signs of increased filling pressure (increased jugular venous pressure, crackles, S3, pulmonary edema, cardiomegaly), and skin hypoperfusion (Chapter 107). Some patients who have acute myocardial infarction and cardiogenic shock have features of SIRS

PREVENTION

TREATMENT  Respiratory Therapy

All patients in septic shock require oxygen initially, and many require mechanical ventilation. Mechanical ventilation is required in most patients who have septic shock because acute lung injury is the most common complication. Lung-protective ventilation (mechanical ventilation that minimizes lung injury by using a relatively low tidal volume, such as >6 mL/kg of

CHAPTER 108  Shock Syndromes Related to Sepsis  

689

TABLE 108-2  POTENTIAL ANTIBIOTIC REGIMENS FOR PATIENTS WITH SEPTIC SHOCK* SOURCE OF SEPSIS

INITIAL ANTIBIOTIC REGIMEN

ALTERNATIVE ANTIBIOTIC REGIMEN

Community-acquired pneumonia

Third-generation cephalosporin (cefotaxime 2 g IV q6h; ceftriaxone 2 g IV q12h; ceftizoxime 2 g IV q8h) plus Fluoroquinolone (e.g., ciprofloxacin 400 mg IV q12h, levofloxacin 750 mg IV q24h, moxifloxacin 400 mg IV q24h) or Macrolide (azithromycin 500 mg IV q24h)

Piperacillin-tazobactam 3.375 g IV q6h plus Fluoroquinolone or Macrolide

Hospital-acquired pneumonia

Imipenem 0.5 g IV q6h or Meropenem 1 g IV q8h

Fluoroquinolone (ciprofloxacin 400 mg IV q12h) plus Vancomycin 1.5 g IV q12h or Piperacillin-tazobactam 3.375 g IV q6h plus Tobramycin 1.5 mg/kg q8h plus Vancomycin

Abdominal (mixed aerobic/anaerobic)

Piperacillin-tazobactam 3.375 g IV q6h or Imipenem 0.5 g IV q6h (or meropenem 1 g IV q8h)

Ampicillin 2 g IV q4h plus Metronidazole 500 mg IV q8h plus Fluoroquinolone (ciprofloxacin 400 mg IV q12h)

Urinary tract

Fluoroquinolone (ciprofloxacin 400 mg IV q12h)

Ampicillin 2 g IV q4h plus Gentamicin 1.5 mg/kg IV q8h or Third-generation cephalosporin (cefotaxime 2 g IV q6h, ceftriaxone 2 g IV q12h, or ceftizoxime 2 g IV q8h)

Necrotizing fasciitis

Imipenem 0.5 g IV q6h

Penicillin G (if confirmed group A streptococci)

Primary bacteremia (normal host)

Piperacillin-tazobactam 3.375 g IV q6h plus Vancomycin 1.5 g IV q12h

Imipenem 0.5 g IV q6h plus Vancomycin 1.5 g IV q12h

Primary bacteremia (intravenous drug user)

Vancomycin 1.5 g IV q12h plus Fluoroquinolone (ciprofloxacin 400 mg IV q12h)

Piperacillin-tazobactam 3.375 g IV q6h plus Vancomycin 1.5 g IV q12h

Febrile neutropenia

Cefepime 2 g IV q8h plus Vancomycin 1.5 g IV q12h

Piperacillin-tazobactam 3.375 g IV q6h plus Gentamicin 1.5 mg/kg q8h or Imipenem 0.5 g IV q6h plus Gentamicin 1.5 mg/kg q8h

Bacterial meningitis

Ceftriaxone 2 g IV q12h plus Ampicillin 3 g IV q6h plus Vancomycin 1.5 g IV q12h plus Dexamethasone 0.15 mg/kg IV q6h for 2-4 days

Gram-positive cocci: vancomycin plus ceftriaxone 2 g IV q12h Gram-negative diplococci: cefotaxime 2 g IV q4-6h Gram-positive bacilli: ampicillin 3 g IV q6h plus gentamicin Gram-negative bacilli: ceftazidime 2 g IV q8h plus gentamicin 1.5 mg/kg IV q8h All above plus dexamethasone

Cellulitis

Ciprofloxacin 400 mg IV q12h plus Clindamycin 900 mg IV q8h

Imipenem 0.5 g IV q6h

*Most antibiotic doses must be adjusted if there is hepatic or renal dysfunction. Some antibiotics require adjustment based on levels (e.g., gentamicin). In selecting a drug, carefully consider the patient’s history of antibiotic (especially penicillin) allergy.

predicted body weight) decreases mortality from acute lung injury and acute respiratory distress syndrome (Chapter 105). A1  Patients who require ventilation need adequate but not excessive sedation, which can worsen hemodynamic instability, prolong ventilation, and increase the risk for development of nosocomial pneumonia. Sedation should be titrated by objective assessment. Daily interruption of sedation decreases the duration of mechanical ventilation and intensive care. Weaning from mechanical ventilation is often associated with fluid overload from prior fluid resuscitation and from the reduction in intrathoracic pressure. Patients whose weaning is guided by brain natriuretic peptide levels are weaned more quickly and have more ventilator-free days because they generally receive more aggressive diuretic therapy, without a concomitant increased need for vasopressors, an increased risk of renal dysfunction, or more electrolyte abnormalities. A2 

Circulatory Therapy

Early therapy is the cornerstone of emergency management, but such therapy need not achieve specific central hemodynamic targets or require the placement of a central venous catheter or the administration of inotropic agents or blood transfusions. A3  A4  Standard therapies should have the goal of increasing tissue oxygen delivery by increasing profoundly low blood pressure, increasing inadequate blood flow, increasing low arterial oxygen saturation, and increasing mixed venous oxygen saturation. Although oxygen delivery is higher in survivors than in nonsurvivors, it is not clear that a specific oxygen delivery target is more beneficial than clinical end points. Several trials have shown that supernormal global oxygen delivery does not decrease mortality rates in sepsis and septic shock. A mean arterial blood pressure goal of 65 to 70 mm Hg is as good as a goal of 80 to 85 mm Hg. A5  However, a higher mean arterial pressure target (80 to 85 mm Hg) may decrease the risk of renal injury and the need for renal replacement therapy in patients with preexisting hypertension. In patients who have acute lung injury, no difference in outcomes is seen with  ,

management using a pulmonary artery catheter versus a central venous catheter. Patients whose acute lung injury is managed after 24 to 48 hours with a conservative fluid strategy (compared with a liberal fluid strategy) have significantly improved lung function and shorter duration of ventilation and ICU stay. Fluids should be used to maintain central venous pressure at 8 to 12 mm Hg; at present, no convincing data indicate that albumin is better than normal saline solution.5, A6  In patients with severe sepsis, large randomized trials confirm that modified lactated Ringer solution or albumin is preferred to 10% hetastarch (a colloid) because of lower rates of acute kidney injury, less need for renal replacement therapy, and fewer deaths. A7  As a result, hetastarch should not be used in septic shock. If central venous oxygen saturation is less than 70%, packed red cell transfusions should be used to maintain a hematocrit greater than 30%. Vasopressors (e.g., norepinephrine, 1 to 50 µg/minute; epinephrine, 1 to 30 µg/minute) should be added if the mean arterial pressure is less than 65 mm Hg. Dobutamine (2.5 to 20 µg/kg/minute) is required if central venous pressure, mean arterial pressure, and hematocrit are optimized but the central venous oxygen saturation remains less than 70%. In a randomized trial of patients with septic shock, the combination of norepinephrine plus dobutamine resulted in a mortality similar to that with epinephrine alone, with no differences in organ dysfunction, time to resolution of shock, or adverse events. A8  In another randomized trial, norepinephrine was slightly but not significantly better than dopamine for reducing mortality when used as the first-line vasopressor for patients with septic shock A9 ; however, norepinephrine was associated with a lower rate of arrhythmias, especially atrial fibrillation. These accumulated data suggest that norepinephrine may be preferable to dopamine as the first vasopressor in septic shock. Clinicians can use epinephrine alone, norepinephrine alone, or norepinephrine plus dobutamine in patients with low cardiac output. As a strategic approach to persistent hypotension despite adequate fluid resuscitation, a

690

CHAPTER 108  Shock Syndromes Related to Sepsis  

vasopressor such as norepinephrine (1 to 50 µg/minute) can be added first. If the cardiac index is low or if the mixed venous oxygen saturation is low (>70%) despite an adequate central venous pressure, an inotropic agent such as dobutamine should be added, initially at approximately 2 to 5 µg/kg/minute and increasing until the mixed venous oxygen saturation is adequate. In some patients in septic shock, the cardiac index is inadequate, as reflected by a low mixed venous oxygen saturation despite a high central venous pressure (>12 mm Hg) or pulmonary artery wedge pressure (>18 mm Hg) because of underlying cardiovascular dysfunction or because of acute left ventricular dysfunction resulting from sepsis. In such patients, earlier use of an inotropic agent such as dobutamine should be considered to increase left ventricular contractility. The overall goal is to achieve an adequate mean arterial pressure (>65 mm Hg), central venous pressure, and mixed venous oxygen saturation while other indices of adequate perfusion are monitored, such as hourly urine output (>0.5 mL/kg/hour), arterial lactate levels ( 70% (optional) Consider pulmonary artery catheter or echocardiogram especially if known cardiovascular disease; goals include: • Wedge pressure 8-15 mm Hg • Cardiac index: normal or increased

D. Drugs: • Antibiotics: Narrow spectrum to cause of infection • Hydrocortisone (if evidence of relative adrenal insufficiency [see text]): hydrocortisone 50 mg intravenously every 6 hours and fludrocortisone 50-µg tab orally or per NG tube daily for 7 days

Other Organ Support • Renal function: Continuous renal replacement • DVT prophylaxis: Low-dose heparin 5000 IU subcutaneously every 12 hours • Stress ulcer prophylaxis: H2-receptor antagonist (e.g., ranitidine 50 mg intravenously every 8 hours) • Nutrition: Enteral preferred • Sedation: Intermittent with daily awakening FIGURE 108-2.  Ongoing critical care support and management in septic shock. CVP = central venous pressure; DVT = deep venous thrombosis; Hg = hemoglobin; IBW = ideal body weight; MAP = mean arterial pressure; NG = nasogastric; ScvO2 = central venous oxygen saturation.

PROGNOSIS

The 28-day mortality of septic shock has decreased during the past 20 years from about 50% to about 25 to 35%,7 especially in academic centers,8 probably because of the earlier initiation of appropriate therapies at appropriate doses for limited periods. Early deaths (in the first 72 hours) are usually the result of refractory, progressive shock despite escalating life support. Later deaths from septic shock (after day 3) are usually secondary to multiple organ dysfunction. The number of dysfunctional organs and the progression or lack of improvement of organ dysfunction are indicators of increased risk of death. Other factors that portend a poor prognosis are increased age, underlying medical conditions, more severe illness, increased arterial lactate concentrations, and the need for high-dose vasopressors. Furthermore, a delay in achieving adequate resuscitation is associated with increased mortality. As the number of survivors of septic shock has increased, so have the numbers with significant long-term sequelae, including cognitive dysfunction,9 depression, and post-traumatic stress disorder. Survivors of septic shock who also had acute lung injury (Chapter 104) can have weakness, fatigue, and dyspnea on exertion after hospital discharge due to pulmonary dysfunction, neuromuscular dysfunction, or other persistent organ dysfunction. Patients who have an episode of acute kidney injury during septic shock have a significantly decreased long-term survival than patients without it.10 Overall, the survival and quality of life after hospital discharge after septic shock remain poorer than expected for at least the next 10 years.11,12

Grade A References A1. Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342:1301-1308. A2. Mekontso Dessap A, Roche-Campo F, Kouatchet A, et al. Natriuretic peptide–driven fluid management during ventilator weaning: a randomized controlled trial. Am J Respir Crit Care Med. 2012;186:1256-1263.

A3. Yealy DM, Kellum JA, Huang DT, et al. A randomized trial of protocol-based care for early septic shock. N Engl J Med. 2014;370:1683-1693. A4. Peake SL, Delaney A, Bailey M, et al. Goal-directed resuscitation for patients with early septic shock. N Engl J Med. 2014;371:1496-1506. A5. Asfar P, Meziani F, Hamel JF, et al. High versus low blood-pressure target in patients with septic shock. N Engl J Med. 2014;370:1583-1593. A6. Caironi P, Tognoni G, Masson S, et al. Albumin replacement in patients with severe sepsis or septic shock. N Engl J Med. 2014;370:1412-1421. A7. Rochwerg B, Alhazzani W, Sindi A, et al. Fluid resuscitation in sepsis: a systematic review and network meta-analysis. Ann Intern Med. 2014;161:347-355. A8. Annane D, Vignon P, Renault A, et al. Norepinephrine plus dobutamine versus epinephrine alone for management of septic shock: a randomized trial. Lancet. 2007;370:676-684. A9. De Backer D, Biston P, Devriendt J, et al. Comparison of dopamine and norepinephrine in the treatment of septic shock. N Engl J Med. 2010;362:779-789. A10. Jones AE, Shapiro NI, Trzeziak S, et al. Lactate clearance vs. central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial. JAMA. 2010;303:739-746. A11. Schortgen F, Clabault K, Katsahian S, et al. Fever control using external cooling in septic shock: a randomized controlled trial. Am J Respir Crit Care Med. 2012;185:1088-1095. A12. Holst LB, Haase N, Wetterslev J, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med. 2014;371:1381-1391. A13. Brunkhorst FM, Oppert M, Marx G, et al. Effect of empirical treatment with moxifloxacin and meropenem vs meropenem on sepsis-related organ dysfunction in patients with severe sepsis: a randomized trial. JAMA. 2012;307:2390-2399. A14. Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358:111-124. A15. Annane D, Cariou A, Maxime V, et al. Corticosteroid treatment and intensive insulin therapy for septic shock in adults: a randomized controlled trial. JAMA. 2010;303:341-348. A16. Ranieri VM, Thompson BT, Barie PS, et al. Drotrecogin alfa (activated) in adults with septic shock. N Engl J Med. 2012;366:2055-2064. A17. Russell JA, Walley KR, Singer J, et al. Vasopressin versus norepinephrine infusion in patients with septic shock. N Engl J Med. 2008;358:877-887. A18. Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283-1297. A19. Palevsky PM, Zhang JH, O’Connor TZ, et al. Intensity of renal support in critically ill patients with acute kidney injury. N Engl J Med. 2008;359:7-20. A20. Cook D, Meade M, Guyatt G, et al. Dalteparin versus unfractionated heparin in critically ill patients. N Engl J Med. 2011;364:1305-1314. A21. Harvey SE, Parrott F, Harrison DA, et al. Trial of the route of early nutritional support in critically ill adults. N Engl J Med. 2014;371:1673-1684. A22. Rice TW, Wheeler AP, Thompson BT, et al. Initial trophic vs full enteral feeding in patients with acute lung injury: the EDEN randomized trial. JAMA. 2012;307:795-803.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 108  Shock Syndromes Related to Sepsis  

GENERAL REFERENCES 1. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med. 2013;369:840-851. 2. Funk DJ, Jacobsohn E, Kumar A. The role of venous return in critical illness and shock—part I: physiology. Crit Care Med. 2013;41:255-262. 3. Vincent JL, De Backer D. Circulatory shock. N Engl J Med. 2013;369:1726-1734. 4. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41:580-637. 5. Myburgh JA, Mythen MG. Resuscitation fluids. N Engl J Med. 2013;369:1243-1251. 6. Desai SV, McClave SA, Rice TW. Nutrition in the ICU: an evidence-based approach. Chest. 2014;145:1148-1157. 7. Kaukonen KM, Bailey M, Suzuki S, et al. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA. 2014;311:1308-1316.

691.e1

8. Walkey AJ, Wiener RS. Hospital case volume and outcomes among patients hospitalized with severe sepsis. Am J Respir Crit Care Med. 2014;189:548-555. 9. Pandharipande PP, Girard TD, Jackson JC, et al. Long-term cognitive impairment after critical illness. N Engl J Med. 2013;369:1306-1316. 10. Linder A, Fjell C, Levin A, et al. Small acute increases in serum creatinine are associated with decreased long-term survival in the critically ill. Am J Respir Crit Care Med. 2014;189:1075-1081. 11. Iwashyna TJ, Ely EW, Smith DM, et al. Long-term cognitive impairment and functional disability among survivors of severe sepsis. JAMA. 2010;304:1787-1794. 12. Linder A, Guh D, Boyd J, et al. Long term (10 year) mortality of younger previously healthy severe sepsis patients is worse than nonseptic patients and general population. Crit Care Med. 2014;42: 2211-2218.

691.e2

CHAPTER 108  Shock Syndromes Related to Sepsis  

REVIEW QUESTIONS 1. Which of the following statements about dopamine in septic shock is correct? A. Low-dose dopamine is effective in decreasing the risk of acute kidney injury. B. Dopamine infusion was associated with increased cardiovascular adverse effects compared with norepinephrine in a large mortalitypowered randomized controlled trial. C. Dopamine increases mean arterial pressure excessively compared with norepinephrine. D. Dopamine alters dopaminergic receptor neurologic functions and alters mental status in septic shock. E. Randomized controlled trial evidence shows that dopamine infusion shortens the duration of vasopressor support compared with norepinephrine. Answer: B  Dopamine infusion is associated with increased cardiovascular adverse effects compared with norepinephrine (De Backer D, Biston P, Devriendt J, et al. Comparison of dopamine and norepinephrine in the treatment of septic shock. N Engl J Med. 2010;362:779-789; Patel GP, Grahe JS, Sperry M, et al. Efficacy and safety of dopamine versus norepinephrine in the management of septic shock. Shock. 2010;33:375-380.). Low-dose dopamine does not decrease the risk of acute kidney injury (Bellomo R, Chapman M, Finfer S, et al. Australian and New Zealand Intensive Care Society Clinical Trials Group. Low dose dopamine in patients with early renal dysfunction: a placebo controlled randomised trial. Lancet. 2000;356:2139-2143.) and is not recommended in the surviving sepsis guidelines (Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41:580-637.) for this reason. There is no convincing evidence that dopamine increases mean arterial pressure excessively compared with norepinephrine; both are given as continuous infusions that can be effectively titrated to a target mean arterial pressure. Although there are dopaminergic neurons, there is no evidence that dopamine infusion alters dopaminergic receptor neurologic functions or alters mental status in septic shock. There is no evidence that dopamine shortens the duration of vasopressor support compared with norepinephrine (De Backer D, Biston P, Devriendt J, et al. Comparison of dopamine and norepinephrine in the treatment of septic shock. N Engl J Med. 2010;362:779-789.).

2. Septic shock has not been shown to be associated with which of the following complications during post-hospitalization recovery? A. Cognitive dysfunction B. Decreased long-term survival C. Neuromuscular dysfunction D. Impaired quality of life E. Recurrent seizure disorder Answer: E  Cognitive function, long-term survival, neuromuscular function, and quality of life are all reduced among survivors of severe sepsis. However, survivors of septic shock do not have increased risk of recurrent seizures. (Pandharipande PP, Girard TD, Jackson JC, et al. Long-term cognitive impairment after critical illness. N Engl J Med. 2013;369:1306-1316. Herridge MS, Tansey CM, Matte A, et al. Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med. 2011;364:1293-1304. Iwashyna TJ, Ely EW, Smith DM, et al. Long-term cognitive impairment and functional disability among survivors of severe sepsis. JAMA. 2010;304:1787-1794.)

