Equine internal medicine [PDF] | Online Book Share (2024)

1,841 Pages • 928,768 Words • PDF • 44.8 MB

Equine Internal Medicine Third Edition Stephen M. Reed, DVM, Dipl ACVIM Associate • Rood & Riddle Equine Hospital • Lexington, Kentucky Warwick M. Bayly, BVSc, MS, PhD, Dipl ACVIM Provost and Executive Vice President • Washington State University • Pullman, Washington Debra C. Sellon, DVM, PhD, Dipl ACVIM Professor, Department of Veterinary Clinical Sciences • College of Veterinary Medicine, Washington State University • Pullman, Washington

SAUNDERS

Front Matter Equine Internal Medicine THIRD EDITION

Stephen M. Reed, DVM, Dipl ACVIM Associate • Rood & Riddle Equine Hospital • Lexington, Kentucky Warwick M. Bayly, BVSc, MS, PhD, Dipl ACVIM Provost and Executive Vice President • Washington State University • Pullman, Washington Debra C. Sellon, DVM, PhD, Dipl ACVIM Professor, Department of Veterinary Clinical Sciences • College of Veterinary Medicine, Washington State University • Pullman, Washington

Copyright 3251 Riverport Lane St. Louis, MO 63043 EQUINE INTERNAL MEDICINE ISBN: 978-1-4160-5670-6 Copyright © 2010, by Saunders, an imprint of 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 photocopy, 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 eld 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 identi ed, 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 the practitioner, relying on their own experience and knowledge of their patient, 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. The Publisher

Previous editions copyrighted 2004, 1998 Library of Congress Cataloging-in-Publication Data Equine internal medicine/[edited by] Stephen M. Reed, Warwick M. Bayly, Debra C. Sellon. -- 3rd ed. p.; cm. Includes bibliographical references and index. ISBN 978-1-4160-5670-6 (pbk.: alk. paper) 1. Horses--Diseases. 2. Veterinary internal medicine. I. Reed, Stephen M. II. Bayly, Warwick M. III. Sellon, Debra C. [DNLM: 1. Horse Diseases. SF 951 E638 2010] SF951.E565 1998 636.1’0896--dc22 2009036106 Vice President and Publisher: Linda Duncan Acquisitions Editor: Penny Rudolph Associate Developmental Editor: Lauren Harms Publishing Services Manager: Patricia Tannian Project Managers: Jonathan Taylor, Sharon Corell Design Direction: Karen Pauls Printed in the United States of America Last digit is the print number: 9 8 7 6 5 4 32

Contributors Dorothy M. Ainsworth, DVM, PhD, DACVIM, Professor of Medicine, Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, New York Frank M. Andrews, DVM, MS, DACVIM, LVMA Equine Committee, Professor and Director, Equine Health Studies Program, School of Veterinary Medicine, Louisiana State University, Baton Rouge, Louisiana Bonnie S. Barr, VMD, DACVIM, Department of Internal Medicine, Rood and Riddle Equine Hospital, Lexington, Kentucky Michelle Henry Barton, DVM, PhD, DACVIM, Fuller E. Callaway Chair and Professor, Department of Large Animal Medicine, University of Georgia, Athens, Georgia Warwick M.. Bayly, BVSc, MS, PhD, Dipl ACVIM, Provost and Executive Vice President, Washington State University, Pullman, Washington Laurie A. Beard, DVM, MS, DACVIM, Clinical Associate Professor, Department of Clinical Sciences, Veterinary Medical Teaching Hospital, Kansas State University, Manhattan, Kansas Joseph J. Bertone, DVM, MS, DACVIM, Professor, College of Veterinary Medicine, Western University of Health Sciences, Pomona, California Anthony T. Blikslager, DVM, PhD, DACVS, Professor, Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina John D. Bonagura, DVM, MS, DACVIM, Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Barbara A. Byrne, DVM, PhD, DACVIM, Assistant Professor, Department of Pathology, Microbiology, and Immunology, College of Veterinary Medicine, University of California, Davis, California Elaine M. Carnevale, DVM, PhD, Department of Biomedical Sciences, Animal Reproduction and Biotechnology Laboratory, Colorado State University, Fort Collins, Colorado Jonathan Cheetham, Vet MB, PhD, DACVS, Department of Comparative Biomedical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, New York John B. Chopin, BVSc, MACVSc (Equine Medicine), PhD, FACVSc (Equine Reproduction), Coolmore Stud, Jerry's Plains, New South Wales, Australia Noah D. Cohen, VMD, MPH, PhD, DACVIM, Professor, Department of Large Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, Texas Marco A. Coutinho da Silva, DVM, PhD, DACT, Assistant Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Mark V. Crisman, DVM, MS, DACVIM, Professor, Department of Large Animal Clinical Sciences, College of Veterinary Medicine, Virginia Maryland Regional College of Veterinary Medicine, Blacksburg, Virginia Jennifer L. Davis, DVM, PhD, DACVIM, DACVCP, Assistant Professor, Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina Patricia M. Dowling, DVM, MSc, DACVIM, DACVCP, Department of Veterinary Biomedical Sciences, Western College of Veterinary Medicine, Saskatoon, Saskatchewan, Canada Susan C. Eades, DVM, PhD, DACVIM, Professor, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University, Baton Rouge, Louisiana Jonathan H. Foreman, DVM, MS, DACVIM, Associate Dean for Academic and Student Affairs, Professor, Department of Veterinary Clinical Medicine, Veterinary Teaching Hospital, University of Illinois, Urbana, Illinois Nicholas Frank, DVM, PhD, DACVIM, Associate Professor, Department of Large Animal Clinical Sciences, University of Tennessee, Knoxville, Tennessee, Associate Professor, School of Veterinary Medicine and Sciences, University of Nottingham, Sutton Bonington, United Kingdom

Grant S. Frazer, BVSc, MS, DipACT, MBA, Professor of Veterinary Reproduction (Theriogenology), Director, Veterinary Clinical and Diagnostic Services, School of Veterinary Science, The University of Queensland, St. Lucia, Queensland, Australia Martin Furr, DVM, PhD, DACVIM, Adelaide C. Riggs Professor of Medicine, Virginia-Maryland Regional College of Veterinary Medicine, Leesburg, Virginia Katherine S. Garrett, DVM, Department of Diagnostic Imaging, Rood & Riddle Equine Hospital, Lexington, Kentucky Ray J. Geor, BVSc, MVSc, PhD, DACVIM, Professor and Chairperson, Department of Large Animal Clinical Sciences, College of Veterinary Medicine, Michigan State University, East Lansing, Michigan Caroline N. Hahn, DVM, MSc, PhD, DECVN, DECEIM, MRCVS, Director, Neuromuscular Disease Laboratory, Royal (Dick) School of Veterinary Studies, Easter Bush Veterinary Centre, University of Edinburgh, Midlothian, United Kingdom Bernard D. Hansen, DVM, MS, DACVIM, DACVECC, Associate Professor, Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina Joanne Hardy, DVM, PhD, DACVS, Clinical Associate Professor, Department of Large Animal Clinical Science, College of Veterinary Medicine, Texas A&M University, College Station, Texas Kenneth W. Hinchcliff, BVSc, PhD, DACVIM, Dean of Veterinary Science, University of Melbourne, Werribee, Victoria, Australia Melissa T. Hines, DVM, PhD, Associate Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University, Pullman, Washington Siddra A. Hines, DVM, Equine Internal Medicine Resident, Veterinary Teaching Hospital, Washington State University,Pullman, Washington David W. Horohov, MS, PhD, William Robert Mills Chair and Professor, Department of Veterinary Sciences, University of Kentucky, Lexington, Kentucky Laura H. Javsicas, VMD, DACVIM, Lecturer, Department of Large Animal Medicine, College of Veterinary Medicine, University of Florida, Gainesville, Florida Samuel L. Jones, DVM, PhD, DACVIM, Professor, Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina Eduard Jose-Cunilleras, DVM, DACVIM, Clinical Instructor, Department of Veterinary Clinical Science, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Thomas R. Klei, BS, PhD, Boyd Professor of Parasitology and Veterinary Science, Associate Dean for Research and Academic Affairs, Louisiana State University, Baton Rouge, Louisiana Catherine W. Kohn, VMD, DACVIM, Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Véronique A. Lacombe, DVM, PhD, DACVIM, Research Assistant Professor, College of Pharmacy, Adjunct Assistant Professor, Department of Veterinary Clinical Sciences, Investigator, Davis Heart and Lung Research Institute, The Ohio State University, Columbus, Ohio Katharina L. Lohmann, Dr.med.vet., PhD, DACVIM, Associate Professor, Department of Large Animal Clinical Sciences, Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada Maureen T. Long, DVM, PhD, DACVIM, Associate Professor, Department of Infectious Disease and Pathology, College of Veterinary Medicine, University of Florida, Gainesville, Florida D. Paul Lunn, BVSc, MS, PhD, MRCVS, DACVIM, Professor and Head, Department of Clinical Sciences, James L. Voss Veterinary Teaching Hospital, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado Jennifer M. MacLeay, DVM, PhD, DACVIM, Associate Professor, Department of Clinical Sciences, Colorado State University, Fort Collins, Colorado Peggy S. Marsh, DVM, DACVIM, DACVECC, Internist, McGee Medicine Center, Hagyard Equine Medical Center, Lexington, Kentucky Dianne McFarlane, DVM, PhD, DACVIM, Assistant Professor, Department of Physiological Sciences, Center for Veterinary Health Sciences, Oklahoma State University, Stillwater, Oklahoma

Robert H. Mealey, DVM, PhD, DACVIM, Associate Professor, Department of Veterinary Microbiology and Pathology, College of Veterinary Medicine, Washington State University, Pullman, Washington Elizabeth S. Metcalf, MS, DVM, DACT, Owner, Honahlee, PC, Sherwood, Oregon Rustin M. Moore, DVM, PhD, DACVS, Bud and Marilyn Jenne Professor and Chair, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State Unviersity, Columbus, Ohio Peter R. Morresey, BVSc, DACT, DACVIM, Department of Internal Medicine, Rood and Riddle Equine Hospital, Lexington, Kentucky William W. Muir, DVM, PhD, DACVA, DACVECC, Regional Director, American Academy of Pain Management, Veterinary Clinical Pharmacology Consulting Services, Columbus, Ohio Yvette S. Nout, DVM, MS, PhD, DACVIM, DACVECC, Assistant Researcher, Department of Neurosurgery, College of Veterinary Medicine, University of California, San Francisco, California J. Lindsay Oaks, DVM, PhD, DACVM, Associate Professor, Department of Veterinary Microbiology and Pathology, College of Veterinary Medicine, Washington State University, Pullman, Washington Dale L. Paccamonti, DVM, MS, DACT, Professor and Head, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University, Baton Rouge, Louisiana Nigel R. Perkins, BVSc (Hons), MS, PhD, DACT, FACVSc, Director, AusVet Animal Health Services, Toowoomba, Queensland, Australia Carlos R.F. Pinto, DVM, PhD, DACT, Associate Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Michael B. Porter, DVM, PhD, DACVIM, Clinical Assistant Professor, Department of Large Animal Clinical Sciences, College of Veterinary Medicine, University of Florida, Gainesville, Florida Nicola Pusterla, DVM, DACVIM, Associate Professor, Department of Medicine and Epidemiology, College of Veterinary Medicine, University of California, Davis, California Stephen M. Reed, DVM, Dipl ACVIM, Associate, Rood & Riddle Equine Hospital, Lexington, Kentucky Virginia B. Reef, DVM, Director of Large Animal Cardiology and Diagnostic Ultrasonography, Executive Board Member, Center for Equine Sport Medicine, Chief, Section of Sports Medicine and Imaging, Mark Whittier and Lila Griswold Allam Professor of Medicine, Department of Clinical Studies, School of Veterinary Medicine, New Bolton Center, University of Pennsylvania, Kennett Square, Pennsylvania Christine A. Rees, DVM, DACVD, Veterinary Dermatology, Veterinary Specialists of North Texas, Dallas, Texas Bonnie R. Rush, DVM, MS, DACVIM, Professor, Department of Clinical Sciences, College of Veterinary Medicine, Kansas State University, Manhattan, Kansas Juan C. Samper, DVM, MSc, PhD, DACT, Veterinary Reproductive Services, Abbotsford, British Columbia, Canada L. Chris Sanchez, DVM, PhD, DACVIM, Assistant Professor, Department of Large Animal Clinical Sciences, University of Florida, Gainesville, Florida William J. Saville, DVM, Professor and Chair, Department of Veterinary Preventative Medicine, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Harold C. Schott, II, DVM, PhD, DACVIM, Professor, Department of Large Animal Clinical Sciences, College of Veterinary Medicine, Michigan State University, East Lansing, Michigan Colin C. Schwarzwald, Dr.med.vet., PhD, DACVIM, Senior Lecturer, Internal Medicine Section, Equine Department, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland Kathy K. Seino, DVM, MS, PhD, Assistant Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University, Pullman, Washington Debra C. Sellon, DVM, PhD, DACVIM, Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University, Pullman, Washington

Daniel C. Sharp, III, BS, MS, PhD, Professor Emeritus, Department of Large Animal Medicine, College of Veterinary Medicine, University of Florida, Gainesville, Florida Carla S. Sommardahl, DVM, PhD, DACVIM, Assistant Professor, Department of Large Animal Medicine, College of Veterinary Medicine, University of Tennessee, Knoxville, Tennessee Ashley M. Stokes, DVM, PhD, Associate Professor and Extension Veterinarian, Department of Human Nutrition and Food and Animal Sciences, College of Tropical Agriculture and Human Resources, University of Hawaii, Honolulu, Hawaii Patricia Ann Talcott, MS, PhD, DVM, DABVT, Associate Professor, Department of Veterinary and Comparative Anatomy, Pharmacology and Physiology, College of Veterinary Medicine, Washington State University, Pullman, Washington, Veterinary Diagnostic Toxicologist, Washington Animal Disease Diagnostic Laboratory, Pullman, Washington Ramiro E. Toribio, DVM, MS, PhD, DACVIM, Assistant Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio Bryan M. Waldridge, DVM, MS, DACVIM, DABVP, Department of Internal Medicine, Rood & Riddle Equine Hospital, Lexington, Kentucky David A. Wilkie, DVM, MS, DACVO, Professor, Department of Veterinary Clinical Sciences, The Ohio State University, Columbus, Ohio Pamela Anne Wilkins, DVM, PhD, DACVIM-LA, DACVECC, Professor and Section Head, Veterinary Teaching Hospital, College of Veterinary Medicine, University of Illinois, Urbana, Illinois W. David Wilson, BVMS, MS, Professor, Director of the William R. Pritchard Veterinary Medical Teaching Hospital, Associate Dean for Clinical Programs, School of Veterinary Medicine, University of California, Davis, California L. Nicki Wise, DVM, Resident, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University, Pullman, Washington Dana N. Zimmel, DVM, DACVIM, ABVP, Clinical Assistant Professor, Associate Chief, Department of Large Animal Clinical Sciences, College of Veterinary Medicine, University of Florida, Gainesville, Florida

Preface This is the third edition of Equine Internal Medicine. Like its two predecessors, it has been written and edited with the aim of promoting a clearer comprehension of the principles of medical disease and/or problem development by focusing on the basic pathophysiologic mechanisms that underlie the development of various equine diseases. As with previous editions, basic information is presented and then related to the clinical characteristics of each disease and its therapy and management. All the chapters that appeared in the rst two editions have been updated, and a number of them have been extensively revised or rewritten. Although the bulk of the chapters address speci c diseases along systems-based lines, we realize that the practitioner is initially confronted with a speci c problem that may have its origin in one or more of the body's systems. The rst section of the book is therefore devoted to an in-depth discussion of the basic mechanisms by which problems might develop and the principles underlying the treatment of many of them. The reader can build on this foundation by reading about speci c disorders in the second section of the book, which is divided into chapters dealing with problems of a particular body system or of a specific nature. Many true experts have contributed to this text. Their depth of knowledge about all aspects of equine internal medicine is encyclopedic and daunting. We are grateful for their e orts and diligence in helping us to produce what we hope will come to be regarded as the de nitive text on medical diseases of horses. We are indebted to them for their e orts. We trust that they derive a sense of pride from the part they have played in producing what we hope represents the gold standard in equine medical textbooks. In these days of progressive globalization of the world's societies and associated growth in the international movement of horses for breeding, recreational, and competitive purposes, there has also a worldwide increase in expectations relating to the standard of veterinary care and evaluation of sick horses. The sophistication of specialist training programs and the increased number of equine internists also taking advantage of postgraduate doctoral opportunities have resulted in a wealth of new information and the maturing of an increasingly complex and challenging discipline—equine internal medicine. The delivery of superior health care and increased client expectations that have been associated with the growth of this discipline have led to the appearance of extremely well-informed and astute equine general practitioners everywhere and specialist equine internists on most continents. More than ever before, equine internal medicine now stands as an autonomous specialty in the veterinary profession. We trust that the third edition of Equine Internal Medicine will prove to have as much universal appeal and application as those that preceded it. Finally, we would be remiss if we did not thank the many people at Elsevier for their persistence and e orts. Penny Rudolph and Lauren Harms in particular deserve our gratitude. They and many others have assisted in manuscript preparation, correspondence, and all the other tasks that must be accomplished to get a book like this into print. Without them and the generosity of our colleagues, this book would not have been published. We think that everyone's efforts have been worthwhile. Stephen M. Reed, DVM, Dipl ACVIM Warwick M. Bayly, BVSc, MS, PhD, Dipl ACVIM Debra C. Sellon, DVM, PhD, Dipl ACVIM

Table of Contents Front Matter Copyright Contributors Preface PART I: Mechanisms of Disease and Principles of Treatment Chapter 1: The Equine Immune System Chapter 2: Mechanisms of Infectious Disease Chapter 3: Clinical Approach to Commonly Encountered Problems Chapter 4: Pharmacologic Principles Chapter 5: Aspects of Clinical Nutrition Chapter 6: Recognizing and Treating Pain in Horses Chapter 7: Critical Care Chapter 8: Epidemiology PART II: Disorders of Specific Body Systems Chapter 9: Disorders of the Respiratory System Chapter 10: Cardiovascular Diseases Chapter 11: Disorders of the Musculoskeletal System Chapter 12: Disorders of the Neurologic System Chapter 13: Disorders of the Skin Chapter 14: Disorders of the Hematopoietic System Chapter 15: Disorders of the Gastrointestinal System Chapter 16: Disorders of the Liver Chapter 17: Equine Ophthalmology Chapter 18: Disorders of the Reproductive Tract Chapter 19: Disorders of the Urinary System Chapter 20: Disorders of the Endocrine System Chapter 21: Disorders of Foals Chapter 22: Toxicologic Problems Index

PART I Mechanisms of Disease and Principles of Treatment

CHAPTER 1 The Equine Immune System Paul Lunn, David Horohov

EQUINE IMMUNOLOGY Although much of modern immunology has focused on humans and murine models of human diseases, the horse has played a signi cant role in our understanding of immunologic processes. These contributions include the earliest work on serotherapy and passive transfer; immunoglobulin structure and function; immunity to infectious agents; immunode ciencies; and, more recently, reproductive immunology. Work in the horse continues in many of these areas to the benefit of equine medicine and comparative immunology. The overall organization and function of the equine immune system are similar to those of other mammalian species, although there are di erences. The reader is referred to any one of a number of texts1-3 for a more in-depth review of basic immunology. Here we shall focus on those aspects of the immune system that may be of most interest to equine researchers and clinicians. When possible, pertinent references to equine work will be provided.

INNATE IMMUNITY AND THE ACUTE INFLAMMATORY RESPONSE Immune defenses include both innate responses and adaptive responses, each of which is mediated by cellular and soluble components. Although we often regard the innate and adaptive responses as separate, they are in fact intimately related, sharing many of the same processes and components. The major di erence lies in the speci city and recall capability that characterizes the adaptive response. It is both the speci city of adaptive responses, mediated by antibodies or by e ector cells such as cytotoxic T-lymphocytes (CTLs), and the phenomena of immunological memory that are responsible for the capacity to completely protect an animal against a particular pathogen. Nevertheless, the role innate responses play in both prompting the adaptive response as well as providing valuable time for speci c adaptive responses to develop cannot be overstated. The horse, like every other species, is under constant assault from a variety of microbes that share its living space. Although most of these organisms are thought to be harmless, their disease-causing potential is evident when they cause opportunistic infections in individuals with compromised immune systems.4 Mammals, in general, have evolved a variety of defensive measures to prevent infections. The rst line of defense includes the physical barriers provided by the skin and the mucosal surfaces of the digestive, respiratory, and urogenital tracts. In addition to providing a barrier to penetration, the surface of the skin contains various enzymes, fatty acids, and oils that inhibit the growth of bacteria, fungi, and viruses. Mucous membranes and mucosal secretions contain bacteriolytic enzymes, bacteriocidal basic polypeptides, mucopolysaccharides, and antibodies that prevent colonization and penetration of these surfaces. Mucus also provides a physical barrier that entraps invading organisms and leads to their eventual disposal.5 Particles trapped in the mucous secretions of the respiratory tract, for example, are transported upwards through the action of ciliary cells to the trachea, where they are swallowed.6 Once they are swallowed, the acidic secretions and digestive enzymes of the stomach destroy most organisms. Normal epithelial and tissue architecture is essential for successful exclusion of bacteria, and the disruption of this mechanism makes the host susceptible to infection by bacteria that normally colonize the upper airway.7,8

Acute Phase Proteins, Pro-inflammatory Cytokines, and Complement Once breached, the host presents a variety of internal defenses to contain and eliminate the invaders. Invading organisms can initiate an in ammatory response either via the activation of plasma protease systems directly, such as by bacterial cell wall components, or by the secretion of toxins or other proteins that can directly activate the in ammatory response.9 The cell walls and membranes of bacteria contain various proteins and polysaccharides with characteristic, often repeating, molecular structures. These pathogen-associated molecular patterns (PAMPs) include such molecules as lipopolysaccharides (LPSs), peptidoglycans, lipoteichoic acid, and agellins.10 Other PAMPs include single- and double-stranded ribonucleic acid (RNA) found on viruses and unmethylated deoxyribonucleic acid (DNA) characteristic of bacteria. These PAMPs are recognized by a class of receptors known as toll-like receptors (TLRs), initially identi ed in Drosophila melanogaster (the fruit y).10 This ancient family of receptors recognizing di erent PAMPs is widely distributed on the various cells of the body. Not surprisingly, a number of different TLRs are found on the cells of the immune system, particularly those cells involved in the initial encounter with invading microbes. The binding of a PAMP to its speci c TLR leads to an intracellular signaling event, culminating in the expression of various accessory proteins that provide the co-stimulatory signals for the developing adaptive immune response. Injured cells also release products that initiate plasma protease cascades or produce pro-in ammatory cytokines that augment the in ammatory process.

Resident macrophages that encounter the invader add to the genesis of the in ammatory response through the production of proin ammatory cytokines such as interleukin (IL)-1, IL-6, and tumor necrosis factor alpha (TNF-α).9 Cytokines are hormonelike proteins that mediate a variety of cellular responses. A vast number of cytokines are involved in the regulation of innate and adaptive immune responses. IL-1, for example, is a pleiotropic mediator of the host response to infections and injurious insults (Box 1-1). Many of the e ects of IL-1 are mediated through its capacity to increase the production of other cytokines, such as granulocyte colony-stimulating factor (G-CSF), TNF-α, IL6, IL-8, platelet-derived growth factor (PDGF), and IL-11 (cytokines, chemokines, and interleukins are discussed later). IL-6 is responsible for the increased production of acute phase proteins (Table 1-1) by hepatocytes. Although the function of some of the acute phase proteins remains unclear, many of these proteins and the cytokines that elicited them are responsible for the characteristic physical signs of in ammation, including increased blood ow and vascular permeability, migration of leukocytes from the peripheral blood into the tissues, accumulation of leukocytes at the in ammatory focus, and activation of the leukocytes to destroy any invading organisms.11 The acute phase proteins include a number of complement proteins. The complement system is an interacting series of proteases and their substrates, resulting in the production of physiologically active intermediaries that can damage membranes, attract neutrophils and other cells, increase blood ow and vascular permeability, and opsonize bacteria and other particles for phagocytosis.12 The complement cascade can be activated in two ways (Figure 1-1). The classical pathway involves the recognition and binding of C1 to antigen-antibody complexes. Bound C1 is proteolytic and cleaves C4. This cleavage of C4 leads to the binding of C2 to C4b. C2 is in turned cleaved by C1 into C2a. The C4bC2a complex is referred to as the classical pathway C3 convertase because it is a protease capable of cleaving C3 into C3a and C3b. Another C3 convertase is generated via the alternate pathway. The activation of complement via the alternate pathway does not involve antibodies; instead, certain microbial products (zymosan and LPS) stimulate the association of Factor D, a proteolytic enzyme, with the complex of Factor B and C3b leading to the formation of the C3bBb complex, which is the alternative pathway C3 convertase. C3a, produced by the cleavage of C3 by the C3 convertases can bind to mast cells, causing them to degranulate, and is thus referred to as an anaphylatoxin, as is C4a. C3b serves as an opsonin for C3b receptor-bearing phagocytic cells. C3b is also required for the formation of the membrane attack complex by the terminal complement components, C5 through C9. In this process C5 is cleaved by either the C4b2a3b (Classic pathway C5 convertase) or C3b, Bb, and properidin (alternate pathway C5 convertase). C5 is cleaved into C5a and C5b. C5a is a chemoattractive factor for neutrophils and monocytes.13 C5b forms a complex with C6, C7, and C8 on cell surfaces. This leads to the insertion and polymerization of C9 that forms a pore in the membrane, leading to cell lysis. BOX 1-1 BIOLOGIC ACTIVITIES OF INTERLEUKIN 1 Activates T cells Activates B cells Enhances NK cell killing Fibroblast growth factor Stimulates PGE synthesis Stimulates bone resorption Chemotactic for neutrophils Activates osteoclasts

Induces fever Cytotoxic for some tumor cells Cytostatic for other tumor cells Stimulates collagen production Stimulates keratinocyte growth Stimulates mesangial cell growth Activates neutrophils Induces IL-6 production

NK, Natural killer; IL, interleukin. TABLE 1-1 Acute Phase Proteins Name C3, C4, and Factor B C-reactive protein Fibrinogen Kininogen Alpha1-acid glycoprotein Ceruloplasmin, Ferritin Haptoglobulin Hemopexin Serum amyloid A (SAA) Serum amyloid P (SAP) α1-antichymotrypsin

Function Opsonins Opsonin and complement activator Fibrin precursor, clotting factor Kinin precursor Function unknown, immunomodulatory Iron restriction Binds free hemoglobulin Binding free heme Lipid transporter, inhibitor of neutrophil function Lipid transporter, inhibitor of neutrophil function Protease inhibitor

α2 -macroglobulin Cysteine protease inhibitor

FIGURE 1-1

Protease inhibitor Protease inhibitor

Classical and alternate pathways of complement activation (see text for explanation). LPS, Lipopolysaccaride.

Lipid Mediators Prostanoids are lipid mediators that regulate the in ammatory response.14,15 The prostanoids group includes the prostaglandins (PGs), leukotrienes (LTs), and prostacyclin (PGI2), and they are the product of cyclooxygenase cleavage of arachidonic acid followed by endoperoxidation (Figure 1-2). The major sources of prostanoids in acute in ammation are the phagocytes, endothelial cells, and platelets. Prostanoids, in general, mediate the cardinal e ects of pain, fever, and edema characteristic of the acute in ammatory response, but their particular roles are somewhat confounding and can be either pro- or anti-in ammatory (Table 1-2).16 Prostanoid production depends on the activity of the two isoforms of the cyclooxygenase enzymes within cells: COX-1, which is present in most cells and its expression is generally constitutive, and COX-2, whose expression is low or undetectable in most cells, but its expression increases dramatically upon stimulation, particularly in cells of the immune system. Increased COX-2 expression by in ammatory stimuli likely accounts for the high levels of prostanoids found in chronic in ammatory lesions and is the basis for the development of COX-2–speci c inhibitors for treating chronic in ammatory diseases.17 However, studies using mice have indicated that the earliest prostanoid response to deleterious environmental stimuli depends on COX-1, and only as the in ammatory process progresses does COX-2 become the major source of prostanoids.18 Further, evidence for increased cardiovascular risk associated with COX-2 inhibitors has called into question the use of COX-2 inhibitors as treatments for inflammatory diseases.19,20

FIGURE 1-2

Lipid mediators of inflammation (see text for explanation). HPETE, Hydroperoxyeicosatetraenoic acid; HETE, hydroxyeicosatetraenoic acid; LTA, leukotriene A

(From Davies P, Bailey PJ, Golderberg MM, et al: The role of arachidonic acid oxygenation products in pain and inflammation, Annu Rev Immunol 21:337, 1984).

TABLE 1-2 Physiologic Effects of Lipid Mediators

Both COX isoforms produce PGH2, which is the common substrate for a series of speci c synthase enzymes that produce PGD2, PGE2, PGF2, PGI2, and TXA2 (see Figure 1-2). It is the di erential expression of these enzymes within cells present at sites of in ammation that will determine the pro le of prostanoid production. For example, mast cells predominantly generate PGD2, whereas resting macrophages produce TXA2 in excess of PGE2, although this ratio changes to favor PGE2 production after activation. Likewise, the biological e ect of a prostanoid depends on its binding to G protein–coupled cell surface receptors. The receptors for PGF2, PGI2, and TXA2 are called FP, IP, and TP, respectively. In contrast, PGD2 acts through two receptors, the DP receptor and the recently identi ed CRTh2 receptor, and there are four subtypes of receptors for PGE2, termed EP1–EP4. The prostanoid receptors themselves are coupled to various G protein–coupled intracellular signaling pathways. The DP, EP2, EP4, IP, and one isoform of the EP3 receptor can couple to Gs and thus increase intracellular cAMP concentration, which in T cells and other in ammatory cells is generally associated with inhibition of e ector cell functions. By contrast, the EP1, FP, IP, and TP receptors, as well as other EP3 isoforms, couple to Gq, and activation of these receptors leads to increased intracellular calcium levels and immune cell activation. Finally, TP, CRTh2, and yet another EP3 receptor isoform can each couple to Gi, causing cAMP levels to decline while also mobilizing intracellular calcium. Many cells of the immune system express multiple receptors that couple to these apparently opposing pathways. The impact of prostanoids present during an in ammatory response is thus determined by the array of receptors the cells express and the intracellular pathways to which they are coupled. Activation of these receptors, even when coupled to similar pathways, might evoke di erent responses because of di erences in the levels of expression (both constitutive and induced) or in the patterns of desensitization. The role of prostanoids in a given inflammatory response depends not only on the presence of the lipid mediators in the lesion but also on the receptor pro le on immune cells and the biochemical signaling pathways of these receptors.18 Thus PGE2 is considered proin ammatory because it promotes vasodilation by activating cAMP-coupled EP2 receptors on vascular smooth muscle and increases vascular permeability indirectly by enhancing the release of histamine and other mediators from tissue leukocytes such as mast cells. PGE2 is also the prostanoid responsible for fever production. However, as in ammation progresses, PGE2 synthesis by macrophages is enhanced as a result of increased expression of COX-2 and PGE-synthase, and the resulting increased levels of PGE2 inhibit leukocyte activation, mast cell degranulation, and relax smooth muscle contractions. In the lung PGE2 promotes bronchodilation through activation of Gs-coupled EP2 and EP4 receptors. Thus in these situations PGE2 may be considered anti-inflammatory.

Chemotaxis and Leukocyte Trafficking One of the initial and most crucial aspects of the acute in ammatory response is the recruitment of leukocytes (primarily neutrophils) to the site of injury. Neutrophils constitute the rst line of the cellular defense and are the initial cells involved in an in ammatory response. These phagocytic cells are derived from multipotent stem cells located chie y in the bone marrow. Under the in uence of a variety of signals provided from both within and outside of the bone marrow, these stem cells become committed to developing into cells of the granulocyte lineage. The critical signal is provided by a family of growth factors known as colony-stimulating factors (CSFs) that provide both proliferative and di erentiative signals leading to the development of granulocytes and other leukocytes. Once released into the circulation, these cells must nd their way to the site of the in ammatory response. The production of various chemotactic factors by host cells, bacteria,

and other invaders causes various leukocytes to enter the circulation and be carried to the site of the injury.21 Chemokines are soluble proteins produced by host cells that induce the directional migration and activation of leukocytes, as well as other somatic cell types, and thus play a major role in the in ammatory response.22 Interleukin 8 (IL-8) plays a central role in this process. Other chemokines promote humoral and cell-mediated immune reactions; regulate cell adhesion, angiogenesis, leukocyte tra cking, and homing; and contribute to lymphopoiesis and hematopoiesis.23 The speci c tra cking of leukocytes from the blood to in ammatory sites depends on both the production of chemotactic factors and the interaction of speci c receptors on the leukocytes with corresponding adhesion molecules on the endothelial surface of the blood vessels. Neutrophil adherence is a two-step process rst involving endothelial cell surface molecules known as selectins.9 Small venular endothelium overlying a site of in ammation and exposed to thrombin, platelet activating factor (PAF), IL1, histamine, or other mediators released by clotting, platelet activation, or mast-cell activation express P-selectin.24 P-selectin mediates the process in which neutrophils initially interact with the endothelial surface in a process known as “rolling,” in which the circulating neutrophil interacts with the endothelial cell before the actual adherence.25 Selectins function by binding to carbohydrate ligands present on the cell surface. In the case of neutrophils, the ligand is sialylated Lewis-X antigen for the endothelial E-selectin. The second part of the adherence process is the tight binding of integrins on the neutrophil surface with intracellular adhesion molecules (ICAMs) on the endothelial cell surface. Leukocyte integrins are heterodimeric proteins with distinct α and shared β polypeptide chains. The α and β chains can combine in di erent heterodimers to form multiple shared and unique speci cities. Neutrophil expression of αMβ2 and αXβ2 is activation dependent. Neutrophils can be activated by a number of soluble proteins, including formylmethionylleucylphenylalanine (fMLP), N-formulated peptides present in bacterial but not eukaryotic proteins. Host factors present at the site of in ammation can also activate neutrophils, notably the complement proteins (C5a, C3a) and cytokines such as IL-8 and tumor necrosis factor (TNF), and immune complexes.26 Expression of integrins by activated neutrophils allows them to become tethered to the endothelial surface. The migration of neutrophils through the vascular wall is less well understood than these initial events leading to rm adhesion. The β2 integrins, as well as αvβ3, PECAM-1 and integrin-associated protein (IAP), appear to play a role in this process. Endothelial cell–produced IL-8 also is believed to have a critical role in this process. Once through the endothelium, the phagocytes will follow chemotactic signals and migrate toward the point of injury. They may adhere to other cells during migration to the site of in ammation, and these interactions also depend on αMβ2 and αXβ2 integrins. Migration through the extracellular matrix is mediated by β1, β3, and β5 integrins recognizing specific protein ligands. Neutrophils recruited and activated in this manner will actively phagocytose microscopic invaders and attempt to destroy them using reactive oxygen products generated via an NADPH-oxidase–dependent “respiratory burst.”25,27 In the process the neutrophils release additional pro-in ammatory mediators, thus amplifying this response. Among those cells attracted to the area are natural killer (NK) cells capable of lysing virus-infected and other abnormal cells. The production of interferon-α/β by macrophages and other cells enhances the cytolytic activity of the NK cells. The NK cells themselves can be the source of interferon-γ, another pro-inflammatory cytokine. Depending on the magnitude of the initial insult and the susceptibility of the invader to neutophil-mediated destruction, the in ammatory response may be either acute or chronic. Acute in ammation is thus a rapid response to an injury that is characterized by accumulations of uid, plasma proteins, and neutrophils that rapidly resolves once the initial in ammatory stimulus is removed. Deactivation signals include PGE2, cortisol, IL-10, and transforming growth factor-β (TGF-β). Some of those chemotactic agents responsible for initiating the response (IL-8, FMLP, C5a, LTB4, and PAF) also serve to downregulate its intensity by inducing the shedding of IL-1 receptors from neutrophils.28 The shedding of this decoy receptor may have anti-in ammatory e ects as it e ectively binds and neutralizes this cytokine. Likewise, many of the acute phase proteins are thought to have immunomodulatory activity downregulating neutrophil function.29 Acute in ammatory responses may often be subclinical and resolve without complications. However, if the invader is resistant to neutrophil-mediated destruction or the degree of injury is large, the response may become more chronic with the added recruitment of macrophages and lymphocytes, and fibroblast growth. The essential characteristic of the innate immune response is that it does not exhibit speci city for the invading organism. Thus the induction of an innate immune response does not require prior exposure to the invading organism nor is it augmented by repeated exposure to the same organism. Whereas resistance may be genetically controlled, the genes encoding resistance are not found within the gene complex that controls adaptive immune responses. In most instances these mechanisms are adequate for eliminating casual invaders. However, pathogenic organisms have evolved various methods for avoiding elimination. In response to these organisms, the specialized cells and products of the adaptive immune response are mobilized.

ADAPTIVE IMMUNITY The adaptive immune response is initiated in response to an encounter with a foreign agent and depends on antigen-speci c immune responses mediated by di erent divisions of the lymphocyte family (Figure 1-3). In contrast to the nonspeci c nature of the innate immune response, an important characteristic of the adaptive immune response is the speci city of this interaction. Thus exposure of the host to a particular microbe or parasite results in the induction of immune responses that are directed against speci c components of the invading organism that do not a ect unrelated organisms. The speci city of the adaptive immune response is the result of the interaction of speci c

molecular structures or antigens of the invader with antigen-speci c receptors on lymphocytes. All types of chemical structures can serve as antigens, but not all antigens can induce an immune response. Immunogens, those antigens that can stimulate an immune response, are usually high molecular–weight, chemically complex molecules. Proteins, nucleic acids, lipids, and polysaccharides can all serve as immunogens. Large immunogens, such as proteins, contain multiple antigenic determinants or epitopes, which interact with lymphocytes via their antigen-speci c receptors. Haptens consist of single antigenic determinants and can e ectively combine with the binding site of antibody molecules. However, because they consist only of a single antigenic determinant, they cannot cross link B-cell receptors (antibody molecules), and they are also unable to stimulate T cell responses. Haptens therefore cannot stimulate an immune response unless multiple haptens are physically attached to a larger molecule, known as a carrier. Although these distinctions between antigens, haptens, and immunogens appear minor, they provide the underlying basis for our understanding of many allergic and autoimmune responses.

FIGURE 1-3

Major divisions of the lymphocyte family. To the left of the diagram, di erent populations of lymphocytes are distinguished by expression of di erent cell surface molecules. To the right of the diagram, the distinctions are functional.

Like the innate response, the adaptive immune response to a speci c antigen consists of both humoral and cellular e ector mechanisms. The humoral component is mediated by immunoglobulins or antibodies found in plasma and tissue uids. Antibodies are produced by B lymphocytes, small lymphoid cells characterized by the cell surface expression of immunoglobulin molecules. B cells represent fewer than 15% of the circulating peripheral blood mononuclear cells but are present in higher proportions in lymph nodes and the spleen. B cells are derived from the fetal liver and bone marrow of mammals and the bursa of Fabricius of birds. In the bone marrow B cells are the products of a putative lymphoid stem cell derived from the pluripotent stem cell. Under the in uence of various cytokines produced by bone marrow stromal cells, the B cell precursor undergoes its 3-day development into a mature B cell. Upon stimulation with speci c antigen, B cells di erentiate into plasma cells that produce enormous quantities of speci c antibody. The activation, proliferation, and di erentiation of B lymphocytes into plasma cells depends on other cells, including T lymphocytes, which represent the cellular component of the adaptive immune response. The T lymphocyte is also derived from the multipotent stem cell and lymphoid precursor in the bone marrow, although its subsequent development into the mature T cell occurs in the thymus. Within the thymic environment the prothymocyte undergoes a developmental and selective process while emigrating through the cortex into the medullary region of the thymus. Fewer than 3% of all the immature thymocytes found in the cortex survive to become peripheral T cells. Although the induction of an antibody response requires the interaction of B and T lymphocytes, these cells recognize di erent epitopes on the same antigen. Indeed, antigen recognition by B cells and T cells is fundamentally quite di erent. B cells, and antibodies, recognize antigens in solution or on cell surfaces in their native conformation, whereas T cells recognize antigen only in association with self molecules known as major histocompatability complex antigens found on most cells’ surfaces. The adaptive immune response thus di ers from innate immunity in that it is antigen driven and those cells that mediate the adaptive immune responses, T and B lymphocytes, express speci c receptors for the antigen. Because the immune system will respond to the antigens of both live and killed pathogens, it is possible to stimulate immunity without causing infection, which is the basis of vaccination. Although this principle appears to be straightforward, vaccination does not always yield the expected result. Why some vaccines work and others fail is a complex issue, a major component of which is the nature of the antigen-specific receptors of lymphocytes.

Immunoglobulin: Antigen-specific Receptor of B Lymphocytes The antigen-speci c receptor of the B cell is cell surface–bound antibody. An antibody molecule is composed of two identical light chains and two identical heavy chains that form a disulfide-linked Y-shaped molecule (Figure 1-4). The light chain can be divided into two domains,

a conserved carboxy-terminal domain and a highly variable amino-terminal domain. Analysis of heavy chains reveals a similar domain structure, with the amino-terminal domain being highly variable and the presence of three constant domains. The antigen-binding region of an antibody molecule is formed by the association of the amino ends of a light and a heavy chain, whereas the carboxyl end of the heavy chain determines the isotype of the molecule. Five di erent isotypes, or classes, of antibody molecules have been identi ed in most species, including the horse: IgD, IgM, IgG, IgA, and IgE (Table 1-3).30,31 Additionally, the IgG isotype can be subdivided into subclasses based on physico-chemical properties. Analysis of equine genomic DNA has indicated the existence of one IgM, one IgD, one IgE, one IgA, and seven IgG genes.31 Before the availability of genetic characterization of the equine immunoglobulin heavy chain gene loci, at least four IgG subclasses were identi ed by physico-chemical means and de ned serologically by monoclonal antibodies as IgGa, IgGb, IgGc, and IgG(T).32 Each of these “classical” (older) IgG subclasses has been identi ed as a gene product of one or more of the equine heavy chain gene loci.30,31 It appears that both IgGb and IgG(T) are encoded by two loci each, raising the possibility that these classical subclasses may each comprise two distinct subclasses. At the time of writing, this remains uncertain, although ongoing studies are starting to resolve the issue.30 The gene product of at least one IgG heavy chain locus is not de ned by the classical IgG subclasses, and it remains to be determined whether it is expressed as a protein and what its role may be.31 Given the current remaining uncertainties as to the role of equine IgG subclasses as defined by nomenclature based on heavy chain gene locus, this edition of this text will continue to use the classical nomenclature.

FIGURE 1-4

Molecular structure of secretory immunoglobulin (Ig) A. This schematic illustrates the major features of immunoglobulin molecules. The illustrated IgA molecule is dimeric, with the two immunoglobulin units joined by a “J-chain” and a series of disulphide bonds, and IgG molecules are monomeric. Each immunoglobulin unit consists of two heavy chains and two light chains. The heavy chains have four subunits and the light chains two. One end of the immunoglobulin unit has a highly variable protein structure and is involved in antigen recognition, whereas the remainder of the immunoglobulin unit has a constant structure in each immunoglobulin class and subclass and determines the functional characteristics of the molecule, such as binding complement, or recognition by macrophages or neutrophil Fc receptors. This specialized dimeric IgA molecule also has a secretory piece that increases its stability in the harsh mucosal environment. TABLE 1-3 Immunoglobulin Isotypes Isotype Immunologic Function IgD Antigen receptor of naïve B lymphocytes. An IgD heavy chain gene has been identified in the horse, and it is probably expressed. IgM Surface IgM is found on naïve, activated, and memory B cells. Secreted IgM is a pentamer and represents the major antibody produced during a primary response. IgM efficiently mediates agglutination, neutralization, opsonization, and complement activation. IgG The principle immunoglobulin found in plasma representing up to 80% of the total immunoglobulin concentration. Various subclasses of IgG have been identified (see text). The classical view is that there are four IgG subclasses in the horse (IgGa, IgGb, IgGc, and IgG(T)) defined by physicochemical properties and monoclonal antibodies, although seven IgG heavy chain genes have been identified and there is evidence that all are expressed. The major functions of IgG include opsonization and neutralization reactions. IgGa and IgGb are effective in fixing complement and participates in antibody-dependent cellular cytotoxicity (ADCC) while IgGc and IgG(T) are not, although they appear to play an important role in exotoxin neutralization and immunity to parasites. As our understanding of IgG heavy chain gene expression and function increases, a new nomenclature (i.e., IgG1–7) will replace the classical nomenclature (i.e., IgGa). IgA IgA, the most abundant antibody in secretions (e.g., tears, mucus, saliva, colostrum) is a dimer composed of two IgA molecules joined by a J chain. IgA in the plasma is predominantly monomeric. IgA antibodies can be neutralizing but only activate complement via the alternative pathway. IgE Most IgE is found associated with the surface of mast cells and basophils and only very small amounts are present in the plasma. The cross-linking of two IgE moleulces with specific antigen results in the degranulation of the mast cells and basophils. Thus IgE is the primary antibody responsible for Type I hypersensitivity reactions and appears to play a central role in immunity to parasites. Ig, immunoglobulin. Data from Wagner B. Immunoglobulins and immunoglobulin genes of the horse. Dev Comp Immunol 2006;30:155-164; Lewis MJ, Wagner B, Woof JM. The di erent e ector function capabilities of the seven equine IgG subclasses have implications for vaccine strategies. Mol Immunol 2008;45:818-827; and Wagner B, Miller DC, Lear TL, Antczak DF. The complete map of the Ig heavy chain constant gene region reveals evidence for seven IgG isotypes and for IgD in the horse. Journal of Immunology 2004;173:3230-3242.

Membrane-bound IgM and IgD serve as the antigen-speci c receptors for B lymphocytes. Each contains a membrane-spanning region near its carboxy end that is inserted into the mRNA during di erential splicing of the heavy chain exons. Although rarely detectable in the circulation, IgD is present in large quantities on the surface of naïve B lymphocytes. Following activation, the surface expression of IgD is lost, although the cell may continue to express the membrane form of IgM. Early in an immune response, the B cell secretes large amounts of the pentameric form of IgM. As the immune response proceeds, the B cell will switch the isotype of its heavy chain. Isotype switching involves the substitution of one heavy chain–constant region in place of another. The genes encoding the ve di erent constant regions of the heavy chain are sequentially arranged on the chromosome (Cδ , C , Cγ, C , and Cα). Initially, the rst two constant region genes encoding the δ and constant regions are used to form the heavy chain. The 5’ region of each constant region gene segment contains repetitive regions of DNA known as switch sequences.33 The switch sequences appear to play a role in this rearrangement and may serve as the target for speci c recombinases. When switching occurs, a new constant region segment is selected and the intervening genes are removed either by splicing or looping out. Isotype switching a ects only the heavy chain–constant domains and has no e ect on the antigen speci city of the immunoglobulin molecule. The signal for B cells to undergo isotype switching is provided by T lymphocytes in the form of various cytokines.34 For example, IL-4 induces isotype switching to the IgE isotype, whereas interferon-γ blocks this induction and augments IgG production.35,36 IgA is produced in response to the combination of the cytokines IL-4, IL-5, and transforming growth factor-ß (TGF-β).37 The antigen speci city of a particular antibody molecule (and the B cell that produces it) is determined by the combination of the variable domains of the light and heavy chains. The association of these two domains results in the formation of an antigen-binding groove or pocket that contains regions of hypervariability that de ne the speci city of a particular antibody molecule. It has been estimated that more than 108 di erent antibody speci cities are possible. The generation of this tremendous amount of diversity in antibody speci city occurs during B cell ontogeny in the bone marrow.38 Within a given B cell, the genes encoding the heavy and light chains of an antibody molecule are organized into speci c gene segments. Thus the light chain is formed from variable (Vl), joining (Jl), and constant (Cl) gene segments that together form the variable and constant domains of the light chain. In the germ line of an undi erentiated cell, several hundred di erent Vl and several dozen Jl gene segments can be found. Likewise, the heavy chain of a B lymphocyte is composed of VH, diversity (D), and JH segments that form the variable domain, and these join to the constant region genes to form the complete heavy chain molecule. Similarly, in the germ line a large number of VH gene segments and a smaller number of D and JH segments are found. During the di erentiation of a B cell (Figure 1-5), there is the sequential selection and rearrangement of a VL segment with a JL segment and the accompanying deletion of intervening VL and JL segments (Figure 1-6). The rearranged VJC sequence is then transcribed into mRNA and translated into the light chain. A somewhat similar sequence follows for heavy chains except that two rearrangements are necessary, a D to JH rearrangement followed by a VH to DJH rearrangement. Once completed, the VDJ segment is brought into the proximity of the appropriate CH segment and transcribed. Not all of the gene segment rearrangements produce functional genes. Because a B cell has two sets of heavy chain genes, one on each chromosome, and most species, including the horse, have two di erent sets of light chain genes,39,40 there are several chances to form appropriate heavy and light chains. Once the heavy and light chain gene segments are successfully recombined, the genes on the sister chromosome neither recombine nor are they expressed. This process of allelic exclusion ensures that the B cell produces antibodies of a single speci city. Although this random assortment of gene segments accounts for much of the diversity in antibody speci city, additional mechanisms are also involved, including junctional diversity, which results from the imprecise joining of gene segments and somatic mutations. Somatic mutations are point mutations in the hypervariable region of either the heavy or light chain that occur during the proliferation of antigen-activated B lymphocytes. Such mutations appear to play a role in increasing antibody a nity for its antigen. Thus fewer than 1000 genes can give rise to more than 108 molecules of the various speci cities needed to recognize the vast number of antigens the host may encounter.

FIGURE 1-5

B cell di erentiation. Di erent stages of B lymphocyte development can be recognized by expression of immunoglobulin molecules. This maturation requires a series of gene rearrangements in order to select the genes that will encode the antigen binding part of the immunoglobulin molecule (variable region) and subsequently to select the genes that determine the class or subclass of the antibody molecule. Initially immature B cells express IgM (the majority of peripheral blood B cells), but after antigen exposure the B cell becomes activated and may express any of the immunoglobulin classes or subclasses. This decision depends in large part on cytokine signals from T-helper cells. Finally activated B cells either mature into short-lived antibody secreting plasma cells or become long-lived memory B cells.

FIGURE 1-6

Immunoglobulin gene rearrangement–somatic recombination process for production of an immunoglobulin heavy chain. The gure shows a hypothetical series of V, D, and J variable heavy chain genes, positioned 5 to the known equine heavy chain constant region gene loci. In the rst step in somatic recombination a D and a J gene segment are joined, and in the second step a V gene segment is joined to complete the VDJ recombination and form a gene capable of encoding the variable region. Subsequently one of the seven equine γ heavy chain constant regions, labeled with their corresponding IgG subclass when known, was selected to complete the gene rearrangement. Because the Cγ4 heavy chain constant region gene was selected, this leads to production of an IgGb heavy chain.

TcR and CD3 Complex: Antigen-specific Receptor of T Cells T lymphocytes can be di erentiated from B lymphocytes in that they do not express surface immunoglobulins but instead express the T cell receptor (TcR). T cells also express another antigen called CD3. (The designation CD stands for cluster designate and is the result of an international workshop to standardize the terminology used to describe leukocyte surface antigens recognized by monoclonal antibodies.) The TcR and CD3 form a multimeric complex on the T cell surface, and this complex is involved in antigen-speci c recognition.41 The TcR structure was rst identi ed using antibodies that recognized a surface antigen expressed on a cloned T lymphoma cell line. This antibody recognized a disul de-linked heterodimer composed of an acidic (α) and a basic (β) protein of 40 to 45,000 molecular weight. Similar heterodimers were found on a variety of antigen-speci c T cell lines but not on B cells. Peptide mapping studies of the α and ß chains from many di erent T cell lines demonstrated that they contained variable and constant domains reminiscent of immunoglobulin structure. Further analysis indicated that, like immunoglobulin genes, the TcR genes underwent gene rearrangements during T cell development. Subsequently, two additional TcR genes were identi ed, the γ chain and δ genes corresponding to a second heterodimer. Thus two TcR exist, an α/β heterodimer that constitutes the TcR on almost 90% of all T cells and a γ/δ heterodimer present on approximately 10% of the peripheral T cells. The signi cance of these two di erent TcR heterodimers has not yet been determined. It should be noted that γ/δ T cells have not yet been identified in the horse. It is clear that γ/δ T cells represent a functionally distinct population of T cells typically associated with mucosal surfaces.42 As such they are thought to play an important role in immunologic surveillance. Analysis of the predicted amino acid sequences for the TcR proteins con rmed a structural similarity with antibody molecules. One peculiarity in the structure of the TcR was observed from the amino acid sequence analysis. Whereas both the α and β chains of the TcR contained a transmembrane region, both proteins had very short cytoplasmic tails. It therefore seemed unlikely that TcR itself could transmit any cytoplasmic signal in response to antigen binding. This led to the search for other proteins associated with the TcR. Solubilization of the T cell membranes revealed that ve other proteins could be immunoprecipitated with the TcR. Similar results were obtained when anti-CD3 antibodies were used. Thus the TcR heterodimer is noncovalently associated with the CD3 complex of proteins. The ve proteins of the CD3 43 Unlike the TcR α and β proteins, the CD3 complex (γ, δ, , , and ) are involved in signal transduction following TcR binding to antigen. proteins have large intracellular domains, some of which are phosphorylated in response to stimulation of the TcR. In addition to providing a signaling mechanism for the TcR, the CD3 complex is also required for the expression of the TcR heterodimer on the cell surface.41 The generation of diversity in the TcR during T cell ontogeny employs a mechanism quite similar to that used to generate immunoglobulin diversity. The TcR α and γ chains resemble immunoglobulin light chains in that they are composed of V, J, and C gene segments. The particular V, J, and C segments used are selected from a germ line con guration containing a few (C region) to several hundred (V region) gene segments. The selection and rearrangement of the gene segments are similar to those employed by the immunoglobulin light chain and

appear to involve the same recombinase. Likewise, the β and δ chains resemble heavy chains, each being composed of V, D, J, and C gene segments, and their selection and rearrangement from germ line genes also parallel immunoglobulin heavy chain rearrangement. Thus the generation of diversity is the result of the combination of multiple gene segments and junctional diversity. However, unlike immunoglobulins, the TcR genes do not undergo somatic mutations.

T Lymphocyte Subsets Mature thymocytes and T lymphocytes can be further divided into two distinct populations on the basis of their expression of either the CD4 or CD8 antigen.44 The expression of these antigens is directly correlated with the speci city of the T cell. The expression of either CD4 or CD8 also correlates to some extent with the T cell’s function. Thus those cells that express the CD8 antigen are typically CTLs, whereas those that express the CD4 antigen are typically helper cells that produce those cytokines that enhance antibody and cell-mediated immune responses. Whereas the T lymphocytes in the periphery express either CD4 or CD8 antigens, cortical thymocytes express both antigens. During the process of thymic selection, these cells convert to either CD4+ or CD8+ cells or they are eliminated (Figure 1-7). It is at this stage of their development that T cells are said to “learn” to recognize antigen. It is also at this stage that autoreactive T cells are eliminated. Although experimental studies have shown that both positive and negative selection of the T cells is occurring, the exact mechanism of these selective processes remains unknown. Interestingly, although T cells expressing the α/β heterodimer of the TcR can be either CD4+ or CD8+, γ/δ cells are either CD8+ or CD4- CD8. These results suggest that the γ/δ cells undergo a di erent developmental process than do the α/β cells. Like the CD3 complex, both the CD4 and the CD8 antigen are involved in the intracellular signaling event following TcR engagement with its speci c antigen. Unlike B cells, and antibodies, that recognize antigens in solution or on cell surfaces in their native conformation, T cells recognize only processed antigen in association with self molecules known as major histocompatability complex (MHC) antigens.

FIGURE 1-7

Thymic development.

Major Histocompatibility Antigens and Antigen Presentation The MHC was originally de ned in terms of its role in allograft rejection. Following the rejection of a primary allograft, antibodies that reacted with the allograft could be found in the recipient’s sera. These antibodies could be used to identify or type tissues to determine the suitability of a donor for transplantation. It was also determined that multiparous females had similar antibodies in their sera as a result of the exposure to paternal MHC antigens on the fetus.45 Through the use of these sera, it was possible to identify a large number of serologically de ned transplantation antigens. Subsequent genetic analysis of the MHC region demonstrated that there were a number of closely linked genes encoding several di erent, though related, antigens that were involved in allograft rejection. These closely related genes are collectively referred to as MHC I genes and their products as MHC I antigens. In addition to the serologically de ned MHC I antigens, another group of antigens was identi ed within the MHC; these antigens were involved in the stimulation of mixed lymphocyte responses and the control of immune responsiveness. These MHC II antigens are structurally and functionally distinct from the MHC I antigens, except that both are involved in T cell recognition of antigen. MHC I antigens are cell surface glycoproteins consisting of two noncovalently associated proteins, an MHC-encoded transmembrane protein of approximately 44 kd (α chain) and ß2-microglobulin, a 12 kd protein encoded outside of the MHC.3 MHC I antigens are expressed on the surface of most nucleated cells. The highest level of expression is on lymphoid cells with lower expression on broblasts, muscle cells, and neural cells. MHC I antigens are not detectable on early embryonal cells, placental cells, and some carcinomas. The level of expression of MHC I antigen can be modi ed by treatment with cytokines or infection with viruses. Interferons and TNF-α augment MHC I antigen expression. This augmented expression is the result of increased production of MHC I mRNA, and the regulatory region of the MHC I antigen genes has been shown to contain interferon and TNF-α responsive elements that control the transcriptional activity of these genes. The MHC I region of most animal species, including the horse, contains a number of MHC I α chain genes, some of which are pseudogenes

and are not expressed.46 The known total of di erent equine MHC class I genes (loci) expressed as mRNA is seven.47 In the horse these genes are located on chromosome 20, and those genes that are expressed exhibit a great deal of polymorphism.48,49 Much of this polymorphism is localized in the α1 and α2 domains, the α3 domain being more conserved. The polymorphism of these two domains is related to their role in presenting antigen to T cells. The physiologic role of MHC I antigens was de ned when it was discovered that cytotoxic T cell (CTL) lysis of virus-infected cells was restricted to target cells expressing the same MHC I antigen as the CTL.50 This observation led to the realization that T cells recognized the combination of self-MHC and foreign antigen. Furthermore, those T cells that recognized MHC I antigens invariably expressed the CD8 co-receptor. The nature of the association between MHC I and the foreign antigen remained unclear until X-ray crystallographic studies of human MHC I antigen were performed.51,52 In addition to revealing the structural organization of the domains of the MHC I antigen, the image also revealed a cleft that lay between the α1 and α2 domains. It was proposed that this cleft binds the processed peptide epitopes for presentation to the T cell receptor. Indeed, the cleft of the cyrstalized protein used for the X-ray di raction studies was found to contain a contaminating peptide.51 Other experiments showed that the incubation of cells with puri ed viral peptides resulted in the lysis of the cells by virus-speci c, MHC I–restricted CTL.52 Together these results support the notion that the endogenous processing of viral antigens leads to the association of viral peptides with MHC I antigens on the surface of the infected cell, and this is recognized by the TcR-CD3 complex in association with CD8.52 How these viral antigens get to the cell surface is the result of a peptide transport system whose function is to transport processed peptides from the cytosol to the ER.53 Once in this compartment, peptides are handed o to newly formed MHC class I molecules and stabilize a trimolecular complex with β 2 microglobulin. This complex is then transported to the cell surface, where antigen presentation occurs. Because this is a normal cellular process for eliminating degraded proteins from the cell, it is not surprising that MHC I antigens are normally loaded with these self-peptides. Indeed, it is this encounter with MHC I loaded with self-peptides in the thymus that is responsible for the deletion of autoreactive clones during T cell ontogeny. This unique peptide-binding characteristic of MHC-I molecules has led to their use as immunologic reagents (tetramers) for the identi cation and enumeration of antigen-speci c CD8+ T cells.54 In the horse tetramers based on the equine MHC class I molecule 7-6, associated with the ELA-A1 haplotype, have been used to identify and enumerate equine infectious anemia virus (EIAV)–speci c cytotoxic T cells.55 In the future similar approaches may be used to analyze CTL responses to a variety of viruses and provide important information on the role these cells play in protection from these infections. MHC II antigens are heterodimeric, transmembrane glycoproteins composed of an acidic α chain (25 to 35 kd) and a basic β chain (25 to 30 kd).56 A third chain, the invariant chain, is associated with the MHC II antigen during assembly in the endoplasmic reticulum but is not expressed on the cell surface. Both the α and β polypeptides are encoded within the MHC region. Both polypeptides possess two extracellular domains. The α chain has a single disulfide bond located in its membrane proximal (α2) domain, and the β chain has a disulfide bond in both of its extracellular domains. Structurally, the MHC II antigens resemble MHC I antigens and are also members of the immunoglobulin superfamily, a group of proteins that have structural similarities to immunoglobulin molecules (Figure 1-8).57

FIGURE 1-8

The immunoglobulin superfamily. Immunoglobulin serves as the prototype model for the superfamily. Both the heavy and light chains of an immunoglobulin molecule can be divided into variable (VH and VL) and constant domains (C H and C L). Analogous regions have been identi ed on a variety of other molecules involved in immune recognition including class I and class II antigens, the T cell antigen receptor (TcR), and the CD4 and CD8 antigens found on T cells (see text). Disul de bonds forming the domains are not shown.

The MHC II genes are functionally and structurally distinct from the MHC I genes. Unlike MHC I antigens, the MHC II antigens are

restricted in their expression to certain cells of the immune system: B lymphocytes, dendritic cells, macrophages, and activated T lymphocytes of some species. Other cells may also express MHC II antigens after treatment with various cytokines.58-60 Interferon-γ, TNF-α, 1,25dihydroxyvitamin-D3, and granulocyte-macrophage colony stimulating factor can induce MHC II antigen expression on monocytes and macrophages and other cells. IL-4 enhances MHC II antigen on B cells. A number of agents downregulate MHC II antigen expression, including glucocorticoids, prostaglandins, and α-fetoprotein. Although MHC II antigen expression is also regulated at the transcriptional level, no interferon or TNF-α response elements have been identi ed in the regulatory regions of MHC II genes. In fact, the regulatory region of MHC I and MHC II genes are quite di erent, and this fact is probably responsible for the di erences in tissue distribution for these two MHC antigens.46 Like the MHC I genes, the MHC II region contains genes for multiple MHC II antigens, some of which appear to be pseudogenes and are not expressed. Those α and β chains that are expressed exhibit a high degree of allelic variability, although typically the β chain exhibits the most polymorphism. Unlike the MHC I genes, the variability in the MHC II genes is the result of point mutations. There is also correspondingly less polymorphism in the MHC II genes when compared with the MHC I genes. The MHC II region of other species, including the horse, have been studied using human DNA probes, and extensive polymorphism involving several genes has been identified. Whereas antigens processed via the endogenous pathway are associated with MHC I antigens, antigen processed via the exogenous pathway is associated with MHC II antigens (Figure 1-9).52 Here endocytosed antigen, such as that phagocytosed by a macrophage, is partially degraded in a prelysosomal compartment of low pH and limited proteolytic activity. The processed protein associates with a peptide binding site at the junction of the α1 and β 1 domains of the MHC II molecule. This association of the epitope with the MHC II molecule protects it from further degradation. The MHC II molecule is then re-expressed on the cell surface for subsequent presentation to the T cell. The immune system contains a distinct group of antigen-presenting cells called dendritic cells (DCs) that are specialized to capture antigens and initiate T cell immunity and move freely from epithelial surfaces to adjoining lymph nodes.3 DCs can be found in a variety of locations in the body and are often named according to their microscopic appearance. Hence interdigitating cells found in lymph nodes, veiled cells in lymphatics, and Langerhan’s cells in skin are all DCs. Immature DCs can take up antigens by micropinocytosis using their extensive cellular processes or receptor mediated phagocytosis. This results in activation and migration to a regional lymph node where antigen presentation to T lymphocytes occurs. Mature DCs have high levels of MHC II expression on their surfaces and are no longer phagocytic but are extremely efficient stimulators of both MHC I– and MHC II–restricted T cell responses in the draining lymph node (Figure 1-10).

FIGURE 1-9

Antigen processing pathways. This gure depicts major histocompatibility complex (MHC) I antigen presentation to the left of the diagram and MHC II antigen presentation to the right. In MHC I antigen presentation (a) peptides generated by degradation of proteins in the cytoplasm are transported into the endoplasmic reticulum (b). In this location MHC I molecules bound by a membrane protein calnexin bind the peptides, which allows release of the MHC I molecules by the calnexin and transport through the Golgi complex to the cell surface (c). In MHC II antigen presentation antigen is taken up by phagocytosis (1) into the endosome compartment and routed to lysosomes for degradation. Vesicles containing MHC II molecules produced in the endoplasmic reticulum fuse with the endosomes (2) and the MHC II molecules bind with the degraded peptides for transport back to the cell surface (3). The MHC II molecules are prevented from binding the endogenous peptides in the endoplasmic reticulum by the presence of invariant chain, which is only lost in the acidic endosomal environment.

FIGURE 1-10

The role of professional antigen-presenting cells (APCs). In this figure pathogen invasion is followed by antigen uptake by a dendritic cell, the most potent of the APC family. The dendritic cells become activated and migrate to a local lymph node, where they are extremely e ective at stimluating naïve T cells including both T helper cells and CTLs.

In a complex immunogen certain antigenic determinants are particularly e ective at stimulating an antibody response. These immunodominant epitopes are often located at exposed areas of the antigen such as in polypeptide loops. These types of structures are often quite mobile and may allow for easier access to the antibody binding site. T cell epitopes possess a particular structural characteristic of amphipathic helices. However, structure alone does not determine the immunogenicity of a particular antigen, and T cell recognition of foreign antigen requires more than just the expression of the processed antigen on the surface of the antigen presenting cell. Additional signals provided by the antigen presenting cell are also required for the activation of the T lymphocytes. Among these are signals provided by other accessory molecules found on the antigen presenting cell and various cytokines present in the extracellular environment.

Signaling Through the Antigen-specific Receptors The encounter of speci c antigen either by a T cell or a B cell’s antigen-speci c receptor results in an intracellular signaling cascade that eventually leads to the production of various proteins and the proliferation of the stimulated cell. T cell recognition of antigen involves the engagement of a TcR-CD3-CD4 or TcR-CD3-CD8 complex with processed peptide in the cleft of a MHC II or MHC I molecule (Figure 1-11).3 The engagement of the TcR-CD3 complex with the appropriate MHC antigen containing peptide results in the binding of CD4 or CD8, depending on the MHC antigen, with the TcR-CD3 complex. In doing so the Lck protein tyrosine kinase associated with the cytoplasmic tail of CD4/CD8 phosphorylates the cytoplasmic regions of the CD3 proteins in regions known as immunoreceptor tyrosine-based activation motifs (ITAMs; Figure 1-12). These ITAMs serve as docking sites for other kinases, including ZAP70 and Fyn. Recruitment of ZAP70 to CD3 results in its subsequent phosphorylation and activation by Lck. Once activated, the ZAP70 can subsequently phosphorylate other signal proteins, including phospholiase C (PLC; Figure 1-13). Activation of PLC leads to the cleavage of phosphatidylinositol bisphosphate (PIP2) into inositol 3-phosphate (IP3) and diacylglycerol (DG). Both IP3 and DAG are second messengers with IP3, causing the release of stored CA2+ from the endoplasmic reticulum and DAG-activating protein kinase C. The increase in intracellular Ca levels and the activation of protein kinase C lead to the phosphorylation of various transcriptional factors. These transcriptional factors regulate the expression of the genes for various cytokines and/or their receptors (see Figure 1-13). The process is subsequently downregulated by various phosphatases that are recruited to and subsequently dephosphorylate the CD3 ITAMs. A similar process occurs in a B cell when its surface immunoglobulin receptor is cross-linked upon binding to specific antigen.

FIGURE 1-11

Class I and class II restricted T cell recognition: the role of T cell CD4 and CD8 molecules. T cell use their T-cell receptors to recognize processed antigen presented in combination with either MHC I or MHC II molecules. T cells exclusively express either CD4 (T-helper cells) or CD8 (cytotoxic lymphocytes; CTLs), and the CD4 molecule is required for interaction with MHC II molecules, whereas CD8 is required for MHC I interaction. As a result, T-helper cells recognize antigen presented by MHC II, and CTLs recognize only antigen presented by MHC I molecules.

FIGURE 1-12

Intracellular signaling by the TcR-CD3 receptor. TcR recognition of its speci c peptide in the peptide binding groove of a MHC molecule on an antigenpresenting cell results in the attraction of CD4/CD8 to the complex (1) and the phosphorylation of CD3 proteins by lck associated with CD4/CD8 (2). The phosphorylation of these sites ( ∗) on CD3 leads to the attraction and binding of other kinases (fyn and ZAP 70) to CD3, where they are in turn phosphorylated and activated. Activation of ZAP70 leads to the subsequent activation of phospholipase C (4), and activation of fyn ultimately leads to the MAP kinases pathway and cell division (5) (see also Figure 1-13).

FIGURE 1-13

Intracellular signaling pathway. Following activation of ZAP70, and other receptor-associated kinases, there is a subsequent propagation of the signal as subsequent kinases and target proteins are phosphorylated. Increases in intracellular CA2+ lead to the activation of calcineurin that is necessary for NF-AT activation. This step is the target for cyclosporin A, a potent and specific immunosuppressive agent. Activation of the transcriptional factors NF-AT and fos/jun leads to their translocation into the nucleus and the binding to regulatory DNA sequences upstream of the promoter for the IL-2 gene.

Co-stimulatory Signals In addition to the interaction of TcR-CD3 and CD4/CD8, other cell surface antigens are involved in the signaling pathways.3 Of greatest importance is the interaction of CD28 on the T cell with B7 on the antigen presenting cell. In the absence of CD28/B7, co-stimulation T cells are rendered functionally inactive or anergic. Upon restimulation, these anergic T cells failed to proliferate or produce cytokines such as IL-2. The induction of anergy can be prevented either by the addition of exogenous IL-2 or, more important, by interaction of the CD28 cell surface antigen with its ligands, B7-1(CD80) and B7-2(CD86). Stimulation of CD28 appears to be necessary for subsequent intracellular signaling events following TcR stimulation as CD28 cross-linking enhances various biochemical events triggered by TCR-mediated signaling, including the activation of PLC, Lck, and Raf-1 kinase, as well as inducing the in ux of Ca2+ and generation of phosphoinositides. Other molecules, including the TNF-receptor family member CD40, regulate either T cell growth or cell death. The engagement of CD40 on the T cell with its ligand, CD40L, on the antigen presenting cell leads to NF-кB activation and thus promotes cell survival and cell cycle progression. The binding of other members of this family, notably TNF-α, to their receptor on activated T cells typically results in the activation of a biochemical cascade of caspases that lead to apoptosis. The cytocidal activity of these receptors is the result of their intracytoplasmic portion of the receptor containing death e ector domains. By contrast, CD40 lacks the intracellular death domains and instead has amino acid motifs that bind TNF-R–associated factors (TRAFs) and promote NF-kB activation. In addition to their role in promoting T cell activation and growth, both the CD28/B7 and TNF-receptor pathways may also play a dominant role in the induction of specific T helper cell subsets.

CYTOKINES, CYTOKINE RECEPTORS, AND T HELPER CELL SUBSETS Frequent mention has been made in this chapter of the role of cytokines in regulating immune responses. Indeed, this particular area of immunology has seen impressive growth over the past several years. This rapid acquisition of knowledge is the result of the application of modern molecular biology techniques to the identi cation and characterization of speci c cytokines. Initial studies of the role of soluble factors in the regulation of immune responses were often confounded by the heterogeneous nature of the culture supernatants used as the source of the cytokine activity. Furthermore, the use of biological assays to identify speci c cytokines resulted in the practice of assigning descriptive names to newly discovered cytokines (e.g., lymphocyte activating factor, T cell growth factor).61 This quickly led to confusion because individual cytokines often exhibited multiple biological activities and the biological assays were not speci c for a particular cytokine. Once the genes for the cytokines had been cloned and the resulting proteins identi ed, it was possible to eliminate much of this confusion. The adoption of the interleukin (IL) terminology for naming cloned immunoregulatory cytokines has further clari ed the biological function and role of particular cytokines.62 Once a new cytokine’s gene is identi ed and the biological activity of the puri ed protein characterized, it is assigned an interleukin designation. To date, more than 50 di erent cytokines and chemokines have been cloned, sequenced, and synthesized in bacterial and eukaryotic expression systems. This has led to both a better understanding of cytokine function and to their application in a variety of clinical settings. Table 1-4 contains a list of interleukins and their known biological activity. Not all cytokines have been given interleukin designation. Interferons, certain growth factors (platelet derived growth factor, TGF-β), and TNF-α have retained their original names. It should also be emphasized that other cells besides T cells produce cytokines and interleukins. For example, monocytes and macrophages are the major source of IL-1, IL-6, and TNF-α. Thus the term lymphokine, which was originally used to describe immunoregulatory products of lymphocytes, has been replaced with cytokine, which denotes the more varied sources of immunoregulatory molecules. TABLE 1-4 Interleukins IL 1 2 3 4 5 6

Biologic Activities and Source Lymphocyte activating factor; multiple biologic activities affecting a variety of lymphoid and nonlymphoid cells T cell growth factor; provides proliferative signal for T cells; also affects B cells, macrophages, and NK cells; high concentrations of IL-2 stimulate cytolytic activity in NK cells and T cells; produced by activated Th1 and some CD8+ cells Multi-CSF; promotes the growth of various hematopoietic cell precursors; produced by T cells and myelomonocytic cell lines B cell stimulatory factor 1; stimulates growth, maturation, and differentiation of B cells; also provides proliferative and differentiation signals for some T cells; produced by Th2 cells T cell–replacing factor; stimulates B cell proliferation and immunoglobulin synthesis; also stimulates T cell proliferation and differentiation as well as eosinophil formation in the bone marrow; produced by Th2 cells B cell differentiation factor; promotes maturation and immunoglobulin production by B cells; stimulates T cell growth and IL-2 synthesis; induces the production of acute

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

phase proteins by hepatocytes; produced by macrophages, T cells, stromal cells, fibroblasts, and a variety of other cell lines Pre-B cell growth factor; stimulates proliferation and maturation of early B and T cells as well as mature T cells; produced by bone marrow–derived stromal cells Neutrophil chemokine produced by monocytes and hepatocytes Also known as P40; supports the growth of certain T cell clones; produced by CD4+ T cells Cytokine synthesis inhibitory factor; inhibits the production of IL-2 and interferon-γ by Th1 cells; produced by Th2 cells An IL-6–like factor produced by bone marrow stromal cells NK cell differentiation factor; augments NK cell function and stimulates generation of Th1 cells; produced by macrophages Produced by Th2 cells; downregulates cytokine production by macrophages/monocytes while activating B cells A high-molecular-weight B cell growth factor produced by T cells and some B cell lines A T cell growth factor similar in function to IL-2 Chemokine for CD4+ T cell subset; produced by T cells, mast cells, and eosinophils A family of related cytokines; enhances expression of the ICAM-1 on fibroblasts; also stimulates epithelial, endothelial, or fibroblastic cells to secrete IL6, IL8, and G-CSF and PGE2 An inducer of IFN-gamma production by T-cells A hom*olog of IL-10 An autocrine factor for keratinocytes that regulates their participation in inflammation Stimulates proliferation of B-cells stimulated by cross-linking of the CD40 antigen, bone marrow progenitor cells, and naïve T cells A proinflammatory cytokine that increases the production of acute phase proteins Produced by dendritic cells; stimulates the production of IFN-γ by T-cells Selectively suppresses the growth of tumor cells by promoting cell death by apoptosis Produced by stromal cells and supports proliferation of cells in the lymphoid lineage Induces the expression of IL6, IL8, and ICAM-1 in primary bronchial epithelial cells Similar to IL-17; expressed in brain Type III interferons; similar to Type I interferons in function and induction Another Type III interferon Similar to IL-27, an early product of antigen presenting cells that stimulates IFN-γ production Produced by Th2 cells and is related to the IL-6 family of cytokines; possible role in pruritis, alopecia, and other allergic skin lesions Inducer of inflammatory cytokine production by monocytes; highly expressed in synovial tissue of rheumatoid arthritis Inducer of Th2 cytokines

IL, Interleukin; NK, natural killer; CSF, colony-stimulating factor; ICAM-1, intracellular adhesion molecule-1.

Many cytokines have similar structures and can be grouped into like families. Hence helical cytokines have alpha helices as the predominant structure with IL-2 serving as the prototypical cytokine for this family (Figure 1-14). This family can be further divided into two subclasses according to the length of the helices: long helical (many growth factors including IL-3 and IL-7) and short helical (IL-2, IL-4, and IL-13). IL-1 is a β-trefoil cytokine whose overall structure is composed of 12 antiparallel β strands that form a bowl-like structure. Most chemokines and other smaller cytokines contain both α helices and β sheets, typically a single α helix and more than two β sheets. TNF-α is the prototype for the β-sandwich family whose structure characteristic consists of five antiparallel strands with an overall jelly roll structure.

FIGURE 1-14

Cytokine structural families. (a) helical cytokine, (b) β-trefoil, (c) α/β, and (d) β sandwich.

The availability of cloned cytokines has also permitted the subsequent identi cation and characterization of cytokine-speci c receptors. Cytokine receptors can also be grouped into major families: Class I or Class II receptor families, immunoglobulin superfamily receptors, the TNF receptor family, and TLRs (IL-1, IL-18).63 The best characterized of these is the Class I receptor for IL-2, which is composed of three subunits: α, β, and γ. Whereas the α and β subunits are involved in speci c binding to IL-2, the γ chain is involved in signal transduction once IL-2 is attached to the receptor. Five di erent immunologically important cytokines share this common cytokine receptor γ chain, though each has its own unique α and/or αβ binding subunits (Figure 1-15). Other common signaling chains of this type of receptor include βc (IL-3, IL-5, and GM-CSF) and gp130 (IL-6 and IL-11). Class II receptors are illustrated by the interferons whose receptors consist of at least two chains. In contrast to the chains being denoted α and β, analogous to the nomenclature for type I cytokine receptors, the chains are called IFNAR-1 and IFNAR-2 for α/β interferons and IFNGR-1 and IFNGR-2 for interferon-γ. A third receptor, CRF2-4, is a component of the IL-10 receptor. The TNF receptor family is composed of two separate receptors, TNF-RI and TNF-RII. Though both can bind TNF, no structural hom*ology is found in their intracellular domains, indicating that they signal by distinct mechanisms. TNF-RI is thought to be the main signaling receptor because many biological actions of TNF, including cytotoxicity, broblast proliferation, and the activation of NF-кB, can be elicited in the absence of TNF-RII. The IL-1 receptor is a member of the Toll-like receptor family. As discussed previously, this receptor superfamily represents an ancient signaling system that was initially identi ed in Drosophila melanogaster. The introduction of a pathogen into Drosophila spp. leads to the activation of proteases that cleave a precursor and generates an extracellular ligand of a receptor called Toll, the intracellular part of which is hom*ologous to the IL-1 receptor cytoplasmic tail. Other Toll-like receptors are involved in other signaling pathways involved in innate immune and in ammatory responses, indicating that this receptor superfamily represents an ancient signaling system.64

FIGURE 1-15

Type I cytokine receptors. These receptors are characterized as having a cytokine speci c α and β chain involved in ligand binding and a shared or common γ chain that is used for intracellular signaling. The Janus kinases (JAK) are associated with the cytoplasmic tails of these receptors and are responsible for the signal transmission.

A common feature of all these receptor families is that signaling is initiated through the recruitment of protein tyrosine kinases and other cytosolic proteins to the receptor.65,66 Although most cytokine receptors lack intrinsic kinase activity, they do have a family of Janus protein tyrosine kinases (JAKs) associated with their cytoplasmic tails. Following binding of a ligand to its cognate receptor, receptor-associated JAKs are activated. A family of transcriptional factors known as STAT (signal transducers and activators of transcription) are in turn activated by tyrosine phosphorylation by the activated JAKs, allowing the STAT to dimerize. After dimerization the STATs translocate into the nucleus and bind to the DNA sequence it recognizes via a DNA binding domain on the protein. The binding of the STAT proteins to DNA subsequently modulates gene expression. It is the sharing of receptor subunits, combined with a similar sharing of JAKs and STATs, that accounts for similar biological functions of many cytokines (see Table 1-4). In addition to the JAKs and STATs, other transcriptional factors can activate multiple genes involved in in ammatory responses and apoptosis. One of these transcriptional factors, NF-кB, regulates many pro-in ammatory cytokines, including TNF-α, IL-1, and IL-8. NF-кB itself is activated by a number of cytokine receptor–signaling cascades, including TNF receptors.67 In the cytoplasm NF-кB is associated with an inhibitory protein, IкB, which prevents its translocation to the nucleus. Phosphorylation of IкB leads to its degradation and the translocation of NF-кB to the nucleus, where it binds to its corresponding DNA motif, altering gene transcription. NF-kB activation is also associated with resistance to apoptosis, probably as the result of its e ect on IL-8 transcription because this chemokine is anti-apoptopic.68 Increased levels of IL-8 in in ammatory lung lesions and the increase in NF- B activation likely account for the neutrophil accumulation seen in some forms of human asthma69 and equine recurrent airway obstruction.70

Immunoregulation The generation of an immune response requires the interaction of multiple leukocyte subsets, including macrophages, dendritic cells, B cells, and both CD4+ and CD8+ T cells. Whereas the initial interactions of B and T cells involves the recognition of speci c epitopes, which in the case of the T cell are presented in the context of MHC antigens, subsequent interactions are mediated by the cytokines produced by the various cells. Although macrophages, B cells, and even nonhematopoietic cells produce a variety of cytokines with immunoregulatory activity, it is the T helper cell that plays a central role in regulating immune responses. Much e ort over the past decade has focused on the characterization of helper T cells and the soluble factors they produce. It is now apparent that CD4+ helper cells may be further divided into distinct helper cell subsets on the basis of the cytokines they produce (Figure 1-16). Thus Th1 cells produce interferon-γ and IL-2, two cytokines involved in the induction of cell-mediated immune responses. Th2 cells, on the other hand, produce IL-4, IL-13, and IL-5, cytokines involved in the induction of antibody responses. The best evidence for separate T helper cell populations comes from the study of intracellular parasite infections in mice. Those strains of mice resistant to Listeria donovani infection develop a cell-mediated immune response characterized by activated macrophages and Th1 helper cells. By contrast, the susceptible BALB/c strain of mice generates a vigorous antibody response and Th2 helper cells. Th1 cells have also been implicated in various autoimmune diseases characterized by the induction of self-reactive cytotoxic cells. Th2 cells in turn play a central role in the resistance to extracellular parasites, such as intestinal helminthes, and in the induction of allergic diseases. A similar contribution of Th1 and Th2 responses in protective and pathologic responses in the horse has been described.71,72 Regulatory T cells (Tregs) are specialized subsets of T cells that play a central role in the prevention of hyperimmune responses and autoimmunity and are thought to represent traditional suppressor cells.73 Although naturally occurring Tregs constitute 5% to 10% of the CD4+ T cell population, antigen-speci c or adaptive Tregs, also known as Th3 cells, are also present. These

di erent Tregs subsets can be identi ed on the basis of the expression of cell surface markers, production of cytokines, and mechanisms of action. Most Tregs express the CD25 surface antigen and produce various immunoregulatory cytokines and growth factors, including IL-10 and/or TGF-β.74 Besides CD25, Tregs can express several other activation markers, such as the glucocorticoid-induced TNF-receptor–related protein (GITR), OX40 (CD134), L-selectin (CD62 ligand [CD62L]), and cytotoxic T lymphocyte-associated antigen 4 (CTLA-4 or CD152).74 However, it should be noted that these markers can also be expressed to various degrees on activated T cell subsets and various antigenpresenting cells.74 A speci c subpopulation of Tregs can be identi ed on the basis of the expression of the transcriptional factor FoxP3. The FoxP3+ Tregs are thought to be important regulators of autoimmunity insofar as loss of function or mutation in this factor leads to the development of autoimmune disease.74 A population of NK cells and natural killer T (NKT) cells with regulatory function has also been described whose immune suppressive function is mediated by secretion of various cytokines (IL-13, IL-4, IL-10) or by direct cell-cell contact. 74 Although the suppressive function of most Tregs is likely mediated by the production of suppressive cytokines, direct contact with Th1 cells, which limits interaction with dendritic cells, is another mechanism.74,75

FIGURE 1-16

Th1and Th2 regulation. The Th1 lymphocyte subsets provide help for macrophage activation, cytolytic activity, and production of a subset of IgG subclasses. The Th2 promotes antibody responses, including IgA, IgE, and the remainder of the IgG subclasses. This is mediated by production of cytokines, which have a regulatory e ect on each other.

Although it remains unclear as to what determines whether a helper cell will be a Th1or Th2 cell, it has been proposed that it is the initial encounter with the antigen during the innate immune response that may determine its fate. Multiple factors are probably involved in this process, but the single most important factor is the type and amounts of cytokines present at the time of the initial encounter with the antigen. Among the cytokines that may play a role, IL-12 and interferon-γ are the main inducers of Th1 responses, and IL-4 and IL-10 play a similar role for Th2 responses. IL-12, produced by macrophages and dendritic cells, is a potent inducer of interferon-γ that inhibits Th2 cell induction. The evidence from a variety of models to date suggests that IL-12 is the single most important factor in regulating the di erentiation and magnitude of the Th1 response; however, this should not preclude the possibility that additional factors, such as IL-18, may have an equally important role in regulating Th1 responses in some situations. It is also apparent that IL-4 plays a similar crucial role in the induction of Th2 immune responses. IL-4 production by mast cells and IL-10 production by macrophages favor Th2 development in part by inhibiting Th1 cells. Other signals (e.g., PGE2) can induce di erentiation of potent antigen presenting cells with dendritic morphology that produce low levels of IL-12 and high levels of IL-10 preferentially inducing Th2 differentiation. In addition to cytokines, another in uence on T helper cell di erentiation is the interaction of the B7/CD28 and CD40/CD40L costimulatory pathways.3 In particular, these co-stimulatory pathways may regulate the di erentiation of Th1 and Th2 cells by a ecting the intensity and the strength of signals through the CD3/TCR complex and those provided through the co-stimulators. High-intensity stimulation favors Th2 development, whereas lower intensity signals favor Th1 cells. The association of stronger co-stimulation with Th2 responses contrasts with the ability of higher concentrations of antigen to mostly induce Th1 responses, whereas lower doses induce Th2 responses suggesting that Th di erentiation may be in uenced quite di erently by the strength of signals through the CD3/TCR complex versus signals delivered through co-stimulators and their associated enzymes.3

Th Paradigms The role of cytokines in regulating immune responses can best be illustrated in two scenarios, the rst involving the induction of a Th1 immune response in response to viral infection and the second an allergic response to inhaled mold antigens. In the rst scenario, viral antigen present at the site of an ongoing infection in the respiratory tract is processed by a resident macrophage via the exogenous pathway. The processed epitope is presented on the surface of the macrophage or a dendritic cell in the context of a MHC II antigen to a CD4+ T cell in a regional lymph node. Additionally, these cells produce IL-12 that induces NK cells, attracted to the site of the infection, to produce interferon-γ that, along with antigen presentation, activates the T cell and drives it toward a Th1 phenotype. Meanwhile, CD8+ T cells encounter viral antigen on the surface of virus-infected cells that has been processed via the endogenous pathway and is now associated with the MHC I antigens on the infected cell’s surface. Once antigen-activated, these T cells express the high a nity form of the IL-2 receptor. The CD4+ Th1 cell produces IL-2 and interferon-γ.The interaction of IL-2 with its receptor drives the clonal proliferation of the activated CD8+

T cells. The interferon-γ also stimulates the CD8+ cell to di erentiate into CTLs that produce additional interferon-γ. These CTLs can lyse the target cells either through the production of TNF-α or via the activation of the fas receptor on the target cell via the fas ligand (fasL) expressed on the activated CTL. Both pathways lead to target cell apoptosis via the activation of cytoplasmic caspases in the target cell. Meanwhile, virus-speci c B cells have also encountered antigen and in the presence of interferon-γ di erentiate into IgG-secreting plasma cells. This combination of IgG antibodies and CTL cells serves to eliminate the virus. PGE2 and IL-10 production by macrophages exert antiin ammatory e ects on the response, and that, coupled with the production of soluble cytokine receptors, dampens the immune response as the invader is eliminated. In the second scenario, the introduction of mold antigens into the respiratory tract leads to the processing of the antigen by macrophages and dendritic cells, as before. However, in the absence of IL-12 and interferon-γ, and perhaps the presence of IL-3, IL-4, IL-9, IL-10, or PGE2, there is the induction of Th2 cells that produce additional IL-4 and IL-13. These cytokines cause those B cells recognizing the allergens to isotype switch to IgE antibodies that bind to mast cells. Subsequent degranulation of these mast cells ensues, as the result of antigen binding to the IgE, leading to the production of other mediators, including IL-4 and PGE2 that exacerbate this response. In the continued presence of the allergen, a secondary inflammatory response characteristic of recurrent airway obstruction occurs.

The Role of Cytokines in the Horse The eld of equine immunology continues to expand with the development of better reagents. Recent advances in gene cloning technology have led to the cloning and expression of a number of equine cytokines. Thus the cDNA sequences for a number of equine cytokines are known, and speci c protocols are now available to measure their expression (Table 1-5). Through use of these procedures, it has been possible to identify the role of Th1 and Th2 cells in both protective and pathologic responses in the horse (Table 1-6).71,76,77 The results from these and other studies con rming the role of in ammatory cytokines in equine sepsis and in joint and airway diseases emphasize the similarities between equine and human immune systems. As such, the potential for manipulating these responses using recombinant cytokines or anticytokine reagents is as applicable to equine medicine as it is to human medicine. TABLE 1-5 Cloned Equine Cytokines

TABLE 1-6 T Helper Cell Paradigm in the Horse Protection Th1 Th2

Immunopathology Rhodococcus equi Equine recurrent uveitis Strongylus vulgaris Insect bite hypersensitivity

Th, T helper cell.

Lymphocyte Trafficking Pathways Leukocyte tra cking has been reviewed previously, with a particular emphasis on the innate immune response. Lymphocytes involved in adaptive immune responses di er in their migration from most other cells in that they recirculate instead of making one-way trips. Memory and naïve T lymphocytes, with their di erent capacities for response to antigens, di er also in their migration pathways through the body. Two general pathways of lymphocyte recirculation have been demonstrated. Naïve T lymphocytes take the most common route, which involves entry into the lymph node by extravasation from the high endothelial venule (HEV) and return to the peripheral circulation via the e erent lymphatic. The endothelial cells of HEVs have a distinctive appearance and specialized receptors and can support a great deal of lymphocyte migration. This allows rapid repeated circulation of naïve lymphocytes through lymph nodes where there is the greatest chance of exposure to their speci c antigens. Memory lymphocytes, on the other hand, leave the bloodstream in peripheral vascular beds, particularly in in amed tissues, and return to lymph nodes via a erent lymphatics. This leads to the exposure of primed memory lymphocytes to the most likely early sites of antigenic encounter and allows for an early response to recall antigens. Thus memory lymphocytes are most common in in ammatory lesions and in the epithelial surfaces of the lung and gut wall. Di ering expression of the adhesion and homing molecules may play an important role in mediating these different migration pathways. For lymphocytes to follow the previously outlined maturation and migration pathway, the rst step is for the naïve lymphocyte to get into a lymph node so that it can meet its antigen on a professional antigen-presenting cell. To achieve this, the T-lymphocyte needs to exit in the HEV. The naïve lymphocyte expresses L-selectin, and this can bind to the vascular addressins GlyCAM-1, CD34, and MAdCAM-1, which are expressed on HEVs and promote rolling similar to that mediated by P- and E-selectin when they bind to phagocytes. These molecules are expressed on a variety of tissues, but in HEVs they have speci c patterns of glycosylation that makes them bind L-selectin. These di erences represent the key to the speci city of the migration of lymphocytes to HEVs. This weak interaction initiates the process of extravasation that is promoted by locally bound chemokines (e.g., IL-8), which increase the affinity of the lymphocyte integrins for their ligands. Approximately 25% of lymphocytes passing through an HEV leave, and this could mean 1.4 × 104 cells in a single lymph node every second, and in the body 5 × 106 lymphocytes may extravasate through HEVs every second (human). The “sticking” process (rolling, activation, arrest) takes a few seconds, with transendothelial migration and passage through the HEV basem*nt membrane occurring in about 10 minutes. After leaving the blood, most T cells travel through the lymph node uneventfully and leave via e erent lymphatics; however, in rare events a naïve T cell recognizes its speci c peptide/MHC complex and becomes activated, eventually leading to formation of e ector and memory T-cells. That process takes 4 or 5 days, and, once activated, the migration pathway of memory T cells di ers considerably from naïve cells. All activated T cells lose the L-selectin molecules that mediated homing to lymph nodes and increase the expression of other adhesion molecules. The homing of individual lymphocytes to speci c sites is regulated by expression of speci c adhesion molecules. Memory cells are speci cally attracted to areas of in ammation as a result of the increased levels of adhesion receptor ligands expressed on vascular endothelium in these regions. This is typically a result of TNF-α production by regional macrophages encountering infections. Sometimes infections do not result in TNF-α production, but memory cells also migrate randomly throughout the body. When they encounter their antigens, they can produce cytokines like TNF-α themselves, which in turn causes local endothelial cells to increase expression of Eselectin and VCAM-1 and ICAM-1. This will subsequently recruit more effector and memory cells to the region.

Mucosal Immunity The mucosal immune system comprises a series of distinct compartments within the imune system, which are adapted to immunologic response in unique environments such as the gut or respiratory or urino-genital tract. The mucosal immune system is perhaps the most important component of our adaptive immune system, and the reader is referred elsewhere for an appropriately detailed description of its general features,1 and of its role in equine immunity in the context of respiratory disease.78 The mucosal immune system may represent the original vertebrate immune system, and it certainly protects the largest vulnerable area of the mammalian body and composes a large proportion of the total lymphocyte populations and immunoglobulin pool.79,80 The mucosal immune system consists of organized and dispersed lymphoid tissues that are closely associated with mucosal epithelial surfaces, and mucosal immune responses generated in one location are transferred throughout the mucosal immune system by lymphocytes programmed to home to regional e ector sites. The principal immunoglobulin produced by the mucosal immune system is secretory IgA, which in humans is the most abundant immunoglobulin class in the body. Secretory IgA has unique adaptations that promote transport out onto mucosal surfaces, where it protects the body from bacteria and viruses principally by immune exclusion (i.e., by physically preventing attachment to mucosal surfaces). The importance of mucosal IgA has already been demonstrated in immunity to numerous equine diseases.78 Secretory IgA (sIgA) is formed by dimerization of two IgA monomers, which are attached by means of disulphide bonds to a J-chain also produced by the same plasma cell that secretes the IgA. This confers the advantage of increased valency to sIgA, which can bind up to four of its targets, thus increasing its agglutinating ability. Secretory IgA protects the body from bacteria and viruses principally by immune exclusion

(i.e., by physically preventing attachment to mucosal surfaces). Immunoglobulin A is relatively nonin ammatory (i.e., it does not x complement as e ectively as IgGa or IgGb), consistent with a role in defense by immune exclusion.3 Similarly, although myeloid cells possess Fc receptors for IgA, it is not clear that IgA functions as an efficient opsonin or promotes phagocytosis. Coordination of the mucosal immune response depends on organized mucosal-associated lymphoid tissue (MALT) principal examples of which are the pharyngeal tonsils and the intestinal Peyer’s patches. In the gastrointestinal tract MALT is distributed throughout the gut, but in the respiratory tract these tissues are found only in the nasopharynx and oropharynx. MALT consists of lymphoid follicles containing IgAcommitted B cells, surrounded by interfollicular T cell areas with APCs and high endothelial venules (HEVs), with an overlying follicleassociated epithelium (FAE). Naïve lymphocytes enter the MALT by extravasation from the HEVs (there are no a erent lymphatics in MALT), and, after antigen encounter in the MALT, they leave through e erent lymphatics. The FAE is specialized for antigen sampling, by having reduced secretion of mucus, and through the presence of specialized antigen uptake cells termed microfold or M cells. These M cells are typically closely associated with underlying aggregates of lymphocytes, often within large basolateral membrane pockets, and play a critical role in mucosal immune surveillance. Adherent macromolecules or particles bound to the apical M cell membrane undergo endocytosis or phagocytosis and are released at the pocket membrane, where antigen presentation is initiated by dendritic cells resulting in activation of antigen-speci c B cell (Figure 1-17). Subsequent tra cking and recirculation of memory IgA-positive B cells to the other components of the mucosal immune system (e.g., respiratory tract, intestinal tract), is responsible for the dissemination of local mucosal IgA responses throughout what is termed the common mucosal immune system. After homing of these B cells to e ector sites, such as the lamina propria of the gut and respiratory tract, and extravasation into the lamina propria from HEVS, further antigen encounter and second signals from antigen presentation cells and from T helper cells result in further di erentiation into IgA producing plasma cells. The short half-life of IgA secreting plasma cells requires a constant generation of precursors in induction sites and ow to e ector sites. Antigen sampling and presentation are not restricted to organized MALT because throughout the mucosal surfaces dendritic cells play a key role in antigen uptake and presentation, subsequently migrating to local lymph nodes or MALT to initiate immune responses.

FIGURE 1-17

Initiation of mucosal immune responses. Respiratory mucosal immune responses typically originate after antigenic encounter at inductive sites, which are the tonsils of the nasopharynx and oropharynx in the horse. Naïve lymphocytes enter the inductive sites from high endothelial venules (HEV) via the specialized cuboidal endothelium of those vessels in response to speci c molecular signals. Antigens, such as microbes, are taken up by microfold or M cells, which are part of the highly specialized follicle associated epithelium present at these sites. Antigenic material is transported across the M cell, and antigen presentation to B and T lymphocytes is accomplished by dendritic cells in the underlying tissues. The underlying lymphoid follicle is composed primarily of B lymphocytes, surrounded by T lymphocytes areas. Antigen-speci c B lymphocytes become committed primarily to IgA production at these sites, although some IgG B lymphocytes are also generated. Subsequently, the primed lymphocyte populations exit the inductive site via e erent lymphatics, eventually reaching the blood circulation through the thoracic duct. Then these cells tra c to HEVs of e ector sites throughout the respiratory epithelium and extravasate to make up the intraepithelial lymphocyte and lamina propria lymphocyte population and to give rise to lymphoid aggregates. Subsequent antigen encounter results in terminal di erentiation to plasma cells, primarily IgA producing, although some IgG plasma cells are also formed. IgG is largely restricted to tissues, but secretory IgA is transported to the respiratory epithelial surface, where it can aggregate infectious organisms. (From McGorum BC, Dixon PM, Robinson NE, et al. Equine respiratory medicine and surgery, Edinburgh, Saunders, 2007).

After release of secretory IgA by plasma cells into the interstitium, it is bound by the polymeric Ig receptor on the abluminal surface of epithelial cells. Subsequently, the sIgA is transported across the epithelial cell and released at the luminal surface together with secretory

component formed by cleavage of part of the polymeric Ig receptor. Secretory component can also be found in a free form in mucosal secretions. Secretory component confers resistance to proteolytic enzymes found in the respiratory and gastroenteric environment, some of which are secreted by pathogens, and prolongs the longevity of sIgA. During its transit through the epithelial cell sIgA can neutralize intracellular infections encountered in the endosomal compartments of cells.81 In addition, sIgA can bind antigens in the submucosa and literally transport or excrete them to the mucosa by this mechanism. The majority of the IgA in the mucosa is dimeric sIgA, whereas the bone marrow–derived IgA in circulation is predominantly monomeric. In the horse our understanding of the architecture and functions of the mucosal lymphoid system is best developed for respiratory lymphoid tissues.78 Although lymphoid tissues are distributed throughout the respiratory tract, the greatest masses comprise the nodular lymphoid tissue of the nasopharynx and oropharynx, which can have an overlying lymphoepithelium specialized for antigen uptake and processing, as in the case of tonsillar tissues. Additional nodular lymphoid tissues are typically present at sites where antigen-laden mucus and air currents converge throughout the trachea and bronchi and are called bronchus-associated lymphoid tissue (BALT). Tonsils represent the most complex mucosal nodular lymphoid tissues. Horses possess all of the various tonsillar tissues that are recognized in other species, and they are anatomically complex.78 The nasopharyngeal tonsil is the largest mass of lymphoid tissue in the respiratory tract of horses of all ages, and its epithelium has been extensively characterized.82 This epithelium has a classical FAE and is heavily folded, forming crypts; it also contains M cells. The nasopharyngeal tonsil exists in the dorsal recess of the nasopharynx, and extends ventrally toward the opercula on either side of the nasopharynx. Therefore it is ideally placed for the sampling of antigens before entry to the airways or alimentary tract and may serve as an important target for intranasal vaccines. This tissue appears to be most abundant in young foals and atrophies with age, though many lymphoid follicles remain throughout the nasopharynx. The nasopharyngeal epithelium also contains numerous lymphocytes. Immmunohistochemical studies indicate that the majority of these lymphocytes are CD8+ T lymphocytes, although B lymphocytes are also present.82 The contribution of these lymphocytes to mucosal cellular immune defenses of the upper respiratory tract is poorly studied. However, following intranasal challenge of yearling and 2-year-old horses with EHV-1, virus-speci c cytotoxic activity is detectable in several mucosal lymphoid tissues of the upper respiratory tract, as well as the local draining lymph nodes, and is particularly evident in the nasopharyngeal lining.83 This cellular immune response is presumably mediated by CD8+ T lymphocytes found in the nasopharyngeal epithelium and underlying lamina propria and may provide an important contribution to the clearance of infectious virus from the upper respiratory tract.

Ontogeny of the Equine Immune System Few studies of the prenatal development of the equine immune system have been conducted. As in other species, the thymus is the rst lymphoid organ to develop, and mitogen responsive cells can be identi ed there from day 80 of the 340 day gestational period of the horse.84 Subsequently, these cells appear in peripheral blood at 120 days, lymph nodes at 160 days, and the spleen at 200 days. Cells responsive in mixed lymphocyte reactions are detectable in the thymus from 100 days and in the spleen at 200 days. Immunoglobulin production is detectable before 200 days, and newborn foals typically have IgM concentrations in their serum of approximately 165 g/ml. Overall, it appears that functional T lymphocytes are present by day 100 and B lymphocytes by day 200 of gestation. Immunologic competence of the equine fetus has been assessed in terms of speci c antibody responses. In utero immunization of foals in late gestation with keyhole limpet hemocyanin in an alum adjuvant results in detectable speci c antibody production and T cell responsiveness at the time of birth.85 In addition, the equine fetus can respond to coliphage T2 at 200 days, and Venezuelan equine encephalitis virus at 230 days.86,87 Detailed studies of the appearance of lymphocyte subpopulations de ned by monoclonal antibodies have not been performed in the equine fetus. However, some information regarding the maturation of thymocytes in young horses is available. During thymic maturation of T cells, stem cells migrate into the thymus and mature into T cells under the in uence of the epithelial microenvironment.88,89 In this process di erent patterns of cell surface di erentiation and antigen expression distinguish successive stages of thymocyte maturation. In humans the earliest thymic precursor cells express low levels of CD4.90 This CD4 expression is lost as early thymocytes become double negative CD4CD8-cells and then demonstrate their T cell commitment by TCRβ gene rearrangement, which is an essential trigger for subsequent events and leads to low levels of expression of a cell surface TCRβ-CD3 complex.91 Intermediate thymocytes are CD4loCD8lo, but after TCRα gene rearrangement and expression of cell surface TCRαβ, they rapidly become CD4hiCDhiTCR-CD3hi.90 Subsequently, thymocytes selected on the basis of productive TCR gene rearrangement and lack of self-reactivity become mature T cells expressing either CD4 or CD8 (single positive) in combination with high levels of TCR-CD3. Using two-color FACS analysis, it is possible to demonstrate similar patterns of EqCD3, EqCD4, and EqCD8 antigen expression in the equine thymus.92,93

Immunocompetence in Foals Infectious disease in neonatal foals is associated with high morbidity and mortality. Although failure of passive transfer is a major cause of this problem, immaturity of the immune system has also been considered a potential contributing factor. As a result, a number of studies of neonatal immunocompetence have been completed and reviewed.94

Innate Immune Responses in Foals

A number of studies have reported neutrophils to be fully functional from birth,95-97 but their function is signi cantly impaired before absorption of colostral antibodies, which are required for opsonization.97,98 A recent study of foal neutrophil development over the rst 8 months of life demonstrated killing (measured by chemiluminescence) to be reduced in the rst 2 weeks of life, as was phagocytic ability when assays were performed using autologous serum.99 When serum from adult horses was used, neutrophil phagocytic ability in foals was normal. This latter di erence may have been due to absence of either adequate immunoglobulin or complement in foal serum. A similar study of foals younger than 7 days of age con rms that phagocytosis and oxidative burst activity of neutrophils is reduced in foals of this age, although the use of adult serum did not improve phagocytosis.100 Similarly, alveolar macrophages recovered from bronchoalveolar lavages may be low in number in foals up to 2 weeks of age and have impaired chemotactic function. 101 The importance of complement in foals is illustrated by the nding that the opsonic capacity of foal serum for bacteria is halved by heat inactivation.12 Interestingly, complement activity in the rst week of life is considerably elevated in colostrum-deprived foals, possibly as an alternative defense mechanism.102 In bovine colostrum–fed foals, serum complement levels reach adult levels by 1 to 3 weeks of age.103

Adaptive Cell–mediated Immunity in Foals Recent studies have measured lymphocyte numbers and subpopulations in foals.104-106 Foals are born with B and T lymphocytes and with CD4+ and CD8+ T lymphocyte subsets. Lymphocyte counts rise in the rst 4 months of life, and the proportion of B lymphocytes increases. A comprehensive study of lymphoproliferative responses in foals from the day of birth through 4 months of age found no di erence between foals and adults.106 Another study reported foal lymphoproliferation as low on the day of birth, possibly as a result of high serum cortisol levels.107 Foals do exhibit reduced levels of IFNγ production throughout the prenatal period, which could account for their susceptibility to Rhodococcus equi and other intracellular pathogens.108 Decreased production of this cytokine could be due to dysregulation of its production or re ect the fact that most T lymphocytes in the foal are naïve. Currently, markers for the development of memory lymphocytes are unavailable in horses, although increased expression of MHC II antigen on T lymphocytes throughout the rst year of life may identify a developing population of memory cells.59 Although there is evidence for the capacity of foals to mount immune responses in utero,85-87 there have been few studies of antigen speci c immune responses in the rst days of life, except in the context of the immunosuppressive e ect of passive transfer of immunity.109 When foals are immunized with antigens against which they have no maternally derived speci c antibodies, it is clear that they can mount immune responses from at least 3 months of age, and possibly sooner.110

Antibody-mediated Immunity in Foals Passively Transferred Maternal Antibody During the rst 1 to 2 months of life, foals depend on passively transferred immunity for protection from infectious disease. The di use epitheliochorial nature of the equine placenta does not allow for in utero immunoglobulin transfer to foals. Although minor concentrations of some immunoglobulins can be detected at birth, the foal is born essentially agammaglobulinemic and acquires passive immunity by the ingestion and absorption of colostrum from the dam.111,112 Colostrum is a specialized form of milk containing immunoglobulins, which are produced during the last 2 weeks of gestation under hormonal in uences.113 Colostrum contains primarily IgGa, IgGb (IgGa plus IgGb is the equivalent of IgG), and IgG(T), with smaller quantities of IgA and IgM, all of which have been concentrated into mammary secretions from the mare’s blood.114,115 Colostrum is produced only one time each pregnancy and is replaced by milk that contains negligible immunoglobulins within 24 hours of the initiation of lactation.111,114 This extremely rapid decline in immunoglobulin concentrations in mammary secretions is consistent with equine colostrum production ending at or even before parturition.115 The absorptive capacity of the foal’s gastrointestinal tract for immunoglobulins is greatest during the rst 6 hours after birth, and then steadily declines until immunoglobulins can no longer be absorbed when the foal is 24 hours old. This “closure” of the gut to absorption of large intact molecules is due to replacement of specialized enterocytes by more mature cells.116

De Novo Antibody Production in Foals Few studies of de novo antibody production have been conducted in foals without the e ect of passively transferred maternal antibody. In a study of 10 pony foals fed only bovine colostrum, endogenous equine antibody production measured by radial immunodi usion (RID) resulted in serum concentrations of IgG of 200 mg/dl by 2 weeks of age, 400 mg/dl by 1 month, and 1000 mg/dl by 3 months of age.117 In a smaller study of two colostrum-deprived pony foals, comparing them with 18 colostrum-fed foals, and measuring serum γ-globulin levels by immunoelectrophoresis, very similar results were obtained, although it was apparent that the colostrum-deprived foals achieved higher serum γ-globulin levels between 6 weeks and 3 months of age than did colostrum-fed foals.111 In a third study, antibody concentrations in six colostrum-deprived foals were substantively higher than in ve control foals between 3 and 5 months of age.102 These three studies provide evidence for substantial endogenous production of IgG in the rst month of life in foals deprived of equine colostrum and suggest that the onset of production is earlier, and the rate is higher, in foals deprived of colostrum. This observation is consistent either with nonspeci c immunosuppression in colostrum-fed foals or to stimulation of immunoglobulin production in colostrum-deprived foals. In another study of foals from mixed breed horses fed only bovine colostrum, endogenous IgG production started later and was rst detected at 1 month of age

in the majority of foals, reaching similar levels to foals fed equine colostrum by 2 months of age.118 In colostrum-fed foals, serum IgG concentration falls to its lowest level at 1 to 2 months of age as a result of catabolism of maternally transferred immunoglobulin, subsequently rising toward adult levels as a result of endogenous immunoglobulin production.99,105,106,112 A study by Sheoran et al.114 extended these observations and extensively documented changes in serum IgG subclass concentrations in ve Quarter Horse foals in the rst 9 weeks of life. This study showed that IgG (the equivalent of IgGa plus IgGb) concentrations were lowest at 1 month of age. However, the subsequent increase in IgG concentration was due to de novo IgGa production, not IgGb. At the end of this study at 9 weeks of age, there was still no clear evidence of IgGb production, although IgGa and IgG(T) concentrations had reached or exceeded adult levels. In adult horse serum, IgGb comprises more than 60% of total serum IgG and is by far the dominant subclass in foal serum after passive transfer of immunity.114 IgGb has also been shown to have a critical role in immunity to a variety of pathogens,119,120 and it is possible that the naturally late onset of endogenous production may be a factor in the increased susceptibility of foals to infections such as bacterial respiratory disease at this age.121,122 This possibility was investigated in a study by Grondahl et al.,12 in which the opsonic capacity of foal serum was measured during the rst 42 days of life using the foal pathogens Escherichia coli and Actinobacillus equuli. No di erences were detected over time, and foal serum was as e ective as adult horse serum. Although this study did not provide evidence of decreased opsonization in serum of older foals, the studies were extended only to samples from 42-day-old foals, and in this in vitro system immunoglobulin concentrations may not have been rate limiting. Finally, Holznagel et al.123 studied immunoglobulin concentrations throughout the rst year of life in foals and demonstrated that all foal immunoglobulin concentrations measured showed decreasing serum levels at 4 weeks of age, after which point all but IgGb began to rise. This pattern was the result of catabolism of maternal immunoglobulins and the di erent times and rates of onset of endogenous production of immunoglobulin classes and subclasses; IgA and IgG(T) levels stabilized at 8 to 12 weeks of age. IgGa levels peaked at 8 weeks of age and then slowly declined throughout the duration of the study. IgGb levels reached their nadir at 2 to 5 months of age and did not begin to rise until after 16 weeks of age. This factor could provide a basis for reduced endogenous antibody-mediated immunity in the rst year of life. This study also showed that at 1 year of age, serum immunoglobulin concentrations had still not yet reached adult levels. A factor that signi cantly a ects de novo immune responses in foals is the suppressive e ect of passively transferred antigen-speci c maternal antibodies. The rate of decline of these antibodies varies for both individuals and di erent infectious agents. The half life for maternal IgG in foals is estimated at 20 to 30 days.111 Studies of antigen speci c antibodies demonstrated similar half lives for anti-in uenza virus and antitetanus antibodies of 27 to 29 days for IgGa, 35 to 39 days for IgGb, and 35 days for IgG(T).124 For many important pathogens the concentration of maternal antibodies in foals falls to nonprotective levels by 2 to 3 months of age.125,126 However, the remaining antibody can still render the foal unresponsive to vaccination for weeks or even months to come. In the case of equine in uenza virus127,128 and tetanus toxin, maternal antibodies can persist until 6 months of age and prevent immune responses in foals vaccinated before reaching that age.124 When foals are vaccinated against antigens against which they have no passively transferred antibody, normal antibody responses have been documented from at least 3 months of age.110

Implications for Immunocompetence in Foals The previously presented evidence suggests that the foal’s immune system is competent in many regards, with the innate immune system completely functional at least by the second week of life and with the full complement of lymphocytes present from birth. Antibody is entirely provided by passive transfer at rst, although endogenously produced immunoglobulin is detectable within a few weeks of birth and predominates from 1 to 2 months of age. Nevertheless, there are some key features of the foal immune system that can limit its ability to defend against infection. A critical factor is antigen specific and nonspecific immunosuppression resulting from transferred maternal antibody. As the foal ages, the continuing immunomodulatory e ect of maternal antibody may limit foal immunoresponsiveness while no longer providing comprehensive protection itself. Of similar importance is the fact that although the lymphocytic immune system is complete from the time of birth, it is naïve. Neonates can mount normal immune responses but require appropriate presentation of antigen and costimulatory signals.129 Antigen presentation in the absence of co-stimulatory second signals (e.g., T helper cells) can induce immune deviation or a failure to mount the appropriate immune response, and particularly so in neonates.130-132 The absence of memory responses and a well-developed repertoire of immune responses is a serious handicap that only appropriate antigenic encounters can overcome. Most important, recent studies have documented that maturation of the adaptive immune system of the foals in terms of both the humoral123 and cellular arms108,133 is incomplete throughout the first year of life.

HYPERSENSITIVITY AND AUTOIMMUNITY Hypersensitivity refers to an altered state of immunoreactivity resulting in self-injury. Four di erent types of hypersensitivity can be de ned by the type of immunologic process underlying the tissue injury, as was originally proposed by Gell and Coombs.1 The general features of this classi cation system are presented in Table 1-7. The most common and important type of hypersensitivity disease, at least in humans, is Type I hypersensitivity, mediated by IgE. In these diseases some individuals produce IgE antibodies against a normally innocuous antigen, which is called an allergen.2 Exposure to the allergen triggers mast cell degranulation as described later, and a series of responses result that

are characteristic of allergy. Allergic diseases are so important that more is known about the function of IgE in this hypersensitivity disease than about its normal role in host defense. In this de nition, and throughout this chapter, the term allergy refers only to Type I hypersensitivity diseases mediated by IgE,2 whereas in other de nitions allergy can refer to the entire spectrum of hypersensitivity diseases.3 Other forms of hypersensitivity disease depend on IgG antibodies (Types II and III hypersensitivities) or T cells (Type IV hypersensitivity). Each of these disease processes can play a role in the immunopathogenesis of autoimmune disease, in which the body mounts an adaptive immune responses to self tissue antigens. TABLE 1-7 Four Types of Hypersensitivity∗

Clinical hypersensitivity diseases, such as recurrent airway obstruction (RAO) or purpura hemhorrhagica, can involve more than one type of hypersensitivity reaction simultaneously, which limits the utility of this classi cation for clinical diagnosis. Alternative strategies for classifying these diseases may have greater clinical utility. For example, antibody mediated hypersensitivity diseases (Types I, II, and III) are immediate in onset if preformed antibody exists in circulation or tissues, with some variation in time course dependent on the antibody isotype involved. Cell mediated hypersensitivity (Type IV) is delayed, even in sensitized individuals, for 1 to 3 days, while e ector cells are recruited to the site of antigen exposure.3 The goals of this section are as follows: • Review the classical hypersensitivity types in order to explain the immunopathogenesis of hypersensitivity diseases. • Describe immediate and delayed hypersensitivities of horses and their immunologic basis. • Identify autoimmune conditions of horses with a known immunologic basis. Detailed descriptions of clinical aspects of hypersensitivity and autoimmune diseases and their diagnosis and management are presented elsewhere in this book. Detailed explanations of many immunologic mechanisms involved in these disease processes are provided previously in this chapter.

CLASSICAL TYPES OF HYPERSENSITIVITY REACTION Type I Hypersensitivity As described previously, Type I hypersensitivity, or allergy, is mediated by IgE antibody speci c for allergens, which are extrinsic antigens normally not recognized by the healthy immune system.2 IgE is predominantly found in tissues, where it is bound to mast cells through the high-affinity IgE receptor, which is called FcεRI and has been identi ed in the horse.4 When antigen binds to IgE on the surface of mast cells, cross-linking two or more IgE molecules and their Fc RI receptors, this triggers the release of chemical mediators from the mast cells, which cause Type I hypersensitivity reactions. Basophils and eosinophils (when activated) also possess Fc RI receptors and therefore can participate in the same process. In addition to Fc RI receptors, there is an unrelated low-a nity IgE receptor called CD23, which is present on many lymphocytes, monocytes, eosinophils, platelets, and follicular dendritic cells. The role of CD23 appears to be to enhance IgE responses to speci c antigens when those antigens are complexed with IgE. Thus CD23 on antigen presenting cells can capture IgE bound antigens. In the horse CD23 has been identified, and its expression is upregulated by equine IL-4.5 The selective stimulation of IgE responses depends on characteristics of the antigen (allergen), the individual a ected (genetic factors such

as MHC antigens), and the mechanism of antigen presentation. The antigen must be capable of eliciting a Th2 immune response in order to stimulate IgE production. Small, soluble proteins, frequently enzymes, containing peptides suitable for MHC II antigen presentation and presented to mucosal surfaces at low doses, are particularly e cient at generating IgE responses. Low doses of antigen speci cally favor Th2 over Th1 responses, and exploiting this relationship is the basis of some therapeutic hyposensitization strategies (see under Immunomodulators). These processes are thought to be regulated by regulatory T cells (Tregs). Natural Tregs are CD4/CD25 positive, and in healthy individuals they suppress Th2 cytokine production.2 When Tregs are de cient, atopy can result. When CD4 T helper cells are exposed to IL-4, as opposed to IL-12, during antigen presentation by dendritic cells, they are driven toward becoming Th2 cells. This process is critical to promoting IgE responses and may be favored at enteric and respiratory mucosal surfaces, or skin, where parasite invasion typically occurs. This makes teleologic sense insofar as IgE responses are very important for antiparasitic immunity.6 The dendritic cells at such locations are frequently programmed to stimulate Th2 responses.2 Cross-linking of Fc RI receptors on granulocytes also results in CD40L expression and IL-4 secretion, which further promotes IgE production by B lymphocytes and sustains allergic reactions. Some individuals maintain IgE responses to a wide variety of allergens, a condition called atopy. A ected individuals have high levels of IgE in the blood and increased eosinophil populations. In humans this condition depends partly on genetic factors, including genetic variations in the IL-4 promoter sequence or association with particular MHC II genes. Nevertheless, environmental factors are also important, as atopy is increasingly common in humans in economically developed parts of the world. Four possible explanations for this are decreased exposure to infectious disease during childhood, environmental pollution, allergen levels, and dietary change. The rst explanation is currently favored, and its basis is the proposal that many infectious diseases bias the immune system toward Th1 responses2 and that their decreased prevalence results in an increased tendency to mount Th2 responses, which may be the natural bias of the neonatal immune system.7

EFFECTOR MECHANISMS IN TYPE I HYPERSENSITIVITY ALLERGIC REACTIONS When triggered by antigen cross-linking of IgE bound to Fc RI cell surface receptors, activated mast cells release chemical mediators stored in preformed granules and synthesize leukotrienes and cytokines. In Type I hypersensitivity reactions the outcome of this reaction can vary from anaphylactic shock to minor localized in ammation. Mast-cell degranulation causes an immediate allergic reaction within seconds, but there is also a sustained late-phase response that develops up to 8 to 12 hours later as a result of recruitment of Th2 lymphocytes, eosinophils, and basophils. Mast cells are highly specialized cells of the myeloid lineage that are common in mucosal and epithelial tissues near small blood vessels. The range of in ammatory mediators released by degranulating mast cells is wide and includes enzymes that can remodel connective tissues; toxic mediators such as histamine and heparin; cytokines such as IL-4, -5, -13, and TNF-α; and lipid mediators, including leukotrienes and PAF.2 Histamine causes an increase in local blood ow and permeability. Enzymes activate matrix metalloproteinases that cause tissue destruction, and TNF-α increases expression of adhesion molecules and attracts in ammatory leukocytes. These reactions are all appropriate when the mast cell is reacting to an invasive pathogen, but in allergy this is the basis of the immediate in ammatory response and also the initiating step in the late-phase response. The role of eosinophils in in ammation is tightly controlled at several levels. Synthesis in the bone marrow depends on IL-5 produced by Th2 cells in the face of infection or other immune stimulation. Transit of the eosinophils to tissues depends on two chemokines, eotaxin 1 and eotaxin 2. Activation of eosinophils by cytokines and chemokines induces them to express Fc RI and complement receptors and primes the eosinophil to degranulate if it encounters antigen that can cross-link IgE on its surface. Mast cell degranulation and Th2 activation recruit and activate large numbers of eosinophils at the site of antigen encounter. Basophils are similarly recruited, and together their presence is characteristic of chronic allergic in ammation. Eosinophils can trigger mast cells and basophil degranulation by release of major basic protein. This late-phase response is an important cause of long-term illnesses such as chronic asthma in humans.

CLINICAL MANIFESTATIONS OF TYPE I HYPERSENSITIVITY REACTIONS DEPEND ON THEIR SITEs The clinical outcome of Type I hypersensitivity reactions depends on the amount of IgE present, doses of allergen, and the sites of allergen introduction. Direct introduction of allergen into the bloodstream or rapid enteric absorption can lead to widespread activation of connective tissue mast cells associated with blood vessels. This potentially disastrous event is called systemic anaphylaxis and can cause catastrophic loss of blood pressure and airway obstruction owing to bronchoconstriction and laryngeal swelling. This leads to anaphylactic shock and can follow administration of drugs against which an individual has an established IgE response. Treatment with epinephrine can control these potentially fatal events. Penicillin is one example of a drug that can cause Type I hypersensitivity reactions in humans, although it is less certain that it can induce this type of hypersensitivity reaction in the horse. Penicillin can act as a hapten (see the section on equine immunology). Penicillin alone can elicit antibody formation by B cells but cannot elicit T helper cell responses because it is not a protein. However, the β-lactam ring of penicillin can react with amino groups on host proteins to form covalent conjugates, and the modi ed self-peptides can generate Th2 responses in some individuals. The Th2 cells can in turn release cytokines, which activate penicillin-binding B cells to produce IgE. In this

scenario penicillin is a B cell antigen and becomes a T cell antigen by modifying self-peptides. Intravenous penicillin results in protein modification and recognition and cross-linking of mast cell IgE, leading to anaphylaxis.2 Allergen inhalation, in contrast, induces local in ammation of the respiratory tract—for example, in the upper airways, as in allergic rhinitis, or in the lower airways, as in human asthma. Similarly, allergen introduction into the skin causes local histamine release initially and a wheal-and- are reaction, followed by a late-phase response several hours later. When allergens are ingested and reach the skin from the bloodstream, a disseminated form of the wheal-and- are reaction occurs that is called urticaria or hives. Prolonged in ammation of the skin results in eczema or atopic dermatitis in some individuals. Ingestion of allergens causes activation of gastrointestinal mast cells, resulting in uid loss across the bowel and smooth muscle contraction. The clinical presentation is diarrhea and vomiting. Sometimes ingestion of allergens can lead to systemic anaphylaxis, if they are absorbed rapidly, or urticaria, as is sometimes seen after oral penicillin administration.

Type II Hypersensitivity Type II hypersensitivity disease occurs when the causal antigen is associated with cells or tissue components of the body and there is an IgG antibody response to this antigen. Phagocytes, or other cells expressing Fcγ receptors, mediate destruction of the a ected tissue or removal from the circulation by the reticulo-endothelial system in the case of antibody-positive erythrocytes or platelets. Antibody-mediated hemolytic anemia and thrombocytopenia are examples of drug-associated Type II hypersensitivities. In the case of the horse, penicillin is an established cause of hemolytic anemia.8 Diagnosis can be accomplished using a Coombs’ test (Figure 1-18). Penicillin binds to the erythrocyte surface and is targeted by antipenicillin antibodies of the IgG isotype. Interestingly, quite large numbers of horses have antipenicillin antibodies of the IgM isotype, but this does not lead to disease.

FIGURE 1-18

Direct Coombs’ test. Ab, Antibody; RBCs, red blood cells.

Type III Hypersensitivity In Type III hypersensitivity the antigen is soluble and present in the circulation. Disease results from formation of antibody-antigen aggregates or immune complexes under certain speci c conditions.2 Although immune complexes are generated in all antibody responses, they are generally harmless. Large complexes x complement and are removed from circulation by the reticulo-endothelial system. However, small complexes can form at antigen excess (Figure 1-19), and these can deposit in blood vessel walls and tissues, where they ligate Fc receptors on leukocytes, causing an in ammatory response, increased vascular permeability, and tissue injury. Complement activation also contributes to this process. Local injection of antigen can sometimes lead to a necrotizing skin lesion caused by Type III hypersensitivity, and this is called an Arthus reaction.

FIGURE 1-19

Antibody-antigen precipitation. Antibody can precipitate soluble antigen in the form of immune complexes. This is most e cient when concentrations of antibody and antigen reach equivalence, and large immune complexes are formed. However, when antigen is in excess some immune complexes are too small to precipitate and can produce pathologic changes such as are seen in Type III hypersensitivities.

The classical example of a Type III hypersensitivity reaction is serum sickness, which is seen after administration of horse antiserum in humans (e.g., in treating snake bites). After an IgG response to the horse, serum is generated (7-10 days), and signs of fever, urticaria, arthritis, and sometimes glomerulonephritis result. The foreign antigen is cleared as part of this process, which makes this condition ultimately selflimiting. Alternative scenarios for induction of Type III hypersensitivity reactions include persistent infectious diseases in which pathogens are not completely cleared from tissues or autoimmune diseases in which antigen persists. Inhaled antigens that induce IgG responses can lead to immune complex formation in the alveolar wall, as occurs in farmer’s lung, compromising lung function.2 Any such circ*mstance in which immune complexes are deposited in tissues can lead to this type of pathology.

Type IV Hypersensitivity Cell-mediated Type IV hypersensitivities cause delayed hypersensitivity reactions. A variety of cutaneous hypersensitivity reactions are seen, such as the contact hypersensitivity seen after absorption of haptens such as pentadecacatechol in poison ivy or the local Th1 response seen in the diagnostic tuberculin reaction. When Type IV hypersensitivity results in a Th2 response, the principle outcome is eosinophil activation and recruitment such as in chronic asthma.

IMMEDIATE HYPERSENSITIVITY DISEASES IN THE HORSE Previously, a major limitation of our ability to study hypersensitivity disease in the horse has been the lack of reagents capable of detecting equine IgE. Although equine IgE has long been known to exist,9,10 and the genetic sequence has been known since 1995,11-13 the only reagents for studying it have historically been conventional polyclonal antisera produced by vaccination with physico-chemically puri ed IgE14,15 or made in chickens after vaccination with recombinant fragments of the IgE heavy chain.16 Although many valuable studies have been performed using these reagents,17-20 the recent availability of well-characterized monoclonal antibodies recognizing equine IgE holds much promise for future studies.21,22 It has already been demonstrated that the onset of IgE production in horses does not occur before 9 to 11 months of age and does not reach adult levels before 18 months.23 This may help explain why hypersensitivity disease is uncommon in horses before puberty. The following section describes a series of equine diseases with characteristics of immediate hypersensitivity disease. This is not an exhaustive list of equine hypersensitivity diseases, and additional examples will be found throughout this book.

Systemic Anaphylaxis The incidence of true systemic anaphylaxis in horses is unknown, although the condition has been reported in association with administration of a wide range of compounds, including serum, vaccines, vitamin E–selenium preparations, thiamine, iron dextrans, and antibiotics such as penicillin.24,25 Target organs in experimental equine anaphylaxis are the lung and the intestine.24 Sudden dyspnea; hypotension, as evidenced by poor peripheral pulse character; rapid onset of urticaria; and collapse are cardinal signs of the onset of systemic anaphylaxis. The therapeutic goals in treating systemic anaphylaxis are to (1) prevent or reverse the complications caused by mediator release, (2) maintain respiratory integrity, and (3) maintain cardiovascular stability. Not all anaphylactic reactions require therapy. However, rapid recognition of those that do is critical to patient survival. Intravenous access via an indwelling intravenous catheter and airway patency should be established immediately because cardiovascular collapse and upper airway obstruction caused by angioedema can occur rapidly. Because the conscious horse does not tolerate tracheal intubation, emergency tracheotomy may be required. Oxygen should be administered if available because bronchoconstriction and cardiovascular collapse result in hypoxemia. The uid requirement of horses in anaphylactic shock is not known, but large volumes of balanced polyionic fluid should be administered rapidly. The principal therapeutic agent is epinephrine, which is a potent sympathetic stimulant. Epinephrine administration may cause excitement in the horse. Epinephrine should be administered intramuscularly (10 to 20 g/kg, equivalent to 5 to 10 ml of 1:1000 dilution of epinephrine for a 450-kg horse) if dyspnea or hypotension are mild. Epinephrine should not be administered subcutaneously because its potent vasoconstriction can lead to poor absorption and tissue necrosis. If dyspnea or hypotension is severe, epinephrine should be administered intravenously or endotracheally if there is no venous access (3 to 5 g/kg or 1.5 to 2.25 ml of 1:1000 dilution of epinephrine for a 450-kg horse). Epinephrine doses can be repeated every 15 to 20 minutes until hypotension improves. The side e ects of epinephrine therapy are tachyarrhythmias and myocardial ischemia, which can be life threatening. Alternatively, an epinephrine or norepinephrine drip can be instituted for cases of refractory hypotension. Other therapeutic agents, such as antihistamines, β-agonists, or other pressors, may be indicated, although their value is less certain. Though its e ects may be delayed, glucocorticoid therapy is indicated to help reverse persistent bronchospasm and angioedema and break the cycle of mediator-induced in ammation triggered during hypersensitivity reactions. Ideally, a rapid acting glucocorticoid should be used, such as prednisolone sodium succinate 0.25 to 10.0 mg/kg, administered intravenously. Glucocorticoid therapy during the acute phase will aid in preventing the late-phase reaction.

Insect Bite Hypersensitivity Horses commonly su er from hypersensitivity to salivary antigens of Culicoides and Simulium species leading to an intensely pruritic skin disease with characteristics of both immediate and delayed-type hypersensitivity.26 The clinical sign of urticaria, combined with the presence of increased numbers of IgE positive cells in the skin and high levels of Culicoides-speci c IgE in serum are all evidence of immediate (Type I) hypersensitivity in the immunopathogenesis of this disease.17,18 This pathogenesis has received further support from studies employing the newly available monoclonal antibodies to equine IgE.22 Interestingly, one recent study demonstrated that in addition to IgE, the IgG(T) subclass can also bind skin mast cells and elicit clinical signs consistent with hypersensitivity.27 It remains to be seen if IgG(T) plays a role in the pathogenesis of naturally occuring insect bite hypersensitivity. In some breeds a genetic predisposition based on an MHC-linkage has been demonstrated.28,29

Recurrent Airway Obstruction

Recurrent airway obstruction is a severe in ammatory disease of middle-aged and older horses induced by exposure of susceptible horses to inhaled organic dust, generally from hay, although a summer pasture-associated form is also observed in the southern United States.30 Hay dust contains a mixture of mold spores, forage mites, particulates, and endotoxins, which can induce and exacerbate airway in ammation. Removal of the hay dust by returning the horse to pasture leads to decreased in ammation within a few days. In RAO-susceptible horses, exposure to hay dust leads to invasion of the lungs and airways by neutrophils within 4 to 6 hours and concurrent airway obstruction resulting from bronchospasm, in ammation, and increased mucus viscosity, which principally a ect the bronchioles. RAO-a ected horses develop nonspeci c airway hyper-responsiveness, which is a bronchospasm in response to a wide variety of stimuli, including in ammatory mediators and neurotransmitters. Horses a ected by RAO demonstrate increased histologic lesions and worsening airway function with increasing age. In addition, significant histopathologic changes are present before abnormal airway function can be detected. The immunologic basis of RAO remains poorly elucidated. Two pieces of evidence suggest a role for Type I hypersensitivity in this disease. First, IgE levels are increased in bronchoalveolar (BAL) uid of RAO-a ected horses,14 and second, allergen-speci c IgE is increased in a ected horses.20,31,32 However, the immediate onset of airway obstruction typical of a Type 1 reaction to exposure to allergens is rarely observed because clinical signs of RAO only develop several hours after antigenic exposure.30 A study of immunoregulatory cytokines in RAO demonstrated evidence for a Th-2 bias in RAO, with increased levels of IL-4 and IL-5 and decreased IFN-γ mRNA in BAL cells.33 However, other investigators have documented increased IFN-γ levels in RAO.34 It seems likely that a number of immunologic processes, including IgE mediated pathology, increased expression of Toll-like receptor 4,35 and dysregulation of cytokine responses,36 may be involved in this disease.

IgG-mediated Diseases IgG-mediated diseases, which broadly correspond to Type II and III hypersensitivities, have also been termed immune-complex diseases in the horse.3 The examples described here are distinguished from the other immediate hypersensitivities of the horse described previously in that there is no evidence for the involvement of IgE in their pathogenesis.

NEONATAL ISOERYTHROLYSIS AND ALLOIMMUNE THROMBOCYTOPENIA Neonatal isoerythrolyis is a common condition of foals and is extensively reviewed elsewhere in this text. The condition results from the passive transfer of maternal antibodies in colostrum that recognize allogenic foal erythrocyte antigens principally of the Aa and Qa haplotype inherited from the sire. A similar condition occurs in mules as a result of the inheritance of a donkey-speci c erythrocyte antigen.37,38 A severe, potentially life-threatening anemia results as the antibody-positive erythrocytes are removed by the reticulo-endothelial system or, less commonly, lysed by complement. A similar condition less commonly a ects platelets, causing severe neonatal thrombocytopenia in horses39 and mules.40 These conditions are typical of Type II hypersensitivities and are mediated by circulating IgG recognizing cell surface antigens on erythrocytes. Diagnosis can be performed using a variation of the Coombs’ test (Figure 1-20).

FIGURE 1-20

Neonatal isoerythrolysis test. Ab, Antibody; RBCs, red blood cells.

PURPURA HEMORRHAGICA Purpura hemorrhagica is an acute disease of the horse characterized by edema of the head and limbs; leucocytoclastic vasculitis; petechial

hemorrhages in mucosae, musculature, and viscera; and sometimes glomerulonephritis.41 It is usually associated with Streptococcus equi subsp. equi infection of the upper respiratory tract disease. Serum of a ected horses contains immune complexes of S. equi subsp. equi–speci c antigens with equine IgA.41 The glomerulonephritis sometimes seen in association with purpura has been attributed to deposition of similar immune complexes containing streptococcal antigens and IgG.42

DELAYED HYPERSENSITIVITY DISEASES IN THE HORSE Documented immunologic characterization of delayed hypersensitivity conditions of the horse are lacking, although contact hypersensitivities have been reported in horses.3 One very well-characterized example of this type of condition is recurrent uveitis.

Equine Recurrent Uveitis Equine recurrent uveitis (ERU), also known as moon blindness or periodic ophthalmia, is the most important cause of blindness in horses.43 The disease results in both acute and chronic ocular in ammatory disease, and chronic sequelae include development of posterior and anterior synechiae, cataracts, lens opacities, secondary glaucoma, and blindness. Eyes of a ected horses contain IgG antibodies and autoreactive T cells speci c for retinal antigens.44 A speci c cause has not been identi ed. However, sensitization to a variety of pathogens, and in particular to Leptospira spp.,45,46 is thought to induce the immune mediated pathology that is central to the disease.47 Treatment with corticosteroids and other anti-in ammatory agents is essential to prevent visual debility or blindness. However, treatment failures are common, and the disease frequently recurs with further ocular damage months after the initial event, commonly leading to euthanasia.43 Our understanding of the immunologic basis of ERU has been extended by studies of the immunoregulatory events in the eyes of a ected horses. It has been shown that the T lymphocytes that invade the iris-ciliary body during this disease produce a pattern of IFN-γ cytokine production typical of a T helper 1 (Th1) response.48 These studies indicate that ERU is an equine example of a Type IV hypersensitivity disease mediated by Th1 cells.

AUTOIMMUNITY Although a number of equine diseases are considered to be autoimmune in etiology, few have been extensively studied.49 Much of the explanation given in the previous sections for the immunopathologies involved in hypersensitivity disease can be applied to autoimmune disease. Well-described equine autoimmune diseases include neonatal isoerythrolysis and alloimmune thrombocytopenia, which are described previously and have characteristics of Type II hypersensitivities, as does immune-mediated anemia in adults.50 Less well-described entities include systemic lupus erythematosus51 and pemphigous foliaceus,52 which have characteristics of both Type II and III hypersensitivity diseases. As described previously, equine recurrent uveitis appears to represent a Type IV hypersensitivity, and there is some morphologic and immunologic evidence for similarly classifying cauda equine syndrome (polyneuritis equi).53-55 With the exception of a few conditions, such as neonatal isoerythrolysis and penicillin-associated hemolytic anemia, there are few autoimmune conditions in which the cause is well understood. One exception, however, is the anemia that can develop subsequent to administration of human recombinant erythropoietin to horses.56,57 There is substantial evidence that horses mount an antibody response to the exogenous erythropoietin that cross-react with the endogenous hormone and result in erythroid hypoplasia. The lesson of this example may be that in the modern world, with increasing availability of recombinant drugs that mimic natural biologic compounds, we would do well to remember that the immune system has an exquisite ability to distinguish what is foreign and to reject it vigorously.

IMMUNODEFICIENCY Immunode ciencies occur in both primary and secondary forms and have been extensively reviewed.1,2 Primary immunode ciencies have a genetic basis, whereas secondary immunode ciencies result from failure of passive transfer in foals, immunosuppressive infections or drug treatments, neoplasia, or malnutrition. Immunode ciencies can a ect speci c components of the immune system, such as the lymphoid or phagocytic system. Typically, immunodeficiency is suspected in any of the following circ*mstances:3 • Onset of infections in the first 6 weeks of life • Repeated infections that are poorly responsive to therapy • Infections caused by commensal organisms or organisms of low pathogenicity • Disease resulting from the use of attenuated live vaccines • Failure to respond to vaccination • Marked neutropenia or lymphopenia that persists for several days Equine immunode ciency is most commonly suspected on the basis of the rst three reasons—that is, because of increased susceptibility

to infection. The most common immunode ciency recognized in clinical practice is failure of passive transfer in foals.4-6 Other causes of immunode ciency vary from well-de ned clinical entities, such as severe combined immunode ciency of Arabian foals,7 to cases in which immunode ciency is suspected on clinical grounds, but the speci c cause or nature of the problem is di cult or impossible to de ne.8 Regardless of their cause, immunode ciencies result in increased susceptibility to infections, which are in turn poorly responsive to appropriate therapy. Defects in antibody production tend to predispose to pyogenic infection, whereas de ciencies in cell-mediated responses lead to infections with organisms normally nonpathogenic in horses such as Candida albicans, Cryptosporidium spp., or adenovirus. When any immunode ciency is suspected, speci c diagnostic tests are indicated to de ne the de ciency. The aim of the next section is to identify tests that clinicians can apply practically in such cases and to explain their merits and limitations.

TESTS OF EQUINE IMMUNE FUNCTION Tests of components of the immune system (e.g., lymphocytes, immunoglobulins) generally can either quantitate that component or measure its functional capacity. In Table 1-8 the components of the immune system that currently can be analyzed in this manner are identi ed, and corresponding quantitative and functional tests are listed. The table also identifies those tests that are likely to be commercially available, and it should be noted that few of the functional tests are available unless the clinician is able to identify a sympathetic and capable equine immunologic research laboratory. Despite these limitations, the available tests do permit the identification of many of the well-defined causes of immunodeficiency in horses. TABLE 1-8 Components of the Immune System and Tests for Quantitative or Functional Analysis∗

Components of the immune system that can be evaluated in horses and appropriate tests for quantitative or functional analyses of each component. The list of tests is not exhaustive but restricted to tests of likely practical value for which normal data are available. Tests listed in bold are routinely available to clinicians. Component Quantitative Tests Functional Tests Immunoglobulin Radial immunodiffusion, membrane-ELISA, electrophoresis, Response to vaccination † precipitation tests Lymphocytes Complete blood cell count, DNA-PKcs genetic evaluation,‡ Response to vaccination, intradermal PHA test, in FACS analysis of lymphocyte subsets using monoclonal vitro lymphoproliferation assays antibodies Neutrophils and Complete blood cell count Chemiluminesence and bactericidal assays, flow macrophages cytometric evaluation of phagocytosis and oxidative burst Eosinophils and Complete blood cell count No commonly available tests basophils Complement No commonly available tests No commonly available tests Acute phase Electrophoresis No commonly available tests proteins ELISA, Enzyme-linked immunosorbent assay; DNA-PK, DNA-protein kinase catytic subunit; FACS, florescence-activated cell sorter; PHA, phyohemagglutinin. ∗ Components of the immune system that can be evaluated in horses and appropriate tests for quantitative or functional analyses of each component. The list of tests is not exhaustive but restricted to tests of likely practical value for which normal data are available. Tests listed in bold type are routinely available to clinicians. † Zinc sulfate turbidity and glutaraldehyde coagulation. ‡ See SCID for description of DNA protein kinase catalytic subunit genetic testing.

Tests of Antibody-mediated Immunity Some assays of B lymphocyte function and number are described later, but the principal tests of humoral immunity are quantitative assays of immunoglobulin concentration and measurements of speci c antibody responses to vaccination. The variety of classes of immunoglobulins in the horse is complex and reviewed earlier in this chapter.9 For practical purposes our attention is generally focused on IgG (representing the combination of two subclasses: IgGa and IgGb), IgG(T), IgA, and IgM. The current gold standard for measurement of concentrations of immunoglobulin classes is the RID assay. The disadvantage of this test is its cost and the time required to perform the assay (24 hours or more), which makes it generally unsuitable for screening for failure of passive transfer of immunity in foals. Nevertheless, this form of test remains the single most valuable assay available to the clinician trying to

measure total antibody concentrations in the horse. Currently, test kits are available for IgG, IgG(T), IgA, and IgM (VMRD Inc., Pullman, Wash.), although speci c IgG subclass RID tests are available for research use only that extend this range to include all the well-characterized IgG subclasses (i.e., IgGa, IgGb, IgGc, and IgG(T); Bethyl Laboratories, Montgomery, Tex.). The RID test is based on the ability of antigen and antibody to precipitate at equivalence when combined in proportion in agar gel plates. The serum being tested is added to punched-out wells in agar, impregnated with antibody to the speci c immunoglobulin class being measured, and allowed to di use outward and bind with the anticlass-speci c antisera. A precipitate forms when equivalence is reached and the area within the precipitate ring is directly proportional to the concentration of the patient’s immunoglobulin class. Normal ranges of serum immunoglobulin concentrations are typically provided with commercial kits, and normal serum, milk, and colostrum concentrations of equine immunoglobulins have been described in numerous published studies. These results have been summarized and are available in tabular form in two sources.3,10 More recent studies of foal and adult horse serum IgG and IgM concentrations using currently available RID assays measured considerably higher normal values in some instances.11,12 In addition, an extensive study of immunoglobulin concentrations in adult and foal serum and nasal secretions and in colostrum and milk using an experimental monoclonal antibody based–system has been reported.13 By far the most common question that a clinician seeks to answer with regard to a horse’s immune status is whether a foal has achieved adequate passive transfer of immunity. Whatever test is chosen must be able to distinguish serum concentrations of IgG of 800 mg/dl in order to permit diagnosis of total or partial failure of passive transfer. The test should be rapid, in order to allow early initiation of therapy, which decreases the utility of the RID test. A wide variety of tests have been used for this purpose: zinc sulfate turbidity, latex bead agglutination tests, ELISA assay, turbidometric analysis, glutaraldehyde coagulation, or infrared spectroscopy.14-17 Addition of serum to zinc sulfate solution causes precipitation of immunoglobulins, principally IgG. Although the degree of resultant turbidity is usually proportional to the IgG concentration, turbidity may be increased by hemolysis in the sample, poor operating conditions, and poor-quality reagents. In the glutaraldehyde coagulation test, glutaraldehyde forms insoluble complexes with basic proteins in the serum.18 Gel formation in 10 minutes or less is equated with a serum IgG concentration of 800 mg/dl or greater, whereas a positive reaction in 60 minutes is indicative of at least 400 mg IgG/dl serum. Like the zinc sulfate turbidity test, hemolysis may falsely overestimate the IgG concentration. In the latex agglutination test (Foalcheck, Haver Mobay Corp., Shawnee, Kan.), the patient’s serum is mixed with the anti-equine IgG absorbed to latex particles. Macroscopic agglutination is proportional to serum IgG. Currently, for rapid diagnosis, the most convenient test system may be membrane lter–based ELISA systems (e.g., SNAP, Idexx, Westbrook, Me.; Figure 1-21). This test can be performed “foal-side” with whole blood. Tests such as the glutaraldehyde coagulation test are simpler and cheaper, although they have the disadvantage that serum is required. While some data suggest that the glutaraldehyde coagulation test may be more sensitive than membranelter ELISAs in detecting failure of passive tranfer,15,22 particularly in di erentiating normal foals (>800 mg/dl IgG) from partial failure of passive transfer (400 to 800 mg/dl), speci city can be relatively poor. The latter problem a ects many of the rapid diagnostic tests for failure of passive transfer, and a more extensive discussion of test selection for this condition is presented later in the section covering this disease.

FIGURE 1-21

Membrane-based ELISA system (SNAP; Idexx, Westbrook, Me.) for measuring serum IgG concentration. The diluted test equine serum sample is applied to a

“patient spot” on a membrane impregnated with a capture antibody recognizing equine IgG. Calibration spots corresponding to speci c concentrations of equine IgG (400 and 800 mg/dl) are adjacent to the patient spot. An enzyme-conjugated second antibody against equine IgG is applied to the entire membrane, and nally the device is triggered to release an enzyme substrate that produces a colored reaction corresponding to the amount of enzyme-conjugated antibody on the membrane. By comparison with the calibration spots, the test sample IgG concentration can be estimated.

Alternative available tests that give information about serum immunoglobulin content include electrophoresis and immunoelectrophoresis (IEP). IEP analysis can demonstrate the presence of all the currently recognized equine immunoglobulin classes. However, the test has a relatively poor sensitivity and gives no quantitative information, as might be obtained from rocket electrophoresis.23 Serum electrophoresis gives quantitative information about albumin and α, β, and γ globulin concentrations (Figure 1-22), and its utility is demonstrated when detecting the monoclonal gammopathies that accompany plasma cell myelomas.24 Nevertheless, in the diagnosis of immunodeficiencies electrophoresis should be seen as an adjunct to RID assays, which are superior in terms of specificity and sensitivity.

FIGURE 1-22

Serum protein electrophoresis. The complex mixture of serum proteins are separated by migration through19-21 an agarose gel slab in response to an electric eld. Proteins are stained, and the intensity of staining of di erent bands is measured by densitometric scanning. These measurements are used to identify di erent types of globulins and albumin corresponding to stained bands.

Tests of Cellular Immunity The simplest test of the cellular arm of the immune response is a total and di erential white blood cell count, and this should be the starting point for any evaluation. Identi cation of an absolute lymphopenia, for example, is a critical nding in a suspected case of severe combined immunode ciency (SCID) in an Arabian foal, although the result must be repeatable in a series of tests given the variability of blood lymphocyte counts. Such a nding would logically lead to genetic testing to con rm the diagnosis.7 The evaluation of a lymph node biopsy for normal architecture, including the presence or absence of cells in either the B lymphocyte– or T lymphocyte–dependent areas, is another powerful test of the immune system. However, in profound immunode ciencies such as SCID, lymphoid organs may be impossible to locate ante mortem. Beyond these readily available conventional techniques, three other more complex types of analysis can be of value: ow cytometric analysis (primarily of lymphocytes, although other cell types can be analyzed), lymphocyte function testing, and functional analysis of phagocytic cells.

FLOW CYTOMETRY It is currently feasible to take the equine di erential white blood cell count a step further, because monoclonal antibodies are now available that can di erentiate the morphologically identical equine lymphocyte family into distinct subsets with speci c functions (Table 1-5).25 Flow cytometry allows rapid measurements to be made of individual cells in a uid stream. Flow cytometers utilize lasers to measure multiple parameters, including light scatter and uorescence characteristics of cells, and are complex instruments to construct, but the principles of their operation are relatively simple (Figure 1-23). The uidics system of the ow cytometer delivers cells one by one to a point in space intersected by a laser beam. The laser beam emits light of a de ned wavelength to illuminate the cell, which results in both scattered light of

the same wavelength and fluorescent light of a different wavelength that is collected by photodetectors and converted into electronic signals.

FIGURE 1-23

Flow cytometric analysis (see text for key); FITC, Fluoroscein isothiocyanate; RPE, R-phycoerythrin.

Light is collected by the forward collection lens on the other side of the ow chamber from the laser source. Here light scattered from 1 to 20 degrees from the laser-beam axis is collected as “forward scatter,” and its amount depends on the size of the cell being analyzed. At 90 degrees (orthogonal) to the laser beam, path light is collected for the purpose of measuring “side scatter” and uorescent emission. Optical ltration separates scattered light and uorescent light to permit their independent measurement. Side scatter light depends on the granularity of cells. Fluorescent light can be detected independently for a number of uorochromes of di erent wavelengths, and typical examples would be fluorescein (FITC) or phycoerythrin (PE). Signals from the di erent detectors can be processed directly or after logarithmic ampli cation. The advantage of log ampli cation for the uorescent signals is that it ampli es weak signals and compresses large signals, allowing their simultaneous display. By these means signals with a 10,000-fold di erence in intensity can be displayed. Signals are typically displayed as either histograms or dual parameter correlated plots (dot plots), and statistical analysis is completed by computer. Histograms are analyzed by setting markers in particular channels and dot plots by drawing rectangular or polygonal boxes around data points. The software also allows the setting of “gates” for determining which events are collected in the rst instance or which events are to be included in later analyses. Typically, these gating techniques are employed using forward and side scatter to di erentiate cell types, such as lymphocytes, monocytes, and granulocytes. The nal key characteristic of ow cytometers is their capacity to analyze large numbers of “events” (cells) in a short time, making it possible to analyze many thousands of cells in a matter of seconds. An example of such an analysis is shown in Figure 1-23. In this instance the goal was to identify lymphocytes expressing the equine hom*ologs of the CD4 and CD8 molecules using two monoclonal antibodies independently labeled with the FITC (CD4) and PE (CD8) uorochromes. For this analysis a peripheral blood leukocyte population was prepared by simply lysing the erythrocytes in a blood sample using distilled water. Subsequently, the whole leukocyte population was stained in solution using the monoclonal antibodies. Alternatively, lymphocytes could have been puri ed from blood using di erential centrifugation techniques before staining. However, identical analytical results are obtained in the horse using the more rapid technique of whole leukocyte preparation and exploiting the capacity of the ow cytometer to distinguish lymphocytes from other cells.26 During the ow cytometric analysis the rst step was to identify the lymphocytes using the di erent forward and side scattering characteristics of di erent leukocyte populations. Panel A in Figure 1-23 shows a dot-plot of forward scatter versus side scatter, with each dot representing a cell. Granulocytes (G) can be distinguished from lymphocytes (L) by their greater granularity (side scatter) and size (forward scatter). The dotted line is a “gate” drawn around the lymphocyte population. Subsequent analysis of uorescence is directed toward only those cells that fall within this gate. After establishing the physical characteristics of the cells to be analyzed, their uorescence can be examined. In Panel B of Figure 1-23, a histogram is shown depicting CD4 (FITC) staining. A vertical marker identi es a gate set based on staining by a negative control antibody. Therefore all cells to the right of this marker are staining

positively for CD4. For two-color staining dot plots are again used. In Panel C each dot represents a cell, and its position relative to the two axes illustrates its staining characteristics; the dotted lines represent the cut-o points for negative or positive staining. Therefore all cells in quadrants 1 and 2 are positively stained for CD4, and all cells in quadrants 2 and 4 are positively stained for CD8. E ectively, quadrant 1 contains all the helper T lymphocytes (CD4 positive), quadrant 4 contains all the cytotoxic T lymphocytes (CD8 positive), quadrant 3 contains the B lymphocyte population, and quadrant 2 is empty because there are no “double-positive” T lymphocytes in the blood of horses. The sum of the cells in quadrants 1 and 4 represents the T lymphocyte population. This technique has revolutionized immunobiologic studies in recent years, nding its most obvious clinical application in enumerating human CD4-positive lymphocytes in cases of acquired immune de ciency27 and in classifying leukemias and lymphomas.28,29 A large number of antibodies are now available for use in the horse25 (see Table 1-5), and recent reports describe the use of this approach in charting the response to microbial infection,11,30,31 and in di erentiating equine leukemia.29,32,33 Several recent studies provide examples of normal values for peripheral blood analysis.11,12,34,35

LYMPHOCYTE FUNCTION TESTING Unfortunately, tests of lymphocyte function are generally limited in their availability in the field. In vitro tests of lymphocyte function include lymphocyte proliferation responses to mitogens such as pokeweed (B cell dependent), phytohemagglutinin (T cell dependent), or concanavilin A (T and B cell dependent). These assays are generally not commercially available, although they are commonly performed by immunologic researchers. In addition, some caution should be used because it may not always be valid to draw inferences about the intact animal from the results of these in vitro tests,3 and because of signi cant variability in results, it is essential to perform parallel studies on suitable age-matched control horses. The end point of these tests is usually read by determining the incorporation of radioactive 3Hthymidine into the total population of proliferating cells. Nonradioactive alternatives exist, and one strategy uses intracellular labeling of lymphocytes with 5-carboxy uorescein diacetate-succinimidyl ester (CFSE). Labeled cells uoresce, and after subsequent divisions in response to mitogens, this uorescence decreases by half for each cycle of cell division. This allows for measurement of equine lymphocyte proliferation using ow cytometry, and simultaneous two-color staining allows for measurement of proliferation in speci c lymphocyte subsets.36 Two tests that can be of value and are readily available in practice are response to vaccination, as measured by rising serum titers, and response to intradermal phytohemagglutinin, which is dependent on a delayed-type hypersensitivity T lymphocyte response and develops in normal animals without prior sensitization.37 A 50- g dose of phytohemagglutinin (PHA) in 0.5 ml of phosphate-bu ered saline is injected intradermally while 0.5 ml of phosphate-bu ered saline is administered intradermally at a distant site. At the PHA site an increase in wheal size of 0.6 mm or less indicates a defect in cell-mediated immunity. Response to vaccination has proved to be a potent means of identifying immunode ciency in such conditions as juvenile llama immunode ciency syndrome.38 Similarly, the equine immune response to a polyvalent inactivated bovine vaccine has been used to document the immunosuppressive e ects of corticosteroid administration.39 For practical purposes, response to rabies or tetanus vaccination may be the most suitable available tests provided that no routine vaccination had been administered in the immediate past. Equine rabies or tetanus antibody titer determination is typically commercially available, and the majority of available vaccines are sufficiently potent to provoke a fourfold increase in titer in normal horses.

PHAGOCYTE FUNCTION TESTING Testing of equine neutrophil migration, phagocytic function, and bactericidal activity has been reported by several investigators.40-43 The techniques employed are typically only available in research laboratories and are not well adapted to investigations of individual animals unless adequate age-matched control animals are also examined. More quantitative information can be obtained by adapting assay systems to ow cytometric analysis. Two reports describe ow cytometric analysis of neutrophil phagocytosis of either uorescent microspheres44 or yeast cells.45 Raidal et al. have described testing of alveolar macrophage and blood neutrophil phagocytic function using uorescent-labeled bacteria and oxidative burst activity using oxidation of dichloro uorescein.46-49 These ow cytometric approaches have great promise and have been applied to various studies, including measurements of the effect of age 48-50 and of exercise51 on neutrophil function.

TESTS OF INNATE IMMUNITY Components of the innate immune response that have been measured in the horse include the numbers of granulocytes and monocytes in peripheral blood and their phagocytic function (see previous discussion); natural cytotoxicity in terms of lymphokine-activated killer (LAK) cell activity;41,52,53 and measurement of soluble factors, including several acute-phase proteins, 54-58 and complement. Equine complement activity can be measured using a hemolytic assay in which antibody sensitized chicken erythrocytes are used as target cells, and the amount of serum required to lyse 50% of these targets is expressed as CH50 units.59 More recently, a ow cytometric assay has been described.60 These tests have been used in some instances to detect relative immunode ciency in terms of LAK cell activity in exercised horses53 or complement

activity in foals.61 Currently, these techniques have very limited availability.

PRIMARY IMMUNODEFICIENCIES Severe Combined Immunodeficiency SCID is a lethal primary immunode ciency, a ecting Arabian foals, humans, mice, and dogs, characterized by failure to produce functional B and T lymphocytes, and resulting in lack of any antigen-speci c immune responses.1,62-64 The vast majority of a ected foals are Arabians, in which the condition is inherited as an autosomal recessive trait, and results from a lack of DNA protein kinase activity that prevent V(D)J recombination.7,65 In studies conducted in the United States and reported in 1977, the incidence of SCID among Arabian foals was at least 2% to 3%,66 suggesting a carrier prevalence rate between 25% and 26%. However, in more recent studies conducted in the United States using a precise molecular diagnosis of the carrier state, carrier prevalence was consistently 8%.67,68

CLINICAL SIGNS AND LABORATORY FINDINGS A ected foals are clinically normal at birth but develop signs of infection during the rst 1 to 3 months of life. The age of onset of infection depends on the adequacy of passive transfer and degree of environmental challenge. As maternal antibodies are catabolically eliminated, SCID foals are increasingly susceptible to infections with bacterial, viral, fungal, and protozoal agents. Bronchopneumonia is a common disease, often caused by adenovirus, which is the most signi cant pathogen of SCID foals, a ecting two thirds of all animals.69 Adenoviral infection frequently extends to the gastrointestinal and urogenital system and causes pancreatic disease, leading to loss of endocrine and exocrine tissue and possibly contributing to the impaired growth and weight loss observed in SCID foals.1 Other common infections in SCID foals include Pneumocystis carinii pneumonia; Rhodococcus equi infections; enteritis, frequently caused by Cryptosporidium parvum70; arthritis; and omphalophlebitis. Clinical signs in SCID foals may include nasal discharge, coughing, dyspnea, diarrhea, fever, and weight loss. Although antibiotics, plasma, and supportive care prolong the course of disease, death invariably occurs before 5 months of age. The only exception to this rule was a single foal experimentally treated with a bone marrow transplant from a histocompatible donor that lived until 5 years of age before dying of an unrelated cause.71 A consistent hematologic nding is absolute lymphopenia (500 cells/ l) during the rst 5 days of clinically apparent infections. In experimental infections of EHV1, a decrease in neutrophils and monocytes occurs on day 3 after infection, but the absolute values of both cell types remain within normal limits.358 Diagnosis 1. Viral isolation. The virus has been recovered from peripheral blood mononuclear cells for up to 2 weeks after the infection or from nasopharyngeal swabs that have been collected in viral transport media. However, nasopharyngeal swabs should be obtained in the early febrile phase of the disease and transported to the laboratory as soon as possible in a sterile, cold-transport medium.333 Viral isolation in cell culture is limited by time constraints and the need to con rm viral identity by immunostaining or PCR. In contrast, direct PCR on clinical materials is rapid and enables distinction between EHV1 and EHV4.365 Recent work reveals poor sensitivity of a real-time PCR assay in detecting EHV1 DNA in bu y coat leukocytes.366 A PCR assay of materials obtained from nasopharyngeal swabs is more sensitive.352 However, because of the short duration of nasal shedding, viral DNA may not be detected, and thus diagnosis should also be supported by serologic assessment. A PCR has been developed that can distinguish between neuropathogenic and nonneuropathogenic strains of EHV1.367 2. Serologic diagnosis is via virus neutralization, immunoassay, complement xation, or a radial immunodi usion enzyme assay that detects antibodies against EHV1. A fourfold change in antibody titer is suggestive of infection.368 A commercially available ELISA (Svanovir, Svanova Biotech AB, Uppsala, Sweden) based on the structural di erences of the viral glycoprotein G of EHV1 and EHV4 distinguishes between antibodies against these two viruses.352 3 . Postmortem examination of foals that died from EHV1 or EHV4 infection revealed interstitial pneumonia, pleural and peritoneal e usions, hypoplasia of the thymus and spleen, focal necrosis of the liver, and viral inclusion bodies within the hepatic parenchyma. Similar pathologic ndings are observed in aborted fetuses. Immunohistochemistry, in situ hybridization, and PCR can be used on tissues to establish the diagnosis of EHV1. Treatment Data available from experimental or clinical trials using antiviral drugs (nucleoside analogs) for the treatment of equine herpesvirus infections are limited. The key protein target for many of the antiviral compounds is the virus-encoded DNA polymerase. The activity of the second-generation nucleoside analogs such as acyclovir, valacyclovir (its prodrug), or ganciclovir is dependent on the drug being phosphorylated by the viral thymidine kinase. During latent infections, the virus does not express genes encoding for viral proteins (thymidine kinase, DNA polymerase), thus latent virus is unaffected by any of the conventional nucleoside analogs.328 Nevertheless, in human medicine, routine use of these antivirals can suppress recurrent infections in patients for periods up to 10 years. In a recent investigation, the in vitro activity of six nucleoside analogs in reducing plaque number or plaque size associated with EHV1infected cells was examined.369 The investigators used three di erent abortigenic and three neuropathic strains of EHV1. They found that the ED50 (de ned as the 50% e ective concentration for inhibition of plaque formation) of ganciclovir was less than that of acyclovir and that no di erences in susceptibility of the six di erent EHV1 strains to these two drugs existed. Based on their studies, the researchers in the investigation suggested that an e ective concentration for ganciclovir was 0.1 g/ml and that of acyclovir was at least 2.0 g/ml. Given that ganciclovir would be quite expensive to use clinically in the horse, it was more likely that acyclovir would be used. Indeed, during outbreaks

of EHV1 involving neurologic disease or neonatal deaths, acyclovir has been administered at doses ranging between 8 to 16 mg/kg orally three times daily to 20 mg/kg orally three times daily.370,371 The e cacy of acyclovir in treating respiratory diseases has not been demonstrated, and its poor bioavailability (3%) when given orally372 may preclude it from achieving the requisite viricidal plasma levels (>2 μg/ml). Even when administered intravenously at 10 mg/kg,373 the in vitro data suggest that the drug would have to be given either as a constant infusion or as a single treatment eight times a day to achieve plasma concentrations between 1.7 and 3.0 g/ml.369 In humans, the major reported adverse e ect of parenteral administration of acyclovir is renal toxicity caused by precipitation of the acyclovir crystals in the renal tubules when maximal solubility is exceeded.374 Because of its improved bioavailability, interest in the acyclovir prodrug valacyclovir has prompted recent pharmaco*kinetic investigations. Garré, Shebany, Gryspeerdt, et al. (2007) reported that when valacyclovir was administered to horses at 20 mg/kg, its bioavailability was much improved, as compared with acyclovir (26%). However, plasma concentrations were maintained above the EC50 value for only 1.5 to 2.0 hours. Based on computer modeling, the investigators suggested dosing valacyclovir at 40 mg/kg orally three times daily in an e ort (1) to maintain plasma concentrations greater than 1.7 g/ml during the entire treatment interval and (2) to exceed plasma concentrations of 3.0 g/ml during 30% of the treatment interval.374 One potential drawback of acyclovir (or valacyclovir) is the development of viral resistance in strains lacking thymidine kinase, although this circ*mstance, to date, has not been realized. Another antiviral drug, (s)-1-[(3-hydroxy-2-phosphonyl methoxyl)-propyl] cytosine (HPMPC), has been shown to be e cacious when given before EHV1 challenge in naïve foals but has not been used clinically in North America.375 Because evidence of immunosuppression during herpetic infections exists, the clinician may consider administering broad-spectrum antimicrobials for 7 to 10 days to affected horses. In addition, horses should be rested for 4 to 6 weeks before exercise training commences. Management Recommendations during an Outbreak During an outbreak of EHV1, measures should be implemented to reduce the potential sequelae, such as abortion and neurologic disease. A ected animals should be isolated, and no horse should be permitted to leave the property or track until 3 weeks after recovering from the last clinical sign.333 Stalls should be cleaned thoroughly of contaminated materials and then disinfected. Prevention Commercially available vaccines include two single-component inactivated (killed) vaccines that are marketed for the prevention of EHV1induced abortions in pregnant mares and respiratory infections (Pneumabort K, Fort Dodge Animal Health; Prodigy, Intervet Inc.). Three intramuscular doses are recommended for primary immunization followed by an annual booster. Multicomponent inactivated vaccines include products such as Prestige (Intervet Inc.) or Calvenza (Boehringer Ingelheim) that contain both EHV1 and EHV4 or Fluvac Innovator (Fort Dodge Animal Health) that contains in uenza A1 and A2 with EHV1 and EHV4. A modi ed-live product (Rhinomune, P zer Animal Health) is also commercially available, although it is not recommended for use in pregnant mares. None of the products provides complete protection against clinical disease. On the premises with con rmed clinical EHV1 infections, booster vaccination of horses that are likely to have been exposed to EHV1 is not recommended. Booster vaccinating nonexposed horses and horses that would come onto the premises is rational. This approach assumes that vaccination induces an anamnestic and not a primary response. The question of whether vaccination during an outbreak could induce nasal shedding (and thus preclude identi cation of clinically infected horses during an outbreak) was investigated. In a pilot study, 14 horses were vaccinated with either a killed product (Pneumabort K + 1b) or a modi ed-live product (Rhinomune). The investigators found that EHV1 DNA was not detected by PCR in nasopharyngeal swabs or in whole blood samples obtained from horses at 1-week intervals for 4 weeks after the vaccination.376 These results suggested that in healthy unstressed horses, one should not detect vaccine DNA or reactivated latent EHV1 DNA after the vaccination. A recent product that has been marketed to prevent EHV1 or EHV4 (or both) infections during disease outbreaks or during times in which the horse is stressed is Zylexis (P zer Animal Health), an inactivated Parapox ovis virus was purported to have immunomodulator activity. The manufacturer’s recommendation are to administer the preparation to healthy horses 4 months of age or older via a series of three intramuscular injections given on days 0, 2, and 9. No scienti c data establishing the e cacy of Zylexis have been published by independent groups.