691

CHAPTER 109  Disorders Due to Heat and Cold  

109  DISORDERS DUE TO HEAT AND COLD MICHAEL N. SAWKA AND FRANCIS G. O’CONNOR

TEMPERATURE REGULATION

Body temperature is regulated through two parallel processes that modify body heat balance: behavioral (clothing, shelter, physical activity) and physiologic (skin blood flow, sweating, shivering). Both peripheral (skin) and central (core) thermal receptors provide afferent input to a central nervous system integrator (hypothalamic thermoregulatory center), and any deviation between the controlled variable (body temperature) and a theoretical reference variable (“set point” temperature) results in a heat loss or conservation response (E-Fig. 109-1). Humans normally regulate body (core) temperature at about 37° C (98.6° F), and fluctuations within the narrow range of 35° C to 41° C (95° F to 105.8° F) can be tolerated by healthy acclimatized persons; core temperatures outside this range can induce morbidity and mortality. There is no single core temperature because it varies at different deep body sites and during rest and physical exercise. Arterial blood temperature, which provides the best invasive measurement of core temperature, is slightly lower than brain temperature. The most accurate noninvasive index of core temperature is esophageal temperature, followed in order of preference by rectal, gastrointestinal tract (telemetry pill), and oral temperature. Ear (tympanic and auditory meatus) or scanned temporal artery temperature should not be relied on for clinical judgment. Rectal temperatures are most commonly recommended because they are easy to measure and are not biased by environmental conditions.

CHAPTER 109  Disorders Due to Heat and Cold  

Hypothalamic temperature

Thermal comfort and effector signal for behavior

Other deep temperatures

691.e3

Cerebral cortex

Tsk Tc

+ –

Tset

+ –

Thermal error signal Pyrogens

Integration of thermal signals

Effector signal for sweating and vasodilation

Sweat glands

Effector signal for vasoconstriction

Skin arterioles

Exercise training & heat acclimation Biological rhythms

Effector signal for heat production

Superficial veins

Skeletal muscle E-FIGURE 109-1.  Schematic diagram of human thermoregulatory control system. Tc = core temperature; Tset = set point temperature; Tsk = skin temperature. (From Sawka MN, Leon LR, Montain SJ, et al. Integrated physiological mechanisms of exercise performance, adaptation, and maladaptation to heat stress. Compr Physiol. 2011;1:1883-1928.)

692

CHAPTER 109  Disorders Due to Heat and Cold  

HEAT ILLNESS DEFINITION

Minor heat-related illnesses include miliaria rubra, heat syncope, and heat cramps. Serious heat illness represents a continuum from heat exhaustion to heat injury and heatstroke.

EPIDEMIOLOGY

Heat illness accounts for considerable morbidity and mortality in the world today. Serious heat illness is associated with a variety of individual factors, health conditions, drugs, and environmental factors (Table 109-1). Exertional heat illness is among the leading causes of death in young athletes,1 and its incidence appears to be increasing in the United States. Classic heat illness caused by high environmental temperatures remains a problem especially in homebound elderly persons without air conditioners.2 Anticholinergic and sympathomimetic poisoning (Chapter 110) can induce hyperthermia. Malignant hyperthermia (Chapter 432) is a rare disorder occurring in genetically predisposed individuals; rapid and massive skeletal muscle contraction from exposure to certain anesthetic agents, most commonly halothane and succinylcholine, can trigger core temperature elevations well above 43° C (110° F). Neuroleptic malignant syndrome (Chapter 434) is an idiosyncratic hyperthermic reaction caused by skeletal muscle rigidity from treatment with neuroleptic medications (e.g., antipsychotics, antidepressants, antiemetics). Both malignant hyperthermia and neuroleptic malignant syndrome are potentially fatal without prompt recognition and early intervention.

TABLE 109-1  FACTORS PREDISPOSING TO SERIOUS HEAT ILLNESS INDIVIDUAL FACTORS Lack of acclimatization Low physical fitness Excessive body weight Dehydration Advanced age Young age HEALTH CONDITIONS Inflammation and fever Viral infection Cardiovascular disease Diabetes mellitus Gastroenteritis Rash, sunburn, and previous burns to large areas of skin Seizures Thyroid storm Neuroleptic malignant syndrome Malignant hyperthermia Sickle cell trait Cystic fibrosis Spinal cord injury DRUGS Anticholinergic properties (atropine) Antiepileptic (topiramate) Antihistamines Glutethimide (Doriden) Phenothiazines Tricyclic antidepressants Amphetamines, cocaine, Ecstasy Ergogenic stimulants (e.g., ephedrine, ephedra) Lithium Diuretics β-Blockers Ethanol ENVIRONMENTAL FACTORS High temperature High humidity Little air motion Lack of shade Heat wave Physical exercise Heavy clothing Air pollution (nitrogen dioxide)

Heat illness can also occur in low-risk individuals who have taken appropriate precautions relative to situations to which they have been exposed in the past. Historically, such unexpected cases were attributed to dehydration (which impairs thermoregulation and increases hyperthermia and cardiovascular strain), but it is now suspected that a previous heat exposure or a concurrent event (e.g., sickness or injury) might make these individuals more susceptible to serious heat illness. One theory is that previous heat injury or illness primes the acute phase response and augments the hyperthermia of exercise, inducing unexpected serious heat illness. Another theory is that previous infection produces proinflammatory cytokines that deactivate the cells’ ability to protect against heat shock.

PATHOBIOLOGY

Body temperature can increase from a number of mechanisms: exposure to environmental heat (impeded heat dissipation); physical exercise (increased heat production); fever from systemic illness (elevated set point with subsequent activation of shivering); and medications (neuroleptic malignant syndrome and malignant hyperthermia). In addition, febrile persons have accentuated elevations in core temperature when they are exposed to high ambient temperature, physical exercise, or both. Environmental temperature and humidity, medications, and exercise heat stress in turn challenge the cardiovascular system to provide high blood flow to the skin, where blood pools in warm, compliant vessels such as those found in the extremities. When blood flow is diverted to the skin, reduced perfusion of the intestines and other viscera can result in ischemia, endotoxemia, and oxidative stress (E-Fig. 109-2). In addition, excessively high tissue temperatures (heat shock: >41° C [105.8° F]) can produce direct tissue injury; the magnitude and duration of the heat shock influence whether cells respond by adaptation (acquired thermal tolerance), injury, or death (apoptotic or necrotic). Heat shock, ischemia, and systemic inflammatory responses can result in cellular dysfunction, disseminated intravascular coagulation, and multiorgan dysfunction syndrome. In addition, reduced cerebral blood flow, combined with abnormal local metabolism and coagulopathy, can lead to dysfunction of the central nervous system.

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Minor heat illness is common and can be recognized by its clinical features. Miliaria rubra (heat rash) results from the occlusion of eccrine sweat gland ducts and can be complicated by secondary staphylococcal infection. Heat syncope (fainting) is caused by temporary circulatory insufficiency as a result of blood pooling in the peripheral veins, especially the cutaneous and lower extremity veins. Skeletal muscle cramps most commonly occur during and after intense exercise and are probably related to dehydration, loss of sodium or potassium, and neurogenic fatigue rather than to overheating itself. Serious heat illness includes heat exhaustion, which can progress to heat injury, which then can progress to heatstroke. In many patients, the degree of severity of heat illness often is not initially clear. Patients who exhibit symptoms (e.g., dizziness, unsteady gait, ataxia, headache, confusion, weakness, fatigue, nausea, vomiting, diarrhea) should have an immediate assessment of their mental status, core (rectal) temperature, and other vital signs. Until it is proved otherwise, heatstroke should be the initial working diagnosis in anyone who is a heat casualty and has an altered mental status. Heat exhaustion is defined as a syndrome of hyperthermia (temperature at time of event usually ≤40° C or 104° F) and debilitation that occur during or immediately after exertion in the heat, accompanied by no more than minor central nervous system dysfunction (headache, dizziness), which resolves rapidly with intervention. It is primarily a cardiovascular event (insufficient cardiac output) frequently accompanied by sweaty hot skin, dehydration, and collapse. Heat injury is a moderate to severe illness characterized by evidence of damage to organs (e.g., liver, renal, gut) and tissues (e.g., rhabdomyolysis) without sufficient neurologic symptoms to be diagnosed as heat stroke. It is usually associated with body temperatures above 40° C (104° F). Heatstroke is a severe illness characterized by profound mental status changes with high body temperatures, usually but not always higher than 40° C (104° F). However, patients with a core temperature higher than 40° C do not universally have a heat injury or heatstroke, and core temperatures this high can be seen transiently after stressful exercise in the heat. To establish the diagnosis of heatstroke, the entire clinical picture, including mental status and laboratory results, must be considered. Heatstroke is often categorized as classic or exertional; classic heatstroke is observed primarily in otherwise sick and compromised individuals, and exertional heatstroke is observed

CHAPTER 109  Disorders Due to Heat and Cold  

692.e1

Heat stress/exercise

Cerebral blood flow

Core temperature

Skin blood flow

Gut blood flow

Hypothalamic damage Hypothermia Recurrent hyperthermia

Ischemia Nitrosative/ Oxidative stress Vascular endothelium damage

Gut epithelial membrane permeability Endotoxin leakage

Coagulation Microvascular thrombosis Consumptive coagulation

Innate immune system

TLR4

Adaptive immune system

SIRS Cytokines Soluble cytokine receptors Multi-organ system dysfunction

E-FIGURE 109-2.  Summary of the pathophysiologic process of heat stroke that culminates in multiorgan system dysfunction and death. SIRS = systemic inflammatory response syndrome; TLR4 = toll-like receptor 4. (From Leon LR, Helwig BG. Heat stroke: role of the systemic inflammatory response. J Appl Physiol (1985). 2010;109:1980-1988.)

CHAPTER 109  Disorders Due to Heat and Cold  

TABLE 109-2  COMPARISON OF CLASSIC AND EXERTIONAL HEATSTROKE PATIENT CHARACTERISTICS

CLASSIC

EXERTIONAL

Age

Young children or elderly

15-55 years

Health

Chronic illness

Usually healthy

Fever

Unusual

Common

Prevailing weather

Frequent in heat waves

Variable

Activity

Sedentary

Strenuous exercise

Drug use

Diuretics, antidepressants, anticholinergics, phenothiazines

Ergogenic stimulants or cocaine

Sweating

Often absent

Common

Acid-base disturbances

Respiratory alkalosis

Lactic acidosis

Acute renal failure

Uncommon

Common (≈15%)

Rhabdomyolysis

Uncommon

Common (≈25%)

CK

Mildly elevated

Markedly elevated (500-1000 U/L)

ALT, AST

Mildly elevated

Markedly elevated

Hyperkalemia

Uncommon

Common

Hypocalcemia

Uncommon

Common

DIC

Mild

Marked

Hypoglycemia

Uncommon

Common

ALT = alanine aminotransferase; AST = aspartate aminotransferase; CK = creatine kinase; DIC = disseminated intravascular coagulation.

primarily in apparently healthy and physically fit individuals during or after vigorous exercise (Table 109-2). In heatstroke, neuropsychiatric impairments (e.g., marked confusion, disorientation, combativeness, and seizures) develop early and universally3 but are readily reversible with early cooling. In addition, heatstroke can be complicated by liver damage, rhabdomyolysis, disseminated intravascular coagulation, water and electrolyte imbalance, and renal failure. In fulminant heat stroke, patients have the full spectrum of abnormalities associated with the systemic inflammatory response syndrome (Chapter 108).

693

TABLE 109-3  MANAGEMENT OF HEAT ILLNESS HEAT EXHAUSTION Rest and shade Loosen and remove clothing Supine position and elevate legs Actively cool skin Fluids by mouth Monitor core temperature Monitor mental status HYPERTHERMIA Protect the airway Insert at least two large-bore intravenous lines Monitor core temperature; options include rectal, pulmonary artery, esophageal probe Actively cool the skin until core temperature reaches 20%§

Cholinesterase|   Serum (butyrylcholinesterase)   Red blood cell (acetylcholinesterase)

3100-6500 U/L 26.7-49.2 U/g of hemoglobin

80 mg/dL**

Ethylene glycol

None measured

>25 mg/dL

Iron

50-175 µg/dL

>350 g/dL

Lead

25 g/dL

Lithium

0.6-1.2 mEq/L

>1.2 mEq/L††

Methanol

None measured

>25 mg/dL

Methemoglobin

1-2%

>15%

Phenobarbital

15-40 µg/mL

>40 g/mL

Phenytoin

10-20 µg/mL

>20 g/mL

Salicylates

≤30 mg/dL

>30 mg/dL

Theophylline

8-20 µg/mL

>20 g/mL

Valproic acid

50-100 µg/mL

>100 g/mL

Normal

SOURCE: URINE

Toxic*

Arsenic

< 50 µg/day

>100 g/24-hr urine††

Mercury

< 20 µg/L

>20 g/L††

Thallium

< 5.0 µg/L

>200 g/L††

*The “toxic” level is provided for perspective. For many toxicants, simply being above this value does not imply a specific need for therapy or a necessarily poor prognosis. It does, however, suggest a need for additional evaluation, observation, or monitoring. † False-positive levels of 16 to 28 µg/mL have been reported in patients with bilirubin levels greater than 17 mg/dL. ‡ Levels drawn more than 4 hours after ingestion should be plotted on the nomogram provided by Rumack and Matthew (Rumack BH, Matthew H. Acetaminophen poisoning and toxicity. Pediatrics. 1975;55:871-876) to assess the potential for toxicity. § Lower levels may be toxic in pregnant patients and in those with prolonged exposure to carbon monoxide. | Consult a reference laboratory for normal values; results are assay dependent. ¶ Some patients may require levels above the therapeutic range to control symptoms. **The value of 80 mg/dL for ethanol is the statutory limit for operating a motor vehicle. Toxic clinical effects are uncommon with concentrations below 200 mg/dL. †† Lower values may indicate toxicity if appropriate clinical findings are present.

ingested, and time required to produce toxic effects) is more important than a specific level in determining the need for treatment. Occult acetaminophen ingestion with toxic serum levels occurs in 0.3 to 1.9% of intentional ingestions. Given that these patients may be asymptomatic until hepatotoxicity develops and that administration of an antidote can prevent this hepatotoxicity, the current recommendation is to test the serum for acetaminophen in all patients with intentional self-harm ingestions.

(Amanita smithiana and Cortinarius sp), paraquat, radiocontrast agents, solvents (e.g., carbon tetrachloride, trichloroethylene, tetrachloroethylene, toluene), and sulfonamides. Agents that decrease glomerular perfusion by reducing renal blood flow include amphotericin, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, cocaine, cyclosporine, mannitol (excessive chronic doses), methotrexate, and nonsteroidal anti-inflammatory drugs.

Other Blood Tests

A computed tomographic scan of the head can detect life-threatening cerebral edema secondary to toxicant-induced hepatic failure, ethylene glycol, and methanol. It also detects intracranial bleeding caused by anticoagulants, scorpion venom, and sympathomimetics (e.g., amphetamines, cocaine, phenylpropanolamine). An abdominal radiograph can reveal radiopaque ferrous sulfate tablets or metals such as arsenic, lead, mercury, and thallium.

Anion gap metabolic acidosis resulting from primary lactic acidosis can be caused by cyanide, hydrogen sulfide, iron, isoniazid, metformin, nucleoside reverse transcriptase inhibitors, phenformin, sodium azide, and, rarely, acetaminophen with high serum levels. Anion gap metabolic acidosis not related to lactic acidosis occurs with diethylene glycol, ethylene glycol, nonsteroidal antiinflammatory drugs, methanol, salicylates, and toluene. In poisonings resulting from ibuprofen, methanol, propylene glycol, and salicylates, lactic acid can also be produced, but the level is insufficient to account for the anion gap. Anion gap metabolic acidosis can also develop in patients with ongoing agitation, hyperthermia, and muscle rigidity, such as in neuroleptic malignant syndrome (Chapter 434), or in some cases of rhabdomyolysis (Chapter 113) secondary to toxicants such as doxylamine, phencyclidine, strychnine, cocaine, and amphetamines. Elevated serum creatinine and blood urea nitrogen levels indicative of declining renal function may be seen with numerous toxicants. Direct toxicity occurs with acetaminophen, aminoglycosides, cadmium, Chinese weight-loss botanicals (containing Stephania tetrandra or Magnolia officinalis), chromium, Crotalus durissus venom, diethylene glycol, diquat, ethylene glycol, fluorinated anesthetics, gold, heroin, lithium (diabetes insipidus), mercury salts, mushrooms

Imaging

Diagnostic Syndromes

Given the myriad combinations of signs, symptoms, and laboratory findings, making the correct diagnosis in a noncommunicative patient can be daunting. A thorough history from bystanders, friends, and prehospital medical personnel may yield crucial information. In addition, the diagnostic possibilities can be narrowed by findings that can narrow the differential diagnosis with modest certainty. For example, consider a patient with sudden loss of consciousness, anion gap metabolic acidosis, and bradycardia without hypoxemia. Among the possible causes of anion gap metabolic acidosis (see earlier) and sudden loss of consciousness are hydrogen sulfide, cyanide, and severe poisoning with sodium azide; however, sinus bradycardia in the absence of acute ischemic cardiac injury is typical only of cyanide poisoning.

704

CHAPTER 110  Acute Poisoning  

TREATMENT  Initial Stabilization

Intubation and Respiratory Support

Appropriate airway management should be instituted to correct hypoxemia and respiratory acidosis and to protect against pulmonary aspiration (Fig. 110-2); intubation should be considered if the patient has depressed consciousness and a decreased gag reflex. Rapid-sequence intubation facilitates airway management. Anatomic difficulties should be anticipated in patients with caustic ingestions (e.g., hypopharyngeal burns that may perforate); angioedema caused by angiotensin-converting enzyme inhibitor therapy or envenomation by some rattlesnakes, such as the canebrake (Crotalus horridus atricaudatus) and eastern diamondback (Crotalus adamanteus; Chapter 112); and swelling secondary to direct tissue injury (e.g., huffing compressed hydrocarbons, smoking crack) or secondary to anaphylactoid and anaphylactic

reactions. Endotracheal intubation by flexible fiberoptic nasopharyngoscopy may be indicated in these cases. Hypoxemia can occur with toxicants that produce CNS depression, such as antidepressants, barbiturates, sedativehypnotics, and central α2-adrenergic receptor agonists (clonidine), or agents causing peripheral neuromuscular impairment, such as nicotine, organophosphorus compounds, strychnine, tetrodotoxin (puffer fish, blue-ringed octopus), botulinum, or envenomation from elapids (coral snake), Mojave rattlesnakes, or certain coelenterates (box jellyfish; Chapter 112). Respiratory acidosis can rapidly worsen the toxicities of cyclic antidepressants and salicylates; sedation of these patients should be accompanied by immediate airway support. Intoxicated patients may have an increased risk for pulmonary aspiration because of concomitant CNS depression, attenuated airway reflexes, full stomachs, and delayed gastric emptying. Succinylcholine can cause prolonged paralysis in patients with organophosphorus poisoning and can exacerbate hyperkalemia from cardioactive  steroids (e.g., digoxin), hydrofluoric acid, or rhabdomyolysis (Chapter 113). 

Patient stable? No

Yes

Assess airway Intubate to correct or avoid: Hypoxemia Respiratory acidosis Pulmonary aspiration Initiate ALS Modifications: Atropine: often ineffective for bradycardia due to BARAs, CCAs, cardiac glycosides Benzodiazepines: cocaine-induced tachycardia Calcium: CCAs, HF, hypermagnesemia Glucagon: BARAs, CCAs Digoxin-specific Fab: cardiac glycosides High-dose insulin-glucose: BARAs, CCAs Nitroprusside: drug-induced hypertension NaHCO3: myocardial sodium-channel blockers Phentolamine: reverses cocaine-induced α-adrenergic agonism Avoid BARAs: in cocaine-induced ischemia

Decontamination can be performed simultaneously with stabilization therapies

Administer antidote Indicated for specific toxins

Correct hypovolemia Initiate/continue vasopressors Consider circulatory assist, e.g., balloon pump, heart-lung bypass

No

Patient unstable Continue resuscitation

Hemodynamic instability may prevent use of extracorporeal modalities

Patient hemodynamically stable?

Decontaminate Oral 1. AC 1g/kg (maximum 100 g) Indications: Toxin with potential for serious toxicity Toxin adsorbs to AC Contraindications: Nonprotected airway Bowel obstruction/perforation Ingestion of pure aliphatic hydrocarbon or caustics 2. Gastric emptying (large-bore orogastric tube lavage; nasogastric tube aspiration of liquid toxin) Indications: Toxins nonadsorbent to AC and with potential for consequential toxicity; ideally performed ≤1 hour post-ingestion Contraindications: Same as for AC; also ingestion of sharp objects or presence of bleeding diathesis 3. Other: Whole bowel irrigation with PEG Surgical removal of drug packets Dermal Wash with soap and water Ocular Irrigate with NS

Yes

Patient stable

Consider use of MDAC for toxins with known or potential enhanced elimination Indications: Definite—carbamazepine, dapsone, phenobarbital, quinine, salicylates, theophylline Potential—amitriptyline, dextropropoxyphene, digitoxin, digoxin, disopyramide, nadolol, phenylbutazone, phenytoin, piroxicam, sotalol Contraindications: Same as for single-dose AC

Is toxin eliminated by kidneys?

Yes

Consider urinary alkalinization Indications: Chlorpropamide, 2,4-dichlorphenoxyacetic acid, formic acid, methotrexate, phenobarbital, salicylates Contraindications: Volume overload, pulmonary or cerebral edema

Is toxin removed by extracorporeal device?

Yes

Institute appropriate extracorporeal modality (see Table 110-7)

FIGURE 110-2.  Algorithm for the management of acute poisoning. AC = activated charcoal; ALS = advanced life support; BARAs = β-adrenergic receptor antagonists; CCAs = L-type calcium-channel antagonists; HF = hydrofluoric acid; MDAC = multidose activated charcoal; NS = 0.9% saline solution; PEG = nonabsorbable polyethylene glycol solution.

CHAPTER 110  Acute Poisoning  

Rhabdomyolysis has been reported with adrenergic agents, doxylamine, phencyclidine, heroin, Tricholoma equestre mushrooms, and envenomation by crotaline snakes, scorpions, or widow spiders (Latrodectus sp); short-acting nondepolarizing agents, such as vecuronium and rocuronium, are preferable in these cases.

Advanced Life Support

Standard emergency cardiovascular care algorithms (Chapter 63) must be modified for effects caused by specific poisons. Atropine often does not reverse bradycardia secondary to β-adrenergic receptor antagonists, L-type calciumchannel antagonists, or cardiac glycosides, and it may actually impair the ability to do adequate gastrointestinal decontamination. In these cases, more specific therapy with intravenous calcium (calcium-channel antagonists), high doses of glucagon (β-adrenergic receptor antagonists, calcium-channel antagonists), or digoxin-specific Fab antibody (cardiac glycosides) is indicated. High-dose insulin-glucose therapy can successfully reverse myocardial depression and conduction abnormalities in humans poisoned with β-adrenergic receptor antagonists and calcium-channel antagonists. Intravenous sodium bicarbonate may reverse cardiac conduction delays caused by antiarrhythmic drugs with sodium-channel blockade recovery rates of greater than 1 second (VaughnWilliams classification IA and IC), cocaine, cyclic antidepressants, diphenhydramine, and quinine. β-Adrenergic receptor antagonists are contraindicated in patients with cocaine-induced myocardial syndromes because they can result in unopposed α-adrenergic–mediated vasoconstriction, but phentolamine can reverse the agonistic effects of cocaine on α-adrenergic receptors. Benzodiazepines can reverse significant sinus tachycardia from sympathomimetic agents. Calcium may also be life-saving in systemic hydrofluoric acid poisoning and severe hypermagnesemia, and it is indicated for symptomatic hypocalcemia caused by ethylene glycol toxicity. Drug-induced hypertension may be transitory; nitroprusside should be used if treatment is clinically indicated. In patients with toxicant-induced circulatory collapse refractory to maximal therapy, including vasopressors, circulatory assist devices may support the patient until sufficient toxicant is eliminated (Chapter 107).

Decontamination Activated Charcoal

Single-dose activated charcoal without prior gastric emptying has been the preferred method of treatment for the ingestion of substances that have the potential to cause moderate to life-threatening toxicity and are known to adsorb to activated charcoal. The absence of clinical signs and symptoms does not preclude administration of activated charcoal because drug absorption and toxicity can be delayed. Activated charcoal can also be administered when the ingested toxicant cannot be identified but significant toxicity is a concern. Activated charcoal consists of pyrolysis products that have been specially cleaned to produce an internal pore structure to which substances can adsorb, thereby limiting their systemic absorption. Activated charcoal can be administered with antiemetic drugs or given through a nasogastric tube, when necessary. The oral dose is approximately 1 g/kg body weight, with a maximum single dose of 100 g. Efficacy in preventing toxicant absorption declines with time, so activated charcoal should be given as soon as possible after ingestion. However, the documented efficacy of activated charcoal for reducing toxicant blood levels has not translated into reduced mortality in reports4 or in randomized trials. A1  A2  The decision to administer activated charcoal should be based on a risk/benefit assessment that includes nature of the exposure, clinical effects displayed during evaluation, and abilities of the medical facility and staff. For patients likely to have a good outcome, the risk and effort associated with activated charcoal administration are not worthwhile. Its use is justified in patients who present early (1 to 2 hours) after exposures to a large amount of a concerning toxin that is likely to be adsorbed to charcoal. Activated charcoal should not be used in patients at risk for aspiration until the airway is secure to minimize aspiration; the patient’s head should also be elevated unless it is contraindicated. Activated charcoal is contraindicated in patients with a perforated bowel, functional or mechanical bowel obstruction, ingestion of a pure aliphatic hydrocarbon such as gasoline or kerosene (no benefit and increased risk for aspiration), and ingestion of caustic acid and alkali (no benefit and obscures endoscopy). Certain agents, such as lithium, iron, metals, and ethanol, do not adsorb significantly to activated charcoal, but its use is not precluded if the patient has ingested other toxicants that do adsorb to activated charcoal. Pulmonary aspiration and bowel obstruction from inspissated activated charcoal are the most common complications; both occur more frequently when multidose activated charcoal is administered, but they can be avoided by withholding treatment in patients who have suboptimal bowel function or decreased fecal elimination. ,

Gastric Emptying

Two methods of gastric emptying, syrup of ipecac5 and orogastric lavage through a large-bore tube, are no longer routinely used. Both are relatively ineffective therapies that potentially increase the risk for aspiration. No welldesigned study has documented any benefit of gastric emptying, either by lavage or by syrup of ipecac, compared with the use of activated charcoal alone. Gastric emptying by lavage or, rarely, by syrup of ipecac may be of benefit and should be performed in patients who have ingested toxicants that

705

do not adsorb to activated charcoal and are known to produce significant morbidity or for which aggressive decontamination may offer the best chance for survival (e.g., colchicine, sodium azide, sodium fluoroacetate). Removal of a liquid toxicant, such as ethylene glycol, may be accomplished by aspiration of gastric contents through a nasogastric tube. Contraindications to gastric emptying include those for activated charcoal, a bleeding diathesis, and the ingestion of sharp objects. Placement of an endotracheal tube before gastric lavage may be necessary to protect the airway in patients who have a decreased level of consciousness and impaired gag reflex but is not required in all cases. Major complications of gastric emptying include pulmonary aspiration, esophageal tears and perforations, and laryngospasm (with lavage).

Whole Bowel Irrigation

Whole bowel irrigation with a nonabsorbable polyethylene glycol solution has been recommended for iron and sustained-release medications, for agents not adsorbed to activated charcoal, and for body packers (smugglers who swallow packets of illicit drugs). The most common complication is vomiting, and whole bowel irrigation is contraindicated in patients with bowel perforation, obstruction, hemorrhage, or hemodynamic or respiratory instability.  The initial recommended dose is 500 mL/hour given orally or by nasogastric tube, with titration to 2000 mL/hour as tolerated; treatment continues until the rectal effluent clears. Rarely, surgery may be necessary to remove packets in smugglers who have symptoms of cocaine toxicity or are obstructed; endoscopic removal of these packets should never be attempted because of the risk of packet rupture.

Antidotes

Few toxicants have specific therapies (Table 110-6). Although antidotes may be essential in treating patients exposed to certain toxicants, their use does not preclude the need for ongoing supportive care and, in some cases, extracorporeal elimination.

Enhanced Elimination

Methods to accelerate the elimination of toxicants or drugs from the body include multiple doses of activated charcoal, urinary alkalinization, and extracorporeal removal. Another method, using the oral ion exchange resins sodium polystyrene sulfonate and cholestyramine, has experimentally enhanced the elimination of lithium, digoxin, digitoxin, and organochlorines but has limited clinical usefulness.

Multiple Doses of Oral Activated Charcoal

The rationale for administering multiple doses of oral activated charcoal includes the adsorption of any toxic agent remaining in the gastrointestinal tract (e.g., sustained-release drugs or drugs that retard their absorption, such as anticholinergics); interference with the enterohepatic and enteroenteric recirculation of toxicants; and enhancement of the elimination of drugs with a long half-life, a volume of distribution less than 1 L/kg body weight, and low protein binding (termed gastrointestinal dialysis). The existing evidence shows enhanced elimination of carbamazepine, dapsone, phenobarbital, quinine, salicylates, and theophylline, but multiple doses of activated charcoal may  also be effective for amitriptyline, dextropropoxyphene, digitoxin, digoxin, disopyramide, nadolol, phenylbutazone, phenytoin, piroxicam, and sotalol. Whether enhanced elimination provided by repeated doses of activated charcoal translates into decreased morbidity and mortality has not been adequately examined in large controlled clinical trials, except for yellow oleander and organophosphate ingestion, for which it has shown no benefit. A1  The usual recommendations are an average dose of 12.5 g of activated charcoal (after the initial dose of 1 g/kg body weight, with a maximum single dose of 100 g) administered every 4 to 6 hours after the previous dose. The contraindications to single-dose activated charcoal also apply to multidose activated charcoal. Reported complications include pulmonary aspiration, bowel obstruction from inspissated charcoal, and fluid and electrolyte imbalance from multiple doses of a simultaneously administered cathartic.

Urinary Alkalinization

Alkalinization of the urine, which increases the renal elimination of weak acids, is used primarily to enhance the elimination of salicylates, but the elimination of chlorpropamide, 2,4-dichlorophenoxyacetic acid, formic acid, methotrexate, and phenobarbital may be increased with this method. Urinary alkalinization is accomplished by an intravenous bolus of 1 to 2 mEq of sodium bicarbonate per kilogram body weight, followed by three ampules (150 mL) of sodium bicarbonate (44 mEq/50 mL) in 850 mL of 5% dextrose in water infused at two to three times the normal maintenance fluid rate. Urinary pH should be checked hourly, and the infusion should be adjusted to maintain a urine pH of 7.5 to 8.0. Potassium should be administered simultaneously to avoid hypokalemia, which prevents urinary alkalinization because the distal tubule excretes hydrogen ion in exchange for potassium (Chapters 116 and 117). Serum pH should be monitored and kept at 7.55 or lower to avoid excessive alkalemia. Contraindications to this therapy include volume overload and cerebral or pulmonary edema. Urinary acidification is not recommended to enhance the elimination of weak bases, such as amphetamines, because of the danger of precipitating tubular myoglobin in patients with rhabdomyolysis.

706

CHAPTER 110  Acute Poisoning  

Extracorporeal Removal

Extracorporeal techniques enhance the elimination of a few drugs and toxicants, especially those that exist in the blood and are otherwise poorly eliminated. Such drugs generally have single-compartment kinetics, a volume  of distribution less than 1 L/kg, and endogenous clearance of less than  4 mL/minute/kg (Table 110-7). For hemodialysis, the toxicant must be water soluble, have a molecular weight less than 500, and exhibit low protein binding. For continuous renal replacement therapy, the toxicant must have a

molecular weight less than the permeability limit of the filter membrane.  As a group, these latter forms of elimination are slow and likely to be of  limited benefit to most patients with acute poisoning.6 Rarely, extracorporeal removal has been used for aminoglycosides, atenolol, bromide, carbamazepine, diethylene glycol, isopropanol, magnesium, metformin, methotrexate, N-acetylprocainamide, phenobarbital, procainamide, sotalol, and trichloroethanol (chloral hydrate).

TABLE 110-6  ANTIDOTES AND INDICATIONS FOR USE ANTIDOTE

INDICATION FOR USE

DOSE*

TREATMENT END POINT

COMMENTS

Antivenom, Crotalidae (Fab)†

Crotaline snake (e.g., 4-6 vials; repeat for persistent or Halt in progression of rattlesnakes, copperhead) worsening clinical condition; circumferential and proximal repeated doses of 2 vials at 6, 12, swelling and 18 hours after initial Resolving systemic effects antivenom dose are recommended

Antivenom, Latrodectus (equine)†

Black widow spider (Latrodectus sp)

1 vial diluted in 50-100 mL NS, infused during 1 hour; can repeat

Resolution of symptoms, vital signs Dilution and slow infusion rate are normal critical to avoid anaphylactoid reaction Indications include severe pain unresponsive to opioids and severe hypertension Serum sickness can occur IV calcium is ineffective

Atropine

Carbamates Nerve agents Organophosphorus compounds

2 mg IV; double the dose every 5 minutes to achieve atropinization and hemodynamic stability; then start continuous infusion of 10-20% of total stabilizing dose per hour

Cessation of excessive oral and pulmonary secretions, >80 beats/min, systolic blood pressure >80 mm Hg

Doubling of the dose every 5 minutes (e.g., 2 mg, 4 mg, 8 mg, 16 mg) estimated to achieve atropinization within 30 minutes Stop infusion when patient develops concerning signs or symptoms of anticholinergic toxidrome (see Table 110-1); restart infusion at lower rate when signs or symptoms abate

Calcium salt‡

Calcium-channel antagonists

Calcium chloride 10%, 10 mL (1 g) during 10 minutes; can be given in 1 minute if critically ill Calcium gluconate 10%, 30 mL (3 g) during 10 minutes; can be given in 1 minute if critically ill

Reversal of hypotension; may not reverse bradycardia

Hydrofluoric acid

Systemic toxicity: calcium gluconate 10%, 1-3 g (10-30 mL) per dose IV during 10-minute period; repeat as needed every 5-10 minutes Calcium gluconate 10%, 1 g (10 mL) per dose IV during 10-minute period; repeat as needed every 5-10 minutes Calcium gluconate 10%, 1-2 g (10-20 mL) per dose IV during 10-minute period; repeat as needed every 5-10 minutes Calcium gluconate 10%, 0.5-1.0 g (5-10 mL) per dose during 10-minute period; repeat as needed every 10 minutes

Reversal of life-threatening manifestations of hypocalcemia and hyperkalemia

All indications: Monitor ionized calcium levels IV extravasation causes tissue necrosis, especially with calcium chloride Can administer at faster than stated rates for immediate life-threatening conditions (i.e., in 1 minute) Calcium chloride contains three times more elemental calcium than calcium gluconate does Can dilute and give intra-arterially or IV with a Bier block for extremity exposures and burns

Hyperkalemia (except cardiac glycosides) Hypermagnesemia

Hypocalcemia (e.g., ethylene glycol)

L-Carnitine

Valproate-induced hyperammonemia or hepatotoxicity

Cyanide antidote kit Cyanide   Amyl nitrite   Sodium nitrite   Sodium thiosulfate [Hydroxocobalamin is preferred if available, see below]

Better safety profile than historical equine-derived antivenom Repetitive dosing indicated for recurrent soft tissue swelling Less effective at correcting hematologic (i.e., coagulation and platelet) disorders

Reversal of myocardial depression and conduction delays

May precipitate ventricular arrhythmias

Reversal of respiratory depression, hypotension, and cardiac conduction blocks

Simultaneous therapies to increase magnesium elimination should be instituted

Reversal of tetany

Correct symptomatic hypocalcemia; avoid excessive administration that may increase production of calcium oxalate crystals in ethylene glycol poisoning

100 mg/kg (maximum 6 g) IV during 30 minutes, then 15 mg/ kg IV during 30-minute period q4h (maximum 6 g/day)

Treat until clinical improvement occurs

Levocarnitine is active form Adjust dose for end-stage renal disease

Amyl nitrite: 0.3-mL pearls, crush and inhale during 30-second period Sodium nitrite 3%: 10 mL IV during 10-minute period Sodium thiosulfate 25%: 50 mL (12.5 g) IV during 10-minute period

Resolution of lactic acidosis and moderate to severe clinical signs and symptoms: seizures, coma, dyspnea, apnea, hypotension, bradycardia

Coordinate amyl nitrite with continued oxygenation and give only until sodium nitrite infusion is begun; nitrites may produce hypotension and excess methemoglobinemia Sodium nitrite dose must be adjusted if patient has hemoglobin 350 g/dL Prolonged therapy can cause pulmonary toxicity

Digoxin-specific antibody Digoxin fragments (Fab) Digitalis and related plants (e.g., oleander, lily of the valley) Other cardiac glycosides (e.g., bufadienolides [Bufo toads])

Unknown digoxin dose or serum Resolution of hyperkalemia, Each vial binds 0.5 mg of digoxin or level, or for plant or toad source: symptomatic bradydysrhythmias, digitoxin acute toxicity—10-20 vials; ventricular arrhythmias, Mobitz Monitor ECG and potassium levels chronic toxicity—3-6 vials II or third-degree heart block Digoxin serum levels unreliable after Digoxin dose known: number of antidote administered unless test is vials = (mg ingested × 0.8) ÷ 0.5 specific for free serum digoxin Digoxin serum level known: number of vials = [serum level (ng/mL) × weight (kg)] ÷ 100 Infuse dose during 30 minutes

Dimercaprol (BAL)

Arsenic: 3-5 mg/kg IM q4h Lead: 75 mg/m2 (4 mg/kg) IM q4h for 5 days Inorganic mercury: 5 mg/kg IM, then 2.5 mg/kg IM q12h for 10 days or until patient is clinically improved

Arsenic Lead Mercury, elemental and inorganic salts

Arsenic: 24-hour urinary arsenic 12 hours after the onset of symptoms) can convert a closed injury into an open wound and thereby increase the risk of uncontrollable infection, late fasciotomy is relatively contraindicated.