EQUINE HERPESVIRUS 2 EHV2 is a slowly replicating virus, often known as equine cytomegalovirus. It is believed to establish latency in B-lymphocytes but can be found in lymphoid and neural tissue.377 Epidemiology EHV2 is endemic in horse populations tested worldwide with approximately 90% of horses tested being seropositive.378,379 Because the prevalence of seropositivity and the magnitude of these titers are not di erent between healthy and diseased horses, opinions vary as to the

true pathogenicity of EHV2.142,380 In surveys of healthy adult horses, viral DNA can be recovered from the peripheral blood leukocytes in the majority of animals, with frequencies ranging between 68% to 71% 381 to 89%.332 The virus has been isolated from nasopharyngeal swabs obtained in young racehorses with PLH, implicating EHV2 in the development of this condition.144 The suggestion has been made that most foals are exposed to EHV2 at 2 to 3 months of age and shed the virus for 2 to 6 months until a humoral antibody response develops and eventually is associated with low viral recovery.380 EHV2 has been isolated from 2- to 3-month-old foals during outbreaks of pneumonia in Hungary, Japan, Australia, New Zealand, and the United States.380,382-385 In a longitudinal study of 27 foals in a Hungarian stable that experienced respiratory disease outbreaks, peripheral blood leukocytes and nasopharyngeal swabs were obtained from young foals until they reached 23 weeks of age.381 By 6 weeks of age, 93% of the nasopharyngeal swabs were positive for EHV2 DNA, and by 8 weeks, 100% of the peripheral blood leukocytes were positive for EHV2 DNA. In some investigations of respiratory disease outbreaks in foals, EHV2 was the sole infectious agent isolated from the respiratory tract, whereas in other outbreaks, bacterial agents—S. zooepidemicus and Rhodococcus equi—complicated the pneumonitis.384 This nding has led some researchers to speculate that if the foals are stressed or infected with a large dose of the virus during a period in which maternal antibodies are waning, then the virus may invade the cells of the immune system and cause further immunosuppression. Blakeslee, Olsen, McAllister, et al.142 reported a dose-related immunosuppression after in vitro infection of lymphocytes with EHV2, and this effect has been documented in vivo using a rabbit model. Pathogenesis After natural exposure, EHV2 is recoverable from nasal and nasopharyngeal swabs; from the kidney, bone marrow, spleen, mammary gland, salivary gland, vagin*, tracheal mucus, neural tissue; and from the cornea and conjunctiva.378,386 Leukocytes and lymph nodes draining the respiratory tract are major reservoirs of EHV2 DNA.378 Clinical Signs In foals that are experimentally infected with EHV2, pharyngitis and lymphoid hyperplasia develop, implicating the EHV2 as a causative agent of chronic lymphoid hyperplasia in young horses.142 However, the role of EHV2 in the natural production of pharyngitis or lymphoid hyperplasia remains uncertain. The role of EHV2 in causing keratoconjunctivitis also remains equivocal, with clinical investigations both supporting387 and refuting its involvement.379 Diagnosis EHV2 diagnosis depends on isolation of the virus, serologic evidence of an active viral infection, and the presence of clinical signs. EHV2 is highly cell associated but more slowly cytopathic than the equine alpha-herpesviruses. Its slow growth in culture may hinder its recognition if alpha-herpesviruses are present, increasing the time required for laboratory diagnosis.353 Diagnosis is aided by PCR methods. Treatment Treatment is symptomatic. Efficacy of antiviral agents has not been examined in EHV2 infections. Prevention No commercial vaccine is available.

EQUINE HERPESVIRUS 5 Causes Originally classified as EHV2, examination of restriction endonuclease profiles of some slowly growing equine herpesviruses eventually led to its identification as a separate DNA virus in the gamma-herpesvirus subfamily. Epidemiology The initial isolates of EHV5 were obtained from peripheral blood leukocytes and nasal swabs from Australian horses,388 but the virus has been subsequently identi ed in horses in Switzerland, Germany, New Zealand, and the United States.381,389,390 In a longitudinal study of mares and their foals that were monitored from birth to 6 months of age, latent EHV5 infections were detected in the peripheral blood mononuclear cells of 88% of the foals at 6 months of age. The presence of the virus was not associated with respiratory disease in that study.390 Pathogenesis Little is known about the natural course of disease in horses infected with EHV5. The initial isolates of EHV5 were from horses with upper respiratory tract disease and, similar to EHV2, has been implicated in the production of lymphadenopathy, immune suppression, malaise, and respiratory tract disease.391 Clinical Signs

In a retrospective study of 24 horses with clinical signs and radiographic evidence of interstitial lung disease, EHV5 was detected by PCR in 79% of the para n- xed lung samples obtained from these horses.392 However, causality between EHV5 and multinodular interstitial lung disease remains to be established because EHV5 can be isolated from peripheral blood mononuclear cells (PBMCs) or nasal swabs of 24% to 33% of healthy horses.381,389 Diagnosis The diagnosis is con rmed by viral isolation and identi cation of the virus by PCR. Serologic tests are problematic because EHV2 and EHV5 share considerable cross-reactivity in ELISA and radioimmunoprecipitation because of common epitopes.391 Treatment Treatment is supportive if indicated. (See the later section on interstitial pneumonia.) Prevention No vaccine is currently available.

EQUINE VIRAL ARTERITIS Causes Equine arteritis virus (EAV), an enveloped single-stranded RNA virus, is a member of the family Arteriviridae (genus Arterivirus, order Nidovirales). EAV is genetically similar to corona viruses (also a member of Arteriviridae) but has a dissimilar viral structure and complement- xing antigen.393 The virus contains six envelope proteins (E, GP2b, GP3, GP4, GP5, and M) and a nucleocapsid protein (N). The major neutralization determinants of EAV are expressed by GP5, and although considerable variation in the sequence of this protein exists in eld strains of the virus, only one serotype of EAV (the Bucyrus strain) has been found.394 All strains of EAV evaluated thus far are neutralized by polyclonal antiserum raised against this highly virulent Bucyrus strain.395 The virus, which has a worldwide distribution, may cause sporadic and sudden death in foals, abortion in mares, and mild or subclinical infections in adult horses.393 Epidemiology Before the 1984 epizootic of equine viral arteritis (EVA) in Kentucky, little attention was paid to this virus.396 Most infections were assumed to be transmitted by inhalation until the important role of venereal transmission was demonstrated in those outbreaks. The virus is maintained in the equine population by long-term and short-term stallion carriers. The duration of the carrier status ranges between several weeks to years. Testosterone is essential for maintenance of persistence because long-term infections do not occur in colts exposed to the virus before the onset of peripubertal development or in mares.396-398 During the carrier state, the virus is present solely in the reproductive tract, principally in the ampulla of the vas deferens. In the carrier stallion, genetic and antigenic variation (diversity of viral isolates) is generated in the course of persistence.395 Primary exposure to the virus is presumed to be via the venereal route. Susceptible mares bred to carrier stallions almost always become infected with EAV. Viral shedding into the respiratory tract allows secondary horizontal transmission of the virus to occur. The latter type of transmission (aerosol or fomites) would be more typical of that occurring in young racehorses at training facilities. The mode of transmission to neonatal foals is not known and might entail both transplacental infection during late gestation or inhalation during or after the birth.399,400 In addition to the outbreak in Kentucky in 1984, several other outbreaks have been reported in North America, Europe, Australia, New Zealand, and South Africa.395 A signi cant di erence in seropositivity to EAV exists not only between countries, but also among the horse breeds within a given country. Approximately 84% of Standardbreds and 93% of Austrian Warmblood stallions have circulating antibodies to EAV, whereas less than 5% of Thoroughbred and less than 1% of Quarter Horses are seropositive.401 No evidence has been found of any breed-speci c variation in susceptibility to EAV infection or in establishment of the carrier state, thus the number of actively shedding carrier stallions likely determines the prevalence of EAV infection in the individual horse breeds.395 The virulence of the strains of EAV associated with individual horse breeds may not be constant, and those shed by carrier Standardbred stallions are often very highly attenuated and cause minimal if any disease in susceptible horses (regardless of breed).395 Pathogenesis After intranasal challenge (aerosolization), the virus invades the respiratory tract epithelium and the alveolar macrophages. By 72 hours after infection, replicating viruses are detectable in the bronchopulmonary lymph nodes, endothelium, and circulating macrophages. With venereal transmission (by natural cover or by arti cial insemination), the virus can be isolated from swabs of the rectal and vagin*l mucosa during the febrile periods. Dissemination of the virus by hematogenous routes allows infection of mesenteric lymph nodes; spleen; liver; kidneys; nasopharyngeal, pleural, and peritoneal uid; and urine.396 By 6 to 8 days after infection, the virus has localized within the endothelium and medial myocytes of blood vessels, where it causes a necrotizing arteritis, a panvasculitis.393 Thus the clinical manifestations of EVA re ect vascular injury with increased permeability and leakage of uid.395 The vascular lesion may be caused by a direct cytopathic e ect of the

virus on the endothelium and medial myocytes, from the e ects of anoxia or thrombosis induced by cell damage, or from virus-induced macrophage-derived vasoactive and in ammatory cytokines.395 In neonatal foals, the virus, by virtue of its ability to cause lymphoid depletion in the spleen and lymph nodes, enhances the neonate’s susceptibility to secondary bacterial infections.400 Clinical Signs The vast majority of EAV infections in equids (horses, donkeys, and mules) are inapparent or subclinical, but occasional outbreaks of EVA occur that are characterized by persistent infections in stallions, in uenza-like illness in adult horses (with nasal and ocular discharge), abortion in pregnant mares, and fatal interstitial pneumonia and enteritis in neonatal foals.395 The variation in clinical signs may be a function of host factors, such as age and immune status, and virus factors, including the virulence, the amount of infective virus, and the route of infection.402 Thus very young, very old, debilitated, or immunosuppressed horses are predisposed to severe EVA. The incubation period ranges between 3 to 14 days (6 to 8 days if transmitted venereally). In the acute disease, horses are febrile (105° F) for 1 to 5 days, anorectic, depressed, and may cough. They may exhibit a serous nasal discharge, congestion of the nasal mucosa, mandibular lymphadenopathy, conjunctivitis, lacrimation, and, less frequently, corneal opaci cation. Edema of the sheath, scrotum, ventral midline, limbs, and eyelids occurs secondary to the vasculitis. Other signs may include respiratory distress, sti ness, soreness, diarrhea, icterus, skin rash on the neck, and papular eruptions on the inside of the upper lip. Most adult horses recover uneventfully; mortality rarely, if ever, occurs in natural outbreaks of EVA. Note that the highly virulent horse-adapted Bucyrus strain of EAV, which does cause high mortality in healthy adult horses, is not representative of field strains of the virus and is best regarded as a laboratory strain.395 Abortions occur between 10 to 34 days after exposure (during the third to tenth months of gestation), but the mechanism is not known. Abortion can occur with or without preceding clinical evidence of infection.403 EVA has not been associated with teratologic abnormalities in fetuses or foals. Neonates may die suddenly or develop severe respiratory distress followed by death.396,399 In such cases, the dam of the a ected foal may be completely asymptomatic.393,400 Clinical Pathology Infection causes leukopenia, lymphopenia, and thrombocytopenia. Diagnosis In North America, EVA is a reportable disease. Although the virus may persist in the bu y coat for up to 36 days after infection, viral isolation can be di cult.402 Samples to be submitted for viral isolation include citrated blood samples, nasopharyngeal and conjunctival swabs, vagin*l swabs, and sem*n from an infected stallion.404 PCR enhances viral detection and speci city, but the serum should also be submitted for detection of antibodies. Virus neutralization, complement xation, immunodi usion, and immuno uorescence are techniques used to demonstrate changes in antibody titers. A fourfold or greater rise in titer or a change in status from seronegative to seropositive may indicate a recent infection. Postmortem examinations of aborted fetuses demonstrate edema, excessive pleural uid, and petechiation on the mucosal surfaces of the respiratory and gastrointestinal tracts, but focal necrosis of the liver or intranuclear inclusions are not features of the disease.394 Foals that die acutely of EVA exhibit interstitial pneumonia characterized by a predominant macrophage in ltrate, hyaline membrane formation, and vasculitis of the small muscular arteries.400 Infective virus particles or viral antigen can be identi ed by immunofluorescence in the small muscular arteries of the lung, kidney, spleen, thymus, and heart. Treatment Treatment is symptomatic with the primary objectives being to keep the horse well hydrated and to provide analgesics (nonsteroidal antiin ammatory drugs) as needed. Fever in stallions can lead to sperm damage and temporary infertility; thus one should administer nonsteroidal antiinflammatory drugs. The horses should be isolated for 3 to 4 weeks to minimize the chances of transmission. Prevention EAV infections are readily prevented through serologic and virologic screening of horses coupled with sound management practices that include appropriate quarantine and strategic vaccination.395 Vaccination of seronegative stallions with a commercially available modi edlive vaccine (ARVAC, Fort Dodge Animal Health) is credited with preventing the establishment of carrier states in the stallion or with further increases in the carrier state during the 1984 epizootic in Kentucky. (A modi ed-live vaccine is licensed for use in the United States and in Canada, whereas a killed vaccine [Artervac; Fort Dodge Animal Health] is available in the United Kingdom, Ireland, France, Denmark, and Hungary.405) For example, in the Kentucky outbreak, stallions were bred only to mares that were vaccinated or were seropositive from previous natural exposure to EVA. After vaccination, clinical immunity was found to develop rapidly and to last for 1 to 3 years.396 Some vaccinates develop mild febrile reactions and transient lymphopenia, and the vaccine virus can be sporadically isolated from the bu y coat for 7 days (but occasionally up to 32 days). Thus mares should be vaccinated 3 to 4 weeks before breeding to an EAV carrier stallion and isolated.403 Vaccinated stallions do not shed the virus in either sem*n or urine. Thus vaccination of prepubertal colts prevents them from

becoming persistently infected carriers and is central to any e ective control program. Vaccination is not recommended for pregnant mares or for foals younger than 6 weeks. Fetal infections have been documented after murine leukemia virus (MLV) vaccination of pregnant mares.405 Passively derived antibodies to EAV decrease to undetectable amounts by 8 months postpartum for foals born to seropositive mares. Vaccination should be e ective at this time if necessary.406 Although the current MLV vaccine is e ective, a recombinant subunit vaccine coexpressing the two major envelope proteins of EAV (GP5 and M) that experimentally has been shown to e ectively induce neutralizing antibodies and protective immunization, may find commercial success.405

EQUINE RHINITIS VIRUSES Causes Formerly termed equine rhinoviruses, at least four serotypes of equine rhinitis viruses have been identi ed to date. These viruses have been isolated from horses in the United Kingdom, mainland Europe, the United States, Australia, and Japan. Serotype 1, formerly called equine rhinovirus 1, has been renamed equine rhinitis A virus (ERAV) and is now reclassi ed as a member of the genus Aphthovirus, family Picornaviridae. The reclassi cation is based on (1) the hom*ology of the nucleotide sequence of the ERAV genome with that of foot-andmouth disease virus, (2) the physicochemical properties of ERAV, and (3) the production of viremia and persistent nasal, urine, and fecal shedding after ERAV infection.407 Serotype 2, formerly called equine rhinovirus 2, has been renamed equine rhinitis B virus 1 (ERBV1) and has been reclassi ed to the genus Erbovirus (from equine rhinitis B virus), family Picornaviridae.408 A third serotype, formerly designated P313/75 or equine rhinovirus 3 has also been assigned to the genus Erbovirus. It is now called equine rhinitis B virus 2 (ERBV2).409 A fourth antigenically distinct, acid-stable equine picornavirus has also been assigned to the genus Erbovirus and is called equine rhinitis B virus 3 (ERBV3).410,411 This virus had been originally classi ed with the ERBV1 isolates and exhibits weak serologic cross-reactivity with antibodies against ERBV1.411 Clinical Signs Infection of horses with ERAV causes acute fever, nasal discharge, coughing, anorexia, pharyngitis, laryngitis, and mandibular lymphadenitis.412,413 Infection produces moderate increases in the neutrophil-to-lymphocyte ratios (from 50:49 in healthy horses to 57:43 in infected horses) and in plasma brinogen.413 The importance of ERAV as a cause of respiratory disease has not been recognized previously because clinical signs may be mild and because viral isolation is di cult.414 The incubation period is 3 to 8 days, with shedding of viral particles from pharyngeal secretions, urine, and feces being a feature of the infection. Young horses are infected more commonly than older horses. For example, the prevalence of serum neutralizing titers in 24-month-old Thoroughbreds just entering a training facility had increased from 18% to 63% when sampled 7 months later.415 Other studies have found prevalence rates of neutralizing antibodies to ERAV to range between 73% to 90%.413,416 ERAV has also been isolated from horses without clinical evidence of respiratory disease.415 Equine rhinovirus A has a broad host range that includes rabbits, guinea pigs, monkeys, and human beings, although the virus does not seem to spread horizontally in each of these species. Intranasal inoculation of a human volunteer with ERAV resulted in severe pharyngitis, lymphadenitis, fever, and viremia.412 Experimental and epidemiologic evidence of ERAV infection of human beings exists. High antibody titers were found in the sera of 3 of 12 stable workers, whereas no ERAV antibody was found in the sera of 159 non–stable workers.417 In a serologic survey of 137 samples obtained from mixed-practice veterinarians, only 3% were seropositive for ERAV, suggesting that the risk of infection is low.416 One can diagnose ERAV infections by serologic testing (fourfold changes in titers between acute and convalescent samples) or by PCR identification of the virus in nasopharyngeal swabs.414 Equine rhinitis viruses assigned to the genus Erbovirus are also respiratory tract pathogens. In a survey involving horses with acute respiratory disease, researchers isolated ERBV from 30% of the horses sampled. Of the 28 horses from which the virus was recovered, a serologic diagnosis was made in only 6 horses. Researchers attributed the low diagnostic rate to the di culty of initially collecting samples during the acute phase: Antibody levels rise rapidly from day 6 after infection and peak between days 14 and 18.418 Thus, in contrast to ERAV, viral isolation may be the more successful diagnostic method. In a survey involving Australian horses, serum neutralizing antibodies against ERBV1 were detected in 77% of the 2- to 3-year-old Thoroughbred horses in training and in nearly 94% of stable horses aged 10 to 22 years.415 Interestingly, approximately 4% of mixed-animal veterinarians develop serum neutralizing antibodies against ERBV1, suggesting that, similar to ERAV, humans are at low risk for becoming infected with this virus. The clinical signi cance of ERBV2 (formerly P313/75) is currently uncertain. ERBV2 has been isolated from one horse with a 6-day history of intermittent fever that began 4 days after an umbilical hernia operation and castration419 and from another horse with signs of acute febrile respiratory disease and high titers against ERBV2.340 In the epidemiologic survey conducted by Black, Wilcox, Stevenson, et al. (2007), 89% of 2- to 3-year-old Thoroughbreds at a training facility in Australia exhibited serum neutralizing antibodies to ERBV2. This prevalence rate contrasts with that found in older stable horses of 45%.415 The fourth rhinovirus, ERBV3 (originally classi ed as ERVA), was initially isolated from horses with acute respiratory disease. Additional studies are required to ascertain the severity of clinical disease that occurs with ERBV3 infections, as well as the seroprevalence in equine populations. No vaccine is currently available against ERAV or ERBV1, 2, or 3.

EQUINE ADENOVIRUSES Two separate nonenveloped DNA viruses belonging to the family Adenoviridae have been isolated from horses and are recognized as equine adenovirus 1 (EADV1) and EADV2.420 EADV1 is endemic in most horse populations, and the belief is that most foals acquire the infection during the suckling period from the mares. Disease develops if maternal antibodies wane or if active immune responses are impaired. Susceptible foals include Arabian foals with combined immunode ciency or Fell pony foals with an unde ned immunode ciency.421 The virus replicates in the respiratory tract and can cause conjunctivitis, rhinitis, and bronchopneumonia. In one recent survey, EADV1 was detected by PCR in the nasal swabs of 40% of hospitalized foals and in 58% of healthy control foals.422 EADV2 has been isolated from the lymph nodes and feces of foals with upper respiratory tract disease and diarrhea.423,424 For a more detailed description, the reader is referred to the section on foal pneumonia.

EQUINE HENDRA VIRUS Cause Originally called equine morbillivirus but later renamed equine Hendra virus (HeV), this enveloped RNA virus is a prototype of a new genus, Henipavirus, within the subfamily Paramyxovirinae.425 HeV is a pleomorphic (spherical or lamentous) structure, the genome of which exceeds 18 kbp. The latter encodes for eight di erent proteins, including the cell-attachment (G) and fusion (F) proteins that protrude from the viral envelop.426 Epidemiology The virus was rst recognized as a causative agent of severe respiratory disease in Thoroughbred racehorses in a suburb of Brisbane, Australia, called Hendra.427 During the 1995 outbreak in which 13 of 20 horses died or were euthanized, two humans also became infected, one fatally. However, in a retrospective analysis of postmortem tissues, it was later determined that in 1994, HeV was the cause of death of two other horses and a person assisting with their necropsy in Mackay, a city approximately 800 kilometers from the Brisbane area. Since that time, several additional outbreaks have occurred in Australia involving fewer numbers of horses or humans. In Australia, most equine cases have been documented to occur in the spring or summer and involve horses that were housed near food or the roost trees of ying foxes. Based on testing of wildlife, ying foxes (fruit bats, Pteropus spp.) were subsequently shown to be the source of infection of the horses. Despite the presence of antibody to HeV in fruit bats throughout their geographic range in Australia, suggesting that infection is quite common in that species,428 HeV disease in horses in Australia has thus far been uncommon and sporadic. Pathogenesis Natural and experimental infection in horses suggests an incubation period of 8 to 16 days or 4 to 10 days, respectively.427,429 The means by which HeV infects horses is still unknown. The virus can be isolated from the oral cavity and urine of some infected horses but not from rectal or nasal swabs.425 One hypothesized route of infection is via consumption of pasture or feed contaminated with bat urine, aborted bat fetuses, or reproductive uids.425 (The birthing season of the bats is consistent with the timing of the HeV cases that were recorded in horses between August and January.) In experimental infections, horses can be infected by eating material contaminated with the virus. Transmission to horses from exposure to infected cat urine is also possible. Horse-to-horse or horse-to-human transmission has been suggested to occur by direct contact with frothy nasal discharges, urine, blood, or pleural uids of a viremic horse. No human-to-human transmission has been reported. In experimental infections, horses kept in stalls adjacent to an infected horse fail to become infected, suggesting that the virus is not highly contagious and that aerosol transmission is ine cient. One additional hypothesis regarding transmission of the virus from bats to horses involves the Australian paralysis tick Ixodes holocylus. This tick is a blood feeder that feeds both on bats and horses (as well as other mammals, including humans).430 The mode of viral attachment and cell infectivity of HeV is di erent from that of other paramyxoviruses. It does not engage the sialic acid– containing receptors for attachment but rather depends on the viral G protein to interact with a cell-surface glycoprotein called ephrin-B2.426 The ephrin-B2 receptors are highly concentrated on endothelial cells (hence the tropism of the virus for those cells), as well as neurons and smooth muscle cells. Fusion of the viral membrane is accomplished by the F glycoprotein. This process not only enables the virus to gain entrance into the cell but is also responsible for the formation of syncytial giant cells.426 The virus replicates within the upper and lower respiratory tract epithelium and causes an interstitial pneumonia. Clinical Signs Horses acutely develop signs of respiratory distress that are characterized by fever (>104° F), depression, tachycardia, tachypnea, sweating, poor capillary re ll, and abnormal lung sounds (caused by pulmonary edema). Neurologic de cits such as ataxia, head pressing, and recumbency can also occur.431 The time between the onset of signs until death is usually between 1 and 3 days.425 Humans die as a result of

respiratory and renal failure and relapsing encephalitic disease.430 Diagnosis Di erential diagnoses to be considered in a ected horses include heart failure, acute African horse sickness, highly virulent in uenza, and plant intoxications.431 Serologic testing (ELISA or virus neutralization tests) can be used to detect antibody titers in a ected horses (or humans). Gross abnormalities in the lungs of a ected horses are highly suggestive of but not pathognomonic for the disease. Histopathologic ndings include alveolar edema, hemorrhage, lymphatic dilation, and vasculitis of pulmonary capillaries and arterioles. Within the walls of the blood vessels are the characteristic syncytial giant cells that contain cytoplasmic inclusion bodies.425 From tissues, the virus can be isolated in cell culture, visualized by electron microscopy, and con rmed by indirect immunoperoxidase test, immuno uorescence test, or PCR. Because of the susceptibility of humans, the high virulence of the virus, and the absence of therapeutic modalities and vaccines, HeV is classi ed as a biosecurity level 4 pathogen and requires the highest level of caution and preparedness when working with the tissues obtained from infected animals.426 Such precautions include wearing e ective eye, nose, and mouth protection from fomites and aerosols, complete covering of the body with plastic overalls, and double gloving with gloves tied by adhesive tape to the cuffs of the overalls.431 Treatment No antiviral agent exists that is e ective against Hendra virus. The outbreak of the disease in Brisbane was controlled by a stamping-out program involving slaughter of all known infected horses, quarantine of premises, controls on the movement of horses within a de ned disease control zone, and serologic surveillance to determine the extent of infection.431 Vaccination No vaccine is available. Prevention In endemic areas, one should avoid con ning horses to paddocks that contain roost trees for bats. The virus is susceptible to several di erent disinfectants, including 3% Lysol for 3 minutes at 20° C, 1% glutaraldehyde for 10 minutes at 20° C, irradiation, or formaldehyde fumigation.

AFRICAN HORSE SICKNESS VIRUS Cause The African horse sickness virus (AHSV) is a double-stranded RNA nonenveloped member of the family Reoviridae (genus Orbivirus). It is morphologically similar to other orbiviruses, including bluetongue virus and equine encephalosis virus. Nine antigenically distinct serotypes of AHSV have been found: serotypes 1 to 8 are highly pathogenic for horses (90%-95% mortality rate), whereas infection with serotype 9 is associated with a mortality rate of 70%. Serotypes 1 to 8 are typically found only in restricted areas of sub-Saharan Africa, whereas type 9 is more widespread, being responsible for nearly all epidemics outside of Africa. The one exception to this report was in the 1987 to 1991 Spanish-Portuguese outbreaks resulting from infection with AHSV4.432 The virus is transmitted by hematophagous arthropods, the most important being Culicoides spp. midges. Because of its severity and its ability to expand rapidly and without apparent warning out of its endemic areas, AHS is classi ed as a reportable disease by the World Organization for Animal Health, formerly known as the O ce International des Epizooties (OIE). Epidemiology Zebras are considered to be the natural vertebrate host and reservoir of the AHSV and thus play a role in its persistence in Africa. The single exception to this observation involves serotype 9, which is found in West Africa where zebras no longer occur.432 Indeed, in the early and mid-1960s, serotype 9 spread across the Middle East (Saudi Arabia, Syria, Lebanon, Jordan, Iraq, Turkey, and Cyprus) and Central Asia (Iran, Afghanistan, Pakistan, and India) and later through North Africa into Spain. AHS was believed to have been e ectively eradicated after its appearance in these regions. Then, in 1987, after the import of zebras from Namibia into Madrid, Spain again experienced an outbreak of AHS attributed to serotype 4. Unfortunately, because the climate was su ciently moderate to permit overwintering of the virus or vector, outbreaks continued into 1991. Culicoides imicola is considered to be the most important eld vector for AHSV, and although this biting midge is present in southern Europe (Spain, Portugal, Italy, Greece, and southern Switzerland), concern exists that with global warming and climatic change, the vector may extend its habitat to northern Europe.433,434 Even more worrisome is that Culicoides variipennis, the North American vector of bluetongue virus, is an e cient laboratory vector of AHSV,435 suggesting that should viremic equids gain entrance into North America, transmission of the virus could occur. Although AHSV typically a ects equids, dogs are experimentally susceptible to infection via ingestion of infected horse meat. This route is not considered to be an important means of transmission of the virus to susceptible equids.

Pathogenesis Once the virus gains entry into the host, it replicates in the regional lymph nodes before being spread hematogenously to most organs and tissues in the body. In the lungs, spleen, lymphoid tissues, and in certain endothelial cells, a secondary viral replication phase ensues. The incubation period from inoculation to secondary viremic phase is approximately 9 days.432 Clinical Signs Clinical signs arise as a result of increased capillary permeability and secondary circulatory impairment leading to serous e usion and hemorrhage in various tissues and organs. Four clinical syndromes of AHS infection have been identi ed (listed in order of decreasing severity): pulmonary or acute form, mixed form (cardiopulmonary), cardiac or subacute form, and the mild (horse sickness fever) form.436 In the acute form, horses initially develop a fever (up to 107° F) and severe respiratory distress characterized by tachypnea, base-wide stance, extended neck, nostril aring, and coughing. Frothy white and, occasionally, blood-tinged uid may be evident at the nostrils. The course of the disease is usually 4 to 5 days. The mortality rate for this form is 95%. In the cardiac or subacute form, horses exhibit fever (104° F) and subcutaneous edema in the supraorbital fossa, head, neck, and chest regions. The conjunctiva may be congested, petechial hemorrhages may be evident, and the horse may develop signs of colic. Mortality rates may exceed 50%. The mixed form, the most common syndrome, represents a combination of the two syndromes, with death occurring within 3 to 6 days of fever. Mortality rates are 70% for this form. The AHS fever form typically a ects mules, donkeys, zebras, and partially immune horses and is characterized by a mild to moderate fever and the development of edema in the supraorbital fossa. Affected animals recover. Diagnosis Di erential diagnoses include EVA, purpura hemorrhagica, equine infectious anemia, and babesiosis (early stages). Lesions are not pathognomonic for AHS, thus viral isolation or identi cation con rms the diagnosis. Samples of whole blood collected in an anticoagulant (preferably ethylenediaminetetraacetic acid) during the febrile phase or of pulmonary, lymphoid, or splenic tissues collected at postmortem enable viral isolation. AHSV can be detected in tissues by PCR or by ELISA.432 Treatment Speci c therapy for this viral infection is not available, and treatment of individual a ected horses is supportive and symptomatic. (See respiratory distress section later in this text.) Prevention This noncontagious disease is spread by bites of infected midges, Culicoides spp. Control should be directed at restricting import of infected animals, slaughter of viremic animals to prevent them from serving as a source of the virus, reducing vector access to susceptible animals (by stabling), and implementing vector control by eliminating breeding sites of Culicoides and applying insecticides in and around stables and directly to equids.432,437 Polyvalent-attenuated vaccines (containing serotypes 1, 2, 4 and 2, 6, 7, 8) are available in Africa, where most animals are vaccinated twice in the rst and second years of life and annually thereafter. A monovalent-attenuated vaccine containing serotype 9 is available for use in West Africa and outside of sub-Saharan Africa. The live vaccines are not recommended for use in pregnant mares.

TRACHEAL DISORDERS Definition Tracheal disorders are not common in the horse. Tracheitis often accompanies viral upper respiratory tract diseases (see viral infections) and is characterized by areas of hyperemia and epithelial desquamation on endoscopy. Tracheitis may also develop after smoke inhalation or barn res. Tracheal perforation or rupture can occur as a result of trauma. Tracheal compression may result from extraluminal masses such as streptococcal abscesses or neoplasms such as lymphosarcoma or lipoma. Intraluminal obstructions may follow brotic stricture formation secondary to trauma, tracheostomy, or pressure necrosis from in ated cu s of endotracheal tubes or may be caused by granulomatous reactions within the trachea.438 Foreign bodies, such as twigs, may be inhaled.439 Tracheal collapse may occur as a congenital defect in miniature horses and ponies.440 Clinical Signs Clinical signs of tracheal disorders include subcutaneous emphysema and harsh tracheal sounds. Tracheal perforation can lead to pneumomediastinum and pneumothorax. Horses with viral infections or with tracheitis secondary to res may have chronic and persistent coughing, bilateral nasal discharge, and a foul odor. Mild manipulation of the trachea may induce paroxysmal coughing. Diagnosis A thorough physical examination or endoscopic evaluation of the trachea will identify these disorders. Radiographs may aid in the

visualization of foreign bodies, as well as in the detection of a concomitant pneumonia, pneumomediastinum, or pneumothorax. Treatment Treatment is directed toward the primary cause. Primary in ammation after viral infection is not treated unless a persistent fever and a secondary bacterial tracheitis and pneumonia develop (see later discussion). In horses with tracheitis and pneumonitis secondary to smoke inhalation, a tracheostomy may be indicated to circumvent swollen soft tissues of the nasopharynx and to aid in the removal of casts and airway debris in the large airways. Foreign bodies may be retrieved by endoscopy snare but may require a tracheotomy to retrieve them if the endoscope is not long enough. In the adult horse, a 2-m endoscope is necessary to reach the level of the carina located at the fth to sixth intercostal space. Extraluminal masses require careful drainage or excision, with the potential for spread of infection into the mediastinum. Minor tracheal tears usually resolve within 48 hours and may be managed conservatively. Large tears should be managed surgically with débridement and repair of the defect. If complete rupture occurs between tracheal rings or trauma is extensive, then tracheal resection and anastomosis is indicated.441,442 Rupture of the trachea may require a temporary tracheotomy distal to the rupture to ensure patency of air ow. Broad-spectrum antimicrobial therapy may be indicated, especially if the mucociliary clearance mechanism is compromised. Complications such as subcutaneous emphysema, cellulitis, and pneumomediastinum usually resolve over 2 weeks. Horses with tracheal defects have a poor to guarded prognosis for return to athletic potential.

EMERGENCY TRACHEOSTOMY An emergency tracheotomy may be necessary in horses with acute airway obstruction of any origin. Common indications include arytenoid chondritis, strangles, and anaphylactic reactions. The procedure is most readily performed in the standing horse. A site on the ventral midline at the junction of the proximal and middle third of the neck is selected and the trachea palpated between the paired sternothyrohyoid muscles. If time allows, the area is prepared for surgery, and the skin and subcutaneous tissues are in ltrated along the midline with local anesthetic. A 10-cm skin incision is made along the midline. The subcutaneous tissue and paired sternothyrohyoid muscles are separated bluntly until the trachea can be seen. A curved hemostat is placed on the midline and used to retract the musculature laterally to the left. A stab incision is made into the trachea between two tracheal rings and extended to the tracheal midline. This procedure is repeated on the right side so that the incisions connect. It is important to re ect the muscles and to avoid extending each incision beyond the midline to prevent inadvertent transaction of the common carotid artery or jugular vein. Removing a tracheal ring is not necessary for this emergency procedure. A tracheostomy tube is placed in the incision to provide an airway. Two types of tubes are in common use. The older style metal tracheostomy tube is inserted in two pieces, locks, and is self-retaining. This tube has the advantage of not occluding the trachea if its lumen becomes blocked. The other type of tube is made of a synthetic material (commonly silicon) and has an in atable cu . The tube should be inserted with the lumen down (toward the lungs) and secured with umbilical tape to skin sutures and around the neck. The cu should not be in ated because doing so will occupy the space around the tube and lead to airway obstruction should the tube’s lumen become occluded with inflammatory exudate. Barrier cream should be placed below the surgical site to prevent serum scaling of the skin.

NORMAL LUNG Anatomy The lungs of the horse di er from those of other domestic species in that they lack deep interlobar ssures and distinct lung lobes. Super cially, however, the left lung consists of a cranial and caudal lobe, whereas the right lung contains a cranial, an intermediate, and a caudal lobe.443 The intrathoracic trachea bifurcates into the right and left mainstem bronchi at the level of the fth or sixth intercostal space and enters the hilum of each lung. At its division from the trachea, the right bronchus assumes a straighter, more horizontal position relative to the left bronchus, a con guration that may predispose the horse to develop right-sided pulmonary disorders. Each bronchus divides into lobar, segmental, and subsegmental bronchi with the eventual formation of bronchioles. In the distal part of the bronchial tree, the terminal bronchioles lead into poorly developed respiratory bronchioles or open directly into alveolar ducts.443,444 The tracheobronchial lining consists of tall columnar, pseudostrati ed epithelium interspersed with serous and goblet cells. The goblet cells and the underlying submucosal mucous glands function to produce mucus consisting of an outer gel and an inner sol layer.445 Mucus serves to prevent epithelial dehydration, contains protective factors that guard against infectious agents, and is an integral part of the mucociliary apparatus. The rapid beating of the cilia moves mucus and any particulate matter to the pharynx to be swallowed. Approximately 90% of material deposited on the mucinous layer is cleared within 1 hour. Ciliated cells decrease in frequency with successive divisions of the airways so that at the level of the bronchioles the epithelium is composed predominantly of low ciliated and taller nonciliated Clara cells.446 Glands are also absent at this level. The bronchiolar epithelium becomes contiguous with that of the alveoli, which are characterized by two distinct cell types. Type I pneumocytes cover most of the alveolar surface with thin cytoplasmic extensions 0.2 to 0.5 m thick.445 Type II cells are considered to be stem cells, replacing the type I cells when damaged. Type II cells are cuboidal and contain the characteristic lamellar cytoplasmic inclusions thought to contain surfactant components. Surfactant is a complex mixture of phospholipids, surfactant-speci c proteins (SP-A, -B, -C, and -D), and neutral lipids that functions to lower alveolar surface tension (SP-B, SP-C) and to enhance innate immune defenses (SP-A, SP-D). Age-related di erences in the phospholipid components of surfactant exist. The concentrations of phosphatidylglycerol (a marker of lung maturity) are signi cantly lower in the bronchoalveolar lavage fluid (BALF) of term foals as compared with adult horses.447 Lymphocytes are scattered throughout the pulmonary epithelium, are associated with the bronchioles and bronchi, and occur in discrete nodules or patches.448 These cells, along with alveolar macrophages, provide an integral component of the pulmonary immune surveillance system. Macrophages are derived from blood monocytes via the interstitium and are cleared continuously from the alveoli. They are the predominant cell type recovered from the bronchoalveolar lavage in normal horses (see Table 9-1). Pulmonary interstitial macrophages are able to ingest particles introduced within the pulmonary circulation rapidly.449 The ability of these macrophages to localize antigenic particles within the pulmonary vasculature may predispose the equine lung to development of acute pulmonary in ammation (e.g., during endotoxemia). Additional cells within the lung parenchyma include subepithelial and free mast cells. These cells bear IgE on their cell surface and release biogenic amines in response to speci c antigen stimulation. They appear to be important in the pathogenesis of several equine pulmonary disorders, including anaphylaxis and lungworm infections. Tracheobronchial secretions consist of substances secreted from mucous and serous glands, as well as serum transudates. The principal component of the secretions is water, with approximately equal amounts of protein, carbohydrate, lipid, and inorganic material. Contained within the protein fraction of the tracheobronchial secretions are albumin, IgA, IgG (the predominant immunoglobulin in the lower lung), lysozyme, lactoferrin, haptoglobulin, transferrin, and complement components.445

Vascular Supply The lung has two sources of blood ow. The major source of blood is the pulmonary circulation, a low-pressure, low-resistance system that serves primarily to deliver blood to the alveoli for participation in gas exchange and secondarily to provide nutrients to the alveolar constituents. The distribution of the pulmonary arterial ow to the various lung regions depends largely on mechanical forces: gravity and pulmonary arterial, pulmonary venous, and alveolar pressures.450 In the standing horse, a vertical perfusion gradient of the lung has been demonstrated,451 the most dorsal part of the equine lung being less well perfused than the ventral dependent regions. The distribution of pulmonary blood ow is also in uenced by vasoactive compounds such as catecholamines, histamine, and eicosanoids and by changes in alveolar oxygen and carbon dioxide. Alterations in pulmonary vascular resistance may help match ventilation with perfusion and optimize gas exchange. The bronchial circulation, the second source of blood ow to the lungs, provides nutrient support to the lymphatics and vascular and airway components. Bronchial circulation provides arterial blood to the pleural surface and anastomotic connections with the alveolar

capillary bed derived from the pulmonary circulation.444 The magnitude of the anastomotic ow depends on the relative pressure in the bronchial and pulmonary microvasculature and on the alveolar pressure. Thus, in the dorsal part of the lung where pulmonary arterial ow is poor, blood ow from the bronchial circulation may be favored.452 The degree of anastomotic bronchial blood ow through the alveolar capillaries increases with systemic arterial hypoxemia and alveolar hypoxia.450 The bronchial circulation undergoes hypertrophy (angiogenesis) in response to inflammatory pulmonary diseases such as chronic bronchitis, bronchiolitis, and bronchiectasis.

Innervation The autonomic nervous system supplies innervation to the pulmonary structures. This nervous system in uences (1) airway smooth-muscle tone, (2) secretion of mucus from the submucosal glands, (3) transport of uid across the airway epithelium, (4) permeability and blood ow in the bronchial circulation, and (5) release of mediators from mast cells and other in ammatory cells.453 In the horse, the autonomic nervous system can be classi ed functionally into three categories: (1) parasympathetic, (2) sympathetic, and (3) nonadrenergic noncholinergic (NANC) pathways.454 The vagus nerve is an integral component of the parasympathetic and NANC systems. The vagus nerve not only contains the e erent bers of these pathways, which help alter airway resistance, lung volume, and compliance, but also contains the a erent bers, the central input of which regulates the pattern of breathing. Stimulation of the parasympathetic bers within the vagus nerve (i.e., the e erent limb) releases acetylcholine from the postganglionic bers and combines with the muscarinic receptors of the airway smooth muscle to cause bronchoconstriction. Nonetheless, because atropinization (antagonism of muscarinic receptors) does not normally change airway resistance, the parasympathetic system appears to exert little in uence on resting airway tone in the healthy horse.455 However, vagal efferents are important in reflex bronchoconstriction during pulmonary disease. As described previously, the vagus nerve also contains a erent nerve bers that transmit information detected by three types of pulmonary mechanoreceptors: (1) the slowly adapting receptors, (2) the rapidly adapting receptors, and (3) the nonmyelinated C bers. The receptors relay information to the central respiratory neurons located within the ventral medulla and pons. Changes in breathing depth and frequency occur in response to mechanical deformations or chemical stimulation of these receptors. Thus these receptors mediate the tachypneic breathing pattern that occurs in response to inhalation of irritant substances. Stimulation of these receptors also in uences airway smooth-muscle tone,456 inducing reflex bronchoconstriction in response to inhaled irritants. In the lung, α- and β-adrenergic receptors mediate sympathomimetic e ects. Postsynaptic bers course from the sympathetic ganglion to airway smooth muscle where released norepinephrine interacts with α- and β-receptors. Postganglionic sympathetic bers also innervate the parasympathetic ganglia. The released norepinephrine inhibits cholinergic neurotransmission.454 However, numerous β 2-adrenergic receptors are distributed throughout the pulmonary parenchyma that lack innervation by the sympathetic bers. Circulating catecholamines are probably important in activating these receptors and in causing subsequent bronchodilation. Interestingly, β 2-adrenergic stimulation does not alter airway diameter in the healthy horse457 but does increase airway caliber in horses with recurrent airway disease. β-Agonists also promote ion transport and water secretion across human airway epithelium (in vitro) and may cause similar e ects in vivo in the horse. This e ect would bene t mucociliary clearance by increasing the sol component of the mucus. In addition, β-agonists stimulate surfactant secretion by type II pneumocytes, inhibit antigen-induced mast cell degranulation, and modulate cholinergic neural transmission.453 Such effects remain to be demonstrated in the horse, however. α-Adrenergic receptors also exist in the equine lung, but their stimulation induces bronchoconstriction only in horses with reactive airway obstruction and not in normal horses.458 This nding suggests that, in health, these receptors are probably unimportant in the regulation of bronchom*otor tone. An additional autonomic pathway, the NANC inhibitory system, has been demonstrated in the equine lung, and its bers course within the vagus nerve.454 Although the mediators involved in neural transmission have not been identi ed de nitively, some researchers speculate that vasoactive intestinal peptide or peptide histidine isoleucine or both may be the neurotransmitters.459 Stimulation of this pathway causes a 50% reduction in smooth-muscle tone, thus exerting a vasodilating e ect.460 Interestingly, horses with recurrent airway disease appear to lack NANC innervation in the bronchioles, and this dysfunction may contribute to the airway hyperactivity observed.454

Lymph Drainage Lymph drainage from the lung is accomplished by (1) the deep lymphatics, which begin at the level of the alveolar ducts and run with the conducting airway and arteries toward the hilar lymph nodes, and (2) the super cial lymphatics, which drain the visceral pleura through a plexus converging on the hilum.445

Pulmonary Physiology The major function of the lung is gas exchange: di usion of oxygen and elimination of carbon dioxide. During eupnea, an adult horse breathes at a frequency of approximately 12 breaths per minute and a tidal volume near 6.5 L, which represents a total minute ventilation of

77 L/min, or more than 100,000 L/day. During this same period, the resting horse consumes approximately 2.1 L/min of oxygen (3000 L/day) and produces approximately 1.7 L/min of carbon dioxide (2400 L/day).67 Thus the lung provides an important means by which normal arterial oxygen and carbon dioxide tensions and arterial pH are maintained. In the absence of lung disease, pulmonary arterial oxygenation (Pao2) depends on the inspired fraction of oxygen (normally 0.21) and the e ective level of alveolar ventilation, which is approximated by a version of the alveolar gas equation:

where PAo2 is the alveolar oxygen tension, PIo2 is the inspired oxygen tension, PAco2 is the alveolar carbon dioxide tension, and R is the respiratory exchange quotient (the ratio of carbon dioxide production to oxygen consumption). Because an admixture of venous blood with the pulmonary arterial circulation occurs, alveolar oxygen tension exceeds arterial oxygen tension by approximately 10 mm Hg. In the absence of alterations in inspired oxygen tensions, arterial hypoxemia (Pao2 < 85 mm Hg) may result from basically four processes, including (1) hypoventilation, (2) diffusion impairment, (3) ( ) inequality, and (4) right-to-left shunts.461 Alveolar and thus arterial carbon dioxide levels depend on the rate of carbon dioxide elimination relative to its production, given by the following equation:

where VCO2 is the rate of CO2 production and VA is alveolar ventilation. Hypoventilation is de ned as inadequate CO2 elimination relative to production, resulting in an elevated Paco2 (hypercapnia, hypercarbia). Hypoventilation can result from a variety of abnormalities. A convenient diagnostic approach follows the control of breathing from (1) the respiratory center in the medulla, (2) the e erent nerves (phrenic), (3) the bellows (diaphragm and chest wall muscles), (4) the pleural space, and (5) the airways. Some situations in which hypoventilation may develop include (1) drug administration (barbiturates, diazepam, xylazine); (2) diseases of the brainstem after infection (encephalitis), trauma, hemorrhage, or neoplasia; (3) diseases of the respiratory muscles, including botulism, trauma, or fatigue; (4) pleuritis and space-occupying lesions; and (5) choke and upper airway obstruction. In patients su ering from hypoventilation, arterial oxygenation improves with oxygen administration (by increasing inspired oxygen tension), but the most e cacious mode of correcting the hypercapnia and consequent hypoxemia is by mechanical ventilation. Di usion impairment occurs when time is inadequate for equilibration of alveolar oxygen tensions with pulmonary capillary oxygen tensions. Under normal resting conditions, equilibration of oxygen tensions occurs within 0.25 seconds, approximately one third of the contact time of the blood in the pulmonary circulation.461 Some evidence suggests that in the exercising horse, the arterial hypoxemia that normally develops at high exercise intensities is caused in part by the decrease in time available for oxygen di usion.462,463 During rest, di usion impairment is unlikely to occur. Arterial hypoxemia caused by diffusion impairment can be corrected by administration of supplemental oxygen. Inequalities are the primary cause of arterial hypoxemia and occur when alveolar ventilation and pulmonary blood ow are mismatched despite the existence of several re exes that normally tend to prevent this problem. For example, a fall in the ( ) ratio causes alveolar hypoxia, which induces pulmonary vasoconstriction and decreases perfusion to that lung region. (However, this re ex also increases pulmonary vascular resistance and thus the work of the right side of the heart.) A second re ex is hypocapnic bronchoconstriction. When the ( ) ratio increases, regional ventilation of those lung units is reduced by smooth-muscle constriction. ( ) inequalities interfere with oxygen and carbon dioxide transfer such that hypoxemia and hypercapnia may ensue. Usually the consequent hypercapnia increases chemoreceptor drive and thus increases alveolar ventilation, which restores normocapnia. However, because of the shape of the oxyhemoglobin dissociation curve, the increase in oxygen tension in the normal alveoli caused by the hyperventilation does not signi cantly increase the oxygen content of the blood coming from these units. Several factors can exacerbate the hypoxemia of ( ) inequalities, including hypoventilation (sedation) and decrements in cardiac output (mechanical ventilation with positive end-expiratory pressure). ( ) inequalities respond to oxygen therapy, although this may lead to increases in arterial carbon dioxide tension by (1) a reduction of the chemoreceptor (hypoxic) drive, (2) an increase in ( ) inequalities, and (3) a shift in the carboxyhemoglobin dissociation curve to the right, decreasing its a nity for carbon dioxide.464 Mechanical ventilation may be indicated in severe ( ) inequalities if hypercapnia is progressive. Passage of blood through abnormal cardiovascular communications (atrial septal defects, ventricular septal defects, patent ductus arteriosus) or through pulmonary capillaries within the walls of atelectatic or uid- lled alveoli causes shunts and consequent hypoxemia.

Such defects may be considered as one extreme of ( ) inequality ( 0) and are resistant to correction by oxygen therapy.464 In shunts caused by pulmonary disease, mechanical ventilation and positive end-expiratory pressure increase end-expiratory lung volume and thus the alveolar surface area available for gas exchange. The incremental increase in end-expiratory lung volume may also help redistribute the excessive extravascular lung water from the alveoli to the interstitium.

Metabolic Functions The entire cardiac output passes through the pulmonary circulation, thus providing an ideal means by which hormones and drugs may be metabolized to inactive compounds. Indeed, relative to the hepatic blood ow (approximately 25% of the cardiac output), the contribution of the lungs to the total body clearance of drugs or xenobiotics may be signi cant. The lung contains hydrolytic enzymes that cleave peptides such as bradykinin and angiotensin I, thus serving to inactivate (bradykinin) or to bioactivate (angiotensin) compounds. The enzymatic activity responsible is concentrated within the caveolae of the pulmonary capillary endothelium or in pouchings in the plasma membranes of these cells.465 Other pulmonary enzymes exist that are capable of cleaving phosphate groups from nucleotides (adenosine triphosphate, adenosine diphosphate, and adenosine monophosphate), of oxidizing steroid hormones (testosterone), of inactivating prostaglandins E and F, and of metabolizing biogenic amines such as 5-hydroxytryptamine, norepinephrine, and tyramine.450,465,466

Defense Against Infection The sterile environment of the lung results from several mechanisms, including the mucociliary apparatus, which clears particulate debris from the lung at a rate of approximately 17 mm/min,56 vagally mediated re exes, which serve to initiate coughing and concomitant bronchoconstriction, and lung surfactant proteins, SP-A and SP-D, that act as collections against a large number of pathogens.447,467 However, the predominant line of defense against noxious agents and bacteria are the alveolar macrophages and dendritic cells of the distal airways and alveoli. Recent investigations have shown depression in alveolar macrophage function after exercise or long transport.468,469 Viral infections also depress macrophage function. Helping to maintain immunosurveillance and the sterile pulmonary environment are polymorphonuclear cells that respond to chemotactic stimuli and lymphocytes.

DISORDERS OF THE LOWER RESPIRATORY TRACT Disorders of the lower respiratory tract are common in horses of all ages. Presenting signs and symptoms vary with these disorders and may include exercise intolerance, cough, nasal discharge, fever, respiratory distress, increased respiratory rate or e ort, or generalized depression, inappetence, and weight loss. Careful clinical examination that includes auscultation while rebreathing often con rms the anatomic site of the problem. Ancillary diagnostic testing such as bronchoalveolar lavage, transtracheal wash, thoracic ultrasound, or thoracic radiography are often indicated to confirm a suspected diagnosis.

Bacterial Pneumonia CAUSES Normally the lungs contain only small numbers of bacteria (colony-forming units) that are considered to be transient contaminants in the process of being removed by clearance mechanisms.470 When pulmonary defense mechanisms are overwhelmed, aspirated bacteria from the oropharynx may proliferate and cause pneumonia. The most common gram-positive bacteria involved are Streptococcus equi subsp. zooepidemicus (β-hemolytic), Staphylococcus aureus, and S. pneumoniae (α-hemolytic). The most frequent gram-negative isolates are Pasteurella and Actinobacillus spp., Escherichia coli, Klebsiella pneumoniae, and Bordetella bronchiseptica.40,41,471-474 Pseudomonas sp. is rarely a cause of equine pneumonia, and its isolation from tracheobronchial aspirates suggests environmental contamination of equipment (endoscope). Nocardia organisms also have been isolated from horses with pulmonary infections, but these are rare and appear to require signi cant derangements of host defense mechanisms.475 The anaerobic bacteria most commonly isolated are Bacteroides fragilis, Peptostreptococcus anaerobius, and Fusobacterium spp. Polymicrobic infections are not uncommon in cases of equine pneumonia and may represent a synergy between aerobic or facultatively anaerobic and anaerobic bacteria. Mechanisms of synergy may involve protection from phagocytosis and intracellular killing, production of essential growth factors, and lowering of local oxygen concentrations.472

PATHOGENESIS Bacterial pneumonia may develop after viral infections, athletic events (races), trailer rides, and general anesthesia. It occurs in horses that are overcrowded, maintained on poor nutrition, or exposed to inclement weather. Laryngeal and pharyngeal dysfunction may predispose the horse to develop pneumonia by aspiration of oropharyngeal bacteria. Such predisposition occurs in horses (1) with primary neuropathies of the ninth and tenth cranial nerves (e.g., equine protozoal myeloencephalitis, botulism, S. equi infections, guttural pouch mycoses, neoplasia), (2) with primary myopathies of the pharyngeal, laryngeal, or esophageal musculature (e.g., vitamin E and selenium de ciency,

megaesophagus), (3) after laryngeal surgery, or (4) with esophageal obstructions (choke). The pathogenic mechanisms described subsequently apply to the development of lung abscesses and pleuropneumonia. Viral-induced modi cations of respiratory tract defenses include (1) enhanced susceptibility to bacterial attachment and colonization after damage to the epithelial cells, (2) diminished mucociliary clearance and physical translocation of bacterial particles deposited on the bronchial ciliated epithelium, and (3) decreased surfactant levels and collapse of the airways because of viral destruction of alveolar type II cells.476 The ensuing anaerobic environment predisposes the area to macrophage dysfunction. In addition, the alveolar exudate that accompanies viral pneumonitis may provide a nutrient medium for multiplication of aspirated bacteria. During high-intensity exercise, the horse may aspirate track debris and oropharyngeal secretions. This, coupled with exercise-induced pulmonary hemorrhage, which provides a nutrient source, predisposes the horse to developing pneumonia.477 An exercise-associated increase in bacterial contaminants of the lower respiratory tract has been demonstrated experimentally. Strenuously exercised horses exhibit a tenfold and a 100-fold increase, respectively, in the number of aerobic and anaerobic bacteria isolated from transtracheal aspirates relative to preexercise samples.478 The bacterial contamination, along with exercise-associated increases in bronchoalveolar cortisol concentrations, decreased pulmonary alveolar macrophage viability, and impaired phagocytic function, enables bacteria to proliferate and cause pneumonia.468,469,477,479,480 Transportation remains one of the most common causes of pneumonia and pleuropneumonia in the horse because physically restraining the head of the horse and preventing postural drainage enhances bacterial colonization of the lower respiratory tract.477,481 In ammation and increased numbers of bacteria are found in the transtracheal aspirates of horses within 6 to 12 hours of con nement with the head elevated.482 Furthermore, although the stress response of horses during transport contributes to immunosuppression and enhanced susceptibility to pneumonia, the increase in serum cortisol levels are greater when horses are cross-tied compared with when left unrestricted in the box during transport.483 Another hypothesis is that dehydration, associated with reduced uid consumption before or during transportation, reduces mucociliary clearance and contributes to bacterial proliferation. Neither the prophylactic administration of antibiotics nor the intermittent release from the con ned head posture reliably reduces tracheal bacterial numbers or prevents the accumulation of purulent tracheobronchial secretions in horses confined with their heads elevated.481,482 Horses undergoing general anesthesia may develop pneumonia when endotracheal intubation introduces oropharyngeal contaminants to the lower respiratory tract. Anesthesia may compromise mucociliary clearance and cause compression atelectasis and vascular congestion that results in regional ischemia and necrosis of lung regions. This circ*mstance provides local conditions suitable for bacterial multiplication.477 In addition, excessive extension and exion of the neck (e.g., during myelography) with secondary movement of the endotracheal tube may cause tracheal erosions that impair mucociliary clearance and predispose the horse to developing pleuropneumonia.484

CLINICAL SIGNS In the early stages of bacterial pneumonia, clinical signs may not be obvious, being limited to a gurgling sound of exudates in the trachea, fever, or depression. As pneumonia progresses, horses may exhibit intermittent fever, tachypnea or respiratory distress, nasal discharge, coughing, inappetence, exercise intolerance, and weight loss. Nasal discharge is usually mucopurulent but may be hemorrhagic in some cases.472 Epistaxis may be the result of bacterial hemolysin production (Actinobacillus equuli) or ischemic necrosis of lung tissue.485 Auscultation of the thorax reveals increased harsh breath sounds dorsally, crackles, wheezes, and dullness of respiratory sounds ventrally. Manipulation of the trachea or larynx may induce a cough. Halitosis and a foul-smelling nasal discharge suggest an anaerobic infection. Mandibular lymphadenopathy may be apparent in aspiration pneumonia associated with strangles or with viral infections.

DIAGNOSIS Clinical signs and history aid in the diagnosis. Clinical pathologic data consistent with bacterial pneumonia include leukocytosis caused by absolute neutrophilia with or without a left shift. Neutropenia may be evident if gram-negative organisms are involved. Hyper brinogenemia (>500 mg/dl), hyperglobulinemia, hypoalbuminemia, and anemia of chronic in ammation are compatible with the diagnosis of chronic bacterial pneumonia. Endoscopic evaluation of the upper respiratory tract may demonstrate a defect in laryngeal or pharyngeal function if this is the inciting cause. One may observe a mucopurulent exudate with or without traces of blood by endoscopy in the lower respiratory tract. Thoracic radiographs demonstrate a radiopacity in the anteroventral thorax and a loss of clarity in the lung elds caudal to the heart. Air bronchograms are occasionally found in the adult horse with bacterial pneumonia (Figure 9-13). Ultrasound may demonstrate consolidation of ventral lung lobes and extension of the pneumonia to the pleural surfaces as evident by the presence of comet tails. Tracheobronchial aspirates yield degenerative neutrophils, damaged epithelial cells, and microorganisms. The presence of squamous epithelial cells supports a diagnosis of aspiration pneumonia, provided that the tracheal catheter was not misplaced in the pharynx during sampling. The clinician should perform aerobic and anaerobic cultures on the tracheal aspirates.

FIGURE 9-13

Two-year-old Thoroughbred with pneumonia. A patchy area of pulmonary consolidation is notable in the ventral dependent portion of the lung silhouetting the heart and diaphragm. (Courtesy D. S. Biller, Manhattan, Kan., 1991.)

TREATMENT The clinician should direct treatment at the causative organism, but in the absence of culture and sensitivity results, one should administer broad-spectrum antimicrobials. Appropriate therapy might include intravenous aqueous sodium or potassium penicillin and an aminoglycoside or third-generation cephalosporin (Table 9-5). The aminoglycosides are e cacious against most gram-negative aerobes, but they lack e cacy against anaerobes. However, metronidazole is e ective against most anaerobes, including the penicillin-resistant B. fragilis and is routinely included in the treatment protocol. Depending on the culture and sensitivity results, one may also administer potentiated sulfonamides. The clinician should use aminoglycosides judiciously in animals that have renal compromise or are dehydrated. (See the section on treatment for parapneumonic e usions.) Antimicrobial therapy may also be administered by nebulization. In general, one third of the calculated systemic dose of the antimicrobial (ceftiofur, gentamicin, marbo oxacin) is administered via an ultrasonic nebulizer two or three times daily.486,487 With the nebulizate, one may also add 5 mg albuterol to enhance bronchodilation and 10 ml 20% acetylcysteine. TABLE 9-5 Antimicrobials and Other Drugs Antimicrobial Amikacin Ampicillin sodium Ceftiofur Chloramphenicol Doxycycline Enrofloxacin

Gentamicin Metronidazole Oxytetracycline Potassium penicillin G Procaine penicillin G Rifampin Sodium penicillin G Trimethoprim-sulfonamide Other Agents Used Acetylcysteine Albuterol Clenbuterol Fenoterol

Dose 16-18 mg/kg, IV, q 24 h 11-22 mg/kg, IV or IM, q 6-8 h 2.2 mg/kg, IV or IM, q 12 h 50 mg/kg, PO, q 6 h 10 mg/kg, PO, q 12 h 7.5 mg/kg, PO or IV, q 24 h 5.0 mg/kg, PO or IV, q 12 h 4.4-6.6 mg/kg, IV or IM, q 24 h 15-25 mg/kg, PO, q 6-8 h 6.6 mg/kg, IV, q 12-24 h 22,000 IU/kg, IV q 6 h 22,000 IU/kg, IM, q 12 h 5-10 mg/kg, PO, q 12 h 22,000 IU/kg, IV, q 6 h 15-20 mg/kg, PO or IV, q 12 h 10-20 ml 20% solution (nebulized) 360-720 μg via inhalation 0.7-0.8 μg/kg, PO, q 12 h or nebulized 1-2 mg via inhalation

Ipratropium bromide Salmeterol Fluticasone

2-3 μg/kg (nebulized) 0.5-1 μg/kg via inhalation 2-4 μg/kg via inhalation

IM, Intramuscularly; IV, intravenously; PO, orally; q, every.

With gram negative infections and the potential for endotoxemia, small doses of unixin meglumine (0.25 mg/kg three times daily) may be given to inhibit arachidonic acid metabolism. Other treatment modalities that have been advocated include prophylactic measures against laminitis. The goal of supportive care should be to minimize stress and ensure adequate ventilation and hydration. Ideally, horses should be bedded on paper or on other materials free of dusts or molds and should be fed forages of excellent quality. One should also direct attention to correcting the primary cause of the pneumonia. Depending on the chronicity of the pneumonia, one should note clinical improvement in 48 to 72 hours. The prognosis can be excellent if the pneumonia is treated aggressively, but the clinician should forewarn the owner of potential complications (see the following discussion). The clinician should administer treatment for 2 to 6 weeks, depending on the extent of the pneumonia and the underlying inciting cause. (See the section on treatment for parapneumonic effusions.) Preventive measures that help deter the development of pneumonia include (1) adequate immunization protocols with vaccination against EIV, EHV1, and EHV4 every 4 to 6 months (in the performance horse); (2) the minimization of stressors such as long van rides in which the head is restrained constantly; and (3) the use of management or husbandry methods that minimize dust or noxious gas accumulations within the stall, prevent exposure to inclement weather, and provide adequate nutrition for the horse.

Lung Abscesses CAUSES In foals younger than 6 months, R. equi and S. equi subsp. zooepidemicus are the most common bacterial isolates recovered.488 In adult horses, Streptococcus and Actinobacillus species are the most common gram-positive and gram-negative organisms implicated in the development of pulmonary abscesses.488-490 This discussion addresses predominantly lung abscesses in the adult horse.

PATHOGENESIS Lung abscesses may develop as a consequence of inhaling bacterial organisms resident to the oropharynx or contaminating the environment (R. equi). The abscess may develop as a consequence of a focal pneumonia or may be a component of pleuropneumonia complex. (See discussion elsewhere in this chapter.) Racehorses may be at an increased risk for developing lung abscesses when bacteria proliferate in the blood in the airways and alveoli after episodes of EIPH. The regional location of lung abscesses in these cases supports this hypothesis.490

CLINICAL SIGNS Clinical signs may vary, but most horses have periods of pyrexia, inappetence, lethargy, mild tachycardia, and respiratory-related signs ranging from tachypnea to distress.489,490 Other signs that may be evident include cough, purulent nasal discharge, thoracic pain, epistaxis, and halitosis.

DIAGNOSIS The adult horse may have a history of transport, strenuous exercise, surgery, prior pneumonia, or administration of medications. In most cases, evidence of an infection is found on the CBC and serum chemistry panel: mature neutrophilia, hyper brinogenemia, hyperglobulinemia, and anemia. Radiography and ultrasound are useful imaging modalities, and one may use them concurrently to detect lung abscesses (Figures 9-14 and 9-15). Pulmonary abscesses appear on ultrasonography as encapsulated cavitated areas lled with uid or echogenic (white) material. Cardiac structures or an air- lled lung may overlie an abscess, obscuring its detection. In one study, two thirds of the abscesses were located in the caudodorsal lung field, whereas the remaining abscesses were located in the caudoventral region.490

FIGURE 9-14

Nine-year-old saddle horse. Multiple cavitating pulmonary abscesses are present within the lungs. Arrows indicate air-fluid interfaces within the abscesses.

(Courtesy D. S. Biller, Manhattan, Kan., 1991.)

FIGURE 9-15

A large air-capped pulmonary abscess. Small arrows indicate an abscess; large arrow indicates the air-fluid interface.

(Courtesy D. S. Biller, Manhattan, Kan., 1991.)

Transtracheal aspirates or percutaneous aspirates of lung abscesses adhered to the body wall may help the clinician decide the choice of antimicrobials. One should culture isolates under aerobic and anaerobic conditions.

TREATMENT Long-term antimicrobial therapy (8 to 10 weeks) along with prolonged periods of rest (5 to 6 months) before resuming strenuous exercise is recommended. (See the section on treatment for pneumonia and parapneumonic e usions for speci c antimicrobial recommendations.) In human beings receiving antimicrobial treatment for pulmonary abscesses, an 80% reduction in the size of abscess cavities less than 3 cm in diameter usually occurs within 1 month of appropriate therapy.491 Furthermore, 70% of cavitary lesions in human cases of lung abscessation resolve completely by 3 months.492 For lung abscesses unresponsive to therapy, the clinician should consider drainage via a thoracotomy or thoracoscopy.493 The prognosis for resumption of athleticism is good if treatment is initiated early. In one study, 23 of 25 Standardbreds and 13 of 20 Thoroughbreds raced after the diagnosis and treatment of a lung abscess with no significant effect on race performance.490

Parapneumonic Effusions and Septic Pleuritis CAUSES Between two thirds and three quarters of cases of septic pleuritis arise as an extension of pneumonia or pulmonary abscessation.494-496 Septic pleuritis also may occur in horses with thoracic trauma, esophageal rupture, or penetration of the esophagus or stomach by a foreign body.477,497-499 The aerobic or facultatively anaerobic organisms most often isolated from horses with pleuropneumonia are bacterial species

that normally reside in the oropharyngeal cavities: Streptococcus spp., Pasteurella and Actinobacillus spp., E. coli, and Enterobacter spp. Anaerobic organisms frequently isolated include Bacteroides spp., Peptostreptococcus spp., Fusobacterium spp., and Clostridium spp.499-501

EPIDEMIOLOGY Risk factors for the development of pleuropneumonia are the same as those associated with pneumonia (see the previous discussion) and include long-distance transport, strenuous exercise, viral respiratory tract disease, surgery, dysphagia, general anesthesia, and systemic illness (enteritis). These conditions may enhance aspiration of oropharyngeal organisms or impair clearance of such organisms.477,495,502

PATHOGENESIS The causative factors of equine pleuropneumonia are those that suppress the pulmonary defense mechanisms and allow bacterial contamination of the lower respiratory tract to progress to pneumonia or abscess formation. The subsequent extension of the infectious process into the pleural space causes pleuritis. The distribution of the pulmonary lesions—cranioventral with the right cranial and middle lung lobes more severely a icted—is consistent with inhalation or aspiration of bacteria rather than infection from a hematogenous spread. Fluid accumulates within the pleural cavity as the parenchymal in ammation increases the permeability of the capillaries in the overlying visceral pleura, causing an outpouring of protein and cells. Bacteria may also invade the pleural uid. (See previous discussion of pneumonia.)

CLINICAL SIGNS Depending on the chronicity of disease, clinical signs may vary and may be confused with signs of colic or rhabdomyolysis. In the acute stages, horses are febrile and lethargic, have a slight nasal discharge, and exhibit a guarded cough, shallow breathing pattern, and painful, stilted gait.501 The right hemithorax is a ected more often than the left, presumably because of the more direct route of the right mainstem bronchus.477 Thoracic auscultation may be abnormal, as evidenced by pleural friction rubs and ventral dullness. In severe acute cases, the horse may exhibit nostril aring; tachycardia; jugular pulsations; toxic mucous membranes; a guarded soft, moist cough; and a serosanguineous fetid nasal discharge. In chronic cases of pleuropneumonia (duration longer than 2 weeks), horses may have bouts of intermittent fever and exhibit weight loss and substernal and limb edema.

DIAGNOSIS The diagnosis is based on historical information, clinical examination, imaging results, and microbiologic and cytologic analysis of tracheal and pleural uid aspirates. On auscultation of the thorax, one may hear vesicular sounds only dorsally, with an absence of lung sounds ventrally. One may hear bronchial or tracheal sounds if lung consolidation exists. Cardiac sounds radiate over a wider area of the lung eld than normal, a nding distinct from clinical cases of pericarditis. On percussion of the chest, the clinician may elicit a painful response (pleurodynia) and detect an area of dullness or decreased resonance ventrally. Laboratory assessment may reveal a normal or toxemic leukogram and chemistry ndings in acute cases. In chronic cases, one may nd anemia, neutrophilia, hyperfibrinogenemia, and hyperproteinemia. Ultrasound is the diagnostic technique of choice in horses with suspected pleuropneumonia or pleural e usion. Using a 3.5- to 5-MHz transducer (sector scanner or linear probe), one can detect free or loculated uid, pleural thickening, pulmonary and mediastinal abscesses, pulmonary consolidation, inundation of airways with uid, brinous adhesions, and concurrent pericarditis. Pleural uid may displace the lungs axially and dorsally. The uid may appear anechoic or hypoechoic, depending on the relative cellularity.501 Free gas echoes within the pleural uid may re ect (1) the presence of anaerobic organisms,48 (2) the presence of air introduced during a thoracocentesis, or (3) the presence of air introduced by a bronchopleural stula. Ultrasonography enables accurate placement of the catheter during thoracocentesis and ensures productive yields during placement of a chest drain (Figures 9-16 to 9-18). Ultrasound examination fails to demonstrate deep parenchymal lesions if the overlying lung is normally aerated.

FIGURE 9-16

Ultrasound of pleural e usion. Anechoic (black) area represents pleural e usion (PE). The echogenic (white) tortuous brin strand is attached to the parietal pleura of the diaphragm. V, Ventral; D, dorsal; arrow, diaphragm. (Courtesy D. S. Biller, Manhattan, Kan., 1991.)

FIGURE 9-17

Normal thoracic ultrasound demonstrates a highly echogenic interface (arrows) between the chest wall and normal lung. A reverberation artifact is notable deep to the chest wall–lung interface. (Courtesy D. S. Biller, Manhattan, Kan., 1991.)

FIGURE 9-18

Ultrasound of ventrally consolidated lung (CL) surrounded by a small amount of pleural effusion. V, Ventral; D, dorsal.

(Courtesy D. S. Biller, Manhattan, Kan., 1991.)

Thoracic radiography also remains a useful technique for evaluating horses with pleuropneumonia, permitting detection of a pneumothorax or an abscess located deep in aerated lung tissue. Pneumothorax may develop after transtracheal aspiration, thoracocentesis, thoracic drainage, or pleuroscopy. Pneumothorax may also develop as a sequela to gas-producing organisms in the pleural cavity or to air leaks from bronchopleural stulae.477 Although radiographs may detect pulmonary abscesses, they may fail to detect lesions obscured by the cardiac silhouette. Thoracocentesis of both sides of the chest is indicated if one notes uid bilaterally. Not only is thoracocentesis diagnostic and prognostic, it may also be a therapeutic life-saving procedure in horses with severe respiratory distress.503 In healthy horses, the fenestrated caudal mediastinum allows communication of the uid between the two sides of the chest. In horses with pleuropneumonia, brin may close

mediastinal fenestrations, allowing for di erences to develop in the pleural uid of the two hemithoraces. The clinician should submit uid for cytologic examination and for aerobic and anaerobic culture. Pleural uid in healthy horses contains fewer than 10,000 cells/ l, approximately 60% of which are neutrophils, and has a total protein of less than 2.5 g/dl. In cases of pleuropneumonia, pleural uid white blood cell counts and protein are elevated and glucose levels may be low (25% of the total nucleated cell count). In the BALF, the percentage of neutrophils may be as great as 90%, far exceeding the normal 5% to 15% found in healthy horses that are stabled and fed hay. In most a ected horses, the percentages of eosinophils or mast cells are not increased. Albumin and immunoglobulin levels within the BAL may not increase.719 Gram positive organisms may be retrieved by tracheobronchial aspirates if a concurrent septic inflammation exists, but this is rare. Pollen, fungal hyphae, and Curschmann’s spirals (inspissated mucus plugs) also may be visible. Thoracic radiography may demonstrate an increase in the interstitial and bronchial pattern throughout the lung elds. However, these changes may be di cult to interpret relative to the normal aging changes that occur. One may also detect exudate within the trachea by thoracic radiography (Figure 9-20). Radiographs are useful in ruling out thoracic neoplastic or infectious disorders in horses that have clinical signs resembling RAO. In severely a ected horses with RAO that repeatedly fail to respond to bronchodilators and glucocorticoids, radiographs should be obtained to rule out the existence of bronchiectasis (dilation or deformation of the bronchi or bronchioles).720 This relatively uncommon complication of RAO causes irreversible airway obstruction as a result of collapse of the a ected airways during expiration.

FIGURE 9-20

Tracheal fluid or exudate (arrow) is visible ventrally in the trachea at the level of the thoracic inlet.

(Courtesy D. S. Biller, Manhattan, Kan., 1991.)

Ultrasound evaluation of RAO-a ected horses con rms a shift of the caudal lung border caused by alveolar hyperin ation. However, gas in the large colon (left side) or in the cecum (right side) may preclude complete imaging of the most caudal extent of the lung elds at the sixteenth intercostal space. On ultrasound examination, small hypoechoic areas of irregularity on the lung surface may be detected, but, in general, this imaging modality does not detect airway inflammation.721 Echocardiographic ndings of RAO-a ected horses demonstrate an increase in the pulmonary artery diameter, an increase in the ratio of the diameters of the pulmonary artery to the aorta, and abnormalities in ventricular septal motion.717 However, no changes in serum troponin I concentrations occur during experimentally induced RAO.717 Intradermal skin testing has been occasionally advocated for diagnosing RAO and for detecting sensitizing allergens. Theoretically, RAOa ected horses should react if sensitized to a given allergen, but the results supporting this theory have been inconclusive. For example, in some studies, clinically normal horses have positive skin test reactions to the allergens, whereas RAO-a ected horses lack positive reactions.722-724 In other studies, the percentage of RAO-a ected horses exhibiting positive skin reactions (to any antigen) exceeds that of controls725 or the response of RAO-prone horses to histamine challenge (measured at 0.5 hours after injection) or to Aspergillus spp. antigens (measured at 24 hours after injection) is enhanced.726 Apparently, positive skin tests and serum precipitins to fungal and thermophilic actinomycete antigens are found in many normal and in RAO-a ected horses and probably re ect a level of exposure of the horse rather than a susceptibility to RAO. In general, intradermal skin testing does not allow one to identify RAO-a ected horses or the speci c inhaled particulate that is inciting the inflammation. A CBC and serum chemistry panel are of limited usefulness in diagnosing RAO because no abnormalities are detectable in the leukogram or in serum biochemistries.684 These tests may be used to rule out the existence of a pneumonia or a neoplastic process. Arterial blood gas analyses are normal in approximately 25% of the horses with RAO, but in others, the development of arterial hypoxemia (Pao2 < 85 mm Hg) in the absence of arterial hypercapnia is noted.727

PATHOLOGIC SIGNS Postmortem examination reveals a di use bronchiolitis with goblet cell metaplasia, airway smooth-muscle hypertrophy and hyperplasia, excess mucus, and inflammatory cells in the small airways.728,729 Eosinophilic infiltration around the bronchioles is a variable finding present in approximately one third of a ected horses. Varying degrees of alveolar overin ation and atelectases also are found, as well as evidence of brosis of the alveolar septa.730 The larger airways—the bronchi and trachea—may show variable changes, including epithelial hyperplasia with loss of ciliated cells, goblet cell hyperplasia, and inflammatory cell infiltration.731

TREATMENT Elimination of the source of mold or dust is not only the most bene cial step in the treatment regimen but also the most di cult management change to implement and achieve long term.684 Horses should be switched to a diet consisting of pelleted feeds or hay cubes, making the dietary transition over 3 to 5 days. Some horses will tolerate hay that has been soaked in water, but this practice reduces the nutrient value of the hay and is often forsaken when the weather becomes cold. Unless pasture exacerbates the condition, the owner should keep horses outside, blanket them in the winter, and allow access to a three-walled shelter a ording protection from the wind and rain. For chronically a ected horses that have been turned out to pasture, clinical remission and normalization of conventional pulmonary function tests takes 4 to 8 weeks, depending on weather conditions. Recent data suggest that when RAO-susceptible horses are maintained in a lowdust environment (outside, no hay) for a prolonged period (years), they still exhibit a reduction in forced expiratory ow despite the absence of clinical signs. This finding suggests the existence of irreversible airway remodeling in affected animals.732 If the horse must be stabled, tall bedding should be shredded paper or shavings to reduce inhaled dusts and molds. Surrounding stalls should be bedded similarly. Although such stabling changes improve lung function somewhat, airway hyperreactivity still remains.733,734 In a recent study of 12 RAO-a ected horses that were stabled and fed a pelleted ration, pulmonary function tests and BALF neutrophil percentages improved within 4 weeks of implementing environmental changes. However, in four horses, pulmonary health (as assessed by BALF neutrophil percentages, maximal pleural pressure changes, and in ammatory cytokine pro les of the BALF cells) had not returned to normal by the conclusion of the study.735 Medical management is recommended in horses with clinical signs of RAO because environmental change alone takes several weeks to quiet the pulmonary in ammation. Corticosteroids are e cacious in reducing the in ammatory reaction in the lungs, whereas nonsteroidal antiin ammatory drugs provide no bene t. One potential regimen is to give oral prednisolone at 2.2 mg/kg once daily in the morning for 7 to 10 days, 1.1 mg/kg once daily for 7 to 10 days, 0.50 mg/kg once daily for 7 to 10 days, and 0.50 mg/lb every other day for 7 to 10 days. Alternatively, one may institute a course of parenteral or oral dexamethasone. For the 500-kg horse, the course is 40 to 50 mg intramuscularly once daily every other day for three treatments followed by 35 to 45 mg intramuscularly once daily every other day for three

treatments, 30 to 40 mg intramuscularly once daily every other day for three treatments, and so forth until the horse is weaned o of the dexamethasone. A similar dosing schedule may be used when the parenteral dexamethasone formulation is given orally except that the initial dosage is 0.165 mg/kg for three treatments, followed by 0.084 mg/kg for three to four treatments, and 0.042 mg/kg for three to four treatments. In horses that remain stabled, are fed a pelleted feed, and receive oral dexamethasone protocol, a normalization of pulmonary function and a signi cant reduction in airway epithelial and luminal cell in ammatory gene expression (IL-8, IL-1β) is seen by 4 weeks after implementing the regimen.735 Other parenteral glucocorticoids that have been evaluated in RAO-a ected horses maintained in dusty environments and fed hay include triamcinolone acetonide (0.09 mg/kg intramuscularly once) and iso upredone (0.03 mg/kg intramuscularly once daily for several days).736,737 The latter, because of its mineralocorticoid e ects, can cause hypokalemia; therefore serum electrolytes should be supplemented and monitored. Corticosteroid use may be contraindicated in horses predisposed to laminitis or exhibiting endocrinopathies or those that have gastritis or gastric ulceration. Inhaled steroids have been recommended to treat the pulmonary in ammation of RAO but may not provide a therapeutic bene t for 24 to 72 hours. For horses in respiratory distress, intravenous dexamethasone can provide an improvement within 4 to 6 hours. This application may be followed by treatment with inhaled steroids. A metered-dose inhaler containing uticasone can be attached to a spacer on the Aeromask (Trudell Medical, London, Ontario, Canada) or to another drug delivery system (Equinehaler, Equine Health Care, Inc.) and administered to the horse during inspiration. The recommended dose is 2 mg twice daily. In RAO-a ected horses that were still fed hay, uticasone administration was associated with a marked improvement in pulmonary function, a normalization of BALF neutrophil counts, and a reduction in IL-8 copy numbers in BALF cells.700 Bronchodilators also are recommended for treating RAO. Three types of compounds are used to relax airway smooth muscle: βadrenergics, methylxanthines, and anticholinergics. β 2-Adrenergic compounds include clenbuterol, albuterol, salmeterol, fenoterol, and trimetoquinol. Experimental investigations with clenbuterol suggest that when given at a dosage of 0.8 g/kg once daily by mouth, the drug may be e cacious in alleviating some of the signs of RAO. In addition to a direct smooth-muscle e ect, clenbuterol may stabilize mast cells, increase mucociliary clearance, improve airway secretions, and decrease in ammatory cytokine production.738,739 If e ects are not obtained at 0.8 g/kg orally, then the dose may be increased step-wise by 25% increments. Adverse e ects (tachycardia, tachypnea, sweating) are signs of toxicity. Albuterol is absorbed poorly from the gastrointestinal tract; thus inhalation is the recommended method of administration. Salmeterol is purported to be a long-acting β 2-agonist. However, in one study in RAO-a ected horses, the investigators found that signi cant improvements in pulmonary function lasted for only 6 hours after its administration.740 Aerosolized trimetoquinol (1000 g administered via an ultrasonic nebulizer) signi cantly improved pulmonary function tests that lasted for 4 hours.741 Some suggested doses of β 2-agonists that can be administered by metered-dose inhalers through the Aeromask are these: albuterol, 720 g three to four times daily; fenoterol, 1 to 2 g three to four times daily; and salmeterol, 210 g three to four times daily. Administration of the β2-agonist before corticosteroid dosing enhances the deposition of the latter in the smaller airways.742 Long-term use of β 2-agonists has been associated with a down-regulation of βreceptors in humans, but whether this occurs in horses is unknown. The methylxanthines include aminophylline (theophylline) and pentoxifylline, which dilate smooth muscle by increasing cyclic adenosine monophosphate levels intracellularly. Cyclic adenosine monophosphate also inhibits degranulation of mast cells and subsequent mediator release. One may give aminophylline at 4 to 6 mg/kg orally three times daily. However, in a recent study, aminophylline at 18 mg/kg intravenously provided clinical improvement in only 50% of horses.743 Pentoxifylline, another methylxanthine derivative and phosphodiesterase inhibitor, is well absorbed orally and improves pulmonary function tests in RAO-a ected horses when administered at 16 g twice daily by mouth.744 It does not reduce BALF neutrophilia. The third class of compounds used to a ect smooth-muscle dilation is the anticholinergics, of which atropine, glycopyrrolate, and ipratropium are representatives. Atropine provides clinical improvement when administered intravenously at 0.02 mg/kg, but the duration of the e ect is short lived (2 hours), and atropine may be associated with the development of ileus, abdominal pain, tachycardia, mydriasis, and thickening of airway secretions.743 Atropine is generally used for emergency relief of airway obstruction. Glycopyrrolate is e cacious at a dose of 0.007 mg/kg, but it may also cause colic. One can also administer ipratropium bromide by inhalation at a dose of 180 to 360 mg three times daily with a low risk of inducing systemic side effects.679 The powder form may be nebulized (dose of 2 to 3 μg/kg). Disodium cromoglycate acts to stabilize mast cell degranulation and inhibit the vagal e erent component of histamine response. A suggested dose688 is 80 mg once daily for 4 days administered by nebulization. Mucolytics (acetylcysteine, dembrexine) and muco*kinetics (iodides, bromhexine) may provide some relief.745 Antimicrobials are indicated if microorganisms are isolated on the tracheobronchial aspirate.

OTHER THERAPIES In horses that develop respiratory distress caused by bronchospasm and airway in ammation, additional therapies are required.

Administering nasal oxygen at ow rates as low as 5 L/min improves arterial oxygen tensions by as much as 30 mm Hg in some severely a ected horses.746 As the total nasal oxygen ow increases, so too will the mean arterial oxygen tension, although not to the same degree that occurs in healthy horses. Flow rates of 30 L/min (delivered by two nasal cannulae at 15 L/min) are associated with coughing and gagging in the horse. Nasal oxygen supplementation does not reduce breathing frequency, suggesting that stimulation of vagally mediated afferents—perhaps responding to inflammatory mediators—is responsible for the tachypnea. Furosemide (1.0 mg/kg intravenously or nebulized) provides bene cial e ects to RAO-a ected horses within 20 minutes of its administration, decreasing total lung resistance (RL) and increasing Cdyn without affecting Pao2.747 Bene cial e ects appear to be mediated by prostaglandin E2, derived from either the renal or airway epithelium, that promotes smooth muscle relaxation. Prior treatment with the cyclooxygenase inhibitor flunixin meglumine prevents furosemide-induced bronchoconstriction.748 Therapies that are not e cacious in the treatment of RAO-a ected horses include a single acupuncture treatment,749 the oral administration of an herbal preparation containing thyme and Primula,750 or the rapid intravenous administration of 30 L of isotonic saline.751

PROGNOSIS RAO-prone horses are likely to develop acute exacerbations of the disease when husbandry practices lapse or when weather conditions become adverse. Horses do not outgrow the disorder, and, in fact, some clinicians believe that the in ammation becomes more di cult to manage as the horse ages. The basis for the lack of clinical response is unknown but might be the result of lung remodeling—development of bromuscular hyperplasia in the alveolar septae,716 excessive smooth-muscle hyperplasia,729 bronchiectasis,720 or potential down-regulation of the glucocorticoid receptor. If proper environmental management changes are implemented, and if the appropriate therapy is initiated, clinical remission occurs in most cases. In a survey of RAO-a ected horses that had been examined and treated at a referral center 2 to 4 years previously, 20% of the original cases (3 of 15) had been euthanized, 33% were still receiving bronchodilators on an as-needed basis, and athleticism had not decreased in 92% of the horses.752 In a similar survey,684 13% of horses diagnosed with RAO 3 to 4 years earlier had been euthanized. However, more than 50% of the respondents in that survey stated that the athletic performance of the horse had been compromised by the disease. Perhaps this di erence between the two surveys simply re ects a failure of the owners to comply with environmental management recommendations. When queried about the husbandry practices, 77% of the respondents still stabled the horse for part of the day, and 84% still fed dry hay. Because the e ects of repetitive episodes on the development of irreversible changes in lung structure have yet to be determined, clients should be encouraged to implement effective husbandry changes that minimize the recurrences of RAO.

PREVENTION Once the clinician has diagnosed RAO, the horse will always be susceptible to recurrences of disease. Husbandry changes should be implemented to decrease environmental dust and organic mold exposure. During periods of hot, humid, or dusty conditions, prophylactic administration of a steroid and bronchodilator via the metered-dose inhaler may be indicated. Routine immunizations against the viral respiratory pathogens and good management practices are logical suggestions.

Summer Pasture-Associated Obstructive Pulmonary Disease DEFINITION In the southeastern United States, a condition clinically similar to RAO, called summer pasture-associated obstructive pulmonary disease (SPAOPD), develops in mature horses that are maintained at pasture. Clinical signs typically show during late spring, summer, and early autumn.683 This disorder has also been reported to occur in horses in the United Kingdom that are out on pasture.753,754 Horses with SPAOPD exhibit airway neutrophilia, excessive mucus production, and accentuated breathing e orts,755 although the inciting environmental factors and the immunopathogenesis of this disorder have yet to be determined. Because of its similarity to RAO, investigators have sought to determine if SPAOD re ects a TH2 immune response to inhaled molds or antigens by examining BALF antibody isotypes and BAL cell cytokine pro les. Neither the total nor the antigen-speci c IgE and IgG titers to Aspergillus spp., Cladosporium herbarum, Penicillium spp., F. rectivirgula, Saccharomonospora viridis, Thermoactinomyces thalpophilus, and the forage mite Lepidoglyphus destructor in tracheal lavage uid are increased in SPAOD-a ected horses relative to controls.756 During the winter months, when horses are free of clinical signs, A. fumigatus–speci c IgG titers in SPAOPD-a ected horses exceed those of controls, but whether this increase re ects ongoing production or decreased consumption cannot be determined. Cytokine pro les of BALF cells isolated from SPAOPD-susceptible and -a ected horses have also been examined. When horses are symptomatic, the mRNA copy numbers of both IL-4 and interferon-gamma in BALF cells are increased relative to controls.757 This di erence is not apparent when horses are

asymptomatic, although the copy number of IL-4 in the SPAOPD-prone horses does not decrease during the asymptomatic periods either.

TREATMENT Recommendations for treatment of RAO-a ected horses include environmental management changes such as moving the horse from the pasture to a clean stable with wood shavings for bedding. Dusty and hot environments should be avoided because these conditions exacerbate SPAOD, although this task may be di cult to achieve in the southeastern United States.758 As with RAO, corticosteroids and bronchodilators may be administered to reduce inflammation, airway hyperreactivity, and bronchoconstriction.

Inflammatory Airway Disease DEFINITION In a recent ACVIM Consensus Statement, the following criteria were listed to de ne the IAD phenotype: (1) poor performance, exercise intolerance or coughing, with or without excess tracheal mucus; (2) nonseptic in ammation detected by cytologic examination of BALF; and (3) pulmonary dysfunction based on evidence of lower airway obstruction, airway hyperresponsiveness, or impaired blood gas exchange at rest or during exercise.759 Excluded from this group are horses with evidence of systemic infection (fever, depression, inappetence) and those with increased respiratory e orts at rest (RAO, SPAOPD). In de ning the syndrome, the panel stated that airway in ammation should be determined by assessing inflammatory cell percentages in the BALF as opposed to tracheal washes.

EPIDEMIOLOGY Although IAD is encountered predominantly in young performance horses,65,144 those of any age may be a ected.760 A prevalence of 20% to 80% among racehorses in training has been reported.144,761 The variation in prevalence may reflect di erences in methods used. The amount of tracheal exudates detected endoscopically increases as a function of exercise intensity,144 thus one may miss the diagnosis of IAD if one performs endoscopic examination in the resting horse.762 A second factor contributing to the range in prevalence rates may be the age of the horse evaluated. In an epidemiologic survey of young racehorses in Britain, the prevalence rate of IAD decreased with age: In 4-year-old horses, it was 2% compared with 9% in the 2-year-old horses.763 The relationship between IAD in young horses and RAO in mature horses is unknown. Current thinking holds that IAD progresses to RAO.679 In nonracehorses or in sport horses, IAD is usually diagnosed at an older age than in racehorses.764

ETIOPATHOGENESIS The de nitive cause of lower airway in ammation observed in horses with IAD is currently unknown. It may be multifactorial, with the relative importance of a causative agent dependent on the age of the horse, its environment, and its athletic use. IAD may re ect the response of the lung to the presence of one or all of the following factors: (1) low-grade persistent bacterial or viral infections; (2) autologous blood from exercise-induced pulmonary hemorrhage; and (3) inhaled stable dusts, molds, endotoxin, particulate matter, or environmental pollutants such as hydrogen sulfide, ammonia, ozone, sulfur dioxide, nitrogen oxide, and carbon monoxide. Studies in the United Kingdom and Australia suggest that bacterial agents may be involved in the pathogenesis of IAD.144,765-770 Using an in ammatory score based on the amount of mucus, total nucleated cell count, and neutrophil percentages in endoscopically obtained tracheobronchial aspirates, investigators found that as the in ammatory score increased, the percentage of positive bacterial cultures also increased.768 The most commonly isolated organisms were S. equi subsp. zooepidemicus, S. pneumoniae, A. equuli, and Pasteurella spp. Mycoplasma spp. were also isolated from some horses with IAD.617,763 That the prevalence of IAD decreased with increasing age of the horse was attributed to the development of an effective immune response. In studies in the United Kingdom and in Australia, no association between the onset of IAD and seroconversion to EHV1, ERAV, ERBV1, or equine influenza existed,765,767,769 suggesting that IAD is not associated with viral infection. Nevertheless, a viral cause had been proposed for IAD based on the cytologic improvements in BALF constituents in horses treated with oral interferon-α. In a study of 32 Standardbred horses with a history of poor performance referable to the respiratory tract, the BALF total nucleated cell count and the percentages of BALF neutrophils, lymphocytes, and monocytes in horses with IAD were signi cantly increased as compared with control horses.771-773 Lymphocytosis and monocytosis were consistent with a low-grade viral infection. A 5-day course of oral interferon-α (50 U/day) lowered total nucleated cell counts and converted the cell distribution to a nonin ammatory pro le for at least 15 days. The investigators did not describe whether the amelioration of pulmonary in ammation was associated with an improvement in athletic performance. They also did not evaluate horses endoscopically beyond 2 weeks to determine if cessation of interferon-α therapy was associated with a return of IAD. Autologous blood, derived from rupture of pulmonary capillaries during intense exercise, has also been suggested as a cause of IAD. Instillation of blood causes a neutrophilic pulmonary in ammatory reaction in the lungs, suggesting that pulmonary hemorrhage may contribute to IAD.774 Inoculation of blood into the bronchial tree of retired racehorses induced an in ammatory response 8 and 14 days later

characterized by increased macrophage activity, alveolar septal thickness, and collagen content.775 Similarly, in horses (2 to 4 years of age), a single intrabronchial instillate of autologous blood was associated with increased numbers of alveolar macrophages by 24 hours, 48 hours, and 7 days after blood inoculation. In contrast, by day 14, no evidence of alveolar fibrosis was found in lung tissues.776 Inhalation of stable dusts, environmental particulates, mites, endotoxin, and fungal wall components may also cause or contribute to IAD. The risk of IAD increases in horses that are stabled.777,778 Episodes of IAD are shorter in horses bedded on shredded paper.765 Environmental pollutants also have been suggested as a cause of IAD. Many racetracks and training facilities are located in metropolitan areas with signi cant accumulations of smog. No data exist verifying this agent as a cause of IAD, although Mills et al. (1996) found that when healthy horses were exposed to ozone, indices of oxidant stress in the pulmonary epithelial lining uid increased (e.g., the amounts of reduced glutathione or glutathione disul de, the glutathione redox ratio, the total iron concentrations), which were similar to those found in horses with poor performance. However, these changes can also occur in pulmonary disease not induced by ozone. Another environmental factor that may contribute to development of IAD is the temperature of inspired air. In experimental studies of horses performing submaximal exercise, chilling of the inspired air to –5° C was associated with a modest increase in BALF neutrophil percentages (16% versus 11% in controls) measured 24 hours after the exercise cessation. Signi cant increases were found in the expression of IL-1, IL-5, IL-6, IL-8, and IL-10 in BALF relative to the control horses.779 Finally, Hare and Viel63 have also suggested that a type I (IgE-mediated) hypersensitivity reaction to inhaled environmental allergens could be the cause of IAD. They studied young Standardbred racehorses with a history of poor performance and found evidence of peripheral blood and BALF eosinophilia (12% eosinophils). Normally, the percentage of eosinophils in BALF is less than 2%. Because they found no evidence of pulmonary or intestinal parasitism in this group of a ected horses, the investigators attributed the eosinophilic IAD to an allergic reaction.

CLINICAL SIGNS In ammatory airway disease is frequently subclinical but in some cases may be associated with poor athletic performance, a chronic intermittent cough, and increased mucoid airway secretions. Serous to mucopurulent nasal discharge is commonly observed in young horses.765 Thoracic auscultation is usually unremarkable. A ected horses may occasionally have increased respiratory rates and enhanced abdominal breathing. However, if maximal pleural pressure changes are measured, they are within the normal reference range (2%) and eosinophils (>1%) in the BALF are increased. Although it has been recommended that the diagnosis of IAD be based on BALF cytologic ndings, a recent study demonstrated that this approach alone would miss the diagnosis of IAD in approximately one third of the a ected horses.781 These investigators suggested that the diagnosis of airway in ammation should be based on tracheal (>20% neutrophils) or BALF (>5% neutrophils) cytologic evaluations obtained 1 to 2 hours after the horse has performed a standardized treadmill exercise test. The exercise test is performed for two reasons: (1) the exercise hyperpnea enhances endoscopic and cytological evidence of IAD,762,782 and (2) the exercise test eliminates (or identi es) other causes (musculoskeletal, cardiovascular) of poor performance. During the standardized treadmill exercise test, one should assess the upper respiratory tract with videoendoscopy; the cardiovascular system with electrocardiogram, post-exercise echocardiogram, and pre- and postexercise serum troponin concentrations; and the musculoskeletal system with lameness examination and pre- and postexercise serum creatine kinase and aspartate transaminase concentrations. During the treadmill exercise testing, arterial blood gases should also be obtained for analysis. In exercising horses, the Pao2 of select horses with IAD alone or in combination with EIPH was significantly less than in healthy control horses.783 Thoracic radiographs are a low-yield diagnostic modality in IAD cases because no correlation exists among radiographic score, BALF cytologic ndings, lung function, and measures of airway hyperreactivity.784 Thoracic radiographs can be used to exclude alternative diagnoses but not to confirm the diagnosis of IAD. Specialized pulmonary function tests such as forced expiration and forced oscillation mechanics have been used to identify horses with IAD.65,785,786 Measures of airway hyperreactivity, performed by determining dose-dependent alterations in respiratory mechanics after nebulization of methacholine or histamine, have demonstrated that some horses with IAD have hyperresponsive airways or airway narrowing

or both.63-65

TREATMENT Because the cause of IAD remains uncertain, most treatment recommendations aim at environmental alterations to decrease exposure to dust, molds, and allergens. Horses should have pasture turnout whenever possible. When stabled, horses should be maintained in a well-ventilated barn free of noxious gases. Horses with bacterial infections should receive a 7- to 10-day course of an antibiotic, the selection of which is based on culture and sensitivity results from the tracheobronchial aspirate. Horses may bene t from a 1-week period of rest coupled with a 5-day course of orally administered interferon-α (natural human interferon-α 50 U or recombinant human interferon-α 90 U). In a clinical trial of 34 young Standardbreds diagnosed with IAD, treatment with interferon-α was associated with recurrence of cough 4 weeks later in only 23% of the horses as compared with relapse in 75% of controls.787 Interferon-α could reduce in ammation via elimination of persistent viral infections, modulation of an allergic response, or attenuation of proximal mediators of inflammation.787 One should also implement e orts to decrease the severity or frequency of EIPH by having horses train and race with furosemide. In the absence of an infectious cause, horses may also bene t from a 2- to 4-week course of corticosteroids: prednisolone at 2.2 mg/kg orally once daily for 7 to 10 days, followed by 1.0 mg/kg orally once daily for 7 to 10 days, and then 1.0 mg/kg orally once every other day for 7 to 10 days. As is the case with glucocorticoids, controlled studies examining the e cacy of bronchodilators in horses with IAD are lacking. Aerosolized albuterol (900 g) administered to eight Standardbred horses (four of which exhibited airway hyperreactivity without clinical evidence of IAD) did not alter maximal oxygen consumption, maximal carbon dioxide production, or treadmill speed at which horses exhibited a heart rate of 200 beat/min.788 Whether horses with overt signs of IAD bene t from treatment with bronchodilators remains to be determined.

Exercise-Induced Pulmonary Hemorrhage CAUSES Strenuous exercise is associated with hemorrhage from the pulmonary vasculature into the alveoli and airways, primarily in the caudal dorsal lung segments. Histologically, EIPH is characterized by the presence of edema, rupture of the pulmonary capillaries, and intraalveolar hemorrhage.

EPIDEMIOLOGY Exercise-induced pulmonary hemorrhage has been detected in most breeds of horses undergoing strenuous exercise. The prevalence of EIPH is estimated to be between 44% and 75% in Thoroughbreds, 26% and 77% in Standardbreds, 62% in racing Quarter Horses, 50% in racing Appaloosas, 68% in steeplechasers, 67% in timber racing horses, 40% in 3-day event horses, 10% in pony club event horses, and 11% in polo ponies.789-796 Indeed, EIPH probably occurs in any breed of horse that is strenuously exercised, but, apparently, the intensity of exercise rather than the duration of exercise is more important in its development. No clear correlation exists between EIPH and the location of stables, the condition of the track, or the track type.797 In a recent study of Australian racing Thoroughbreds (ages 1 to 10 years) that were not permitted to receive furosemide prerace or to wear nasal dilator strips during the race,798 EIPH was associated with the presence of excess tracheal mucus or tracheal dirt and with concentration of airborne particulates. EIPH was not associated with the age of the horse, gender, weight carried, and race distance, although previous studies have indicated a greater prevalence of EIPH in older horses.

PATHOGENESIS Numerous mechanisms have been suggested to cause EIPH including ventilation inhom*ogeneities caused by in ammatory airway disease (IAD),799,800 upper airway obstruction,801,802 mechanical concussive forces associated with locomotion,803,804 redistribution of blood ow in the lung,805 and stress failure of the pulmonary capillaries.806 Robinson and Derksen799 proposed that poor collateral ventilation in the horse coupled with small airway disease, and thus altered time constants for alveolar lling, caused underventilation of certain lung units. Extreme uctuations in the alveolar pressure of these underventilated regions during exercise produced parenchymal tearing or alveolar capillary rupture. Small airway disease (IAD) has been detected in a large percentage of horses with EIPH.800 However, its absence in the lungs of young racehorses that exhibit evidence of EIPH suggests that small airway disease, at least, is not an inciting cause of EIPH.807 The role of small airway disease in propagating the cycle of EIPH cannot be dismissed. Clarke803 speculated that visceral constraint of the diaphragm caused greater mechanical forces or stresses to develop in the dorsal thorax. Thus, in the caudodorsal lung, these mechanical forces would be borne over a narrow area, leading to parenchymal tearing or rupture of capillaries during inspiration. However, evidence of widespread hemorrhage in the dorsal portions of the lung suggests that the distribution of extremely negative intrapleural pressures is more complex and not simply restricted to the region in close proximity to the

caudal dorsal lung. An alternative mechanical theory has proposed that forelimb locomotory impact forces are transmitted through the chest wall, setting up waves converging caudodorsally in the lung parenchyma.804 Although support for locomotory-impact forces as a cause of EIPH comes from the higher percentage of steeplechasers that bleed as compared with atracers, the di erence could simply re ect a greater exercise intensity and hence cardiac output and pulmonary artery pressures generated in the jumpers.808 Furthermore, because horses still develop EIPH during swimming when concussive forces would be minimal, it seems less likely that mechanical forces alone cause EIPH.809 Currently, the most widely accepted hypothesis for the development of EIPH is stress failure of pulmonary capillaries.806,810 Stress failure is thought to occur after development of high transmural pressures (the pressure di erence between the pulmonary capillary bed and the adjacent alveoli) that disrupt capillary endothelial and alveolar epithelial tight junctions, leading to hemorrhage within the interstitium and alveoli. Elevated pulmonary capillary pressures, which may exceed 70 mm Hg in strenuously exercised horses, are a consequence of the high cardiac output, lack of su cient pulmonary vascular vasodilation, and increased blood viscosity that occurs during exercise. The second component of transmural pressures is very negative intrapleural and alveolar pressures that are required to generate air ows exceeding 120 L/s through high resistance passages (nasal cavity, larynx).811 More negative pressures would be generated when nasal, laryngeal, or pulmonary passages are narrowed from disease or if larger tidal volumes are generated, for example during exercise on an incline.802,812 Evidence supportive of the stress failure theory is provided by electron micrographs prepared from lung segments taken from strenuously exercised horses. These micrographs demonstrate a breakdown of the endothelial and epithelial tight junctions and exudation of red blood cells into alveoli.806 Although some investigators have found no correlation between pulmonary capillary pressures in exercising horses and the development of EIPH ascertained endoscopically,813 others have found a correlation between mean pulmonary artery pressures in exercising horses and the number of erythrocytes in postexercise BALF samples.814 Thus the etiopathogenesis of EIPH may represent a complex interplay of many factors relating to the physiologic and pathologic features of pulmonary function and exercise.815

CLINICAL SIGNS Physical examination of the resting horse that develops EIPH is unremarkable. In one study of 20 Standardbred racehorses, thoracic percussion, but not thoracic auscultation, identified abnormalities in the caudodorsal region of 50% (n = 5) of horses with EIPH.816 Epistaxis is an infrequent nding, occurring in less than 7% of Standardbreds, Thoroughbreds, or Quarter Horses after racing.789-791,794,795 In one retrospective study of Japanese racehorses examined 30 minutes after racing either on the at track or over hurdles, epistaxis was detected in 0.13% of the Thoroughbreds (>246,000 starts) and in 0.10% of the Anglo-Arabs (>4000 starts).817 Similar prevalence rates were reported by Weideman et al. (2003) who found epistaxis in 0.17% of South African Thoroughbred racehorses (51,465 horses in 778,532 races). In the study by Takahashi et al.,817 epistaxis occurred more frequently in horses that were older (0.04% in 2-yearolds versus 0.27% in horses > 5 years), raced over hurdles (0.11% atracers versus 0.76% steeplechasers), and sprinted (0.13% in races < 1600 m long versus 0.11% in races 1601-2000 m long). They also found that the rate of recurrence of epistaxis was 13%. The lower frequency of epistaxis reported in this study by Takahashi et al.817 may be because horses were examined 30 minutes after racing, and the horses were young (64% of horses were < 3 years of age). Sudden death attributed to EIPH is reported to occur in 82% to 87% of racehorses that die suddenly.818,819 Additional clinical signs that may be exhibited by some horses with EIPH after competition include di cult or labored breathing, coughing, or excessive swallowing. Although long suspected to cause decreased athletic performance, it had been di cult to objectively ascertain because investigations had been confounded by any or all of the following: concurrent use of medications in horses (furosemide), statistical methods, and the numbers of horses evaluated. Nevertheless, owners or trainers often report that an a ected horse does not perform up to expectations or that it exhibits a sudden loss of speed during competition.809 In a prospective study of 744 Australian Thoroughbred racehorses (2396 starts) that competed without the bene t of furosemide or nasal dilator strips, an association between the presence and severity of tracheal blood and race performance was found.798 Horses with a tracheal score at or above 2 (based on a scale of 0-4; see later discussion) had lower odds of winning or nishing in the rst three positions and a lower likelihood of being in the 90th percentile or higher for race earnings. Although the exact pathophysiologic e ect of EIPH on performance was not known, the investigators suggested that it may have re ected a deleterious effect of the bleeding on arterial oxygenation or on maximal oxygen consumption or both.783,820

DIAGNOSIS Endoscopic detection of frank blood within the trachea is the usual method of diagnosis of EIPH. Optimal time for endoscopic examination is within 90 minutes of a race or a strenuous workout. An endoscopic scoring system for tracheal blood accumulation has been developed and achieves a high degree of concordance among evaluators.798 The numeric scores range from 0 to 4, with a 0 assigned to cases for which no blood is detected in the pharynx, larynx, trachea, or mainstem bronchi and a 4 assigned to cases with multiple coalescing streams of blood

covering more than 90% of the tracheal surface with blood pooling at the thoracic inlet. One limitation of the scoring system is that it does not provide a quantitative measure of the amount of hemorrhage.798 Furthermore, con rmation of EIPH may require multiple endoscopic examinations of the same horse after exercise bouts of the same intensity. Whether, in a given horse, the tracheal blood score increases in severity over time is currently unknown. Cytologic analysis of tracheobronchial or BAL constituents can aid one in the diagnosis of EIPH. Although total nucleated cell counts in BALF do not di er between racehorses and yearlings, the number of free erythrocytes and hemosiderophages in the BALF is greater in horses in training compared with horses not being trained.821 Concern has surfaced that simple enumeration of erythrocytes recovered in the BALF to semiquantify the severity of EIPH may not be accurate and reproducible.822 When EIPH positive and negative racehorses are compared, the total hemosiderin score (a re ection of cytoplasmic iron in alveolar macrophages) agrees well with EIPH status. Speci cally, a total hemosiderin score above 75 has a sensitivity and speci city of 94% and 88%, respectively, in identifying EIPH-positive horses.823 Hemosiderin scores should be interpreted cautiously because hemorrhage from prior episodes of pneumonia, pleuropneumonia, or abscessation will increase hemosiderin scores. Pulmonary biopsy specimens obtained from the seventh or eighth intercostal space (10 cm above the level of the point of the elbow) from the lungs of EIPH-positive horses do not demonstrate histologic abnormalities that would distinguish a ected horses from EIPH negative horse.816 In contrast, histologic evaluation of lung samples obtained from the caudal dorsal lung regions (postmortem) of horses with EIPH reveals bronchiolitis, hemosiderophages, focal eosinophilic infiltration, and increased connective tissue.800 Radiographs are of limited usefulness in diagnosing horses with EIPH, given that the signs found in a ected horses—an accentuated bronchial, bronchointerstitial, or interstitial pattern824—can also be observed in racehorses that do not exhibit evidence of pulmonary bleeding.823 In severe cases of hemorrhage, a radiopaque region in the caudal dorsal lung lobe that obscures the pulmonary vasculature is detected825,826 (Figure 9-21). Nuclear scintigraphy has not been useful in detecting areas of pulmonary hemorrhage after exercise.

FIGURE 9-21

Focal area of increased interstitial pulmonary opacity (arrows) in the dorsal caudal lung field represents exercise-induced pulmonary hemorrhage.

(Courtesy D. S. Biller, Manhattan, Kan., 1991.)

Hemogram, serum biochemistry, and arterial blood gas values obtained from resting horses with EIPH are within normal limits. The mean platelet counts of EIPH-positive horses (128 × 109 cells/L) are signi cantly lower than those of EIPH negative horses (159 × 109 cells/L), although the total counts are within the normal range.816

TREATMENT Drugs that reduce pulmonary capillary pressure during exercise have gained the most widespread use in the treatment and prevention of EIPH. Furosemide, 250 to 300 mg, is given intravenously 4 hours before the horse is strenuously worked or raced. Furosemide decreases plasma volume and decreases exercise-associated pulmonary artery and pulmonary capillary pressures.827-829 Although furosemide administration has failed to prevent the development of EIPH in horses that were previously EIPH negative,830,831 some evidence indicates that it reduces the red blood cell counts in BALF or the EIPH endoscopic score.832 Furosemide administration is also associated with improved performance in Thoroughbred racehorses.833 Nitric oxide, a potent vasodilator, has been administered to horses with EIPH. In one laboratory investigation of strenuously exercised horses, intravenous nitric oxide failed to reduce pulmonary artery or pulmonary capillary pressures and failed to prevent EIPH.834 In another study, inhaled nitric oxide administered to horses maximally exercised on the treadmill reduced pulmonary arterial pressures but worsened erythrocyte counts in the BALF.835 Nasal dilators (Flair nasal strips) have been advocated as a treatment for horses with EIPH. These nasal dilators are adhesive strips

containing springs that are applied externally 3 cm above the nostrils with the purpose of maintaining upper airway patency and reducing airway resistance during inspiration. In laboratory studies of exercising horses, the application of nasal strips failed to change arterial blood gases, pulmonary artery pressures, peak esophageal pressures, or tracheal blood accumulation compared with controls.836,837 In one study, the nasal strips reduced erythrocyte counts in the BALF.837 The nding of concomitant small airway disease (IAD) in some horses suggests that one should attempt to minimize environmental dusts and molds. Corticosteroids and bronchodilators, inhaled or administered parenterally, have not been shown to reduce EIPH but may lessen the severity of small airway disease. In a laboratory study simulating airway responses in horses with EIPH (produced by intrapulmonary inoculation of 50 ml of autologous blood), each of three treatments consisting of oral prednisolone (1 mg/kg once daily), inhaled beclomethasone dipropionate (3.75 mg twice daily), or oral clenbuterol hydrochloride (0.8 g/kg) decreased total erythrophage count in BALF compared to nontreated controls. This study suggested that the treatments suppressed numbers and possibly phagocytic function of macrophages.838 In contrast, the antiin ammatory cromoglycate, which is believed to stabilize mast cell membranes, has not been e cacious in the treatment of EIPH.839 Other formulations that have been advocated for the treatment of EIPH include intratracheal immunoglobulin, oral herbal remedies (e.g., Yunnan paiyao, Single Immortal, Red Lung, and Red Lung Plus), intravenous aminocaproic acid, bio avinoids, aspirin, and estrogens, but no scientific data exist that demonstrates their efficacy.840 Rest has been recommended, but pulmonary hemorrhage will likely recur once the horse resumes training. In severe cases of EIPH, a short course of broad-spectrum antimicrobials may be indicated to lessen the chance of secondary pneumonia or abscess formation.

PREVENTION World wide, EIPH is of concern because of nancial losses resulting from decreased performance, suspension of horses from racing, lost training days, and the cost of prerace mediation.841 The e cacy of treatment regimens in the prevention of EIPH under race conditions has been di cult to determine even when speed handicapping methods are used. Variables such as the administration of drugs unknown to the investigators or the inability to diagnose or reproduce EIPH within a given horse on consecutive days make interpretation of these studies difficult.

Tumors of the Respiratory System CAUSES Based on postmortem surveys, the most frequently encountered tumors in the horse, regardless of the body system, are sarcoids, squamous cell carcinomas, bromas, melanomas, papillomas, brosarcomas, and lymphomas.653,842,843 Neoplastic involvement of the thoracic structures or pulmonary parenchyma per se is rare in horses and occurs at a prevalence rate of 0.15 to 0.62.524,843 Tumors in the thoracic cavity arise as either primary neoplasms or from metastases and may involve extrapulmonary structures or the lung parenchyma. Examples of primary thoracic tumors are lymphomas and mesotheliomas.524,761,844-847 In addition to being the most common thoracic neoplasm, lymphoma is also the most common tumor of the equine hematopoietic system (see the discussion in the hematopoietic Chapter 14).845,847 Lymphoma is classi ed according to its anatomic distribution as mediastinal, alimentary, cutaneous, multicentric, and generalized, although several forms may exist in the same horse. Horses a ected with mediastinal lymphoma are young to middle-aged (mean age of 10 years) and exhibit weight loss, ventral edema, jugular distention caused by the presence of a mediastinal mass, tachypnea, or respiratory distress caused by pleural e usion secondary to compression of vessels.845,848 Ultrasonography allows one to de ne the size and extent of the tumor, as well as to assess potential responses to treatment. Most mediastinal lymphomas are hom*ogeneous hypoechoic masses greater than 4 × 4 cm in size, but some tumors exhibit focal hyperechoic areas.849 As the size of the mass increases, the heart is displaced caudally causing it to be imaged in the sixth through the eighth intercostal spaces. Anechoic uid surrounding the mass—indicative of pleural e usion—enhances the ultrasonographic image but may contribute to lung compression and breathing di culties. Di erential diagnoses for a cranial mediastinal mass include abscess, in amed mediastinal lymph node, fatty tissue, and the thymus, which may be present in horses up to 4 years of age.850 With the horse well sedated and its forelimb extended cranially, an ultrasound-guided biopsy specimen of the mass may be obtained aseptically through the third intercostal space.850 Short-term intravenous anesthesia provides an alternative method of obtaining the ultrasound-guided biopsy if the horse will not remain stationary with sedation. In horses with concurrent pleural e usion, thoracocentesis may yield a modi ed transudate or nonseptic exudate, but neoplastic cells— lymphoblasts with mitotic gures—are not consistently found. After drainage of the pleural e usion, thoracic radiographs should be obtained to determine if the lung parenchyma is involved. Thoracoscopy may be another option to determine the extent of involvement of intrathoracic structures and possibly to obtain a biopsy, although the cranial and ventral position of the mass may preclude its clear visualization.75,77,850,851 In a ected horses, rectal examination and ultrasonographic evaluation of the abdomen should always be performed to detect involvement of abdominal organs or structures. The clinical course of the disease, from onset of signs until euthanasia, is often rapid and may be less than 8 weeks for most cases.845 A recent report describes the successful treatment of a mixed-cell mediastinal lymphoma using cytarabine (170 mg/m2 intramuscularly), cyclophosphamide (142 mg/m2 intravenously), and prednisolone (86 mg/m2 orally).852 Mesothelioma is the second most common primary thoracic tumor in the horse. This tumor arises from the serosal lining cells of the pleura and pericardium. In contrast to human disease, its development in horses has not been linked to asbestos exposure. A ected horses are often young to middle aged (2 to 13 years) and may exhibit respiratory distress caused by pleural e usion.844,846,853-855 Thoracocentesis usually yields a dark to blood-tinged uid of normal cellularity but with an increased total protein concentration. Cytologic examination of pleural uid may reveal pleomorphic mesothelial cells with mitotic gures.846 However, in other cases, distinguishing mesothelioma from other neoplasms or from reactive mesothelial cells may be impossible. Thoracoscopy may be used to con rm the presence of neoplastic nodules on the pleura, which should be biopsied for histologic examination and classi cation.856 Malignant mesothelioma can be classi ed into three main histologic types: epithelioid, biphasic, and sarcomatoid. An immunohistochemical marker for epithelioid-type mesotheliomas that di erentiates them from metastatic adenocarcinomas is calretinin, a protein that is related to the calcium-binding proteins and may be helpful in the diagnosis.857 Treatment is not usually attempted in horses with mesothelioma. Whether pleurodesis would reduce the e usion and prolong the horse’s life is unknown. Primary equine lung tumors include granular cell tumors,858-868 bronchial myxomas,869 bronchial and pulmonary carcinomas,653,870,871 bronchial and pulmonary adenocarcinomas,872,873 bronchogenic squamous cell carinomas,871 bronchioalveolar carinoma,872 pulmonary chondrosarcomas,874,875 and pulmonary leiomyosarcoma.876 The prevalence rate of primary lung tumors (0.15%) is much less than that of metastatic pulmonary tumors. Of the primary lung tumors, the most common one is granular cell tumor (myoblastoma or Schwann cell tumor), and it is the only primary lung tumor that is mesenchymal rather than epithelial in origin. Based on a review of 26 reported cases,651,858,860-866,868,877-880 the mean age of a ected horses is 13 years; the majority of cases occur in mares (22 of 26 horses), and the tumor is often located in the right lung (16 of 26 horses). In a small number of horses, the tumor occurred in both lung lobes. In some horses, the tumor may be an incidental nding detected at necropsy; however, in most cases, clinical signs of chronic respiratory disease—coughing, increased respiratory rate or e ort, and weight loss—develop. In the absence of an extensive workup, these signs may lead to a misdiagnosis of RAO. However, because the tumor or tumors protrude into the bronchus, air ow into that lung lobe may be reduced or completely

lacking, making lung sounds di cult or impossible to detect. (This nding is in contrast to RAO in which adventitious lungs sounds may be heard over an expanded lung eld.) Secondary bacterial infections may develop in the a ected lung presumably because of impaired clearance of secretions caused by the tumor. In a few cases, hypertrophic osteopathy (Marie’s disease) develops concurrently.858,866,877,879 This condition is characterized by enlargement of the joints and the presence of bony swellings in the distal limbs caused by subperiosteal bony proliferation. As a result, a ected horses exhibit generalized sti ness and shifting limb lameness. The diagnosis of granular cell tumor is based on imaging and endoscopic ndings. On thoracic radiographs, single or multiple radioopaque nodules are evident. Endoscopic examination reveals mucosal-covered nodules protruding into the bronchi. A de nitive diagnosis is obtained after histologic evaluation of endoscopically obtained biopsies. The cells are globoid or stellate and contain eosinophilic, PAS-positive cytoplasmic granules.651,862 Immunohistochemical staining of the cells is positive for vimentin, an intermediate lament most commonly associated with cells of mesenchymal origin, and for S-100, a calcium-binding protein most commonly found in cells of neural crest origin.865 Currently, no medical treatment has been developed for pulmonary granular cell tumors. In two cases, the tumor was resected successfully by transendoscopic electrosurgery881 or via lung resection.880 In general, the tumor does not usually metastasize to other sites, but bilateral lung involvement is a possibility. The limited number of case reports in the literature describing clinical features of other primary lung tumors653,869-876 precludes a detailed description of these tumors. In general, middle-aged to older horses are a ected and exhibit weight loss, chronic cough, and tachypnea. In the absence of a thorough evaluation, the chronicity of the clinical signs may lead to a misdiagnosis of RAO, pneumonia, or even EIPH (epistaxis). Thoracic radiographs demonstrate the presence of pulmonary nodules that are con rmed to be neoplastic in lung biopsy obtained by ultrasound guidance. Cytologic examination of tracheal aspirates is not usually diagnostic. Treatment has not been attempted. Thoracic tumors involving the pleural surfaces, lymph nodes, mediastinum, esophagus, diaphragm, or lungs may also arise from metastases from other neoplastic sites. Examples of metastatic tumors that a ect the thoracic cavity include adenocarcinoma (primary sites of origin are in the kidney, uterus, ovary, mammary gland, and pancreas),521,522,882-884 squamous cell carcinoma,516,885-888 hepatoblastoma,517 hymoma,889 fibrosarcoma,515 melanoma,890 hemangiosarcoma,518,524,891-903 and lymphoma.524,845,904,905 Hemangiosarcoma remains a rare tumor, occurring in 2 of 1322 horses in a study by Sundbery, Burnstein, Page, et al.842 and in 1 of 4739 cases in a survey reported by Hargis and McElwain.892 It is a malignant tumor of endothelial cells but does not appear to develop from a preexisting hemangioma.895 A comprehensive review896 provides a description of the clinical aspects of 35 cases of disseminated hemangiosarcoma, 21 of which were diagnosed at a participating veterinary hospital and 14 of which were previously reported in the literature.∗ The median age of horses was 12 years, and most clinical signs were referable to the respiratory or to the musculoskeletal systems. For example, respiratory distress and epistaxis were noted in 26% of cases, and muscular or subcutaneous swellings were noted in 24%. The majority of horses exhibited mucous membrane pallor either alone or in combination with icterus, re ecting prior hemorrhage. Ultrasonographic evaluation detected pleural e usion or hemothorax in four cases. Cytologic analysis of either the pleural or tracheobronchial aspirates (in horses that developed epistaxis) failed to demonstrate neoplastic cells. Antemortem diagnosis is di cult and may require thoracoscopy or pleuroscopy to visualize neoplasms and to obtain samples for histopathologic examination.74,894 Microscopically, one may see vascular channels lined with plump, spindle-shaped cells. In some cases, well-di erentiated channels are less conspicuous or are lacking, and cells have epithelioid morphologic appearance. When present, the tumor is classi ed as epithelioid (histiocytoid) hemangiosarcoma.902 Immunohistochemical staining of tumor samples demonstrates cells that can be poorly reactive against factor VIII reactive antigen, positive for UEA-agglutinin I (a receptor of Ulex europaeus I) and negative for cytokeratin.901,902 In a ected horses, supportive care, including thoracocentesis, blood transfusions, and intravenous uids, is generally unrewarding. Clinical deterioration is rapid, with death occurring, on average, 17 days from the onset of clinical signs. In a study of 11 young horses (younger than 4 years) with hemangiosarcoma, the investigators reported that the majority of cases had cutaneous masses or di use limb and joint swellings.903 Although only one horse in this study developed disseminated hemangiosarcoma, the exact percentage of horses with cutaneous neoplasms that develop disseminated disease and over what time period is unknown. In general, horses with metastatic thoracic neoplasia exhibit nonspeci c clinical signs of depression, inappetence and weight loss, or clinical signs referable to the site of the primary neoplasm. In some cases, respiratory distress may occur if pleural e usion develops, or coughing may occur if the tumor compresses an airway.847 Diagnostic examination, which requires a combination of imaging techniques including endoscopy, thoracic radiography, ultrasonography (and possibly thoracoscopy), and cytologic evaluation of respiratory secretions along with evaluation of the abdominal cavity and other body systems should be undertaken. Usually, hematologic and serum biochemical abnormalities are nonspeci c and include anemia, leukocytosis, neutrophilia, and hyper brinogenemia.846 The prognosis is poor for horses with neoplasms, and most patients are euthanized. In horses in which the tumor is localized and has a low metastatic potential, as exists with granular cell tumors, lung resection may be an option.880 In cases with pleural e usion, the clinician may achieve temporary stabilization by

thoracocentesis, but the therapeutic benefits of pleurodesis have not been established for equine patients (see the section on pleural effusion).

REFERENCES 1. Banks K.L., McGuire T.C., Jerrells T.R. Absence of B lymphocytes in a horse with primary agammaglobulinemia. Clin Immunol Immunopathol. 1976;5:282. 2. Deem D.A., Traver D.S., Thacker H.L., et al. Agammaglobulinemia in a horse. J Am Vet Med Assoc. 1979;175:469. 3. Davis D.M., Honnas C.M., Hedlund C.S., et al. Resection of a cervical tracheal bronchus in a foal. J Am Vet Med Assoc. 1991;198:2097. 4. Bacon C.L., Davidson H.J., Yvorchuk K., et al. Bilateral Horner’s syndrome secondary to metastatic squamous cell carcinoma in a horse. Equine Vet J. 1996;28:500. 5. Simoens P., Lauwers H., De Muelenare C., et al, Horner’s syndrome in the horse: a clinical, experimental and morphological study. Equine Vet J, 10, 1990, 62. 6. Kotlikoff M.I., Gillespie J.R. Lung sounds in veterinary medicine I. Terminology and mechanisms of sound production. Compend Contin Educ Pract Vet. 1983;5:634. 7. Kotlikoff M.I., Gillespie J.R. Lung sounds in veterinary medicine part II. Deriving clinical information from lung sounds. Compend Contin Educ Pract Vet. 1984;6:462. 8. Viel L., Harris F.W., Curtis R.A. Terminology of lung sounds in large animal veterinary medicine. In: Deegen E., Beadle R.E., editors. Lung Function and Respiratory Diseases. Stuttgart, Germany: Hippiatrika, 1986. 9. Forgacs P. Lung Sounds. London: Bailliere Tindall, 1978. 10. Bakos Z., Voros K. Thoracic percussion to determine the caudal lung border in healthy horses. J Vet Intern Med. 2007;21:504. 11. Roudebush P., Sweeney C.R. Thoracic percussion. J Am Vet Med Assoc. 1990;197:714. 12. Darien B.J., Brown C.M., Walker R.D., et al. A tracheoscopic technique for obtaining uncontaminated lower airway secretions for bacterial culture in the horse. Equine Vet J. 1990;22:170. 13. Franklin S.H., Naylor J.R., Lane J.G., Videoendoscopic evaluation of the upper respiratory tract in 93 sport horses during exercise testing on a high-speed treadmill. Equine Vet J, 36, 2006, 540. 1 4 . Lane J.G., Bladon B., Little D.R., et al. Dynamic obstructions of the equine upper respiratory tract. Part 1: observations during high-speed treadmill endoscopy of 600 Thoroughbred racehorses. Equine Vet J. 2006;38:393. 15. Tan R.H., Dowling B.A., Dart A.J. High-speed treadmill videoendoscopic examination of the upper respiratory tract in the horse: the results of 291 clinical cases. Vet J. 2005;170:243. 16. Worster A.A. Equine sinus endoscopy using a exible endoscope: diagnosis and treatment of sinus disease in the standing sedated horse. Proc Am Assoc Equine Pract. 1999;45:128. 17. Ruggles A.J., Ross M.W., Freeman D.E. Endoscopic examination and treatment of paranasal sinus disease in 16 horses. Vet Surg. 1993;22:508. 18. Morrow K.L., Park R.D., Spurgeon T.L., et al. Computed tomographic imaging of the equine head. Vet Radiol Ultrasound. 2000;41:491. 19. Kraft S.L., Gavin P.R. Physical principles and technical considerations for equine computed tomography and magnetic resonance imaging. Vet Clin North Am Equine Pract. 2001;17:115. 20. Tucker R.L., Sande R.D. Computed tomography and magnetic resonance imaging of the equine musculoskeletal conditions. Vet Clin North Am Equine Pract. 2001;17:145. 21. Smallwood J.E., Wood B.C., Taylor W.E., et al. Anatomic reference for computed tomography of the head of the foal. Vet Radiol Ultrasound. 2002;43:99. 22. Chalmers H.J., Cheetham J., Dykes N.L., et al. Computed tomographic diagnosis–stylohyoid fracture with pharyngeal abscess in a horse without temporohyoid disease. Vet Radiol Ultrasound. 2006;47:165. 23. Pease A.P., van Biervliet J., Dykes N.L., et al. Complication of partial stylohyoidectomy for treatment of temporohyoid osteoarthropathy and an alternative surgical technique in three cases. Equine Vet J. 2004;36:546. 24. Probst A., Henninger W., Willmann M. Communications of normal nasal and paranasal cavities in computed tomography of horses. Vet Radiol Ultrasound. 2005;46:44. 25. Walker A.M., Sellon D.C., Cornelisse C.J., et al. Temporohyoid osteoarthropathy in 33 horses (1993-2000). J Vet Intern Med. 2002;16:697. 26. O’Brien R.T., Biller D.S. Dental imaging. Vet Clin North Am Equine Pract. 1998;14:259. 27. Wion L., Perkins G., Ainsworth D.M., et al. Use of computerised tomography to diagnose a Rhodococcus equi mediastinal abscess causing severe respiratory distress in a foal. Equine Vet J. 2001;33:523. 28. Arencibia A., Vazquez J.M., Jaber R., et al. Magnetic resonance imaging and cross sectional anatomy of the normal equine sinuses and nasal passages. Vet Radiol Ultrasound. 2000;41:313. 29. Ferrell E.A., Gavin P.R., Tucker R.L., et al. Magnetic resonance for evaluation of neurologic disease in 12 horses. Vet Radiol Ultrasound. 2002;43:510. 30. Chiesa O.A., Vidal D., Domingo M., et al. Cytological and bacteriological findings in guttural pouch lavages of clinically normal horses. Vet Rec. 1999;144:346. 31. Chiesa O.A., Garcia F., Domingo M., et al. Cytological and microbiological results from equine guttural pouch lavages obtained percutaneously: correlation with histopathological findings. Vet Rec. 1999;144:618. 32. Derksen F.J., Brown C.M., Sonea I., et al. Comparison of transtracheal aspirate and bronchoalveolar lavage cytology in 50 horses with chronic lung disease. Equine Vet J. 1989;21:23. 33. Larson V.L., Busch R.H. Equine tracheobronchial lavage: comparison of lavage cytologic and pulmonary histopathologic findings. Am J Vet Res. 1985;46:144. 34. Fogarty U. Evaluation of a bronchoalveolar lavage technique. Equine Vet J. 1990;22:174. 35. Sweeney C.R., Sweeney R.W.3rd, Benson C.E. Comparison of bacteria isolated from specimens obtained by use of endoscopic guarded tracheal swabbing and percutaneous tracheal aspiration in horses. J Am Vet Med Assoc. 1989;195:1225. 36. Racklyeft D.J. A tracheoscopic technique for obtaining uncontaminated lower airway secretions for bacterial culture in the horse. Equine Vet J. 1990;22(408):415. 37. Christley R.M., Hodgson D.R., Rose R.J., et al. Comparison of bacteriology and cytology of tracheal uid samples collected by percutaneous transtracheal aspiration or via an endoscope using a plugged, guarded catheter. Equine Vet J. 1999;31:197. 38. Sweeney C.R., Beech J., Roby K.A. Bacterial isolates from tracheobronchial aspirates of healthy horses. Am J Vet Res. 1985;46:2562. 39. Burrell M.H., Mackintosh M.E., Taylor C.E. Isolation of Streptococcus pneumoniae from the respiratory tract of horses. Equine Vet J. 1986;18:183. 40. ME Mackintosh, ST Grant and MH Burrell: Evidence of streptococcus pneumonia as a cause of respiratory disease in young thoroughbred horses in training. In DG Powell, editor: Equine Infectious Diseases V: Proceedings of the Fifth International Conference, Lexington, Kentucky, 1988, University of Kentucky Press. 41. Benson C.E., Sweeney C.R. Isolation of Streptococcus pneumoniae type 3 from equine species. J Clin Microbiol. 1984;20:1028. 42. Reimer J.M. Diagnostic ultrasonography of the equine thorax. Compend Contin Educ Pract Vet. 1990;12:1321.

43. Lamb C.R., O’Callaghan M.W. Diagnostic imaging of equine pulmonary disease. Compend Contin Educ Pract Vet. 1989;11:1110. 44. King G.K. Equine thoracic radiography II. Radiographic patterns of equine pulmonary and pleural diseases using air-gap rare-earth radiography. Compend Contin Educ Pract Vet. 1981;3:S283. 45. Farrow C.S. Radiographic aspects of inflammatory lung disease in the horse. Vet Radiol. 1981;22:107. 46. Rantanen N.W. Diseases of the thorax. Vet Clin North Am Equine Pract. 1986;2:49. 47. Reef V.B., Boy M.G., Reid C.F., et al. Comparison between diagnostic ultrasonography and radiography in the evaluation of horses and cattle with thoracic disease: 56 cases (19841985). J Am Vet Med Assoc. 1991;198:2112. 48. Reimer J.M., Reef V.B., Spencer P.A. Ultrasonography as a diagnostic aid in horses with anaerobic bacterial pleuropneumonia and/or pulmonary abscessation: 27 cases (19841986). J Am Vet Med Assoc. 1989;194:278. 49. Bennett D.G. Evaluation of pleural fluid in the diagnosis of thoracic disease in the horse. J Am Vet Med Assoc. 1986;188:814. 50. Sweeney C.R., Divers T.J., Benson C.E. Anaerobic bacteria in 21 horses with pleuropneumonia. J Am Vet Med Assoc. 1985;187:721. 51. O’Callaghan M.W., Hornof W.J., Fisher P.E., et al. Ventilation imaging in the horse with 99mtechnetium-DTPA radioaerosol. Equine Vet J. 1987;19:19. 52. Attenburrow D.P., Portergill M.J., Vennart W. Development of an equine nuclear medicine facility for gamma camera imaging. Equine Vet J. 1989;21:86. 53. O’Callaghan M.W., Hornof W.J., Fisher P.E., et al. Exercise-induced pulmonary haemorrhage in the horses: results of a detailed clinical, post mortem and imaging study. VII. Ventilation/perfusion scintigraphy in horses with EIPH. Equine Vet J. 1987;19:423. 54. O’Callaghan M.W., Kinney L.M. Pulmonary scintigraphy in horses with chronic obstructive pulmonary disease. Comp Respir Soc. 1989;8:78. 55. O’Callaghan M.W. Scintigraphic imaging of lung disease. In: Beech J., editor. Equine respiratory disorders. Philadelphia, PA: Lea & Febiger, 1991. 56. Nelson R., Hampe D.W. Measurement of tracheal mucous transport rate in the horse. Am J Vet Res. 1983;44:1165. 57. Willoughby R.A., Ecker G., Riddolls L. Mucociliary clearance in the nose and trachea of horses. Comp Respir Soc. 1989;8:36. 58. Barakzai S., Tremaine H., Dixon P. Use of scintigraphy for diagnosis of equine paranasal sinus disorders. Vet Surg. 2006;35:94. 59. Robinson N.E. The physiologic basis of pulmonary function tests. Proc Am Coll Vet Intern Med. 1992;10:403. 60. Robinson N.E. Tests of equine airway function. Proc Am Coll Vet Intern Med. 1992;10:284. 61. Young S.S., Hall L.W. A rapid, non-invasive method for measuring total respiratory impedance in the horse. Equine Vet J. 1989;21:99. 62. van Erck E., Votion D., Art T., et al. Measurement of respiratory function by impulse oscillometry in horses. Equine Vet J. 2004;36:21. 63. Hare J.E., Viel L. Pulmonary eosinophilia associated with increased airway responsiveness in young racing horses. J Vet Intern Med. 1998;12:163. 64. Hoffman A.M., Mazan M.R., Ellenberg S. Association between bronchoalveolar lavage cytologic features and airway reactivity in horses with a history of exercise intolerance. Am J Vet Res. 1998;59:176. 65. Couëtil L.L., Rosenthal F.S., DeNicola D.B., et al. Clinical signs, evaluation of bronchoalveolar lavage uid, and assessment of pulmonary function in horses with in ammatory respiratory disease. Am J Vet Res. 2001;62:538. 66. Mazan M.R., Ho man A.M., Manjerovic N. Comparison of forced oscillation with the conventional method for histamine bronchoprovocation testing in horses. Am J Vet Res. 1999;60:174. 67. Art T., Anderson L., Woakes A.J., et al. Mechanics of breathing during strenuous exercise in Thoroughbred horses. Respir Physiol. 1990;82:279. 68. Savage C.J., Traub-Dargatz J.L., Mumford E.L. Survey of the large animal diplomates of the American College of Veterinary Internal Medicine regarding percutaneous lung biopsy in the horse. J Vet Intern Med. 1998;12:456. 69. Raphel C.F., Gunson D.E. Percutaneous lung biopsy in the horse. Cornell Vet. 1981;71:439. 70. Venner M., Schmidbauer S., Drommer W., et al. Percutaneous lung biopsy in the horse: comparison of two instruments and repeated biopsy in horses with induced acute interstitial pneumopathy. J Vet Intern Med. 2006;20:968. 71. Pusterla N., Watson J.L., Madigan J.E., et al. Technique and diagnostic value of percutaneous lung biopsy in 66 horses with di use pulmonary diseases using an automated biopsy device. Equine Vet Educ. 2007;19:157. 72. Perkins G., Ainsworth D.M., Yeager A. Hemothorax in 2 horses. J Vet Intern Med. 1999;13:375. 73. Mansmann R.A., Bernardstrother S. Pleuroscopy in horses. Mod Vet Pract. 1985;66:9. 74. Rossier Y., Sweeney C.R., Heyer G., et al. Pleuroscopic diagnosis of disseminated hemangiosarcoma in a horse. J Am Vet Med Assoc. 1990;196:1639. 75. Vachon A.M., Fischer A.T. Thoracoscopy in the horse: diagnostic and therapeutic indications in 28 cases. Equine Vet J. 1998;30:467. 76. Lugo J., Stick J.A., Peroni J., et al. Safety and efficacy of a technique for thoracoscopically guided pulmonary wedge resection in horses. Am J Vet Res. 2002;63:1232. 77. Pollock P.J., Russell T. Standing thoracoscopy in the diagnosis of lymphosarcoma in a horse. Vet Rec. 2006;159:354. 78. Kirschvink N. The challenge of assessing inflammatory and structural changes in lower equine airways: a chance for thoracoscopic-guided pulmonary biopsy? Vet J. 2006;172:202. 79. Peroni J.F., Robinson N.E., Stick J.A., et al. Pleuropulmonary and cardiovascular consequences of thoracoscopy performed in healthy standing horses. Equine Vet J. 2000;32:280. 80. Freeman D.E. Sinus disease. Vet Clin North Am Equine Pract. 2003;19:209. 81. Hillman D.J. The skull. In: Getty R., editor. Sisson and Grossman’s the anatomy of the domestic animals. Philadelphia: WB Saunders, 1975. 82. McCann J.L., Dixon P.M., Mayhew I.G. Clinical anatomy of the equine sphenopalatine sinus. Equine Vet J. 2004;36:466. 83. Pirie M., Pirie H.M., Wright N.G. A scanning electron microscopic study of the equine upper respiratory tract. Equine Vet J. 1990;22:333. 84. Head K.W., Dixon P.M. Equine nasal and paranasal sinus tumours. Part 1: review of the literature and tumour classification. Vet J. 1999;157:261. 85. Tremaine W.H., Dixon P.M. A long-term study of 277 cases of equine sinonasal disease. Part 1: details of horses, historical, clinical and ancillary diagnostic ndings. Equine Vet J. 2001;33:274. 86. Tremaine W.H., Dixon P.M. A long-term study of 277 cases of equine sinonasal disease. Part 2: treatments and results of treatment. Equine Vet J. 2001;33:283. 87. Lane J.G., Longstaffe J.A., Gibbs C. Equine paranasal sinus cysts: a report of 15 cases. Equine Vet J. 1987;19:537. 88. Mason B.J. Empyema of the equine paranasal sinuses. J Am Vet Med Assoc. 1975;167:727. 89. Prichard M.A., Hackett R.P., Erb H.N. Long-term outcome of tooth repulsion in horses. A retrospective study of 61 cases. Vet Surg. 1992;21:145. 90. Gibbs C., Lane J.G. Radiographic examination of the facial, nasal and paranasal sinus regions of the horse. II. Radiological findings. Equine Vet J. 1987;19:474. 91. Tremaine W.H., Clarke C.J., Dixon P.M. Histopathological findings in equine sinonasal disorders. Equine Vet J. 1999;31:296. 92. Dixon P.M., Head K.W. Equine nasal and paranasal sinus tumours: part 2: a contribution of 28 case reports. Vet J. 1999;157:279.

93. Tucker R.L., Farrell E. Computed tomography and magnetic resonance imaging of the equine head. Vet Clin North Am Equine Pract. 2001;17:131. 94. Henninger W., Frame E.M., Willmann M., et al. CT features of alveolitis and sinusitis in horses. Vet Radiol Ultrasound. 2003;44:269. 95. Schumacher J., Dutton D.M., Murphy D.J., et al. Paranasal sinus surgery through a frontonasal flap in sedated, standing horses. Vet Surg. 2000;29:173. 96. Freeman D.E., Orsini P.G., Ross M.W., et al. A large frontonasal bone flap for sinus surgery in the horse. Vet Surg. 1990;19:122. 97. Boulton C.H. Equine nasal cavity and paranasal sinus disease: a review of 85 cases. J Equine Vet Sci. 1985;5:268. 98. Hilbert B.J., Little C.B., Klein K., et al. Tumours of the paranasal sinuses in 16 horses. Aust Vet J. 1988;65:86. 99. Marriott M.R., Dart A.J., Hodgson D.R. Treatment of progressive ethmoidal haematoma using intralesional injections of formalin in three horses. Aust Vet J. 1999;77:371. 100. Cook W.R., Littlewort M.C. Progressive haematoma of the ethmoid region in the horse. Equine Vet J. 1974;6:101. 101. Schumacher J., Yarbrough T., Pascoe J., et al. Transendoscopic chemical ablation of progressive ethmoidal hematomas in standing horses. Vet Surg. 1998;27:175. 102. Rothaug P.G., Tulleners E.P. Neodymium:yttrium-aluminum-garnet laser-assisted excision of progressive ethmoid hematomas in horses: 20 cases (1986-1996). J Am Vet Med Assoc. 1999;214:1037. 103. Greet T.R. Outcome of treatment in 23 horses with progressive ethmoidal haematoma. Equine Vet J. 1992;24:468. 104. Laing J.A., Hutchins D.R. Progressive ethmoidal haematoma in horses. Aust Vet J. 1992;69:57. 105. Specht T.E., Colahan P.T., Nixon A.J., et al. Ethmoidal hematoma in nine horses. J Am Vet Med Assoc. 1990;197:613. 106. Colbourne C.M., Rosenstein D.S., Ste cek B.A., et al. Surgical treatment of progressive ethmoidal hematoma aided by computed tomography in a foal. J Am Vet Med Assoc. 1997;211:335. 107. Frees K.E., Gaughan E.M., Lillich J.D., et al. Severe complication after administration of formalin for treatment of progressive ethmoidal hematoma in a horse. J Am Vet Med Assoc. 2001;219:950. 108. Baptiste K.E., Naylor J.M., Bailey J., et al. A function for guttural pouches in the horse. Nature. 2000;403:382. 109. Manglai D., Wada R., Endo H., et al. Macroscopic anatomy of the auditory tube diverticulum (guttural pouch) in the thoroughbred equine–a silicon mold approach. Okajimas Folia Anat Jpn. 2000;76:335. 110. Habel R.E., editor. Applied veterinary anatomy. Ithaca, NY: WB Saunders, 1975. 111. Blazyczek I., Hamann H., Deegen E., et al. Retrospective analysis of 50 cases of guttural pouch tympany in foals. Vet Rec. 2004;154:261. 112. Blazyczek I., Hamann H., Ohnesorge B., et al. Inheritance of guttural pouch tympany in the arabian horse. J Hered. 2004;95:195. 113. Freeman D.E. Diagnosis and treatment of diseases of the gutteral pouch (part 1). Compend Contin Educ Pract Vet. 1980;2:S3. 114. Tetens J., Tulleners E.P., Ross M.W., et al. Transendoscopic contact neodymium:yttrium aluminum garnet laser treatment of tympany of the auditory tube diverticulum in two foals. J Am Vet Med Assoc. 1927;204:1994. 115. Greet T.R. Outcome of treatment in 35 cases of guttural pouch mycosis. Equine Vet J. 1987;19:483. 116. Cook W.R., Campbell R.S., Dawson C. The pathology and aetiology of guttural pouch mycosis in the horse. Vet Rec. 1968;83:422. 117. Freeman D.E. Long-term follow-up on a large number of horses that underwent transarterial coil embolisation (TCE) for guttural pouch mycosis (GPM). Equine Vet J. 2006;38:271. 118. Walmsley J.P. A case of atlanto-occipital arthropathy following guttural pouch mycosis in a horse. The use of radioisotope bone scanning as an aid to diagnosis. Equine Vet J. 1988;20:219. 119. Dixon P.M., Rowlands A.C. Atlanto-occipital joint infection associated with guttural pouch mycosis in a horse. Equine Vet J. 1981;13:260. 120. Millar H. Guttural pouch mycosis in a 6-month-old filly. Can Vet J. 2006;47:259. 121. Cook W.R. The clinical features of guttural pouch mycosis in the horse. Vet Rec. 1968;83:336. 122. Colles C.M., Cook W.R. Carotid and cerebral angiography in the horse. Vet Rec. 1983;113:483. 123. Lane J.G. The management of guttural pouch mycosis. Equine Vet J. 1989;21:321. 124. Lepage O.M., Piccot-Crezollet C. Transarterial coil embolisation in 31 horses (1999-2002) with guttural pouch mycosis: a 2-year follow-up. Equine Vet J. 2005;37:430. 125. Leveille R., Hardy J., Robertson J.T., et al. Transarterial coil embolization of the internal and external carotid and maxillary arteries for prevention of hemorrhage from guttural pouch mycosis in horses. Vet Surg. 2000;29:389. 126. Freeman D.E., Ross M.W., Donawick W.J., et al. Occlusion of the external carotid and maxillary arteries in the horse to prevent hemorrhage from guttural pouch mycosis. Vet Surg. 1989;18:39. 127. Freeman D.E., Staller G.S., Maxson A.D., et al. Unusual internal carotid artery branching that prevented arterial occlusion with a balloon-tipped catheter in a horse. Vet Surg. 1993;22:531. 128. Freeman D.E., Donawick W.J. Occlusion of internal carotid artery in the horse by means of a balloon-tipped catheter: clinical use of a method to prevent epistaxis caused by guttural pouch mycosis. J Am Vet Med Assoc. 1980;176:236. 129. Caron J.P., Fretz P.B., Bailey J.V., et al. Balloon-tipped catheter arterial occlusion for prevention of hemorrhage caused by guttural pouch mycosis: 13 cases (1982-1985). J Am Vet Med Assoc. 1987;191:345. 130. Hardy J., Robertson J.T., Wilkie D.A. Ischemic optic neuropathy and blindness after arterial occlusion for treatment of guttural pouch mycosis in two horses. J Am Vet Med Assoc. 1990;196:1631. 131. Freeman D.E., Ross M.W., Donawick W.J. Steal phenomenon” proposed as the cause of blindness after arterial occlusion for treatment of guttural pouch mycosis in horses. J Am Vet Med Assoc. 1990;197:811. 132. Bacon Miller C., Wilson D.A., Martin D.D., et al. Complications of balloon catheterization associated with aberrant cerebral arterial anatomy in a horse with guttural pouch mycosis. Vet Surg. 1998;27:450. 133. Judy C.E., Chaffin M.K., Cohen N.D. Empyema of the guttural pouch (auditory tube diverticulum) in horses: 91 cases (1977-1997). J Am Vet Med Assoc. 1999;215:1666. 134. Verheyen K., Newton J.R., Talbot N.C., et al. Elimination of guttural pouch infection and in ammation in asymptomatic carriers of Streptococcus equi. Equine Vet J. 2000;32:527. 135. Seahorn T.L., Schumacher J. Nonsurgical removal of chondroid masses from the guttural pouches of two horses. J Am Vet Med Assoc. 1991;199:368. 136. Gehlen H., Ohnesorge B. Laser fenestration of the mesial septum for treatment of guttural pouch chondroids in a pony. Vet Surg. 2005;34:383. 137. Hawkins J.F., Frank N., Sojka J.E., et al. Fistulation of the auditory tube diverticulum (guttural pouch) with a neodymium:yttrium-aluminum-garnet laser for treatment of chronic empyema in two horses. J Am Vet Med Assoc. 2001;218:405.

138. Hobo S., Matsuda Y., Yoshida K. Prevalence of upper respiratory tract disorders detected with a exible videoendoscope in thoroughbred racehorses. J Vet Med Sci. 1995;57:409. 139. Holcombe S.J., Robinson N.E., Derksen F.J., et al. Effect of tracheal mucus and tracheal cytology on racing performance in Thoroughbred racehorses. Equine Vet J. 2006;38:300. 140. Raker C.W., Boles C.L. Pharyngeal lymphoid hyperplasia in the horse. J Equine Med Surg. 1978;2:202. 141. Holcombe S.J., Jackson C., Gerber V., et al. Stabling is associated with airway inflammation in young Arabian horses. Equine Vet J. 2001;33:244. 142. Blakeslee J.R.Jr., Olsen R.G., McAllister E.S., et al. Evidence of respiratory tract infection induced by equine herpesvirus, type 2, in the horse. Can J Microbiol. 1940;21:1975. 143. Prickett ME: The pathology of disease caused by equine herpesvirus-1. Proceedings of the Second International Conference on Equine Infectious Diseases, Princeton, NJ, 1969. Veterinary Publications. 144. Burrell M.H. Endoscopic and virological observations on respiratory disease in a group of young Thoroughbred horses in training. Equine Vet J. 1985;17:99. 145. Christley R.M., Hodgson D.R., Rose R.J., et al. Coughing in thoroughbred racehorses: risk factors and tracheal endoscopic and cytological findings. Vet Rec. 2001;148:99. 146. Kannegieter N.J., Dore M.L. Endoscopy of the upper respiratory tract during treadmill exercise: a clinical study of 100 horses. Aust Vet J. 1995;72:101. 147. Holcombe S.J., Derksen F.J., Stick J.A., et al. E ect of bilateral blockade of the pharyngeal branch of the vagus nerve on soft palate function in horses. Am J Vet Res. 1998;59:504. 148. Ducharme N.G. Pharynx. In: Auer J.A., Stick J.A., editors. Equine surgery. St Louis: Elsevier Health Sciences, 2006. 149. Dart A.J., Dowling B.A., Hodgson D.R., et al. Evaluation of high-speed treadmill videoendoscopy for diagnosis of upper respiratory tract dysfunction in horses. Aust Vet J. 2001;79:109. 150. Barakzai S.Z., Johnson V.S., Baird D.H., et al. Assessment of the e cacy of composite surgery for the treatment of dorsal displacement of the soft palate in a group of 53 racing Thoroughbreds (1990-1996). Equine Vet J. 2004;36:175. 151. Parente E.J., Martin B.B., Tulleners E.P., et al. Dorsal displacement of the soft palate in 92 horses during high-speed treadmill examination (1993-1998). Vet Surg. 2002;31:507. 152. Lane J.G., Bladon B., Little D.R., et al. Dynamic obstructions of the equine upper respiratory tract. Part 2: comparison of endoscopic ndings at rest and during high-speed treadmill exercise of 600 Thoroughbred racehorses. Equine Vet J. 2006;38:401. 153. Rehder R.S., Ducharme N.G., Hackett R.P., et al. Measurement of upper airway pressures in exercising horses with dorsal displacement of the soft palate. Am J Vet Res. 1995;56:269. 154. Cook W.R. Some observations on form and function of the equine uppe airway in health and disease. Proc Am Assoc Eq Pract. 1982;198:355. 155. Fogel R.B., Trinder J., White D.P., et al. The effect of sleep onset on upper airway muscle activity in patients with sleep apnoea versus controls. J Physiol. 2005;564:549. 156. Gumery P.Y., Roux-Buisson H., Meignen S., et al. An adaptive detector of genioglossus EMG re ex using Berkner transform for time latency measurement in OSA pathophysiological studies. IEEE Trans Biomed Eng. 2005;52:1382. 157. Tessier C., Holcombe S.J., Derksen F.J., et al. Effects of stylopharyngeus muscle dysfunction on the nasopharynx in exercising horses. Equine Vet J. 2004;36:318. 158. Holcombe S.J., Derksen F.J., Stick J.A., et al. Effect of bilateral tenectomy of the tensor veli palatini muscle on soft palate function in horses. Am J Vet Res. 1997;58:317. 159. Ducharme N.G., Hackett R.P., Woodie J.B., et al. Investigations into the role of the thyrohyoid muscles in the pathogenesis of dorsal displacement of the soft palate in horses. Equine Vet J. 2003;35:258. 160. Holcombe S.J., Beard W.L., Hinchcliff K.W., et al. Effect of sternothyrohyoid myectomy on upper airway mechanics in normal horses. J Appl Physiol. 1994;77:2812. 161. Malhotra A., Pillar G., Fogel R.B., et al. Pharyngeal pressure and flow effects on genioglossus activation in normal subjects. Am J Respir Crit Care Med. 2002;165:71. 162. Brennick M.J., Gefter W.B., Margulies S.S. Mechanical e ects of genioglossus muscle stimulation on the pharyngeal airway by MRI in cats. Respir Physiol Neurobiol. 2007;156:154. 163. Franklin S.H., Naylor J.R., Lane J.G., The effect of a tongue-tie in horses with dorsal displacement of the soft palate. Equine Vet J, Suppll (34), 2002, 430. 164. Harrison I.W., Raker C.W. Sternothyrohyoideus myectomy in horses: 17 cases (1984-1985). J Am Vet Med Assoc. 1988;193:1299. 165. Anderson J.D., Tulleners E.P., Johnston J.K., et al. Sternothyrohyoideus myectomy or staphylectomy for treatment of intermittent dorsal displacement of the soft palate in racehorses: 209 cases (1986-1991). J Am Vet Med Assoc. 1995;206:1909. 166. Ahren T.J. Oral palatopharyngoplasty: a survey of one hundred post-operative raced horses. J Equine Vet Sci. 1993;13:670. 167. Barakzai S.Z., Dixon P.M. Conservative treatment for thoroughbred racehorses with intermittent dorsal displacement of the soft palate. Vet Rec. 2005;157:337. 168. Woodie J.B., Ducharme N.G., Kanter P., et al. Surgical advancement of the larynx (laryngeal tie-forward) as a treatment for dorsal displacement of the soft palate in horses: a prospective study 2001-2004. Equine Vet J. 2005;37:418. 169. Woodie J.B., Ducharme N.G., Hackett R.P., et al. Can an external device prevent dorsal displacement of the soft palate during strenuous exercise? Equine Vet J. 2005;37:425. 170. Goulden B.E., Anderson L.G. Equine laryngeal hemiplegia. Part III. N Z Vet J. 1982;30:1. 171. Lane J.G., Ellis D.R., Greet T.R. Observations on the examination of Thoroughbred yearlings for idiopathic laryngeal hemiplegia. Equine Vet J. 1987;19:531. 172. Dixon P.M., McGorum B.C., Railton D.I., et al. Laryngeal paralysis: a study of 375 cases in a mixed-breed population of horses. Equine Vet J. 2001;33:452. 173. Brown J.A., Hinchcli K.W., Jackson M.A., et al. Prevalence of pharyngeal and laryngeal abnormalities in Thoroughbreds racing in Australia, and their association with performance. Equine Vet J. 2005;37:397. 174. Brakenhoff J.E., Holcombe S.J., Hauptman J.G., et al. The prevalence of laryngeal disease in a large population of competition draft horses. Vet Surg. 2006;35:579. 175. Harrison G.D., Duncan I.D., Clayton M.K. Determination of the early age of onset of equine recurrent laryngeal neuropathy. 1. Muscle pathology. Acta Neuropathol. 1992;84:307. 176. Dixon P, Robinson E, Wade JF. 2003. Workshop summary. Proceedings of a Workshop on Equine Recurrent Laryngeal Neuropathy. Stratford-upon-Avon, United Kingdom, R & W Publications (Newmarket) Ltd. 177. Rakestraw P.C., Hackett R.P., Ducharme N.G., et al. Arytenoid cartilage movement in resting and exercising horses. Vet Surg. 1991;20:122. 178. Ducharme N. 2003. 4-grade system of equine laryngeal function. Proceedings of a Workshop on Equine Recurrent Laryngeal Neuropathy. Stratford-upon-Avon, United KingdomR & W Publications (Newmarket) Ltd. 179. Hammer E.J., Tulleners E.P., Parente E.J., et al. Videoendoscopic assessment of dynamic laryngeal function during exercise in horses with grade-III left laryngeal hemiparesis at rest: 26 cases (1992-1995). J Am Vet Med Assoc. 1998;212:399. 180. Chalmers H.J., Cheetham J., Yeager A.E., et al. Ultrasonography of the equine larynx. Vet Radiol Ultrasound. 2006;47:476. 181. Chalmers HJ, Cheetham J, Mohammed HO, et al. Ultrasonography as an aid in the diagnosis of recurrent laryngeal neuropathy in horses. Proceedings 2006 ACVS Veterinary Symposium. 2006. 182. Quinlan T.J., Goulden B.E., Barnes G.R., et al. Innervation of the equine intrinsic laryngeal muscles. N Z Vet J. 1982;30:43.

183. Duncan I.D., Griffths I.R., McQueen A., et al. The pathology of equine laryngeal hemiplegia. Acta Neuropathol. 1974;27:337. 184. Gunn H.M. Histochemical observations on laryngeal skeletal muscle fibres in ‘normal’ horses. Equine Vet J. 1972;4:144. 185. Cahill J.I., Goulden B.E. Equine laryngeal hemiplegia. Part I. A light microscopic study of peripheral nerves. N Z Vet J. 1986;34:161. 186. Dattilo D.J., Drooger S.A. Outcome assessment of patients undergoing maxillofacial procedures for the treatment of sleep apnea: comparison of subjective and objective results. J Oral Maxillofac Surg. 2004;62:164. 187. Curtis R.A., Hahn C.N., Evans D.L., et al. Thoracolaryngeal reflex latencies in Thoroughbred horses with recurrent laryngeal neuropathy. Vet J. 2005;170:67. 188. Cole C.R. Changes in the equine larynx associated with laryngeal hemiplegia. Am J Vet Res. 1946;7:69. 189. Derksen F.J., Stick J.A., Scott E.A., et al. Effect of laryngeal hemiplegia and laryngoplasty on airway flow mechanics in exercising horses. Am J Vet Res. 1986;47:16. 190. Brown J.A., Derksen F.J., Stick J.A., et al. Laser vocal cordectomy fails to effectively reduce respiratory noise in horses with laryngeal hemiplegia. Vet Surg. 2005;34:247. 191. Shappell K.K., Derksen F.J., Stick J.A., et al. E ects of ventriculectomy, prosthetic laryngoplasty, and exercise on upper airway function in horses with induced left laryngeal hemiplegia. Am J Vet Res. 1988;49:1760. 192. Tetens J., Derksen F.J., Stick J.A., et al. E cacy of prosthetic laryngoplasty with and without bilateral ventriculocordectomy as treatments for laryngeal hemiplegia in horses. Am J Vet Res. 1996;57:1668. 193. Ehrlich P.J., Seeherman H.J., Morris E., et al. The e ect of reversible left recurrent laryngeal neuropathy on the metabolic cost of locomotion and peak aerobic power in thoroughbred racehorses. Vet Surg. 1995;24:36. 194. Kidd J.A., Slone D.E. Treatment of laryngeal hemiplegia in horses by prosthetic laryngoplasty, ventriculectomy and vocal cordectomy. Vet Rec. 2002;150:481. 195. Hawkins J.F., Tulleners E.P., Ross M.W., et al. Laryngoplasty with or without ventriculectomy for treatment of left laryngeal hemiplegia in 230 racehorses. Vet Surg. 1997;26:484. 196. Kraus B.M., Parente E.J., Tulleners E.P. Laryngoplasty with ventriculectomy or ventriculocordectomy in 104 draft horses (1992-2000). Vet Surg. 2003;32:530. 197. Dixon R.M., McGorum B.C., Railton D.I., et al. Long-term survey of laryngoplasty and ventriculocordectomy in an older, mixed-breed population of 200 horses. Part 1: Maintenance of surgical arytenoid abduction and complications of surgery. Equine Vet J. 2003;35:389. 198. Brown J.A., Derksen F.J., Stick J.A., et al. Effect of laryngoplasty on respiratory noise reduction in horses with laryngeal hemiplegia. Equine Vet J. 2004;36:420. 199. Schumacher J., Wilson A.M., Pardoe C., et al. In vitro evaluation of a novel prosthesis for laryngoplasty of horses with recurrent laryngeal neuropathy. Equine Vet J. 2000;32:43. 200. Radcli e C.H., Woodie J.B., Hackett R.P., et al. A comparison of laryngoplasty and modi ed partial arytenoidectomy as treatments for laryngeal hemiplegia in exercising horses. Vet Surg. 2006;35:643. 201. Davenport C.L., Tulleners E.P., Parente E.J. The e ect of recurrent laryngeal neurectomy in conjunction with laryngoplasty and unilateral ventriculocordectomy in thoroughbred racehorses. Vet Surg. 2001;30:417. 202. Strand E., Martin G.S., Haynes P.F., et al. Career racing performance in Thoroughbreds treated with prosthetic laryngoplasty for laryngeal neuropathy: 52 cases (1981-1989). J Am Vet Med Assoc. 2000;217:1689. 203. Russell A.P., Slone D.E. Performance analysis after prosthetic laryngoplasty and bilateral ventriculectomy for laryngeal hemiplegia in horses: 70 cases (1986-1991). J Am Vet Med Assoc. 1994;204:1235. 204. Dixon P.M., McGorum B.C., Railton D.I., et al. Long-term survey of laryngoplasty and ventriculocordectomy in an older, mixed-breed population of 200 horses. Part 2: Owners’ assessment of the value of surgery. Equine Vet J. 2003;35:397. 205. Ducharme N.G., Horney F.D., Partlow G.D., et al. Attempts to restore abduction of the paralyzed equine arytenoid cartilage. I. Nerve-muscle pedicle transplants. Can J Vet Res. 1989;53:202. 206. Fulton I.C., Derksen F.J., Stick J.A., et al. Treatment of left laryngeal hemiplegia in standardbreds, using a nerve muscle pedicle graft. Am J Vet Res. 1991;52:1461. 207. Fulton I.C., Derksen F.J., Stick J.A., et al. Histologic evaluation of nerve muscle pedicle graft used as a treatment for left laryngeal hemiplegia in standardbreds. Am J Vet Res. 1992;53:592. 208. Barnes A.J., Slone D.E., Lynch T.M. Performance after partial arytenoidectomy without mucosal closure in 27 Thoroughbred racehorses. Vet Surg. 2004;33:398. 209. Russell T., Wainscott M. Treatment in the field of 27 horses with epiglottic entrapment. Vet Rec. 2007;161:187. 210. McClure S.R., Schumacher J., Snyder J.R. Transnasal incision of restrictive nasopharyngeal cicatrix in three horses. J Am Vet Med Assoc. 1994;205:461. 211. Perkins J.D., Hughes T.K., Brain B. Endoscope-guided, transoral axial division of an entrapping epiglottic fold in fifteen standing horses. Vet Surg. 2007;36:800. 212. Lumsden J.M., Stick J.A., Caron J.P., et al. Surgical treatment for epiglottic entrapment in horses: 51 cases (1981-1992). J Am Vet Med Assoc. 1994;205:729. 213. Ross M.W., Gentile D.G., Evans L.E. Transoral axial division, under endoscopic guidance, for correction of epiglottic entrapment in horses. J Am Vet Med Assoc. 1993;203:416. 214. Greet T.R. Experiences in treatment of epiglottal entrapment using a hook knife per nasum. Equine Vet J. 1995;27:122. 215. Tate L.P., Sweeney C.L., Bowman K.F., et al. Transendoscopic Nd:YAG laser surgery for treatment of epiglottal entrapment and dorsal displacement of the soft palate in the horse. Vet Surg. 1990;19:356. 216. Tulleners E.P. Correlation of performance with endoscopic and radiographic assessment of epiglottic hypoplasia in racehorses with epiglottic entrapment corrected by use of contact neodymium:yttrium aluminum garnet laser. J Am Vet Med Assoc. 1991;198:621. 217. Tulleners E.P. Transendoscopic contact neodymium:yttrium aluminum garnet laser correction of epiglottic entrapment in standing horses. J Am Vet Med Assoc. 1990;196:1971. 218. Haynes P.F., Snider T.G., McClure J.R., et al. Chronic chondritis of the equine arytenoid cartilage. J Am Vet Med Assoc. 1980;177:1135. 219. Behrens E., Pinero L. Laryngeal chondropathy in a group of broodmares. J Equine Vet Sci. 1990;10:113. 220. Kelly G., Lumsden J.M., Dunkerly G., et al. Idiopathic mucosal lesions of the arytenoid cartilages of 21 Thoroughbred yearlings: 1997-2001. Equine Vet J. 2003;35:276. 221. Dean P.W., Cohen N.D. Arytenoidectomy for advanced unilateral chondropathy with accompanying lesions. Vet Surg. 1990;19:364. 222. Hay W.P., Tulleners E. Excision of intralaryngeal granulation tissue in 25 horses using a neodymium:YAG laser (1986 to 1991). Vet Surg. 1993;22:129. 223. Smith R.L., Perkins N.R., Firth E.C., et al. Arytenoid mucosal injury in young Thoroughbred horses: investigation of a proposed aetiology and clinical signi cance. N Z Vet J. 2006;54:173. 224. MacLean A.A., Robertson-Smith R.G. Chronic chondritis of the arytenoid cartilages in a pony mare. Aust Vet J. 1984;61:27. 225. Terry C., Shumpert K., Rashmir-Raven A.M., et al. An unusual case of upper respiratory obstruction in a horse. Vet Radiol Ultrasound. 2002;43:43. 226. Haynes P.F. Dorsal displacement of the soft palate and epiglottic entrapment: diagnosis, management and interrelationships. Compend Contin Educ Pract Vet. 1983;5:S379. 227. Blikslager A.T., Tate L.P., Tudor R. Transendoscopic laser treatment of rostral displacement of the palatopharyngeal arch in four horses. J Clin Laser Med Surg. 1999;17:49. 228. Klein H.J., Deegen E., Stockhofe N., et al. Rostral displacement of the palatopharyngeal arch in a seven-month-old Hanoverian colt. Equine Vet J. 1989;21:382. 229. Goulden B.E., Anderson L.J., Davies A.S., et al. Rostral displacement of the palatopharyngeal arch: a case report. Equine Vet J. 1976;8:95.