Management of Crush Injury

For victims of crush injury (Chapter 111), aggressive on-site hydration with intravenous normal saline is recommended. For massive damage, amputation of the extremity may be required to protect the patient’s overall health. The Mangled Extremity Severity Score can identify nonsalvageable extremities  on the basis of the degree of skeletal and soft tissue injury, the patient’s  blood pressure, the presence of a detectable pulse, and age (see http:// www.mdcalc.com/mangled-extremity-severity-score-mess-score/).

Malignant Hyperthermia

Rhabdomyolysis caused by malignant hyperthermia (Chapter 432) requires rapid diagnosis and aggressive management. Anesthetics should be discontinued, and the patient should be treated with dantrolene sodium, 2.5 to 4 mg/kg intravenously, followed by about 1 mg/kg every 4 hours for up to 48 hours to avoid recrudescence.

PROGNOSIS

Hydration

Hydration is the cornerstone of preserving kidney function in patients with rhabdomyolysis, and delay of fluid administration for more than 6 hours increases the risk for acute kidney injury. Inpatient hydration is indicated for victims of collapse, trauma, or exertional heat injury as well as for patients who have moderate early symptoms, more than mild elevations in CK, or abnormal serum levels of creatinine, potassium, calcium, phosphate, or bicarbonate. In adults, the target urine output is 300 mL/hr for at least 24 hours to prevent acute kidney injury. Hydration is accomplished by the aggressive administration of isotonic intravenous fluids at a rate that results in a urine output of 200 to 300 mL/hour until CK levels begin to decline. If fluid resuscitation fails to correct intractable hyperkalemia and acidosis, renal replacement therapy should be considered (Chapter 120). By comparison, adults with mild symptoms and serum CK levels less than 3000 U/L are considered to be at low risk and may be treated as outpatients with vigorous oral hydration, limited physical activity, and careful follow-up.

Specific Therapeutic Measures

Alkalinization of the urine decreases cast formation, minimizes the toxic effects of myoglobin on the renal tubules, inhibits lipid peroxidation, and decreases the risk for hyperkalemia. However, bicarbonate therapy can cause calcium to precipitate in the soft tissues and contribute to a hyperosmolar state. Mannitol is an osmotic diuretic, volume expander, and free radical scavenger, but it should be used only after adequate kidney function is established and must be used with great caution in patients with marginal cardiac function. To date, no convincing evidence demonstrates that adding sodium bicarbonate or mannitol is superior to fluid therapy alone.8 Sodium bicarbonate should be used only in patients with evidence of systemic acidosis, and mannitol should be used only when it is needed to maintain a urine output of 300 mL/hour. Deposition of calcium, which occurs early in rhabdomyolysis, is directly related to the degree of muscle destruction and to the administration of

The most serious consequence of rhabdomyolysis is acute kidney injury, which occurs in up to 67% of all cases, regardless of cause. Predictors of the risk of needing renal replacement therapy or death in patients with rhabdomyolysis include age older than 50 years; initial serum creatinine level of 1.4 mg/dL or higher; initial serum calcium level below 7.5 mg/dL; initial serum phosphate level above 4.0 mg/dL; initial serum bicarbonate level below 19 mEq/L; and a cause other than syncope, seizures, exercise, statins, or myositis.11 Event rates range from 0% in patients with none of these criteria to 20% or more in patients with four or more criteria. For compartment syndrome, a poor prognosis is associated with an ischemic period lasting longer than 6 hours. The prognosis of patients with rhabdomyolysis improves markedly when treatment is started soon after the diagnosis is made. With mild episodes, the prognosis is customarily excellent, and the patient can typically resume usual activities within several weeks after CK levels have normalized. However, some patients do not return to normal and continue to experience extreme fatigue and muscle pain on exertion. These patients require additional testing (nonischemic forearm test, electromyography, muscle disease enzyme panel, muscle biopsy; Chapter 421) to determine whether an underlying metabolic myopathy exists. The results of these tests will help determine future recommendations, but the patient’s tolerance and response to light and more strenuous exercise are important factors. Most authorities agree that statin-induced rhabdomyolysis is an indication for discontinuation of their use.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 113  Rhabdomyolysis  

GENERAL REFERENCES 1. Zimmerman JL, Shen MC. Rhabdomyolysis. Chest. 2013;144:1058-1065. 2. Update: Exertional rhabdomyolysis, active component, U.S. Armed Forces 2008-2012. MSMR. 2013;120:21-24. 3. Ross EA, Reisfield GM, Watson MC, et al. Psychoactive “bath salts” intoxication with methylenedioxypyrovalerone. Am J Med. 2012;125:854-858. 4. Auer J, Sinzinger H, Franklin B, et al. Muscle- and skeletal-related side-effects of statins: tip of the iceberg? Eur J Prev Cardiol. 2014. [Epub ahead of print] 5. Szczepanik ME, Heled Y, Capacchione J, et al. Exertional rhabdomyolysis: identification and evaluation of the athlete at risk for recurrence. Curr Sports Med Rep. 2014;13:113-119.

726.e1

6. Pasnoor M, Barohn RJ, Dimachkie MM. Toxic myopathies. Neurol Clin. 2014;32:647-670. 7. Deuster PA, Contreras-Sesvold CL, O’Connor FG, et al. Genetic polymorphisms associated with exertional rhabdomyolysis. Eur J Appl Physiol. 2013;113:1997-2004. 8. Scharman EJ, Troutman WG. Prevention of kidney injury following rhabdomyolysis: a systematic review. Ann Pharmacother. 2013;47:90-105. 9. Bosch X, Poch E, Grau JM. Rhabdomyolysis and acute kidney injury. N Engl J Med. 2009;361:62-72. 10. Graziani G, Calvetta A, Cucchiari D, et al. Life-threatening hypercalcemia in patients with rhabdomyolysis-induced oliguric acute renal failure. J Nephrol. 2011;24:128-131. 11. McMahon GM, Zeng X, Waikar SS. A risk prediction score for kidney failure or mortality in rhabdomyolysis. JAMA Intern Med. 2013;173:1821-1827.

726.e2

CHAPTER 113  Rhabdomyolysis  

REVIEW QUESTIONS 1. During the management of a patient who has multiple traumatic injuries, you note that a previously declining serum creatine kinase level now happens to rise. The patient’s urine output is stable on aggressive hydration, and vital signs are normal. Which of the following represents the most likely clinical scenario accounting for the increasing creatine kinase? A. Infection B. Compartment syndrome C. Medication toxicity D. Underlying metabolic myopathy E. Acute respiratory failure Answer: B  In the management of the trauma patient with rhabdomyolysis, serum creatine kinase levels should steadily decline after a peak has been achieved, which typically is 3 to 5 days into effective treatment with aggressive fluid hydration. When the serum creatine kinase level increases, the clinician should suspect a compartment syndrome. Excessive pain is typically a clinical clue, but the use of regional anesthetic blocks or strong analgesics may mask the symptoms. 2. Rhabdomyolysis can be a complication of crush injury in people trapped at the site of a collapsed building. Which of the following field site interventions is most prudent to preserve both life and limb? A. Extremity hypothermia protocol B. Hypotensive resuscitation C. Aggressive fluid hydration D. Low-molecular-weight dextran to increase viscosity E. Extremity hyperthermia protocol Answer: C  Crush injury can cause both early and delayed deaths. The principal culprit for delayed morbidity and mortality is acute renal failure with associated metabolic complications. Early intravenous hydration, preferably within the first 6 hours, should aim to achieve a urine output in adults of 300 mL/hr to prevent acute renal failure.

3. Rhabdomyolysis can be a potentially life-threatening illness, with acute, subacute, and chronic complications, including compartment syndrome and acute kidney injury. Which of the following diagnostic tests is not necessary in the acute management of a significant case of rhabdomyolysis? A. Electrocardiogram B. Muscle biopsy C. Metabolic panel D. Urinalysis E. Measurement of creatine kinase levels Answer: B  In the management of rhabdomyolysis, the immediate focus is on treating metabolic complications and preventing kidney failure. Accordingly, a metabolic panel, measurement of creatine kinase levels, electrocardiogram, and urinalysis are all valuable assessments in detecting abnormalities and assisting in ongoing management. A muscle biopsy is not indicated in the acute management of the patient with rhabdomyolysis, but it can assist in determining whether a patient with recurrent rhabdomyolysis has an underlying inherited and acquired myopathy. 4. A serum creatine kinase level is commonly obtained to aid in both diagnosis and management of the patient with rhabdomyolysis. Which of the following characteristics are important to understand the baseline creatine kinase level in an individual patient? A. Sex B. Race C. Activity level D. A and B E. A, B, and C Answer: E  The difficulty in using serum creatine kinase levels alone to make the diagnosis of mild rhabdomyolysis partly lies in the wide variability of baseline levels. Men, African Americans, and more active people have higher baseline serum creatine kinase levels.

728

CHAPTER 114  Approach to the Patient with Renal Disease  

114  APPROACH TO THE PATIENT WITH RENAL DISEASE DONALD W. LANDRY AND HASAN BAZARI

The prominent functions of the kidney include the excretion of nitrogenous waste; the regulated excretion of water, sodium, potassium, and acid; and the synthesis of a variety of hormones, including 1,25-dihydroxyvitamin D, erythropoietin, and renin. The kidney’s elaboration of a protein-free and cell-free ultrafiltrate is uniquely responsible for the excretion of nitrogenous wastes. The approach to the patient with renal disease is largely focused on disordered ultrafiltration and not on defects in the isolated renal tubular processing of individual ions, water, or acids. A patient may, for example, present with an isolated defect in renal acid excretion (Chapter 118), but in this case the “approach to the patient” is framed for the evaluation of metabolic acidosis, an abnormality for which the kidney is only one among the many causes in a broad differential diagnosis. In contrast, acute kidney injury (Chapter 120) and chronic kidney disease (Chapter 130) refer specifically and exclusively to defects in the filtration function of the kidney. In the context of a diminished magnitude of filtration, many of the other individual functions of the kidney (e.g., hormone synthesis, electrolyte homeostasis) may fail as well. Primary diseases of the tubules, such as acute tubular necrosis (Chapter 120) and tubulointerstitial disease (Chapter 122), also impair the rate of glomerular filtration and cause acute kidney injury and chronic kidney disease. In contrast, an impairment in ultrafiltration may, in early chronic kidney disease, be reflected solely in a decreased quality of glomerular filtration (e.g., the presence of albuminuria) rather than in a decreased quantity of filtrate with increased concentrations of nitrogenous waste. Similarly, the glomerular filtration rate (GFR) may be normal in nephrotic syndrome despite ultrafiltration defects that result in massive proteinuria. Defects in the filter can also allow passage of cells, such as red blood cells (RBCs), as is seen in the acute nephritic syndrome (Chapter 121), with or without heavy proteinuria. The paradox of glomerular hematuria without albuminuria is also possible. For example, in mild forms of immunoglobulin A (IgA) nephropathy (Chapter 121), relatively few defects in the glomerular filter will permit a detectable number of RBCs per high-power field in the urine despite a urine albumin level that still remains within normal limits (Fig. 114-1). In this context, this chapter considers the approach to the patient with acute kidney injury, glomerular syndromes (nephrotic vs. nephritic), tubulointerstitial disease, vasculitis and vascular diseases of the kidney, papillary necrosis, and chronic kidney disease.

PATHOBIOLOGY

The approximately 2 million renal glomeruli normally filter about 180 L/day. The renal glomerulus is not simply a filter but rather a size- and chargedependent ultrafilter that excludes not only cells but also proteins larger than 60 kD from the ultrafiltrate. Smaller proteins are variably filtered at the glomerulus and endocytosed in the proximal tubule so that the protein concentration of the urine is normally low. Kidney disease reflects a failure in the quantity or quality of the glomerular ultrafiltrate. The normal GFR may decline in hours to days in acute kidney injury or during months to years in chronic kidney disease. An acute decline in glomerular filtration is the necessary and sufficient condition for the diagnosis of acute kidney injury, but abnormal urinary findings can assist with elucidating the etiology of the injury. Proteinuria, ranging from microscopic to nephrotic range (Chapter 121), and urinary findings, from a few cells per microscopic high-power field to gross hematuria or pyuria, may be the only evidence of the earliest stages of chronic kidney disease. As chronic kidney disease advances, the decline in the GFR progresses until dialysis or transplantation (Chapter 131) is required to forestall or to treat the syndrome of uremia.

DIAGNOSIS

Measuring Kidney Function

Although the most accurate method of evaluating kidney function is a formal measurement of GFR with iothalamate, iohexol, or similar markers, these

tests are too expensive and time-consuming to be recommended for routine clinical practice. Currently, the most common methods used to estimate GFR are the serum creatinine concentration, the calculated creatinine clearance, and estimation equations based on serum creatinine.1 Serum creatinine is, to a first approximation, neither secreted nor reabsorbed, so the amount appearing in the urine per unit time is a measure of the amount that was filtered at the glomerulus during that period. As a result, the rate of creatinine clearance is a reasonably close estimate of the GFR. A decrement in the GFR diminishes creatinine clearance but has no immediate effect on creatinine production by muscle; as a result, the serum creatinine concentration rises. The change in serum creatinine over time indicates the tempo of the renal disease and can distinguish acute injury from chronic kidney disease. Problems with the routine use of serum creatinine alone to infer GFR stem from the differing rates of creatinine production among individuals, mainly because of variations in muscle mass. Women and the elderly can have deceptively low serum creatinine levels despite significant declines in GFR.2 In addition, the shape of the curve relating the GFR to serum creatinine (Fig. 114-2) has an important and potentially easily overlooked clinical implication, namely, that an initial small absolute rise in creatinine usually reflects a marked fall in GFR. Creatinine clearance can be calculated with a 24-hour urine collection to measure the creatinine concentration. The patient must be instructed to discard the first morning urine before initiating the collection and to conclude the collection by including the next morning void. The formula for calculating creatinine clearance is as follows: CCr = (urine Cr × V )/(plasma Cr) where CCr is creatinine clearance, urine Cr is urine creatinine concentration, V is urine flow rate, and plasma Cr is plasma creatinine. The creatinine clearance overestimates GFR by about 10% owing to tubular secretion of creatinine. Calculation of creatinine clearance from a 24-hour urine collection can be cumbersome for patients and is prone to error because of inaccurate urine collection. Because of the logistical and practical limitations of a 24-hour urine collection, several equations have been developed to estimate GFR on the basis of easily obtainable clinical data and laboratory results. To date, the most widely used equations are the co*ckcroft-Gault, the Modification of Diet in Renal Disease (MDRD) Study, and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations (Table 114-1). Weight estimations or ideal weight estimations can make calculation and reporting of co*ckcroftGault results problematic. The MDRD equations (both the full and abbreviated forms) use data that are readily available to laboratories, but the equations systematically underestimate GFR at higher serum creatinine values, thereby raising concern for false diagnoses of chronic kidney disease. The CKD-EPI equation appears to be more precise and accurate than the MDRD equation, especially at higher GFRs.3 Cystatin C may provide a more accurate and prognostic measurement of GFR in patients whose creatinine levels are in the upper end of the normal range, but it does not replace estimated GFR measurements for most clinical purposes.4,5

Urinalysis

The normal color of the urine is derived from urochromes, which are pigments excreted in the urine. Abnormal color or appearance of the urine may be explained by many conditions (Table 114-2). The basic analysis of the urine sample involves measurements with commercially available dipsticks or microscopic analyses.

Urine Dipstick

The specific gravity of the urine generally is related linearly with osmolality. However, it can be raised by the presence of molecules with relatively high molecular weight, such as glucose or contrast dye. A fixed specific gravity of 1.010, so-called isosthenuria, is characteristic of chronic kidney disease (Chapter 130). Urine pH typically is 5 as a result of daily net acid excretion. An alkaline pH often is noted after meals, when an “alkaline tide” to balance gastric acid excretion increases urine pH. A high urine pH also is seen in patients who are on a vegetarian diet. An exceptionally high urine pH is indicative of an infection with a urea-splitting organism, such as Proteus species (Chapter 284). An inappropriately high urine pH in the setting of systemic non–anion gap metabolic acidosis may be seen in certain forms of renal tubular acidosis (Chapter 118). In a proximal renal tubular acidosis, the urine pH is high until the tubular reabsorption threshold for bicarbonate, which is abnormally low,

CHAPTER 114  Approach to the Patient with Renal Disease  

Filtration Defect

Differential Diagnosis

AKI Quantitative CKD

Scr ( GFR)

729

Diagnostic Approach

Prerenal

BUN/Cr ratio, FENa

Intrarenal

Urinalysis and sediment, serologies, biopsy

Postrenal

Imaging (US, CT)

AKI on CKD D,

Qualitative

ay

3.5 g/d

ay

KI

out A

With

Nephrotic syndrome

C ic M path Idio MN , 1° S FSG

2° - Systemic disease, medication, reflux

Chronic glomerulopathy

Serologies (ANA, SIEP, SFLC, ASLO, cryoglobulins), biopsy

4

C3/C AKI

Acute glomerulonephritis

Serologies (e.g., hepatitis B, hepatitis C, HIV, serum light chains), imaging, biopsy Genetic testing, biopsy

Hematuria With

Biopsy, antibody to M-type phospholipase A2 receptor?

Norm

al C3

/C4

Serologies (ANCA, anti-GBM, SIEP), biopsy

FIGURE 114-1.  Overview of approach to kidney disease. Quantitative defects in filtration, manifested by elevated serum creatinine (Scr) and reduced glomerular filtration rate (GFR), should lead to a query into acute kidney injury (AKI) versus chronic kidney disease (CKD). AKI, in turn, is generally divided into prerenal, postrenal, and intrinsic causes. Qualitative defects in filtration, manifested by proteinuria or hematuria, can occur in the absence of changes in GFR and often require biopsy for diagnosis. Proteinuria of more than 3.5 g/ day signals nephrotic syndrome, which may be idiopathic or secondary to systemic diseases, such as hepatitis B or C, human immunodeficiency virus (HIV) infection, or diabetes. Glomerular hematuria without AKI is consistent with a chronic glomerulopathy, such as IgA nephropathy, or familial diseases, such as thin basem*nt membranes disease. When hematuria accompanies AKI, acute glomerulonephritis should be suspected and can diagnostically be divided into low-complement glomerulonephritides (immune complex– mediated lesions such as lupus nephritis, postinfectious glomerulonephritis, and cryoglobulinemic glomerulonephritis) and normocomplementemic glomerulonephritides (classically seen in the rapidly progressive glomerulonephropathies due to antineutrophil cytoplasmic antibody [ANCA] and anti–glomerular basem*nt membrane [anti-GBM] antibody). ANA = antinuclear antibody; ASLO = antistreptolysin O; BUN = blood urea nitrogen; Cr = creatinine; CT = computed tomography; MCD = minimal change disease; FENa = fractional excretion of sodium; FSGS = focal segmental glomerulosclerosis; MN = membranous nephropathy; SFLC = serum free light chain; SIEP = serum immunoelectrophoresis; US, ultrasonography.