230. Crabill M., Schumacher J., Walker M. What is your diagnosis? Rostral displacement of the palatopharyngeal arch. J Am Vet Med Assoc. 1994;204:1347. 231. Cook W.R. Some observations on diseases of the ear, nose and throat in the horse, and endoscopy using a flexible fibreoptic endoscope. Vet Rec. 1974;94:533. 232. Schumacher J., Hanselka D.V. Nasopharyngeal cicatrices in horses: 47 cases (1972-1985). J Am Vet Med Assoc. 1987;191:239. 233. Chesen A.B., Rakestraw P.C. Indications for and short- and long-term outcome of permanent tracheostomy performed in standing horses: 82 cases (1995-2005). J Am Vet Med Assoc. 2008;232:1352. 234. Timoney J.F. Strangles, Vet Clin North Am Equine Pract. 1993;9:365. 235. Timoney J.F., Artiushin S.C., Boschwitz J.S. Comparison of the sequences and functions of Streptococcus equi M-like proteins SeM and SzPSe. Infect Immun. 1997;65:3600. 236. Jorm L.R., Love D.N., Bailey G.D., et al. Genetic structure of populations of beta-haemolytic Lance eld group C streptococci from horses and their association with disease. Res Vet Sci. 1994;57:292. 237. Sweeney C.R., Benson C.E., Whitlock R.H. Streptococcus equi infection in horses: Part I. Compend Contin Educ Pract Vet. 1987;9:689. 238. Prescott J.F., Srivastava S.K., deGannes R., et al. A mild form of strangles caused by an atypical Streptococcus equi. J Am Vet Med Assoc. 1982;180:293. 239. Grant S.T., Efstratiou A., Chanter N. Laboratory diagnosis of strangles and the isolation of atypical Streptococcus equi. Vet Rec. 1993;133:215. 240. Artiushin S., Timoney J.F. PCR for detection of Streptococcus equi. Adv Exp Med Biol. 1997;418:359-361. 241. Harrington D.J., Sutcliffe I.C., Chanter N. The molecular basis of Streptococcus equi infection and disease. Microbes Infect. 2002;4:501. 242. Galan J.E., Timoney J.F., Lengemann F.W. Passive transfer of mucosal antibody to Streptococcus equi in the foal. Infect Immun. 1986;54:202. 243. Sweeney C.R., Benson C.E., Whitlock R.H., et al. Description of an epizootic and persistence of Streptococcus equi infections in horses. J Am Vet Med Assoc. 1989;194:1281. 244. Todd T.G. Strangles. J Comp Pathol Ther. 1910;23:212. 245. Timoney J.F. The pathogenic equine streptococci. Vet Res. 2004;35:397. 246. Hamlen H.J., Timoney J.F., Bell R.J. Epidemiologic and immunologic characteristics of Streptococcus equi infection in foals. J Am Vet Med Assoc. 1994;204:768. 247. Piche C.A. Clinical observations on an outbreak of strangles. Can Vet J. 1984;25:7. 248. George J.L., Reif J.S., Shideler R.K., et al. Identification of carriers of Streptococcus equi in a naturally infected herd. J Am Vet Med Assoc. 1983;183:80. 249. Newton J.R., Wood J.L., Dunn K.A., et al. Naturally occurring persistent and asymptomatic infection of the guttural pouches of horses with Streptococcus equi. Vet Rec. 1997;140:84. 250. Waller A.S., Jolley K.A. Getting a grip on strangles: recent progress towards improved diagnostics and vaccines. Vet J. 2007;173:492. 251. Sweeney C.R., Timoney J.F., Newton J.R., et al. Streptococcus equi infections in horses: guidelines for treatment, control, and prevention of strangles. J Vet Intern Med. 2005;19:123. 252. Jorm LR. Factors affecting the survival of Streptococcus equi subsp equi. Sixth International Conference on Equine Infectious Disease. Newmarket, England, 1991. 253. Yigezu L.M., Roger F., Kiredjian M., et al. Isolation of Streptococcus equi subspecies equi (strangles agent) from an Ethiopian camel. Vet Rec. 1997;140:608. 254. Breiman R.F., Silverblatt F.J. Systemic Streptococcus equi infection in a horse handler: a case of human strangles. West J Med. 1986;145:385. 255. Evers W.D. Effect of furaltadone on strangles in horses. J Am Vet Med Assoc. 1968;152:1394. 256. Muhktar M.M., Timoney J.F. Chemotactic response of equine polymorphonuclear leucocytes to Streptococcus equi. Res Vet Sci. 1988;45:225. 257. Hamilton A., Robinson C., Sutcli e I.C., et al. Mutation of the maturase lipoprotein attenuates the virulence of Streptococcus equi to a greater extent than does loss of general lipoprotein lipidation. Infect Immun. 2006;74:6907. 258. Tiwari R., Qin A., Artiushin S., et al. Se18.9, an anti-phagocytic factor H binding protein of Streptococcus equi. Vet Microbiol. 2007;121:105. 259. Anzai T., Timoney J.F., Kuwamoto Y., et al. In vivo pathogenicity and resistance to phagocytosis of Streptococcus equi strains with di erent levels of capsule expression. Vet Microbiol. 1999;67:277. 260. Kelly C., Bugg M., Robinson C., et al. Sequence variation of the SeM gene of Streptococcus equi allows discrimination of the source of strangles outbreaks. J Clin Microbiol. 2006;44:480. 261. Anzai T., Sheoran A.S., Kuwamoto Y., et al. Streptococcus equi but not Streptococcus zooepidemicus produces potent mitogenic responses from equine peripheral blood mononuclear cells. Vet Immunol Immunopathol. 1999;67:235. 262. Sheoran A.S., Sponseller B.T., Holmes M.A., et al. Serum and mucosal antibody isotype responses to M-like protein (SeM) of Streptococcus equi in convalescent and vaccinated horses. Vet Immunol Immunopathol. 1997;59:239. 263. Timoney J.F., Timoney P.J., Strickland K.L. Lysogeny and the immunologically reactive proteins of Streptococcus equi. Vet Rec. 1984;115:148. 264. Newton J.R., Verheyen K., Talbot N.C., et al. Control of strangles outbreaks by isolation of guttural pouch carriers identi ed using PCR and culture of Streptococcus equi. Equine Vet J. 2000;32:515. 265. Sweeney C.R. Strangles: Streptococcus equi infection in horses. Equine Vet Educ. 1996;8:317. 266. Ramey D. Does early antibiotic use in horses with ‘strangles’ cause metastatic Streptococcus equi bacterial infections? Equine Vet Educ. 2007;19:14. 267. Sweeney C.R., Benson C.E., Whitlock R.H., et al. Streptococcus equi infection in horses. Part 2. Compend Contin Educ Pract Vet. 1987;9:845. 268. Rumbaugh G.E., Smith B.P., Carlson G.P. Internal abdominal abscesses in the horse: a study of 25 cases. J Am Vet Med Assoc. 1978;172:304. 269. Spoormakers T.J., Ensink J.M., Goehring L.S., et al. Brain abscesses as a metastatic manifestation of strangles: symptomatology and the use of magnetic resonance imaging as a diagnostic aid. Equine Vet J. 2003;35:146. 270. Koblik P.D., Lofstedt J., Jakowski R.M., et al. Use of 111In-labeled autologous leukocytes to image an abdominal abscess in a horse. J Am Vet Med Assoc. 1985;186:1319. 271. Pusterla N., Whitcomb M.B., Wilson W.D. Internal abdominal abscesses caused by Streptococcus equi subspecies equi in 10 horses in California between 1989 and 2004. Vet Rec. 2007;160:589. 272. Zicker S.C., Wilson W.D., Medearis I. Di erentiation between intra-abdominal neoplasms and abscesses in horses, using clinical and laboratory data: 40 cases (1973-1988). J Am Vet Med Assoc. 1990;196:1130. 273. Pusterla N., Watson J.L., Affolter V.K., et al. Purpura haemorrhagica in 53 horses. Vet Rec. 2003;153:118. 274. Heath S.E., Geor R.J., Tabel H., et al. Unusual patterns of serum antibodies to Streptococcus equi in two horses with purpura hemorrhagica. J Vet Intern Med. 1991;5:263. 275. Galan J.E., Timoney J.F. Immune complexes in purpura hemorrhagica of the horse contain IgA and M antigen of Streptococcus equi. J Immunol. 1985;135:3134. 276. Yelle M.T. Clinical aspects of Streptococcus equi infection. Equine Vet J. 1987;19:158. 277. Todhunter R.J., Brown C.M., Stickle R. Retropharyngeal infections in five horses. J Am Vet Med Assoc. 1985;187:600.

278. Golland L.C., Hodgson D.R., Davis R.E., et al. Retropharyngeal lymph node infection in horses: 46 cases (1977-1992). Aust Vet J. 1995;72:161. 279. De Clercq D., van Loon G., Nollet H., et al. Percutaneous puncture technique for treating persistent retropharyngeal lymph node infections in seven horses. Vet Rec. 2003;152:169. 280. Rigg D.L., Ramey D.W., Reinertson E.L. Tracheal compression secondary to abscessation of cranial mediastinal lymph nodes in a horse. J Am Vet Med Assoc. 1985;186:283. 281. Bell R.J., Smart M.E. An unusual complication of strangles in a pony. Can Vet J. 1992;33:400. 282. Finno C., Pusterla N., Aleman M., et al. Streptococcus equi meningoencephalomyelitis in a foal. J Am Vet Med Assoc. 2006;229:721. 283. Kaese H.J., Valberg S.J., Hayden D.W., et al. Infarctive purpura hemorrhagica in five horses. J Am Vet Med Assoc. 2005;226:1893. 284. Sponseller BT, Valberg SJ, Tennent-Brown BS, et al: Severe acute rhabdomyolysis associated with Streptococcus equi infection in four horses, J Am Vet Med Assoc 227:1800, 1753. 285. Valaberg SJ, Bullock P, Hgetvedt W, et al. Muopathies associated with Streptococcus equi infections in horses. Proceedings of the American Association of Equine Practitioners, 1996. 286. Galan J.E., Timoney J.F. Mucosal nasopharyngeal immune responses of horses to protein antigens of Streptococcus equi. Infect Immun. 1985;47:623. 287. Wood J.M. Antigenic variation of equine influenza: a stable virus. Equine Vet J. 1988;20:316. 288. Kemp-Symonds J., Kemble T., Waller A. Modi ed live Streptococcus equi (‘strangles’) vaccination followed by clinically adverse reactions associated with bacterial replication. Equine Vet J. 2007;39:284. 289. Prescott J.F., Timoney J.F. Could we eradicate strangles in equids? J Am Vet Med Assoc. 2007;231:377. 290. Newton R., Waller A., King A. Investigation of suspected adverse reactions following strangles vaccination in horses. Vet Rec. 2005;156:291. 291. Morley P.S., Townsend H.G., Bogdan J.R., et al. Descriptive epidemiologic study of disease associated with in uenza virus infections during three epidemics in horses. J Am Vet Med Assoc. 2000;216:535. 292. Newton J.R., Verheyen K., Wood J.L., et al. Equine influenza in the United Kingdom in 1998. Vet Rec. 1999;145:449. 293. Martella V., Elia G., Decaro N., et al. An outbreak of equine in uenza virus in vaccinated horses in Italy is due to an H3N8 strain closely related to recent North American representatives of the Florida sub-lineage. Vet Microbiol. 2007;121:56. 294. Minke J.M., Toulemonde C.E., Coupier H., et al. E cacy of a canarypox-vectored recombinant vaccine expressing the hemagglutinin gene of equine in uenza H3N8 virus in the protection of ponies from viral challenge. Am J Vet Res. 2007;68:213. 295. Higgins W.P., Gillespie J.H., Holmes D.R., et al. Surveys of equine influenze outbreaks during 1983 and 1984. J Equine Vet Sci. 1986;6:15. 296. Kemen M.J., Frank R.A., Babish J.B. An outbreak of equine influenza at a harness horse racetrack. Cornell Vet. 1985;75:277. 297. Crawford P.C., Dubovi E.J., Castleman W.L., et al. Transmission of equine influenza virus to dogs. Science. 2005;310:482. 298. Daly J.M. Equine influenza in dogs: too late to bolt the stable door? Vet J. 2006;171:7. 299. McChesney A.E. Viral respiratory infections of horses: structure and function of lungs in relation to viral infection. J Am Vet Med Assoc. 1975;166:76. 300. Carr C.M., Kim P.S. Flu virus invasion: halfway there. Science. 1994;266:234. 301. Paillot R., Hannant D., Kydd J.H., et al. Vaccination against equine influenza: quid novi? Vaccine. 2006;24:4047. 302. Chambers T.M., Holland R.E., Lai A. Equine influenza: current veterinary perspectives, part 1. Equine Pract. 1995;17:19. 303. Nelson K.M., Schram B.R., McGregor M.W., et al. Local and systemic isotype-speci c antibody responses to equine in uenza virus infection versus conventional vaccination. Vaccine. 1998;16:1306. 304. Gross D.K., Morley P.S., Hinchcli K.W., et al. Pulmonary ultrasonographic abnormalities associated with naturally occurring equine in uenza virus infection in standardbred racehorses. J Vet Intern Med. 2004;18:718. 305. Gross D.K., Hinchcli K.W., French P.S., et al. E ect of moderate exercise on the severity of clinical signs associated with in uenza virus infection in horses. Equine Vet J. 1998;30:489. 306. Chambers T.M., Holland R.E., Lai A. Equine influenza: current veterinary perspectives, Part 2. Equine Pract. 1995;27:26. 307. Britton A.P., Robinson J.H. Isolation of influenza A virus from a 7-day-old foal with bronchointerstitial pneumonia. Can Vet J. 2002;43:55. 308. Peek S.F., Landolt G., Karasin A.I., et al. Acute respiratory distress syndrome and fatal interstitial pneumonia associated with equine in uenza in a neonatal foal. J Vet Intern Med. 2004;18:132. 309. van Maanen C., van Essen G.J., Minke J., et al. Diagnostic methods applied to analysis of an outbreak of equine in uenza in a riding school in which vaccine failure occurred. Vet Microbiol. 2003;93:291. 310. Garin B., Plateau E., Gillet-Forin S. Serological diagnosis of in uenza A infections in the horse by enzyme immunoassay. Comparison with the complement xation test. Vet Immunol Immunopathol. 1986;13:357. 311. Yamanaka T., Tsujimura K., Kondo T., et al. In vitro efficacies of oseltamivir carboxylate and zanamivir against equine influenza A viruses. J Vet Med Sci. 2006;68:405. 312. Yamanaka T., Tsujimura K., Kondo T., et al. Efficacy of oseltamivir phosphate to horses inoculated with equine influenza A virus. J Vet Med Sci. 2006;68:923. 313. Yamanaka T., Yamada M., Tsujimura K., et al. Clinical pharmaco*kinetics of oseltamivir and its active metabolite oseltamivir carboxylate after oral administration in horses. J Vet Med Sci. 2007;69:293. 314. Baker D.J. Rationale for the use of influenza vaccines in horses and the importance of antigenic drift. Equine Vet J. 1986;18:93. 315. Morley P.S., Townsend H.G., Bogdan J.R., et al. E cacy of a commercial vaccine for preventing disease caused by in uenza virus infection in horses. J Am Vet Med Assoc. 1999;215:61. 316. Newton J.R., Daly J.M., Spencer L., et al. Description of the outbreak of equine in uenza (H3N8) in the United Kingdom in 2003, during which recently vaccinated horses in Newmarket developed respiratory disease. Vet Rec. 2006;158:185. 317. Soboll G., Horohov D.W., Aldridge B.M., et al. Regional antibody and cellular immune responses to equine in uenza virus infection, and particle mediated DNA vaccination. Vet Immunol Immunopathol. 2003;94:47. 318. Lunn D.P., Soboll G., Schram B.R., et al. Antibody responses to DNA vaccination of horses using the influenza virus hemagglutinin gene. Vaccine. 1999;17:2245. 319. Edlund Toulemonde C., Daly J., Sindle T., et al. E cacy of a recombinant equine in uenza vaccine against challenge with an American lineage H3N8 in uenza virus responsible for the 2003 outbreak in the United Kingdom. Vet Rec. 2005;156:367. 320. Paillot R., Kydd J.H., Sindle T., et al. Antibody and IFN-gamma responses induced by a recombinant canarypox vaccine and challenge infection with equine in uenza virus. Vet Immunol Immunopathol. 2006;112:225.

321. Paillot R., Kydd J.H., MacRae S., et al. New assays to measure equine influenza virus-specific Type 1 immunity in horses. Vaccine. 2007;25:7385. 322. Chambers T.M., Holland R.E., Tudor L.R., et al. A new modi ed live equine in uenza virus vaccine: phenotypic stability, restricted spread and e cacy against heterologous virus challenge. Equine Vet J. 2001;33:630. 323. Lunn D.P., Hussey S., Sebing R., et al. Safety, e cacy, and immunogenicity of a modi ed-live equine in uenza virus vaccine in ponies after induction of exercise-induced immunosuppression. J Am Vet Med Assoc. 2001;218:900. 324. van Maanen C., Bruin G., de Boer-Luijtze E., et al. Interference of maternal antibodies with the immune response of foals after vaccination against equine in uenza. Vet Q. 1992;14:13. 325. Wilson W.D., Mihalyi J.E., Hussey S., et al. Passive transfer of maternal immunoglobulin isotype antibodies against tetanus and in uenza and their e ect on the response of foals to vaccination. Equine Vet J. 2001;33:644. 326. Daly J.M., Whitwell K.E., Miller J., et al. Investigation of equine influenza cases exhibiting neurological disease: coincidence or association? J Comp Pathol. 2006;134:231. 327. Kleiboeker S.B., Schommer S.K., Johnson P.J., et al. Association of two newly recognized herpesviruses with interstitial pneumonia in donkeys (Equus asinus). J Vet Diagn Invest. 2002;14:273. 328. Field H.J., Biswas S., Mohammad I.T. Herpesvirus latency and therapy—from a veterinary perspective. Antiviral Res. 2006;71:127. 329. Patel J.R., Heldens J. Equine herpesviruses 1 (EHV-1) and 4 (EHV-4)—epidemiology, disease and immunoprophylaxis: a brief review. Vet J. 2005;170:14. 330. Blunden A.S., Smith K.C., Binns M.M., et al. Replication of equid herpesvirus 4 in endothelial cells and synovia of a eld case of viral pneumonia and synovitis in a foal. J Comp Pathol. 1995;112:133. 331. Browning G.F., Ficorilli N., Studdert M.J. Asinine herpesvirus genomes: comparison with those of the equine herpesviruses. Arch Virol. 1988;101:183. 332. Kleiboeker S.B., Turnquist S.E., Johnson P.J., et al. Detection and nucleotide sequencing of a DNA-packaging protein gene of equine gammaherpesviruses. J Vet Diagn Invest. 2004;16:67. 333. van Maanen C. Equine herpesvirus 1 and 4 infections: an update. Vet Q. 2002;24:58. 334. Telford E.A., Watson M.S., Perry J., et al. The DNA sequence of equine herpesvirus-4. J Gen Virol. 1998;79:1197. Pt 5 335. Kinyili J.H., Thorsen J. Antigenic comparisons between herpesviruses isolated from fallow deer in Alberta and the viruses of infectious bovine rhinotracheitis, equine rhinopneumonitis and DN-599, a non-IBR bovine herpesvirus. J Wildl Dis. 1979;15:339. 336. Rebhun W.C., Jenkins D.H., Riis R.C., et al. An epizootic of blindness and encephalitis associated with a herpesvirus indistinguishable from equine herpesvirus I in a herd of alpacas and llamas. J Am Vet Med Assoc. 1988;192:953. 337. Bildfell R., Yason C., Haines D., et al. Herpesvirus encephalitis in a camel (Camelus bactrianus). J Zoo Wildlife Med. 1996;27:409. 338. Pagamjav O., Yamada S., Ibrahim el-S.M., et al. Molecular characterization of equine herpesvirus 1 (EHV-1) isolated from cattle indicating no speci c mutations associated with the interspecies transmission. Microbiol Immunol. 2007;51:313. 339. Nugent J., Birch-Machin I., Smith K.C., et al. Analysis of equid herpesvirus 1 strain variation reveals a point mutation of the DNA polymerase strongly associated with neuropathogenic versus nonneuropathogenic disease outbreaks. J Virol. 2006;80:4047. 340. Dynon K., Black W.D., Ficorilli N., et al. Detection of viruses in nasal swab samples from horses with acute, febrile, respiratory disease using virus isolation, polymerase chain reaction and serology. Aust Vet J. 2007;85:46. 341. Bryans J.T., Allen G.P. Herpesviral disease of the horse. In: Wittman G., editor. Herpes virus disease of cattle, horses and pigs. Boston: Kluvar, 1989. 342. O’Keefe J.S., Alley M.R., Jones D., et al. Neonatal mortality due to equid herpesvirus 4 (EHV-4) in a foal. Aust Vet J. 1995;72:353. 343. Thein P., Darai G., Janssen W., et al. Recent information about the etiopathogenesis of paretic-paralytic forms of herpesvirus infection in horses. Tierarztl Prax. 1993;21:445. 344. Verheyen K., Newton J.R., Wood J.L., et al. Possible case of EHV-4 ataxia in warmblood mare. Vet Rec. 1998;143:456. 345. van Maanen C., Vreeswijk J., Moonen P., et al. Di erentiation and genomic and antigenic variation among fetal, respiratory, and neurological isolates from EHV1 and EHV4 infections in The Netherlands. Vet Q. 2000;22:88. 346. Slater J.D., Borchers K., Thackray A.M., et al. The trigeminal ganglion is a location for equine herpesvirus 1 latency and reactivation in the horse. J Gen Virol. 1994;75:2007. Pt 8 347. Edington N., Welch H.M., Griffiths L. The prevalence of latent Equid herpesviruses in the tissues of 40 abattoir horses. Equine Vet J. 1994;26:140. 348. Welch H.M., Bridges C.G., Lyon A.M., et al. Latent equid herpesviruses 1 and 4: detection and distinction using the polymerase chain reaction and co-cultivation from lymphoid tissues. J Gen Virol. 1992;73:261. Pt 2 349. Pusterla N., Leutenegger C.M., Wilson W.D., et al. Equine herpesvirus-4 kinetics in peripheral blood leukocytes and nasopharyngeal secretions in foals using quantitative real-time TaqMan PCR. J Vet Diagn Invest. 2005;17:578. 350. Foote C.E., Love D.N., Gilkerson J.R., et al. Detection of EHV-1 and EHV-4 DNA in unweaned Thoroughbred foals from vaccinated mares on a large stud farm. Equine Vet J. 2004;36:341. 351. Foote C.E., Love D.N., Gilkerson J.R., et al. EHV-1 and EHV-4 infection in vaccinated mares and their foals. Vet Immunol Immunopathol. 2006;111:41. 352. Brown J.A., Mapes S., Ball B.A., et al. Prevalence of equine herpesvirus-1 infection among Thoroughbreds residing on a farm on which the virus was endemic. J Am Vet Med Assoc. 2007;231:577. 353. Wang L., Raidal S.L., Pizzirani A., et al. Detection of respiratory herpesviruses in foals and adult horses determined by nested multiplex PCR. Vet Microbiol. 2007;121:18. 354. Dolby C.A., Hannant D., Mumford J.A. Response of ponies to adjuvanted EHV-1 whole virus vaccine and challenge with virus of the hom*ologous strain. Br Vet J. 1995;151:27. 355. Campbell T.M., Studdert M.J. Equine herpesvirus type 1 (EHV1). Vet Bull. 1983;53:135. 356. Kydd J.H., Smith K.C., Hannant D., et al. Distribution of equid herpesvirus-1 (EHV-1) in respiratory tract associated lymphoid tissue: implications for cellular immunity. Equine Vet J. 1994;26:470. 357. Frampton A.R.Jr., Stolz D.B., Uchida H., et al. Equine herpesvirus 1 enters cells by two di erent pathways, and infection requires the activation of the cellular kinase ROCK1. J Virol. 2007;81:108. 358. Sutton G.A., Viel L., Carman P.S., et al. Pathogenesis and clinical signs of equine herpesvirus-1 in experimentally infected ponies in vivo. Can J Vet Res. 1998;62:49. 359. Tearle J.P., Smith K.C., Boyle M.S., et al. Replication of equid herpesvirus-1 (EHV-1) in the testes and epididymides of ponies and venereal shedding of infectious virus. J Comp Pathol. 1996;115:385. 360. Hebia I., Fiéni F., Duchamp G., et al. Potential risk of equine herpes virus 1 (EHV-1) transmission by equine embryo transfer. Theriogenology. 2007;67:1485. 361. Slater J.D., Lunn D.P., Horohov D.W., et al. Report of the equine herpesvirus-1 Havermeyer Workshop, San Gimignano, Tuscany, June 2004. Vet Immunol Immunopathol. 2006;111:3. 362. Hamir A.N., Vaala W., Heyer G., et al. Disseminated equine herpesvirus-1 infection in a two-year-old filly. J Vet Diagn Invest. 1994;6:493.

363. Del Piero F., Wilkins P.A., Timoney P.J., et al. Fatal nonneurological EHV-1 infection in a yearling filly. Vet Pathol. 2000;37:672. 364. Mason D.K., Watkins K.L., McNie J.T., et al. Haematological measurements as an aid to early diagnosis and prognosis of respiratory viral infections in thoroughbred horses. Vet Rec. 1990;126:359. 365. Sharma P.C., Cullinane A.A., Onions D.E., et al. Diagnosis of equid herpesviruses-1 and -4 by polymerase chain reaction. Equine Vet J. 1992;24:20. 366. Hussey S.B., Clark R., Lunn K.F., et al. Detection and quanti cation of equine herpesvirus-1 viremia and nasal shedding by real-time polymerase chain reaction. J Vet Diagn Invest. 2006;18:335. 367. Allen G.P. Development of a real-time polymerase chain reaction assay for rapid diagnosis of neuropathogenic strains of equine herpesvirus-1. J Vet Diagn Invest. 2007;19:69. 368. Gradil C., Joo H.S. A radial immunodiffusion enzyme assay for detection of antibody to equine rhinopneumonitis virus (EHV-1) in horse serum. Vet Microbiol. 1988;17:315. 369. Garre B., van der Meulen K., Nugent J., et al. In vitro susceptibility of six isolates of equine herpesvirus 1 to acyclovir, ganciclovir, cidofovir, adefovir, PMEDAP and foscarnet. Vet Microbiol. 2007;122:43. 370. Murray M.J., del Piero F., Jeffrey S.C., et al. Neonatal equine herpesvirus type 1 infection on a thoroughbred breeding farm. J Vet Intern Med. 1998;12:36. 371. Friday P.A., Scarratt W.K., Elvinger F., et al. Ataxia and paresis with equine herpesvirus type 1 infection in a herd of riding school horses. J Vet Intern Med. 2000;14:197. 372. Bentz B.G., Maxwell L.K., Erkert R.S., et al. Pharmaco*kinetics of acyclovir after single intravenous and oral administration to adult horses. J Vet Intern Med. 2006;20:589. 373. Wilkins P.A., Papich M., Sweeney R.W. Pharmaco*kinetics of acyclovir in adult horses. J Vet Emerg Crit Care. 2005;15:174. 374. Garré B., Shebany K., Gryspeerdt A., et al. Pharmaco*kinetics of acyclovir after intravenous infusion of acyclovir and after oral administration of acyclovir and its prodrug valacyclovir in healthy adult horses. Antimicrob Agents Chemother. 2007;51:4308. 375. Gibson J.S., Slater J.D., Field H.J. The activity of (S)-1-[(3-hydroxy-2-phosphonyl methoxy) propyl] cytosine (HPMPC) against equine herpesvirus-1 (EHV-1) in cell cultures, mice and horses. Antiviral Res. 1992;19:219. 376. Pusterla N., Chaney K.P., Maes R., et al. Investigation of the molecular detection of vaccine-derived equine herpesvirus type 1 in blood and nasal secretions from horses following intramuscular vaccination. J Vet Diagn Invest. 2007;19:290. 377. Drummer H.E., Reubel G.H., Studdert M.J. Equine gammaherpesvirus 2 (EHV2) is latent in B lymphocytes. Arch Virol. 1996;141:495. 378. Rizvi S.M., Slater J.D., Wol nger U., et al. Detection and distribution of equine herpesvirus 2 DNA in the central and peripheral nervous systems of ponies. J Gen Virol. 1997;78:1115. Pt 5 379. Borchers K., Ebert M., Fetsch A., et al. Prevalence of equine herpesvirus type 2 (EHV-2) DNA in ocular swabs and its cell tropism in equine conjunctiva. Vet Microbiol. 2006;118:260. 380. Fu Z.F., Robinson A.J., Horner G.W., et al. Respiratory disease in foals and the epizootiology of equine herpesvirus type 2 infection. N Z Vet J. 1986;34:152. 381. Nordengrahn A., Merza M., Ros C., et al. Prevalence of equine herpesvirus types 2 and 5 in horse populations by using type-specific PCR assays. Vet Res. 2002;33:251. 382. Pálfi V., Belák S., Molnár T. Isolation of equine herpesvirus type 2 from foals, showing respiratory symptoms. Zentralbl Veterinarmed B. 1978;25:165. 383. Sigiura T., f*ckuzawa Y., Kamada M., et al. Isolation of equine herpesvirus type 2 from foals with pneumonitis. Bull Equine Res Inst. 1983;20:148. 384. Ames T.R., O’Leary T.P., Johnston G.R. Isolation of equine herpesvirus type 2 from foals with respiratory disease. Compend Contin Educ Pract Vet. 1986;8:664. 385. Murray M.J., Eichorn E.S., Dubovi E.J., et al. Equine herpesvirus type 2: prevalence and seroepidemiology in foals. Equine Vet J. 1996;28:432. 386. Borchers K., Wol nger U., Ludwig H., et al. Virological and molecular biological investigations into equine herpes virus type 2 (EHV-2) experimental infections. Virus Res. 1998;55:101. 387. Kershaw O., von Oppen T., Glitz F., et al. Detection of equine herpesvirus type 2 (EHV-2) in horses with keratoconjunctivitis. Virus Res. 2001;80:93. 388. Browning G.F., Studdert M.J. Genomic heterogeneity of equine betaherpesviruses. J Gen Virol. 1987;68:1441. Pt 5 389. Dunowska M., Meers J., Wilks C.R. Isolation of equine herpesvirus type 5 in New Zealand, N Z Vet J. 1999;47:44. 390. Bell S.A., Balasuriya U.B., Nordhausen R.W., et al. Isolation of equine herpesvirus-5 from blood mononuclear cells of a gelding. J Vet Diagn Invest. 2006;18:472. 391. Holloway S.A., Lindquester G.J., Studdert M.J., et al. Identification, sequence analysis and characterisation of equine herpesvirus 5 glycoprotein B. Arch Virol. 1999;144:287. 392. Williams K.J., Maes R., Del Piero F., et al. Equine multinodular pulmonary fibrosis: a newly recognized herpesvirus-associated fibrotic lung disease. Vet Pathol. 2007;44:849. 393. Del Piero F. Equine viral arteritis. Vet Pathol. 2000;37:287. 394. Mumford J.A. Preparing for equine arteritis. Equine Vet J. 1985;17:6. 395. MacLachlan N.J., Balasuriya U.B. Equine viral arteritis. Adv Exp Med Biol. 2006;581:429. 396. Timoney P.J., McCollum W.H. Equine viral arteritis. Can Vet J. 1987;28:693. 397. Hedges J.F., Balasuriya U.B., Timoney P.J., et al. Genetic divergence with emergence of novel phenotypic variants of equine arteritis virus during persistent infection of stallions. J Virol. 1999;73:3672. 398. Holyoak G.R., Little T.V., McCollam W.H., et al. Relationship between onset of puberty and establishment of persistent infection with equine arteritis virus in the experimentally infected colt. J Comp Pathol. 1993;109:29. 399. Del Piero F., Wilkins P.A., Lopez J.W., et al. Equine viral arteritis in newborn foals: clinical, pathological, serological, microbiological and immunohistochemical observations. Equine Vet J. 1997;29:178. 400. Szeredi L., Hornyak A., Denes B., et al. Equine viral arteritis in a newborn foal: parallel detection of the virus by immunohistochemistry, polymerase chain reaction and virus isolation. J Vet Med B Infect Dis Vet Public Health. 2003;50:270. 401. Hullinger P.J., Gardner I.A., Hietala S.K., et al. Seroprevalence of antibodies against equine arteritis virus in horses residing in the United States and imported horses. J Am Vet Med Assoc. 2001;219:946. 402. Traub-Dargatz J.L., Ralston S.L., Collins J.K., et al. Equine viral arteritis. Compend Contin Educ Pract Vet. 1985;7:S490. 403. Timoney P.J., Klingeborn B., Lucas M.H. A perspective on equine viral arteritis (infectious arteritis of horses). Rev Sci Tech. 1996;15:1203. 404. Chirnside E.D. Equine arteritis virus: an overview. Br Vet J. 1992;148:181. 405. MacLachlan N.J., Balasuriya U.B., Davis N.L., et al. Experiences with new generation vaccines against equine viral arteritis, West Nile disease and African horse sickness. Vaccine. 2007;25:5577. 406. Hullinger P.J., Wilson W.D., Rossitto P.V., et al. Passive transfer, rate of decay, and protein speci city of antibodies against equine arteritis virus in horses from a Standardbred herd with high seroprevalence. J Am Vet Med Assoc. 1998;213:839. 407. Hartley C.A., Ficorilli N., Dynon K., et al. Equine rhinitis A virus: structural proteins and immune response. J Gen Virol. 2001;82:1725. 408. King A.M.Q., Brown F., Christian P., et al. Family picornaviridae. In: Fauquet C.M., Mayo M.A., Manilo J., et al, editors. Virus Taxonomy: VIIIth Report of the International

Committee on Taxonomy of Viruses. Burlington, MA: Elsevier, 2005. 409. Huang J.A., Ficorilli N., Hartley C.A., et al. Equine rhinitis B virus: a new serotype. J Gen Virol. 2001;82:2641. 410. Black W.D., Hartley C.A., Ficorilli N.P., et al. Sequence variation divides Equine rhinitis B virus into three distinct phylogenetic groups that correlate with serotype and acid stability. J Gen Virol. 2005;86:2323. 411. Black W.D., Studdert M.J. Formerly unclassified, acid-stable equine picornaviruses are a third equine rhinitis B virus serotype in the genus Erbovirus. J Gen Virol. 2006;87:3023. 412. Plummer G., Kerry J.B. Studies on an equine respiratory virus. Vet Rec. 1962;74:967. 413. Klaey M., Sanchez-Higgins M., Leadon D.P., et al. Field case study of equine rhinovirus 1 infection: clinical signs and clinicopathology. Equine Vet J. 1998;30:267. 414. Li F., Drummer H.E., Ficorilli N., et al. Identi cation of noncytopathic equine rhinovirus 1 as a cause of acute febrile respiratory disease in horses. J Clin Microbiol. 1997;35:937. 415. Black W.D., Wilcox R.S., Stevenson R.A., et al. Prevalence of serum neutralising antibody to equine rhinitis A virus (ERAV), equine rhinitis B virus 1 (ERBV1) and ERBV2. Vet Microbiol. 2007;119:65. 416. Kriegshäuser G., Deutz A., Kuechler E., et al. Prevalence of neutralizing antibodies to Equine rhinitis A and B virus in horses and man. Vet Microbiol. 2005;106:293. 417. Feng L., Browning G.F., Studdert M.J., et al. Equine rhinovirus 1 is more closely related to foot and mouth disease virus than to other picornaviruses. Proc Nat Acad Sci USA. 1996;93:990. 418. Carman S., Rosendal S., Huber L., et al. Infectious agents in acute respiratory disease in horses in Ontario. J Vet Diagn Invest. 1997;9:17. 419. Steck F, Hofer B, Schaeren B, et al: Equine rhinoviruses: new serotypes. In Proceedings of the Fourth International Conference on Equine Infectious Disease, Princeton, NJ, 1978, Veterinary Publications. 420. Studdert M.J. Equine adenovirus infections. In: Studdert M.J., editor. Virus Infections of Equines. New York: Elsevier, 1996. 421. Richards A.J., Kelly D.F., Knottenbelt D.C., et al. Anaemia, diarrhoea and opportunistic infections in Fell ponies. Equine Vet J. 2000;32:386. 422. Bell S.A., Leclere M., Gardner I.A., et al. Equine adenovirus 1 infection of hospitalised and healthy foals and horses. Equine Vet J. 2006;38:379. 423. Horner G.W., Hunter R. Isolation of two serotypes of equine adenovirus from horses in New Zealand. N Z Vet J. 1982;30:62. 424. Studdert M.J., Blackney M.H. Isolation of an adenovirus antigenically distinct from equine adenovirus type 1 from diarrheic foal feces. Am J Vet Res. 1982;43:543. 425. Barclay A.J., Paton D.J. Hendra (equine morbillivirus). Vet J. 2000;160:165. 426. Eaton B.T., Broder C.C., Middleton D., et al. Hendra and Nipah viruses: different and dangerous. Nat Rev Microbiol. 2006;4:23. 427. Murray K., Selleck P., Hooper P., et al. A morbillivirus that caused fatal disease in horses and humans. Science. 1995;268:94. 428. Westbury H. Hendra virus: a highly lethal zoonotic agent. Vet J. 2000;160:165. 429. Baldock F.C., Douglas I.C., Halpin K., et al. Epidemiological investigations into the 1994 equine morbillivirus outbreaks in Queensland, Australia. Sing Vet J. 20, 1996. 430. Barker S.C. The Australian paralysis tick may be the missing link in the transmission of Hendra virus from bats to horses to humans. Med Hypotheses. 2003;60:481. 431. Westbury H.A. Hendra virus disease in horses. Rev Sci Tech. 2000;19:151. 432. Mellor P.S., Hamblin C. African horse sickness. Vet Res. 2004;35:445. 433. Englund L., Pringle J. New diseases and increased risk of diseases in companion animals and horses due to transport. Acta Vet Scand Suppl. 2003;100:19. 434. Wittmann E.J., Baylis M. Climate change: effects on culicoides–transmitted viruses and implications for the UK. Vet J. 2000;160:107. 435. Boorman J., Mellor P.S., Penn M., et al. The growth of African horse-sickness virus in embryonated hen eggs and the transmission of virus by Culicoides variipennis Coquillett (Diptera, Ceratopogonidae). Arch Virol. 1975;47:343. 436. Coetzer J.A.W., Erasmus B.J.. Coetzer J.A.W., Thommson G.R., Tustin R.C., editors, Infections diseases of livestock with special reference to southern Africa. African horse sickness. Vol. 1, 1994. Oxford University Press, Cape Town, South Africa. 437. Sinclair M., Bührmann G., Gummow B. An epidemiological investigation of the African horsesickness outbreak in the Western Cape Province of South Africa in 2004 and its relevance to the current equine export protocol. J S Afr Vet Assoc. 2006;77:191. 438. Mair T.S., Lane J.G. Tracheal obstructions in two horses and a donkey. Vet Rec. 1990;126:303. 439. Urquhart K.A., Gerring E.L., Shepherd M.P. Tracheobronchial foreign body in a pony. Equine Vet J. 1981;13:262. 440. Couetil L.L., Gallatin L.L., Blevins W., et al. Treatment of tracheal collapse with an intraluminal stent in a miniature horse. J Am Vet Med Assoc. 2004;225:1727. 441. Kirker-Head C.A., Jakob T.P. Surgical repair of ruptured trachea in a horse. J Am Vet Med Assoc. 1990;196:1635. 442. Tate L.P.Jr., Koch D.B., Sembrat R.F., et al. Tracheal reconstruction by resection and end-to-end anastomosis in the horse. J Am Vet Med Assoc. 1981;178:253. 443. Hare W.C.D. Equine respiratory system. In: Getty R., editor. Sisson and Grossman’s the anatomy of the domestic animals. Philadelphia: W.B. Saunders, 1975. 444. McLaughlin R.F., Tyler W.S., Canada R.O. A study of the subgross pulmonary anatomy in various mammals. Am J Anat. 1961;108:149. 445. Breeze R., Turk M. Cellular structure, function and organization in the lower respiratory tract. Environ Health Perspect. 1984;55:3. 446. Pirie M., Pirie H.M., Cranston S., et al. An ultrastructural study of the equine lower respiratory tract. Equine Vet J. 1990;22:338. 447. Christmann U., Livesey L.C., Taintor J.S., et al. Lung surfactant function and composition in neonatal foals and adult horses. J Vet Intern Med. 2006;20:1402. 448. Mair T.S., Batten E.H., Stokes C.R., et al. The histological features of the immune system of the equine respiratory tract. J Comp Pathol. 1987;97:575. 449. Frevert C.W., Warner A.E., Adams E.T., et al. Pulmonary intravascular macrophages are an important part of the mononuclear phagocyte system in the horse (Abstract). J Vet Intern Med. 1991;5:145. 450. Taylor A.E., Rehder K., Hyatt R.E., et al. Clinical respiratory physiology. Philadelphia: W.B. Saunders; 1989. 451. Amis T.C., Pascoe J.R., Hornof W. Topographic distribution of pulmonary ventilation and perfusion in the horse. Am J Vet Res. 1984;45:1597. 452. Robinson N.E. Exercise induced pulmonary haemorrhage (EIPH): could Leonardo have got it right? Equine Vet J. 1987;19:370. 453. Barnes P.J. Neural control of human airways in health and disease. Am Rev Respir Dis. 1986;134:1289. 454. Derksen FJ, Broadstone RV. Bronchodilation therapy and the autonomic nervous system in horses with airway obstruction. Proceedings of the American College of Veterinary Internal Medicine, 1991. 455. Derksen F.J., Robinson N.E., Slocombe R.F. Ovalbumin-induced lung disease in the pony: role of vagal mechanisms. J Appl Physiol. 1982;53:719. 456. Sant’Ambrogio G. Nervous receptors of the tracheobronchial tree. Annu Rev Physiol. 1987;49:611. 457. Derksen F.J., Scott J.S., Slocombe R.F., et al. Effect of clenbuterol on histamine-induced airway obstruction in ponies. Am J Vet Res. 1987;48:423. 458. Scott J.S., Garon H., Broadstone R.V., et al. Alpha 1-adrenergic-induced airway obstruction in ponies with recurrent pulmonary disease. J Appl Physiol. 1988;65:687.

459. Sonea I., Bowker R.M., Broadstone R., et al. Presence and distribution of vasoactive intestinal peptide-like and peptide histidine isoleucine-like immunoreactivity in the equine lung. Am Rev Respir Dis. 1991;143:A355. (Abstract) 460. Robinson N.E., Wilson R. Airway obstruction in the horse. J Equine Vet Sci. 1989;9:155. 461. West J.B. Pulmonary pathophysiology. Baltimore: Williams & Wilkins; 1982. 462. Bayly W., Grant B.D., Breeze R.G., et al. The e ects of maximal exercise on acid-base balance and arterial blood gas tension in thoroughbred horses. In: Snow D.H., Persson S.G.N., Rose R.J., editors. Equine exercise physiology. Cambridge, England: Granta, 1983. 463. Wagner P.D., Gillespie J.R., Landgren G.L., et al. Mechanism of exercise-induced hypoxemia in horses. J Appl Physiol. 1989;66:1227. 464. Dantzker D.R. Physiology and pathophysiology of pulmonary gas exchange. Hosp Pract (Off Ed). 1986;21:135. 465. Roth R.A. Biochemistry, physiology and drug metabolism–implications regarding the role of the lungs in drug disposition. Clin Physiol Biochem. 1985;3:66. 466. Gillis C.N., Pitt B.R. The fate of circulating amines within the pulmonary circulation. Annu Rev Physiol. 1982;44:269. 467. Wright J.R. Immunoregulatory functions of surfactant proteins. Nat Rev Immunol. 2005;5:58. 468. Wong C.W., Thompson H.L., Thong Y.H., et al. Effect of strenuous exercise stress on chemiluminescence response of equine alveolar macrophages. Equine Vet J. 1990;22:33. 469. Huston L.H., Bayly W.M., Liggitt H.D., et al, Alveolar macrophage function in thoroughbreds after strenuous exercise. Gillespie J.R., Robinson N.E., editors. Equine exercise physiology, Volume 2. Davis, CA: ICEEP Publications, 1987. 470. Blunden A.S., Mackintosh M.E. The microflora of the lower respiratory tract of the horse: an autopsy study. Br Vet J. 1991;147:238. 471. Spurlock S.L., Spurlock G.H., Donaldson L.L. Consolidating pneumonia and pneumothorax in a horse. J Am Vet Med Assoc. 1988;192:1081. 472. Racklyeft D.J., Love D.N. Bacterial infection of the lower respiratory tract in 34 horses. Aust Vet J. 2000;78:549. 473. Garcia-Cantu M.C., Hartmann F.A., Brown C.M., et al. Bordetella bronchiseptica and equine respiratory infections: a review of 30 cases. Equine Vet Educ. 2000;12:45. 474. Anzai T., Walker J.A., Blair M.B., et al. Comparison of the phenotypes of Streptococcus zooepidemicus isolated from tonsils of healthy horses and specimens obtained from foals and donkeys with pneumonia. Am J Vet Res. 2000;61:162. 475. Biberstein E.L., Jang S.S., Hirsh D.C. Nocardia asteroides infection in horses: a review. J Am Vet Med Assoc. 1985;186:273. 476. Jakab G.J. Viral-bacterial interactions in pulmonary infection. Adv Vet Sci Comp Med. 1982;26:155. 477. Raidal S.L. Equine pleuropneumonia. Br Vet J. 1995;151:233. 478. Raidal S.L., Love D.N., Bailey G.D. Effect of a single bout of high intensity exercise on lower respiratory tract contamination in the horse. Aust Vet J. 1997;75:293. 479. Bayly W.M. Stress and equine respiratory immunity. Proc Am Coll Vet Intern Med. 1990. 480. Raidal S.L., Love D.N., Bailey G.D., et al. The e ect of high intensity exercise on the functional capacity of equine pulmonary alveolar macrophages and BAL-derived lymphocytes. Res Vet Sci. 2000;68:249. 481. Raidal S.L., Taplin R.H., Bailey G.D., et al. Antibiotic prophylaxis of lower respiratory tract contamination in horses con ned with head elevation for 24 or 48 hours. Aust Vet J. 1997;75:126. 482. Raidal S.L., Love D.N., Bailey G.D. In ammation and increased numbers of bacteria in the lower respiratory tract of horses within 6 to 12 hours of con nement with the head elevated. Aust Vet J. 1995;72:45. 483. Stull C.L., Rodiek A.V. Effects of cross-tying horses during 24 h of road transport. Equine Vet J. 2002;34:550. 484. Rainger J.E., Hughes K.J., Kessell A., et al. Pleuropneumonia as a sequela of myelography and general anaesthesia in a Thoroughbred colt. Aust Vet J. 2006;84:138. 485. Pusterla N., Jones M.E., Mohr F.C., et al. Fatal pulmonary hemorrhage associated with RTX toxin producing Actinobacillus equuli subspecies haemolyticus infection in an adult horse. J Vet Diagn Invest. 2008;20:118. 486. McKenzie H.C. Characterization of antimicrobial aerosols for administration to horses. Vet Ther. 2003;4:110. 487. Art T., de Moffarts B., Bedoret D., et al. Pulmonary function and antimicrobial concentration after marbofloxacin inhalation in horses. Vet Rec. 2007;161:348. 488. Lavoie J.P., Fiset L., Laverty S. Review of 40 cases of lung abscesses in foals and adult horses. Equine Vet J. 1994;26:348. 489. Mair T.S., Lane J.G. Pneumonia, lung abscesses and pleuritis in adult horses: a review of 51 cases. Equine Vet J. 1989;21:175. 490. Ainsworth D.M., Erb H.N., Eicker S.W., et al. E ects of pulmonary abscesses on racing performance of horses treated at referral veterinary medical teaching hospitals: 45 cases (1985-1997). J Am Vet Med Assoc. 2000;216:1282. 491. Schachter E.N. Suppurative lung disease: old problems revisited. Clin Chest Med. 1981;2:41. 492. Lubitz R.M. Resolution of lung abscess due to Pseudomonas aeruginosa with oral ciprofloxacin: case report. Rev Infect Dis. 1990;12:757. 493. Colahan P.T., Knight H.D. Drainage of an intrathoracic abscess in a horse via thoracotomy. J Am Vet Med Assoc. 1979;174:1231. 494. Smith B.P. Pleuritis and pleural effusion in the horse: a study of 37 cases. J Am Vet Med Assoc. 1977;170:208. 495. Raphel C.F., Beech J. Pleuritis secondary to pneumonia or lung abscessation in 90 horses. J Am Vet Med Assoc. 1982;181:808. 496. Schott H.C., Mansmann R.A. Thoracic drainage in horses. Compend Contin Educ Pract Vet. 1990;12:251. 497. Fenno C.H. Severe equine pleuritis due to wire penetration. Vet Med Small Anim Clin. 1975;70:458. 498. Tremaine W.H., Dixon P.M., McGorum B.C., et al. Pleuropulmonary abscessation in a horse caused by a gastric foreign body. Vet Rec. 1995;136:637. 499. Collins M.B., Hodgson D.R., Hutchins D.R. Pleural e usion associated with acute and chronic pleuropneumonia and pleuritis secondary to thoracic wounds in horses: 43 cases (1982-1992). J Am Vet Med Assoc. 1994;205:1753. 500. Hudson N.P.H., McClintock S.A., Hodgson D.R. Case of pleuro pneumonia with complications in a thoroughbred stallion. Equine Vet Educ. 1999;11:285. 501. Sprayberry K.A., Byars T.D. Equine pleuropneumonia. Equine Vet Educ. 1999;11:290. 502. Austin S.M., Foreman J.H., Hungerford L.L. Case-control study of risk factors for development of pleuropneumonia in horses. J Am Vet Med Assoc. 1995;207:325. 503. Chaffin M.K. Thoracocentesis and pleural drainage in horses. Equine Vet Educ. 1998;10:106. 504. Foreman J.H. Equine respiratory pharmacology. Vet Clin North Am Equine Pract. 1999;15:665. 505. Orsini J.A., Perkins S. The fluoroquinolones: clinical applications in veterinary medicine. Compend Contin Educ Pract Vet. 1992;14:1491. 506. Giguère S., Belanger M. Concentration of enrofloxacin in equine tissues after long-term oral administration. J Vet Pharmacol Ther. 1997;20:402. 507. Gustafsson A., Båverud V., Gunnarsson A., et al. The association of erythromycin ethylsuccinate with acute colitis in horses in Sweden. Equine Vet J. 1997;29:314. 508. Thoracotomy Grant B. Proceedings of the 1997 Dubai International Equine Symposium. Dubai, United Arab Emirates: Neyenesch Printers; 1997. 509. Byars T.D. Pleuropneumonia: treatment and prognosis. In The diagnosis and treatment of respiratory diseases. Proceedings of the 1997 Dubai International Equine Symposium.

Dubai, United Arab Emirates: Neyenesch Printers; 1997. 510. Frevert C.W., Warner A.E. Respiratory distress resulting from acute lung injury in the veterinary patient. J Vet Intern Med. 1992;6:154. 511. Seltzer K.L., Byars T.D. Prognosis for return to racing after recovery from infectious pleuropneumonia in thoroughbred racehorses: 70 cases (1984-1989). J Am Vet Med Assoc. 1996;208:1300. 512. Byars T.D., Dainis C.M., Seltzer K.L., et al. Cranial thoracic masses in the horse: a sequel to pleuropneumonia. Equine Vet J. 1991;23:22. 513. Byars T.D., Becht J.L. Pleuropneumonia. Vet Clin North Am Equine Pract. 1991;7:63. 514. Chaffin M.K. Diagnostic assessment of pleural effusion in horses. Compend Contin Educ Pract Vet. 1994;16:1035. 515. Jorgensen J.S., Geoly F.J., Berry C.R., et al. Lameness and pleural effusion associated with an aggressive fibrosarcoma in a horse. J Am Vet Med Assoc. 1997;210:1328. 516. Wrigley R.H., Gay C.C., Lording P., et al. Pleural effusion associated with squamous cell carcinoma of the stomach of a horse. Equine Vet J. 1981;13:99. 517. Prater P.E., Patton C.S., Held J.P. Pleural effusion resulting from malignant hepatoblastoma in a horse. J Am Vet Med Assoc. 1989;194:383. 518. Valentine B.A., Ross C.E., Bump J.L., et al. Intramuscular hemangiosarcoma with pulmonary metastasis in a horse. J Am Vet Med Assoc. 1986;188:628. 519. Murray M.J., Cavey D.M., Feldman B.F., et al. Signs of sympathetic denervation associated with a thoracic melanoma in a horse. J Vet Intern Med. 1997;11:199. 520. Mair T.S., Hillyer M.H., Brown P. Mesothelioma of the pleural cavity in a horse: diagnostic features. Equine Vet Educ. 1992;4:59. 521. Foreman J.H., Weidner J.P., Parry B.W., et al. Pleural effusion secondary to thoracic metastatic mammary adenocarcinoma in a mare. J Am Vet Med Assoc. 1990;197:1193. 522. Morris D.D., Acland H.M., Hodge T.G. Pleural effusion secondary to metastasis of an ovarian adenocarcinoma in a horse. J Am Vet Med Assoc. 1985;187:272. 523. Thatcher C.D., Roussel A.J., Chickering W.R., et al. Pleural effusion with thoracic lymphosarcoma in a mare. Compend Contin Educ Pract Vet. 1985;7:S726. 524. Sweeney C.R., Gillette D.M. Thoracic neoplasia in equids: 35 cases (1967-1987). J Am Vet Med Assoc. 1989;195:374. 525. Ogilvie T.H., Rosendal S., Blackwell T.E., et al. Mycoplasma felis as a cause of pleuritis in horses. J Am Vet Med Assoc. 1983;182:1374. 526. Burbidge H.M. Penetrating thoracic wound in a Hackney mare. Equine Vet J. 1982;14:94. 527. Mair T.S., Pearson H., Waterman A.E., et al. Chylothorax associated with a congenital diaphragmatic defect in a foal. Equine Vet J. 1988;20:304. 528. Sahn S.A. The pathophysiology of pleural effusions. Annu Rev Med. 1990;41:7. 529. Tarn A.C., Lapworth R. Biochemical analysis of pleural fluid: what should we measure? Ann Clin Biochem. 2001;38:311. 530. Das D.K. Serous effusions in malignant lymphomas: a review. Diagn Cytopathol. 2006;34:335. 531. Yildirim E., Dural K., Yazkan R., et al. Rapid pleurodesis in symptomatic malignant pleural effusion. Eur J Cardiothorac Surg. 2005;27:19. 532. Kuzniar T.J., Blum M.G., Kasibowska-Kuzniar K., et al. Predictors of acute lung injury and severe hypoxemia in patients undergoing operative talc pleurodesis. Ann Thorac Surg. 2006;82:1976. 533. Baron R.D., Milton R., Thorpe J.A. Pleurodesis using small talc particles results in an unacceptably high rate of acute lung injury and hypoxia. Ann Thorac Surg. 2007;84:2136. 534. Antony V.B., Loddenkemper R., Astoul P., et al. Management of malignant pleural effusions. Eur Respir J. 2001;18:402. 535. Stather D.R., Tremblay A. Use of tunneled pleural catheters for outpatient treatment of malignant pleural effusions. Curr Opin Pulm Med. 2007;13:328. 536. Putnam J.B.Jr., Light R.W., Rodriguez R.M., et al. A randomized comparison of indwelling pleural catheter and doxycycline pleurodesis in the management of malignant pleural effusions. Cancer. 1999;86:1992. 537. Carter G.R., Changappa M.M., Roberts A.W. Systemic mycoses. In Carter G.R., Changappa M.M., Roberts A.W., editors: Essentials of Veterinary Mycology, ed 5, Philadelphia: Lea & Febiger, 1995. 538. De Groote M.A., Bjerke R., Smith H., et al. Expanding epidemiology of blastomycosis: clinical features and investigation of 2 cases in Colorado. Clin Infect Dis. 2000;30:582. 539. Toribio R.E., Kohn C.W., Lawrence A.E., et al. Thoracic and abdominal blastomycosis in a horse. J Am Vet Med Assoc. 1999;214:1357. 540. Fisher M.C., Koenig G.L., White T.J., et al. Molecular and phenotypic description of Coccidioides posadasil sp nov, previously recognized as the non-California population of Coccidioides immitis. Mycologia. 2002;94:73. 541. Ziemer E.L., Pappagianis D., Madigan J.E., et al. Coccidioidomycosis in horses: 15 cases (1975-1984). J Am Vet Med Assoc. 1992;201:910. 542. Stoltz J.H., Johnson B.J., Walker R.L., et al. Coccidioides immitis abortion in an Arabian mare. Vet Pathol. 1994;31:258. 543. Walker R.L., Johnson B.J., Jones K.L., et al. Coccidioides immitis mastitis in a mare. J Vet Diagn Invest. 1993;5:446. 544. Casadevall A., Rosas A.L., Nosanchuk J.D. Melanin and virulence in Cryptococcus neoformans. Curr Opin Microbiol. 2000;3:354. 545. Carter G.R., Changappa M.M., Roberts A.W. Mycoses caused by yeasts or yeast-like fungi. In Carter G.R., Changappa M.M., Roberts A.W., editors: Essentials of Veterinary Mycology, ed 5, Philadelphia: Lea & Febiger, 1995. 546. Belay T., Cherniak R., O’Neill E.B., et al. Serotyping of Cryptococcus neoformans by dot enzyme assay. J Clin Microbiol. 1996;34:466. 547. Sebghati T.S., Engle J.T., Goldman W.E. Intracellular parasitism by Histoplasma capsulatum: fungal virulence and calcium dependence. Science. 2000;290:1368. 548. Rezabek G.B., Donahue J.M., Giles R.C., et al. Histoplasmosis in horses. J Comp Pathol. 1993;109:47. 549. Cornick J.L. Diagnosis and treatment of pulmonary histoplasmosis in a horse. Cornell Vet. 1990;80:97. 550. Slocombe R.F., Slauson D.O. Invasive pulmonary aspergillosis of horses: an association with acute enteritis. Vet Pathol. 1988;25:277. 551. Green S.L., Hager D.A., Mays M.B.C., et al. Acute diffuse mycotic pneumonia in a 7-month old colt. Vet Radiol. 1987;28:216. 552. Blomme E., Del Piero F., La Perle K.M.D., et al. Aspergillosis in horses: a review. Equine Vet Educ. 1998;10:86. 553. Sweeney C.R., Habecker P.L. Pulmonary aspergillosis in horses: 29 cases (1974-1997). J Am Vet Med Assoc. 1999;214:808. 554. Perryman L.E., McGuire T.C., Crawford T.B. Maintenance of foals with combined immunodeficiency: causes and control of secondary infections. Am J Vet Res. 1978;39:1043. 555. Ainsworth D.M., Weldon A.D., Beck K.A., et al. Recognition of Pneumocystis carinii in foals with respiratory distress. Equine Vet J. 1993;25:103. 556. Blue J., Perdrizet J., Brown E. Pulmonary aspergillosis in a horse with myelomonocytic leukemia. J Am Vet Med Assoc. 1987;190:1562. 557. Pusterla N., Pesavento P.A., Leutenegger C.M., et al. Disseminated pulmonary adiaspiromycosis caused by Emmonsia crescens in a horse. Equine Vet J. 2002;34:749. 558. Carrasco L., Mendez A., Jensen H.E. Chronic bronchopulmonary aspergillosis in a horse with Cushing’s syndrome. Mycoses. 1996;39:443. 559. Ho P.L., Yuen K.Y. Aspergillosis in bone marrow transplant recipients. Crit Rev Oncol Hematol. 2000;34:55. 560. Thirion-Delalande C., Guillot J., Jensen H.E., et al. Disseminated acute concomitant aspergillosis and mucormycosis in a pony. J Vet Med A Physiol Pathol Clin Med. 2005;52:121. 561. Moore B.R., Reed S.M., Kowalski J.J., et al. Aspergillosis granuloma in the mediastinum of a non-immunocompromised horse. Cornell Vet. 1993;83:97. 562. Carrasco L., Tarradas M.C., Gómez-Villamandos J.C., et al. Equine pulmonary mycosis due to Aspergillus niger and Rhizopus stolonifer. J Comp Pathol. 1997;117:191.

563. Santamauro J.T., Stover D.E. Pneumocystis carinii pneumonia. Med Clin North Am. 1997;81:299. 564. Ceré N., Polack B. Animal pneumocystosis: a model for man. Vet Res. 1999;30:1. 565. Morris A.M., Swanson M., Ha H., et al. Geographic distribution of human immunode ciency virus-associated Pneumocystis carinii pneumonia in San Francisco. Am J Respir Crit Care Med. 2000;162:1622. 566. Morris A., Beard C.B., Huang L. Update on the epidemiology and transmission of Pneumocystis carinii. Microbes Infect. 2002;4:95. 567. Gigliotti F., Harmsen A.G., Haidaris C.G., et al. Pneumocystis carinii is not universally transmissible between mammalian species. Infect Immun. 1993;61:2886. 568. Thomas C.F.Jr., Limper A.H. Pneumocystis pneumonia. N Engl J Med. 2004;350:2487. 569. Pusterla N., Holmberg T.A., Lorenzo-Figueras M., et al. Acremonium strictum pulmonary infection in a horse. Vet Clin Pathol. 2005;34:413. 570. Fincher R.M., Fisher J.F., Lovell R.D., et al. Infection due to the fungus Acremonium (cephalosporium). Medicine (Baltimore). 1991;70:398. 571. England D.M., Hochholzer L. Adiaspiromycosis: an unusual fungal infection of the lung. Report of 11 cases. Am J Surg Pathol. 1993;17:876. 572. Hattel A.L., Drake T.R., Anderholm B.J., et al. Pulmonary aspergillosis associated with acute enteritis in a horse. J Am Vet Med Assoc. 1991;199:589. 573. Long J.R., Mitchell L. Pulmonary aspergillosis in a mare. Can Vet J. 1971;12:16. 574. Cooper V.L., Kennedy G.A., Kruckenberg S.M., et al. Histoplasmosis in a miniature Sicilian burro. J Vet Diagn Invest. 1994;6:499. 575. Foley J.P., Legendre A.M. Treatment of coccidioidomycosis osteomyelitis with itraconazole in a horse. A brief report. J Vet Intern Med. 1992;6:333. 576. Petrites-Murphy M.B., Robbins L.A., Donahue J.M., et al. Equine cryptococcal endometritis and placentitis with neonatal cryptococcal pneumonia. J Vet Diagn Invest. 1996;8:383. 577. Begg L.M., Hughes K.J., Kessell A., et al. Successful treatment of cryptococcal pneumonia in a pony mare. Aust Vet J. 2004;82:686. 578. Shively J.N., Dellers R.W., Buergelt C.D., et al. Pneumocystis carinii pneumonia in two foals. J Am Vet Med Assoc. 1973;162:648. 579. Flaminio M.J., Rush B.R., Cox J.H., et al. CD4+ and CD8+ T-lymphocytopenia in a filly with Pneumocystis carinii pneumonia. Aust Vet J. 1998;76:399. 580. Marrs G.E. Pneumocystis carinii pneumonia in a Paso Fino colt. Vet Med. 1987;82:1172. 581. Ewing P.J., Cowell R.L., Tyler R.D., et al. Pneumocystis carinii pneumonia in foals. J Am Vet Med Assoc. 1994;204:929. 582. Whitwell K. Pneumocystis carinii infection in foals in the UK. Vet Rec. 1992;131:19. 583. Perron Lepage M.F., Gerber V., Suter M.M. A case of interstitial pneumonia associated with Pneumocystis carinii in a foal. Vet Pathol. 1999;36:621. 584. Prescott J.F., Ogilvie T.H., Markham R.J. Lymphocyte immunostimulation in the diagnosis of Corynebacterium equi pneumonia of foals. Am J Vet Res. 1980;41:2073. 585. MacNeill A.L., Alleman A.R., Franklin R.P., et al. Pneumonia in a Paso-Fino mare. Vet Clin Pathol. 2003;32:73. 586. Pearson E.G., Watrous B.J., Schmitz J.A., et al. Cryptococcal pneumonia in a horse. J Am Vet Med Assoc. 1983;183:577. 587. Kramme P.M., Ziemer E.L. Disseminated coccidioidomycosis in a horse with osteomyelitis. J Am Vet Med Assoc. 1990;196:106. 588. Riley C.B., Bolton J.R., Mills J.N., et al. Cryptococcosis in seven horses. Aust Vet J. 1992;69:135. 589. Benbrook E.A., Bryant J.B., Saunders L.Z. A case of blastomycosis in the horse. J Am Vet Med Assoc. 1948;112:475. 590. Wilson J.H., Olson E.J., Haugen E.W., et al. Systemic blastomycosis in a horse. J Vet Diagn Invest. 2006;18:615. 591. Langham R.F., Beneke E.S., Whitenack D.L. Abortion in a mare due to coccidioidomycosis. J Am Vet Med Assoc. 1977;170:178. 592. DeMartini J.C., Riddle W.E. Disseminated coccidioidomycosis in two horses and a pony. J Am Vet Med Assoc. 1969;155:149. 593. Cho D.Y., Pace W., Beadle R.E. Cerebral cryptococcosis in a horse. Vet Pathol. 1986;23:207. 594. Welsh R.D., Stair E.L. Cryptococcal meningitis in a horse. J Equine Vet Sci. 1995;15:80. 595. Chandna V.K., Morris E., Gliatto J.M., et al. Localised subcutaneous cryptococcal granuloma in a horse. Equine Vet J. 1993;25:166. 596. Blanchard P.C., Filkins M. Cryptococcal pneumonia and abortion in an equine fetus. J Am Vet Med Assoc. 1992;201:1591. 597. Lenard Z.M., Lester N.V., O’hara A.J., et al. Disseminated cryptococcosis including osteomyelitis in a horse. Aust Vet J. 2007;85:51. 598. Katayama Y., Kuwano A., Yoshihara T. Histoplasmosis in the lung of a race horse with yersiniosis. J Vet Med Sci. 2001;63:1229. 599. Johnson P.J., Moore L.A., Mrad D.R., et al. Sudden death of two horses associated with pulmonary aspergillosis. Vet Rec. 1999;145:16. 600. Nunes J., Mackie J.T., Kiupel M. Equine histoplasmosis presenting as a tumor in the abdominal cavity. J Vet Diagn Invest. 2006;18:508. 601. Richter M., Hauser B., Kaps S., et al. Keratitis due to Histoplasma spp. in a horse. Vet Ophthalmol. 2003;6:99. 602. Pace L.W., Wirth N.R., Foss R.R., et al. Endocarditis and pulmonary aspergillosis in a horse. J Vet Diagn Invest. 1994;6:504. 603. Franklin R.P., Long M.T., MacNeill A., et al. Proliferative interstitial pneumonia, Pneumocystis carinii infection, and immunode ciency in an adult Paso Fino horse. J Vet Intern Med. 2002;16:607. 604. Higgins J.C., Leith G.S., Pappagianis D., et al. Treatment of Coccidioides immitis pneumonia in two horses with fluconazole. Vet Rec. 2006;159:349. 605. Higgins J.C., Leith G.S., Voss E.D., et al. Seroprevalence of antibodies against Coccidioides immitis in healthy horses. J Am Vet Med Assoc. 1888;226:2005. 606. Higgins J.C., Pusterla N., Pappagianis D. Comparison of Coccidioides immitis serological antibody titres between forms of clinical coccidioidomycosis in horses. Vet J. 2007;173:118. 607. Furcolow M.L., Menges R.W. Comparison of histoplasmin sensitivity rates among human beings and animals in Boone County, Missouri. Am J Public Health Nations Health. 1952;42:926. 608. Schar D.L., Sage A.M., Hayden D.W., et al. What is your diagnosis? A large mass dorsal to the bifurcation of the trachea with tracheobronchial compression. J Am Vet Med Assoc. 2004;224:1757. 609. Guillot J., Sarfati J., de Barros M., et al. Comparative study of serological tests for the diagnosis of equine aspergillosis. Vet Rec. 1999;145:348. 610. deJaham C., Paradis M., Papich M.G. Antifungal dermatologic agents: Azones and allylamines. Comp Cont Educ Pract Vet. 2000;22:548. 611. Gaboriau F., Chéron M., Petit C., et al. Heat-induced superaggregation of amphotericin B reduces its in vitro toxicity: a new way to improve its therapeutic index. Antimicrob Agents Chemother. 1997;41:2345. 612. Clark-Price S.C., Cox J.H., Bartoe J.T., et al. Use of dapsone in the treatment of Pneumocystis carinii pneumonia in a foal. J Am Vet Med Assoc. 2004;224:407. 613. Antal T., Szabó I., Antal V., et al. Respiratory disease of horses associated with Mycoplasma infection. Zentralbl Veterinarmed B. 1988;35:264. 614. Rosendal S., Blackwell T.E., Lumsden J.H., et al. Detection of antibodies to Mycoplasma felis in horses. J Am Vet Med Assoc. 1986;188:292. 615. Hoffman A.M., Baird J.D., Kloeze H.J., et al. Mycoplasma felis pleuritis in two show-jumper horses. Cornell Vet. 1992;82:155. 616. Antal A., Szabó I., Vajda G., et al. Immunoglobulin concentration in the blood serum of foals su ering from pneumonia associated with mycoplasma infection. Arch Exp Veterinarmed. 1989;43:747.

617. Wood J.L., Chanter N., Newton J.R., et al. An outbreak of respiratory disease in horses associated with Mycoplasma felis infection. Vet Rec. 1997;140:388. 618. Newton J.R., Wood J.L., Chanter N. A case control study of factors and infections associated with clinically apparent respiratory disease in UK Thoroughbred racehorses. Prev Vet Med. 2003;60:107. 619. Dungworth D.L. Interstitial pulmonary disease. Adv Vet Sci Comp Med. 1982;26:173. 620. Lopez A. Respiratory system. In McGavin M.D., Zachary J.F., editors: Pathologic Basis of Veterinary Disease, ed 4, St Louis: Mosby Elsevier, 2007. 621. Buergelt C.D. Interstitial pneumonia in the horse: a fledgling morphological entity with mysterious causes. Equine Vet J. 1995;27:4. 622. Lakritz J., Wilson W.D., Berry C.R., et al. Bronchointerstitial pneumonia and respiratory distress in young horses: clinical, clinicopathologic, radiographic, and pathological findings in 23 cases (1984-1989). J Vet Intern Med. 1993;7:277. 623. Nout Y.S., Hinchcliff K.W., Samii V.F., et al. Chronic pulmonary disease with radiographic interstitial opacity (interstitial pneumonia) in foals. Equine Vet J. 2002;34:542. 624. Sharma O.P. Interstitial lung disease. Curr Opin Pulm Med. 1995;1:345. 625. O’Sullivan B.M. Crofton weed (Eupatorium adenophorum) toxicity in horses. Aust Vet J. 1979;55:19. 626. Oelrichs P.B., Calanasan C.A., MacLeod J.K., et al. Isolation of a compound from Eupatorium adenophorum (Spreng.) [Ageratina adenophora (Spreng.)] causing hepatotoxicity in mice. Nat Toxins. 1995;3:350. 627. Gibson J.A., O’Sullivan B.M. Lung lesions in horses fed mist flower (Eupatorium riparium). Aust Vet J. 1984;61:271. 628. Nobre D., Dagli M.L., Haraguchi M. Crotalaria juncea intoxication in horses. Vet Hum Toxicol. 1994;36:445. 629. Lindley W.H. Acute pulmonary edema. Mod Vet Pract. 1978;59:64. 630. Breeze R.G., Legreid W.W., Bayly W.M., et al. Perilla ketone toxicity: a chemical model for the study of equine restrictive lung disease. Equine Vet J. 1984;16:180. 631. Schmidbauer S.M., Venner M., von Samson-Himmelstjerna G., et al. Compensated overexpression of procollagens alpha 1(I) and alpha 1(III) following perilla mint ketoneinduced acute pulmonary damage in horses. J Comp Pathol. 2004;131:186. 632. Scarratt W.K., Moon M.L., Sponenberg D.P., et al. Inappropriate administration of mineral oil resulting in lipoid pneumonia in three horses. Equine Vet J. 1998;30:85. 633. Davis J.L., Ramirez S., Campbell N., et al. Acute and chronic mineral oil pneumonitis in two horses. Eq Vet Educ. 2001;13:230. 634. Bos M., de Bosschere H., Deprez P., et al. Chemical identification of the (causative) lipids in a case of exogenous lipoid pneumonia in a horse. Equine Vet J. 2002;34:744. 635. Henninger R.W., Hass G.F., Freshwater A. Corticosteroid management of lipoid pneumonia in a horse. Eq Vet Educ. 2006;18:205. 636. Field H.E., Barratt P.C., Hughes R.J., et al. A fatal case of Hendra virus infection in a horse in north Queensland: clinical and epidemiological features. Aust Vet J. 2000;78:279. 637. Mair T.S., Taylor F.G., Gibbs C., et al. Generalized avian tuberculosis in a horse. Equine Vet J. 1986;18:226. 638. Gunnes G., Nord K., Vatn S., et al. A case of generalised avian tuberculosis in a horse. Vet Rec. 1995;136:565. 639. Monreal L., Segura D., Segales J., et al. Diagnosis of Mycobacterium bovis infection in a mare. Vet Rec. 2001;149:712. 640. Pavlikel I., Jahn P., Dvorska L., et al. Mycobacterial infections in horses: a review of the literature. Vet Med (Czech). 2004;49:427. 641. Muser R., Nassal J. Tuberculosis, tuberculin reaction any mycobacteria in horse. Rindertuberkulose Brucell. 1962;11:118. 642. Winder C., Ehrensperger F., Hermann M., et al. Interstitial pneumonia in the horse: two unusual cases. Equine Vet J. 1988;20:298. 643. Buergelt C.D., Hines S.A., Cantor G., et al. A retrospective study of proliferative interstitial lung disease of horses in Florida. Vet Pathol. 1986;23:750. 644. Derksen F.J., Slocombe R.F., Brown C.M., et al. Chronic restrictive pulmonary disease in a horse. J Am Vet Med Assoc. 1982;180:887. 645. Kelly D.F., Newsholme S.J., Baker J.R., et al. Diffuse alveolar damage in the horse. Equine Vet J. 1995;27:76. 646. Donaldson M.T., Beech J., Ennulat D., et al. Interstitial pneumonia and pulmonary fibrosis in a horse. Equine Vet J. 1998;30:173. 647. Dixon P.M., McGorum B.C., Long K.J., et al. Acute eosinophilic interstitial pulmonary disease in a pony. Vet Rec. 1992;130:367. 648. Pusterla N., Pesavento P.A., Smith P., et al. Idiopathic granulomatous pneumonia in seven horses. Vet Rec. 2003;153:653. 649. Uhlhorn M., Hurst M., Demmers S. Disseminated eosinophilic pulmonary granulomas in a pony. Eq Vet Educ. 2006;18:178. 650. Wright J.L. Consequences of aspiration and bronchial obstruction. In Thurbeck W.M., Churg A.M., editors: Pathology of the Lung, ed 2, New York: Thieme Medical Publishing Co, 1995. 651. Turk J.R., Brown C.M., Johnson G.C. Diffuse alveolar damage with fibrosing alveolitis in a horse. Vet Pathol. 1981;18:560. 652. Schwarzwald C.C., Stewart A.J., Morrison C.D., et al. Cor pulmonale in a horse with granulomatous pneumonia. Equine Vet Educ. 2006;18:182. 653. Bastianello S.S. A survey on neoplasia in domestic species over a 40-year period from 1935 to 1974 in the Republic of South Africa. IV. Tumours occurring in Equidae. Onderstepoort J Vet Res. 1983;50:91. 654. Schwartz L.W., Knight H.D., Whittig L.D., et al. Silicate pneumoconiosis and pulmonary fibrosis in horses from the Monterey-Carmel peninsula. Chest. 1981;80:82. 655. Berry C.R., O’Brien T.R., Madigan J.E., et al. Thoracic radiographic features of silicosis in 19 horses. J Vet Intern Med. 1991;5:248. 656. Konyha L.D., Kreier J.P. The significance of tuberculin tests in the horse. Am Rev Respir Dis. 1971;103:91. 657. American Thoracic Society (ATS)the European Respiratory Society (ERS). Idiopathic pulmonary fibrosis: Diagnosis and treatment. Am J Respir Crit Care Med. 2000;161:646. 658. Raghu G. Idiopathic pulmonary fibrosis: treatment options in pursuit of evidence-based approaches. Eur Respir J. 2006;28:463. 659. Lyons E.T., Tolliver S.C., Drudge J.H., et al. Parasites in lungs of dead equids in Kentucky: emphasis on Dictyocaulus arnfieldi. Am J Vet Res. 1985;46:924. 660. Lyons E.T., Tolliver S.C., Drudge J.H., et al. Lungworms (Dictyocaulus arnfieldi): prevalence in live equids in Kentucky. Am J Vet Res. 1985;46:921. 661. Jorgensen R.J., Andersen S. Spread of equine lungworm (Dictyocaulus arnfieldi) larvae from faeces by Pilobolus fungi. Nord Vet Med. 1984;36:162. 662. Clayton H.M. Lung Parasites. Philadelphia: W.B. Saunders; 1983. 663. George L.W., Tanner M.L., Roberson E.L., et al. Chronic respiratory disease in a horse infected with Dictyocaulus arnfieldi. J Am Vet Med Assoc. 1981;179:820. 664. Clayton H.M., Duncan J.L. Natural infection with Dictyocaulus arnfieldi in pony and donkey foals. Res Vet Sci. 1981;31:278. 665. Lyons E.T., Drudge J.H., Tolliver S.C. Ivermectin: treating for naturally occurring infections of lungworms and stomach worms in equids. Vet Med. 1985;80:58. 666. Coles G.C., Hillyer M.H., Taylor F.G., et al. Activity of moxidectin against bots and lungworm in equids. Vet Rec. 1998;143:169. 667. Humber K.A. Near drowning of a gelding. J Am Vet Med Assoc. 1988;192:377. 668. Austin S.M., Foreman J.H., Goetz T.E. Aspiration pneumonia following near-drowning in a mare: a case report. J Equine Vet Sci. 1988;8:313. 669. Sembrat R., Di Stazio J., Reese J., et al. Acute pulmonary failure in the conscious pony with Escherichia coli septicemia. Am J Vet Res. 1978;39:1147. 670. Goer R.J., Ames T.R. Smoke inhalation injury in horses. Compend Contin Educ Pract Vet. 1991;13:1162.

671. Davis J.L., Gardner S.Y., Schwabenton B., et al. Congestive heart failure in horses: 14 cases (1984-2001). J Am Vet Med Assoc. 2002;220:1512. 672. Kollef M.H., Schuster D.P. The acute respiratory distress syndrome. N Engl J Med. 1995;332:27. 673. Ware L.B., Matthay M.A. The acute respiratory distress syndrome. N Engl J Med. 2000;342:1334. 674. Lewis J.F., Jobe A.H. Surfactant and the adult respiratory distress syndrome. Am Rev Respir Dis. 1993;147:218. 675. Kemper T., Spier S., Barratt-Boyes S.M., et al. Treatment of smoke inhalation in five horses. J Am Vet Med Assoc. 1993;202:91. 676. Johansson A.M., Gardner S.Y., Levine J.F., et al. Furosemide continuous rate infusion in the horse: evaluation of enhanced e cacy and reduced side e ects. J Vet Intern Med. 2003;17:887. 677. Johansson A.M., Gardner S.Y., Levine J.F., et al. Pharmaco*kinetics and pharmacodynamics of furosemide after oral administration to horses. J Vet Intern Med. 2004;18:739. 678. Mitten L.A., Hinchcli K.W., Holcombe S.J., et al. Mechanical ventilation and management of botulism secondary to an injection abscess in an adult horse. Equine Vet J. 1994;26:420. 679. Robinson N.E. International Workshop on Equine Chronic Airway Disease. Michigan State University 16-18 June 2000. Equine Vet J. 2001;33:5. 680. Muylle E., Oyaert W. Lung function tests in obstructive pulmonary disease in horses. Equine Vet J. 1973;5:37. 681. Derksen F.J., Scott D., Robinson N.E., et al. Intravenous histamine administration in ponies with recurrent airway obstruction (heaves). Am J Vet Res. 1985;46:774. 682. Derksen F.J., Robinson N.E., Scott J.S., et al. Aerosolized Micropolyspora faeni antigen as a cause of pulmonary dysfunction in ponies with recurrent airway obstruction (heaves). Am J Vet Res. 1988;49:933. 683. Seahorn T.L., Beadle R.E. Summer pasture-associated obstructive pulmonary disease in horses: 21 cases (1983-1991). J Am Vet Med Assoc. 1993;202:779. 684. Aviza G.A., Ainsworth D.M., Eicker S.W., et al. Outcome of horses diagnosed with and treated for heaves (recurrent airway obstruction). Equine Vet Educ. 2001;13:243. 685. Marti E., Gerber H., Essich G., et al. The genetic basis of equine allergic diseases. 1. Chronic hypersensitivity bronchitis. Equine Vet J. 1991;23:457. 686. Jost U., Klukowska- Rötzler J., Dolf G., et al. A region on equine chromosome 13 is linked to recurrent airway obstruction in horses. Equine Vet J. 2007;39:236. 687. Ewart S.L., Robinson N.E. Genes and respiratory disease: a first step on a long journey. Equine Vet J. 2007;39:270. 688. McPherson E.A., Thomson J.R. Chronic obstructive pulmonary disease in the horse. 1: Nature of the disease. Equine Vet J. 1983;15:203. 689. McPherson E.A., Lawson G.H., Murphy J.R., et al. Chronic obstructive pulmonary disease (COPD) in horses: aetiological studies: responses to intradermal and inhalation antigenic challenge. Equine Vet J. 1979;11:159. 690. McGorum B.C., Dixon P.M., Halliwell R.E. Phenotypic analysis of peripheral blood and bronchoalveolar lavage uid lymphocytes in control and chronic obstructive pulmonary disease affected horses, before and after ‘natural (hay and straw) challenges’. Vet Immunol Immunopathol. 1993;36:207. 691. Pirie R.S., Dixon P.M., Collie D.D., et al. Pulmonary and systemic effects of inhaled endotoxin in control and heaves horses. Equine Vet J. 2001;33:311. 692. Pirie R.S., Collie D.D., Dixon P.M., et al. Inhaled endotoxin and organic dust particulates have synergistic proin ammatory e ects in equine heaves (organic dust-induced asthma). Clin Exp Allergy. 2003;33:676. 693. Halliwell R.E., McGorum B.C., Irving P., et al. Local and systemic antibody production in horses a ected with chronic obstructive pulmonary disease. Vet Immunol Immunopathol. 1993;38:201. 694. Eder C., Crameri R., Mayer C., et al. Allergen-speci c IgE levels against crude mould and storage mite extracts and recombinant mould allergens in sera from horses a ected with chronic bronchitis. Vet Immunol Immunopathol. 2000;73:241. 695. Lavoie J.P., Maghni K., Desnoyers M., et al. Neutrophilic airway in ammation in horses with heaves is characterized by a Th2-type cytokine pro le. Am J Respir Crit Care Med. 2001;164:1410. 696. Cordeau M.E., Joubert P., Dewachi O., et al. IL-4, IL-5 and IFN-gamma mRNA expression in pulmonary lymphocytes in equine heaves. Vet Immunol Immunopathol. 2004;97:87. 697. Dixon P.M., McGorum B.C., Marley C., et al. E ects of equine in uenza and tetanus vaccination on pulmonary function in normal and chronic obstructive pulmonary disease affected horses. Equine Vet J. 1996;28:157. 698. Corry D.B., Kheradmand F. Induction and regulation of the IgE response. Nature. 1999;402:B18. 699. Ainsworth D.M., Appleton J.A., Antczak D.F., et al. Immune responses in horses with recurrent airway obstruction. Am J Respir Crit Care Med. 2000;161:A842. 700. Giguère S., Viel L., Lee E., et al. Cytokine induction in pulmonary airways of horses with heaves and e ect of therapy with inhaled uticasone propionate. Vet Immunol Immunopathol. 2002;85:147. 701. Kleiber C., McGorum B.C., Horohov D.W., et al. Cytokine pro les of peripheral blood and airway CD4 and CD8 T lymphocytes in horses with recurrent airway obstruction. Vet Immunol Immunopathol. 2005;104:91. 702. van der Haegen A., Künzle F., Gerber V., et al. Mast cells and IgE-bearing cells in lungs of RAO-affected horses. Vet Immunol Immunopathol. 2005;108:325. 703. Couëtil L.L., Art T., de Mo arts B., et al. DNA binding activity of transcription factors in bronchial cells of horses with recurrent airway obstruction. Vet Immunol Immunopathol. 2006;113:11. 704. Franchini M., Gilli U., Akens M.K., et al. The role of neutrophil chemotactic cytokines in the pathogenesis of equine chronic obstructive pulmonary disease (COPD). Vet Immunol Immunopathol. 1998;66:53. 705. Debrue M., Hamilton E., Joubert P., et al. Chronic exacerbation of equine heaves is associated with an increased expression of interleukin-17 mRNA in bronchoalveolar lavage cells. Vet Immunol Immunopathol. 2005;105:25. 706. Ainsworth D.M., Wagner B., Franchini M., et al. Time-dependent alterations in gene expression of interleukin-8 in the bronchial epithelium of horses with recurrent airway obstruction. Am J Vet Res. 2006;67:669. 707. Ainsworth D.M., Wagner B., Erb H.N., et al. E ects of in vitro exposure to hay dust on expression of interleukin-17, -23, -8, and -1beta and chemokine (C-X-C motif) ligand 2 by pulmonary mononuclear cells isolated from horses chronically affected with recurrent airway disease. Am J Vet Res. 2007;68:1361. 708. Art T., Kirschvink N., Smith N., et al. Indices of oxidative stress in blood and pulmonary epithelium lining uid in horses su ering from recurrent airway obstruction. Equine Vet J. 1999;31:397. 709. Bowie A., O’Neill L.A. Oxidative stress and nuclear factor-kappaB activation: a reassessment of the evidence in the light of recent discoveries. Biochem Pharmacol. 2000;59:13. 710. Barnes P.J., Karin M. Nuclear factor-kappaB: a pivotal transcription factor in chronic inflammatory diseases. N Engl J Med. 1997;336:1066. 711. Bureau F., Bonizzi G., Kirschvink N., et al. Correlation between nuclear factor-kappaB activity in bronchial brushing samples and lung dysfunction in an animal model of asthma. Am J Respir Crit Care Med. 2000;161:1314. 712. Koivunen A.L., Maisi P., Konttinen Y.T., et al. Collagenolytic activity and its sensitivity to doxycycline inhibition in tracheal aspirates of horses with chronic obstructive pulmonary disease. Acta Vet Scand. 1997;38:9.

713. Raulo S.M., Maisi P. Gelatinolytic activity in tracheal epithelial lining fluid and in blood from horses with chronic obstructive pulmonary disease. Am J Vet Res. 1998;59:818. 714. Raulo S.M., Sorsa T.A., Maisi P.S. Concentrations of elastinolytic metalloproteinases in respiratory tract secretions of healthy horses and horses with chronic obstructive pulmonary disease. Am J Vet Res. 2000;61:1067. 715. Dixon P.M. Pulmonary artery pressures in normal horses and in horses affected with chronic obstructive pulmonary disease. Equine Vet J. 1978;10:195. 716. Sage A.M., Valberg S., Hayden D.W., et al. Echocardiography in a horse with cor pulmonale from recurrent airway obstruction. J Vet Intern Med. 2006;20:694. 717. Johansson A.M., Gardner S.Y., Atkins C.E., et al. Cardiovascular e ects of acute pulmonary obstruction in horses with recurrent airway obstruction. J Vet Intern Med. 2007;21:302. 718. Robinson N.E., Olszewski M.A., Boehler D., et al. Relationship between clinical signs and lung function in horses with recurrent airway obstruction (heaves) during a bronchodilator trial. Equine Vet J. 2000;32:393. 719. Derksen F.J., Scott J.S., Miller D.C., et al. Bronchoalveolar lavage in ponies with recurrent airway obstruction (heaves). Am Rev Respir Dis. 1985;132:1066. 720. Lavoie J.P., Dalle S., Breton L., et al. Bronchiectasis in three adult horses with heaves. J Vet Intern Med. 2004;18:757. 721. Bakos Z., Vörös K., Kellokoski H., et al. Comparison of the caudal lung borders determined by percussion and ultrasonography in horses with recurrent airway obstruction. Acta Vet Hung. 2003;51:249. 722. Evans A.G., Paradis M.R., O’Callaghan M. Intradermal testing of horses with chronic obstructive pulmonary disease and recurrent urticaria. Am J Vet Res. 1992;53:203. 723. Lorch G., Hillier A., Kwochka K.W., et al. Results of intradermal tests in horses without atopy and horses with chronic obstructive pulmonary disease. Am J Vet Res. 2001;62:389. 724. Lorch G., Hillier A., Kwochka K.W., et al. Comparison of immediate intradermal test reactivity with serum IgE quantitation by use of a radioallergosorbent test and two ELISA in horses with and without atopy. J Am Vet Med Assoc. 2001;218:1314. 725. Jose-Cunilleras E., Kohn C.W., Hillier A., et al. Intradermal testing in healthy horses and horses with chronic obstructive pulmonary disease, recurrent urticaria, or allergic dermatitis. J Am Vet Med Assoc. 2001;219:1115. 726. Wong D.M., Buechner-Maxwell V.A., Manning T.O., et al. Comparison of results for intradermal testing between clinically normal horses and horses a ected with recurrent airway obstruction. Am J Vet Res. 2005;66:1348. 727. Dixon P.M., Railton D.I., McGorum B.C. Equine pulmonary disease: a case control study of 300 referred cases. Part 3: Ancillary diagnostic findings. Equine Vet J. 1995;27:428. 728. Slocombe R. Pathology of the airways. (In International Workshop on Equine Chronic Airway Disease, Michigan State University 16-18 June 2000.). Equine Vet J. 2001;33:6. 729. Herszberg B., Ramos-Barbón D., Tamaoka M., et al. Heaves, an asthma-like equine disease, involves airway smooth muscle remodeling. J Allergy Clin Immunol. 2006;118:382. 730. Kaup F.J., Drommer W., Damsch S., et al. Ultrastructural ndings in horses with chronic obstructive pulmonary disease (COPD). II: Pathom*orphological changes of the terminal airways and the alveolar region. Equine Vet J. 1990;22:349. 731. Kaup F.J., Drommer W., Deegen E. Ultrastructural ndings in horses with chronic obstructive pulmonary disease (COPD). I: Alterations of the larger conducting airways. Equine Vet J. 1990;22:343. 732. Miskovic M., Couëtil L.L., Thompson C.A. Lung function and airway cytologic pro les in horses with recurrent airway obstruction maintained in low-dust environments. J Vet Intern Med. 2007;21:1060. 733. Jackson C.A., Berney C., Jefcoat A.M., et al. Environment and prednisone interactions in the treatment of recurrent airway obstruction (heaves). Equine Vet J. 2000;32:432. 734. Vandenput S., Votion D., Duvivier D.H., et al. E ect of a set stabled environmental control on pulmonary function and airway reactivity of COPD a ected horses. Vet J. 1998;155:189. 735. DeLuca L., Erb H.N., Young J.C., et al. The effect of adding oral dexamethasone to feed alterations on the airway cell inflammatory gene expression in stabled horses affected with recurrent airway obstruction (RAO). J Vet Intern Med. 2008;22:427. 736. Lapointe J.M., Lavoie J.P., Vrins A.A. E ects of triamcinolone acetonide on pulmonary function and bronchoalveolar lavage cytologic features in horses with chronic obstructive pulmonary disease. Am J Vet Res. 1993;54:1310. 737. Picandet V., Léguillette R., Lavoie J.P. Comparison of e cacy and tolerability of iso upredone and dexamethasone in the treatment of horses a ected with recurrent airway obstruction (‘heaves’). Equine Vet J. 2003;35:419. 738. Genetzky R.M., Loparco F.V. Clinical efficacy of clenbuterol with COPD in horses. J Equine Vet Sci. 1985;5:320. 739. Laan T.T., Bull S., Pirie R.S., et al. The anti-inflammatory effects of IV administered clenbuterol in horses with recurrent airway obstruction. Vet J. 2006;171:429. 740. Henrikson S.L., Rush B.R. Efficacy of salmeterol xinafoate in horses with recurrent airway obstruction. J Am Vet Med Assoc. 1961;218:2001. 741. Camargo F.C., Robinson N.E., Berney C., et al. Trimetoquinol: bronchodilator e ects in horses with heaves following aerosolised and oral administration. Equine Vet J. 2007;39:215. 742. Rush B.R., Hoskinson J.J., Davis E.G., et al. Pulmonary distribution of aerosolized technetium Tc 99m pentetate after administration of a single dose of aerosolized albuterol sulfate in horses with recurrent airway obstruction. Am J Vet Res. 1999;60:764. 743. Pearson E.G., Riebold T.W. Comparison of bronchodilators in alleviating clinical signs in horses with chronic obstructive pulmonary disease. J Am Vet Med Assoc. 1989;194:1287. 744. Léguillette R., Désévaux C., Lavoie J.P. E ects of pentoxifylline on pulmonary function and results of cytologic examination of bronchoalveolar lavage uid in horses with recurrent airway obstruction. Am J Vet Res. 2002;63:459. 745. Matthews A.G., Hackett I.J., Lawton W.A. The mucolytic effect of Sputolosin in horses with respiratory disease. Vet Rec. 1988;122:106. 746. Wilson D.V., Schott H.C.2nd, Robinson N.E., et al. Response to nasopharyngeal oxygen administration in horses with lung disease. Equine Vet J. 2006;38:219. 747. Broadstone R.V., Robinson N.E., Gray P.R., et al. Effects of furosemide on ponies with recurrent airway obstruction. Pulm Pharmacol. 1991;4:203. 748. Rubie S., Robinson N.E., Stoll M., et al. Flunixin meglumine blocks frusemide-induced bronchodilation in horses with chronic obstructive pulmonary disease. Equine Vet J. 1993;25:138. 749. Wilson D.V., Berney C.E., Peroni D.L., et al. The effects of a single acupuncture treatment in horses with severe recurrent airway obstruction. Equine Vet J. 2004;36:489. 750. van den Hoven R., Zappe H., Zitterl-Eglseer K., et al. Study of the e ect of Bronchipret on the lung function of ve Austrian saddle horses su ering recurrent airway obstruction (heaves). Vet Rec. 2003;152:555. 751. Jean D., Vrins A., Lavoie J.P. Respiratory and metabolic e ects of massive administration of isotonic saline solution in heaves-a ected and control horses. Equine Vet J. 2004;36:628. 752. Naylor J.M., Clark E.G., Clayton H.M. Chronic obstructive pulmonary disease: Usefulness of clinical signs, bronchoalveolar lavage, and lung biopsy as diagnostic and prognostic aids. Can Vet J. 1992;33:591. 753. Dixon P.M., McGorum B. Pasture-associated seasonal respiratory disease in two horses. Vet Rec. 1990;126:9.

754. Mair T.S. Obstructive pulmonary disease in 18 horses at summer pasture. Vet Rec. 1996;138:89. 755. Costa L.R., Seahorn T.L., Moore R.M., et al. Correlation of clinical score, intrapleural pressure, cytologic ndings of bronchoalveolar uid, and histopathologic lesions of pulmonary tissue in horses with summer pasture-associated obstructive pulmonary disease. Am J Vet Res. 2000;61:167. 756. Seahorn T.L., Beadle R.E., McGorum B.C., et al. Quanti cation of antigen-speci c antibody concentrations in tracheal lavage uid of horses with summer pasture-associated obstructive pulmonary disease. Am J Vet Res. 1997;58:1408. 757. Beadle R.E., Horohov D.W., Gaunt S.D. Interleukin-4 and interferon-gamma gene expression in summer pasture-associated obstructive pulmonary disease a ected horses. Equine Vet J. 2002;34:389. 758. McGorum B.C., Dixon P.M. Summer pasture associated obstructive pulmonary disease (SPAOD): an update. Eq Vet Educ. 1999;11:121. 759. Couëtil L.L., Hoffman A.M., Hodgson J., et al. Inflammatory airway disease of horses. J Vet Intern Med. 2007;21:356. 760. Robinson N.E. Inflammatory airway disease: defining the syndrome. Conclusions of the Havemeyer Workshop. Equine Vet Educ. 2003;15:61. 761. Sweeney C.R., Humber K.A., Roby K.A. Cytologic findings of tracheobronchial aspirates from 66 thoroughbred racehorses. Am J Vet Res. 1992;53:1172. 762. Martin B.B.Jr., Beech J., Parente E.J. Cytologic examination of specimens obtained by means of tracheal washes performed before and after high-speed treadmill exercise in horses with a history of poor performance. J Am Vet Med Assoc. 1999;214:673. 763. Wood J.L., Newton J.R., Chanter N., et al. Inflammatory airway disease, nasal discharge and respiratory infections in young British racehorses. Equine Vet J. 2005;37:236. 764. Allen K.J., Tremaine W.H., Franklin S.H. Prevalence of in ammatory airway disease in national hunt horses referred for investigation of poor athletic performance. Equine Vet J. 529(36), 2006. Suppl 765. Burrell M.H., Wood J.L., Whitwell K.E., et al. Respiratory disease in thoroughbred horses in training: the relationships between disease and viruses, bacteria and environment. Vet Rec. 1996;139:308. 766. Wood J.L., Burrell M.H., Roberts C.A., et al. Streptococci and Pasteurella spp. associated with disease of the equine lower respiratory tract. Equine Vet J. 1993;25:314. 767. Christley R.M., Hodgson D.R., Rose R.J., et al. A case-control study of respiratory disease in Thoroughbred racehorses in Sydney, Australia. Equine Vet J. 2001;33:256. 768. Chapman P.S., Green C., Main J.P., et al. Retrospective study of the relationships between age, in ammation and the isolation of bacteria from the lower respiratory tract of thoroughbred horses. Vet Rec. 2000;146:91. 769. Burrell M.H., Whitwell K.E., Wood J.L., et al. Pyrexia associated with respiratory disease in young thoroughbred horses. Vet Rec. 1994;134:219. 770. Wood J.L., Newton J.R., Chanter N., et al. Association between respiratory disease and bacterial and viral infections in British racehorses. J Clin Microbiol. 2005;43:120. 771. Moore B.R., Krakowka S., Robertson J.T., et al. Cytologic evaluation of bronchoalveolar lavage uid obtained from standardbred racehorses with in ammatory airway disease. Am J Vet Res. 1995;56:562. 772. Moore B.R., Krakowka S., Cummins J.M., et al. Changes in airway in ammatory cell populations in standardbred racehorses after interferon-alpha administration. Vet Immunol Immunopathol. 1996;49:347. 773. Moore B.R., Krakowka S., Mcvey D.S., et al. In ammatory markers in bronchoalveolar lavage uid of standardbred racehorses with in ammatory airway disease: response to interferon-alpha. Equine Vet J. 1997;29:142. 774. Tyler WS, Pascoe JR, Aguilera-Tejero E, et al: Morphological e ects of autologous blood in airspaces of equine lungs. In Proceedings of the 10th Veterinary Respiratory Symposium, Michigan State University, September 22-24, 1991. 775. McKane S.A., Slocombe R.F. Alveolar fibrosis and changes in equine lung morphometry in response to intrapulmonary blood. Equine Vet J. 451(34), 2002. Suppl 776. Derksen F.J., Williams K.J., Uhal B.D., et al. Pulmonary response to airway instillation of autologous blood in horses. Equine Vet J. 2007;39:334. 777. Tremblay G.M., Ferland C., Lapointe J.M., et al. Effect of stabling on bronchoalveolar cells obtained from normal and COPD horses. Equine Vet J. 1993;25:194. 778. Gerber V., Robinson N.E., Luethi S., et al. Airway inflammation and mucus in two age groups of asymptomatic well-performing sport horses. Equine Vet J. 2003;35:491. 779. Davis M.S., Williams C.C., Meinkoth J.H., et al. In ux of neutrophils and persistence of cytokine expression in airways of horses after performing exercise while breathing cold air. Am J Vet Res. 2007;68:185. 780. Koch C., Straub R., Ramseyer A., et al. Endoscopic scoring of the tracheal septum in horses and its clinical relevance for the evaluation of lower airway health in horses. Equine Vet J. 2007;39:107. 781. Hughes K.J., Malikides N., Hodgson D.R., et al. Comparison of tracheal aspirates and bronchoalveolar lavage in racehorses. 1. Evaluation of cytological stains and the percentage of mast cells and eosinophils. Aust Vet J. 2003;81:681. 782. Malikides N., Hughes K.J., Hodgson J.L. Comparison of tracheal aspirates before and after high-speed treadmill exercise in racehorses. Aust Vet J. 2007;85:414. 783. Sanchez A., Couëtil L.L., Ward M.P., et al. Effect of airway disease on blood gas exchange in racehorses. J Vet Intern Med. 2005;19:87. 784. Mazan M.R., Vin R., Hoffman A.M. Radiographic scoring lacks predictive value in inflammatory airway disease. Equine Vet J. 2005;37:541. 785. Hoffman A., Mazan M. Programme of lung function testing horses suspected with small airway disease. Eqine Vet Educ. 1999;11:322. 786. Pirrone F., Albertini M., Clement M.G., et al. Respiratory mechanics in Standardbred horses with sub-clinical in ammatory airway disease and poor athletic performance. Vet J. 2007;173:144. 787. Moore I., Horney B., Day K., et al. Treatment of inflammatory airway disease in young standardbreds with interferon alpha. Can Vet J. 2004;45:594. 788. Mazan M.R., Hoffman A.M. Effects of aerosolized albuterol on physiologic responses to exercise in standardbreds. Am J Vet Res. 1812;62:2001. 789. Pascoe J.R., Ferraro G.L., Cannon J.H., et al. Exercise-induced pulmonary hemorrhage in racing thoroughbreds: a preliminary study. Am J Vet Res. 1981;42:703. 790. Raphel C.F., Soma L.R. Exercise-induced pulmonary hemorrhage in Thoroughbreds after racing and breezing. Am J Vet Res. 1982;43:1123. 791. Speirs V.C., van Veenendaal J.C., Harrison I.W., et al. Pulmonary haemorrhage in standardbred horses after racing. Aust Vet J. 1982;59:38. 792. MacNamara B., Bauer S., Iafe J. Endoscopic evaluation of exercise-induced pulmonary hemorrhage and chronic obstructive pulmonary disease in association with poor performance in racing Standardbreds. J Am Vet Med Assoc. 1990;196:443. 793. Lapointe J.M., Vrins A., McCarvill E. A survey of exercise-induced pulmonary haemorrhage in Quebec standardbred racehorses. Equine Vet J. 1994;26:482. 794. Hillidge C.J., Lane T.J., Johnson E.L., et al. Preliminary investigations of exercise-induced pulmonary hemorrhage in racing quarter horses. J Equine Vet Sci. 1984;4:21. 795. Hillidge C.J., Lane T.J., Whitlock T.W. Exercise-induced pulmonary hemorrhage in the racing Appaloosa horse. J Equine Vet Sci. 1985;5:531. 796. Voynick B.T., Sweeney C.R. Exercised-induced pulmonary hemorrhage in polo and racing horses. J Am Vet Med Assoc. 1986;188:301. 797. Mason D.K., Collins E.A., Watkins K.L. Exercise-induced pulmonary hemorrhage in horses. In: Snow D.H., Persson S.G.B., Rose R.J., editors. Equine Exercise Physiology. Cambridge, England: Burlington Press, 1983. 798. Hinchcli K.W., Jackson M.A., Morley P.S., et al. Association between exercise-induced pulmonary hemorrhage and performance in Thoroughbred racehorses. J Am Vet Med

Assoc. 2005;227:768. 799. Robinson N.E., Derksen F.J. Small airway obstruction as a cause of exercise-associated pulmonary hemorrhage: an hypothesis. Am Assoc Equine Pract. 1980;26:421. 800. O’Callaghan M.W., Pascoe J.R., Tyler W.S., et al. Exercise-induced pulmonary haemorrhage in the horse: results of a detailed clinical, post mortem and imaging study. VIII. Conclusions and implications. Equine Vet J. 1987;19:428. 801. Cook W.R., Williams R.M., Kirker-Head C.A., et al. Upper airway obstruction (paratial asphyxia) as the possible cause of exercise-induced pulmonary hemorrhage in the horse: an hypothesis. J Equine Vet Sci. 1988;8:11. 802. Ducharme N.G., Hackett R.P., Gleed R.D., et al. Pulmonary capillary pressure in horses undergoing alteration of pleural pressure by imposition of various upper airway resistive loads. Equine Vet J Suppl. 1999;30:27. 803. Clarke A.F. Review of exercise induced pulmonary haemorrhage and its possible relationship with mechanical stress. Equine Vet J. 1985;17:166. 804. Schroter R.C., Marlin D.J., Denny E. Exercise-induced pulmonary haemorrhage (EIPH) in horses results from locomotory impact induced trauma–a novel, unifying concept. Equine Vet J. 1998;30:186. 805. Erickson H.H., Bernard S.L., Glenny R.W., et al. Effect of furosemide on pulmonary blood flow distribution in resting and exercising horses. J Appl Physiol. 1999;86:2034. 806. West J.B., Mathieu-Costello O., Jones J.H., et al. Stress failure of pulmonary capillaries in racehorses with exercise-induced pulmonary hemorrhage. J Appl Physiol. 1993;75:1097. 807. Slocombe R: EIPH: the role of airways. Proceedings of the World Equine Airway Symposium, Edinburgh, Scotland, 2001. 808. Newton J.R., Rogers K., Marlin D.J., et al. Risk factors for epistaxis on British racecourses: evidence for locomotory impact-induced trauma contributing to the aetiology of exercise-induced pulmonary haemorrhage. Equine Vet J. 2005;37:402. 809. Birks E.K., Durando M.M., McBride S. Exercise-induced pulmonary hemorrhage. Vet Clin North Am Equine Pract. 2003;19:87. 810. Birks E.K., Mathieu-Costello O., Fu Z., et al. Very high pressures are required to cause stress failure of pulmonary capillaries in thoroughbred racehorses. J Appl Physiol. 1997;82:1584. 811. Poole D.C., Epp T.S., Erickson H.H. Exercise-induced pulmonary haemorrhage (EIPH): mechanistic bases and therapeutic interventions. Equine Vet J. 2007;39:292. 812. Kindig C.A., Ramsel C., McDonough P., et al. Inclined running increases pulmonary haemorrhage in the Thoroughbred horse. Equine Vet J. 2003;35:581. 813. Manohar M., Goetz T.E. Pulmonary vascular pressures of exercising thoroughbred horses with and without endoscopic evidence of EIPH. J Appl Physiol. 1996;81:1589. 814. Meyer T.S., Fedde M.R., Gaughan E.M., et al. Quantification of exercise-induced pulmonary haemorrhage with bronchoalveolar lavage. Equine Vet J. 1998;30:284. 815. Newton J.R., Wood J.L. Evidence of an association between inflammatory airway disease and EIPH in young Thoroughbreds during training. Equine Vet J. 417(34), 2002. Suppl 816. Doucet M.Y., Viel L. Clinical, radiographic, endoscopic, bronchoalveolar lavage and lung biopsy ndings in horses with exercise-induced pulmonary hemorrhage. Can Vet J. 2002;43:195. 817. Takahashi T., Hiraga A., Ohmura H., et al. Frequency of and risk factors for epistaxis associated with exercise-induced pulmonary hemorrhage in horses: 251,609 race starts (1992-1997). J Am Vet Med Assoc. 2001;218:1462. 818. Gunson D.E., Sweeney C.R., Soma L.R. Sudden death attributable to exercise-induced pulmonary hemorrhage in racehorses: nine cases (1981-1983). J Am Vet Med Assoc. 1988;193:102. 819. Boden L.A., Charles J.A., Slocombe R.F., et al. Sudden death in racing Thoroughbreds in Victoria, Australia. Equine Vet J. 2005;37:269. 820. Art T., Tack S., Kirschvinck N., et al. E ect of instillation into lung of autologous blood on pulmonary function and tracheobronchial wash cytology. Equine Vet J. 442(34), 2002. Suppl 821. McKane S.A., Canfield P.J., Rose R.J. Equine bronchoalveolar lavage cytology: survey of thoroughbred racehorses in training. Aust Vet J. 1993;70:401. 822. Hinchcliff K.W. Counting red cells–is it an answer to EIPH? Equine Vet J. 2000;32:362. 823. Doucet M.Y., Viel L. Alveolar macrophage graded hemosiderin score from bronchoalveolar lavage in horses with exercise-induced pulmonary hemorrhage and controls. J Vet Intern Med. 2002;16:281. 824. Wisner E.R., O’Brien T.R., Lakritz J., et al. Radiographic and microscopic correlation of di use interstitial and bronchointerstitial pulmonary patterns in the caudodorsal lung of adult thoroughbred horses in race training. Equine Vet J. 1993;25:293. 825. Pascoe J.R., O’Brien T.R., Wheat J.D., et al. Radiographic aspects of exercise-induced pulmonary hemorrhage in racing horses. Vet Radiol. 1983;24:85. 826. O’Callaghan M.W., Goulden B.E. Radiographic changes in the lungs of horses with exercise-induced epistaxis. N Z Vet J. 1982;30:117. 827. Manohar M., Hutchens E., Coney E. Furosemide attenuates the exercise-induced rise in pulmonary capillary blood pressure in horses. Equine Vet J. 1994;26:51. 828. Gleed R.D., Ducharme N.G., Hackett R.P., et al. Effects of furosemide on pulmonary capillary pressure in horses exercising on a treadmill. Equine Vet J. 1999;30:102. Suppl, 1999 829. Lester G., Clark C., Rice B., et al. E ect of timing and route of administration of furosemide on pulmonary hemorrhage and pulmonary arterial pressure in exercising thoroughbred racehorses. Am J Vet Res. 1999;60:22. 830. Sweeney C.R., Soma L.R. Exercise-induced pulmonary hemorrhage in thoroughbred horses: response to furosemide or hesperidin-citrus bio avinoids. J Am Vet Med Assoc. 1984;185:195. 831. Sweeney C.R., Soma L.R., Maxson A.D., et al. Effects of furosemide on the racing times of Thoroughbreds. Am J Vet Res. 1990;51:772. 832. Pascoe J.R., McCabe A.E., Franti C.E., et al. E cacy of furosemide in the treatment of exercise-induced pulmonary hemorrhage in Thoroughbred racehorses. Am J Vet Res. 2000;46:1985. 833. Gross D.K., Morley P.S., Hinchcli K.W., et al. E ect of furosemide on performance of Thoroughbreds racing in the United States and Canada. J Am Vet Med Assoc. 1999;215:670. 834. Manohar M., Goetz T.E. Pulmonary vascular pressures of strenuously exercising Thoroughbreds during intravenous infusion of nitroglycerin. Am J Vet Res. 1999;60:1436. 835. Kindig C.A., McDonough P., Finley M.R., et al. NO inhalation reduces pulmonary arterial pressure but not hemorrhage in maximally exercising horses. J Appl Physiol. 2001;91:2674. 836. Goetz T.E., Manohar M., Hassan A.S., et al. Nasal strips do not a ect pulmonary gas exchange, anaerobic metabolism, or EIPH in exercising Thoroughbreds. J Appl Physiol. 2001;90:2378. 837. Kindig C.A., McDonough P., Fenton G., et al. Efficacy of nasal strip and furosemide in mitigating EIPH in Thoroughbred horses. J Appl Physiol. 2001;91:1396. 838. Walker H.J., Evans D.L., Slocombe R.F., et al. E ect of corticosteroid and bronchodilator therapy on bronchoalveolar lavage cytology following intrapulmonary blood inoculation. Equine Vet J. 516(36), 2006. Suppl 839. Hillidge C.J., Whitlock T.W., Lane T.J. Failure of inhaled disodium cromoglycate aerosol to prevent exercise-induced pulmonary hemorrhage in racing quarter horses. J Vet Pharmacol Ther. 1987;10:257.

840. Erickson H.H., Hidreth T.S., Poole D.C., et al. Management of exercise-induced pulmonary hemorrhage in nonracing performance horses. Comp Cont Educ Pract Vet. 2001;23:1090. 841. Weideman H., Schoeman S.J., Jordaan G.F., et al. Epistaxis related to exercise-induced pulmonary haemorrhage in south African Thoroughbreds. J S Afr Vet Assoc. 2003;74:127. 842. Sundberg J.P., Burnstein T., Page E.H., et al. Neoplasms of Equidae. J Am Vet Med Assoc. 1977;170:150. 843. Cotchin E., Baker-Smith J. Correspondence: Tumours in horses encountered in an abattoir survey. Vet Rec. 1975;97:339. 844. Colbourne C.M., Bolton J.R., Mills J.N., et al. Mesothelioma in horses. Aust Vet J. 1992;69:275. 845. Mair T.S., Lane J.G., Lucke V.M. Clinicopathological features of lymphosarcoma involving the thoracic cavity in the horse. Equine Vet J. 1985;17:428. 846. Mair T.S., Brown P.J. Clinical and pathological features of thoracic neoplasia in the horse. Equine Vet J. 1993;25:220. 847. Mair T.S., Rush B.R., Tucker R.L. Clinical and diagnostic features of thoracic neoplasia in the horse. Equine Vet Educ. 2004;16:30. 848. Mair T.S., Hillyer M.H. Clinical features of lymphosarcoma in the horse: 77 cases. Equine Vet Educ. 1991;4:108. 849. Garber J.L., Reef V.B., Reimer J.M. Sonographic findings in horses with mediastinal lymphosarcoma: 13 cases (1985-1992). J Am Vet Med Assoc. 1994;205:1432. 850. De Clercq D., van Loon G., Lefere L., et al. Ultrasound-guided biopsy as a diagnostic aid in three horses with a cranial mediastinal lymphosarcoma. Vet Rec. 2004;154:722. 851. Peroni J.F., Horner N.T., Robinson N.E., et al. Equine thoracoscopy: normal anatomy and surgical technique. Equine Vet J. 2001;33:231. 852. Saulez M.N., Schlipf J.W.Jr., Cebra C.K., et al. Use of chemotherapy for treatment of a mixed-cell thoracic lymphoma in a horse. J Am Vet Med Assoc. 2004;224:733. 853. Kramer J.W., Nickels F.A., Bell T. Cytology of diffuse mesothelioma in the thorax of a horse. Equine Vet J. 1976;8:81. 854. Straub R., Tscharner C., Pauli B., et al. Mesothelioma of the pleura in a horse. Schweiz Arch Tierheilkd. 1974;116:207. 855. Kolbl V.S. Pleural mesothelioma as a cause of death in a horse. Wein tierarztl Mschr. 1979;66:22. 856. Fry M.M., Magdesian K.G., Judy C.E., et al. Antemortem diagnosis of equine mesothelioma by pleural biopsy. Equine Vet J. 2003;35:723. 857. Stoica G., Cohen N., Mendes O., et al. Use of immunohistochemical marker calretinin in the diagnosis of a di use malignant metastatic mesothelioma in an equine. J Vet Diagn Invest. 2004;16:240. 858. Nickels F.A., Brown C.M., Breeze R.G. Myoblastoma. Equine granular cell tumor. Mod Vet Pract. 1980;61:593. 859. Turk M.A., Breeze R.G. Histochemical and ultrastructural features of an equine pulmonary granular cell tumour (myoblastoma). J Comp Pathol. 1981;91:471. 860. Misdorp W., Nauta-van Gelder H.L. ‘Granular-cell myoblastoma’ in the horse. A report of 4 cases. Pathol Vet. 1968;5:385. 861. Parodi A.L., Tassin P., Rigoulet J. Myoblastome a cellules granuleuses: Trois nouvelles observations a localisation pulmonair chez le cheval. Rec Med Vet. 1974;150:489. 862. Parker G.A., Novilla N.M., Brown A.C., et al. Granular cell tumour (myoblastoma) in the lung of a horse. J Comp Pathol. 1979;89:421. 863. Inoue S., Okada N., Midoro K., et al. An equine case of granular cell tumor with chondroplasia. Nippon Juigaku Zasshi. 1987;49:581. 864. Kelly L.C., Hill J.E., Harner S., et al. Spontaneous equine pulmonary granular cell tumors: morphologic, histochemical, and immunohistochemical characterization. Vet Pathol. 1995;32:101. 865. Bouchard P.R., Fortna C.H., Rowland P.H., et al. An immunohistochemical study of three equine pulmonary granular cell tumors. Vet Pathol. 1995;32:730. 866. Sutton R.H., Coleman G.T. A pulmonary granular cell tumour with associated hypertrophic osteopathy in a horse (abstract). N Z Vet J. 1995;43:123. 867. Kagawa Y., Hirayama K., Tagami M., et al. Immunohistochemical analysis of equine pulmonary granular cell tumours. J Comp Pathol. 2001;124:122. 868. Pusterla N., Norris A.J., Stacy B.A., et al. Granular cell tumours in the lungs of three horses. Vet Rec. 2003;153:530. 869. Murphy J.R., Breeze R.G., McPherson E.A. Myxoma of the equine respiratory tract. Mod Vet Pract. 1978;59:529. 870. Dill S.G., Moise N.S., Meschter C.L. Cardiac failure in a stallion secondary to metastasis of an anaplastic pulmonary carcinoma. Equine Vet J. 1986;18:414. 871. Schultze A.E., Sonea I., Bell T.G. Primary malignant pulmonary neoplasia in two horses. J Am Vet Med Assoc. 1988;193:477. 872. Anderson J.D., Leonard J.M., Zeliff J.A., et al. Primary pulmonary neoplasm in a horse. J Am Vet Med Assoc. 1992;201:1399. 873. Uphoff C.S., Lyncoln J.A. A primary pulmonary tumor in a horse. Equine Pract. 1987;9:19. 874. Sullivan D.J. Cartilaginous tumors (chondroma and chondrosarcoma) in animals. Am J Vet Res. 1960;21:531. 875. Clem M.R., O’Brien T.R., Feeney D.A., et al. Pulmonary chondrosarcoma in a horse. Compend Contin Educ Pract Vet. 1986;8:S964. 876. Rossdale P.D., Greet T.R.C., McGladdery A.J., et al. Pulmonary leiomyosarcoma in a 13 year-old Thoroughbred stallion presenting as a di erential diagnosis to recurrent airway obstruction. Equine Vet Educ. 2004;16:21. 877. Godber L.M., Brown C.M., Mullaney T.P. Polycystic hepatic disease, thoracic granular cell tumor and secondary hypertrophic osteopathy in a horse. Cornell Vet. 1993;83:227. 878. Heinola T., Heikkila M., Ruohoniemi M., et al. Hypertrophic pulmonary osteopathy associated with granular cell tumour in a mare. Vet Rec. 2001;149:307. 879. Alexander J.E., Keown G.H., Palotay J.L. Granular cell myoblastoma with hypertrophic pulmonary osteoarthropathy in a mare. J Am Vet Med Assoc. 1965;146:703. 880. Facemire P.R., Chilcoat C.D., Sojka J.E., et al. Treatment of granular cell tumor via complete right lung resection in a horse. J Am Vet Med Assoc. 2000;217:1522. 881. Ohnesorge B., Gehlen H., Wohlsein P. Transendoscopic electrosurgery of an equine pulmonary granular cell tumor. Vet Surg. 2002;31:375. 882. Rendle D.I., Hewetson M., Barron R., et al. Tachypnoea and pleural effusion in a mare with metastatic pancreatic adenocarcinoma. Vet Rec. 2006;159:356. 883. Chaffin M.K., Fuentealba I.C., Schmitz D.G., et al. Endometrial adenocarcinoma in a mare. Cornell Vet. 1990;80:65. 884. Gunson D.E., Gillette D.M., Beech J., et al. Endometrial adenocarcinoma in a mare. Vet Pathol. 1980;17:776. 885. Whiteley L.O., Leininger J.R., Wolf C.B., et al. Malignant squamous cell thymoma in a horse. Vet Pathol. 1986;23:627. 886. Vaala W.E. Pleuritis and pleural effusion in a mare secondary to disseminated squamous cell carcinoma. Comp Cont Educ Pract Vet. 674, 1987. 887. Ford T.S., Vaala W.E., Sweeney C.R., et al. Pleuroscopic diagnosis of gastroesophageal squamous cell carcinoma in a horse. J Am Vet Med Assoc. 1987;190:1556. 888. Hovda L.R., Shaftoe S., Rose M.L., et al. Mediastinal squamous cell carcinoma and thyroid carcinoma in an aged horse. J Am Vet Med Assoc. 1990;197:1187. 889. Furuoka H., Taniyama H., Matsui T., et al. Malignant thymoma with multiple metastases in a mare. Nippon Juigaku Zasshi. 1987;49:577. 890. Milne J.C. Malignant melanomas causing Horner’s syndrome in a horse. Equine Vet J. 1986;18:74. 891. Frye F.L., Knight H.D., Brown S.I. Hemangiosarcoma in a horse. J Am Vet Med Assoc. 1983;182:287. 892. Hargis A.M., McElwain T.F. Vascular neoplasia in the skin of horses. J Am Vet Med Assoc. 1984;184:1121. 893. Freestone J.F., Williams M.M., Norwood G. Thoracic haemangiosarcoma in a 3-year-old horse. Aust Vet J. 1990;67:269. 894. Johnson J.E., Beech J., Saik J.E. Disseminated hemangiosarcoma in a horse. J Am Vet Med Assoc. 1988;193:1429. 895. Jean D., Lavoie J.P., Nunez L., et al. Cutaneous hemangiosarcoma with pulmonary metastasis in a horse. J Am Vet Med Assoc. 1994;204:776.

896. Southwood L.L., Schott H.C.2nd, Henry C.J., et al. Disseminated hemangiosarcoma in the horse: 35 cases. J Vet Intern Med. 2000;14:105. 897. Collins M.B., Hodgson D.R., Hutchins D.R., et al. Haemangiosarcoma in the horse: three cases. Aust Vet J. 1994;71:296. 898. Waller T., Rubarth S. Haemangioendothelioma in domestic animals. Acta Vet Scand. 1967;8:234. 899. Gruys E., Kok H.A., Van Der Werff Y.D. Dyspnoea due to intrathoracic haemorrhage and haemangiosarcoma in a horse (author’s transl). Tijdschr Diergeneeskd. 1976;101:310. 900. Reinacher V.M. Hamangioenothelome in der skellettmuskulatur eines pferds. Berl Munch Tierarztl Wochenschr. 1978;91:121. 901. Katayama Y., Oikawa M.A., Yoshihara T., et al. Clinical and immunohistochemical observation of hemangiosarcoma in a racing thoroughbred. Equine Pract. 1996;18:24. 902. Warren A.L., Summers B.A. Epithelioid variant of hemangioma and hemangiosarcoma in the dog, horse, and cow. Vet Pathol. 2007;44:15. 903. Johns I., Stephen J.O., Del Piero F., et al. Hemangiosarcoma in 11 young horses. J Vet Intern Med. 2005;19:564. 904. Rebhun W.C., Bertone A. Equine lymphosarcoma. J Am Vet Med Assoc. 1984;184:720. 905. Keen J.A., Swain J.M., Rhind S.M., et al. Lymphoproliferative disease resembling lymphomatoid granulomatosis in a thoroughbred mare. J Vet Intern Med. 2004;18:904. 906. Waugh S.L., Long G.G., Uriah L., et al. Metastatic hemangiosarcoma in the equine: report of two cases. J Equine Med Surg. 1977;1:311. ∗

References: 74, 518, 891, 893, 894, 897-900, and 906.

CHAPTER 10 Cardiovascular Diseases John D. Bonagura, Virginia B. Reef, Colin C. Schwarzwald∗

Cardiovascular (CV) system function is critical to exercise, thermoregulation, and the blood ow–dependent functions of the lungs, kidneys, gut, and reproductive system. Heart and vascular diseases are common in horses, but fortunately the underlying lesion is often minor and well tolerated. However, clinically signi cant CV disease can develop in horses with clinical signs that can include arrhythmia, exercise intolerance, congestive heart failure (CHF), weakness or collapse, systemic infection, or sudden death. The clinical evaluation of the equine CV system is often perplexing. Horses are renowned for a variety of physiologic murmurs and arrhythmias. Furthermore, only when horses are undergoing exercise at the highest levels do they reach the limits of normal CV function. Clinical assessment is best served by an awareness of normal variation, the appreciation of relevant diseases, diagnostic studies and prognostic indicators, and an understanding of available management options. Communication of these issues constitutes the focus of this chapter. Equine cardiology has advanced from a study of physiologic variation and speculation to one of accurate diagnosis and focused therapy, although many clinical issues remain unresolved. Much of the important groundwork in clinical equine cardiology was established by studies of normal CV physiology, cardiac catheterization, pathology, cardiac auscultation, electrocardiography, and echocardiography. These data, de ning the normal anatomic and psychologic features of the equine CV system, constitute the backdrop of the clinical examination. Other examinations of importance to CV assessment include ambulatory electrocardiography, functional exercise testing, and biochemical tests of cardiac injury. Appropriate selection and interpretation of these tests allows the clinician to identify and quantify most diseases of the heart and circulation. No recent comprehensive studies of CV disease prevalence are available. Holmes, Darke, and Else33 observed that 2.5% of hospitalized horses were in atrial brillation (AF).33 Else and Holmes27-29 noted myocardial brosis in 14.3% of horses examined at necropsy and evidence of chronic valvular disease in approximately 25% of the hearts examined. Various CV lesions were considered important in 8.5% of 480 consecutive losses in a necropsy study conducted by Baker and Ellis.52 CV diseases are probably the third most common cause of poor performance after musculoskeletal and respiratory diseases.53-56 Occult heart disease, cardiac arrhythmias, and vascular lesions are considered important reasons for unexplained sudden death.52,57-69 Certainly, most clinicians encounter manifestations of CV disease or dysfunction on a regular basis. The assessment of CV disease in a horse is predicated on a competent clinical examination, the clinician’s knowledge, and the ability to order and evaluate diagnostic studies. Incomplete information may impede an accurate diagnosis, foster miscommunication of risks to the client, or delay the proper course of management. Fundamentals of CV anatomy, physiology, and electrophysiology are reviewed elsewhere.9,20,70-77 This chapter o ers a framework for understanding the lesions, pathophysiology, diagnosis, and management of important congenital and acquired conditions of the heart and major vessels. Clinical aspects of circulatory shock are described in this volume, and the reader is referred elsewhere for the management of cardiopulmonary arrest.78,79

ANATOMIC CORRELATES OF CARDIOVASCULAR DISEASES Most diagnostic techniques, including cardiac auscultation, electrocardiography, cardiac catheterization, and echocardiography, are predicated on an understanding of cardiac anatomy and physiology. The CV system is divided into two separate circulations: (1) systemic and (2) pulmonary. The systemic circulation has a greater venous capacitance, ventricular pumping pressure, arterial pressure, and vascular resistance.73,75-77 Despite these di erences, the functions of these two circulations are interdependent, as the following examples illustrate. Systemic and pulmonary circulations are arranged in series, therefore cardiac output (CO) from the left ventricle (LV) and the right ventricle (RV) must be equivalent. Accordingly, failure of either ventricle limits CO. Isolated LV failure, as with severe mitral regurgitation (MR), can cause right-sided failure; this is explained by the increased pulmonary venous pressure causing pulmonary hypertension and imparting a pressure load on the RV. A third example is the case of isolated RV failure with marked RV dilatation. Here the leftward bulging of the ventricular septum impairs lling of the LV. This last situation also can develop in chronic pericarditis. Arrhythmias also a ect both sides of the heart so that the development of AF in the setting of structural heart disease can promotes biventricular heart failure. The heart consists of unique active and passive components,75 and di erent diagnostic methods are needed to evaluate these structures and associated functions. Normal heart action requires coordination of electrical activity, muscular contraction and relaxation, and valve motion. When reviewing heart anatomy, and subsequently cardiac pathology, it is useful to consider the anatomic integrity of the pericardium,

myocardium, endocardium, and valves, specialized impulse-forming and conduction systems, and blood vessels.70 Using this approach, the causes of CV disease can be conveniently subdivided into anatomic, physiologic (functional), and causative diagnoses (Boxes 10-1 and 10-2). BOX 10-1 CARDIAC DIAGNOSES Anatomic Diagnosis Cardiac malformation Valvular (endocardial) disease Myocardial disease Pericardial disease Cor pulmonale (pulmonary disease leading to secondary heart disease) Disorder of the impulse-forming or conduction system Vascular disease

Physiologic Diagnosis Systemic: pulmonary shunting Left to right Right to left Valvular insufficiency Valvular stenosis Myocardial (systolic) dysfunction Diastolic dysfunction Cardiac rhythm disturbance Cardiac-related syncope Heart insufficiency or failure (limited cardiac output) Congestive heart failure Shock Sudden cardiac death Cardiopulmonary arrest

Etiologic Diagnosis Malformation (genetic) Degenerative disease Metabolic or endocrine disease Neoplasia Nutritional disorder Inflammatory disease Infective or parasitic Noninfective Immune-mediated

Idiopathic Ischemic injury Idiopathic disorder Iatrogenic disease Toxic injury Traumatic injury

BOX 10-2 CAUSES OF CARDIOVASCULAR DISEASE Congenital Cardiac Malformation Simple systemic-to-pulmonary shunts (left to right) Atrial septal defect Ventricular septal defect Paramembranous defect Ventricular inlet defect Subarterial (subpulmonic) defect Muscular defect Patent ductus arteriosus Patent foramen ovale (permitting right-to-left shunting) Valvular dysplasia Mitral stenosis/atresia Pulmonic stenosis (bicuspid pulmonary valve) Pulmonary atresia (leading to a right-to-left shunt) Tricuspid stenosis/atresia (leading to a right-to-left shunt) Aortic stenosis/insufficiency (bicuspid or quadricuspid valve) Subaortic rings with stenosis Tetralogy of Fallot Pulmonary atresia with ventricular septal defect (pseudotruncus arteriosus) Double-outlet right ventricle Subaortic stenosis Hypoplastic left-side of the heart Other complex malformations

Valvular Heart Disease Causing Valve Insufficiency or Stenosis Congenital valve malformation Semilunar valve fenestrations causing valve insufficiency Degenerative (fibrosis) or myxomatous disease causing valve insufficiency Vavular prolapse Bacterial endocarditis causing valve insufficiency with or without stenosis Rupture of a chorda tendinea causing mitral or tricuspid valve insufficiency Rupture of a valve leaflet causing flail leaflet and valve insufficiency

Noninfective valvulitis Valvular regurgitation following dilation of the heart or a great vessel Papillary muscle dysfunction causing valvular insufficiency

Myocardial Disease Idiopathic dilated cardiomyopathy: ventricular dilation and myocardial contractility failure Myocarditis Myocardial fibrosis Ischemic (embolic?) myocardial fibrosis Parasitic (Strongylus) embolization Myocardial degeneration/necrosis Myocardial ischemia Myocardial infarction Toxic injury (e.g., monensin) Nutritional deficiencies (e.g., selenium deficiency) Myocardial neoplasia Lymphosarcoma Melanoma Hemangioma/hemangiosarcoma Pulmonary carcinoma Infiltrative myocardial disease (e.g., amyloidosis)

Pericardial Disease Pericardial effusion with or without cardiac tamponade Infective: bacterial or viral Idiopathic pericardial effusion Constrictive pericardial disease Mass lesion (intrapericardial or extrapericardial) compressing the heart

Pulmonary Hypertension and Cor Pulmonale Pulmonary hypertension following left-sided heart disease Pulmonary vascular disease following left-to-right shunt Immature pulmonary circulation Primary bronchopulmonary or pulmonary vascular disease Alveolar hypoxia with reactive pulmonary arterial vasoconstriction Severe acidosis Pulmonary thromboembolism

Cardiac Arrhythmias (see Box 10-9) Atrial arrhythmias Junctional (nodal) arrhythmias Ventricular arrhythmias

Conduction disturbances

Vascular Diseases Congenital vascular lesions Rupture of the aorta, pulmonary artery, or systemic artery Aneurysm of the aortic sinus of Valsalva Aortic or aortoiliac degenerative disease Arteritis Infective Immune-mediated (Jugular) Venous thrombosis/thrombophlebitis Thrombophlebitis Pulmonary embolism Mass lesion or tumor obstructing blood flow Aorto-iliac thrombosis

Pericardial Disease The pericardium limits cardiac dilatation, acts as a barrier against contiguous infection, and contributes to the diastolic properties of the heart. The pericardial space is formed by the re ection of the two major pericardial membranes—(1) the parietal pericardium and (2) the visceral pericardium (the epicardium)—and normally contains such a small amount of serous uid that it cannot be seen by echocardiography. Pericardial e usion leading to cardiac compression (tamponade) impairs ventricular lling and diastolic function, typically causing right-sided CHF. Some cases of pericarditis progress to constrictive pericardial disease, which severely limits ventricular filling. Pericardial e usion can develop as a primary disorder or secondary to pleuropneumonia. Infective pericarditis can produce an e usion su cient to cause cardiac tamponade or eventual constriction of the heart.55,80-97 Sterile, idiopathic pericardial e usion also has been reported in horses. The volume of e usion can be substantial and can lead to cardiac decompensation.87 Cardiac mass lesions and intrapericardial tumors have been reported sporadically.89,92,98,99 Cranial mediastinal tumors (lymphosarcoma) or abscesses secondary to pleuropneumonia also can compress the heart and mimic pericardial disease.100 Clinical aspects of pericardial disease are discussed later in this chapter.

Myocardial Disease The myocardium forms the bulk of the atrial and ventricular muscular walls. The right atrium (RA) communicates with the RV inlet through the right atrioventricular (AV) or tricuspid valve. The RV appears crescent shaped on cross-sectional echocardiographic examination and is functionally U shaped. The RV inlet is located in the right hemithorax and the outlet, pulmonary valve, and main pulmonary artery (PA) on the left side of the chest. The left atrium (LA) is caudal to the RA and separated by the atrial septum. The LA is dorsal to the LV through which it communicates across the left atrioventricular (mitral) valve. The LV is circular in cross-section when viewed by echocardiography and separated from the RV by the ventricular septum. The septum and free walls are thicker than the RV free wall (by approximately 2.5 to 3 times). Persistent embryologic openings in the cardiac septa are known as septal defects, with the ventricular septal defect (VSD) representing the most common cardiac anomaly in most equine practices (see Box 10-2). The LV is functionally V shaped with an inlet and outlet separated by the cranioventral (septal or “anterior”) lea et of the mitral valve (Figure 10-1). The aorta originates in the LV outlet, continuous with the ventricular septum cranially and in brous continuity with the septal mitral lea et caudally. This great vessel exits from near the center of the heart and to the right of the main PA.

FIGURE 10-1

Sagittal view of the equine heart. The thicknesses of the ventricles, the position of the atria relative to the ventricles, and the relationship of the left ventricular (LV) inlet and outlet are evident. The bicuspid valve referred to in this gure is the mitral valve. The circular appearance of the left atrium (LA) and the relationship of the septal cusp of the mitral valve to the LV inlets and outlets are notable. These aspects are important when examining the heart by echocardiography. v.a., Segment of aortic valve. (From Sisson S, Grossman JD: Anatomy of the domestic animals, ed 4, Philadelphia, 1953, WB Saunders.)

The myocardium may dilate or hypertrophy in response to exercise,101,102 increased work caused by structural cardiac disease, or as a consequence of a noncardiac disorder. Ventricular or atrial dilatation is recognized echocardiographically or at necropsy by distention and rounding of the a ected chambers, including a “double-apex” sign when marked RV enlargement occurs. Lesions causing systolic pressure overload lead to concentric hypertrophy.103 More common in horses are lesions such as incompetent valves or shunts that cause ventricular volume overload with dilatation and eccentric ventricular hypertrophy. Increased cardiac work also occurs in response to exercise, severe anemia, and infections. In these situations, compensatory increases in CO, sympathetic activation, and peripheral vasodilation occur to maintain oxygen delivery to the tissues.104,105 The overall prevalence of myocardial disease is unknown; however, multifocal areas of brosis are commonly found at necropsy.28,29,106-110 Whether these areas indicate prior in ammation, toxic injury, or ischemic necrosis caused by intramural coronary disease is uncertain. Cases of multifocal or di use myocarditis have been observed. Myocardial in ammation and myocardial failure can lead to cardiac arrhythmias and heart failure.4,111,112 Idiopathic dilated cardiomyopathy develops sporadically and is recognized echocardiographically or by nuclear studies as a dilated, hypokinetic LV or RV.90,113 Ingestion of monensin or other ionophores can cause mild to severe myocardial injury.114-120 Neoplastic in ltration is considered rare.59,98,99,121 Impaired myocardial function as a consequence of regional ischemia has been sought using stress echocardiography immediately after treadmill exercise or pharmacologic stress, but this diagnosis requires further de nition.122 Myocardial contraction is dictated by electrical activity of the myocardium; accordingly, cardiac arrhythmias—especially AF or ventricular tachycardia (VT)—can limit CO and cause exercise intolerance in performance animals (see later discussion). Clinical aspects of myocardial disease are discussed later in this chapter.

Valvular and Endocardial Diseases The cardiac chambers are lined by the endocardium, which also covers the four cardiac valves and is continuous with the endothelium of the great vessels. Normal valves govern the one-way ow of blood through the heart by preventing signi cant regurgitation of blood from higher to lower pressure zones. The AV inlet valves—the tricuspid and the mitral—are anchored by the collagenous chordae tendineae and papillary muscles and are supported by a valve “annulus” and the caudal atrial walls (see Figures 10-1 and 10-2).70 The mitral valve consists of two major cusps and several accessory cusps.45 The tricuspid valve is the largest valve and consists of three well-de ned lea ets. Lesions of any portion of the AV valve apparatus or dilatation of the ventricle can lead to valvular insu ciency (see Box 10-2). The aortic and pulmonary valves each consist of three semilunar lea ets that close during diastole to protect the ventricles from the higher arterial blood pressure (ABP). Aortic valvular tissue in horses is not simply passive and will contract in response to a number of adrenergic and vascular agonists, such as angiotensin II and endothelin.123 The left and right main coronary arteries originate within the aortic valve sinuses (of Valsalva).

FIGURE 10-2

Anatomy of the left atrioventricular (mitral) valve. Opened left atrium (LA) and LV viewed from the caudal perspective. The large anterior (cranioventral or septal) lea et in the center of the gure is notable. Chordae tendineae attach the valve to the papillary muscles. The ventricle has been cut so that the multiple cusps of the posterior (caudodorsal or mural) leaflet are visible to the left and the right of the anterior leaflet.

Valvular disorders in horses are common.∗ Congenital valve stenosis, dysplasia, or atresia are recognized sporadically in foals.4,111,127,155-168 Degenerative diseases of the aortic, mitral, and tricuspid valves are very common in mature horses,4,127,155-168 and endocarditis can develop on any cardiac valve† (see Box 10-2). Valvular lesions of obscure cause, including nonseptic valvulitis, have been recognized sporadically. Tricuspid and MR, of unspeci ed cause, is often detected in high performance animals, including Standardbred and National Hunt horses.182-185 MR caused by rupture of a chorda tendineae is recognized in both foals and mature animals.38,132,147,186 Clinical aspects of valvular heart disease are discussed later in this chapter.

Disease of the Impulse-Forming and Conduction Systems The specialized cardiac tissues consist of the sinoatrial (SA) node, internodal pathways, AV node, bundle of His, bundle branches, fascicles, and Purkinje system (Figure 10-3). The SA node, a relatively large, crescent-shaped structure, is located subepicardially at the junction of the right auricle and cranial vena cava. Well-documented sinus node disease, although suggested,187 is rare; in contrast, vagally induced sinus arrhythmias are common.9,187-189 The equine atrial muscle mass is large and predisposes the horse to development of re-entrant rhythms and brillatory conduction.190 The AV node, situated in the ventral atrial septum, and the bundle of His, which continues into the bundle branches, are sites for AV block, both physiologic (vagal), and infrequently, pathologic in nature. Conduction is slow across the normal AV node.191-193 The His-Purkinje system in the ventricular septum and ventricular myocardium can act as substrates for junctional and ventricular ectopic impulses and tachycardias. Because the horse has relatively complete penetration of Purkinje bers in the ventricles— except for a small portion of the LV free wall—the substantial equine ventricles are electrically activated in a relatively short time (≈110 msec).7

FIGURE 10-3

Impulse-forming and conduction systems of the heart. The impulse originates in the sinoatrial node (SAN) and is propagated across the right atrium (RA) and left atrium (LA), generating the P wave. Specialized internodal and interatrial (Bachmann’s bundle) pathways facilitate impulse conduction. The impulse is delayed in the atrioventricular node (AVN) and rapidly conducted through the bundle of His (H), bundle branches and Purkinje network (top). Electrical activation of ventricular myocytes generates the QRS complex. The automaticity of the SA node and conduction across the AV node are modulated by the autonomic nervous system (bottom). (Courtesy Dr. Robert L. Hamlin. From Schwarzwald CC, Bonagura JD, Muir WW: The cardiovascular system. In Muir WW, Hubbell JA, editors: Equine anesthesia: monitoring and emergency therapy, ed 2, St Louis, 2009, WB Saunders).

The autonomic nervous system extensively innervates the heart and in uences cardiac rhythms. 4,111,134,194-199 Interplay between the sympathetic and parasympathetic branches normally controls heart rate and rhythm in response to changes in ABP.9,200 The vagus nerve innervates supraventricular tissues extensively and probably a ects proximal ventricular tissues to a minor extent. Vagal in uence generally depresses heart rate, AV conduction, excitability, and myocardial inotropic (contractile) state. However, because vagotonia also shortens the action potential and refractory period of atrial myocytes, high vagal activity is a predisposing factor in the development of AF.201 Innervation of the stimulatory sympathetic nervous system is extensive throughout the heart and has e ects generally opposite to those of the parasympathetic system. β 1-Adrenergic receptors dominate in the equine heart,202 but presumably, other autonomic subtype receptors exist, including α-adrenergic receptors and small numbers of β 2-adrenoceptors.76,203 The notable increase in heart rate that attends exercise is related to increased sympathetic e erent activity and withdrawal of parasympathetic tone.9 Increases in heart rate to 220 to 240 beats/min are not uncommon with maximal exercise.204-208 The exact role of dysautonomia in the genesis of cardiac arrhythmias has not been determined; however, infusion of autonomic receptor agonists and antagonists can be associated with direct or baroreceptor-induced changes in heart rate and rhythm.209-212 Cardiac arrhythmias are discussed later in this chapter.

Vascular Diseases There are three major subdivisions of the circulation exist: (1) systemic, (2) coronary, and (3) pulmonary. The arteries and veins consist of three layers: (1) adventitia, (2) media, and (3) intima. The overall structure and function of each layer varies with the vessel and location. Vascular receptors20,73,75,76 and anatomic lesions in uence vascular resistance and blood ow. α-Adrenoceptors dominate in the systemic vasculature, and blood pressure (BP) is generally raised by vasoconstriction after stimulation of postsynaptic α-adrenergic receptors by norepinephrine, epinephrine, or infused α-adrenergic receptor agonists like phenylephrine.213,214 The presence of vasodilator β 2-adrenergic receptors is clinically relevant, insofar as infused β 2-agonists cause vasodilation in circulatory beds that contain high β-agonist adrenergic receptor density. Many vascular beds also dilate after the production of local vasodilator substances, such as nitric oxide, released during exercise, stress, or metabolic activity.73,75 Dopaminergic receptors, when present in vascular walls, may be stimulated, causing vasodilation, provided vasoconstricting α-adrenergic activity does not dominate. Stimulation of histamine (H1) receptors or serotonin (5-HT) receptors causes arteriolar dilatation, venular constriction, and increased capillary permeability.75 Infusion of endothelin215 or of calcium salts causes arterial vasoconstriction,216 whereas administration of calcium channel antagonists (e.g., verapamil, diltiazem) causes vasodilation of vascular smooth muscle.217 Various vascular lesions have been reported in horses (see Box 10-2). Rupture of the aorta, PA, or middle uterine artery is devastating and often lethal.∗ The aorta may also rupture into the heart, creating an aortic to cardiac stula.220,221 Although parasitic arteritis may predispose

to vascular injury, the cause of most vascular lesions, including aortic-iliac thrombosis, is unknown.222-227 Causes of vasculitis include Strongylus vulgaris infestation of the cranial mesenteric artery, infective thrombophlebitis of the jugular veins, equine viral arteritis, and suspected immune-mediated disease.106,108,228 Neoplasms can obstruct blood ow by external compression or through invasion, more often a ecting the right side of the circulation. Examples include obstruction of the PA by a lung tumor and obstruction of venous return by neoplastic compression or invasion of the vena cava. Clinical features of vascular disease are discussed later in this chapter.

CLINICAL CARDIOVASCULAR PHYSIOLOGY The clinician must appreciate elementary aspects of normal heart function to perform a clinical CV examination and understand the abnormalities associated with heart disease and CHF.229 Central to this are the electrical-mechanical correlates of Wiggers cardiac cycle.

Cardiac Cycle The association between electrical and mechanical events of the heart rst described by Wiggers has been reviewed in standard physiology textbooks (Figure 10-4).73,75-77 From a study of this cycle, it is evident that cardiac electrical activity precedes pressure and volume changes; therefore arrhythmias can exert deleterious hemodynamic e ects, especially during exercise, illness, or anesthesia. Relevant aspects of this cycle are now considered.

FIGURE 10-4

A, The cardiac (Wiggers) cycle of the horse. This drawing integrates the electric, pressure, mechanical, and ow events of diastole and systole and demonstrates the origins of the heart sounds. Electric activity precedes mechanical events. See the text for a full description. AVC, Closure of the mitral (atrioventricular) valve; AVO, opening of the mitral (atrioventricular) valve; SLO, opening of the aortic (semilunar) valve; SLC, closure of the aortic (semilunar) valve. B, Determinants of cardiac output (CO) and blood pressure (BP). (A, Modi ed from Detweiler DK, Patterson DF: The cardiovascular system. In Cattcott EJ, Smithcors JF, editors: Equine medicine and surgery, ed 2, Santa Barbara, CA, 1972, American Veterinary Publications. B, From Muir WW, Hubbell JA: Equine anesthesia, ed 2, St Louis, 2009, Saunders.)

The P wave of the electrocardiogram (ECG) stems from electrical activation of the atria, late in ventricular diastole and after the ventricles have been largely lled. During the ensuing atrial contraction, the atrial sound (fourth heart sound or S4) is generated and the ventricle is lled to its end diastolic volume. The increase in atrial pressure, the atrial a wave, is re ected as a normal jugular pulse in the ventral cervical region. The magnitude of the atrial contribution to ventricular lling generally placed at 15% to 20% at rest, but increases dramatically during high heart rates. Therefore atrial tachyarrhythmias such as AF have the greatest e ect on CO during exercise or tachycardia. The QRS complex heralds ventricular systole. After depolarization of the ventricular myocytes, calcium enters the cell to trigger release of calcium stores in the sarcoplasmic reticulum. Increased cytosolic calcium interacts with the cardiac troponin complex on actin and myosin laments to shorten the myo laments and develop tension. These events are enhanced by sympathetic activity or drugs such as digoxin or dobutamine and depressed by anesthetics and drugs that impair calcium entry into cells. The abrupt increase in ventricular wall tension and chamber pressure closes the AV valves (coinciding with the vibrations of the rst heart sound, S1) and increases intraventricular pressure

(isovolumetric period) until the semilunar valves open.23,25 At this instant the ventricular walls move inward (Figure 10-5, B and C) and blood is ejected into the great vessel as the ventricular pressure increases to peak value and creates a similar peak ABP (see Figures 10-5, D, and 10-6). The contracting heart twists during systole, and the LV strikes the chest wall caudal to the left olecranon causing the cardiac impulse or “apex beat.” This early systolic movement, coincident with opening of the aortic valve, is a useful timing clue for cardiac auscultation and for identifying the mitral valve area for auscultation. The delay between the onset of the QRS and the opening of the semilunar valves, termed the pre-ejection period, can be measured by Doppler echocardiography and is an index of ventricular myocardial contractility such that positive inotropic drugs shorten the pre-ejection period.46,230-237 Blood is ejected into the aorta and PA with an initial velocity that generally peaks near 1 ms/sec and can be measured by Doppler echocardiography.149,230 The aortic ejection time usually exceeds 400 ms in a horse at rest, and reductions of either ejection velocity or ejection time are suggestive of reduced LV function. A functional systolic ejection murmur is often heard during ejection (see Figure 10-4, A). Such murmurs, by de nition, must begin after the rst sound and end before the second sound. The di erence between the diastolic and systolic pressure (pulse pressure) and rate of rise of pressure contribute to a palpable arterial pulse during midsystole (see Figures 10-4, A, and 10-6). The precise timing of the pulse depends on the proximity of the palpation site relative to the heart. At the end of the ejection period, as ventricular pressures fall below those of the corresponding arteries, the semilunar valves close coincident with the high-frequency second heart sound (S2) or sounds and the incisura of the arterial pressure curves.19,23,25 The pulmonary valve may close either after or before the aortic valve.3,4,238 Asynchronous valve closure may lead to audible splitting of S2, which is normal but can be extreme in some horses with lung disease. During the ejection period the ventricular volume curve graphs a marked reduction from the end diastolic volume: this volume ejected is de ned as the stroke volume. The ratio of the stroke volume to the end diastolic volume is the ejection fraction, a commonly used index of systolic heart function and correlated to the often-used shortening fraction of the M (motion)-mode echocardiogram (see Figure 10-5). Contraction of the ventricles causes the atrial pressures to decline leading to the x descent of the atrial pressure curve and a brief systolic collapse of the jugular vein. Subsequent to atrial lling during ventricular systole, a positive pressure wave, the v wave, occurs in the atrial and venous pressure curves. Severe TR accentuates this wave and may lead to pathologic systolic pulsations extending up the jugular furrow. Finally, a decline in ventricular pressure (isovolumetric relaxation) occurs, related to o -loading of calcium from the troponin apparatus and resequestration into the sarcoplasmic reticulum. This active ventricular relaxation is associated initially with closure of the semilunar valves and eventually by opening of the AV valves. Relaxation in healthy hearts is enhanced by sympathetic activity. Conversely, myocardial ischemia can impair active relaxation, and it is likely that subendocardial ischemia combined with reduced diastolic lling time contribute to the marked elevations in LA pressures observed during galloping or other high-intensity exercise.239,240

FIGURE 10-5

Ventricular function and echocardiography. A, Derivation of the M-mode echocardiogram. The lines demonstrate typical paths of M-mode recording planes (1,

ventricular/papillary muscle; 2, chordae tendineae; 3, anterior mitral valve [AMV]; 4, aortic root and left atrium [LA]/auricle; TW, thoracic wall; RVW, right ventricular wall; RV, right ventricle; S, septum; LV, left ventricle; AV, aortic valve; AO, aortic out ow tract; LVW, left ventricular wall; PMV, posterior mitral valve; LA, left atrium). B, The drawing demonstrates the appearance of the M-mode echocardiogram at each level (PER, pericardium; RS, LS, right and left sides of the ventricular septum; EN, endocardium). C, M-mode echocardiogram demonstrating the method of measuring left ventricular shortening fraction (LVSF), in which D is diastolic dimension and S is systolic dimension (LVSF = D – S/D). The prominent thickening of the walls during systole is notable. The end systolic excursion of the LV wall is visible (arrow). In practice, the systolic dimensions of the ventricular septum and the LV wall generally are measured along the same line as demonstrated for the LV lumen in systole (S) (W, LV wall; VS, ventricular septum). D, Left: Doppler echocardiographic recordings of LV lling (left) and ejection (right). Transmitral in ow is characterized by an early diastolic rapid- lling wave (E), low-velocity mid-diastolic lling (diastasis), and a presystolic atrial lling wave (A). Right: The velocity pro le of aortic ejection is characterized by a roughly triangular appearance with rapid acceleration of blood ow into the ascending aorta in early to midsystole and termination of ow at the time of aortic valve closure. The area under the velocity spectrum curve (velocity-timeintegral) correlates directly with ventricular stroke volume. The pre-ejection period (the time between the start of the QRS and beginning of ejection) and the ejection times (ET, arrows) are load-dependent indices of LV function.

FIGURE 10-6

Compressed electrocardiogram (ECG) with simultaneous arterial blood pressure (ABP) recording in a horse with second-degree atrioventricular block. The progressive increase in ABP triggers a baroreceptor re ex leading to atrioventricular conduction block (upper arrows) and a corresponding fall in the ABP (lower arrows). Presumably this mechanism, along with sinus arrhythmia and sinus arrest, represents vagally induced mechanisms for controlling ABP in the standing horse.

Ventricular lling commences just as the AV valves open. As shown in Figure 10-5, D, ventricular diastole can be subdivided into three general phases: (1) rapid ventricular lling, (2) diastasis, and (3) atrial contraction.73,75,76 These phases are readily observed using pulsedwave Doppler echocardiography or Doppler tissue imaging.241 Once the ventricles have relaxed and the atrial pressure exceeds the corresponding ventricular pressure, the AV valves open. At that instant, rapid lling ensues with a peak velocity of about 0.5 to 1 m/sec, but varying directly with the heart rate.149 The ventricular pressures increase only slightly during this phase, whereas the ventricular volume curves change dramatically from the venous return. Rapid lling may be associated with a functional protodiastolic murmur, which is concluded by the third heart sound (S3), the low-frequency vibrations occurring near the termination of rapid ventricular lling. The loss of atrial volume and corresponding decline in the atrial pressure (the y descent) is re ected in the jugular furrow as the vein collapses. After rapid lling, a period of greatly reduced low-velocity– lling diastasis ensues. This period may last for seconds during vagal arrhythmias such as sinus bradycardia, pronounced sinus arrhythmia, or second-degree AV block. With markedly exaggerated pauses, the jugular vein may begin to ll prominently. The last phase of diastole is the contribution to ventricular lling caused by the atrial contraction. A functional presystolic murmur has been associated with this period between the fourth and first heart sounds. During the cardiac cycle the atrium functions as a reservoir for blood (ventricular systole), a conduit for venous return (early to middiastole), and as a pressure pump (atrial systole). Mechanical atrial function of the LA can be studied using two-dimensional (2D) echocardiography and advanced Doppler echocardiographic methods. Impaired electrical and mechanical function of the atria may predispose to recurrent atrial arrhythmias such as AF.190,242-245

ASSESSMENT OF VENTRICULAR FUNCTION The ability of the ventricles to eject blood depends on both systolic and diastolic ventricular function, as well as heart rate and rhythm (Box 10-3). The most commonly used measurements of overall ventricular performance and circulatory function are invasively or noninvasively determined ABP, rate of ventricular pressure change (dp/dt), CO, stroke volume, ejection fraction, LV shortening fraction, systolic time intervals, central venous pressure, PA or wedge pressures, and arteriovenous oxygen difference (A-V DO2).∗ CO, the amount of blood pumped by the left (or right) ventricle in 1 minute (L/min), is the product of ventricular stroke volume (ml/beat) multiplied by the heart rate (beats/min; see Figure 10-4, B). Cardiac index refers to the CO divided by (indexed to) the body surface area (or body mass). CO coupled to systemic vascular resistance determines the mean ABP: an increase in either variable raises mean arterial pressure. Values for CO vary widely with the size and activity of the horse and are often in uenced by drug therapy or anesthesia.210,260-275 A noninvasive estimate of CO can also be obtained using Doppler techniques.233,234,257,276-285 BOX 10-3 DETERMINANTS OF CARDIAC FUNCTION Systolic Function: Determinants of Ventricular Stroke Volume Preload [+]—ventricular end-diastolic volume Determinants of diastolic function (see below) Plasma volume

Venous pressure/venous return Myocardial inotropism [+]—contractility of the myocardium Sympathetic activity Myocardial disease Drugs (positive or negative inotropic agents) Myocardial perfusion Ventricular afterload [–]—wall tension required to eject blood Aortic impedance Vascular resistance Ventricular volume (tension increases with dilation) Ventricular wall thickness (thin walls have higher tension) Cardiac lesions increasing workload [–] Valvular regurgitation (common) Valvular stenosis (rare) Septal defects and shunts

Diastolic Function: Determinants of Ventricular Filling Pleural/mediastinal factors (pressure, mass lesions) Pericardial function (intrapericardial pressure, constriction) Myocardial relaxation Myocyte stiffness Ventricular wall distensibility (chamber and myocyte compliance) Venous pressure and venous return (must be matched with compliance) Heart rate and ventricular filling time (shortened by tachycardia) Myocardial perfusion (ischemia impairs relaxation) Atrial contribution to filling (lost in atrial fibrillation) Cardiac rhythm (arrhythmias can alter atrioventricular contraction sequencing) Atrioventricular valve function

Cardiac Output Cardiac output = Stroke volume [+] × heart rate [+]

Arterial Blood Pressure Arterial blood pressure = Cardiac output [+] × vascular resistance [+]

Ventricular stroke volume depends on myocardial contractility and loading conditions (preload and afterload; see Figure 10-4, D).24,26 Although traditionally considered independent determinants of myocardial function, these variables are all interconnected and in uence force, velocity, and duration of ventricular contraction.76 Ultimately, the availability of calcium to the sarcomere is modulated by the inotropic state, the initial myocardial stretch (preload), and the tension that must be generated to eject blood into the vascular system (afterload). Myocardial contractility is increased by catecholamines, calcium, digitalis glycosides, and phosphodiesterase inhibitors.∗ Contractility is di cult to measure in the clinical setting but can be estimated noninvasively by observing directional changes in load-dependent pre-ejection or ejection phase indices of ventricular function. These include shortening and ejection fractions by M-mode and 2D echocardiography; preejection period, ejection time, ow acceleration, velocity time integral of aortic or pulmonic ejection, and peak myocardial velocity by Doppler echocardiography; and myocardial deformation or strain by computerized analysis of 2D echocardiograms or tissue Doppler studies

(see Figure 10-5).49,50,51,237,303-306 Measured variables will be in uenced by physiologic state, altered mildly by day-to-day variation307,308 or sedatives,231 and affected markedly by general anesthesia. Ventricular ber length or preload is a positive determinant of ventricular systolic function that depends on venous return and ventricular size and distensibility. The normal ventricle is highly preload dependent, such that increases in preload increase stroke volume. Dehydration, venous pooling, loss of atrial contribution to lling (AF), and recumbency all reduce ventricular lling and decrease stroke volume. When hypotension develops consequent to decreased ventricular lling, intravenous (IV) crystalloid or colloid is often administered initially to increase venous pressure and ventricular preload. Increased preload also can be observed in horses with heart disease as a consequence of impaired pump function, cardiac lesions, or uid retention. Moderate to severe valvular insu ciency increases ventricular lling pressures and preload.38,105,309-311 The increased ventricular diastolic dimension serves as a compensatory mechanism that maintains forward stroke volume in the setting of a failing ventricle or regurgitant heart valve. Ventricular preload can be estimated by determining ventricular end diastolic volume or size, as measured by echocardiography, or by measuring venous lling pressures.310 The measurement of venous lling pressures (central venous pressure, pulmonary diastolic or pulmonary capillary wedge pressure)310-314 provides an accurate gauge of preload provided that heart rate and ventricular compliance (distensibility) are normal and ventilation is relatively stable. Myocardial ischemia, which impairs myocardial relaxation, and pericardial diseases, which constrict the ventricles, both reduce ventricular compliance; in such cases, the venous lling pressures may not accurately reflect chronic changes in ventricular preload. Ventricular afterload is represented by the ventricular wall tension that must be developed to eject blood and relates to forces impeding this ejection. Peak wall tension occurs immediately before aortic valve opening. Vascular impedance during ejection also relates to the elastic, resistive, and dynamic properties of the connected great vessel and vascular tree. Increases in ventricular chamber size, arterial sti ness, or resistance, as well as marked increases in hematocrit, increase the impedance to ventricular ejection and reduce stroke volume. Afterload is di cult to measure clinically, and although BP is not identical to afterload, it may be used to estimate directional changes in afterload. LV failure, alveolar hypoxia-induced pulmonary vasoconstriction, atelectasis, and PA or venoocclusive diseases are important causes of increased RV afterload. Arterial vasodilators such as hydralazine and angiotensin converting enzyme (ACE) inhibitors decrease afterload.268,315 Ventricular synergy refers to the normal method of ventricular activation and contraction. Normal electrical activation causes a burst of activation of great mechanical advantage.20 Cardiac arrhythmias, especially ventricular rhythm disturbances, can cause dyssynergy (dyssynchrony), with a resultant decrease in stroke volume. Coronary occlusions leading to ischemic myocardial necrosis also cause dyssynergy but are considered relatively rare.57,108,316 Structural and functional competency of the cardiac valves and the ventricular septa in uence ventricular systolic function. Valvular insu ciency (or the rare stenosis) reduces ventricular stroke volume unless adequate compensation from ventricular dilatation and hypertrophy. Ventricular remodeling, combined with heart rate reserve, often allows small septal defects or mild to moderate valvular lesions to be well tolerated even during exercise. However, large defects or severe valvular lesions can create signi cant volume overload of the left heart, progressive myocardial dysfunction, and heart failure. Development of CHF is particularly likely when an arrhythmia such as AF is superimposed on a serious structural lesion. Ventricular diastolic function determines ventricular lling and preload.73,75,76,259 Factors that a ect diastolic function are indicated in Box 10-3. When diastolic function is abnormal, often greater heart rate and higher venous pressure dependencies occur for maintenance of CO. A well-recognized cause of diastolic dysfunction is constriction or compression of the heart caused by pericardial disease. Marked ventricular chamber dilatation or hypertrophy also decreases ventricular compliance and requires higher ventricular distending pressures for lling. LV diastolic dysfunction as a consequence of severe RV dilatation or hypertrophy can be explained by bulging of the ventricular septum into the LV, which impedes left-sided lling. This in uence of ventricular interdependence is observed clinically with chronic pericardial disease and severe pulmonary hypertension. Ventricular diastolic function also is a ected by arrhythmias. Persistent tachycardia shortens diastole, cardiac lling time, and coronary perfusion. With AF, the atrial contribution to lling is lost. Junctional and ventricular arrhythmias lead to AV dissociation preventing normal AV sequencing and can also create marked dyssynchrony of ventricular contraction. Objective measures of diastolic function are complicated, and no good clinical indicator of diastolic function is currently available for horses. However, diastolic dysfunction may be assumed when one of the aforementioned conditions is recognized. It is possible to measure transmitral and tricuspid in ow using Doppler techniques, but these methods are crude and depend on atrial pressure. Tissue Doppler imaging and rate of myocardial deformation also may provide insight into diastolic cardiac function but currently represent investigational techniques in horses. Imbalance between myocardial oxygen demand and delivery can reduce both ventricular systolic and diastolic function and may a ect cardiac rhythm as well. This relationship is also relevant when there is airway obstruction or bronchopulmonary disease, which can reduce arterial oxygenation.317 Myocardial oxygen demand is augmented by increasing myocardial inotropic state, heart rate, and ventricular wall tension (related to preload and afterload).76 Oxygen delivery depends on coronary anatomy and vasomotion (degree of vessel constriction),

diastolic ABP, diastolic (coronary perfusion) time, and metabolic activity of the myocardium.10,75,318-325 Normal coronary ow is highest to the LV myocardium in the ventricular septum and LV wall.325 The immediate subendocardial layer of myocardium is probably most vulnerable to ischemic injury,9,318 and altered ventricular depolarization may develop secondary to an imbalance in oxygen delivery. This probably accounts in part for the ST-T depression and changes in the T waves observed in hypotensive animals and in normal horses during sinus tachycardia. Coronary vasomotion is e ective in augmenting coronary perfusion even at high heart rates (up to 200/min in ponies); however, coronary autoregulation is not as effective if diastolic perfusing pressure decreases in the aorta.319,320,322 The clinician may use the “double product” of ABP × heart rate as a general estimate of myocardial oxygen demand.76 Persistent ST segment depression or elevation, especially at rest and normal heart rates, suggests de cient myocardial perfusion. However, ST segment and T wave changes are normal in horses examined during treadmill exercise and are therefore difficult to interpret.