10

Plasma creatinine, mg/dL

8

6

4

2

0 0

40 80 120 160 Glomerular filtration rate, mL/min

FIGURE 114-2.  Relationship between plasma creatinine and glomerular filtration rate measured by inulin clearance in 171 patients (circles). The continuous line reflects the idealized relationship between these parameters if creatinine were excreted solely by glomerular filtration; the dashed line represents an upper limit of “normal” for the creatinine concentration of 1.4 mg/dL. (Redrawn from Shemesh O, Golbetz H, Kriss JP, et al. Limitations of creatinine as a filtration marker in glomerulopathic patients. Kidney Int. 1985;28:830-838.)

is reached. At this point, the urine pH decreases to 5. In distal renal tubular acidosis, the inability to create a sufficient gradient for hydrogen ions results in a urine pH that is always higher than 5.5. In type 4 renal tubular acidosis, the urine pH is often 5, and the urine net charge is often positive, thereby confirming the absence of significant amounts of ammonium in the urine; this defect is exacerbated by the accompanying hyperkalemia.

Glucose in the urine is detected by an assay using dipsticks impregnated with the enzyme glucose oxidase. Glycosuria is seen in diabetes mellitus (Chapter 229), when pregnancy causes the tubular threshold for glucose reabsorption to change, and in tubular diseases that affect the proximal convoluted tubule and cause tubular glycosuria. Evidence for pan–proximal tubular dysfunction (e.g., glycosuria, aminoaciduria, phosphaturia) indicates that Fanconi syndrome is present. The dipstick for protein is a sensitive assay based on color change induced by the presence of proteins at a given pH. It is most sensitive to the presence of albumin and is much less sensitive to other proteins, such as the light chains of Bence Jones protein (Chapter 187). The presence of 1+ protein correlates with about 30 mg/dL of albuminuria, and 3+ protein correlates with more than 500 mg/dL of proteinuria. Because the dipstick is not a quantitative measurement, small amounts of proteinuria in an oliguric patient may give the false appearance of high-grade proteinuria. The excretion of abnormal quantities of albumin below the level detectable by the urine dipstick is called microalbuminuria. Normal albumin excretion, which is less than 30 mg/day, is best detected by radioimmunoassay or enzyme immunoassay. Microalbuminuria is the earliest clinically detectable stage of diabetic nephropathy (Chapter 124). Proteinuria of increasing severity is associated with a more rapid decline in the GFR, regardless of the GFR,6 except in minimal change disease (Chapter 121). The dipstick for heme uses the peroxidase-like activity of hemoglobin and myoglobin molecules to detect the presence of heme pigment. The reaction occurs on exposure to hemoglobin, myoglobin, or intact RBCs. The presence of myoglobin, which is found in patients with rhabdomyolysis (Chapter 113), or free hemoglobin, which is seen in patients with intravascular hemolytic anemias (Chapter 160), is suspected if the heme reaction is intensely positive and there is a paucity of cellular elements in the sediment. Persistent, isolated, asymptomatic, microscopic hematuria in adolescents and young adults is associated with a nearly 20-fold increased risk of subsequent endstage renal disease.7 The dipstick detection of leukocytes depends on the presence of leukocyte esterase. Leukocyte esterase is usually present in infections (Chapter 284) and in inflammatory conditions.

730

CHAPTER 114  Approach to the Patient with Renal Disease  

TABLE 114-1  EQUATIONS FOR ESTIMATION OF GLOMERULAR FILTRATION RATE co*ckCROFT-GAULT CCr (mL/min) =

Male

CCr (mL/min) =

Female

(140 − age) × lean body wt (kg) SCr (mg/dL) × 72 (140 − age) × lean body wt (kg) × 0.85 SCr (mg/dL) × 72

MODIFICATION OF DIET IN RENAL DISEASE 1 Black male

GFR = 170 × SCr−0.999 × age−0.176 × BUN−0.170 × Albumin0.318 × 1.18

Black female

GFR = 170 × SCr−0.999 × age−0.176 × BUN−0.170 × Albumin0.318 × 1.18 × 0.762

White male

GFR = 170 × SCr−0.999 × age−0.176 × BUN−0.170 × Albumin0.318

White female

GFR = 170 × SCr−0.999 × age−0.176 × BUN−0.170 × Albumin0.318 × 0.762

MODIFICATION OF DIET IN RENAL DISEASE 2 (ABBREVIATED) Black male

GFR = 186 × SCr−1.154 × age−0.203 × 1.21

Black female

GFR = 186 × SCr−1.154 × age−0.203 × 1.21 × 0.742

White male

GFR = 186 × SCr−1.154 × age−0.203

White female

GFR = 186 × SCr−1.154 × age−0.203 × 0.742

CHRONIC KIDNEY DISEASE EPIDEMIOLOGY COLLABORATION Black male, SCr ≤0.9 mg/dL

GFR = 163 × (SCr/0.9) −0.411 × 0.993age

Black male, SCr >0.9 mg/dL

GFR = 163 × (SCr/0.9) −1.209 × 0.993age

Black female, SCr ≤0.7 mg/dL

GFR = 166 × (SCr/0.7) −0.329 × 0.993age

Black female, SCr >0.7 mg/dL

GFR = 166 × (SCr/0.7) −1.209 × 0.993age

White male, SCr ≤0.9 mg/dL

GFR = 141 × (SCr/0.9) −0.411 × 0.993age

White male, SCr >0.9 mg/dL

GFR = 141 × (SCr/0.9) −1.209 × 0.993age

White female, SCr ≤0.7 mg/dL

GFR = 144 × (SCr/0.7) −0.329 × 0.993age

White female, SCr >0.7 mg/dL

GFR = 144 × (SCr/0.7) −1.209 × 0.993age

BUN = blood urea nitrogen; GFR = glomerular filtration rate; SCr = serum creatinine.

TABLE 114-2  MACROSCOPIC APPEARANCE OF URINE APPEARANCE

CAUSE

Milky

Acid urine: urate crystals Alkaline urine: insoluble phosphates Infection: pus Spermatozoa Chyluria

Smoky pink

Hematuria (>0.54 mL blood/L urine)

Foamy

Proteinuria

Blue or green

Pseudomonas urinary tract infection Bilirubin Methylene blue

Pink or red

Aniline dyes in sweets Porphyrins (on standing) Blood, hemoglobin, myoglobin Drugs: phenindione, phenolphthalein Anthocyaninuria (beetroot, “beeturia”)

Orange

Drugs: anthraquinones (laxatives), rifampicin Urobilinogenuria

Yellow

Mepacrine Conjugated bilirubin Phenacetin Riboflavin

Brown or black

Melanin (on standing) Myoglobin (on standing) Alkaptonuria

Green or black

Phenol Lysol

Brown

Drugs: phenazopyridine, furazolidone, l-dopa, niridazole Hemoglobin and myoglobin (on standing) Bilirubin

From Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.

Urine Sediment

RBCs, white blood cells (WBCs), tubular cells, transitional cells, and squamous epithelial cells may be seen in the urine. Casts are formed in tubules, may contain cells or cellular debris, or may be acellular. RBCs may originate from intrarenal vessels, glomeruli, tubules, or anywhere in the urogenital tract (Fig. 114-3). Dysmorphic RBCs are cells that have been deformed by transit through the glomerulus and through the medullary interstitium, as opposed to RBCs from the remainder of the genitourinary tract (Figs. 114-4 and 114-5); these cells are often lysed and less refractile than nonglomerular RBCs. Dysmorphic RBCs often fragment with poikilocytosis and with blebs, forming so-called “Mickey Mouse” RBCs. Phase contrast microscopy aids in the identification of dysmorphic RBCs. The presence of a majority of dysmorphic RBCs in a urine sediment points to a glomerular origin of the hematuria. The presence of RBC casts is often conclusive evidence for the presence of glomerulonephritis. WBCs are seen most commonly in urinary tract infections, but they also can be seen in acute interstitial nephritis, infections with Legionella (Chapter 314) and Leptospira (Chapter 323) species, chronic infections such as tuberculosis (Chapter 324), allergic interstitial nephritis (Chapter 122), atheroembolic diseases (Chapter 125), granulomatous diseases such as sarcoidosis (Chapter 95), IgG4-related interstitial nephritis, and tubulointerstitial nephritis and uveitis syndrome. Mononuclear cells often appear with transplant rejection. Tubular cells, which are seen in many conditions involving tubulointerstitial diseases, also are seen in ischemic and nephrotoxic injury, such as with myeloma kidney (Chapter 187) or cast nephropathy. Eosinophils require special stains, with the Giemsa stain being much less sensitive than the Hansel stain (Chapter 122). Urine eosinophils classically are seen in allergic interstitial nephritis (Chapter 122), but they also are seen in atheroembolic disease (Chapter 125), prostatitis (Chapter 129), and vasculitis. Casts, which are formed in tubules, are characterized by the arrangement of the cells in a clearly formed matrix composed of Tamm-Horsfall protein. Because casts are formed in the renal parenchyma, they may give a clue to the origin of accompanying cellular elements. Hyaline casts are composed of Tamm-Horsfall proteins that are formed normally and are seen in increased numbers after exercise (Fig. 114-6).

CHAPTER 114  Approach to the Patient with Renal Disease  

731

Persistent dipstick positive for heme Infectious

Renal or urologic disorders

Hematologic or metabolic disorders

Positive leukocytes

Positive for red cells

Negative for red cells Rule out myoglobinuria or hemoglobinuria

Rule out UTI

Urologic disorders Calculi, BPH, urinary tract cancer

Intrinsic renal disease Decreased GFR or glomerular hematuria (RBC casts + dipstick protein or protein/creatinine > 0.5)

Non-glomerular hematuria

CT scan/MRI Urine cytology x 3 Nephrology referral Negative findings and patient with low risk of cancer

Positive findings or negative findings in a patient with high risk of cancer

Observe

Urology referral

FIGURE 114-3.  Algorithm for the evaluation of asymptomatic hematuria. BPH = benign prostatic hyperplasia; CT = computed tomography; GFR = glomerular filtration rate; MRI = magnetic resonance imaging; RBC = red blood cell; UTI = urinary tract infection. (Courtesy Ali Gharavi, MD. Modified from Cohen RA, Brown RS. Microscopic hematuria. N Engl J Med. 2003;348:2330-2338).

FIGURE 114-4.  Dysmorphic erythrocytes. These dysmorphic erythrocytes vary in size, shape, and hemoglobin content and reflect glomerular bleeding. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

Granular casts are degenerated tubular cell casts that are seen in the setting of tubular injury (Fig. 114-7). Pigmented granular casts are seen in rhabdomyolysis (Chapter 113) with myoglobinuria or, rarely, hemoglobinuria. RBC casts (Fig. 114-8) are rarely seen in allergic interstitial nephritis and diabetic nephropathy, but they are frequently seen in acute glomerulonephritis (Chapter 121). The presence of RBC casts in a patient with microscopic hematuria can narrow the focus of the evaluation to a glomerular lesion. WBC casts are seen commonly in pyelonephritis (Chapter 284) and in acute and chronic nonbacterial infections. They also are seen in other conditions in which WBCs are associated with parenchymal renal processes, such as allergic interstitial nephritis (Chapter 122), atheroembolic diseases (Chapter 125), and granulomatous diseases such as sarcoidosis (Chapter 95). Rarely, WBC casts can be a dominant feature of many diseases that traditionally are thought of as glomerular diseases, such as lupus nephritis (Chapter 266) and antineutrophil cytoplasmic antibody (ANCA)–associated glomerulonephritis (Chapter 270). Tubular cell casts are seen with any acute tubular injury and

FIGURE 114-5.  Isomorphic erythrocytes. These erythrocytes are similar in size, shape, and hemoglobin content. Isomorphic cells reflect nonglomerular bleeding from lesions such as calculi and papillomas or hemorrhage from cysts in polycystic renal disease. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

are the dominant cellular casts in ischemic acute tubular necrosis (Chapter 120). They also can be seen with nephrotoxic injury, such as with aminoglycosides and cisplatin. Some casts may contain both leukocytes and tubular cells. Crystals can be a normal finding in the urine or serve as clues to pathophysiologic processes. Certain crystals, such as the hexagonal crystals seen with cystinuria (Chapter 128), are always abnormal (Fig. 114-9). Others, such as the octahedral calcium oxalate crystals (Fig. 114-10), may be a normal finding or may be evidence for ethylene glycol intoxication (Chapter 110). Triple phosphate crystals, which are composed of ammonium magnesium phosphate and are coffin shaped (Fig. 114-11), are seen in urinary tract infections with urea-splitting organisms (Chapter 284). Uric acid crystals,

732

CHAPTER 114  Approach to the Patient with Renal Disease  

FIGURE 114-6.  Hyaline cast of the type seen in small numbers in normal urine. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.) FIGURE 114-10.  Oxalate crystals. A pseudocast of calcium oxalate crystals accompanied by crystals of calcium oxalate dehydrate. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

FIGURE 114-7.  Number and type of granules and their density in the cast vary in different casts. The presence of erythrocytes in this cast may mean that the granules are derived partly from disrupted erythrocytes. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

FIGURE 114-11.  Coffin-lid crystals of magnesium ammonium phosphate (struvite). (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

FIGURE 114-8.  A cast composed entirely of erythrocytes reflects heavy hematuria and active glomerular disease. Crescentic nephritis is likely to be present if erythrocyte cast density is greater than 100/mL. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

sodium urate crystals (Fig. 114-12), and calcium phosphate amorphous crystals are common and do not usually have pathologic significance. Other Elements

Bacteria may be seen in the urine sediment. A spun urine sediment may show rods or cocci in chains, but bacteria are identified best by Gram staining of the urine sediment. Budding yeast forms (which are highly refractile), trichom*onads, and spermatozoa also may be seen in the urinary sediment.

  SPECIFIC RENAL SYNDROMES

This chapter considers the approach to the patient with acute kidney injury (Chapter 120), glomerular syndromes (nephrotic vs. nephritic; Chapter 121), tubulointerstitial disease (Chapter 122), vasculitis and vascular diseases of the kidney (Chapter 125), papillary necrosis, and chronic kidney disease (Chapter 130).

Acute Kidney Injury FIGURE 114-9.  Typical hexagonal cystine crystal. A single crystal provides a defini-

tive diagnosis of cystinuria. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

Acute kidney injury (Chapter 120) is a syndrome in which glomerular filtration declines during a period of hours to days. The serum creatinine level is elevated in both acute and chronic kidney disease, but an actively rising serum creatinine level confirms an acute or acute-on-chronic insult to kidney

CHAPTER 114  Approach to the Patient with Renal Disease  

733

FIGURE 114-13.  Normal findings on sagittal renal ultrasound. The cortex is hypoechoic compared with the echogenic fat containing the renal sinus. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.) FIGURE 114-12.  Urate crystals. Complex crystals suggestive of acute urate nephropathy or urate nephrolithiasis. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

function. As a blood filtration organ, the kidney is susceptible to an acute compromise of renal arterial perfusion (Chapter 125), such as prerenal kidney injury, or blockage in urine outflow, such as urinary obstruction due to benign prostatic hypertrophy (Chapter 129). Thus, the patient with acute renal failure is best approached by evaluation for prerenal, renal, and postrenal causes. The intrarenal causes of acute kidney injury include acute tubular necrosis (Chapter 120), acute interstitial nephritis (Chapter 122), acute glomerulonephritis (Chapter 121), and acute vasculitis and vascular disease (Chapters 121 and 125). The careful and systematic evaluation of the patient should start with a thorough history and physical examination, which should be followed by selected laboratory tests and often an imaging test, such as renal ultrasonography. Most cases of acute renal failure in the hospital have hemodynamic or toxic causes, so prerenal azotemia and acute tubular necrosis must be considered carefully and distinguished from one another.

ETIOLOGY

Prerenal Kidney Injury

Prerenal kidney injury can be caused by shock or renal hypoperfusion from a variety of conditions, including arterial underfilling secondary to edematous states (e.g., severe heart failure, decompensated cirrhosis) or, more variably, cases of nephrotic syndrome. History relevant to renal hypoperfusion states, such as a history of acute gastroenteritis, should be sought. Patients should also be asked about use of nonsteroidal anti-inflammatory drugs or blockers of the renin-angiotensin-aldosterone system (e.g., angiotensionconverting enzyme inhibitors, angiotensin receptor blockers) that can exacerbate prerenal injury. Relative hypotension compared with a patient’s baseline blood pressure and orthostatic changes in blood pressure and pulse indicate arterial underfilling. Relatively minor orthostatic hypotension may explain the acute decompensation of kidney function in a patient with chronic kidney disease (Chapter 130) or renal artery stenosis (Chapter 125). Lower extremity edema is common in cirrhosis (Chapter 153), heart failure (Chapter 58), and nephrotic syndrome (Chapter 121).

Acute Tubular Necrosis

Acute tubular necrosis can arise from ischemic or toxic injury to the kidneys. Prerenal azotemia can progress to acute tubular necrosis, particularly if frank hypotension occurs in the setting of infection and persists. The transition of prerenal renal failure to acute tubular necrosis may be revealed by a rise in the fractional excretion of sodium to a value greater than 1%. Alternatively, acute tubular necrosis may arise from a toxic effect, so a medication and ingestion history is critical to the evaluation of the patient.

DIAGNOSIS

Laboratory Testing

The normal concentration of blood urea nitrogen (BUN), which is a product of protein catabolism, is about 10-fold higher than the creatinine concentration. Because the BUN-to-creatinine ratio commonly rises with arterial underfilling, BUN typically is used as a marker of effective volume status. Classically, the BUN-to-creatinine ratio will be higher than 15 to 20 in prerenal azotemia but 10 or close to it in acute tubular necrosis. However, the

BUN concentration (and hence its ratio to creatinine concentration) may be inappropriately high in other circ*mstances, such as with high protein intake, gastrointestinal bleeding, or the use of steroids or tetracyclines. The BUN concentration and its ratio to creatinine concentration may be low in patients who have a poor dietary intake of protein, malnutrition, or liver disease. The excretion of sodium in the setting of oliguria and acute kidney injury (Chapter 120) often gives insight into the appropriateness of tubular function. The fractional excretion of sodium (FeNa) is calculated as follows: FE Na = (urine Na/plasma Na)/(urine Cr/plasma Cr) × 100 where Na is the sodium concentration (in mmol/L) and Cr is the creatinine concentration (in mmol/L or mg/dL). In the setting of oliguria, FeNa below 1% often denotes prerenal azotemia, whereas FeNa above 1% suggests intrinsic renal damage. Although this measurement is generally useful, FeNa below 1% may be seen without evidence of a prerenal component, including contrast nephropathy (Chapter 120), hepatorenal syndrome (Chapter 154), obstructive uropathy (Chapter 123), interstitial nephritis (Chapter 122), glomerulonephritis (Chapter 121), and rhabdomyolysis (Chapter 113). Conversely, a high FeNa can be seen in cases in which there is a prerenal component, including diuretic use, adrenal insufficiency (Chapter 227), cerebral salt wasting, and salt-wasting nephropathy (Chapter 116). The FeNa must be evaluated in the context of the clinical situation because it can be low or high in a normal patient or in a patient with chronic kidney disease. Ultimately, a patient’s volume status is best at the bedside and should not be deduced solely from a measurement of electrolytes.