CARDIOVASCULAR EXAMINATION OF THE HORSE General Approach A general approach to the recognition and diagnosis of heart disease and an assessment of its severity is summarized in Box 10-4. Undoubtedly, history and auscultation are the most important initial evaluation procedures in the CV examination of the horse. With the exception of abnormalities in ventricular function, normal cardiac auscultation in a horse with good exercise tolerance practically precludes signi cant heart disease. The initial CV physical examination should include a thorough auscultation of the heart at all valve areas, auscultation of both lung elds, palpation of the precordium, evaluation of the arterial pulses (head and limbs), inspection of the veins, and evaluation of the mucous membranes for pallor, re ll time, and cyanosis, which may develop secondary to a right-to-left cardiac shunt or severe respiratory disease (see Box 10-4). An accurate resting heart rate and respiratory rate also should be recorded. BOX 10-4 CLINICAL EXAMINATION OF THE EQUINE CARDIOVASCULAR SYSTEM Modified from Bonagura JD: Clinical evaluation and management of heart disease, Equine Vet Educ 2:31–37, 1990.

History and Physical Examination Work history and identification of possible hemodynamic dysfunction Heart rate and rhythm Examination of the arterial and venous pulses, refill, and pressure Inspection of the mucous membranes Evaluation for abnormal fluid accumulation: pulmonary, pleural, peritoneal, and subcutaneous Auscultation of the heart and lungs Measurement of arterial blood pressure (indirect method)

Electrocardiography Heart rate, rhythm, and conduction sequence P-QRS-T complexes: configuration, amplitude, duration, and axis (axis has limited value in horses) Postexercise and exercise electrocardiography†‡ Holter (digitally recorded) electrocardiogram Postexercise and exercise electrocardiography†‡

Echocardiography M-mode echocardiography: cardiac dimensions and systolic ventricular function, cardiac anatomy and valve motion, and estimation of cardiac output Two-dimensional echocardiography: cardiac anatomy and size, systolic ventricular function, identification of lesions, and estimation of cardiac output Doppler echocardiography†: identification of normal and abnormal flow, estimation of intracardiac pressures, estimation of cardiac output, and ventricular function Postexercise echocardiography†: identification of regional or global wall dysfunction or valve dysfunction exacerbated by exercise

Thoracic Radiography† Evaluation of pleural space, pulmonary parenchyma, and lung vascularity Estimation of heart size (more beneficial in foals)

Clinical Laboratory Tests Complete blood count and fibrinogen to identify anemia or inflammation Serum biochemical tests, including electrolytes, renal function tests, and muscle enzymes: These studies may be useful for assessing arrhythmias, identifying low cardiac output (azotemia), and recognizing myocardial cell injury (cardiac isoenzymes of creatine kinase or lactic dehydrogenase) Cardiac troponin-I concentration (cTnI): can be measured at rest and after exercise testing to identify myocardial damage Serum protein to identify hypoalbuminemia and hyperglobulinemia Arterial pH and blood gas analysis to evaluate pulmonary and renal function and to assess acid-base status Urinalysis to identify renal injury from heart failure or endocarditis Blood cultures for bacteremia and diagnosis of endocarditis Serum/plasma assays for digoxin, quinidine, and other drugs

Radionuclide Studies§ Detection of abnormal blood flow or lung perfusion and assessment of ventricular function

Cardiac Catheterization and Angiocardiography§ Diagnosis of abnormal blood flow and identification of abnormal intracardiac and intravascular pressures ∗Most important part of the cardiac evaluation. † May be needed

to identify paroxysmal atrial fibrillation.

‡ Generally done after referral. § Not commonly performed.

It is worth emphasizing that physical examination and a stethoscope can detect most serious cardiac disorders initially. Sustained or recurrent cardiac arrhythmias are easily discovered through cardiac auscultation and palpation of the arterial pulse, and diagnosis of the rhythm can be veri ed through electrocardiography. Pericarditis and cardiac tamponade are characterized by mu ed heart sounds or pericardial friction rubs, jugular distension, and often RV failure. Signi cant myocardial disease is usually associated with heart failure, arrhythmias, or a cardiac murmur, especially when ventricular dilatation or dysfunction causes insu ciency of the mitral or tricuspid valves. The presence of a cardiac murmur is the essential finding that leads one to consider degenerative or infective valvular disease or a congenital heart malformation.111,156 Laboratory studies, electrocardiography, and echocardiography are additional tests that are particularly useful in recognizing the underlying basis and severity of CV disease. Mild or subtle CV disease may require a detailed examination, including exercise testing before abnormalities can be objectively detected.326-332

History Cardiovascular disease may be suspected from the history or a serendipitous nding during the course of a routine examination.∗ The horse with CHF may be presented for generalized venous distention, jugular pulsations, or edema. Conversely, other cardiac problems such as arrhythmia or murmur can be incidental ndings, detected during a routine physical, prepurchase, or insurance examination. Once an abnormality has been found, a complete CV examination is aimed at determining the lesion and the signi cance of disease in terms of safety, performance capabilities, and expected longevity. The horse with clinically apparent CV disease may have subtle performance problems that are only apparent at peak performance levels. For example, slowing in the last quarter to three eighths of a race is a common historical complaint. In many cases, performance may deteriorate by only 2 to 3 seconds. In other horses, particularly in cases of AF, racing performance may decline greatly by 20 to 30 seconds or more. Horses with malignant VT may stop abruptly or even collapse. Horses with CV disease also may demonstrate excessively high heart and respiratory rates during and after exercise or may take a longer than normal time to return to a resting rate (or “cool out”). Coughing, either at rest or during exercise, tachypnea, and exercise-induced pulmonary hemorrhage are respiratory signs reported in some horses with

heart disease. CV disease must always be considered along with musculoskeletal, respiratory, metabolic, and neurologic problems in the di erential diagnosis of poor performance (Box 10-5).54-56,122,317,336,337 Other performance-related problems with CV disease can include weakness, ataxia, collapse, and sudden death (Box 10-6).52,57-68,219 BOX 10-5 CARDIOVASCULAR ASSOCIATION OF POOR PERFORMANCE Arrhythmias Atrial premature complexes Ventricular premature complexes Atrial fibrillation Supraventricular tachycardia Ventricular tachycardia Advanced second-degree atrioventricular block Complete third-degree atrioventricular block

Congenital, Valvular, or Myocardial Heart Diseases Associated with Murmurs Ventricular septal defect Mitral regurgitation Tricuspid regurgitation Aortic regurgitation Cardiomyopathy with secondary atrioventricular valvular regurgitation

Occult Heart Disease Pericardial disease Cardiomyopathy or myocarditis Ischemic myocardial disease (?)

Vascular Disorders Aortic-iliac thrombosis Jugular vein thrombosis/thrombophlebitis (bilateral) Aortic root rupture (aorto-cardiac fistula) Peripheral vein thrombosis/thrombophlebitis

BOX 10-6 CAUSES OF SUDDEN CARDIOVASCULAR DEATH Electric Disorders of the Heart (Arrhythmias) Ventricular tachycardia, flutter, or fibrillation Complete atrioventricular block Asystole

Toxic Injury to the Heart Acute myocardial failure

Anesthetics Drug- or toxin-induced arrhythmia Toxic plants Myocardial toxins Systemic toxin secondarily affecting the heart

Cardiac Tamponade Bacterial pericarditis Idiopathic pericarditis Viral pericarditis Trauma

Hemorrhage Rupture of the heart (with cardiac tamponade) Rupture of the aorta or pulmonary artery (with or without cardiac tamponade) Arterial rupture Middle uterine artery Mesenteric, omental, or other large arteries Severe pulmonary hemorrhage Rupture of the spleen or liver Brain hemorrhage

Embolism Carotid air embolism Coronary embolism or thrombosis

Electrocution Lightning Alternating current electrocution

Cardiac Trauma Cardiac catheterization or needle puncture of a ventricle leading to ventricular fibrillation Penetrating thoracic wound

Auscultation CLINICAL METHOD Cardiac auscultation is the systemic examination of the heart using a stethoscope. Auscultation is expedient and relatively sensitive for detection of serious heart disease when performed by a knowledgeable and experienced examiner. It provides information about heart rate, persistent arrhythmias, and presence or absence of congenital and acquired heart diseases. Pericardial diseases also may be recognized by auscultation. Auscultation should be conducted within the context of a medical history and general physical examination, which includes assessment of the precordium and pulses, auscultation of the thorax, and inspection for edema and abnormal ventilatory patterns.

E ective auscultation requires an understanding of anatomy, physiology, pathophysiology, and sound. Extensive clinical experience data regarding cardiac auscultation in the horse, have been published∗ and experience with Doppler echocardiography has re ned the clinician’s understanding of heart sounds and murmurs.130,145,354 Experience and training are also signi cant factors in e ective cardiac auscultation,355,356 and this examination method should be considered an acquired skill that can be constantly honed. The overall sensitivity of auscultation is high for identi cation of congenital heart diseases, significant valvular disease, and persistent cardiac arrhythmias. Sensitivity is lower for primary myocardial or pericardial diseases, unless obvious associated abnormalities exist (e.g., murmur, arrhythmia, prominent friction rub). The speci city of auscultation in the horse (e.g., the ability to distinguish a functional from a pathologic murmur or identify a speci c ow disturbance) has not been su ciently studied but most certainly depends on the clinician’s knowledge and experience,355 as well as one’s opinion related to the physiologic versus pathologic nature of valvular regurgitation in some highperformance horses. A prerequisite for auscultation is an appreciation of the normal heart sounds, the genesis of which has already been described (see Clinical Cardiovascular Physiology). The examiner must be familiar with the causes and clinical features of arrhythmias and murmurs (Tables 10-1, 10-2, and 10-3) and the areas for auscultation (Figure 10-8).∗ Auscultation is best carried out in a very quiet area because extraneous noise makes detection of soft to moderate murmurs quite di cult. The horse or foal should be su ciently restrained so that the examiner can concentrate on listening. The venous pulse should be inspected and then the arterial pulse and the precordium palpated before commencing auscultation. Although uncommonly used today, practiced examiners can use percussion of the precordial area e ectively to identify the region of cardiac dullness to identify heart size.357 TABLE 10-1 Auscultation of Cardiac Arrhythmias

TABLE 10-2 Causes of Cardiac Murmurs Cardiac Murmur Functional murmurs∗ Congenital heart disease murmurs Mitral regurgitation Tricuspid regurgitation Aortic regurgitation Pulmonary insufficiency

Lesion Identified by Echocardiography, Catheterization, or Necropsy No identifiable lesions Defect(s) in the atrial or ventricular septa; patent ductus arteriosus; atresia/stenosis of the tricuspid or pulmonic valve; valve stenosis; other malformations of the heart High-level training; degenerative thickening of the valve; bacterial endocarditis; mitral valve prolapse into left atrium; rupture of a chorda tendinea; dilated-hypokinetic ventricle (dilated cardiomyopathy); papillary muscle lesion or dysfunction; valvulitis; malformation High-level training†; same causes as for as mitral regurgitation; also pulmonary hypertension from severe left-sided heart failure or chronic respiratory disease Degeneration of the aortic valve; congenital fenestration of the valve; bacterial endocarditis‡; aortic prolapse into a ventricular septal defect or the left ventricle; flail aortic valve leaflet; valvulitis; malformation; ruptured aorta or aortic sinus of Valsalva Bacterial endocarditis‡; pulmonary hypertension; flail pulmonic valve leaflet; valvulitis; malformation; rupture of the pulmonary artery

Murmur associated with vegetative endocarditis

Insufficiency of the affected valve‡

Functional murmur may be innocent (unknown cause) or physiologic (suspected physiologic cause). Functional murmurs are common in foals and trained athletes (athletic murmur), associated with fever and high sympathetic nervous system activity (pain, stress, sepsis), and often are heard in anemic horses. Functional murmurs depend on the physiologic state and can be altered by changing the heart rate. Such dynamic auscultation is useful in detecting functional murmurs. ∗

† Doppler echocardiography can identify “silent” regurgitation across a right-sided cardiac valve in some horses; this is probably a normal finding of no clinical significance. ‡ Anatomic stenosis generally is caused by a large vegetation and also should be associated with a diastolic murmur of valvular insu ciency; increased ow across the valve, even in the absence of a true stenosis, may generate a murmur of “relative” valvular stenosis (e.g., with aortic regurgitation a systolic ejection murmur may occur because of an increased stroke volume). Modified from Bonagura JD: Clinical evaluation and management of heart disease, Equine Vet Educ 2:31–37, 1990.

TABLE 10-3 Diagnostic Algorithm for Cardiac Auscultation in the Horse

FIGURE 10-8

Cardiac auscultation areas in the horse from the left (A) and the right (B) side of the thorax. Note the size and anatomic position of the heart within the thorax. The cardiac apex (apical area) is usually located slightly above the level of the olecranon and can be identi ed by palpation of the apical beat. The cardiac base (basilar area) is located more cranially and at the level of the scapulohumeral joint. The shaded areas represent the respective valve areas (P, pulmonic; A, aortic; M, mitral; T, tricuspid). The right atrium (RA) (above T), tricuspid valve, and right ventricular (RV) inlet (in ow region, below T) are located on the right side of the thorax. Paramembranous ventricular septal defects (VSDs) that communicate with the RV inlet (below the tricuspid valve) are usually heard best along the lower right hemithorax (below T) and typically radiate from the right ventricle (RV) into the pulmonary artery (left hemithorax). Most (but not all) murmurs of tricuspid valve disease are heard best over the right chest wall, usually heard best more dorsally as they radiate into the RA (above T). The RV outlet projects to the left side of the thorax and continues into the pulmonary artery, which is located at the left dorsal cardiac base (above P). Thus murmurs originating in the RV outlet (e.g., murmurs from subpulmonic VSDs), systolic murmurs of relative pulmonic stenosis (from cardiac shunting), murmurs of pulmonic insu ciency (rare), and functional pulmonary arterial murmurs are heard best over the left chest wall (P). The aortic valve is located centrally, and diastolic murmurs of aortic insu ciency may be heard at either hemithorax, although they are usually loudest on the left (A). Functional murmurs generated in the ascending aorta are heard over the cranial basilar region of the heart. The systolic murmur of mitral regurgitation (MR) radiates to the left apex and is usually heard across the left ventricular (LV) inlet (area caudoventral of M). The regurgitant mitral jet can be directed craniodorsal or caudodorsal into the left atrium (LA) and can account for variability in murmur radiation. The functional protodiastolic murmurs associated with ventricular filling are usually evident over the ventricular inlets and may be heard on either side of the thorax. (From Reef VB: Cardiovascular system. In Orsini JA, Divers TJ, editors: Manual of equine emergencies: treatment and procedures, ed 2, Philadelphia, 2003, WB Saunders).

The stethoscope examination generally commences on the left side. Both stethoscope chest pieces—the diaphragm (applied tightly) and the bell (applied lightly)—should be used. When using a single-piece tunable diaphragm, the pressure on the chest piece should be gradually increased to optimize the sounds of interest. All auscultatory areas should be examined. The locations of the cardiac valve areas can be identified using the following method: • The left apical impulse located at the left thoracic wall is adjacent to or under the olecranon near the fth intercostal space; this location identi es the ventral region of the LV inlet. The mitral valve is dorsal to the cardiac impulse, but mitral sounds and murmurs often project well to the apex while also radiating dorsally into the LA. The rst heart sound (S1) is best heard at this location, and the ventricular lling or third heart sound (S3) is also heard well over this point. • The aortic valve area is located one (or two) intercostal spaces cranially and is dorsal to the left apical impulse. The second sound (S2) is loudest at this point, and aortic valve murmurs are heard best over this valve. The murmur of MR may also radiate dorsally and cranially to this area. Because of the central location of the aortic valve and the orientation of the ascending aorta to the right, aortic ow murmurs and the murmur of aortic regurgitation (AR) generally can be heard bilaterally, just medial to the triceps muscles. • The pulmonic valve is located slightly cranioventral (generally one intercostal space) to the aortic valve, and the pulmonary component of S2 is loudest at this point. This is also the location where splitting of the second sound is most obvious. The main PA extends dorsally from the pulmonic valve, high on the left cranial base. Murmurs that radiate into the PA are heard best at this location, including some functional murmurs and the murmur of patent ductus arteriosus (PDA). The murmur associated with an atrial septal defect (ASD) or a subarterial VSD, as well as the rare murmur of pulmonic stenosis, are typically loudest over the pulmonary valve and PA. • The tricuspid valve area is located over a wide area on the right hemithorax, dorsal to the sternum and just cranial to the mitral valve; sounds and murmurs associated with tricuspid valve disease are usually heard best over the right hemithorax. Murmurs of paramembranous VSD are often detected ventral and slightly cranial to the tricuspid valve area, dorsal to the sternum. It is worthwhile to concentrate rst on the individual heart sounds when assessing the heart rhythm, because the generation of cardiac sounds depends on the underlying electrical rhythm and heart murmurs are timed relative to the heart sounds (Figure 10-9). The atrial sound (S4) is heard after the P wave. The rst and second heart sounds (S1 and S2) encompass systole, indicating the presence of a QRS complex. A

murmur detected between the rst and second sounds is termed systolic. In contrast, a murmur heard after S2 is designated as diastolic. The distinctive atrial sound (S4) is absent in AF. Normal variation in the P-R interval causes gradual changes in the S4 to S1 interval. Absence of a QRS complex, which occurs with second-degree AV block and with nonconducted atrial premature complexes (APCs), causes a pause in which the rst and second heart sounds are absent. The heart sounds and precordial movements often can be palpated, especially over the left thoracic wall. Frequently the cardiac movements and low-frequency vibrations corresponding to the atrial contraction (S4), onset of ventricular contraction (S1), and closure of the semilunar valves (S2) can be detected. With ventricular volume overloading or vigorous normal lling, an accentuated apex beat and third sound (S3) will be palpable. Cardiac enlargement or displacement of the heart by an intrathoracic mass may lead to an abnormal location of the cardiac impulse. A loud cardiac murmur is often associated with a palpable vibration or precordial thrill.

FIGURE 10-9

Schematic diagram of the normal heart sounds (Phono) in relation to a surface electrocardiogram (ECG) and an arterial blood pressure tracing (ABP) recorded from the transverse facial artery. S1, First (systolic) heart sound; S2, second (diastolic) heart sound; S3, third heart sound; S4, fourth (atrial) heart sound. B-lub dup-uh describes the sounds heard on auscultation. Note the timing of the heart sounds and the (peripheral) pulse pressure wave relative to the ECG. (From Schwarzwald CC, Bonagura JD, Muir WW: The cardiovascular system. In Muir WW, Hubbell JAE, editors: Equine anesthesia: monitoring and emergency therapy, ed 2, St Louis, 2009, WB Saunders).

HEART SOUNDS Heart sounds should be readily heard on each side of the thorax, although some variability is found based on body type, and sounds are louder over the left thoracic wall. All four heart sounds can be detected in healthy horses, but all may not be present or evident at the same location (see Figures 10-9 and 10-10).3,23,25 The ventricular lling sound (S3) is most localized and variable, and may be di cult to detect unless the bell is placed lightly over the left apex. The intensity of the heart sounds should be consistent when the rhythm is regular, but variation of heart sound intensity occurs with arrhythmias (irregular cardiac lling). Mu ed heart sounds are heard with pericardial e usions or pericardial abscesses (the mu ing may only be on one side of the thorax) but also occur in some horses with large pleural e usions and cranial mediastinal masses. Accentuation of all heart sounds, especially the third sound, may be detected with volume-loaded ventricles or with marked sympathetic activity. Projection of heart sounds over a wider area is sometimes evident in cases of pleuropneumonia with pulmonary consolidation.

FIGURE 10-10

Phonocardiograms (PCG). A, The four normal heart sounds (S1 through S4) are evident in this recording. (The numbers 1 through 4 indicate the various components of the rst heart sound; ECG, electrocardiogram.) B, Systolic murmur recorded from a yearling with a ventricular septal defect (VSD) (P waves on the ECG are negative in this tracing). C, Decrescendo, musical, holodiastolic murmur recorded from a horse with aortic regurgitation (AR). The murmur begins in early diastole (after the T wave) and ends at the QRS complex. (A recording courtesy of D. Smetzer, R.L. Hamlin, and C.R. Smith.)

The rst heart sound varies with arrhythmias and often becomes louder (or sometimes softer) after prolonged diastolic periods. This in

itself is not diagnostic of an abnormality. Splitting of the rst heart sound, if pronounced, may indicate abnormal ventricular electrical activation or ventricular premature complexes.238 Close splitting of the rst sound may be more obvious when AF is present, and in the absence of an atrial sound, the split S1 may be misinterpreted as a closely timed S4 to S1 complex. The second sound is loudest normally over the aortic valve area and may be audibly split over the pulmonic valve area.3,4,238 This sound can be very soft or absent after a premature beat, and it may be obscured by a holosystolic murmur. Audible splitting of S2 occurs commonly in normal horses, varies with heart rate or respiration, and only infrequently is associated with pulmonary hypertension. The relative closure of the semilunar valves probably varies with the heart rate and PA pressure,3,238 although the pulmonic component is most often detected after the aortic component in the healthy, resting horse. If the pulmonary component of the second sound develops a tympanic quality, becoming equal to or louder than the aortic component of S2, then the clinician should consider pulmonary hypertension. Identi cation of a loud pulmonic S2 is a useful clinical nding in cases of right-sided heart failure because it often indicates that a lesion on the left side of the heart has led to pulmonary hypertension and right-sided CHF. Diastolic transient sounds are normal.16,17,19,350 The atrial sound (S4) should be detected in virtually all horses, and this sound may be quite loud in some cases. Isolated atrial sounds are commonly ausculted at slow resting heart rates caused by second-degree AV block; however, multiple isolated S4 sounds indicate high-grade AV block. The ventricular sound (S3) is normal and as a low-pitched sound may be di cult to hear. The sound often increases transiently in the setting of sinus tachycardia and enhanced rapid lling. This sound also achieves greater intensity when ventricular dilatation and elevated filling pressures develop, as with heart failure. Other sounds may be detected with heart disease. Systolic clicks are uncommonly heard over the great vessels where they are thought to be benign. A systolic click loudest over the mitral area may be a marker for mitral valve disease or prolapse of the valve into the LA. Constrictive pericardial disease can create a loud early diastolic ventricular lling sound, a ventricular “knock,” indicating abrupt termination of rapid lling under high venous pressures. In addition to mu ing of sounds, pericarditis is often associated with a pericardial friction rub. These rubs are classically detected during three portions of the cardiac cycle (systole, early diastole, late diastole). These sounds must be distinguished from pleural friction rubs or respiratory mediated sounds, which are occasionally associated with the cardiac cycle. Pleural and pulmonary rubs tend to be heard only during one or two phases of the cardiac cycle and may not correlate closely.

HEART RATE AND RHYTHM Heart rate can change rapidly and dramatically, varying with autonomic e erent tra c and level of physical activity. Changes in the rate and rhythm are re ected in cardiac auscultation and palpation of the pulse (see Table 10-1).344 The normal arterial pulse and cardiac rhythm is regular or cyclically irregular. Heart rate in mature horses varies between 28 and 44 beats/min, although slightly slower heart rates can be detected in some very t racehorses and higher heart rates are often present in foals,348 yearlings, draft horses,358 and healthy but nervous horses. An elevated resting heart rate is commonly detected with late pregnancy, fever, pain, hypovolemia, severe anemia, infection, or shock. Persistent, otherwise unexplained tachycardia is also typical of heart failure wherein the standing heart rate usually exceeds 60 beats/min. Physiologic arrhythmias associated with high resting vagal tone must be di erentiated from pathologic arrhythmias. Vagal-induced arrhythmias are most often observed in healthy

FIGURE 10-7

Cranial view of the heart demonstrating the U shape of the right ventricle (RV). The tricuspid valve and the right ventricular inlet are located on the right side, whereas the outlet and pulmonary artery (PA) are on the left side. Murmurs of tricuspid regurgitation and perimembranous ventricular septal defects are typically loudest over the right hemithorax, whereas flow murmurs, and those caused by subaortic or subpulmonic septal defects are loudest over the left cardiac base. RA, Right atrium.

(From Bonagura JD, Muir WW: The cardiovascular system. In Muir WW, Hubbell JAE, editors: Equine anesthesia, St. Louis, 1991, Mosby-Year Book.)

relaxed horses. The most common of these are sinus arrhythmia and second-degree AV block (see Figure 10-6), which is reported in 15% to 18% of normal horses at rest and has been detected in up to 44% of normal horses with 24-hour continuous electrocardiographic monitoring.194,197,199,359,360 This arrhythmia is most common in t racehorses or in other high-performance animals and disappears with stimulation of sympathetic nervous system activity. Sinus arrhythmia also occurs regularly in horses and is often associated with sinus bradycardia and resting heart rates between 24 to 28 beats/min. Sinus arrhythmia may wax and wane with respiration; however, synchronization with ventilation is not a consistent nding in the horse, and it is more likely related to changes in baroreceptor activity. SA block and SA arrest occur sporadically in fit horses. Because these physiologic arrhythmias are associated with high vagal tone, it is typical for the heart rate to range from a low-normal value to overt bradycardia. Maneuvers that reduce vagal activity and increase sympathetic tone can be undertaken to ensure that the rhythm becomes normal. Successful methods include leading the horse through three or four tight circles, jogging, or lunging. It should be recognized, however, that some horses redevelop physiologic AV block within 10 to 60 seconds after such maneuvers. If the arrhythmia persists after exercise, or if the auscultatory ndings suggest another arrhythmia (see Table 10-1), then an ECG should be obtained. The arrhythmias associated with heart disease occur most often with normal to increased heart rates, with the exception of advanced second-degree and complete (third-degree) AV block.∗ Atrial premature beats are characterized by a regular cardiac rhythm, which is suddenly interrupted by earlier than normal beats. Because the sinus node is reset during an atrial premature beat, the following pause is usually incomplete or less than compensatory (provided the underlying sinus rhythm is regular). Premature atrial beats can also be blocked in the AV node, leading to a sudden pause in the rhythm without a premature rst heart sound. A ventricular premature beat is often followed by a fully compensatory pause unless the extrasystole is interpolated between two normal sinus beats. Pulse de cits (more rst sounds than arterial pulses) are common with premature beats. Atrial brillation is characterized by an irregularly irregular pulse and ventricular rhythm, with some beats occurring sooner than expected, and with pauses in the rhythm demonstrating no consistent diastolic interval. The atrial contraction and hence the fourth sound are absent in AF. Infrequently, AF develops with recurring periodicity of cycle lengths that can cause di culty in distinguishing this from second-degree AV block.191,193,201 Another diagnostic pitfall can be the presence of a split rst sound, because the clinician may mistake this for a closely timed S4 to S1 complex. Conversely, atrial tachycardia can be conducted with patterns leading to a rapid but regular heart rhythm. However, most horses with atrial tachycardia develop irregular rhythms owing to frequent block of atrial impulses in the AV node, especially at rest. Heart rates are usually higher than normal at rest. Auscultation of the horse in a quiet area may reveal isolated audible fourth heart sounds associated with the second-degree AV block. Junctional or nodal tachycardia and VT are usually manifested by a rapid regular rhythm, although some VTs cause an irregular rhythm. Owing to associated AV dissociation or abnormal ventricular activation, the heart sounds may sound split or variable in intensity.

CARDIAC MURMURS Cardiac murmurs are prolonged audible vibrations developing in a usually quiet portion of the cardiac cycle. In general, murmurs are manifestations of either normal (functional) or abnormal (pathologic) blood ow in the heart and blood vessels. Although many heart murmurs are functional (physiologic, athletic, innocent), other murmurs provide evidence of heart disease and may require further investigation. Most functional cardiac murmurs are caused by vibrations that attend the ejection of blood from the heart during systole or the rapid filling of the ventricles during early diastole.∗ Causes of abnormal murmurs include incompetent cardiac valves, septal defects, vascular lesions, and (very rarely) valvular stenosis (see Table 10-2 and 10-3). The continuous murmur of PDA is a normal nding in full-term foals for up to 3 days after parturition, but it occasionally persists for almost 1 week after foaling.265,343,348,367 One of the challenges of physical diagnosis is determination of the cause and clinical signi cance of a cardiac murmur. With knowledge and experience, the clinician can accurately determine the likelihood that a murmur is physiologic or pathologic in origin and e ciently select animals requiring additional studies. The clinician should describe the timing, duration, quality or pitch, grade, point of maximal murmur intensity, and radiation of a murmur, as well as the e ect of changing heart rate on the sounds. Determination of timing is made relative to the heart sounds so that murmurs are designated as systolic, diastolic, or continuous. When the overall timing is not obvious, the clinician can either palpate the apical impulse, which occurs in early systole, or the systolic pulse in the brachial or median artery to identify the timing of the murmur. Skilled auscultators subdivide the timing of murmurs into proto-(early), meso-(middle), or tele-(late) systole or diastole, because the timing and duration of the murmur often correlate with speci c ow disturbances. Shorter murmurs, especially those occurring in early systole or protodiastole, are more likely to be functional, although this is not always the case. Experience also teaches that relatively loud or long murmurs are more likely to be associated with cardiac pathology (see Figure 10-10 and Table 10-3), but again exceptions occur. The pitch or quality of the murmur provides additional insight. For example, mixed frequency (harsh or blowing) murmurs are typical of cardiac disease. A brief vibratory or musical murmur is typically functional; however, a prolonged musical murmur is

suggestive of a valvular lesion. Applying the characteristics of murmurs to the clinical setting requires study, an organized scheme (see Table 10-3), and practice. Functional, innocent, or physiologic heart murmurs are very common and are not attributed to cardiac pathology. Such murmurs may be related to the large size and tremendous in ow and out ow volumes of the equine heart.364 Physiologic causes of functional murmurs include fever, high sympathetic activity (e.g., colic, exercise, pain), moderate to severe anemia, and peripheral vasodilation. These murmurs are typically soft (grades 1 to 3/6, of a possible 6), localized, relatively short, crescendo-decrescendo in timing, and labile. Increasing heart rate usually increases the intensity and duration of a functional murmur, although in some horses the murmur becomes less intense. Functional murmurs are neither holosystolic nor holodiastolic; therefore the heart sounds should still be evident. The most common functional murmur is the systolic ejection murmur heard over the aortic and pulmonic valves and projecting into their respective arteries at the left cardiac base (see Figure 10-8).3 The functional ejection murmur is generated by ow into the great vessels, and by de nition it must start after S1 and end before S2. Nonetheless, in some horses, the functional ejection murmur can achieve substantial intensity (rarely, a grade 4 or 5) and may seem holosystolic at higher heart rates. At times a remarkable change in functional murmur intensity is seen from day to day. An example of this is the varying murmurs identi ed in some horses with emergent colic. Because a loud functional murmur can radiate caudally, it may be confused with MR. In terms of diagnostic considerations, the clinician should not mistake a functional ejection murmur for that of aortic or pulmonic stenosis. Pulmonic stenosis is rare in the horse, and aortic stenosis even less common. Accordingly, these valvular malformations are rarely entertained in the di erential diagnosis. The murmurs resulting from subpulmonic (subarterial) VSDs or the stenosis associated with tetralogy of Fallot may be loudest over the great vessels, but such murmurs are usually loud and prompt echocardiographic investigation. Functional diastolic murmurs are also common, especially in young horses and in the Thoroughbred breed. These murmurs are generally soft and are detected over the LV or RV inlets—from the dorsal atrium to the ventricular apex. The functional protodiastolic murmur is an early diastolic murmur detected between S2 and S3. The murmur may be musical, vibratory, or “squeaky” and typically accentuates with increased heart rate.3 It most reasonably represents the rapid early lling of the ventricles. The presystolic functional murmur is heard between S4 and S1, and the vibrations may sound like a rub or long heart sound after the atrial contraction. It can be confused with an early systolic murmur or a friction rub. The most commonly detected murmurs associated with structural or organic heart disease are those generated by tricuspid regurgitation (TR), MR, AR, and VSD. Typical causes and auscultatory features of these murmurs are indicated in Tables 10-2 and 10-3 and described in detail later in this chapter. When a murmur is believed to represent underlying cardiac pathology, further investigation is required. The overall signi cance of the hemodynamic abnormality should include consideration of the work history, general physical examination, presence or absence of heart enlargement or cardiac failure, and results of ancillary tests such as echocardiography, electrocardiography, and exercise performance (or testing). Certainly, the history should be considered, because the horse with excellent performance and good exercise tolerance is less likely to have serious heart disease. Auscultation, however, is insu cient to distinguish trivial from signi cant heart disease in the poorly performing horse or in a horse with a cardiac arrhythmia. Echocardiography is helpful in these cases to quantify heart size and objectively assess ventricular function. Underlying lesions such as a valve vegetation, ruptured valve chorda tendineae, or dilated cardiomyopathy can be identi ed or discounted. Although there are no well-de ned echocardiographic or Doppler correlates to functional heart murmurs, these studies can document abnormal blood flow and can pinpoint the cause of a pathologic murmur.

PULMONARY AUSCULTATION Auscultation of the lungs should reveal normal breath sounds with the horse at rest, while ventilating into a rebreathing bag, and after exercise. Decreased or absent airway sounds or large airway sounds in the ventral portions of the thorax indicate a pleural e usion, a common nding in biventricular or right-sided heart failure. Moist or bubbling ( uid) sounds or crackles (i.e., rales) are uncommonly auscultated in the lungs of horses with pulmonary edema and left-sided heart failure. Instead, tachypnea associated with harsh bronchovesicular breath sounds is usually heard, because horses with chronic left-sided CHF seem to develop more interstitial than alveolar pulmonary edema. When alveolar edema does develop, respiratory distress may be severe, crackles may become evident, and free uid may be auscultated in the trachea. On rare occasion, primarily with peracute left-sided heart failure, froth will be visible at the nares and the horse will cough and expel large quantities of pulmonary edema (Figure 10-11). Such horses demonstrate severe respiratory distress (marked tachypnea and dyspnea), anxiety, and agitation.

FIGURE 10-11

Peracute pulmonary edema in a horse with a ruptured chorda tendinea. The photo was taken immediately after euthanasia. An important note is that tracheal froth is often a postmortem artifact, particularly when observed hours after death.

Examination of the Peripheral Vasculature An evaluation of the peripheral vasculature is part of a CV workup and should include examination of the arteries and veins in the head, forelimbs, and hindlimbs. When important, ABP also should be measured. The genesis of the arterial and venous pulses has been described previously. The heart sounds should be correlated with both the jugular and the arterial pulses. Normal-quality arterial pulses should be palpable in the facial, median, carotid, great metatarsal, coccygeal, and digital arteries. The arterial pulse represents the pulse pressure (i.e., the di erence between peak systolic and diastolic pressures), the rate of rise of arterial pressure, and the physical characteristics of the artery and surrounding tissues. Palpating the facial artery pulse may identify heart rate and rhythm, as well as altered hemodynamic states. The arterial pulse can be described as normal, hypokinetic (weak), hyperkinetic, or variable. Irregularity often indicates a cardiac arrhythmia. Thready or hypokinetic arterial pulses are associated with reduced stroke volume and peripheral vasoconstriction and therefore may be detected in CHF and in diseases associated with volume depletion or profuse hemorrhage. Thready arterial pulses may also be detected only in the hindlimbs with aortoiliac thrombosis. Bounding, hyperkinetic arterial pulses are palpable with clinically signi cant AR, PDA, and aortic to pulmonary or aortic to right-sided heart stulas. The pulse in septic animals may be weak, owing to reduced pulse pressure, or normal to rapidly declining if peripheral vasodilation occurs. Marked variation in the intensity of the pulse usually occurs with arrhythmias, particularly AF and rapid or multiform VT. A pulse de cit ( rst heart sound without palpable pulse) occurs when developed LV pressure does not exceed aortic pressure. De cits are likely to be palpated in association with pathologic arrhythmias, particularly premature beats, or after very short diastolic periods of tachyarrhythmias. Palpation of the arterial pulse cannot quantify the actual ABP, which must be measured directly by arterial puncture or cannulation or indirectly using various auscultatory, Doppler, or oscillometric techniques.71,253,269,368-393 Percutaneous placement of an arterial catheter in the facial or transverse tibial artery is frequently used to monitor pressure invasively in critically ill or anesthetized horses. Indirect methods have been used successfully to monitor pressure in the coccygeal artery; however, these methods are less sensitive in hypotensive animals and may lag in response during rapid changes in BP.371,373,375 Attention must be directed to placement and diameter of the occluding cu when indirect methods are used,385 and the optimal width of the cu (bladder) is between approximately one quarter to one fth of the circumference of the tail when measuring pressure in the middle coccygeal artery.379,384 The arterial pulse wave itself varies, depending on the site of measurement, and the distal arterial systolic pressure may be higher, and the diastolic pressure lower, than the corresponding central aortic pressures. ABP monitoring includes determination of systolic, diastolic, and mean pressures. Arterial pressure is determined by the interplay between stroke volume, heart rate, and vascular resistance (see Box 10-3 and Figure 10-4, B). Marked increases in CO can lead to signi cant increases in ABP with systolic pressures exceeding 200 mm Hg.246,390,394 Normal reported values for indirect arterial pressure are 111.8 mm Hg systolic (ranging between 79 to 145 mm Hg) and 67.7 diastolic (ranging between 49 to 106 mm Hg).∗ BP is lowest in neonates and rises during the rst month of life to the normal adult range.253,373 Some variation is found among breeds. Draft breeds tend to have lower pressures than racehorses, and Standardbreds have lower pressures than Thoroughbreds.389 Arterial pressures uctuate slightly with ventilation and signi cantly under positive pressure ventilation or during cyclic changes in heart rate (see Figure 10-6). Posture imposes a signi cant in uence on arterial pressure, because raising the head from the feeding position necessitates a higher aortic pressure to maintain cerebral perfusion. Obviously, when the head is lowered, the hydrostatic pressure imposed by an elevated head position is minimized. Mean arterial pressure measured in the middle coccygeal artery can vary approximately 20 mm Hg with the changing head position.386 Abnormalities of arterial pressure and perfusion can result in changes in mucous membrane color and capillary re ll time. Re ll time is prolonged during hypotension or peripheral vasoconstriction; conversely, it may be shortened with vasodilation exists.

The systolic arterial pressure is generated by the LV and is consequently a ected by the interplay between the stroke volume, heart rate, aortic compliance, and previous diastolic BP.46 Arterial pulse pressure is highly dependent on stroke volume and the peripheral arteriolar resistance that determines the run o of diastolic pressure. Ventricular failure or reduced venous return reduces pulse pressure, creating a hypokinetic pulse; whereas abnormal diastolic run o (AR, generalized vasodilation) widens the pulse pressure, producing a stronger peripheral pulse. Diastolic and mean pressures are better estimates of perfusion pressure in anesthetized or critical equine patients; these variables are increased by arteriolar vasoconstriction or by higher CO. Mild variation in pulse pressure is normal across the respiratory cycle related to variation in ventricular lling. However, marked di erences can be associated with cardiac tamponade, and the dramatic fall in pressure during inspiration is termed pulsus paradoxus. The cutaneous veins should be examined for distensibility, re ll, thrombosis, and estimated venous pressure. The jugular veins in normal horses are collapsed, but the veins of the limbs and torso are visible and somewhat filled. Venous pressure in the jugular vein is normally less than 10 cm of water above the phlebostatic point, and jugular pulsations are normally con ned to the thoracic inlet and the ventral one third (10 cm) of the neck. These pulsations are re ections of the RA pressure changes that range from positive to subatmospheric over the cardiac cycle (discussed previously). Pulsations are normally visible along the entire length of the jugular vein when the head is lowered below the level of the heart and should not be misinterpreted. A pathologic jugular pulse also can be misdiagnosed if transmission of carotid arterial pressure into the jugular furrow occurs. Occluding the jugular vein ventrally can demonstrate if the pulses are actually originating within the RA or not, because venous pulse transmission, or prominent collapse, will be prevented by light digital pressure over the vein. Pronounced jugular pulsations are observed occasionally in excited but otherwise normal horses with high sympathetic tone. This might be caused by either vigorous RA contraction. Alternatively, an exaggerated venous collapse can create a “jugular pulse” that is simply a re ection of normal venous re lling during periods of positive RA pressure. When doubt persists, this nding can be veri ed by simple ultrasound imaging of the ventral portion of the vein. Jugular abnormalities include abnormal pulses or elevated venous pressure. An abnormal jugular pulse is one that extends proximally up the jugular vein for more than 10 cm in systole (with the horse’s head held in a normal position) or that demonstrates retrograde lling from the heart when the vein is occluded dorsally. Duplex Doppler studies of the jugular vein can distinguish a prominent pulse caused by retrograde ow from an apparent pulse caused by prominent collapse and normal lling. Abnormal jugular pulses are observed with arrhythmias causing AV dissociation (nonconducted atrial premature beats, junctional tachycardia, VT); with diseases of the tricuspid valve (moderate to severe TRor the rare case of tricuspid stenosis); and with RV failure from any cause including cardiac tamponade or constrictive pericardial disease. Elevated jugular venous pressure generally indicates right-sided CHF, pericardial disease, or hypervolemia. Generalized venous distention, particularly when accompanied by subcutaneous edema, is characteristic of CHF (Figures 10-12 and 10-13). Isolated distention of the veins cranial to the thoracic inlet is suggestive of a cranial mediastinal or pulmonary mass (or abscess) with obstruction of the cranial vena cava.100 Prolonged re ll of the saphenous vein indicates the possibility of aortoiliac thrombosis and decreased arterial supply to the a ected limb.222-227,396,397 A distended vein that is rm to palpation, and associated with marked distention of associated veins, indicates probable venous thrombus.398 Pain or heat associated with venous swelling is suggestive of thrombophlebitis.

FIGURE 10-12

Jugular venous distention in a Shire foal with biventricular congestive heart failure (CHF) caused by congenital heart disease.

FIGURE 10-13

Ascites, ventral edema, and weight loss are evident in this mare with right-sided congestive heart failure (CHF). The distended lateral thoracic vein (arrow) is evident caudal to the triceps.

Congestive Heart Failure Congestive heart failure in horses is relatively uncommon; however, the diagnosis often can be made from a thorough physical examination.399 Clinical ndings include persistent tachycardia (generally at 60/min or greater), jugular venous distention and pulsations, abnormal arterial pulses, uid retention (ventral subcutaneous edema, pleural e usion, or pulmonary edema), and loss of body condition (see Figure 10-13). LV failure secondary to valvular or myocardial heart disease causes pulmonary venous congestion with pulmonary interstitial or, uncommonly, alveolar edema. Tachypnea or coughing associated with left-sided CHF can be misdiagnosed as pneumonia or chronic obstructive pulmonary disease. Exercise intolerance, lethargy, collapse (or syncope), anorexia, depression, and weight loss are other potential signs of heart failure. This subject and management are discussed more fully later in this chapter.

ODIAGNOSTIC AND LABORATORY STUDIES Exercise Testing Because most horses function as athletes, exercise testing represents an important method for identifying CV and other disorders in the underperforming horse. Additionally, some form of exercise testing should be included in the prepurchase CV examination, unless the animal is too young to perform or is to be used solely for breeding. Even in these situations, evaluation of a foal, weanling, or yearling after a period of free exercise is optimal. Furthermore, it should be determined that a stallion has su cient stamina to perform in the breeding shed. For these reasons, it is recommended that CV system be evaluated at rest, immediately after exercise, and under the anticipated form of work required by the prospective owner. Having advanced this recommendation, it is also acknowledged that more data and information are needed to establish the reliability and predictive values of such studies relative to CV disease. High-speed treadmill and other forms of exercise testing are increasingly used to detect subtle clinical abnormalities that limit peak performance.∗ The exercise examination provides valuable information about musculoskeletal diseases, airway diseases, and CV disorders. Treadmill exercise is also useful when evaluating the horse with a history of collapse or syncope, because the cardiac response to exercise can be evaluated without risk to a rider. A number of components to a standardized treadmill examination exist. Gait and lameness are evaluated and pre- and postexercise serum creatine kinase (CK) measured to identify subclinical myopathy. Videoendoscopy is used to identify laryngeal and other dynamic airway disorders. Arterial blood gases are often analyzed to assess pulmonary function. An exercise and postexercise ECG is recorded to identify heart rhythm disturbances. Pre- and postexercise echocardiography may be used to identify regional or global LV dysfunction. Doppler studies can be added to evaluate valvular insufficiency.184 Changes in heart rate with exercise have been well studied.205,326-330,403,426 The normal horse develops sinus tachycardia, shortening of conduction intervals, and marked ST-T alterations associated with exercise. The maximal heart rate achieved depends on the level of exercise performed. At heart rates of less than 100 to 120 beats/min, SA node discharge is very labile and subject to psychologic in uences. Once exercise has commenced, the heart rate should accelerate and then stabilize at a level appropriate for the work being performed. As a general guideline, heart rates of 70 to 120 beats/min are normal at the trot, 120 to 150 beats/min are normal at the canter, 150 to 180 beats/min are normal at the gallop, and more than 180 beats/min are normal when galloping at maximal exertion. The maximal heart rate for most horses is between 210 and 240 beats/min, although some develop higher rates. A linear relationship relates the heart rate and the velocity or work e ort of exercise when the heart rate is between 120 and 210 beats/min. Heart rate usually recovers rapidly and falls to less than 100 beats/min within 2 to 5 minutes of cessation of exercise. The recovery of heart rate to the resting level is in uenced by many factors, including the humidity, ambient temperature, training tness, work performed, psychologic factors, and state of CV health. After maximal work, such as a race, 1 hour may pass before the heart rate completely recovers to the resting rate. An inappropriately high heart

rate for a given level of exercise may simply denote a healthy, but un t, horse; however, this nding can also indicate the presence of CV, pulmonary, or musculoskeletal disease. Information about the heart rate is particularly useful when a horse has the heart rate monitored routinely during exercise. Exercise-induced arrhythmias may be observed during or after exercise. An exercising ECG obtained by radiotelemetry is necessary to determine if an arrhythmia is present during exercise. A heart rate monitor may provide some indication regarding arrhythmias during exercise; however, the detection of erratic heart “blips” or “beeps” does not adequately characterize the rhythm disturbance. Exercise may induce supraventricular premature complexes, AF, ventricular extrasystoles, or VT. Exercise-induced (paroxysmal) AF should be included in the di erential diagnosis of horses with poor performance during high-intensity exercise.39,427,428 Exercise-induced ventricular arrhythmias are a particular concern because ventricular rhythm disturbances can lead to poor performance, sudden stopping, or even falling, and they have been suspected as one cause of sudden death.60,62-65 Repetitive ventricular activity (paroxysmal VT) is particularly worrisome and may indicate preexisting myocardial disease or myocardial ischemia with altered cardiac electrical activity. In some horses, a maximal work e ort is required to elicit heart rhythm disturbances. Thus when the history indicates performance problems only at the peak of exercise, or if other indicators of heart disease are present, a treadmill exercise test to exhaustion may be needed to reveal cardiac arrhythmias. Horses demonstrate marked uctuations in sympathetic and parasympathetic tone during the recovery period after exercise.420,429,430 Arrhythmias are especially common during this time,317,337 even if the resting and peak exercise heart rhythms were normal. Marked sinus arrhythmia, second-degree AV block, supraventricular ectopia, and ventricular premature complexes can be evident in the immediate postexercise period. With the exception of vagally mediated arrhythmias, postexercise arrhythmias should be considered suspicious. AF, ectopic supraventricular tachycardias, and VTs are considered abnormal at any point in the exercise test. However, the importance of single (isolated) ectopic complexes identified only in the immediate postexercise phase is less certain,337 which is not surprising considering similar issues persist in assessing human athletes. In assessing these horses, attention also should be directed to the presence of coexisting respiratory disease, baseline auscultation and echocardiographic examinations, the heart rhythm during exercise, measurements of serum cardiac troponin I (cTnI) before and after exercise, and the so-called stress echo immediately after work. Echocardiography after the stress of exercise can be used in an attempt to identify global and regional myocardial dysfunction.122,317,331,337 This examination must occur within 2 to 3 minutes after high-intensity exercise and with the heart rate in excess of 100 beats/min. However, the optimal heart rates and the in uence of rapidly decreasing heart rates on echocardiographic measurements such as shortening fraction require better de nition. Dynamic changes can be viewed from the short axis tomograms, and prominent thickening of the ventricular septum and LV free wall identi ed from long axis images. Persistent abnormalities in regional myocardial contraction are suggestive of vascular disease with ischemia, preexistent myocardial brosis, or a localized conduction disturbance. LV shortening fraction and LV fractional area change (two estimates of LV systolic function) are commonly measured during stress echocardiography. It is important to scrutinize multiple segments across the entire shortening area of the LV, because errors in cursor placement (e.g., too dorsal) or inconsistent orientation of the M-mode cursor placement between examinations can produce misleading results. Other methods used for stress echocardiographic assessment of LV function are the manual tracking of endocardial motion and subjective evaluation of regional wall motion. Newer echocardiographic modalities, including tissue Doppler imaging and 2D strain (2D speckle tracking) imaging, may have some advantages over conventional methods and could prove useful for quantitative assessment of regional and global LV function during stress echocardiography (CC Schwarzwald, unpublished observations), but more data and standardizations are needed before methods can be widely accepted. Undoubtedly, CV disorders can be induced or worsened by the stress of exercise, and the likelihood of detecting disease is increased. However, the assessment is chellenging because physiologic variability is also magni ed. The immediate postexercise period involves rapidly changing autonomic tra c420,429,430 that can result in confusing ECG and echocardiographic ndings.184,337 The presence of signi cant respiratory disease and associated arterial hypoxemia also creates the potential for secondary rhythm disturbances, unrelated to primary cardiac disease.317 Thus the reliability and overall sensitivity, speci city, and predictive values of the exercise test (as it pertains to detection of arrhythmia and organic heart disease) are incompletely de ned for a number of reasons. The “standardized” exercise test may not be identical across equine medical centers, so it may be di cult to compare results. The examiner’s own reference for normal variation further quali es the examination results. For example, the presence of isolated postexercise premature complexes may de ne the examination as abnormal for some317 or suspicious or uncertain for others. Similarly, if the LV shortening fraction does not increase, but the LV fractional shortening area does, then one must ensure that the issue is not simply technical (i.e., cursor placement) as opposed to pathologic. Finally, the assessment of cardiac valve function after exercise or pharmacologic stress requires far better de nition before any conclusions can be made. Although the standardized treadmill examination is the most commonly used cardiac “stress test” at equine centers, some attention has been directed on performing pharmacologic stress testing. This examination involves increasing the heart rate and myocardial contractile force with dobutamine, either with or without atropine or glycopyrrolate, followed by echocardiographic assessment of ventricular contraction.332,413,418,423,431 Pharmacologic stress echocardiography is used to identify regions of reduced myocardial perfusion or ischemic myocardial injury in human patients with coronary heart disease; the study is used in people who cannot perform physical exercise to stress

the heart and increase myocardial oxygen demand. Whether this form of stress testing should be used in horses is subject to debate. First, it must be emphasized that no compelling evidence from pathology studies indicates that horses develop signi cant coronary arterial disease. Admittedly, myocardial brosis is found in some horses at necropsy29,432; however, the pathogenesis or clinical relevance of these often-microscopic lesions is unresolved. In human patients, the diagnostic target is well-de ned extramural coronary artery disease, and stress echocardiography generally predicts normal or abnormal ndings at coronary arteriography. No evidence has been published at this time for a similar situation in horses, and this represents a major contextual issue when considering “stress echo” with either exercise- or pharmacologic-induced cardiac work. Furthermore, pharmacologic stress testing may not achieve the identical exercise endpoints in terms of heart rate or altered vascular resistances. The combination of dobutamine and atropine involves the use of proarrhythmic drugs,413 making the heart rhythm assessment (a pivotal component of the exercise test) quite problematic. Pharmacologic stress testing does not provide an opportunity for dynamic assessment of the gait and airways.417 Lastly, assessment of left heart valve function is complicated by changes in systemic ABP that may develop during drug-induced tachycardia (but have been poorly characterized in studies thus far.) Thus although pharmacologic stress testing carries the bene t of not requiring a treadmill, a number of outstanding issues must be addressed before this examination can be advocated for wider use.

Electrocardiography NORMAL CARDIAC CELL ELECTRICAL ACTIVITY Cardiomyocytes are excitable and capable of responding to electrical stimuli regardless of extrinsic innervation. Specialized cardiac tissues such as those in the SA node and the His-Purkinje system demonstrate spontaneous depolarization and can serve as pacemakers for the heart. The electrical processes responsible for the ECG are caused by ion uxes across the cell membrane (Figure 10-14)73,76,77 A general understanding of cellular activity, generation, and spread of the cardiac electrical impulse; e ects of autonomic innervation; and electrocardiographic lead systems is required to interpret the equine ECG.

FIGURE 10-14

The cardiac action potential. The diagram demonstrates the phases of depolarization and repolarization, as well as membrane conductance (g) to important ions (see the text for details). (From Berne RM, Levy MN: Cardiovascular physiology, St Louis, 1986, CV Mosby.)

The partially selective nature of the cardiac cell membrane and the presence of various cell membrane pumps leads to an unequal partitioning of ions across the cell membrane.73 This results in very high potassium and relatively low sodium concentrations intracellularly when compared with the extracellular uid. Other ions including chloride and magnesium (but most notably calcium) are important to cellular electrical activity. Calcium also is essential for contraction of the cardiomyocytes. Sudden changes in cardiac cell membrane permeability or conductance to sodium, calcium, potassium, and chloride are responsible for the processes of depolarization, muscular contraction, and repolarization (see Figure 10-14). Serum electrolyte concentration, acid-base status, autonomic tra c, myocardial perfusion and oxygenation, heart disease, and drugs, in turn, a ect these processes. The basic processes of cell depolarization, calcium in ux, and cell repolarization form the basis for, respectively, the P and QRS complexes of the ECG, myocardial contraction, and the ST-T wave of the ECG. The resting membrane potential is determined primarily by the partitioning of potassium ions and proteins across the cell membrane and the relative impermeability of the membrane to sodium. Depolarization of atrial and ventricular myocytes and Purkinje bers is caused by the rapid in ux of extracellular sodium into the cell.76,77 This current is represented by phase 0 of the cardiac cell action potential recording. Normal cardiac tissues depolarized in this manner conduct electrical impulses at high velocity. In contrast, the cells of the SA and AV nodes, as well as ischemic cells, demonstrate less negative diastolic membrane potentials. Nodal cells depend on a slow-inward current of depolarization that is carried mainly by calcium ions across the transient and long-lasting calcium channels. In ischemic cells, because of the less negative resting potential, some sodium channels remain in the inactivated state and thus are unavailable for activation and membrane depolarization during phase 0. Therefore cells in the SA node, AV node, and ischemic tissue conduct electrical impulses very slowly and can be involved in blocks, abnormal automatic mechanisms, or re-entrant pathways that promote cardiac arrhythmias. Antiarrhythmic drugs and calcium channel blockers act di erently on tissues that are “fast” versus “slow” current dependent (see Cardiac Arrhythmias later in this

chapter). The in ux of extracellular calcium into the cell during the plateau phase (phase 2) of the action potential triggers the release of intracellular calcium from the sarcoplasmic reticulum. Increases in cytosolic calcium cause myocardial contraction. Hypoxia, acidosis, anesthetics, and many other drugs can a ect myocardial contractility by interfering with calcium entry into cells, with the release of calcium from the sarcoplasmic reticulum, or with the binding of calcium to contractile laments.73,76,104 Conversely, digitalis glycosides and dobutamine increase calcium influx and myocardial contractility. Cellular repolarization is initiated by decreases in the cell membrane conductance to sodium and to calcium coincident with increased conductance to potassium. As intracellular potassium moves out of the cell, along its concentration gradient, phase 3 of the cardiac action potential occurs. Repolarization is complex, involves a number of ion currents,433 and is a ected by genetics and by drugs such as quinidine or amiodarone. Sympathetic stimulation may activate currents that facilitate repolarization at higher heart rates, explaining the shortening of the action potential under sympathetic drive. Paradoxically, vagotonia also enhances repolarization by opening potassium channels, an e ect that is very prominent in the atria and may predispose to development of AF. Hypoxia and abnormalities in serum potassium or calcium also alter repolarization. Hyperkalemia accelerates repolarization and shortens the ST-T wave of the ECG. This occurs because hyperkalemia increases membrane permeability to potassium and the high intracellular potassium (exceeding 100 mmol/L) is more than adequate to drive potassium out of the cell (because extracellular potassium rarely exceeds 10 mmol/L). Spontaneous depolarization or automaticity is a property of select cardiac tissues. Such activity is most prominent in the SA node but may also be encountered in cells around the AV node and in the His-Purkinje network. Normal automaticity is observed in these cells during phase 4 of the cardiac action potential. Spontaneous pacemaker activity is generated in normal pacemaker tissues by the background inward sodium current and a time-related decrease in membrane permeability to potassium ion e ux, as well as a transient inward calcium current.76 Once membrane threshold is reached, ion ow across the long-lasting calcium channels (“slow current”) predominates and leads to cell depolarization (phase 0). Spontaneous depolarization is modi ed markedly by autonomic tra c. Vagal activity opens potassium channels and hyperpolarizes the membrane, depressing automaticity. With sympathetic stimulation or membrane hyperpolarization, the depolarizing funny current (IF), becomes activated, enhancing pacemaker activity. Autonomic innervation has profound e ects on electrical activity9,189,434 but is not distributed equally through the heart. The right and left vagus have preferential innervation to SA and AV nodes, and parasympathetic tra c is more extensive in supraventricular than ventricular myocardium. Sympathetic activity derived from paired ganglia is also not bilaterally symmetrical, but does innervate both atria and ventricles extensively. Parasympathetic activity depresses SA nodal activity, enhances intra-atrial conduction by shortening atrial action potential duration, and slows AV nodal conduction. Conversely, sympathetic e erent tra c increases heart rate and shortens AV conduction time.76 Sympathetic activity also increases cellular excitability, predisposes to some cardiac arrhythmias, and increases myocardial oxygen consumption by augmenting the heart rate, force of myocardial contraction, and myocardial wall tension. Parasympathetic e erent tra c dominates in the resting, standing horse and frequently uctuates with changes in BP. The pronounced sinus arrhythmia, SA block, and second-degree AV block so often encountered in the normal horse is caused by changing vagal tone and serves to regulate ABP at rest (see Figure 10-6).71 Depression of normal SA automaticity or increased activity in other tissues in the atria, Purkinje system, or in cells around the AV node may lead to an abnormal or “ectopic” cardiac rhythm. This situation is often observed in healthy horses under general anesthesia. When the principal abnormality resides in depression of SA node function or blockage of the impulse in the AV node, the slow discharge of a subsidiary pacemaker is termed an escape mechanism. The purpose of these latent cardiac pacemakers is to rescue the heart from extreme bradycardia or asystole. However, drugs, in ammatory mediators, abnormal sympathetic activity, electrolyte disturbances, or ischemia also can enhance ectopic pacemakers abnormally, leading to ectopic rhythms that manifest as premature complexes or tachycardias. Additional abnormal electrophysiologic mechanisms, including abnormal automaticity in tissues other than nodal and conduction tissues (i.e., ischemic myocardium), after depolarizations, and re-entry account for the development of other arrhythmias.

GENESIS OF THE ELECTROCARDIOGRAM The SA node is located near the right auricle; therefore initial cardiac muscle depolarization crosses the RA (see Figure 10-3). Activation waves spread through the atrial myocardial cells to the LA and also in the direction of the AV node. Specialized atrial muscle cells comprising internodal pathways and Bachman’s bundle facilitate transmission of current across the atria.8,435 These specialized pathways—as well as the SA node—are relatively resistant to high serum potassium concentrations and may function during hyperkalemia to cause a sinoventricular rhythm, even while normal atrial myocytes are inexcitable (atrial standstill). Conduction continues across the AV tissues by rst entering the AV node. Current transmission across the AV nodal cells is especially slow because this tissue depends on inward calcium currents and is subject to physiologic blockade from vagal e erent tra c.192 Conduction proceeds at a greater velocity through the bundle of His and bundle branches. Propagation of the impulse through the ventricles is enhanced by a rapidly conducting Purkinje system that penetrates relatively completely through the ventricular myocardium.6,7

The ECG graphs the time-voltage activity of the heart. The average electrical potential generated by the heart muscle is recorded throughout the phases of the cardiac cycle with time displayed along the x-axis and electrical potential inscribed vertically (Figure 10-15). The normal waveforms are the P wave (atrial depolarization), the PR (or PQ) interval that is due mostly to slow AV nodal conduction, the QRS complex (ventricular depolarization), and the ST-T wave (ventricular repolarization). A prominent atrial repolarization wave (Ta wave) is often noted in the PR segment of the equine ECG, particularly at faster heart rates. The QT interval represents total electrical activationrepolarization time.

CLINICAL ELECTROCARDIOGRAPHY The clinical application of electrocardiography to the horse has been studied extensively.∗ The principles of recording and interpreting the equine ECG are similar to those used for humans, dogs, and other species.111,445 The lead systems used are identical, although some modi ed leads, such as the base to apex lead, are most useful for monitoring the cardiac rhythm. A number of semiorthogonal lead systems have been evaluated experimentally but are rarely used in clinical practice. The modi ed Einthoven’s lead system, consisting of leads frontal planes I, II, III, aVR, aVL, and aVF, and the precordial lead V10, is quite applicable for ECG studies of horses (Box 10-7). The base-apex lead or chest leads are most often used for rhythm analysis. BOX 10-7 ELECTROCARDIOGRAPHIC LEADS Frontal Plane Leads Bipolar leads Lead I = Left foreleg [+] – right foreleg [–] Lead II = Left hindleg [+] – right foreleg [–] Lead III = Left hindleg [+] – left foreleg [–] Unipolar augmented limb leads Lead aVR = Right foreleg [+] – left foreleg and left hindleg [–] Lead aVL = Left foreleg [+] – right foreleg and left hindleg [–] Lead aVF = Left hindleg [+] – right foreleg and left hindleg [–]

Precordial Leads Base-apex monitor lead Positive electrode over the left apex compared with the negative electrode over the right jugular furrow Precordial V leads Positive electrode over the selected precordial site; -the lead is named based on the location of the exploring (V or C) electrode Lead V3: near the left apex Lead V10: over the dorsal spinous processes (interscapular) The most common method of obtaining the base-apex lead is to place the left foreleg electrode over the left apex, the right foreleg electrode over the jugular furrow (or at the top of the right scapular spine), and select lead I on the electrocardiograph. This lead is an excellent choice to monitor the cardiac rhythm.

The value of continuous, ambulatory (Holter) ECG monitoring is obvious for identi cation of infrequent rhythm disturbances, quantifying the severity of an arrhythmia, or objectively evaluating drug therapy (Figure 10-16). In one study it was observed that many horses that were historically and clinically without evidence of cardiac disease had supraventricular or (less often) ventricular arrhythmias on Holter ECG.359 Ambulatory monitoring can be performed with contact electrodes using a bipolar lead system similar to that used for the equine heart rate monitors. Electrodes are usually placed over the left saddle area and sternum and are kept moist and in contact with the horse using a surcingle and padding material. This type of continuous 24-hour rhythm monitor can be quite useful to evaluate the horse with a history of syncope or an arrhythmia, but in which an arrhythmia cannot be induced during resting, exercise, and postexercise ECG examinations. An alternative is an ECG event recorder that continuously records the rhythm and stores the ECG after an observed event, such as a sudden fall. These event recorders can be worn for longer periods of time.

FIGURE 10-15

A normal equine base-apex lead electrocardiogram (ECG) (25 mm/sec; 1 cm = 1 mV). The waveforms are indicated. The normal P wave in this lead is positive and notched (bi d), and a depression occurs in the baseline after the P wave that indicates atrial repolarization (Ta wave). The QRS complex often lacks an R wave in this lead (QS complex). The T wave is labile and may be negative, biphasic, or positive. Increases in heart rate or sympathetic tone usually lead to positive T waves in this lead.

FIGURE 10-16

Examples of extrasystoles. Atrial (P′ at top) and ventricular (bottom) extrasystoles in a mare with heart failure, sinus tachycardia, and premature beats. The recordings are from a 24-hour tape-recorded (Holter) electrocardiogram (ECG). Two transthoracic leads recorded simultaneously. The waveforms are indicated. The dark circles indicate the premature complexes. The P-R interval of the APC is longer because of physiologic refractoriness in the atrioventricular node. The ventricular ectopic follows the sinus P wave (late diastolic) but discharges the ventricle before the sinus impulse can cause a normal QRS complex. The T wave of the ventricular extrasystole is abnormal (secondary T wave change). Paper speed is 25 mm/sec.

Telemetry-based ECG recordings are commonly used in exercise testing (see Exercise Testing) and in monitoring of critical patients in hospital settings.317,415 The lead systems used are often similar to those used for Holter ECG recordings; however, the degree of artifact that occurs during exercise is considerable. Those interested in consistent recordings must experiment with various electrode positions and lead systems. The orientation, amplitude, and duration of the ECG waveforms depend on many factors, including the age of the horse,348,464 the lead examined, the size of the cardiac chambers, the degree of training, and even the phase of ventilation.439,440 The principal use of the ECG is to diagnose the heart rhythm, because in horses this examination is insensitive for detecting cardiomegaly, especially in horses with mild to moderate heart enlargement. A normal ECG does not exclude heart disease; moreover, the ECG is not a test of myocardial function. A systematic approach to ECG analysis should be undertaken and compared with reference values (Table 10-4). TABLE 10-4 Electrocardiography: Approximate Limits for Normal Frontal Plane Leads Variable Heart rate Cardiac rhythm P wave P-R interval QRS complex QT interval R wave lead I R wave leads 2, aVF Frontal axis ∗

Maximal Value 44-50 beats/min — 0.16 seconds 0.48 seconds 0.14 seconds 0.575 seconds 0.85 mV 2.3 mV 0-100 degrees

Comments Resting mature horses; higher in foals, colts, and ponies Sinus mechanisms, sinus arrhythmia, vagal-induced first- and second-degree atrioventricular block∗ Atrial repolarization (Ta wave) possibly making measurement difficult Varies; longer when vagal tone is high May vary with the size of the heart (often longer in the base-apex lead) Inversely related to the heart rate

Axis variable in foals and yearlings

When evaluating the electrocardiographic rhythm, one should consider the following points:

Technical aspects: paper speed, calibration, and lead(s) Artifacts: electric, motion, twitching, and muscle tremor Heart rate per minute: atrial and ventricular Cardiac rhythm: atrial, atrioventricular conduction sequence, and ventricular conduction Arrhythmias

Site or chamber of abnormal impulse formation, myocardial fibrillation, or conduction disturbance Rate of abnormal impulse formation Conduction of abnormal impulses Patterns or repeating cycles P wave: morphology, duration, amplitude, and variation P-R interval: duration, variation, and conduction block (of P wave) QRS: morphology, duration, and amplitude Frontal plane mean electric axis ST segment: depression or elevation T wave: changes in morphology or size; QT interval Miscellaneous: electrical alternans, synchronous diaphragmatic contraction Atrial depolarization generates the P wave. Normal activation proceeds from right-to-left and craniad to caudad, leading to positive P waves in left-right lead I and also in craniocaudal leads II and aVF.8,11 The normal P wave is notched or bi d; however, single peaked, diphasic, and polyphasic P waves may be encountered in normal horses. A negative/positive P wave is often recorded if the focus of pacemaker activity shifts to the caudal RA near the coronary sinus (Figure 10-17). The initial peak of the common bi d P wave is reportedly caused by depolarization of the middle and caudal one third of the RA.6 The second peak represents activation of the atrial septum and the medial surface of the LA. The P wave peaks can be subdivided with P1 reported to be as high as 0.25 mV (mean of 0.14) and the second peak, P2, reported to be as high as 0.5 mV (mean of 0.28).445 Even subtle di erences exist among breeds of horses. The P wave morphology can change cyclically with waxing and waning of vagal tone during sinus arrhythmia. During tachycardia, the P wave shortens, becomes more peaked, and is followed by a prominent atrial repolarization (Ta) wave that deviates the P-Q segment downwards. Such features make the diagnosis of atrial enlargement by electrocardiography very difficult.

FIGURE 10-17

P waves of the horse are demonstrated. On the left is a normal, bi d P wave morphology with two distinct peaks (designated as P1 and P2) and also a physiologic second-degree atrioventricular block (arrow). The center panel demonstrates a negative/positive P wave of coronary sinus origin. This is a normal variation. The right panel shows increased amplitude P waves recorded in a horse after conversion from atrial brillation (AF). The second peak is particularly large and may indicate atrial enlargement; however, such voltage criteria correlate poorly with cardiomegaly in horses. Echocardiography is a more accurate method for evaluating atrial size. Lead 2 paper speed is 25 mm/sec.

The time required for conduction across the AV node and His-Purkinje system is estimated by measuring the P-R interval. Because physiologic AV block is so common, signi cant variation exists in the P-R interval, even within the same horse; thus normal maximal values for the P-R interval are di cult to state. Values that persistently exceed 0.5 seconds are probably abnormal. Variation in the P-R interval is not usually related to changes in ventilation42 but is often correlated with changes in BP and baroreceptor activation.71 The morphology of the QRS complex is variable. The relatively complete penetration of the conduction system into the free walls of the ventricle causes these chambers to be activated simultaneously with a burst of depolarization, which cancels much of the divergent electromotive forces.6,7 Consequently, the normal electrical axis may vary and the amplitude of the QRS complex can be quite small in the frontal plane leads. A substantial dorsally oriented vector causes a prominent positive terminal de ection in lead V10, while the (− electrode) right base to (+ electrode) left apex lead exhibits a prominent S wave (see Figures 10-15 and 10-18). The normal slur in the ST segment makes determination of QRS duration di cult in many horses.35,471 The mean amplitude for the R wave in lead II for normal racehorses is about 0.8 to 1.1 mV.∗ Clinical experience suggests that R wave amplitudes exceeding 2.2 mV in lead II or 1.7 mV in lead I are often abnormal. However, occasionally the R wave amplitude in a normal horse exceeds even these limits.

FIGURE 10-18

Multiple lead recordings in a normal horse. The varying configuration of the P-QRS-T complexes across the leads is notable.

The frontal plane leads can be inspected to estimate the mean electrical axis of depolarization, the “average” wave of depolarization (Figure 10-19). This determination is usually reported by direction, using a quadrant orientation (i.e., right or left, craniad or caudad), or is stated in degrees (e.g., lead I is 0 degrees, lead II is 60 degrees, lead aVF is 90 degrees, lead III is 120 degrees, lead aVR is minus 150 degrees, and lead aVL is minus 30 degrees). The general direction of the QRS axis can be quickly estimated by surveying the height of the R waves in the frontal leads and selecting the lead with the greatest net-positive QRS complex. The axis in foals and yearlings is variable and frequently oriented craniad, whereas the frontal axis in most mature horses is directed left-caudad. Abnormal axis deviations have been observed with cardiomegaly, cor pulmonale, conduction disturbances, and electrolyte imbalance.467

FIGURE 10-19

Diagram of the normal frontal plane lead axis. The predominant vectors for the QRS complex in Standardbred (SB) and Thoroughbred (TB) horses are indicated. The orientation of the two components of the normal P wave (P1 and P2; see Figure 10-17) differ slightly. (From Fregin GF: The cardiovascular system. In Mansmann RA, McAllister ES, Pratt PW, editors: Equine medicine and surgery, Santa Barbara, CA, 1982, American Veterinary Publications.)

After ventricular activation, repolarization of the ventricles is recorded. This period is measured from the end of the QRS complex (the “J point”) and extends to the end of the T wave.35,451,472 The T wave vector is most often directed toward the right caudal quadrant, resulting in a positive T wave in lead III and a negative or isoelectric T wave in lead I in resting horses.445 Although some clinical surveys have suggested that abnormalities of the ST-T indicate cardiac dysfunction in performance animals, marked deviation of the ST segment and increased amplitude of the T wave are anticipated in healthy horses during exercise or even with excitement-induced tachycardia. No compelling evidence indicates that T wave abnormalities can be interpreted consistently. Progressive J point or ST segment deviation in the horse with hypovolemia or shock may indicate myocardial ischemia, whereas enlargement of the T wave may develop with myocardial hypoxia or hyperkalemia. Diagnosis and management of cardiac arrhythmias are discussed later in this chapter.

Thoracic Radiography There are signi cant limitations related to the use of thoracic radiography in the evaluation of the equine heart the large size of the mature horse and the ability to obtain only a standing lateral thoracic radiograph in all except the smallest foals.113 While recumbent lateral and, on occasion, ventrodorsal or dorsoventral projections can be obtained on neonatal foals, the stress of restraint or requirement for heavy sedation can make such positioning contraindicated. Thoracic radiographs may be useful to identify areas of pulmonary or pleural disease and assist in the differential diagnosis of respiratory problems. Gross changes in cardiac size and shape may be detected in a lateral thoracic radiograph of a foal or horse with signi cant cardiomegaly.113,342,473-475 A normal thoracic radiograph does not necessarily indicate that the heart is normal in size. Mild to moderate increases in the size of the cardiac chambers may go undetected, particularly in adult horses. Generalized enlargement of the cardiac silhouette is observed in cases of signi cant pericardial e usion or with CHF (Figure 10-20). Dorsal displacement of the trachea may be detected in some horses with LA and LV enlargement. In some horses with LA enlargement, the caudodorsal border of the cardiac silhouette bulges caudally. Increased contact between the ventral border of the heart and the sternum may be detected with RV enlargement but is usually di cult to appreciate. A 50% decrease in the spinotracheal angle (the angle between the dorsal border of the trachea and the ventral border of the adjacent thoracic vertebrae) was demonstrated in young horses with cardiomegaly because of congenital cardiac disease.113

FIGURE 10-20

Thoracic radiography. A, Signi cant cardiomegaly in a Quarter Horse foal with complex congenital heart disease, including multiple ventricular septal defects (VSDs). B, Cardiomegaly and alveolar pulmonary edema in a lly with mitral regurgitation (MR) and left-sided congestive heart failure (CHF). C, Increased pulmonary density and pleural e usion in a Standardbred gelding with atrial brillation (AF), atrioventricular valvular regurgitation, myocardial failure, and biventricular CHF. D, Angiocardiogram obtained from an Arabian foal with pulmonary atresia and a large VSD (pseudotruncus arteriosus). Contrast medium was injected in the left ventricle (LV), and this medium opaci es a dilated, overriding aorta. Right-to-left shunting across the VSD results in dilution of contrast medium. The right arrow shows the subaortic LV out ow tract; the widened aorta is delineated by the arrows at the left. Ao, Aorta; LA, left atrium; LV, left ventricle; RCA, right coronary artery.Thoracic radiography. A, Signi cant cardiomegaly in a Quarter Horse foal with complex congenital heart disease, including multiple ventricular septal defects (VSDs). B, Cardiomegaly and alveolar pulmonary edema in a lly with mitral regurgitation (MR) and left-sided congestive heart failure (CHF). C, Increased pulmonary density and pleural e usion in a Standardbred gelding with atrial brillation (AF), atrioventricular valvular regurgitation, myocardial failure, and biventricular CHF. D, Angiocardiogram obtained from an Arabian foal with pulmonary atresia and a large VSD (pseudotruncus arteriosus). Contrast medium was injected in the left ventricle (LV), and this medium opaci es a dilated, overriding aorta. Right-to-left shunting across the VSD results in dilution of contrast medium.

Evaluation of the pulmonary vasculature and pulmonary parenchyma is also di cult and very dependent on radiographic technique. Enlarged pulmonary vessels associated with pulmonary overcirculation can occasionally be observed in left-to-right shunts; the opposite is also true in congenital right-to-left shunts. Pulmonary edema causes generalized increased radiopacity, particularly in the hilar regions; the characteristic air bronchograms are more readily identified there when alveolar edema is present. Angiocardiography is only practical in foals less than about 115 kg and usually requires general anesthesia, which may be contraindicated in the foal with a severely compromised CV system. Both selective and nonselective positive contrast angiocardiograms have been performed in foals and adult horses (see Figure 10-20, D),156,473,476 but these techniques have been almost exclusively replaced by echocardiography and Doppler studies.

Nuclear medicine imaging has been developed in some equine referral institutions and has been used to assess cardiac function. First-pass nuclear angiocardiography permits the visualization of the cardiac chambers during sequential phases of the cardiac cycle.113 Application of first-pass studies may be useful for identification of cardiac shunting, but again, this diagnosis is more simply attained by echocardiography.

Echocardiography Ultrasound is the mechanical vibration of sound waves within a medium at frequencies greater than 20,000 c/s. These inaudible sound waves follow the laws of optics, can be transmitted through tissue, and can also be refracted and reflected. This last property permits the clinician to image the heart and other organs.477 Echocardiography is the application of ultrasound diagnostics to the heart. When combined with Doppler studies,478 transthoracic echocardiography has evolved to become the most important diagnostic study currently available for evaluation of the equine heart.∗ Transesophageal echocardiography (TEE) has been used in horses,234,235,285 but (except in foals) it requires specially made endoscope-mounted transducers and is only practical in anesthetized or heavily sedated horses because of the fragile (and expensive) nature of the TEE probe. Successful examination of the equine heart requires the use of crystals that vibrate at frequencies between 1.5 to 5.0 MHz, a detailed knowledge of ultrasound anatomy, thorough appreciation of equine cardiac diseases, and technical skills with imaging and instrumentation. Ultimately, a thorough echocardiographic examination yields information about the presence and nature of heart lesions, the size of the heart and great vessels, ventricular function, valvular function, blood ow, and the relevant hemodynamics of a cardiac lesion. Examination ndings are relatively consistent when performed by experienced examiners, but day-to-day, subject, and examiner variation can be significant (>10%) for some variables and must be considered when performing serial examinations.303,307,308,518 The essential principle of echocardiography is straightforward: when ultrasound is directed into the chest and toward the heart, some of the sound energy re ects back to the transducer. This occurs because tissue interfaces with di erent acoustic impedance, such as muscle, collagen and blood, are encountered by the echobeam and ultrasound is reflected from these tissue interfaces. A hand-held transducer acts to both send and receive ultrasound.486,520 The echocardiograph computer is capable of determining the spatial orientation and distance of the returning echoes, processing the signals, and displaying an image of the heart. The image is usually displayed, by convention, with that part of the heart closest to the transducer at the top of the display and adjacent to the transducer artifact. Two imaging modalities, the M-mode and 2D (B-mode or cross-sectional) formats, are in widespread use in equine practices. When Doppler studies or contrast echocardiography is added, blood ow can be detected relative to the 2D and M-mode images (Box 10-8; Figures 10-5 and 10-21 to 10-24; see Figure 10-5).149,303 Mmode, 2D, and Doppler studies can be used to quantify heart size and mass,49,50,481 as well as estimate ventricular function. The in uence of training on cardiac size can be assessed.102,515,516 The technique also can be applied to assess the fetal heart.521 BOX 10-8 INTERPRETATION OF ECHOCARDIOGRAPHIC AND DOPPLER STUDIES General Principles of Interpretation • Evaluate the electrocardiogram relative to cardiac motion and cardiac rhythm disturbances; arrhythmias can alter measures of ventricular function, cause aortic valvular regurgitation, and alter flow and pressure gradients. • Remember that persistent tachyarrhythmias can lead to reversible form of dilated cardiomyopathy (tachycardia-induced cardiomyopathy). • Determine the initial image planes needed for examination based on clinical examination and your standard operating procedures. • Identify the general situs of the heart and cardiac structures; identify the atria, ventricles, cardiac septa, great vessels, and cardiac valves. • Note any dilation, attenuation, or absence of the aorta or pulmonary artery and identify their origins and relationships to the atrioventricular valves and to the ventricles. • Note the presence of unanticipated or lack of expected structures. • Identify pleural effusion. • Rule out extrapericardial mass lesions of the lung or thorax.

Pericardium and Pericardial Space • Rule out pericardial effusion and pericardial mass lesion. • Mixed echoic effusions or shaggy tags may indicate highly cellular or inflammatory effusate. • In pericardial e usion: Identify diastolic collapse of the right atrium or right ventricle indicating cardiac tamponade or a large bilateral pleural e usion; protracted collapse or ventricular or atrial inversion are more reliable signs of tamponade. Exuberant swinging of heart may be seen.

• In pericardial effusion: Identify any heart-related lesions using multiple complementary planes. • If there is concern about constrictive pericardial disease, evaluate the motion of the ventricular septum (for accentuated utter or “septal bounce”) and perform Doppler studies of the left and right sides of the ventricular inlets; large E-waves with abrupt termination of lling may be observed; evaluate the e ects of ventilation because marked variation may be observed in E-waves; tissue Doppler may also be instructive in cases of constrictive disease.

Left Atrium, Pulmonary Veins, and Atrial Septum • Identify pulmonary veins and pulmonary venous entry. • Examine for attenuation or small size: Rule out volume depletion, right-to-left shunt, or low cardiac output. • Examine for enlargement or rounding/turgid appearance: rule out left-to-right shunt, mitral valve disease, left ventricular systolic or diastolic failure, chronic arrhythmia (e.g., atrial fibrillation). • Measure left atrial diameter and/or left atrial area by 2D imaging. • Examine the atrial septum for abnormal bowing to the left or right atrium (high atrial pressure). • Examine the atrial septum for septal defects or patent foramen ovale (perform Doppler and “bubble” contrast studies if necessary). • Examine the blood pool for spontaneous contrast. • Examine the atrioventricular groove for dilated vascular structure: Rule out dilated coronary sinus and left cranial vena cava. • 2D Echocardiography and tissue Doppler can be used to evaluate atrial wall motion and function.

Mitral Valve • Identify two valve lea ets and cusps in long and short axis planes; if evidence of mitral regurgitation, also examine mitral valve from left caudal imaging windows in two or more imaging planes. • Observe motion during cardiac cycle by 2D and M-mode echocardiogram. • Examine the support apparatus (chordae tendineae, papillary muscles). • Increased valve echogenicity: Rule out degenerative thickening, valvulitis, vegetation (infective endocarditis), malformation (rare). • Cleft or common septal leaflet: Rule out endocardial cushion defect/primum atrial septal defect (rare defects). • M-mode: Reduced diastolic (E-F) slope: decreased transvalvular flow. • Lack of diastolic separation or increased mitral E point to septal distance: Rule out aortic regurgitation (regurgitant jet impinging on the valve); left ventricular dilation with left ventricular failure (reduced transmitral flow), or mitral stenosis/tethering (rare defects or acquired from endocarditis). • Prolapse of mitral leaflet into left atrium: Rule out degenerative disease, elongated or ruptured chordae tendineae, lesion of papillary muscle. • Flail leaflet: Rule out ruptured chordae tendineae or avulsion of papillary muscle. • Double line of mitral closure (mobile): Rule out ruptured chordae tendineae (flail leaflet). • Diastolic mitral valve fluttering: Rule out aortic regurgitation. • Systolic mitral fluttering: Rule out mitral regurgitation (musical). • Chaotic valve motion: Rule out arrhythmia (atrial flutter, premature ventricular complexes); ruptured chordae tendineae. • Premature (diastolic) closure: Rule out severe semilunar valve insufficiency, long P-R interval, or atrioventricular block. • Delayed (systolic) closure (B-shoulder): rule out left ventricular failure and elevated atrial pressure. • Systolic anterior motion of the mitral valve (systolic anterior motion, systolic mitral–septal contact): rule out dynamic left ventricular out ow tract obstruction due to valve malformation, subaortic stenosis, volume depletion, pulmonic stenosis, or hypertrophic or infiltrative cardiomyopathy (all very rare). • Doppler studies can be used to interrogate the mitral valve for regurgitation or abnormal ow; multiple planes should be obtained including short-axis images at the level of the

left atrium immediately dorsal to the mitral valve. • Mitral regurgitation is common. Eccentric jets are often observed. Do not confuse “back ow” color noise with true regurgitation. Horses can develop mid-to-late systolic mitral regurgitation (as with mitral prolapse). Multiple jets of mitral regurgitation are common. • Always time ow events of mitral regurgitation (color M-mode and spectral Doppler); do not misdiagnose diastolic mitral regurgitation. Do not overemphasize receiving chamber color coding in mitral regurgitation because red blood cell entrainment and a “spray e ect” can result in overestimation of mitral regurgitation severity, whereas wall-hugging jets underestimate severity of mitral regurgitation. Attempt to measure jet width at origin and correlate to left atrial and left ventricular chamber size.

Left Ventricle • Evaluate from long and short axis tomograms; inspect contour and walls; dorsal septum is normally slightly thicker than left ventricle free wall. • Echogenic smoke/contrast: Some spontaneous contrast may be normal; rule out low output states, bradycardia, systemic inflammation (?). • Evaluate left ventricle dimensions (internal dimensions, thickness of interventricular septum and free wall) and left ventricle systolic function (fractional shortening, ejection fraction) using M-mode or 2D imaging. • Left ventricle dilation: Rule out causes of volume overload or left ventricle failure: mitral or aortic valvular regurgitation, dilated cardiomyopathy, myocarditis, left-to-right shunt, persistent tachyarrhythmia. • Left ventricle wall thinning: Rule out infarct, prior myocarditis, congenital aneurysm (rare). • Left ventricle wall thickening: Consider “athletic heart”; rule out pseudohypertrophy from volume depletion, in ltrative cardiomyopathy (amyloid), myocarditis, chronic systemic hypertension, or left ventricle outflow obstruction (very rare). • Hyperkinesis of left ventricle: Rule out mitral or aortic insufficiency, bradycardia, sympathetic stimulation. • Hypokinesis or dyskinesis of the left ventricle: Rule out cardiomyopathy or other myocardial disease, ischemia, infarct, arrhythmia, regional wall infiltration, or myocarditis. • Hyperechoic myocardium or subendocardium: Rule out recurrent ischemia, fibrosis, infiltration, infarction, myocarditis, amyloidosis. • Tissue Doppler and 2D strain (2D speckle imaging) may be used to evaluate global and regional left ventricle function; transmitral ow can be obtained but is limited with respect to evaluation of left ventricle diastolic function

Left Ventricular Outlet, Aortic Valve and Aorta • Identify aortic valve lea ets and motion during cardiac cycle: Identify abnormal lea et morphology (e.g., thickenings, fenestrations) or motion. Examine in both long-axis and short-axis image planes to see all leaflets. • Thickened leaflets: diffuse or nodular: Rule out degeneration, endocarditis. • Prolapse of aortic valve into the left ventricle: Rule out degenerative semilunar valve disease (common), ventricular septal defect, or bacterial endocarditis with torn leaflet • Diastolic fluttering of the aortic valve: Rule out aortic insufficiency. • Systolic fluttering or the aortic valve: normal or high flow state. • Lack of systolic separation: Rule out valvular cardiac output, arrhythmia, stenosis (rare). • Decreased aortic diameter: low cardiac output, hypotension. • Aortic dilation: Rules out aortic regurgitation tetralogy of Fallot, pulmonary artery atresia, truncus arteriosus, patent ductus arteriosus (rare). • Aortic sinus of Valsalva aneurysm: focal dilation or ballooning of affected sinus: Evaluate for rupture into ventricular septum or right atrium. • Doppler studies may be used to identify blood ow patterns in the left ventricle out ow tract, aortic regurgitation, paramembranous ventricular septal defect and rare cases of aortic stenosis. Doppler studies also can identify abnormal blood flow paths in rare cases of aortic sinus ruptured into the heart.

Ventricular Septum • Septal hyperkinesis: as per the left ventricle. • Septal hypokinesis: as per left ventricle; also rule out right ventricular pressure or volume overload.

• Paradoxical ventricular septal motion: rule out moderate to severe right ventricular volume overload such as atrial septal defect, severe tricuspid regurgitation, severe pulmonary hypertension with tricuspid regurgitation. • Flat ventricular septum: rule out right ventricular volume overload (diastolic flattening) or right ventricular pressure overload (systolic flattening). • Exaggerated and disparate ventricular filling rates with diastolic septal fluttering: Rule out constrictive pericardial disease. • Examine for ventricular septal defects in both long-axis and short-axis image planes (perform Doppler studies if necessary). • Echocardiogram drop-out or discontinuity of the ventricular septum: Rule out ventricular septal defect (peri/paramembranous adjacent to right/noncoronary cusps of the aortic valve and septal lea et of tricuspid valve; inlet septal ventricular septal defect (ventral to tricuspid valve); muscular or trabecular ventricular septal defect; subpulmonic ventricular septal defect (also termed outlet, supracristal, subarterial, or doubly committed ventricular septal defect; directly underneath pulmonic and aortic valves); tetralogy of Fallot; pulmonary artery atresia; truncus arteriosus, or a false defect caused by angle of the ascending aorta or aortic dilation. • Malalignment of the ventricular septum and anterior aortic root: examine for tetralogy of Fallot, malalignment type ventricular septal defect. • Examine for dissection or “track” that may suggest rupture of the aorta into the ventricular septum with subsequent fistula. • High-frequency fluttering of the ventricular septum: Rule out aortic regurgitation. • Doppler studies, especially color Doppler, are useful for recognizing ventricular septal defects; continuous wave Doppler should be used to measure velocity across shunts to evaluate for “restrictive” versus unrestrictive septal defects. High velocity ow across a ventricular septal defect (>4.75 m/s) generally suggests a restrictive defect more likely to be well tolerated.

Right Atrium and Atrial Septum • Examine for attenuation: If small, rule out volume depletion, external compression, or mass lesion impairing venous return. • Examine for enlargement: If enlarged, rule out tricuspid regurgitation, tricuspid stenosis or atresia (rare), right heart failure, atrial septal defect, moderate to severe anemia or chronic arrhythmia. • Examine the atrial septum for abnormal bowing into the left or right atrium (high atrial pressure). • Examine the atrial septum for septal defects or patent foramen ovale (double mobile lines in foals). • Doppler studies can be used to identify atrial septal defect, but streaming of normal flow patterns (i.e., caudal vena cava flow) is confusing. • “Bubble studies” (by injecting agitated normal saline into the jugular vein) should be used to identify right-to-left shunting of blood. Caution: Similar to Doppler studies, streaming of normal blood from the caudal vena cava may produce negative contrast in the right atrium, falsely suggesting left-to-right shunting of blood.

Tricuspid Valve • Identify valve leaflets and cusps in long- and short-axis image planes. • Observe motion during cardiac cycle. • Identify support apparatus (chordae tendineae, papillary muscles). • Increased valve echogenicity: Rule out degenerative thickening, vegetation, malformation, thrombus on valve. • Common septal leaflet: Rule out endocardial cushion defect/primum atrial septal defect (rare). • Flail leaflet: Rule out ruptured chordae tendineae or avulsion of papillary muscle; rule out infective endocarditis. • Chaotic valve motion: Rule out arrhythmia, ruptured chordae tendineae. • Premature (diastolic) closure: Rule out severe pulmonic insufficiency, long P-R interval, or atrioventricular block. • Doppler studies can interrogate the tricuspid valve for abnormal ow or regurgitation; most horses have some small jets of tricuspid regurgitation by color Doppler evaluation; use color M-mode and spectral Doppler to evaluate timing and peak velocity. High velocity tricuspid regurgitation (>3 m/s): Also rule out pulmonary hypertension, large ventricular septal defect, right ventricular outflow obstruction.

Right Ventricle • Dilation: Rule out severe tricuspid insu ciency, chronic right ventricular pressure overload, biventricular congestive heart failure, atrial septal defect, large ventricular septal defect, severe pulmonary insufficiency, persistent tachyarrhythmia. • Hypertrophy: Rule out large (unrestrictive) ventricular septal defect, pulmonary hypertension, pulmonic stenosis (rare), tetralogy of Fallot (rare), or complex congenital heart disease. • Midventricular bands: Rule out double chambered right ventricle (examine for ventricular septal defect); remember that the right ventricular moderator band can be very prominent in horses. • Hyperechoic tissue: Rule out myocardial fibrosis.

Right Ventricular Outlet, Pulmonary Valve, and Pulmonary Artery • Identify valve leaflets and motion during cardiac cycle; if pulmonary valve lesion is suspected, also interrogate from a left cranial transducer position. • Thick, fused, or hypoplastic leaflets: Rule out congenital stenosis or dysplasia (rare). • Diastolic fluttering of pulmonic valve: Rule out valvar insufficiency.

• Systolic fluttering of pulmonic valve: normal or high flow state. • Lack of systolic separation of the pulmonic valve: Rule out low cardiac output, arrhythmia, stenosis. • Pulmonary artery, absence or attenuation: Rule out pulmonary atresia or pulmonary artery hypoplasia. • Pulmonary artery, dilation of: Rule out pulmonary hypertension, left-to-right shunt, poststenotic dilation of pulmonic stenosis (rare); rupture of the aorta into the right heart or pulmonary artery (rare). • Doppler studies can be used to identify abnormal ow in the pulmonary artery, including the rare pulmonic stenosis (high velocity), increased velocity ow from a left-to-right shunt, and patent ductus arteriosus in equine neonates. Pulmonary regurgitation is normal; high velocity pulmonary regurgitation indicates pulmonary hypertension.

FIGURE 10-21

Two-dimensional (2D) echocardiograms obtained from the right side of the chest. An ECG is recorded simultaneously for timing. A, Four-chamber view. B, Leftventricular out ow tract view. C, Right-ventricular in ow and out ow tract view. These images, recorded in a right-parasternal long-axis view, allow subjective assessment of cardiac structures and myocardial function and measurement of selected cardiac dimensions. D, M-mode echocardiogram of the normal left ventricle (LV) performed in a rightparasternal short-axis view at the level of the chordae tendineae. The myocardial wall motion (y-axis) along the cursor line (arrowheads) is displayed over time (x-axis). An ECG is recorded simultaneously for timing. This view allows subjective assessment of right- and left-ventricular dimensions and left-ventricular systolic function, measurement of the leftventricular internal dimensions in systole and diastole, and calculation of the left-ventricular fractional shortening (percentage of change in the internal dimension; FS = [LVIDd – LVIDs]/LVIDd × 100). The latter provides an index of systolic left-ventricular function. LV, Left ventricle; LA, left atrium; RV, right ventricle; RA, right atrium; IVS, interventricular septum; Ao, aorta; PA, pulmonary artery; LVFW, left-ventricular free wall; LVIDd, left-ventricular internal diameter at end diastole; LVIDs, left-ventricular internal diameter at peak systole. (From Schwarzwald CC, Bonagura JD, Muir WW: The cardiovascular system. In Muir WW, Hubbell JAE, editors: Equine anesthesia: monitoring and emergency therapy, ed 2, St Louis, 2009, WB Saunders).

FIGURE 10-22

M-mode echocardiography. A, M-mode tracing demonstrating a hyperdynamic left ventricle (LV) in a horse with acute mitral regurgitation (MR). The signi cant change in LV dimensions from diastole (D) to systole (S) is notable and is typical of volume overloading with preserved ventricular function. Furthermore, the reduced resistance to ejection of blood into the left atrium (LA) enhances ventricular shortening. Depth calibration in millimeters is shown on the left. B, Recording through the ventricles demonstrating decreased contractility with a reduced LV shortening fraction. This pattern of contraction can be caused by myocarditis, myocardial injury (e.g., monensin), idiopathic dilated cardiomyopathy, chronic volume overload, protracted ventricular tachycardia (VT), or administration of negative inotropic drugs. Depth calibration in millimeters is shown on the left (RV, right ventricle). C, M-mode echocardiogram across the ventricles recorded from a horse with cardiac amyloidosis. Global LV systolic function is mildly reduced. The IVS and LVPW are thickened. The IVS also appears hyperechoic relative to the RV wall. The LVPW in the far eld is less echogenic owing to attenuation of echoes in the far eld. The right ventricle (RV) contained spontaneous echocontrast. Depth calibration in millimeters is shown on the left (RV, right ventricle; IVS, interventricular septum; LV, left ventricle; LVPW, left ventricular peripheral wall). D, Recording from a horse with aortic regurgitation (AR) and atrial brillation (AF), demonstrating ne diastolic uttering of an aortic valve lea et ( small arrows). The aortic root (AO), valve opening (O), and valve closing (C) are indicated. The murmur (m) of AR is evident in the previous phonocardiogram. First (1) and second (2) heart sounds are labeled. Diastolic uttering of the mitral valve (most common), aortic valve, ventricular septum, or walls of the aortic root may be observed in horses with this hemodynamic abnormality. The M-mode sampling rate (approximately 1000 pulses/sec) is ideal for detecting these high-frequency events.

FIGURE 10-23

Two-dimensional (2D) echocardiograms. A, Long-axis image from the right thorax in a 12-year-old Quarter Horse gelding demonstrating signi cant right ventricular (RV) and right atrial (RA) dilation after tricuspid regurgitation (TR) and elevated pulmonary artery pressure. In this case a tumor that obstructed ow in the main pulmonary artery caused the pulmonary hypertension. The ventricular septum is at and bulges slightly into the left ventricle (LV). B, Images of the ascending aorta and pulmonary artery obtained from the right hemithorax (left panel) and left cranial hemithorax. An evaluation of the pulmonary artery diameter relative to the aorta is useful when identifying pulmonary hypertension. C, Biatrial dilation in a Thoroughbred colt with mitral regurgitation (MR), pulmonary hypertension, atrial brillation (AF), and congestive heart failure (CHF). Both atria are rounded and appear turgid. The cause of mitral disease in this case was idiopathic lymphocytic-plasmacytic mitral valvulitis (RA, right atrium; LV, left ventricle; LA, left atrium). D, Contrast echocardiogram demonstrating right-to-left shunting at the level of the foramen ovale in a foal with severe respiratory disease. The saline generates echocontrast, opaci es the right atrium (RA) and right ventricle (RV), and visibly lls the left atrium (LA) (arrows), although the LV has not yet been opaci ed. This technique is easy and practical for demonstrating right-to-left shunts across the cardiac septa.

FIGURE 10-24

A to C, Black-and-white representations of color Doppler echocardiograms demonstrating mapping of blood through the heart. A, Perimembranous ventricular septal defect (VSD) in a foal (arrow). The ow pattern was coded red and aliased as the velocity increased across the defect. B, Wide spray of mitral regurgitation (MR) with ow moving from left ventricle (LV) into the left atrium (LA) in a horse with mitral valve prolapse (RV, right ventricle; RA, right atrium). C, Color M-mode study from a horse with tricuspid regurgitation (TR). The bright systolic ow pattern, following the QRS of the electrocardiogram (ECG), was mapped as turbulence in the color study. This is an excellent method for timing abnormal ow events. D, Example of continuous wave (CW) spectral Doppler echocardiogram obtained from a horse with a VSD and mild aortic regurgitation (AR). Time and the ECG waveforms are shown along the x-axis and velocity of ow along the y-axis with a meters per second scale (m/sec). Direction of ow relative to the transducer is also shown; ow toward the transducer is displayed as a positive signal (relative to the zero velocity baseline) and away from the transducer as a negative signal. A low-velocity ow pattern is evident in diastole related to diastolic shunting across the VSD, aortic regurgitant ow moving across the cursor line “o angle” (with parallel alignment, the AR velocity would be higher [i.e., >3.75 m/sec]), or contaminated transtricuspid ow. In systole a high positive velocity signal is recorded (>5 m/sec) compatible with a small (restrictive) left-to-right shunting VSD. However, the maximum velocity is not evident because the signal reaches the upper limit of the velocity scale (arrow). The negative signal (lower arrow) is either an aliased signal (if antialiasing lters are o ) or VSD turbulence in the other direction. Moving the baseline down or increasing the velocity scale would have allowed for faithful recording of the peak velocity. (note: For color images of color Doppler echocardiograms, see case examples included on the DVD.)

The M-mode echocardiogram is a single-crystal ice pick image of the heart (see Figures 10-5 and 10-22).13,14,53,522 The movement of the cardiac structures (vertical axis) is displayed over time along the horizontal axis. The ECG can be recorded to provide a timing reference if desired, and the depth of the cardiac structures from the transducer (y-axis) is displayed in centimeters. Visualization of the characteristic movements of cardiac structures permits the experienced viewer to evaluate and quantify cardiac anatomy and function (see Figure 10-21). The high sampling rate of the M-mode study makes it excellent for visualizing rapidly vibrating structures, such as the oscillating mitral leaflet in AR.138 The 2D echocardiogram generates a cardiac eld by mechanically or electrically sweeping one or more piezoelectric crystals across the heart.∗ In conventional transthoracic echocardiography, the operator hand-directs the imaging probe to achieve a suitable tomographic plane. The pie-shaped image obtained has both depth and breadth but has no signi cant thickness. Accordingly, di erent image planes must be used to interrogate the three-dimensional (3D) heart. These imaging planes are designated long-axis (sagittal), short axis (coronal), apical (when the transducer is near the left apex), and angled (hybrid) views. Apical images are difficult to obtain except in foals. The 2D eld is constantly updated to visualize cardiac motion in real time, and this is done at typical sampling rates of 20 to 40 frames/sec, with the update rate inversely related to penetration depth and angle of eld. Fully digital echocardiographs can display much faster frame rates, often exceeding 60 frames/sec. In most cases the 2D study is watched in “real time” and also recorded digitally or on videotape for subsequent playback and analysis. Human interpretation is generally limited to 32 frames/sec so that recordings made at higher frame rates require slow-motion playback for detailed analysis. 2D echocardiography allows assessment of cardiac anatomy, detection of macroscopic structural lesions, subjective evaluation and measurement of chamber and vessel dimensions, and evaluation of LA and LV function. Doppler echocardiography relies on the Doppler principle to measure the direction and velocity of red blood cells (RBCs) in the heart (see Figures 10-24 and 10-5).∗ In Doppler echocardiography, a portion of the ultrasound emitted by the transducer strikes moving RBCs. These targets cause ultrasound to be re ected back to the transducer. Because the RBCs constitute a moving “source” of (re ected) ultrasound, the returning sound waves attain a frequency that is slightly di erent from that originally transmitted (the “carrier frequency”). When the echocardiograph unit records the Doppler frequency shift, calculation of RBC velocity and direction relative to the transducer is possible. The information is displayed as a Doppler spectrum showing time along the horizontal axis, ow direction relative to the transducer as above

(toward) or below (away from) a zero baseline, and calculated RBC velocity along the y-axis. Thus pulsed Doppler methods measure direction and velocity of RBCs within a discrete area of the heart or circulation. Disturbed ow, as might be recorded across a regurgitant orifice or VSD, results in a high-velocity and broadened-velocity spectrum. Color-coded Doppler imaging is a more re ned example of pulsed-wave technology, whereby ow toward the transducer is coded in red and ow away is represented in blue. Calculated velocity is displayed in relative shades of these colors, and green or yellow are added to the ow mapping to identify “turbulence.” Color-coding permits superimposition of ow information onto the 2D or M-mode image (see Figure 10-24). Although substantial technical challenges exist in terms of penetration and frame rate, color-Doppler imaging is useful in horses because a large area of the heart can be screened for flow disturbances. For example, it is much easier to find a jet of MR using color Doppler imaging than with other Doppler methods. A pivotal limitation of color Doppler regards temporal resolution. Depending on the system used, the frame rates of interrogation can be very slow (often 2 to 3 mg/kg) has occurred most frequently, but salinomycin (0.6 mg/kg or higher) and lasalocid (21.5 mg/kg or higher) also have caused myocardial injury and death in horses.677 Ionophores react with polar cations to form lipid-soluble complexes, leading to cation transport across myocardial cell membranes. Various ionophores demonstrate particular a nities for di erent cations, although lasalocid may complex with a variety of ions. Exposure of horses to the ionophore antibiotics usually stems from accidental contamination of equine feedstu s at the mill or through accidental delivery of poultry or cattle feed to horses. In most outbreaks, a recently acquired ration was fed. Clinical signs of horses with ionophore toxicity vary with the speci c type, quantity, and concentration of ionophore ingested and the preexisting health and body condition of the exposed horses. A wide range of clinical signs has been observed by one of the authors in exposed horses ranging from none to clinical signs involving almost all body systems. Weakness, lethargy, depression, anorexia, ataxia, colic, diarrhea, profuse sweating, and recumbency have been seen. The cardiac ndings are similar to those previously described for myocardial diseases. If sudden death occurs, it is usually within 12 to 36 hours of ingestion of the contaminated feed. A single dose may lead to death from cardiac arrhythmias before the development of myocardial necrosis. A recent report indicates that sublethal toxicosis may be well tolerated by some but not all horses, and echocardiography and exercise testing may help to identify horses a ected more severely.120 Polyuria and hematuria are other signs that have been reported in ponies after ionophore exposure. A diagnosis of ionophore toxicity is based on the detection of the ionophore in the feed or stomach contents of exposed horses. Various clinicopathologic abnormalities are observed with monensin toxicity, including decreased serum calcium, potassium, magnesium, and phosphorus, as well as increases in serum urea, creatinine, unconjugated bilirubin, aspartate aminotransferase (AST), and muscle enzymes. Elevated concentrations of serum cardiac troponin cTnI and isoenzyme patterns of CK and LDH have indicated cardiac, skeletal, and RBC damage. Elevated packed cell volume (PCV) and total solids have been associated with dehydration. Echocardiographic evaluation of a ected horses has revealed marked decreases in shortening fraction with segmental wall motion abnormalities that range from mild to severe. Prognostically, horses that exhibit decreased shortening fraction and dyskinesis shortly after exposure to monensin generally are unlikely to survive. Horses with mild decreases in shortening fraction survive and may be useful breeding animals, but most do not return to previous performance levels. Horses with normal echocardiograms typically survive and return to work at their previous levels. Postmortem ndings range from no visible lesions to myocardial pallor and signs of CHF. Severe myocardial necrosis and fibrosis has been observed in horses with

decreased shortening fraction or ventricular dyskinesis on echocardiography. Treatment for a ected horses is largely symptomatic (see previous discussion), unless very recent exposure is known. Vitamin E has been suggested to have a protective e ect in other species and may be bene cial in a ected horses. Digoxin and calcium channel blockers are contraindicated in acutely a ected horses. If ingestion of the contaminated feedstu is recent, then treatment with activated charcoal or mineral oil is indicated to reduce absorption of the ionophore. IV uid and electrolyte replacement therapy may be indicated, as well as antiarrhythmic drugs, for any life-threatening arrhythmias. Stall rest in a quiet environment for up to 8 weeks after exposure is most important, because echocardiograms recorded after trivial exercise or excitement can reveal residual disease characterized by marked decreases in fractional shortening and myocardial dyskinesis.

Dilated Cardiomyopathy Idiopathic dilated cardiomyopathy is a disorder of the myocardium characterized by global reduction of LV systolic function that cannot be explained by valvular, vascular, coronary, or congenital heart disease. The inciting cause of dilated cardiomyopathy is generally undetermined, although myocarditis or prior toxic injury is often suspected. Relentless junctional or VT also can lead to a dilated cardiomyopathic state that is reversible with control of the tachyarrhythmia. The clinical signs of cardiomyopathy are similar to those described earlier for myocardial diseases. Echocardiography is diagnostic, revealing cardiomegaly with biatrial and biventricular dilatation and depressed shortening fraction (see Figure 10-22, B). Symptomatic therapy with digoxin and diuretics may temporarily stabilize CHF and lead to transient improvement, but most horses deteriorate in the 3 to 12 months after diagnosis and are humanely destroyed. Neither the history nor the postmortem examination reveals the cause, and generally only di use or multifocal myocardial degeneration, necrosis, and fibrosis are observed. A form of dilated cardiomyopathy also can be caused by vitamin E and selenium de ciency and is observed primarily in fast-growing foals from mares with a marginal or de cient selenium status raised in selenium-de cient areas. A ected foals are usually younger than 6 months of age and experience an acute onset of weakness, recumbency, respiratory distress, pulmonary edema, tachycardia, murmurs, and arrhythmias. The prognosis for foals a ected with the myocardial manifestations of white muscle disease is poor, and most die within 24 to 48 hours after the onset of clinical signs. Laboratory abnormalities in a ected foals include marked elevations of CK (including the MB fraction in foals with myocardial involvement), AST, and LDH, as well as hyperkalemia, hyponatremia, and hypochloremia. Myoglobinuria may occur. Echocardiography demonstrates the severity of myocardial involvement. Whole blood selenium, RBC glutathione peroxidase, and vitamin-E levels may be helpful in the diagnosis; however, tissue samples provide a more accurate indication of selenium stores. Treatment with vitamin E and selenium might be successful, but typically the myocardial necrosis is extensive and incompatible with life. Postmortem ndings reveal pale streaking of the myocardium with intramuscular edema, myodegeneration, myocardial necrosis, and brosis or calci cation. Prevention of white muscle disease is important in selenium-de cient areas. Supplementation of pregnant mares should occur during gestation based on individual blood and tissue selenium levels and should be continued during lactation, because more selenium is passed to the foal through the milk than across the placenta.

VASCULAR DISEASES Acquired disorders of blood vessels include a variety of conditions, varying in cause and ranging from subclinical to devastating.∗ Jugular vein thrombosis or thrombophlebitis is probably the most common vascular problem encountered in clinical practice (Figure 10-48). Catheter embolization is observed periodically and has been amenable to surgical and percutaneous catheter retrieval.699,700 Venous aneurysms,680 vascular and lymphatic malformations, and angiomatous lesions are rare.612 Rupture of the aorta or PA is usually a lifethreatening event or leads to severe sequelae. Rupture of aortic sinus aneurysm into the ventricular septum or right heart chambers (Figures 10-49 and 10-50), aortic rupture into the pericardium, aortic aneurysms, and acquired aortic-pulmonary stula have been reported.34,58,219,221,698 Degenerative lesions (see Figure 10-50, A) including calci cation may be observed in the aorta or PA701 Degenerative and calci c lesions a ecting the bifurcation of the right and left pulmonary arteries also have been reported,702,703 although the clinical importance of these is not well de ned. Arteriosclerosis and arterial thrombosis may be caused by parasite migration in the mesenteric arteries704 and possibly at the terminal aortic quadrifurcation (Figures 10-51 and 10-52). Aortic in ammation of unknown cause has been observed.508,513 Rupture of the PA is a potential consequence of longstanding pulmonary hypertension and left heart failure. Arteriovenous communications occur rarely and may develop after vascular rupture or subsequent to growth of vascular tumors. Pulmonary vascular disease and associated hypertension may place a load on the RV, a condition called cor pulmonale. Although both acute and chronic airway obstruction can negatively a ect right heart and overall cardiac function,555,705 there is no compelling evidence to incriminate cor pulmonale as common cause of clinically important heart disease in horses. Idiopathic pulmonary hypertension has been reported as a cause of AF in horses; however, left heart dysfunction was not completely excluded in the reported cases.62 Tumors of the great arteries and veins

are rare,612 but neoplasms can compress or invade arteries and veins, including the PA and venae cavae. Mycotic disease within the guttural pouch is a well-recognized cause of arteritis and rupture of the affected arteries.

FIGURE 10-48

A, Postmortem specimen of a thrombosed jugular vein in cross-section. B and C, Ultrasonographic images obtained form a horse with jugular vein thrombosis in a cross-sectional (B) and a longitudinal (C) plane. The images show a mixed echogenic thrombus within the jugular vein (arrowheads). The common carotid artery (CCA) is displayed adjacent to the jugular vein. Residual blood ow through the a ected area can be evaluated using duplex Doppler ultrasonography or newer vascular imaging technologies that directly visualize blood reflectors (e.g., red blood cells [RBCs]) in a gray-scale display (see DVD).

FIGURE 10-49

Aortic to cardiac stula. O -angle long-axis image of the proximal aorta (Ao) shows the origin (arrow) of a ruptured sinus of Valsalva aneurysm. This channel communicated with the right atrium (RA) in another plane. The right ventricle (RV) and pulmonary artery (PA) are slightly dilated because of left-to-right shunting.

FIGURE 10-50

Aortic disease. A, Severe atheromatous change with secondary Streptococcus spp. infection of the proximal aorta. The intimal and subintimal change begins just distal to the coronary ostia. Representative sections of the ascending aorta are shown. Strongylus vulgaris was not found. The most distal segment was necrotic and severely narrowed ow across the region. B, Dissecting lesion communicating the root of the aorta (top panel, arrow) to the right side of the heart across the intraventricular septum (IVS). The distorted tricuspid valve is retracted to demonstrate the lesion on the right (arrow).

FIGURE 10-51

Aorto-iliac thrombosis. A, Postmortem image showing the thrombus in the terminal aorta (Ao) and the iliac arteries. B, Images obtained by rectal ultrasonographic examination in cross-sectional (left) and longitudinal (right) direction. The thrombus is evident as hyperechoic mass within the vessel (arrow).

FIGURE 10-52

Verminous arteritis and thrombosis in the cranial mesenteric artery caused by Strongylus vulgaris.

The echocardiographic observation of intravascular, spontaneous contrast within blood vessels and cardiac chambers has been suggested to indicate CV or rheologic disease,490,706 as well as a normal phenomenon caused by rouleaux formation or platelet aggregates. This observation is common during ultrasound examinations and contrast may be observed in a variety of vessels, as well as in the heart (see Figure 10-47). Pronounced spontaneous echo contrast is especially likely when low ow rates exist as a result of bradycardia, arrhythmia, or myocardial failure, as well as when actual obstruction to blood ow in vessels is noted. Contrast is often prominent in valvular endocarditis. Lesions, thrombosis, or even venipuncture of the jugular vein can lead to spontaneous echo contrast in systemic veins and in blood ow entering the right side of the heart, often becoming become more obvious if venous return suddenly increases. The identi cation of

spontaneous contrast is also dependent on the operator with respect to transducer, gain and gray-scale contrast settings. The clinical signi cance of isolated intravascular contrast cannot be stated with certainty; it has been observed in healthy horses, as well as those with CV disease. Serious vascular disease often can be suspected from the clinical examination, although it may not be obvious until a catastrophic vessel rupture or fatal hemorrhage occurs. In general, signi cant arterial disease reduces perfusion to the a ected vascular bed, impairing tissue metabolism and function. If severe or complete, then ischemia may lead to tissue necrosis (infarction) and even perforation of a hollow organ. A bruit (auscultatory evidence of vascular narrowing or rupture) may be noted over the a ected artery (or obvious hemorrhage if a super cial the vessel has ruptured). Ultrasound and Doppler examination can demonstrate serious arterial disease provided an acoustic window for examination is available. Rarely angiography or nuclear scintigraphy is required to demonstrate arterial thrombosis or disruption. In major venous disease, the typical sign is swelling of soft tissues drained by the a ected vessel. Palpable evidence of local in ammation and thrombosis may be found in super cial venous disease. Right-sided endocarditis, embolic pneumonia, and pulmonary embolism are potential consequences of systemic venous disease.

Thrombophlebitis Thrombophlebitis, the in ammation and thrombosis of a vein, is most commonly caused by IV catheterization or IV injection (see Figure 10-48).398,689 The jugular veins are most often a ected. Thrombosis can develop in the absence of preexisting vessel trauma subsequent to coagulation disorders. Clinical signs of phlebitis are straightforward, including swelling over the a ected vein and pain on palpation of the involved tissues. Thrombotic occlusion results in distention of the veins and possibly subcutaneous edema adjacent to the a ected area. Marked swelling, heat, and pain on palpation, combined with fever, hyper brinogenemia, or a neutrophilic leukocytosis, indicate infection of the thrombus. The diagnosis and assessment of thrombophlebitis is based on the detection of the previously described clinical signs and examination by duplex Doppler ultrasonography (see Figures 10-48, B and C). Imaging of the a ected vein reveals lling of the lumen with an anechoic, hypoechoic, hyperechoic (if gas lled), or mixed echogenic material that partially or completely occludes the vessel. Ultrasonographic evaluation may also reveal thickening of the vessel wall, perivascular swelling, tracts extending from the vein to the subcutaneous tissues or skin, or subcutaneous abscesses. Cavitation in the center of a thrombus is suggestive of infection and represents a target for aspiration and bacterial culture. Stagnant or turbulent ow may be manifested by increased spontaneous venous contrast. Flow across the a ected zone may be evaluated by contrast venography using agitated saline as a contrast agent, with pulsed-wave or color-coded Doppler studies, or with newer vascular imaging technologies that directly visualize blood re ectors (e.g., RBCs) in a gray-scale display. Venous drainage proximal and distal to the area of thrombosis also should be assessed, because collateral circulation may still enter the vein distally. A careful evaluation of the communicating large veins and of the heart should be performed in horses with infective thrombophlebitis. Bacterial endocarditis may develop secondary to embolization of infective thrombus or bacteremia. Infections with multiple organisms should be anticipated in cases of infective thrombophlebitis, and high doses of broad spectrum antimicrobials should be administered until culture and sensitivity results of the aspirate are obtained. Metronidazole (15 mg/kg, PO every 6 hours, or 25 mg/kg, PO every 12 hours) should be considered if anaerobic infection is suspected. Flunixin meglumine (1 mg/kg every 12 hours) may be added to reduce in ammation. Local therapy including hot compresses and topical ichthammol (ammonium bituminosulfonate), NSAIDs (diclofenac sodium), heparinoids (chondroitin polysulfate), or dimethyl sulfoxide may be helpful in the treatment of thrombophlebitis. If antimicrobial therapy is unsuccessful, then surgical resection of the a ected tissue may be indicated. Recanalization of an a ected vein often occurs and can be documented by Doppler or contrast ultrasound studies demonstrating blood ow between the vessel wall and the thrombus. Occasionally, loculation may be detected within the ends of the thrombus. Such loculation may result in persistent brous webs within the vein that restrict venous return. Venous stricture also may occur because of longstanding thrombophlebitis. Complete brous occlusion of the jugular vein may impede venous return from the head such that if adequate collateral circulation does not develop, impaired venous drainage of the head and neck may limit performance. Vascular surgery may be required to regain venous drainage. Alternatively, an expert in interventional radiology or medicine can be consulted regarding other options, such as intravascular stenting.

Aortoiliac Thrombosis Aortoiliac thrombosis is an uncommon, but potentially serious, disorder. The gross appearance, histologic features, and clinical ndings of this vaso-occlusive disorder have been reviewed extensively, and the interested reader is directed to these and other related reports.∗ The terminal aorta and the iliac arteries, including their proximal branches, are the focus of involvement in this disease. Vessels become partially to completely occluded by multifocal ingrowths of brous tissue, laminated thrombi, or brous plaques.222 Histologic lesions

include organized, brous connective tissue, hemosiderin-laden macrophages, irregular vascular channels, and disruption of the intima and internal elastic lamina. The lesions are generally devoid of fat. Arteriosclerotic and atherosclerotic aortic lesions have also been observed at other sites as well (see Figure 10-50, A). Suggested causes of these arterial disorders include Strongylus vulgaris infection, systemic infections, embolization, and vasculitis. No convincing evidence indicates that aortoiliac thrombosis is consistently caused by Strongylus spp. infection, and Azzie224,226 has refuted this contention. Although aortoiliac thrombosis is uncommon, it can be associated with severe performance problems, including reproductive failure.690 The typical features include exercise-associated, typically unilateral, hindlimb lameness, ataxia, or collapse.222,224,226,227 Aortoiliac thrombosis has also been diagnosed as a cause of breeding failure in stallions.224,226,227 Physical examination of an a ected horse at rest may reveal weak metatarsal arterial pulses or delayed saphenous re ll in the a ected limb. The temperature of the limb in the resting animal is usually normal, unless complete arterial occlusion has occurred, in which case the limb is cold and painful and may become edematous. Exercise in a ected animals results in an exercise-associated gait abnormality (lameness, ataxia, or weakness), with a decreased or absent metatarsal and digital pulse and delayed or absent saphenous re ll. Claudication may cause the horse to become very uncomfortable and reluctant to bear weight on the a ected limb. Marked hyperpnea, other signs of distress, and profuse, generalized sweating are often present with trembling of the a ected limb. A rectal examination may reveal fremitus, a weak or absent pulse, or aneurysmal dilatation of the affected artery or arteries. These abnormalities may be more evident after exercise and may help to confirm the diagnosis. The diagnosis can be con rmed by ultrasonographic evaluation of the terminal aorta and iliac arteries (see Figure 10-51), using a highfrequency (5 or 10 MHz) rectal transducer,223,225,397 or by nuclear techniques.711,714 Doppler studies may indicate abnormal blood ow in the femoral arteries.715 Essential abnormalities include a hypoechoic to echogenic mass protruding into the arterial lumen. An estimation of the degree of obstruction can be made based on the percentage of the artery occluded. Many cases are longstanding, and hyperechoic areas, even tissues su ciently echo-dense to cast acoustic shadows, may be imaged within the aortic or iliac thrombus (these ndings suggest mature scar tissue and calcification). The prognosis for this disorder is guarded. No controlled studies have evaluated therapy for this condition. A case survey709 suggests no treatment consistently improves outcome, but potential bene t of surgical or catheter-based thrombectomy or anticoagulation should be considered. Various medical treatments have been reported including IV sodium gluconate, larvicidal dewormings, phenylbutazone, low molecular weight dextran, anticoagulants,713 and a controlled exercise program. Ultrasound-guided balloon thrombectomy712 has been successful in some cases but is probably most useful in cases of active thrombosis in which the obstruction can be partially relieved. Early diagnosis is essential if therapy is to be beneficial. Treatment is aimed at improving collateral circulation and preventing additional thrombus formation.

Parasitic Infection The migrating larval forms, particularly L4, of Strongylus vulgaris, are known causes of arterial disease, particularly involving the aorta and the cranial mesenteric artery and its branches (see Figure 10-52). Lesions have been described as far forward as the bulbous aorta and the aortic sinuses in infected horses. Rounded brous plaques and mural thrombi have been reported in the thoracic and cranial abdominal aorta in 9.4% of horses examined immediately after death by Cranley and McCullagh.106 These investigators reported a statistical association between the occurrence of proximal aortic S. vulgaris lesions and the presence of focal ischemic lesions in the myocardium. They hypothesized that these lesions were consequent to microembolism from parasitic lesions. With the advent of ivermectin and other new anthelmintics, as well as aggressive deworming programs currently recommended by practicing veterinarians, heavy S. vulgaris larval migration damage occurs much less frequently. However, migration of S. vulgaris L4 larvae must be considered in horses with poor deworming histories, high potential of exposure, high fecal egg counts, and palpable abnormalities on rectal examination, as well as when fremitus or aneurysmal dilatation of the aorta is found, particularly in the region of the cranial mesenteric artery or iliac system. An ultrasonographic evaluation of the cranial mesenteric artery is possible and can be used to con rm the diagnosis of thrombosis in these vessels.687,709 Treatment should consist of larvicidal deworming, combined with a rigorous individual and environmental parasite control program.

Arteriovenous Fistulas Arteriovenous stulas occur uncommonly and are most often detected with large vascular tumors (hemangiomas, hemangiosarcomas) or as a rare congenital defect or post-traumatic sequelae.686,716 Classic features include a continuous thrill and murmur over the a ected area, localized edema, slowing of the heart rate after occlusion of the shunt, and signs of increased CO if the shunt ow is great. Clinically important stulas are more likely to develop where large arteries and veins run in parallel, as with the jugular vein and carotid artery. A diagnosis can be made with Doppler ultrasound, including pulsed-wave and color- ow Doppler to demonstrate continuous and pulsatile ow from the artery, through the associated communication, into the distended veins. A complete CV examination should be performed

because a large arteriovenous communication could lead to cardiac failure. Treatment depends on the underlying cause, and an experienced surgeon should be consulted. No treatment may be indicated with small, uncomplicated arteriovenous communications. Large vascular neoplasms are usually inoperable at the time when they are diagnosed or complicated by the possibility of metastasis.

Arterial Rupture Rupture of the aorta or its branches or of the PA is reported sporadically. Rupture of the aortic root has been commonly reported in older breeding stallions but may also occur in older mares or in younger horses.58 Rupture typically involves the right aortic sinus of Valsalva (see Figures 10-49 and 10-50, B). The aortic root can rupture into the RA, RV, intraventricular septum, or pericardial space.58,152,220,221 Dystrophic changes in the media of the aorta have been reported, and the hypertension associated with breeding has been implicated as a cause of aortic root rupture. Aneurysms of the right sinus of Valsalva have also been reported and are related to congenital defects in the media of the aorta near the right aortic sinus. Rupture of the middle uterine artery can lead to fatal peripartum hemorrhage.218 Ruptures of the extrapericardial aorta or other blood vessels leading to fatal hemorrhage or a systemic to pulmonary shunt have been reported.52,57-60,62,64-68 Rupture of an artery is often a catastrophic event, and most diagnoses are made at the necropsy oor. Antemortem diagnosis of an aortic sinus of Valsalva aneurysm is possible by echocardiography. Aside from breeding, horses with diagnosed aortic aneurysms or aortocardiac stulas should not be used for athletic endeavor, because rupture of the aneurysm could occur at any time. When aortic root rupture or rupture of a sinus of Valsalva aneurysm communicates with the RA or RV, right-sided heart failure can occur. Rupture of the more distal aorta into the pulmonary artery also can occur.34 These types of stulas typically cause a continuous heart murmur. Cardiac arrhythmias (usually VT) may also develop, especially if dissection progresses into the intraventricular septum.220 The prognosis for life in these horses is generally guarded, depending on the site and size of the communication. Some horses survive, and other may require supportive treatment for CHF. Pulmonary artery rupture is most often caused by chronic pulmonary hypertension and PA dilatation. This is most commonly detected in conjunction with severe MR and pulmonary hypertension. The potential for PA rupture should be considered whenever a large dilated main PA or left and right PA is detected echocardiographically. A ected horses should be considered unsafe for riding or driving and should not be used, except for breeding, because of the possibility of sudden death. Focal medial calci cation of the PA is not uncommon at necropsy506,703 but is of uncertain clinical significance.

CARDIAC ARRHYTHMIAS Cardiac arrhythmias are disturbances in heart rate, rhythm, or conduction and can be classi ed based on atrial and ventricular rate, anatomic origin of the impulse, method of impulse formation, and conduction sequence. A wide variety of cardiac arrhythmias have been recognized.566,717 Of principal concern to the clinician are the hemodynamic consequences of arrhythmias (reduced pressure, ow, perfusion) and the potential for further electrical instability (myocardial fibrillation, sudden death). The electrophysiologic basis of cardiac arrhythmias is not discussed in this section; however, other sections of the chapter provide information regarding abnormal automaticity, re-entry, and other mechanisms of arrhythmogenesis. Cardiac arrhythmias are classi ed based on the anatomic origin of the manifest ECG mechanism (Box 10-9). The reader should be aware that some arrhythmias, especially those originating in the AV junction, may mimic either atrial rhythm disturbances or “high” ventricular rhythms. For purposes of this discussion, the authors have elected to distinguish atrial from junctional; however, the term supraventricular may also be applied to these rhythm disturbances. BOX 10-9 CARDIAC RHYTHMS OF THE HORSE Sinus Mechanisms∗ Normal sinus rhythm Sinus arrhythmia Sinoatrial block or arrest Sinus bradycardia Sinus tachycardia

Atrial Rhythm Disturbances

Atrial escape complexes† Atrial premature complexes Atrial tachycardia: nonsustained and sustained Atrial flutter Atrial fibrillation Reentrant supraventricular tachycardia

Junctional Rhythm Disturbances Junctional escape complexes† Junctional escape rhythm† (idionodal rhythm) Junctional premature complexes Junctional (“nodal”) tachycardia Re-entrant supraventricular tachycardia

Ventricular Rhythm Disturbances Ventricular escape complexes† Ventricular escape rhythm† (idioventricular rhythm) Ventricular premature complexes Accelerated ventricular rhythm (idioventricular tachycardia) Ventricular tachycardia Ventricular flutter Ventricular fibrillation

Conduction Disturbances Sinoatrial block∗ Atrial standstill (hyperkalemia) Atrioventricular blocks: first degree,∗ second degree,∗ third degree (complete) Ventricular pre-excitation † Escape complexes develop ∗

following another rhythm disturbance.

Generally physiologic.

The clinical evaluation of the horse with an arrhythmia is reviewed in Box 10-10. Routine electrocardiography, clinical chemistry, hematology, echocardiography, exercising electrocardiography, postexercise electrocardiography, and continuous 24-hour electrocardiographic (Holter) monitoring can all play a role in the complete evaluation of a horse suspected of having intermittent cardiac arrhythmias. These points are discussed further following. BOX 10-10 EVALUATION OF THE HORSE WITH A CARDIAC ARRHYTHMIA (Modified from Bonagura JD: Clinical evaluation and management of heart disease, Equine Vet Educ 2:31–37, 1990.)

General medical history, past illnesses, past and current medications Work history and exercise tolerance General physical examination Physical examination: evidence of congestive heart failure Subcutaneous edema Jugular venous distention and pulsations Pulmonary edema Pleural and peritoneal effusion Weight loss, poor condition Auscultation of the heart Assessment of heart sounds (rhythm, third heart sound) Assessment of heart murmurs Resting electrocardiogram Heart rate and cardiac rhythm Analysis of P-QRS-T waveforms Postexercise electrocardiogram Exercise (or treadmill) electrocardiogram Clinical laboratory studies Complete blood count (rule out anemia, infection) Blood culture (in cases of suspected endocarditis)

Serum electrolytes (particularly potassium and possibly magnesium) Serum biochemical tests (especially renal function) Skeletal muscle enzymes (cardiac isoenzymes) Serum troponin concentration Red blood cell potassium Fractional excretion of potassium Echocardiogram/Doppler echocardiogram Examine for cardiomegaly or reduced myocardial function Identify predisposing cardiac lesions and abnormal flow Serum/plasma concentrations of cardioactive drugs Serum digoxin concentration Plasma quinidine concentration Response to medications

Sinus Rhythms A number of physiologic sinus rhythms are recognized. These can be explained by the e ect of autonomic nervous system tra c on the SA node (Figure 10-53). Normal horses at rest demonstrate vagal-mediated sinus bradycardia, sinus arrhythmia, sinus block, and sinus arrest; however, fear or a sudden stimulus may provoke sympathetically driven sinus tachycardia. Exercise leads to pronounced sinus tachycardia, with heart rates often exceeding 200 beats/min. AV conduction, as a general rule, tends to follow sinus activity. During sinus tachycardia, the P-R interval shortens, whereas during periods of progressive sinus node slowing, the P-R interval generally prolongs. Mobitz type I seconddegree AV block (Wenckebach periodicity) often follows a progressive prolongation of the P-R interval. Sinus rate and AV conduction usually change in parallel; however, some horses appear to control BP during high sinus tachycardia by blocking impulses in the AV node. Seconddegree AV block is more likely in a standing horse, in a horse that has suddenly stopped submaximal exercise, or after a brief surge of sinus tachycardia in an anxious animal.

FIGURE 10-53

Sinus rhythms. A, Sinus bradycardia (top) and tachycardia (bottom). Base-apex lead recorded at 25 mm/ sec. Whereas two consecutively blocked P-waves are considered normal on a 24-hour ECG, these phenomenon are observed infrequently (in approximately 1% of normal horses). B, Sinus arrhythmia with sinus arrest/sinus block recorded at 25 mm/sec paper speed. C, Sinus arrhythmia with second-degree atrioventricular block (arrows). An ambulatory electrocardiogram (ECG) is shown; transthoracic leads recorded at 25 mm/ sec. Whereas two consecutively blocked Pwaves are considered normal on a 24-hour ECG, these phenomena are observed infrequently (in approximately 1% of normal horses).

Sinus rate and rhythm should be carefully monitored in the critically ill or anesthetized horse because heart rate is a major determinant of CO and ABP (see Figure 10-4, B). Sedative drugs and anesthetics can cause sinus bradycardia or sinus arrest. Anesthetic drugs, hypoxia, traction on an abdominal viscus, ocular manipulation, hypothermia, increased intracranial pressure, and hypertension can depress sinus node function. Concurrent depression of the sinus node with stimulation of latent pacemakers in the coronary sinus or AV junction can lead to

ectopic rhythms in the anesthetized horse. Conversely, an increasing sinus rate may indicate inadequate depth of anesthesia, pain, hypotension, hypovolemia, hypercarbia and hypoxemia, anemia, endotoxemia or sepsis, pyrexia, systemic in ammatory response syndrome (SIRS), anaphylaxis, or excessive catecholamine administration. Speci c therapy for sinus tachycardia is rarely required because it represents a physiologic response to stress. However, when sinus tachycardia is identified, the cause must be sought and treated as appropriate. Sinus bradycardia is generally a benign rhythm in standing horses; however, when encountered during sedation or anesthesia, this rhythm can lead to hypotension. Treatment of symptomatic sinus bradycardia can include the infusion of catecholamines (dobutamine, dopamine, epinephrine) or the administration of anticholinergic drugs (atropine, glycopyrrolate). Dopamine or dobutamine can be infused to increase heart rate, contractility, and ABP; however, re ex AV block may develop in some horses.209,213,236,294 Excessive administration of catecholamines causes sinus tachycardia and ectopic beats. Anticholinergic drugs may not be e ective in the setting of anesthetic-induced depression of SA function, particularly if vagal e erent tra c is low. Gastrointestinal complications including ileus and colic may develop after anticholinergic drug therapy; thus it should not be chosen for trivial rate problems.

Atrial Arrhythmias Rhythms originating in the atria are common. These disturbances often develop as functional disorders with no overt structural cardiac lesion. Autonomic imbalance (including high sympathetic activity or high vagal tone), hypokalemia, drugs (catecholamines, anesthetics), infections, fever, anemia, hypoxia, and colic can be associated with APCs. Atrial rhythm disturbances are also very common in the setting of structural lesions of the heart valves, myocardium, and pericardium. Cardiac conditions known to predispose to atrial arrhythmias include mitral or tricuspid insu ciency, endocarditis, myocarditis, cardiac (atrial) enlargement, myocardial brosis, and myocardial ischemia. APCs can precipitate sustained atrial arrhythmias, including atrial tachycardia, atrial utter, and AF. The large size of the equine atria and the frequent presence of microscopic atrial brotic lesions likely predispose the horse to these sustained arrhythmias. The high resting vagal tone present in most horses serves to shorten the duration of the action potential of atrial myocytes and also facilitates development of sustained atrial tachyarrhythmias, which probably depend on re-entry. Atrial arrhythmias are the most common abnormal rhythms detected in horses (Figures 10-54 and 10-55).∗ APCs are the least complex of these rhythm disturbances; these can be clinically insigni cant or associated with exercise intolerance or other signs of cardiac disease. In contrast, atrial tachycardia, atrial utter, and AF are more likely to become clinically important. The overall importance of APCs is often di cult to ascertain. Routine ECG rhythm strips recorded from over 950 horses and interpreted by one of the authors indicated that atrial arrhythmias, overall, were present in less than 3% of the horses studied. However, when clinically normal horses were examined by Holter monitor, APCs were recorded in 28% of the horses. Thus it appears that the incidence of atrial arrhythmias depends not only on the population examined but also on the methods used for identi cation. These rhythm disturbances should be assessed in light of other clinical findings.

FIGURE 10-54

Electrocardiograms (ECGs) with atrial rhythm disturbances(recorded at 25mm/sec). A, Premature atrial complexes (arrows). The premature P waves of

di erent morphology, the slightly prolonged P-R interval indicating atrioventricular nodal refractoriness, and the normal-appearing QRS complex indicating a supraventricular origin are notable. A slight conduction aberrancy is evidence in the second premature QRS complex (see also Figure 10-16). B, Premature atrial complexes occurring during the QT interval of the preceding complexes. The rst premature complex is nonconducted (arrow), the second premature complex is conducted with rst-degree AV block (arrowheads). C, Sustained atrial tachyarrhythmias. The top picture shows atrial tachycardia with rapid, regular P waves (P′) and a variable ventricular rate response to conducted P waves. The baseapex lead recorded at 25 mm/sec. The lower trace shows atrial flutter (F) with variable ventricular response (S). Lead 3 recorded at 25 mm/sec.

FIGURE 10-55

Atrial arrhythmias. A and B, Successful conversion of atrial brillation (AF) using a combination of digoxin and quinidine (see text for details). A, Incomplete conversion of AF to atrial tachycardia with rapid, regular atrial activity (arrows). B, Conversion to normal sinus rhythm during combination therapy. Notice the high amplitude of the second part of the P wave, indicating possible atrial enlargement. C, Quinidine toxicosis can be manifested as abnormalities on the electrocardiogram (ECG). In this horse, AF (top; base-apex lead recorded at 25 mm/sec) was not converted. The lower recording shows persistent AF with a rapid atrioventricular conduction response, electrical alternans, and widening of the QRS complex (lead 2 recorded at 25 mm/sec; right panel, at 50 mm/sec). The rapid rate response is related to the vagolytic e ect of quinidine. The electrical alternans is common with regular supraventricular tachycardias with rapid ventricular response rates; this nding indicates varying conduction into the ventricle. The widened QRS is a sign of quinidine toxicosis. An idiosyncratic reaction leading to polymorphous ventricular tachycardia (VT) also has been recognized (see Figure 10-59) . D, Hemodynamic consequences of increasing ventricular response rate in AF. The short QRS-QRS intervals do not generate e ective arterial pulsations recorded in the arterial blood pressure (ABP) tracing, which is related to inadequate ventricular filling time during short cycles. Pulse deficits are detectable by physical examination.

ATRIAL PREMATURE COMPLEXES Atrial premature complexes are detected by auscultation and are documented by an ECG. Auscultation usually reveals a regular sinus rhythm that is interrupted by an obviously premature beat or—in case of a blocked APC—by an abnormal pause. Premature atrial contractions may be fully interpolated (continuation of the regular sinus rhythm after the premature atrial contraction is noted), or a pause may occur (if the sinus node is reset or the premature complex is nonconducted). The ECG is characterized by a premature (usually narrow) QRS complex, which is preceded by an abnormal, premature P wave (P′ [P prime wave]) that is often buried within the preceding T wave. If the impulse arrives in the AV node before complete repolarization, then the P-R interval is longer than normal (physiologic rst-degree AV block). If the ectopic P′ wave is nonconducted (physiologic second-degree AV block), then a pause will be evident in the ventricular rhythm (see Figure 10-54). Premature atrial impulses can also be conducted aberrantly through the ventricle as a result of incomplete repolarization or persistent refractoriness of ventricular conducting tissues; this causes the QRS-T complex to be wider than normal or atypical in con guration. Care must be taken not to overdiagnose APCs. Sinus arrhythmia and sinus bradycardia often lead to variations in the P-P intervals, and often a “wandering atrial pacemaker” gradually alters the P wave morphology. Exercise or excitement will abolish such physiologic rhythms. APCs, as a general rule, are more likely to be clinically signi cant in the following circ*mstances: When they (1) are frequent at rest, (2) associated with runs of atrial tachycardia, (3) related to poor performance (if other causes are excluded), (4) precipitate paroxysmal atrial utter or brillation, or (5) develop in conjunction with structural cardiac disease. Documentation of atrial arrhythmias during exercise may be critical for determining if paroxysmal atrial tachycardia or brillation is likely to be the cause of poor performance.428 Clinical judgment must be used; however, because supraventricular premature complexes are most likely to occur in the immediate postexercise period, probably associated with autonomic imbalance. If these postexercise arrhythmias are not associated with clinical signs and are not detected during exercise, then they are unlikely to be clinically significant. Cardiac rhythm monitoring during exercise may be necessary to be certain.

Isolated APCs generally are not treated. The hemodynamic consequences of these rhythm disturbances are minor, unless sustained abnormal activity develops. Consideration of antiarrhythmic therapy might be appropriate if frequent atrial extrasystoles are documented to precipitate AF. Unfortunately, neither quinidine nor procainamide are practical for chronic use, and the effectiveness of other antiarrhythmics including digoxin has not been established for this indication. Maintenance of normal serum potassium and magnesium may be important in suppressing atrial ectopic activity. In cases of suspected myocarditis, anti-in ammatory doses of dexamethasone might be considered as empiric therapy with consideration of adverse effects of such treatment.

ATRIAL TACHYCARDIA Atrial tachycardia can be de ned as a series of ectopic APCs. Atrial tachycardia may be sustained or nonsustained (paroxysmal) and is generally precipitated by a single APC. The atrial rate is rapid and regular; however, because ectopic P′ waves can be blocked physiologically in the AV node, an irregular ventricular rate response often results (see Figure 10-54, C). Atrial rates of 120 to 300 beats/min are typical in horses with sustained atrial tachycardia. At lower atrial rates, 2:1 AV conduction may yield a regular, relentless heart rate. At the higher atrial rates, the rhythm may be indistinguishable from atrial utter. Di erentiation of atrial tachycardia from utter may not be critical because both arrhythmias seem to carry the same clinical signi cance and are treated in the same general way. Sustained atrial tachycardia is most often recognized during treatment of horses with quinidine sulfate. Before conversion of AF to sinus rhythm, atrial tachycardia may be observed (see Figure 10-55, A); thus in this setting, rhythm indicates a partial therapeutic e ect of quinidine on the atrial myocardium. When this atrial tachycardia occurs as an isolated nding, structural or underlying myocardial disease should be suspected and the horse should be considered predisposed to the development or recurrence of AF. Treatment is the same as that described for AF. If the ventricular response rate is especially rapid (>120 beats/min), then consideration can be given to administration of either digoxin to block AV nodal conduction or possibly to diltiazem if digoxin is ineffective (see Table 10-4). ABP should be monitored carefully if diltiazem is administered.243,732,733

ATRIAL FLUTTER Isolated atrial utter is probably less common than atrial tachycardia and represents a form of atrial circuit movement or macroreentry. The clinical circ*mstances and assessment of this rhythm disturbance are identical to those of atrial tachycardia. The ECG in atrial utter is characterized by a very rapid, abnormal, but regular atrial activity that is usually manifested as a “saw-toothed” ECG baseline. The atrial frequency often exceeds 300 beats/min. The RR intervals are usually irregular because of variable AV conduction, and fewer QRS and T complexes than utter waves exist (see Figure 10-54, C).The ventricular rate depends on the refractory period of the AV node and on the strength of the atrial stimulus. Atrial utter often alternates with AF in horses, creating so-called utter- brillation. Treatment is similar to that for AF.

Atrial Fibrillation Atrial brillation is the most common atrial arrhythmia associated with poor performance and exercise intolerance.∗ AF is most common in adult horses and has been reported infrequently in foals, weanlings, yearlings, or ponies. Several reports indicate a higher incidence in Standardbreds, draft horses, and Warmblood horses compared with the general hospital population. CO at rest is normal in most horses with AF546; however, maximal CO during exercise is limited because the atrial contribution to lling is most important at higher heart rates. As expected, exercise intolerance is most common in high-performance horses (e.g., racehorses, advanced combined training horses, polo ponies, and some Grand Prix jumpers) and less common in show hunters and in pleasure, dressage, and endurance horses. The high LA pressures present in heavily exercising horses240,553 is further exacerbated by the loss of active atrial transport function313 and may contribute to exercise intolerance or other associated signs. Exercise-induced pulmonary hemorrhage, respiratory distress, CHF, ataxia or collapse, and myopathy have all been reported with AF; conversely, the arrhythmia is often detected as an incidental finding in horses with no overt signs.363 AF can be acute or chronic, and it can be paroxysmal, persistent, or permanent. Paroxysmal AF is often associated with a single episode of poor performance, with the horse often decelerating suddenly during a race. The arrhythmia usually disappears spontaneously within 24 to 48 hours. Paroxysmal AF may be associated with transient potassium depletion, particularly in horses treated with furosemide or bicarbonate solutions, and it is most often unrelated to other clinical or echocardiographic abnormalities of cardiac disease (lone AF). Although less common than paroxysmal AF, both persistent and permanent AF are easier to diagnose (persistent AF terminates after treatment, whereas permanent AF is sustained and resistant to therapy). Many horses with sustained AF have no other evidence of signi cant underlying cardiac disease when examined by physical examination and echocardiography. However, ultrastructural and functional myocardial pathology may be present, predisposing to AF. Furthermore, AF can induce atrial electrical, structural, and functional remodeling that may be responsible for the self-perpetuating, progressive, and recurrent nature of AF.242,244,740 These remodeling changes are more likely to be reversible if AF is treated promptly. Recurrent episodes of AF are not uncommon and are more likely in the presence of

concurrent structural or functional cardiac disease. Horses with AF generally have a normal resting heart rate, although the rate decreases (and stroke volume tends to increase) after successful conversion of AF to normal sinus rhythm.741 The presence of resting tachycardia in a horse with AF should give the clinician pause to consider intercurrent cardiac lesions or a disorder that increases sympathetic tone such as pain, anemia, fever, or infection. A number of factors interact in the prognosis of a horse with AF. Of these, the presence of CHF represents the overall worst prognostic indicator and is invariably associated with resting tachycardia. Severe structural heart disease, recurrent bouts of AF, long-standing AF, and failure to convert with therapeutic serum concentrations of quinidine represent negative prognostic factors inasmuch as a ected horses can be more di cult to convert to sinus rhythm or more likely to revert back to AF even when therapy is successful. The diagnosis of AF is initiated by recognition of an irregularly irregular heart rhythm, variable intensity heart sounds, and an absent fourth heart sound during cardiac auscultation. Arterial pulses vary in intensity, and pulse de cits may be present, especially when the ventricular rate is high (see Figure 10-55, D). The ECG is characterized by an absence of P waves; instead, brillation or f waves are seen in the baseline. These f waves may be coarse (large) or ne (small), and the number of atrial impulses per minute cannot be easily counted but usually exceeds 500 per minute. The QRS-T complexes are normal in morphology and duration, but ventricular rate response is quite irregular, although periodicity may be observed infrequently.191 As for all atrial tachyarrhythmias, the ultimate ventricular rate response (and examination heart rate) depends on the refractory period of the AV node and the frequency and strength of the atrial stimuli. In the otherwise healthy horse in AF, vagal tone will be high and sympathetic tone low when standing; consequently, the ventricular rate will be close to normal or slightly increased. If sympathetic activity is increased for any reason, or if vagal activity is blocked (as occurs with quinidine sulfate therapy), then the ventricular rate response will increase as the AV nodal refractory period shortens. This explains the clinician’s simple, but very useful, dependence on measuring pretreatment, resting heart rate in horses with AF.363,574 Because the horse with structural heart disease will be more likely to require sympathetic support to maintain CO and ABP, persistent resting tachycardia, higher than 60 to 70 beats/min, indicates underlying cardiac disease and is associated with a poorer prognosis. Laboratory studies in horses with AF are usually normal. Infrequently, a horse is found to have low serum potassium, fractional excretion of potassium, or RBC potassium. Chest radiographs are usually normal unless there is concurrent pulmonary disease Thoracic radiography is not a high-yield procedure in horses without compatible signs of lung disease or structural heart disease. The echocardiogram is usually normal, unless concurrent valvular or ventricular myocardial disease is noted. It is not uncommon for an otherwise normal horse to demonstrate a slightly reduced LV shortening fraction (usually in the 24% to 32% range), which returns to normal once the horse has been converted to sinus rhythm.495,731 This decrease in fractional shortening is probably multifactorial in origin, probably related in part to decreased preload from loss of the atrial contribution to ventricular lling and irregular cardiac cycle. Doppler studies are useful in evaluating horses with concurrent valvular disease. The most common treatment for AF involves conversion to normal sinus rhythm, unless concurrent CHF is noted or the horse is aged and therapy is deemed of little bene t. Conversion of AF to sinus rhythm can be accomplished using a number of drugs or through electrocardioversion, generally delivered by intracardiac catheters.628,742-748 No prospective study has compared the two methods, and both quinidine and electrocardioversion are generally safe and e ective treatments. Quinidine sulfate is the standard drug against which other drug treatments should be compared, although this agent is becoming more di cult to obtain in some countries. Quinidine cardioversion of AF is successful in many horses, and a number of di erent treatment plans can be followed (see Table 10-6).319,717,739 The use of other drugs including flecainide,738,749-753 propafenone,754 and amiodarone755-758 may be considered in unresponsive cases or when quinidine cannot be tolerated or obtained (see following). As a general principle, when AF occurs in the setting of CHF, the treatments should be aimed at controlling CHF and resting heart rate with furosemide, digoxin, and possibly an ACE inhibitor, as opposed to conversion back to sinus rhythm. Based on the work of Reef, Levitan, and Spencer,574 an excellent prognosis for quinidine conversion (>95% conversion rate) may be given for horses with a heart rate of less than or equal to 60 beats/min, murmurs less than or equal to grade 3/6, and AF lasting less than 4 month. Recurrences a ect approximately 25% of these horses. Quinidine conversion to sinus rhythm represents the standard therapy for most equine practices and has been used for the longest period of time. In general the use of other drugs (e.g., ecainide, amiodarone) is considered mainly for cases of resistant or recurrent AF or in situations in which quinidine or electrocardioversion are unavailable. In a study of electrocardioversion, the results were also very positive with a greater than 98% conversion rate (71 of 72 episodes in 63 horses).747 Horses with longer duration of AF or signi cant structural cardiac disease may be more di cult to convert to sinus rhythm using quinidine or electrocardioversion, and they are more likely to have a higher recurrence rate independent of the treatment modality.

QUINIDINE THERAPY OF ATRIAL FIBRILLATION Quinidine is typically administered by nasogastric tube because of its irrigating e ects on the mucous membranes (see Table 10-6). Intravenous quinidine gluconate can be successful in conversion of horses with AF of recent onset or when nasogastric delivery is not

feasible,727 but failure to respond does not predict the response to oral treatment. Many successful approaches have been used to convert AF in horses, and the clinician should appreciate that the mean quinidine elimination half-life after an oral 10-g dose is about 6.7 hours.294,759 One treatment approach for AF involves administration of a loading dose of 22 mg/kg quinidine sulfate by nasogastric tube every 2 hours for two to four doses, followed by every 6 hour dosing until the horse converts or develops initial signs of toxicosis. Aggressive every-2-hours dosing exceeding 88 to 132 mg/kg is especially likely to induce adverse e ects. Therapeutic plasma quinidine concentrations for conversion from AF to sinus rhythm are 2 to 5 g/ml (6.2 to 15.4 mol/L) and should be measured if a horse fails to convert after an appropriate dosing regimen. If conversion to sinus rhythm has not occurred after 24 hours of therapy, then digoxin at 0.0055 to 0.011 mg/kg orally twice a day may be added to the treatment regimen for 24 to 48 hours. As in other species, a digoxin-quinidine interaction occurs that can e ectively double the serum concentration of digoxin.760 Thus combination therapy beyond 24 hours should be continued only with drug level monitoring of the serum digoxin (see Table 10-6) and consideration of using the lower end of the digoxin dosage range. Even horses that do not convert on the “standard” administration regimen may convert after the combined use of quinidine with digoxin. The value of using such a treatment plan every 6 hours is that steady-state levels are reached, su cient time elapses to attain myocardial concentrations, and quinidine toxicity is less frequent when compared with the every-2-hour regimen. When prolonged treatment over more than 12 to 24 hours is necessary, adequate hydration and electrolyte balance (particularly potassium and magnesium) must be ensured by oral or IV administration of crystalloid solutions, because most horses undergoing oral quinidine treatment will become depressed and inappetent, will show reduced water intake, and may develop mild diarrhea. Careful clinical and continuous electrocardiographic monitoring should be performed on horses with AF during conversion to sinus rhythm. The QRS duration should be monitored and compared with the pretreatment QRS duration before each additional treatment is administered. Prolongation of the QRS duration to greater than 25% of the pretreatment value is an indication of quinidine toxicity and should prompt discontinuation of therapy. The simplest ECG change is an acceleration of AV nodal conduction related to the vagolytic e ect of quinidine (see Figure 10-55, C). Rapid supraventricular tachycardias with ventricular rate responses of up to 300 beats/min have been seen in several horses receiving quinidine sulfate. These horses have been treated with IV digoxin at 0.002 mg/kg to slow the ventricular response rate, IV replacement uids to improve perfusion, IV sodium bicarbonate at 1 mEq/kg to reverse the sodium channel blocking e ects of quinidine (probably by a combined e ect of increasing extracellular sodium concentration and alkalinization, leading to increased protein binding of quinidine and decreased extracellular concentrations of ionized calcium), and if needed, a phenylephrine drip to restore BP if critical hypotension develops. Ventricular arrhythmias (torsades de pointes, multiform VT, and ventricular premature complexes) have also been detected with quinidine toxicosis (Figure10-56). Intravenous sodium bicarbonate is also indicated in these horses to bind freecirculating quinidine, whereas IV magnesium sulfate (up to 25 g in a 450- to 500-kg horse) is the treatment of choice for quinidine-induced ventricular arrhythmias (see Table 10-6). Lidocaine HCl may also be used if needed, starting with an IV bolus of 0.5 to 1.5 mg/kg, IV slowly. Conversely, administration of digoxin is contraindicated in horses with ventricular arrhythmias induced by quinidine.

FIGURE 10-56

Polymorphous ventricular tachycardia (VT) after administration of quinidine sulfate for attempted conversion of atrial brillation (AF). In this case, normal sinus rhythm was established (lower tracing) after treatment with lidocaine, bicarbonate (to reverse sodium channel–blocking e ects of quinidine), and phenylephrine (administered to maintain arterial blood pressure [ABP]).

Clinical markers of quinidine toxicosis include ataxia, colic, and nasal edema causing upper respiratory tract stridor. Most toxic reactions

to quinidine sulfate are associated with higher plasma concentrations of the drug (>5 g/ml) and can be avoided with careful clinical and electrocardiographic monitoring. Some of the adverse e ects of quinidine administration, particularly polymorphic VT, are likely an idiosyncratic reaction. These adverse e ects should prompt discontinuation of therapy or altering treatment intervals. Depression and paraphimosis occur in most horses treated with quinidine but disappear with the discontinuation of the drug. Diarrhea often develops with the administration of higher doses of quinidine, but this sign also disappears with discontinuation of the drug. Convulsions, hypotension because of vasodilation, CHF, laminitis, urticaria, and sudden death have rarely been reported with quinidine sulfate administration. A ventricular rate response in excess of 100 beats/min is an adverse reaction that may be more common in nervous horses or those prone to the vagolytic e ects of quinidine. These horses may bene t from administration of digoxin to blunt the ventricular rate response, particularly if no other signs of quinidine toxicosis or elevated plasma concentration are present. If an increased ventricular response rate is anticipated, then one may even consider pretreatment with digoxin before initiation of quinidine therapy. Alternatively, diltiazem could be used for ventricular rate control, provided that BP can be closely monitored (see Table 10-6). However, the clinical experience with the use of diltiazem is limited, and doses should be carefully titrated to effect.

OTHER DRUGS FOR ATRIAL FIBRILLATION Other drugs have been studied recently for treatment of AF in horses. Amiodarone administered as a constant-rate infusion has been used successfully for treating experimentally induced and natural AF in horses, but it cannot be recommended as a rst-line therapy at this time.755,757,758 Flecainide, a class Ic antiarrhythmic drug, may become more widely used if quinidine becomes unavailable. Although IV administration was generally ine ective at a dose of 2 to 3 mg/kg,753 oral administration has demonstrated some success. Pharmaco*kinetics of ecainide have been studied in horses, and it has been successfully administered for treatment of AF, including recurrent AF.738,749-753 The drug can be dissolved in water and delivered orally or by nasogastric tube. The approximate dosage is 4 to 6 mg/kg PO at intervals of 2 to 4 hours. Like quinidine, ecainide can prolong the QRS complex and become proarrhythmic, so the ECG must be carefully monitored from the rst dose. Appreciation of the overall adverse e ect pro le of the drug in horses will require more clinical experience. IV propafenone also has been suggested for conversion of chronic AF,754,761 but it has recently been shown to be ine ective at a dose of 2 mg/kg IV in horses with naturally occurring (and with pacing-induced) AF. Therefore the current knowledge does not support the use of this class Ic antiarrhythmic drug for treatment of AF.

ELECTROCARDIOVERSION OF ATRIAL FIBRILLATION Electrical cardioversion of AF to sinus rhythm has been used at a number of referral centers as either the primary method of treatment or for management of horses that respond adversely or inadequately to quinidine therapy. As stated previously, electrocardioversion is very e ective therapy, especially for AF of recent onset, but it requires special equipment and trained personnel. The procedure involves percutaneous placement of two specialized electrode catheters transvenously, with one catheter tip located within the left PA and the other within the RA cavity. Catheter placement is done under sedation and local anesthesia. The catheters are guided by pressure monitoring through the catheter lumen and 2D echocardiographic imaging (Figure 10-57). Radiography is used to verify the placement of the catheters either before or preferably after induction of general anesthesia. Although the catheters can be placed under local anesthesia, electrical cardioversion is very painful and must be performed under general anesthesia with the horse well padded because the shock results in a sudden jolt of the body and limbs.

FIGURE 10-57

A, Electrocardioversion procedure in a Thoroughbred horse. The horse is under general anesthesia. Two specialized electrode catheters (arrows) were previously placed percutaneosly under local anesthesia into the right atrium (RA) and left pulmonary artery under guidance from two-dimensional (2D) echocardiography and intravascular pressure monitoring. The biphasic cardioverter (top, center of image) is connected to the catheters and is used to deliver the synchronized cardioversion shock, as well as monitor electrocardiogram (ECG) rhythm, invasive blood pressure (BP), and pulse oximetry. Inset: 2D image showing short axis image at base of heart. The catheter inserted into the left pulmonary artery (LPA) is indicated (arrows) (RA, right atrium; RV, right ventricle; Ao, aorta; RPA, right pulmonary artery). B, Transvenous electrical cardioversion for treatment of atrial brillation (AF) in an 2-year-old Standardbred racehorse under general anesthesia. A surface ECG (25 mm/sec) and an arterial blood pressure (ABP) tracing are displayed. The QRS complexes are automatically detected by the de brillator unit and marked by small triangles. Biphasic electrical shocks (larger triangles on top) are applied at increasing energy levels. Delivery of the shocks is synchronized to the QRS complex to avoid the vulnerable period (T wave) and prevent induction of ventricular arrhythmias. (A) Unsuccessful attempt at an energy level of 125J. (B) Successful cardioversion at an energy level of 225J. Immediately after the shock, the baseline attens and normal sinus rhythm resumes. No further treatment is required at this point. (From Schwarzwald CC, Bonagura JD, Muir WW: The cardiovascular system. In Muir WW, Hubbell JAE, editors: Equine anesthesia: monitoring and emergency therapy, ed 2, St Louis, 2009, WB Saunders).

The speci c procedure used for cardioversion is beyond the scope of this textbook but has been well described in several In these reports, cardioversion was achieved with a mean energy level of 160 J. Experience indicates that some horses require higher energy doses, sometimes exceeding 300 J, and as with quinidine, not all cases result in conversion. The complications associated with electrocardioversion appear to be quite low but are nite related to general anesthesia or electrical shock. A case of transient complete AV block has been reported.762 studies.743-747

FOLLOW-UP CARE Horses can usually be returned to training within 24 to 48 hours after conversion. However, persistent sinus tachycardia, recurrent APCs (as detected by follow-up 24-hour Holter ECG), or persistent LV or LA dysfunction (as detected by follow-up echocardiography)740 observed after successful conversion of AF to normal sinus rhythm may indicate subtle myocardial disease or AF-induced atrial remodeling. In addition, it may potentially indicate a higher risk of recurrent AF (although this contention has not been proven so far). These horses should probably be given more time to recover if possible (i.e., 1 to 2 weeks). Conversion of horses with AF generally results in a return to their previous performance level.363,574 The exception is the horse with persistent LV dysfunction whose functional status will be improved following conversion to sinus rhythm, but whose performance will be less than its previous best. Horses with repeated episodes of AF are often converted numerous times with quinidine sulfate or electrocardioversion. Most horses that experience a recurrence of AF do so within 1 year of initial conversion, but much longer intervals have occurred between episodes of AF in some horses. If the duration of sinus rhythm becomes shorter, then repeated treatments may no longer be practical and a career change may be indicated. Some horses eventually become refractory to drug or electrical cardioversion, probably because of progressive atrial brosis or underlying myocardial disease. However, in the absence of signi cant, detectable cardiac disease, horses with persistent or permanent AF may still be used at lower levels of exercise (as long as the expected level of performance can be achieved), or they can be kept as breeding animals or pasture horses.

Junctional and Ventricular Arrhythmias Cardiac arrhythmias that originate within the AV conducting tissues, the ventricular specialized conducting tissues, or myocardium are

classi ed as junctional (AV nodal or bundle of His) or ventricular in origin. Unlike sinus or atrial arrhythmias, these arrhythmias are not preceded by a conducted P wave. When sustained, junctional and ventricular rhythms often lead to dissociation between the SA (P wave) activity and that of the ventricle (QRS-T), resulting in AV dissociation.447 In these cases an independent atrial rhythm is superimposed on the ectopic rhythm (see Figure 10-58, C and D). AV dissociation in these cases develops because the premature AV junctional or ventricular depolarization causes interference to the conduction of normal SA impulses. It is important to note that escape rhythms (see following) also cause AV dissociation. Therefore AV dissociation is a purely descriptive term of an ECG nding and neither characterizes the type and pathophysiologic mechanism of the arrhythmia nor determines the therapeutic approach.

FIGURE 10-58

Ventricular arrhythmias. A, The ectopic complex (arrow) is premature and abnormal in morphology. A compensatory pause follows the extrasystole because the next sinus P wave is blocked in the atrioventricular node. The e ect of the premature complex on arterial blood pressure (ABP) in the lower tracing (arrow) is noticeable. B, Base-apex lead electrocardiogram (ECG) recorded from a 15-year-old Arab mare with ventricular bigeminy. Normal sinus beats alternate with slightly larger and wider ventricular ectopic beats. SA node discharge is not a ected by the ectopic beats, as indicated by the presence of nonconducted P waves immediately before the ectopic beats (arrowheads) (paper speed 25 mm/sec). C, Base-apex lead ECG recorded from an 18-year-old Arab mare recovering from acute diarrhea and endotoxemia. The ECG shows an intermittent accelerated idioventricular rhythm at a rate of 50 beats/min. P wave intervals are indicated (arrowheads). The recording demonstrates that the ectopic focus is suppressed at higher rates of SA node discharge. The ventricular rhythm only becomes manifest when the sinoatrial (SA) rate drops below the rate of the ventricular pacemaker. SA node discharge is not a ected by the ectopic rhythm, resulting in AV dissociation. A fusion beat is present (arrow), resulting from summation of a conducted sinus impulse with an ectopic ventricular beat (paper speed 25 mm/sec, voltage calibration 0.5 cm/mV). D, Base-apex lead ECG recorded from a 3-year-old Clydesdale gelding. The top recording shows a regular tachycardia at a rate of 120 beats/min. The appearance of the QRS-T complexes does not allow conclusive distinction between a supraventricular rhythm with rapid ventricular response and a ventricular rhythm. However, as the rate slows down (bottom strip), AV dissociation because of a ventricular tachycardia (VT) becomes apparent. P waves (arrowheads) and a capture beat (arrow) are indicated (paper speed 25 mm/sec, voltage calibration 0.25 cm/mV). E, Base-apex ECG recorded from a 5-year-old Clydesdale stallion with acute myocardial necrosis of unknown cause. The serum cardiac troponin I concentrations were severely elevated (404 ng/ml; normal 120 beats/min) may lead to CHF because of reversible myocardial failure (tachycardia-induced cardiomyopathy).110 VTs can also progress to ventricular utter and ventricular fibrillation, characterized by chaotic ventricular activation patterns leading to uncoordinated undulations of the electrical baseline. These rhythms often represent terminal rhythms and usually result in cardiac arrest (see Figure 10-59). VT may be life threatening if R on T complexes are detected, if the arrhythmia is rapid (>180-200 beats/min), if multiform VT is detected, or if polymorphic tachycardia (torsades de pointes) is evident. Immediate treatment of CV collapse may be required.717 IV therapies for VT include lidocaine, procainamide, quinidine magnesium salts, amiodarone, and propafenone (for refractory VT). Administration of antiarrhythmic drugs (see Table 10-6) can be associated with side e ects, including proarrhythmia, seizures, and sudden death. Lidocaine causes central nervous system (CNS) excitation; however, boluses and infusions can be well tolerated in horses, and the drug causes minimal hemodynamic or electrophysiologic alterations at nontoxic doses.623,624,772 Quinidine may be used for treatment of ventricular arrhythmias but is a myocardial depressant and vagolytic drug. Procainamide773 has been studied to a limited degree, but it is generally well tolerated when given in graded doses. It has similar electrophysiologic e ects to quinidine, but seems less vagolytic and possibly has less gastrointestinal side e ects. Cumulative doses of up to 10 mg/kg are generally well tolerated; additional dosing should be guided by ECG, QRS duration, and ABP. Magnesium sulfate is an alternative antiarrhythmic that can be e ective in both normomagnesemic and hypomagnesemic horses and is generally well tolerated at therapeutic doses. Amiodarone has been evaluated in a small number of horses756,757 and can be considered as a second-line drug for serious, drug-resistant VT. Other antiarrhythmics, including IV propafenone761

or phenytoin may be tried in refractory ventricular arrhythmias. A period of rest (4 to 8 weeks) followed by electrocardiographic and echocardiographic re-evaluation (including an exercising ECG) is indicated for follow-up of horses that have developed sustained VT. A substantial number of these horses can be successfully returned to performance after treatment with antiarrhythmics and rest, so long as signi cant and untreatable structural heart disease is not evident. The use of anti-inflammatory drugs in horses with VT is empiric.

Conduction Disturbances Once a cardiac electrical impulse is formed, it is conducted rapidly throughout the heart. The sequence of cardiac electric activation is usually dictated by the specialized conducting tissues in the atria, the AV node, the bundle of His, the bundle branches, and the Purkinje ber system. This conduction system permits orderly activation of atrial and ventricular muscle and facilitates e ective mechanical activity of the heart. A variety of conduction disorders are recognized, including SA nodal exit block, atrial standstill (usually because of hyperkalemia), AV block, bundle branch block, and ill-defined ventricular conduction disturbances (Figures 10-60 and 10-61).∗

FIGURE 10-61

Conduction disturbances. A, Ventricular pre-excitation. The short P-R interval (arrow) and small de ection in the P-R segment (delta wave) indicate an accessory pathway around the atrioventricular node with premature activation of a portion of the ventricle (delta wave and initial portion of the QRS). The large QRS and secondary T wave changes can be explained by the loss of normal cancellation of ventricular electrical forces. B, A sinus arrhythmia and intraventricular conduction disturbance. The sudden change and widening of the ventricular conduction pattern are notable, despite the consistent P-R interval and probably represent a bundle branch block, although this diagnosis is difficult to make in horses (base-apex lead recorded at 25 mm/sec). C, Hyperkalemia in a foal with a ruptured bladder. The top strip demonstrates atrial standstill (no P waves), signi cant widening of the QRS complexes, and large T waves with a shortened ST segment. The lower tracing shows the e ects after initial medical therapy for hyperkalemia, which included treatment with saline and sodium bicarbonate. The normalization of the QRS complexes and the appearance of low-amplitude P waves (arrows) are notable. (Tracing courtesy of Ron Hilwig, DVM, PhD.)

Rarely, accelerated conduction occurs in the heart, which involves a pathway around the normally slow-conducting AV node and results in early excitation of the ventricles.775 These syndromes are termed pre-excitation and have various associated labels related to similar human disorders (e.g., Wolff-Parkinson-White syndrome). A shortened P-R interval is typically found.

ATRIOVENTRICULAR CONDUCTION BLOCK Delays in AV conduction are the most common conduction blocks. These delays are classi ed as rst-, second-, and third-degree (or complete). First-degree AV block occurs when the PR (or PQ) interval exceeds a certain value (approximately 500 ms in large-breed horses and 320 ms in small breeds and ponies) while the atrial impulse still transmits through the AV conduction system and activates the ventricle, causing a QRS complex. Some P waves are not conducted to the ventricles during second-degree AV block, which results in occasional P waves not followed by a QRS-T complex (see Figure 10-60, A). Second-degree AV block after progressive prolongation of the PQ interval is classi ed as Mobitz type I (Wenckebach) block. Conversely, the AV block is termed Mobitz type II if the PQ interval is constant. Occurrence of two or more consecutive second-degree AV blocks in presence of a normal or slow sinoatrial rate is called high-grade (advanced) AV

block, although this can be a normal (vagally induced) nding in approximately 1% of healthy horses, based on 24-hour ECG ndings of one of the authors (VBR). Third-degree or complete AV block is characterized by an absence of atrial to ventricular conduction. P waves are not followed by or related to QRS complexes, and to prevent ventricular asystole, a junctional or ventricular escape rhythm must develop below the level of the AV block (see Figure 10-60, B). First- and second-degree AV block are considered normal variations.764 These rhythms are most often associated with high vagal tone and may be seen with sinus bradycardia or sinus arrhythmia. Second-degree AV block can usually be abolished with exercise, stress, or vagolytic drugs such as atropine or glycopyrrolate (although drugs are rarely needed). Complete AV block is indicative of organic heart disease, severe drug toxicity, or abnormally high vagal activity. Generally speaking, third-degree AV block caused by vagal-e erent tra c is short-lived. Sudden development of high-grade or complete AV block may require the administration of atropine or a catecholamine, or it may require rapid transvenous insertion of a pacing wire into the RV. Chronic third-degree AV block requires treatment with a pacemaker;507,776-779 and pacemakers have been successfully placed epicardially and transvenously in several horses and in a number of Miniature Donkeys with possible congenital complete AV block (see Figure 10-60, C, and 10-62).

FIGURE 10-62

A lateral radiograph of a permanent transvenous pacing system in a Miniature Donkey. The pacemaker is evident at the upper left of the radiograph. The thin transvenous pacing wire extends from the device through the jugular vein and vena cava and into the right ventricular (RV) apex.

INTRAVENTRICULAR CONDUCTION BLOCKS Intraventricular conduction blocks, such as bundle branch block, are less common and more di cult to diagnose. Widening of the QRS complex and axis deviation are typical features. These abnormalities may also be found after APCs, from overdose with quinidine sulfate, from severe hyperkalemia, or secondary to supraventricular tachycardias with rapid ventricular response.

PRE-EXCITATION Pre-excitation, or accelerated AV conduction, has been reported sporadically,775 but the clinical signi cance of this electrocardiographic abnormality has yet to be determined in this species. Ventricular pre-excitation in humans and in dogs is often due to an anomalous conducting pathway around the AV node, which serves as a path for re-entrant supraventricular tachycardias. These rhythm disturbances may cause hypotension and syncope. Whether pre-excitation syndromes are important has yet to be determined. Nevertheless, ECG traces occasionally show evidence of ventricular pre-excitation and are characterized by a P-QRS-T relationship but with an extremely short PR interval, early excitation of the ventricle characterized by a slurring of the initial QRS complex (a delta wave), and an overall widening of the QRS complex (see Figure10-61, A).

HYPERKALEMIA Hyperkalemia can cause signi cant depression of atrial, AV, and ventricular conduction and can shorten ventricular repolarization. Serum potassium is most likely to be markedly elevated after renal failure and oliguria, during shock, with severe metabolic acidosis, in foals with ruptured bladder or ureter and uroperitoneum, or after excessive IV potassium replacement. Hyperkalemia also occurs in Quarter Horses with hyperkalemic periodic paralysis. Changes in the ECG are usually evident at potassium serum concentrations greater than 6 mEq/L, with severe changes evident when serum concentrations are between 8 and 10 mEq/L.198,780-782 Broadening and attening of the P wave are the most consistently observed change. Prolongation of the PQ interval and bradycardia develop, excitability decreases, and atrial standstill (sinoventricular rhythm)— characterized by complete absence of P waves—may be observed. Either inversion or enlargement (tenting) of the T waves is also likely. Marked widening of the QRS complex may be noted as near-lethal concentrations of potassium are approached. Ventricular asystole or brillation can develop. The QT interval is not a reliable indicator of induced hyperkalemia, and other electrolyte and acid-base alterations,

including serum calcium and sodium, influence the effect of hyperkalemia on the heart. Therapy for hyperkalemia includes correction of the underlying problem, infusion of isotonic crystalloid solutions and dextrose, and administration of calcium salts and sodium bicarbonate. Lactated Ringer’s solution (LRS) can be used in patients with hyperkalemia; the small amount of potassium contained in LRS is generally not considered problematic, because the bene cial e ects of volume replacement and treatment of metabolic acidosis largely overweigh the potential negative e ects of additional potassium supply. Alternatively, sodium chloride (NaCl) or NaCl/dextrose solutions may be used, but maintenance solutions should not be used because of their relatively high potassium content. Dextrose (or glucose) can be administered as a bolus of 0.25 to 0.5 g/kg, IV over 15 minutes. The administration of regular insulin (0.1 IU/kg) in conjunction with dextrose infusion (0.5 to 1.0 g/kg, IV over 15 minutes) may be considered in cases of lifethreatening hyperkalemia. Sodium bicarbonate may be used in cases of severe metabolic acidosis (provided that ventilation is adequate) at a dose of 1 to 2 mEq/kg, IV over 15 minutes (or depending on the base excess obtained by blood gas analyses [mEq sodium bicarbonate = 0.3 (−0.5) × BE × kg BWT]). In cases of severe arrhythmias, 1 ml/kg calcium gluconate 23% can be administered over a 10-minute period to effect.

REFERENCES 1. Detweiler D.K. Electrocardiogram of the horse. Fed Proc. 1952;11:34. 2. Detweiler D.K. Auricular fibrillation in horses. J Am Vet Med Assoc. 1955;126:47. 3. Patterson D.F., Detweiler D.K., Glendinning S.A. Heart sounds and murmurs of the normal horse. Ann N Y Acad Sci. 1965;127:242. 4. Detweiler D.K., Patterson D.F. The cardiovascular system. In: Catcott E.J., Smithcors J.F., editors. Equine medicine and surgery. ed 2. Santa Barbara, CA: American Veterinary Publications; 1972:277-347. 5. Bishop S.P., Cole C.R., Smetzer D.L. Functional and morphologic pathology of equine aortic insufficiency. Vet Pathol. 1966;3:137. 6. Hamlin R.L., Smetzer D.L., Smith C.R. Analysis of QRS complex recorded through a semiorthogonal lead system in the horse. Am J Physiol. 1964;207:325. 7. Hamlin R.L., Smith C.R. Categorization of common domestic mammals based on their ventricular activation process. Ann N Y Acad Sci. 1965;127:195. 8. Hamlin R.L., Smetzer D.L., Senta T., et al. Atrial activation paths and P waves in horses. Am J Physiol. 1970;219:306. 9. Hamlin R.L., Klepinger W.L., Gilpin K.W., et al. Autonomic control of heart rate in the horse. Am J Physiol. 1972;222:976. 1 0 . Hamlin R.L., Levesque M.J., Kittleson M.D. Intramyocardial pressure and distribution of coronary blood ow during systole and diastole in the horse. Cardiovasc Res. 1982;16:256. 11. Illera J.C., Illera M., Hamlin R.L. Unipolar thoracic electrocardiography that induces QRS complexes of relative uniformity from male horses. Am J Vet Res. 1987;48:1700. 12. McKeever K.H., Hinchcliff K.W., Reed S.M., et al. Splenectomy alters blood pressure response to incremental treadmill exercise in horses. Am J Physiol. 1993;265:R409. 13. Pipers F.S., Hamlin R.L. Echocardiography in the horse. J Am Vet Med Assoc. 1977;170:815. 14. Pipers F.S., Hamlin R.L., Reef V.B. Echocardiographic detection of cardiovascular lesions in the horse. J Equine Med Surg. 1979;3:68. 15. Senta T., Smetzer D.L., Smith C.R. Effects of exercise on certain electrocardiographic parameters. Cornell Vet. 1970;60:552. 16. Smetzer D.L., Smith C.R., Hamlin R.L. The fourth heart sound in the equine. Ann N Y Acad Sci. 1965;127:306. 17. Smetzer D.L., Smith C.R. Diastolic heart sounds of horses. J Am Vet Med Assoc. 1965;146:937. 18. Smetzer D.L., Bishop S.P., Smith C.R. Diastolic murmur of equine aortic insufficiency. Am Heart J. 1966;72:489. 19. Smetzer D.L., Hamlin R.L., Smith C.R. Cardiovascular sounds. In: Swenson M.J., editor. Dukes’ physiology of domestic animals. ed 8. Ithaca, NY: Comstock Publishing Company Inc; 1970:159-168. 20. Smith C.R., Hamlin R.L. Regulation of the heart and blood vessels. In: Swenson M.J., editor. Dukes’ physiology of domestic animals. ed 8. Ithaca, NY: Comstock Publishing Company Inc; 1970:169-195. 21. Brown C.M., Holmes J.R. Haemodynamics in the horse. II. Intracardiac, pulmonary arterial, and aortic pressures. Equine Vet J. 1978;10:207. 22. Brown C.M., Holmes J.R. Haemodynamics in the horse. I. Pressure pulse contours. Equine Vet J. 1978;10:188. 23. Brown C.M., Holmes J.R. Phonocardiography in the horse. I. The intracardiac phonocardiogram. Equine Vet J. 1979;11:11. 24. Brown C.M., Holmes J.R. Assessment of myocardial function in the horse. II. Experimental findings in resting horses. Equine Vet J. 1979;11:248. 25. Brown C.M., Holmes J.R. Phonocardiography in the horse. II. The relationship of the external phonocardiogram to intracardiac pressure and sound. Equine Vet J. 1979;11:183. 26. Brown C.M., Holmes J.R. Assessment of myocardial function in the horse. I. Theoretical and technical considerations. Equine Vet J. 1979;11:244. 27. Else R.W., Holmes J.R. Cardiac pathology in the horse. III. Clinical correlations. Equine Vet J. 1972;4:195. 28. Else R.W., Holmes J.R. Cardiac pathology in the horse. I. Gross pathology. Equine Vet J. 1972;4:1. 29. Else R.W., Holmes J.R. Cardiac pathology in the horse. II. Microscopic pathology. Equine Vet J. 1972;4:57. 30. Gattland L., Holmes J.R. ECG recording at rest and during exercise in the horse. Equine Vet Educ. 1990;2:28. 31. Hillyer M.H., Mair T.S., Holmes J.R. Treatment of bacterial endocarditis in a Shire mare. Equine Vet Educ. 1990;2:5. 32. Holmes J.R. The equine heart: problems and difficulties in assessing cardiac function on clinical examination. Equine Vet J. 1968;1:10. 33. Holmes J.R., Darke P.G.G., Else R.W. Atrial fibrillation in the horse. Equine Vet J. 1969;1:212. 34. Holmes J.R., Rezakhani A., Else R.W. Rupture of a dissecting aortic aneurysm into the left pulmonary artery in a horse. Equine Vet J. 1973;5:65. 35. Holmes J.R., Rezakhani A. Observations on the T wave of the equine electrocardiogram. Equine Vet J. 1975;7:55. 36. Holmes J.R. Prognosis of equine cardiac conditions. Equine Vet J. 1977;9:181. 37. Holmes J.R. Sir Frederick Smith Memorial Lecture. A superb transport system—the circulation. Equine Vet J. 1982;14:267. 38. Holmes J.R., Miller P.J. Three cases of ruptured mitral valve chordae in the horse. Equine Vet J. 1984;16:125. 39. Holmes J.R., Henigan M., Williams R.B., et al. Paroxysmal atrial fibrillation in racehorses. Equine Vet J. 1986;18:37. 40. Holmes J.R. Electrocardiography in the diagnosis of common cardiac arrythmias in the horse. Equine Vet Educ. 1990;2:24.

41. Holmes J.R. The development of clinical cardiology. Equine Vet J (Suppl 19). 2, September 1995. 42. Miller P.J., Holmes J.R. Effect of cardiac arrhythmia on left ventricular and aortic blood pressure parameters in the horse. Res Vet Sci. 1983;35:190. 43. Miller P.J., Holmes J.R. Beat-to-beat variability in QRS potentials recorded with an orthogonal lead system in horses with second-degree partial A-V block. Res Vet Sci. 1984;37:334. 44. Miller P.J., Holmes J.R. Interrelationship of some electrocardiogram amplitudes, time intervals and respiration in the horse. Res Vet Sci. 1984;36:370. 45. Miller P.J., Holmes J.R. Observations on structure and function of the equine mitral valve. Equine Vet J. 1984;16:457. 46. Miller P.J., Holmes J.R. Relationships of left side systolic time intervals to beat-by-beat heart rate and blood pressure variables in some cardiac arrhythmias of the horse. Res Vet Sci. 1984;37:18. 47. Miller P.J., Holmes J.R. Computer processing of transaortic valve blood pressures in the horse using the rst derivative of the left ventricular pressure trace. Equine Vet J. 1984;16:210. 48. Miller P.J., Holmes J.R. Observations on seven cases of mitral insufficiency in the horse. Equine Vet J. 1985;17:181. 49. Voros K., Holmes J.R., Gibbs C. Left ventricular volume determination in the horse by two-dimensional echocardiography: an in vitro study. Equine Vet J. 1990;22:398. 50. Voros K., Holmes J.R., Gibbs C. Anatomical validation of two-dimensional echocardiography in the horse. Equine Vet J. 1990;22:392. 51. Voros K., Holmes J.R., Gibbs C. Measurement of cardiac dimensions with two-dimensional echocardiography in the living horse. Equine Vet J. 1991;23:461. 52. Baker J.R., Ellis C.E. A survey of post-mortem findings in 480 horses 1958 to 1980. I. Causes of death. Equine Vet J. 1981;13:43. 53. Pipers F.S. Applications of diagnostic ultrasound in veterinary medicine. Equine Vet J. 1982;14:341. 54. Morris E.A., Seeherman H.J. Clinical evaluation of poor performance in the racehorse: the results of 275 evaluations. Equine Vet J. 1991;23:169. 55. Mitten L.A. Cardiovascular causes of exercise intolerance. Vet Clin North Am Equine Pract. 1996;12:473. 56. Reef V.B. Clinical approach to poor performance in horses. Proc Am Coll Vet Internal Med San Diego. 1989:566. 57. Cronin M.T.L., Leader G.H. Coronary occlusion in a Thoroughbred colt. Vet Rec. 1952;64:8. 58. Rooney J.R., Prickett M.E., Crowe M.W. Aortic ring rupture in stallions. Vet Pathol. 1967;4:268. 59. Pascoe R.R., O’Sullivan B.M. Sudden death in a Thoroughbred stallion. Equine Vet J. 1980;12:211. 60. Platt H. Sudden and unexpected deaths in horses: a review of 69 cases. Br Vet J. 1982;138:417. 61. Hughes P.E., Howard E.B. Endocardial fibroelastosis as a cause of sudden death in the horse. Vet Clin North Am Equine Pract. 1984;6:23. 62. Gelberg H.B., Zachary J.F., Everitt J.I., et al. Sudden death in training and racing Thoroughbred horses. J Am Vet Med Assoc. 1985;187:1354. 63. Kiryu I., Nakamura T., Kaneko M., et al. Cardiopathology of sudden death in the racehorse. Heart Vessels Suppl. 1987;2:40. 64. Lucke V.M. Sudden death. Equine Vet J. 1987;19:85. 65. Brown C.M., Kaneene J.B., Taylor R.F. Sudden and unexpected death in horses and ponies: an analysis of 200 cases. Equine Vet J. 1988;20:99. 66. Allen J.R., Heidel J.R., Hodgson D.R., et al. Spontaneous rupture of the great coronary vein in a pony. Equine Vet J. 1987;19:145. 67. Leblond A., Villard I., Leblond L., et al. A retrospective evaluation of the causes of death of 448 insured French horses in 1995. Vet Res Commun. 2000;24:85. 68. Schiff P., Knottenbelt D.C. Sudden death in a 11-year-old Thoroughbred stallion. Equine Vet Educ. 1990;2:8. 69. Boden L.A., Charles J.A., Slocombe R.F., et al. Sudden death in racing Thoroughbreds in Victoria, Australia. Equine Vet J. 2005;37:269. 70. Sisson S., Grossman J.D. Anatomy of the domestic animals. Philadelphia: WB Saunders, 1953. 71. Geddes L.A., Hoff H.E., McCrady J.D. Some aspects of the cardiovascular physiology of the horse. Cardiovasc Res Cent Bull. 1965;3:80. 72. Miller M.S., Bonagura J.D. Genesis of the equine electrocardiogram and indications for electrocardiography in clinical practice. J Equine Vet Sci. 1985;5:23. 73. Guyton A.C. The circulation textbook of medical physiology, ed 7. Philadelphia: WB Saunders; 1986. pp 205-346 74. Jones W.E. Equine sports medicine. Philadelphia: Lea & Febiger, 1988. 75. Shepherd J.T., Vanhoutte P.M. The human cardiovascular system. Lippincott: Williams, and Wilkins, 1992. 76. Opie L.H. Heart physiology: from cell to circulation, 2004. 77. Schwarzwald C.C., Bonagura J.D., Muir W.W.3rd. The cardiovascular system. In: Muir W.W.3rd, Hubbell J.A.E., editors. Equine anesthesia: monitoring and emergency therapy. ed 2. St Louis: WB Saunders; 2009:37-100. 78. Muir W.W.3rd. Cardiopulmonary resuscitation and the prevention of hypotensive emergencies in horses. Proc Am Assoc Equine Pract. 1984;30:117. 79. Muir W.W.3rd. Anesthetic complications and cardiopulmonary resuscitation in the horse. In: Muir W.W., Hubbell J.A.E., editors. Equine anesthesia. St Louis: Mosby Year Book; 1992:461-484. 80. Rainey J.W. A specific arthritis with pericarditis affecting young horses in Tasmania. Aust Vet J. 1944;20:204. 81. Ryan A.F., Rainey J.W. A specific arthritis with pericarditis affecting horses in Tasmania. Aust Vet J. 1945;21:146. 82. Wagner P., Miller R., Merritt F., et al. Constrictive pericarditis in the horse. J Equine Med Surg. 1977;1:242. 83. Reef V.B. Advances in echocardiography. Vet Clin North Am Equine Pract. 1991;7:435. 84. Rantanen N.W. Diseases of the heart. Vet Clin North Am Equine Pract. 1986;2:33. 85. Marr C.M. Equine echocardiography—sound advice at the heart of the matter. Br Vet J. 1994;150:527. 86. Hardy J., Robertson J.T., Reed S.M. Constrictive pericarditis in a mare: attempted treatment by partial pericardiectomy. Equine Vet J. 1992;24:151. 87. Freestone J.F., Thomas W.P., Carlson G.P., et al. Idiopathic effusive pericarditis with tamponade in the horse. Equine Vet J. 1987;19:38. 88. Dill S.G., Simoncini D.C., Bolton G.R., et al. Fibrinous pericarditis in the horse. J Am Vet Med Assoc. 1982;180:266. 89. Carnine B.L., Schneider G., Cook J.E., et al. Pericardial mesothelioma in a horse. Vet Pathol. 1977;14:513. 90. Bonagura J.D., Herring D.S., Welker F. Echocardiography. Vet Clin North Am Equine Pract. 1985;1:311. 91. Bonagura J.D., Blissitt K.J. Echocardiography. Equine Vet J. 1995;19(Suppl 1):5. 92. Birks E.K., Hultgren B.D. Pericardial haemangiosarcoma in a horse. J Comp Pathol. 1988;99:105. 93. Wijnberg I.D., Vink-Nooteboom M. Sloet van Oldruitenborgh-Oosterbaan MM: Idiopathic pericardial e usion with tamponade in a Friesian gelding. Tijdschr Diergeneeskd. 1997;122:216. 94. Worth L.T., Reef V.B. Pericarditis in horses: 18 cases (1986-1995). J Am Vet Med Assoc. 1998;212:248.

95. Bolin D.C., Donahue J.M., Vickers M.L., et al. Microbiologic and pathologic findings in an epidemic of equine pericarditis. J Vet Diagn Invest. 2005;17:38. 96. Perkins S.L., Magdesian K.G., Thomas W.P., et al. Pericarditis and pleuritis caused by Corynebacterium pseudotuberculosis in a horse. J Am Vet Med Assoc. 2004;224:1133. 97. Seahorn J.L., Slovis N.M., Reimer J.M., et al. Case-control study of factors associated with brinous pericarditis among horses in central Kentucky during spring 2001. J Am Vet Med Assoc. 2003;223:832. 98. Baker D., Kreeger J. Infiltrative lipoma in the heart of a horse. Cornell Vet. 1987;77:258. 99. Dill S.G., Moise N.S., Meschter C.L. Cardiac failure in a stallion secondary to metastasis of an anaplastic pulmonary carcinoma. Equine Vet J. 1986;18:414. 100. Byars T.D., Dainis C.M., Seltzer K.L., et al. Cranial thoracic masses in the horse: a sequel to pleuropneumonia. Equine Vet J. 1991;23:22. 101. Evans D.L. Cardiac responses to exercise and training. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:32-46. 102. Buhl R., Ersboll A.K., Eriksen L., et al. Changes over time in echocardiographic measurements in young Standardbred racehorses undergoing training and racing and association with racing performance. J Am Vet Med Assoc. 2005;226:1881. 103. Rugh K.S., Garner H.E., Sprouse R.F., et al. Left ventricular hypertrophy in chronically hypertensive ponies. Lab Anim Sci. 1987;37:335. 104. Braunwald E., Sonnenblick E.H., Ross J.Jr. Mechanisms of cardiac contraction and relaxation. In Libby P., Bonow R., Zipes D., et al, editors: Braunwald’s heart disease: a textbook of cardiovascular medicine, ed 8, Philadelphia: WB Saunders, 2008. 105. Braunwald E. Pathophysiology of heart failure. In Libby P., Bonow R., Zipes D., et al, editors: Braunwald’s heart disease: a textbook of cardiovascular medicine, ed 8, Philadelphia: WB Saunders, 2008. 106. Cranley J.J., McCullagh K.G. Ischaemic myocardial fibrosis and aortic strongylosis in the horse. Equine Vet J. 1981;13:35. 107. Dudan F., Rossi G.L., Luginbuhl H. Cardiovascular study of the horse: relationships between vascular and tissue lesions in the myocardium. II. Schweiz Arch Tierheilkd. 1984;126:527. 108. Dudan F., Rossi G.L., Luginbuhl H. Cardiovascular study in the horse: relationship between vascular and myocardial lesions. III. Schweiz Arch Tierheilkd. 1985;127:319. 109. Reef V.B., Bain F.T., Spencer P.A. Severe mitral regurgitation in horses: clinical, echocardiographic and pathological findings. Equine Vet J. 1998;30:18. 110. Traub-Dargatz J.L., Schlipf J.W.Jr., Boon J.A., et al. Ventricular tachycardia and myocardial dysfunction in a horse. J Am Vet Med Assoc. 1994;205:1569. 111. Fregin G.F. The cardiovascular system. In Mansmann R.A., McCallister E.S., Pratt P.W., editors: Equine medicine and surgery, ed 3, Santa Barbara, CA: American Veterinary Publications Inc, 1982. 112. Bonagura J.D. Equine heart disease. An overview. Vet Clin North Am Equine Pract. 1985;1:267. 113. Koblik P.D., Hornof W.J. Diagnostic radiology and nuclear cardiology. Their use in assessment of equine cardiovascular disease. Vet Clin North Am Equine Pract. 1985;1:289. 114. Doonan G.R., Brown C.M., Mullaney T.P., et al. Monensin poisoning in horses—an international incident. Can Vet J. 1989;30:165. 115. Mollenhauer H.H., Rowe L.D., Witzel D.A. Effect of monensin on the morphology of mitochondria in rodent and equine striated muscle. Vet Hum Toxicol. 1984;26:15. 116. Amend J.F., Mallon F.M., Wren W.B., et al. Equine monensin toxicosis: some experimental clinicopathologic observations. Compend Cont Educ Pract Vet. 1980;2:S172. 117. Amend J.F., Nichelson R.L., Freeland L.R., et al. Clinical toxicology of an antibiotic ionophore (monensin) in ponies and horses: diagnostic markers and therapeutic considerations. In: Comparative veterinary pharmacology, toxicology and therapy. Lancaster, UK: MTP Press Ltd; 1986:381-389. 118. Bezerra P.S., Driemeier D., Loretti A.P., et al. Monensin poisoning in Brazilian horses. Vet Hum Toxicol. 1999;41:383. 119. Bila C.G., Perreira C.L., Gruys E. Accidental monensin toxicosis in horses in Mozambique. J S Afr Vet Assoc. 2001;72:163. 120. Hughes K.J., Hoffmann K.L., Hodgson D.R. Long-term assessment of horses and ponies post exposure to monensin sodium in commercial feed. Equine Vet J. 2009;41:47. 121. Sweeney R.W., Hamir A.N., Fisher R.R. Lymphosarcoma with urinary bladder infiltration in a horse. J Am Vet Med Assoc. 1991;199:1177. 122. Reef V.B. Stress echocardiography and its role in performance assessment. Vet Clin North Am Equine Pract. 2001;17:179. 123. Bowen I.M., Marr C.M., Chester A.H., et al. In-vitro contraction of the equine aortic valve. J Heart Valve Dis. 2004;13:593. 124. Deprez P., Sustronck B., Vanroy M., et al. A case of mitral-valve insufficiency due to a ruptured chorda tendinea in a horse. Vlaams Diergeneesk Tijdschr. 1993;62:180. 125. Brown C.M. Acquired cardiovascular disease. Vet Clin North Am Equine Pract. 1985;1:371. 126. Critchley K.L. An interventricular septal defect, pulmonary stenosis and bucuspid pulmonary valve in a Welsh pony foal. Equine Vet J. 1976;8:176. 127. Clark E.S., Reef V.B., Sweeney C., et al. Aortic valve insufficiency in a one-year-old colt. J Am Vet Med Assoc. 1987;191:841. 128. Stadler P., Weinberger T., Kinkel N., et al. B-mode-, M-mode- and Dopplersonographic findings in mitral valve insufficiency (MVI) in horses. Am J Vet Med. 1992;39:704. 129. Blissitt K.J., Bonagura J.D. Colour flow Doppler echocardiography in normal horses. Equine Vet J Suppl. 1995;19(Suppl 1):5. 130. Blissitt K.J., Bonagura J.D. Colour flow Doppler echocardiography in horses with cardiac murmurs. Equine Vet J Suppl. 82, 1995. 131. Ecke P., Malik R., Kannegieter N.J. Common atrioventricular canal in a foal. N Z Vet J. 1991;39:97. 132. Marr C.M., Love S., Pirie H.M., et al. Confirmation by Doppler echocardiography of valvular regurgitation in a horse with a ruptured chorda tendinea of the mitral valve. Vet Rec. 1990;127:376. 133. Amada A. Diagnosis and treatment of valvular disease of the heart in the horse. Advances in Animal Cardiology. 1987;20:7. 134. Mill J., Hanak J. Diagnosis of heart valve defects, arrhythmia and functional disorders of cardiac conduction in competition horses and racehorses. Arch Exp Veterinarmed. 1985;39:319. 135. Nilsfors L., Lombard C.W., Weckner D., et al. Diagnosis of pulmonary valve endocarditis in a horse. Equine Vet J. 1991;23:479. 136. Yamaga Y., Too K. Diagnostic ultrasound imaging in domestic animals: two-dimensional and M-mode echocardiography. Nippon Juigaku Zasshi. 1984;46:493. 137. Reimer J.M., Reef V.B., Sommer M. Echocardiographic detection of pulmonic valve rupture in a horse with right-sided heart failure. J Am Vet Med Assoc. 1991;198:880. 138. Reef V.B., Spencer P.A. Echocardiographic evaluation of equine aortic insufficiency. Am J Vet Res. 1987;48:904. 139. Bonagura J.D., Pipers F.S. Echocardiographic features of aortic valve endocarditis in a dog, a cow, and a horse. J Am Vet Med Assoc. 1983;182:595. 140. Wingfield W.E., Miller C.W., Voss J.L., et al. Echocardiography in assessing mitral valve motion in 3 horses with atrial fibrillation. Equine Vet J. 1980;12:181. 141. Stadler P, Deegen E: Echocardiography in horses, I. Mitral valve insuffi ciency. II. Aortic valve insuffi ciency. Wiesbaden Proceedings, 1990, p. 11. 142. Stadler P., Hoch M., Fruhauf B., et al. Echocardiography in horses with and without heart murmurs in aortic regurgitation. Pferdeheilkunde. 1995;11:373. 143. Stadler P., Hoch M., Radu I. Echocardiography in the horse with special regard to color-flow Doppler technique. Prakt Tierarzt. 1995;76:1015. 144. Buergelt C.D., Cooley A.J., Hines S.A., et al. Endocarditis in six horses. Vet Pathol. 1985;22:333. 145. Reef V.B. Heart murmurs in horses: determining their significance with echocardiography. Equine Vet J. 1995;19(Suppl):71. 146. Brumbaugh G.W., Thomas W.P., Hodge T.G. Medical management of congestive heart failure in a horse. J Am Vet Med Assoc. 1982;180:878.

147. Reef V.B. Mitral valvular insufficiency associated with ruptured chordae tendineae in three foals. J Am Vet Med Assoc. 1987;191:329. 148. Travers C.W., van den Berg J.S. Pseudomonas spp. associated vegetative endocarditis in two horses. J S Afr Vet Assoc. 1995;66:172. 149. Reef V.B., Lalezari K., deBoo J., et al. Pulsed-wave Doppler evaluation of intracardiac blood flow in 30 clinically normal Standardbred horses. Am J Vet Res. 1989;50:75. 150. Stadler P., Weinberger T., Deegen E. Pulsed Doppler-echocardiography in healthy warm-blooded horses. J Vet Med A Physiol Pathol Clin Med. 1993;40:757. 151. Wilson R.B., Haffner J.C. Right atrioventricular atresia and ventricular septal defect in a foal. Cornell Vet. 1987;77:187. 152. Roby K.A.W., Reef V.B., Shaw D.P., et al. Rupture of an aortic sinus aneurysm in a 15-year-old broodmare. J Am Vet Med Assoc. 1986;189:305. 153. Habermehl K.H., Schmack K.H. The topography of the heart valves in horses, cattle and dogs. Anat Histol Embryol. 1986;15:240. 154. Lescure F., Tamzali Y. TM echocardiography in the horse. II. Acquired valvular heart disease. L’echocardiographie TM chez le cheval. II. Les cardiopathies ori cielles acquises. Point Veterinaire. 1983;15:9. 155. Rooney J.R., Franks W.C. Congenital cardiac anomalies in horses. Vet Pathol. 1964;1:454. 156. Bayly W.M., Reed S.M., Leathers C.W., et al. Multiple congenital heart anomalies in five Arabian foals. J Am Vet Med Assoc. 1982;181:684. 157. McClure J.J., Gaber C.E., Watters J.W., et al. Complete transposition of the great arteries with ventricular septal defect and pulmonary stenosis in a Thoroughbred foal. Equine Vet J. 1983;15:377. 158. Tadmor A., Fischel R., Tov A.S. A condition resembling hypoplastic left heart syndrome in a foal. Equine Vet J. 1983;15:175. 159. Musselman E.E., LoGuidice R.J. Hypoplastic left ventricular syndrome in a foal. J Am Vet Med Assoc. 1984;185:542. 160. Crowe M.W., Swerczek T.W. Equine congenital defects. Am J Vet Res. 1985;46:353. 161. Physick-Sheard P.W., Maxie M.G., Palmer N.C., et al. Atrial septal defect of the persistent ostium primum type with hypoplastic right ventricle in a Welsh pony foal. Can J Comp Med. 1985;49:429. 162. Reef V.B. Cardiovascular disease in the equine neonate. Vet Clin North Am Equine Pract. 1985;1:117. 163. Reef V.B., Mann P. Echocardiographic diagnosis of tricuspid atresia in 2 foals. J Am Vet Med Assoc. 1987;191:225. 164. Sojka J.E. Persistent truncus arteriosus in a foal. Vet Clin North Am Equine Pract. 1987;9:19. 165. Zamora C.S., Vitums A., Nyrop K.A., et al. Atresia of the right atrioventricular ori ce with complete transposition of the great arteries in a horse. Anat Histol Embryol. 1989;18:177. 166. Leipold H.W., Saperstein G., Woollen N.E. Congenital defects in foals. In: Smith B.P., editor. Large animal internal medicine. St Louis: CV Mosby; 1990:137-146. 167. Cargile J., Lombard C.W., Wilson J.H., et al. Tetralogy of Fallot and segmental uterine aplasia in a three-year-old Morgan filly. Cornell Vet. 1991;81:411. 168. Hinchcliff K.W., Adams W.M. Critical pulmonary stenosis in a newborn foal. Equine Vet J. 1991;23:318. 169. Innes J.M., Berger J., Francis J. Subacute bacterial endocarditis with pulmonary embolism in a horse. Br Vet J. 1950;106:245. 170. McCormick B.S., Peet R.L., Downes K. Erysipelothrix rhusiopathiae vegetative endocarditis in a horse. Aust Vet J. 1985;62:392. 171. Wagenaar G., Kroneman J., Breukink H. Endocardtis in the horse. Blue Book Vet Prof. 1986;12:38. 172. Dedrick P., Reef V.B., Sweeney R.W., et al. Treatment of bacterial endocarditis in a horse. J Am Vet Med Assoc. 1988;193:339. 173. Hamir A.N., Reef V.B. Complications of a permanent transvenous pacing catheter in a horse. J Comp Pathol. 1989;101:317. 174. Collatos C., Clark E.S., Reef V.B., et al. Septicemia, atrial brillation, cardiomegaly, left atrial mass, and Rhodococcus equi septic osteoarthritis in a foal. J Am Vet Med Assoc. 1990;197:1039. 175. Ewart S., Brown C., Derksen F.J., et al. Serratia marcescens endocarditis in a horse. J Am Vet Med Assoc. 1992;200:961. 176. Ball M.A., Weldon A.D. Vegetative endocarditis in an Appaloosa gelding. Cornell Vet. 1992;82:301. 177. Pace L.W., Wirth N.R., Foss R.R., et al. Endocarditis and pulmonary aspergillosis in a horse. J Vet Diagn Invest. 1994;6:504. 178. deGroot J., Sloet van Oldruitenborgh-Oosterbaan M.M., van der Linde Sipman J.S., et al. Heart diseases in foals. A literature review exemplified by 2 case reports. Tijdschrift voor Diergeneeskunde. 1996;121:382. 179. Maxson A.D., Reef V.B. Bacterial endocarditis in horses: ten cases. Equine Vet J. 1984-1995;29(394):1997. 180. Church S., Harrigan K.E., Irving A.E., et al. Endocarditis caused by Pasteurella caballi in a horse. Aust Vet J. 1998;76:528. 181. Little D., Keene B.W., Bruton C., et al. Percutaneous retrieval of a jugular catheter fragment from the pulmonary artery of a foal. J Am Vet Med Assoc. 2002;220:212. 182. Patteson M.W., Cripps P.J. A survey of cardiac auscultatory findings in horses. Equine Vet J. 1993;25:409. 183. Patteson M.W., Blissitt K.J. Evaluation of cardiac murmurs in horses. I. Clinical examination. In Pract. 1996;18:367. 184. Buhl R., Ersboll A.K. Effect of light exercise on valvular regurgitation in Standardbred trotters. Equine Vet J. 2006;36(Suppl):178. 185. Lightfoot G., Jose-Cunilleras E., Rogers K., et al, An echocardiographic and auscultation study of right heart responses to training in young national hunt thoroughbred horses. Equine Vet J, Suppl 36, August 2006, 153-158. 186. Brown C.M., Bell T.G., Paradis M.R., et al. Rupture of mitral chordae tendineae in two horses. J Am Vet Med Assoc. 1983;182:281. 187. Kiryu K., Kaneko M., Kanemaru T., et al. Cardiopathology of sinoatrial block in horses. Nippon Juigaku Zasshi. 1985;47:45. 188. Matsui K., Sugano S., Amada A. Heart rate and ECG response to twitching in Thoroughbred foals and mares. Nippon Juigaku Zasshi. 1986;48:305. 189. Matsui K., Sugano S. Relation of intrinsic heart rate and autonomic nervous tone to resting heart rate in the young and the adult of various domestic animals. Nippon Juigaku Zasshi. 1989;51:29. 190. Moore E.N., Spear J.F. Electrophysiological studies on atrial fibrillation. Heart Vessels (Suppl. 1987;2:32). 191. Meijler F.L., Kroneman J., van der Tweel I., et al. Nonrandom ventricular rhythm in horses with atrial fibrillation and its significance for patients. J Am Coll Cardiol. 1984;4:316. 192. Meijler F.L. Atrioventricular conduction versus heart size from mouse to whale. J Am Coll Cardiol. 1985;5:363. 193. Meijler F.L., van der Tweel I. Comparative study of atrial fibrillation and AV conduction in mammals. Heart Vessels (Suppl. 1987;2:24). 194. Raekallio M. Long-term ECG recording with Holter monitoring in clinically healthy horses. Acta Vet Scand. 1992;33:71. 195. Tschudi P. Electrocardiography in the horse. II. Disorders of impulse formation and impulse conduction. Tierarztl Prax. 1985;13:529. 196. Littlewort M.C.G. The equine heart in health and disease. In: Hickman J., editor. Equine surgery and medicine. London: Academic Press, 1986. 197. Reef V.B. Heart murmurs irregularities and other cardiac abnormalities. In: Brown C., editor. Problems in equine medicine. New York: Lea & Febiger, 1992. 198. Bonagura J.D., Miller M.S. Common conduction disturbances [ECG in the horse]. J Equine Vet Sci. 1986;6:23. 199. Reef V.B. Twenty-four hour rhythm monitoring. In: Mayhew I., editor. Equine medicine and surgery IV. ed 4. Santa Barbara, CA: American Veterinary Publications; 1991:213-214.

200. Slinker B.K., Campbell K.B., Alexander J.E., et al. Arterial baroreflex control of heart rate in the horse, pig. calf, Am J Vet Res. 1926;43:1982. 201. Meijler F.L. Atrial fibrillation: a new look at an old arrhythmia. J Am Coll Cardiol. 1983;2:391. 202. Horn J., Bailey S., Berhane Y., et al. Density and binding characteristics of beta-adrenoceptors in the normal and failing equine myocardium. Equine Vet J. 2002;34:411. 203. Badino P., Odore R., Re G. Are so many adrenergic receptor subtypes really present in domestic animal tissues? A pharmacological perspective. Vet J. 2005;170:163. 204. Evans D.L., Rose R.J. Determination and repeatability of maximum oxygen uptake and other cardiorespiratory measurements in the exercising horse. Equine Vet J. 1988;20:94. 205. Evans D.L., Rose R.J. Cardiovascular and respiratory responses in Thoroughbred horses during treadmill exercise. J Exp Biol. 1988;134:397. 206. Landgren G.L., Gillespie J.R., Fedde M.R., et al. o 2 transport in the horse during rest and exercise. Adv Exp Med Biol. 1988;227:333. 207. Littlejohn A., Snow D.H. Circulatory, respiratory and metabolic responses in Thoroughbred horses during the first 400 meters of exercise. Eur J Appl Physiol. 1988;58:307. 208. Marsland W.P. Heart rate response to submaximal exercise in the Standardbred horse. J Appl Physiol. 1968;24:98. 209. Donaldson L.L. Retrospective assessment of dobutamine therapy for hypotension in anesthetized horses. Vet Surg. 1988;17:53. 210. Swanson C.R., Muir W.W.3rd, Bednarski R.M., et al. Hemodynamic responses in halothane-anesthetized horses given infusions of dopamine or dobutamine. Am J Vet Res. 1985;46:365. 211. Trim C.M., Moore J.N., White N.A. Cardiopulmonary effects of dopamine hydrochloride in anaesthetized horses. Equine Vet J. 1985;17:41. 212. Whitton D.L., Trim C.M. Use of dopamine hydrochloride during general anesthesia in the treatment of advanced atrioventricular heart block in four foals. J Am Vet Med Assoc. 1985;187:1357. 213. Hinchcli K.W., McKeever K.H., Muir W.W.3rd. Hemodynamic e ects of atropine, dobutamine, nitroprusside, phenylephrine, and propranolol in conscious horses. J Vet Intern Med. 1991;5:80. 214. Hardy J., Bednarski R.M., Biller D.S. Effect of phenylephrine on hemodynamics and splenic dimensions in horses. Am J Vet Res. 1994;55:1570. 215. Benamou A.E., Marlin D.J., Lekeux P. Equine pulmonary and systemic haemodynamic responses to endothelin-1 and a selective ET(A) receptor antagonist. Equine Vet J. 2001;33:337. 216. Gasthuys F., deMoor A., Parmentier D. Cardiovascular e ects of low dose calcium chloride infusions during halothane anaesthesia in dorsally recumbent ventilated ponies. Zentralbl Veterinarmed A. 1991;38:728. 217. Schwarzwald C.C., Bonagura J.D., Luis-Fuentes V. Effects of diltiazem on hemodynamic variables and ventricular function in healthy horses. J Vet Intern Med. 2005;19:703. 218. Rooney J.R. Internal hemorrhage related to gestation in the mare. Cornell Vet. 1964;54:11. 219. van der Linde Sipman J.S., Kroneman J., Meulenaar H., et al. Necrosis and rupture of the aorta and pulmonary trunk in four horses. Vet Pathol. 1985;22:51. 220. Marr C.M., Reef V.B., Brazil T.J., et al. Aorto-cardiac fistulas in seven horses. Vet Radiol Ultrasound. 1998;39:22. 221. Sleeper M.M., Durando M.M., Miller M., et al. Aortic root disease in four horses. J Am Vet Med Assoc. 2001;219:491. 222. Maxie M.G., Physick-Sheard P.W. Aortic-iliac thrombosis in horses. Vet Pathol. 1985;22:238. 223. Reef V.B., Roby K.A.W., Richardson D.W. Use of ultrasonography for the detection of aortic-iliac thrombosis in horses. J Am Vet Med Assoc. 1987;190:286. 224. Azzie M.A.J. Clinical diagnosis of equine aortic iliac thrombosis and its histopathology as compared with that of the strongyle aneurysm. Proc Am Assoc Equine Pract San Francisco. 1972;43:43. 225. Edwards G.B., Allen W.E. Aorto-iliac thrombosis in two horses: clinical course of the disease and use of real-time ultrasonography to con rm diagnosis. Equine Vet J. 1988;384:1988. 226. Azzie M.A.J. Aortic/iliac thrombosis of Thoroughbred horses. Equine Vet J. 1969;1:113. 227. Physick-Sheard P.W., Maxie M.G. Aortoiliofemoral arteriosclerosis. In: Robinson N.E., editor. Current therapy in equine medicine. ed 1. Phildelphia: WB Saunders; 1983:153-155. 228. Reef V.B. Vasculitis in Robinson NE. In: Current Therapy in Equine Medicine II. Philadelphia: WB Saunders 1987; 1992:312-314. 229. Elliott J. Control of Cardiovascular Function: Physiology and Pharmacology. In: Marr C.M., editor. Cardiology of the Horse. London: WB Saunders; 1999:15-31. 230. Blissitt K.J., Bonagura J.D., Pulsed wave Doppler echocardiography in normal horses. Equine Vet J, Suppl 19, 1995, 18. 231. Patteson M.W., Gibbs C., Wotton P.R., et al. E ects of sedation with detomidine hydrochloride on echocardiographic measurements of cardiac dimensions and indices of cardiac function in horses. Equine Vet J Suppl. 33, 1995. 232. Raisis A.L., Young L.E., Blissitt K.J., et al. A comparison of the haemodynamic effects of isoflurane and halothane anaesthesia in horses. Equine Vet J. 2000;32:318. 233. Young L.E., Blissitt K.J., Clutton R.E., et al, Feasibility of transoesophageal echocardiography for evaluation of left ventricular performance in anaesthetised horses. Equine Vet J, Suppl 19, 1992, 225. 234. Young L.E., Blissitt K.J., Clutton R.E., et al. Temporal effects of an infusion of dopexamine hydrochloride in horses anesthetized with halothane. Am J Vet Res. 1997;58:516. 235. Young L.E., Blissitt K.J., Clutton R.E., et al. Temporal effects of an infusion of dobutamine hydrochloride in horses anesthetized with halothane. Am J Vet Res. 1998;59:1027. 236. Young L.E., Blissitt K.J., Clutton R.E., et al. Haemodynamic e ects of a sixty minute infusion of dopamine hydrochloride in horses anaesthetised with halothane. Equine Vet J. 1998;30:310. 237. Lightowler C., Piccione G., Fazio F., et al. Systolic time intervals assessed by 2-D echocardiography and spectral Doppler in the horse. Anim Sci J. 2003;74:505. 238. Welker F.H., Muir W.W.3rd. An investigation of the second heart sound in the normal horse. Equine Vet J. 1990;22:403. 239. Smith B.L., Jones J.H., Pascoe J.R., et al, Why are left atrial pressures high in exercising horses?. Physiologist, 1992. ∗∗∗. 240. Jones J.H., Smith B.L., Birks E.K., et al. Left atrial and pulmonary arterial pressures in exercising horses. Physiologist. 1992;35:A2020. 241. Long K.J. Doppler echocardiography in the horse. Equine Vet Educ. 1990;2:15. 242. DeClercq D., VanLoon G., Tavernier R., et al. Atrial and ventricular electrical and contractile remodeling and reverse remodeling owing to short-term pacing-induced atrial fibrillation in horses. J Vet Intern Med. 2008;22:1353. 243. Schwarzwald C.C., Hamlin R.L., Bonagura J.D., et al. Atrial, SA nodal, and AV nodal electrophysiology in standing horses: normal ndngs and electrophysiologic e ects of quinidine and diltiazem. J Vet Intern Med. 2007;21:166. 244. Schwarzwald C.C., Schober K.E., Bonagura J.D. Methods and reliability of echocardiographic assessment of left atrial size and mechanical function in horses. Am J Vet Res. 2007;68:735. 245. Schwarzwald C.C., Schober K.E., Bonagura J.D. Echocardiographic evidence of left atrial mechanical dysfunction after conversion of atrial brillation to sinus rhythm in 5 horses. J Vet Intern Med. 2007;21:820. 246. Rugh K.S., Garner H.E., Miramonti J.R., et al. Left ventricular function and haemodynamics in ponies during exercise and recovery. Equine Vet J. 1989;21:39.

247. Beglinger R., Becker M. Comparative study of the contractility measurement max. dp/dt in the horse, cow, swine, dog, cat and man. Schweiz Arch Tierheilkd. 1984;126:265. 248. Hillidge C.J., Lees P. Left ventricular systole in conscious and anesthetized horses. Am J Vet Res. 1977;38:675. 249. Koblik P.D., Hornof W.J., Rhode E.A., et al. Left ventricular ejection fraction in the normal horse determined by rst-pass nuclear angiocardiography. Vet Radiol Ultrasound. 1985;26:53. 250. Hillidge C.J., Lees P. Studies of left ventricular isotonic function in conscious and anaesthetised horse. Br Vet J. 1977;133:446. 251. Robine FJ: Morphological and functional measurements on the equine heart by means of two-dimensional echocardiography, Dissertation, Teirarztliche Hochscule, Hanover, Germany, 1991. 252. Lescure F., Tamzali Y. Reference values for echocardiography applied to sport horses (English Thoroughbreds and French riding horses). Rev Med Vet (Toulouse). 1984;135:405. 253. Lombard C.W., Evans M., Martin L., et al. Blood pressure, electrocardiogram and echocardiogram measurements in the growing pony foal. Equine Vet J. 1984;16:342. 254. Stewart J.H., Rose R.J., Barko A.M. Echocardiography in foals from birth to three months old. Equine Vet J. 1984;16:332. 255. Van-Aarde M.N., Littlejohn A., Van der Walt J.J. The ratio of cardiopulmonary blood volume to stroke volume as an index of cardiac function in horses. Vet Res Commun. 1984;8:293. 256. Reef V.B. Evaluation of the equine cardiovascular system. Vet Clin North Am Equine Pract. 1985;1:275. 257. Goldberg S.J., Allen H.D., Marx G.R., et al. Doppler echocardiography, 2nd edition. 1988. Lea & Febiger, Philadelphia. ∗∗∗. 258. Wetmore L.A., Derksen F.J., Blaze C.A., et al. Mixed venous oxygen tension as an estimate of cardiac output in anesthetized horses. Am J Vet Res. 1987;48:971. 259. Bright J.M. Ventricular function. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:3-14. 260. Hillidge C.J., Lees P. Cardiac output in the conscious and anaesthetised horse. Equine Vet J. 1975;7:16. 261. Muir W.W.3rd, Skarda R., Milne D. Estimation of cardiac output in the horse by thermodilution techniques. Am J Vet Res. 1976;37:697. 262. Thomas D.P., Fregin G.F., Gerber N.H., et al. Effects of training on cardiorespiratory function in the horse. Am J Physiol. 1983;245:R160. 263. Evans D.L. Cardiovascular adaptations to exercise and training. Vet Clin North Am Equine Pract. 1985;1:513. 264. Swanson C.R., Muir W.W.3rd. Dobutamine-induced augmentation of cardiac output does not enhance respiratory gas exchange in anesthetized recumbent healthy horses. Am J Vet Res. 1986;47:1573. 265. Thomas W.P., Madigan J.E., Backus K.Q., et al. Systemic and pulmonary haemodynamics in normal neonatal foals. J Reprod Fertil Suppl. 1987;35:623. 266. Ward D.S., Fessler J.F., Bottoms G.D. In vitro calibration and surgical implantation of electromagnetic blood ow transducers for measurement of left coronary blood ow and cardiac output in the pony. Am J Vet Res. 1987;48:1120. 267. Weber J.M., Dobson G.P., Parkhouse W.S., et al. Cardiac output and oxygen consumption in exercising Thoroughbred horses. Am J Physiol. 1987;253:R890. 268. Bertone J.J. Cardiovascular effects of hydralazine HCl administration in horses. Am J Vet Res. 1988;49:618. 269. Gasthuys F., deMoor A., Parmentier D. Haemodynamic changes during sedation in ponies. Vet Res Commun. 1990;14:309. 270. Hinchcliff K.W., McKeever K.H., Schmall L.M., et al. Renal and systemic hemodynamic responses to sustained submaximal exertion in horses. Am J Physiol. 1990;258:R1177. 271. Schmall L.M., Muir W.W.3rd, Robertson J.T. Haemodynamic effects of small volume hypertonic saline in experimentally induced haemorrhagic shock. Equine Vet J. 1990;22:273. 272. Lepiz M.L., Keegan R.D., Bayly W.M., et al. Comparison of Fick and thermodilution cardiac output determinations in standing horses. Res Vet Sci. 2008;85:307. 273. Durando M.M., Corley K.T., Boston R.C., et al. Cardiac output determination by use of lithium dilution during exercise in horses. Am J Vet Res. 2008;69:1054. 274. Valverde A., Giguere S., Morey T.E., et al. Comparison of noninvasive cardiac output measured by use of partial carbon dioxide rebreathing or the lithium dilution method in anesthetized foals. Am J Vet Res. 2007;68:141. 275. Giguere S., Bucki E., Adin D.B., et al. Cardiac output measurement by partial carbon dioxide rebreathing, 2-dimensional echocardiography, and lithium-dilution method in anesthetized neonatal foals. J Vet Intern Med. 2005;19:737. 276. Long K.J., Young L.E., Utting J.E., et al. Determination of cardiac output in the standing horse by Doppler echocardiography and thermodilution. Proc 30th British Equine Vet Assoc Congress. 30, 1991. 277. Mizuno Y., Aida H., Hara H., et al. Comparison of methods of cardiac-output measurements determined by dye dilution, pulsed Doppler-echocardiography and thermodilution in horses. J Vet Med Sci. 1994;56:1. 278. Stadler P., Kinkel N., Deegen E. Evaluation of systolic heart function of the horse with PW-Doppler-echocardiography compared with thermodilution. Dtsch Tierarztl Wochenschr. 1994;101:312. 279. Aida H., Hara H., Fujinaga T., et al. Comparison of methods of cardiac output measurements determined by dye dilution, pulsed Doppler echocardiography and thermodilution in horses. J Vet Med Sci. 1994;56:1. 280. Young L.E., Blissitt K.J., Bartram D.H., et al. Measurement of cardiac output by transoesophageal Doppler echocardiography in anaesthetized horses: comparison with thermodilution. Br J Anaesth. 1996;77:773. 281. Cipone M., Pietra M., Gandini G., et al. Pulsed wave-Doppler ultrasonographic evaluation of the common carotid artery in the resting horse: physiologic data. Vet Radiol Ultrasound. 1997;38:200. 282. Blissitt K.J., Young L.E., Jones R.S., et al. Measurement of cardiac output in standing horses by Doppler echocardiography and thermodilution. Equine Vet J. 1997;29:18. 283. Gratopp M: Estimation of cardiac output by Doppler echocardiography in healthy horses and horses with heart disease, Dissertation, Hanover, Tierarztl Hochsch, Hanover, Germany, 1996. 284. Kinkel N: Determination of cardiac minute volume in horses by Doppler ultrasonography in comparison with conventional methods, 1993, 169 pp 169 pp:1998. 285. Linton R.A., Young L.E., Marlin D.J., et al. Cardiac output measured by lithium dilution, thermodilution, and transesophageal Doppler echocardiography in anesthetized horses. Am J Vet Res. 2000;61:731. 286. Brumbaugh G.W., Thomas W.P., Enos L.R., et al. A pharmaco*kinetic study of digoxin in the horse. J Vet Pharmacol Ther. 1983;6:163. 287. Button C., Gross D.R., Johnston J.T., et al. Digoxin pharmaco*kinetics, bioavailability, efficacy, and dosage regimens in the horse. Am J Vet Res. 1980;41:1388. 288. Dyson D.H., Pascoe P.J. In uence of preinduction methoxamine, lactated Ringer solution, or hypertonic saline solution infusion or postinduction dobutamine infusion on anesthetic-induced hypotension in horses. Am J Vet Res. 1990;51:17. 289. Francfort P., Schatzmann H.J. Pharmacological experiments as a basis for the administration of digoxin in the horse. Res Vet Sci. 1976;20:84. 290. Gasthuys F., deMoor A., Parmentier D. In uence of digoxin followed by dopamine on the cardiovascular depression during a standard halothane anaesthesia in dorsally recumbent, ventilated ponies. Zentralbl Veterinarmed A. 1991;38:585. 291. Grubb T.L., Foreman J.H., Benson G.J., et al. Hemodynamic effects of calcium gluconate administered to conscious horses. J Vet Intern Med. 1996;10:401.

292. Keen P. The use of drugs in the treatment of cardiac disease in the horse. Equine Vet Educ. 1990;2:81. 293. Meijler F.L., van der Tweel I. Digitalis and atrial fibrillation in 1985. Ned Tijdschr Geneeskd. 1985;129:729. 294. Muir W.W.3rd, McGuirk S.M. Pharmacology and pharmaco*kinetics of drugs used to treat cardiac disease in horses. Vet Clin North Am Equine Pract. 1985;1:335. 295. Muir W.W.3rd, McGuirk S. Cardiovascular drugs. Their pharmacology and use in horses. Vet Clin North Am Equine Pract. 1987;3:37. 296. Muir W.W.3rd. Cardiovascular effects of dopexamine HCl in conscious and halothane-anaesthetised horses. Equine Vet J. 1992;24:24. 297. Pedersoli W.M., Belmonte A.A., Purohit R.C., et al. Pharmaco*kinetics of digoxin in the horse. J Equine Med Surg. 1978;2:384. 298. Pedersoli W.M., Ravis W.R., Belmonte A.A., et al. Pharmaco*kinetics of a single, orally administered dose of digoxin in horses. Am J Vet Res. 1981;42:1412. 299. Staudacher G. Individual glycoside treatment by means of serum concentration determination in cardiac insufficiency in horses. Berl Munch Tierarztl Wochenschr. 1989;102:1. 300. Sweeney R.W., Reef V.B., Reimer J.M. Pharmaco*kinetics of digoxin administered to horses with congestive heart failure. Am J Vet Res. 1993;54:1108. 301. Trim C.M. Inotropic agents and vasopressors in equine anesthesia. Compend Cont Educ Pract Vet. 1991;13:118. 302. Muir W.W.3rd. The haemodynamic effects of milrinone HCl in halothane anaesthetised horses. Equine Vet J. 1995;9(Suppl):108. 303. Long-Blissitt K.J., Bonagura J.D., Darke P.G.G. Standardized imaging technique for guided M-mode and Doppler echocardiography in the horse. Equine Vet J. 1992;24:226. 304. Patteson M.W., Gibbs C., Wotton P.R., et al, Echocardiographic measurements of cardiac dimensions and indices of cardiac function in normal adult Thoroughbred horses. Equine Vet J, Suppl 19, September 1995, 18-27. 305. Slater J.D., Herrtage M.E., Echocardiographic measurements of cardiac dimensions in normal ponies and horses. Equine Vet J, Suppl 19, September 1995, 28-32. 306. Bakos Z., Voros K., Jarvinen T., et al. Two-dimensional and M-mode echocardiographic measurements of cardiac dimensions in healthy standardbred trotters. Acta Vet Hung. 2002;50:273. 307. Kriz N.G., Rose R.J. Repeatability of standard transthoracic echocardiographic measurements in horses. Aust Vet J. 2002;80:362. 308. Gehlen H., Marnette S., Rohn K., et al. Precision-controlled echocardiographic left ventricular function parameters by repeated measurement on three consecutive days in trained and untrained Warmblood horses. Dtsch Tierarztl Wochenschr. 2005;112:48. 309. Nuytten J., Deprez P., Picavet T., et al. Heart failure in horses: hemodynamic monitoring and determination of LDH1 concentration. J Equine Vet Sci. 1988;8:214. 310. Fruhauf B., Stadler P., Deegen E. Evaluation of pulmonary wedge pressure in horses with and without left-heart abnormalities detected by echocardiography. Pferdeheilkunde. 1996;12:544. 311. Gehlen H., Bubeck K., Stadler P. Pulmonary artery wedge pressure measurement in healthy Warmblood horses and in Warmblood horses with mitral valve insu ciencies of various degrees during standardized treadmill exercise. Res Vet Sci. 2004;77:257. 312. Milne D.W., Muir W.W.3rd, Skarda R.T. Pulmonary artery wedge pressure blood gas tensions and pH in the resting horse. Am J Vet Res. 1975;36:1431. 313. Gehlen H., Bubeck K., Rohn K., et al. Pulmonary artery wedge pressure during treadmill exercise in Warmblood horses with atrial fibrillation. Res Vet Sci. 2006;81:134. 314. Gehlen H., Groner U., Rohn K., et al. Day-to-day variability of cardiac pressure values in horses measured with right heart catheterization on 3 consecutive days. Berl Munch Tierarztl Wochenschr. 2006;119:400. 315. Muir W.W.3rd, Sams R.A., Hubbell J.A.E., et al. E ects of enalaprilat on cardiorespiratory, hemodynamic, and hematologic variables in exercising horses. Am J Vet Res. 2001;62:1008. 316. Rugh K.S., Garner H.E., Hatfield D.G., et al. Ischemia induced development of functional coronary collateral circulation in ponies. Cardiovasc Res. 1987;21:730. 317. Martin B.B.Jr., Reef V.B., Parente E.J., et al. Causes of poor performance of horses during training, racing, or showing: 348 cases (1992-1996). J Am Vet Med Assoc. 2000;216:554. 318. Parks C.M., Manohar M. Distribution of blood flow during moderate and strenuous exercise in ponies (Equus caballus). Am J Vet Res. 1861;44:1983. 319. Parks C., Manohar M., Lundeen G. Regional myocardial blood ow and coronary vascular reserve in unanesthetized ponies during pacing-induced ventricular tachycardia. J Surg Res. 1983;35:119. 320. Parks C.M., Manohar M. Transmural coronary vasodilator reserve and flow distribution during severe exercise in ponies. J Appl Physiol. 1983;54:1641. 321. Manohar M. Transmural coronary vasodilator reserve and flow distribution during maximal exercise in normal and splenectomized ponies. J Physiol Lond. 1987;387:425. 322. Manohar M., Parks C. Transmural coronary vasodilator reserve in ponies at rest and during maximal exercise. J Appl Physiol. 1983;54:1641. 323. Parks C.M., Manohar M. Transmural distribution of myocardial blood ow during graded treadmill exercise in ponies. In: Persson S.G.B., Rose R.J., editors. Equine exercise physiology. Cambridge, England: Granta; 1983:105-120. 324. Parks C.M., Manohar M. Regional blood flow changes in response to near maximal exercise in ponies: a review. Equine Vet J. 1985;17:311. 325. Reddy V.K., Kammula R.G., Graham T.C., et al. Regional coronary blood flow in ponies. Am J Vet Res. 1976;37:1261. 326. Thornton J.R. Exercise testing. Vet Clin North Am Equine Pract. 1985;1:573. 327. Evans D.L., Rose R.J. Dynamics of cardiorespiratory function in Standardbred horses during different intensities of constant-load exercise. J Comp Physiol. 1988;157:791. 328. Foreman J.H., Bayly W.M., Grant B.D., et al. Standardized exercise test and daily heart rate responses of Thoroughbreds undergoing conventional race training and detraining. Am J Vet Res. 1990;51:914. 329. Rose R.J., Hendrickson D.K., Knight P.K. Clinical exercise testing in the normal thoroughbred racehorse. Aust Vet J. 1990;67:345. 330. Seeherman H.J., Morris E.A. Comparison of yearling, two-year-old and adult Thoroughbreds using a standardised exercise test. Equine Vet J. 1991;23:175. 331. Reef V.B. Ambulatory and exercise electrocardiography and post-exercise echocardiography. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:150-160. 332. Durando M.M., Slack J., Reef V.B., et al, Right ventricular pressure dynamics and stress echocardiography in pharmacological and exercise stress testing. Equine Vet J, Suppl 36, August 2006, 183-192. 333. Glendinning S.A. The clinician’s approach to equine cardiology. Equine Vet J. 1977;9:176. 334. King J.M. Anomalous epicardial lymphatics. Vet Med. 1993;88:512. 335. Glazier B. Clinical aspects of equine cardiology. Pract. 1987;9:98. 336. Fregin G.F. Medical evaluation of the cardiovascular system. Vet Clin North Am Equine Pract. 1992;8:329. 337. Jose-Cunilleras E., Young L.E., Newton J.R., et al, Cardiac arrhythmias during and after treadmill exercise in poorly performing Thoroughbred racehorses. Equine Vet J, Suppl 36, August 2006, 163-170. 338. Gerring E.L. Auscultation of the equine heart. Equine Vet Educ. 1990;2:22. 339. Kammerer H. Auscultation of the horse’s heart, using a new stethoscope. Dtsch Tierarztl Wochenschr. 1983;90:521.

340. Littlejohn A., Button C. When is a murmur not a murmur? J S Afr Vet Assoc. 1982;53:130. 341. Littlewort M.C.G. The clinical auscultation of the equine heart. Vet Rec. 1962;74:1247. 342. Lombard CW: Cardiovascular diseases. In Koterba A, editor: Equine clinical neonatology, pp. 240-261; Philadelphia, 1990. 343. Machida N., Yasuda J., Too K. Auscultatory and phonocardiographic studies on the cardiovascular system of the newborn Thoroughbred foal. Jpn J Vet Res. 1987;35:235. 344. McGuirk S.M., Muir W.W.3rd. Diagnosis and treatment of cardiac arrhythmias. Vet Clin North Am Equine Pract. 1985;1:353. 345. Moore B.R. Lower respiratory-tract disease. Vet Clin North Am Equine Pract. 1996;12:457. 346. Reef V.B. The significance of cardiac auscultatory findings in horses—insight into the age-old dilemma. Equine Vet J. 1993;25:393. 347. Reimer J.M. Performing cardiac auscultation on horses. Vet Med. 1993;88:660. 348. Rossdale P.D. Clinical studies on the newborn Thoroughbred foal. II. Heart rate. Br Vet J. 1967;123:521. 349. Savage C.J. Evaluation of the equine respiratory system using physical examination and endoscopy. Vet Clin North Am Equine Pract. 1997;13(3):443-462. 350. Vanselow B., McCarthy M., Gay C.C. A phonocardiographic study of equine heart sounds. Aust Vet J. 1978;54:161. 351. Blissitt K.J. Auscultation. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:73-92. 352. Young L.E., Helwegen M.M., Rogers K., et al, Associations between exercise-induced pulmonary haemorrhage, right ventricular dimensions and atrioventricular valve regurgitation in conditioned national hunt racehorses. Equine Vet J, Suppl 36, August 2006, 193-197. 353. Naylor J.M., Wolker R.E., Pharr J.W. An assessment of the terminology used by diplomates and students to describe the character of equine mitral and aortic valve regurgitant murmurs: correlations with the physical properties of the sounds. J Vet Intern Med. 2003;17:332. 354. Marr C.M., Reef V.B., Physiological valvular regurgitation in clinically normal young racehorses: prevalence and two-dimensional colour ow Doppler echocardiographic characteristics. Equine Vet J, Suppl 19, September 1995, 56-62. 355. Naylor J.M., Yadernuk L.M., Pharr J.W., et al. An assessment of the ability of diplomates, practitioners, and students to describe and interpret recordings of heart murmurs and arrhythmia. J Vet Intern Med. 2001;15:507. 356. Abbott J. Auscultation: what type of practice makes perfect? J Vet Intern Med. 2001;15:505. 357. Bakos Z., Voros K. Comparative examination of percussional and echocardiographic determination of the cardiac dullness area in healthy horses. Acta Vet Hung. 2007;55:277. 358. Hintz H.F., Collyer C., Brant T. Resting heart rates in draft horses. Vet Clin North Am Equine Pract. 1989;11:7. 359. Reef VB: Frequency of cardiac arrhythmias and their signi cance in normal horses, In Proceedings of the American College of Veterinary Internal Medicine, 1989, San Diego, pp 506. 360. Bonagura J.D., Miller M.S. Electrocardiography. In: Jones W.E., editor. Equine sports medicine. Philadelphia: Lea & Febiger; 1985:89-106. 361. Bertone J.J. Atrial fibrillation in the horse: diagnosis; prognosis, treatment. Vet Clin North Am Equine Pract. 1984;6:6. 362. Bertone J.J., Wingfield W.E. Atrial fibrillation in horses. Compend Cont Educ Pract Vet. 1987;9:763. 363. Deem D.A., Fregin G.F. Atrial fibrillation in horses. J Am Vet Med Assoc. 1982;180:261. 364. Li J.K. Laminar and turbulent flow in the mammalian aorta: Reynolds number. J Theor Biol. 1988;135:409. 365. Marr C.M. Cardiac murmurs: congenital heart disease. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:210-232. 366. Marr C.M. Cardiac murmurs: acquired valvular disease. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:231-255. 367. Machida N., Yasuda J., Too K., et al. A morphological study on the obliteration processes of the ductus arteriosus in the horse. Equine Vet J. 1988;20:249. 368. Andersson B., Augustinsson O., Bademo E., et al. Systemic and centrally mediated angiotensin II effects in the horse. Acta Physiol Scand. 1987;129:143. 369. Bailey J.E., Dunlop C.I., Chapman P.L., et al. Indirect Doppler ultrasonic measurement of arterial blood-pressure results in a large measurement error in dorsally recumbent anesthetized horses. Equine Vet J. 1994;26:70. 370. Branson K.R. A clinical-evaluation of an oscillometric blood-pressure monitor on anesthetized horses. J Equine Vet Sci. 1997;17:537. 371. Ellis P.M. The indirect measurement of arterial blood pressure in the horse. Equine Vet J. 1975;7:22. 372. Erickson B.K., Erickson H.H., Co man J.R., Pulmonary artery, aortic and oesophageal pressure changes during high intensity treadmill exercise in the horse: a possible relation to exercise-induced pulmonary haemorrhage. Equine Vet J, Suppl 9, June 1990, 47-52. 373. Franco R.M., Ousey J.C., Cash R.S., et al. Study of arterial blood pressure in newborn foals using an electronic sphygmomanometer. Equine Vet J. 1986;18:475. 374. Fritsch R., Bosler K. Monitoring circulation in the horse during sedation and anesthesia by indirect blood pressure measurement. Berl Munch Tierarztl Wochenschr. 1985;98:166. 375. Fritsch R., Hausmann R. Indirect blood pressure determination in the horse with the Dinamap 1255 research monitor. Tierarztl Prax. 1988;16:373. 376. Gasthuys F., Muylle E., deMoor A., et al. In uence of premedication and body position during halothane anaesthesia on intracardial pressures in the horse. Zentralbl Veterinarmed A. 1988;35:729. 377. Will J.A., Bisgard G.E. Cardiac catheterization of unanesthetized large domestic animals. J Appl Physiol. 1972;33:400. 378. Gay C.C., McCarthy M., Reynolds W.T., et al. A method for indirect measurement of arterial blood pressure in the horse. Aust Vet J. 1977;53:163. 379. Geddes L.A., Chaffee V., Whistler S.J., et al. Indirect mean blood pressure in the anesthetized pony. Am J Vet Res. 1977;38:2055. 380. Glen J.B. Indirect blood pressure measurement in anesthetized animals. Vet Rec. 1970;87:349. 381. Hall L.W. Cardiovascular and pulmonary effects of recumbency in two conscious ponies. Equine Vet J. 1984;16:89. 382. Johnson J.H., Garner H.E., Hutcheson D.P. Ultrasonic measurement of arterial blood pressure in conditioned Thoroughbreds. Equine Vet J. 1976;8:55. 383. Kvart C. An ultrasonic method for indirect blood pressure measurement in the horse. J Equine Med Surg. 1979;3:16. 384. Latshaw H., Fessler J.F., Whistler S.J., et al. Indirect measurement of mean blood pressure in the normotensive and hypotensive horses. Equine Vet J. 1979;11:191. 385. Parry B.W., McCarthy M.A., Anderson G.A., et al. Correct occlusive bladder width for indirect blood pressure measurement in horses. Am J Vet Res. 1982;43:50. 386. Parry B.W., Gay C.C., McCarthy M.A. Influence of head height on arterial blood pressure in standing horses. Am J Vet Res. 1980;41:1626. 387. Parry B.W. Resting blood pressure values in various equine clinical cases. J Equine Vet Sci. 1984;4:49. 388. Parry B.W., Anderson G.A. Importance of uniform cuff application for equine blood pressure measurement. Equine Vet J. 1984;16:529. 389. Parry B.W., McCarthy M.A., Anderson G.A. Survey of resting blood pressure values in clinically normal horses. Equine Vet J. 1984;16:53. 390. Physick-Sheard P.W. Cardiovascular response to exercise and training in the horse. Vet Clin North Am Equine Pract. 1985;1:383. 391. Taylor P.M. Techniques and clinical application of arterial blood pressure. Equine Vet J. 1981;13:271.

392. Wagner A.E., Brodbelt D.C. Arterial blood-pressure monitoring in anesthetized animals. J Am Vet Med Assoc. 1997;210:1279. 393. Wens H.M. Catheterization of the heart by J.B.A. Chauveau in 1861. Tierarztliche Umschau. 1989;44:90. 394. Bayly W.M., Gabel A.A., Barr S.A. Cardiovascular e ects of submaximal aerobic training on a treadmill in Standardbred horses, using a standardized exercise test. Am J Vet Res. 1983;44:544. 395. Muir W.W.3rd, Wade A., Grospitch B. Automatic noninvasive sphygmomanometry in horses. J Am Vet Med Assoc. 1983;182:1230. 396. Tillotson P.J., Kooper P.H. Treatment of aortic thrombus in a horse. J Am Vet Med Assoc. 1966;149:766. 397. Tithof P.K., Rebhun W.C., Dietze A.E. Ultrasonographic diagnosis of aorto-iliac thrombosis. Cornell Vet. 1985;75:540. 398. Dickson L.R., Badcoe L.M., Burbidge H., et al. Jugular thrombophlebitis resulting from an anesthetic induction technique in the horse. Equine Vet J. 1990;22:177. 399. Marr C.M. Heart failure. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:289-311. 400. Hall M.C., Steel J.D., Stewart G.A. Cardiac monitoring during exercise tests in the horse. II. Heart rate responses to exercise. Aust Vet J. 1976;52:1. 401. Maier-Bock H., Ehrlein H.J. Heart rate during a defined exercise test in horses with heart and lung diseases. Equine Vet J. 1978;10:235. 402. Rose R.J., Allen J.R., Hodgson D.R., et al. Responses to submaximal treadmill exercise and training in the horse: changes in haematology, arterial blood gas and acid base measurements, plasma biochemical values and heart rate. Vet Rec. 1983;113:612. 403. Rose R.J., Lovell D.K. Symposium on exercise physiology. Vet Clin North Am Equine Pract. 1985;1:437. 404. Evans D.L., Rose R.J. Method of investigation of the accuracy of four digitally displaying heart rate meters suitable for use in the exercising horse. Equine Vet J. 1986;18:129. 405. Evans D.L., Rose R.J. Cardiovascular and respiratory responses to submaximal exercise training in the Thoroughbred horse. Pflugers Arch. 1988;411:316. 406. Erickson B.K., Erickson H.H., Co man J.R. Pulmonary artery and aortic pressure changes during high intensity treadmill exercise in the horse: e ect of furosemide and phentolamine. Equine Vet J. 1992;24:215. 407. Evans D.L., Harris R.C., Snow D.H. Correlation of racing performance with blood lactate and heart rate after exercise in Thoroughbred horses. Equine Vet J. 1993;25:441. 408. Christley R.M., Hodgson D.R., Evans D.L., et al. Cardiorespiratory responses to exercise in horses with di erent grades of idiopathic laryngeal hemiplegia. Equine Vet J. 1997;29:6. 409. Hackett R.P., Ducharme N.G., Gleed R.D., et al. Oral nitroglycerin paste did not lower pulmonary capillary pressure during treadmill exercise. Equine Vet J Suppl. 1999;30:153. 410. Betros C.L., McKeever K.H., Kearns C.F., et al, Effects of ageing and training on maximal heart rate and VO2max. Equine Vet J, Suppl 34, September 2002, 100-105. 411. Durando M.M., Reef V.B., Birks E.K., Right ventricular pressure dynamics during exercise: relationship to stress echocardiography. Equine Vet J, Suppl 34, 2002, 472-477. 412. Ohmura H., Hiraga A., Matsui A., et al, Physiological responses of young Thoroughbreds during their rst year of race training. Equine Vet J, Suppl 34, September 2002, 140146. 413. Frye M.A., Bright J.M., Dargatz D.A., et al. A comparison of dobutamine infusion to exercise as a cardiac stress test in healthy horses. J Vet Intern Med. 2003;17:58. 414. Gehlen H., Bubeck K., Stadler P. Pulmonary wedge pressure and heart frequency measurements during standardized treadmill exercise for extension of left atrial function diagnosis in warmblood horses. Dtsch Tierarztl Wochenschr. 2003;110:280. 415. Zucca E., Ferrucci F., Di F.V., et al. The use of electrocardiographic recording with Holter monitoring during treadmill exercise to evaluate cardiac arrhythmias in racehorses. Vet Res Commun. 2003;27(Suppl 1):811. 416. Meyer C., Gerber R., Guthrie A.J. The use of the standard exercise test to establish the clinical signi cance of mild echocardiographic changes in a Thoroughbred poor performer. J S Afr Vet Assoc. 2004;75:100. 417. Durando M. Diagnosing cardiac disease in equine athletes: the role of stress testing. Equine Vet J. 2005;37:101. 418. Gehlen H., Marnette S., Rohn K., et al. Stress echocardiography in Warmblood horses: comparison of dobutamine/atropine with treadmill exercise as cardiac stressors. J Vet Intern Med. 2006;20:562. 419. Gramkow H.L., Evans D.L., Correlation of race earnings with velocity at maximal heart rate during a eld exercise test in Thoroughbred racehorses. Equine Vet J, Suppl 36, August 2006, 118-122. 420. Hada T., Ohmura H., Mukai K., et al. Utilization of the time constant calculated from heart rate recovery after exercise for evaluation of autonomic activity in horses. Equine Vet J. 2006;36(Suppl):141. 421. Mukai K., Ohmura H., Hiraga A., et al. Effect of detraining on cardiorespiratory variables in young thoroughbred horses. Equine Vet J. 2006;36(Suppl):210. 422. Nankervis K.J., Williams R.J., Heart rate responses during acclimation of horses to water treadmill exercise. Equine Vet J, Suppl 36, August 2006, 110-112. 423. Sandersen C., Detilleux J., Art T., et al, Exercise and pharmacological stress echocardiography in healthy horses. Equine Vet J, Suppl 36, August 2006, 159-162. 424. Sandersen C.F., Detilleux J., de Mo arts B., et al. E ect of atropine-dobutamine stress test on left ventricular echocardiographic parameters in untrained Warmblood horses. J Vet Intern Med. 2006;20:575. 425. Vincen T.L., Newton J.R., Deaton C.M., et al, Retrospective study of predictive variables for maximal heart rate (HRmax) in horses undergoing strenuous treadmill exercise. Equine Vet J, Suppl 36, August 2006, 146-152. 426. Poggenpoel D.G. Measurements of heart rate and riding speed on a horse during a training program for endurance rides. Equine Vet J. 1988;20:224. 427. Amada A., Kurita H. Five cases of paroxysmal atrial fibrillation in the racehorse. Exp Rep Equine Hlth Lab. 1975;12:89. 428. Hiraga A., Kubo K. Two cases of paroxysmal atrial fibrillation during exercise in horses. Equine Vet Educ. 1999;11:6. 429. Physick-Sheard P.W., Marlin D.J., Thornhill R., et al. Frequency domain analysis of heart rate variability in horses at rest and during exercise. Equine Vet J. 2000;32:253. 430. Thayer J.F., Hahn A.W., Pearson M.A., et al. Heart rate variability during exercise in the horse. Biomed Sci Instrum. 1997;34:246. 431. Sandersen C.F., Detilleux J., Delguste C., et al. Atropine reduces dobutamine-induced side e ects in ponies undergoing a pharmacological stress protocol. Equine Vet J. 2005;37:128. 432. Coudry V., Jean D., Desbois C., et al. Myocardial fibrosis in a horse with polymorphic ventricular tachycardia observed during general anesthesia. Can Vet J. 2007;48:623. 433. Finley M.R., Li Y., Hua F., et al. Expression and coassociation of ERG1, KCNQ1, and KCNE1 potassium channel proteins in horse heart. Am J Physiol Heart Circ Physiol. 2002;283:H126. 434. Kuwahara M., Hiraga A., Kai M., et al. In uence of training on autonomic nervous function in horses: evaluation by power spectral analysis of heart rate variability. Equine Vet J Suppl. 1999;30:178. 435. Glomset D.J., Glomset A.T.A. A morphologic study of the cardiac conduction system in ungulates, dog, and man. Part I: The sinoatrial node. Am Heart J. 1940;20:389. 436. Gross D.R. Practical electrocardiography in the equine subject. J Am Vet Med Assoc. 1971;159:1335. 437. Grauerholz H., Jaeschke G. Construction of main and reference vectors from limb leads in the ECG of the horse. Berl Munch Tierarztl Wochenschr. 1988;101:376.

438. Grauerholz H., Jaeschke G. Training-induced changes of reference vectors in the QRS complex of the EKG of young trotting horses. Berl Munch Tierarztl Wochenschr. 1990;103:329. 439. Grauerholz H. In uence of respiration on the QRS complex of the ECG in clinically healthy horses and in horses with respiratory problems. Berl Munch Tierarztl Wochenschr. 1990;103:293. 440. Grauerholz H., Jaeschke G. Alterations induced by training in reference vectors of the electrocardiographic QRS complex of young trotting horses. Berl Munch Tierarztl Wochenschr. 1990;103:329. 441. Hartley J.W., Hahn A.W., DeLorey M., et al. Digital processing of equine exercise electrocardiograms. Biomed Sci Instrum. 1990;26:11. 442. Hilwig R.W. Cardiac arrhythmias in the horse. J Am Vet Med Assoc. 1977;170:153. 443. Hanak J., Zert Z. Some ECG characters in Thoroughbred horses, common to parents and their offspring. Vet Med (Praha). 1982;27:87. 444. Hanak J., Jagos P. Electrocardiographic lead system and its vector verification. Acta Vet Brno. 1983;52:67. 445. Fregin G.F. The equine electrocardiogram with standardized body and limb positions. Cornell Vet. 1982;72:304. 446. Fregin G.F. Electrocardiography. Vet Clin North Am Equine Pract. 1985;1:419. 447. Bonagura J.D., Miller M.S. Electrocardiography. What is your diagnosis? Junctional and ventricular arrhythmias. J Equine Vet Sci. 1985;5:347. 448. Amory H., Rollin F., Genicot B., et al. Bovine vectocardiography: a comparative study relative to the validity of four tridimensional lead systems. J Vet Med. 1992;39:453. 449. Clark D.R., McCrady J.D. Clinical use of the electrocardiogram in animals. I. Fundamentals of ECG examination. Vet Med Small Anim Clin. 1966;61:751. 450. Cornick J.L., Seahorn T.L. Cardiac arrhythmias identified in horses with duodenitis/proximal jejunitis: six cases (1985-1988). J Am Vet Med Assoc. 1990;197:1054. 451. Costa G., Illera M., Garcia-Sacristan A. Electrocardiographical values in non-trained horses. Zentralbl Veterinarmed A. 1985;32:196. 452. Illera J.C., Illera M. Physiological electrocardiograms as the basis for diagnosis of heart diseases in horses. Arch Med Vet. 1986;3:239. 453. Illera J.C., Illera M. Electrocardiography and heart score of horses competing in an endurance ride. Aust Vet J. 1987;64:88. 454. Illera J.C., Illera M. Precordial heart score. Aust Vet J. 1988;65:355. 455. Irie T. A study of arrhythmias in Thoroughbred newborn foals immediately after birth. Jpn J Vet Res. 1990;38:57. 456. Kuwahara M., Hashimoto S., Ishii K., et al. Assessment of autonomic nervous function by power spectral analysis of heart rate variability in the horse. J Auton Nerv Syst. 1996;60:43. 457. Landgren S., Rutqvist L. Electrocardiogram of normal cold blooded horses after work. Nord Vet Med. 1953;5:905. 458. Lannek N., Rutqvist L. Normal area of variation for the electrocardiogram of horses. Nord Vet Med. 1951;3:1094. 459. Matsui K. Fetal and maternal heart rates in a case of twin pregnancy of the Thoroughbred horse. Nippon Juigaku Zasshi. 1985;47:817. 460. Polglaze K., Evans D.L. The relationship between racing performance and electrocardiographic findings in the Standardbred racehorse. Austr Equine Vet. 1992;10:88. 461. Rose R.J., Davis P.E. The use of electrocardiography in the diagnosis of poor racing performance in the horse. Aust Vet J. 1978;54:51. 462. Stewart J.H., Rose R.J., Davis P.E., et al. A comparison of electrocardiographic ndings in racehorses presented either for routine examination or poor racing performance. In: Persson S.G.B., Rose R.J., editors. Equine exercise physiology. Cambridge, England: Granta; 1983:135-143. 463. Studzinski T., Czarnecki A. Relationship between the QRS duration (heart score) and ventricular weight in horses. Ann Univ Mariae Curie Sklodowska Med. 1980;35/36:33-43. 464. Tovar P., Escabias M.I., Santisteban R. Evolution of the ECG from Spanish bred foals during the post natal stage. Res Vet Sci. 1989;46:358. 465. Tschudi P. Electrocardiography in the horse (1). Principles and normal picture. Tierarztl Prax. 1985;13:181. 466. Tschudi P. Electrocardiography in the horse. (3). Tierarztl Prax. 1986;14:365. 467. White N.A., Rhode E.A. Correlation of electrocardiographic findings to clinical disease in the horse. J Am Vet Med Assoc. 1974;164:46. 468. Yamamoto K., Yasuda J., Too K. Electrocardiographic ndings during parturition and blood gas tensions immediately after birth in Thoroughbred foals. Jpn J Vet Res. 1991;39:143. 469. Yamamoto K., Yasuda J., Too K. Arrhythmias in newborn Thoroughbred foals. Equine Vet J. 1992;24:169. 470. Yamaya Y., Kubo K., Amada A., et al. Intrinsic atrioventricular conductive function in horses with a 2nd-degree atrioventricular block. J Vet Med Sci. 1997;59:149. 471. Persson S.G.B., Forssbergy P. Exercise tolerance in Standardbred trotters with T-wave abnormalities. In: Gillespie J.R., Robinson N.E., editors. Equine exercise physiology. Davis, CA: ICEEP Publications; 1987:772-780. 472. Matsui K., Sugano S. Species di erences in the changes in heart rate and T-wave amplitude after autonomic blockade in Thoroughbred horses, ponies, cows, pigs, goats and chickens. Nippon Juigaku Zasshi. 1987;49:637. 473. Carlsten J., Kvart C., Jeffcott L.B. Method of selective and non-selective angiocardiography for the horse. Equine Vet J. 1984;16:47. 474. Marr C.M. Ancillary diagnostic aids in equine cardiology. Equine Vet Educ. 1990;2:18. 475. O’Brien R.T., Biller D.S. Field imaging of the respiratory tract radiology and ultrasonography. Vet Clin North Am Equine Pract. 1997;13:487. 476. Scott E.A., Chaffee A., Eyster G.E., et al. Interruption of aortic arch in two foals. J Am Vet Med Assoc. 1978;172:347. 477. Patteson M.W. Two-dimensional and M-mode echocardiography. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:93-116. 478. Marr C.M. Doppler echocardiography. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:117-134. 479. Lombard C.W., Scarratt W.K., Buergelt C.D. Ventricular septal defects in the horse. J Am Vet Med Assoc. 1983;183:562. 480. Gerlis L.M., Wright H.M., Wilson N., et al. Left ventricular bands. A normal anatomical feature. Br Heart J. 1984;52:641. 481. O’Callaghan M.W. Comparison of echocardiographic and autopsy measurements of cardiac dimensions in the horse. Equine Vet J. 1985;17:361. 482. Pipers F.S., Reef V.B., Wilson J. Echocardiographic detection of ventricular septal defects in large animals. J Am Vet Med Assoc. 1985;187:810. 483. Yamaga Y., Shibui I., Yasuda J., et al. Echocardiographic and ultrasonographic observations in a horse with mitral regurgitation and “intrahepatic cholangiocellular fibroadenomatosis. Advances in Animal Cardiology. 1985;18:65-75. 484. Bertone J.J., Paull K.S., Wingfield W.E., et al. M-mode echocardiographs of endurance horses in the recovery phase of long-distance competition. Am J Vet Res. 1987;48:1708. 485. Reef V.B., Klumpp S., Maxson A.D., et al. Echocardiographic detection of an intact aneurysm in a horse. J Am Vet Med Assoc. 1990;197:752. 486. Reef V.B. Echocardiographic examination in the horse: the basics. Compend Cont Educ Pract Vet. 1990;12:1312. 487. Reef V.B. Echocardiographic findings in horses with congenital cardiac disease. Compend Cont Educ Pract Vet. 1991;1:109. 488. Weinberger T: Doppler echocardiography in horses, Hannover, 1991, Germany, Tierarztliche Hochsc, p 237 489. Lester G.D., Lombard C.W., Ackerman N. Echocardiographic detection of a dissecting aortic root aneurysm in a Thoroughbred stallion. Vet Radiol Ultrasound. 1992;33:202.

490. Mahony C., Rantanen N.W., DeMichael J.A., et al. Spontaneous echocardiographic contrast in the Thoroughbred: high prevalence in racehorses and a characteristic abnormality in bleeders. Equine Vet J. 1992;24:129. 491. Long K.J. Echocardiographic studies of valvular and ventricular function in horses, Dissertation. Scotland: University of Edinburgh, 1993. 492. Long K.J. Doppler echocardiography—clinical applications. Equine Vet Educ. 1993;5:161. 493. Young L.E., Long K.J., Clutton R.E., et al. The use of two dimensional and Doppler echocardiography for haemodynamic monitoring during general anaesthesia: preliminary findings in halothane-anaesthetised horses. Vet Anaesth Analg. 1993;20:42. 494. Bonagura J.D. Echocardiography. J Am Vet Med Assoc. 1994;204:516. 495. Stadler P., Deegen E., Kroker K. Echocardiography and therapy of atrial-fibrillation in horses. Dtsch Tierarztl Wochenschr. 1994;101:190. 496. Cipone M., Pietra M., Guglielmini C., et al. Insu cienca aortica nel cavallo. Rilievi elettrofonocardiogra ci, ecocardiogra cied eco-Doppler carotidei in due casi [Aortic insu ciency in horses. Results of electrophonocardiography, ultrasonic cardiography and carotid pulsed-wave Doppler echocardiography in two cases]. Obiettivi e Documenti Veterinari. 1995;16:37. 497. Stadler P., Fruhauf B., Deegen E. Echocardiographic determination of diastolic heart function and measurement of pulmonary wedge pressure in horses. Pferdeheilkunde. 1995;11:109. 498. Tucker R.L., Wickler S.J., London C., et al. Echocardiographic and right-sided cardiac pressure comparison of the mule and horse. J Equine Vet Sci. 1995;15:404. 499. Blissitt K.J., Marr C.M., Rossdale P.D., et al. Equine cardiovascular medicine. Equine Vet J. (Suppl):1996. 500. Darke P.G.G., Bonagura J.D., Kelly D.F. Colour Atlas of Veterinary Cardiology. London: Mosby, 1996. 501. Stadler P., Wohlsein P., Gratopp M., et al. Echocardiographic and radiographic imaging of aortic root and aortic arch aneurysm in the horse. Pferdeheilkunde. 1996;12:91. 502. Stadler P., Robine F.J. B-mode echocardiographic measurement of heart dimensions in warm-blooded horses without heart disease. Pferdeheilkunde. 1996;12:35. 503. Hoch M. Colour-coded Doppler echocardiography in horses. Dissertation. Germany: University of Hanover, 1998. 504. Reef V.B. Evaluation of ventricular septal defects in horses using two-dimensional and Doppler echocardiography. Equine Vet J Suppl. 86, 1995. 505. Reef V.B., Heart murmurs in horses: determining their significance with echocardiography. Equine Vet J, Suppl 19, September 1995, 71-80. 506. Cranley J.J. Focal medial calcification of the pulmonary artery: a survey of 1066 horses. Equine Vet J. 1983;15:278. 507. van Loon G., Fonteyne W., Rottiers H., et al. Dual-chamber pacemaker implantation via the cephalic vein in healthy equids. J Vet Intern Med. 2001;15:564. 508. Diaz O.S., Sleeper M.M., Reef V.B., et al. Aortitis in a Paint gelding. Equine Vet J. 2000;32:354. 509. Young L.E., Wood J.L. Effect of age and training on murmurs of atrioventricular valvular regurgitation in young Thoroughbreds. Equine Vet J. 2000;32:195. 510. Schober K.E., Kaufhold J., Kipar A. Mitral valve dysplasia in a foal. Equine Vet J. 2000;32:170. 511. Karlstam E., Ho S.Y., Shokrai A., et al. Anomalous aortic origin of the left coronary artery in a horse. Equine Vet J. 1999;31:350. 512. Southwood L.L., Tobias A.H., Schott H.C., et al. Cyanosis and intense murmur in a neonatal foal. J Am Vet Med Assoc. 1996;208:835. 513. Reppas G.P., Canfield P.J., Hartley W.J., et al. Multiple congenital cardiac anomalies and idiopathic thoracic aortitis in a horse. Vet Rec. 1996;138:14. 514. Lord P.F., Croft M.A. Accuracy of formulae for calculating left ventricular volumes of the equine heart. Equine Vet J Suppl. 1990;9:53-56. 515. Hoffmann K.L., Wood A.K., Kirby A.C. Use of Doppler ultrasonography to evaluate renal arterial blood flow in horses. Am J Vet Res. 1997;58:697. 516. Rivas L.J., Hinchcliff K.W. Effect of furosemide and subsequent intravenous fluid administration on right atrial pressure of splenectomized horses. Am J Vet Res. 1997;58:632. 517. Young L. Transoesophageal echocardiography. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:135-149. 518. Buhl R., Ersboll A.K., Eriksen L., et al. Sources and magnitude of variation of echocardiographic measurements in normal Standardbred horses. Vet Radiol Ultrasound. 2004;45:505. 519. Zucca E., Ferrucci F., Croci C., et al. Echocardiographic measurements of cardiac dimensions in normal Standardbred racehorses. J Vet Cardiol. 2008;10:45. 520. Reimer J.M. Cardiac evaluation of the horse using ultrasonography. Vet Med. 1993;88:748. 521. Adams Brendemuehl C., Pipers F.S., Antepartum evaluations of the equine fetus. J Reprod Fertil, Suppl 35, 1987, 565. 522. Stadler P., Rewel A., Deegen E. M-mode echocardiography in dressage horses, class S jumping horses and untrained horses. Zentralbl Veterinarmed A. 1993;40:292. 523. Carlsten J.C. Two-dimensional, real-time echocardiography in the horse. Vet Radiol Ultrasound. 1987;28:76. 524. Stadler P., D’Agostino U., Deegen E. Real-time, two-dimensional echocardiography in horses. Pferdeheilkunde. 1988;4:161. 525. Schwarzwald C.C., Bonagura J.D. Methods and reliability of tissue Doppler imaging for assessment of left ventricular radial wall motion in horses. J Vet Intern Med. 2009. in press 526. Schwarzwald C.C., Schober K.E., Berli A.S.J., et al. Left ventricular radial and circumferential wall motion analysis in horses using strain, strain rate, and displacement by 2D speckle tracking. J Vet Intern Med. 2009;23:890. 527. Rantanen N.W. Diseases of the thorax. Vet Clin North Am Equine Pract. 1986;2:49. 528. Bonagura J.D., Pipers F.S. Diagnosis of cardiac lesions by contrast echocardiography. J Am Vet Med Assoc. 1983;182:396. 529. Buhl R., Ersboll A.K., Larsen N.H., et al. The e ects of detomidine, romi dine or acepromazine on echocardiographic measurements and cardiac function in normal horses. Vet Anaesth Analg. 2007;34:1. 530. Amend J.F., Garner H., Rosborough J.P., et al. Hemodynamic studies in conscious domestic ponies. J Surg Res. 1975;19:107. 531. Bove A.A. Effects of strenuous exercise on myocardial blood flow. Med Sci Sports Exerc. 1985;17:517. 532. Davis J.L., Manohar M. Effect of splenectomy on exercise-induced pulmonary and systemic hypertension in ponies. Am J Vet Res. 1988;49:1169. 533. Drummond W.H., Sanchez I.R., Kosch P.C., et al. Pulmonary vascular reactivity of the newborn pony foal. Equine Vet J. 1989;21:181. 534. Dunlop C.I., Hodgson D.S., Chapman P.L., et al. Thermodilution estimation of cardiac output at high flows in anesthetized horses. Am J Vet Res. 1893;52:1991. 535. Fisher E.W., Dalton R.G. Determination of cardiac output of cattle and horses by the injection. Br Vet J. 1961;117:141. 536. Goetz T.E., Manohar M. Pressures in the right side of the heart and esophagus (pleura) in ponies during exercise before and after furosemide administration. Am J Vet Res. 1986;47:270. 537. Grossman W., Barry W.H. Cardiac catheterization. In: Braunwald E., editor. Heart disease: a textbook of cardiovascular medicine. ed 4. Philadelphia: WB Saunders; 1992:180-203. 538. Hall L.W., Nigam J.M. Measurement of central venous pressure in horses. Vet Rec. 1975;97:66. 539. Hellyer P.W., Dodam J.R., Light G.S. Dynamic barore ex sensitivity in anesthetized horses, maintained at 1.25 to 1.3 minimal alveolar concentration of halothane. Am J Vet Res. 1991;52:1672.

540. Hillidge C.J., Lees P. Influence of general anaesthesia on peripheral resistance in the horse. Br Vet J. 1977;133:225. 541. Hlastala M.P., Bernard S.L., Erickson H.H., et al. Pulmonary blood-flow distribution in standing horses is not dominated by gravity. J Appl Physiol. 1996;81:1051. 542. Manohar M. Right heart pressures and blood-gas tensions in ponies during exercise and laryngeal hemiplegia. Am J Physiol. 1986;251:H121. 543. Manohar M., Goetz T.E., Hutchens E., et al. Atrial and ventricular myocardial blood flows in horses at rest and during exercise. Am J Vet Res. 1994;55:1464. 544. Manohar M. Pulmonary vascular pressures of strenuously exercising Thoroughbreds after administration of flunixin meglumine and furosemide. Am J Vet Res. 1994;55:1308. 545. Manohar M., Goetz T.E., Hutchens E., et al. Effects of graded-exercise on pulmonary and systemic hemodynamics in horses. Vet Clin North Am Equine Pract. 1995;17:17. 546. Muir W.W.3rd, McGuirk S.M. Hemodynamics before and after conversion of atrial fibrillation to normal sinus rhythm in horses. J Am Vet Med Assoc. 1984;184:965. 547. Schatzmann U., Battier B. Factors influencing central venous pressure in horses. Dtsch Tierarztl Wochenschr. 1987;94:147. 548. Sheridan V., Deegen E., Zeller R. Central venous pressure (CVP) measurements during halothane anaesthesia. Vet Rec. 1972;90:149. 549. Sinha A.K., Gleed R.D., Hakim T.S., et al. Pulmonary capillary pressure during exercise in horses. J Appl Physiol. 1996;80:1792. 550. Sporri H., Denac M. The ventricular pressure increase velocity as a parameter of myocardial strength development. Schweiz Arch Tierheilkd. 1969;111:239. 551. Staddon G.E., Weaver B.M.G., Webb A.I. Distribution of cardiac output in anaesthetized horses. Res Vet Sci. 1979;27:38. 552. West J.B., Mathieu-Costello O. Stress failure of pulmonary capillaries as a mechanism for exercise-induced pulmonary hemorrhage in the horse. Equine Vet J. 1994;26:441. 553. Manohar M. Pulmonary artery wedge pressure increases with high-intensity exercise in horses. Am J Vet Res. 1993;54:142. 554. Klein L., Sherman J. Effects of preanesthetic medication, anesthesia, and position of recumbency on central venous pressure in horses. J Am Vet Med Assoc. 1977;170:216. 555. Bisgard G.E., Orr J.A., Will J.A. Hypoxic pulmonary hypertension in the pony. Am J Vet Res. 1975;36:49. 556. Gelberg H.B., Smetzer D.L., Foreman J.H. Pulmonary hypertension as a cause of atrial fibrillation in young horses: four cases (1980-1989). J Am Vet Med Assoc. 1991;198:679. 557. Corley K.T., Donaldson L.L., Furr M.O. Comparison of lithium dilution and thermodilution cardiac output measurements in anaesthetized neonatal foals. Equine Vet J. 2002;34:598. 558. Corley K.T., Donaldson L.L., Durando M.M., et al. Cardiac output technologies with special reference to the horse. J Vet Intern Med. 2003;17:262. 559. Wilkins P.A., Boston R.C., Gleed R.D., et al. Comparison of thermal dilution and electrical impedance dilution methods for measurement of cardiac output in standing and exercising horses. Am J Vet Res. 2005;66:878. 560. Hallowell G.D., Corley K.T. Use of lithium dilution and pulse contour analysis cardiac output determination in anaesthetized horses: a clinical evaluation. Vet Anaesth Analg. 2005;32:201. 561. Orr J.A., Bisgard G.E., Forster H.V., et al. Cardiopulmonary measurements in nonanesthetized resting normal ponies. Am J Vet Res. 1975;36:1667. 562. Schwarzwald C.C., Hardy J., Buccellato M. High cardiac troponin I serum concentration in a horse with multiform ventricular tachycardia and myocardial necrosis. J Vet Intern Med. 2003;17:364. 563. Holbrook T.C., Birks E.K., Sleeper M.M., et al. Endurance exercise is associated with increased plasma cardiac troponin I in horses. Equine Vet J (Suppl 36). August 2006:27-31. 564. Divers T.J., Krans M.S., Jesty S.A., et al. Clinical Findings and serum cardiac troponin I concentrations in horses after intragastric administration of sodium monensin. J Vet Diagn Invest. 2009;21:338. 565. Davis J.L., Gardner S.Y., Schwabenton B., et al. Congestive heart failure in horses: 14 cases (1984-2001). J Am Vet Med Assoc. 2002;220:1512. 566. Bonagura J.D. Diagnosis of cardiac arrhythmias. In Robinson N.E., editor: Current therapy in equine medicine, ed 6, St Louis: WB Saunders, 2009. 567. Imhasly A., Tschudi P.R., Lombard C.W., et al. Clinical and echocardiographic features of mild mitral valve regurgitation in 108 horses. Vet J. March 7, 2009. [Epub ahead of print] 568. Belgrave J.O.S. A case of atrial fibrillation with congestive heart failure. Equine Vet Educ. 1990;2:2. 569. Bonagura J.D. Clinical evaluation and management of heart disease. Equine Vet Educ. 1990;2:31. 570. Divers T.J., Whitlock R.H., Byars T.D., et al. Acute renal failure in six horses resulting from haemodynamic causes. Equine Vet J. 1987;19:178. 571. Glazier D.B. Congestive heart failure and congenital cardiac defects in horses. Vet Clin North Am Equine Pract. 1986;8:20. 572. Hinchcliff K.W., Muir W.W.3rd. Pharmacology of furosemide in the horse: a review. J Vet Intern Med. 1991;5:211. 573. Marr C.M. Treatment of cardiac arrhythmias and cardiac failure. In Robinson N.E., editor: Current therapy in equine medicine, ed 6, St Louis: WB Saunders, 2009. 574. Reef V.B., Levitan C.W., Spencer P.A. Factors affecting prognosis and conversion in equine atrial fibrillation. J Vet Intern Med. 1988;2:1. 575. Reimer J.M., Marr C.M., Reef V.B., et al. Aortic origin of the right pulmonary artery and patent ductus arteriosus in a pony foal with pulmonary hypertension and right-sided heart failure. Equine Vet J. 1993;25:466. 576. Seahorn T.L., Hormanski C.E. Ventricular septal defect and atrial fibrillation in an adult horse—a case report. J Equine Vet Sci. 1993;13:36. 577. Smith H.J., Nuttall A. Experimental models of heart failure. Cardiovasc Res. 1985;19:181. 578. Taylor F.G., Wotton P.R., Hillyer M.H., et al. Atrial septal defect and atrial fibrillation in a foal. Vet Rec. 1991;128:80. 579. Mogg T.D. Equine cardiac disease. Clinical pharmacology and therapeutics. Vet Clin North Am Equine Pract. 1999;15:523. 580. Wijnberg I.D., van der Kolk J.H., van Garderen E., et al. Atrial fibrillation associated with central nervous symptoms and colic in a horse: a case of equine cardiomyopathy. Vet Q. 1998;20:73. 581. Garber J.L., Reef V.B., Reimer J.M. Sonographic findings in horses with mediastinal lymphosarcoma—13 cases (1985-1992). J Am Vet Med Assoc. 1994;205:1432. 582. Southwood L.L., Schott H.C., Henry C.J., et al. Disseminated hemangiosarcoma in the horse: 35 cases. J Vet Intern Med. 2000;14:105. 583. Reef V.B., Freeman D., Gentile D. Successful treatment of pericarditis in a horse. J Am Vet Med Assoc. 1984;185:94. 584. Bernard W., Reef V.B., Clark E.S., et al. Pericarditis in horses: six cases (1982-1986). J Am Vet Med Assoc. 1990;196:468. 585. Harkins J.D., Hackett R.P., Ducharme N.G. Effect of furosemide on physiological variables in exercising horses. Am J Vet Res. 1993;54:2104. 586. Manohar M., Hutchens E., Coney E. Frusemide attenuates the exercise-induced rise in pulmonary capillary blood pressure in horses. Equine Vet J. 1994;26:51. 587. Johansson A.M., Gardner S.Y., Levine J.F., et al. Pharmaco*kinetics and pharmacodynamics of furosemide after oral administration to horses. J Vet Intern Med. 2004;18:739. 588. Johansson A.M., Gardner S.Y., Levine J.F., et al. Furosemide continuous rate infusion in the horse: evaluation of enhanced e cacy and reduced side e ects. J Vet Intern Med. 2003;17:887. 589. Baggot J.D., The pharmacological basis of cardiac drug selection for use in horses. Equine Vet J, Suppl 19, September 1995, 97-100. 590. Pearson E.G., Ayers J.W., Wood G.L., et al. Digoxin toxicity in a horse. Compend Cont Educ Pract Vet. 1987;9:958. 591. Gardner S.Y., Atkins C.E., Sams R.A., et al. Characterization of the pharmaco*kinetic and pharmacodynamic properties of the angiotensin-converting enzyme inhibitor, enalapril, in

horses. J Vet Intern Med. 2004;18:231. 592. Sleeper M.M., McDonnell S.M., Ely J.J., et al. Chronic oral therapy with enalapril in normal ponies. J Vet Cardiol. 2008;10:111. 593. Luciani A., Civitella C., Santori D., et al. Haemodynamic effects in healthy horses treated with an ACE-inhibitor (ramipril). Vet Res Commun. 2007;31(Suppl1):297. 594. Gehlen H., Vieht J.C., Stadler P. E ects of the ACE inhibitor quinapril on echocardiographic variables in horses with mitral valve insu ciency. J Vet Med A Physiol Pathol Clin Med. 2003;50:460. 595. Guglielmini C., Giuliani A., Testoni S., et al. Use of an ACE inhibitor (ramipril) in a horse with congestive heart failure. Equine Vet Educ. 2002;14:297. 596. Bonagura J.D. Congenital heart disease. In Robinson N.E., editor: Current therapy in equine medicine, ed 6, St Louis: WB Saunders, 2009. 597. Buergelt C.D., Carmichael J.A., Tashjian R.J., et al. Spontaneous rupture of the left pulmonary artery in a horse with patent ductus arteriosus. J Am Vet Med Assoc. 1970;157:313. 598. Cha n M.K., Miller M.W., Morris E.L. Double outlet right ventricle and other associated congenital cardiac anomalies in an American Miniature horse foal. Equine Vet J. 1992;24:402. 599. Cottrill C.M., Rossdale P.D. A comparison of congenital heart disease in horses and man. Equine Vet J. 1992;24:338. 600. Cox V.S., Weber A.F., Lima A., et al. Left cranial vena cava in a horse. Anat Histol Embryol. 1991;20:37. 601. Glazier D.B., Farrelly B.T., Neylon J.F. Patent ductus arteriosus in an eight-month-old foal. Ir Vet J. 1974;28:12. 602. Glazier D.B., Farrelly B.T., O’Connor J. Ventricular septal defect in a 7-year-old gelding. J Am Vet Med Assoc. 1975;167:49. 603. Greene H.J., Wray D.D., Greenway G.A. Two equine congenital cardiac anomalies. Ir Vet J. 1975;29:115. 604. Guarda F., Rattazzi C., Appina S. Pathology of cardiac aneurysms in horses. Pferdeheilkunde. 1992;8:241. 605. Huston R., Saperstein G., Leipold H.W. Congenital defects in foals. J Equine Med Surg. 1977;1:146. 606. Johnson J.W., DeBowes R.M., Cox J.H., et al. Diaphragmatic hernia with a concurrent cardiac defect in an Arabian foal. J Equine Vet Sci. 1984;4:225. 607. King J.M., Flint T.J., Anderson W.I. Incomplete subaortic stenotic rings in domestic animals—a newly described congenital anomaly. Cornell Vet. 1988;78:263. 608. Koblik P.D., Hornof W.J. Use of first-pass nuclear angiocardiography to detect left-to- right cardiac shunts in the horse. Vet Radiol Ultrasound. 1987;28:177. 609. Koterba A.M., Drummond W.H., Kosch P.C.E. Equine clinical neonatology. Philadelphia: Lea & Febiger, 1990. 610. Kvart C., Carlsten J., Jeffcott L.B., et al. Diagnostic value of contrast echocardiography in the horse. Equine Vet J. 1985;17:357. 611. Lescure R., Tamzali Y. L’echocardiographie TM chez le cheval. III. Cardiopathie congenitale et troubles du rythme [TM echocardiography in the horse. III. Congenital heart disease and arrhythmias]. Point Veterinaire. 1983;15:373. 612. Platt H. Vascular malformations and angiomatous lesions in horses: a review of 10 cases. Equine Vet J. 1987;19:500. 613. Reef V.B. Equine pediatric ultrasonography. Compend Cont Educ Pract Vet. 1991;13:1277. 614. Staller G.S., Reef V.B. Aortic insufficiency associated with ventricular septal defect in a. J Am Vet Med Assoc. 1992. (submitted for publication) 615. Steyn P.F., Holland P., Hoffman J. The angiocardiographic diagnosis of a persistent truncus arteriosus in a foal. J S Afr Vet Assoc. 1989;60:106. 616. Vitums A., Bayly W.M. Pulmonary atresia with dextroposition of the aorta and ventricular septal defect in three Arabian foals. Vet Pathol. 1982;19:160. 617. Zamora C.S., Vitums A., Foreman J.H., et al. Common ventricle with separate pulmonary outflow chamber in a horse. J Am Vet Med Assoc. 1985;186:1210. 618. Zamora C.S., Vitums A., Sande R.D., et al. Multiple cardiac malformation in a horse. Anat Histol Embryol. 1988;17:95. 619. Patterson-Kane J.C., Harrison L.R. Giant right atrial diverticulum in a foal. J Vet Diagn Invest. 2002;14:335. 620. Gehlen H., Bubeck K., Stadler P. Valvular pulmonic stenosis with normal aortic root and intact ventricular and atrial septa in an Arabian horse. Equine Vet Educ. 2001;13:286. 621. Anderson R.H. The pathological spectrum of pulmonary atresia. Equine Vet Educ. 1997;9:128. 622. Young L.E., Blunden A.S., Bartram D.H., et al. Pulmonary atresia with an intact ventricular septum in a thoroughbred foal. Equine Vet Educ. 1997;9:123. 623. Anderson R. Nomenclature and classi cation: sequential segmental analysis. In: Moller J.H., Ho man H.J., editors. Pediatric cardiovascular medicine. New York: Churchill Livingstone, 2003. 624. Van Praagh R. Nomenclature and classi cation: morphologic and segmental approach to diagnosis. In: Moller J.H. H.J., editor. Pediatric cardiovascular medicine. New York: Churchill Livingstone, 2003. 625. Sleeper M.M., Palmer J.E. Echocardiographic diagnosis of transposition of the great arteries in a neonatal foal. Vet Radiol Ultrasound. 2005;46:259. 626. Valdes-Martinez A., Eades S.C., Strickland K.N., et al. Echocardiographic evidence of an aortico-pulmonary septal defect in a 4-day-old thoroughbred foal. Vet Radiol Ultrasound. 2006;47:87. 627. Cottrill C.M., O’Connor W.N., Cudd T., Rantanen N.W. Persistence of foetal circulatory pathways in a newborn foal. Equine Vet J. 1987;19:252. 628. McGurrin M.K., Physick-Sheard P.W., Southorn E. Parachute left atrioventricular valve causing stenosis and regurgitation in a Thoroughbred foal. J Vet Intern Med. 2003;17:579. 629. Coumbe K.M. Cardiac disease: endocardial fibroelastosis. Equine Vet Educ. 2002;14:81. 630. Seco D.O., Desrochers A., Hoffmann V., et al. Total anomalous pulmonary venous connection in a foal. Vet Radiol Ultrasound. 2005;46:83. 631. Cottrill C.M., Ho S.Y., O’Connor W.N., Embryological development of the equine heart. Equine Vet J, Suppl 24, June 1997, 14-18. 632. Stephen J.O., Abbott J., Middleton D.M., et al. Persistent truncus arteriosus in a Bashkir Curly foal. Equine Vet Educ. 2000;12:251. 633. Petrick S.W., Roos C.J., van N.J. Persistent right aortic arch in a horse. J S Afr Vet Assoc. 1978;49:355. 634. van der Linde-Sipman J.S., Goedegebuure S.A., Kroneman J. Persistent right aortic arch associated with a persistent left ductus arteriosus and an interventricular septal defect in a horse. Tijdschr Diergeneeskd. 1979;104(Suppl 4):189-194. 635. Kutasi O., Voros K., Biksi I., et al. Common atrioventricular canal in a newborn foal—case report and review of the literature. Acta Vet Hung. 2007;55:51. 636. Meurs K.M., Miller M.W., Hanson C., et al. Tricuspid valve atresia with main pulmonary artery atresia in an Arabian foal. Equine Vet J. 1997;29:160. 637. Button C., Gross D.R., Allert J.A., et al. Tricuspid atresia in a foal. J Am Vet Med Assoc. 1978;172:825. 638. Honnas C.M., Puckett M.J., Schumacher J. Tricuspid atresia in a Quarter Horse foal. Southwestern Vet. 1987;38:17. 639. Rooney J.R., Robertson J.L. Equine Pathology. Ames, Iowa, Iowa State University Press, 1996. p 15 640. Littlewort M.C.G. Cardiological problems in equine medicine. Equine Vet J. 1977;9:173. 641. Buergelt C.D. Equine cardiovascular pathology: an overview. Anim Health Res Rev. 2003;4:109. 642. Buhl R., Ersboll A.K., Eriksen L., et al. Use of color Doppler echocardiography to assess the development of valvular regurgitation in Standardbred trotters. J Am Vet Med Assoc. 2005;227:1630. 643. Stevens K.B., Marr C.M., Horn J.N., et al. E ect of left-sided valvular regurgitation on mortality and causes of death among a population of middle-aged and older horses. Vet

Rec. 2009;164:6. 644. Kriz N.G., Hodgson D.R., Rose R.J. Prevalence and clinical importance of heart murmurs in racehorses. J Am Vet Med Assoc. 2000;216:1441. 645. Haldar S.M., O’Gara P.T. Infective endocarditis: diagnosis and management. Nat Clin Pract Cardiovasc Med. 2006;3:310. 646. Bertone J.J., Dill S.G. Traumatic gastropericarditis in a horse. J Am Vet Med Assoc. 1985;187:742. 647. Wagner P.C. Pericarditis. In: Robinson N.E., editor. Current therapy in equine medicine. ed 1. Philadelphia: WB Saunders; 1983:149-151. 648. Robinson J.A., Marr C.M., Reef V.B., et al. Idiopathic, aseptic, e usive, brinous, nonconstrictive pericarditis with tamponade in a Standardbred lly. J Am Vet Med Assoc. 1992;201:1593. 649. Bradfield T. Traumatic pericarditis in a horse. Southwestern Vet. 1970;23:145. 650. Voros K., Felkai C., Szilagyi Z., et al. Two-dimensional echocardiographically guided pericardiocentesis in a horse with traumatic pericarditis. J Am Vet Med Assoc. 1953;198:1991. 651. Buergelt C.D., Wilson J.H., Lombard C.W. Pericarditis in horses. Compend Cont Educ Pract Vet. 1990;12:872. 652. Morley P.S., Chirinotrejo M., Petrie L., et al. Pericarditis and pleuritis caused by Mycoplasma felis in a horse. Equine Vet J. 1996;28:237. 653. Wilson J.H., Olson E.J., Haugen E.W., et al. Systemic blastomycosis in a horse. J Vet Diagn Invest. 2006;18:615. 654. May K.A., Cheramie H.S., Howard R.D., et al. Purulent pericarditis as a sequela to clostridial myositis in a horse. Equine Vet J. 2002;34:636. 655. Marcus L.C., Ross J.N. Microscopic lesions in the hearts of aged horses and mules. Vet Pathol. 1967;4:162. 656. King J.M., Roth L., Haschek W.M. Myocardial necrosis secondary to neural lesions in domestic animals. J Am Vet Med Assoc. 1982;180:144. 657. Fujii Y., Watanabe H., Yamamoto T., et al. Serum creatine kinase and lactate dehydrogenase isoenzymes in skeletal and cardiac muscle damage in the horse. Bull Equine Res Institute. 1983;20:87-96. 658. Shawley R.V., Rolf L.L.J. Experimental cantharidiasis in the horse. Am J Vet Res. 1984;45:2261. 659. Hulland T.J. Leptomeric fibrils in the myocardial fibers of a foal. Vet Pathol. 1988;25:175. 660. Schmitz D.G. Cantharidin toxicosis in horses. J Vet Intern Med. 1989;3:208. 661. Freestone J.F., Williams M.M., Norwood G. Thoracic haemangiosarcoma in a 3-year-old horse. Aust Vet J. 1990;67:269. 662. Reef V.B. Myocardial disease. In: Robinson N.E., editor. Current therapy in equine medicine. ed 3. Philadelphia: WB Saunders; 1992:393-395. 663. Weldon A.D., Step D.L., Moise N.S. Lymphosarcoma with myocardial infiltration in a mare. Vet Med. 1992;87:595. 664. Perkins J.D., Bowen I.M., Else R.W., et al. Functional and histopathological evidence of cardiac parasympathetic dysautonomia in equine grass sickness. Vet Rec. 2000;146:246. 665. Slack J.A., McGuirk S.M., Erb H.N., et al. Biochemical markers of cardiac injury in normal, surviving septic, or nonsurviving septic neonatal foals. J Vet Intern Med. 2005;19:577. 666. Delesalle C., VanLoon G., Nollet H., et al. Tumor-induced ventricular arrhythmia in a horse. J Vet Intern Med. 2002;16:612. 667. Nout Y.S., Hinchcliff K.W., Bonagura J.D., et al. Cardiac amyloidosis in a horse. J Vet Intern Med. 2003;17:588. 668. Cornelisse C.J., Schott H.C., Olivier N.B., et al. Concentration of cardiac troponin I in a horse with a ruptured aortic regurgitation jet lesion and ventricular tachycardia. J Am Vet Med Assoc. 2000;217:231. 669. O’Brien P.J., Landt Y., Ladenson J.H. Differential reactivity of cardiac and skeletal muscle from various species in a cardiac troponin I immunoassay. Clin Chem. 1997;43:2333. 670. Smith G.W., Constable P.D., Foreman J.H., et al. Cardiovascular changes associated with intravenous administration of fumonisin B1 in horses. Am J Vet Res. 2002;63:538. 671. Phillips W., Giguere S., Franklin R.P., et al. Cardiac troponin I in pastured and race-training Thoroughbred horses. J Vet Intern Med. 2003;17:597. 672. Diana A., Guglielmini C., Candini D., et al. Cardiac arrhythmias associated with piroplasmosis in the horse: a case report. Vet J. 2007;174:193. 673. Galey F.D., Holstege D.M., Plumlee K.H., et al. Diagnosis of oleander poisoning in livestock. J Vet Diagn Invest. 1996;8:358. 674. Tiwary A.K., Puschner B., Kinde H., et al. Diagnosis of Taxus (yew) poisoning in a horse. J Vet Diagn Invest. 2005;17:252. 675. Beasley V.R., Wolf G.A., Fischer D.C., et al. Cantharidin toxicosis in horses. J Am Vet Med Assoc. 1983;182:283. 676. Dickinson C.E., Traub-Dargatz J.L., Dargatz D.A., et al. Rattlesnake venom poisoning in horses: 32 cases (1973-1993). J Am Vet Med Assoc. 1996;208:1866. 677. Aleman M., Magdesian K.G., Peterson T.S., et al. Salinomycin toxicosis in horses. J Am Vet Med Assoc. 2007;230:1822. 678. Vitums A. Origin of the aorta and pulmonary trunk from the right ventricle in a horse. Vet Pathol. 1970;7:482. 679. Vitums A., Grant B.D., Stone E.C., et al. Transposition of the aorta and atresia of the pulmonary trunk in a horse. Cornell Vet. 1973;63:41. 680. Hilbert B.J., Rendano V.T. Venous aneurysm in a horse. J Am Vet Med Assoc. 1975;167:394. 681. Harrington D.D., Page E.H. Acute vitamin D3 toxicosis in horses: case reports and experimental studies of the comparative toxicity of vitamins D2 and D3. J Am Vet Med Assoc. 1983;182:1358. 682. Lombardo de Barros C.S. Aortic body adenoma in a horse. Aust Vet J. 1983;60:61. 683. Knezevic P.F., Fessl L. Thrombectomy of the descending aorta in the horse. Tierarztl Prax (Suppl. 1985;1:94). 684. Spier S. Arterial thrombosis as the cause of lameness in a foal. J Am Vet Med Assoc. 1985;187:164. 685. Laging C., Grabner A. Malignant haemangioendothelioma in a horse. Pferdeheilkunde. 1988;4:273. 686. Parks A.H., Guy B.L., Rawlings C.A., et al. Lameness in a mare with signs of arteriovenous fistula. J Am Vet Med Assoc. 1989;194:379. 687. Wallace K.D., Selcer B.A., Tyler D.E., et al. In vitro ultrasonographic appearance of the normal and verminous equine aorta, cranial mesenteric artery, and its branches. Am J Vet Res. 1989;50:1774. 688. Johnson B., Baldwin C., Timoney P., et al. Arteritis in equine fetuses aborted due to equine viral arteritis. Vet Pathol. 1991;28:248. 689. Gardner S, Reef VB: Ultrasonographic evaluation of jugular vein thrombophlebitis in horses. 1992. 690. McDonnell S.M., Love C.C., Martin B.B., et al. ejacul*tory failure associated with aortic-iliac thrombosis in two stallions. J Am Vet Med Assoc. 1992;200:954. 691. Welch R.D., Dean P.W., Miller M.W. Pulsed spectral Doppler evaluation of a peripheral arteriovenous fistula in a horse. J Am Vet Med Assoc. 1992;200:1360. 692. Markey B.K.J., Carter M.E., Quinn P.J. Notes on equine viral arteritis—recent outbreaks in Britain. Ir Vet J. 1993;46:104. 693. Timoney P.J., McCollum W.H. Equine viral arteritis. Vet Clin North Am Equine Pract. 1993;9:295. 694. West J.B., Mathieu-Costello O., Jones J.H., et al. Stress failure of pulmonary capillaries in racehorses with exercise-induced pulmonary hemorrhage. J Appl Physiol. 1993;75:1097. 695. Yanai T., Masegi T., Ishikawa K., et al. Spontaneous vascular mineralization in the brain of horses. J Vet Med Sci. 1996;58:35. 696. Birks E.K., Mathieu-Costello O., Fu Z., et al. Very high pressures are required to cause stress failure of pulmonary capillaries in Thoroughbred racehorses. J Appl Physiol.

1997;82:1584. 697. Carr E.A., Carlson G.P., Wilson W.D., et al. Acute hemorrhagic pulmonary infarction and necrotizing pneumonia in horses—21 cases (1967-1993). J Am Vet Med Assoc. 1997;210:1774. 698. Shirai W., Momotani E., Sato T., et al. Dissecting aortic aneurysm in a horse. J Comp Pathol. 1999;120:307. 699. Hoskinson J.J., Wooten P., Evans R. Nonsurgical removal of a catheter embolus from the heart of a foal. J Am Vet Med Assoc. 1991;199:233. 700. Lees M.J., Read R.A., Klein K.T., et al. Surgical retrieval of a broken jugular catheter from the right ventricle of a foal. Equine Vet J. 1989;21:384. 701. Fales-Williams A., Sponseller B., Flaherty H. Idiopathic arterial medial calcification of the thoracic arteries in an adult horse. J Vet Diagn Invest. 2008;20:692. 702. Imaizumi K., Nakamura T., Kiryu K., et al. Morphological changes of the aorta and pulmonary artery in Thoroughbred racehorses. J Comp Pathol. 1989;101:1. 703. Arroyo L.G., Hayes M.A., DeLay J., et al. Arterial calcification in race horses. Vet Pathol. 2008;45:617. 704. Guglick M.A., MacAllister C.G., Ewing P.J., et al. Thrombosis resulting in rectal perforation in a horse. J Am Vet Med Assoc. 1996;209:1125. 705. Johansson A.M., Gardner S.Y., Atkins C.E., et al. Cardiovascular e ects of acute pulmonary obstruction in horses with recurrent airway obstruction. J Vet Intern Med. 2007;21:302. 706. Rantanen N.W., Byars T.D., Hauser M.L., et al. Spontaneous contrast and mass lesions in the hearts of racehorses: ultrasound diagnosis—preliminary data. J Equine Vet Sci. 1984;4:220. 707. Barrelet A. Aorto-iliac thrombosis in a breeding stallion and an eventer mare. Equine Vet Educ. 1993;5:86. 708. Mouchot E., Desbrosse F. Aorto-iliac thrombosis in a horse. Clinical examination and use of ultrasonography as a diagnostic aid. Pratique Vétérinaire Équine. 1994;26:147. 709. Dyson S.J., Worth L. Aortoiliacofemoral thrombosis. In: Robinson N.E., editor. Current therapy in equine medicine. ed 4. Philadelphia: WB Saunders; 1997:267-268. 710. Moore L.A., Johnson P.J., Bailey K.L. Aorto-iliac thrombosis in a foal. Vet Rec. 1998;142:459. 711. Duggan V.E., Holbrook T.C., Dechant J.E., et al. Diagnosis of aorto-iliac thrombosis in a quarter horse foal using Doppler ultrasound and nuclear scintigraphy. J Vet Intern Med. 2004;18:753. 712. Hilton H., Aleman M., Textor J., et al. Ultrasound-guided balloon thrombectomy for treatment of aorto-iliac-femoral thrombosis in a horse. J Vet Intern Med. 2008;22:679. 713. Trachsel D., Cohausz O., Scharf G., et al. Aorto-iliac thrombosis in a gelding treated with the anticoagulant phenprocoumon (Marcoumar). Schweiz Arch Tierheilkd. 2008;150:613. 714. Ross M.W., Maxson A.D., Stacy V.S., et al. First-pass radionuclide angiography in the diagnosis of aortoiliac thromboembolism in a horse. Vet Radiol Ultrasound. 1997;38:226. 715. Warmerdam E.P. Ultrasonography of the femoral artery in six normal horses and three horses with thrombosis. Vet Radiol Ultrasound. 1998;39:137. 716. Guglielmini C., Bernardini D. Echo-Doppler findings of a carotid-jugular fistula in a foal. Vet Radiol Ultrasound. 2003;44:310. 717. Reef V.B. Arrhythmias. In: Marr C.M., editor. Cardiology of the horse. London: WB Saunders; 1999:179-209. 718. Glazier D.B., Nicholson J.A., Kelly W.R. Atrial fibrillation in the horse. Ir Vet J. 1959;13:47. 719. Glendinning S.A. The use of quinidine sulfate for the treatment of atrial fibrillation in horses. Vet Rec. 1965;77:951. 720. Glendinning S.A. Significance of clinical abnormalities of the heart in soundness. Equine Vet J. 1972;4:21. 721. Rose R.J., Davis P.E. Treatment of atrial fibrillation in three racehorses. Equine Vet J. 1977;9:68. 722. Oka Y. Studies on uses of quinidine sulfate for treatment of atrial fibrillation in heavy horses. Jpn J Vet Res. 1985;33:89. 723. Cipone M., Venturoli M. Atrial fibrillation in five horses. Changes in the ECG during quinidine therapy. Summa Phytopathol. 1986;3:53. 724. Petch M.C. Atrial fibrillation: bad news for man and horse? Equine Vet J. 1986;18:3. 725. Shaftoe S., McGuirk S.M. Valvular insufficiency in a horse with atrial fibrillation. Compend Cont Educ Pract Vet. 1987;9:203-208. 726. Machida N., Yasuda J., Too K. Three cases of paroxysmal atrial fibrillation in the Thoroughbred newborn foal. Equine Vet J. 1989;21:66. 727. Muir W.W.3rd, Reed S.M., McGuirk S.M. Treatment of atrial fibrillation in horses by intravenous administration of quinidine. J Am Vet Med Assoc. 1990;197:1607. 728. Stewart G.A., Fulton L.J., McKellar C.D. Idiopathic atrial fibrillation in a champion Standardbred racehorse. Aust Vet J. 1990;67:187. 729. Matsuda H. Treatment of atrial fibrillation in horses. Jpn J Vet Res. 1992;40:44. 730. Marr C.M., Reef V.B., Reimer J.M., et al. An echocardiographic study of atrial fibrillation in horses: before and after conversion to sinus rhythm. J Vet Intern Med. 1995;9:336. 731. Reef V.B., Reimer J.M., Spencer P.A. Treatment of atrial fibrillation in horses—new perspectives. J Vet Intern Med. 1995;9:57. 732. Schwarzwald C.C., Sams R.A. Determination of plasma protein binding of diltiazem in horses by ultrafiltration. J Vet Pharmacol Ther. 2006;29:579. 733. Schwarzwald C.C., Sams R.A., Bonagura J.D. Pharmaco*kinetics of the calcium-channel blocker diltiazem after a single intravenous dose in horses. J Vet Pharmacol Ther. 2006;29:165. 734. Amada A., Kiryu K. Atrial fibrillation in the race horse. Heart Vessels Suppl. 1987;2:2. 735. Bertone J.J., Traub-Dargatz J.L., Wingfield W.E. Atrial fibrillation in a pregnant mare: treatment with quinidine sulfate. J Am Vet Med Assoc. 1987;190:1565. 736. Blissitt K.J. Diagnosis and treatment of atrial fibrillation. Equine Vet Educ. 1999;11:11. 737. Kuwahara M., Hiraga A., Nishimura T., et al. Power spectral analysis of heart rate variability in a horse with atrial fibrillation. J Vet Med Sci. 1998;60:111. 738. Ohmura H., Hiraga A., Aida H., et al. Determination of oral dosage and pharmaco*kinetic analysis of flecainide in horses. J Vet Med Sci. 2001;63:511. 739. van Loon G., Jordaens L., Muylle E., et al. Intracardiac overdrive pacing as a treatment of atrial flutter in a horse. Vet Rec. 1998;142:301. 740. Schwarzwald C.C., Schober K.E., Bonagura J.D. Echocardiographic evidence of left atrial mechanical dysfunction after conversion of atrial brillation to sinus rhythm in 5 horses. J Vet Intern Med. 2007;21:820. 741. Gehlen H., Stadler P. Comparison of systolic cardiac function before and after treatment of atrial brillation in horses with and without additional cardiac valve insu ciencies. Vet Res Commun. 2004;28:317. 742. Frye M.A., Selders C.G., Mama K.R., et al. Use of biphasic electrical cardioversion for treatment of idiopathic atrial brillation in two horses. J Am Vet Med Assoc. 2002;220:1039. 743. McGurrin M.K., Physick-Sheard P.W., Kenney D.G., et al. Transvenous electrical cardioversion in equine atrial brillation: technique and successful treatment of 3 horses. J Vet Intern Med. 2003;17:715. 744. McGurrin M.K., Physick-Sheard P.W., Kenney D.G. How to perform transvenous electrical cardioversion in horses with atrial fibrillation. J Vet Cardiol. 2005;7:109. 745. McGurrin M.K., Physick-Sheard P.W., Kenney D.G., et al. Transvenous electrical cardioversion of equine atrial brillation: technical considerations. J Vet Intern Med. 2005;19:695.

746. Bellei M.H., Kerr C., McGurrin M.K., et al. Management and complications of anesthesia for transvenous electrical cardioversion of atrial brillation in horses: 62 cases (20022006). J Am Vet Med Assoc. 2007;231:1225. 747. McGurrin M.K., Physick-Sheard P.W., Kenney D.G. Transvenous electrical cardioversion of equine atrial brillation: patient factors and clinical results in 72 treatment episodes. J Vet Intern Med. 2008;22:609. 748. DeClercq D., VanLoon G., Schauvliege S., et al. Transvenous electrical cardioversion of atrial brillation in six horses using custom made cardioversion catheters. Vet J. 2008;177:198. 749. Birettoni F., Porciello F., Rishniw M., et al. Treatment of chronic atrial fibrillation in the horse with flecainide: personal observation. Vet Res Commun. 2007;31(Suppl 1):273. 750. Ohmura H., Nukada T., Mizuno Y., et al. Safe and efficacious dosage of flecainide acetate for treating equine atrial fibrillation. J Vet Med Sci. 2000;62:711. 751. Ohmura H., Hiraga A., Aida H., et al. Influence of quinidine and flecainide on autonomic nervous activity in Thoroughbred horses. Vet Rec. 2003;152:114. 752. Risberg A.I., McGuirk S.M. Successful conversion of equine atrial fibrillation using oral flecainide. J Vet Intern Med. 2006;20:207. 753. VanLoon G., Blissitt K.J., Keen J.A., et al. Use of intravenous flecainide in horses with naturally occurring atrial fibrillation. Equine Vet J. 2004;36:609. 754. DeClercq D., VanLoon G., Tavernier R., et al. Use of propafenone for conversion of chronic atrial fibrillation in horses. Am J Vet Res. 2009;70:223. 755. DeClercq D., VanLoon G., Baert K., et al. Intravenous amiodarone treatment in horses with chronic atrial fibrillation. Vet J. 2006;172:129. 756. DeClercq D., VanLoon G., Baert K., et al. Treatment with amiodarone of refractory ventricular tachycardia in a horse. J Vet Intern Med. 2007;21:878. 757. DeClercq D., VanLoon G., Baert K., et al. Effects of an adapted intravenous amiodarone treatment protocol in horses with atrial fibrillation. Equine Vet J. 2007;39:344. 758. Trachsel D., Tschudi P., Portier C.J., et al. Pharmaco*kinetics and pharmacodynamic e ects of amiodarone in plasma of ponies after single intravenous administration. Toxicol Appl Pharmacol. 2004;195:113. 759. McGuirk S.M., Muir W.W.3rd, Sams R.A. Pharmaco*kinetic analysis of intravenously and orally administered quinidine in horses. Am J Vet Res. 1981;42:938. 760. Parraga M.E., Kittleson M.D., Drake C.M., Quinidine administration increases steady state serum digoxin concentration in horses. Equine Vet J, Suppl 19, September 1995, 114119. 761. Puigdemont A., Riu J.L., Guitart R., et al. Propafenone kinetics in the horse. Comparative analysis of compartmental and noncompartmental models. J Pharmacol Methods. 1990;23:79. 762. VanLoon G., DeClercq D., Tavernier R., et al. Transient complete atrioventricular block following transvenous electrical cardioversion of atrial brillation in a horse. Vet J. 2005;170:124. 763. Cornick J.L., Hartsfield S.M., Miller M. ECG of the month. Premature ventricular complexes in an anesthetized colt. J Am Vet Med Assoc. 1990;196:420. 764. Gabriel F., Lekeux P. Cardiac arrhythmias encountered in 159 Belgian riding horses. Ann Med Vet. 1986;130:205. 765. Kiryu K., Machida N., Kashida Y., et al. Pathologic and electrocardiographic findings in sudden cardiac death in racehorses. J Vet Med Sci. 1999;61:921. 766. Marr C.M., Reef V.B. ECG of the month [Multifocal ventricular tachycardia in a horse]. J Am Vet Med Assoc. 1991;198:1533. 767. Miller P.J., Rose R.J., Hoffman K., et al. Idioventricular tachycardia in a horse. Aust Vet J. 1987;64:55. 768. Nielsen I.L. Ventricular tachycardia in a Thoroughbred racehorse. Aust Vet J. 1990;67:140. 769. Reimer J.M., Reef V.B., Sweeney R.W. Ventricular arrhythmias in horses: 21 cases (1984-1989). J Am Vet Med Assoc. 1992;201:1237. 770. Vrins A., Doucet M., DeRoth L. Paroxysmal ventricular tachycardia in a horse. Medecin Veterinaire du Quebec. 1989;19:79. 771. Garcia-Lopez J.M., Provost P.J., Rush J.E., et al. Prevalence and prognostic importance of hypomagnesemia and hypocalcemia in horses that have colic surgery. Am J Vet Res. 2001;62:7. 772. Meyer G.A., Lin H.C., Hanson R.R., et al. Effects of intravenous lidocaine overdose on cardiac electrical activity and blood pressure in the horse. Equine Vet J. 2001;33:434. 773. Ellis E.J., Ravis W.R., Malloy M., et al. The pharmaco*kinetics and pharmacodynamics of procainamide in horses after intravenous administration. J Vet Pharmacol Ther. 1994;17:265. 774. Gasthuys F., Parmentier D., Goossens L., et al. A preliminary study on the e ects of atropine sulphate on bradycardia and heart blocks during romi dine sedation in the horse. Vet Res Commun. 1990;14:489. 775. Muir W.W.3rd, McGuirk S.M. Ventricular preexcitation in two horses. J Am Vet Med Assoc. 1983;183:573. 776. Nihouannen J.C., Sevestre J., Dorso Y., et al. Implantation of a cardiac pacemaker into horses. I. Equipment and techniques. Rev Med Vet (Toulouse). 1984;135:91. 777. Nihouannen J.C., Sevestre J., Dorso Y., et al. Implantation of a cardiac pacemaker into horses. II. Postoperative monitoring of a pacemaker with epicardial and myocardial electrodes in a pony. Rev Med Vet (Toulouse). 1984;135:165. 778. Reef V.B., Clark E.S., Oliver J.A., et al. Implantation of a permanent transvenous pacing catheter in a horse with complete heart block and syncope. J Am Vet Med Assoc. 1986;189:449. 779. van Loon G., Laevens H., Deprez P. Temporary transvenous atrial pacing in horses: threshold determination. Equine Vet J. 2001;33:290. 780. Castex A.M., Bertone J.J. ECG of the month. Sinus tachycardia and hyperkalemia in a horse. J Am Vet Med Assoc. 1989;194:654. 781. Epstein V. Relationship between potassium administration, hyperkalaemia and the electrocardiogram: an experimental study. Equine Vet J. 1984;16:453. 782. Hardy J. ECG of the month. Hyperkalemia in a mare. J Am Vet Med Assoc. 1989;194:356. When appears, a corresponding video can be found on the DVD inside this book. References 5, 27, 28, 38, 45, 48, 56, 91, 109, 124-154. † References 4, 31, 83, 84, 90, 91, 109, 111, 125, 139, 135, 144, 148, 169-181. ∗ References 4, 34, 58, 111, 152, 218, 219. ∗ References 13, 24, 26, 46, 47, 246-259. ∗ References 209, 210, 213, 216, 234, 259, 264, 286-302. ∗ References 32, 111, 112, 196, 256, 333-335. ∗ References 3, 16, 19, 23, 25, 27, 32, 36, 48, 77, 111, 112, 134, 182, 196, 238, 256, 338-353. ∗ ∗

References 3, 19, 52, 57-65, 77, 219, 333-335, 340, 341, 343. References 2, 4, 33, 111, 344, 361-363. ∗ References 18, 19, 32, 36, 134, 256, 333, 340, 343, 351, 364-366. ∗ References 371, 374, 375, 379, 380, 382-386, 389, 391, 395. ∗ References 122, 204, 273, 311, 313, 317, 327-330, 332, 337, 394, 400-425. ∗ References 1, 4, 6-8, 11, 30, 33, 35, 40, 43, 72, 77, 111, 188, 194, 195, 336, 344, 360, 436-470. ∗ References 1, 4, 111, 360, 445, 446, 467. ∗ References 14, 41, 49, 51, 77, 83, 85, 90, 91, 94, 109, 122, 128-130, 137-139, 142, 143, 145, 149, 150, 163, 179, 183, 201, 220, 221, 230, 231, 233, 241, 257, 262, 276-280, 282, 283, 303-305, 342, 354, 251-254, 306-308, 479-519. ∗ References 50, 51, 90, 303, 403, 523, 524. ∗ References 149, 150, 233, 241, 257, 278, 354, 492. ∗ References 37, 42, 46, 47, 71, 214, 258, 260, 261, 269, 270, 276, 280, 282, 284, 288, 309, 321, 368, 372, 376, 377, 381, 386, 393, 406, 498, 265-267, 277-279, 318-320, 530-552. ∗ References 239, 240, 321, 406, 530-533, 553. ∗ References 265, 312, 406, 497, 536, 553. ∗ References 260, 261, 265-267, 272-280, 282, 284, 285, 532, 534, 535, 551, 557-560. ∗ References 87, 88, 105, 109, 112, 137, 146-148, 152, 186, 294, 299, 300, 309, 342, 504, 556, 564, 565, 570-580. ∗ References 146, 286, 287, 289, 290, 294, 295, 297, 299, 363, 573, 582, 589, 590. ∗ References 4, 14, 27, 28, 61, 90, 111, 112, 126, 127, 131, 134, 145, 151, 155-168, 171, 178, 265, 319, 334, 342, 367, 473, 479, 480, 482, 487, 504, 513, 528, 571, 575, 576, 578, 595-622. ∗ References 131, 145, 151, 157, 162, 165, 479, 482, 504, 513, 528, 576, 596, 602, 608, 614-617. ∗ References 131, 159, 161, 163, 513, 528, 578. ∗ References 156, 159, 165, 168, 265, 367, 575, 597, 601. ∗ References 5, 18, 19, 38, 45, 48, 124, 127, 128, 130, 132, 136, 141, 149, 182, 183, 186, 366, 483, 496, 614, 639-641. † References 31, 90, 91, 109, 125, 135, 138, 139, 142-144, 147, 148, 169, 170, 172-178, 185, 509, 642. ∗ References 129, 130, 145, 278, 346, 354, 509. ∗ References 4, 38, 45, 48, 77, 109, 111, 124, 125, 128, 130, 132, 136, 138, 139, 141, 143, 144, 147, 148, 186, 366, 483. ∗ References 129, 130, 354, 505, 509, 510. ∗ References 4, 18, 77, 111, 125, 127, 130, 136, 138, 139, 141, 142, 366, 496, 614, 640. ∗ References 84, 130, 136, 138, 139, 141, 142, 152, 496, 501, 504. ∗ References 4, 77, 111, 125, 130, 149, 179, 180, 182, 184, 230, 354, 639, 644. † References 151, 156, 161-163, 165, 366, 637, 638. ∗ References 4, 31, 77, 90, 109, 111, 125, 135, 138, 139, 142-144, 147, 148, 169, 170, 172-180. ∗ References 4, 80-84, 86-95, 100, 125, 583, 584, 646-653. ∗ References 4, 28, 29, 77, 98, 106-108, 110, 111, 114, 115, 121, 125, 432, 604, 655-664. ∗ References 4, 34, 52, 55-58, 62, 64, 66, 103, 106-108, 111, 125, 137, 152, 171, 207, 218, 219, 221-228, 316, 364, 368, 396-398, 476, 489, 501, 506, 549, 597, 600, 612, 678, 698. ∗ References 56, 222-227, 396, 397, 683, 690, 707-713. ∗ References 2, 33, 39, 42, 46, 55, 109, 134, 140, 174, 190, 191, 193, 195, 201, 293, 363, 469, 495, 546, 556, 568, 574, 576, 578, 718-731. ∗ References 2, 33, 39, 42, 46, 55, 109, 134, 140, 146, 174, 190, 191, 193, 201, 293, 361-363, 427, 428, 469, 495, 546, 556, 568, 574, 576, 578, 580, 718, 719, 721-731, 734-739. ∗ References 110, 199, 359, 432, 447, 469, 489, 668, 672, 763-770. ∗ References 173, 192, 198, 344, 470, 764, 774. ∗ ∗

CHAPTER 11 Disorders of the Musculoskeletal System Jennifer M. MacLeay

FUNDAMENTALS OF THE MUSCULOSKELETAL SYSTEM, DIAGNOSTIC TECHNIQUES, AND CLASSIFICATION OF MYOPATHIES Jennifer M. MacLeay The innate ability of the horse to learn and trust is a special attribute that allows human beings to take advantage of their extraordinary strength and agility. Horses have contributed to human life by assisting in agriculture, business, travel, and recreation. Although e ective locomotion depends on the coordination of many body systems, movement ultimately depends on skeletal muscle. This chapter endeavors to review the many disorders that affect the muscular system of the horse.

STRUCTURE Mammalian skeletal muscle consists of approximately 75% water, 18% to 22% protein, 1% carbohydrate, and 1% mineral, with variable lipid content. Depending on the breed and type of horse, between 44% and 53% of the live weight of a mature 500-kg horse has been estimated to be muscle.1 The myo ber, or individual muscle cell, is the fundamental building block of muscle, and each is a fusiform, multinucleated cell. Combined, myo bers constitute 75% to 90% of the volume of muscle. The myo bers are arranged in parallel along the length of the muscle such that the force of contraction is additive. The rest of the muscle tissue is made up of broblasts, capillaries, adipose cells, nerves, and connective tissue bers. The composition of any muscle varies depending on the muscle surveyed and the overall tness level, age, and breed of the horse. Muscles begin and end with tendons of various sizes, which attach the muscle to bone. Golgi tendon organs act as end organs of muscle sense and are found in the major tendinous origins and insertions. The nerve and blood vessels that supply an individual muscle typically enter near the midpoint of the muscle belly at a region called the neurovascular hilum. The nerve bundle splits o into individual nerves once it enters the muscle. A single nerve contacts several myo bers such that they contract in unison when stimulated (motor unit). A single nerve innervates each myofiber at a single point known as the motor end plate. Each myo ber is bounded by a complex membrane called the sarcolemma, which invagin*tes into the muscle ber at numerous points to form the T, or transverse, tubules. The T tubules terminate within each muscle cell in proximity to the sarcoplasmic reticulum of the cell, contacting the myo brils between the A and I band twice within each sarcomere. The T tubule lies between two terminal cisterns of the sarcoplasmic reticulum. Together these three structures form the triad. A basem*nt membrane surrounds the sarcolemma and attaches closely to a layer of connective tissue called the endomysium. The endomysium is continuous with the perimysium, which surrounds groups or bundles of muscle bers. The perimysium, in turn, is continuous with the epimysium, which surrounds the entire muscle. Satellite cells, which consist of a simple cell membrane around a nucleus with a minimal amount of cytoplasm and few mitochondria, lie in shallow indentations on the myo ber surface between the sarcolemma and the basem*nt membrane. Satellite cells give rise to new myo bers after rhabdomyolysis (dissolution of muscle bers). The nuclei of the satellite cells and mature myo bers are oriented to the long axis of the muscle ber. The muscle ber nuclei lie at the periphery of the cells just within the sarcolemma. Within each myo ber are parallel arrays of myo brils, the basic contractile units of muscle. Myo brils are composed of end-to-end stacks of rodlike structures or contractile units called sarcomeres. Each stack of sarcomeres forms a long lament, and arrays of laments form the myo brils. Sarcomeres are composed of the contractile proteins actin, myosin, tropomyosin, and troponin. Troponin is a complex of troponin T, troponin I, and troponin C, whereas myosin is made of heavy and light chains and a globular head that contains the myosin adenosine triphosphatase (ATPase). Figure 11-1 shows an electron micrograph of several myo brils. The alternating bands of light and dark are notable. The dark A band is composed of the thick myosin laments and is transected by a darker line called the M line. Each myosin lament is surrounded by, and interdigitates with, six thin laments composed of actin, tropomyosin, and troponin. Areas where the thin filaments are not overlapping with myosin form the lighter I band. The dark line running across the I band is called the Z line. A sarcomere is de ned as the region between each Z line. At the Z lines the protein α-actinin anchors actin. Because of the di ering way these structures absorb light, the regular alternation of A bands and I bands in adjacent myo brils produces the striated appearance characteristic of skeletal muscle bers as viewed with light microscopy. Mitochondria t into the small spaces around the myo brils and produce the adenosine triphosphate (ATP) necessary for the contraction-relaxation cycle. The number, size, and shape of mitochondria vary depending on the tness of the individual and the muscle fiber type predominant within the muscle sampled.1-3

FIGURE 11-1

Electron micrograph of several muscle myofibrils (×45,000.)

MUSCLE CONTRACTION To understand the pathophysiologic features of many muscle disorders, de ning the sequential steps that lead to contraction is valuable. This process is known as excitation-contraction coupling.

Excitation Excitation begins with the origination of a nerve impulse within the central nervous system. The impulse exits the spinal cord, travels along a motor neuron, and arrives at the motor end plate of an individual myo ber. Motor end plates are located in deep primary clefts or folds of the sarcolemma. The nerve impulse arrives at the motor end plate, releasing acetylcholine, which traverses the synaptic cleft and binds to nicotinic receptors. The binding of acetylcholine produces a conformational change in the nicotinic receptor, which results in the opening of ion channels, producing increased sodium and potassium conductance in the end plate membrane and causing depolarization and generation of an end plate action potential. The T tubule system propagates the action potential across the sarcolemma in all directions and carries it deep into the muscle ber. Di usion of acetylcholine from the receptors or hydrolysis by acetylcholine esterase causes depolarization to cease.

Excitation-contraction coupling Coupling of excitation with contraction begins with the arrival of the action potential at the triad region, which causes opening of sodium channels embedded in the T tubule membrane and rapid in ux of sodium into the cell. The rapid in ux through the sodium channel results in a change of polarity across the membrane that we associate with depolarization. As a result the inside of the sarcolemma becomes more positive relative to the outside of the membrane. Depolarization of the T tubule membrane triggers a voltage-gated calcium channel known as the dihydropyridine receptor, which is also located within the T tubule membrane. Activation of the dihydropyridine receptor triggers the closely related ryanodine receptor in the terminal cisternae of the sarcoplasmic reticulum, causing it to release calcium into the sarcoplasm.

Contraction Calcium released into the sarcoplasm binds to troponin C, resulting in a conformational change of the tropomyosin molecule and unmasking part of the actin lament responsible for binding to myosin: the G-actin monomer. With binding of myosin to the G-actin monomers, a 90degree cross-bridge forms. As soon as the cross-bridge forms, myosin ATPase cleaves ATP to form adenosine diphosphate (ADP) and adenosine monophosphate (AMP), resulting in another conformational change converting the 90-degree angle to a 45-degree angle. The contractural force is generated by the movement of the cross-bridge myosin head from a 90-degree to a 45-degree angle with a coincident sliding of the myo laments relative to one another toward the center of the sarcomere. The hydrolytic products of ATP then detach from the myosin head enabling the cycle to recommence. The addition of new ATP to the myosin molecule results in the rapid dissociation of the actin and myosin filaments.4,5 Sustained contraction occurs with the rapid repetition of this mechanicochemical cycle.5 Each sarcomere shortens in unison in successive ratchetlike movements composed of multiple acts of relaxation and reengagement depending on the degree of shortening demanded by the nervous system. With input from the central nervous system to relax opposing muscles, the process of excitation-contraction coupling produces smooth coordinated movement.

Relaxation For the cell to relax and return to baseline polarity, the sarcoplasmic reticulum and the plasmalemma must reaccumulate calcium, thereby lowering the concentration of calcium in the sarcoplasm and the T tubule while sarcolemmal membranes begin to repolarize. Sarcoplasmic

calcium concentrations decrease when calcium-magnesium ATPase pumps return calcium to the sarcoplasmic reticulum and sodiumpotassium ATPase pumps move sodium extracellularly and potassium intracellularly. Energy-dependent reaccumulation of calcium by the sarcoplasmic reticulum results in the release of calcium from troponin C, restoration of the resting con guration of the troponin-tropomyosin complexes, and ultimately disruption of the linkage between the myosin and actin laments. Further details on muscle structure and contraction are available.4-7 An important note is that all of the processes necessary for relaxation are active, meaning they require energy in the form of ATP for them to occur. In the absence of ATP, actin and myosin remain engaged, and the muscle remains in the shortened or contracted position known as rigor. Several terms are frequently used when referring to the shortening of muscle in veterinary medicine. Contraction implies a shortening or tightening of the muscle that is initiated by the central nervous system. This electrical activity is recorded readily using electromyography. A contracture is considered to be the shortening of a muscle because of abnormal activity within the muscle cell or cells. A contracture originates within the muscle ber itself and is therefore electrically silent on electromyography. Contractures occur in disorders characterized by muscle cramping such as polysaccharide storage myopathy or recurrent exertional rhabdomyolysis in which muscle shortening is typically electrically silent. The term spasm is more generic and may refer to a transient or sustained contraction or contracture. If a spasm is painful, the term cramp is often used. Fibrous tissue within a muscle may prevent complete stretching and shortening of a muscle, thereby limiting the range of motion of a limb or joint. This process occurs in fibrotic myopathy as described in a subsequent section.

Source of Energy for Contraction Understanding the sources of energy for contraction is important because most metabolic myopathies arise from derangements in these processes. ATP stores in muscle are very minimal and are quickly exhausted. Thereafter, ATP must be generated to sustain muscle function. In short, ATP is generated by either aerobic (oxidative) or anaerobic mechanisms. The primary substrates for aerobic metabolism include circulating nonesterified fatty acids and glucose or intramuscular stores (or both) of glycogen and triglycerides. When the phosphate bond of ATP is cleaved, the resulting by-products include ADP, inorganic phosphate, and energy. At the onset of muscle activity, cleavage of the phosphate bonds of creatine phosphate (or phosphorylcreatine) in the presence of ADP by the enzyme creatine kinase produces creatine and ATP. This anaerobic pathway is quickly supplanted by other pathways to produce energy (Table 11-1). TABLE 11-1 Energy Sources in Muscle∗

Numerous di erent processes within the muscle ber generate ATP after creatine phosphate and include glycogenolysis, glycolysis, Krebs cycle, β-oxidation of free fatty acids, and purine nucleotide deamination. Which pathways are most active depends on factors that include the underlying muscle ber composition, the number of motor units recruited, capillarization of the muscle, the underlying oxidative and glycolytic capacities of the muscle bers, and the delivery of oxygen and other energy substrates to the muscle. In turn, the breed and age of the horse, the length or stage of exercise, the intensity of exercise, and the fitness level of the horse influence these factors. For speci cs on glycolysis and generation of ATP via the Embden-Meyerhof pathway or the hexose monophosphate shunt, the reader is directed to a physiology text. Glycolysis or the breakdown of glucose (a 6-carbon molecule) to carbon dioxide and water in the presence of oxygen liberates 38 moles of ATP per mole of glucose. Whereas the fatty acid oxidation of a similar 6 carbon fatty acid yields 44 moles of ATP. The primary sources of glucose include circulating glucose in the blood, small quantities of free glucose within the sarcoplasm, and stored glycogen within the muscle cell. Glycogen is a branched polymer of individual glucose molecules consisting of 1:4α and 1:6α linkages. Breakdown of glycogen, or glycogenolysis, results in the liberation of 37 moles of ATP per mole of glycogen. A large, extramuscular storage site of glycogen is the liver. With exercise, the liver begins to liberate free glucose into the blood. Muscle cells take up glucose by glucose transporters located in the cell membrane. Athletes who have exercised frequently over time develop a denser network of capillaries about each muscle cell to facilitate the delivery of oxygen, free fatty acids, and glucose to the cell and the removal of by-products. Glycogen is made through the action of glycogen synthase and glycogen branching enzyme. The breakdown of glycogen’s 1:4α linkages is mediated by phosphorylase, whereas the breakdown at the 1:6α linkages is mediated by glycogen debranching enzyme. To prevent a futile

cycle of formation and degradation, these sets of enzymes are reciprocally regulated. Glycogen synthase is stimulated when glucose-6phosphate is plentiful, whereas the presence of glucose-6-phosphate and ATP inhibit phosphorylase. Phosphorylase is activated when levels of AMP are elevated. Hormones may also activate phosphorylase. Epinephrine stimulates phosphorylase in both the liver and skeletal muscle, leading to elevations in blood lactate and glucose. Glucagon acts only on liver phosphorylase, resulting in elevations in blood glucose and not lactate. Glycolysis and glycogenolysis are rapid processes and produce adequate energy for aerobic exercise. When exercise intensity increases and oxygen demand cannot be met, glucose is metabolized anaerobically. Anaerobic glycolysis of glycogen yields 3 moles of ATP and lactic acid as a by-product. The accumulation of lactic acid within the cell lowers the pH of the sarcoplasm, inhibits the function of many intracellular enzymes, and contributes to fatigue. After a period of light aerobic exercise (this varies between individuals but in many is after approximately 15 to 20 minutes), the muscle begins to rely more on the β-oxidation of free fatty acids than on glycolysis for the production of ATP. Small stores of fat are located within the muscle. However, most of the free fatty acids are liberated from body fat stores or the liver and are taken up by the muscle cell during exercise. The longer the aerobic exercise, the more energy is derived from metabolism of free fatty acids. More intensive aerobic exercise uses a combination of carbohydrate and fatty acids as energy sources. The amount of energy derived from fatty acid metabolism appears also to be partially in uenced by diet. The higher the intake of fat, the more muscle appears to rely on free fatty acids for energy. The optimal ratio of fat in the diet of the horse has yet to be determined. The onset of fatigue during aerobic exercise is di erent from that during anaerobic exercise. During aerobic exercise, glycogen depletion, hyperthermia, and electrolyte depletion appear to be major factors, whereas during anaerobic exercise, lactic acidosis and depletion of creatine phosphate and ATP appear to be important factors initiating fatigue. Accumulation of lactic acid within the muscle cell leads to decreased intracellular pH and inhibition of the enzymes important in glycolysis. A low intracellular pH also inhibits excitation-contraction coupling. In aerobic and anaerobic forms of exercise, eventual onset of myalgia and decreased motivation appear to be additional factors in fatigue. When most forms of energy within the muscle are exhausted the last source of energy within the cell is the formation of ATP from two molecules of ADP in a process known as the myokinase reaction. A by-product of this reaction is ammonia. Increasing ammonia concentrations in the circulation are well correlated with depletion of ATP. Depleted bers may develop a painful, electrically silent contracture that is similar to rigor, and severely a ected bers may die. The death of muscle cells related to exercise is known as exertional rhabdomyolysis. Rhabdomyolysis is often assessed by the measurement of enzymes such as creatine kinase or lactate dehydrogenase that are released into the blood when muscle cells are severely stressed (the cell membranes become permeable) or the cells lyse.

FIBER TYPE Differentiation Under light microscopy using simple staining techniques, skeletal muscle appears to be a hom*ogeneous tissue. However, muscle is composed of many bers with di ering characteristics. Myo bers may be grouped into two categories: type 1 and type 2. Table 11-2 gives a summary of the characteristics of di erent ber types. The type 1 and 2 classi cation is broad, and a spectrum of bers has characteristics of both categories. The di erences between bers arise from variability in the proteins that comprise the contractile elements. Several di erent isoforms have been identi ed for the myosin heavy and light chains, tropomyosin, troponin I, troponin T, and troponin C. The greatest variability is in the isoform of the myosin heavy chain. In addition, the calcium pumping capacity of the sarcoplasmic reticulum may vary from cell to cell. TABLE 11-2 Fiber Type Characteristics

All types of bers are found in most skeletal muscles. Type 1 bers are high in oxidative capacity and therefore rely on glucose, free fatty acids, and a high amount of oxygen to supply energy for contraction. Their sarcoplasmic reticula have a moderately slow reuptake of calcium capacity. Therefore type 1 bers are best suited to moderate contraction speeds and sustained contractions over long periods. As expected, these bers predominate in muscles involved in posture. Type 2 bers have a fast myosin ATPase rate and a rapid capacity to reuptake calcium into the sarcoplasmic reticulum. They are large-diameter bers compared with type 1 bers and have a higher glycolytic capacity, making them suited to powerful fast contractions. These bers predominate in the skeletal muscles involved in locomotion and rapid nemotor movements such as around the eye. Type 2 bers that have activities closer to type 1 bers are called type 2A, whereas those that exhibit high anaerobic capacity are characterized as type 2B. Because of the high oxidative capacity of type 1 bers, these muscles often appear to have a deeper red color than fibers high in type 2A and 2B fibers. Histologically, di erentiation of muscle ber types relies on the measurement of myo brillar ATPase activity (the enzyme responsible for the breakdown of ATP at the actin-myosin cross-bridges). Staining for this enzyme distinguishes two distinct ber types at pH 9.4. The slowtwitch type 1 bers have a low activity at this pH and so appear lighter in color than the fast-twitch type 2 bers. The type 2 bers can be divided further by preincubation at a more acidic pH into types 2A, 2B, and even 2C (the pH values required for this di erentiation vary with the laboratory but are around 4.5 and 4.3). The varying capacities of muscle ber types is largely the result of variations in myosin heavy-chain isoforms. These isoforms are identi ed through antibody and electrophoresis techniques. In mammals, at least nine have been identi ed. Two are developmental (embryonic and neonatal), two are found in cardiac muscle (alpha and beta), three are common in skeletal muscle (2a, 2b, and 2x), and two are rare with greatest expression in jaw, extraocular, and laryngeal muscles. Beta is also found in skeletal muscle and is an isoform with properties of slow twitch, high oxidative capacity, low glycolytic capacity, and low ATPase activity similar to the histologically identi ed type 1 myo bers. Type 2a myosin heavy-chain isoforms are typically found in bers exhibiting fast twitch, high glycolytic, high ATPase activity, and high oxidative capacity and as such resist fatigue. Type 2b and 2x isoforms are found in muscles that are fast twitch, produce strong contractions, have high glycolytic and ATPase activity but have low oxidative capacity and, as such, fatigue faster than type 2a rich myofibers.8-11

Innervation A neuron and all of its innervated myo bers constitute a motor unit. Each motor neuron may innervate a few or several hundred myo bers, and all of those myo bers exhibit similar contraction characteristics. Each myo ber, however, is innervated only by a branch of one motor neuron. Two types of motor nerve ber are found, slow and fast. The type and pattern of electrical activity on a muscle cell is related to the type of myosin isoforms expressed: The large phasic motor nerve bers innervate the type 2 myo bers, whereas the small tonic bers innervate the type 1 fibers.11 Muscle spindles are a bundle of specialized muscle bers that run parallel to nonsensory muscle bers and which contact sensory axon endings. Muscle spindle cells are important in telling the nervous system about the tone and length of the muscle. They are also important in maintaining posture. Discharges are relayed directly to the ipsilateral spinal motor neurons that innervate the muscle in which the spindle is located, resulting in a re ex arc. Two types of sensory neurons innervate each spindle: the primary endings are from the large-diameter group 1A a erent axons, whereas the less elaborate secondary endings are from the smaller group 2 a erent axons. The spindle cell bers also receive motor neuron innervation from small γ-motor neurons, which helps keep the spindle taut when the muscle contracts, which, in turn, enables the sensory endings to respond to a wide range of muscle lengths. Muscle spindle cells are di erent from the muscle Golgi tendon organ. Golgi tendon organs are found at the ends of the muscle bers at the musculotendinous junction and are formed from a common tendon to which several muscle bers are attached. A sensory nerve ending

wraps around each Golgi tendon organ. The sensory nerve innervating the Golgi tendon organ responds to tension generated by the contraction of any of these muscle bers, and the discharge is relayed to the ipsilateral α-motor neurons, where it is inhibitory. This action forms another reflex arc responding to muscle tension and helps ensure that muscles of flexion and extension do not contract simultaneously.