Imaging

Ultrasonography, which is the most commonly used renal imaging study (Fig. 114-13), provides reliable information about obstruction, kidney size, presence of masses, and renal echotexture. Ultrasonography has only a 90% sensitivity for the detection of hydronephrosis and hence is not sufficient to exclude obstruction (Chapter 123) with certainty. In addition, its inability to detect stones in the ureters and bladder limits its utility in the evaluation for kidney stones (Chapter 126). Ultrasonography can detect vascular disease, and Doppler imaging permits evaluation of the renal vessels with resistive indices. Resistive indices are crucial in ascribing renal dysfunction to the detected vascular disease (Chapter 125). A high resistive index reflects parenchymal disease with scarring and indicates that intervention on the vascular disease itself is unlikely to improve renal function. A computed tomography (CT) scan stone protocol to assess the kidneys, ureters, and bladder is the study of choice for detecting kidney stones (Chapter 126) because of its ability to detect stones of all kinds, including uric acid stones and nonobstructing stones, as well as stones in the ureters (Fig. 114-14). Masses in the kidney can be evaluated with either contrast CT or a renal ultrasound examination. CT angiography with iodinated contrast material can assess possible renal artery stenosis (Chapter 125) with an accuracy comparable to that of magnetic resonance (MR) angiography.

Glomerular Syndromes: Nephrotic versus Nephritic The nephrotic syndrome (Chapter 121) is characterized by the presence of proteinuria of more than 3.5 g/day/1.73 m2, with accompanying edema, hypertension, and hyperlipidemia. Other consequences include a predisposition to infection and hypercoagulability. In general, the diseases associated

734

CHAPTER 114  Approach to the Patient with Renal Disease  

Serologies

FIGURE 114-14.  Delayed excretion in the left kidney secondary to a distal calculus. Contrast-enhanced computed tomography scan shows dilated left renal pelvis. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

with nephrotic syndrome do not cause acute kidney injury, although acute kidney injury may be seen with minimal change disease, human immunodeficiency virus (HIV)–associated nephropathy, and bilateral renal vein thrombosis (Chapter 125). The causes of primary idiopathic nephrotic syndrome, in decreasing order of prevalence, are focal and segmental glomerulosclerosis, membranous nephropathy, minimal change disease, and membranoproliferative glomerulonephritis. Membranous nephropathy has been associated with antibodies to the M-type phospholipase A2 receptor. Secondary causes of the nephrotic syndrome include diabetic nephropathy (Chapter 124), amyloidosis (Chapter 188), and membranous lupus nephritis (Chapters 121 and 266). The acute nephritic syndrome is an uncommon but dramatic presentation of an acute glomerulonephritis (Chapter 121). The hallmark of the acute nephritic syndrome is the presence of dysmorphic RBCs and RBC casts, but their absence does not exclude the syndrome. The acute nephritic syndrome can be caused by any of the rapidly progressive glomerulonephropathies with ANCA-associated vasculitis (granulomatosis with polyangiitis, microscopic polyangiitis, and eosinophilic granulomatosis with polyangiitis), anti–glomerular basem*nt membrane (anti-GBM) glomerulonephritis, and immune complex–mediated glomerulonephritis (including systemic lupus erythematosus, cryoglobulinemia, postinfectious glomerulonephritis, endocarditis, IgA nephropathy, and Henoch-Schönlein purpura). The rapid decline in renal function often warrants urgent and usually inpatient evaluation.

DIAGNOSIS

Laboratory Testing

Proteinuria (as albuminuria) of more than 3.5 g in 24 hours generally indicates glomerular disease (Chapter 121). Lesser quantities do not preclude glomerular disease, and electrophoresis gives valuable insight into the composition of the proteinuria (Chapter 187). On occasion, overflow proteinuria of a low-molecular-weight protein, such as light chains in Bence Jones proteinuria, can be higher than 3.5 g/day without any of the manifestations or implications of the nephrotic syndrome; a urine protein electrophoresis study is important in making the distinction. A comparison of the microalbumin-to-creatinine ratio with the protein-to-creatinine ratio will give an insight into the presence of Bence Jones protein because of the absence of albuminuria despite significant proteinuria. Collection must be done by discarding the first morning void and collecting all urine output for the next 24 hours, including the first morning void the next day. The 24-hour urine collection for protein excretion is cumbersome and subject to inaccuracies. Instead, a spot urine sample for protein and creatinine can be used to estimate the amount of protein excreted. A protein-to-creatinine ratio of 3 translates to a 24-hour protein excretion of about 3 g. The ratio is most accurate when the first morning urine collection is used and may be inaccurate in patients with orthostatic proteinuria. The evaluation of proteinuric renal dysfunction, particularly when glomerular diseases are suspected, should follow a stepwise progression from noninvasive serologic evaluation to a definitive or confirmatory diagnostic evaluation, such as a renal biopsy.8 Sometimes an expeditious diagnosis is needed, and a biopsy may be done relatively early in the evaluation.

An antinuclear antibody (ANA) titer can be useful to evaluate glomerular disease in either nephrotic or nephritic presentations. A high ANA titer (e.g., 1 : 320), especially if it is accompanied by a more specific finding such as anti–double-stranded DNA antibody or anti-Smith antibody, can be highly specific for the diagnosis of lupus nephritis (Chapter 266), which usually requires a renal biopsy. Lower titers (e.g., 1 : 80 or 1 : 40) are nonspecific. A rheumatoid factor titer will usually be elevated in patients with rheumatoid arthritis (Chapter 264), but vasculitis is a relatively late and rare event. Rheumatoid factor can be detected in some forms of cryoglobulinemia (Chapter 187); for example, IgM, which is present in type II and type III cryoglobulinemia, has rheumatoid factor activity. Rheumatoid factor also can be seen as a nonspecific finding in bacterial endocarditis (Chapter 76) and systemic vasculitis (Chapter 270). The levels of complement components C3 and C4 and the 50% hemolyzing dose of complement (CH50) usually are measured to evaluate suspected rapidly progressive glomerulonephritis (Chapter 121). Complement levels are usually low in active systemic lupus erythematosus (Chapter 266), poststreptococcal glomerulonephritis (Chapter 121), endocarditis (Chapter 76), membranoproliferative glomerulonephritis, cryoglobulinemia (Chapter 187), shunt nephritis with infection of a ventriculoatrial shunt, and glomerulonephritis associated with visceral abscesses. A particularly depressed C4 compared with C3 should raise the suspicion of cryoglobulinemia. Serum immunoelectrophoresis will detect elevated polyclonal IgA levels in about 50% of cases of IgA nephropathy (Chapter 121) and Henoch-Schönlein purpura (Chapter 121). Polyclonal elevation of IgG may occur in a variety of systemic diseases and is a nonspecific finding. The presence of a monoclonal protein in the serum should raise the suspicion for a monoclonal gammopathy–associated disease (Chapter 187). The differential diagnosis includes monoclonal gammopathy of uncertain significance, myeloma kidney, lymphomas (Chapter 185), amyloidosis (Chapter 188), light chain deposition disease, heavy chain deposition disease, immunotactoid glomerulonephritis, and cryoglobulinemia. The concentration of the monoclonal protein is higher when the diagnosis of multiple myeloma is made, but even small quantities of Bence Jones proteins in the serum can have clinical significance. A urine immunoelectrophoresis always should be obtained concomitantly if myeloma is suspected. Because a substantial fraction of multiple myelomas can have no heavy chain excretion and small quantities of light chains may be difficult to detect by serum immunoelectrophoresis, a urine immune electrophoresis test for Bence Jones protein complements the serum immunoelectrophoresis. In light chain myeloma, patients may have Bence Jones proteinuria even in the absence of an M component in the serum immunoelectrophoresis. Bence Jones proteinuria may be present in myeloma kidney, amyloidosis, light chain deposition disease, lymphoma, or, occasionally, monoclonal gammopathy of uncertain significance. However, some patients with systemic AL (light chain) amyloidosis have a normal serum immunoelectrophoresis and no Bence Jones proteinuria (Chapter 187). More sensitive assays for serum free light chains and an assessment of the ratio of κ to λ lights chains increase the sensitivity for detection of monoclonal gammopathies. The antineutrophil cytoplasmic antibody (ANCA) assay has allowed earlier and more definitive recognition of vasculitic causes of rapidly progressive glomerulonephritis (Chapter 270), especially granulomatosis with polyangiitis, microscopic polyangiitis, and eosinophilic granulomatosis with polyangiitis, when it is confirmed by enzyme-linked immunosorbent assay. The antibodies cause two different patterns of staining: perinuclear staining (p-ANCA) and cytoplasmic staining (c-ANCA). Both antigens actually have a cytoplasmic distribution, and the perinuclear staining pattern is an artifact of the fixation method. In most cases, the antigen for p-ANCA is myeloperoxidase (MPO), whereas the antigen for c-ANCA is proteinase 3 (PR3). Anti-MPO antibodies are associated with microscopic polyangiitis, idiopathic crescentic glomerulonephritis, or Churg-Strauss syndrome (eosinophilic granulomatosis with polyangiitis; Chapter 270). Anti-PR3 antibodies often correlate with the classic disease of granulomatosis with polyangiitis (formerly known as Wegener granulomatosis) (Chapter 270). Anti–glomerular basem*nt membrane (anti-GBM) antibodies are autoantibodies to the Goodpasture antigen (Chapter 121), which resides in a domain of the α chain of type 4 collagen. An early and accurate diagnosis of Goodpasture syndrome can be made by immunofluorescence and confirmed by Western blot analysis. Anti-GBM antibody staining also may occur in the

CHAPTER 114  Approach to the Patient with Renal Disease  

presence of a positive ANCA. In these cases, the theory is that exposure of the Goodpasture antigen, as a result of the glomerular injury, leads to antiGBM antibody formation as a secondary process. Cryoglobulins (Chapter 187) are thermolabile immunoglobulins. They are a single monoclonal type in type I cryoglobulinemia. In type II and type III cryoglobulinemia, however, the mixture of immunoglobulins includes one with rheumatoid factor activity against IgG. Type I and type II cryoglobulins are more likely to be associated with clinical disease, especially at higher titers. Type III cryoglobulinemia is often of less clinical significance. Type I cryoglobulinemia is seen with Waldenström macroglobulinemia and multiple myeloma (Chapter 187); type II, with hepatitis C infection (Chapters 148 and 149), Sjögren syndrome (Chapter 268), lymphomas (Chapters 185 and 186), and systemic lupus erythematosus (Chapter 266); and type III, with hepatitis C (Chapters 148 and 149), chronic infections, and inflammatory conditions. When cryoglobulinemia is associated with hepatitis C, the hepatitis C virus (HCV) RNA is concentrated in the cryoprecipitate; the diagnosis can be made by an RNA assay of the cryoprecipitate at 37° C. Membranous nephropathy is associated with chronic hepatitis B infection with hepatitis B surface antigenemia (Chapter 149). Classic polyarteritis nodosa (Chapter 270) occasionally is seen with chronic hepatitis B infection, often with surface antigenemia and hepatitis B e antigenemia. M-type phospholipase A2 receptor antibodies also have been detected as autoantibodies in idiopathic membranous nephropathy. Hepatitis C serology is associated with a variety of renal diseases, including cryoglobulinemia, membranoproliferative glomerulonephritis, and membranous nephropathy. The evaluation may include the antibody test and an assay for HCV RNA. On occasion, the HCV RNA analysis may have to be conducted on the cryoprecipitate at 37° C. HIV-associated nephropathy (Chapter 121) is associated with nephrotic syndrome and acute kidney injury. In the appropriate clinical setting, HIV serology and viral titers are warranted tests for both clinical syndromes. Streptococcal infection can be confirmed as the cause of postinfectious glomerulonephritis (Chapter 121) with an anti-DNase or antistreptolysin assay. Acute and convalescent serology assays are used to confirm recent infection. The erythrocyte sedimentation rate (ESR) is a relatively nonspecific test in the evaluation of renal disease. However, a high ESR often points to systemic vasculitis (Chapter 270), multiple myeloma (Chapter 187), or malignant disease as the underlying cause. However, the ESR often is elevated in the nephrotic syndrome (Chapter 121), including diabetic nephropathy (Chapter 124).

Renal Biopsy

No formal guidelines exist for the indications to perform a renal biopsy. Most nephrologists will perform a biopsy for adults with idiopathic nephrotic syndrome and for children with steroid-dependent or steroid-resistant nephrotic syndrome. In addition, acute kidney injury without an identifiable inciting cause is a clear indication for biopsy. Notably, patients with hospital-acquired kidney failure rarely meet this indication. Other abnormal clinical findings, such as gross or microscopic hematuria or subnephrotic proteinuria, often but not always lead to a kidney biopsy. Renal biopsy usually is performed percutaneously with real-time ultrasound or CT guidance. About 1 to 2% of patients without an underlying coagulopathy will develop bleeding that requires a transfusion. The transjugular approach can be used in patients in whom the risks for bleeding are high. The decision to pursue a kidney biopsy should be individualized for each patient, but a renal biopsy generally is justified for most patients with two or more of the following four findings: hematuria, proteinuria above 1 g/day, renal insufficiency, or positive serologies for systemic diseases with known potential for kidney involvement (e.g., hepatitis B or C virus infection, systemic lupus erythematosus, ANCA seropositivity). The decision about whether to perform a renal biopsy in diabetic patients with suspected diabetic nephropathy should be individualized and is usually driven by the presence of atypical features or an active urine sediment.9 In addition, in patients with renal transplants (Chapter 131) and acute or chronic renal failure, biopsy of the allograft kidney provides crucial information in guiding diagnosis and treatment.

Tubulointerstitial Diseases Tubulointerstitial diseases (Chapter 122) vary in presentation from acute kidney injury to chronic kidney dysfunction that initially is manifested as asymptotic mild renal insufficiency (Table 114-3). The urine sediment often

735

TABLE 114-3  MAJOR CAUSES OF TUBULOINTERSTITIAL DISEASE Ischemic and toxic acute tubular necrosis Allergic interstitial nephritis Interstitial nephritis secondary to immune complex–related collagen vascular disease, such as Sjögren disease or systemic lupus erythematosus Granulomatous diseases: sarcoidosis, tubulointerstitial nephritis with uveitis IgG4-related interstitial nephritis Pigment-related tubular injury: myoglobulinuria, hemoglobinuria Hypercalcemia with nephrocalcinosis Tubular obstruction: drugs such as indinavir, uric acid in tumor lysis syndrome Myeloma kidney or cast nephropathy Infection-related interstitial nephritis: Legionella, Leptospira species Infiltrative diseases, such as lymphoma

contains small to moderate amounts of proteinuria, usually less than 1 g/day, as well as WBCs, RBCs, tubular cells, and WBC casts. RBC casts are rare in acute interstitial nephritis and are more characteristic of glomerular disease.

Vasculitis and Vascular Diseases of the Kidney Vascular diseases of the kidney can be divided into large-vessel obstruction and medium- to small-vessel diseases (Chapter 125). Renovascular disease is a common cause of hypertension, heart failure, and renal insufficiency. About 90% of renal artery stenosis is atherosclerotic in origin, with most of the remaining caused by fibromuscular dysplasia, which is more common in women 20 to 50 years of age. Medium-sized arterial vessel diseases include polyarteritis nodosa, which is seen in patients with hepatitis B (Chapters 148 and 149), HIV infection (Chapter 121), or, rarely, hepatitis C (Chapters 148 and 149). Symptoms include abdominal pain, hypertension, and mild renal insufficiency, often with a benign sediment; diagnostic findings include microaneurysms at the bifurcation of medium-sized arteries. Other diseases involving small vessels include atheroembolic disease (Chapter 125), which is seen either spontaneously or after arteriography or surgery. This syndrome typically affects the kidneys, gastrointestinal tract, and lower extremities, but it can also involve the central nervous system when the aortic arch is affected. The thrombotic microangiopathies include hemolytic-uremic syndrome (HUS) and thrombotic thrombocytopenic purpura (Chapter 172). Thrombocytopenic purpura is associated with an acquired inhibitor to or the congenital inherited absence of a protease that cleaves large-molecular-weight von Willebrand multimers. HUS is caused by endothelial injury. In diarrheapositive (or typical) HUS, the endothelial injury is induced by Shiga toxin from Escherichia coli O157:H7 infection. In diarrhea-negative (atypical) HUS, dysregulation of the alternative complement pathway is the underlying cause of endothelial injury. The antiphospholipid antibody syndrome (Chapter 176) can cause large-vessel thrombosis and stenosis as well as a thrombotic microangiopathy with proteinuria, hypertension, and renal insufficiency. Scleroderma renal crisis, which is a manifestation of systemic sclerosis (Chapter 267), often leads to an inexorable progression to end-stage renal insufficiency if untreated. A systemic vasculitis may be manifested in a variety of ways, including skin manifestations such as petechial rash, purpura, digital gangrene, and splinter hemorrhages. Otitis, sinusitis, epistaxis, hemoptysis, and nasal septal ulcers are common manifestations of granulomatosis with polyangiitis (Chapter 270). Pulmonary hemorrhage can be a catastrophic manifestation of Goodpasture syndrome (Chapter 121) or anti-GBM disease as well as the ANCAassociated vasculitis (Chapter 270). Abdominal pain and tenderness and gastrointestinal hemorrhage may be observed in Henoch-Schönlein purpura and classic polyarteritis nodosa (Chapter 270). Neurologic symptoms may be a manifestation of vasculitis, such as microscopic polyangiitis (Chapter 270) and cryoglobulinemia (Chapter 187).

DIAGNOSIS

Radiologic Evaluation

Magnetic resonance imaging (MRI) with MR angiography (Fig. 114-15) is highly sensitive for detecting atherosclerotic renovascular disease (Chapter 125), but it tends to overestimate the degree of stenosis. Its accuracy in detecting fibromuscular dysplasia, however, is less well validated. MRI also can be used to evaluate renal masses. MRI does not require iodinated contrast material, but gadolinium-based contrast agents for vascular studies are

736

CHAPTER 114  Approach to the Patient with Renal Disease  

Lethargy, fits, coma Frost Red eye

Epistaxis Anemia (mucosal pallor)

Anorexia Nausea Vomiting

Sallow pigmentation

Hypertension Pericarditis Heart failure

Pruritic excoriations

Pleurisy Dyspnea on exertion

Bruising Amenorrhea Impotence Infertility

Nail changes Bone pain

Myopathy (muscle weakness)

FIGURE 114-15.  Magnetic resonance angiography. Coronal three-dimensional image shows right renal artery stenosis (arrow). (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

Peripheral neuropathy

associated with the syndrome of nephrogenic systemic fibrosis in patients with advanced renal failure (Chapter 267). Renal arteriography, which is the “gold standard” in the evaluation of renal artery stenosis (Chapter 125), also is used for the evaluation of arteriovenous malformations, polyarteritis nodosa, and other vascular lesions of the kidneys. This invasive study uses iodinated contrast material and incurs a small risk for atheroembolic disease (Chapter 125). Therapeutic angioplasty and stenting can be done at the time of angiography.

Papillary Necrosis Acute necrosis of the renal papilla is associated with sickle cell anemia (Chapter 163), analgesic nephropathy (Chapter 122), diabetic nephropathy (Chapter 124), and obstructive pyelonephritis (Chapter 284). In sickle cell disease (Chapter 163)10, the hypoxic and hypertonic milieu of the inner medulla promotes sickling, and chronic sickling at the vasa recta results in medullary ischemia. Massive and prolonged consumption of analgesics, particularly the combination of aspirin, caffeine, and acetaminophen, is associated with chronic interstitial nephritis and a predisposition to papillary necrosis (Chapter 122); medullary ischemia is thought to be caused by inhibition of synthesis of vasodilatory prostaglandins by aspirin, and direct toxicity is attributed to metabolites of phenacetin. Similarly, medullary perfusion is thought to be compromised in diabetic nephropathy (Chapter 124) and obstructive pyelonephritis (Chapter 123). The clinical manifestations of papillary necrosis can include flank pain and hematuria. If the papilla is sloughed, obstruction may occur at the renal pelvis or ureter of the affected kidney, with referred pain migrating from the flank to the groin. A sloughed papilla may precipitate frank renal failure if the function of the contralateral kidney is impaired or if obstruction occurs at the level of the bladder or urethra (Chapter 123). Classically, papillary necrosis is diagnosed on an excretory pyelogram as a calyceal defect after sloughing of a papilla, but CT with contrast enhancement is as good for advanced lesions. If the necrotic papilla is retained, however, the defect will be more subtle. Transitional cell carcinoma (Chapter 197) can occur in the setting of papillary necrosis or can mimic its appearance. Obstruction, if present, must be relieved, but treatment otherwise is limited to pain control and hydration.

Chronic Kidney Disease Chronic kidney disease, which is defined as either kidney damage or a GFR of less than 60 mL/min/1.73 m2 for longer than 3 months, includes five stages (Table 114-4). Kidney damage is defined as pathologic abnormalities or markers of kidney damage, including abnormalities in the composition of blood or urine or abnormalities on imaging tests. The excretion of 30 to 300 mg of albumin in a 24-hour period defines microalbuminuria. An estimated 12% of the adult U.S. population has abnormal albumin excretion in the urine, and the frequency increases with age. Kidney failure is defined as

Edema

FIGURE 114-16. Common symptoms and signs of chronic renal failure. (Redrawn from Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.)

TABLE 114-4  STAGES OF CHRONIC KIDNEY DISEASE* STAGE

DESCRIPTION

GFR (mL/min/1.73 m2)

1

Kidney damage with normal or ↑GFR

2

Kidney damage with mild or ↓GFR

≥90 60-89

3

Moderate ↓GFR

30-59

4

Severe ↓GFR

15-29

5

Kidney failure

1.020), an elevated osmolality (>400 mOsm/ kg), and a sodium concentration of less than 20 mmol/L because of enhanced renal tubule reabsorptive activity. More complex indices of the appropriate renal response to hypovolemia include fractional excretion of sodium of less than 1% and fractional excretion of urea of less than 30 to 35%. Intrinsic renal injury confounds the diagnostic value of these urinary indices.

Differential Diagnosis

Relative hypovolemia secondary to arterial vasodilation mimics some of the clinical manifestations of absolute hypovolemia. With vasodilation, as seen, for example, in sepsis (Chapter 108), tachycardia and hypotension are common, but the extremities may be warm. However, tissues are actually underperfused, as reflected by reduced renal and cerebral function and lactic acidosis.

745

TREATMENT  Absolute Hypovolemia

The major goal in treatment of hypovolemia is to restore hemodynamic integrity and tissue perfusion. The management approach includes treatment of the underlying disease state when possible, replacement of the volume deficit, and fluid administration to maintain ECF volume in the event of continuing losses. Irrespective of specific treatments, the mainstay of therapy involves fluid administration. The important issues are the volume, rate of administration, and composition of the replacement and maintenance fluids. These factors may vary during different stages of treatment and should be adjusted according to the patient’s response as determined by closely monitored clinical parameters. The choice of oral or intravenous replacement fluids (or both) for hypovolemic states is dictated by the integrity of gastrointestinal absorptive function, by the magnitude of the volume deficit, and by the disturbances in other electrolyte and acid-base parameters. The rate of replacement is a function of the urgency of the threat to circulatory integrity and consideration of complications related to overzealous or too rapid correction. Fluid therapy for hypovolemic states sometimes begins with a diagnostic fluid challenge. In situations in which clinical parameters do not permit a firm diagnosis of hypovolemia, the response to a fluid challenge can be informative and serve as the initial treatment step. For example, a patient with known long-standing compensated heart failure who is being maintained on a therapeutic regimen that includes diuretics may have tachycardia, reduction in blood pressure from baseline values, poor cognition, and renal dysfunction. Such a clinical scenario could have a number of different explanations, including superimposed volume depletion with inadequate left ventricular filling volume. CVP, whether measured directly or assessed by jugular venous pressure, may be misleading in the face of right ventricular dysfunction, but direct measurement of pulmonary capillary wedge pressure does not significantly improve clinical outcomes.3 Conversely, interventional hemodynamic monitoring to guide fluid resuscitation has been shown, in randomized controlled trials, to lead to lower mortality and a reduced incidence of acute kidney injury in certain postsurgical settings. A1  Another example is a patient with hyponatremia in the setting of suspected volume depletion. The degree of volume depletion is often too subtle to be detected by clinical examination, and a therapeutic challenge with fluid of the appropriate composition may be the only option. The initial volume and rate of therapeutic replacement fluid should be determined by ongoing monitoring of clinical parameters rather than by a priori estimates of volume deficit.3 In some settings, the clinical state will dictate rapid fluid replacement, as in a patient with unambiguous hypovolemic shock and life-threatening circulatory collapse. In such cases, fluids can be administered at the most rapid rate possible, limited only by intravenous access, until blood pressure and tissue perfusion are restored. However, in most cases, much slower rates are indicated, especially in elderly patients, patients whose medical background is unclear, or those with known comorbid conditions. Replacement fluids of different compositions have disparate volumes of distribution in the body fluid compartments and therefore differ in their efficiency of restoring ECF volume. Crystalloid solutions with sodium as the principal cation are the mainstay in fluid replacement therapy for hypovolemic states and are indicated primarily for hypovolemic states that are caused by renal, gastrointestinal, or sweat-based sodium losses. These solutions also are useful initial agents and adjuncts to therapy for the hypovolemia of hemorrhage and burns. Isotonic saline is confined to the ECF compartment (except in cases of severe dysnatremia). Thus, retention of 1 L of infused isotonic saline increases plasma volume by about 300 mL, with the remaining portion distributed in the interstitial subcompartment of the ECF. In contrast, a solution of 5% dextrose in water (D5W) is equivalent to administering solute-free water and distributes uniformly throughout all body fluid compartments (one third of the retained volume of infusate remains in the ECF compartment and only approximately 10 to 15% in the intravascular compartment). Infusing a given volume of half-isotonic saline (0.45% sodium chloride plus 5% glucose) can be considered equivalent to infusing half that volume as solute-free water (distributed throughout body fluid compartments) and the other half as isotonic saline (confined to the ECF compartment). The retained solute-free volume reduces body tonicity and the plasma sodium concentration, potentially beneficial in the follow-up treatment of patients whose hypovolemia is accompanied by hypertonicity and hypernatremia but detrimental for patients with normotonic or hypotonic hypovolemia.4 When hypovolemia is accompanied by hypobicarbonatemia (metabolic acidosis), it may be appropriate to design a solution in which a portion of  the sodium is accompanied by bicarbonate (Chapter 118). For example, it is possible to add a given quantity of hypertonic sodium bicarbonate to a solution of half-isotonic saline (in which chloride is the anion accompanying sodium) to obtain an isotonic replacement fluid appropriate for the given acid-base status of the patient. Similarly, in patients with concomitant potassium depletion (Chapter 117), especially when it is accompanied by metabolic alkalosis, addition of potassium chloride to the replacement solution may be

746

CHAPTER 116  Disorders of Sodium and Water Homeostasis  

indicated. A number of crystalloid solutions with predetermined concentrations of potassium, lactate (converted to bicarbonate by the liver), and other electrolytes are commercially available, but it is more appropriate to begin with a sodium chloride–containing solution at a concentration appropriate to body tonicity, then to add other solutes as indicated or at a separate intravenous administration site. This approach provides maximal flexibility in tailoring individualized fluid replacement therapy to the patient’s needs. Administration of chloride-restricted intravenous fluids rather than of chloride-liberal intravenous fluids may decrease the incidence of acute kidney injury in critically ill patients.5 Colloid-containing solutions include albumin or high-molecular-weight carbohydrate molecules (e.g., hydroxyethyl starch or dextran) at concentrations that exert a colloid osmotic pressure equal to or greater than that of plasma. Banked human plasma is also considered a colloid solution. Because large molecules such as albumin and high-molecular-weight carbohydrates do not readily cross the transcapillary barrier, they are thought to expand the intravascular compartment more rapidly and efficiently than crystalloid  solutions. However, randomized trials have not shown any benefit of colloids compared with crystalloids for fluid resuscitation. A2  Moreover, some largemolecular-weight carbohydrates, such as hydroxyethyl starch, appear to  be nephrotoxic and probably should not be used. A3  In patients with multiorgan system failure and capillary leakage, albumin is both rapidly catabolized and redistributed into the interstitial compartment, so it can aggravate interstitial edema without providing the benefit of intravascular volume repletion. Nevertheless, albumin-containing solutions may be useful in hypovolemia associated with burns (Chapter 111), when cutaneous protein losses are  appreciable. Furthermore, because of the capacity for rapid intravascular volume expansion with just a small volume of replacement fluid, colloidcontaining solutions are frequently used when rapid intravascular expansion is desired, such as at trauma sites outside of the hospital setting. Overall, crystalloid-containing solutions should be the mainstay of volume replacement therapy. In theory, blood products can be used for volume replacement in hypovolemic states, and a unit of packed red blood cells remains entirely in the vascular compartment. However, erythrocytes are actually considered part of the intracellular compartment and do not contribute to organ plasma flow. The role of packed red cells in the treatment of hemorrhage is to restore the principal function of the erythrocyte in oxygen carriage and delivery, not as a means of ECF volume replacement. In addition to replacement fluids, maintenance fluids must be provided to counteract ongoing losses. Such ongoing losses may be a continuation of the underlying disease state (e.g., continued vomiting, diarrhea, polyuric states, or severe burns). The volume, rate of administration, and composition of these replacement fluids are best determined by actual measurements of the corresponding ongoing fluid losses, with appropriate adjustments for the patient’s clinical assessment parameters.

Relative Hypovolemia

The treatment approach to relative hypovolemia is more complex than for absolute hypovolemia. When relative hypovolemia is the result of peripheral vasodilation, therapy should be directed toward reversal of the underlying cause and restoration of normal vascular reactivity. Bridging to maintain circulatory integrity until the underlying cause is successfully reversed can be achieved by infusion of an isotonic crystalloid solution such as normal saline. In such situations, selection of volumes and rates must be done with extreme caution because there is no absolute deficit and the administered volume will have to be excreted or removed once systemic vascular resistance and vascular capacitance are restored to normal. Furthermore, it is more difficult to estimate an increase in vascular capacitance than it is to estimate an absolute volume deficit. On occasion, it is appropriate to consider the use of vasoconstrictor agents.

Hypervolemia

DEFINITION

Hypervolemia refers to expansion of ECF volume, which varies, even in normal individuals, with dietary sodium intake. Thus, an individual in steady state with low daily dietary sodium intake (e.g., 20 mmol/day, corresponding to approximately 1.2 g of table salt per day) will have correspondingly low urinary sodium excretion, equivalent to dietary intake minus extrarenal losses. A shift to much higher sodium intake (e.g., 200 mmol/day, corresponding to approximately 12 g of table salt per day) will bring the individual to a new steady state characterized by a correspondingly higher urinary sodium excretion rate. This shift is accompanied by an increase in ECF volume, which triggers the sensor and effector mechanisms for increased urinary sodium excretion (described earlier). In most individuals, this increase in ECF volume

TABLE 116-4  PRIMARY AND SECONDARY RENAL SODIUMRETAINING STATES PRIMARY Oliguric renal failure Chronic kidney disease Glomerular disease, including nephrotic syndrome Severe bilateral renovascular obstruction Mineralocorticoid excess Inherited sodium-retaining tubulopathies SECONDARY Cardiac failure Cirrhosis Idiopathic edema

is not clinically detectable and does not have pathologic consequences. In some individuals, however, this upward shift in ECF volume increases systemic arterial blood pressure. When the sodium surfeit expands the ECF volume beyond the range necessary for the adjustment needed to restore sodium balance, a state of pathologic hypervolemia ensues.

EPIDEMIOLOGY

Primary and secondary renal sodium retention (Table 116-4) can lead to hypervolemia. Patients with oliguric renal failure of any cause (Chapters 120 and 130) have a limited ability to excrete both sodium and water. Urinary sodium retention can be one of the cardinal manifestations of primary glomerular diseases (Chapter 121), even when the GFR is well preserved. States of mineralocorticoid excess (Chapter 227) or enhanced activity are associated with a phase of sodium retention; however, because of the phenomenon of “mineralocorticoid escape,” the clinical manifestation is generally that of hypertension rather than hypervolemia. Both heart failure (Chapter 58) and cirrhosis (Chapter 153) are associated with renal sodium retention.

PATHOBIOLOGY

Two pathophysiologic mechanisms can lead to sodium retention with ECF volume expansion. The first involves renal sodium retention that is primary and unrelated to the activation of afferent sensor mechanisms. This category includes primary renal diseases and endocrine disorders characterized by excess mineralocorticoid action. In the second category, EABV is reduced, and afferent sensory mechanisms activate effector responses that drive renal sodium retention. In these conditions, total ECF volume is expanded, but intravascular volume is contracted. Therefore, the volume homeostatic mechanisms of the body mimic those of hypovolemia because of the perception of reduced EABV. The degree of solute-free water retention that accompanies the sodium surfeit has a relatively small influence on the extent of hypervolemia but influences the accompanying tonicity state and determines whether the hypervolemia is hypotonic or isotonic. When the ECF volume is expanded, the relative distribution between the intravascular and extravascular (interstitial) compartments depends on a number of factors. When cardiac and hepatic functions are normal and peripheral transcapillary Starling forces are intact, the excess ECF volume is evenly distributed between the intravascular and interstitial fluid compartments. In such cases, edema does not occur until there is a substantial surfeit of sodium, and hypertension is expected. In contrast, concomitant disruption of transcapillary Starling forces in a given microcirculatory bed would favor the accumulation of retained fluid at one or more such interstitial locations (e.g., dependent edema progressing to anasarca, ascites, pleural effusion, pulmonary congestion).

Primary Renal Sodium Retention

Patients who retain ingested or administered sodium and water loads expand their ECF volume. In patients with chronic kidney disease, the filtered load of sodium remains well above dietary intake until very late stages of severely reduced GFR; even when the GFR is decreased by as much as 90%, the daily filtered load of approximately 2400 mmol still greatly exceeds dietary intake. Nevertheless, the relationship between tubular reabsorption and filtered load may be disrupted in kidney disease. Monogenic disorders that cause or mimic enhanced mineralocorticoid activity or are associated with enhanced activity of the distal nephron

CHAPTER 116  Disorders of Sodium and Water Homeostasis  

sodium reabsorptive pathways include Liddle syndrome and pseudohy­ poaldosteronism type 2 (Chapters 67, 117, and 128). In these conditions and in other causes of mineralocorticoid excess, the only clue to mild hypervolemia may be hypertension, which can be severe. Mineralocorticoid excess, glucocorticoid-remediable hypertension, apparent mineralocorticoid excess, and Liddle syndrome are associated with hypokalemia, whereas pseudohypoaldosteronism type 2 (Gordon syndrome) is often accompanied by hyperkalemia.

Secondary Renal Sodium Retention

With both low-output and high-output heart failure and both systolic and diastolic dysfunction, sodium retention is typical (Chapter 58). Low cardiac output, diversion of cardiac output away from arterial intravascular volume– sensing sites, or a high cardiac output that still is not sufficient to meet tissue demands appears to be a necessary and sufficient condition for initiating renal sodium retention. In the case of cirrhosis with ascites (Chapter 153), hepatic intrasinusoidal hypertension is a sufficient and necessary condition for initiating renal sodium retention. These pathophysiologic disturbances in cardiac or hepatic function disrupt afferent signals that govern normal sodium homeostasis and trigger effector mechanisms that lead to enhanced tubular reabsorption of sodium at multiple nephron sites. At the very earliest stages of disease, sodium retention occurs independently of any measurable or detectable reduction in the volume of the intravascular compartments or any of its measurable subcompartments. At more advanced stages of disease, reduced intravascular volume serves as the overriding stimulus for renal sodium retention and thereby leads to a decompensated state of intractable ECF volume accumulation. The more advanced stages, which often are accompanied by a disproportionate degree of positive water balance and consequent hyponatremia, herald imminent compromise of the GFR. Among the many neuronal and humoral abnormalities that characterize the sodium retention associated with heart failure and cirrhosis are endothelial dysfunction, enhanced sympathetic nerve activity, activation of the reninangiotensin-aldosterone axis, and resistance to natriuretic peptides. In cirrhosis with ascites (Chapter 153), portosystemic shunting together with translocation of intravascular volume to the splanchnic and venous circulation further compromises EABV. In addition, synthetic dysfunction with resulting hypoalbuminemia favors transudation of fluid into the interstitial compartment. At the level of intrahepatic hemodynamics, intrasinusoidal hypertension results in enhanced hepatic lymph formation. When the rate of enhanced hepatic lymph formation exceeds the capacity for return to the intravascular compartment through the thoracic duct, hepatic lymph accumulates in the form of ascites, and the intravascular compartment is further compromised.

CLINICAL MANIFESTATIONS

In addition to the clinical manifestations of the underlying disease, the clinical manifestations of hypervolemia depend on the amount and relative distribution of accumulated fluid in the various ECF subcompartments, including the venous and arterial components of the intravascular compartment (jugular venous distention and hypertension), the interstitial spaces of the extremities, the subcutaneous tissues of the lower back and the periorbital region (peripheral pitting edema, the predominant location of which depends on the patient’s position), the peritoneal and pleural spaces (ascites and pleural effusion, respectively), and the alveolar space (pulmonary edema). When cardiac and hepatic function are normal and the transcapillary Starling forces are not disrupted, the excess volume is distributed proportionately throughout the ECF compartment. Hypertension may be an early manifestation, depending on cardiac function and the state of systemic vascular resistance. Jugular venous distention (see Fig. 51-3) and peripheral edema (see Fig. 51-7) may be present. Clinically detectable pitting peripheral edema usually signifies the accumulation of at least 3 L of excess interstitial volume. Because intravascular plasma volume is itself only 3 L, any state of generalized peripheral edema must signify ECF volume expansion and therefore past or ongoing renal sodium retention or both. When cardiac function is impaired because of myocardial disease, valvular disease, or pericardial disease, pulmonary and systemic venous hypertension predominates and systemic arterial pressure may be low as a result of disproportionate accumulation of intravascular volume in the venous as opposed to the arterial circulation (Chapter 58). The presence of transudative ascites (see Fig. 146-4) signifies the substantial accumulation of excess ECF volume in the peritoneal cavity, most commonly secondary to disruption of intrahepatic hemodynamics in the setting of liver disease. Pleural effusions can also

747

be a manifestation of hypervolemia, particularly in the setting of heart failure or advanced cirrhosis with ascites.

DIAGNOSIS

Hypervolemia usually is easily detected by findings of generalized edema, ascites, elevated jugular venous pressure, inspiratory pulmonary crackles, or evidence of the presence of pleural effusion. The prevailing systemic arterial blood pressure often provides a clue about whether the hypervolemic state is secondary to reduced EABV or instead due to primary renal sodium retention. The history and physical examination are often sufficient to yield the diagnosis of an underlying secondary cause of sodium retention, such as heart failure or cirrhosis. Adjunctive laboratory tests providing evidence of cardiac dysfunction or liver disease may be helpful. The presence of glomerular-range proteinuria with hypoalbuminemia indicates a glomerular cause of the sodium retention and hypervolemia. Elevated creatinine points to renal failure, which can be intrinsic or may occur in association with advanced stages of some of the aforementioned conditions, such as heart failure (cardiorenal syndrome) or hepatic cirrhosis (hepatorenal failure). Hypoalbuminemia is characteristic of both cirrhosis and nephrotic syndrome. A low urine sodium concentration and low fractional excretion of sodium confirm renal sodium retention secondary to a perceived decrease in EABV in the edema states, even in the face of overall hypervolemia. More recently, elevated concentrations of brain natriuretic peptide have been used to support the diagnosis of hypervolemia, particularly in the setting of cardiac failure and renal disease.