Fiber Recruitment Under neural control, an orderly selection of bers occurs with increasing demands. When just maintaining posture or walking the horse uses only the nerves supplying the slow-twitch type 1 and possibly a few of the intermediate type fast-twitch high-oxidative type 2A bers. As the speed or intensity of work increases, more and more bers are recruited. Thus, at a medium trot, approximately 50% of the muscle cells within a muscle belly are contracting, whereas at a gallop, most or all bers are involved. The bers are recruited in a set order: type 1, type 2A, and then type 2B. The gradation in response of a muscle is known as the recruitment of motor units. A motor unit consists of a motor nerve and the cells it innervates. The smoothness of a muscular contraction occurs because of the asynchronous contraction of the various motor units.

Distribution The proportions of fiber types present within a muscle vary according to individual traits of the horse, the muscle location, the breed, and the age of the horse.1,8,10,12-15 In some muscles (in particular the middle gluteal muscle, which is sampled most commonly) the distribution also depends on the sampling site or depth because the distribution of ber types is nonhom*ogeneous.1,10,15,16 The suggestion has been made, however, that within a speci c muscle of an individual, the variation in ber types is small if samples are taken from the same site or an identical contralateral site under controlled conditions.15,17 This arrangement is why it is generally recommended to biopsy speci c muscles to diagnose certain disorders. It provides a relatively hom*ogeneous baseline from which to evaluate a muscle. The exception is when a specific muscle is affected (e.g., exhibits atrophy) wherein it should be biopsied.

GROWTH, TRAINING, AND PERFORMANCE Fiber type predominance does a ect performance. However, with growth and training muscle cells adapt and may change ber type over time.1,7,18-22 Training results in (1) an increase in the ability of a ber to use oxygen by increasing the number of mitochondria within the sarcoplasm, (2) a decrease in the use of muscle glycogen and blood glucose with a greater reliance on fat oxidation to supply ATP, and (3) a decrease in the amount of lactate produced per given intensity of exercise. Variable results have been reported on the e ects of exercise on ber size and capillarization. The extent and nature of the changes appear to depend on the duration, intensity, and type of exercise involved, as well as the age of the animal. Over the last decades, much information has been gained in the eld of equine exercise physiology, and multiple texts are available on the subject both in print and online that extend beyond the scope of this chapter.23-26

PATHOLOGIC CHANGES General Response of Muscle to Injury Muscle responds to injury or damage in a limited number of ways. A growing body of evidence indicates that myopathic conditions share a similar nal pathway of muscle ber degeneration, although the number and type of bers a ected and the degree of damage varies.27 The nal common pathway involves failure to sequester calcium ions from the sarcoplasm. Prolonged calcium accumulation within the sarcoplasm leads to activation of cellular enzymes and prolonged activation of myo laments. This process is visible histologically as hypercontraction. Consequently, mitochondria take up excess calcium in an attempt to prevent cellular damage. Accumulation of calcium within the mitochondria leads to failure of energy production (i.e., ATP) within the cell and eventually failure of all intracellular energydependent mechanisms, which leads to failure of membrane pumps and ultimately cell swelling and lysis.27,28 The exact nature of these calcium-activated degenerative pathways is still unknown. A key step may be calcium-induced membrane phospholipid hydrolysis via the activation of phospholipase enzymes, resulting in the production of tissue-damaging metabolites. Other processes, however, involving nonenzymatic lipid peroxidation also may be activated. During exercise the ow of oxygen increases. At the same time an overall depletion of ATP sources occurs. The resultant metabolic stress leads to a greatly increased rate of oxygen free radical production that may exceed the scavenger and antioxidant defense systems of the cell, leading to a loss of cell viability and damage. The increase in free radical activity can lead to a failure of calcium homeostasis and consequent muscle damage, or, alternatively, calcium overloading may lead to an activation of free radical–mediated processes.29 Multiple initiators likely lead to this nal common pathway, depending on the type of exercise, tness level, presence or absence of underlying genetic defects in muscle metabolism, and nutritional status of the horse. Free radical–induced skeletal muscle damage may be especially important if reperfusion follows a period of ischemia.30

Atrophy and Hypertrophy Atrophy may be de ned as a decrease in muscle ber diameter or cross-sectional area. Atrophy can occur in a variety of circ*mstances,

including denervation, disuse, and cachexia, in many muscle diseases, after circulatory disturbances, and after extensive myolysis. Within 2 or 3 weeks after peripheral denervation, up to two thirds of the original muscle mass may be lost, although this may not always be obvious clinically because of the presence of super cial and intramuscular fat deposits. Histologic changes re ect the a ected nerve and the muscle ber supplied. In disuse (e.g., atrophy resulting from a tenotomy) a preferential atrophy of type 1 bers is apparent, often with hypertrophy of type 2 bers. In contrast, with cachexia and malnutrition, type 2 bers are preferentially a ected, especially in the postural muscles. Not all muscles show the same degree of atrophy, even within one disease process. In cachexia, the back and thigh muscles tend to be the rst a ected, and the loss of muscle is usually symmetric, with a concurrent loss of fat deposits. A localized asymmetric atrophy is associated with paralysis, immobilization, and denervation.31 As indicated previously, hypertrophy may occur through training. A compensatory hypertrophy may also occur in the bers surrounding an area where bers have been lost or have decreased greatly in size. Often, large bers are visible histologically in such conditions, with evidence of incomplete longitudinal division.31

Repair after Denervation Denervation results in atrophy, as remarked previously, but also leads to abnormal muscle excitability. Muscle becomes more sensitive to circulating acetylcholine. The result is the presence of ne, irregular contractions or brillations detectable using electromyography. Fibrillations are not visible clinically and should not be confused with fasciculations that appear grossly as jerky muscle contractions in response to pathologic spinal motor neuron discharges.11 With reinnervation the brillations cease. Reinnervation can occur in two ways. Damage without severing the nerve is called axonotmesis. The axon is damaged but the Schwann cell and endoneural brous sheaths remain. Axons from the proximal part of the damaged nerve may reestablish connections within the empty residual Schwann cell sheath to the distal portion of the original nerve. Initially, atrophy of scattered muscle bers is visible, but after reinnervation a proportion of the muscle bers are restored to their normal size and function. Even if the a ected muscle is not fully restored, other surrounding una ected muscles may be able to compensate so that overall function is maintained, although slight gait abnormalities may remain. The second type of reinnervation occurs when some nerve fascicles are severed completely or when the nerve lesion is located a great distance from the muscle, for example, when lesions involve the motor neuron cell bodies or nerve roots or both. In this case, collateral reinnervation from adjacent, una ected axons can occur. Given that the nerve type characterizes the muscle type, this action results in regrouping of muscle bers so that clusters of bers of the same histochemical type are found rather than the normal checkerboard pattern. This phenomenon is known as fiber type grouping.

Repair after Injury Regeneration where a break in the continuity of the muscle ber has occurred di ers considerably from neurologic damage. In such cases, repair involves multiplication of nuclei and formation of new internal structures and organelles followed by fusion and alignment into a new multinuclear myo ber. Regeneration can be swift when parts of the a ected ber remain intact. In this case, regeneration occurs at the healthy end of the severed ber (continuous or budding regeneration). When damage is more severe, regeneration occurs by fusion of mononuclear myoblasts to form a myotube, which develops in a similar way to fetal bers (discontinuous or embryonic regeneration). The origin of myoblasts is primarily satellite cells, and their origins and function have been well reviewed.32,33 If the sca olding, basem*nt membrane, and supporting tissues remain intact, and if the initiating disease process subsides, new bers tend to orient in a way similar to the original bers. This type of regeneration usually occurs with segmental necrosis in which necrosis of the whole diameter of the ber has occurred involving several sarcomeres. Fibroblastic and vascular reactions are minimal in this type of regeneration. Therefore full function of the muscle is restored without residual dysfunction. Massive trauma, hemorrhage, infection, or infarction can result in damage to the basem*nt membrane and other supporting structures resulting in complete disorientation of the regenerating bers with signi cant proliferation of broblasts and vessels. In such cases, regeneration may result in signi cant deformation of the muscle with disruption of its normal function. Regenerated cells often have centrally located nuclei and larger diameters than older myofibers.

DIAGNOSIS OF SKELETAL MUSCLE DISORDERS Because damage to muscle tissue often occurs along similar pathways, recognizing that muscle disorders can only rarely be diagnosed based on muscle histopathology alone is important. A thorough history that includes genetic background, exercise, nutrition, patient signalment, and clinical description of the muscular disorder is mandatory. Skeletal muscle disorders broadly fall into two categories: acquired or inborn. Acquired myopathies include those of traumatic or infectious origin and ones arising secondarily from endocrine or nutritional disease or electrolyte imbalances (acquired metabolic myopathies).34 Inborn errors of metabolism that manifest as metabolic myopathies originate primarily from defects in muscle energy metabolism. As such, they can be grouped as errors in carbohydrate, lipid, or purine metabolism; as mitochondrial defects; or as primary disorders of ion channels (channelopathies).

Acute trauma typically produces obvious clinical signs such as an open wound, severe lameness, heat, pain, swelling, or any combination. Healed traumatic injuries may be evident by abnormal gait or scar tissue, or both, within the muscle as primary complaints. Infectious myositis may also cause heat, pain, swelling, or crepitus, in addition to systemic signs of colic, endotoxemia, and shock if the infection is severe and widespread. In general the metabolic myopathies reveal a more chronic history involving some of the following client complaints: episodic weakness or muscle cramping, progressive weakness, atrophy, muscle wasting, reluctance to move, rapid muscle fatigue with exercise, intermittent or consistent gait abnormalities, muscle fasciculations, or discolored urine. Modalities used to diagnose equine myopathies may include a thorough history (including use, diet, genetic background, signalment, and client complaint), physical examination, serum chemistry pro le, complete blood count, urinalysis (including analysis for myoglobin and fractional excretions of sodium, potassium, phosphorus, and chloride), muscle biopsy (histology, histochemistry, electron microscopy), electromyography (EMG), thermography, genetic testing, nuclear scintigraphy, or any combination. In horses with suspected metabolic myopathies, diagnostics may be strategically put together as part of an exercise challenge test that includes the history, observation of the horse at work, pre- and postexercise measurement of serum muscle enzyme activity and lactate, urinalysis with fractional excretions of electrolytes, genetic testing, and muscle biopsy.

Plasma and Serum Enzyme Activities The enzymes that are most useful in evaluating the equine muscular system are creatine kinase (CK), aspartate transferase (AST), and lactate dehydrogenase (LDH). A change in the plasma activity of any enzyme can occur for a variety of reasons, including alteration in the permeability of the enclosing cell membrane, cell necrosis, impaired removal or clearance of the enzyme, and increased or impaired synthesis. Decreases in plasma enzyme activities are not usually clinically signi cant. Often, no one speci c organ of elimination is responsible, although most elimination occurs via the liver, kidneys, and lungs. Therefore, under most circ*mstances, the elimination rate of an enzyme from the plasma remains fairly constant, and the rate of influx to the plasma is the crucial factor. Increases most commonly occur because of a defect in the integrity of the membrane containing the enzyme-rich sarcoplasm.6,35 The defect results from complete disruption of the cell or a transient change in the permeability of the membrane without cell death. A complete explanation for the pattern of release of intracellular constituents from diseased muscles cannot be given here. Most of the enzymes that are detectable in increased concentrations in the blood with the various muscle disorders are the major soluble (sarcoplasmic) enzymes, although one can also detect the mitochondrial form of AST with severe injury. The degree of muscle damage does not correlate with the magnitude of increase in serum muscle enzyme activity; a small amount of severely damaged muscle may release a similar amount of enzyme as a larger quantity of mildly damaged muscle. Under general circ*mstances, the rate of e ux of an enzyme most likely depends not only on its molecular weight and intracellular localization but also on its binding to various intracellular structures and relative intracellular and extracellular concentration. No explanation for the signi cant differences in half-lives of the various plasma muscle enzymes reported for human beings and horses has been given.

Creatine Kinase Creatine kinase is the enzyme responsible for breaking down creatine phosphate to creatine and phosphate, releasing energy for muscular contraction. This reaction is the sole source of energy in muscle at the initiation of exercise. Thereafter, energy is supplied by oxidation of glucose and free fatty acids as described previously. Because this enzyme is responsible for the breakdown of creatine phosphate, it is often called creatine phosphokinase. This term, however, is inappropriate, and one should use creatine kinase. In the horse, CK is found mainly in skeletal muscle, the myocardium, and the brain.36 Little or no exchange of CK between the cerebrospinal uid and plasma appears to occur. A signi cant increase in total plasma CK activity therefore is caused by cardiac or skeletal muscle damage. CK (molecular weight, 80,000 daltons [Da]) does not enter the bloodstream directly after its release from the muscle cell but transits through the lymph via the interstitial uid. The total quantity of circulating CK in the horse under normal conditions is estimated to be equivalent to the quantity of CK in approximately 1 g of muscle; a threefold to fivefold increase in plasma CK activity corresponds to the apparent myolysis of approximately 20 g of muscle.37 However, as described previously, making assumptions as to the amount of muscle damaged based on serum CK activity is difficult because muscle cells appear to be able to release significant quantities of CK without necessarily being lysed. In human beings, two monomers of CK are known and are designated M and B. The enzyme is dimeric, and three possible primary forms exist: MM, MB, and BB. In simpli ed terms, MM is found mainly in skeletal muscle, BB in the brain and epithelial tissues, and MB in the myocardium. In the horse, some confusion over CK isoenzymes exists, with workers reporting di erent electrophoretic bands and tissue activities, perhaps because of the di erent techniques used.38-41 In one study the heart and skeletal muscle were found to contain predominantly the MM dimer; the brain, pancreas, and kidney, mainly the BB dimer; and the intestine, the MB and BB dimers.39 This work suggests that in the horse, CK isoenzymes on their own could not be used to di erentiate between skeletal and cardiac muscle damage. This problem is alleviated largely by the current availability of tests for cardiac troponin I. Unfortunately, commercial tests for skeletal troponin I are not available. In the absence of clinical cardiac disease, elevations in CK primarily can be attributed to skeletal muscle. The plasma halflife of CK in the horse is short (108 minutes, 123 ± 28 min with a plasma clearance of 0.36 ± 0.1 ml/kg/min)37,42 in contrast to reports of

12 hours in human beings.43

A

A

Aspartate Aminotransferase AST is found mainly in skeletal muscle, liver, and heart, although lower activities are present in several other tissues. Therefore AST is not tissue speci c.36,44,45 Two isoenzymes have been identi ed by electrophoresis: MAST (found exclusively in the mitochondria) and CAST (originating from the cytoplasm or sarcoplasm). The ratio of cytosolic to mitochondrial enzyme in horse serum is signi cantly greater than that found in human beings and many other mammals.46 In the horse, the ratio of these two forms varies between tissues, and no tissue speci city is apparent for either isoenzyme. The plasma half-life of AST in the horse is 7 to 10 days,42 far longer than the 11.8 hours in human beings.43

Lactate Dehydrogenase Lactate dehydrogenase is a tetrapeptide that occurs in ve di erent isozymic forms produced by combinations of two kinds of subunits (H [heart] and M [muscle]) in groups of four to produce the functional tetrameric molecule. The ve isoenzymes are labeled as LDH1 to LDH5, or H4, MH3, M2H2, M3H, and M4, respectively. Similar to AST, LDH is found in most tissues and is therefore not organ speci c. However, tissues contain various amounts of the LDH isoenzymes, and the isoenzyme pro le obtained by electrophoretic separation has been used to identify speci c tissue damage.47 For the most part, LDH5 (plus some LDH4) is found in the locomotor muscles, the liver contains mainly LDH3 (with some LDH4 and LDH5), the heart contains mainly LDH1 (with LDH2 and LDH3), and all types have been found in certain nonlocomotor muscles.38,41 Training has been shown to increase the percentage of LDH1 to LDH4 and decrease that of LDH5 in skeletal muscle.48 The practitioner should use nonhemolyzed samples for LDH determinations because red blood cells contain large amounts of LDH. In general, though, assays for CK and AST are most commonly used to evaluate skeletal muscle.

Urinalysis: Myoglobinuria Myoglobin (molecular weight 16,500 Da) is essential for the transport of oxygen into and within muscle cells. Most mammalian muscles contain approximately 1 mg myoglobin per gram of fresh tissue, and the suggestion has been made that acute destruction of at least 200 g of muscle must occur before serum myoglobin levels rise su ciently for detection in the urine.49 Serum myoglobin has a short half-life, and therefore measuring serum myoglobin is of limited diagnostic value. In human beings, myoglobinuria occurs in a variety of conditions, including myocardial infarction, crush and burn injuries, malignant hyperthermia, idiopathic and exertional rhabdomyolysis, and certain genetic metabolic abnormalities.6,49-52 In the horse, myoglobinuria has been seen in equine exercise or trauma-induced rhabdomyolysis,53,54 white muscle disease in foals,55 and postanesthetic myositis.56 Pigmenturia or dark urine may be caused by increased myoglobin, hemoglobin, or whole blood in the urine. Low levels of myoglobin, hemoglobin, or whole blood may occur in the urine and can be detected by urine dipstick or laboratory test before visible changes in the color of urine. Therefore considerable amounts of myoglobin, hemoglobin, or whole blood must be present for visible pigmenturia to occur. Many causes of hemolysis in the horse can result in hemoglobinuria, including oxidative damage to erythrocytes,57 neonatal isoerythrolysis, hepatic disease, and renal disease.59 Unfortunately, stored urine, concentrated urine, and urine containing myoglobin, hemoglobin, or other porphyrins can appear similar in color57,59,60 (Figure 11-2). Therefore myoglobinuria cannot be distinguished by color alone. The orthotoluidine-impregnated strips commonly used by veterinary clinicians (urine dipstick tests such as BM-Test-8, Boehringer Ingelheim Pharmaceuticals, Ridge eld, Conn.) are insensitive to the presence of myoglobin and hemoglobin.54,61 The di erential salting out of hemoglobin with ammonium sulfate gives false results in human beings49,50,59,62 and the horse but is used commonly in many laboratories.49,50,59,61,62 Visual inspection of the serum or plasma may indicate the cause of the pigmenturia. The low a nity of myoglobin for haptoglobin means that myoglobin is excreted at plasma concentrations around 0.2 g/L, whereas hemoglobin only appears in the urine at plasma concentrations greater than 1.0 g/L. At this concentration a pink discoloration of the plasma occurs, indicating that hemoglobin is present.9,49,50,63

FIGURE 11-2

Four urine samples. A, With 3 mg/ml hemoglobin. B, With 3 mg/ml myoglobin. C, Normal stored for 1 month at 18° C. D, Control (fresh urine).

On electrophoretic separation on cellulose acetate, myoglobin migrates as a β 2-globulin and hemoglobin as an α2-globulin. This method can distinguish the two proteins in urine containing high concentrations (>125 g/ml) of either protein, provided no other proteins, apart from albumin, are present. Immunoassays are the most sensitive and speci c tests and can be used for detecting small amounts of myoglobin in blood and urine.61,64,65 Sequelae of acute tubular necrosis and acute renal failure are associated with myoglobinuria in human beings66 and horses.44,63 Renal disease associated with myoglobinemia in the horse seems to be associated most with horses that are highly stressed (causing decreased renal blood ow), dehydrated, or have lactic acidosis associated with rhabdomyolysis. Therefore horses su ering from “capture” myopathy (e.g., cast in a stall or trailer accident and struggling for a prolonged period) and endurance horses are most at risk.

Muscle Biopsy The practitioner can obtain muscle biopsy samples by surgical excision under general or local anesthesia63 or by percutaneous needle biopsy from the standing horse.9,14,67 Because of the variation in ber composition in muscles, which muscle is chosen, as well as the position and depth of sampling, is important. Good specimen preparation for frozen sections is vital to prevent artifacts such as ice crystals.6,9,68 For laboratories that prefer examination of frozen sections, specimens are typically shipped chilled overnight so that the laboratory may freeze the sample to minimize artifacts. Other laboratories prefer formalin- xed specimens. The practitioner should discuss sample procurement, preparation, and shipment procedures with the laboratory before biopsy. Muscle biopsy allows the morphologic, biochemical, and physiologic properties of the myofibers to be examined with the animal still alive and causes little morbidity. Typical sites used for percutaneous biopsy with a 6-mm modi ed Bergstrom needle in cases of generalized muscle disease or for research purposes include the semimembranosus, the biceps femoris, and most commonly the middle gluteal muscle. Although convenient, the necessity to have a modi ed Bergstrom needle means that most practitioners opt for open biopsy of the semimembranosus muscle over percutaneous middle gluteal muscle biopsy. Instructions on how to obtain a muscle biopsy are available online.69 To biopsy the semimembranosus a 6- to 10-cm2 area is clipped and surgically prepared slightly lateral to and below the anus. At this location, if a scar occurs after the procedure the tail will hide it. An inverted L lidocaine block of the skin and subcutaneous tissues is performed. Local anesthetic is not injected into the muscle itself, which may induce artifact into the biopsy. The muscle belly is innervated poorly for pain, and reaction to transection of a small piece of muscle is minimal. A piece of muscle measuring approximately 1 × 1 × 2 cm is dissected free, and sutures are used to close the subcutaneous layer and skin. Buried skin sutures that do not need to be removed or skin staples appear to be preferable because they are less likely to cause skin irritation and therefore are less likely to be rubbed out by the patient. Should the horse open the biopsy site, the wound typically heals well by second intention with general wound care. This procedure rarely has complications; however, the horse should be current on tetanus prophylaxis as a general precaution. The biopsy is wrapped loosely in saline-dampened gauze and placed in a sealed container for shipment on icepacks to a laboratory skilled in muscle biopsy assessment. Histochemical staining of frozen muscle sections aids in the detection of glycolytic enzyme de ciencies and storage of various atypical metabolic compounds.6,70-72 Pathologic changes that can be found in muscle include presence of degenerate or regenerate myo bers, abnormal cytoplasmic inclusions, accumulations of polysaccharide, changes in ber volume, number, shape, malformation, degeneration or proliferation of organelles, and disruption of the basic architecture of the ber. Aggregates of in ammatory cells, reactive changes in vessel walls, occlusion of blood vessels, and increased amounts of brous tissue are other possible pathologic ndings. In some conditions, such as tetanus or botulism, no significant changes may be seen. Further information on the types of changes that occur is available.6,31 Various histologic stains have been developed to highlight the di erent morphologic changes and inclusions that occur. The most important of these is the periodic acid–Schi stain that highlights glycogen stores within a cell and membrane structures containing mucopolysaccharide, glycoproteins, mucoproteins, glycolipids, or phospholipids. Large aggregates of abnormal polysaccharide occur in Quarter Horses and Draft-breed horses with polysaccharide storage myopathy. Histochemical staining for ber type enables the pathologic changes to be recognized as a ecting all bers or those of one type only. In cases of reinnervation, large groups of a single ber type are visible as opposed to the more interspersed pattern typically observed. Immunocytochemical studies have been used to investigate neuromuscular disorders in human beings and to diagnose autoimmune streptococcal-associated myositis in the horse.73,74 Specialized staining techniques help to localize speci c enzymes and intracellular and extracellular muscle components such as the various collagen and myosin types, complement, and bronectin. Electron microscopic studies enable investigation into the ultrastructure of muscle fibers.

Electromyography Needle EMG is the study of the electrical activity of muscles. EMG may be useful in horses with unusual gait or muscle changes and can be used to discriminate between neurogenic and myogenic causes and more speci cally to identify areas or regions of an abnormality.75 A

recording needle electrode is placed into the muscle, and the electrical activity is ampli ed, recorded on an oscilloscope, and projected audibly into a loudspeaker. The electrical status of muscle membranes depends on the integrity of the whole motor unit. Thus an EMG evaluates the function of the ventral motor horn cell, its axon, axon terminals, and neuromuscular junctions, as well as the muscle bers it innervates. Investigations are usually carried out in two phases. First the practitioner assesses the electrical potentials associated with the physical disruption of the sarcolemma resulting from insertion of the needle. These insertional potentials, if relayed through a loudspeaker, tend to emit a sound similar to short bursts of loud static. They vary among muscles, probably because of differences in the size and number of motor units present. The second phase involves assessment of electrical potentials when the needle electrode is at rest in the muscle. Normally, electrical silence occurs with cessation of needle movement, unless the needle is located in proximity to a nerve branch or the end plate zone when miniature end plate potentials are recorded continuously; these sound similar to low-intensity static. In relaxed, diseased muscles, di erent types of abnormal electrical activity have been recognized. In cases of denervation, insertional activity may persist after needle movement has stopped because of increased excitability of the muscle ber membranes. Positive, sharp waves are slow monophasic waves, rapid in onset, with a slow decay to the baseline, which occur repeatedly with variable amplitude (100 V to 20 mV). Their cause is uncertain, they often occur with denervation, and they may represent a nonpropagated depolarization region in the muscle bers near to the tip of the electrode. When occurring in trains, these positive sharp waves sound similar to a waning brrrr. Fibrillation potentials are electric signals generated by a single muscle ber. Long, random volleys of mono- or biphasic (occasionally triphasic) potentials of short duration (0.5 to 5.0 ms), with amplitudes of usually less than 200 V, commonly occur in denervation (depending on the stage). Constant and repetitive brillation potentials, which sound similar to rain on a tin roof or the sizzling of frying eggs, can be found especially in the early stages of denervation. Fibrillation potentials can also be seen in myopathic disorders in which segmental muscle necrosis may have caused, for example, the separation of a muscle fiber and its nerve supply. Myotonic discharges—high-frequency (up to 1000 Hz) repetitive discharges with waxing and waning of potentials is seen and heard with a characteristic, musical, dive-bomber, or more precisely, revving motorcycle–like sound—are found in the myotonias (myotonia congenita, myotonia dystrophica, and hyperkalemic periodic paralysis) if the practitioner moves the exploring EMG electrode or percusses the muscle externally. Bizarre, high-frequency discharges (often called pseudomyotonia) may produce a dive bomber–like sound. No true waxing and waning occurs, although the amplitude and frequency of the potentials may change abruptly to mimic a revving motorcycle–like sound. The discharges are often in couplets or triplets and are likely to terminate abruptly. These, or similar discharges, occur in long-standing denervations, ventral horn cell disease, polymyositis, and certain myopathies. Unlike the myotonic discharges, pseudomyotonic discharges may be abolished by curare and therefore are believed to originate presynaptically. Fasciculations, visible muscle twitching, are caused by the spontaneous contraction of some or all of the constituent bers of a motor unit. They can be local or generalized and occur primarily with any cause of muscular weakness (myasthenia), especially in neurogenic disorders such as tetanus and certain debilitating and metabolic disorders. In addition, one can assess the electric activity induced by electric stimulation of nerves or associated with voluntary or induced muscle contraction. Further information on EMG is available.76

Thermography An infrared thermographic scanner converts the radiated thermal energy of the skin to electric signals that can be ampli ed and displayed on a video screen. By using isotherm colors of known temperature, the examiner can obtain two-dimensional, graphic, and quantitative information regarding the precise temperature of the skin surface. The practitioner can use this technique noninvasively as a means of detecting changes in skin temperature resulting primarily from changes in peripheral blood ow. In ammation, atrophy, neoplasia, and neurologic lesions, particularly of the autonomic supply to the skin, can alter local blood flow. Abnormalities have been found in exercise-exacerbated focal thoracolumbar gait abnormalities, as well as disuse atrophy.77 The technique has been useful in documenting hindlimb muscle strain as a cause of lameness in horses.

Scintigraphy Bone-seeking radiopharmaceuticals have been used to detect and localize skeletal muscle involvement, especially in poor performance cases, but are of limited routine value in the field. Accumulation in damaged muscle may be related to the deposition of calcium in damaged fibers, binding by tissue hormones or enzyme receptors, tagging to denatured proteins, or altered capillary permeability. Uptake of labeled phosphates appears to occur only when muscle damage is ongoing and does not occur in areas of repair. Three main types of muscle uptake were identi ed in one study of horses with skeletal muscle damage: a di use, severe, and generalized uptake; bilateral symmetric uptake involving muscle groups that perform synergistic functions; and asymmetric radioisotope uptake in one or more muscle groups on one side of the animal.78 The reasons for the various patterns have not been elucidated fully. Scintigraphy can be used as part of a series of tests to identify atypical or difficult-to-pinpoint lamenesses in horses that may have primary muscle lesions among other problems.79,80

Exercise Tests Certain physiologic changes can result in a transient alteration in cell membrane permeability. Hypoxia, catecholamines, hypoglycemia, changes in pH, and altered ionic concentrations have been reported as causing such a change in membrane permeability.45,81,82 Many of these changes are believed to act by decreasing the amount of ATP available for the maintenance of cell integrity, which becomes especially important during exercise.83 Measuring the CK and AST activities before and after a controlled period of exercise has been suggested as an aid to diagnosing certain muscle disorders. A major di culty has been to establish exactly the normal enzyme response to exercise. Much confusion exists in the literature regarding this parameter, partly because of the di erences in the intensity and duration of the exercise undertaken, the varying sampling intervals used in the reports, and the possible inclusion of individuals with muscular problems. The majority of workers have suggested that an increase in CK activities occurs with hard exercise,83-85 whereas with slower work, others have shown no signi cant increase. This di erence suggests that intensity could be an important factor.86,87 Such a conclusion is supported by work in dogs, which shows that CK activities correlate with the intensity of muscular activity.85 Another study in the horse,88 however, suggested that CK elevations did not vary according to the intensity of the work, and one researcher89 proposed that the duration of exercise was the more important factor. Another study suggested that when the duration of exercise was kept constant, the intensity of the exercise in fact did have an effect on the extent of CK activity increase.90 Although no signi cant increase in CK activity was found after trotting exercise in conditioned animals by one worker, others recorded signi cant increases when horses performed the same exercise after 1 or more days of rest.87 Increases in AST activities of 35% have been reported after a 1500-m canter91 and of 50% after strenuous exercise in previously rested animals.84 Most other workers have found little increase in AST after different types of exercise.42,46,89,92 Therefore the e ects of exercise on plasma muscle enzyme activities may depend on the tness of the animal and the intensity and duration of the exercise, as well as on the environment.38,63,89 However, concurrent lameness does not appear to alter postexercise CK activity signi cantly in healthy horses without myopathy.93 In horses, as in human beings, large intersubject variability may occur in the postexercise rise in CK activities that must be taken into account.37 Plasma volume changes may a ect the activities recorded, especially if measured immediately after exercise. As discussed previously, the physiologic increase in CK activity after exercise is believed to be caused by a change in cell membrane permeability, possibly caused by hypoxia, although other factors are likely involved. Hypoxia may occur at lower workloads in unconditioned horses, and these horses may be expected to show higher postexercise activities than a t horse given the same work. The suggestion has been made that the magnitude of the exercise-induced increase in enzymes lessens with training.42,88,89 Some workers have found no signi cant changes in the AST and CK responses to exercise during a training program,92 whereas others have found that after an endurance ride the ttest animals (indicated by the speed of heart rate recovery after an endurance ride) had lower increases in CK activities.86 The magnitude of exercise-induced changes in CK activities increases with detraining. This study concluded that increases of more than 100% in AST activity after exercise likely are abnormal, regardless of the intensity of the exercise or the tness of the animal. Also, if a short, submaximal exercise test is carried out the serum CK and AST activities at 2 hours after exercise should not rise to more than 250% and 50% of the pre-exercise values, respectively, regardless of fitness.90 The point has been stressed that, although exercise might result in statistically signi cant changes in CK and AST activity, these may not always be of biologic or clinical signi cance.94,95 The practitioner must always take into consideration the clinical history and clinical presentation when interpreting enzyme values. For example, a young racehorse given its rst gallop often has activity changes greater than those described (e.g., from a pre-exercise level of 40 U/L to a 2-hour postexercise level of 350 U/L), although this is unlikely to be clinically signi cant. However, in cases of recurrent exertional rhabdomyolysis (RER) or polysaccharide storage myopathy (PSSM), similar changes may indicate ongoing subclinical muscular dysfunction. For example, in the author’s experience, horses with RER or PSSM may have 4-hour postexercise serum CK activities of 10,000 to 15,000 U/L without outward clinical signs of rhabdomyolysis. Currently the author recommends a standardized submaximal exercise test wherein the practitioner can interpret the results in the light of the horse’s history and presentation. Box 11-1 shows the work up for a horse with chronic exertional rhabdomyolysis, and Box 11-2 illustrates a normal response to a standardized submaximal exercise test. Horses that have recently ridden in a trailer a long distance for evaluation may require 24 hours of hospitalization to ensure a normal CK value before the examination. BOX 11-1 WORKUP FOR A HORSE WITH CHRONIC EXERTIONAL RHABDOMYOLYSIS 1. History: genetic background, diet, use, description of previous episodes, housing conditions 2. Physical examination 3. Collection of urine for fractional excretions (Na+, K +, Cl –, and PO 4, if desired)

4. Collection of blood for baseline CK, AST, electrolytes, and lactate (if desired) a. A full profile is recommended in horses without a recent one performed. b. A complete blood count may be performed if inflammatory disease is suspected. 5. Exercise at a trot for 30 minutes either under saddle or on a lunge line; after 15 minutes, stop every 5 minutes for 1 minute to observe if sti ness or gait abnormalities are developing. If so, stop exercising the horse. 6. Postexercise CK analysis 4 to 6 hours after exercise. A 24-hour postexercise AST may be examined if desired. 7. Muscle biopsy. Middle gluteal, semimembranosus or other muscle if indicated. AST, Aspartate transferase; CK, creatine kinase; Cl, chloride; K +, potassium; Na+, sodium; P, phosphorus.

BOX 11-2 CRITERIA FOR A NORMAL RESPONSE TO A SUBMAXIMAL EXERCISE TEST DESIGNED FOR A GIVEN HORSE 1. Pre-exercise: CK activity 1 IU/L) most often was associated with EPM in horses and may be helpful in di erentiating compressive spinal cord disease from EPM. Furthermore, persistently increased CSF CK activity may be associated with a poor prognosis in horses with EPM.19 Lactate dehydrogenase activity may be increased in spinal lymphosarcoma.

Lactic Acid Concentration CSF lactic acid concentration may be an indicator of neurologic disease (see Table 12-2). CSF lactic acid concentration increases in eastern equine encephalomyelitis (4.10 ± 0.6 mg/dl), head trauma (5.40 ± 0.9 mg/dl), and brain abscess (4.53 mg/dl). Lactic acid concentration may be the only CSF parameter increased in horses with brain abscess.20

TABLE 12-2 Cerebrospinal Fluid Values from Atlantooccipital and Lumbosacral Spaces of Normal Healthy Adult Horses∗

Red blood cell (RBC) count (per μl) White blood cell count (per μl) Total protein (mg/dl) ∗ Albumin (mg/dl) Albumin quotient (AQ) Immunoglobulin G (IgG) (mg/dl) IgG index Creatine kinase (CK) (IU/L) Lactate dehydrogenase (IU/L) Aspartate aminotransferase (AST) (IU/L) Glucose (mg/dl) Lactate (mg/dl) Sodium (mEq/L) Potassium (mEq/L) ∗

Atlantooccipital Space (Mean ± SD [Range]) 51.0 ± 160 (0-558) 0.33 ± 0.49 (0-1) 87.0 ±± 17.0 (53 ± 11.6) (59-118) (35-74) 35.8 ± 9.7 (24-51) 1.4 ± 0.4 (1.0-2.0) 5.6 ± 1.4 (3-8) 0.19 ± 0.046 (0.12-0.27) 0-8 0-8 4-16 35%-70% of blood glucose 1.92 ± 0.12 140-150 2.5-3.5

Lumbosacral Space (Mean ± SD [Range]) 36.8 ± 59.7 (0-167) 0.83 ± 1.11 (0-3) 93.0 ± 16.0 (58.0 ± 11.0) (65-124) (39-78) 37.8 ± 11.2 (24-56) 1.5 ± 0.4 (1.0-2.0) 6.0 ± 2.1 (3-10) 0.19 ± 0.5 (0.12-0.26) 0-8 0-8 0-16 2.30 ± 0.21 140-150 2.5-3.5

Total protein concentration next to the mean ± SD in parentheses is the value expected using a total protein standard.

summary Normal CSF ndings do not always rule out the presence of neurologic disease. CSF values may be normal with lesions outside the CNS that are not bathed in the CSF, such as extradural, ventral root, and peripheral nerve lesions. Normal CSF values also may occur early or late in the disease process and in CSF samples obtained from a site distant from the lesion. Acute neurologic disease, especially if multifocal, may not have su cient time to cause signi cant damage to the blood-brain barrier and alter CSF constituents, whereas in chronic CNS disease the blood-brain barrier may be repaired and functional but with nervous tissue replaced by brous tissue. Fibrosis of nervous tissue may result in signi cant neurologic gait de cits and normal CSF constituents. CSF taken away from the site of the lesion shows normal ndings despite signi cant neurologic gait de cits. For example, CSF in a horse with a cervical spinal cord abscess may show a suppurative in ammation in the lumbosacral CSF and a normal atlantooccipital CSF. This discrepancy is caused by the caudad flow of CSF. CSF may be helpful in supporting the diagnosis of neurologic disease in horses and is part of the diagnostic workup. Because CSF evaluation is an ancillary diagnostic test, one should use it with, and not as a substitute for, a thorough history, physical examination, neurologic examination, and other diagnostic tests.

Electrodiagnostic Aids and Selected Neurologic Diseases Frank M. Andrews, V.A. Lacombe Localizing lesions to and within the nervous system can be di cult in some horses using only clinical and neurologic examinations. Generally, neurologic disease is characterized by changes in cell electric activity; because electric activity has amplitude and frequency, one can measure it by electronic equipment. The observed electric activity may be helpful in de ning and localizing lesions of the nervous system. Electromyography (EMG), auditory brainstem response (ABR) testing, and electroencephalography (EEG) are electrodiagnostic aids that may be helpful in the localization, diagnosis, and prognosis of neurologic disease in horses. Needle EMG and nerve conduction studies are helpful in localizing and de ning diseases of the lower motor neuron or motor unit. ABR testing is helpful in localizing lesions to cranial nerve VIII and auditory pathways along the brainstem. EEG is helpful in diagnosing focal and di use intracranial lesions. Use of these diagnostic aids as an extension of the neurologic examination, separately or collectively, can provide valuable information about nervous system function and help in diagnosing neurologic disease. These techniques are noninvasive and in many cases can be done on the awake horse with mild sedation. Even when these techniques do not provide the information necessary to arrive at a diagnosis, they may provide a more complete understanding of the disease process.

Electromyography Needle EMG is the graphic recording of muscle cell electric activity during contraction or at rest from a recording electrode placed in the muscle. Electric activity is recorded by means of an ampli er on an oscilloscope.1-3 Nerve conduction studies consist of stimulating a

peripheral nerve with electric current and recording the resultant physiologic electric activity from other segments of the nerve or from the muscles innervated by those nerves.4 Together, EMG and nerve conduction studies may help in the localization, diagnosis, and prognosis of diseases of the lower motor unit. The motor unit consists of the ventral horn cell bodies (located in the ventral horn of the spinal cord), its axon, the peripheral nerve, the myoneural junction, and the muscle fibers it innervates (Figure 12-3).

FIGURE 12-3

Illustration of the motor unit, including the ventral horn cell in the spinal cord, ventral root, peripheral nerve, neuromuscular junction, and muscle.

Electromyographic Examination A history and physical and neurologic examination always should precede the EMG examination; these aid in localizing the lesion, shortening the examination time, and minimizing trauma to the horse. Initially, one examines the standing horse that is under mild sedation. One can tranquilize the horse with 0.2 to 0.5 mg/kg xylazine administered intravenously or 0.2 to 0.5 mg/kg xylazine and 2 to 10 mg butorphanol administered intravenously. Examination of the awake horse aids in evaluating individual motor unit action potentials (MUAPs), summated MUAPs, and interference pattern. One can evaluate normal and abnormal MUAPs and measure their amplitudes. Unfortunately, in the awake horse an interference pattern (many MUAPs) sometimes can obscure abnormal low-amplitude EMG potentials. In this case, further examination may require general anesthesia. In the needle EMG examination, one should thrust the exploring electrode briskly into the muscle and hold it until the animal completely relaxes. To enable relaxation, one can force the animal to bear weight on the opposite limb. Once relaxation has occurred, one can evaluate the resting activity or any postinsertional activity of the muscle. One should evaluate at least four areas and depths of smaller skeletal muscles and six to eight areas and depths of larger skeletal muscles when possible. One should examine the horse systematically so as not to miss a lesion. Needle EMGs can also be performed in many of the major extrinsic muscles of the horse (Table 12-3). In addition, they can be performed on facial, laryngeal, esophageal, pectoral, and external anal sphincter muscles when indicated by neurologic examination. One may perform nerve conduction studies and collect muscle biopsy specimens to define suspected lesions further or confirm a diagnosis. TABLE 12-3 Muscles, Nerves, and Nerve Roots Evaluated during Routine Electromyographic Examination of Horses Muscles REAR LIMB Long digital extensor Gastrocnemius Deep digital flexor Semimembranosus Vastus lateralis Biceps femoris Middle gluteal PARAVERTEBRAL Paravertebral muscles (segmentally) THORACIC LIMB

Peripheral Nerve

Spinal Nerve Root

Peroneal nerve Tibial nerve Tibial nerve Ischiatic nerve Femoral nerve Caudal gluteal, ischiatic, and peroneal nerves Cranial and caudal gluteal nerves

L6-S1 S1-S2 S1-S2 L5-S2 L3-L5 L6-S2 L5-S2

Dorsal branches of ventral spinal nerves (L6-C1)

L6-C1

Subclavius Supraspinatus Infraspinatus Deltoideus Biceps brachii Triceps Extensor carpi radials Superficial digital flexor Deep digital flexor

Pectoral nerve Suprascapular nerve

C6-C7, T1 C6-C8

Axillary nerve Musculocutaneous nerve Radial nerve Radial nerve Ulnar nerve Ulnar and median nerve

C6-C8 C6-C8 C7-T1 C7-T1 C8-T2 C7-T1, T2

Nerve Conduction Studies The evaluation of nerve conduction velocities requires knowledge of the topographic anatomy of nerves and muscles, plus a stimulator capable of delivering up to 150 V at durations of 0.1 to 3.0 ms at variable frequencies, up to 100 Hz. Most standard EMGs have built-in stimulators with adequate parameters to do nerve conduction studies. The peripheral nerve to be assessed can be located by palpation or by anatomic landmarks and then can be stimulated. One can palpate or observe the resultant muscle contraction and can view the evoked muscle action potential, which has a thumping sound, on the oscilloscope. Nerve conduction studies are di cult to do in horses and therefore are not done routinely. However, the technique for radial and median nerve conduction studies in the horse has been reported.5,6 One must perform nerve conduction studies with the horse under general anesthesia and may use them to evaluate the speed of conduction of large myelinated motor nerves. To do this, the horse should be placed in lateral recumbency, with the a ected side up. One stimulates the appropriate motor nerve by monopolar needle electrodes inserted at or near the nerve and records an evoked MUAP from an innervated muscle (Figures 12-4 and 12-5). The examiner may see or palpate the contraction of appropriate muscles and insert needle electrodes until a repeatable response is obtained. The una ected limb can be used as a control.

FIGURE 12-4

Illustration of anatomy and electrode placement used in median nerve conduction velocities. DDF, Deep digital flexor.

(From Henry RW, Diesem CD, Wiechers MD: Evaluation of equine radial and median nerve conduction velocities, Am J Vet Res 40:1406–1410, 1979.)

FIGURE 12-5

Illustration of anatomy and electrode placement used in radial nerve conduction velocities. ECR, Extensor carpi radialis; CDE, common digital extensor; LDE, long digital extensor; ECO, extensor carpi obliquus. (From Henry RW, Diesem CD, Wiechers MD: Evaluation of equine radial and median nerve conduction velocities, Am J Vet Res 40:1406–1410, 1979.)

In horses, one can obtain radial nerve recordings from the extensor carpi radialis and abductor digiti longus (extensor carpi obliquus) muscles (see Figure 12-4) and can obtain median nerve recordings from the humeral and radial heads of the deep digital exor tendon5,6 (see Figure 12-5). Facial nerve recordings can be obtained from the levator nasolabialis muscle by stimulating the buccal branch of the facial nerve just ventral to the facial crest.7 Often a supramaximal stimulus can be obtained at 70 to 90 V for 0.1-ms duration. Nerve conduction studies are helpful in diagnosing radial nerve, median nerve, and possibly facial nerve injury.

Normal Electromyographic Potentials Normally occurring EMG potentials and nerve stimulation studies are described next. One can examine the muscle at rest, under submaximal contraction, under maximal contraction, and after direct nerve stimulation.

Insertional Activity Insertional activity consists of short bursts of high-amplitude, moderate- to high-frequency (17 mol/L) may exclude EGS.15 Most horses with EGS and healthy co-grazed horses had signi cantly decreased plasma sulfur amino acid (cysteine and methionine) concentrations compared with healthy control animals. When watching a horse eat, one may notice signs of abnormal esophageal function such as esophageal spasm and reverse peristalsis. Contrast radiography and endoscopy of the esophagus may reveal abnormalities such as megaesophagus and esophageal erosions.6 Examination of an ileal biopsy, collected via laparotomy, is the only method to con rm EGS ante mortem.17,18 Neuronal degeneration is present in the enteric plexuses from the esophagus to the rectum, but the most severe changes are present in the terminal ileum. Examination of a rectal mucosal biopsy is specific but not sensitive for EGS. The most important di erential diagnoses for acute EGS are a small intestinal strangulating lesion and duodenitis-proximal jejunitis. Clinical signs and neuronal lesions in chronic EMND and chronic EGS may be similar, and these similarities have led to the speculation that these two diseases are related.19 Further examination, however, reveals important di erences such as the fact that EMND occurs in older horses that have not had access to pasture, and EGS occurs in young horses with access to pasture. EGS sometimes occurs along with EMND.2,19

OTREATMENT AND PREVENTION No e ective cure exists for EGS. Horses with acute EGS may respond initially to gastric decompression and intravenous uid therapy. Management of horses with chronic EGS includes nursing care and the use of the prokinetic drug cisapride and the appetite stimulant diazepam.6

OPROGNOSIS EGS is usually fatal. The case fatality rate for acute and subacute cases is 100%. If horses with acute EGS are not euthanized within 48 hours, then they die from circulatory failure or gastric rupture. Horses with subacute grass sickness usually are euthanized within 7 days.7 Horses with chronic EGS often are euthanized because of weakness and emaciation; however, when given appropriate care at a referral hospital, approximately 50% of horses may survive.6 Survivors may return to work, but some display residual abnormalities such as mild dysphagia, sweating, and coat changes.

LYME DISEASE IN HORSES Stephen M. Reed, Ramiro E. Toribio Lyme disease is the most common vector-borne infectious disease in human beings in the United States. Initially recognized in the mid-

1970s in Lyme, Connecticut, as the cause of unexplained rheumatoid arthritis in children, the causative agent of Lyme disease was discovered to be a spirochete, Borrelia burgdorferi (sensu lato). In addition to human beings, Lyme disease a ects domestic animals such as dogs, cats, cattle, and horses.1-7 Human beings and animals acquire Lyme disease by transmission of B. burgdorferi through the bite of infected hard ticks (Ixodes spp.). In the Eastern and Midwestern United States the vector is the blacklegged tick or deer tick, Ixodes scapularis (formerly I. dammini), whereas in the Western United States the vector is the western blacklegged tick, I. pacificus. In Europe the sheep tick, I. ricinus is the vector of Lyme borreliosis. Not all ticks are infected with the spirochete, and infection varies by tick species and geographic region. These ticks have a 2-year, threestage life cycle and feed once during each stage.8,9 The larvae hatch in the spring and summer and are usually noninfective because transovarial transmission is rare.10 The tick may become infected at any stage of the life cycle by feeding on small mammalian hosts, typically the white-footed mouse (Peromyscus leucopus), which is a natural host for the spirochete. The nymphal stage emerges the next spring and is most likely to transmit the disease because it is small, di cult to see, and engorges faster; engorgement is necessary to transmit B. burgdorferi. The white-tailed deer (Odocoileus virginianus) is the host for adult ticks. The life cycle of I. ricinus in Europe involves birds and mammals because immature stages feed on birds and reptiles. Nymphal I. scapularis are the main vectors of B. burgdorferi in human beings. These ticks are more active from May to July. Once the tick has engorged, B. burgdorferi is transmitted to the host via lymphatics or blood. Although larval and nymph stages are responsible for B. burgdorferi transmission to other mammals, the stage responsible for transmitting Lyme disease in horses is unknown.

OEPIDEMIOLOGY In endemic areas of the Northeast and Midwestern United States, approximately 20% of nymphal stages and 30% to 40% of adult stages of Ixodes scapularis are infected with Borrelia burgdorferi. In contrast, I. paci cus often feeds on lizards that are poor reservoirs for B. burgdorferi, and only 1% to 3% of these ticks, including the nymphal stages, are infected with the spirochete.10 This di erence in the number of infected ticks may explain the di erence in the prevalence of Lyme borreliosis between the Eastern and Western United States. Equine seroprevalence is high in the Northeastern United States, but limited information is available for other parts of the country.2,6 In Europe, clinical cases and equine serorologic surveys have documented Lyme borreliosis as a cause of equine disease in Poland, Germany, Scandinavia, and England.5,11,12 The apparent increasing incidence of disease in human beings and animals may result from an increased deer population, increased number of ixodid ticks, expansion of human and horse populations into previously rural woodland areas, or increased recognition of the disease manifestations. The disease has a seasonal prevalence and is most common in spring, summer, and fall, with a peak incidence in June and July. In some climates such as California, ticks may be active throughout the year. The organism is maintained in a complex life cycle of small wild mammals and immature stages of the blacklegged tick. Larval and nymphal stages of the tick acquire the infection when they feed on infected mice. The white-tailed deer (Odocoileus virginianus) is not a reservoir for Lyme disease but rather the host for the adult stages of the tick. This is relevant to tick and Lyme disease control programs, because regional reductions in the number of I. scapularis has only been possible when the deer population has been decreased.8

OCLINICAL SIGNS Clinical signs associated with Borrelia burgdorferi in horses, as in human beings, are often nonspeci c and involve multiple body systems. In human beings, clinical signs frequently include annular rash, erythema migrans, myalgia, aseptic meningitis, arthritis, and seventh cranial nerve palsy.10 The disease usually begins as a skin rash and often progresses to involve joints and the nervous and cardiac systems but may involve other body systems as well. In human beings the skin rash may be slowly progressive and sometimes may take up to 1 month before becoming apparent. In horses, reported clinical signs of Lyme borreliosis include chronic weight loss, sporadic lameness, laminitis, low-grade fever, swollen joints, muscle tenderness, anterior uveitis, encephalitis, and abortion.3,4,13 Some horses may demonstrate clinical illness and depression and may go o feed in a short time. The bacteria may enter the CNS within a short time after exposure. Chronic arthritis may develop as a result of autoimmune mechanisms, although this is not fully understood. Ixodid spp. ticks are often co-infected with Borrelia burgdorferi and Anaplasma phagocytophila; however, dual infections remain to be documented in the horse. An important consideration is that limb swelling and response to tetracycline treatment, indicating Lyme borreliosis could be the result of A. phagocytophila infection.

DIAGNOSIS

The diagnosis of Lyme disease is often di cult. History of tick exposure or living in a Lyme disease endemic area is helpful; when combined with the identi cation of clinical signs and the elimination of other diseases, this information allows the clinician to make a presumptive diagnosis. Examination of blood work for other diseases may be more bene cial than a positive blood test for Lyme, although one should complete such a test as well. In human beings an early increase in serum IgM often occurs. Response to therapy also helps to support a presumptive diagnosis of Lyme disease. Diagnosis of Lyme disease in horses is di cult,14 and in many cases one bases a presumptive diagnosis on history, clinical signs, and response to antibiotic therapy in an animal with probable cause to be infected (i.e., possible exposure). In human beings the skin lesions are obvious. Blood tests are of limited value; however, ELISA, kinetic ELISA, Western blot testing, and PCR on blood samples and synovial uid from suspect animals have been evaluated.2,5-7,14,15 One may test samples of joint uid, CSF, and tissues from a ected patients for presence of organisms. Several pathologic conditions should be considered in the di erential diagnosis of equine borreliosis, including Anaplasma phagocytophila infection, chronic diseases, vasculitis, and immune-mediated arthritis if limb or periarticular swelling are present, as well as causes of neurologic disease.

OPATHOGENESIS Borrelia burgdorferi organisms are capable of nonspeci cally activating monocytes, macrophages, and synovial lining cells, as well as natural killer cells, B cells, and complement, leading to production of host proin ammatory mediators. These proin ammatory mediators seem to localize in joints, leading to chronic arthritis and lameness. B. burgdorferi has developed strategies to interact with the mammalian host, including adhesion to host cells and components of the extracellular matrix such as bronectin, β 3 integrins, and glycosaminoglycans. The current thinking in the development of arthritis is that outer-surface proteins (Osps) of B. burgdorferi trigger an autoimmune reaction because antibodies to OspA and OspB have been detected in 50% to 80% of human patients with arthritis.16,17 Natural killer cells also play a central role in joint inflammation by producing excessive amounts of TNF-α and IL-8.16

OTREATMENT AND PREVENTION Many horses are treated annually for a presumptive diagnosis of Lyme disease. Recommended treatment includes oxytetracycline (6.6 mg/kg IV every 24 hours) or doxycycline (10 mg/kg PO every 12 hours).18 Treatment can be started with tetracycline for 1 week followed with doxycycline for 3 to 4 weeks. Ceftiofur (2.2 to 4.4 mg/kg IV every 12 hours) has also been evaluated.18 In animals infected with B. burgdorferi, a rapid clinical response (2 to 4 days) is expected with tetracyclines. In some horses the clinical signs may show an initial worsening as a response to toxins released after death of the organisms. Currently, no licensed vaccine is available to prevent Lyme disease in horses, although vaccines are available for dogs and human beings. A recombinant OspA vaccine was evaluated in horses with promising results.19 Aids to prevention include daily grooming with removal of ticks, along with the use of tick repellents that contain permethrin. One should apply tick repellents to the head, neck, legs, abdomen, and under the tail. Keeping pastures mowed and removing brush and woodpiles makes the environment less hospitable for rodents, which in turn decreases the tick population. Regional programs to control Lyme disease are based on reducing the deer population.

HEAD SHAKING Robert H. Mealey

OCLINICAL SIGNS Head shaking is a widely recognized disorder characterized by persistent or intermittent, spontaneous, and frequently repetitive vertical, horizontal, or rotary movements of the head and neck.1 The a ected horse shakes, jerks, or icks its head uncontrollably in the absence of obvious physical stimuli, and the condition can be severe enough to render the horse unusable and dangerous.2-5 Head shaking often is accompanied by snorting and sneezing, and many horses rub their noses on stationary objects or on the ground while moving. In addition to shaking the head up and down, ipping the upper lip, or shaking the head side to side, horses also may act as if a bee has own up the nostril or may strike at the face with the forelimbs.2,6,7 Head-shaking horses also may exhibit avoidance behavior, with low head carriage, corner seeking, and nostril clamping.1,8 To characterize the severity of clinical signs, a grading system of 1 to 5 has been described, with grade 1 being a rideable horse with intermittent and mild signs (facial muscle twitching) and grade 5 being an unrideable, uncontrollable, dangerous horse with bizarre behavior patterns.1 Clinical signs are usually worse during exercise; however, head shaking can a ect horses at rest.1-57 Mean age of onset is typically 7 to 9 years, and Thoroughbreds and geldings appear to be over-represented in some studies.2,3,6 Head shaking can be seasonal or nonseasonal. Early literature suggested an increased incidence of head shaking during the warmer months of the

year,5 and a later study indicated that the peak periods of onset were spring and early summer.3 Recent studies have found that 64% 6 and 59% 2 of head-shaking horses are affected seasonally, with the majority developing signs in the spring or early summer.

OPATHOGENESIS Early work suggested 58 possible causes for head shaking, including two that have received considerable attention recently: (1) photophobia and (2) trigeminal neuralgia.9 Other reported and suggested causes include behavioral resentment to rider-induced head and neck exion, exercise-induced hypoxia, ear mites, cranial nerve dysfunction, otitis media or otitis interna, cervical injury, ocular disease, guttural pouch mycosis, dental periapical osteitis, vasomotor rhinitis, allergic rhinitis, Trombicula autumnalis (harvest mite) larval infestation, maxillary osteoma, and EPM.3,4,9-14 A study of 100 head-shaking horses revealed a potential speci c cause in only 11 horses, and elimination of the abnormality resulted in resolution of the head shaking in only two of them.3 Idiopathic head shaking has been used to describe the majority of cases in which no specific underlying cause is found, and necropsy results in these horses reveal no lesions. The clinical signs exhibited by most head-shaking horses now generally are thought to result from sharp, electric-like, burning pain involving the trigeminal nerve.1,2 Trigeminal nerve involvement is supported by the observations that some horses improve after blocking the infraorbital nerve (part of the maxillary branch),15,16 and that most horses improve after blocking the posterior ethmoidal nerve (part of the ophthalmic branch).1 Horses that head shake in response to light stimulation have been described as photic head shakers.8 Most of these horses are a ected seasonally (spring and summer), improve at night or when brought indoors, and improve when blindfolded or when gray lenses are placed over their eyes. Photic head shaking is hypothesized to be caused by optic-trigeminal summation (stimulation of the optic nerve that results in referred sensation to parts of the nose innervated by the trigeminal nerve), a mechanism similar to that proposed for photic sneezing in human beings.2,8 Many features of head shaking in the horse are similar to trigeminal neuralgia in human beings—a severe chronic pain syndrome characterized by dramatic, brief stabbing or electric shocklike pain paroxysms felt in one or more divisions of the trigeminal distribution, spontaneously or on gentle tactile stimulation of a trigger point on the face or oral cavity. The disease is associated with microvascular compression and pathologic changes in the trigeminal root and trigeminal ganglion.17 The basic underlying cause of head shaking in the horse may be trigeminal neuralgia,1 and recent work suggests a degenerative disorder of the brainstem nucleus of the trigeminal nerve (D.C. Knottenbelt, personal communication). The trigeminal nerve is postulated to be hypersensitive and to re in response to a variety of trigger factors including wind, airway turbulence, increased blood ow, pollen, dust, warmth, cold, insects, allergies, or other irritations. These trigger factors appear to act intranasally in many horses but could act in any trigeminal sensory region.1

ODIAGNOSIS AND TREATMENT One should perform a complete physical examination, including neurologic, dental, ophthalmic, and otoscopic examinations to rule out potential underlying causes. Other diagnostics could include endoscopy of the nasal passages, pharynx, and guttural pouches; radiography of the skull and cervical spine; CBC; and serum biochemistry pro le. These examinations reveal no abnormalities in most cases. A recent survey suggests that one should consider idiopathic head shaking if two of the following three clinical signs are present: (1) vertical ipping of the head, (2) acting as if an insect has own up the nostril, or (3) rubbing the muzzle on objects.2 For photic head-shaking horses, clinical signs should improve after blindfolding or after placement of a mask shielding the eyes from the sun. Cyproheptadine, an antihistamine (histamine1) and serotonin antagonist with anticholinergic e ects, has been used (0.3 mg/kg orally b.i.d.) to treat head-shaking horses, resulting in improvement in 5 of 7 horses in one study8 and 43 of 61 horses in another.2 Horses that respond do so within 1 week, and some may respond within 24 hours. Although the mechanism for e cacy of cyproheptadine in these cases is unknown, blocking of serotonin-mediated pain sensation could be involved.8 Other investigators have found cyproheptadine alone to be ine ective, but the addition of carbamazepine (4 to 8 mg/kg PO q6-8h) resulted in 80% to 100% improvement in seven of nine cases, with horses responding within 3 to 4 days.1 Carbamazepine is a sodium channel–blocking antiepileptic drug and is the drug of choice for treating trigeminal neuralgia in human beings.18 Carbamazepine alone was reported to be e ective in head shaking horses, but results were unpredictable.1 The elimination half-life of carbamazepine in the horse is less than 2 hours, making sustaining therapeutic concentrations di cult; the drug is therefore of more bene t diagnostically than therapeutically in head-shaking horses (D.C. Knottenbelt, personal communication). Other medications and therapies including NSAIDs, corticosteroids, antihistamines, acupuncture, chiropractic manipulation, homeopathy, and feed supplements are generally ineffective in most horses.2,19 Diagnostic blockade of trigeminal nerve branches can be useful for identifying trigger points in head-shaking horses. Blockade of the infraorbital nerve may improve some horses and might identify candidates for infraorbital neurectomy.15,16 However, results of infraorbital blockade and infraorbital neurectomy do not correlate, and because of neuroma formation, self-trauma, and low success rate, infraorbital neurectomy is not a recommended procedure.15 A high percentage of horses improve after blockade of the posterior ethmoidal nerve, and sclerosis of this nerve results in temporary improvement in some horses.1

Therapies aimed at decreasing the response to trigger factors can be e ective and include tinted contact lenses; face masks or hoods that block sunlight, wind, insects, and dust, with mesh or dark eye cups; and nose nets.1,2,8,19,20 A recent owner survey found that nose nets that cover the nostrils with mesh and include a drawstring or elastic band that applies pressure to the upper lip resulted in some improvement in 70% of head-shaking horses, and that 70% or more improvement occurred in about 30% of the horses.20 Finally, permanent tracheostomy is e ective in some horses,1,4 presumably because air ow then bypasses the nasal cavity where the majority of trigger factors are thought to act.1

VERMINOUS ENCEPHALOMYELITIS Eduard Jose-Cunilleras Aberrant migration of helminth and y larvae through the CNS of horses and donkeys is a rare but important cause of severe neurologic disease. Parasites that have been reported to a ect the brain and spinal cord of equids include rhabditid nematodes (Halicephalobus [Micronema] deletrix [syn. H. gingivalis]), strongyloid nematodes (Strongylus vulgaris, S. equinus, and Angiostrongylus cantonensis), protostrongyloid nematodes (Parelaphostrongylus tenuis), spiruroid nematodes (Draschia megastoma), larid nematodes (Setaria digitata and other Setaria spp.), and warble fly larvae (Hypoderma spp.). Antemortem diagnosis is often impossible; however, a high index of suspicion may be warranted for certain clinical signs (acute onset or rapidly progressive asymmetric, focal, or multifocal brain or spinal cord signs) and changes in CSF analysis, in which case the treatment regimen should include speci c antiparasitic drugs (e.g., fenbendazole) in addition to the more routinely used symptomatic and antiinflammatory treatments (e.g., NSAIDs, DMSO, corticosteroids, antiprotozoal drugs).

OCAUSES Halicephalobus (Micronema) Deletrix (Halicephalobus Gingivalis) Halicephalobus deletrix is synonymous with H. gingivalis, the latter being the most recent terminology.1 The parasite was referred to as Micronema deletrix until 1970s and 1980s with use of the new nomenclature beginning in 1990. The life cycle of Halicephalobus spp. nematodes is unknown. This small roundworm generally is considered a saprophytic organism that occasionally acts as a facultative parasite in horses and human beings. The likely route of entry is through nasal and oral mucosa, followed by possible hematogenous spread to organs with high vascularization such as the brain, spinal cord, and kidneys. H. deletrix recently has been shown to be transmitted from a mare to her foal because of mammary infestation.2 Organs a ected by migration of Halicephalobus spp. include the brain, spinal cord, nasal and oral cavities, pituitary gland, and kidneys and less commonly the lymph nodes, heart, lungs, stomach, liver, and bones.3,4 Organisms a ecting the CNS have been reported to have a predilection for the basilar, pituitary region of the brainstem.5 Characteristic histopathologic lesions in the CNS include malacia, granulomatous and lymphohistiocytic in ammatory in ltration, meningitis, and vasculitis in addition to identi cation of the nematodes. Other clinical signs include oral and nasal granulomata, renal involvement (H. deletrix is visible in the urine), granulomatous osteomyelitis,6 and granulomatous chorioretinitis.7 Most of the cases of Halicephalobus encephalomyelitis in adult horses reported in the literature during the last 30 years describe simultaneous renal granulomatous lesions encapsulating the nematodes (Table 12-14). Granulomata within or adjacent to the renal parenchyma were observed on postmortem examination in 13 of 16 horses with neurologic disease; of the other three horses, in one case lesions were con ned to the sacral spinal cord and cauda equine, and in another case only the skull and brain were examined at postmortem examination. In contrast, all three cases reported to date in foals showed no renal involvement but did show pulmonary granulomata.2,5 TABLE 12-14 Signalment, Organism Identified, Clinical Signs, and Organs Affected in Cases of Verminous Encephalomyelitis

Strongylus Vulgaris, Strongylus Equinus, and Angiostrongylus Cantonensis Aberrant strongyle larval migration is a much less common cause of neurologic signs because of routine broad spectrum antiparasitic treatment with ivermectin and moxidectin. The pathogenic mechanisms described for strongyle encephalomyelitis include aberrantly migrating fourth- and fth-stage larvae in the intima of the aorta or left ventricle, which causes endothelial damage, stimulates the clotting cascade, and results in formation of a thrombus that often contains the parasitic larva.21 Lesions are generally asymmetrical because of random migration in the brain, although migration along the spinal cord has been reported in a donkey.22

Parelaphostrongylus tenuis The meningeal worm Parelaphostrongylus tenuis causes neurologic disease in cervids, ovids, bovids, and camelids. Recently, there have been multiple reports of similar disease in horses ranging in age between 6 months to 9 years.28-30 A ected horses have a history of sudden onset of scoliosis with no history or evidence of trauma.

Draschia megastoma Adult Draschia megastoma worms are found in the stomach of equids where they cause mucosal granulomatous masses and mild chronic gastritis. The life cycle of Draschia is indirect because the organism uses ies as the intermediate host. Eggs and larvae are released into the gastric lumen, and the rst larval stage passes in the feces and is ingested by y larvae of the genus Musca in which Draschia spp. organisms develop to infective larvae. Finally, horses become infected when the third larval stage is deposited on the host by adult ies. Infective larvae that are ingested and reach the stomach develop into adults and complete the life cycle. Larvae deposited in damaged skin result in local inflammation and development of extensive granulation tissue, which is the typical lesion of cutaneous habronemiasis. Stray D. megastoma larvae may be found anywhere throughout the body, and a case has been reported of D. megastoma migration in the brainstem of a horse in Southern United States, resulting in asymmetrical brainstem disease.23

Setaria Species Infestation with the larial nematode Setaria is common in the abdominal cavity of ungulates where the organism does not cause signi cant clinical e ects. S. labiatopapillosa is found in cattle, and S. equina is found in horses. Micro lariae gain access to peripheral circulation and then are transmitted to other potential hosts via mosquitoes. In a postmortem examination study of 305 horses in Japan, S. equina was recovered from 66 of those horses (≈22%). The cattle parasite S. digitata occurs only in Asia, and infestation in unnatural hosts (horses, sheep, goats, camel, and human beings) has been associated with cerebrospinal nematodiasis. This condition can be enzootic in India, Burma, and Sri Lanka, is called Kumri (weak back), and has a seasonal occurrence, usually during the fly season (late summer and fall).31 Two cases of aberrant S. digitata CNS migration have been reported in Japanese racehorses, and one case of Setaria spp. larval migration

in the brainstem and cervical spinal cord has been reported in a 12-year-old Quarter Horse mare in the Midwestern United States.20,32

Angiostrongylus cantonensis Adult parasites are found in the right ventricle and pulmonary artery of rats. The pulmonary circulation carries eggs released by adult worms to the alveoli, where the rst larval stage develops and migrates up the trachea, is swallowed, and passes in feces. Snails and slugs are intermediate hosts, and ingestion of the snail results in ingestion of the infective larvae that migrate through CNS, and finally, the larvae reach the heart via the circulation. Aberrant CNS migration of Angiostrongylus cantonensis larvae has been reported in two foals with tetraparesis in Australia,24 and aberrant migration is a recognized cause of eosinophilic meningoencephalitis in human beings and dogs.33

Hypoderma Species Cattle are normal hosts of Hypoderma spp., but horses are accidental hosts of warble y larvae. In cattle, Hypoderma larvae penetrate the skin after hatching from eggs attached to hair in the rump and lower legs. These larvae migrate through connective tissues to reach the esophagus, in the case of H. lineatum, or to the vertebral canal around epidural fat, in the case of H. bovis. Finally, the larvae migrate back to the skin over the back where they create a breathing hole. Cutaneous myiasis is uncommon in horses, and warble y larvae occasionally may migrate aberrantly into the brain. Tissue damage causing necrosis and hemorrhage is extensive because of the big size of the instars.25 Instars can enter through large natural foramina such as the foramen magnum, optic foramen, and occasionally intervertebral foramina.

OCLINICAL SIGNS Severity and duration of clinical signs vary from mild, transient, and insidious to severe and fatal. In some cases, clinical signs progress over 2 to 4 months, with periods of improvement or stabilization.4,8,23 Variability of clinical signs depends on the number of parasites (thromboembolic shower of Strongylus vulgaris), on the size of the migrating organism (Hypoderma spp. larvae are large and cause severe necrosis and hemorrhage as they migrate in the CNS parenchyma), and neuroanatomic localization of the lesions. Horses su ering from larval migrations in the brain (S. vulgaris, Halicephalobus deletrix, Draschia megastoma, and Hypoderma spp.) may display head tilt, circling, recumbency, blindness, hyperesthesia, sti neck, ataxia, head pressing, recumbency, seizures, and coma. In those cases in which larvae cause lesions in the spinal cord (S. vulgaris, H. deletrix, Setaria spp., Angiostrongylus cantonensis), clinical signs may include focal or multifocal asymmetrical ataxia, weakness, dog-sitting posture caused by paraparesis, increased patellar re exes, atonic bladder, poor tail tone, and poor rectal tone with impaction of feces (see Table 12-13). Recently, a case of cauda equine neuritis has been reported that was caused by H. gingivalis migration in the sacral spinal cord and cauda equina nerves.15 Clinical signs of Parelaphostrongylus tenuis infection are quite characteristic and include acute onset of scoliosis (observed in 90% of reported cases), with progressive gait de cits. Scoliosis is most common in the cervical area.28-30 The a ected area of the vertebral column is usually C shaped, with accid muscles and decreased cutaneous sensation on the convex side of the curve. Tetraparesis and ataxia is mild to moderate and predominantly ipsilateral to the convex side of the curve. If bilateral gait de cits are observed, then they are worse on the convex side. The proposed pathogenesis for clinical signs involves unilateral weakness of the paraspinal epaxial muscles on the convex side of the curve.30 A linear lesion that extends for several segments, most likely in the dorsal gray column, is most likely to cause the observed signs.

ODIFFERENTIAL DIAGNOSIS One should consider verminous encephalomyelitis in all cases of acute or progressive asymmetrical disease of the spinal cord, cerebrum, cerebellum, or brainstem. If brain involvement is evident without other localizing signs, then the di erential diagnosis list may include equine togaviral encephalomyelitis, rabies, equine protozoal myelitis (EPM), trauma, cerebral abscess or basilar epidural empyema, bacterial meningitis, hepatic encephalopathy, neoplasia, and leukoencephalomalacia. If the neurologic signs are limited to spinal cord involvement, then other diseases to include in the di erential diagnosis list may be cervical stenotic myelopathy, EPM, EHV-1 myeloencephalopathy, trauma, WNV meningoencephalitis, EDM, trauma, spinal osteomyelitis or discospondylitis, vertebral fracture, and neoplasia. If the only signs present are related to cauda equina syndrome, then other conditions to consider are polyneuritis equi, sacral or coccygeal fracture, EPM, EHV1 myeloencephalopathy, sorghum or Sudan grass toxicity, epidural abscess from tail blocking, and neoplasia. Initial diagnostics should include CBC, serum chemistry pro le, urinalysis, and CSF collection for cytologic evaluation. In addition, serologic examination and viral isolation and CSF Western blot for EPM may help eliminate several more common viral and protozoal myeloencephalitides. Radiographs of the cervical or lumbosacral spine and myelography may be required to rule out more common causes of spinal cord disease (cervical stenotic myelopathy, trauma, fractures). Although only routinely available at referral hospitals, advanced

imaging techniques (e.g., computerized axial tomography, MRI) may be useful in diagnosing parasitic encephalomyelitis or other conditions with similar clinical signs. CSF analysis in cases of parasitic encephalomyelitis can be normal; however, CSF changes are common and include xanthochromia, increased protein, and neutrophilic and mononuclear pleocytosis, but eosinophils rarely occur (Table 12-15). Only in a case of Angiostrongylus cantonensis were eosinophils the predominant cell type on CSF cytologic examination. In addition, Halicephalobus deletrix eggs are visible on microscopic examination of a CSF sample subjected to cytocentrifugation.5 Similarly, H. deletrix larvae may be visible on microscopic examination of urine sediment or sem*n in cases of renal and testicular involvement.4 Analysis of CSF obtained from horses with Parelaphostrongylus tenuis neurologic disease has been within normal limits.30 TABLE 12-15 Cerebrospinal Fluid Analysis in Verminous Encephalomyelitis

Antemortem diagnosis may be possible in those cases in which renal or bony involvement is detected and nematodes are identi ed in biopsies of the a ected tissues.6 As described previously, renal involvement with granulomatous lesions encapsulating Halicephalobus deletrix worms is observed in most cases of H. deletrix encephalomyelitis. Therefore transabdominal renal ultrasound examination and ultrasound-guided biopsy (in those cases with renal lesions present), as well as microscopic examination of urine sediment, may prove useful in cases of acute onset progressive asymmetrical neurologic disease, especially if one suspects cerebrospinal nematodiasis caused by H. deletrix. Additionally, a PCR–based diagnostic test has been developed for confirmation of Setaria encephalomyelitis in goats, sheep, and horses.34 This test is based on specific amplification of Setaria spp. filarial DNA from a blood sample from the host.

OPATHOLOGIC FINDINGS The gross and histopathologic lesions depend on the parasite involved. Thorough postmortem examination with sectioning and histopathologic examination of all areas of the CNS relevant to the antemortem clinical signs are important to identify migrating parasites. Strongylus vulgaris and Hypoderma spp. larvae are easy to see, but other nematodes are only visible on microscopic examination. Some reports describe how one may recover and examine whole, xed nematodes for distinctive morphologic features by centrifugation of formalin solution in which the brain had been xed.9,18 Gross examination of other tissues for evidence of S. vulgaris thrombi or presence of Halicephalobus deletrix granulomatous lesions may help establish a postmortem diagnosis. On histopathologic examination, one generally sees extensive tissue necrosis with mixed in ammatory response. H. deletrix migration in the CNS typically results in histiolymphocytic in ltrates, malacia, glial proliferation, and perivascular lymphocytic cu ng. Migrations of warble fly and S. vulgaris larvae result in severe hemorrhage, large malacic tracts, and edema caused by their relative larger size.

OTREATMENT Treatment of verminous encephalomyelitis in horses is often unrewarding. None of the cases reported in the literature have responded favorably to anti-in ammatory drugs and antihelmintics. Only in one case of Halicephalobus deletrix granuloma limited to the prepuce, therapy with ivermectin and diethylcarbamazine was successful.35 However, numerous cases of cerebrospinal nematodiasis likely respond favorably to anti-in ammatory drugs and antihelmintics, but one never reaches a de nitive diagnosis because diagnosis requires a postmortem examination. Therapeutic recommendations depend on the severity, progression, and localization of neurologic signs, as well as on consideration of possible contraindications (e.g., if one suspects bacterial or viral cause, then one should avoid using corticosteroids). In cases of acute neurologic disease use of the following anti-in ammatory drugs may be warranted: NSAIDs such as phenylbutazone (2.2 mg/kg b.i.d.), unixin meglumine (1 mg/kg b.i.d.), or ketoprofen (2 mg/kg b.i.d.); intravenously administered DMSO (1 g/kg as a 10% solution once daily for 3 to 5 days); and corticosteroids (dexamethasone at 0.1 to 0.25 mg/kg once daily or prednisolone at 0.2 to 4.4 mg/kg once daily). If cerebral signs are observed, then one may administer mannitol (0.25 to 2 g/kg as a 20% solution once daily), or hypertonic saline may be warranted in an attempt to minimize cerebral edema. In countries in which EPM is known to occur, antiprotozoal treatment is recommended. Speci c antiparasitics suggested for treating verminous encephalomyelitis include benzimidazole compounds (oxfendazole, thiabendazole, fenbendazole, and mebendazole), diethylcarbamazine and ivermectin for the treatment of nematodes, and organophosphates (trichlorfon and dichlorvos) for the treatment of warble y larvae. Although ivermectin is e ective against most equine parasites, it may not be the best choice because of its delayed method of killing, which may take as long as 10 to 14 days. Some authors have suggested administration of fenbendazole at 50 mg/kg by mouth once daily for 3 days36,37; however, speci c data on the e cacy of antihelmintics in treating nematode or warble fly larvae migration through the CNS are not available.

MISCELLANEOUS NEUROLOGIC DISORDERS Debra C. Sellon In addition to the numerous neurologic problems discussed in detail earlier in this chapter, horses can develop a wide variety of acute or chronic neurologic disorders with clinical signs related to brain, spinal cord, or peripheral nerve dysfunction. Clinical signs can occur because of primary disease or dysfunction of the CNS or secondary to a variety of metabolic disorders or disorders of other body systems. This section will mention the most important of the miscellaneous CNS disorders and refer the reader to relevant chapters for more detailed information.

ONEUROLOGIC NEOPLASIA Many types of neoplastic lesions have been identi ed in the brain and spinal cord of horses. A comprehensive review of the clinical and pathologic aspects of equine CNS neoplasia, however, is lacking in the peer-reviewed veterinary literature. Available case reports and summaries of pathologic studies suggest that CNS neoplasia is quite uncommon in horses. In one Australian survey of 450 horses with neurologic disease, the prevalence of neurologic disease secondary to neoplasia was less than 2%.1 This percentage is likely to be much lower in horses in the United States because of the much higher incidence of encephalitic disorders and EPM. Tumors of the nervous system can be classi ed on the basis of their cellular origin as tumors of nerve cells, neuroepithelium, glia, peripheral nerves and nerve sheaths, mesodermal structures, and endocrine organs. Neoplasms of nerve cell origin appear to be extremely uncommon in horses with only a few reports of ganglioneuromas, complex tumors arising in peripheral ganglia and composed of welldi erentiated neurons, nerve processes, Schwann cells, and glial cells.2,3 These tumors may cause intestinal obstruction and signs of colic in affected horses. Tumors of neuroepithelial origin in horses include ependymoma, choroid plexus papilloma, neuroepithelial tumor of the optic nerve, malignant medulloepithelioma, ocular medulloepithelioma, and pineoblastoma.4-19 Glial cell tumors include optic disc astrocytoma, retinoblastoma, and microglioma.13,19,20 Neoplasms of peripheral nerves and nerve sheaths appear to be more common than CNS neoplasia and include neuro bromas and neuro brosarcomas.21-26 The most common sites for neuro bromas are in the cutaneous tissues of the pectoral region, abdomen, neck, and face but gastrointestinal tract lesions have also been described.21 There are also reports of intracranial and mediastinal schwannomas in horses.27,28 The most common mesodermal neoplasm in horses appears to be meningioma.5,18,29 Neoplastic reticulosis, lipoma in the mesencephalic aqueduct, angioma in the cervical spinal cord, primary meningeal lymphoma, and melanoblastoma of the cerebellar meninges have also been described.30-32 Older texts often refer to pituitary pars intermedia dysfunction (PPID) of older horses as a neoplastic lesion. This disorder is most likely a

manifestation of oxidative damage to dopamine-producing neurons in the central nervous system with subsequent dysregulation of hormone production within the pars intermedia of the pituitary gland. PPID is discussed in detail in the Endocrinology chapter of this text. Cholesterol granulomas (cholesteatomas, cholesterinic granulomas) are likewise not neoplastic lesions, though they may have been described as such in the past.33-41 These lesions are found in the choroid plexus of up to 20% of older horses.42 Although more common in the choroid plexus of the fourth ventricle, lesions in the lateral ventricles may be more likely to result in clinical signs. Lesions may represent a chronic granulomatous reaction to deposition of cholesterol crystals associated with chronic choroid plexus leakage. Grossly the granulomas are circ*mscribed, rm, granular, and yellow-brown with a glistening cut surface. They are most often recognized as incidental ndings at necropsy of horses without noticeable clinical signs. If they are large enough, however, they might cause CNS signs as a result of direct pressure on cerebral tissues or obstructive hydrocephalus. In a ected horses, reported clinical signs have included altered behavior, depression, somnolence, seizures, ataxia, weakness, and coma. CSF from a ected horses may have an elevated protein and appear xanthochromic.33-35 Secondary neoplastic conditions a ecting the CNS may penetrate the cranial vault or vertebrae, extend through osseous foramina, or metastasize via the vascular system. Clinical signs vary depending on the type, site, and extent of the lesion. Lymphosarcoma is probably the most common secondary neoplasm of the CNS in horses.32,43-48 Lesions may cause forebrain signs if present in the cranial vault or ataxia and paresis if causing compression of the spinal cord. Melanoma of the spinal cord, meninges, and brain may occur with metastasis of primary cutaneous lesions.49-51 Alternatively, clinical signs may represent spread from affected sublumbar lymph nodes.32,52 Melanoma of the CNS is most common in gray horses. Hemangiosarcoma and undi erentiated sarcoma have been reported to a ect the CNS of horses32,53-55 There are numerous descriptions of adenocarcinomas and carcinomas (including squamous cell carcinoma) a ecting the CNS of horses by direct spread from a primary site in the head or by metastasis from a distant site.32,56-62 Tumors of the bone or bone marrow may involve the skull or vertebrae resulting in compression of the brain or spinal cord, respectively.32,63-65 Malformation tumors, including epidermoids, dermoids, teratomas, and teratoids, may also a ect the CNS.32 They are often incidental findings at necropsy but may occasionally be clinically significant. Supportive evidence for the presence of CNS neoplasia can be obtained with advanced imaging (i.e., computed tomography, magnetic resonance imaging) of the brain or spinal cord in select equine patients. De nitive diagnosis of primary CNS neoplasia in horses requires cytologic or histopathologic identi cation of neoplastic cells or tissues. This diagnosis necessitates biopsy, cytologic evaluation of an aspirate from a suspect lesion, or identi cation of neoplastic cells in cerebrospinal uid (CSF). Analysis of CSF is rarely diagnostic of neoplasia in horses because of the rarity of the presence of neoplastic cells in the uid. Given the practical di culties inherent in obtaining biopsy samples from the brain or spinal cord, de nitive diagnosis of primary CNS neoplasia in horses rarely occurs antemortem. De nitive diagnosis is more likely in horses with secondary CNS neoplasia in which a diagnosis is made by biopsy or cytologic evaluation of primary lesions identified external to the CNS.

OTOXIC DISORDERS A wide variety of toxic substances can a ect the CNS of horses. Most of these are discussed in detail in the Toxicology chapter of this text. Three toxic neurologic conditions merit mention in this chapter because of their common occurrence in some geographic areas and their distinctive clinical appearance as demonstrated in the referenced video segments on the DVD accompanying this text.

Leukoencephalomalacia Equine leukoencephalomalacia is a seasonal disorder of horses, ponies, donkeys, and mules, occurring most commonly in late fall through early spring. Most outbreaks are associated with a dry growing period followed by a wet period. The disorder is caused by ingestion of the mycotoxin fumonisin B1, a metabolite of Fusarium moniliforme.66-88 Two clinical syndromes are associated with fumonisin B1 intoxication. The more common is a neurologic syndrome characterized initially by incoordination, aimless walking, intermittent anorexia, lethargy, depression, blindness, and head pressing. These signs may be followed by hyperexcitablity, belligerence, extreme agitation, profuse sweating, and delirium. Recumbency and clonic-tetanic seizures may occur before death. Less commonly, horses develop a hepatotoxic syndrome with swelling of the lips and nose, somnolence, severe icterus and petechiation of mucous membranes, abdominal breathing, and cyanosis. Gross lesions include liquefactive necrosis and degeneration of the cerebral hemispheres; changes may also occur in the brainstem, cerebellum, and spinal cord.

Nigropallidal Encephalomalacia Yellow star thistle and Russian knapweed grow over much of the western United States in nonirrigated pastures during dry seasons of

summer and fall. Ingestion of these plants result in unilateral or more commonly bilateral softening and necrosis in areas of the globus pallidus and substantia nigra.89-98 Lesions are often sharply de ned and may be cavitary. Although the plants are generally considered unpalatable by most horses, some horses may develop a craving for the plants and selectively seek them out. The exact toxic principle in these plants has not been determined. A ected horses demonstrate variable degrees of impairment of eating and drinking with lack of coordination of movements of prehension, mastication, and deglutition. Most horses appear to be able to swallow if food or water is placed in the posterior pharynx. Severely a ected horses might attempt to drink by immersing their muzzle deeply into the water in an apparent attempt to force water into the posterior pharynx. Hypertonicity of facial muscles is common with the horse often holding the mouth partially open with the lips retracted. The tongue may protrude from the mouth and many horses display constant chewing movements.

Fluphenazine Toxicity Fluphenazine administration to horses may result in characteristic clinical signs of toxicity.99-103 Fluphenazine is a highly potent phenothiazine neuroleptic that is widely used in human medicine for a variety of psychological disturbances. In horses, uphenazine decanoate has been used to provide a persistent sedative e ect. It binds avidly to dopamine D2 receptors in the brain. The very slow dissociation of uphenazine from these receptors is associated with a greater risk of adverse extrapyramidal signs than is observed with newer atypical antipsychotic medications.99 Its administration to some horses can result in severe extrapyramidal e ects and Parkinsonism with clinical signs including agitation, profuse sweating, hypermetria, aimless circling, intense pawing and striking with the thoracic limbs, and rhythmic swinging of the head and neck alternating with periods of severe stupor. There are no published reports of the pharmaco*kinetics of uphenazine in horses. A long-acting depot formulation for IM administration is available and appears to be most commonly used. One published report suggests a dosage of 0.05 to 0.08 mg/kg IM every 2 weeks with a warning to beware of idiosyncratic reactions.104 Horses have exhibited signs of extrapyramidal e ects of uphenazine decanoate after receiving doses as low as 40 mg IM.99 Given that recommended doses for adult humans range between 25 to 50 mg, it is reasonable to assume that at least some horses are much more susceptible than humans to the adverse effects of fluphenazine decanoate. The severity of clinical signs observed in some affected horses is su cient to pose considerable safety risks for veterinarians and handlers treating these horses. This author is aware of at least one horse that was euthanized because of the extreme danger of clinical signs after administration of uphenazine; other authors have indicated similar adverse outcomes.99 Treatment consists of discontinuation of drug administration and enhancing cholinergic function by administering anticholinergic medications. Diphenhydramine chloride results in signi cant improvement in some, but not all, a ected horses.99,101 Benztropine mesylate (0.018 mg/kg IV every 8 hours) has also been helpful for treatment of some a ected horses.99 Other horses apparently bene ted from sodium pentobarbital (2 mg/kg IV followed by a constant rate infusion at 2.5 mg/kg/h) in an attempt to control maniacal behavior without inducing recumbency or anesthesia.99

OMETABOLIC DISORDERS A wide variety of metabolic disorders can cause clinical signs indicative of central nervous system dysfunction. These signs include hepatic dysfunction, hypoglycemia, hypoxemia and ischemia, and severe abnormalities in plasma electrolyte concentrations.

Hepatoencephalopathy An excellent review of the etiology, pathogenesis, clinical signs, diagnosis, and treatment of liver disease is included elsewhere in this text. A normally functioning liver is necessary to maintain normal brain neuron and astrocyte function. With acute liver disease hepatoencephalopathy is most often a result of astrocyte swelling, acute cytotoxic cerebral edema and intracranial hypertension. In chronic hepatic disease astrocytes are swollen but also show evidence of Alzheimer type II changes. Horses with hepatoencephalopathy show signs of cerebral cortex dysfunction. The earliest recognizable clinical signs in horses are often depression, lethargy, mild ataxia, and various forms of inappropriate behavior. These signs can progress to head pressing, circling, somnolence, and diminished responsiveness to external stimuli. Eventually a ected horses become recumbent and comatose. Seizures may occasionally be observed but they are not common. The type and severity of clinical signs do not correlate with type or reversibility of the underlying hepatic disease.

Hypoglycemia, Hypoxemia, and Ischemia Severe hypoglycemia is not common in adult horses but is common in neonates, often as a complication of sepsis, prematurity, hypoxic ischemic encephalopathy, or hypothermia. When hypoglycemia does occur it may result in weakness, depression, ataxia, and eventually loss of consciousness. Seizure activity is not common in adult horses but may occasionally be observed in a ected foals. Iatrogenic hypoglycemia can occur in horses treated with insulin or with abrupt cessation of intravenous glucose therapy. Clinical signs are rapidly reversed with intravenous administration of glucose-containing solutions. Guidelines for intravenous glucose therapy are included in the chapter on foal

diseases elsewhere in this text (Chapter 21). Because of the brain’s high and continuous demand for oxygen, the CNS is extremely susceptible to hypoxic or ischemic damage. This is most apparent in the neonate with hypoxic ischemic encephalopathy, discussed in detail in the chapter on foal diseases. As nervous tissue switches to anaerobic glycolysis to meet its energy needs brain glucose is depleted and localized lactic acidosis occurs. Neurons swell and cytotoxic edema exacerbates ischemia. The end result is irreversible cell damage and neuronal death. Clinical signs depend on what areas of the brain are a ected. Localized hypoxia may occur with thromboembolic disorders or intracarotid injections. Deprivation of oxygen to the entire CNS results in more dramatic clinical signs of impaired cerebral function ultimately leading to seizures, coma, and death. This type of generalized severe tissue hypoxia may result from decreased inspired oxygen, ventilation/perfusion abnormalities, right to left shunting of blood, impairment of gas di usion within the lungs, hypoventilation, severe anemia, impaired oxygen utilization by tissues, or profound hypotension secondary to sepsis, hemorrhage, anaphylaxis, or cardiac failure. These disorders are discussed in detail in the appropriate chapters elsewhere in this text.

Electrolyte Abnormalities Severe hyponatremia or hypernatremia can result in CNS dysfunction. Hyponatremia most often results from water intoxication because of excessive IV or oral administration or consumption of hypotonic solutions. A common clinical scenario resulting in hyponatremia is oral consumption of large quantities of water by a patient with a disorder that results in signi cant concurrent electrolyte loss (e.g., diarrhea, excessive sweating, polyuric renal failure). Clinical signs are most common in horses with a serum sodium concentration of 1 cm in diameter) or large vesicles. Occasionally, bullae fill with blood and are red-purple.

Viral Causes of Vesicular Eruptions EQUINE HERPES COITAL EXANTHEMA Coital exanthema is a rare contagious venereal disease of horses cause by equine herpesvirus type 3, which occurs worldwide. The disease is transmitted via coitus, insects, fomites, and inhalation. Early lesions begin as papules that rapidly progress to vesicles or bullae or pustules on the vulva or perineum of mares and the penis and prepuce of stallions.4 Vesicles and bullae also may be present in the mouth or nostril or on the lips. A ected areas commonly become eroded and ulcerative. Lesions are more often pruritic rather than painful. Depigmentation often occurs in areas where lesions have developed and healed. Stress may precipitate recurrences. The diagnosis is based on the appearance of lesions on the vulva, perineum, penis, or prepuce; skin biopsy; and virus isolation. Histologic ndings include hyperplastic super cial and deep perivascular dermatitis with ballooning degeneration and eosinophilic intranuclear inclusion bodies. Treatment is symptomatic, but corticosteroids are contraindicated. Owners should isolate a ected horses and remove broodmares and stallions from the breeding program for a minimum of 4 weeks. The disease has no known effect on fertility.

VESICULAR STOMATITIS (SORE NOSE AND SORE MOUTH) Vesicular stomatitis is caused by a rhabdovirus with two major serotypes: New Jersey and Indiana.21 The disease a ects horses, cattle, and swine and is enzootic in North, Central, and South America. In North America the disease is reportable. The disease has a seasonal occurrence in summer and fall and is believed to be transmitted to horses by insect bites. Within a group of horses, however, the disease may be transmitted via direct contact. The incubation period is short (24 to 72 hours), and lesions may last for only 3 to 4 days. Infected horses rapidly develop vesicles up to 2 cm in diameter in the mouth and on the lips. These lesions rupture, leaving large painful erosions and ulcers. A ected animals salivate profusely, and fever and anorexia are common. Rarely, lesions develop on the hooves, prepuce, and teats. Con rmation of the diagnosis is obtained by serologic examination. Histologic ndings are nonspeci c and include a hyperplastic epidermis, intercellular and intracellular edema of the epidermis, reticular degeneration, spongiotic microvesicles, and focal necrosis. Superficial and deep perivascular dermatitis is present. Owners should feed a ected horses a soft gruel until oral lesions heal, usually within a few days. Permanent depigmentation may occur in areas of former ulceration. Infection results in immunity for up to 6 months.

HORSEPOX Horsepox is uncommon and has been reported only in Europe. Horsepox is a benign disease, caused by an unclassified poxvirus. The virus gains access to the body by the respiratory tract or the skin.101 Systemic spread occurs by the lymphatics. Poxvirus inoculated into the skin may multiply locally, directly enter the blood, and create a primary viremia. The virus causes degenerative changes in the epithelium as a result of virus replication and results in development of vesicular lesions. Poxviruses also cause epithelial hyperplasia by stimulating host-cell DNA synthesis before the onset of cytoplasmic virus-related DNA replication. Ischemia and necrosis occur in the dermis and subcutaneous areas as a result of vascular damage. Clinical lesions of equine poxvirus infections develop as follows. An erythematous maculopapular eruption develops, followed by the development of vesicles. This stage is transient and may not be observed. Vesicles develop into umbilicated pustules with a depressed center and a raised erythematous border. When the pustule ruptures a crust develops and the lesion heals, often with scarring. Horsepox may a ect horses of any age. The disease is transmitted by direct contact with an infected host or with contaminated grooming tools or tack. Lesions of horsepox are typically vesicles and umbilicated pustules with crusts. Three clinical presentations (oral, leg, and vulvar) have been described.64 Oral horsepox (buccal horsepox, contagious pustular stomatitis) is characterized by the development of lesions on the inner lips and buccal mucosa. In severe forms, pox lesions may develop in the pharynx, larynx, nostrils, or all of these sites. Leg pox usually develops on the pasterns and fetlocks and often is confused with “scratches” or “greasy heel.” Pain and lameness are common. Vulvar pox (genital horsepox) is uncommon, occurring predominantly in severely a ected animals. One may observe pyrexia and anorexia early in the course of the three forms of horsepox.

The diagnosis is based on history and clinical signs and con rmed by histologic examination. Intracytoplasmic inclusion bodies demonstrated by routine histopathologic examination of tissue or electron microscopy are diagnostic. Other ndings include ballooning degeneration of the epidermis (stratum spinosum), reticular degeneration, acantholysis (loss of cohesion of cell in the granular area), intraepidermal microvesicles, a superficial and deep perivascular dermatitis, and intraepidermal microabscesses and pustules. Treatment is symptomatic. Most horses recover in 2 to 4 weeks. Occasionally the disease is fatal in severely a ected young horses. Recovery produces lifelong immunity. This disease affects human beings and cattle.

Autoimmune Diseases Causing Vesicular Lesions Bullous pemphigoid is a rare autoimmune skin disease of horses characterized by the development of vesicles and ulceration of the oral mucosa and occasionally the skin. Bullous pemphigoid is caused by production of an autoantibody (pemphigoid antibody) against a component in the basem*nt membrane of the skin and mucous membranes.56 The pemphigoid antibody binds to the basem*nt membrane zone and complement xation occurs, resulting in production of in ammatory mediators chemotactic for neutrophils and eosinophils. These cells release proteolytic enzymes that disrupt the dermal-epidermal cohesion; separation occurs, and a vesicle forms. In horses, vesicles and bullae develop at the mucocutaneous junction in the inguinal region or in the oral cavity.85 Skin lesions occur in the axilla and groin. Vesicles and bullae rupture easily and are often crusted and secondarily infected with bacteria. Pain, pruritus, anorexia, fever, and salivation are common. A de nitive diagnosis is made by eliminating other causes of vesicular lesions from the di erential list and by skin biopsy. Biopsy of a vesicle or intact bullae is essential; biopsy of crusted or ulcerative lesions will be nondiagnostic. Bullous pemphigoid is characterized by subepidermal vacuolar alterations and subepidermal clefts and vesicles.1 Neutrophilic and eosinophilic in ltration of the super cial epidermis is common. If direct immuno uorescence testing is performed and the results are positive, one observes a linear deposition of immunoglobulin at the basem*nt membrane zone. Direct immunofluorescence testing is not recommended. There are few reports of the treatment of bullous pemphigoid in horses.1,57,85 Treatments have included an initial large dose of prednisolone (1 mg/kg orally every 12 hours) or dexamethasone (0.2 mg/kg orally every 24 hours) to induce remission. After the arrest of new lesion development and healing, one institutes maintenance therapy with alternate-day administration of glucocorticoids at the smallest e ective dose. Therapy must be individualized and will most likely be lifelong. The use of adjuvant therapy (e.g., gold salts or azathioprine) has not been reported in horses with bullous pemphigoid but may be useful.

Irritant and Toxic Causes of Mucocutaneous Vesicles Irritant reactions, toxic chemicals, or allergic reaction to drugs may result in oral mucocutaneous vesicles. One should suspect these causes when lesions develop acutely in the absence of signs of systemic illness, in single animals within a herd, or in horses receiving medications or counterirritants. Mercurial compounds (commonly used in counterirritants) can also cause generalized hair loss, lameness, and emaciation in chronically intoxicated animals. Creosol and cantharidin beetles are common causes of oral ulcers. Foreign bodies such as grass awns may cause mucocutaneous ulceration and vesicles.

Vasculitis Oral ulcers may develop in horses with vasculitis (see the section on Vasculitis and Purpura Hemorrhagica).

NECROTIZING SKIN DISORDERS Necrotizing skin diseases develop from a wide range of causes. The unifying characteristic is that each of these diseases may cause local or widespread full-thickness death of the skin, sloughing of tissue, ulceration, and exudation. A de nitive diagnosis of the underlying cause may be difficult in some cases.

Decubital Ulcers (Pressure Sores, Setfasts, Saddle Galls, and Saddle Sores) Decubital ulcers or sores are caused by prolonged application of pressure to an area. Thin and emaciated animals are at greatest risk for development of decubital ulcers. Pressure sores may develop from poorly tting tack on the back, neck, or girth of the horse; from recumbency during surgery, even while on suitable bedding; and from casts, leg wraps, or elastic bandages that have become wet or are unevenly wrapped, too tight, or ill tting. The amount of pressure required to create a decubital ulcer is not great. Tack and saddles may cause lesions over bony prominences or other areas subjected to extended pressure. Capillary circulation is stopped or severely limited, resulting in tissue anoxia and retention of metabolic wastes with tissue damage and death. Most lesions begin as small areas of tissue necrosis and progress to areas of ulceration. Secondary infections are common.1

Clinically, early lesions begin with hair loss, swelling, and erythema. A red-purple area of tissue discoloration may be visible. Within a few days, oozing, necrosis, and ulceration become visible. In severe cases the skin loses its elasticity, subcutaneous tissues slough, and the skin may harden. Strands of living tissue may be attached to necrotic tissue. Lesions are often malodorous. Healed lesions tend to scar, and leukotrichia or leukoderma are common sequela.82 The diagnosis is based on history and clinical examination. Areas of skin that have lost sensation or have a parchmentlike feel should be suspect. Treatment of these lesions is di cult. Because of signi cant capillary and venous congestion, systemic antibiotics are often not effective.1,82 Wounds are managed best topically. One should clean the wounds daily with copious amounts of water to remove tissue debris and stimulate circulation and should use topical antibiotic creams or ointments (i.e., iodine and chlorhexidine) to prevent secondary infection and to prevent the wound from becoming desiccated. Re-epithelialization and wound healing greatly improve with the use of surgical dressings, but bandaging some lesions may be difficult. Surgical débridement and skin grafting may be indicated in some cases.

Gangrene Gangrene is a clinical term used to describe wet or dry tissue necrosis. Gangrene may result from external pressure, severe edema, burns, frostbite, snake bites, vasculitis, ergotism, fescue toxicosis, or bacterial or viral infections.1 The characteristic lesion results from occlusion of the venous or arterial blood supply. Dry gangrene occurs when the arterial blood supply to an area is occluded but the venous or lymphatic drainage is intact. Wet or moist gangrene is caused by occlusion or impairment of lymphatic and venous drainage plus putrefaction caused by a bacterial infection. Lesions of dry gangrene are dry, leathery, discolored, sunken, and cold to the touch. The skin may take a long time to slough. Lesions of wet gangrene are swollen, discolored, and malodorous. The diagnosis is most often made by clinical examination and con rmed by biopsy. One must identify and treat the primary cause of gangrene.

Thermal Injuries BURNS Excessive heat (burns) can occur from barn res, brush res, accidental spillage of hot solutions, lightning, electrocution, rope burns, counterirritants, or radiation. Barn fires cause most burns in horses. Large thermal injuries in horses are di cult to manage.21,102-104 The large surface area of the burn dramatically increases the potential for uid, electrolyte, and caloric losses. Burns on 50% or more of the body are usually fatal, but this depends on the depth of the burn.93,95 Massive wound contamination is almost impossible to prevent because of the impossibility of maintaining a sterile environment.21,101 Horses require long-term restraint to prevent continued trauma; wounds are often pruritic, and self-mutilation is common.103 Burned horses are frequently disfigured, preventing them from returning to full function. Burns are classi ed by the depth of the injury.21,102-104 First-degree burns involve only the most super cial layers of the epidermis. These burns are characterized by erythema, edema, desquamation of super cial layers of the skin, and pain. The germinal layer of the epidermis is spared, and these burns heal without complication.21 Second-degree burns involve the entire epidermis and may be super cial or deep. Super cial second-degree burns involve the stratum corneum, stratum granulosum, and a few cells of the basal layer. Deep seconddegree burns involve all layers of the epidermis. Clinically, these burns are characterized by erythema, edema at the epidermal-dermal junction, necrosis of the epidermis, accumulation of white blood cells at the basal layer of the burn, eschar (slough produced by a thermal burn) formation, and minimal pain. The only germinal cells spared are those within the ducts of sweat glands and hair follicles. Seconddegree burns heal well with good wound care.21 Third-degree burns are characterized by loss of epidermal and dermal components. Fluid loss occurs, along with signi cant cellular response at the margins and deeper tissue, eschar formation, lack of pain, shock, wound infection, and possible bacteremia and septicemia. Healing is by contraction and epithelization and occurs only from the wound margins. Infection and problems with wound healing frequently complicate these burns.21 Fourth-degree burns involve all the skin and underlying muscle, bone, and ligaments. Burns cause local and systemic e ects.21,27,103,104 Local tissue damage results from massive protein coagulation and cellular death. In the immediate area of the burn, arteries and venules constrict and capillary beds dilate. Capillary wall permeability increases in response to vasoactive amines released as a result of tissue damage and in ammation.104 These vascular responses result in uid, protein, and in ammatory cells accumulating in the wound. Vascular sludging, thrombosis, and dermal ischemia occur, resulting in further tissue damage. The tissue ischemia continues for 24 to 48 hours after the injury and is believed to be caused by the local release of thromboxane A2.21,104 Lipid layers in the skin are destroyed, and a fourfold increase in the loss of uid occurs. Fluid losses result in increased heat loss from

evaporation and an increased metabolic rate. The full extent and depth of the burn may not be evident for several days. Neutrophil function and chemotaxis greatly decrease, predisposing the wound to local infection, bacteremia, and septicemia. The following microorganisms frequently colonize burns: Pseudomonas aeruginosa, Staphylococcus, Escherichia coli, Klebsiella spp., nonhemolytic Streptococcus spp., Proteus spp., Clostridium spp., and Candida spp.21,101-103 Systemic e ects are life-threatening and include hypovolemia, uid and electrolyte losses, protein loss, pulmonary edema, anemia, increased basal metabolic rate, increased caloric needs, and depressed cell-mediated and humoral immune responses. Hypovolemia exacerbates a decrease in cardiac output after a circulating myocardial-depressant factor.27 The practitioner should treat first-degree burns and superficial second-degree burns immediately with ice or cold water to prevent further tissue necrosis.21,101 One should apply aloe vera cream or a water-soluble antibacterial cream to the wound to prevent infection. In addition, nonsteroidal analgesics should be administered to alleviate pain and to help reduce dermal ischemia. Aspirin at 10 to 20 mg/kg orally once or twice daily decreases thromboxane production, may halt further dermal ischemia, and may be the initial drug of choice.21 Deep second-degree burns are initially treated as described previously; however, these burns tend to form blisters that should be left intact as long as possible because the vesicular uid is a good medium for reepithelialization.21,102,103 One should trim ruptured blisters, clean the area with copious amounts of water, and cover the wound with an antibacterial dressing or xenograft or allow an eschar to form.21,102-104 The bandage should allow drainage and should be changed at least daily. Full-thickness burns (third and fourth degree) are treated by occlusive dressing (closed technique), eschar production (exposure technique), continuous wet dressings (semiopen technique), or excision and grafting technniques.104 The most practical therapy for large burns in horses is the semiopen method, leaving the eschar intact with continuous application of moist bandages and antibacterial agents. The moist dressings help prevent heat and moisture loss from the eschar, provide protection of the eschar, and help prevent bacterial invasion in the wound. Routine use of systemic antibiotics is not recommended in burn patients. Short-term systemic antibiotics may be useful in the initial 3 to 5 days after the burn to minimize bacterial colonization of burns and systemic sepsis.104 In the absence of sepsis, systemic antibiotics are contraindicated. Extensive use of antibiotics may cause altered microbial ora in the gut and in mucous membranes, which may predispose the patient to infections from antibiotic-resistant gram-negative bacteria or from fungi.105 Topical medications should be water-based, should be applied and removed easily, should not interfere with wound healing, and should be excreted or metabolized readily. Silver sulfadiazine and aloe vera are e ective. Silver sulfadiazine is e ective against gram-negative bacteria, causes no discomfort to the horse, penetrates eschar, and has a 24-hour duration of action. Aloe vera is reported to relieve pain, decrease in ammation, stimulate cell growth, and kill bacteria and fungi. Aloe vera is most useful in the early treatment of wounds. Several excellent references for further information on long-term medical and surgical management of horses with large thermal burns are available.21,102-104

FROSTBITE Frostbite occurs when tissue is exposed to extreme cold. Sick, debilitated, and neonatal animals are at increased risk. Cold temperatures inhibit cell metabolism and cause tissue dehydration, cell disruption by ice crystals, ischemia, and vascular damage. The most commonly affected areas are the glans penis, ear tips, coronary bands, and heels.1 The initial lesion of frostbite is paleness of the skin. Erythema, scaling, and hair loss follow. Pigment loss may occur. In severe cases, necrosis and dry gangrene occur. Mild cases of frostbite do not require treatment. More severe cases require rapid thawing in warm water (41° to 44° C).1 After rewarming, one should apply antibiotic ointments without steroids to the area. In severe cases with necrosis and sloughing, topical wet soaks and symptomatic therapy with antibiotics may be needed to prevent sepsis. One should not attempt surgical débridement until an obvious demarcation is present between viable and nonviable tissue. Previously frostbitten areas may be more susceptible to cold injury.

Chemical Toxicosis In addition to causing disorders of the hair coat, selenosis may result in necrosis and sloughing of the hoof (see the section Chemical and Plant Causes of Coat Abnormalities).

Stachybotryotoxicosis Stachybotryotoxicosis is a mycotoxicosis caused by the toxins of the fungus Stachybotrys atra.1 This fungus grows on hay and straw and produces toxins referred to as macrocytic trichothecenes. These toxins cause bone marrow suppression, profound neutropenia, thrombocytopenia, and necrotic-ulcerative lesions of the skin and mucous membranes. Commonly, lesions begin at the mucocutaneous junction as focal areas of necrosis and ulceration. Petechiae, ulcers, large areas of necrosis, catarrhal rhinitis, suppurative rhinopharyngitis, and laryngitis follow. Skin lesions occur as early as 24 hours after ingestion of the toxin. Systemic signs include lethargy, anorexia, weight loss, hyperactivity, proprioceptive deficits, colic, muscular stiffness, and second-degree atrioventricular block. Affected animal develops hemorrhagic diathesis, hemorrhagic enteritis, and septicemia and die.

The diagnosis is based on history, clinical signs, and nding the toxin in the feed. If one recognizes toxicosis early in the course of the disease, withdrawal of a ected feed has resulted in resolution of signs. Animals with extensive lesions and chronic exposure have a poor prognosis.

Bites and Stings of Venomous Insects, Spiders, and Reptiles Many species of snakes, spiders, and insects have venomous bites and stings. Except for snake bites, these animals lack su cient quantity of venom to cause more than transient pain or local in ammation. Venoms contain a variety of substances, including enzymes, peptides, polypeptides, amines, and glycosides, which act locally by causing tissue necrosis and vascular thrombosis and hemorrhage or systemically by causing widespread hemolysis and neurotoxicity.

SNAKE BITE Of the poisonous snakes in the United States, bites from rattlesnakes, water moccasins, and copperheads occur most commonly in horses.106 The venom from snakes is hemotoxic and proteolytic and produces extreme local swelling with signi cant tissue and red blood cell destruction. Most bites occur in the spring and summer. Horses are most commonly bitten on the nose, head, neck, and legs. Snake bites may or may not involve envenomation and may be dry or wet. Bites in which no venom is injected swell minimally and are only slightly painful (dry bite). When envenomation occurs (wet bite), rapid swelling, pain, and local hemorrhage develop, usually within 60 minutes of the bite. Fang marks are often di cult to nd because of tissue swelling. Edema, erythema, and tissue necrosis develop over days, and the skin may slough (Figure 13-35). Bites occurring on the face or head are serious because of the risk of respiratory and nasal edema. If respiratory distress is severe, the horse may need a tracheotomy.

FIGURE 13-35

Face of a horse after being bitten by an Eastern diamondback rattlesnake.

Snake bites should be treated symptomatically. One should clean the wound and provide hydrotherapy with cold water to minimize swelling. If swelling is already present, warm hydrotherapy may stimulate circulation and removal of tissue edema. Whether hydrotherapy increases the absorption of venom in the early stages of the bite is unknown, however. Broad-spectrum antibiotics are indicated to prevent secondary infection. The use of glucocorticoids is controversial but may be bene cial in decreasing in ammation and pain. Nonsteroidal antiin ammatory agents also may be bene cial. Surgical débridement and wound closure may be necessary if severe necrosis and sloughing occur. Although speci c antivenoms are available, they are of limited use. Maximal bene t is attained only when they are administered within hours of the bite. Additionally, the volume of antivenom necessary to treat snake bites in horses may be cost prohibitive.

FIRE ANT BITES Fire ants, Solenopsis spp., are common in the southern United States. Single or small numbers of stings are acutely painful and rapidly develop into a pustule or crust. Horses are bitten most commonly on the legs, nose, and ventrum, but massive exposure can occur if the horse rolls on an anthill and is bitten by hundreds or thousands of ants. Complications of massive exposure include infection, anaphylactic shock, bronchospasms, and sloughing of the epidermis. Single or multiple small numbers of solitary stings usually require little or no treatment. Applying an antibiotic ointment to minimize the chance of myiasis is prudent; otherwise, wounds heal without complication. If massive exposure has occurred, one may need to use systemic antibiotics and nonsteroidal anti-inflammatory drugs to decrease pain and swelling.

SPIDER BITES Black, brown, and red widow spiders (Latrodectus spp.) and the brown recluse (Loxosceles reclusa) spider are common in America. In

Australia the black house spider (Ixeuticus) causes painful bites.106 Spider bites are characterized by hot, edematous, painful swellings in the area of the bite. The brown recluse spider has a dermal necrotoxin that may cause severe dermal necrosis, but this has not been reported in horses. Treatment is symptomatic. Cold ice packs applied to the area, and systemic glucocorticoids or antihistamines, or both, may be helpful. Spraying the environment of the horse with a parasiticide is important. Owners should remove discarded furniture, newspapers, old cloths, and trash from the premises because these are favorite living areas for venomous spiders.

Vasculitis and Purpura Hemorrhagica Vasculitis is an in ammatory reaction occurring in the wall of the blood vessel (Figure 13-36). Vasculitides are classi ed by in ammatory cell type. Neutrophilic vasculitides are subdivided further into leukocytoclastic (neutrophil nuclei undergo karyorrhexis) or nonleukocytoclastic.

FIGURE 13-36

Horse with vasculitis resulting from strangles.

The immunologic mechanisms involved are typically type I and type II hypersensitivity reactions.21 Equine viral arteritis, equine in uenza, C. pseudotuberculosis, and Streptococcus spp. (especially S. zooepidemicus subsp. equi) infections may result in vasculitis.35 Vasculitis also may occur after Rhodococcus equi pneumonia, cholangiohepatitis, and some antibiotic treatments. The underlying cause is often not identified. Idiopathic vasculitis has no known breed, sex, or age predilection. The most common locations for lesions are the distal limbs, ears, lips, and periocular areas.1 Oral ulcers and bullae may be present. Skin lesions consist of purpura, edema, and erythema. Systemic signs may include pyrexia, depression, anorexia, weight loss, and lameness. Purpura hemorrhagica is an acute noncontagious disease of horses characterized by extensive edema and hemorrhage of the subcutaneous tissue. Hemorrhage in the mucosae and viscera are common. Most cases occur after strangles or equine in uenza.1 Clinical signs usually develop within 2 to 4 weeks of the respiratory infection. Urticaria followed by pitting edema of the distal limbs, head, and ventral abdomen is common. Severe edema of the head may compromise breathing.17 Tissue exudation and sloughing may occur. Pain and pruritus are rare. Affected horses usually are depressed, reluctant to move, and often anorectic. The diagnosis of vasculitis is made by skin biopsy. Obtaining skin biopsy specimens from lesions 8 to 24 hours of age is important because these lesions tend to have the most diagnostic changes. Lesions more than 24 hours old may be nondiagnostic because of intense secondary cellular in ltrates or necrosis. Skin biospy reveals neutrophilic, eosinophilic, lymphocytic, or mixed cellular in ltrates in the vessel wall. Fibrinoid degeneration and hemorrhage are common.64 Direct immuno uorescence testing is occasionally useful as an adjunctive diagnostic aid. Prognosis for horses with vasculitis is unpredictable.27 Therapy of vasculitis is discussed elsewhere in this text.

Panniculitis and Fat Necrosis

Panniculitis refers to in ammation of the subcutaneous fat. This condition is rare in horses and results from widespread death of lipocytes. Fat cells are vulnerable to trauma, ischemia, and neighboring in ammation. When lipocytes are damaged, lipid is released and undergoes hydrolysis into glycerol and fatty acids. Fatty acids are potent inflammatory agents that elicit further inflammatory reactions.21 Panniculitis may be precipitated by a wide range of causes, including trauma, infections, autoimmune disease, pancreatic disease, glucocorticoid therapy, vasculitis, vitamin E de ciency, and idiopathic causes. In the horse, few cases have been reported, and the causes of those were obscure. Vitamin E deficiency was suspected in a few horses.76 Clinically, horses with panniculitis show deep-seated nodules and plaques.21 Lesions may be single or multiple and vary in size. Nodules may be hard and well-de ned or soft and ill-de ned. Initially, lesions are not xed to the overlying skin, but as the disease progresses, nodules become cystic and rupture onto the skin surface. The ulcerating nodules drain a yellow to brown to bloody oily material. Pain varies. Healed lesions may leave depressed scars. Affected animals may be febrile, depressed, lethargic, and anorectic. A de nitive diagnosis is made by skin biopsy. Samples should be obtained for histopathologic evaluation by deep excisional biopsy using a scalpel blade because skin biopsy punches do not obtain su ciently deep samples to be diagnostic. One should request special stains (i.e., periodic acid–Schi , Gomori’s methenamine silver, Brown and Brenn) for causative agent. Skin biopsy reveals lobular to di use pyogranulomatous inflammation in the panniculitis. The practitioner should identify and treat underlying causes appropriately. Idiopathic panniculitis may respond well to prednisolone at 1 to 2 mg/kg orally once daily for 7 to 14 days or to a single treatment of dexamethasone at 20 to 30 mg intramuscularly.21,76 Clinical improvement usually occurs within 7 to 14 days. Relapses occur, and the horse may require lifelong therapy.

DERMATOSIS OF THE LOWER LIMB Skin diseases of the lower limbs of horses are common dermatologic problems. These diseases may be extensions of generalized dermatoses or unique clinical syndromes.

Infectious Diseases of the Lower Limb BACTERIAL PASTERN FOLLICULITIS Bacterial folliculitis of the pastern is caused by Staphylococcus aureus, Staphylococcus hyicus, or β-hemolytic streptococci.82,106 This disease is considered a primary pyoderma, but the mechanism by which organisms initiate the disease is unknown. Lesions are limited to the posterior aspect of the pastern and fetlock region and may a ect single or multiple limbs. The initial lesion consists of papules and pustules that eventually coalesce and produce large areas of ulceration and suppuration. The disease is not associated with systemic signs. The diagnosis is based on clinical signs, skin scrapings, cytologic examination of the exudate, Gram stain, and bacterial culture and sensitivity. Cytologic examination of pustule contents usually reveals large numbers of neutrophils engulfing bacteria. Sedation may be necessary for initial treatment because lesions are painful.106 One should clip hair from a ected areas and wash the area daily in an antimicrobial scrub or shampoo (e.g., chlorhexidine shampoo or scrub) and should apply an appropriate antibiotic ointment without corticosteroids twice daily. Systemic antibiotic therapy is not usually necessary.

ULCERATIVE LYMPHANGITIS Ulcerative lymphangitis is a bacterial infection of the cutaneous lymphatics of horses. Lesions are most common on the hindlegs, especially distal to the hock, and consist of hard to uctuant nodules that abscess, ulcerate, and drain pus. Individual nodules may heal, but new lesions develop. Cording of the regional lymphatics, edema, and fibrosis are common. Regional lymph nodes usually are not involved. A de nitive diagnosis is based on direct smear, Gram stain, and culture of a nodule. Skin biopsy reveals super cial and deep perivascular dermatitis that may be suppurative or pyogranulomatous. Special stains (Brown and Brenn, and Gram) may reveal the organism. If one treats the disease early, therapy may be e ective and may prevent permanent dis gurement and debilitation. The drug of choice pending culture and sensitivity is procaine penicillin 20,000 to 80,000 IU/kg intramuscularly every 12 hours for 30 days or longer. Adjunct hydrotherapy may be bene cial. After brosis develops, prognosis is poor.106 One should clean the a ected area daily with copious amounts of water and wash it with an antimicrobial scrub (e.g., chlorhexidine).

VIRAL PAPILLOMATOSIS Equine viral papillomatosis (warts) is caused by a DNA papovavirus.101 Viral papillomatosis is common in horses and occurs most frequently in horses less than 3 years of age. No sex or breed predilection exists. Transmission appears to be by direct contact. Groups of young horses are often a ected more frequently than horses of the same age housed individually. Lesions are most common on the muzzle, genitalia, and

distal legs and are usually multiple and resemble papillomatosis of other species. A definitive diagnosis is made by skin biopsy. Epithelial proliferation without connective tissue proliferation is typical. Benign neglect is the best treatment for these lesions; spontaneous remission almost always occurs. One should consider surgical removal only when the lesion interferes with function. Anecdotal reports that surgical removal or damage to part of the lesion induces remission of the lesion are unproven.83 In fact, controlled studies suggest that such intervention might increase the duration of the lesions.83,106 E cacy of autogenous vaccines is unproven.

SPOROTRICHOSIS The dimorphic fungus S. schenckii causes sporotrichosis, which was discussed in detail under Infectious Causes of Nodules. A cutaneous lymphatic form of sporotrichosis may localize to the distal extremities of horses. The lymphatics become corded, and large nodules ulcerate and drain a thick brown-red discharge. Regional lymph nodes are not involved. Edema is rare. Diagnosis and treatment have been discussed previously.

Parasitic Diseases of the Lower Limb The common parasitic diseases of horses have been discussed previously (see Parasitic Skin Diseases). Infestations of lice, Chorioptes spp., and chiggers are the most common parasitic dermatoses that cause lower limb pruritus.

Neoplastic and Nonneoplastic Proliferation Diseases of the Lower Limbs KELOIDS A keloid is a nonneoplastic broblastic response that occurs on the pastern of horses. Clinically the keloid resembles a cluster of grapes. A de nitive diagnosis is determined by biopsy. These lesions do not respond to surgical excision, and in fact surgery may exacerbate them.21,106 The best treatment for keloids is by intralesional injection of triamcinolone acetonide (not to exceed 20 mg per horse). Unfortunately, most lesions are so massive that response to therapy is poor.

HEMANGIOMATA Hemangiomata are benign tumors of the endothelial cells of blood vessels. Lesions occur most commonly in horses less than 1 year of age, and some horses are born with them.83 Hemangiomata tend to be solitary tumors occurring on the distal limbs. The clinical appearance varies and may be circ*mscribed, nodular, rm to uctuant, blue to black in color, dermal or subcutaneous, hyperkeratotic, or verrucous. Ulceration and bleeding are common. Con rmation of the diagnosis is made by skin biopsy, which reveals proliferation of blood- lled vascular spaces lined by single layers of well-di erentiated endothelial cells.37,83 Equine hemangiomata are characterized by a multinodular capillary hemangioma with hyperplasia and hyperkeratosis of the overlying epidermis. Surgical excision is the treatment of choice. Equine verrucous hemangioma is di cult to excise, and cryosurgery may be bene cial. Recurrence is common.

Autoimmune Diseases of the Lower Limbs PEMPHIGUS FOLIACEUS Pemphigus foliaceus was discussed under Immunologic Causes of Exfoliation. In some horses the condition a ects only the coronary band.1 The coronary band of all four limbs is crumbly, degenerating, and exudative. Pain, lameness, and edema of the lower limb may occur.

SYSTEMIC LUPUS ERYTHEMATOSUS Systemic lupus erythematosus is a multisystemic autoimmune disease that is rare in horses. Profound lymphedema of the lower limbs may be the only clinical sign. The disease is treated with immunosuppressive doses of prednisolone (2 mg/kg orally every 12 hours). The prognosis is grave.

VASCULITIS Vasculitis was discussed earlier under Vasculitis and Purpura Hemorrhagica. In some horses, cutaneous vasculitis is restricted to the white skin of the pasterns and face of horses in the summer.1 Characteric lesions are erythema, swelling, pain, and exudation, and lesions may resemble photodermatitis, which suggests a photoinduced cause. Edema of the pastern and face is significant.

LEUKOCYTOCLASTIC VASCULITIS

Leukocytoclastic vasculitis or photoaggravated vasculitis is a relatively common but poorly understood disease.80 Immune deposits (immunoglobulin G or complement component 3 or both) around the vessel walls have been detected in early skin lesions using direct immuno uorescence. The fact that unpigmented skin appears to be a ected suggests a role of ultraviolet light in the development of this dermatologic problem. Drug reactions may play a role and a recent report suggested that Staphlycoccus spp. bacterial infection as an underlying cause. This dermatologic condition generally a ects mature horses. The skin lesions are con ned to the lower limb (Figure 1337). A lack of pigment in the a ected area is the most common nding, but pigmented skin has also been reportedly diagnosed with this condition. Lesions are usually multiple and well marked. Initially, erythema and exudation occur, which progresses to crusting and open sores or ulcerated areas. The a ected area may ooze, and swelling is present. Chronic cases may develop a rough or “warty“ surface. This condition is more painful than pruritic. The diagnosis for this condition is made by skin biopsy. On histopathologic examination, leukocytoclastic skin lesions will have in ammation around the blood vessels with blood vessel degeneration and clost involving the small vessels in the super cial dermis. Treatment consists of systemic corticosteroids at relatively high doses for 1 week and reduced doses for another 4 to 6 weeks. A reduction in ultraviolet light exposure by bandaging the a ected leg or stabling the horse indoors during daylight hours or both is thought to be beneficial. Application of sun block (minimal sun protection factor of 20) is thought to be beneficial.

FIGURE 13-37

Horse with leukocytoclastic vasculitis.

PASTERN DERMATITIS (GREASY HEEL AND SCRATCHES) Skin disease of the lower limbs is a common and frustrating problem in horses. Pastern dermatitis, scratches, greasy heel, and mud fever are colloquial terms for a moist exudative dermatitis of the caudal heel and pastern area. The reader should remember that these terms are nonspecific and describe a variety of inflammatory skin conditions of the lower limbs of horses. The pathogenesis of this syndrome is not understood completely. The initial lesion may be a primary infection or may follow a predisposing factor such as PF or mites. Horses with long fetlock hair or horses housed in muddy paddocks, unsanitary conditions, or rough stubbly pasture are at risk for pastern dermatitis. The problem is common in horses worked on tracks consisting of grit particles, which can cause microtrauma to the skin. Clinical signs depend greatly on whether the condition is acute or chronic and whether the owner has treated the lesions. Many horse owners do not recognize the acute development of skin diseases, minimizing the opportunity to identify the underlying cause. To complicate matters further, many owners treat these lesions before seeking veterinary care. Many common over-the-counter medications can induce irritant or allergic reactions, making distinguishing between these reactions and the original skin disease almost impossible. Regardless of the cause, clinical signs are similar (Figure 13-38).106 Acute lesions usually begin at the heel. Pain, swelling, moist exudation, and hair loss are common. As the disease process continues, lesions spread proximally and anteriorly. Matting of the hair occurs, and the horse may be noticeably lame. Crusting is common. If the underlying disease process involves a vasculitis, ulceration may be present. If the disease is left untreated, a foul odor develops. Because of the constant exion in the area, ssures often develop. In Draft horses,

vegetative granulomatous growths commonly result. Lesions may occur on one or multiple limbs or just on extremities with white markings.

FIGURE 13-38

Pastern dermatitis.

A de nitive diagnosis requires a complete medical history. Liver function tests are essential in any horse in which the lesions are limited to the unpigmented areas of the skin. One should perform Dermatophilus preparations, fungal cultures, and skin scrapings in all cases. In horses with long hairs on the fetlocks, the hair should be combed thoroughly with a ne-toothed metal comb, which is often the only successful method for finding lice. In difficult cases, tissue samples should be submitted for bacterial or fungal culture. Correct therapy requires identifying the underlying cause. The reader can nd speci c therapy for most of the diseases elsewhere in this chapter or in the references. In all cases, one should clip the long hairs of the fetlock area and thoroughly wash the a ected area with an antimicrobial scrub (e.g., 2% chlorhexidine). Iodine preparations should be avoided because they can be irritating. One should remove all crusts and exudation, daily if necessary, and should avoid exposing the horse to moistures and irritants. Horses with idiopathic greasy heel often respond well to cleaning of the area, improved hygiene of the stall, and systemic corticosteroid therapy (assuming bacteria are not the cause). Large doses of prednisolone (e.g., 1 mg/kg orally once daily) may be necessary to induce remission of clinical signs. One should not decrease the dose of glucocorticoids too rapidly or relapse may occur. If oral prednisolone does not induce remission, dexamethasone at 0.02 to 0.04 mg/kg orally once daily for 3 to 5 days may be effective.

SYSTEMIC DISEASES WITH CUTANEOUS MANIFESTATIONS Reviewing all the skin diseases that may have systemic clinical signs is beyond the scope of this chapter. The author has made an e ort to include the most common systemic findings in this chapter. Table 13-2 provides a brief summary. TABLE 13-2 Skin Diseases with Systemic Manifestations Disease ENVIRONMENTAL Gangrene

Cutaneous Signs Wet: moist swelling, discoloration, malodorous, tissue decomposition

Systemic Signs Depends on underlying cause; fever

Dry: dry, discolored, leathery skin Burns

Selenosis Arsenic poisoning Mercury poisoning Iodism Hepatogenous photosensitization

Ergotism Leucenosis Hairy vetch toxicosis BACTERIAL DISEASES

Superficial: erythema, edema, pain, vesicles Deep: necrosis, ulceration, anesthesia, scarring Painful coronary band, sloughing and necrosis of hoof, rough hair coat, progressive loss of long hairs of mane and tail Severe seborrhea, ulcer, nonhealing wounds, hypertrichosis Progressive alopecia Severe dry seborrhea with or without hair loss Erythema, edema, pruritus, pain, in white or light-skinned area; vesicles and bullae may progress to oozing, necrosis, and sloughing

Swelling of coronary band, necrosis of feet, sloughing of ears, tail, feet Hoof dystrophies, shedding of hair Cutaneous plaques and papules that ooze yellow pus; pruritus; hair loss

Shock, respiratory compromise

Lameness, weight loss Gastroenteritis, emaciation, variable appetite Gastroenteritis, lameness, emaciation Cough, variable appetite, joint pain, seromucoid nasal discharge, lacrimation Acute: hepatic encephalopathy, icterus, depression, decreased appetite Chronic: weight loss, depression neurologic signs Lameness of hind legs, fever, weight loss, poor appetite Laminitis, lameness Conjunctivitis, anorexia, pyrexia, weight loss

Strangles

Limb edema; edema of lips, eyelids; petechial hemorrhages of mucous membranes and sclera

Corynebacterium pseudotuberculosis abscesses

Single or multiple deep abscesses that develop slowly or rapidly; 50% occurring in the pectoral or ventral abdominal area; ventral midline edema Exudative crusted lesions on dorsal trunk, face, pastern, or coronets

Dermatophilosis Actinobacillosis Clostridial infections

Glanders (farcy)

Depression, fever, lethargy, poor appetite, weight loss, lymphadenopathy; lesions on legs may cause edema, pain, and lameness. Thick-walled abscess of soft tissue In newborn foals, disease is a highly fatal septicemic disease, and skin lesions are rare. High fever, anorexia, muscle tremors, acute death possible Malignant edema: swelling at site of infection, pitting edema, local within 24-48 hours erythema; skin may become hot to touch, painful, or slough; crepitus Blackleg: hot, painful, swelling that progresses to a cold painful swelling with edema and subcutaneous emphysema Subcutaneous nodules begin most commonly on medial aspect of hock; Respiratory infection that rapidly leads to death lesions rapidly ulcerate and drain a honey-colored material; lymphadenopathy and cording of lymphatics is common

Burkholderia mallei (not in United States) FUNGAL DISEASE Histoplasma farciminosa (epizootic Unilateral nodules on face, head, neck and occasionally trunk; nodules lymphangitis) initially are firm but rupture and exude light-green, blood-tinged exudate; large ulcer may form and lesions may spread bilaterally PARASITES Lice Pruritus, scaling, alopecia Black flies Painful papules, and wheals that may become vesicular, hemorrhagic, and necrotic; lesion may be localized to ears or intermandibular areas IMMUNE-MEDIATED DISEASES Atopy Pemphigus foliaceus

Chronic pruritic urticaria, excoriations, alopecia, lichenification Crusts, scales, oozing, annular eruption, matting of the coat; lesions may be limited to coronary band Bullous pemphigoid Vesicles and bullae in mouth, groin, and axilla; crusts, ulcers Systemic lupus erythematosus Lymphedema, panniculitis, alopecia, leukoderma; scaling of face, neck, and trunk Transfusion reactions and graft-versus- Exfoliative to ulcerative dermatitis, ulcerative stomatitis host disease (may occur in horse after unmatched blood transfusion) Erythema multiforme Symmetric maculopapular lesions, urticaria that results in annular arciform or polycyclic shapes; wheals that do not disappear Vasculitis Purpura, edema, erythema, necrosis, crusts; purpura hemorrhagica causes edema and hemorrhagic swelling in tissue, mucosa, and viscera Equine exfoliative eosinophilic Scaling and crusting that progress to generalized exfoliation, alopecia, dermatitis and stomatitis ulceration, and exudation; variable pruritus Equine cutaneous amyloidosis Papules, nodules, and plaques over the head and neck that develop rapidly ENDOCRINE DISEASES Hypothyroidism Dull rough hair coat, delayed shedding of coat, edema of face and limb Pituitary paors intermedia dysfunction SWEAT GLAND DISORDERS Anhidrosis MISCELLANEOUS DISEASES

History of acute contagious upper respiratory infection; pyrexia, mucopurulent nasal discharge; abscesses in the mandibular or retropharyngeal lymph nodes Pitting edema, depression, fever, lameness, internal abscess, prolonged fever, abortion

Long shaggy hair that fails to shed; mane and tail are unaffected

Acute episode: none Chronic: dry hair coat, excessive scaling, partial alopecia, pruritus

Lacrimation, conjunctivitis, and respiratory signs may occur

Anemia in severe infestations with sucking lice Toxin in bite can cause increased capillary permeability; depression, weakness, staggers, tachypnea, tachycardia, weak pulse, shock, and possible death Respiratory difficulty, especially on expiration Depression, weight loss, poor appetite, pyrexia Anorexia, depression, fever Polyarthritis, thrombocytopenia, proteinuria, fever, depression, weight loss Diarrhea, increased heart and respiratory rates, lacrimation, muscle fasciculation Occurs after pregnancy, drugs, neoplasia, connective tissue disease, and infections or may be idiopathic May occur after strangles or influenza; depression, fever, reluctance to move, colic, diarrhea Severe, progressive weight loss; no diarrhea, ravenous appetite Diffuse nodules in upper respiratory tract may cause severe dyspnea Anecdotal reports of laminitis, infertility, anhidrosis, anemia, myopathy; weight gain and decreased food intake; skeletal limb disorders have been reported in foals Polydipsia, polyuria, muscle wasting, weight loss, lethargy, swayback appearance, pendulous abdomen, blindness, chronic infections, neurologic disorders or signs Acute: labored breathing, fever, flared nostrils, lack of sweating, collapse, death Chronic: polydipsia, polyuria, poor appetite, loss of body condition

Panniculitis NEOPLASTIC DISEASES Hemangioma and hemangiosarcoma Lymphosarcoma

Firm to fluctuant nodules most commonly found on trunk in subcutaneous tissue that ruptures and drains an oily yellow-brown to bloody discharge

Anorexia, depression, lethargy, pyrexia

Two types: (1) well-circ*mscribed nodules that are blue-black in appearance; (2) dark hyperkeratotic and verrucous lesions that bleed easily Single or multiple dermal-to-subcutaneous nodules, especially on trunk

Anemia Internal organ involvement; usually fatal

REFERENCES 1. Scott D.W. Large animal dermatology. Philadelphia: WB Saunders; 1988. 2. Talukdar A.H., Calhoun M.L., Stinson A.W., et al. Microscopic anatomy of the skin of the horse. Am J Vet Res. 1972;33:2365. 3. Sisson S., Grossman J.D. Anatomy of domestic animals. Philadelphia: WB Saunders; 1975. 4. Fitzpatrick T.B., Eisen A.Z., Wolff K., et al. Dermatology in general medicine, ed 3. New York: McGraw-Hill; 1993. 5. Suter M.M., Crameri F.M., Olivry T., et al. Keratinocyte biology and pathology. Vet Dermatol. 1997;8(2):67-100. 6. Rook A.J., Walton G.S., editors. Comparative physiology and pathology of the skin. Blackwell: Oxford, 1965. 7. Perris E.E. Parasitic dermatoses that cause pruritus in horses. Vet Clin North Am Equine Pract. 1995;11(April):11-28. 8. Ettinger S.J., Feldman, EC. Textbook of veterinary internal medicine, ed 6. 2005. Saunders, St. Louis 9. Murray M.D. Influence of skin temperature on populations of Linognathus pedialis. Aust J Zool. 1960;8:357. 10. Bergvall K. Advances in acquisition, identification and treatment of equine parasites. Clinc Tech Equine Pract. 2005;4:296-301. 11. Hugnet C., Cadore J.L., Bourdoiseaq G. Intert du fiprinol a 0.25 pour cent anspray dans le traitement de la philtirose. Vet Equine. 1999;31:65-68. 12. Littlewood J.D. Control of ectoparasites in horses. Pract. 1999;21:418-424. 13. Martineau G.P. Pathophysiology of sarcoptic mange in swine, part 2. Compend Cont Educ Pract Vet. 1987;9:F93. 14. , ARCO. M.H. Hayes. 1968. Veterinary notes for horse owners, New York. 15. Kunkle G.A., Greiner E.C. Dermatitis in horses and man caused by straw itch mite. Am J Vet Med Assoc. 1982;181:467. 16. Dvorak H.F. Cutaneous basophil hypersensitivity. J Allergy Clin Immunol. 1976;58:229. 17. Pascoe R.R. The nature and treatment of skin conditions observed in horses in Queensland. Aust Vet J. 1979;49:35. 18. Rabalasis F.C., Votava C.L. Cutaneous distribution of Onchocerca cervicalis in horses. Am J Vet Res. 1974;35:1369. 19. Foil L., Foil C. Parasitic skin diseases. Vet Clin North Am Equine Pract. 1983;5:529. 20. Stannard A.A., Cello R.M. Onchocerca cervicalis infection in horses from the western United States. Am J Vet Res. 1975;36:1029. 21. Robinson N.E., Sprayberry K.A., editors. Current therapy in equine medicine, ed 6, St. Louis: Saunders, 2009. 22. Fadok V.A., Mullowney P.C. Dermatologic diseases of horses. 1. Parasitic dermatoses of the horse. Compend Cont Educ Pract Vet. 1983;5:S615. 23. Lavach J.P. Large animal opththalmology. St Louis: Mosby-Year Book; 1989. 24. Anderson R.R. The use of ivermectin in horses: research and clinical observations. Compend Cont Educ Pract Vet. 1984;6:S516. 25. Pusterla N., Watson J.L., Wilson W.D., et al. Cutaneous and ocular habronemiasis in horses: 63 cases. Am Vet Med Assoc. 1985-2002;222:978-982. 2003 26. Vasey J.R., et al. Equine cutaneous habronemiasis. Compend Contin Educ Pract Vet. 1981;3:S290-S295. 27. PT Colahan, IG Mayhew, AM Merritt, editors. Equine medicine and surgery. SGoleta, Calif: American Veterinary Publications, 1991. 28. Stannard A.A. Nodular diseases. Vet Dermatol. 2000;11:179-186. 29. Mathison P.T. Eosinophilic nodular dermatoses. Vet Clin North Am Equine Pract. 1995;11 (April):83-86. 30. Herd RP, Donham JC: E cacy of ivermectin against “summer sores” due to Draschia and Habronema infection in horses. Proceedings of the twenty-sixth annual meeting of the Association of Veterinary Parasitologists, July 19-20, 1981, St Louis. p 8. 31. Bowman D.D. Helminths. In Bowman D.D., editor: Georgis’ parasitology for veterinarians, ed 7, Philadelphia: WB Saunders, 1999. 32. Greiner E.C. Entomologic evaluation of insect hypersensitivity in horses. Vet Clin North Am Equine Pract. 1995;11 (April):29-46. 33. Rosenkrantz W, Gri n C: Treatment of equine urticaria and pruritus with hyposensitization and antihistamines. Proceedings of the annual meeting of the American Academy of Veterinary Dermatology and the American College of Veterinary Dermatology, New Orleans, 1986. 34. Friberg C.A., Logas D. Treatment of Culicoides hypersensitive horses with high-dose n-3 fatty acids: a double-blinded crossover study. Vet Dermatol. 1999;10:117-122. 3 5 . O’Neill W, McKee S, Clarke AF: Flaxseed as a Potential Treatment for Allergic Skin Disease in Horses, Nutraceutical Alliance Research Web site: http://www.nutraceuticalalliance.com/ research_flaxseed3.htm. 36. Ackermann L.J. Practical equine dermatology. Philadelphia: WB Saunders; 1988. 37. Fadok V.A. Overview of equine pruritus. Vet Clin North Am Equine Pract. 1995;11:1-10. 38. Rees C.A. Response to immunotherapy in six related horses with urticaria secondary to atopy. J Am Vet Med Assoc. 2001;218:753-755. 39. White S.D. Advances in equine atopic dermatitis:serologic and intradermal allergy testing. Clin Tech. Equine Pract. 2005;4:311-313. 40. Fadok V.A. Overview of equine papular and nodular dermatoses. Vet Clin North Am Equine Pract. 1995;11:61-74. 41. Wagner I.P., Rees C.A., Dunstan R.W., et al. Evaluation of systemic immunologic hypersensitivity after intradermal testing in horses with chronic laminitis. Am J Vet Res. 2003;3:279-283. 42. Lorch G., Hillier A., Kwochka K.W., et al. Comparison of immediate intradermal test reactivity with serum IgE quantitation by use of radioallergosorbent test and two ELISA in horses with and without atopy. J Am Vet Med Assoc. 2001;218:1314-1322. 43. Magnan A.O., Mely L.G., Camilla C.A., et al. Assessment of Th1/Th2 paradigm in whole blood in atopy and asthma: increased IFN-gamma producing CD8 (+) T cells in asthma. Am J Respir Crit Care Med. 2000;161:1790-1796. 44. Karl S., Ring J. Pro and contra of specific hyposensitization. Eur J Dermatol. 1999;9:325-331.

45. Kwochka KW, Willemse T, Tscharner CV, et al, editors. Advances in veterinary dermatology, ed 3, Boston: Butterworth Heinemann, 1998. 46. Plumb D.C., editor. Veterinary drug handbook, ed 3, Ames, Iowa: Iowa State University Press, 1999. 47. White P. Essential fatty acids: use in management of canine atopy. Compend Cont Educ Pract Vet. 1993;15:451. 48. Lloyd D.H., Sellers K.C. Dermatophilosis infection in domestic animals and man. New York: Academic Press; 1976. 49. Burd E.M., Juzych L.A., Rudrik J.T., et al. Pustular dermatitis cause by Dermatophilus congolensis. J Clin Microbiol. 2007;45:1650-1658. 50. Hermoso M.J., Arenas A., Rey J., et al. In vitro study of Dermatophilus congolensis antimicrobial inhibitory and bacteriocidal concentrations. Br Vet J. 1994;150:189-196. 51. Mackie R.M., editor. Current perspectives in immunodermatology. Edinburgh: Churchill Livingstone, 1984. 52. Stannard A.A. Alopecia in the horse: an overview. Vet Dermatol. 2000;11:191-203. 53. White S.D. Equine bacterial and fungal diseases. A diagnostic and therapeutic update. Clin Tech Equine Pract. 2005;4:302-310. 54. Buffington C.A.T. Nutrition and the skin.Proceedings of the eleventh annual KalKan Symposium. Columbus, Ohio: Ohio State University; 1987. p 11 55. Fadok V.A., Wild S. Suspect cutaneous iodism in a horse. J Am Vet Med Assoc. 1983;183:1104. 56. Halliwell R.E.W., Gorman N.T.L. Veterinary clinical immunology. Philadelphia: WB Saunders; 1989. 57. Scott D.W., Walton D.K., Slater M.R. Immune-mediated dermatoses in domestic animals: ten years after, part 2. Compend Cont Educ Pract Vet. 1987;9:S39. 58. Vandenabeele S.I.J., White S.D., Affolter V.K., et al. Pemphigus foliaceous in the horses: a retrospective study of 20 cases. Vet Dermatol. 2004;15:381-388. 59. Brenner S., Wolf R., Ruocoo V. Drug-induced pemphigus. Clin Dermatol. 1993;11:501-505. 60. Brenner S., Bialy-Golan A. Ruoccov: Drug-induced pemphigus. Clin Dermatol. 1998;16:393-397. 61. Peroni D.L., Stanely S., Kollias-Baker C., et al. Prednisone per os is likely to have limited efficacy in horses. Equine Vet J. 2002;34:283-287. 62. Manning T., Sweeny C. Immune-mediated equine skin disease. Compend Cont Educ Pract Vet. 1986;12:979. 63. Mullowney P.C. Dermatologic diseases of horses. 4. Environmental, congenital and neoplastic diseases. Compend Cont Educ Pract Vet. 1985;7:S22. 64. Jubb K.V., Kennedy J., editors. Pathology of domestic animals, ed 3, vol. 1. New York: Academic Press, 1985. 6 5 . Wilkie J.S.N., Yager J.A., Nation P.N. Chronic eosinphilic dermatitis: a manifestation of a multisystemic, eosinophilic, epitheliotropic disease in ve horses. Vet Pathol. 1985;22:297. 66. Lindberg R. Clinical and pathophysiological features of granulomatosis in the horse. Zentralbl Vet Med Assoc. 1985;32:536. 67. Hillyer M.H., Mair T.S. Multisystemic eosinophilic epitheliotrophic disease in a horse: attempted treatment with hydroxy urea and dexamethason. Vet Rec. 1992;130:392. 68. Kerdel F.A., Moschella S. Sarcoidosis: an updated review. J Am Acad Dermatol. 1984;11:1. 69. Mullowney P.C., Fadok V.A. Dermatological diseases of horses. Part 2. Bacterial and viral diseases. Compend Cont Educ Pract Vet. 1984;6:S16. 70. Miers K.C., Ley W.B. Corynebacterium pseudotuberculosis infection in the horse: study of 117 clinical cases and consideration of etiopathogenesis. J Am Vet Med Assoc. 1980;117:250. 71. Blackford J. Superficial and deep mycoses in horses. Vet Clin North Am Large Anim Pract. 1984;6:47. 72. Mullowney P.C., Fadok V.A. Dermatologic diseases of horses. 3. Fungal skin diseases. Compend Cont Educ Pract Vet. 1984;6:S324. 73. Hubert J.D., Grooters A.M. Treatment of equine phytiosis. Compen Contin Edu Pract Vet. 2002;24:812-815. 74. Miller R.I., Campbell R.S.F. The comparative pathology of equine cutaneous phycomycosis. Vet Pathol. 1984;21:325. 75. Chaffin M.K., Schumacher J., McMullan W.C. Cutaneous pythiosis in the horse. Vet Clin North Am Equine Pract. 1995;11:91-103. 76. Thomsett L.R. Noninfectious skin diseases of horses. Vet Clin North Am Large Anim Pract. 1984;6:57. 77. Sullins K.E., Lavach J.D., Roberts S.M., et al. Equine sarcoid. Equine Pract. 1986;8:21-27. 78. Lazary S., Gerber H., Glatt P.A. Equine leukocyte antigens in sarcoid affected horses. Equine Vet J. 1985;17:283. 79. Otten N., von Tscharner C., Lazary S., et al. DNA of bovine papillomavirsu type 1 and 2 in equine sarcoids: PCR detection and direct sequencing. Arch Virol. 1993;132:121. 80. Pascoe R.R.R., Knottenbelt D.C. Neoplastic conditions. In: Pascoe R.R., Knottenbelt D.C., editors. Manual of equine dermatology. London: WB Saunders, 1999. 81. Tuthill R.E., Clark W.H. Jr., Levene, A: Equine melanotic disease: a unique model for human dermal melanocytic disease (abstract). Lab Invest. 1982;46:85A. 82. Stannard A.A., Pulley L.T., Tumors of the skin and soft tissue. Moulton J.E., editor. Tumors in domestic animals, vol. 2. Berkeley: University of California Press, 1978. 83. Goetz T.E. Cimetidine for treatment of melanoma in three horses. J Am Vet Med Assoc. 1990;196:449. 84. Sheahan B.J., Atkins G.J., Russell R.J., O’Connor J.P. Histiolymphocytic lymphosarcoma in the subcutis of two horses. Vet Pathol. 1980;17:123. 85. Gallagher R.D., Ziola B., Chelack B.J. Immunotherapy of equine cutaneous lymphosarcoma using low dose cyclophosphamide and autologous tumor cells infected with vaccinia virus. Can Vet J. 1993;34:371. 86. Pascoe R.R., Summers R.M. Clinical survey of tumors and tumor-like lesions in horses in southeast Queensland. Equine Vet J. 1981;13:235. 87. Pascoe R.R. Equine dermatoses (no. 22). In Post Graduate Foundation in Veterinary Science. Sydney, Australia: University of Sydney; 1981. 88. Lerner D.J., McCracken M.D. Hyperelastosis cutis in 2 horses. J Equine Med Surg. 1978;2:350-352. 89. Hardy M.H., Fischer K.R.S., Vrablic O.E., et al. An inherited connective tissue disease in the horse. Lab Invest. 1988;59:253-262. 90. Bridges C.H., McMullan W.C. Dermatosparaxis in Quarter horses. Proceedings of the thirty- fth annual meeting of the American College of Veterinary Pathologists. Toronto: Canada; November 12-16, 1984. p 22 91. White S.D., A olter V.K., Bannasch D.L., et al. Hereditary equine regional dermal asthenia (‘hyperelastosis cutis’) in 50 horses: clinical, histological, immunohistological and ultrastructual findings. Vet Dermatol. 2004;15:207-217. 92. Brounts S.H., Rashimir-Raven M., Black S.S. Zonal dermal seperation: a distinctive histopathological lesion associated with hyperelastosis cutis in a Quarter horse. Vet Dermatol. 2001;12:219-223. 93. Jones R.J. Toxicity of Leucaena leucocephala. Aust Vet J. 1978;54:387. 94. Feld J.R., Wolf C. Cushing’s syndrome in a horse. Equine Vet J. 1988;20:301-304. 95. vander Kolk H. Diagnosis of equine hyperadrenocorticism. Equine Pract. 1995;17:24-27. 96. Kolk J.H., Van der Kalsbeek H.C., Wensing T., et al. Urinary concentration of corticoids in normal horses and horses with hyperadrenocorticism. Res Vet Sci. 1994;56:126-128. 97. Auer D.E., Wilson R.G., Groenendick S., et al. Glucose metabolism in a pony with a tumour of the pituitary gland pars intermedia. Aust Vet J. 1987;64:379-382. 98. Munoz M.C., Doresle F., Ferrer O., et al. Pergolide treatment for Cushing’s syndrome in a horse. Vet Rec. 1996;139:41-43. 99. Shaver J.R., Fretz P., Doige C.E., et al. Skeletal manifestations of suspected hypothyroidism in two foals. J Equine Med Surg. 1979;3:269.

100. Vivrette S.L. Skeletal disease of a hypothyroid foal. Cornell Vet. 1984;74:373. 101. Howard J.C., editor. Current veterinary therapy. Philadelphia: WB Saunders, 1981. food animal practice 102. Gillespie J.H., Timoney J.F. Hagen and Bruner’s infectious diseases of domestic animals. Ithaca, New York: Cornell University Press; 1981. 103. Geiser D.R., Walker R.D. Management of large animal thermal injuries. Compend Cont Educ Pract Vet. 1985;7:S69. 104. Fox S.M. Management of a large thermal lesion in a horse. Compend Cont Educ Pract Vet. 1988;10:88. 105. Asch M.J., Meserol P.M., Mason A.D.Jr., et al. Systemic and pulmonary hemodynamic changes accompanying thermal injuries. Ann Surg. 1973;178:218. 106. Swaim S.F., Lee A.H. Topical wound medications: a review. J Am Vet Med Assoc. 1988;190:588. ∗

The editors acknowledge and appreciate the contributions of Karen A. Moriello, Douglas J. DeBoer, and Susan D. Semrad, former authors of this chapter. Their work has been incorporated into this edition.

CHAPTER 14 Disorders of the Hematopoietic System Debra C. Sellon, L. Nicki Wise

The hematopoietic system includes the blood and blood-forming tissues of the body. Peripheral blood cells are essential for tissue oxygenation, immune surveillance and clearance of foreign antigens, coagulation, and in ammatory reactions. Because of the interactions of blood with other body tissues and organs, alterations in blood parameters often re ect dysfunctions elsewhere in the body, and evaluation of the blood and its constituents has become an essential component of many diagnostic efforts. This chapter discusses the basics of hematopoiesis and hematopoietic metabolism as they relate to speci c disease conditions in the horse. Included is a discussion of the erythron, leukon, platelets, and hemostatic mechanisms of the horse.

ERYTHRON Erythropoietic tissue of the bone marrow and the circulating erythrocytes may be referred to as the erythron, to emphasize their organlike function. Erythrocytes are essential for delivery of oxygen to all tissues of the body. The complex maturation and development of erythrocytes and their pivotal role in survival of all body tissues make them uniquely re ective of many pathologic conditions. An understanding of the erythron and its responses to perturbations in homeostatic mechanisms throughout the body can provide the astute clinician with invaluable clues in unraveling difficult diagnostic dilemmas.

Physiology Erythropoiesis refers to the process of development and maturation of erythrocytes. Knowledge of the metabolic processes of these cells is critical to understanding pathologic processes that interfere with their primary function of tissue oxygenation. In the fetus, the cellular elements of blood are produced almost exclusively in the liver and spleen. As the body matures and di erentiates in utero, hematopoiesis gradually shifts to the marrow cavities so that at birth the bone marrow is the main organ of hematopoiesis. As an animal ages, bone marrow hematopoiesis decreases and fat in ltrates much of the previously active marrow. In the older adult animal, active hematopoiesis is limited to the marrow of the vertebrae, ribs, sternum, skull, and pelvis, as well as to the epiphyseal marrow of the humerus and femur. Erythrocyte production begins from a pluripotent colony-forming unit stem cell capable of di erentiating into erythroid-, myeloid-, megakaryocytoid-, or lymphoid-producing cell lines. The stem cell is capable of self-renewal to provide a continuing supply of pluripotent stem cells. The direction of di erentiation is determined in part by the types and quantities of cytokine mediators to which the stem cell is exposed at the time it begins to divide. The exact combinations of mediators necessary to direct di erentiation into each line are not understood entirely. In erythrocyte development the stem cell undergoes sequential mitotic divisions to produce the committed erythrocyte progenitor burst-forming unit, followed by erythroid colony-forming units. These two stages are capable of limited replicative self-renewal but are committed ultimately to di erentiate into erythrocytes (Figure 14-1). Burst-promoting activity depends on the presence of various cytokines, including interleukin 3 (IL-3), IL-4, and granulocyte-macrophage colony-stimulating factor. The erythroid colony-forming unit cell is the immediate precursor stage of the first morphologically recognizable erythrocyte precursor, the proerythroblast.

FIGURE 14-1

Schematic representation of the progenitor basis of hematopoiesis. Maturation is depicted from left to right with circulating blood cells as the nal product on the far right of the drawing. Progressive ampli cation of progenitors and precursors as they mature and di erentiate is not shown. CFU-S, Colony-forming unit, spleen; CFU-G/M, CFU-granulocyte/macrophage; CFU-Eo, CFU-eosinophil; BFU-E, burst-forming unit, erythroid; CFU-mega, CFU-megakaryocyte; CFU-G, CFU-granulocyte; CFU-M, CFU-monocyte; CFUE, CFU-erythrocyte. (From Wyngaarden JB, Smith LH, editors: Cecil textbook of medicine, ed 18, Philadelphia, 1988, WB Saunders.)

Initial stages of erythrocyte development depend greatly on the presence of the glycoprotein hormone erythropoietin, which the kidney produces in response to renal hypoxia and is an absolute requirement for erythrocyte progenitor cell maturation and di erentiation. The absence of erythropoietin, as in patients with chronic renal disease, is associated with potentially severe nonregenerative anemia. Erythrocytosis, an increase in the circulating red blood cell (RBC) mass, is associated with excess secretion of erythropoietin in certain disease conditions. Bone marrow proerythrocytes undergo four successive divisions followed by a period of maturation to produce 16 erythrocytes. Sequential divisions produce basophilic erythroblasts, polychromatophilic erythroblasts I and II, and orthochromic erythroblasts. The orthochromic erythroblast becomes a reticulocyte by ejecting its nucleus. Reticulocyte production requires approximately 72 hours; 24 to 48 hours later the reticulocyte will have matured into an erythrocyte. Iron is an essential component for hemoglobin synthesis and thus for erythropoiesis. The diet of the horse normally contains an abundance of readily available iron. Absorption is most e cient in the duodenum but may occur at almost any part of the gastrointestinal tract. The amount of iron absorbed from the gastrointestinal tract varies with the systemic iron status of the animal. Absorption increases in irondepleted animals and decreases in animals with ample iron stores. Unabsorbed iron passes through the remainder of the tract and is lost in feces. Horses, especially young foals, supplemented with excessive dietary iron can develop hemochromatosis, hepatic cirrhosis, and liver failure.1-4 After absorption the blood transports iron complexed with the protein transferrin. In the liver, spleen, and bone marrow, iron is transferred to molecules of the storage protein ferritin. Approximately 30% of body iron is present in the storage proteins ferritin and hemosiderin (a complex aggregation of ferritin molecules). Only minute amounts of iron are complexed to the transport protein transferrin at any given time. Minor quantities of iron are also present in myoglobin and various electron transport molecules and enzymes. The hemoglobin of senescent erythrocytes removed from the peripheral circulation is degraded in cells of the mononuclear phagocyte system in the spleen, liver, and bone marrow. Iron ultimately is recycled to the bone marrow or liver to be used in synthesis of hemoglobin for new erythrocytes. Almost 70% of total body iron is present in hemoglobin. Hemoglobin is synthesized in the mitochondria of the developing erythroblast and consists of four polypeptide chains (two α and two β in the adult) and a heme moiety. Heme is the functional portion of the hemoglobin molecule and contains an iron atom held in place by the four pyrrole rings of a porphyrin molecule. The iron in hemoglobin must be in the ferrous form (Fe2+) to bind oxygen reversibly. Heme iron oxidized to the ferric form (Fe3+) forms methemoglobin, and the molecule is no longer capable of transporting oxygen. The normal life span of equine erythrocytes in circulation is approximately 150 days.5,6 Mononuclear phagocytes of the spleen, liver, and bone marrow remove senescent RBCs from circulation. Transferrin transports released iron to RBC precursors in the marrow. Heme in

phagocytic cells undergoes enzymatic degradation to biliverdin and then bilirubin. Phagocytic cells release unconjugated bilirubin into the circulation, where it binds to plasma albumin. Hepatocytes take up unconjugated bilirubin, which is conjugated and excreted via the bile into the gastrointestinal tract, where it is converted to urobilinogen and then stercobilin. Intravascular or extravascular hemolysis often results in increased serum unconjugated bilirubin concentrations. During intravascular hemolysis, ruptured erythrocytes release hemoglobin, which combines with haptoglobin, an acute phase protein normally present in plasma, to be carried back to the liver, where hemoglobin is converted to bilirubin and excreted. If the level of hemoglobin exceeds the carrier capacity of hemoglobin, then the result is free hemoglobin in the plasma. Free plasma hemoglobin is ltered readily across the renal glomerulus and reabsorbed by renal tubular epithelial cells. Hemoglobin and other related pigment molecules, including myoglobin, are potentially nephrotoxic, and excessive intravascular hemolysis may result in acute renal failure. The mature erythrocyte is an anucleate cell incapable of de novo protein synthesis to replace enzymes or other proteins that have been used during normal metabolism. Binding, transport, and delivery of oxygen to tissues does not require energy expenditure by the erythrocyte, but maintaining iron and various cellular proteins in a reduced state essential for proper function does require energy. Maintenance of appropriate electrolyte gradients across the RBC membrane also requires energy. Glucose is the main in vivo energy source for erythrocytes. RBCs di er from other cell types in the absence of the Krebs (tricarboxylic acid) cycle. Glucose must be metabolized via the anaerobic glycolytic (Embden-Meyerhof) pathway or via the hexose monophosphate shunt. The enzymes involved in these metabolic pathways are critical to RBC survival and rare hereditary de ciencies have been described in the horse. The de ciencies of either glucose-6-phospahte dehydrogenase (G6PD) or avin adenine dinucleotide (FAD) resulted in hemolytic anemia and persistent methemoglobinemia, respectively.7,8 The body must maintain iron, hemoglobin, some critical RBC enzymes, and several membrane proteins in a reduced form for appropriate activity. High intracellular concentrations of glutathione (GSH), a sulfhydryl-containing tripeptide, are important for reduction of sulfhydryl groups of hemoglobin and other erythrocyte proteins. Reduction reactions involve the conversion of GSH to its oxidized form (GSSG), which then is returned to the reduced form (GSH) by the action of GSH reductase. When the functional ferrous form of iron (Fe2+) in hemoglobin is oxidized to the ferric form (Fe3+), the resulting methemoglobin is incapable of carrying oxygen. Small amounts of methemoglobin (1% to 2% of total hemoglobin) are produced normally in erythrocytes. This methemoglobin is reduced to functional hemoglobin primarily via the action of the enzyme methemoglobin reductase.

Evaluation of the Erythron Laboratory evaluation of the blood and bone marrow provides diagnostic clues that may aid the practitioner in diagnosing and treating a variety of disorders. Knowing the tests that are available and developing an understanding of how and why they are performed aids one in interpreting the results.

PERIPHERAL BLOOD EVALUATION Most clinical laboratories now use automated cell counters for evaluation of peripheral blood. These counters are capable of determining RBC counts, hemoglobin, packed cell volume (PCV), and various erythrocyte indices. Tables 14-1 and 14-2 list normal values for the horse.9,10 These values are intended as guidelines only. Each laboratory should establish its own normal values for each species. Table 14-1 Reference Ranges for Equine Hematologic Parameters∗

Table 14-2 Influence of Breed on Normal Erythron Values (Mean ± Standard Deviation) in Adult Horses

Red blood cell count, hemoglobin, and PCV are the most frequently used parameters for assessing the quantity of erythrocytes in circulation. In the normal horse, hemoglobin concentration is approximately one third of the PCV. This value increases in horses with intravascular hemolysis and hemoglobinemia. RBC count, hemoglobin concentration, and PCV decrease rapidly during the rst weeks of life in the equine neonate. This decrease has been attributed to decreased erythrocyte production, a shorter RBC life span in the neonate, and hemodilution. During this period, mean RBC volume also decreases to the point that cells would be classi ed as microcytic if compared with normal adult values.11 Evaluation of RBC indices can be helpful in characterizing anemia. If these parameters are not included in automated blood analyses, then they may be calculated from the PCV, RBC count (in millions), and hemoglobin (in g/dl) as presented in Box 14-1. Some automated cell counters also report RBC distribution width. Box 14-1 EVALUATION OF RED BLOOD CELL INDICES: CALCULATION FROM THE PACKED CELL VOLUME, RED BLOOD CELL COUNT (MILLIONS), AND HEMOGLOBIN (G/DL)

Expressed as femtoliters (fl) Increased in some horses with regenerative anemia Decreased with iron deficiency anemia Increased in older horses∗ ∗ Data From McFarlane D, Sellon DC, Ga ney D, et al: Hematologic and serum biochemical variables and plasma corticotropin concentration in healthy aged horses, Am J Vet Res 59:1247, 1998. RBC, Red blood cell; PCV, packed cell volume; MCV, mean corpuscular volume; MCHC, mean corpuscular hemoglobin concentration.

Expressed as picograms (pg) Increased with intravascular hemolysis Decreased with iron deficiency anemia

Expressed as grams per deciliter (g/dl) Increased with intravascular hemolysis Decreased with iron deficiency anemia

Peripheral blood evaluation is not complete without a thorough examination of a stained blood smear to assess erythrocyte morphology. The normal equine erythrocyte is a biconcave disc; however, in contrast to other species, most equine RBCs lack distinct central pallor. Equine RBCs normally exhibit a strong tendency toward rouleau formation, a coinlike stacking or grouping of erythrocytes. This tendency causes equine erythrocytes to sediment rapidly after collection, and one should mix all samples thoroughly, immediately before any evaluation. Excellent pictures of normal and abnormal equine erythrocyte morphologic condition are available in a variety of texts, and the reader is referred to those to aid in recognition of the following cell types12,13: Poikilocyte: Any abnormally shaped erythrocyte. Poikilocytosis usually is reserved for description of RBCs that exhibit a variety of morphologies. If one particular shape predominates, then one should use a more specific description. Anisocytosis: A variability in the size of erythrocytes, usually associated with an increased RBC distribution width. Polychromasia: A variability in the color of erythrocytes usually caused by a variable hemoglobin and RNA content. Spherocyte: Spherical erythrocyte observed in some horses with immune-mediated hemolysis. Echinocyte: Burr cell with short, regularly spaced spicules projecting from the RBC surface. Echinocytes may be associated with uremia. Acanthocyte: Spur cell with irregularly shaped spicules extending from the RBC surface. Acanthocytes may be associated with liver disease or gastrointestinal malabsorption. Elliptocyte: Ellipsoid or oval erythrocyte found in animals with iron deficiency or myelophthisic anemia. Leptocyte: Thin, flat RBC frequently associated with hepatic disease or iron deficiency. Codocyte: Target cell with a dense central area of hemoglobin surrounded by a pale zone. Codocytes may be associated with hypochromic anemias or hepatic disease. Howell-Jolly bodies: Basophilic nuclear remnants seen in the cytoplasm of erythrocytes. Approximately 10 in 10,000 erythrocytes contain Howell-Jolly bodies in the normal horse.13 Heinz bodies: Oxidized precipitated hemoglobin indicating oxidative damage to the RBC, usually resulting in intravascular or extravascular hemolysis. One may see Heinz bodies best using new methylene blue stain but may observe them in smears using Wright’s stain as round structures protruding from the edge of the RBC membrane. Occasionally, specialized tests of erythrocyte function or stability are indicated. Osmotic fragility tests measure the resistance of erythrocytes to in vitro hemolysis when incubated in increasingly hypotonic sodium chloride (NaCl) solutions. Erythrocytes from horses with immunemediated hemolytic anemia (IMHA) frequently are more fragile than RBCs from normal horses.14,15 The direct antiglobulin or Coombs’ test is also helpful for diagnosing IMHA. Coombs’ test detects immunoglobulin (Ig) or complement on the surface of circulating erythrocytes. The Coombs’ reagent should contain antisera to immunoglobulin G (IgG), immunoglobulin M (IgM), and the third component of complement (C3). One incubates washed erythrocytes from the patient with Coombs’ reagent and observes for agglutination. The cells can be incubated at 10° C to detect cold reactive antibodies or at 30° C to detect warm reactive antibodies.13 Because agglutination is the end point of the direct Coombs’ test, autoagglutination of blood is considered diagnostic of immune mediated hemolysis. One must di erentiate autoagglutination from rouleau formation by dilution of an RBC suspension with isotonic saline. Saline disperses rouleau formation but does not a ect true autoagglutination. The indirect Coombs’ test detects the presence of circulating anti-RBC antibody in the serum by incubating patient serum with normal equine erythrocytes and monitoring for agglutination.

BONE MARROW EVALUATION Examination of the bone marrow of a horse is indicated if one identi es or suspects a disorder of the hematopoietic system but cannot diagnose it from information gathered by history, physical examination, and routine laboratory tests.16 Bone marrow aspirates or core biopsies may be useful in characterizing anemias, evaluating iron stores, or explaining quantitative or qualitative abnormalities of blood cells. One may collect equine bone marrow aspirates or biopsies from the sternum,16-18 tuber coxae,16,19 or proximal ribs.16,20 The sternal aspirates are obtained from the ventral midline between the front legs. This site is preferable in most horses because hematopoietic activity persists throughout life, bones are not covered by a large muscle mass, and the marrow cavity is covered by only a thin layer of bone.16 One

may attempt an aspirate of the tuber coxae in young horses by directing the needle toward the opposite coxofemoral joint. The costal marrow is collected from the proximal portion of an easily palpable cranial rib, usually beneath the latissimus dorsi and serratus posticus muscles.20 Any large-gauge (16 gauge or larger) bone marrow needle with stylet can be used to collect marrow aspirates. The needle should be at least 2 inches long. The veterinarian clips the appropriate area and surgically scrubs it, injects a small amount of local anesthetic into the subcutaneous and periosteal tissues, and makes a stab incision through the skin and subcutaneous fascia. Forcing the bone marrow needle through the bone and into the marrow cavity may require considerable pressure and rotational movement. The veterinarian removes the stylet from the needle; uses a sterile syringe containing a small quantity of anticoagulant to aspirate redcolored marrow, discontinuing aspiration when blood is visible in the hub of the syringe; withdraws the needle; and smears marrow from the hub of the syringe and the needle on glass microscope slides as for blood smears. Alternatively, one injects the marrow sample into a Petri dish. On visualization, the clinician transfers spicules to a glass slide for staining. One may obtain core marrow biopsies using a 10-gauge Jamshidi needle and fixing the sample for routine histopathologic analysis.16,21 Marrow samples should rst be examined at low magni cation for evidence of cellularity, distribution of fat cells, and heterogeneity of progenitor cell populations. At higher magni cation, one may evaluate individual progenitor series and estimate a bone marrow myeloid-toerythroid ratio (M:E) by counting 500 to 800 cells.12 Descriptions and pictures of bone marrow progenitor cells are reported elsewhere, and the reader is referred to these for identi cation of cell types present in bone marrow aspirates or biopsies.12,13,19 Normal M:E in the horse has been reported to range between 0.5 and 3.76.17,18,20 Ratios of less than 0.5 are considered indicative of erythrocyte regeneration or myeloid suppression.22 Special histochemical stains for iron (Prussian blue stain) are available to evaluate peripheral iron stores in the horse.23 In many horses with of hemolytic anemia, bone marrow macrophages may be visible with phagocytosed RBCs. Myelophthisis is a reduction in the cellular elements in the bone marrow, frequently a result of myelo brosis, proliferation of brous tissue in the marrow. Myelodysplasia implies accumulation of abnormal cells in the bone marrow. These cells are often not dysplastic in the truest sense of the word but rather result from myeloproliferative neoplasia.

IRON STATUS EVALUATION Adult horses do not often develop abnormalities in iron status because iron is readily available in most forages and horses have a natural ability to store iron. However, evaluation of systemic iron status may aid in characterizing hematologic abnormalities, especially in the investigation of anemias. Several laboratory parameters can aid in the identi cation and classi cation of abnormalities of iron metabolism. The peripheral blood smear may provide initial clues. Microcytic, hypochromic erythrocytes often are seen in animals with disturbances of iron metabolism. One may con rm these observations by calculating the mean corpuscular volume (MCV) and mean corpuscular hemoglobin concentration (MCHC) of peripheral blood erythrocytes. Measurement of serum ferritin provides a reliable index of stored hepatic and splenic iron in the horse. Serum ferritin concentration in 28 normal horses was 152 ± 54.6 ng/ml. A bone marrow aspirate examined after special iron stains is also an excellent means of assessing the pool of storage iron. Serum iron concentration re ects the total quantity of transport iron in the plasma (i.e., the quantity of iron bound to transferrin). Normal equine values of 120 ± 5.0 g/dl24 and 108 g/dl25 have been reported. The total iron-binding capacity (TIBC) re ects the amount of iron that plasma transferrin could bind if fully saturated. Normal TIBC in the horse has been reported as 388 ± 8.1 g/dl.24 One calculates the percentage of transferrin saturation from the serum iron concentration and the TIBC. Percentage transferrin saturation is normally approximately 30%. Serum iron concentration increases dramatically for 48 to 72 hours after administration of corticosteroids, but TIBC and serum ferritin concentrations do not.26 In healthy foals at birth, serum iron and ferritin concentrations are lower than those of adult horses but very rapidly increase over the rst 24 hours of life to levels greater than that of adults because of absorption of colostral iron. This is accompanied by a very high-percent saturation of transferrin with iron.11,27 The high transferrin saturation makes foals quite susceptible to iron toxicity in the rst 24 to 72 hours of life. A small amount of orally administered iron can be absorbed and exceed the iron-binding capacity of serum with free or unbound iron reaching tissues, especially the liver, resulting in acute hepatic failure.28 Iron, ferritin, and transferrin saturation values gradually decrease to less than normal adult concentrations, reaching a minimum concentration at 3 weeks and returning to normal adult values by 6 months of age. The PCV of neonatal foals correlates closely with these changes: PCV is relatively high in the rst few days of life followed by a decrease in PCV to less than adult values by 2 weeks of age. Iron supplementation does not change these hematologic parameters.

Disorders of the Erythron ERYTHROCYTOSIS Erythrocytosis or polycythemia is a real or apparent increase in the circulating RBC mass and may be classi ed as relative, caused by a decrease in plasma volume, or absolute, caused by a real increase in RBC numbers. Animals with persistent erythrocytosis have muddy red to blue mucous membranes, prolonged capillary re ll time, weakness, lethargy, and exercise intolerance. Despite the large RBC mass in the circulation, oxygen delivery to the peripheral tissues often decreases because of increased blood viscosity and sludging in small vessels. Persistent erythrocytosis may lead to complications including hypertension, tissue hypoxia, thrombosis, and hemorrhage.29

Relative Erythrocytosis Relative erythrocytosis or polycythemia may occur because of hemoconcentration or splenic contraction. Hemoconcentration occurs when total plasma volume decreases without any change in total RBC numbers in circulation. Clinical dehydration is usually evident, with slow capillary re ll time, dry mucous membranes, and prolonged skin tenting. Dehydration may occur with excessive water losses accompanying diuresis, diarrhea, or excessive sweating, as well as with decreased water intake. Endotoxic shock results in dehydration and hemoconcentration by causing a shift of water from plasma to the interstitial space. An increase in total plasma protein usually accompanies relative erythrocytosis caused by hemoconcentration. An exception to this occurs with concurrent loss of protein and body water (e.g., in horses with severe protein-losing diarrhea or glomerulonephritis). In these horses, total plasma protein may be normal or decreased in the presence of a high PCV. The PCV of hemoconcentrated horses usually returns to normal after intravenous uid therapy, unless water losses continue. The spleen of the resting horse may harbor up to one third of the total circulating erythrocyte volume. Exercise, endogenous epinephrine release caused by excitement or stress, and exogenous epinephrine administration results in splenic contraction that may increase the PCV by as much as 50%.30-33 This relative erythrocytosis is not accompanied by a signi cant increase in total plasma protein. If the horse is allowed to relax in a nonstressful environment, then PCV usually returns to normal within a few hours. Absolute Erythrocytosis Absolute erythrocytosis occurs when increased erythropoiesis causes an increase in PCV, RBC count, and hemoglobin concentration. Plasma volume and plasma protein concentration remain normal. Primary absolute erythrocytosis (including polycythemia vera) is rare and considered a myeloproliferative disorder of the bone marrow.29 Serum erythropoietin concentrations are decreased to normal and Pao2 is normal. This condition may be accompanied by thrombocytosis or leukocytosis. Severe erythrocytosis has been described in two horses with normal serum erythropoietin levels.29,34 Secondary absolute erythrocytosis may be appropriate or inappropriate, depending on the presence or absence, respectively, of tissue hypoxia. Appropriate secondary absolute erythrocytosis is associated with increased circulating erythropoietin levels caused by chronic hypoxia from right-to-left shunting of blood in the heart or great vessels, chronic pulmonary disease, or adaptation to high altitude. The most common cardiac anomalies associated with erythrocytosis are complex defects such as tetralogy or pentalogy of Fallot, although a number of other defects, including ventricular septal defect, eventually may result in right-to-left shunting and secondary erythrocytosis.35 Horses with this condition frequently have cyanotic mucous membranes and an audible heart murmur. One can document chronic hypoxia by measuring the Pao2. Oxygen saturation concentrations of less than 92% usually are required for stimulation of erythropoiesis.29,36 One may con rm the diagnosis with cardiac radiography, ultrasonography, angiocardiography, or cardiac catheterization with blood gas and pressure analyses. Erythrocytosis is rare in horses with chronic lung disease. Auscultation of animals with chronic lung disease severe enough to result in hypoxia likely would reveal abnormal lung sounds such as wheezes and crackles. Thoracic radiographs aid in assessing the extent and severity of the problem. One should direct treatment of these animals toward correction or alleviation of the primary problem. If this is not possible, then periodic phlebotomy and removal of several liters of blood helps decrease blood viscosity and may improve the quality of life of these horses. Inappropriate secondary absolute erythrocytosis occurs after increased erythropoietin or other hormone release in the absence of tissue hypoxia. Although reported in horses with metastatic carcinoma and thoracic lymphoma,37,38 this condition is most common in horses with hepatic neoplasms or renal abnormalities. These animals have a normal or nearly normal Pao2. In human beings, hydronephrosis, renal cysts, and embryonal nephromata have been associated most frequently with this paraneoplastic syndrome. In the horse, secondary absolute erythrocytosis has been described in association with hepatocellular carcinoma and hepatoblastoma. All a ected horses have been less than 3 years of age, and concurrent elevations in serum erythropoietin concentrations, α-fetoprotein concentrations, or both have been described.39-41 These horses had a persistently elevated PCV that did not decrease with intravenous uid therapy, normal plasma protein concentrations, and mild to moderate elevations in hepatic enzymes. A horse with erythrocytosis that occurred after metastatic carcinoma also has been described. Most horses with primary or secondary erythrocytosis have a guarded to poor prognosis. If an underlying disease exists, then one should treat it appropriately. Intermittent phlebotomy may provide supportive therapy.29

ANEMIA Anemia is a decrease in the circulating RBC mass caused by an imbalance in the rate of loss or destruction of erythrocytes and the rate of their production in the bone marrow. Anemia is not considered a primary diagnosis but rather a hematologic abnormality resulting from an underlying disease process. Anemia is defined most frequently as a decrease in PCV or RBC numbers. Hemoglobin also is decreased, except in cases of intravascular hemolysis. A systematic approach to characterization of the anemia can provide valuable clues in the diagnosis of an underlying disease process. All anemias may be classi ed as regenerative or nonregenerative based on bone marrow response to the decrease

in circulating RBC mass. Regenerative anemia results from loss of intact erythrocytes from circulation (hemorrhage) or accelerated destruction of RBCs (hemolysis) and is characterized by an increase in e ective erythropoiesis in the bone marrow. Nonregenerative anemia occurs after systemic abnormalities or because of intrinsic bone marrow disease and results from a lack of appropriate marrow erythropoiesis in response to normal or accelerated RBC senescence or destruction. Anemia also may be characterized on the basis of erythrocyte size (MCV) and hemoglobin content (MCHC). Normocytic normochromic anemia accompanies many chronic systemic disease processes, including renal and hepatic failure, endocrine abnormalities, neoplastic conditions, and chronic infections. Microcytic hypochromic anemia (low MCV and MCHC) is associated classically with iron de ciency anemia. Small erythrocytes, characterized by a low MCV, are normal in foals. Macrocytic anemia (increased MCV) occasionally occurs in horses after a severe hemolytic or hemorrhagic crisis. Large erythrocytes, characterized by a high MCV, are normal in older horses, especially those over 20 years of age. Clinical signs of severe anemia relate to decreased tissue oxygenation and the physiologic compensatory mechanisms intended to alleviate this hypoxia. Signs include pallor of the mucous membranes, tachycardia, polypnea, weakness, lethargy, and a systolic heart murmur caused by decreased viscosity and increased turbulence of the blood as it ows through the heart and great vessels. Horses with mild to moderate anemia may have no obvious clinical signs or may have only lethargy and slightly pale mucous membranes. Other clinical signs, including fever, icterus, and hemoglobinuria, may be present in anemic horses and reflect the primary pathophysiologic process involved. Regenerative Anemia When erythrocytes are lost from the circulation at an accelerated rate by hemorrhage or hemolysis, the bone marrow responds by increasing its rate of erythrocyte production and release into the peripheral circulation. Documentation of an active regenerative response is important in diagnosis and prognosis of many cases of anemia. An increase in the number of reticulocytes in the peripheral circulation and an increase in the MCV are considered accurate indicators of a regenerative response in the bone marrow of most species. Unfortunately, in the horse, reticulocytes are not routinely released into the peripheral circulation, even during a strong regenerative bone marrow response. An equinespeci c, automated hematology instrument that was used in a horse with severe hemolytic anemia revealed the presence of a low level of reticulocytes in the peripheral circulation. This analysis was based on detection of nucleic acids within the RBCs by ow cytometry.42,43 Increases in MCV are inconsistent but tend to be slightly more common after hemolysis than after acute blood loss.44-46 A slight anisocytosis that is quantitatively assessed through changes in RBC distribution width can be observed in equine peripheral blood after acute hemorrhage or hemolysis.46-48 Because of the di culty in documenting a regenerative bone marrow response with conventional methods, several erythrocyte parameters have been investigated as markers of RBC regeneration in the equid.49-51 G6PD, creatine,49 and adenosine-5-triphosphate52 increase in equine erythrocytes during a regenerative response, but tests for these are not available in most veterinary laboratories. The most practical, reliable method of assessing the erythrocyte regenerative response in an anemic horse, aside from serial PCV analysis, is bone marrow analysis. A bone marrow M:E of less than 0.5 is considered evidence of erythrocyte regeneration.22 Reticulocyte counts in the bone marrow are also indicative of a regenerative response in the horse. Normal equine bone marrow contains approximately 3% reticulocytes, but this may increase to as high as 66% in response to severe blood loss.21,22 Hemorrhage Blood loss anemia may develop acutely or chronically. In many cases the source of the hemorrhage is obvious but in others is not. Internal hemorrhage (into body cavities) permits the body to reuse blood components. Approximately two thirds of the erythrocytes lost into the abdomen or thorax are autotransfused back into the circulation within 24 to 72 hours. The other one third are lysed or phagocytized, and the iron and protein are reused. External hemorrhage (including hemorrhage into the gastrointestinal tract) prevents reuse of these components. Accelerated bone marrow erythropoiesis is usually evident by 3 days after acute hemorrhage and is maximal by 7 days.12 Box 14-2 lists differential diagnoses for blood loss in the horse. BOX 14-2 DIFFERENTIAL DIAGNOSES FOR BLOOD LOSS IN THE HORSE Epistaxis Guttural pouch mycosis Pulmonary abscess Exercise-induced pulmonary hemorrhage Ethmoid hematoma Paranasal sinus abscess or infection

Traumatic nasogastric intubation Upper respiratory tract neoplasm Coagulopathy Trauma Pneumonia/pleuritis Hemothorax Thoracic trauma Fractured rib Lacerated heart or vessels Ruptured pulmonary abscess Ruptured great vessel Neoplasia Coagulopathy

Hematuria Pyelonephritis Cystitis/urolithiasis Neoplasia Trauma Urethral ulceration Coagulopathy

Hemoperitoneum Trauma Splenic rupture Hepatic rupture Mesenteric vessel rupture Verminous arteritis Uterine artery rupture Abdominal abscess Neoplasia Coagulopathy

Gastrointestinal Hemorrhage Ulcerations Nonsteroidal antiinflammatory drug (NSAID) toxicity Parasites Strongylus vulgaris Small strongyles

Granulomatous intestinal disease Histoplasmosis Tuberculosis Granulomatous enteritis Neoplasia Squamous cell carcinoma Lymphosarcoma Coagulopathy

External Hemorrhage Trauma Surgical complication Coagulopathy External parasites

Clinical signs of hemorrhage in the horse vary depending on the duration and severity of blood loss. Horses may lose up to one third (approximately 10 to 12 L) of their blood volume acutely without dying. Acute loss of large quantities of blood results in severe hypovolemic shock with tachycardia, polypnea, pale mucous membranes, poor venous distention, weakness, and oliguria. Initially, RBC parameters in the remaining circulating blood appear normal because all blood components have been lost in equal volumes. Physiologic compensatory mechanisms induce redistribution of interstitial fluid into the vasculature and eventually result in decreased RBC numbers and total protein in the peripheral blood. This redistribution may take up to 24 hours after acute hemorrhage. The di culty in estimating the severity of blood loss during the rst 24 hours after hemorrhage is compounded by the many erythrocytes that can be stored in the equine spleen and released into the circulation as a consequence of endogenous catecholamine release. After vascular equilibration, hematologic parameters should reveal a decrease in PCV, total RBC count, and hemoglobin with no change in associated RBC indices (MCV, MCHC, and MCH). Anemia caused by hemorrhage usually is accompanied by panhypoproteinemia because of signi cant loss of plasma proteins. A neutrophilic leukocytosis commonly is apparent by 3 hours after hemorrhage; platelets may increase as well if they have not been consumed by excessive coagulation.12 One may diagnose blood loss into the thoracic cavity by thoracic radiography, ultrasound, or thoracocentesis and may diagnose blood loss into the abdominal cavity by ultrasound or abdominocentesis. Horses with severe hemorrhage into the abdominal cavity may exhibit signs of colic or abdominal discomfort. One may di erentiate intra-abdominal hemorrhage from inadvertent splenic aspiration or laceration of a subcutaneous blood vessel by the presence of erythrophagocytosis and absence of platelets and may di erentiate recent hemorrhage into the abdominal cavity from diapedesis across compromised bowel or old hemorrhage by the presence of platelets in fresh blood. One may observe gastrointestinal blood loss as melena or hematochezia. More commonly, gastrointestinal hemorrhage is not severe enough to result in these overt signs, and feces appear normal. Brood mares in the periparturient period can su er from rupture of the middle uterine, utero-ovarian, or external iliac arteries.53,54 The ruptured vessel can lead to blood loss into the peritoneal cavity, into the broad ligament or serosa of the uterus, or into the lumen of the uterus iteslf.55 This condition should be considered possible in any mare up to 48 hours after foaling exhibiting the aforementioned clinical signs of acute blood loss, as well as lethargy, colic, and sweating. Older mares53,56,57 or multiparous mares58 are at the greatest risk and a link between age and declining serum copper concentrations, vascular degeneration, and fatal hemorrhage has been proposed.54 Therapy for hemorrhage discussed in the following sections pertains to this form of hemorrhage as well. Oxytocin can also be used to enhance uterine involution. E ective treatment of acute blood loss anemia must begin with nding and eliminating the source of hemorrhage. In many animals, this source is readily apparent, and direct pressure or surgical ligation of ruptured vessels is indicated. Internal hemorrhage or postpartum hemorrhage is more di cult to control because immediate surgery to access the site of hemorrhage is usually not an option. In horses with acute severe hemorrhage the administration of isotonic crystalline intravenous uids may aid tissue oxygenation by increasing peripheral blood volume and perfusion pressures. Administration of hypertonic saline solution (4 to 5 ml/kg of 7.5% NaCl) or hydroxyethyl starch (5 to 10 ml/kg) may be bene cial for rapid volume expansion after achieving adequate hemostasis.59 However, volume expansion has the potential risk of exacerbating hemorrhage if one cannot achieve hemostasis. Use of hypertonic or colloidal solutions should be followed by administration of appropriate isotonic crystalline fluid therapy. Polymerized ultrapuri ed bovine hemoglobin is an alternative oxygen-carrying uid that has been administered to horses, including

postpartum mares, in lieu of whole blood transfusion.60,61 Although expensive and used with variable success, this product has the advantages of compatibility without crossmatching and long shelf life. Naloxone, a -speci c opioid antagonist, has been recommended for use in horses with acute hemorrhage. Evidence suggests that endogenous opioids are involved in the pathophysiology of shock and that administration of naloxone may decrease some of the associated cardiovascular derangements.62 Aminocaproic acid is an anti brinolytic drug that has been used in human patients to reduce hemorrhage during various surgical procedures. In horses, it has been suggested that aminocaproic acid may be useful if a hyper brinolytic state exists.63 This drug has been used to treat mares su ering from periparturient hemorrhage.55 Pharmaco*kinetic studies in the horse suggest that a loading dose of 70 mg/kg intravenously (diluted in 1 L of saline and given over 20 minutes) followed by a constant-rate infusion at 15 mg/kg/hr is safe and achieves target blood concentrations. However, clinical efficacy research is lacking.64 The Chinese herb Yunnan Baiyao has been used with some evidence of e cacy in human patients with hemorrhage. The mechanism of action for this inexpensive herb is unclear but proposed e ects include decreased bleeding and clotting times, platelet e ect, and vasoconstriction. The drug decreased template bleeding times and activated clotting times in anesthetized ponies,65 and it is advocated as an adjunctive treatment protocol for mares with periparturient hemorrhage.55 A dose of 8 mg/kg orally every 6 hours has been suggested, but clinical trials are needed. If 20% to 30% of the total blood volume is lost acutely, then blood transfusion from a compatible donor horse is indicated. More detailed discussion of the principles of blood and blood product therapy occur later in this chapter. Hemorrhage into a large body cavity frequently is followed by autotransfusion of erythrocytes back into the peripheral circulation. Therefore one should not remove blood from these spaces unless necessary to stop continued bleeding (because the blood is interfering with vital organ function) or to lavage potentially septic sites. Horses with subacute hemorrhage (over a few days) usually begin to show clinical signs of severe anemia when the PCV declines to 15% to 20%. The same clinical signs may not be noticeable in horses with chronic blood loss (over weeks to months) until the PCV reaches 12% or less. Blood lactate concentration and central venous pressure may be used as indicators of the need for transfusion or to monitor cardiovascular responses to transfusion in horses with acute blood loss. Increased blood lactate concentrations and decreased central venous pressures are early indicators of blood loss. Post-transfusion, the pressures and blood lactate concentrations should rapidly return to near normal.66 Because these laboratory parameters are so variable and potentially misleading, one should base treatment decisions on the clinical signs in the individual patient and not necessarily on defined target clinical laboratory parameters. A horse with subacute to chronic blood loss anemia severe enough to result in peripheral tissue hypoxia as a result of inadequate RBC mass to supply needed oxygen exhibits tachycardia, polypnea, pale mucous membranes, weakness and lethargy, and possibly a mild to moderate systolic heart murmur caused by reduced viscosity and increased turbulence of the blood. If these signs are severe, then blood transfusion is indicated. As with acute hemorrhage, one must identify the source of the blood loss. Other treatment is supportive and should be designed to eliminate the primary problem. If external blood loss has occurred over weeks to months, then body iron stores may be depleted, a particularly common problem with chronic gastrointestinal bleeding. Depletion of iron slows or stops e ective bone marrow erythropoiesis so that these animals have laboratory parameters typical of iron de ciency anemia. One may identify low serum iron, low marrow iron stores, and increased TIBC, accompanying a hypochromic, microcytic anemia. Dietary iron may be supplemented orally with ferrous sulfate at a dose of 2 mg/kg body mass. One should not administer iron dextran parentally to horses because it has been associated with sudden death in this species. If iron must be administered intravenously, then iron cacodylate solutions are available. Administration of human recombinant erythropoietin to horses results in increased RBC mass and tissue oxygenation.67 After experimentally induced anemia, horses receiving concurrent recombinant erythropoietin and iron supplementation had RBC counts and reticulocyte parameters that indicated enhanced regeneration.43 However, repeated administration of human recombinant erythropoietin may result in production of antibodies against the drug that cross-react with, and inhibit the activity of, naturally produced equine erythropoietin. The end result is a potentially fatal nonregenerative aplastic anemia that persists until antibody levels are depleted. Because of the lack of peripheral signs of a bone marrow regenerative response in the horse, accurately assessing the regenerative e orts of bone marrow after hemorrhage may be di cult. One may evaluate serial bone marrow aspirates if this information is critical, but in most cases, monitoring increases in PCV over time is adequate. After experimental phlebotomy to mimic blood loss anemia in the horse, PCV increased approximately 0.672% per day, more slowly than in other species.52 Experimentally, the bone marrow erythroid response peaked at 9 days after phlebotomy, coinciding with the lowest marrow M:E ratio. Regeneration of the erythroid compartment was incomplete 31 days after hemorrhage.68 Hemolysis Intravascular or extravascular destruction of erythrocytes can occur in a variety of disorders. During intravascular hemolysis, hemoglobin released from destroyed erythrocytes combines with plasma haptoglobin, and tissue mononuclear phagocytes remove the haptoglobin-

hemoglobin complex. Plasma haptoglobin levels decrease as intravascular hemolysis increases.69,70 When plasma haptoglobin binding is exceeded, free hemoglobin accumulates in the plasma and is eliminated via the kidneys. Thus the hallmarks of intravascular hemolysis are hemoglobinemia and hemoglobinuria. Horses recovering from a severe hemolytic episode are more likely to have a demonstrable increase in MCV than are horses recovering from acute blood loss.44,45,71 The most common causes of hemolysis in horses are immune-mediated disease, oxidant-induced damage to erythrocytes, and infectious diseases. However, acute hemolytic anemia of adult horses has been associated with a wide variety of pathologic conditions (Box 14-3). Dimethyl sulfoxide administered intravenously at concentrations of 50% or greater results in severe intravascular hemolysis.72 Hemolyticuremic syndrome is characterized by acute renal failure, microangiopathic hemolytic anemia, and intravascular coagulation. Erythrocytes become damaged as they pass between brin strands deposited in the lumen of small renal vessels. Thrombocytopenia may occur concurrently.73,74 Hemolysis has been reported in horses ingesting leaves of the northern red oak (Quercus rubra L. var. borealis). Other associated clinical signs included abdominal pain, constipation, and increased coagulation times.75 Administration of signi cantly hypotonic or hypertonic solutions to horses also may result in intravascular hemolysis. All these conditions are uncommon in the horse. BOX 14-3 DIFFERENTIAL DIAGNOSES FOR HEMOLYSIS IN THE HORSE Infectious Diseases Piroplasmosis Equine infectious anemia (EIA)

Immune-Mediated Disease Autoimmune disease Bacterial infection Clostridium perfringens Streptococcal infections Viral infection EIA Neoplasia Lymphosarcoma Drug reaction Penicillin Neonatal isoerythrolysis

Oxidative Injury Phenothiazine Onion Red maple leaf Familial methemoglobinemia

Iatrogenic Conditions Hypotonic solutions Hypertonic saline

Miscellaneous Conditions Hepatic disease Hemolytic uremic syndrome

Disseminated intravascular coagulation (DIC)

Other Toxicities Intravenous dimethyl sulfoxide Bacterial toxins (Clostridium) Oak Burn injury

Immune-Mediated Hemolytic Anemia Immune-mediated hemolytic anemia develops when an animal produces antibodies that attach to the surface of RBCs. Primary IMHA is an autoimmune process in which antibodies are directed against surface antigens and occurs when a normally suppressed B-lymphocyte clone proliferates and produces an antibody directed against normal RBCs.76 Secondary IMHA is more common than true autoimmune disease. Antibodies attach to the surface of erythrocytes for several possible reasons: (1) alterations in the RBC membrane produced by a primary viral, bacterial, or neoplastic disease process; (2) antigen-antibody complex deposition on the surface of RBCs; or (3) drugs that cause immunoproteins to react indirectly with RBCs. Drug-induced immune-mediated hemolysis may occur via three mechanisms77: 1. The drug may combine with RBC membranes and may be recognized as foreign by the body. An antibody to this new antigen develops and destroys the drug-coated erythrocytes. These animals have a positive direct Coombs’ test. 2. The drug may complex with a carrier molecule in the blood and induce an immune response, and the drug-carrier-antibody complex attaches to RBC membranes in a complement-mediated process that leads to hemolysis. 3. Occasionally, a drug may induce true autoantibody production that leads to RBC destruction. Antibody-coated RBCs are unable to pass through the microcirculation of the spleen, become sequestered there, and are destroyed or phagocytized. If RBC membrane is lost in excess of intracellular contents, then spherocytes with an increased osmotic fragility form. Most immune-mediated hemolysis is extravascular; however, if the antibody xes and activates complement, then intravascular complementmediated hemolysis may result.13 IMHA may be classi ed as warm or cold hemagglutinin disease based on the optimal temperature at which the autoantibody agglutinates the host RBCs. Many clinical pathologists believe that these designations can be arbitrary and re ect an in vitro phenomenon rather than in vivo pathogenesis. Classically, in cold hemagglutinin disease, the antibody involved is IgM, complement is activated, and the liver is the primary site of removal of injured RBCs from the circulation. In warm hemagglutinin disease, IgG more commonly is involved, and the spleen is the primary site of RBC sequestration.13,78,79 Warm hemagglutinins are frequently incomplete (i.e., they do not cause autoagglutination).78 Warm80,81 and cold78,82 hemagglutinin anemia have been described in the horse. Primary and secondary IMHA are uncommon in the adult horse.48 Onset of disease is usually insidious, and many horses have fever, lethargy, and weight loss or have signs referable to a primary disease process. Routine hematologic and biochemical analyses frequently reveal decreased RBC numbers, spherocytosis, increased MCV, anisocytosis, and an increased total and indirect bilirubin. Most cases of immune-mediated hemolysis involve extravascular RBC destruction; however, if intravascular hemolysis is occurring, then hemoglobinuria will occur. One con rms the diagnosis by documenting a regenerative erythropoietic response in the bone marrow, increased erythrocyte fragility in hypotonic saline, autoagglutination, and a positive direct antiglobulin or Coombs’ test. The Coombs’ reagent used should contain antisera to IgG, IgM, and C3. One must di erentiate autoagglutination from rouleau formation by dilution of a cell suspension with isotonic saline. Saline disperses rouleau formation but does not a ect true autoagglutination. Flow cytometry has been used to con rm a diagnosis of IMHA in an affected horse by identifying and quantifying the amount of erythrocyte surface-bound antibody.83 In human beings, IMHA has been associated with lymphoreticular neoplasms, a variety of viral and bacterial diseases, in ammatory or granulomatous diseases, and generalized autoimmune disorders such as systemic lupus erythematosus. One may have di culty distinguishing between primary autoimmune hemolytic anemia (AIHA), in which autoantibodies are directed against an abnormal epitope on the erythrocyte membrane, and nonspeci c immune complex deposition on the surface of erythrocytes, usually occurring via attachment to Fc or complement receptors. In the horse, immune-mediated hemolysis has been reported in association with Clostridium perfringens septicemia and myositis,42,48,79 an unclassi ed respiratory infection,48 purpura hemorrhagica,82,84 a streptococcal abscess,82 and lymphosarcoma.14,82,84 Regardless of the exact sequence of events leading to antibody deposition on the surface of erythrocytes, a ected horses show clinical signs similar to those described for primary AIHA, and clinical pathologic data and diagnosis are similar. IMHA that occurs after treatment with penicillin has been documented in horses.85-88 Most a ected horses have not been hemoglobinuric,

indicating predominantly extravascular hemolysis. IMHA in the horse also has been associated with administration of trimethoprimsulfamethoxazole.89 Although not documented as causing immune-mediated hemolysis in the horse, drugs that have caused this problem in other species include tetracycline, rifampin, cephalosporins, chlorpromazine, and various nonsteroidal anti-in ammatory drugs (NSAIDs). In all reported cases of drug-induced IMHA in horses, the problem has resolved with discontinuation of administration of the o ending therapeutic agent and appropriate supportive care. Neonatal isoerythrolysis is an important form of immune-mediated hemolysis in the foal90 and is discussed in detail elsewhere in this text. Immune-mediated anemia resulting in intravascular and extravascular hemolysis may be present in horses with equine infectious anemia (EIA) (see the following discussion). Infected horses may have a positive result on the Coombs’ test or exhibit autoagglutination, especially during acute disease. Immune-mediated anemia also has been described in a horse with systemic lupus erythrematosus.91 Therapy of immune-mediated hemolysis is similar regardless of the initiating factor (or factors) involved. One should discontinue any previously administered drug and treat speci c underlying primary disease processes. If hemolysis is severe and life threatening, then the veterinarian should administer blood transfusions from a compatible donor (see the discussion on blood transfusion). If the horse has severe intravascular hemolysis with hemoglobinuria, then one should initiate intravenous uid therapy to protect against pigment-induced nephropathy. Several cases of immune-mediated hemolysis in the horse have been at least transiently responsive to corticosteroid therapy.14,48,80 One horse experienced complete disease remission after a 10-week regimen of glucocorticoids.76 Parenterally administered corticosteroids are recommended initially, and dexamethasone is probably the drug of choice. One should titrate the amount and frequency of administration to the response of the individual animal, but an initial dose of 30 to 40 mg dexamethasone may be tried in an adult horse. Ideally, administration of corticosteroids in the morning minimizes interference with the circadian rhythm of endogenous corticosteroid release from the adrenal gland. However, in an acute hemolytic crisis, twice-daily administration of dexamethasone is recommended until RBC numbers cease to decline. After stabilization, one should decrease corticosteroid doses to once daily, then every other day, and gradually discontinue the therapy. If the horse requires long-term therapy, then one may administer prednisolone orally with a declining every-other-day dosage. Potential complications of corticosteroid therapy in the horse include laminitis and secondary infections. Exacerbation of primary infectious conditions also may occur in cases of secondary autoimmune hemolysis treated with corticosteroids. Caution is recommended, especially when using corticosteroids at a high dose or for a prolonged period. One horse with IMHA refractory to corticosteroid therapy was treated successfully with cyclophosphamide at 1.1 mg/kg intramuscularly once daily and azathioprine at 1.1 mg/kg intramuscularly once daily.92 Oxidative injury to RBCs is evident from Heinz body formation, hemolytic anemia, and methemoglobinemia. Heinz bodies are aggregations of oxidized precipitated hemoglobin. They are small, round, blue-black, refractile granules near the cellular margin of erythrocytes stained with new methylene blue.93 Heinz bodies damage the erythrocyte membrane, disturbing intracellular tonicity and resulting in rupture of the cell (intravascular hemolysis). Cells of the mononuclear-phagocyte system of the spleen and liver may recognize damaged RBC membranes as abnormal and remove the RBCs from the peripheral circulation (extravascular hemolysis). One mechanism by which the RBCs protect themselves is through the production of reduced nicotinamide adenine dinucleotide phosphate (NADPH) via the pentose pathway. Several enzymes are essential to this pathway, including the rate-controlling enzyme G6PD. Oxidation of iron in the hemoglobin molecule from the normal ferrous state (Fe2+) to the ferric state (Fe3+) forms methemoglobin. Erythrocytes normally produce methemoglobin, which cellular enzymes like FAD rapidly convert back to hemoglobin. In addition, cellular GSH acts to decrease the amount of methemoglobin formed by competing with hemoglobin for oxidizing agents.94 Methemoglobin is incapable of carrying oxygen, and excessive accumulation results in a brownish discoloration of blood. Methemoglobin formation alone does not result in hemolysis,95 but toxins simultaneously may oxidize the sulfhydryl groups of the globin moiety of methemoglobin. This oxidation causes methemoglobin denaturation and precipitation, Heinz body formation, and intravascular hemolysis. Pathologic methemoglobin accumulation in the blood most frequently results from exposure to an oxidative toxin that overwhelms the protective mechanisms of the body but also may result from a decreased rate of reduction of physiologically produced methemoglobin. The aforementioned enzymes involved in erythrocyte metabolism are essential, and a de ciency of any of these enzymes can cause a decrease in the erythrocyte life span and a form of anemia. Several genetic alterations that lead to these de ciencies have been identi ed in human beings and dogs; however, only three have been identified in the horse thus far, and all result in clinical signs typical of oxidative RBC injury. A de ciency of G6PD in human beings is the result of a well-described X-linked genetic mutation that is relatively common. When subjected to oxidative insult, a ected individuals develop hemolytic anemia and hyperbilirubinemia.96 A 6-month-old American Saddlebred colt was diagnosed with this disorder via a series of tests including G6PD assays, electron microscopy, and numerous other erythrocyte assays. The colt had no history of exposure to exogenous oxidants but was su ering from a signi cant dermatitis. This was assumed to be the source of endogenous oxidants, and the hemolytic anemia improved as the dermatitis resolved. Management of these cases may simply include prevention of in ammation and other forms of oxidative stress. Although most clinicians believe that this disorder is inherited in horses as it is in human beings, this has not been confirmed.7

Two adult horses, a Spanish Mustang and a Kentucky Mountain Saddle Horse, were diagnosed with FAD de ciencies through a series of tests including avin assays. Both horses had persistent methemoglobinemia, eccentrocytosis, and pyknocytosis on laboratory analysis. Hemoglobin crystals were identified in the Mustang. Neither horse was profoundly anemic nor were exogenous oxidants identified.8 Familial methemoglobinemia and hemolytic anemia caused by decreased erythrocyte GSH reductase and intracellular GSH have been described in two Trotter mares.97 These mares displayed findings similar to those with FAD deficiency. Acute hemolytic anemia and methemoglobinemia can develop in horses after ingestion of wilted red maple (Acer rubrum) leaves.71,98-101 The toxic principle is uncertain, but gallic acid, a strong oxidant, has been implicated in conjunction with another unidenti ed agent.102 Freshly harvested leaves are not dangerous but become toxic when dried and remain so for at least 30 days. Overnight freezing of the leaves does not a ect toxicity. A ected horses present most commonly in late summer and fall, often with brownish discoloration of the blood (massive methemoglobinemia) and peracute death within 12 to 18 hours of exposure. A more prolonged hemolytic syndrome characterized by icterus, methemoglobinemia, hemoglobinuria, bilirubinemia, bilirubinuria, fever, and possibly death may occur 5 days or more after ingestion of wilted leaves. Horses may initially present for evaluation of colic.101 Laboratory evaluation of horses with the hemolytic syndrome of red maple leaf toxicosis frequently reveals Heinz body anemia, depletion of RBC-reduced GSH, and increased RBC osmotic fragility.99 Normal methemoglobin concentration in equine blood is 1.77% of total hemoglobin.97 Horses with red maple leaf toxicosis may have methemoglobin concentrations of close to 50% of total hemoglobin.100 Treatment of horses with red maple leaf toxicity includes eliminating access to the toxic leaves, whole blood or packed cell transfusions as needed, and general nursing care. The veterinarian should treat all exposed animals with activated charcoal via nasogastric tube to decrease absorption of toxin, even if several days have elapsed since exposure. Massive intravascular hemolysis may result in pigment-induced nephropathy. Fluid diuresis is indicated to prevent or treat this complication. One should use potentially nephrotoxic drugs with caution. Some clinicians have suggested dexamethasone administration to stabilize cellular membranes, but administration of corticosteroids is associated with an increased mortality rate in a ected horses.98,101,103 Methylene blue therapy has been attempted in some horses without success.71 Because methylene blue may potentiate hemolysis, its use is not advisable. Two horses with severe hemolysis and methemoglobinemia caused by ingestion of red maple leaves were treated with large doses of ascorbic acid (vitamin C), but administration of this drug is not signi cantly associated with survival.101,103 Ascorbic acid is thought to convert methemoglobin to reduced hemoglobin and may supplement endogenous protective mechanisms.104 One may give adult horses a dose of ascorbic acid at 30 mg/kg twice daily in intravenous fluids.103 Prognosis for horses with red maple leaf toxicosis is guarded to poor and is not associated with initial physical examination ndings or laboratory abnormalities at the time of presentation.101 Horses are at risk for pigment nephropathy and acute anuric renal failure, colic, diarrhea, and laminitis.100 Mortality rates have been reported at approximately 60%. Outbreaks of Heinz body anemia have been reported in horses with access to unharvested onions105 or overgrazed pasture covered with wild onions (Allium canadense).93 The toxic principle in onions is n-propyl disul de. Treatment should include removal from access to the plants, activated charcoal via nasogastric tube to reduce absorption of the toxin, and blood transfusions and supportive care as indicated. Heinz body anemia in horses also has been reported after phenothiazine administration. Toxicity appears to depend on individual susceptibility and some as yet unidenti ed environmental factors. Groups of animals may be a ected at a dose that is generally considered safe, and animals in poor condition may be more susceptible to toxicity.106 Although a 30-g dose for a 1000-lb horse is considered safe, one report described poisoning in six of nine horses treated with 25 g of phenothiazine.107 Because phenothiazine is no longer widely used as an equine anthelmintic, reports of its toxicity are becoming rare. Nitrate poisoning is associated with methemoglobin formation in horses but only rarely with intravascular hemolysis. Although it has been speculated that the horse might be more susceptible to nitrate poisoning than other simple-stomach species because the cecum and colon provide an optimal environment for microbial reduction of nitrate to nitrite,71,108,109 nitrate toxicity appears to be extremely rare in horses. Heinz body anemia has been reported in one horse with lymphosarcoma.110 The authors could not con rm access to known oxidative toxins and presumed the anemia to be caused by the neoplastic process by an unidenti ed mechanism. Several horses that endured severe subcutaneous burns developed intravascular hemolytic anemia because of oxidative damage to RBCs. The mechanism of the damage to the RBC is not fully understood but does seem to correlate with severity of burns.111 Equine infectious anemia virus (EIAV) is a viral disease of horses known colloquially in the United States as swamp fever because of its high prevalence in Gulf Coast states where climatic conditions are favorable for transmission. The causative agent is a member of the lentivirus genus of the family Retroviridae. Infected horses may have one of three clinical syndromes: (1) acute infection, (2) chronic infection, or (3) inapparent carrier.112,113 Acutely infected horses show fever, lethargy, and anorexia within 30 days of exposure. The most consistent hematologic abnormality in these horses is thrombocytopenia, although anemia also may be present. Many acutely infected animals show few if any recognizable clinical signs, and de nitive diagnosis may be di cult because some of these horses are not yet seropositive. Most horses seroconvert by 40 days after infection.

The horse chronically infected with EIAV has classic signs of recurrent fever, weight loss, ventral edema, and anemia. These animals are seropositive by the agar gel immunodi usion (AGID or Coggins) test or by competitive enzyme-linked immunosorbent assay (C-ELISA). Each cycle of disease is associated with the emergence of a new antigenic strain of the virus, temporary evasion of the host immune response, and replication of the virus to high titer. The anemia from which the disease derives its name results from intravascular and extravascular hemolysis of complement-coated RBCs and an impaired bone marrow response.114-117 Each febrile episode is associated with thrombocytopenia, but platelet counts rebound rapidly as temperature returns to normal. Thrombocytopenia may result from immune118 or nonimmune119 platelet destruction or decreased platelet production in the bone marrow.120 Most EIAV seropositive horses are clinically normal and never show any recognizable clinical signs. However, they have immune-related blood abnormalities (hyperglobulinemia, decreased percentage of CD5+ and CD4+ lymphocytes) consistent with ongoing viral activity.121 These animals are inapparent carriers of the virus and remain infected for life. They have circulating infectious virus in their blood and remain a threat to other horses for the rest of their lives. Regardless of disease stage (acute, chronic, or inapparent carrier), virus is found in vivo primarily in tissue macrophages and endothelial cells.122-124 Transmission of EIAV occurs predominantly by the intermittent feeding of hematophagous insects such as horse ies and deer ies. Flies are strictly mechanical vectors, and virus does not appear to survive more than 30 to 120 minutes on their mouthparts. Chances of transmission are greatest when ies feed on horses undergoing a febrile, viremic episode. However, horse ies can transmit virus from inapparent carrier horses to uninfected animals under field conditions, so all seropositive horses must be considered infectious for life. EIAV also can be transmitted iatrogenically with blood product transfusions and previously used or improperly sterilized needles, surgical instruments, tattooing instruments, dental equipment, or any other blood-contaminated materials. Virus occasionally is transmitted across the placental barrier from an infected mare to its foal. Mares experiencing a febrile episode during gestation are more likely to give birth to infected foals than are asymptomatic mares. Of foals born to chronic carrier mares, approximately 10% are virus and antibody positive.125,126 Virus transmission to foals may occur by ingestion of colostrum and milk from infected mares.126,127 Foals born to antibodypositive mares usually are seropositive for EIAV at 24 hours of age because of absorption of colostral immunoglobulins. This colostral immunity is usually undetectable by 6 months of age.126,127 The Coggins test and C-ELISA are recognized by the U.S. Department of Agriculture as valid and reliable for the diagnosis of EIAV. Good correlation has been reported between Coggins test results and C-ELISA results.128 A few horses have been identi ed, however, with consistently negative or equivocal AGID results that were proved subsequently to be infected.129,130 No specific antiviral therapy for EIAV is available. Treatment of an animal requires supportive therapy as indicated during febrile episodes. Minimizing environmental stress may be helpful in decreasing the severity and recurrence of clinical signs. All EIA reactor horses (seropositive) must be permanently identi ed using the National Uniform Tag code number assigned by the United States Department of Agriculture (USDA) to the state in which the reactor was tested, followed by the letter “A”. A hot brand, chemical brand, freezemarking, or lip tattoo may be used and must be applied by a USDA representative. Brands must be at least 2 inches high and applied to the left shoulder or left side of the neck. Lip tattoos should not be less than 1 inch high and 3⁄4 inches wide and should be applied to the inner surface of the upper lip of the reactor. Reactor horses must be separated from other horses by quarantine at the premises of origin, euthanasia, or movement to a federally approved diagnostic or research facility. The ramining horses on the premises must be serologically tested for EIA with repeat testing at 30 to 60 day intervals until no new cases are found. The premises are released from quarantine when all tests have been negative for a minimum of 60 days after the last reactor equid was removed.131 Federal law prohibits interstate travel of horses that have tested positive for EIAV. Interstate movement of infected horses is allowed under three conditions: (1) back to the farm of origin, (2) to slaughter, and (3) to a diagnostic laboratory or approved research facility. Before interstate movement, reactors must be identi ed o cially using the national uniform tag code number assigned by the U.S. Department of Agriculture to the state in which the reactor was tested, followed by the letter A. A hot iron, chemical brand, or freeze-marking may be used. Markings must be at least 2 inches high and applied to the left shoulder or left side of the neck. Despite the signi cant decline in the incidence of EIAV in the United States since these control measures were implemented in the mid1970s, propagating epizootics still occur.132 Veterinarians should advise horse owners to do the following: 1. Require a negative EIAV test as part of every prepurchase examination. 2. Require all new arrivals on a farm to have documentation of a recent negative EIAV test, and test all horses on the farm annually. 3. Practice excellent fly control. 4. Encourage all events involving the congregation of horses to require documentation of a recent negative EIAV test. 5. Thoroughly disinfect any surgical items contacting equine blood before use on another horse.

6. Do not share or reuse needles and syringes between horses. Equine piroplasmosis results from infection with one or both of two species of hemoprotozoan parasite: Babesia caballi and B. equi. These intraerythrocytic parasites are found in subtropical locales and transmitted predominantly by tick vectors. Although classi ed in the genus Babesia, B. equi may be related more closely to the theilerial organisms.133 Babesia equi undergoes some developmental stages in lymphocytes and apparently lacks transovarial tick transmission typical of most Babesia organisms. Iatrogenic transmission has occurred through administration of contaminated blood products and use of unsterilized or shared needles or surgical instruments. Piroplasmosis is only enzootic in those areas where the tick vector can survive the winter such as South and Central America, the Caribbean, and areas within Africa, Europe, and the Middle East. Babesia caballi infection has previously been diagnosed in horses in Florida and is spread by the tropical horse tick Dermacentor nitens. In the United States, D. nitens has been found in southeastern Florida and occasionally in Texas. Only rarely has B. equi infection been con rmed in horses in the United States, and none of the ticks commonly found in this country are known to transmit this organism. The United States is currently considered Babesia free. Horses raised in Babesia-endemic areas frequently are infected with the organisms without ever showing recognizable clinical signs. Clinically recovered horses remain infected, asymptomatic carriers of the organism as well, and stress may precipitate clinical relapse. Horses with B. caballi infection may clear the organism spontaneously after 12 to 42 months, whereas horses infected by B. equi do not appear to clear the organism spontaneously.134 Infected animals develop a strong active immunity that depends on the continuing presence of the organism (premunity). They may be reinfected readily, soon after the organism is eliminated from the body. An infected dam will passively transfer antibodies to o spring but antibody levels are undetectable by 63 to 77 days of age.135 No evidence of cross-protection exists between the two species of Babesia.136 Previously unexposed adult horses develop clinical signs of disease within 1 to 4 weeks of exposure. The horse may have fever, depression, dyspnea, pale or icteric mucous membranes, ecchymoses of the nictitating membrane, constipation, colic, and dependent edema. As anemia worsens, a ected horses may develop diarrhea. Massive intravascular destruction of parasitized erythrocytes occasionally occurs, resulting in hemoglobinuria. Cardiac arrhythmias may develop as a result of myocardial damage.137 Clinical disease with B. caballi lasts a few days to a few weeks, and mortality is usually low. Horses infected with B. equi generally have a more severe clinical course and may die within 24 to 48 hours of initial signs. Diagnosis can be made by identi cation of the Babesia organism in blood smears stained with a Giemsa-type stain. The absence of Babesia organisms in the peripheral blood does not exclude the diagnosis of piroplasmosis because parasitemia may be brief and occur before the onset of recognizable clinical signs. Several diagnostic tests are also available. A complement xation (CF) test was previously recommended as a means of identifying both the actively infected and the carriers. The indirect uorescent antibody (IFA) test is a more sensitive test than the CF test, especially for latent infections.138 However, the most sensitive test, and therefore the most appropriate test for screening asymptomatic carriers, is the competitive inhibition enzyme-linked immunosorbent assay (CI-ELISA).139 Because native horses in the United States are rarely infected with this organism, the United States has instituted strict regulations for import of horses from endemic areas. Before import, all horses must undergo mandatory testing by the Animal and Plant Health Inspection Servise using the CI-ELISA. This test is considered the standard for identi cation of inapparent carriers and is now commercially available to horse owners and veterinarians.140 The World Organization for Animal Health lists the ELISA and IFA tests as prescribed tests for equine piroplasmosis with the CF test as an alternative test. Treatment of piroplasmosis varies depending on the location of the horse and the desired goal of treatment. In animals that reside in Babesia-endemic areas, suppressing clinical signs without eliminating the organism from the body is desirable, because premunition depends on the continued presence of the parasite at low levels. Clinical signs exhibited by carriers usually subside after one intramuscular injection of imidocarb dipropionate (Burroughs Wellcome Co., Research Triangle Park, N.C.) at 2.2 mg/kg. Owners who wish to move their horses to or enter Babesia-free areas should isolate them from all tick vectors and should have them treated to eliminate the organism. Administration of imidocarb dipropionate organism for treatment of B. caballi has shown variable results. Elimination of the organism was achieved at 2 mg/kg intramuscularly once daily for 2 days.141 However, recent research has shown that long-term elimination was not achieved with doses as high as 4.7 mg/kg every 72 hours for ve doses.142 B. equi is a more di cult organism to eliminate from the horse, and four doses of imidocarb at 4 mg/kg intramuscularly at 72-hour intervals had variable e cacy in eliminating the carrier state.143-145 In one study by Frerichs, Allen, and Holbrook,144 the foregoing treatment regimen successfully cleared 13 of 14 horses of the infection, whereas in a more recent study by Kuttler, Zaugg, and Gipson,143 this treatment regimen cleared none of nine geldings of infection. This wide di erence in e cacy may be because of strain di erences in drug susceptibility. Potential adverse e ects of imidocarb administration include salivation, restlessness, colic, and gastrointestinal tract hypermotility. Donkeys appear to be sensitive to the toxic e ects of imidocarb, and eight donkeys treated with the drug died.144 Although the antitheilerial drug buparvaquone at 4 to 6 mg/kg intravenously or intramuscularly is therapeutically effective in horses acutely infected with B. equi, it is not consistent in clearing infection in carrier horses.146 Severe Hepatic Disease Acute intravascular hemolysis and anemia may develop in the terminal stages of acute or chronic hepatic failure in horses and is characterized by hemoglobinemia, hemoglobinuria, and icterus.147 Onset and progression are rapid, and the condition is usually fatal;

however, signs of hemolysis may subside if the liver disease is treated sucessfully.148 A ected erythrocytes have increased osmotic fragility. Hemolysis may result from decreased structural integrity of the erythrocyte membrane caused by alterations in exchangeable RBC membrane lipoproteins and the effect of bile acids on erythrocyte metabolism during liver failure.13 Microangiopathic Hemolysis This disorder occurs after thrombosis or brinoid change within the lumen of small blood vessels,149 is typical of chronic disseminated intravascular coagulation (DIC), and has been reported in horses.73 The resulting hemolysis is usually mild. Nonregenerative Anemia A variety of intrinsic or extrinsic factors may suppress normal bone marrow erythropoiesis, resulting in anemia caused by a failure to replace senescent RBCs adequately as they are removed from circulation (Box 14-4). The most common disorders associated with nonregenerative anemia in the horse are iron de ciency; chronic in ammatory, endocrine, or neoplastic diseases; and generalized bone marrow failure. Dietary factors often are incriminated in these abnormalities but only rarely are involved. Severe protein deprivation may result in decreased erythropoiesis as the body becomes de cient in synthesizing hemoglobin and other cellular proteins. Folic acid and cobalamin (vitamin B12) de ciencies cause macrocytic hypochromic anemia in human beings, but they are associated rarely with anemia in domestic animals, including the horse. Horses do not have an absolute dietary requirement for vitamin B12, which is produced by bacterial action in the gut and absorbed from the lower gastrointestinal tract.150 Anemia associated with hypothyroidism is thought to be functional, following a lowered metabolic rate. Normocytic normochromic anemia responsive only to thyroid replacement therapy has been reported in one horse.151 BOX 14-4 DIFFERENTIAL DIAGNOSES FOR NONREGENERATIVE ANEMIA IN THE HORSE Iron Deficiency Chronic hemorrhage Nutritional deficiency (rare)

Chronic Disease Chronic infection/inflammation Pleuritis/pneumonia Peritonitis/enteritis Bacterial endocarditis Internal abscess Chronic viral disease (e.g., equine infectious anemia [EIA]) Neoplasia Endocrine disorders

Bone Marrow Failure Myelophthisis Myeloproliferative disease Bone marrow toxins Phenylbutazone Chloramphenicol Radiation Idiopathic pancytopenia

Miscellaneous Conditions Administration of human recombinant erythropoietin Chronic hepatic disease

Chronic renal disease Recent hemorrhage or hemolysis

Iron Deficiency Iron de ciency in the horse is associated only rarely with low dietary iron intake or absorption. More commonly, iron de ciency results from chronic external blood loss. In initial stages of iron de ciency, the only detectable abnormalities re ect decreased iron storage pools: a low serum ferritin concentration and decreased stainable iron in the bone marrow. Progression of the condition a ects erythropoiesis adversely. These animals have a decrease in the percentage saturation of plasma transferrin, an increased TIBC, and increased numbers of hypochromic erythrocytes (decreased MCHC). Fulminant iron deficiency anemia leads to abnormal development of erythrocytes in the bone marrow late in the maturation process. Cells continue to divide in the late rubricyte stage, without su cient iron to continue heme synthesis. The result is the release of small cells with decreased hemoglobin concentration into the peripheral circulation (decreased MCHC and MCV). Treatment of iron de ciency anemia should concentrate initially on the identi cation and elimination of the source of chronic iron loss. In the horse, occult blood loss occurs most frequently from the gastrointestinal tract and may be veri ed with a fecal occult blood test. Chronic gastrointestinal ulceration after phenylbutazone administration is a well-recognized phenomenon and is more frequent and severe in ponies than in horses. Gastrointestinal parasites, especially Strongylus vulgaris and small strongyles, occasionally may result in chronic blood loss and subsequent iron de ciency. Many iron-containing hematinics are available commercially, and oral supplementation with a product containing ferrous sulfate is probably best. Parenteral iron dextran solutions have been associated with fatal anaphylactoid reactions in horses. If parenteral administration is essential, then 1 g of iron cacodylate administered intravenously for an adult horse generally is considered safe. Neonatal foals in the rst few days of life have high serum iron concentrations and a high percentage saturation of transferrin.11,27 As a result they are quite susceptible to iron toxicity if they receive oral or intravenous supplements. Administration of oral digestive inoculants containing ferrous fumarate to neonatal foals was associated with a fatal toxic hepatopathy characterized by icterus, hepatic atrophy, bile duct hyperplasia, lobular necrosis, and intrahepatic cholestasis.2,28 Researchers have concluded that little bene cial e ect occurs on hematologic variables when giving healthy foals oral iron supplementation.152 In contrast, older foals may be more susceptible than adults to the development of iron de ciency anemia because of their naturally lower levels of iron during growth (after the rst week of life). In a survey of iron status in hospitalized horses, only six animals with iron de ciency were identi ed on the basis of low serum ferritin and serum iron concentrations. All six were foals younger than 5 weeks of age.153 Foals, although technically iron de cient, rarely develop anemia as a result. A severe anemia attributed to a signi cant iron de ciency was identi ed in a foal that concurrently su ered from septicemia.154 The source of the de ciency was not identi ed and considered mulitfactorial, including lack of access to dirt and pasture and low dietary iron intake. Chronic Disease Chronic in ammatory conditions and neoplasms frequently are associated with a mild to moderate normocytic, normochromic, nonregenerative anemia. This anemia has been attributed to several abnormalities. A block of iron released from reticuloendothelial storage (ferritin and hemosiderin) results in an unavailability of iron for heme synthesis. Horses with anemia of chronic disease typically have a decreased serum iron concentration, normal to decreased TIBC, decreased transferrin saturation with iron, normal to increased serum ferritin concentration, and normal to increased bone marrow storage iron. In addition to altered iron mobilization, a defective response of the bone marrow to circulating erythropoietin and a decrease in RBC life span during many chronic diseases are apparent. The anemia of chronic disease rarely is associated with clinical signs of decreased tissue oxygenation. Therapy is directed at the primary disease process. Oral iron supplementation is not indicated because systemic iron stores are usually normal to increased. Bone Marrow Suppression Anemia caused by selective bone marrow suppression of erythropoiesis is unusual in the horse. Selective erythroid hypoplasia has been reported as a sequela to administration of recombinant human erythropoietin to horses.155,156 Anemia is thought to result from production of antibodies that cross-react with erythropoietin, inhibiting erythropoiesis. Treatment is with blood transfusions and corticosteroids. Prognosis for recovery is fair to guarded. Pancytopenia (decreased circulating erythrocytes, leukocytes, and platelets) has been described in horses with aplastic bone marrow disorders 157-159 and with myelophthisic diseases. Myelophthisis is a reduction in the cellular elements of the bone marrow, frequently a result of metastatic neoplasia or myelo brosis, a proliferation of brous tissue in the marrow. Myelodysplasia implies accumulation of abnormal cells in the bone marrow. These cells are not usually dysplastic in the truest sense of the word, but rather they result from myeloproliferative neoplasia. Hematopoietic neoplasia is rare in the horse and is discussed in more detail later in this chapter. These conditions usually result in concurrent anemia, leukopenia, and thrombocytopenia. Diagnosis is by identi cation of abnormal cells in bone marrow aspirates.

Bone marrow from horses with aplastic anemia is generally devoid of hematopoietic precursor cells because of a congenital or acquired failure of stem cell function. Acquired aplasia may be associated with bacterial and viral infections, chronic renal or hepatic disease, neoplasia, irradiation, or drug therapy. In most cases a predisposing factor is not identi ed.157-162 Because of the shorter life spans of the cells, leukopenia and thrombocytopenia usually precede anemia.12 Diagnosis depends on bone marrow biopsy. If one cannot obtain adequate marrow by serial aspirates, then one must obtain a core biopsy sample. Hypoplastic marrow is yellow because of fatty in ltration with a mixture of cell types, but the number of erythroid, myeloid, and megakaryocytoid cells is decreased.12 Aplastic anemia is thought to be an autoimmune disorder in some cases, and some horses have apparently bene ted from therapy with corticosteroids and anabolic steroids.157 These drugs stimulate erythropoietin production and increase sensitivity of the stem cell receptors to erythropoietin. Temporary or permanent bone marrow aplasia has been associated with the administration of a variety of drugs in other species. Transient hypoplastic anemia after phenylbutazone administration has been reported in one horse.82 Aplastic anemia has been induced experimentally by feeding of trichloroethylene-extracted soybean oil meal to horses or by irradiation.163 Chloramphenicol produces transient and irreversible forms of bone marrow dysfunction in human beings, and similar reactions are possible in other species. One always should handle chloramphenicol with caution because human toxic reactions appear to be idiosyncratic and may be associated with absorption of low amounts of the drug. Any ongoing drug therapy should be discontinued in horses with aplastic anemia, and supportive therapy should be administered. Some cases of drug-induced bone marrow failure are temporary, and hematopoietic function returns to normal with time if secondary complications are not severe. One may administer compatible whole blood transfusions to horses with severe anemia or thrombocytopenia, and systemic antibiotics and strict isolation from other animals may be considered for horses with severe leukopenia. Bone marrow erythropoiesis frequently is suppressed in horses with chronic renal disease. Suppression generally is attributed to a decrease in erythropoietin production by the kidneys. Recombinant erythropoietin has been used successfully in cats and human beings with severe anemia associated with renal disease.

LEUKON The leukon consists of circulating leukocytes, their precursor cells, and the tissues that produce them. This includes the granulocytes (neutrophils, eosinophils, and basophils), monocytes and macrophages, and lymphocytes. The leukon provides the primary e ector cells for immune surveillance and clearance. This discussion provides a brief description of the production of these cells and their quantitative abnormalities in peripheral blood. Excellent pictures of the various developmental stages of equine leukocytes in the bone marrow and peripheral blood are available in other texts, and the reader is referred to these for identification of cell types.12,13

Neutrophils PHYSIOLOGY Granulocytes and mononuclear phagocytes originate from a common committed progenitor cell in the bone marrow. Under the in uence of cytokines—including granulocyte-macrophage colony-stimulating factor, monocyte colony-stimulating factor, granulocyte colony-stimulating factor (G-CSF), and various interleukins—this stem cell proceeds along a path to granulocyte or monocyte production. Cytokine signals for constitutive leukocyte production di er from those during periods of increased demand for speci c cell types. The progenitor cell undergoes successive mitotic divisions from myeloblast (the first recognizable precursor of neutrophils) to promyelocyte and myelocyte. Mitosis does not occur as the cell matures from metamyelocyte to band cell to mature neutrophil. Mature neutrophils are stored in the bone marrow until needed. The peripheral neutrophil pool is equally divided between circulating cells and cells adhered to the endothelium of small vessels (the marginated pool).164 Circulating neutrophils have a half-life of 10 1/2 hours in the horse165 and eventually migrate into peripheral tissues, where they live several more days. Neutrophils alter their distribution between storage, circulating, and marginating pools in response to various endogenous and exogenous stimuli. The marginating neutrophils are mobilized into the circulating pool in response to exercise, epinephrine, or stress. Glucocorticoids increase the rate of neutrophil egress from the bone marrow storage pool and decrease egress from the circulation.166,167 Neutrophils are attracted to sites of infection and in ammation by soluble chemotactic factors released during pro-in ammatory reactions. Neutrophils ingest and kill invading microorganisms and release additional factors that further propagate in ammation at the site of tissue injury. The morphology of neutrophils on a stained smear of blood or body uids (e.g., peritoneal or pleural uid) aids in interpretation of the signi cance and severity of many disorders. Toxic changes in circulating neutrophils occur in response to in ammation and include cytoplasmic basophilia, cytoplasmic granulation and vacuolation, and appearance of Döhle’s bodies (slate gray inclusions caused by retention and aggregation of rough endoplasmic reticulum).13 Degenerative changes result from altered cell membrane permeability and include hydropic degeneration of the nucleus and a spreading out of the nuclear chromatin so that it more completely lls up the cytoplasm of the cell. Aged neutrophils may exhibit hypersegmented, pyknotic nuclei with round, tightly clumped chromatin. Aged neutrophils most commonly are visible in tissue uids. Idiopathic hypersegmentation of blood neutrophils of one Quarter Horse, unrelated to any clinical

disease, has been described.168

DISORDERS OF NEUTROPHILS Neutropenia Neutropenia is a decrease in the number of circulating neutrophils and may be acute (occurring transiently over 24 to 48 hours) or chronic (lasting several days to months). Acute neutropenia most commonly results from a shift of neutrophils from circulating to marginating pools. Endotoxin is a potent stimulus for margination of circulating neutrophils with sequestration in pulmonary capillaries.164 Experimentally, administration of endotoxin to horses results in neutropenia within 90 minutes and return to baseline numbers within 6 to 18 hours.169 Endotoxemia is probably the common denominator in the neutropenia associated with various equine gastrointestinal disturbances including strangulating obstruction, peritonitis, enteritis, and salmonellosis. Neutropenia is a common nding in acute septicemia in the adult and is considered an aid for diagnosis of sepsis in the neonate.170 A variety of bacterial, rickettsial, and viral disorders also may be associated with neutropenia in the horse. Chronic neutropenia may result from increased peripheral use of neutrophils or decreased bone marrow production. Neutropenia may accompany severe infectious or in ammatory diseases such as pleuritis, pneumonia, peritonitis, internal abscessation, enteritis, burns, vasculitis, or immune-mediated diseases. Neutropenia caused by increased use often is accompanied by appearance of immature cells in circulation. As tissue demand increases, the bone marrow releases progressively more immature band cells into circulation.12 This regenerative left shift is an appropriate response to high tissue demand for neutrophils. In animals with a degenerative left shift, the number of immature band cells or metamyelocytes in circulation exceeds the number of mature neutrophils because tissue use exceeds the capacity of the bone marrow to increase production.13 A degenerative left shift is considered a poor prognostic indicator. Neutropenia caused by bone marrow suppression may occur in horses with pancytopenia from a variety of causes (discussed previously). Myeloproliferative disorders including granulocytic leukemia also may result in chronic neutropenia.171 Alloimmune neonatal neutropenia (ANN), an immune-mediated neutropenia in newborn foals, has been described in several breeds.172-174 This disease is con rmed by ow cytometric identification of increased neutrophil surface-bound antibody. Affected foals require treatment with prophylactic antibiotics during the recovery period. They may also bene t from administration of G-CSF as described below. The neutrophil count generally returns to normal within 1 month. A syndrome of neutropenia and thrombocytopenia is described in related Standardbreds. A heritable cyclic neutropenia was suspected. Most affected horses died from complications of infectious diseases, thrombocytopenia, or both.175 Administration of canine or bovine recombinant G-CSF to normal newborn foals results in a profound increase in blood neutrophil count after a single subcutaneous injection of 6 g/kg.174,176,177 The factor may also bene t foals recovering from septicemia. Recombinant G-CSF is reported to improve recovery of adult horses when administered after experimental bowel resection.178 Neutrophilia Many of the same disorders associated with neutropenia may be associated alternatively with neutrophilia. Endogenous or exogenous glucocorticoids or epinephrine, excitement, exercise, or stress may result in neutrophilia.13,179 Any infections or in ammatory process in any part of the body may result in neutrophilia. Many neoplastic conditions also are accompanied by peripheral neutrophilia. A rebound neutrophilia is common in later stages of endotoxemia. The balance between bone marrow production and tissue use determines the magnitude of a neutrophilic response. A regenerative left shift is not uncommon in animals with neutrophilia because of excessive tissue demand. A severe neutrophilia with signi cant left shift including metamyelocytes and myelocytes indicates serious in ammatory disease and is termed a leukemoid response because of its similarity to granulocytic leukemia.12

Eosinophils Eosinophils arise from the same bone marrow precursor as neutrophils and mononuclear phagocytes. They may be distinguished rst from neutrophils at the early myelocyte stage and eventually develop characteristic bright red–staining cytoplasmic granules. IL-5 is critical in the di erentiation and maturation of eosinophils. The prominent cytoplasmic granules of eosinophils contain a variety of substances, including major basic protein, peroxidase, and various hydrolytic enzymes. Eosinophils are important in parasite immunity and are involved in some hypersensitivity reactions. Eosinopenia may result from acute infections, corticosteroid or epinephrine administration or release, or stress. Peripheral eosinophilia most commonly results from parasitic infections, including habronemiasis, strongylosis, and pediculosis. Occasionally, allergic reactions also may result in peripheral eosinophilia. Eosinophilic myeloproliferative disease has been described in the horse.180 A marked eosinophilia also has been seen in horses with lymphosarcoma and transitional cell carcinoma.181

Basophils Basophils also originate from the common granulocyte-macrophage precursor cell and mature into cells with basophilic cytoplasmic granules. They are the least common of the circulating granulocytes, and basopenia is not clinically signi cant. Basophils live in the blood

approximately 6 hours and then migrate into tissues, where they exist another 10 to 12 days.13 Basophils are involved in mediating some hypersensitivity reactions. Increased numbers of circulating basophils may occur with allergic, in ammatory, or neoplastic diseases or in association with lipemia.

Mononuclear Phagocytes Monoblasts are the rst recognizable bone marrow precursors of peripheral blood monocytes, originating from a pluripotent marrow precursor. After progressing through the promonocyte stage, monocytes are released into the peripheral blood where they circulate for a few days before migrating into tissues to mature into tissue macrophages. The role of mononuclear phagocytes in orchestrating immune and in ammatory reactions has long been underplayed. These cells are critical in regulating in ammation through release of proin ammatory cytokines such as tumor necrosis factor (TNF), IL-1, and platelet-activating factor. Mononuclear phagocytes phagocytose microbial organisms, particulate debris, and possibly neoplastic cells. They process these foreign antigens and present them to T lymphocytes in a form that initiates speci c immune responses. They produce cytokines, including granulocyte-macrophage colony-stimulating factor, that are critical for regulation of hematopoiesis. They are responsible for removal of senescent cells and activated coagulation factors from circulation. Quantitative abnormalities of monocytes rarely are encountered in equine medicine, but monocytosis may be observed in some cases of chronic inflammation.

Lymphocytes Lymphocytes are the primary mediators of humoral and cell-mediated immune responses. They originate from an uncommitted bone marrow progenitor cell that is capable of di erentiating into a committed lymphopoietic precursor or a granulocyte-macrophage precursor cell. Committed lymphopoietic progenitor cells di erentiate into T lymphocytes that mature during migration through the thymus or B cells that appear to di erentiate in the marrow and then migrate to lymph nodes. Circulating lymphocytes are predominantly T cells and represent only a small fraction of the total lymphocyte pool. Most lymphocytes reside in the spleen, lymph nodes, and other lymphoid tissue of the body. The spleen contains the greatest number of lymphocytes in the adult horse. The spleen plays a major role in immune defense. The large number of phagocytic cells in the spleen facilitates ltering of senescent blood cells, particulate debris, and microorganisms from the blood. Hemoglobin is degraded, and iron is stored in splenic phagocytes pending reuse for erythropoiesis. The spleen is also important as a reservoir of RBCs and platelets, as discussed previously. Lymphopenia is associated with glucocorticoid administration, stress, many viral infections, and combined immunode ciency of Arabian foals. Lymphocytosis is associated with epinephrine administration, excitement, exercise, lymphocytic leukemia, and chronic immune stimulation. As horses age, the number of lymphocytes in their peripheral blood gradually decreases.182,183

Hematopoietic Neoplasia MYELOID NEOPLASIA Neoplasia that results in unregulated proliferation of leukocytes can be categorized as myeloid neoplasia and lymphoid neoplasia. Lymphoid forms, including lymphosarcoma, are more common. Myeloid neoplasia is characterized by uncontrolled proliferation of one or more marrow-derived blood cell lines. When neoplastic cells are observed in peripheral circulation, the disease process is termed leukemia. Forms of myeloid neoplasia that have been described in the horse include acute and chronic granulocytic leukemia,171,184,185 myeloblastic leukemia,186 myelomonocytic leukemia,187-190 monocytic leukemia,191 eosinophilic myeloproliferative disorder,180 and malignant histiocytosis.192 Regardless of the involved cell line, the eventual outcome is significant myelophthisis with loss of normal marrow elements. No age predilection is apparent for equine myeloid neoplasia.171,180 Common clinical signs include depression, weight loss, edema, anemia, and mucosal petechial hemorrhages. Fever,180,187,189,190 peripheral lymphadenopathy,187,190 nonhealing wounds,185 hemorrhagic diathesis,180 and oral ulcers 187,188 were identified in some of the horses. Most a ected horses are anemic and thrombocytopenic; total white blood cell counts may be decreased, normal, or increased. Common clinical signs and hematologic parameters can be explained by severe destruction of normal bone marrow resulting in inadequate production of erythrocytes, platelets, and normal leukocytes. This predisposes a ected horses to hemorrhagic diathesis, infection, and inadequate oxygenation of tissues. De nitive diagnosis can often be made by observation of abnormal cells on a peripheral blood smear. Abnormal leukocytes typically predominate in bone marrow aspirates and may be identi ed in peripheral lymph node aspirates. Acute leukemia is de ned as more than 20% blast cells in the peripheral circulation, whereas chronic leukemias lack a signi cant number of blast forms in the blood.185 Exact identi cation of cell type can be di cult because of abnormal cellular architecture in neoplastic cells. As a result, additional staining and cytochemical analysis is often needed for a specific diagnosis. The cause of myeloid neoplasia remains uncertain. Treatment with cytotoxic190 drugs and with steroids186,184 has been attempted in

a ected horses with no success. Appropriate long-term combination chemotherapy protocols have not been attempted. Compared with lymphoid neoplasia, presentation of clinical signs as a result of myeloid neoplasia is often acute and patients rarely survive for more than a few weeks after diagnosis. Horses that su er from forms that are better di erentiated tend to have a longer duration of clinical signs before presentation, but unfortunately the prognosis remains the same.193 Postmortem examinations have revealed the presence of abnormal leukocytes within various organs, most commonly the liver and spleen.

LYMPHOID NEOPLASIA The most common type of neoplasia to involve the equine hemolymphatic system is lymphosarcoma.194 Four anatomic forms have been described (generalized, intestinal, mediastinal, and cutaneous) based on the major site of tumor involvement194,195; however, the forms overlap substantially, both clinically and pathologically. Rarely, horses can develop lymphocytic leukemia either as a primary neoplastic condition or, much more commonly, as a sequela to lymphosarcoma that has metastasized to the marrow.193 The typical age of onset of lymphoid neoplasia in horses is between 5 and 10 years,196 although lymphosarcoma has been documented in horses ranging in age from birth through 25 years.197-199 No breed or gender predilection exists. The incidence of disease is unknown; however, reported prevalence of a ected horses at autopsy is between 2% and 5%, and less than 3% of all tumors are classi ed as lymphoma.200 Lymphocytic or lymphoblastic leukemia is quite rare in horses, so conclusions regarding age, breed, and gender predilections are difficult to interpret.193,201-203 Cause and Pathogenesis A viral cause of equine lymphosarcoma has not been documented204,205; however, one report describes viruslike particles in the lymph node of a foal with lymphosarcoma that died shortly after birth.199 A congenital or genetic predisposition is possible, because tumors were observed in an aborted fetus and in two horses less than 1 year of age.206 Clinical signs and laboratory changes of lymphosarcoma generally are caused by loss of normal organ and tissue function after in ltration by lymphocytes or physical obstruction by tumor masses or the excessive generation of tumor cell cytokines.206 Data on immunophenotyping of equine lymphosarcoma are con icting. Histologic examination of tumor cells typically reveals a heterogeneous cell population. Previous research indicates that the majority of equine lymphoma is B-cell lymphoma, with many of these described as T-cell rich B-cell lymphomas. A more recent report suggested the opposite: T-cell lymphoma is more prevalent that B-cell lymphoma.206 Anatomically, T-cell lymphoma is commonly associated with mediastinal tumors and B-cell tumors tend to be more multicentric. Too few cases of lymphocytic leukemia have been described to draw accurate conclusions regarding cytologic classification, but T-cell lymphoma tends to predominate. Decreased serum concentrations of IgM and other indications of immunosuppression have been observed in horses with lymphosarcoma.207-210 A decreased serum IgM concentration is de ned as ≤23 mg/dl. A recent study found that IgM concentrations should not be used as a screening test for equine lymphoma because of extremely poor sensitivity.211 Other neoplastic lymphocytes may arise from autoreactive B-cell clones that produce antibodies responsible for gammopathies and immune-mediated cytopenias. Neoplastic proliferation of large granular lymphocytes with natural killer cell activity was described in one horse.212 Clinical Signs and Laboratory Findings Lymphosarcoma most commonly presents as a chronic, progressive disease with a sudden onset of more signi cant clinical signs. The most common clinical signs of lymphosarcoma are chronic weight loss, ventral subcutaneous edema, and regional lymphadenopathy.213,214 Peripheral lymphadenopathy is not observed frequently. Other clinical manifestations are highly variable depending on the organs involved and the duration of the disease process. Sudden death was attributed to thoracic lymphoma in one racing Standardbred.215 Lymphosarcoma involving the thoracic cavity may cause tachypnea, dyspnea, cough, and pleural e usion.197,216,217 Lymphoma should also be considered in horses with recurrent pleural e usion.218 Ultrasound and thoracic radiography can be useful for the diagnosis if tumors are located in an area that is amenable to imaging with these modalities. If pleural e usion is present, then one frequently identi es neoplastic lymphocytes by routine cytologic evaluation of fluid obtained by thoracocentesis.183 If cytologic ndings are nondiagnostic but indicative of thoracic lymphoma, then immunophenotyping of the uid may be performed to identify markers of lymphosarcoma.218 Alimentary lymphoma, the most common intestinal neoplasia of horses,195 causes clinical signs consistent with an in ltrative, in ammatory disorder of the gastrointestinal tract including weight loss, recurrent colic, and diarrhea.219-221 Weight loss can often be attributed to malabsorptive disease, as well as to the negative energy balance caused by chronic, systemic inflammation. The regional or generalized occurrence of multiple subcutaneous nodules with histopathologic characteristics of lymphosarcoma, unassociated with other lesions of lymphosarcoma, has been reported in horses.194,222 Nodules may appear suddenly, grow slowly, remain static or regress, and recur at a later time. Hormonal factors including estrous cycle, pregnancy, lactation, and foaling may in uence this cutaneous form of lymphosarcoma. In one horse, cutaneous lymphosarcoma cells were positive for progesterone receptors, and lesions regressed completely after surgical removal of an ovarian tumor.223

Tumor masses in localized areas may result in clinical signs referable to dysfunction of involved and adjacent tissues and organs.204,213,216,224-229 Pharyngeal and laryngeal masses may result in dysphagia and unresponsive nasal discharge. Ocular lymphosarcoma may manifest as uveitis or as in ltrates in palpebral conjunctiva, eyelids, third eyelid, at the corneoscleral junction, or in the retrobulbar area.230 Lymphosarcoma of the central nervous system may result in ataxia and cranial nerve de cits.204,229,231 Other manifestations of lymphoma include parotid gland, uterine, bladder, epididymal, and periarticular tumors.232-236 Angiotrophic T-cell lymphoma, which is characterized by proliferation of neoplastic lymphocytes within the lumen of small vessels, was identi ed in a Thoroughbred mare with a history of anemia but no abnormalities in the circulating leukocyte examination.237 The laboratory ndings in horses with lymphosarcoma are highly variable. Hematologic indications of chronic in ammatory disease are common, including neutrophilic leukocytosis, nonresponsive anemia, hyper brinogenemia, thrombocytopenia, and hypergammaglobulinemia.206,214,216 Anemia may result from immune-mediated destruction, decreased production because of some degree of myelophthisis, or a combination of both.206 Horses with lymphocytic leukemia have lymphocytosis and circulating, neoplastic lymphocytes.221,238-241 The total plasma protein concentration may be low, normal, or elevated, but the albumin-to-globulin ratio often is reduced, particularly with gastrointestinal involvement.221,242 The ventral edema observed does not always result from hypoalbuminemia and is more likely caused by lymphatic stasis or obstruction.206 Protein electrophoresis produces variable results, with the possibility of either polyclonal or monoclonal gammopathy. Polyclonal gammopathies are nonspeci c and may be due in part to the production of acute phase proteins, Ig, tumor cytokines, or secondary infection. A mild increase in liver-derived serum enzymes may occur with hepatic involvement. Hyperbilirubinemia usually is caused by anorexia or hemolysis. A variety of paraneoplastic syndromes have been identi ed in association with equine lymphosarcoma. Hypercalcemia occasionally is reported.226,243-245 Pseudohyperparathyroidism, resulting in hypercalcemic nephropathy and polyuria or polydipsia, has been reported in a horse with splenic lymphosarcoma.246 In some patients, Coombs’-positive IMHA, immune-mediated thrombocytopenia (IMTP), or both may occur.14,84,247 Hypereosinophilia has been reported in one pony with intestinal lymphosarcoma.181 Concurrent intestinal lymphosarcoma and eosinophilic epitheliotrophic disease have been reported in a Paso Fino mare.248 These two cases suggest the possibility that clonal proliferation of lymphocytes occasionally may result in hypersecretion of IL-5 with subsequent eosinophil activation. A case of thoracic lymphoma resulted in paraneoplastic polycythemia because of erythropoietin gene expression from the tumor cells.38 Paraneoplastic pruritus and alopecia also has been described in a horse with lymphosarcoma.243 Diagnosis The diagnosis of lymphosarcoma is con rmed by demonstration of neoplastic lymphocytes in a ected tissue. Histologic examination of a biopsy from a tumor mass or a ected lymph node is the most reliable method, and excisional biopsies are much preferred over needle biopsies or aspirates. Without lymph nodes or other masses accessible for biopsy, antemortem diagnosis is often di cult.14,197,228,249 Diagnosis is only rarely possible on a peripheral blood smear238,240 and generally requires careful cytologic evaluation of bone marrow, pleural e usion, or peritoneal uid.216,250,251 Between 38% and 50% of horses with the alimentary form of lymphoma have evidence of neoplastic cells on abdominocentesis.252,253 The clinician rarely is able to locate neoplastic lymphocytes by transcutaneous liver biopsy254; however, laparoscopy may be used to visualize a mass for biopsy more clearly.197 Although the lymphocyte count more commonly is normal or reduced in horses with lymphoma, the presence of atypical or obviously neoplastic lymphocytes on a peripheral blood smear occurs in 30% to 50% of cases.∗ One may palpate splenic enlargement, internal lymphadenopathy, or abdominal masses on rectal examination. The character of these masses can then be de ned further through ultrasonography.256 Abdominocentesis may reveal in ammatory changes in the uid (increased nucleated cell count and protein concentration), but identifying neoplastic cells in abdominal uid of horses with alimentary or abdominal lymphosarcoma is not common. Radiography and ultrasonography may be useful in locating and perhaps enabling biopsies of masses in the thorax or abdomen. Often, diagnosis of lymphosarcoma is possible only by exploratory laparotomy or postmortem examination.14,219,228,249 The morphologic condition of neoplastic lymphocytes in horses is highly variable.198 On cytologic preparations they often appear as large lymphoid cells with a variable nucleus-to-cytoplasm ratio, multiple nucleoli, nuclear chromatic clumping, cytoplasmic basophilia, and vacuolation. Mitotic gures and binucleate cells may be visible. The cytologic diagnosis of lymphosarcoma is best left to those experienced in evaluation of equine uid specimens, because normal reactive lymphocytes and mesothelial cells may be di cult to distinguish from welldifferentiated neoplastic cells. Histologically, the neoplastic cellular morphology also varies, but destruction of normal tissue architecture by a population of lymphoid cells aids the diagnosis. More advanced diagnostic techniques may be used to speci cally analyze the neoplastic cell line and its immunologic characteristics. These tests include cytochemical staining, immunophenotyping using both ow cytometry and immunohistochemistry, and immunologic function testing.193 Treatment

Clinical experience in treating horses with lymphosarcoma is limited, because horses tend to be greatly debilitated by the time of diagnosis. Resection of localized tumors may be curative or at least signi cantly prolong survival.257 Several solitary tumors located in areas such as the perineum and the nasal cavity have been treated successfully with high-dose radiation.258 Remission of generalized disease can be transiently achieved through the use of cytotoxic drugs, immunomodulators, and immunosuppressants. Chemotherapeutic regimens and dosages have been adapted from human and small animal oncology and are now being used with some success in the equine lymphoma patient. Controlled clinical research supporting the use of these drugs is lacking, but case reports and anecdotal evidence support their use where appropriate. The drugs are dosed on a square meter calculation of body surface area, and the protocols generally use a combination of cytotoxic and immunosuppressive drugs administered over several weeks to months depending on the severity. Most commonly used protocols include cytosine arabinoside, cyclophosphamide, vincristine, chlorambucil, doxorubicin, Lasparaginase, and prednisolone.218,259 One report suggests a remission rate of as high as 50% for several months to a year in horses diagnosed with multicentric lymphoma using a protocol of prednisolone, vincristine, and cytosine arabinoside.259 Immunosuppressive doses of dexamethasone have also been e ective in reducing the size of some tumors and severity of clinical signs for up to 18 months.247 The cutaneous form of lymphosarcoma is often responsive to corticosteroid and progestin therapy or a combination of low-dose cyclophosphamide and immunotherapy with an autologous tumor cell vaccine.260 However, abrupt discontinuation or insu cient duration of therapy may result in recurrence of cutaneous lesions in a more aggressive and rapidly progressive form.261 Treatment of lymphocytic leukemia with the aforementioned therapies has been less rewarding. These drugs are typically very expensive, and remission has not been established in the horse with any of various reported protocols. The prognosis for equine generalized lymphosarcoma is grave for long-term survival. Most horses die or are euthanized within 6 months of diagnosis; however, with increased use of chemotherapy in the horse, these survival times may begin to increase. Horses with lymphocytic leukemia typically do not survive for more than several weeks after a diagnosis regardless of treatment strategy.193

Plasma Cell Myeloma Plasma cell myeloma (multiple myeloma), characterized by proliferation of neoplastic plasma cells in the bone marrow, spleen, liver, and lymph nodes, is rare in horses. Clinical signs include weight loss, weakness, recurrent fever, ventral edema, hemorrhage, lameness, and posterior paralysis.262-266 Recurrent infections secondary to the lack of normal immunoglobulins are not uncommon. Renal failure has also been described in a ected horses. In human beings and dogs, skeletal pain and pathologic fractures result from neoplastic invasion of bone that causes osteolytic “punched-out” lesions267,268; osteolysis, however, does not occur consistently in horses. Laboratory ndings include anemia, hypercalcemia, azotemia, and hyperproteinemia with monoclonal gammopathy. A diagnosis of plasma cell myeloma should be considered in horses that demonstrate hypercalcemia with normal parathyroid hormone (PTH), elevated parathyroid hormone–related protein (PTH-rP), and hyperglobulinemia.269 Light-chain proteinuria is variable. If any degree of lameness is noted, then radiographs of the a ected limb are warranted. Criteria for diagnosis of plasma cell myeloma include plasmacytosis in the bone marrow or a soft tissue lesion, evidence of invasiveness, and the presence of monoclonal gammopathy or light-chain proteinuria.267 The degree of plasmacytosis observed in the bone marrow is variable, because marrow involvement may be focal rather than diffuse. Examination of multiple samples from di erent areas of marrow are recommended whenever possible. If protein electrophoresis yields nonspeci c results, then it is recommended to perform radioimmunodi usion to quantify speci c Ig classes.269 AL amyloidosis, which is commonly associated with multiple myeloma in human beings, was diagnosed in one horse with concurrent multiple myeloma.270 Histologic and clinicopathologic ndings in some horses with multiple myeloma can be quite similar to those of lymphosarcoma; therefore achieving a de nitive diagnosis may be di cult.271,272 Of the 15 horses with plasma cell myeloma described to date in the literature, none have been successfully treated. Horses typically survive for no more than 2 years after initial diagnosis. Management of horses with multiple myeloma generally includes careful monitoring and treatment of recurrent infections as needed.273

PLATELETS Platelets are circulating anucleate fragments of bone marrow megakaryocytes that are essential for the formation of primary hemostatic plugs. Platelets adhere to injured vascular endothelium, release substances that initiate and propagate hemostatic events, and provide a phospholipid surface for activation of several coagulation factors.274 Platelets also are important proin ammatory elements that interact with endothelium, mononuclear phagocytes, neutrophils, and fibroblasts in initiation and propagation of inflammation.

Physiology Platelets ultimately originate from the same pluripotent stem cell as erythrocytes and leukocytes. This stem cell undergoes successive divisions, becoming progressively more di erentiated. The colony-forming unit megakaryocytoid cell is committed fully to platelet production. Polyploid megakaryoblasts double their DNA content by a process of endomitosis, becoming megakaryocytes, the largest of the hematopoietic cells of the bone marrow, within 5 days. The highly compartmentalized cytoplasm of the megakaryocyte ultimately disintegrates to release up to 8000 platelets into the peripheral circulation. Various cytokines, including IL-3, thrombopoietin, and IL-6, are

essential for this process of differentiation, maturation, and release. The mature platelet is a complex anucleate cell fragment that accumulates many substances in a variety of secretory granules. These include dense bodies containing adenosine diphosphate (ADP), adenosine triphosphate (ATP), calcium, and serotonin; α-granules that accumulate various procoagulant factors, albumin, platelet-derived growth factor, thrombospondin, and platelet factor 4; and lysosomal granules with acid hydrolases.274 Glycoprotein receptor molecules are embedded in the external lipid bilayer membrane and interact with extracellular proteins that mediate platelet adhesion and aggregation. The external lipid membrane of the platelet invagin*tes into the interior to form a complex canalicular system important for secretory function. A second internal membrane system, the dense tubular system, supplies enzymes for arachidonic acid metabolism and is involved in sequestration and release of calcium.275 When a circulating platelet contacts exposed vascular subendothelial collagen, platelet surface receptors recognize and attach to the damaged endothelium in a process known as adhesion. Normal adhesion requires von Willebrand’s factor (vWF), a glycoprotein found in vascular endothelium and platelets and present in plasma as a multimolecular complex with coagulation factor VIII. (The role of vWF in platelet adhesion is separate from the role of factor VIII in secondary hemostasis, which is discussed later. The two activities are referred to as VIII:vWF for the platelet adhesion mediated by vWF and as VIII:C for the coagulative protein.) The glycoprotein P-selectin is another essential platelet surface receptor that is involved in primary hemostasis by enhancing interactions between platelets, endothelial cells, and leukocytes. Expression of P-selectin offers a possible platelet surface marker for platelet activation.276 Adhesion of platelets attracts additional platelets in a process known as aggregation. Primary aggregation is reversible and not associated with platelet degranulation. Irreversible secondary aggregation begins with release of ADP from platelet granules and production of thromboxane A2 (TXA2) via metabolism of arachidonic acid in the platelet membrane. Substances that promote aggregation usually do so by decreasing the concentration of intracellular cyclic adenosine monophosphate by inhibiting adenylate cyclase or by stimulating phosphodiesterase.274 Equine platelets exhibit reversible aggregation when exposed to serotonin and arachidonic acid and exhibit irreversible aggregation when exposed to ADP.277 Platelet-platelet bridging requires brinogen attachment to a newly exposed binding site on activated platelets.275 Thrombospondin, a glycoprotein released from platelet α-granules, is also integral in the platelet aggregation response. Platelet aggregation is responsible for formation of the initial hemostatic plug (primary hemostasis) at any site of vascular injury. Arachidonic acid release from the membranes of the platelet-dense tubular network and its subsequent metabolism are critical to platelet aggregation. Arachidonic acid is a polyunsaturated fatty acid normally attached to the second glycerol carbon atom of many membrane glycerophospholipids. Arachidonic acid may be released by the direct action of phospholipase A2 or by the sequential actions of phospholipase C and diglyceride lipase. Free arachidonic acid is metabolized via a cyclooxygenase pathway to form prostaglandins and TXA2 or via a lipoxygenase pathway to form various eicosanoids. TXA2 is a potent platelet activator, stimulating platelet aggregation and vasoconstriction. Although platelets preferentially produce TXA2 via the cyclooxygenase pathway, vascular endothelial cells produce prostacyclin (prostaglandin I2), a potent inhibitor of platelet aggregation and a strong vasodilator. The balance between TXA2 production by platelets and prostacyclin production by vascular endothelium appears to regulate platelet aggregation in vivo.274,275 As with erythrocytes, platelets can be sequestered in the spleen and splenic contraction can increase the number of circulating platelets by 30% to 50%. In horses, splenectomy results in substantial and persistent increases in platelet counts.13 Normal equine platelets have a life span of 4 to 5 days.278 As platelets age, tissue macrophages of the spleen, liver, and bone marrow remove them from circulation.279

Disorders of Platelets THROMBOCYTOPENIA Thrombocytopenia is a decrease in the number of circulating platelets. Normal platelet counts in the horse are slightly less than in other species, with higher counts in horses younger than 3 years of age and in male horses.280 Most laboratories de ne thrombocytopenia in the horse as a peripheral platelet count of less than 100,000 per microliter. Clinical signs of thrombocytopenia in the horse re ect abnormal primary hemostasis and include petechial and ecchymotic hemorrhages of mucosal membranes, epistaxis, increased bleeding after venipuncture, melena, or hyphema.281-286 Clinical bleeding usually is associated with platelet counts of less than 30,000 per microliter. Decreased numbers of circulating platelets can result from decreased production of platelets in the bone marrow, increased destruction of platelets, sequestration of platelets, or increased use of platelets during processes of coagulation (Box 14-5). BOX 14-5 DIFFERENTIAL DIAGNOSES FOR THROMBOCYTOPENIA IN THE HORSE Decreased Platelet Production Hereditary defects Myelophthisis

Myeloproliferative disease Idiopathic pancytopenia Myelosuppressive drugs Phenylbutazone Chloramphenicol Estrogens Trichlorethylene-extracted soybean meal Irradiation

Increased Platelet Use Intravascular coagulation Disseminated intravascular coagulation (DIC) Localized Hemolytic uremic syndrome Hemangioma Hemorrhage Thrombosis

Platelet Sequestration Splenomegaly

Increased Platelet Destruction Infectious diseases Equine infectious anemia (EIA) >Anaplasma phagocytophilum Immune-mediated diseases Autoimmune disease Systemic lupus erythematosus Idiopathic disease Secondary disease Neoplasia (lymphosarcoma) Bacterial infection Viral infection Drugs Neonatal alloimmune disease Drugs or toxins Snakebites Laboratory Error

Proper sample collection and platelet counting is essential before one can make a diagnosis of true thrombocytopenia in a horse.279 One can obtain an accurate platelet count in most horses from a blood sample obtained with ethylenediamine tetraacetic acid (EDTA) as the anticoagulant. However, the platelets of some horses consistently clump in EDTA, resulting in an inaccurate count or pseudothrombocytopenia.287 An experienced technician always should examine a stained blood smear visually to assess the presence of platelet clumps. Platelet clumping may occur as a result of in vivo or in vitro platelet activation.279 When in doubt, reassessing the platelet count in a blood sample anticoagulated with sodium citrate instead of EDTA is advisable. Platelet clumping and pseudothrombocytopenia also have been described in equine blood samples collected using low–molecular-weight heparin as an anticoagulant.288 Decreased Platelet Production Thrombocytopenia only infrequently results from decreased platelet production.289 Primary bone marrow disease is diagnosed by bone

marrow analysis, and caution must be used when interpreting megakaryocyte numbers from equine bone marrow aspirates. Megakaryocytes may be absent or present in low numbers in aspirates, even though adequate numbers are present in the intact marrow, possibly because of trapping of these large cells within the subendothelial layer of marrow sinuses. Bone marrow core biopsies are preferred for adequate evaluation of megakaryocyte numbers in the horse. Flow cytometric enumeration of thiazole orange–positive platelets in peripheral blood may be useful as a noninvasive test to assess platelet production.290 Because of the short life span of platelets in the peripheral circulation, thrombocytopenia is a common abnormality in animals with generalized bone marrow suppression of any cause. Thrombocytopenia with dramatically decreased bone marrow megakaryocyte numbers has been reported in horses with a variety of myeloproliferative disorders.171,180,187,264 Thrombocytopenia also is recognized in horses with idiopathic pancytopenia156,160 and drug- or toxin-induced myelosuppression. A syndrome of neutropenia and thrombocytopenia has been described in related Standardbreds. A heritable cyclic neutropenia was suspected. Most affected horses died from complications of infectious diseases, thrombocytopenia, or both.175 Increased Platelet Destruction A variety of pathophysiologic events may increase the rate at which platelets are removed from circulation. Most of these disorders are accompanied by a compensatory bone marrow megakaryocytosis best identi ed with core marrow biopsies. Immunoglobulin deposition on the surface of platelets is the most common cause of accelerated platelet destruction. Cells of the mononuclear phagocyte system, especially splenic macrophages and hepatic Kup er’s cells, remove antibody-coated platelets from circulation. Nonimmune mechanisms of accelerated platelet destruction include drug- or toxin-induced platelet damage and snake bites. Thrombocytopenia of controversial cause has been described in human beings and horses receiving heparin therapy.291,292 Immune-Mediated Thrombocytopenia Most cases of idiopathic thrombocytopenia in the horse likely are the result of an immune-mediated increase in platelet destruction. The initiating factors of this condition often are not identi ed, but immune-mediated destruction of platelets may be associated with viral or bacterial infections, neoplastic conditions such as lymphosarcoma, and with other immunemediated disorders, including IMHA, glomerulonephritis, and systemic vasculitis.281 Circulating platelets from human beings with IMTP have increased quantities of surface-bound IgG, IgM, or complement.293-295 The mechanics of antibody attachment to platelets are similar to those described for IMHA. True autoantibodies may be produced against normal platelet surface antigens or against novel platelet antigens that have developed in response to a primary disease process. Circulating immune complexes may attach to platelets nonspeci cally via Fc or complement receptors. Antibodies also may be directed against foreign molecules, such as drugs, that have attached themselves to platelet membranes. Regardless of the source of the antibody, mononuclear phagocytes remove Ig-coated platelets from the circulation, principally in the spleen and liver.296 Horses with IMTP may have petechial and ecchymotic hemorrhages of mucosal membranes, epistaxis, increased bleeding after venipuncture, melena, or hyphema.281-286 Clinical bleeding usually is associated with platelet counts of less than 30,000 per microliter. Patient history and a careful physical examination are critical for identifying underlying disease. An AGID or Coggins test for EIA is indicated. One should evaluate other coagulation parameters, including prothrombin time (PT) and partial thromboplastin time (PTT), to identify complex coagulative defects such as those seen with DIC (discussed in a subsequent section). A bone marrow analysis usually reveals normal to increased numbers of megakaryocytes. However, if antibodies against platelet membranes cross-react with megakaryocyte membrane antigens, megakaryocytes may be decreased or absent in the bone marrow.297 Con rmation of IMTP in horses can be di cult, requiring demonstration of increased quantities of platelet-bound antibody. This can be accomplished by measuring platelet factor 3 or by performing flow cytometry. Platelet factor 3 is a platelet membrane phospholipid released when platelets are injured. Addition of patient plasma containing released platelet factor 3 to platelet-rich plasma from a normal horse results in accelerated clotting of the sample. Unfortunately, this test is di cult to perform consistently, results vary, and most veterinary laboratories do not o er the assay. Direct uorescent antibody testing and ow cytometric assay for platelet surface IgG and IgM have been described in horses with IMTP.117,298 Direct and indirect immunoradiometric assays and an indirect enzyme-linked immunosorbent assay (ELISA) for detection of platelet-bound antibody also have been described.299,300 These tests are not widely available, however, and require careful control and interpretation. In the absence of an easy, accurate, and reliable method for de nitive diagnosis of IMTP in the horse, one may make a presumptive diagnosis in horses with a low platelet count, normal coagulation parameters (PT and PTT), and no evidence of excessive consumption of platelets (i.e., DIC). Any current medication being administered to horses suspected of having IMTP should be discontinued immediately. If life-threatening hemorrhage is occurring, then transfusions of fresh whole blood or platelet-rich plasma are indicated. Most horses respond to therapy with parenterally administered corticosteroids.281,285 Corticosteroids appear to act by suppressing the phagocytic activity of mononuclear phagocytes, inhibiting antibody synthesis, increasing e ective platelet production, and reducing capillary fragility. One should tailor the corticosteroid dose and route of administration to the individual patient to achieve the best results with the lowest possible dose. Horses with IMTP have been reported to respond to dexamethasone at 0.1 mg/kg intravenously or intramuscularly twice daily.282,285 One always should

taper corticosteroid administration gradually to once-daily morning administration to decrease interference with endogenous corticosteroid release. If long-term maintenance therapy is necessary, every-other-day or every-third-day administration of the lowest e ective dose is indicated. Maintenance therapy may be oral or parenteral. Potential complications associated with corticosteroid administration in the horse include laminitis, secondary infections, and iatrogenic hyperadrenocorticism. Occasionally, horses with IMTP fail to respond to parenteral corticosteroid therapy. Two such horses responded favorably to azathioprine at 3.0 mg/kg orally once daily.286 However, a recent study on the pharmaco*kinetic properties of azathioprine indicates that the drug has poor bioavailability in the horse when administered orally at the previously recommended dose and that clinical trials for e cacy are needed.301 Azathioprine has been used in human beings and dogs with IMTP and acts as an immunosuppressive agent interfering with humoral and cell-mediated immune function. Vincristine is a vinca alkaloid used to treat refractory IMTP in human beings. Vincristine therapy was successful in treating one horse with idiopathic IMTP302 and unsuccessful in treating another.286 Alternative treatments for IMTP that have been e ective in other species include danazol (a synthetic analog of androgenic steroids and progesterone), gamma globulin infusions, and splenectomy.303 Neonatal alloimmune thrombocytopenia has been described in horse and mule foals.299,300,304 A syndrome of presumptive neonatal alloimmune thrombocytopenia, mucosal erosions, and skin lesions has been described in foals of several breeds. These foals respond to supportive care with judicious administration of corticosteroids and transfusions of platelet-rich plasma. Long-term prognosis is good, but platelet counts can take 30 days or longer to stabilize at more than 10,000 per microliter. Equine Infectious Anemia Virus Acute febrile episodes of EIAV are accompanied by a sharp decline in platelet count followed by a rapid rebound as fever and viremia resolve. This thrombocytopenia is probably multifactorial in origin. Circulating platelets have increased quantities of IgG and IgM on the surface117 and show evidence of in vivo activation, possibly indicating enhanced nonimmune-mediated destruction of platelets.118 Bone marrow megakaryocytes are not infected during acute disease117; however, evidence exists of decreased bone marrow production of platelets in horses with EIA, possibly because of altered cytokine production in the bone marrow.119,305,306 The thrombocytopenia of EIAV may contribute to petechia and ecchymoses but rarely is associated with a severe bleeding diathesis. EIAV is discussed in detail in previous sections. Equine Granulocytic Ehrlichiosis Infection with the obligate, intracellular, rickettsia Anaplasma phagocytophilum (formerly Ehrlichia equi) can result in a syndrome of fever, anorexia, depression, petechia and ecchymoses, icterus, ventral edema, and ataxia lasting 3 to 16 days. Involuntary recumbency with seizurelike activity was observed in a horse diagnosed with anaplasmosis.307 Consistent hematologic abnormalities regardless of clinical presentation include thrombocytopenia, leukopenia, and mild anemia.308,309 Clinical signs are more severe in adult horses than in those younger than 4 years of age. Foals younger than 1 year of age often experience only fever.309,310 Most infected horses originate from the foothills of northern California, but infections also have been diagnosed in horses from other parts of the country.310,311 Most cases of ehrlichiosis occur in late fall, winter, and spring, and the disease is known to be transmitted by several species of the Ixodes spp. tick. Tick control is recommended for prevention of disease in endemic areas. Granular inclusion bodies may be visible in the cytoplasm of neutrophils and eosinophils during routine cytologic examination using any Wright’s or Giemsa type of stain. One may see inclusion bodies best under oil immersion, and they appear as pleomorphic, blue-gray to dark-blue spoke-wheel shapes.309 Inclusion bodies represent a cluster of coccobacillary organisms, varying in size between 0.2 to 5 m in diameter, within cytoplasmic membrane-bound vacuoles. The appearance of cytoplasmic inclusion bodies correlates closely with the onset of fever, and they remain visible for approximately 10 days.308 Serologic diagnosis using the IFA test has historically been the standard method for diagnosis of equine anaplasmosis.312 However, as many as 50% of normal horses in endemic areas of California have antibodies against the organism.313 A polymerase chain reaction (PCR) test to detect the organism’s DNA o ers a means to con rm a diagnosis early in the course of the disease with comparatively high sensitivity and speci city. Animals that are experimentally infected become PCR positive on approximately the fth day after infection but do not show clinical signs until day 7. Seroconversion occurs quickly after clinical signs become apparent at approximately days 12 to 16. Extrapolating the results of this study to clinical situations suggests that PCR is likely to be positive before onset of clinical signs and that IFA should be performed early with suspicion of disease so that seroconversion can be accurately assessed.312 Mortality is rare in horses with ehrlichiosis; most untreated animals recover over 2 weeks and acquire a sound immunity against reinfection for at least 2 years.308,309 Death, when it occurs, is usually associated with secondary infections that are the result of immunosuppression or trauma because of ataxia.314 One experimentally infected horse died with signs consistent with systemic in ammatory response syndrome, shock, and DIC.315 Treatment with oxytetracycline at 7 mg/kg intravenously once or twice daily for up to 7 days may hasten recovery. One should provide supportive therapy, including intravenous uids, leg wraps, and stall con nement for the severely ataxic horse, as indicated.308,309,311 In animals that die or are killed, necropsy lesions include petechial and ecchymotic hemorrhages, edema, and icterus.

Histologically, small arteries and veins are in amed, and mild in ammatory vascular or interstitial lesions may be present in kidneys, heart, brain, and lungs.308 Increased Platelet Use Rapid use of circulating platelets occasionally occurs and may be appropriate, as in the case of massive trauma or external hemorrhage, or inappropriate, as in some cases of systemic or localized activation of hemostatic mechanisms. Thrombocytopenia accompanying hemorrhage or trauma is usually mild to moderate and rapidly reversible. DIC is a complex disorder of hemostasis resulting from widespread systemic activation of coagulation mechanisms. This microvascular coagulation rapidly uses circulating platelets, and thrombocytopenia is a common result. A ected animals have a variety of other abnormalities detectable by routine coagulation testing. Horses with DIC and normal platelet counts may have abnormal platelet function, inhibiting the use of those platelets. In one study of horses presented to a veterinary referral hospital, DIC after gastrointestinal or in ammatory diseases was the most common cause of thrombocytopenia.289 In horses evaluated for colitis, approximately half of the horses had platelet counts less than 100,000,316 and in horses that underwent surgery for correction of a large colon torsion, a decreased platelet count was signi cantly associated with increased mortality.317,318 The section on DIC provides more information on the pathophysiology, diagnosis, and treatment of this disorder. Localized activation of coagulation may occur in some disease conditions, including neoplasia and renal disease. Thrombocytopenia has been reported accompanying hemangiosarcoma in the horse319-321 and is associated with vascular tumors in human beings and dogs.322,323 Hemolytic uremic syndrome is an unusual condition in the horse characterized by acute renal failure with microangiopathic intravascular hemolysis and disseminated or renal intravascular coagulation. Thrombocytopenia is a common finding in affected horses.73,74

Thrombocytosis Thrombocytosis is an increase in the number of circulating platelets. Increased bone marrow production of megakaryocytes may occur as a primary myeloproliferative disorder or in association with other neoplastic conditions, including polycythemia vera. Thrombocytosis more commonly is associated with acute or chronic in ammatory disorders or follows acute hemorrhage.324 Mild thrombocytosis also may occur during and immediately after exercise or excitement caused by splenic contraction and release of sequestered platelets into the peripheral circulation. Thrombocytosis is usually asymptomatic, and speci c therapy is not indicated. Thrombocytosis rarely may be associated with venous thrombotic tendencies, and one may consider anticoagulation therapy in these cases.

Functional Defects of Platelets Functional defects of platelets can result in clinical signs similar to those of thrombocytopenic patients, despite a normal to increased platelet count. The most widely used laboratory tests to assess platelet function are bleeding time and platelet aggregation studies. Bleeding time depends on the number and function of platelets, the level of vWF, and vascular integrity. One performs the test by inverting the lower lip and making a vertical incision 1 mm deep by 5 mm long, with a template bleeding time device (Surgicutt, International Technidyne Corp., Edison, N.J.). After 30 seconds the blood is removed with Whatman lter paper. Then the previous step is repeated every 15 seconds until one no longer detects blood on the lter paper, and the total time is recorded.325 Alternatively, one may make an incision on the caudolateral aspect of the foreleg with proximal venostasis using a sphygmomanometer cu above the carpus.326,327 The site should be shaved before testing. Normal bleeding time varies, depending on the technique and site of incision, and one must establish the time for each technique. The veterinarian should evaluate a normal control horse concurrently with the patient. A platelet aggregometer can be used to measure the aggregation of platelets in response to ADP, serotonin, epinephrine, arachidonic acid, thrombin, ristocetin, collagen, and collagen-related peptide.328-330 One should collect whole blood samples with sodium citrate as an anticoagulant and prepare platelet-rich plasma by centrifugation. Congenital defects in platelet function are rare but have been identi ed in horses. Glanzmann thrombasthenia is a rare, heritable platelet disorder that has been thoroughly described in human beings and dogs. Four horses including a Standardbred,331 an Oldenburg,332 a Quarter Horse, and a Thoroughbred cross333 have been diagnosed with the condition. The platelet dysfunction was due to an amino acid change in the gene for the platelet brinogen receptor, integrin α IIb β 3.334 Horses present for clinical bleeding, most commonly in the form of epistaxis. Platelet count and coagulation times are normal, whereas mucosal bleeding time is signi cantly prolonged and clot retraction is markedly reduced. Results of aggregometry testing are consistent with an inability of the platelets to bind to brinogen; ow cytometry reveals decreased surface integrin αIIb β 3.332,333 Treatment was attempted in these horses with tapering doses of dexamethasone with no apparent change in the occasional bleeding and laboratory abnormalities. Another heritable thrombasthenia caused by a platelet brinogen-binding defect has been identi ed in a population of Thoroughbreds.330,335,336 The clinical condition was identi ed in a mare and its o spring, prompting a survey of a population of Thoroughbreds in that area. The syndrome observed in the Thoroughbred mare and its lly resembled Glanzmann thrombasthenia in clinical presentation and basic laboratory analysis. However, aggregometry and ow cytometry results di er in that platelets of the a ected horses bind in reduced amounts to brinogen in the presence of thrombin.336 Aggregometry often demonstrates the inability of the platelets to bind

in the presence of collagen.330 Although the genetic basis of this disorder remains unclear, a secretion defect is believed to be related to either abnormal platelet granule contents or intracellular signaling. The population study reported an incidence of 0.7% (3/444 horses) of this disorder, yet none of the horses in the screening program exhibited clinical signs. When appropriate, ow cytometry should be used to screen for this heritable disease.335 Researchers have described von Willebrand’s disease in the horse.337 Because the disease is not an inherent platelet defect, it is discussed later in this chapter (with hereditary defects of coagulation). Acquired platelet functional defects may be associated with uremia, acute myelodysplastic diseases, and dysproteinemias, as well as after the administration of various drugs. Abnormal responses to platelet aggregating agents and increased bleeding time have been described in normal human and equine neonates,325,329 and neonatal foals may have an increased susceptibility to platelet-associated hemorrhagic diathesis. Many human beings and animals with uremia exhibit abnormal hemostasis, most of which is attributable to abnormal platelet function and likely is caused by abnormalities in the biochemical processes necessary for platelet aggregation and granule release. Human beings with myeloproliferative and lymphoproliferative disorders may have abnormal platelet shapes and decreased aggregation and secretion.338 Dysproteinemia in these patients may result in platelet dysfunction when abnormal proteins coat platelets or endothelium and interfere with normal adhesion and aggregation. A hemorrhagic diathesis caused by platelet dysfunction has been described in one horse with monoclonal gammopathy and lymphoproliferative disease.271 Drugs used in equine medicine and associated with abnormalities of platelet function in human beings include penicillins, cephalosporins, quinidine, chlorpromazine, halothane, and NSAIDs.338 NSAIDs act by inhibiting cyclooxygenase, an enzyme critical for the formation of prostaglandins, prostacyclins, and thromboxane from arachidonic acid. As previously discussed, TXA2 is a strong stimulus for platelet aggregation and release reactions and is a potent vasoconstrictor. Aspirin covalently acetylates cyclooxygenase, and because platelets are anucleate cells, they are incapable of producing new enzyme. The result is an irreversible inhibition of platelet cyclooxygenase that can be corrected only when the body clears aspirin from the circulation and the bone marrow releases new platelets. In contrast, other NSAIDs produce only a temporary inhibition of cyclooxygenase by reversibly binding to the enzyme. Template bleeding time is prolonged in horses that have received aspirin, phenylbutazone, or unixin meglumine.327,328,339,340 The e ect is most pronounced after aspirin therapy, and a dosage of 17 mg/kg once daily for 3 days increases bleeding time for at least 3 days after discontinuation of therapy.327 The minimal effective dose of aspirin to decrease thromboxane generation is 5 mg/kg; the duration of decrease is dose dependent.341

HEMOSTASIS The vascular system provides a closed conduit for the circulation of uid and cellular components of blood. Exchange of substances across the vessel surfaces is controlled carefully. The body must repair any disruption in the closed network of blood vessels rapidly and e ciently to prevent excessive loss of critical blood elements into the extravascular environment. This repair process requires interaction between the injured vessel wall, platelets, and soluble proteins and accessory molecules in the blood. The end result is a stable brin mesh that prevents loss of blood while the vascular endothelium regenerates. When endothelium has been repaired, brinolytic reactions occur to remove brin and re-establish vessel patency. A careful balance of coagulative and brinolytic processes must occur to maintain optimal circulatory capacity without loss of blood. The body also must control procoagulative forces carefully to prevent inappropriate activation that might result in thrombosis of normal vessels with subsequent tissue and organ ischemia.

Physiology Primary hemostasis is the process of platelet interaction with the vessel wall to achieve a temporary plug of the vascular defect. Secondary hemostasis involves the interaction of soluble coagulation factors to produce a stable brin mesh that reinforces the platelet plug. The body carefully regulates hemostatic mechanisms via an intricate network of positive and negative feedback with inhibitors and potentiators of coagulation and fibrinolysis. The initial vascular response to injury is intense vasoconstriction to divert blood from the site. Anticoagulative processes associated with vascular endothelium, including synthesis of prostacyclin and plasminogen activators and uptake and degradation of proaggregating molecules,13 prevent the interaction of normal intact vascular endothelium with circulating platelets. Exposure of subendothelial collagen at the site of vascular injury triggers platelet adherence with interaction between endothelial vWF and platelet receptors. The function of platelets in primary hemostasis has been discussed in detail already. Platelet adhesion stimulates a release reaction involving platelet secretory granules. Contents of these granules (including ADP, ATP, serotonin, platelet factors 3 and 4, TXA2, and acid hydrolases) attract other platelets in a process of irreversible platelet aggregation at the site of vascular injury. Platelet aggregation provides a favorable environment for accumulation and activation of the intrinsic and extrinsic coagulation systems. The classic coagulation pathway has traditionally o ered a concept of hemostasis based on and under the control of protein coagulation factors. The system involves sequential activation of inactive precursor factors (Table 14-3) by limited proteolysis (Figure 14-2). Activated factors are indicated by the letter a after their numeric designation. Most coagulation factors are classi ed as serine proteases because of a critical serine residue at the active site of the enzyme. This serine is exposed after cleavage of a short peptide during activation of factor precursors. Table 14-3 Coagulation Factors Factor I II III IV V VI VII VIII IX X XI XII XIII Prekallikrein High-molecular-weight kininogen

Synonym Fibrinogen Prothrombin Tissue factor, tissue thromboplastin Calcium Proaccelerin, labile factor Not assigned Proconvertin, stable factor Antihemophilic factor Christmas factor Stuart-Prower factor Plasma thromboplastin antecedent Hageman factor Fibrin-stabilizing factor Fletcher factor Fitzgerald factor

FIGURE 14-2

Schematic representation of blood coagulation showing the numerous interactions between coagulation factors in the classic extrinsic and intrinsic pathways. Arrow, Primary action; dashed arrow, actions of secondary importance; a, activated form; HMWK, high-molecular-weight kininogen; TF, tissue factor; PL, phospholipids; Prekal, prekallikrein. (From Morris DD: Recognition and management of disseminated intravascular coagulation in horses, Vet Clin North Am Equine Pract 4:115-143, 1988.)

Although commonly discussed as separate pathways, intrinsic and extrinsic coagulation cascades interact at many levels and are considered more accurately as part of one highly complex system. For the sake of simplicity, however, they are discussed separately. The liver produces most soluble coagulation factors. Mononuclear phagocytes are responsible for clearing activated clotting factors from the circulation. Contact of soluble coagulation factors with negatively charged subendothelial collagen and with the various products of the platelet release reaction initiates the intrinsic coagulation system. This initial contact phase (see Table 14-3) requires interaction of coagulation factor XII, high-molecular-weight kininogen (HMWK), and prekallikrein to accelerate production of the activated form of factor XII (designated XIIa). Factor XIIa in turn catalyzes the limited proteolysis of factor XI to XIa. Elements of the contact phase reaction are also capable of inducing parallel pro-in ammatory reactions by initiating kinin and complement cascades. The importance of this contact phase reaction to in vivo coagulation is questionable. Human beings and animals with hereditary de ciencies of the proteins involved in this reaction (factor XII, HMWK, or prekallikrein) rarely show any clinical bleeding tendencies. In contrast, de ciency of almost any other coagulation factor, including factor XI, results in profound hemorrhagic tendencies. Factor XIa, in the presence of calcium, activates factor IX. Factor IXa forms a molecular complex with factor VIIIa and calcium on a phospholipid membrane surface to catalyze the conversion of factor X to Xa. Platelet membranes are thought to provide the phospholipid surface for this reaction. Activation of factor X is the point of interaction between the extrinsic and intrinsic systems. The extrinsic coagulation system begins with formation of a molecular complex between tissue factor (TF; thromboplastin), factor VIIa, and calcium ions on a platelet phospholipid surface to catalyze the conversion of factor X to Xa. Factor Xa is capable of activating factor VII to VIIa in a feedback loop that increases the production of both. In a slower reaction that appears to be important in vivo, the TF VIIacalcium complex is also capable of activating factor IX.342 TF is present in virtually every tissue of the body, and a variety of insults can initiate the extrinsic coagulation system, including vascular endothelial damage, endotoxin, antigen-antibody complexes, RBC hemolysis, and local areas of tissue necrosis. Formation of factor Xa marks the beginning of what classically is termed the common pathway of coagulation. Factor Xa complexes with factor Va and calcium on a phospholipid membrane to convert prothrombin to thrombin enzymatically. Thrombin in turn is the catalyst for formation of brin from brinogen. Thrombin is also important in the activation of factors V, VIII:C, and XIII, forming positive feedback loops that potentiate the coagulative process and promote platelet aggregation. Fibrin monomers aggregate, and factor XIIIa catalyzes, the formation of covalent cross-linkages between the monomers to stabilize the clot. The cascade concept of the physiology of hemostasis has recently been revisited because several disease processes cannot be adequately explained. Many experts now believe that coagulation should be viewed as three overlapping steps termed initiation, ampli cation, and

propagation.343 This series of events is dependent on two cells: the TF-bearing cell located extravascularly and the platelet. In a normal physiologic state, coagulation is prevented until necessary by limiting contact of these two cell types. Once an injury to a vessel has occurred, the two come into contact and the process of coagulation begins. Initiation begins as a vessel is damaged, plasma is exuded, and it comes in contact with the TF-bearing extravascular cells. Factor VIIa binds to TF and after activation by several proteases forms a factor VIIa-TF complex. This complex in turn activates factors X and IX. Factor Xa goes on to activate factor Va, and together this prothrombinase complex, located on the surface of TF-bearing cells, forms small amounts of thrombin. If factor Xa leaves the cell surface, then it is quickly inhibited. In contrast, factor IXa can move from cell to cell without rapid inhibition. The ampli cation phase also involves the aforementioned damage to the vessel that is signi cant enough to allow exudation of platelets and factorVIII-vWF into the extravascular space. Platelets adhere to the extravascular matrix and interact with the small amount of thrombin that was produced on the surface of the TF-bearing cells. This thrombin enhances platelet adhesion, activates other platelets, and activates factors Va and VIIIa on the platelet surface. This activation allows vWF to be released, further enhancing platelet adhesion and aggregation. These activated platelets with factors Va and VIIIa bound to their surfaces are the basis for the propagation phase. Propagation involves the formation of two additional complexes. Factor IXa that was activated by the factor VII-TF complex during the initiation phase, binds to activated platelets. Factor IXa then forms a complex (tenase complex) with factor VIIIa already on the platelet cell surface. This complex activates factor Xa that is present on the platelet surface. Factor Xa then forms a complex with already platelet bound factor Va (prothrombinase complex) that causes a marked increase in thrombin formation (enough to bind fibrinogen and form a clot).343 Inhibitors of procoagulative elements in part maintain a balance between clot formation at sites of vascular injury and pathologic thrombosis. Several physiologically important systems for in vivo procoagulant inhibition have been described. The most important of these is plasma antithrombin III (AT III), responsible for 50% to 75% of plasma antithrombin activity. AT III inhibits all serine proteases generated during the coagulation process, including thrombin; factors Xa, IXa, XIa, and XIIa; plasmin; kallikrein; and the anticoagulants protein C and protein S. Heparin interacts with AT III, causing a conformational change that increases the activity of AT III 1000-fold.344 Tissue factor pathway inhibitor (TFPI) is a protein that reversibly inhibits factor Xa and then as a complex with Xa, can inhibit the factor VII-TF complex at the level of the endothelial cell. Another of the main anticoagulative systems involves protein C, protein S, and thrombomodulin. As the primary clot is formed, small pieces of thrombin are lost within the vasculature. When this thrombin comes in contact with an intact endothelial cell, it binds to thrombomodulin, which in turn activates proteins C and S. These vitamin K–dependent serine proteases act to inactivate factors Va and VIIIa and prevent inappropriate clotting. This system acts mainly on endothelial cells rather than platelets.345 The brinolytic system is responsible for eventual remodeling and removal of a stable brin clot to restore blood ow (Figure 14-3). Because small amounts of brin production and subsequent brinolysis occur continuously in vivo, the body must maintain a delicate balance between procoagulant and brinolytic processes to prevent widespread thrombosis or hemorrhage. The principal enzyme of the brinolytic system is plasmin, normally present in plasma as an inactive precursor, plasminogen. Plasminogen is converted to plasmin proteolytically by endothelial tissue plasminogen activator, an enzyme present in most normal and neoplastic tissue. Endothelial cells adjacent to clot formations release tissue plasminogen activator in response to vascular stasis and a high local concentration of thrombin. Released tissue plasminogen activator has a high a nity for brin, localizing it to the adjacent clot for plasminogen activation. Circulating antiplasmins, the most important of which is α2-antiplasmin, rapidly inactivate circulating plasmin. Plasmin bound to brin is much less accessible for inactivation because plasmin interactions with fibrin obscure binding sites for inhibitors.346

FIGURE 14-3

Major brinolytic reactions (see text for details). A negative sign (−) indicates inhibition of a speci c reaction. TPA, Tissue plasminogen activator; PAI, plasminogen activator inhibitor. (From Kobluk CN, Ames TR, Geor RJ: The horse: diseases and clinical management, Philadelphia, 1995, WB Saunders.)

Plasminogen also may be activated during the contact phase of the intrinsic coagulation system. Factor XIIa and HMWK interact with prekallikrein to produce kallikrein. Kallikrein activates urokinase, which in turn converts plasminogen to plasmin.347 Because plasmin can activate factor XII, a potential feedback loop ampli cation exists. The importance of contact phase activation in vivo is questionable, however. Plasmin is a serine protease with a high a nity for brinogen and brin, degrading them into brin degradation products (FDPs). Plasmin also may aid in degradation of factors V, VIII, IX, and XI. Fibrin and brinogen are split into the clinically signi cant FDPs including D-dimers, as well as fragments X,Y,O, and E. Increases in the circulating concentrations of these fragments can be attributed to ongoing fibrinolysis. The mononuclear phagocytes of the liver clear FDPs from the circulation.

Evaluation of Hemostasis Disorders of hemostasis usually may be classi ed based on clinical signs as abnormalities of primary or secondary hemostasis. Abnormalities of primary hemostasis include vascular diseases and changes in the number or function of platelets. A ected horses usually have signs of mucosal bleeding, petechia and ecchymoses, epistaxis, hyphema, or melena. Platelet function tests were discussed previously. Diagnosis of the vasculitides is discussed in another section. Disorders of secondary hemostasis usually involve abnormalities in quantity or function of the soluble coagulation factors and present as spontaneous or excessive hemorrhage in response to surgery or trauma. These horses may have hemorrhage into body cavities, hematoma formation, hemarthroses, prolonged bleeding after venipuncture, or external hemorrhage. Most coagulation tests measure the time necessary for clot formation under various circ*mstances. Coagulation times tend to be longer in horses than in human beings or other domestic animals.348 Table 14-4 lists some laboratory tests used to assess equine hemostatic function with normal reference values. Table 14-4 Reference Ranges for Equine Hemostatic Values∗

Activated clotting time measures the time to clot formation on activation of whole blood by contact with diatomaceous earth. This process bypasses the contact phase of coagulation and is useful in evaluating de ciencies of factors VIII and IX, prothrombin, and brinogen. Abnormalities of factor VII and platelets do not appear to alter the activated clotting time.348 One collects blood directly into a tube containing diatomaceous earth, mixes, incubates the contents at 37° or 38° C, and measures the time until clot formation. PTT and activated partial thromboplastin time (APTT) measure the activity of factors XII, XI, X, IX, VIII, and V, prothrombin, and brinogen. The PTT test is performed by adding platelet-poor plasma to a glass tube containing phospholipid emulsion and calcium and determining the time to clot formation. One performs the APTT test similarly, except an activating agent is added. PT, also known as onestage PT, measures the function of the extrinsic and common coagulation pathways, including factors V, VII, and X, prothrombin, and brinogen. In this test, platelet-poor plasma is mixed with thromboplastin and calcium and the time to clot formation is determined. One may use citrated whole blood or plasma for accurate measurement of PT up to 3 days after collection. A control sample from a clinically normal horse should accompany the patient’s sample.349 Commercially available tests for measurement of PT have been evaluated in horses and found to be accurate when the manufacturer’s guidelines were followed and the appropriate reagent was used.350 In horses with colic, APTT is more reliably elevated and more markedly changed than any of the other routine coagulation tests.351,352 A prolonged APTT at admission is associated with a negative prognosis.353 Thrombin time (TT) measures the time to clot formation after addition of thrombin to citrated plasma. TT is only prolonged with abnormalities in fibrinogen quantity or function or in the presence of thrombin inhibitors. Most commonly, FDPs are measured in the serum using antibody-coated latex beads and monitoring macroscopic agglutination. This test re ects only the presence of D and E fragments. One must collect blood into a special tube containing thrombin and a protease inhibitor (ThromboWellcotest; Burroughs Wellcome Co., Research Triangle Park, N.C.). Commercially available latex agglutination kits for FDPs are not suitable for use in the horse.353 However, D-dimers, a brinogen degradation product, can be measured reliably in horses using latex agglutination kits. An increase in D-dimers occurs in horses with colic,317,352 laminitis,354 jugular vein thrombosis, and DIC355 and may indicate a risk factor for development of a coagulopathy. Fibrinogen concentration may be easily estimated by refractometer as the di erence between plasma protein concentration at room temperature and the concentration after heating plasma to 56° to 58° C for 3 minutes and centrifugation to remove precipitated brinogen. Alternatively, brinogen may be estimated from the thrombin clot time, which is inversely proportional to the brinogen concentration, or by immunologic assays. Tests for plasma AT III measure the presence or the activity of the protein. Because inactive AT III antigen may be present, functional tests are preferable. Assays of AT III activity are clotting or chromogenic tests that measure residual thrombin activity after one adds the patient sample to a known quantity of thrombin. Normal AT III activity in the horse has been reported as 218% of normal human pooled plasma356 and 63% to 131% of normal equine pooled plasma.357,358 Widespread in vivo activation of coagulation is the most common cause of decreased plasma AT III activity. This is reliably observed in horses with gastrointestinal disease317 and is a negative prognostic indicator for some populations of horses.359,360 Decreased activity has been associated with protein-losing enteropathies or nephropathies and liver disease. Decreased AT III activity has been attributed to consumption of AT III, but the decrease may actually be the result of the loss of albumin or decreased production as seen with liver disease.361 AT III’s role as an acute phase protein in the horse is controversial.351,352 Increased AT III activity has rarely been reported in horses with hepatic disease.358 Specialized assays for various components of the equine procoagulative and brinolytic systems have been described and are available

only at specialized laboratories. These include assays for thrombin-antithrombin complexes (TAT),317 protein C,362 plasminogen,363α2antitrypsin,364 vWF,337 and factor VIII:C.365 Mean platelet component (MPC) is a measure of platelet density that reliably decreases when platelets become activated and degranulate. This test was found to be an indicator of platelet activation in both adult horses and ill neonates.366

Hereditary Disorders of Hemostasis Inherited disorders of hemostasis are most common in purebred animals that have been highly inbred. Many of these disorders are inherited in an autosomal recessive fashion; both parents must be carriers of the disorder to produce clinically a ected o spring. For practitioners to be able to di erentiate between inherited and acquired disorders of hemostasis is essential to advise owners appropriately. Inherited disorders are potentially treatable, but they are incurable. Owners should not use a ected animals and their carrier parents as breeding stock. Practitioners see most animals with inherited bleeding problems when they are still young. Owners frequently complain of inappropriate bleeding in response to trauma or surgery or apparently spontaneous external hemorrhage or hematoma formation. As with any inherited defect, one should obtain a careful history and examine the patient and any close relatives, including siblings and parents, if possible. Most inherited defects result from a single-factor abnormality. Routine coagulation studies may narrow the di erential diagnoses for these bleeding disorders, but confirmation of diagnosis usually requires specialized tests. In contrast, most acquired defects of hemostasis are characterized by multiple-factor abnormalities, frequently with a severe underlying systemic disorder.

VON WILLEBRAND’S DISEASE von Willebrand’s factor is a glycoprotein of plasma, platelets, and endothelium required for platelet adhesion to exposed subendothelium and subsequent normal platelet plug formation. In plasma, vWF is bound in a multimolecular complex to coagulation factor VIII. Quantitative or qualitative abnormalities of vWF result in spontaneous bleeding from mucosal surfaces and excessive bleeding after surgery or trauma. Researchers have reported von Willebrand’s disease in a Quarter Horse lly337 that presented for episodes of oral bleeding, conjunctival hemorrhage, and prolonged bleeding at injection sites, as well as in a Thoroughbred mare and foal that presented for episodes of epistaxis. Routine complete blood count (CBC), including platelet count and coagulation times were within normal limits. Another 8-dayold Quarter Horse foal had a fever and numerous focal swellings at various sites around its body. This foal was persistently anemic and hypoproteinemic despite repeated transfusions and at necropsy was found to have signi cant hemarthrosis and secondary infection.367 The ELISA for vWF is a sensitive and rapid means for diagnosis of this de ciency. Ristocetin, a co-factor involved in platelet aggregation, is de cient in human beings su ering from type II vWD. This speci c de ciency was also identi ed in the aforementioned related Thoroughbreds.368 This condition is heritable in human beings and dogs, and reports indicate the same is true in horses. Owners should not use affected animals for breeding.

FACTOR VIII DEFICIENCY Classic hemophilia (factor VIII de ciency, hemophilia A) is the most commonly reported inherited defect of hemostasis in the horse. Hemophilia has been diagnosed in Thoroughbred, Standardbred, Quarter Horse, and Arabian colts365,369-376 with severe spontaneous hemorrhage or excessive hemorrhage related to surgery or mild trauma. Coagulation testing reveals abnormal function of the intrinsic coagulation system, with a prolonged APTT and a normal PT. Diagnosis is con rmed by measuring factor VIII:C activity in plasma. In human beings the severity of clinical signs varies inversely with this value, and the same appears to be true for a ected horses. In severely a ected horses, factor VIII activity is less than 10% of normal equine plasma.365,371,373 A less severely a ected colt that survived castration and was undiagnosed until 3 years of age had a factor VIII activity of 20% to 30%.375 As in human, canine, and feline hemophilia, equine factor VIII de ciency is transmitted as an X-linked recessive trait.370 Carrier mares transmit the defect to 50% of male o spring on average. Carrier female horses usually have a factor VIII:C activity of approximately 50% of normal, although this value may vary widely and even overlap normal values. No cure exists for hemophilia, and the disease is progressively debilitating and potentially lethal. A ected human beings are treated with repeated transfusions of fresh plasma and factor concentrates. Factor IX and XI de ciencies have been reported in one Arabian foal with concurrent factor VIII deficiency.372

COAGULATION FACTOR ABNORMALITIES In the classic pathway, initiation of the intrinsic coagulation system involves interaction of factor XII, prekallikrein, and HMWK ultimately to activate factor XI. This process usually begins with contact between factor XII and negatively charged molecules such as subendothelium or collagen. Prekallikrein de ciency has been diagnosed in families of Belgian horses377 and American Miniature Horses.378 Plasma prekallikrein de ciency is characterized by prolonged APTT with a normal PT. A ected horses may or may not show clinical signs of a bleeding tendency.377,378 Con rmation of diagnosis is by demonstrating low prekallikrein activity with normal activity of other intrinsic coagulation factors. One also can correct the prolonged APTT by extended incubation with a contact activator.377,378 In human beings,

prekallikrein de ciency is inherited as an autosomal recessive or autosomal dominant trait.379 The mode of inheritance in horses has not been con rmed, but appearance of the defect in multiple members of some equine families strongly suggests that it is an inherited disorder in this species as well. A similar defect in the contact phase of coagulation, not thought to result from prekallikrein de ciency, was diagnosed in one Quarter Horse gelding with prolonged APTT and normal PT.380 Researchers did not observe a clinical bleeding problem in association with this defect.

Acquired Disorders of Hemostasis VASCULITIS Vasculitis, in ammation of blood vessels, may occur as a primary disease process but more commonly is encountered as a secondary complication of infectious, immunologic, toxic, or neoplastic disorders.381 Hypersensitivity vasculitides are characterized by predominant involvement of small blood vessels in the skin. In most cases the postcapillary venules are involved. The most common in ammatory pattern is leukocytoclasis, de ned by the presence of neutrophilic nuclear debris in and around the involved vessels. Vessel wall necrosis and fibrinoid changes occur. Clinical signs depend on the type of vessel a ected and the severity of the in ammatory response. Vascular occlusion may result in tissue ischemia and necrosis. Increased vascular permeability results in hemorrhage and edema as uid and cellular constituents of the blood escape into the extravascular space. In horses, vasculitis is characterized by well-demarcated areas of cutaneous edema that may involve any portion of the body.382 The distal extremities and ventral body wall are a ected most commonly, and the edematous area is often hot and painful. A ected horses generally are depressed and reluctant to move. Hyperemia, petechial and ecchymotic hemorrhages, and ulcerations are commonly present on mucous membranes of the mouth, nose, and vulva. Other clinical signs may re ect edema, hemorrhage, and infarction in other body systems. Lameness, colic, diarrhea, dyspnea, or ataxia can result from involvement of vessels in joints, muscles, gastrointestinal tract, respiratory tract, and central nervous system. Subclinical renal disease may occur. Secondary complications such as laminitis, thrombophlebitis, and localized infections are common. Affected skin often weeps serum and eventually sloughs. The underlying disease, length of illness, other organ involvement, and secondary complications usually determine hematologic and serum biochemical ndings in vasculitis; none are characteristic. Neutrophilia, mild anemia, hyperglobulinemia, and hyper brinogenemia usually accompany chronic in ammation. The platelet count is often normal to increased but can be low because of a concomitant consumptive coagulopathy or IMTP. With renal involvement, serum creatinine may be elevated or urinalysis may reveal hematuria or proteinuria or both.

INFECTIOUS DISEASES Infectious diseases associated with vasculitis and subsequent petechiation, ecchymoses, and dependent edema include EIAV, equine granulocytic ehrlichiosis, and equine viral arteritis (EVA). All three diseases may result in thrombocytopenia, although this is somewhat less common with EVA. The first two are discussed in detail elsewhere in this chapter; a discussion of EVA follows. An arterivirus of the family Arteriviridae causes EVA. The disease can be characterized by fever, chemosis, lacrimation, rhinitis, focal dermatitis, ventral edema, and occasionally abortion.383,384 However, many infected horses do not show recognizable clinical signs. Horses become infected by aerosol or venereal contact with an infected animal. The virus replicates in the intima and media of most arteries, resulting in an in ammatory reaction and cellular in ux.385 Although mares and geldings generally clear the disease, stallions may become asymptomatic carriers of the disease, harboring virus in the accessory sex glands383,386 and transmitting infection to previously unexposed mares. Infected mares usually make an uneventful recovery from EVA and eliminate the virus from the body. Abortion may occur during or shortly after acute illness or asymptomatic infection as a result of myometrial necrosis and edema leading to placental separation and fetal death.387 Not all infected mares abort, however, and the incidence of abortion may be strain dependent.388 One con rms diagnosis of EVA in mares by serum neutralization assay, demonstrating a fourfold or greater rise in titer between acute and convalescent serum samples. EVA titer may remain elevated for a long time after natural infection, and recovered mares are considered immune to further infection. Diagnosis of persistently infected, asymptomatic, nonvaccinated stallions depends on a positive serum neutralization test of greater than 1:4, with either isolation of the virus from sem*n samples or demonstration of seroconversion of an otherwise unexposed mare within 2 weeks of breeding to the suspect stallion.389 Diagnosis of abortion caused by EVA requires virus isolation from the fetus or the placenta. Recent seroconversion in the mare is highly supportive of a diagnosis of EVA abortion. A modi ed live virus vaccine is available but should be reserved for use only in breeding animals.388 It should be administered annually to mares while open and to stallions or teasers 3 to 4 weeks before the breeding season.390 Vaccinates should be isolated for 21 days after vaccination because of virus shedding during this time frame. Infections caused by the vaccinate strain cannot be distinguished from natural

infection. Most breeding facilities require a negative serologic result before acceptance of sem*n, stallions, or broodmares. Any breeding animal that may be potentially exported must have a negative serologic test followed by an appropriate vaccination record before export.391 Vaccine records should be carefully managed as to provide protection and allow accurate diagnostic testing.392,393 One may o er supportive care to acutely infected animals and should isolate them from uninfected animals, especially pregnant mares.

IMMUNOLOGIC DISEASES Most noninfectious cases of vasculitis in the horse are thought to be immune mediated. Small cutaneous vessels most commonly are a ected, resulting in edema that may progress to skin infarctions, necrosis, and exudation.381 The best characterized of these disorders is purpura hemorrhagica after Streptococcus equi subsp. equi infection. However, vasculitides of undetermined cause also have been described in the horse.284,394 Most of these cases appear to have an immune-mediated cause. The antigenic stimulus of hypersensitivity vasculitis is usually a microbe, drug, toxin, or foreign or endogenous protein. The deposition of circulating immune complexes with subsequent vessel wall damage is probably the major event in the pathogenesis of these vasculitic syndromes. Soluble immune complexes, formed in moderate antigen excess, become deposited in blood vessel walls in areas of increased permeability. Deposited immune complexes activate complement with resulting formation of C5a, a potent chemotactic factor for neutrophils. In ltrating neutrophils release lysosomal and other cytoplasmic enzymes such as elastase and collagenase that directly damage vessel walls. The net e ect is vessel wall leakage and luminal compromise, resulting in edema, hemorrhage, thrombosis, and ischemic changes in supplied tissues. Aside from vessel wall necrosis, vasculitis results in vascular dysfunction characterized by a net increase in vasoconstriction and platelet aggregation, which contribute to tissue ischemia.395 The diseased endothelium releases endothelin, a polypeptide that causes contraction of underlying smooth muscle396 and produces fewer dilator substances such as endothelial-derived relaxing factor.397 Injury to the vessel wall also results in diminished prostacyclin production, which usually serves to maintain vasodilation and platelet nonreactivity.267 Thus endothelial dysfunction contributes ultimately to vessel failure after vasculitis. The factors that determine which individuals develop hypersensitivity vasculitis remain unde ned. Genetic predisposition, altered immunoregulatory mechanisms, and the amount, relative size, and type of complement components in circulating immune complexes are important in determining the potential risk of developing vasculitis (because these factors determine how quickly the mononuclear phagocyte system clears immune complexes).398 Perhaps even more signi cant in this potential risk are genetic and acquired di erences in the number and activity of receptors for complement and IgG on erythrocytes and macrophages. The di erence can a ect immune complex handling and distribution greatly.399,400 Receptors for complement and the Fc of IgG mediate immune complex transport to and removal by the macrophages of the mononuclear phagocyte system. Physical factors such as turbulence of blood ow, hydrostatic pressure within vessels, and previous endothelial damage likely determine the size, type, and location of blood vessels involved in vasculitis. Researchers believe that the propensity for lesion formation in the skin of dependent body portions likely is caused by the hydrostatic pressure in a ected postcapillary venules in these areas.267

DIAGNOSIS The diagnosis of vasculitis depends on demonstration of typical histologic changes (e.g., leukocytoclasis, brinoid necrosis) in involved vessels. One should obtain full-thickness punch biopsies (6 mm in diameter) of skin in an a ected area and preserve it in 10% formalin for histopathologic study. The skin in the designated area may be clipped but not scrubbed before sampling. One can process the samples saved in Michel’s transport medium subsequently for immuno uorescence analysis to detect immune complexes.401 Multiple biopsies from various sites are optimal in reaching a diagnosis, because distribution of histologic lesions and immune complexes is patchy. However, a ected skin is often on the distal limbs where risk increases for cellulitis or exuberant granulation tissue. The irregular distribution of histologic lesions, coupled with di culty in obtaining biopsies, makes the de nitive diagnosis of vasculitis di cult. Often a diagnosis of vasculitis is based on history, clinical signs, and response to therapy. Purpura hemorrhagica is a syndrome of cutaneous vasculitis that can develop several weeks after infection with Streptococcus equi subsp. equi, S. equi subsp. zooepidemicus, in uenza virus, Corynebacterium pseudotuberculosis, or equine herpesvirus. Purpura hemorrhagica also has been described as an adverse side e ect of antistreptococcal vaccination in the horse. Purpura hemorrhagica after infection with S. equi subsp. equi has been attributed to deposition of circulating immune complexes consisting of IgA and S. equi subsp. equi M protein in small subcutaneous vessels.402 In approximately 30% of a ected horses, no viral, bacterial, or vaccine cause is identi ed.381,403 Horses are often febrile with ventral edema, especially of the distal limbs. A ected animals frequently have great pain and are reluctant to move. Petechia and ecchymoses may be present on oral, nasal, and conjunctival mucous membranes. Extensive skin necrosis and sloughing often follow severe swelling. A more severe infarctive form of purpura has also been described.404 A ected horses develop areas of infarcted skin, as well as areas of ischemic damage in the intestines, kidneys, and skeletal muscles. Reports show increased mortality with this form and indicate that

older horses repeatedly vaccinated or exposed to S. equi subsp. equi may be at an increased risk of developing the syndrome. Horses with purpura hemorrhagica are not usually thrombocytopenic but frequently have laboratory evidence of chronic in ammatory disease, including increased brinogen and total plasma protein, as well as neutrophilia.381 Careful monitoring of physical examination parameters and lab values is important to rapidly identify changes in the disease including kidney and skeletal muscle involvement.404 If S. equi subsp. equi is the possible inciting cause, then M protein titers can be measured. Titers in horses su ering from S. equi subsp. equi–induced purpura are often very high, but lower titers can be hard to interpret in horses that have been exposed to or vaccinated against the organism.405 Cutaneous biopsy of a ected areas may con rm the diagnosis. Histopathologically, a leukocytoclastic vasculitis characterized by neutrophilic accumulation in and around small dermal and subcutaneous vessels occurs.

TREATMENT The aims in the therapy of equine vasculitic syndromes are (1) to remove the antigenic stimulus if possible, (2) to reduce vessel wall in ammation, (3) to normalize the immune response, and (4) to provide supportive care. Regardless of the cause, vasculitis in horses generally warrants aggressive nursing care, which one should institute immediately. Hydrotherapy can minimize edema, and pressure wraps are useful on the limbs. Animals that become severely depressed and fail to drink or those with dysphagia caused by pharyngeal edema require uids, intravenously or via nasogastric tube. A tracheostomy may be necessary in horses with stridor and dyspnea that occurs after edema of the upper respiratory tract. Phenylbutazone, unixin meglumine, or other NSAIDs are indicated to reduce vascular in ammation and provide analgesia. The disease also may warrant corticosteroid administration. Long-term antimicrobial therapy is indicated to reduce the incidence or severity of cellulitis or other septic sequelae.403 Because no treatment e ectively eliminates EIA or EVA from the body, neither disease warrants speci c therapy. Although equine anaplasmosis spontaneously resolves, elimination of the organism by oxytetracycline therapy considerably shortens the clinical course of disease. When the cause is unknown, vasculitis can be di cult to treat because the antigenic stimulus remains unde ned and may not be eliminated easily. One should discontinue administration of any drug being used at the time that signs occur. A thorough search for an underlying infection or neoplasia is warranted. Horses with a recent history of strangles or that have been exposed to the disease should receive procaine penicillin G at 22,000 U/kg intramuscularly twice daily or potassium penicillin G intravenously four times daily for at least 2 weeks. One should drain accessible abscesses and examine and culture the guttural pouches via endoscopy. Even though the sensitizing infection usually has resolved by the time signs of purpura hemorrhagica occur, ongoing sepsis and antigen production prolong the allergic vasculitis. Although the use of corticosteroids for treating hypersensitivity vasculitis was once controversial,267 research and clinical experience suggest that horses with purpura hemorrhagica or idiopathic vasculitic syndromes respond favorably to corticosteroid therapy. Of the 53 cases documented, all were treated with corticosteroids, resulting in a 93% recovery rate.381,403 Mild cases of vasculitis may resolve without immunosuppressive therapy, but life-threatening edema involving the upper respiratory tract or causing organ system dysfunction necessitates early aggressive corticosteroid treatment. One should administer dexamethasone (0.05 to 0.2 mg/kg intravenously, intramuscularly, or orally once or twice daily) at the dose and rate necessary to reduce edema. Prednisolone may be substituted at 0.5 to 1.0 mg/kg orally twice daily for dexamethasone, but prednisone should not be used because it has poor oral absorption in the horse. After edema and hemorrhages start to resolve, one may reduce the corticosteroid dose gradually (by 10% to 15% every 1 to 2 days) while carefully monitoring the horse for relapse. It is not uncommon for horses with purpura hemorrhagica to require corticosteroid and antibiotic therapy for a period of 4 to 6 weeks before edema permanently resolves. Occasionally, horses su er disease are-ups that require the corticosteroid dose to be increased over the previously e cacious level. In these instances, unixin meglumine may enhance steroid e cacy. Antimicrobials are indicated throughout the period of systemic corticosteroid administration to reduce the incidence and severity of secondary sepsis.

PROGNOSIS The prognosis for vasculitis depends on the initiating disease process. With early aggressive therapy and supportive care, most horses recover from purpura hemorrhagica within 4 weeks, although numerous sequelae may prolong the convalescence. A 93% recovery rate was reported in one study403;however, horses that su er from the severe infarctive form carry a much poorer prognosis for recovery.404 Dermal infarction in the distal limbs often leads to skin sloughing followed by exuberant granulation tissue and may require excision followed by skin grafting. Laminitis and various infections such as cellulitis, pneumonia, colitis, and thrombophlebitis are common and may be related to long-term corticosteroid therapy. EVA and equine anaplasmosis carry a good prognosis, and disease confers some immunity to reinfection. Horses with EIA are infected persistently and may su er recurrent clinical relapses. Horses with idiopathic vasculitic syndromes have an unpredictable response to

therapy, and some have a poor prognosis. Inadequate resolution of hypersensitivity vasculitis in human beings usually is caused by failure to identify the antigenic stimulus or completely eliminate it from the body.267Although in most horses remission is spontaneous, some have debilitating cutaneous disease or develop systemic necrotizing vasculitis with a poor prognosis.

Thrombocytopenia When circulating platelet counts decrease to less than 30,000 per microliter, a bleeding diathesis frequently occurs. A ected horses have petechial and ecchymotic hemorrhages of mucosal membranes, epistaxis, increased bleeding after venipuncture, melena, or hyphema.281-286 Similar signs may occur in horses with platelet function defects. Disorders involving alterations in platelet number or function are described in previous sections.

Vitamin K Deficiency Vitamin K is a fat-soluble vitamin essential for a nal step, γ-carboxylation, in the hepatic synthesis of many coagulation factors. The serine proteases, including factors II, VII, IX, and X, require γ-carboxylation of glutamate before they are biologically functional. Vitamin K de ciency results in a gradual decrease in functional coagulation factors in circulation that ultimately can end in a clinical syndrome of inappropriate bleeding. The rapidity with which coagulation factor concentrations in circulation decline depends largely on the half-life of those factors in circulation. Factor VII has a plasma half-life in the dog of 4 to 6 hours (compared with 41, 14, and 16 hours for factors II, IX, and X, respectively) and reaches a critical concentration in the circulation more rapidly than other vitamin K–dependent factors.406 As a result, coagulation studies early in the course of vitamin K de ciency may reveal abnormalities in the extrinsic coagulation system (prolonged PT) with a normal intrinsic system (normal APTT). As the condition progresses, however, extrinsic and intrinsic coagulation parameters are abnormal. All green leafy feeds, including hay and fresh pasture, contain high concentrations of vitamin K. For this reason, foals before weaning have decreased vitamin K levels as compared with adult horses.407 Researchers also believe that vitamin K is synthesized by the intestinal micro ora of many species. As a result, vitamin K de ciency only rarely results from an absolute de ciency of the nutrient in the diet. More commonly, de ciency results from an inability of the horse to absorb or use dietary vitamin K. Chronic intestinal malabsorption of lipids can result in clinical bleeding problems. One may assess dietary fat absorption by feeding 150,000 IU of vitamin A (another fat-soluble vitamin) in corn oil with grain and measuring increases in serum vitamin A concentration.408 Concentrations should double over 12 to 24 hours. Fat absorption depends on bile acid secretion from the liver, and chronic cholestatic disorders also can result in decreased vitamin K absorption. Antagonism of vitamin K by warfarin or other dicumarol derivatives is a more common clinical problem than is absolute vitamin-K de ciency. The dicumarol derivative warfarin has been advocated as an oral anticoagulant for use in horses with navicular disease409 or jugular thrombophlebitis.410 Warfarin also is the active ingredient in many rodenticides, and accidental exposure to toxic doses of the compound is possible. Warfarin is an e ective antagonist of vitamin K and inhibits the production of functional coagulation factors II, VII, IX, and X. Warfarin therapy may begin with a daily dose of 0.018 mg/kg orally and increase in increments of 20% to achieve the desired therapeutic effect,409 which usually results in a final dose of 0.012 to 0.75 mg/kg daily. Concurrent administration of other drugs, especially phenylbutazone, may increase the risk of warfarin-induced hemorrhage. Phenylbutazone acts by competing with warfarin for binding sites on plasma albumin, increasing the concentration of free warfarin in circulation. Similarly, hypoalbuminemia can increase the risk of warfarin toxicosis. Thyroxin and corticosteroids lower the necessary therapeutic dose of warfarin by enhancing receptor a nity and increasing clotting factor catabolism. Rapid withdrawal of drugs that induce hepatic microsomal enzyme metabolism, such as rifampin, chloramphenicol, and barbiturates, also can potentiate toxicosis because these drugs enhance warfarin metabolism. Because factor VII has the shortest half-life of the vitamin K–dependent factors, warfarin a ects the extrinsic coagulation system more rapidly than the intrinsic system, and determination of PT is the most commonly used method for monitoring the e ectiveness of warfarin anticoagulative therapy. An increase in PT of 1.5 to 2.5 times baseline is recommended as a therapeutic goal of warfarin anticoagulation.410,411 Toxic doses of warfarin result in spontaneous hemorrhage that may be life threatening. One makes a diagnosis of warfarin toxicity based on a history of exposure, clinical evidence of hemorrhage (bleeding from body ori ces or into body cavities, hematoma formation, or clinical pathologic indications of blood loss anemia), prolonged PT or PTT with normal platelet count and brinogen and FDP concentrations, and response to exogenous vitamin K administration.412 Because warfarin interferes with the nal stages of clotting factor synthesis, administration of vitamin K1 at up to 1.0 mg/kg subcutaneously rapidly reverses toxicity.410 One may repeat this dose every 4 to 6 hours until PT returns to normal, followed by daily monitoring for 3 to 4 days.413 PT returns to normal approximately 5 days after the last dose of warfarin without administration of supportive therapy; administration of vitamin K1 results in a return of PT to baseline within 24 hours.413,414 In horses with severe bleeding, one should

administer whole blood transfusions concurrently to replace plasma clotting factors. The horse should improve within 1 to 2 hours. One may administer smaller doses of vitamin K1 orally, concurrent with warfarin, to titrate the anticoagulant effects of the latter.415 Second-generation anticoagulant rodenticides such as brodifacoum have a prolonged half-life (1.2 days) and lower median lethal dose compared with warfarin.415 As a result, their potential for adverse e ects are greater. History of exposure is important because therapy requires prolonged antidotal therapy.416 Aggressive symptomatic and supportive therapy is often unsuccessful.417 Toxicosis caused by ingestion of moldy sweet clover (Melilotus spp.) hay is reported in all herbivores.418 White sweet clover (Melilotus alba) and yellow sweet clover (M. o cinalis) contain coumarin. When sweet clover hay is cured improperly, numerous fungi (e.g., Penicillium, Mucor) propagate; these fungi convert coumarin to 4-hydroxycoumarin, which condenses to dicumarol.419 Animals that ingest su cient quantities of moldy hay develop severe coagulation dysfunction within 2 to 7 days with internal and external hemorrhaging.420 Continued ingestion of moldy hay can result in a lethal hemorrhagic crisis. One should consider toxicity as a diagnosis in horses with history of access to moldy sweet clover hay and can con rm the diagnosis with coagulation tests described previously for warfarin toxicosis. Suspect hay should be destroyed because toxic concentrations of dicumarol may remain for up to 4 years.419 One may treat clinical bleeding episodes resulting from inadequate vitamin K and subsequent de ciency in the levels of vitamin K– dependent coagulation factors in circulation with parenteral vitamin Kl (phylloquinone) injections at a dosage of 0.3 to 0.5 mg/kg. Subcutaneous injection is recommended because intravenous administration has resulted in a high incidence of adverse reactions in other species. One ultimately must address underlying gastrointestinal or hepatic disorders as well. Parenterally administered vitamin K3 (menadione sodium bisul te) is nephrotoxic in horses at the manufacturer’s recommended dose and should not be administered.421 Horses that receive vitamin K3 may show signs of colic, hematuria, azotemia, and electrolyte abnormalities consistent with acute renal failure.

Hepatic Diseases Severe hepatic disease often is associated with hemorrhagic tendencies caused by decreased production of coagulation factors II, V, VII, IX, and X and brinogen. Impairment of bile acid secretion may impair absorption of the fat-soluble vitamins, resulting in decreased production of the vitamin K–dependent clotting factors. Hepatic Kup er’s cells are responsible for removal of many activated coagulation factors and FDPs from the circulation. Increased circulating FDPs inhibit normal platelet function and generation of brin. These abnormalities may combine to initiate DIC. A severe bleeding diathesis is a poor prognostic indicator in horses with hepatic disease.

Drug-Induced Alterations in Hemostasis A variety of pharmaceutic agents can alter hemostatic mechanisms. Incorrect or poorly monitored administration of these agents may result in hemorrhagic tendencies in the horse.

ASPIRIN Aspirin causes an irreversible inhibition of platelet function. The mechanisms and implications of this e ect are discussed in previous sections.

HEPARIN Heparin is a mucopolysaccharide complex of variable molecular weight that potentiates the activity of AT III in neutralizing coagulation factor X and increasing the rate of inactivation of other serine proteases (factors II, IX, XI, and XII).422,423 Potentiation of AT III activity is attributed primarily to lower-molecular-weight forms of the molecule that inhibit factors X and XII, with minimal inhibitory e ects on thrombin and factors IX and XI.423 Heparin has been advocated for maintenance of the patency of indwelling catheters, prophylaxis for venous thrombosis and laminitis, anticoagulation of whole blood transfusions, prevention of postoperative peritoneal adhesions, and treatment and prevention of DIC.422,424-427 Efficacy remains controversial with reports supporting and denying the effects of heparin.428,429 Bleeding complications associated with unfractionated heparin therapy are uncommon, but administration of protamine sulfate at up to 1 mg for every 90 to 100 U of heparin has been recommended in horses with heparin-associated hemorrhage.422 Protamine sulfate forms a stable salt with heparin and with soluble brinogen monomers. If the FDP concentration increases in a patient, then precipitation of these salts may result in intravascular infarction, and use of protamine sulfate is contraindicated. The most pronounced adverse e ect associated with unfractionated heparin administration in the horse is a signi cant reduction in the circulating RBC mass.291,430 Doses of 160 to 320 U/kg subcutaneously twice daily for nine doses resulted in a signi cant decrease in RBC numbers, PCV, and total hemoglobin, and an increase in MCV, with a rapid return of values to baseline after discontinuation of therapy.291 Because heparin is known to increase the activity of mononuclear phagocytes, decreased RBC numbers originally were hypothesized to result from an enhanced phagocytosis of erythrocytes.430 However, decreased RBC numbers now are thought to result primarily from in vivo erythrocyte agglutination.431,432 One can reverse this agglutination in vitro by adding a trypsin solution to the RBC suspension. Decreases in

RBC numbers and increases in MCV associated with heparin administration are most likely artifactual, resulting from the measurement of multiple agglutinated cells as single units.432 Unfractionated heparin administration in horses (160 U/kg subcutaneously twice daily) also has been associated with development of large plaques of painful edema.433 Low-molecular-weight heparin (LMWH) is derived from unfractionated heparin by solvent extraction or gel ltration. The resulting drug is made up of smaller molecules and has more reliable anticoagulant properties, better bioavailability, and a longer half-life in human beings.434 The pharmaco*kinetic properties of LMWH have been studied in the horse, and appropriate dosages have been identi ed.435 An injection of 50 IU/kg of LMWH once daily was compared with unfractionated heparin at tapering dosages between 150 to 100 IU/kg twice daily in horses with colic.436 Signi cantly more jugular vein problems were observed in horses receiving unfractionated heparin and adverse e ects, including decreased PCV and prolonged APTT and TT, were markedly increased in the unfractionated group as compared with the LMWH group. E cacy trials for the use of LMWH in prevention and treatment of hypercoagulability are lacking, but LMWH may o er fewer adverse e ects as compared with horses treated with the unfractionated form. The use of this drug is often confounded by its expense in the horse.

Disseminated Intravascular Coagulation DIC is a condition of pathologic activation of coagulative and brinolytic systems, ultimately leading to microvascular ischemia and secondary organ dysfunction. Human research indicates that the in ammatory cytokine cascade that results from massive trauma or sepsis is intimately associated with the development of a hypercoagulable state that can ultimately lead to DIC.437 DIC may be acute or chronic, systemic or localized, and frequently is characterized by severe hemorrhage caused by consumption of coagulation factors, but it also may manifest as a pronounced thrombotic tendency due to a hypercoagulable state. The type and severity of clinical signs depend on the initial stimulus for pathologic systemic coagulation activation, the duration and extent of activation, the availability of coagulation factors, and the relative strengths of activation of procoagulant and brinolytic forces. The nal result in horses with severe DIC is often widespread ischemic organ failure, severe hemorrhage, and death. DIC is not considered a primary diagnosis and occurs after a variety of disorders triggered by many di erent mechanisms. In the horse, DIC most commonly is associated with gastrointestinal disease or sepsis.438-448 The common initiating factor in most cases is endotoxin, the external lipopolysaccharide cell wall of gram-negative bacteria. Endotoxin can activate factor XII directly to initiate the intrinsic coagulation pathway and may cause widespread damage to vascular endothelium, exposing subendothelial collagen and releasing tissue thromboplastin to trigger the coagulation pathways. Endotoxin damages platelets, inducing aggregation and activation, and exposing platelet factor 3. Other conditions associated with DIC include intravascular hemolysis, bacteremia (gram positive or gram negative), viremia, neoplasia, circulating immune complexes, immune thrombocytopenia, fetal death in utero, burns, hepatic disease, renal disease, and vasculitides. Virtually any primary disease process may initiate DIC by injury to blood cells, increasing the availability of phospholipid surfaces, injury to vascular endothelium, or tissue injury and subsequent release of thromboplastin.449 As widespread activation of proin ammatory pathways occurs, kinins are generated and the complement cascade is stimulated.449 Cytokines involved in the activation of this cascade include TNF-α, IL-1, IL-6, and IL-8.450 Factor XIIa is responsible for initiating kinin formation by activating kallikrein, which in turn proteolytically converts HMWK to kinins. Bradykinin, the most physiologically important of the kinins, enhances vascular permeability, dilates some blood vessels, and stimulates migration of leukocytes into the extravascular space.451 Plasmin activates the complement cascade and induces neutrophil chemotaxis, increases vascular permeability, and destroys RBCs and platelets, releasing membrane phospholipoprotein and ADP to act as procoagulants. Complement products can activate factor XII. In response to endotoxin, mononuclear phagocytes release procoagulant substances including TF, platelet activating factor, TNF, and IL452 l. These phagocytic cells are responsible for removal of activated clotting factors and FDPs from the circulation. Massive concentrations of activated clotting factors or FDPs, or impairment of phagocytic function, can lead to abnormally high levels of either element in the blood. Activated clotting factors then remain available for continued coagulative events. Elevated FDP levels in the peripheral circulation can inhibit fibrin monomer polymerization and thrombin formation and also coat platelet membranes, interfering with aggregation. The clinical signs associated with DIC in the horse vary widely and re ect the primary disease process, the duration and severity of coagulation activation, and the balance between procoagulant and brinolytic forces. Horses seldom show severe bleeding diathesis after DIC, and less severe forms of DIC in which the horse may be in a subclinical hypercoagulable state may go unnoticed. Careful clinical examination may reveal petechia and ecchymoses of mucous membranes and prolonged bleeding from venipuncture sites. Many horses exhibit an increased tendency to venous thrombosis after venipuncture.442 Other clinical signs may re ect organ dysfunction after microvascular thrombosis and tissue ischemia. In the kidney, organ dysfunction may occur as tubulointerstitial disease and acute renal failure. In the foot, DIC may contribute to digital ischemia and laminitis424,425,438,442; however, experimentally induced laminitis was not associated with signi cant changes in hemostatic parameters.364 Many horses with DIC have clinical signs re ecting poor peripheral perfusion and shock,

including prolonged capillary re ll time, cool extremities, and tachycardia, which may result from the primary disease process or occur after DIC-induced microvascular occlusion and inflammation. Diagnosis of DIC depends on the results of a variety of tests assessing the quantity and function of coagulative substances. The most commonly evaluated parameters in horses are platelet count and function, PT, PTT, activated clotting time, and FDPs. Platelet counts usually are decreased in horses with DIC.439-442,444,447 In horses with normal platelet counts, platelet function frequently is impaired.438 Although coagulation times may be shorter early in the disease, rarely does one examine horses at this time. Most horses have reached a stage of hemorrhagic tendency, largely because of consumption of coagulative elements, by the time DIC is recognizable. These patients have increased PT, PTT, and activated clotting time. Because DIC accelerates brinolytic processes, elevations in circulating FDPs and D-dimers occur commonly.438,442,444,447 In contrast to other species, horses with DIC commonly have normal to increased brinogen concentrations.352,438,442,447 Fibrinogen acts as an acute phase reactant protein in the horse, and in ammation, which accompanies most cases of DIC in the horse, stimulates accelerated hepatic production. Several other laboratory tests may aid in diagnosis of DIC in the horse. Decreased plasma AT III concentrations are considered a sensitive indicator of DIC in human beings.453 Because AT III is bound irreversibly to serine proteases in the plasma, AT III is consumed rapidly during pathologic intravascular coagulation. AT III activity has been shown to decline in horses with experimentally induced large colon torsion.445 This decline was reversible in horses that survived and persisted in horses that died. In horses with naturally occurring colic or acute colitis, plasma AT III activity is decreased signi cantly.357,439,440 One does not always nd decreases in AT III activity, however, and normal activity has been reported in ponies with coagulopathy after equine ehrlichial colitis.443 Because AT III is a small protein, it may be lost from circulation in most disease conditions associated with hypoalbuminemia, including protein-losing enteropathies and nephropathies.454 Increased plasma AT III activity has been associated with hepatic disease in the horse.357 Assays for equine α2-antiplasmin, plasminogen, and protein C have been described, and normal values have been reported.362-364 Plasma α2-antiplasmin concentrations were decreased in ponies with coagulopathy after equine ehrlichial colitis. Plasminogen concentrations were increased in these ponies, possibly indicating a role of plasminogen as an acute phase reactant.443 Plasma protein C concentrations may be decreased in human beings with DIC.455 One may determine individual coagulation factor levels, but such determination rarely provides more information than PT or PTT. Diagnostic criteria for DIC in the horse have not been de ned rigidly. One must assess each patient in the light of the severity and duration of underlying disease processes, clinical appearance, and laboratory hemostatic values. Subclinical DIC may be documented with laboratory evaluation of animals that have no overt clinical signs of pathologic coagulopathy. Multiple hemostatic abnormalities in an animal with thrombotic or hemorrhagic tendencies strongly suggest DIC. Regardless of the de nition of DIC, if a high index of suspicion exists, attempted therapy is warranted. The therapeutic plan for horses with DIC must focus on treatment of the primary underlying disease process. Only by removing the pathologic stimulus for coagulation can procoagulant and brinolytic mechanisms be rebalanced e ectively. One should administer supportive therapy for shock, decreased tissue perfusion, and acid-base and electrolyte abnormalities. If endotoxemia contributes to the disease process, then one should administer NSAIDs. Low-dose unixin meglumine therapy (0.25 mg/kg intravenously three times daily) is e ective in inhibiting endotoxin-induced cyclooxygenase activity.456 One should treat animals with sepsis with appropriate antimicrobial agents. Fresh platelet-rich plasma can be introduced intravenously to horses with life-threatening hemorrhage after DIC. Fresh plasma resupplies soluble coagulation factors, and normal platelets may restore primary hemostatic mechanisms. The recommended dose of plasma for use in DIC in other species is typically 10 to 20 ml/kg or 5 to 10 L in the average horse, which is expensive. Because knowledge regarding the e cacy of this treatment is limited, few patients undergo such aggressive therapy. Administration of whole blood or blood products remains controversial in the treatment of patients with DIC because of the potential for worsening the coagulopathy on administration of fresh coagulation factors. One must consider this possibility, but in the face of life-threatening hemorrhage, bene ts may outweigh risks. Heparin therapy has been advocated for treatment of DIC in other species. Because heparin acts as an anticoagulant by complexing with AT III to inhibit serine proteases, AT III levels must be normal to achieve any bene ts from heparin therapy. Heparin administration is controversial in the horse.422,424-429 As discussed previously in this chapter, the use of LMWH has been advocated over unfractionated heparin because of a tendency for unfractionated heparin to cause erythrocyte agglutination and anemia in vivo.291,432 The recommended dosage of LMWH is 50 to 100 IU/kg for dalteparin and 40 to 80 IU/kg for enoxaparin subcutaneously once daily.435 Aspirin, as mentioned previously, acts as an irreversible inhibitor of platelet cyclooxygenase, which leads to a decreased platelet aggregation. It has been used in horses in a hypercoagulable state. Aspirin can be given at 20 mg/kg either orally once daily or every other day; it can be given rectally once daily when oral administration is contraindicated. Adverse e ects can include bleeding because of platelet dysfunction or gastrointestinal ulceration.457

Pentoxifylline and other methylxanthine derivatives are e ective in decreasing the production of in ammatory cytokines, including TNFα, IL-1, and IL-6, as well as increasing RBC exibility.458 This drug acts as a nonspeci c phosphodiesterase inhibitor and may be helpful in decreasing clinical platelet activation in the horse.459 Administration at 10 mg/kg orally twice daily yields appropriate plasma levels; however, long-term administration might require increasing dosages as the clinical response decreases with ongoing treatment.458 Other drugs, including etamsylate and recombinant hirudin, have been used successfully in other species for their antithrombogenic effects.460,461 Initial investigation of these drugs in horses have yielded promising results, but further research is required. Early recognition of DIC and aggressive therapy of underlying disease conditions remain the most e ective treatments for the horse with pathologic coagulopathy. Understanding the pathogenesis and diagnosis of DIC will aid the practitioner in this process.