TREATMENT  The most important step in ameliorating renal sodium retention is recognition and treatment of the underlying disease. Optimization of hemodynamic parameters in heart failure (Chapter 59), improvement of liver function (Chapter 154), or remission of nephrotic syndrome (Chapter 121) improves or reverses sodium retention. Therapeutic intervention to reduce ECF volume without addressing the underlying disease is often met by complications, especially when ECF volume expansion is associated with decreased intravascular volume or EABV. Nevertheless, three treatment modalities are available to reduce ECF volume directly by inducing negative sodium balance: dietary sodium restriction, diuretics, and extracorporeal fluid removal by ultrafiltration. The modality and the desired rate of sodium removal vary with the clinical setting and depend on the relative distribution of the sodium surfeit and excess volume in the body fluid compartments. Therefore, before initiating any treatment, the clinician should identify the specific disturbances in clinical parameters that are harmful to the patient and monitor the improvement in these parameters during the course of treatment. Harmful manifestations of hypervolemia include hypertension, pulmonary congestion and edema or pleural effusions with compromised respiratory function, hepatic congestion and ascites, and degrees of peripheral edema that compromise skin integrity and predispose the patient to cellulitis. Once ECF volume reduction has removed these threats to the patient’s well-being, rates of sodium removal should be slowed significantly. Thus, a patient with mild peripheral edema, small pleural effusions, minimal ascites, jugular venous distention, and normal blood pressure might be managed with sodium restriction and limited use of natriuretic medications to induce a gradual negative sodium balance during a period of many days to weeks. In contrast, a patient with limb- or lifethreatening anasarca, pulmonary congestion, or hypervolemia-induced hypertension might require the continuous intravenous infusion of natriuretics or in some cases extracorporeal ultrafiltration therapy.

Sodium Restriction

In the management of chronic hypervolemia, other modalities are futile if they are not accompanied by restriction of sodium intake because renal sodium avidity results in the reaccumulation of ECF fluid as soon as the influence of diuretics has ceased. Dietary sodium restriction in the range of 50 to 100 mmol/day is often recommended and requires abstention from added salt as well as from foods rich in sodium. In acute decompensated heart failure, however, sodium restriction does not augment negative fluid balance over what can be achieved by furosemide alone. A4  Sodium substitutes can be useful, although caution needs to be exercised in patients with a tendency  to hyperkalemia because some salt substitutes contain potassium. Calorie intake and nutritional parameters should be monitored to ensure that an overly draconian diet does not induce protein-energy malnutrition. In hospitalized patients, it is particularly important to ensure that the sodium content of administered intravenous fluids and sodium-containing medications is monitored and reduced to the minimum possible. The practice of infusing sodiumcontaining solutions on the one hand and simultaneously treating with diuretics has no sound physiologic or therapeutic basis. In one randomized trial, however, the combination of high-dose furosemide and small doses of

748

CHAPTER 116  Disorders of Sodium and Water Homeostasis  

hypertonic saline was better than furosemide alone for treatment of patients with refractory heart failure A5  or with ascites. A6  Water restriction is not appropriate in hypervolemic edema states unless the plasma sodium concentration is less than 135 mmol/L or symptomatic hyponatremia supervenes.

diuretics. Proximal tubule natriuretics rarely are used as primary therapy but are used as supplements to loop natriuretics when loop natriuretics alone are insufficiently effective.

Diuretics

Loop natriuretics, such as furosemide, bumetanide, torsemide, and ethacrynic acid, induce natriuresis by inhibiting the coupled entry of Na+, Cl−, and K+ across apical plasma membranes in the thick ascending limb of the loop of Henle, which is responsible for the reabsorption of approximately 25% of filtered sodium. Loop diuretics, which are the most potent diuretics, continue to be effective even in patients with relatively compromised kidney function.

Loop Natriuretics

Diuretics and natriuretics (Table 116-5) enhance the urinary excretion of sodium-containing fluid by inhibiting tubular reabsorption at specific nephron sites (Fig. 116-3).

Proximal Tubule Natriuretics

The cardinal example of a proximal tubule natriuretic is acetazolamide, a carbonic anhydrase inhibitor that blocks proximal reabsorption of sodium bicarbonate. Consequently, prolonged use of acetazolamide may lead to hyperchloremic acidosis, in contrast to all other natriuretics, which act at loci before the late distal nephron. Metolazone, a congener of the thiazide class of natriuretics, blocks sodium chloride absorption in the proximal tubule as well as in the early distal tubule. Because the major locus for phosphate absorption is in the proximal nephron, the phosphaturia accompanying metolazone administration considerably exceeds that observed with other thiazide-class

Distal Tubule Natriuretics

Distal tubule natriuretics, such as hydrochlorothiazide, chlorthalidone, and metolazone, interfere primarily with sodium chloride absorption in the earliest segments of the distal convoluted tubule, where they block the sodium chloride cotransport mechanism across apical plasma membranes. Distal tubule natriuretics generally are used in the same conditions as loop natriuretics are, but not in chronic kidney disease and in disorders of calcium metabolism. Whereas loop natriuretics are calciuretic and are valuable for managing acute

TABLE 116-5  DIURETICS AND OTHER NATRIURETIC MEDICATIONS DIURETICS IN COMMON USE

DAILY DOSE RANGE

ADVERSE REACTIONS

Thiazides (oral)   Hydrochlorothiazide   Metolazone   Chlorthalidone

COMMENTS

Rash, neutropenia, thrombocytopenia, hyperglycemia, hyperuricemia

25-100 mg 2.5-5 mg 20-50 mg

Usually not effective below GFR of 30-40 mL/mm (metolazone, 20-30 mL/mm)

Loop diuretics (oral or intravenous)

  Furosemide   Bumetanide   Torsemide

20-320 mg 1-8 mg 20-200 mg

Potassium sparing   Spironolactone   Triamterene   Amiloride   Eplerenone

25-400 mg 25-100 mg 5-20 mg 25-50 mg

Rapid onset, short duration Split doses in normal renal function; give intravenously in acute situations or if reduced gastrointestinal absorption Can use up to 500 mg furosemide (or equivalent) in severe renal insufficiency

Ototoxicity at high doses Hyperkalemia

Not very potent

GFR = glomerular filtration rate.

+

(1) Proximal convoluted tubule (2) Amino acids PARS RECTA

Glucose

(2)

(4) Distal convoluted tubule –

–

Cortical thick ascending limb

Na+

(1) NaHCO3

+ +

(3)

Organic acids

+ 2

– – –

–

Na+ Cl–

Cl– Na+

+ Na+, K+ + +

Diuretics (1)

Afferent arteriole

1. Osmotic diuretics 2. Carbonic anhydrase inhibitors 3. Loop diuretics 4. Thiazides 5. Potassium sparing

CORTEX MEDULLA Medullary thick

Descending limb

(1) (3)

ascending limb (3) + 2 Cl–

+ Na+, K+ H 2O H2O

Thin ascending limb UREA=NaCl

H+ K+

– – –

H2O with ADH

– – (5) – Aldosterone – stimulates – H+

+ + Collecting + duct H2O with ADH

+

UREA NaCl

FIGURE 116-3.  Major transport processes along the nephron segments and primary sites of action of diuretics. The site of action of diuretics is shown by numbers in parentheses in each nephron segment; numbers correspond to those next to the diuretics listed in the lower left section of the figure. ADH = antidiuretic hormone. (From Kokko JP. Diuretics. In: Alexander RW, Schlant RC, Fuster V, eds. The Heart. 9th ed. New York: McGraw-Hill; 1998.)

CHAPTER 116  Disorders of Sodium and Water Homeostasis  

hypercalcemia (Chapter 245), thiazide natriuretics promote hypocalciuria and calcium retention and are useful in managing hypercalciuric states. With the exception of acetazolamide, which impairs bicarbonate absorption, the natriuretics discussed so far can cause hypokalemia and metabolic alkalosis.

Collecting Duct Natriuretics

Spironolactone and eplerenone compete with aldosterone and inhibit sodium absorption in the collecting duct while concomitantly suppressing potassium and proton secretion. Triamterene and amiloride directly block sodium uptake by collecting duct cells and concomitantly suppress potassium and proton secretion. These agents are used in combination with thiazide and loop natriuretics to offset hypokalemia. However, hyperkalemia and hyperchloremic metabolic acidosis may complicate the injudicious use of any of these agents. Spironolactone and eplerenone are useful in managing disorders characterized by secondary hyperaldosteronism (such as cirrhosis with ascites), in promoting natriuresis in hypokalemic patients, and in competitively blocking nonepithelial mineralocorticoid receptors in patients with left ventricular dysfunction (Chapter 59). Nesiritide is the recombinant version of a naturally occurring brain natriuretic peptide with unique vasodilator and natriuretic actions. Nesiritide  given alone may compromise renal function, especially when high levels are achieved after an initial bolus, so its usefulness appears to be limited; frequent monitoring of urine output and of plasma urea and creatinine concentrations is required. However, a recent randomized trial showed therapeutic benefit with the combination of an angiotensin-converting enzyme inhibitor (enalapril) and the endopeptidase inhibitor AHU-377 (a neprilysin inhibitor prodrug, which blocks the breakdown and, hence, augments the levels and activity of endogenous natriuretic peptides) for patients with heart failure. A7  Patients with severe degrees of renal sodium avidity can be resistant to conventionally recommended doses of individual classes of diuretic agents. In such patients, combinations of diuretic agents acting at different sites along the nephron may overcome this resistance and induce a natriuretic response.6 The continuous intravenous infusion of furosemide, sometimes in conjunction with intermittent bolus infusions of albumin, also can overcome natriuretic resistance in some hospitalized patients. Monitoring of plasma sodium, potassium, magnesium, calcium, and phosphate concentrations is mandatory in patients treated with high or frequent doses or continuous infusions of  natriuretic agents. Besides body tonicity and electrolyte disturbances, other potential adverse consequences include a reduction in GFR. Drug-specific idiosyncratic adverse responses, such as allergic cutaneous reactions, interstitial nephritis, pancreatitis, and blood dyscrasias, are much less common.

Extracorporeal Ultrafiltration

In a small subset of patients, either superimposed renal impairment or extreme resistance to natriuretic action may require the direct removal of excess ECF volume by ultrafiltration, hemodialysis, or peritoneal dialysis (Chapter 131). Chronic ambulatory peritoneal dialysis has been used for the symptomatic relief of pulmonary congestion and anasarca in some patients with chronic heart failure who are unresponsive to other therapeutic modalities and are not candidates for cardiac transplantation.

  WATER BALANCE DISORDERS

Water balance disorders generally come to medical attention because of one or more of three clinical manifestations: hyponatremia,7 hypernatremia,8 or polyuria.

Hyponatremia

DEFINITION

Hyponatremia, which is defined as a plasma sodium concentration of less than 136 mmol/L, is the most frequently encountered electrolyte abnormality in hospitalized patients. Hyponatremia, irrespective of the underlying cause, is independently associated with higher mortality.9

EPIDEMIOLOGY AND PATHOBIOLOGY

Hyponatremia may be hypertonic, isotonic, or hypotonic. Hypertonic hyponatremia occurs when there is an accumulation in the ECF compartment of non–sodium-containing effective solutes, such as very high concentrations of glucose in diabetic patients or exogenously administered mannitol or glycerol. These non-sodium solutes lead to a shift of water from the ICF to the ECF compartments and consequent ICF shrinkage. The accumulation of a solute such as urea, which contributes to the measured plasma osmolality but is not an osmotically effective solute in terms of transcellular water shift, should not be included in the category of hypertonic hyponatremic states. Isotonic hyponatremia signifies the laboratory finding of hyponatremia in patients with no disturbances in body fluid tonicity and almost always reflects

749

the interference of marked hyperlipidemia or marked hyperglobulinemia with certain laboratory techniques for the measurement of the plasma sodium concentration; these situations are termed pseudohyponatremia and should always be excluded before embarking on diagnostic or therapeutic measures to alter water balance or body tonicity. True hypotonic hyponatremia always reflects an important underlying disorder that leads to abnormal body water retention and either past or ongoing expansion of ICF volume. Even in chronic hypotonic hyponatremic states in which cell volume has been restored to normal by osmotic adaptive mechanisms, the compensation occurs at the price of loss of intracellular solutes and compromised cell function. Hypotonic hyponatremia can be further classified according to volume status. Heart failure (Chapter 59) and cirrhosis with ascites (Chapter 153) are examples of hypervolemic hyponatremia. In these conditions, reduced EABV stimulates the release of AVP and also may limit the delivery of glomerular ultrafiltrate to the diluting segments of the nephron, thereby leading to impaired water excretion. The hyponatremia that is seen in advanced renal failure (Chapter 130) because of impaired excretion of water may also be associated with hypervolemia. Hypovolemic hyponatremia occurs when relatively more sodium than water is lost through the gastrointestinal tract (e.g., diarrhea) or in urine (e.g., thiazide diuretics). Decreased body tonicity can also develop even when fluid loss is isotonic or hypotonic (e.g., sweat) if the ingested or administered replacement fluid is more hypotonic than the lost fluid (e.g., ingestion of water or intravenous administration of D5W). Another important consideration is the potassium concentration in the lost fluid. For example, diarrheal fluid and natriuretic medication–induced polyuric urine are often rich in potassium as well as in sodium. Because water moves freely across most cell membranes, the ECF and ICF are in osmotic equilibrium. As a result, plasma tonicity is equal to the sum of the effective osmolalities of the ICF and total body water. Given that sodium is the principal determinant of ECF tonicity and potassium is the main contributor to ICF tonicity, plasma sodium is directly proportional to the sum of sodium and potassium concentrations in total body water. Therefore, even if the concentration of lost sodium alone is less than that in the ECF, combined loss of sodium plus potassium can cause hypotonicity.

CLINICAL MANIFESTATIONS

The finding of hyponatremia is often incidental on routine laboratory testing, on laboratory testing of patients with nonspecific complaints, or as part of the investigation of other clinical syndromes. The symptoms of hypotonic hyponatremia depend on its duration, severity, and rate of development. When hyponatremia develops rapidly (hours to days), acute brain swelling or cerebral edema occurs and is manifested as headache, lethargy, seizures, and a progressively decreased level of consciousness that can lead to coma and death. In addition, women between menarche and menopause are particularly susceptible to the life-threatening neurologic manifestations of acute hyponatremia, even of relatively mild degree. In contrast, when the rate of decline in plasma sodium concentration is more gradual, osmotic adaptation ensues, and even severe hyponatremia (plasma sodium concentration 20

Renal Losses Diuretics Mineralocorticoid deficiency Salt-losing nephropathy Bicarbonaturia with RTA and metabolic alkalosis Cerebral salt wasting

Hypervolemia • Edema • Total body water • Total body sodium

Euvolemia • No edema • Total body water • Total body sodium

U[Na] < 20

U[Na] > 20

U[Na] > 20

Extrarenal Losses Vomiting Diarrhea Third spacing of fluids • Burns • Pancreatitis • Trauma

Glucocorticoid deficiency Hypothyroidism Stress Drugs SIADH

Acute or chronic renal failure

U[Na] < 20

Nephrotic syndrome Cirrhosis Cardiac failure

FIGURE 116-4.  Diagnostic approach to hyponatremia. RTA = renal tubular acidosis; SIADH = syndrome of inappropriate antidiuretic hormone secretion. (Modified from Halterman R, Berl T. Therapy of dysnatremic disorders. In: Brady H, Wilcox C, eds. Therapy in Nephrology and Hypertension. Philadelphia: Saunders; 1999:256.)

natriuretic medication use or gastrointestinal fluid losses, suggest hypovolemia, but the absence of these findings does not exclude hypovolemia. Further laboratory tests should include a repeated set of plasma electrolyte concentrations, including potassium and chloride levels, which together with determination of acid-base parameters (pH, Pco2, and bicarbonate) can point to processes not always detectable in the history, such as vomiting, diarrhea, or natriuretic medication exposure. Other laboratory tests should include liver function tests and measurement of plasma urea, creatinine, uric acid, thyroid-stimulating hormone, and cortisol concentrations and, if indicated, an adrenocorticotropic hormone stimulation test. High levels of both urea and creatinine point to intrinsic renal disease, whereas a disproportionate elevation of urea over creatinine might support hypovolemia with a tendency to prerenal azotemia (Chapter 120). In contrast, very low levels of urea and uric acid are typical of both the syndrome of inappropriate antidiuretic hormone secretion (SIADH) and the cerebral salt-wasting syndrome. Marked elevation in the plasma glucose concentration increases both measured and calculated plasma osmolality and indicates a state of hypertonic hyponatremia that should be approached as a state of body fluid hypertonicity with cell shrinkage (see later) rather than hypotonicity. The plasma sodium concentration declines by approximately 1.6 mmol/L for each increase of 100 mg/dL (5.5 mmol/L) in plasma glucose concentration. However, this decline in plasma sodium is variable and is greater in states of progressively severe hyperglycemia. In contrast to hyperglycemia, an elevated urea concentration should not be considered as contributing to plasma or ECF tonicity, even though urea does contribute to the laboratory measurement of plasma osmolality. Thus, a hyponatremic patient with a normal or elevated laboratory measurement of plasma osmolality that can be fully attributed to an increased urea concentration should be considered as having hypotonic hyponatremia. A discrepancy in which measured plasma osmolality exceeds calculated plasma osmolality and that cannot be attributed to either glucose or urea indicates the presence of an unidentified small solute (osmolar gap), including alcohols (e.g., ethanol, methanol, ethylene glycol, and isopropyl alcohol) and the organic anions of weak acids, which raise the plasma anion gap. Because these small molecules are not effective solutes in terms of water movement, the water balance and tonicity status of the patient is determined by the plasma sodium concentration. Just as for urea, a patient with hyponatremia and normal or elevated measured plasma osmolality as a result of one of these small solutes should be approached as having a true hypotonic hyponatremia, notwithstanding the normal or elevated plasma osmolality measurement. However, the finding of such an osmolar gap should prompt a thorough investigation for poisoning, intoxication, or an organic acidosis (Chapter 118). Once a state of true hypotonic hyponatremia has been established, determination of the cause and further diagnostic approach follow a classification into one of three categories based on assessment of the volume status of the patient (Fig. 116-4). Abnormal liver function test results can provide

adjunctive support for hepatic disease and a hypervolemic hyponatremic state. The diagnosis of heart failure should be made clinically, but it can be assisted by a brain natriuretic peptide level, chest radiograph, or echocardiograph (Chapter 58). A radiograph or chest computed tomography scan may help identify intrathoracic lesions that are associated with SIADH. Approximately 85% of hyponatremic inpatients have true hypotonic hyponatremia. Among these patients, about 25% are hypovolemic, about 25% have an edema state, about one third are normovolemic, and most of the remainder have renal failure. In the absence of a clinically obvious edema state, a low urine sodium concentration (