Thrombosis and Thrombophlebitis A thrombus is an intravascular accumulation of brin, platelets, and leukocytes that usually occurs at a site of endothelial damage. Three factors may predispose a horse to vascular thrombosis: (1) vascular endothelial injury, (2) vascular stasis or slowing of blood ow, and (3) abnormalities of coagulation processes. After vascular damage, thrombus formation is the normal end result of procoagulative forces. However, pathologic or abnormal thrombus formation may occur in patients with an imbalance of procoagulative and brinolytic forces, as in DIC. The horse appears to be particularly prone to super cial vein thrombosis during periods of systemic activation of coagulation. Jugular vein thrombosis may occur in these animals after a single, apparently atraumatic, venipuncture. Jugular catheter associated thrombophlebitis has been linked to several risk factors including catheter material and experience of the person placing the catheter,462,463 but most risk factors are related to the patient. Studies suggest that endotoxemic horses and septicemic horses have an increased susceptibility to thrombosis at the site of indwelling venous catheters.464 Bilateral jugular vein thrombosis leads to increased intravascular pressures proximal to the thrombus formation and severe edema of the head. This edema may impair respiratory function, necessitating tracheostomy. If one suspects that sepsis is contributing to thrombus formation, then appropriate antibiotic therapy should be initiated. If hypercoagulation may be a predisposing factor for thrombosis, then aspirin administration at 20 mg/kg every other day may be bene cial to inhibit platelet aggregation and release reactions.338,341 Hot compresses and hydrotherapy may improve collateral blood ow and decrease in ammation at the site of thrombosis. The thrombus eventually should recanalize, and venous ow should return. One may assess recanalization and the size of the thrombus with ultrasound examination of blood ow through the a ected vessel.465 If an underlying disease process is present, then primary care should be directed at correcting this disorder. In horses with unilateral jugular vein thrombosis, one should protect patency of the opposite jugular vein by minimizing prothrombotic trauma to that vessel, including venipuncture or catheter placement.

BLOOD AND BLOOD COMPONENT THERAPY Treatment of a variety of equine disorders involves administration of whole blood, plasma, or other blood constituents. One must base therapeutic decisions on clinical signs, laboratory parameters, availability of suitable donor services, and economic considerations.

Whole Blood Transfusion One most commonly administers blood transfusions to horses with anemia severe enough to impair tissue oxygenation. Tachycardia, tachypnea, lethargy, weakness, cool extremities, and pale mucous membranes indicate signi cant tissue hypoxia and may occur at a PCV as high as 18 or as low as 10, depending on the rapidity of erythrocyte loss. Chronic anemias allow for physiologic adaptation to the low circulating RBC numbers, and severe clinical signs may not be apparent until the PCV is extremely low. Fresh whole blood transfusions also may be indicated in horses with severe hemorrhage associated with DIC to renew supplies of soluble coagulation factors and platelets. Donor animals should lack detectable circulating alloantibodies against erythrocyte antigens that might be present in the recipient.466,467 One can use crossmatching to detect existing antibody before transfusion. EDTA-anticoagulated blood and serum should be collected from the recipient and from several potential donor horses. The major crossmatch checks for compatibility between donor RBCs and any alloantibody that might be present in patient serum. The minor crossmatch assesses compatibility between alloantibody that might be present in donor serum and patient RBCs.13 Both tests are important before transfusion of whole blood; the minor crossmatch is most critical when one contemplates plasma transfusion. Transfusion of RBCs may sensitize the recipient to produce alloantibody against any incompatible erythrocyte antigens present in donor blood. For this reason, one should consider routine use of donors that are negative for factors Aa and Qa to prevent the inadvertent sensitization of brood mares against the two most common alloantigens involved in neonatal isoerythrolysis, Aa and Qa. If crossmatching is not possible, unmatched transfusion for critical patients can be considered. One must weigh the potential adverse e ects of a transfusion reaction against the potential bene ts to the patient. Most horses without previous transfusions may safely receive a

single transfusion or multiple transfusions over 3 to 4 days from an unmatched donor. Chances of an adverse reaction increase when either donors or recipients are mares that have been bred previously. For this reason, male horses or female horses that have never been pregnant are preferred as donor animals for unmatched transfusions. One safely may remove 5 to 10 L of blood from most adult donor horses. All equipment used for blood or blood product collection and administration should be sterile, and all procedures should be performed with strict attention to aseptic technique. Stock solutions of acid citrate dextrose anticoagulant are available commercially. Alternatively, one may use sodium citrate as an anticoagulant, mixing one part of a 3.2% sodium citrate solution to nine parts whole blood. Plastic blood collection bags with premeasured anticoagulant are convenient and e ective for use with horses. Equine RBCs are stable for 35 days after collection when collected and stored in commercially available bags using citrate phosphate dextrose (CPD) as the anticoagulant.468 The volume of blood administered to a patient depends on the severity of RBC depletion and the total blood volume. A recommended formula for calculating this is as follows329:

Normal blood volume in adult horses is approximately 72 ml/kg.469 In neonates, blood volume is 151 ml/kg, decreasing to 93 ml/kg at 4 weeks, 82 ml/kg at 12 weeks, and adult values by 4 to 6 months of age.470 Transfused RBCs survive less than 1 week in the horse, sometimes less than 2 days, and multiple transfusions may be necessary to maintain a patient until bone marrow production can exceed the rate of peripheral RBC destruction or loss.471 Blood products should be administered through an intravenous system with an in-line lter. Even with crossmatching of donor and recipient, one should administer the initial 25 to 50 ml of transfused blood slowly over 15 to 30 minutes with close monitoring of patient heart rate, respiratory rate, and behavior. If these parameters remain stable, then the remainder of the transfusion can be administered at a rate of 15 to 25 ml/kg/hr. One should give transfusions to horses with suspected or confirmed endotoxemia or septicemia more cautiously. Several adverse reactions, immediate or delayed, may be associated with administration of blood or blood components. Mild reactions such as development of urticaria and tachycardia often occur.472 More severe reactions are characterized by increases in heart and respiratory rates, dyspnea, fever, trembling, weakness, hypotension, diarrhea, abdominal pain, anaphylaxis, shock, or pulmonary edema. Acute hemolytic reactions with hemoglobinemia and hemoglobinuria result from transfusions from a donor with an incompatible blood type. DIC may accompany severe acute hemolysis. One should discontinue the transfusion immediately and administer supportive therapy. Intravenous crystalline fluids improve peripheral circulation and maintain renal perfusion. NSAIDs may decrease inflammatory reactions. Febrile reactions are not uncommon during or immediately after transfusions and may result from incompatible leukocyte or platelet antigens or presence of pyrogens in the transfused blood. One should discontinue transfusions if clinical signs are severe. NSAIDs may help in decreasing the febrile response. Occasionally, blood products become contaminated with bacteria or bacterial products, and their administration can result in severe septicemia and endotoxemia. A ected animals develop uncontrollable shaking, hypotension, weakness, tachycardia, tachypnea, and collapse. If this occurs, the transfusion should be discontinued immediately, supportive care should be provided, and the blood being administered should be cultured. A Gram stain may reveal bacteria if contamination is severe. In this case, one should institute appropriate antibiotic therapy. NSAIDs may help in alleviating the adverse effects of endotoxin. More delayed reactions to transfusion include transmission of infectious diseases and allosensitization of the recipient. One should test all donor horses regularly for EIAV. Other infectious agents potentially transmittable via blood transfusions include Anaplasma phagocytophilum and the equine herpesviruses (EHV). Mares that receive blood transfusions are at risk for producing foals that develop neonatal isoerythrolysis later in life.

Plasma Transfusion Plasma transfusion is indicated in horses with hypoproteinemia su cient to result in signi cant uid loss from the blood to the extravascular space and also is indicated for foals with failure of passive transfer (discussed previously in the text), replacement of some coagulation factors (II, V, VII, X, XIII), reversal of warfarin toxicity, endotoxemia-septicemia, and in some cases of DIC. Selection of donor animals, administration of plasma, and potential adverse reactions are similar to those described for whole blood transfusions. One may estimate the volume (in milliliters) to be administered to a hypoproteinemic horse from the following:

Plasma volume in adult horses is estimated as 48 ml/kg.469 In neonates, plasma volume is approximately 95 ml/kg, decreasing to 62

ml/kg at 4 weeks, 53 ml/kg at 12 weeks, and adult values by 4 to 6 months of age.470 The observed increase in plasma protein after plasma transfusions is often not as great as would be predicted from this formula and most likely is caused by equilibration of administered blood proteins between intravascular and extravascular spaces. Plasma transfusions should be administered cautiously in the same manner as whole blood transfusions. Severe anaphylaxis is possible, but most reactions are mild and include only urticaria and edema of the face and limbs. Patients should be monitored for these signs, as well as for increased heart rate, increased respiratory rate, sweating, colic, and agitation. Another less common reaction to plasma that has been described is fatal serum hepatitis. The four horses described had received plasma transfusions at 41 to 60 days before demonstration of clinical signs consistent with severe liver disease. This study estimated the prevalence of this complication to be 0.4% of horses older than 1 year of age receiving plasma transfusions.473 Plasma from suitable donors is commercially available. Large equine practices may consider the purchase of plasmapheresis equipment, capable of economically preparing large quantities of pure plasma. One also may prepare plasma by allowing erythrocytes in whole blood to settle by gravity or, preferably, by centrifugation. The plasma is then removed in an aseptic manner and administered through an intravenous system with an in-line lter. Compared with gravity ow, automated plasmapheresis is the preferred method of plasma collection; the effect on hematologic and coagulation factors is mild and of no consequence to the recepient.474,475

Blood Component Therapy Continuous- ow centrifugation hemapheresis has been used extensively in human medicine to produce relatively pure blood components for administration to patients needing replacement of one speci c blood constituent. This technique is being used increasingly in veterinary medicine and is the preferred method for equine plasma collection. Blood removed from one jugular vein of a donor horse circulates in a closed loop through the blood separation device, with collection of desired blood components and return of unwanted components to the donor via the opposite jugular vein.476,477 By altering the centrifugal force of the collection device, one may collect plasma, RBCs, leukocytes, or platelets preferentially.478 Collection of leukocyte and platelet-rich fractions from horses has been described.476,478 Platelet transfusions may be indicated in selected horses with thrombocytopenia. Leukocyte transfusions have been beneficial in septic human neonates, and the feasibility and efficacy of granulocyte transfusions in the equine neonate have been investigated.478,479

REFERENCES 1. Arnbjerg J. Poisoning in animals due to oral application of iron with a description of a case in a horse. Nord Vet Med. 1981;33:71. 2. Divers T.J., Warner A., Vaala W.E., et al. Toxic hepatic failure in newborn foals. J Am Vet Med Assoc. 1983;183:1407. 3. Edens L.M., Robertson J.L., Feldman B.F. Cholestatic hepatopathy, thrombocytopenia and lymphopenia associated with iron toxicity in a Thoroughbred gelding. Equine Vet J. 1993;25:81. 4. Lavoie J.P., Teuscher E. Massive iron overload and liver fibrosis resembling haemochromatosis in a racing pony. Equine Vet J. 1993;25:552. 5. Cornelius C., Kaneko J., Benson D., et al. Erythrocyte survival studies in the horse, using glycine-2-C14. Am J Vet Res. 1960;21:1123. 6. Marcilese N., Figueiras H., Kremenchuzky S. Red cell survival time in the horse, determined with di-isopropyl-phosphorodluoridate-P32. Am J Physiol. 1966;211:281. 7. Stockham S.L., Harvey J.W., Kinden D.A. Equine glucose-6-phosphate dehydrogenase deficiency. Vet Pathol. 1994;31:518-527. 8. Harvey J.W., Stockham S.L., Scott M.A., et al. Methemoglobinemia and eccentrocytosis in equine erythrocyte flavin adenine dinucleotide deficiency. Vet Pathol. 2003;40:632-642. 9. Harvey R., Hambright M., Rowe L. Clinical biochemical and hematologic values of the American Miniature Horse: reference values. Am J Vet Res. 1984;45:987. 10. Zinkl J., Mae D., Merida P., et al. Reference ranges and the in uence of age and sex on hematologic and serum biochemical values in donkeys (Equus asinus). Am J Vet Res. 1990;51:408. 11. Harvey J., Asquith R., McNulty P., et al. Haematology of foals up to one year old. Equine Vet J. 1984;16:347. 12. Latimer K.S., Mahaffey E.A., Prasse K.W. Duncan and Prasse’s veterinary laboratory medicine, ed 4. Ames, Iowa: Iowa State University Press; 2003. 13. Feldman B.F., Zinkl J.G., Jain N.C. Schalm’s veterinary hematology, ed 5. Philadelphia: Lippincott Williams & Wilkins; 2000. 14. Reef V., Dyson S., Beech J. Lymphosarcoma and associated immune-mediated hemolytic anemia and thrombocytopenia in horses. J Am Vet Med Assoc. 1984;184:313. 15. Perk K., Frei Y., Herz A. Osmotic fragility of red blood cells of young and mature domestic laboratory animals. Am J Vet Res. 1964;25:1241. 16. Russell K.E., Sellon D.C., Grindem C.B. Bone marrow in horses: indications, sample handling, and complications. Compend Contin Educ Pract Vet. 1994;16:1359. 17. Franken P., Wensing T., Schotman A. The bone marrow of the horse. I. The techniques of sampling and examination and values of normal warm-blooded horses,. Zentralbl Veterinarmed A. 1982;29:16. 18. Franken P., Wensing T., Schotman A. The bone marrow of the horse. I. Warm-blooded horses with anaemia,. Zentralbl Veterinarmed A. 1982;29:23. 19. Tschudi P., Archer R., Gerber H. The cells of equine blood and their development,. Equine Vet J. 1975;7:141. 20. Calhoun M. A cytological study of the costal marrow. III. Hemograms of the horse and cow,. Am J Vet Res. 1955;16:297. 21. Tablin F., Weiss L. Equine bone marrow: a quantitative analysis of erythroid maturation. Anat Rec. 1985;213:202. 22. Schalm O. Equine hematology. IV. Erythroid marrow cytology in response to anemia. Vet Clin North Am Equine Pract. 1980;2:35. 23. Tschudi P., Archer R., Gerber H. Cytochemical staining of equine blood and bone marrow cells. Equine Vet J. 1977;9:205. 24. Smith J., Moore K., Cipriano J., et al. Serum ferritin as a measure of stored iron in horses. J Nutr. 1984;114:677. 25. Osbaldiston G., Griffith P. Serum iron levels in normal and anemic horses. Can Vet J. 1972;13:105.

26. Smith J., DeBowes R., Cipriano J. Exogenous corticosteroids increase serum iron concentrations in mature horses and ponies. J Am Vet Med Assoc. 1986;133:1296. 27. Harvey J.W., Asquith R.L., Sussman W.A., et al. Serum ferritin, serum iron, and erythrocyte values in foals. Am J Vet Res. 1987;48:1348. 28. Mullaney T.P., Brown C.M. Iron toxicity in neonatal foals. Equine Vet J. 1988;20:119. 29. McFarlane D., Sellon D.C., Parker B. Primary erythrocytosis in a 2-year-old Arabian gelding. J Vet Intern Med. 1998;12:384. 30. Torton M., Schalm O. Influence of the equine spleen on rapid changes in the concentration of erythrocytes in peripheral blood. Am J Vet Res. 1964;25:500. 31. Archer R., Clabby J. The effect of excitation and exertion on the circulation blood of horses,. Vet Rec. 1965;77:689. 32. Keenan D. Changes in packed cell volume of horses during races. Aust Vet Pract. 1980;10:125. 33. Dalton R. The signi cance of variations with activity and sedation in the haematocrit, plasma protein concentration and erythrocyte sedimentation rate of horses,. Br Vet J. 1972;128:439. 34. Beech J., Bloom J.C., Hodge T.G. Erythrocytosis in a horse. J Am Vet Med Assoc. 1984;184:986. 35. Bayly W.M., Reed S.M., Leathers C.W., et al. Multiple congenital heart anomalies in five Arabian foals. J Am Vet Med Assoc. 1982;181:684. 36. Berlin N.I. Diagnosis and classification of the polycythemias. Semin Hematol. 1975;12:339. 37. Cook T., Divers T.J., Rowland P.H. Hypercalcaemia and erythrocytosis in a mare associated with a metastatic carcinoma. Equine Vet J. 1995;27:316. 38. Koch T.G., Wen X., Bienzle D. Lymphoma, erythrocytosis, and tumor erythropoietin gene expression in a horse. J Vet Intern Med. 2006;20(5):1251-1255. 39. Lennox T.J., Wilson J.H., Hayden D.W., et al. Hepatoblastoma with erythrocytosis in a young female horse. J Am Vet Med Assoc. 2000;216:718. 40. Roby K.A., Beech J., Bloom J.C., et al. Hepatocellular carcinoma associated with erythrocytosis and hypoglycemia in a yearling filly. J Am Vet Med Assoc. 1990;196:465. 41. Jeffcott L.B. Primary liver-cell carcinoma in a young Thoroughbred horse. J Pathol. 1969;97:394. 42. Weiss D.J., Moritz A. Equine immune-mediated hemolytic anemia associated with Clostridium perfringens infection. Vet Clin Pathol. 2003;32(1):22-26. 43. Cooper C., Sears W., Bienzle D. Reticulocyte changes after experimental anemia and erythropoietin treatment of horses. J Appl Physiol. 2005;99:915-921. 44. Lumsden H.J., Valli V.E., McSherry B.J., et al. The kinetics of hematopoiesis in the light horse. III. The hematological response to hemolytic anemia,. Can J Comp Med. 1975;39:332. 45. Lumsden J.H., Valli V.E., McSherry B.J., et al. The kinetics of hematopoiesis in the light horse. II. The hematological response to hemorrhagic anemia,. Can J Comp Med. 1975;39:324. 46. Radin M., Eubank M., Weiser M. Electronic measurement of erythrocyte volume and volume heterogeneity in horses during erythrocyte regeneration associated with experimental anemias. Vet Pathol. 1986;23:656. 47. Easley J.R. Erythrogram and red cell distribution width of Equidae with experimentally induced anemia. Am J Vet Res. 1985;46:2378. 48. Weiser G., Kohn C., Vachon A. Erythrocyte volume distribution analysis and hematologic changes in two horses with immune-mediated hemolytic anemia. Vet Pathol. 1983;20:424. 49. Shull R. Biochemical changes in equine erythrocytes during experimental regenerative anemia. Cornell Vet. 1981;71:280. 50. Wu M.J., Feldman B.F., Zinkl J.G., et al. Using red blood cell creatine concentration to evaluate the equine erythropoietic response. Am J Vet Res. 1983;44:1427. 51. Kaneko J.J., Tanaka S., Nakajima H., et al. Enzymes of equine erythrocytes: changes during equine infectious anemia. Am J Vet Res. 1969;30:543. 52. Smith J.E., Agar N.S. Studies on erythrocyte metabolism following acute blood loss in the horse. Equine Vet J. 1976;8:34. 53. Rooney J.R. Internal haemorrhage related to gestation in the mare. Cornell Vet. 1964;54:11. 54. Lofstedt R. Haemorrhage associated with pregnancy and parturition. Equine Vet Educ. 1994;6:138. 55. Scoggin C.F., McCue P.M. How to assess and stabilize a mare suspected of periparturient hemorrhage in the field. Proceedings Am Assn Eq Pract. 2007;53:342-348. 56. Pascoe R.R. Rupture of the utero-ovarian or middle uterine artery in the mare at or near parturition. Vet Rec. 1979;104:77. 57. Stowe H.D. Effects of age and impending parturition upon serum copper of Thoroughbred mares. J Nutr. 1968;95:179. 58. Arnold C.E., Payne M., Thompson J.A., et al. Periparturient hemorrhage in mares: 73 cases (1998-2005). J Am Vet Med Assoc. 2008;232:1345-1351. 59. Moon P.F., Snyder J.R., Haskins S.C., et al. Effects of a highly concentrated hypertonic saline-dextran volume expander on cardiopulmonary function in anesthetized normovolemic horses. Am J Vet Res. 1991;52:1611. 60. Belgrave R.L., Hines M.T., Keegan R.D., et al. E ects of a polymerized ultrapuri ed bovine hemoglobin blood substitute administered to ponies with normovolemic anemia. J Vet Intern Med. 2002;16:396. 61. Maxson A.D., Giger U., Sweeney C.R., et al. Use of a bovine hemoglobin preparation in the treatment of cyclic ovarian hemorrhage in a miniature horse. J Am Vet Med Assoc. 1993;203:1308. 62. Weld J.M., Kamerling S.G., Combie J.D., et al. The effects of naloxone on endotoxic and hemorrhagic shock in horses,. Res Commun Chem Pathol Pharmacol. 1984;44:227-238. 63. Heidmann P., Tornquist S.J., Qu A., et al. Laboratory measures of hemostasis and brinolysis after intravenous administration of epsilon-aminocaproic acid in clinically normal horses and ponies. Am J Vet Res. 2005;66:313-318. 64. Ross J., Dallap B.L., Dolente B.A., et al. Pharmaco*kinetics and pharmacodynamics of epsilon-aminocaproic acid in horses. Am J Vet Res. 2007;68:1016-1021. 65. Graham L, Farnsworth K, Cary J: The e ect of Yunnan baiyao on the template bleeding time and activated clotting time in healthy halothane anesthetized ponies. In Proceedings of the Eighth International Veterinary Emergency Critical Care Society Symposium, San Antonio, Tex, 2002. 66. Magdesian K.G., Fielding C.L., Rhodes D.M., et al. Changes in central venous pressure and blood lactate concentration in response to acute blood loss in horses. J Am Vet Med Assoc. 2006;229:1458-1462. 67. McKeever K., Kirby K., Hinchcli K.W. E ects of erythropoietin on plasma and red cell volume Vo 2 max and hemodynamics in exercising horses [abstract]. Med Sci Sports Exerc. 1993;25:S23. 68. Malikides N., Kessell A., Hodgson J.L., et al. Bone marrow response to large volume blood collection in the horse. Res Vet Sci. 1999;67:285. 69. Allen B., Archer R.K. Haptoglobins in the horse. Vet Rec. 1971;89:106. 70. McGuire T., Henson J. The detection of intravascular haemolysis in the horse. Br Vet J. 125:v-vi, 1969. 71. Tennant B., Dill S.G., Glickman L.T., et al. Acute hemolytic anemia, methemoglobinemia, and Heinz body formation associated with ingestion of red maple leaves by horses. J Am Vet Med Assoc. 1981;179:143. 72. Alsup E.M., DeBowes R.M. Dimethyl sulfoxide. J Am Vet Med Assoc. 1984;185:1011. 73. MacLachlan N., Divers T. Hemolytic anemia and fibrinoid change of renal vessels in a horse. J Am Vet Med Assoc. 1982;181:716. 74. Morris C.F., Robertson J.L., Mann P.C., et al. Hemolytic uremic-like syndrome in two horses. J Am Vet Med Assoc. 1987;191:1453.

75. Duncan S. Oak leaf poisoning in two horses. Cornell Vet. 1961;51:159. 76. Beck D.J. A case of primary autoimmune haemolytic anaemia in a pony,. Equine Vet J. 1990;22:292. 77. Dodds W. Autoimmune hemolytic disease and other causes of immune-mediated anemia: an overview. J Am Anim Hosp Assoc. 1977;13:437. 78. Moriarty K., Brown M., Sutton R. An anaemic state in a horse associated with a cold-acting antibody. N Z Vet J. 1976;24:85. 79. Reef V.B. Clostridium perfringens cellulitis and immune-mediated hemolytic anemia in a horse. J Am Vet Med Assoc. 1983;182:251. 80. Anderson L. Idiopathic auto-immune haemolytic anaemia in a horse. N Z Vet J. 1974;22:102. 81. Sutton R., Pearce H., Kelley C., et al. Auto-immune haemolytic anaemia in a horse. N Z Vet J. 1978;26:311. 82. Kitchen H, Krehbiel JD: Proceedings of the First International Symposium on Equine Hematology, East Lansing, MI, 1975, American Assoc Equine Practitioners. 83. Davis E.G., Wilkerson M.J., Rush B.R. Flow cytometry: clinical applications in equine medicine. J Vet Intern Med. 2002;16:404-410. 84. Farrelly B., Collins J., Collins S. Autoimmune haemolytic anaemia (AHA) in the horse. Ir Vet J. 1977;20:42. 85. Blue J., Dinsmore R., Anderson K. Immune-mediated hemolytic anemia induced by penicillin in horses. Cornell Vet. 1987;77:263. 86. McConnico R.S., Roberts M.C., Tompkins M. Penicillin-induced immune-mediated hemolytic anemia in a horse. J Am Vet Med Assoc. 1992;201:1402. 87. Step D.L., Blue J.T., Dill S.G. Penicillin-induced hemolytic anemia and acute hepatic failure following treatment of tetanus in a horse. Cornell Vet. 1991;81:13. 88. Robbins R.L., Wallace S.S., Brunner C.J., et al. Immune-mediated haemolytic disease after penicillin therapy in a horse. Equine Vet J. 1993;25:462. 89. Thomas H.L., Livesey M.A. Immune-mediated hemolytic anemia associated with trimethoprim-sulphamethoxazole administration in a horse. Can Vet J. 1998;39:171. 90. Bailey E. Prevalence of anti-red blood cell antibodies in the serum and colostrum of mares and its relationship to neonatal isoerythrolysis. Am J Vet Res. 1917;43:1982. 91. Geor R.J., Clark E.G., Haines D.M., et al. Systemic lupus erythematosus in a filly. J Am Vet Med Assoc. 1990;197:1489. 92. Messer N.T., Arnold K. Immune-mediated hemolytic anemia in a horse. J Am Vet Med Assoc. 1991;198:1415. 93. Pierce K.R., Joyce J.R., England R.B., et al. Acute hemolytic anemia caused by wild onion poisoning in horses. J Am Vet Med Assoc. 1972;160:323. 94. Fettman M. Comparative aspects of glutathione metabolism affecting individual susceptibility to oxidant injury. Compend Contin Educ Pract Vet. 1991;13:1079. 95. Clark B., Morrissey R. Relation of methemoglobin to hemolysis. Blood. 1951;6:532. 96. Harvey J.W. Pathogenesis, laboratory diagnosis, and clinical implications of erythrocyte enzyme deficiencies in dogs, cats and horses. Vet Clin Pathol. 2006;35:144-156. 97. Dixon P.M., McPherson E.A., Muir A. Familial methaemoglobinaemia and haemolytic anaemia in the horse associated with decreased erythrocytic glutathione reductase and glutathione. Equine Vet J. 1977;9:198. 98. Divers T.J., George L.W., George J.W. Hemolytic anemia in horses after the ingestion of red maple leaves. J Am Vet Med Assoc. 1982;180:300. 99. George L.W., Divers T.J., Mahaffey E.A., et al. Heinz body anemia and methemoglobinemia in ponies given red maple (Acer rubrum L.) leaves. Vet Pathol. 1982;19:521. 100. Corriher C.A., Parviainen A.K.J., Gibbons D.S., et al. Equine red maple leaf toxicosis. Compend Contin Educ Pract Vet. 1999;21:74. 101. Alward A., Corriher C.A., Barton M.H., et al. Red maple (Acer rubrum) leaf toxicosis in horses: a retrospective study of 32 cases. J Vet Intern Med. 2006;20:1197-1201. 102. Boyer J.D., Breeden D.C., Brown D.L. Isolation, identi cation, and characterization of compounds from Acer rubrum capable of oxidizing equine erythrocytes. Am J Vet Res. 2002;63:604-610. 103. McConnico R., Brownie C. The use of ascorbic acid in the treatment of 2 cases of red maple (Acer rubrum)—poisoned horses,. Cornell Vet. 1992;82:293. 104. Cullison R. Acetaminophen toxicosis in small animals: clinical signs, mode of action, and treatment. Compend Contin Educ Pract Vet. 1984;6:315. 105. Thorp F., Harsefield G. Onion poisoning in horses. J Am Vet Med Assoc. 1939;94:52. 106. McSherry B., Roe C., Milne F. The hematology of phenothiazine poisoning in horses,. Can Vet J. 1966;7:3. 107. Purchase H. Phenothiazine poisoning in a Thoroughbred racing stable, J S Afr Vet Med Assoc. 1961;32:403. 108. Wright M., Davison K. Nitrate accumulation in crops and nitrate poisoning in animals. Adv in Agronomy. 1964;16:197. 109. Davidson W., Doughty J., Bolton J. Nitrate poisoning of livestock. Can J Comp Med. 1941;5:303. 110. Rollins J.B., Wigton D.H., Clement T.H. Heinz body anemia associated with lymphosarcoma in a horse. Vet Clin North Am Equine Pract. 1991;13:20. 111. Norman T.E., Chaffin M.K., Johnson M.C., et al. Intravascular hemolysis associated with severe cutaneous burn injuries in five horses. J Am Vet Med Assoc. 2005;226:2039-2043. 112. Clabough D. Equine infectious anemia: the clinical signs, transmission, and diagnostic procedures. Vet Med. 1990;85:1007. 113. Clabough D. The immunopathogenesis and control of equine infectious anemia,. Vet Med. 1990;85:1020. 114. McGuire T.C., Henson J.B., Quist S.E. Viral-induced hemolysis in equine infectious anemia. Am J Vet Res. 1969;30:2091. 115. McGuire T.C., Henson J.B., Quist S.E. Impaired bone marrow response in equine infectious anemia. Am J Vet Res. 1969;30:2099. 116. Sentsui H., Kono Y. Complement-mediated hemolysis of horse erythrocytes treated with equine infectious anemia virus. Arch Virol. 1987;95:53. 117. Perryman L.E., O’Rourke K.I., McGuire T.C. Immune responses are required to terminate viremia in equine infectious anemia lentivirus infection. J Virol. 1988;62:3073. 118. Clabough D.L., Gebhard D., Flaherty M.T., et al. Immune-mediated thrombocytopenia in horses infected with equine infectious anemia virus. J Virol. 1991;65:6242. 119. Russell K.E., Perkins P.C., Ho man M.R., et al. Platelets from thrombocytopenic ponies acutely infected with equine infectious anemia virus are activated in vivo and hypofunctional. Virology. 1999;259:7. 120. Crawford T.B., Wardrop K.J., Tornquist S.J., et al. A primary production deficit in the thrombocytopenia of equine infectious anemia,. J Virol. 1996;70:7842. 121. Russell K.E., Walker K.M., Miller R.T., et al. Hyperglobulinemia and lymphocyte subset changes in naturally infected, inapparent carriers of equine infectious anemia virus. Am J Vet Res. 1998;59:1009. 122. Oaks J.L., McGuire T.C., Ulibarri C., et al. Equine infectious anemia virus is found in tissue macrophages during subclinical infection. J Virol. 1998;72:7263. 123. Oaks J.L., Ulibarri C., Crawford T.B. Endothelial cell infection in vivo by equine infectious anaemia virus. J Gen Virol. 1999;80:2393. 124. Clabough D., Perry S., Coggins L., et al. Wild-type equine infectious anemia virus replicates in vivo predominantly in tissue macrophages, not in peripheral blood monocytes. J Virol. 1992;66:5906. 125. Kemen M.J.Jr., Coggins L. Equine infectious anemia: transmission from infected mares to foals. J Am Vet Med Assoc. 1972;161:496. 126. Burns S.J. Equine infectious anemia: plasma clearance times of passively transferred antibody in foals. J Am Vet Med Assoc. 1974;164:64. 127. Tashjian R. Transmission and clinical evaluation of an equine infectious anemia herd and their offspring over a 13-year period. J Am Vet Med Assoc. 1984;184:282. 128. Burki F., Rossmanith E. Comparative evaluation of the agar gel immunodi usion test and two commercial ELISA kits for the serodiagnosis of equine infectious anemia. Zentralbl Veterinarmed A. 1990;37:448. 129. Issel C.J., Adams W.V.Jr. Detection of equine infectious anemia virus in a horse with an equivocal agar gel immunodiffusion test reaction. J Am Vet Med Assoc. 1982;180:276.

130. McConnell S., Katada M., Darnton S. Occult equine infectious anemia in an immunosuppressed serologically negative mare. Vet Clin North Am Equine Pract. 1983;5:32. 131. Tashjian R.J., editor. Equine infectious anemia: a review of policies, programs, and future objectives. Amarillo, Tex: American Quarter Horse Association, 1985. 132. Hall R.F., Pursell A.R., Cole J.R.Jr., et al. A propagating epizootic of equine infectious anemia on a horse farm,. J Am Vet Med Assoc. 1988;193:1082. 133. Moltinann H., Melhourn H., Schein G. Ultrastructural study of the development of Babesia equi (Coccidia: Piroplasmia) in the salivary glands of its vector ticks. J Protozool. 1983;30:218. 134. Holbrook A.A. Biology of equine piroplasmosis. J Am Vet Med Assoc. 1969;155:453. 135. Kumar S., Kumar R., Gupta A.K., et al. Passive transfer of Theileria equi antibodies to neonate foals of immune tolerant mares. Vet Parasitol. 2008;151:80-85. 136. Taylor W.M., Bryant J.E., Anderson J.B., et al. Equine piroplasmosis in the United States—a review. J Am Vet Med Assoc. 1969;155:915. 137. Diana A., Guglielmini C., Candini D., et al. Cardiac arrhythmias associated with piroplasmosis in the horse: a case report. Vet J. 2007;174:193-195. 138. Ogunremi O., Halbert G., Mainar-Jaime R., et al. Accuracy of an indirect uorescent-antibody test and of a complement- xation test for the diagnosis of Babesia caballi in eld samples from horses. Prev Vet Med. 2008;83:41-51. 139. Knowles D.P., Perryman L.E., McElwain T.F., et al. Conserved recombinant antigens of Anaplasma marginale and Babesia equi for serologic diagnosis. Vet Parasitol. 1995;57:9396. 140. Brüning A. Equine piroplasmosis an update on diagnosis, treatment and prevention. Br Vet J. 1996;152:139-151. 141. Kirkham W.W. The treatment of equine babesiosis,. J Am Vet Med Assoc. 1969;155:457. 142. Butler C.M., Nijhof A.M., van der Kolk J.H., et al. Repeated high dose imidocarb dipropionate treatment did not eliminate Babesia caballi from naturally infected horses as determined by PCR-reverse line blot hybridization. Vet Parasitol. 2008;151:320-322. 143. Kuttler K.L., Zaugg J.L., Gipson C.A. Imidocarb and parvaquone in the treatment of piroplasmosis (Babesia equi) in equids. Am J Vet Res. 1987;48:1613. 144. Frerichs W.M., Allen P.C., Holbrook A.A. Equine piroplasmosis (Babesia equi): therapeutic trials of imidocarb dihydrochloride in horses and donkeys. Vet Rec. 1973;93:73. 145. Singh B., Banerjee D., Guatam O. Comparative efficacy of diminazene diaceturate and imidocarb dipropionate against B. equi infection in donkeys. Vet Parasitol. 1980;7:173. 146. Zaugg J.L., Lane V.M. Evaluations of buparvaquone as a treatment for equine babesiosis (Babesia equi). Am J Vet Res. 1989;50:782. 147. Tennant B., Evans C., Kaneko J., et al. Intravascular hemolysis associated with hepatic failure in the horse. Calif Vet J. 1972;26:15. 148. Ramaiah S.K., Harvey J.W., Giguère S., et al. Intravascular hemolysis associated with liver disease in a horse with marked neutrophil hypersegmentation. J Vet Intern Med. 2003;17:360-363. 149. Colman R.W., Hirsch J., Marder V.J., et al, editors. Hemostasis and thrombosis: basic principles and clinical practice, ed 2, Philadelphia: JB Lippincott, 1987. 150. Stillions M., Teeter S., Nelson W. Utilization of dietary B12 and cobalt by mature horses. J Anim Sci. 1971;32:252. 151. Waldron-Mease E. Hypothyroidism and myopathy in racing Thoroughbreds and Standardbreds. J Equine Med Surg. 1979;3:124. 152. Kohn C.W., Jacobs R.M., Knight D., et al. Microcytosis, hypoferremia, hypoferritemia, and hypertransferrinemia in Standardbred foals from birth to 4 months of age. Am J Vet Res. 1990;51:1198. 153. Smith J.E., Cipriano J.E., DeBowes R., et al. Iron deficiency and pseudo-iron deficiency in hospitalized horses. J Am Vet Med Assoc. 1986;188:285. 154. Fleming K.A., Barton M.H., Latimer K.S. Iron deficiency anemia in a neonatal foal. J Vet Intern Med. 2006;20:1495-1498. 155. Woods P.R., Campbell G., Cowell R.L. Nonregenerative anemia associated with administration of recombinant human erythropoietin in a Thoroughbred racehorse. Equine Vet J. 1997;29:326. 156. Piercy R.J., Swardson C.J., Hinchcli K.W. Erythroid hypoplasia and anemia following administration of recombinant human erythropoietin to two horses. J Am Vet Med Assoc. 1998;212:244. 157. Lavoie J.P., Morris D.D., Zinkl J.G., et al. Pancytopenia caused by bone marrow aplasia in a horse. J Am Vet Med Assoc. 1987;191:1462. 158. Archer R., Miller W. A case of idiopathic hypoplastic anaemia in a two-year-old Thoroughbred filly,. Vet Rec. 1965;77:538. 159. Ward M., Mountan P., Dodds W. Severe idiopathic refractory anemia and leukopenia in a horse. Calif Vet J. 1982;4:19. 160. Angel K.L., Spano J.S., Schumacher J., et al. Myelophthisic pancytopenia in a pony mare. J Am Vet Med Assoc. 1991;198:1039. 161. Berggren P.C. Aplastic anemia in a horse. J Am Vet Med Assoc. 1981;179:1400. 162. Milne E.M., Pyrah I.T.G., Smith K.C., et al. Aplastic anemia in a Clydesdale foal: a case report. J Equine Vet Sci. 1995;15:129. 163. Archer R., Jeffcott L. Comparative clinical hematology. Oxford: Blackwell Scientific; 1977. 164. Athens J., Haab O., Raab S., et al. Leukokinetic studies. IV. The total blood, circulating, and marginal granulocyte pools and the granulocyte turnover rate in normal subjects,. J Clin Invest. 1961;40:989. 165. Carakostas M.C., Moore W.E., Smith J.E. Intravascular neutrophilic granulocyte kinetics in horses. Am J Vet Res. 1981;42:623. 166. Carakostas M.C., Moore W.E., Smith J.E., et al. Effects of etiocholanolone and prednisolone on intravascular granulocyte kinetics in horses. Am J Vet Res. 1981;42:626. 167. Dale D., Fauci A., Wolff S. Alternate day prednisone: leukocyte kinetics and susceptibility to infections. N Engl J Med. 1974;291:1154. 168. Prasse K.W., George L.W., Whitlock R.H. Idiopathic hypersegmentation of neutrophils in a horse. J Am Vet Med Assoc. 1981;178:303. 169. Ward D.S., Fessler J.F., Bottoms G.D., et al. Equine endotoxemia: cardiovascular, eicosanoid, hematologic, blood chemical, and plasma enzyme alterations. Am J Vet Res. 1987;48:1150. 170. Brewer B.D., Koterba A.M. Development of a scoring system for the early diagnosis of equine neonatal sepsis. Equine Vet J. 1988;20:18. 171. Searcy G., Orr J. Chronic granlulocytic leukemia in a horse. Can Vet J. 1981;22:148. 172. Jain N.C., Vegad J.L., Kono C.S. Methods for detection of immune-mediated neutropenia in horses, using antineutrophil serum of rabbit origin. Am J Vet Res. 1990;51:1026. 173. Leidl W., Cwik S., Schmid D. Neonatal isoimmune leukopenia in foals. Berl Munch Tierarztl Wochenschr. 1980;93:141. 174. Davis E.G., Rush B., Bain F., Clark-Price S., et al. Neonatal neutropenia in an Arabian foal. Equine Vet J. 2003;35:517-520. 175. Kohn C., Couto G., Swardson S., et al. Myeloid hypoplasia in related Standardbreds. J Vet Intern Med. 1992;6:133. 176. Madigan J.E., Zinkl J.G., Fridmann D.M., et al. Preliminary studies of recombinant bovine granulocyte colony stimulating factor on haematological values in normal neonatal foals. Equine Vet J. 1994;26:159. 177. Zinkl J.G., Madigan J.E., Fridmann D.M., et al. Haematological, bone marrow and clinical chemical changes in neonatal foals given canine recombinant granulocyte-colony stimulating factor. Equine Vet J. 1994;26:313. 178. Sullivan K.E., Snyder J.R., Madigan J.E., et al. Effects of perioperative granulocyte colony stimulating factor in horses with large colon ischemia. Vet Surg. 1994;22:343.

179. Burguez P.N., Ousey J., Cash R.S., et al. Changes in blood neutrophil and lymphocyte counts following administration of cortisol to horses and foals. Equine Vet J. 1983;15:58. 180. Morris D.D., Bloom J.C., Roby K.A., et al. Eosinophilic myeloproliferative disorder in a horse. J Am Vet Med Assoc. 1984;185:993. 181. Duckett W.M., Matthews H.K. Hypereosinophilia in a horse with intestinal lymphosarcoma. Can Vet J. 1997;38:719. 182. McFarlane D., Sellon D.C., Ga ney D., et al. Hematologic and serum biochemical variables and plasma corticotropin concentration in healthy aged horses. Am J Vet Res. 1998;59:1247. 183. McFarlane D., Sellon D.C., Gibbs S.A. Age-related quantitative alterations in lymphocyte subsets and immunoglobulin isotypes in healthy horses. Am J Vet Res. 2001;62:1413. 184. Ringger N.C., Edens L., Bain P., et al. Acute myelogenous leukaemia in a mare. Aust Vet J. 1997;75:329-331. 185. Johansson A.M., Skidell J., Lilliehöök I., et al. Chronic granulocytic leukemia in a horse. J Vet Intern Med. 2007;21:1126-1129. 186. Clark P., Cornelisse C.J., Schott H.C., et al. Myeloblastic leukaemia in a Morgan horse mare. Equine Vet J. 1999;31:446-448. 187. Brumbaugh G.W., Stitzel K.A., Zinkl J.G., et al. Myelomonocytic myeloproliferative diseases in a horse. J Am Vet Med Assoc. 1982;180:313. 188. Boudreaux M.K., Blue J.T., Durham S.K., et al. Intravascular leukostasis in a horse with myelomonocytic leukemia. Vet Pathol. 1984;21:544. 189. Blue J., Perdrizet J., Brown E. Pulmonary aspergillosis in a horse with myelomonocytic leukemia. J Am Vet Med Assoc. 1987;190:1562. 190. Spier S.J., Madewell B.R., Zinkl J.G., et al. Acute myelomonocytic leukemia in a horse. J Am Vet Med Assoc. 1986;188:861. 191. Burkhardt E., Saldern F.V., Huskamp B. Monocytic leukemia in a horse. Vet Pathol. 1984;21:394. 192. Lester G.D., Alleman A.R., Raskin R.E., et al. Pancytopenia secondary to lymphoid leukemia in three horses. J Vet Intern Med. 1993;7:360-363. 193. McClure J.T. Leukoproliferative disorders in horses. Vet Clin North Am Equine Pract. 2000;16:165-182. 194. Theilen G.H., Madewell B.R. Veterinary cancer medicine. Philadelphia: Lea & Febiger; 1979. 195. van den Hoven R., Franken P. Clinical aspects of lymphosarcoma in the horse. Equine Vet J. 1983;15:49. 196. Neufeld J.L. Lymphosarcoma in the horse: a review. Can Vet J. 1973;14:129. 197. Mackey V.S., Wheat J.D. Reflections on the diagnostic approach to multicentric lymphosarcoma in an aged Arabian mare. Equine Vet J. 1985;17:467. 198. Haley P.J., Spraker T. Lymphosarcoma in an aborted equine fetus. Vet Pathol. 1983;20:647. 199. Tomlinson M.J., Doster A.R., Wright E.R. Lymphosarcoma with virus-like particles in a neonatal foal. Vet Pathol. 1983;16:629. 200. Savage C.J. Lymphoproliferative and myeloproliferative disorders. Vet Clin North Am Equine Pract. 1998;14:563-578. 201. Dascanio J.J., Zhang C.H., Antczak D.F., et al. Differentiation of chronic lymphocytic leukemia in the horse. A report of two cases,. J Vet Intern Med. 1992;6:225-229. 202. McClure J.T., Young K.M., Fiste M., et al. Immunophenotypic classification of leukemia in 3 horses. J Vet Intern Med. 2001;15:144-152. 203. Rendle D.I., Durham A.E., Thompson J.C., et al. Clinical, immunophenotypic and functional characterization of T-cell leukaemia in six horses. Equine Vet J. 2007;39:522-528. 204. Kannegieter N.J., Alley M.R. Ataxia due to lymphosarcoma in a young horse. Aust Vet J. 1987;64:377. 205. Marayama K, Swearingen GR, Dmochowsku L, et al: Herpes type virus and type C particles in spontaneous equine lymphoma. In Proceedings of the Twenty-Eighth Annual Meeting of the Electron Microscope Society Of America, 1970, p 162. 206. Meyer J., DeLay J., Bienzle D. Clinical, laboratory, and histopathologic features of equine lymphoma. Vet Pathol. 2006;43:914-924. 207. Perryman L.E., Wyatt C.R., Magnuson N.S. Biochemical and functional characterization of lymphocytes from a horse with lymphosarcoma and IgM de ciency. Comp Immunol Microbiol Infect Dis. 1984;7:53. 208. Dopson L.C., Reed S.M., Roth J.A., et al. Immunosuppression associated with lymphosarcoma in 2 horses. J Am Vet Med Assoc. 1983;182:1239. 209. Furr M.O., Crisman M.V., Robertson J., et al. Immunodeficiency associated with lymphosarcoma in a horse. J Am Vet Med Assoc. 1992;201:307. 210. Ansar Ahmed S., Furr M., Chickering W.R., et al. Immunologic studies of a horse with lymphosarcoma. Vet Immunol Immunopathol. 1993;38:229. 211. Perkins G.A., Nydam D.V., Flaminio M.J., et al. Serum IgM concentrations in normal, t horses and horses with lymphoma or other medical conditions. J Vet Intern Med. 2003;17:337-342. 212. Grindem C.B., Roberts M.C., McEntee M.F. Large granular lymphocyte tumour in a horse. Vet Pathol. 1989;26:86. 213. Rebhun W.C., Bertone A.L. Equine lymphosarcoma. J Am Vet Med Assoc. 1984;184:720. 214. Schalm O.W. Lymphosarcoma in the horse. Vet Clin North Am Equine Pract. 1981;3:23. 215. Lawn K. Sudden death due to thoracic lymphoma in a Standardbred racing horse. Can Vet J. 2005;46:528-529. 216. Mair T.S., Lane J.G., Laucke V.M. Clinicopathological features of lymphosarcoma involving the thoracic cavity in the horse. Equine Vet J. 1985;17:428. 217. Garber J.L., Reef V.B., Reimer J.M. Sonographic findings in horses with mediastinal lymphosarcoma: 13 cases (1985-1992). J Am Vet Med Assoc. 1994;205:1432. 218. Saulez M.N., Schlipf J.W., Cebra C.K., et al. Use of chemotherapy for treatment of a mixed cell thoracic lymphoma in a horse. J Am Vet Med Assoc. 2004;224:733-738. 219. Bertone A.L., Yovich J.V., McIlwraith C.W. Surgical resection of intestinal lymphosarcoma in a mare. Compend Contin Educ Pract Vet. 1985;7:S506. 220. Traub-Dargatz J.L., Bayly W.M., Reed S.M., et al. Intraabdominal neoplasia as a cause of chronic weight loss in the horse. Compend Contin Educ Pract Vet. 1983;5:S526. 221. Wiseman A., Petrie L., Murray M. Diarrhea in the horse as a result of alimentary lymphosarcoma. Vet Rec. 1974;95:454. 222. Sheahan B.J., Atkins G.J., Russell R.J., et al. Histiolymphocytic lymphosarcoma in the subcutis of two horses. Vet Pathol. 1980;17:123. 223. Henson K.L., Alleman A.R., Cutler T.J., et al. Regression of subcutaneous lymphoma following removal of an ovarian granulosa theca cell tumor in a horse. J Am Vet Med Assoc. 1998;212:1419. 224. Lane P.C. Palatine lymphosarcoma in two horses. Equine Vet J. 1985;17:465. 225. Adams R., Calderwood-Mays M.B., Peyton L.C. Malignant lymphoma in three horses with ulcerative pharyngitis. J Am Vet Med Assoc. 1988;193:674. 226. Esplin D.J., Taylor J.L. Hypercalcemia in a horse with lymphosarcoma. J Am Vet Med Assoc. 1977;170:180. 227. Murphy C.J., Lavoie J.P., Groff J., et al. Bilateral eyelid swelling attributable to lymphosarcoma in the horse. J Am Vet Med Assoc. 1989;194:939. 228. Rousseaux C.G., Doige C.E., Tuddenham T.J. Epidural lymphosarcoma with myelomalacia in a seven-year-old Arabian gelding. Can Vet J. 1989;30:751. 229. Shamis L.B., Everitt J.I., Baker G.J. Lymphosarcoma as the cause of ataxia in a horse. J Am Vet Med Assoc. 1984;184:1517. 230. Rebhun W.C., Del Piero F. Ocular lesions in horses with lymphosarcoma: 21 cases (1977-1997). J Am Vet Med Assoc. 1998;212:852. 231. Lester G.D., MacKay R.J., Smith-Meyer B. Primary meningeal lymphoma in a horse. J Am Vet Med Assoc. 1992;201:1219. 232. Freeman S.L., England G.C., Bjornson S., et al. Uterine T cell lymphoma in a mare, with multicentric involvement. Vet Rec. 1997;141:391-393. 233. Held J.P., McCracken M.D., Toal R., et al. Epididymal swelling attributable to generalized lymphosarcoma in a stallion. J Am Vet Med Assoc. 1992;201:1913-1915.

234. Coumbe K.M. Primary parotid lymphoma in a 10 year old Hanoverian gelding. Eq Vet Educ. 1994;6:91-94. 235. Sweeney R.W., Hamir A.N., Fisher R.R. Lymphosarcoma with urinary bladder infiltration in a horse. J Am Vet Med Assoc. 1991;199:1177-1178. 236. Gerard M.P., Healy L.N., Bowman K.F., et al. Cutaneous lymphoma with extensive periarticular involvement in a horse. J Am Vet Med Assoc. 1998;213:391-393. 237. Raidal S.L., Clark P., Raidal S.R. Angiotrophic T-cell lymphoma as a cause of regenerative anemia in a horse. J Vet Intern Med. 2006;20:1009-1013. 238. Madewell B.R., Carlson G.P., MacLachlan N.J., et al. Lymphosarcoma with leukemia in a horse. Am J Vet Res. 1982;43:807. 239. Roberts M.C. A case of primary lymphoid leukaemia in a horse,. Equine Vet J. 1977;9:216. 240. Bernard W.V., Sweeney C.R., Morris C.R., et al. Primary lymphocytic leukemia in a horse. Vet Clin North Am Equine Pract. 1988;10:24. 241. Green P.D., Donovan L.A. Lymphosarcoma in a horse. Can Vet J. 1977;18:257. 242. Roberts M.C., Pinsent P.J.N. Malabsorption in the horse associated with alimentary lymphosarcoma. Equine Vet J. 1975;7:166. 243. Finley M.R., Rebhun W.C., Dee A., et al. Paraneoplastic pruritus and alopecia in a horse with diffuse lymphoma. J Am Vet Med Assoc. 1998;213:102. 244. Mair T.S., Yeo S.P., Lucke V.M. Hypercalcaemia and soft tissue mineralization associated with lymphosarcoma in two horses. Vet Rec. 1990;126:99. 245. Moore B.R., Weisbrode S.E., Biller D.S., et al. Metacarpal fracture associated with lymphosarcoma-induced osteolysis in a horse. J Am Vet Med Assoc. 1995;207:208. 246. Marr C.M., Love S., Pirie H.M. Clinical, ultrasonographic and pathological ndings in a horse with splenic lymphosarcoma and pseudohyperparathyroidism. Equine Vet J. 1989;21:221. 247. Robinson N.E., editor. Current therapy in equine medicine, ed 6, St Louis: WB Saunders, 2009. 248. La Perle K.M., Piercy R.J., Long J.F., et al. Multisystemic, eosinophilic, epitheliotropic disease with intestinal lymphosarcoma in a horse. Vet Pathol. 1998;35:144. 249. McConnel S., Katada M., Fiske R.A., et al. Equine lymphosarcoma diagnosed as equine infectious anemia in a young horse. Equine Vet J. 1982;14:160. 250. Lock P.F., Macy D.W. Equine ovarian lymphosarcoma. J Am Vet Med Assoc. 1979;175:72. 251. Thatcher C.D., Roussel A.J., Chickering W.R., et al. Pleural effusion with thoracic lymphosarcoma in a mare. Compend Contin Educ Pract Vet. 1985;7:S726. 252. Taylor S.D., Pusterla N., Vaughan B., et al. Intestinal neoplasia in horses. J Vet Intern Med. 2006;20:1429-1436. 253. Zicker S.C., Wilson W.D., Medearis I. Di erentiation between intra-abdominal neoplasms and abscesses in horses, using clinical and laboratory data: 40 cases (1973-1988). J Am Vet Med Assoc. 1990;196:1130-1134. 254. Hambright M.B., Meuten D.J., Scrutchfield W.L. Equine lymphosarcoma. Compend Contin Educ Pract Vet. 1983;5:S53. 255. Allen B.V., Wannop C.C., Wright I.M. Multicentric lymphosarcoma with lymphoblastic leukemia in a young horse. Vet Rec. 1984;115:130. 256. Chaffin M.K., Schmitz D.G., Brumbaugh G.W., et al. Ultrasonographic characteristics of splenic and hepatic lymphosarcoma in three horses. J Am Vet Med Assoc. 1992;201:743. 257. Dabareiner R.M., Sullins K.E., Goodrich L.R. Large colon resection for treatment of lymphosarcoma in two horses. J Am Vet Med Assoc. 1996;208:895. 258. Henson F.M.D., Dixon P.M., Dobson J.M. Treatment of 4 cases of equine lymphoma with megavoltage radiation. Equine Vet Educ. 2004;16:312-314. 259. Mair T.S., Couto C.G. The use of cytotoxic drugs in equine practice,. Equine Vet Educ. 2006;18:149-156. 260. Gollagher R.D., Zoila B., Chelack B.J., et al. Immunotherapy of equine cutaneous lymphosarcoma using low dose cyclophosphamide and autologous tumor cells infected with vaccinia virus. Can Vet J. 1993;34:371-373. 261. Littlewood J.D., Whitwell K.E., Day M.J. Equine cutaneous lymphoma: a case report. Vet Dermatol. 1995;6:105. 262. Cornelius C.E., Goodbary R.F., Kennedy P.C. Plasma cell myelomatosis in a horse. Cornell Vet. 1959;49:478. 263. Drew R.A., Greatorex J.C. Vertebral plasma cell myeloma causing posterior paralysis in a horse. Equine Vet J. 1974;6:131. 264. Henry M., Prasse K., White S. Hemorrhagic diathesis caused by multiple myeloma in a three-month-old foal. J Am Vet Med Assoc. 1989;194:392. 265. MacAllister C., Qualls C., Ryler L., et al. Multiple myeloma in a horse. J Am Vet Med Assoc. 1987;191:337. 266. Markel M.D., Dorr T.E. Multiple myeloma in a horse. J Am Vet Med Assoc. 1986;188:621. 267. Goldman L., Ausiello D.A., editors. Cecil textbook of medicine, ed 23, Philadelphia: WB Saunders, 2008. 268. Seide R.K., Jacobs R.M., Dobblestein T.N., et al. Characterization of a hom*ogenous paraprotein from a horse with spontaneous multiple myeloma syndrome. Vet Immunol Immunopathol. 1987;17:69. 269. Barton M.H., Sharma P., LeRoy B.E., et al. Hypercalcemia and high serum parathyroid hormone-related protein concentration in a horse with multiple myeloma. J Am Vet Med Assoc. 2004;225:409-413. 270. Kim D.Y., Taylor H.W., Eades S.C., et al. Systemic AL amyloidosis associated with multiple myeloma in a horse. Vet Pathol. 2005;42:81-84. 271. Jacobs R.M., Kociba G.J., Ruoff W.W. Monoclonal gammopathy in a horse with defective hemostasis. Vet Pathol. 1983;20:643. 272. Thrall M.A. Lymphoproliferative disorders: lymphocytic leukemia and plasma cell myeloma. Vet Clin North Am Small Anim Pract. 1981;11:321. 273. Pusterla N., Stacy B.A., Vernau W., et al. Immunoglobulin A monoclonal gammopathy in two horses with multiple myeloma. Vet Rec. 2004;155:19-23. 274. Jackson M. Platelet physiology and platelet function: inhibition by aspirin. Compend Contin Educ Pract Vet. 1987;9:627. 275. Gerrard J. Platelet aggregation: cellular regulation and physiologic role. Hosp Pract. 1988;23:89. 276. Segura D., Monreal L., Perez-Pujol S., et al. Assessment of platelet function in horses: ultrastructure, ow cytometry, and perfusion techniques. J Vet Intern Med. 2006;20(3):581588. 277. Meyers K.M., Lindner C., Grant B. Characterization of the equine platelet aggregation response. Am J Vet Res. 1979;40:260. 278. Coyne C.P., Kelly A.B., Hornof W.J., et al. Radiolabeling of equine platelets in plasma with 111-In-(2-mercaptopyridine-N-oxide) and their in vivo survival. Am J Vet Res. 1987;48:385. 279. Sellon D.C., Grindem C.B. Quantitative platelet abnormalities in horses. Compend Contin Educ Pract Vet. 1994;16:1335. 280. Finocchio E., Coffman J., Osbaldiston G. Platelet counts in horses. Cornell Vet. 1960;60:518. 281. Byars T.D., Greene C.E. Idiopathic thrombocytopenic purpura in the horse. J Am Vet Med Assoc. 1982;180:1422. 282. Larson V.L., Perman V., Stevens J.B. Idiopathic thrombocytopenic purpura in two horses. J Am Vet Med Assoc. 1983;183:328. 283. Sockett D.C., Traub-Dargatz J., Weiser M.G. Immune-mediated hemolytic anemia and thrombocytopenia in a foal. J Am Vet Med Assoc. 1987;190:308. 284. Werner L.L., Gross T.L., Hillidge C.J. Acute necrotizing vasculitis and thrombocytopenia in a horse. J Am Vet Med Assoc. 1984;185:87. 285. Morris D.D., Whitlock R.H. Relapsing idiopathic thrombocytopenia in a horse. Equine Vet J. 1983;15:73. 286. Humber K.A., Beech J., Cudd T.A., et al. Azathioprine for treatment of immune-mediated thrombocytopenia in two horses. J Am Vet Med Assoc. 1991;199:591. 287. Hinchcliff K.W., Kociba G.J., Mitten L. EDTA-dependent pseudothrombocytopenia in a horse. J Am Vet Med Assoc. 1993;203:1715.

288. Kingston J.K., Bayly W.M., Sellon D.C., et al. E ects of sodium citrate, low molecular weight heparin, and prostaglandin E1 on aggregation, brinogen binding, and enumeration of equine platelets. Am J Vet Res. 2001;62:547. 289. Sellon D.C., Levine J., Millikin E., et al. Thrombocytopenia in horses: 35 cases (1989-1994). J Vet Intern Med. 1996;10:127. 290. Russell K.E., Perkins P.C., Grindem C.B., et al. Flow cytometric method for detecting thiazole orange-positive (reticulated) platelets in thrombocytopenic horses. Am J Vet Res. 1997;58:1092. 291. Duncan S.G., Meyers K.M., Reed S.M. Reduction of the red blood cell mass of horses: toxic effect of heparin anticoagulant therapy. Am J Vet Res. 1983;44:2271. 292. Bell W. Thrombocytopenia occurring during heparin therapy. N Engl J Med. 1976;295:276. 293. Myers T., Kim B., Steiner M., et al. Platelet-associated complement C3 in immune thrombocytopenic purpura. Blood. 1982;59:1023. 294. Nel J.D., Stevens K., Mouton A., et al. Platelet-bound IgM in autoimmune thrombocytopenia. Blood. 1983;61:119. 295. Court W.S., Bozeman J.M., Soong S.J., et al. Platelet surface-bound IgG in patients with immune and nonimmune thrombocytopenia. Blood. 1987;69:278. 296. Neiman J., Mant M., Shnitka T. Phagocytosis of platelets by Kupffer cells in immune thrombocytopenia. Arch Pathol Lab Med. 1987;111:563. 297. Ho man R., Zaknoen S., Yang H., et al. An antibody cytotoxic to megakaryocyte progenitor cells in a patient with immune throbocytopenic purpura. N Engl J Med. 1985;312:1170. 298. McGurrin M.K., Arroyo L.G., Bienzle D. Flow cytometric detection of platelet-bound antibody in three horses with immune-mediated thrombocytopenia. J Am Vet Med Assoc. 2004;224:83-87. 299. Buechner-Maxwell V., Scott M.A., Godber L., et al. Neonatal alloimmune thrombocytopenia in a Quarter Horse foal. J Vet Intern Med. 1997;11:304. 300. Ramirez S., Gaunt S.D., McClure J.J., et al. Detection and e ects on platelet function of antiplatelet antibody in mule foals with experimentally induced neonatal alloimmune thrombocytopenia. J Vet Intern Med. 1999;13:534. 301. White S.D., Maxwell L.K., Szabo N.J., et al. Pharmaco*kinetics of azathioprine following single-dose intravenous and oral administration and e ects of azathioprine following chronic oral administration in horses. Am J Vet Res. 2005;66:1578-1583. 302. Fey K., Sasse H.L. Relapsing immune-mediated thrombocytopenia of unknown origin in a stallion. Equine Vet Educ. 1998;10:127. 303. Sellon D.C. Thrombocytopenia in horses. J Vet Intern Med. 1998;10:133. 304. Traub-Dargatz J.L., McClure J.J., Koch C., et al. Neonatal isoerythrolysis in mule foals. J Am Vet Med Assoc. 1995;206:67. 305. Tornquist S.J., Crawford T.B. Suppression of megakaryocyte colony growth by plasma from foals infected with equine infectious anemia virus. Blood. 1997;90:2357. 306. Wardrop K.J., Baszler T.V., Reilich E., et al. A morphometric study of bone marrow megakaryocytes in foals infected with equine infectious anemia virus,. Vet Pathol. 1996;33:222. 307. Nolen-Walston R.D., D’Oench S.M., Hanelt L.M., Sharkey L.C., et al. Acute recumbency associated with Anaplasma phagocytophilum infection in a horse. J Am Vet Med Assoc. 2004;224(12):1964-1966. 308. Gribble D.H. Equine ehrlichiosis. J Am Vet Med Assoc. 1969;155:462. 309. Madigan J.E., Gribble D. Equine ehrlichiosis in northern California: 49 cases (1968-1981). J Am Vet Med Assoc. 1987;190:445. 310. Ziemer E.L., Keenan D.P., Madigan J.E. Ehrlichia equi infection in a foal. J Am Vet Med Assoc. 1987;190:199. 311. Brewer B.D., Harvey J.W., Mayhew I.G., et al. Ehrlichiosis in a Florida horse. J Am Vet Med Assoc. 1984;185:446. 312. Franzén P., Aspan A., Egenvall A., et al. Acute clinical, hematologic, serologic, and polymerase chain reaction ndings in horses experimentally infected with a European strain of Anaplasma phagocytophilum. J Vet Intern Med. 2005;19:232-239. 313. Madigan J.E., Gribble D. Equine ehrlichiosis in northern California: 49 cases (1968-1981). J Am Vet Med Assoc. 1987;190:445-448. 314. Madigan J.E., Pusterla N. Ehrlichial diseases. Vet Clin North Am Equine Pract. 2000;16:487-499. 315. Franzén P., Berg A.L., Aspan A., et al. Death of a horse infected experimentally with Anaplasma phagocytophilum. Vet Rec. 2007;160:122-125. 316. Dolente B.A., Wilkins P.A., Boston R.C. Clinicopathologic evidence of disseminated intravascular coagulation in horses with acute colitis. J Am Vet Med Assoc. 2002;220:10341038. 317. Dallap B.L. Coagulopathy in the equine critical care patient. Vet Clin North Am Equine Pract. 2004;20:231-251. 318. Dallap B, Dolente B, Boston R: Coagulation pro les in 27 horses with large colon volvulus. In Proceedings of the Eleventh Annual Veterinary Meeting of the American College of Veterinary Surgeons, Chicago, 2001, p 4. 319. Valentine B.A., Ross C.E., Bump J.L., et al. Intramuscular hemangiosarcoma with pulmonary metastasis in a horse. J Am Vet Med Assoc. 1986;188:628. 320. Waugh S., Long G., Uriah L. Metastatic hemangiosarcoma in the equine: report of 2 cases. J Equine Med Surg. 1977;1:311. 321. Johns I., Stephen J.O., Del Piero F., et al. Hemangiosarcoma in 11 young horses. J Vet Intern Med. 2005;19:564-570. 322. Hargis A.M., Feldman B.F. Evaluation of hemostatic defects secondary to vascular tumors in dogs: 11 cases (1983-1988). J Am Vet Med Assoc. 1991;198:891. 323. Shim W.K. Hemangiomas of infancy complicated by thrombocytopenia. Am J Surg. 1968;116:896. 324. Sellon D.C., Levine J.F., Palmer K., et al. Thrombocytosis in 24 horses (1989-1994). J Vet Intern Med. 1997;11:24. 325. Clemmons R.M., Dorsey-Lee M.R., Gorman N.T., et al. Haemostatic mechanisms of the newborn foal: reduced platelet responsiveness. Equine Vet J. 1984;16:353. 326. Cambridge H., Lees P., Hooke R.E., et al. Antithrombotic actions of aspirin in the horse. Equine Vet J. 1991;23:123. 327. Kopp K.J., Moore J.N., Byars T.D., et al. Template bleeding time and thromboxane generation in the horse: e ects of three non-steroidal anti-in ammatory drugs. Equine Vet J. 1985;17:322. 328. Meyers K.M., Lindner C., Katz J., et al. Phenylbutazone inhibition of equine platelet function. Am J Vet Res. 1979;40:265. 329. Koterba A., Drummond W., Kosch P., editors. Equine clinical neonatology. Philadelphia: Lea & Febiger, 1990. 330. Fry M.M., Walker N.J., Blevins G.M., et al. Platelet function defect in a Thoroughbred filly. J Vet Intern Med. 2005;19:359-362. 331. Sutherland R.J., Cambridge H., Bolton J.R. Functional and morphological studies on blood platelets in a thrombasthenic horse. Aust Vet J. 1989;66:366-370. 332. Macieira S., Rivard G.E., Champagne J., et al. Glanzmann thrombasthenia in an Oldenbourg filly. Vet Clin Pathol. 2007;36:204-208. 333. Livesey L., Christopherson P., Hammond A., et al. Platelet dysfunction (Glanzmann’s thrombasthenia) in horses. J Vet Intern Med. 2005;19:917-919. 334. Christopherson P.W., Insalaco T.A., van Santen V.L., et al. Characterization of the cDNA encoding alphaIIb and beta3 in normal horses and two horses with Glanzmann thrombasthenia. Vet Pathol. 2006;43:78-82. 335. Norris J., Pratt S., Auh J.-H., et al. Investigation of a novel, heritable bleeding diathesis of Thoroughbred horses and development of a screening assay. J Vet Intern Med. 2006;20:1450-1456.

336. Norris J.W., Pratt S.M., Hunter J.F., et al. Prevalence of reduced fibrinogen binding to platelets in a population of Thoroughbreds. Am J Vet Res. 2007;68:716-721. 337. Brooks M., Leith G.S., Allen A.K., et al. Bleeding disorder (von Willebrand disease) in a Quarter Horse. J Am Vet Med Assoc. 1991;198:114. 338. George J., Shattil S. The clinical importance of acquired abnormalities of platelet function,. N Engl J Med. 1991;324:27. 339. Judson D., Barton M. Effect of aspirin on haemostasis in the horse. Res Vet Sci. 1981;30:241. 340. Trujillo O., Rios A., Maldonado R., et al. Effect of oral administration of acetylsalicylic acid on haemostasis in the horse. Equine Vet J. 1981;13:205. 341. Baxter G.M., Moore J.N. Effect of aspirin on ex vivo generation of thromboxane in healthy horses. Am J Vet Res. 1987;48:13. 342. Osterud B., Rapaport S. Activation of factor IX by the reaction product of tissue factor and factor VII: additional pathway for initiating blood coagulation. Proc Natl Acad Sci U S A. 1977;74:5. 343. Hoffman M., Monroe D.M.3rd. A cell-based model of hemostasis,. Thromb Haemost. 2001;85:958-965. 344. Rosenberg R., Lam L. Correlation between structure and function of heparin. Proc Natl Acad Sci U S A. 1979;76:1218. 345. Hoffman M. Remodeling the blood coagulation cascade. J Thromb Thrombolysis. 2003;16:17-20. 346. Christensen U., Clemmensen I. Kinetic properties of the primary inhibitor of plasmin from human plasma. Biochem J. 1977;163:389. 347. Morris D.D. Recognition and management of disseminated intravascular coagulation in horses. Vet Clin North Am Large Anim Pract. 1988;4:115. 348. Rawlings C.A., Byars T.D., Van Noy M.K., et al. Activated coagulation test in normal and heparinized ponies and horses. Am J Vet Res. 1975;36:711. 349. Wagner A.E. Transport of plasma for prothrombin time testing in monitoring warfarin therapy in the horse. J Am Vet Med Assoc. 1981;178:306. 350. Mischke R., Junker J., Deegen E. Sensitivity of commercial prothrombin time reagents to detect coagulation factor deficiencies in equine plasma. Vet J. 2006;171:114-119. 351. Topper M.J., Prasse K.W. Analysis of coagulation proteins as acute-phase reactants in horses with colic. Am J Vet Res. 1998;59:542-545. 352. Feige K., Kästner S.B., Demp e C.E., et al. Changes in coagulation and markers of brinolysis in horses undergoing colic surgery. J Vet Med A Physiol Pathol Clin Med. 2003;50:30-36. 353. Stokol T., Erb H.N., De Wilde L., et al. Evaluation of latex agglutination kits for detection of brin(ogen) degradation products and D-dimer in healthy horses and horses with severe colic. Vet Clin Pathol. 2005;34:375-382. 354. Weiss D.J., Geor R.J., Burger K. E ects of pentoxifylline on hemorheologic alterations induced by incremental treadmill exercise in Thoroughbreds. Am J Vet Res. 1996;57(9):1364-1368. 355. Stern-Balestra E. Bestimmung von löslichen Fibrinkomplexen beim Pferd als Möglichkeit der Diagnostik einer aktivierten Blutgerinnung, doctoral thesis. Zurich, Switzerland: University of Zurich; 2000. 356. Bernard W., Morris D.D., Divers T.J., et al. Plasma antithrombin-III values in healthy horses: effect of sex and/or breed. Am J Vet Res. 1987;48:866. 357. Johnstone I.B., Physick-Sheard P., Crane S. Breed, age, and gender differences in plasma antithrombin-III activity in clinically normal young horses. Am J Vet Res. 1989;50:1751. 358. Johnstone I., Petersen D., Crane S., Antithrombin III. (AT III) activity in plasmas from normal and diseased horses, and in normal canine, bovine and human plasmas. Vet Clin Pathol. 1987;16:14. 359. Johnstone I., Crane S. Haemostatic abnormalities in horses with colic—their prognostic value. Equine Vet J. 1986;18:271-274. 360. Holland M., Kelly A., Snyder J., et al. Antithrombin III activity in horses with large colon torsion. Am J Vet Res. 1986;47:897-900. 361. Asakura H., Ontachi Y., Mizutni T., et al. Decreased plasma activity of antithrombin or protein C is not due to consumption coagulopathy in septic patients with disseminated intravascular coagulation. Eur J Haematol. 2001;67:170-175. 362. Welles E.G., Prasse K.W., Duncan A., et al. Antigenic assay for protein C determination in horses. Am J Vet Res. 1990;51:1075. 363. Welles E.G., Prasse K.W., Duncan A. Chromogenic assay for equine plasminogen. Am J Vet Res. 1990;51:1080. 364. Prasse K.W., Allen D.Jr., Moore J.N., et al. Evaluation of coagulation and brinolysis during the prodromal stages of carbohydrate-induced acute laminitis in horses. Am J Vet Res. 1950;51:1990. 365. Littlewood J.D., Bevan S.A., Corke M.J., Haemophilia A. (classic haemophilia, factor VIII deficiency) in a Thoroughbred colt foal. Equine Vet J. 1991;23:70. 366. Segura D., Monreal L., Armengou L., et al. Mean platelet component as an indicator of platelet activation in foals and adult horses. J Vet Intern Med. 2007;21:1076-1082. 367. Laan T.T.J.M., Goehring L.S. Sloet van Oldruitenborgh-Oosterbaan MM: Von Willebrand’s disease in an eight-day-old quarter horse foal. Vet Rec. 2005;157:322-324. 368. Rathgeber R.A., Brooks M.B., Bain F.T., et al. von Willebrand disease in a Thoroughbred mare and foal. J Vet Intern Med. 2001;15:63. 369. Archer R. True haemophilia (haemophilia A) in a Thoroughbred foal. Vet Rec. 1961;73:338. 370. Archer R.K., Allen B.V. True haemophilia in horses. Vet Rec. 1972;91:655. 371. Feldman B., Giacopuzzi R., Hemophilia A. (factor VIII deficiency) in a colt. Vet Clin North Am Equine Pract. 1982;4:24. 372. Hinton M., Jones D.R., Lewis I.M., et al. A clotting defect in an Arab colt foal,. Equine Vet J. 1977;9:1. 373. Henninger R.W. Hemophilia A in two related quarter horse colts. J Am Vet Med Assoc. 1988;193:91. 374. Sanger V., Mairs R., Trapp A. Hemophilia in a foal. J Am Vet Med Assoc. 1964;144:259. 375. Mills J., Bolton J. Haemophilia A in a 3-year-old Thoroughbred horse. Aust Vet J. 1983;60:63. 376. Hutchins D., Lepherd E., Crook I. A case of equine haemophilia,. Aust Vet J. 1967;43:83. 377. Geor R.J., Jackson M.L., Lewis K.D., et al. Prekallikrein deficiency in a family of Belgian horses. J Am Vet Med Assoc. 1990;197:741. 378. Turrentine M.A., Sculley P.W., Green E.M., et al. Prekallikrein deficiency in a family of miniature horses. Am J Vet Res. 1986;47:2464. 379. Williams W., Beutler E., Erslev A., et al, editors. Hematology, ed 4, New York: McGraw Hill, 1990. 380. Ainsworth D.M., Dodds W.J., Brown C.M. Deficiency of the contact phase of intrinsic coagulation in a horse. J Am Vet Med Assoc. 1985;187:71. 381. Morris D.D. Cutaneous vasculitis in horses: 19 cases (1978-1985). J Am Vet Med Assoc. 1987;191:460. 382. Morris DD: Vasculitis in horses. In Proceedings of the Fourth Scientific Forum of the American College of Veterinary Internal Medicine, Washington, DC, 1990, p 3. 383. Huntington P., Ellis P., Forman A., et al. Equine viral arteritis. Aust Vet J. 1990;67:429. 384. Traub-Dargatz J., Ralston S., Collins J., et al. Equine viral arteritis. Compend Contin Educ Pract Vet. 1985;7:S490. 385. Crawford T, Henson J: Immuno uorescent, light microscopic and immunologic studies of equine viral arteritis. In Proceedings of the Third International Conference on Equine Infectious Diseases, Basel, Switzerland, 1972, S. Karger Press, Basel, Switzerland, p 282. 386. Neu S, Timoney P, McCollum W: Persistent infection of the reproductive tract in stallions experimentally infected with equine arteritis virus. In Proceedings of the International Conference on Equine Infectious Diseases, Lexington, Ky, 1988, University Press of Kentucky, Lexington, KY, p 149.

387. Coignoul F.L., Cheville N.F. Pathology of maternal genital tract, placenta, and fetus in equine viral arteritis. Vet Pathol. 1984;21:333. 388. Timoney P.J., McCollum W.H., Roberts A.W., et al. Status of equine viral arteritis in Kentucky. J Am Vet Med Assoc. 1985;191(36):1987. 389. Timoney P., McCollum W., Roberts A., et al. Demonstration of the carrier state in naturally acquired equine arteritis virus infection in the stallion. Res Vet Sci. 1986;41:279. 390. Timoney P J: Equine viral arteritis in perspective: fact vs. fiction. In Proceedings of the Sixth World Equine Veterinary Congress, September 30-October 3, 1999, pp 167-170. 391. Timoney P.J. The increasing significance of international trade in equids and its influence on the spread of infectious diseases,. Ann N Y Acad Sci. 2000;916:55-60. 392. Timoney P.J., McCollum W.H., Vickers M.L. The rationale for greater national control of EVA,. Equine Dis Q. 1998;7:2-3. 393. Glaser A.L., Rottier P.J.M., Horzinek M.C., et al. Equine arteritis virus: a review of clinical features and management aspects. Vet Q. 1996;18:95-99. 394. Morris D.D., Miller W.H.Jr., Goldschmidt M.H., et al. Chronic necrotizing vasculitis in a horse. J Am Vet Med Assoc. 1983;183:579. 395. Conn D.T. Update on systemic necrotizing vasculitis. Mayo Clin Proc. 1989;4:535. 396. Yanagisawa M., Kurihara H., Kimura S., et al. A novel potent vasoconstrictor peptide produced by vascular endothelial cells,. Nature. 1988;332:411. 397. Vanhoutte P.M. The endothelium-modulator of vascular smooth-muscle tone,. N Engl J Med. 1988;319:512. 398. Schifferli J.A., Ng Y.C., Peters D.K. The role of complement and its receptor in the elimination of immune complexes,. N Engl J Med. 1986;315:488. 399. Smiley J.D., Moore S.E. Southwestern Internal Medicine Conference: role of complement and IgG-Fc receptor functions. Am J Med Sci. 1989;298:257. 400. Fearon D.T. Complement, C receptors, and immune complex disease. Hosp Pract. 1988;23:63. 401. Caciolo P.L., Hurvitz A.I., Nesbitt G.H. Michel’s medium as a preservative for immuno uorescent staining of cutaneous biopsy specimen in dogs and cats. Am J Vet Res. 1984;45:128. 402. Galan J.E., Timoney J.F. Immune complexes in purpura hemorrhagica of the horse contain IgA and M antigen of Streptococcus equi. J Immunol. 1985;135:3134. 403. Pusterla N., Watson J.L., Affolter V.K., et al. Purpura haemorrhagica in 53 horses. Vet Rec. 2003;153(4):118-121. 404. Kaese H.J., Valberg S.J., Hayden D.W., et al. Infarctive purpura hemorrhagica in five horses. J Am Vet Med Assoc. 2005;226:1893-1898. 405. Waller A.S., Jolley K.A. Getting a grip on strangles: recent progress towards improved diagnostics and vaccines. Vet J. 2007;173:492-501. 406. Ettinger S.J., Feldman E.C., editors. Textbook of veterinary internal medicine, ed 6, St Louis: WB Saunders, 2005. 407. Siciliano P.D., Warren L.K., Lawrence L.M. Changes in vitamin K status of growing horses. J Equine Vet Sci. 2000;20:726-729. 408. Ralston S. Equine clinical nutrition: specific problems and solutions. Compend Contin Educ Pract Vet. 1988;10:357. 409. Colles C.M. A preliminary report on the use of warfarin in the treatment of navicular disease,. Equine Vet J. 1979;11:187. 410. Scott E.A., Byars T.D., Lamar A.M. Warfarin anticoagulation in the horse. J Am Vet Med Assoc. 1980;177:1146. 411. Scott E.A., Sandler G.A., Byars T.D. Warfarin: effects on anticoagulant, hematologic, and blood enzyme values in normal ponies. Am J Vet Res. 1979;40:142. 412. Vrins A., Carlson G., Feldman B. Warfarin: a review with emphasis on its use in the horse. Can Vet J. 1983;24:211. 413. Byars T.D., Greene C.E., Kemp D.T. Antidotal effect of vitamin K1 against warfarin-induced anticoagulation in horses. Am J Vet Res. 1986;47:2309. 414. Scott E.A., Sandler G.A., Byars T.D. Warfarin: effects of intravenous loading doses and vitamin K on warfarin anticoagulation in the pony. Am J Vet Res. 1888;39:1978. 415. Boermans H.J., Johnstone I., Black W.D., et al. Clinical signs, laboratory changes and toxico*kinetics of brodifacoum in the horse. Can J Vet Res. 1991;55:21. 416. McConnico R.S., Copedge K., Bischoff K.L. Brodifacoum toxicosis in two horses. J Am Vet Med Assoc. 1997;211:882. 417. Ayala I., Rodriguez J., Martos N., et al. Fatal brodifaccum poisoning in a pony. Can Vet J. 2007;48:627-629. 418. Blood D.C., Radostit* O.M., Arundel J.H., et al. Veterinary medicine, ed 7. Philadelphia: Balliere Tindall; 1989. 419. Burrows G.E., Tyrl R.J. Plants causing sudden death in livestock. Vet Clin North Am Food Anim Pract. 1989;5:263. 420. Osweiler G.D., Rurh L.P. Plants affecting blood coagulation. In Howard J.S., editor: Current veterinary therapy food animal practice, ed 2, Philadelphia: WB Saunders, 1986. 421. Rebhun W.C., Tennant B.C., Dill S.G., et al. Vitamin K 3-induced renal toxicosis in the horse. J Am Vet Med Assoc. 1984;184:1237. 422. Byars T.D., Wilson R.C. Clinical pharmacology of heparin. J Am Vet Med Assoc. 1981;178:739. 423. MacHarg M.A., Becht J.L. The pharmacology of heparin prophylaxis,. J Am Vet Med Assoc. 1983;183:129. 424. Hood D. Current concepts of the physiopathology of laminitis. Proc Am Assoc Eq Pract. 1979;25:13-20. 425. Hood D., Gremmel S., Amoss M. Equine laminitis. III. Coagulation dysfunction in the developmental and acute disease. J Equine Med Surg. 1979;3:355. 426. Parker J., Fubini S., Car B., et al. Prevention of intrabdominal adhesions in ponies by low dose heparin therapy. Vet Surg. 1987;16:459. 427. Moore B.R., Hinchcliff K.W. Heparin: a review of its pharmacology and therapeutic use in horses. J Vet Intern Med. 1994;8:26-35. 428. Belknap J.K., Moore J.N. Evaluation of heparin for prophylaxis of equine laminitis: 71 cases (1980-1986). J Am Vet Med Assoc. 1989;195:505. 429. Young D., Richardson D., Markel D. The e ect of low dose heparin therapy on complication and survival rates in horses following exploratory celiotomy,. Equine Vet J Suppl. 1989;7:91. 430. Engelking L.R., Mariner J.C. Enhanced biliary bilirubin excretion after heparin-induced erythrocyte mass depletion. Am J Vet Res. 1985;46:2175. 431. Mahaffey E.A., Moore J.N. Erythrocyte agglutination associated with heparin treatment in three horses. J Am Vet Med Assoc. 1986;189:1478. 432. Moore J.N., Mahaffey E.A., Zboran M. Heparin-induced agglutination of erythrocytes in horses. Am J Vet Res. 1987;48:68. 433. Moore J., Colicchio M., Darien B. Effects of heparin on coagulation times and plasma antithrombin III in normal horses. Symp Equine Colic Res. 1986;2:347. 434. Weitz J.I. Low-molecular-weight heparins. N Engl J Med. 1997;337(10):688-698. 435. Schwarzwald C., Fiege K., Wunderli-Allenspach H., et al. Comparison of pharmaco*kinetic variables for two low molecular weight heparins after subcutaneous administration of a single dose to horses. Am J Vet Res. 2002;63:868-873. 436. Feige K., Schwarzwald C.C., Bombeli T. Comparison of unfractioned and low molecular weight heparin for prophylaxis of coagulopathies in 52 horses with colic: a randomized double-blind clinical trial. Equine Vet J. 2003;35:506-513. 437. Marshall J. Inflammation, coagulopathy, and the pathogenesis of multiple organ dysfunction syndrome. Crit Care Med. 2001;29(Suppl 7):S99-S104. 438. Johnstone I., Blackwell T. Disseminated intravascular coagulation in a horse with postpartum ulcerative colitis and laminitis. Can Vet J. 1984;25:195. 439. Johnstone I.B., Crane S. Hemostatic abnormalities in equine colic. Am J Vet Res. 1986;47:356. 440. Johnstone I.B., Crane S. Haemostatic abnormalities in horses with colic—their prognostic value. Equine Vet J. 1986;18:271. 441. Johnstone I.B., McAndrew K.H., Baird J.D. Early detection and successful reversal of disseminated intravascular coagulation in a Thoroughbred mare presented with a history of diarrhea and colic. Equine Vet J. 1986;18:337.

442. Morris D.D., Beech J. Disseminated intravascular coagulation in six horses. J Am Vet Med Assoc. 1983;183:1067. 443. Morris D.D., Messick J., Whitlock R.H., et al. Effect of equine ehrlichial colitis on the hemostatic system in ponies. Am J Vet Res. 1988;49:1030. 444. Morris D., Vaala W., Sartin E. Protein-losing enteropathy in a yearling lly with subclinical disseminated intravascular coagulation and autoimmune hemolytic disease. Compend Contin Educ Pract Vet. 1982;4:S542. 445. Holland M., Kelly A.B., Snyder J.R., et al. Antithrombin III activity in horses with large colon torsion. Am J Vet Res. 1986;47:897. 446. Lavoie J.P., Madigan J.E., Cullor J.S., et al. Haemodynamic, pathological, haematological and behavioural changes during endotoxin infusion in equine neonates. Equine Vet J. 1990;22:23. 447. Pablo L.S., Purohit R.C., Teer P.A., et al. Disseminated intravascular coagulation in experimental intestinal strangulation obstruction in ponies. Am J Vet Res. 1983;44:2115. 448. Meyers K., Reed S., Keck M., et al. Circulating endotoxin-like substance(s) and altered hemostasis in horses with gastrointestinal disorders: an interim report. Am J Vet Res. 1982;43:2233. 449. Feldman B. Disseminated intravascular coagulation. Compend Contin Educ Pract Vet. 1981;3:46. 450. van der Poll T., de Jonge E., Levi M. Regulatory role of cytokines in disseminated intravascular coagulation. Semin Thromb Hemost. 2001;27:639-651. 451. Habal F., Movat H. Kininogens of human plasma. Semin Thromb Hemost. 1976;3:27. 452. Henry M.M., Moore J.N. Clinical relevance of monocyte procoagulant activity in horses with colic. J Am Vet Med Assoc. 1991;198:843. 453. Bick R., Bick M., Fekete L. Antithrombin III patterns in disseminated intravascular coagulation. Am J Clin Pathol. 1980;73:577. 454. Green R.A., Kabel A.L. Hypercoagulable state in three dogs with nephrotic syndrome: role of acquired antithrombin III deficiency. J Am Vet Med Assoc. 1982;181:914. 455. Clouse L., Camp P. The regulation of hemostasis: the protein C system,. N Engl J Med. 1986;314:1298. 456. Semrad S.D., Hardee G.E., Hardee M.M., et al. Low dose unixin meglumine: e ects on eicosanoid production and clinical signs induced by experimental endotoxaemia in horses. Equine Vet J. 1987;19:201. 457. Broome TA, Brown MP, Gronwall RR, et al: Pharmaco*kinetics and plasma concentrations of acetylsalicylic acid after intravenous, rectal, and intragastric administration to horses, Can J Vet Res 67:297-230 458. Liska D.A., Akucewich L.H., Marsella R., et al. Pharmaco*kinetics of pentoxifylline and its 5-hydroxyhexyl metabolite after oral and intravenous administration of pentoxifylline to healthy adult horses. Am J Vet Res. 2006;67:1621-1627. 459. Dunkel B., Rickards K.J., Page C.P., et al. Phosphodiesterase isoenzymes in equine platelets and their influence on platelet adhesion. Am J Vet Res. 2007;68:1354-1360. 460. Segura D., Monreal L., Pérez-Pujol S., et al. Effects of etamsylate on equine platelets: in vitro and in vivo studies. Vet J. 2007;174:325-329. 461. Feige K., Dennler M., Kästner S.B., et al. Pharmaco*kinetics of recombinant hirudin in healthy horses. Equine Vet J. 2004;36:135-141. 462. Spurlock S.L., Spurlock G.H. Risk factors of catheter-related complications. Compend Contin Educ Pract Vet. 1990;12:214-245. 463. Armstrong C.W., Mayhall C.G., Miller K.B., et al. Prospective study of catheter placement and other factors for infection of hyperalimentation catheters. J Infect Dis. 1986;154:808-814. 464. Dolente B.A., Beech J., Lindborg S., et al. Evaluation of risk factors for development of catheter-associated jugular thrombophlebitis in horses: 50 cases (1993-1998). J Am Vet Med Assoc. 2005;227:1134-1141. 465. Gardner S.Y., Reef V.B., Spencer P.A. Ultrasonographic evaluation of horses with thrombophlebitis of the jugular vein: 46 cases (1985-1988). J Am Vet Med Assoc. 1991;199:370. 466. Stormont C., Suzuki Y., Rhode E. Serology of horse blood groups. Cornell Vet. 1964;54:439. 467. Stormont C.J. Blood groups in animals. J Am Vet Med Assoc. 1982;181:1120. 468. Niinistö K., Raekallio M., Sankari S. Storage of equine red blood cells as a concentrate. Vet J. 2008;176:227-231. 469. Carlson G.P., Rumbaugh G.E., Harrold D. Physiologic alterations in the horse produced by food and water deprivation during periods of high environmental temperatures. Am J Vet Res. 1979;40:982. 470. Spensley M.S., Carlson G.P., Harrold D. Plasma, red blood cell, total blood, and extracellular fluid volumes in healthy horse foals during growth. Am J Vet Res. 1987;48:1703. 471. Kallfelz F.A., Whitlock R.H., Schultz R.D. Survival of 59Fe-labeled erythrocytes in crosstransfused equine blood. Am J Vet Res. 1978;39:617. 472. Hurcombe S.D., Mudge M.C., Hinchcli K.W. Clinical and clinicopathologic variables in adult horses receiving blood transfusions: 31 cases (1999-2005). J Am Vet Med Assoc. 2007;231:267-274. 473. Aleman M., Nieto J.E., Carr E.A., et al. Serum hepatitis associated with commercial plasma transfusion in horses. J Vet Intern Med. 2005;19:120-122. 474. Feige K., Ehrat F.B., Kästner S.B., et al. The e ects of automated plasmapheresis on clinical, haematological, biochemical and coagulation variables in horses,. Vet J. 2005;169:102-107. 475. Feige K., Ehrat F.B., Kästner S.B., et al. Automated plasmapheresis compared with other plasma collection methods in the horse. J Vet Med A Physiol Pathol Clin Med. 2003;50:185-189. 476. Gordon B.J., Latimer K.S., Murray C.M., et al. Evaluation of leukapheresis and thrombocytapheresis in the horse. Am J Vet Res. 1986;47:997. 477. Gordon B.J., Latimer K.S., Murray C.M., et al. Continuous-flow centrifugation hemapheresis in the horse. Am J Vet Res. 1986;47:342. 478. Morris D.D., Bruce J., Gaulin G., et al. Evaluation of granulocyte transfusion in healthy neonatal pony foals. Am J Vet Res. 1987;48:1187. 479. Morris D.D. Blood products in large animal medicine: a comparative account of current and future technology. Equine Vet J. 1987;19:272. ∗

References: 14, 207, 214, 238, 254, and 255.

CHAPTER 15 Disorders of the Gastrointestinal System∗ EXAMINATION FOR DISORDERS OF THE GASTROINTESTINAL TRACT L. Chris Sanchez

PHYSICAL EXAMINATION Examination of patients with disease of the gastrointestinal tract must include evaluation of the metabolic and cardiovascular status of the patient, because acute conditions of the proximal or distal intestinal tract can lead to endotoxemia and sepsis. Examination of the cardiovascular system (heart, peripheral pulse, and mucous membranes), lungs, and abdomen is essential to detect clinical signs of systemic in ammation from endotoxemia, coagulation disorders, dehydration, ileus, shock, and other abnormalities resulting from injury to the small or large intestine. Clinical signs of systemic inflammation from endotoxemia and sepsis are described elsewhere in this chapter. One performs the physical examination of the abdomen primarily by auscultation, transabdominal ballottement, and transrectal palpation. Abdominal distention often indicates distention of the large intestine; however, small intestinal distention also can cause visible abdominal distention if a large proportion of the small intestine is involved. Abdominal palpation can be performed in neonatal foals; after several weeks of age, however, the abdominal wall is too rigid to allow effective palpation of intra-abdominal structures. Abdominal auscultation is particularly useful for assessing the motility of the large intestine. Progressive motility of the small intestine, conversely, is di cult to distinguish by auscultation from nonprogressive motility. The distinct character of the borborygmi produced during propulsive contractions of the cecum and ascending colon allows evaluation of the frequency and strength of retropulsion and propulsion. Propulsive contractions of the cecum and ventral colon occur every 3 to 4 minutes and give rise to prolonged rushing sounds heard over long segments of intestine. Retropulsive sounds presumably are similar to propulsive sounds, but they occur less frequently. Distinguishing between propulsion and retropulsion is not important clinically because both types of contractions signify normal motility. Interhaustral and intrahaustral mixing contractions produce nonspeci c sounds of uid and ingesta movement that are di cult to distinguish from other borborygmi, such as small intestinal contractions or spasmodic contractions.1 Auscultation over the right ank and proceeding along the caudal edge of the costal margin toward the xiphoid allows evaluation of the cecal borborygmi. Auscultation over a similar area on the left side allows evaluation of the pelvic exure and ascending colon. Typical progressive borborygmi heard every 3 to 4 minutes on both sides of the abdomen indicate normal motility of the cecum and ascending colon. Less frequent progressive sounds may indicate a pathologic condition of the large intestine or may result from anorexia, nervousness (sympathetic tone), or pharmacologic inhibition of motility (i.e., α2-adrenergic agonists such as xylazine).2-5 Absolute absence of any auscultable borborygmi suggests abnormal motility and indicates ileus resulting from a serious pathologic condition but is not speci c to any segment of the intestine.3,6 If borborygmi are audible but progressive sounds are not detectable, determining whether a signi cant abnormality exists is di cult.6 Borborygmi heard more frequently than normal may result from increased motility following feeding; from excessive stimulation from irritation, distention, or in ammation; or after administration of parasympathomimetic drugs such as neostigmine. Large intestinal motility increases in the early stages of intestinal distention regardless of the site.7 Mild inflammation or irritation of the large intestinal mucosa also can stimulate motility.3 Parasympathomimetic drugs stimulate contractions and auscultable borborygmi in the large intestine; however, an increase in parasympathetic tone may result in segmental contractions, which actually inhibit progressive motility.2 One can detect sand or gravel in the large intestinal ingesta by auscultation behind the xiphoid process. It is possible to hear sand or gravel particles grinding together during progressive contractions of the ascending colon. In one study characteristic sounds were heard on auscultation (5 minutes’ duration) after daily administration of sand to healthy horses, none of which developed clinical signs of colic or diarrhea.8 If the frequency of progressive contractions is low or absent, detecting sand by auscultation is difficult. Percussion of the abdomen during auscultation can reveal gas in the large intestine. The characteristic ping produced by simultaneous digital percussion and auscultation over a gas- lled viscus often is associated with abnormal accumulation of gas under pressure. This technique is particularly useful in foals, ponies, and miniature horses because of the limitations of rectal palpation. Transabdominal ballottement can be used to detect large, rm masses or an abnormal volume of peritoneal uid (PF). The usefulness of this technique is usually limited to animals too small to palpate rectally. One can detect soft tissue masses or fetuses by bumping the structures with a hand or st. If excessive PF is present, a uid wave can be generated by ballottement; however, this technique is not as useful in horses older than 4 weeks because the abdominal wall is rigid. Transrectal palpation is the most speci c physical examination technique for investigation of intestinal disease and is particularly valuable

when evaluating obstructive diseases.9,10 The primary objectives of transrectal palpation are to assess the size, consistency, and position of the segments of the large intestine; to determine the presence of any distention of the small intestine; and to detect intra-abdominal masses. Evaluation of the wall thickness and texture and the mesenteric structures (blood and lymphatic vessels and lymph nodes) also may aid in diagnosis of large intestinal disease. The interpretation of transrectal palpation ndings in light of clinical signs and laboratory results is an important diagnostic aid for developing appropriate treatment strategies for intestinal diseases manifested by abdominal pain. Enlargement of one or more segments of large intestine detected by transrectal palpation provides evidence of obstruction at or distal to the enlarged segment. By systematically evaluating each segment, one can determine the site of obstruction. Obstruction of the pelvic exure, for instance, results in enlargement of the pelvic exure and ventral colon, but the dorsal and descending colons are of normal size. Enlargement of a segment of the large intestine usually is accompanied by abnormal consistency of the contents. It is possible to distinguish among gas, uid, and ingesta and to detect foreign bodies in palpable segments. Accumulation of gas and uid suggests complete and acute obstruction, whereas accumulation of ingesta suggests chronic and incomplete obstruction. Accumulation of uid usually indicates ileus. The practitioner must evaluate the consistency of the contents in light of the size of the segment; ingesta in the ventral colon of a dehydrated patient may be rm, but the size of the ventral colon will be normal. Conversely, if the ingesta is rm because of a distal obstruction, the ventral colon will be enlarged. Displacement of a segment of the large intestine may create an obstruction detectable by enlargement of the segment and accumulation of gas and uid, even if the site of obstruction is not palpable. Torsion of the ascending colon at the sternal and diaphragmatic exures results in acute accumulation of gas and uid proximal to the torsion, causing distention of the left dorsal and ventral colons. Depending on the degree of torsion, the position of the ventral and dorsal colons may not be signi cantly abnormal. Displacement of a segment of large intestine often results in incomplete obstruction, and the diagnosis relies solely on detection of the displaced segment in an abnormal position. The position of the displaced segment may not be palpable, and the diagnosis then relies on the inability to nd the segment in a normal position. One must take care to ensure that the segment that appears to be displaced is not in a normal position but has become too small to palpate because of a decrease in the volume of ingesta. The cecum, right dorsal and ventral colons, pelvic exure, and descending colon are palpable in most horses. One should palpate the nephrosplenic space to detect the presence of intestine, usually pelvic exure, entrapped within the ligament. Small intestine is not normally palpable in the horse. Distention indicates ileus with gas or fluid retention, usually following a strangulating or nonstrangulating obstruction. Strangulating obstructions result from conditions such as volvulus or torsion, lipoma, and entrapments. Such conditions often are accompanied by severe pain, dehydration, PF changes, and a varying degree of gastric uid accumulation. The small intestine in these cases is turgid and rm on palpation. One should assess the mesentery and wall thickness in the same manner as for large intestinal disorders. Careful palpation of the inguinal rings in stallions with small intestinal distention is crucial for determining inguinal herniation. Evaluation of the wall thickness and mesenteric vessels can reveal venous congestion (mural edema and enlarged blood and lymphatic vessels) or in ammation (mural edema with normal vessels). Disruption of arterial blood ow does not cause venous congestion, but the arterial pulse is not detectable. Mesenteric tears may not be palpable, but the entrapped ischemic intestinal segment may be thickened with edema. It is possible to detect acute or chronic in ammation with cellular in ltration of the intestinal wall as thickening of the wall without edema; enlargement of mesenteric lymph nodes also may be noted. One should interpret abnormalities in the wall or vessels in light of the size, consistency, and position of the segment of intestine and the clinical signs. Several conditions involving small intestinal strangulating lesions do not necessarily cause abnormal rectal examination ndings until the disease has been present for an extended time. These conditions include diaphragmatic herniae and epiploic foramen entrapments. PF analysis can be normal in these cases as well because the fluid is trapped in the thorax or cranial abdomen. Nonstrangulating causes of small intestinal distention can be divided further into intraluminal and extraluminal obstructions. Ileal impactions are probably the most common cause of intraluminal obstruction, and on rare occasions the impaction can be palpated in the upper right quadrant, near the ileocecal opening. Intraluminal masses caused by lymphoma, eosinophilic enteritis, foreign bodies, or ascarid impactions often lead to small intestinal distention and are usually indistinguishable from one another on the basis of palpation alone. Small intestine in these cases can be moderately to severely distended, depending on the degree of obstruction. Extraluminal obstructions include abdominal masses, abscesses or tumors, and large colon displacement. One should always palpate the rest of the abdomen carefully to help rule out these causes. Some cases of small intestinal distention result from a physiologic rather than a mechanical obstruction. Ileus may result postoperatively or after in ammatory diseases of the bowel (proximal enteritis) or peritoneal cavity (peritonitis). The bowel is usually mildly to moderately distended and almost always is accompanied by significant amounts of accumulated gastric fluid. The small colon is easily distinguishable by the presence of normal fecal balls and an antimesenteric band. In cases of impaction of the small colon, a long, hard, tubelike structure is present in the caudal abdomen, and the band is palpable along the length. Fluid stool is often present in the rectum in these cases, as is tenesmus, and the rectal mucosa is often edematous and occasionally roughened. One can detect and carefully evaluate rectal tears by palpation. Also detectable are mural masses in palpable segments of intestine or mesentery; however, if a mass causes obstruction, it is possible to detect the result of the obstruction in proximal segments of intestine even if the mass is

unreachable. Palpation of the mesenteric vessels may reveal thickening and thrombosis, which can lead to ischemia or infarction. One can perform visual inspection of the mucosa of the rectum and descending colon with a speculum or exible endoscope and also can evaluate rectal tears or perforations, mural masses, strictures, or mucosal inflammation. One can also perform guided biopsy of the mucosa or masses. The obvious limitations are the amount of fecal material, which can interfere with the examination, and the distance of the lesion of interest from the anus. These techniques offer little advantage over palpation in many cases, unless the patient is too small to palpate. Examination of the oral cavity in cases of dysphagia or weight loss is a necessary part of the physical examination. One should adequately sedate the horse and use a full-mouth speculum to allow palpation and visualization of all parts of the oral cavity. One should examine the area for abnormal dentition, foreign bodies, fractures, abscesses, and ulceration. The presence of uid accumulation in the stomach indicates a decrease or absence in propulsive motions of the small intestine or obstruction of gastric out ow. Decreased small intestinal motility may result from a functional or mechanical blockage. Masses, feed impactions, or strictures in the pylorus or in the proximal duodenum may obstruct gastric out ow. One routinely assesses uid accumulation in the stomach by siphoning o the gastric contents with a nasogastric tube and examining the uid for amount, color, and any particular odor. Normal uid is green and may contain foamy saliva. The volume obtained by gastric lavage is usually less than 4 L. Large volumes of uid (≥8 to 10 L) accumulate in the stomach of horses with proximal enteritis, and the uid is foul smelling and often has an orange to yellow discoloration. If proximal enteritis is suspected, the uid can be submitted for culture and Gram staining. Salmonella sp. and Clostridium sp. have been cultured in some cases. Patients with postoperative ileus also frequently accumulate large amounts of gastric uid. Horses with strangulating obstructions or luminal obstructions often accumulate moderate amounts of gastric uid, but the amount is generally less than in horses with proximal enteritis or postoperative ileus. Hemorrhage in the gastric uid usually indicates devitalized small intestine, stomach wall, or severe gastric ulceration. Fluid with large amounts of food material can indicate a gastric impaction, and one should lavage the stomach until no more ingesta can be obtained. Horses and foals with chronic gastric ulceration in the glandular mucosa of the stomach or in the duodenum may develop strictures and have accumulated uid in the stomach. Endoscopy or contrast radiography aids in diagnosing gastric outflow obstruction.

DIAGNOSTIC EVALUATION Clinical Pathology Evaluation of the hemogram is essential when assessing conditions of the gastrointestinal tract. However, hematologic alterations associated with diseases of the gastrointestinal tract are often nonspeci c, re ecting systemic response to in ammation, endotoxemia, or sepsis. Neutrophilic leukocytosis and normochromic, normocytic anemia with or without hyper brinogenemia commonly are associated with chronic inflammatory conditions of the intestine. Anemia from chronic blood loss occurs infrequently in adult horses because of the large iron stores and high concentrations of iron in their diet; anemia usually follows chronic in ammation, as do alterations in the leukon and plasma brinogen concentrations. Plasma protein concentrations vary depending on gastrointestinal losses of albumin and globulin and elevation of globulin concentration from antigenic stimulation. Protein-losing enteropathies may manifest predominantly as a hypoalbuminemia or as a panhypoproteinemia. Immunoglobulin quanti cation can be useful in selected cases; immunosuppression with low immunoglobulin M concentration has been shown to occur in some cases of lymphosarcoma.11 Parasitic infections, especially strongylosis, may be characterized by elevated serum immunoglobulin G(T) concentration.12 Signi cant alterations of the hemogram do not accompany acute disease of the intestine unless severe in ammation, dehydration, endotoxemia, or sepsis is present. During the early stages of endotoxemia, elevations in circulating concentrations of in ammatory mediators, epinephrine, and cortisol produce characteristic changes in the hemogram. Leukopenia, with neutropenia and a left shift, toxic changes in the neutrophil cytoplasm, and lymphopenia commonly occur.13 Hemoconcentration and hyper brinogenemia are also common. Thrombocytopenia and other coagulopathies are also features of endotoxemia. Indeed, thrombocytopenia may be the earliest indicator of sepsis.14 Endotoxemia and circulating mediators of in ammation activate the coagulation cascade, causing a hypercoagulable state that can lead to consumption of coagulation factors and coagulation defects manifested as elevated prothrombin time, partial thromboplastin time, brin degradation products, and bleeding time and reduced activity of antithrombin III.15-17 Neutrophilic leukocytosis occurs during the later stages of endotoxemia.15 The most common serum biochemical abnormalities with diseases of the large or small intestine are electrolyte imbalances. Serum calcium concentrations are often low with strangulating obstructions and acute in ammatory diseases.18 In ammation of the mucosa can severely disrupt electrolyte uxes. Diarrhea or gastric re ux greatly exacerbates the loss of sodium, potassium, calcium, magnesium, and bicarbonate. Hypoxia and cellular damage caused by ischemia of the intestine may be re ected by an elevated serum phosphate concentration resulting from phosphate leakage from damaged cells.19 Ischemia and cellular hypoxia in any segment of the intestine also causes a shift in energy metabolism to anaerobic glycolysis, resulting in increased production of lactate and elevated serum lactate concentration. Reduced perfusion of peripheral tissues from hypotensive shock and intestinal ischemia can cause elevations in serum lactate. However, obstruction of the intestine during ischemia may result in absorption of lactate from the lumen.20,21 Anion gap is an indirect measurement of organic acid

production during states of tissue hypoxia and is a reasonable estimate of serum lactate concentration.21 Metabolic acidosis may accompany lactic acidemia, but an inconsistent association exists between the two, especially when mixed acid-base imbalances are present.21,22 Elevations of hepatic enzymes, speci cally γ-glutamyltransferase, may occur with large colon displacements, duodenal strictures, or anterior enteritis. An elevated γ-glutamyltransferase is more suggestive of a right dorsal displacement than a left dorsal displacement.23 Relative polycythemia from hemoconcentration or splenic contraction and changes in red blood cell deformability from hypoxia or hypocalcemia may increase blood viscosity. Blood viscosity increases in patients with acute obstructive disease. Hyperviscosity reduces perfusion of capillary beds, thereby exacerbating ischemia and tissue hypoxia.24 Hyperviscosity is one manifestation (along with lactic acidemia, coagulopathies, and clinical signs of shock) of the pathophysiologic events that take place during acute in ammatory or vascular injury to the large intestine. Laboratory tests designed to re ect the systemic e ects of endotoxemia, ischemia, sepsis, and shock are important to design therapeutic strategies and monitor response to therapy.

PERITONEAL FLUID Abdominocentesis and analysis of PF are diagnostic techniques performed on many patients with disease of the gastrointestinal tract. The veterinarian can quantitate cytologic examination of PF; white blood cell and red blood cell counts; protein, brinogen, lactate, phosphate, and glucose concentrations; lactate dehydrogenase, creatine kinase, and alkaline phosphatase activity; and pH. The results of PF analysis may help establish a speci c diagnosis and, more important, may re ect in ammatory, vascular, or ischemic injury to the intestine requiring surgical intervention. PF re ects a sequence of events that takes place during acute vascular injury to the intestine. The PF protein concentration rst increases, followed by an increase in the red blood cell count and brinogen concentration. A transudative process resulting from vascular congestion and increased endothelial permeability allows small macromolecules (albumin) to escape into the PF, followed by larger macromolecules (globulin and brinogen), and nally diapedesis of cells (red blood cells, and then white blood cells).25,26 If ischemic in ammation of the intestine and visceral peritonitis occurs, an exudative process ensues. Severe in ammation of the intestine and visceral peritoneum causes large quantities of protein and white blood cells, primarily neutrophils, to escape into the PF.24,26 As damage to the bowel progresses, the protein concentration and red blood cell and white blood cell counts continue to rise. As the degree of irreversible damage to the intestine increases, the PF characteristics become more exudative.25,26 Eventually, bacteria begin to translocate across the intestinal wall and appear in the PF as the mucosal barrier breaks down. Neutrophils predominate, their cytoplasm becomes granulated, and Döhle’s bodies often are visible. If perforation occurs, bacteria and particles of ingesta appear in the PF, and the neutrophils become degenerate (i.e., pyknotic), with karyorrhexis, karyolysis, and smudge cells. Elevated PF protein concentration is a sensitive indicator of early in ammation, whereas elevated red blood cell counts in the presence of normal white blood cell counts suggest vascular damage without signi cant tissue ischemia.26 Of note, the anticoagulant potassium EDTA, but not lithium heparin, can cause an increase in total protein as measured by refractometer, relative to the value obtained from the same sample without anticoagulant.27 Elevation of the white blood cell count usually indicates severe tissue in ammation or intestinal injury.20 The gross color of the PF can be helpful in detecting injury and necrosis of the intestine. A serosanguinous appearance indicates vascular injury, whereas orange or brown-red indicates necrosis with the release of pigments such as hemosiderin. Serial samples of PF are most useful in determining the nature and extent of damage to the intestine, but in many cases of ischemia, irreversible tissue damage has occurred by the time PF abnormalities appear. Tissue hypoxia and ischemia cause a rapid elevation of PF lactate dehydrogenase, creatine kinase, and alkaline phosphatase activity and lactate concentration.20,21,28,29 Phosphate concentration increases when cellular disruption occurs.19 PF enzyme activities, phosphate, and lactate concentration increase faster and higher than serum activities.19-21,28,29 PF pH and glucose concentration tend to decrease during intestinal ischemia but are not as low as in septic peritonitis.30 Although biochemical alterations may be early indicators of intestinal ischemia and necrosis, they are nonspeci c and o er no advantage over conventional methods of PF analysis in many cases. PF alkaline phosphatase has been shown to arise predominantly from degenerating white blood cells, and elevations of other enzyme activities may occur with many in ammatory diseases.28 Thus the speci city of many tests performed on PF is questionable. However, in selected cases in which conventional PF analysis and physical examination do not provide su cient information to develop a treatment plan, biochemical analysis of the PF may be useful. Cytologically examined cells of the PF may re ect chronic in ammatory conditions of the large intestine, especially eosinophilic or lymphocytic processes.31 Infectious and in ammatory conditions often cause increases in the neutrophil count and may be indistinguishable unless bacteria are visible. One also may detect neoplastic diseases by PF examination. Chronic infection and in ammation may be associated with elevated PF protein and brinogen concentrations. Culture of PF usually is required to distinguish bacterial infections from noninfectious in ammation unless bacteria are visible on cytologic examination. However, culture of PF is often unrewarding because factors that are found in in ammatory PF inhibit bacterial growth, and leukocytes phagocytose many bacteria in the PF.32 Decreases in PF glucose concentrations (5 mm) in cases of right dorsal colitis. Ultrasound also can be a useful tool for evaluating changes in motility over time in a given location.60 Evaluation of the abdomen always should include assessment of the peritoneal space for any evidence of an increased amount of PF or increased cellularity of the uid, as indicated by an increase in echogenicity. Ultrasonography also can be useful in determining the ideal location for abdominocentesis. One also should evaluate the liver, kidneys, and spleen to detect any choleliths, nephroliths, masses, abscesses, or enlargement. Identifying abscesses or tumors not associated with visceral organs may be di cult and depends on their location and surrounding structures.

Nuclear Scintigraphy Although more commonly used to diagnose lameness and musculoskeletal problems, nuclear scintigraphy has several uses in the evaluation of the gastrointestinal tract. Proper isolation protocols and waste disposal techniques must be strictly observed. The procedure requires special gamma cameras and the injection of radioactive materials into the bloodstream. One of two methods may be used: injection of technetium-99m methylene diphosphonate (99mTc-MDP) directly into the blood or injection of 99mTc-labeled leukocytes.61 The principle of nuclear scintigraphy then lies in increased uptake of the dye or the white blood cells into areas of in ammation. One of the most common uses of nuclear scintigraphy in evaluating the gastrointestinal tract is diagnosis of dental disease. Scintigraphy using 99mTc-MDP proved to be more sensitive in cases of dental disease than was radiography. Scintigraphy was slightly less speci c, however, and therefore should be used with radiography or computed tomography for ultimate accuracy.62 Scintigraphy using radiolabeled white blood cells can support a diagnosis of right dorsal colitis in the horse.63 Images taken of the abdomen 20 hours after injection showed an increased linear uptake of leukocytes in the region of the right dorsal colon in horses with right dorsal colitis compared with normal horses. Other uses of nuclear scintigraphy include evaluation of metastasis of abdominal tumors to bony areas, assessment of biliary kinetics, and determination of liquid- and solid-phase gastric emptying.62-64

Advanced Imaging Computed tomography (CT) is becoming increasingly available, and as such, various references are currently available describing the normal anatomy of the equine head as imaged with this modality.65,66 CT is extremely useful to evaluate dental disease as well as tumors and masses of the head, larynx, pharynx, and proximal esophagus.67,68 CT also has promise for evaluating abdominal disorders in foals. Most equipment can accommodate animals up to 400 lb. Restrictions of CT as a diagnostic aid include expense, availability, and weight and size limitation.

Endoscopy Endoscopic examination of the gastrointestinal tract begins with evaluation of the pharyngeal area for any signs of collapse or dysfunction. One should evaluate the ability of the horse to swallow. The oor of the pharynx should be clean and free of feed material and foreign bodies. The oral cavity should be examined with the horse under heavy sedation or anesthesia and with the help of a full-mouth speculum. One can examine the teeth for any irregularities, obvious cavities, sharp points, or hooks and the hard and soft palates for completeness and any evidence of ulceration, masses, or foreign bodies.

A 3-m exible endoscope should be used to examine the esophagus, which is accomplished best by passing the endoscope into the stomach and viewing the esophagus as one withdraws the endoscope while dilating the lumen with air. The esophageal mucosa normally should be a glistening, light pink color. Ulceration can occur with cases of choke or re ux esophagitis or in horses that have had an indwelling nasogastric tube. Erosions may be punctate, linear, or circumferential. One should evaluate carefully for any ulcers to ensure that no areas of perforation through the entire thickness of the esophageal wall exist. Distinguishing normal peristaltic contractions from areas of stricture requires observation of the area and its motility over time. Diverticula also may be noted as outpouchings of the mucosa, sometimes associated with a stricture distally. Megaesophagus, although rare, appears as a generalized dilation of the esophagus. Endoscopy may reveal food or foreign body impactions. The veterinarian always should re-evaluate the esophagus after removing any obstruction to detect the presence of complications (ulceration, rupture) or initiating causes (strictures, diverticula, and masses). A 3-m exible endoscope also allows examination of the stomach. The horse should be fasted for at least 12 hours before endoscopy. One can examine the cardia and fundus easily, as well as the margo plicatus. With complete emptying of the stomach (aided with gastric lavage, if necessary), one can also can evaluate the pylorus and proximal duodenum. The squamous mucosa should resemble the esophageal mucosa. The glandular mucosa should be glistening red and may have a reticulated pattern. The veterinarian should carefully examine it for evidence of ulceration or masses. Transendoscopic biopsy material can be easily obtained from esophageal, pharyngeal, or gastric masses, and because the biopsy size will be small, several samples should be taken for histopathologic examination. A complete description of gastroscopy and evaluation of gastric and gastroduodenal ulceration is available later in this chapter.

Tests of Absorption and Digestion d-Glucose

or d-xylose absorption tests are useful in determining malabsorption of carbohydrates from the small intestine in horses. The protocol for absorption tests using either carbohydrate is similar. The horse should be fasted for 18 to 24 hours before testing. Increased periods of fasting actually have been shown to decrease absorption of d-xylose and interfere with results.69 A dosage of 0.5 to 1 g/kg of dglucose or d-xylose is administered through a nasogastric tube. Administration of sedatives may falsely increase the blood glucose levels and interfere with gastrointestinal transit times. Blood samples are collected to measure glucose or xylose concentrations at 0, 30, 60, 90, 120, 150, 180, 210, and 240 minutes after administration. Additional samples can be taken up to 6 hours after dosing if the results are questionable. One should measure glucose in blood samples collected with sodium uoride as an anticoagulant and measure xylose in samples collected in heparinized plasma. A normal d-glucose absorption test, also known as an oral glucose tolerance test, should have a peak between 90 and 120 minutes, and this peak should be greater than 85% above the resting glucose value.70 Complete malabsorption is de ned as a peak less than 15% above the resting levels, and partial malabsorption is de ned as a peak between 15% and 85% above the resting level. It is important to remember that gastric emptying, gastrointestinal transit time, length of fasting, cellular uptake and metabolism, age, diet, and endocrine function in uence glucose absorption curves.70,71 Malabsorption demonstrated by the oral glucose tolerance test is sensitive but not speci c. Diseases that may cause a lowered or delayed peak include in ltrative lymphosarcoma, in ammatory bowel disease (lymphocytic-plasmacytic or eosinophilic), cyathostomiasis, chronic colitis (Salmonella sp.), multisystemic eosinophilic epitheliotropic disease, food allergies, and small intestinal bacterial overgrowth.72 d-Xylose absorption tests have some advantages over the oral glucose tolerance test because xylose is not metabolized in the small intestinal mucosa and insulin does not in uence its absorption. Gastric and intestinal motility, intraluminal bacterial overgrowth, and renal function still in uence xylose absorption because the kidneys clear xylose.72 The other main drawback to d-xylose is that it is generally available only in research settings. However, xylose measurements are available at many major universities. A normal dxylose absorption curve should peak between 20 and 25 mg/dl at 60 to 120 minutes after dosing.73 Decreased xylose absorption can occur in horses with inflammatory bowel disease, lymphosarcoma, multisystemic eosinophilic epitheliotropic disease, cyathostomiasis, extensive small intestinal resections, and any cause of villous atrophy.72 Maldigestion is a common occurrence in foals with diarrhea. Bacteria (especially Clostridium sp.) and viruses (especially rotavirus or coronavirus) may invade and destroy the villous epithelial cells that manufacture lactase and other disaccharidases, resulting in an inability to digest lactose. In this case continued ingestion of the mare’s milk may cause an osmotic diarrhea, which may exacerbate the underlying enterocolitis. The veterinarian can perform lactose tolerance testing to assess the degree of maldigestion by administering d-lactose at 1 g/kg as a 20% solution via nasogastric tube and measure glucose concentrations in the blood at 0, 30, 60, 90, 120, 150, 180, 210, and 240 minutes. A normal curve shows doubling of glucose levels compared with baseline by 60 minutes after administration.74

Evaluation of Gastric Emptying Assessment of gastric emptying may be useful in cases of gastric and esophageal ulceration, pyloric stenosis, proximal enteritis, and potentially postoperative ileus. However, accurate measurement of gastric emptying can be difficult to assess. Multiple diagnostic imaging techniques have been used to study gastric emptying times. Contrast radiography can be used to assess gastric emptying in foals. In the normal foal barium remains in the stomach for varying amounts of time, but a signi cant amount should be gone

within 2 hours.48 Gastric emptying of solid, nondigestible, radiopaque markers also has been used in adult horses and ponies, but the results were variable and unpredictable even in the normal horse.75 Nuclear scintigraphy is used commonly in human beings to measure gastric emptying and can be used in horses when available. The technique for measurement of liquid gastric emptying requires oral administration of 99mTc pentenate (10 mCi) and serial images taken of the cranial abdomen. The tracer is usually not visible 1 hour after administration in normal horses.64 An adaptation of this methodology can be used for measurement of solid phase emptying of a 99m Tc–labeled pelleted ration.76 Alternatively, if nuclear scintigraphy is not available, acetaminophen absorption testing can be used as an indirect determination of liquid gastric emptying.77,78 This test is performed by administering 20 mg/kg of acetaminophen orally and measuring subsequent blood values, then calculating the time to reach maximum serum concentrations and the absorption constant. In human beings the proximal small intestine absorbs almost all of the acetaminophen.79 The median time to reach peak plasma levels using acetaminophen absorption in horses was 47.7 minutes.77 The 13C-octane acid breath test o ers an easy, noninvasive method of determining gastric emptying of solids.76,80 This test is performed by feeding a standard 13C-labeled test meal, and then collecting breath samples using a modi ed mask. The breath is then analyzed for the ratio of the novel isotope, 13:CO2, to the normally produced 12:CO2.

HISTOPATHOLOGIC EXAMINATION It is often necessary to perform a histopathologic examination of tissues from the intestine to diagnose chronic in ammatory, in ltrative, or neoplastic conditions, and such examination can be useful in evaluating the extent of injury after obstruction or ischemia. Rectal mucosal biopsies are easy to collect, with few complications. However, to collect a full-thickness biopsy of the intestine requires a surgical approach (flank or ventral midline approach). Laparoscopy offers a safer technique to observe the large intestine and other abdominal structures.81 The practitioner can obtain biopsies of masses, lymph nodes, and mesentery or intestinal serosa using laparoscopy and mucosal biopsies of the upper gastrointestinal tract using endoscopy.

LAPAROSCOPY Other diagnostic tools, speci cally laparoscopy and CT, are available but require specialized equipment and personnel with speci c training. Flexible or rigid endoscopes used for laparoscopic evaluation of the abdomen allow for visualization of visceral organs and potentially for collection of biopsy material from masses or organs. Full-thickness biopsies of the intestines are not routinely possible through the laparoscope and usually require ank or ventral midline laparotomy. The laparoscopic procedure can be done with the horse standing or recumbent. Advantages of this technique over a ank or ventral midline celiotomy include smaller incisions, shorter healing time, and shorter procedure time. Disadvantages include the large amount of equipment needed, the high degree of skill involved, and the limitation as a diagnostic modality, rather than a treatment.82 Clinical applications of diagnostic laparoscopy include the correction of rectal tears; percutaneous abscess drainage, assessment of adhesions, displacements, and integrity of the serosa of various bowel segments; and biopsy of abdominal masses.81

PATHOPHYSIOLOGY OF GASTROINTESTINAL INFLAMMATION Samuel L. Jones The in ammatory response of the gastrointestinal tract is a mechanism ultimately aimed at eliminating pathogens, initiating tissue repair, and restoring the gastrointestinal barrier. Blood ow is altered, endothelial permeability increases, cells are rapidly recruited into the tissue, plasma protein cascades are activated, and myriad soluble products are released that coordinate the response, trigger innate and adaptive immunity, and mobilize reparative elements. Although the cellular and vascular response and the secreted mediators of in ammation are important for killing pathogens and limiting invasion of injured tissues by commensal organisms, they can be quite damaging to host cells and proteins if not tightly regulated. Thus if the inciting stimulus is not eliminated quickly, then the in ammatory response itself will cause signi cant tissue injury. The mechanism regulating in ammation has been the focus of much research to identify therapeutic targets to modulate the damage to host tissues that occurs in many gastrointestinal diseases. Recent work has provided some of the molecular and cellular details of this complex physiology and has led to novel therapeutic strategies for treating inflammation.

INITIATION OF THE INFLAMMATORY RESPONSE Epithelium The gastrointestinal epithelium interfaces with a luminal environment that is inhabited by potentially hostile microbial organisms. The epithelium presents a physical barrier to invasion by the ora of the gastrointestinal tract, consisting of the apical cellular membrane, intercellular tight junctions (the permeability of which is highly regulated), and a secreted layer of mucus. When invading pathogens breach

the mucosal barrier, potent soluble and neural signals are generated that initiate an in ammatory response.1 The epithelium can be conceptualized as a sensory organ that detects pathogen invasion to trigger an appropriate host defense and reparative response. Noninfectious mucosal injury or invasion of epithelial cells by pathogenic organisms such as Salmonella activates the synthesis of proin ammatory chemokines (chemoattractants) by epithelial cells that triggers a robust in ux of neutrophils into the tissue within hours of the damage. 1 Of the chemoattractants produced by epithelium, interleukin-8 (IL-8) has a particularly important role in initiating in ammation by recruiting neutrophils from blood2-4 and regulating neutrophil migration through tissue matrix adjacent to epithelium.5,6 Complement fragments such as C5a and bacteria-derived formylated chemotactic peptides also act as potent “end target” chemoattractants that are fully capable of stimulating a robust in ammatory response in the intestine if the epithelial barrier permits invasion of bacteria or the di usion of bacterial peptides across the mucosa. Epithelial cells activated during infection produce cytokines such as tumor necrosis factor α (TNF-α), arachidonic acid metabolites, and other pro-in ammatory mediators that activate recruited leukocytes.7 Microbial products, particularly lipopolysaccharide and other bacterial cell wall components and microbial nucleic acids, are potent activators of leukocytes recruited into the tissue.8 Mast cells are key sentinel leukocytes that sense microbial invasion, releasing TNF-α that appears to be a critical initiator and regulator of the cellular phase of inflammation.9 Once the in ammatory response has been initiated, TNF-α; IL-1β; and other pro-in ammatory products of neutrophils, monocytes, mast cells, and epithelial cells amplify the inflammatory response. The enteric nervous system has a key role in sensing and regulating in ammatory responses in the intestine. For example, Clostridium difficile toxin A activates a neural pathway that triggers mast cell degranulation and neutrophil in ux into the tissue.10,11 Blockade of this neural pathway is su cient to abolish the profound in ammatory response induced by toxin A as well as many of the e ects of toxin A on enterocyte secretion. Other pathogens and immune-mediated hypersensitivity reactions similarly stimulate in ammation by mechanisms that involve the enteric nervous system. Thus the epithelium interacts in a highly complex manner with the intestinal milieu, the enteric nervous system, and inflammatory cells to regulate the tissue response to injury and infection.

Macrophages Resident macrophages located in the lamina propria, submucosa, and intestinal lymphoid organs are among the rst cells beyond the epithelium to respond to infection or injury. Macrophages are activated by microbial products by way of pattern recognition receptors and begin to produce pro-in ammatory molecules important for recruiting and activating neutrophils and monocytes. Pattern recognition receptors recognize microbial molecules such as lipopolysaccharide, lipoproteins, agellin, peptidoglycan, and nucleic acids to signal the invasion by pathogens.8 Of the pattern recognition receptors, the lipopolysaccharide (LPS) receptor complex is perhaps the best de ned. LPS activates macrophages by way of the CD14-Toll-like receptor 4 complex to initiate transcription of the in ammatory cytokines TNF-α and IL1β, which synergize with LPS to amplify the macrophage response.8 LPS, particularly in concert with in ammatory cytokines, stimulates macrophages to produce copious amounts of nitric oxide, which is both microbicidal and vasoactive.12 Nitric oxide and other nitrogen radicals react with reactive oxygen intermediates (ROIs) generated by the activated oxidase complex to produce some of the most toxic molecules of the host defense system: the peroxynitrites.12 IL-8 is produced as well to recruit neutrophils. As the response progresses, other in ammatory mediators, particularly the arachidonic acid–derived lipids dependent on in ammation-induced cyclooxygenase-2 and 5lipoxygenase activity, are produced that have potent vasoactive and pro-in ammatory e ects through the activation of endothelial cells, neutrophils, and platelets13.

VASCULAR RESPONSE DURING INFLAMMATION Four important changes occur in the intestinal vasculature during in ammation: (1) alteration of blood ow; (2) increased vascular permeability; (3) increased adhesiveness of endothelial cells, leukocytes, and platelets; and (4) exposure of the basem*nt membrane and activation of the complement, contact, and coagulation cascades. A wide range of mediators alter blood ow during in ammation in the intestinal tract, ranging from gases such as nitric oxide (a major vasodilator of the intestinal vasculature) to lipids (prostaglandins, leukotrienes, thromboxanes, and platelet-activating factor [PAF]), cytokines, bradykinin, histamine, and others. The major sources for these mediators include activated leukocytes, endothelial cells, epithelial cells, and broblasts. The primary determinant of blood ow early in in ammation is vascular caliber, which initially decreases in arterioles but then quickly changes to vasodilation coincident with opening of new capillary beds, increasing net blood ow. The increase in blood ow is relatively short lived, as the viscosity of the blood increases because of uid loss and tissue edema resulting from leaky capillaries. Leukocyte margination, platelet adhesion to endothelial cells and exposed matrix, and areas of coagulation protein accumulation further decrease local circulation. Increased vascular permeability is initially caused by in ammatory mediator actions on the endothelial cells. Histamine, leukotrienes, PAF, prostaglandins, bradykinin, and other mediators stimulate endothelial cell contraction, and interendothelial gaps form.14,15 This stage of increased vascular permeability is readily reversible. Concurrently, mediators such as the cytokines TNF-α and IL-1β induce a structural

reorganization of the interendothelial junctions, resulting in frank discontinuities in the endothelial monolayer.16 Cytokines also stimulate endothelial cells to express adhesion molecules that support adhesion of leukocytes and platelets,17 leading to the next and perhaps most devastating event. Leukocytes (primarily neutrophils) and platelets adhere to exposed basem*nt membranes and activated endothelial cells. Adherent neutrophils and platelets are then exposed to the mediators of in ammation present in the surrounding milieu, which activates the cells to release oxidants and proteases (particularly elastase) that injure the endothelium and have the potential to cause irreparable harm to the microvasculature.18-20 Marginated neutrophils begin to transmigrate between endothelial cells (as described in later sections), which, if in sufficiently large numbers, disrupt the integrity of the interendothelial junctions, worsening the vascular leakage.19 These stages of enhanced vascular permeability can be conceptualized as a mechanism to allow plasma proteins to enter the tissues and to potentiate the critical influx of leukocytes into tissues. However, if they are not regulated precisely, alterations in both hydrostatic and oncotic forces and irreversible damage to the vascular bed may have devastating consequences. Moreover, inappropriate activation of plasma protein cascades and leukocytes by activated endothelium and exposed matrix proteins can contribute to systemic in ammation (systemic in ammatory response syndrome, or SIRS; see the section on gastrointestinal ileus for more information) characterized by hypotension, generalized vascular leak syndrome, and multiorgan dysfunction, which may be fatal. Phosphodiesterase inhibitors reduce endothelial permeability in ischemia-reperfusion injury and other models of in ammation-induced vascular leakage21,22 by increasing endothelial tight junction integrity and thus may be a viable therapeutic strategy to prevent or reduce the permeability alterations associated with inflammation.

CELLULAR EFFECTORS OF INFLAMMATION Endothelial Cells Endothelial cells respond to products of activated epithelial cells and macrophages in the intestinal tissue to recruit cells and humoral mediators of in ammation into the tissue. Activated endothelial cells display a range of molecules critical for neutrophil and platelet adhesion. The role of endothelial cells in mediating neutrophil recruitment is discussed in more detail later in this chapter. Intercellular permeability is increased to expose basem*nt membrane proteins that trigger humoral defense systems (complement, coagulation, and contact system cascades) and to provide access for these macromolecules to the tissue. Endothelial cells are an important source of inflammatory mediators that amplify the response and vasoactive substances (particularly nitric oxide) altering blood flow.

Neutrophils RECRUITMENT Infection or injury to the gastrointestinal mucosa causes an in ux of leukocytes from the blood that lay the foundation of the in ammatory response. Neutrophils, the rst to arrive during in ammation, have a dominant role in the acute response. Within minutes neutrophils are recruited into the tissue, where they are activated to release products that not only are lethal to pathogens and pro-in ammatory but also may damage host cells and tissues.23 Not surprisingly, much attention has been paid to the role of neutrophils in the pathophysiology of many in ammatory conditions.24 Neutrophil depletion is protective in many models of gastrointestinal in ammatory disease. Of interest to clinicians, blockade of neutrophil migration into in amed tissues prevents many of the pathophysiologic events associated with infectious enteritis, ischemia-reperfusion injury, and other gastrointestinal diseases.18,25-29 Neutrophil transendothelial migration is a multistep process that is temporally and spatially regulated and has a degree of cell type specificity (Figure 15-2). The predominant sites of neutrophil transendothelial migration are in the postcapillary venules and, in some tissues, capillaries. Endothelial cells in these vessels respond to cytokines and other soluble signals by expressing molecules that promote neutrophil adhesion and transmigration, including selectins and counter receptors for integrins. As neutrophils ow through these vessels, they are rst tethered to activated endothelium. Tethering is mediated by selectin molecules expressed on neutrophils (L-selectin) and on activated endothelial cells (P- and E-selectins), which bind to PSGL-1, ESL-1, and other mucin counter-receptors.30,31 The function of tethering is to increase the exposure of the neutrophil to activating chemokines presented on the surface of the endothelial cells.

FIGURE 15-2

Depiction of neutrophil responses during intestinal in ammation in response to salmonella infection. Salmonellae infect epithelial cells, stimulating the production of chemokines (interleukin-8 [IL-8]), cytokines (IL-1β and tumor necrosis factor-α [TNF-α]), and other proin ammatory mediators. Endothelial cells stimulated by in ammatory mediators produce chemoattactants (such as IL-8) and display adhesion molecules that promote neutophil emigration. The three steps of neutophil (polymorphonuclear [PMN]) emigration—capture/rolling (mediated by selectins), adhesion (mediated by β2 integrins), and transendothelial migration (mediated by integrins and platelet/endothelial cellular adhesion molecule [PECAM])—occur on activated endothelium. Chemoattractant molecules such as IL-8 trigger neutrophil emigration. In in amed tissues cytokines (IL-1β and TNF-α) and a variety of other pro-in ammatory mediators stimulate the neutrophil oxidase complex to produce reactive oxygen intermediates (ROIs; O 2- and H2O 2 and their derivatives). Activated neutrophils degranulate to release proteases and other hydrolases, cationic peptides (defensins), myeloperoxidase, and other products into the tissue. Activated neutophils synthesize a variety of in ammatory mediators, including prostaglandins (PGE 2) that modulate the in ammatory response. The products of activated neutrophils (ROIs, proteases, and mediators) stimulate epithelial secretion and alter tight junction permeability, promoting diarrhea. Neutrophils eventually migrate across the infected epithelium by a mechanism that involves integrins, disrupting tight junction integrity and increasing permeability to bacterial products, thus exacerbating the inflammatory response.

Stimulation of neutrophils by IL-8 and other chemokines activate the second step of transendothelial migration. Chemokine binding to their receptors on the neutrophil generates signals that activate the binding of integrin adhesion receptors to their ligands, called intracellular adhesion molecules (ICAMs) or vascular cell adhesion molecules (VCAMs), expressed on endothelial cells in in amed mucosa. Integrin ligation to ICAMs arrests the tethered neutrophils, resulting in rm adhesion to the endothelium. Of the integrins expressed on neutrophils, the β2 integrins have a particularly important role in transendothelial migration. Calves and people with the disorder leukocyte adhesion de ciency (LAD) illustrate the requirement for β 2 integrin-mediated adhesion in neutrophil function. LAD is a result of an autosomal recessive trait resulting in the lack of the β2 integrin expression. The neutrophils from a ected individuals cannot migrate into most tissues and do not function normally, resulting in poor tissue healing and profound susceptibility to infection, especially at epithelial barriers.32,33 Other integrins also have a role in transendothelial migration. β1 Integrins mediate transendothelial migration in some cells and seem to be particularly important for mediating emigration of monocytes into many tissues.34 Following this rm adhesion step, neutrophils migrate through the endothelium along a chemotactic gradient of IL-8 and other chemoattractants, such as C5a and LTB4.3,19,35 Neutrophils migrate across the endothelial monolayer at intercellular junctions by way of a mechanism involving a series of integrin-ligand interactions mediated by both β2 and β1 integrins and other adhesion molecules31 that is generally capable of maintaining the integrity of the endothelial barrier.36 However, massive ux of neutrophils through the endothelium alter endothelial tight junctions and injure the basem*nt membrane, resulting in increased endothelial permeability to molecules as large as plasma proteins and even endothelial cell detachment from the basem*nt membrane.19,20 Nonintegrin molecules such as platelet-endothelial cell adhesion molecules (PECAMa) also are involved in transendothelial migration of neutrophils.31 hom*otypic binding of PECAMs on adjacent endothelial cells form part of the intercellular junction. Neutrophils express an integrin of the β3 family that can bind PECAM, and through sequential binding of β3 integrins to PECAM, the neutrophil can “unzip” the intercellular junction and migrate through, closing it behind itself.

ACTIVATION A key feature of neutrophils and other leukocytes is the requirement for integrin-mediated adhesion to extracellular matrix (ECM) proteins or other cells to achieve an optimal e ector phenotype.37 Critical components of the ECM in in amed tissues include bronectin, brinogen, and vitronectin, deposited in tissues as a result of plasma leakage and by synthesis of new proteins by stromal cells and resident macrophages in response to in ammatory mediator activation. The changing composition of the matrix proteins deposited in tissues during in ammation

serves as a cue as to the nature of the tissue environment for recruited in ammatory cells as they become activated. Individual gene expression studies have demonstrated that adhesion to matrix proteins induces the expression of cytokines and chemokines and their receptors, arachidonic acid–derived lipid mediator synthases, metalloproteinases, growth factors, transcription factors, and other genes that in uence the di erentiation and activation of in ammatory cells.38 Reactive oxygen intermediate production, phagocytosis, degranulation, and other e ector functions stimulated by in ammatory mediators and bacterial products are optimal only when neutrophils are adherent to the ECM.37 Adhesion to distinct ECM proteins selectively activates signaling pathways and gene expression of neutrophils, monocytes, and other leukocytes with di ering abilities to promote certain functions such that the composition of ECM in many ways controls the development of the ultimate e ector phenotype. Thus integrin-mediated adhesion provides a mechanism whereby neutrophils and other leukocytes can sense the complex tissue environment and respond appropriately. Of the activators of neutrophils at sites of in ammation, complement (C3-opsonized particles), cytokines (TNF-α and IL-1β), PAF, immune complexes, and bacterial products are among the most potent stimuli. Other mediators produced during in ammation may modify neutrophil activity, particularly formylated bacterial peptides, chemokines, complement fragments (C5a), leukotriene B4 (LTB4), and prostaglandins. Activated neutrophils are highly phagocytic; produce large amounts of ROI; degranulate to release myeloperoxidase, cationic antimicrobial peptides (defensins), serine proteases (mainly elastase), and metalloproteinases; and secrete in ammatory mediators (TNF-α, IL-1β, prostaglandins, leukotrienes, and others) (see Figure 15-2).

Mast Cells Mast cells strategically reside in mucosal tissues, including the submucosa and lamina propria of the gastrointestinal tract, and constitute a crucial rst line of defense at epithelial barriers. However, they are also important e ector cells of the pathophysiology of in ammatory gastrointestinal diseases.39 Experimental depletion of mast cells, genetic de ciency in the development of mast cells, or pharmacologic stabilization of mast cells to prevent degranulation all have a protective e ect in a variety of models of gastrointestinal in ammatory disease, including dextran- or trinitrobenzenesulfonic acid–induced colitis40,41, ischemia-reperfusion injury,42,43 and immediate hypersensitivity responses.44 Mast cells are activated by a wide variety of microbial products and host-derived mediators.45 Among the activators of mast cells, the socalled anaphylatoxins (complement fragments C3a, C5a, and C4a) are extremely potent stimuli causing release of mediators of in ammation. In addition, mast cells are the primary e ector cells of IgE-mediated anaphylaxis (Type I hypersensitivity reactions) by virtue of their high a nity receptors for IgE. Cross-linking of receptor-bound IgE on mast cell surface by antigens (i.e., food antigens) causes rapid degranulation, resulting in the explosive release of granule contents.46 Neural pathways in the intestine also regulate mast cells. Mast cells respond to enteric pathogen invasion via neural reflexes that stimulate the release of inflammatory mediators. Activated mast cells release preformed histamine, 5-HT, proteases, heparin, and cytokines from granules. Activation also stimulates de novo synthesis of a range of in ammatory mediators, including prostaglandins, PAF, and leukotrienes. Transcription of a number of peptide mediators, such as the cytokines TNF-α and IL-1β among many others, also increases upon stimulation of mast cells. Mast cell products have profound e ects on the vasculature, increasing endothelial permeability and causing vasodilation.47 Moreover, mast cell–derived mediators markedly enhance epithelial secretion by a mechanism that involves the activation of neural pathways and direct stimulation of epithelial cells.46 In particular, the mast cell granule protease tryptase operating via the protease-activated receptor-2 is a key regulator of gastrointestinal physiologic responses during in ammation, including epithelial secretion and intercellular junction integrity, motility, and pain responses.48,49 Mast cell products signi cantly alter intestinal motility, generally increasing transit and expulsion of intestinal contents. Mast cell–derived leukotrienes and TNF-α also have a crucial role in host defense against bacterial pathogens, acting to recruit and activate neutrophils,50,51 and are crucial players in the mechanism regulating dendritic cell function and adaptive immune responses.52 Mast cells have a newly identi ed role in host defense and in ammatory responses to bacterial pathogens.9 Their role is in part due to the release of pro-in ammatory mediators during bacterial infection, which is critical for recruiting and activating other innate host defense cells such as neutrophils. However, mast cells are also phagocytic, have microbicidal properties, and can act as antigen presenting cells to the adaptive immune system.9 The role for mast cells in host protective responses appears to be as a sensor of bacterial invasion. Unlike IgEmediated responses, bacterial products seem to elicit a highly regulated and selective response from mast cells.

HUMORAL MEDIATORS OF INFLAMMATION Complement The complement cascade is a fundamental part of the in ammatory response. Activation of the complement cascade, either by immune complexes (classical pathway) or by bacteria or bacterial products, polysaccharides, viruses, fungi, or host cells (alternative pathway), results in the deposition of complement proteins on the activating surface and the release of soluble proteolytic fragments of several complement components.53 In particular, activation of either pathway results in the deposition of various fragments of the complement protein C3, which are potent activators of neutrophils and monocytes.53 Opsonization of particles with C3 fragments constitutes a major mechanism of target

recognition and phagocyte activation.54 During the activation of the complement cascade culminating in deposition of C3, soluble fragments of C3 (C3a), C5 (C5a), and C4 (C4a) are liberated. These fragments, termed anaphylatoxins, have potent e ects on tissues and cells during in ammation. Perhaps most notably, they are chemotactic for neutrophils (particularly C5a), activate neutrophil and mast cell degranulation, and stimulate reactive oxygen metabolite release from neutrophils.53 The termination of the complement cascade results in the formation of a membrane attack complex in membranes at the site of complement activation. If this occurs on host cells such as endothelium, the cell may be irreversibly injured. Although the primary source of complement is plasma, epithelial cells of the gastrointestinal tract also produce C3, suggesting that local production and activation of the complement cascade during inflammation occurs in intestinal tissues. It is clear that if the regulatory mechanisms of the complement cascade fail, then the in ammatory response may be inappropriate and tissue injury can occur. The role of complement in gastrointestinal in ammation has been most extensively studied in models of ischemiareperfusion injury. Activation of the complement cascades has a major role in altered endothelial and epithelial permeability in these models. Several lines of evidence support the importance of complement in intestinal injury. Mice de cient in C3 or C4 are protected against ischemia-reperfusion injury.55 Moreover, administration of monoclonal antibodies against C5 reduced local and remote injury and in ammation during intestinal reperfusion injury in a rat model.56 Administration of a soluble form of complement receptor 1, a regulatory protein that halts the complement cascade by dissociating C3 and C5 on host cell membranes, reduced mucosal permeability, neutrophil in ux, and leukotriene B4 production during ischemia-reperfusion injury in rats and mice.55,57 Although neutrophils and mast cells mediate many of the pathophysiologic e ects of the complement cascade, the membrane attack complex may have a primary role in altered vascular permeability during ischemia-reperfusion injury.58

Contact System The contact system of plasma is initiated by four components: Hageman factor (HF), prekallikrein, Factor XI, and high-molecular-weight kininogen. HF is a large plasma glycoprotein that binds avidly to negatively charged surfaces.59 Bacterial cell walls, vascular basem*nt membranes, heparin, glucosaminoglycans, and other negatively charged surfaces in the intestine capture HF and the other three important initiators of the contact system in a large multimolecular complex. Of the surfaces that bind HF, the ECM is an extremely potent activator of the contact system. Once bound, HF is converted to HF-α, which cleaves prekallikrein to kallikrein and Factor XI to Factor XIa. The ultimate result is further cleavage of HF by kallikrein and triggering of the contact system cascade, activation of intrinsic coagulation by Factor XIa, activation of the alternative pathway by HF, and proteolytic cleavage of high-molecular-weight kininogen by kallikrein, releasing biologically active kinins. The products of the contact system, particularly bradykinin, have several important biologic properties that drive many of the vascular and leukocyte responses during in ammation.59 Bradykinin induces endothelial cell contracture and intracellular tight junction alterations that increase vascular permeability to uid and macromolecules. Bradykinin also a ects vascular smooth muscle contracture, resulting in either vasoconstriction or vasodilation, depending on the location. Bradykinin also increases intestinal motility, enhances chloride secretion by the intestinal mucosa, and intensi es gastrointestinal pain. In neutrophils kinins stimulate the release of many in ammatory mediators, including cytokines, prostaglandins, leukotrienes, and reactive oxygen intermediates.60 Kallikrein cleaves C5 to release C5a, a potent chemotactic factor for neutrophils, and thus has a role in recruiting and activating inflammatory leukocytes. The plasma kallikrein-bradykinin system is activated in a variety of acute and chronic in ammatory diseases of the gastrointestinal tract.61,62 Recent evidence has demonstrated that blockade of the pathophysiologic e ects of bradykinin has clinical applications. Oral or intravenous administration of the bradykinin receptor antagonist icatibant reduces the clinical signs, onset of diarrhea, and many of the histopathologic changes in experimental models of colitis in mice.63 Inhibition of kallikrein by oral administration of P8720 attenuated the intestinal in ammation, clinical score, and systemic manifestations in a model of chronic granulomatous enterocolitis.62 Thus the contact system is a viable therapeutic target for inflammatory diseases of the intestine.

TISSUE INJURY DURING INFLAMMATION Changes in blood ow to the mucosa and other regions of the intestine that reduce perfusion of the tissues can potentiate the initial damage caused by infection or injury. For example, reperfusion of ischemic tissues is associated with platelet and neutrophil clumping in the small vessels of the mucosa, which can impede blood ow.64 Platelets are activated and adhere to exposed basem*nt membrane and activated endothelial cells and provide a surface for leukocyte adhesion. The accumulation of platelets and leukocytes can signi cantly reduce vessel diameter and blood flow while potentiating local coagulation and thrombus formation. Soluble mediators released by activated leukocytes and endothelial cells also a ect blood ow. Histamine and the vasoactive lipids derived from arachidonic acid (leukotrienes, prostaglandins, thromboxane, prostacycline, and PAF) have a prominent role in regulating local perfusion during in ammation and may have systemic e ects on blood ow as well. Procoagulant mediators released by in ammatory cells in response to the in ammatory process (i.e., tissue factor produced by macrophages or endothelial cells), exposed basem*nt membrane proteins, and bacterial components can trigger the contact system and the coagulation and complement cascades, the products of which a ect blood ow. Nitric oxide, whether produced by endothelial cells or leukocytes (macrophages), is a potent regulator of blood ow and has a

signi cant role in the control of perfusion during in ammation.65 Many of the mediators that a ect perfusion also a ect endothelial permeability, altering osmotic and hydrostatic balance and tissue edema. In extreme cases local and systemic coagulopathies initiated by vascular injury and absorption of microbial products and in ammatory mediators induce a hypercoagulable state, leading to microthrombus formation, which can reduce blood flow or macrothrombus formation, causing tissue infarction. The cellular mediators of in ammation have the potential to in ict severe injury to intestinal tissues. Neutrophils have an important role in the pathophysiology of many intestinal diseases, including ischemia-reperfusion injury,18,25 infectious enterocolitis,26,28,66 nonsteroidal anti-in ammatory drug–induced mucosal ulceration,29 and others. Depletion of neutrophils, blockade of their emigration into tissues, or inhibition of neutrophil activation reduces the severity of these and other in ammatory diseases.67 Thus many anti-in ammatory therapies are emerging that specifically target neutrophil adhesion, migration, and activation. Migration of neutrophils through endothelium during emigration into in amed tissues is remarkable in that the permeability of the endothelial monolayer is preserved under most circ*mstances. However, there is a limit above which neutrophil migration alters the permeability characteristics of the endothelium. The e ect is in part physical in that the mere movement of large numbers of neutrophils through the endothelium is su cient to mechanically disrupt the tight junctions and in part due to toxic products of neutrophils that damage endothelial cells and basem*nt membranes.64,68 Serine proteases (particularly elastase) and metalloproteinases released by degranulating neutrophils liquefy tissue matrix proteins and cleave cell surface proteins that make up endothelial intercellular junctions to ease neutrophil migration to the site of infection.23 These degradative enzymes are particularly damaging to basem*nt membranes and the cellular barriers of the endothelium, thus contributing to vascular permeability (and local tissue edema) and thrombosis. The permeability may be a ected to the extent that not only water but also macromolecules (e.g., albumin, matrix proteins, complement) leak into the interstitium. Blockade of neutrophil adhesion to endothelium with anti-β2 integrin antibodies has a sparing e ect on the microvasculature in experimental intestinal ischemia-reperfusion injury, reducing the alterations in vascular permeability and histopathologic evidence of microvascular damage.64 Similar to the endothelium of in amed tissues, massive neutrophil transmigration occurs across the epithelium in response to infection or injury. Neutrophil transepithelial migration increases epithelial permeability by disrupting tight junctions.68 Like the endothelium, neutrophils disrupt the epithelial barrier mechanically as they migrate through (see Figure 15-2). Proteases, particularly elastase, degrade basem*nt membrane components and tight junction proteins. Protease activated receptor-2 activated by neutrophil granule serine proteases alter epithelial and endothelial tight junction integrity. Pro-in ammatory products of activated neutrophils (TNF-α and IFN-γ) increase tight junction permeability by direct e ects on enterocytes. Prostaglandins released by activated neutrophils stimulate epithelial secretion, thus contributing to diarrhea. Subepithelial accumulation of neutrophils can lead to de-adhesion of the epithelial cells from the basem*nt membrane and mild to severe ulceration. The physiologic result of the e ects of neutrophils and their products on the epithelial barrier includes protein-losing enteropathy and absorption of bacterial cell wall constituents, which potentiates the local and systemic in ammatory responses. Neutrophils in in amed tissues stimulated by potent host-derived activators (such as IL-1β and TNF-α) and bacterial products (LPS) release copious amounts of reactive oxygen intermediates (see Figure 15-2). Although these oxygen and oxyhalide radicals are important for killing pathogens, they are also potentially toxic to epithelial and endothelial cells and matrix proteins. Reactive nitrogen intermediates (RNIs), produced primarily by macrophages during in ammation, combine with ROI to form peroxynitrites, which are particularly toxic.12 In addition to injury to mucosal tissues, ROI also have an as yet ill-de ned role in recruiting and activating neutrophils, thereby potentiating the in ammatory response.69 In support of the role of ROI in in ammatory diseases of the gastrointestinal tract, administration of inhibitors of ROI production or pharmacologic ROI scavengers can be protective in many models of reperfusion injury or enterocolitis. Many therapies are aimed at inhibiting neutrophil activation, and e ector functions in tissues have been evaluated for use in intestinal diseases. Phosphodiesterase inhibitors, by causing cyclic adenosine monophosphate (cAMP) accumulation in neutrophils, are anti-in ammatory by virtue of their ability to suppress neutrophil activation and ROI production. New phosphodiesterase inhibitors selective for the predominant neutrophil isoform of phosphodiesterase hold promise for use in many inflammatory diseases. Subepithelial mast cells also have an important role in altering epithelial permeability in in amed intestine. During the intestinal hypersensitivity response subepithelial mast cell release of mast cell protease tryptase by degranulation increases epithelial permeability via an e ect on tight junctions.44,70,71 This alteration in tight junction permeability results in enhanced transepithelial ux of macromolecules, including proteins and bacterial products. Cytokines released by mast cells and phagocytes also regulate tight junction permeability. IL-4, a product of mast cells and macrophages, has been demonstrated to increase epithelial permeability.72 Moreover, TNF-α and IFNγ, products of many inflammatory cells, synergistically increase tight junction permeability.73

PATHOPHYSIOLOGY OF DIARRHEA L. Chris Sanchez Acute equine colitis causes rapid, severe debilitation and death in horses (more than 90% of untreated horses die or require euthanasia).1 Since 1919 several reports have described a number of acute diarrheal conditions in the horse that appear to share a common characteristic

clinical presentation.2 Diarrhea associated with acute equine colitis occurs sporadically and is characterized by intraluminal sequestration of uid; moderate to severe colic (abdominal pain); and profuse watery diarrhea with resultant endotoxemia, leukopenia, and hypovolemia.3,4 Causes of acute colitis and therapeutic options are discussed elsewhere in this chapter. Although the mechanisms responsible for the uid losses are not known, in ammatory cells may play an integral role because this condition is characterized by large numbers of granulocytes in ltrating the large intestinal mucosa.5-9 Equine cecal and colonic tissues collected during the acute stages of experimentally induced acute equine colitis (Potomac horse fever, lincomycin with and without Clostridium spp. inoculation, nonsteroidal anti-in ammatory drug administration) reveal the presence of numerous neutrophils and eosinophils in the lamina propria and submucosa.5,8,10,11 Granulocyte-derived reactive oxygen intermediates are crucial to antimicrobial defenses in the gut and stimulate chloride and water secretion by interactions with enterocytes.12,13 Normal equine intestinal tissue is unique compared with that in most other mammalian species for a preponderance of eosinophils located in the intestinal mucosa and submucosa.14,15 Production of ROIs by stimulated phagocytic granulocytes following mucosal barrier disruption may be responsible for the massive fluid secretory response that occurs during the early stages of acute equine colitis. Colitis refers to in ammation and mucosal injury of the colon and cecum (typhlocolitis) that may occur in response to a number of causes.16 The cause of the colonic injury may be well-de ned such as in naturally occurring infectious or experimentally induced colitis. However, many cases of human and animal diarrhea have a speculative or unknown diagnosis or no diagnosis. Irrespective of the underlying or initiating cause of colonic injury, the colon apparently has a limited repertoire of responses to damage because most forms of colitis demonstrate similarities in histopathologic appearance and clinical presentation. Various degrees of mucosal erosion and ulceration, submucosal and mucosal edema, goblet cell depletion, and presence of an in ammatory cellular in ltrate within the mucosa and submucosa are common to many types of human and animal colitis.15,16 Characteristic clinical manifestations include intraluminal uid sequestration; abdominal discomfort; hypovolemia; and most often profuse, watery diarrhea.

PATHOPHYSIOLOGY OF COLITIS Large bowel diarrhea results from abnormal uid and ion transport by cecal and colonic mucosa. Loss of uid by the large intestine can result from malabsorptive or hypersecretory processes and is often a combination of the two.17 Colonic secretory processes are a function of the crypt epithelium, whereas absorptive processes are limited to surface epithelial cells.18,19 Under normal baseline conditions an underlying secretion by crypt epithelium is masked by a greater rate of surface epithelial cell absorption. Abnormal forces in uencing the rates of secretion and absorption can result in massive, uncontrolled secretion and malabsorption by large intestinal mucosal epithelial cells, leading to rapid dehydration and death.17-19 Two intracellular processes control colonic secretion: the cyclic nucleotide cAMP and cyclic guanosine monophosphate (cGMP) and the calcium system.20,21 Agents may activate adenyl cyclase (vasoactive intestinal peptide, prostaglandin E2 [PGE2]) or guanyl cyclase (bacterial enterotoxins) and induce increases in cAMP or cGMP, respectively. This reaction causes phosphorylation of speci c protein kinases that induce the actual apical and basolateral membrane transport events. Increases in intracellular free calcium may arise from cyclic nucleotide– dependent release of stored calcium within the cell or from increased calcium entry across the cell membrane.17-19 Calcium may act through calmodulin, which then can activate membrane-phosphorylating protein kinases. At least four central systems control intestinal secretion: (1) the hormonal system, (2) the enteric nervous system, (3) bacterial enterotoxins, and (4) the immune system.21,22 Hormonal control of colonic electrolyte transport is exerted primarily through the renin-angiotensinaldosterone axis.23 The enteric nervous system controls transport through three separate components: (1) extrinsic nerves of the parasympathetic and sympathetic pathways; (2) intrinsic ganglia and nerves, which secrete a variety of neurotransmitters, including peptides; and (3) neuroendocrine cells (intraepithelial lymphocytes) that reside in the epithelium and release messengers onto the epithelial cells in a paracrine manner.17,21-23 Many bacterial enterotoxins can induce intestinal secretion by cAMP or cGMP signal transduction.24 Bacterial enterotoxins can stimulate release of mediators (such as substance P) from primary a erent neurons, which then a ect enteric neurons, often propagating neurogenic inflammation.25 Preformed in ammatory mediators such as histamine, serotonin, or adenosine and newly synthesized mediators such as prostaglandins, leukotrienes, PAF, various cytokines, the inducible form of nitric oxide, and reactive oxygen metabolites can initiate intestinal secretion by directly stimulating the enterocyte and by acting on enteric nerves indirectly to induce neurotransmitter-mediator intestinal secretion.22 For instance, when added to the T84 colonic cell line, the known mast cell mediators histamine, adenosine, and PGD2 induce chloride secretion.26,27 Prostaglandins of the E and F series can cause an increase in chloride secretion in intact tissue and isolated colonic cells.28,29 Leukotrienes, PAF, and a number of cytokines have been shown to have no e ect on T84 cell secretion but have a signi cant e ect on electrolyte transport in intact tissue, suggesting that intermediate cell types may be involved in these secretory responses.30-32 The epithelial cell chloride secretory response occurs via prostaglandin- and adenosine-mediated increases in cellular cAMP, whereas histamine acts by H1 receptor induction of phosphatidylinositol turnover, production of inositol triphosphate, and mobilization of intracellular calcium stores.22 Lipoxygenase products (leukotrienes) are capable of activating a colonic secretory response and do not appear

to involve the cyclic nucleotides or calcium ions.30 Phagocyte-derived reactive oxygen mediators (ROMs) can induce colonic electrolyte secretion in vitro, suggesting that oxidants may contribute directly to the diarrhea associated with colitis.33 Reactive oxygen species initiate the secretory response by increasing cellular cAMP or stimulating mesenchymal release of PGE2 or prostacyclin, which in turn stimulates the epithelial cell or enteric neuron, respectively.33-36 Sodium nitroprusside, an exogenous source of nitric oxide, stimulated an increase in chloride secretion in rat colon that was mediated by cyclooxygenase products and enteric neurons.37 Table 15-1 summarizes in ammatory mediator–induced epithelial cell chloride secretion. TABLE 15-1 Inflammatory Mediators that Stimulate Epithelial Cell Chloride Secretion Mediator Prostaglandin E2 Vasoactive intestinal peptide Endotoxin Serotonin Interferon-γ Interleukin-1 and interleukin-1β Histamine (H1) Bradykinin Reactive oxygen mediators Thromboxanes Lipoxygenase products Platelet-activating factor Adenosine

Action Increases Cl secretion Decreases neutral NaCl absorption Increases cAMP-mediated NaCl secretion Activates cholinergic nerves Increases Na absorption Increases cell membrane permeability Increases fluid and electrolyte secretion Decreases tight junctions and causes increase in cell membrane permeability Increase prostaglandins E2 and F 1α and thromboxane B2 . Increases Cl secretion via Ca-mediated pathways Increases Cl secretion through prostaglandin-mediated pathways Increase Cl secretion Increase Cl secretion Decrease neutral NaCl absorption Increase Cl secretion via prostaglandin-mediated pathways Increases Isc (Cl secretion) Increases Cl secretion

Reference 38 39 40 41 42 43 44 26, 45 46, 47 33, 48 49 50 51 27

cAMP, Cyclic adenosine monophosphate.

ROLE OF INFLAMMATORY CELLS Acute colitis rarely develops as a result of a simple cause or e ect phenomenon but instead is in uenced by many extrinsic and intrinsic host and microorganism factors. In ammatory mediators released from mast cells and monocytic or granulocytic phagocytes cause intestinal chloride and water secretion and inhibit neutral sodium and chloride absorption.21,22,52 In ammatory cells, particularly the phagocytic granulocytes, play an important role in mucosal pathophysiology in cases of colitis.13,53 Large numbers of these cells are observed on histopathologic examination of tissues from human and animal cases of colitis. Products of cell activation stimulate direct and indirect secretory responses in intestinal cells and tissues.21,22 Products of phagocyte secretion may amplify the in ammatory signal or have e ects on other target cells in intestine such as enterocytes and smooth muscle cells.

ROLE OF PHAGOCYTE-DERIVED REACTIVE OXYGEN METABOLITES The nicotinamide adenine dinucleotide phosphate (NADPH)-oxidase system of phagocytes (neutrophils, eosinophils, monocytes/macrophages) is a potent inducer of superoxide radicals used as a host defense mechanism to kill invading microorganisms.13 During inappropriate stimulation such as in ammation, trauma, or ischemia followed by reperfusion, increased levels of toxic oxygen species are produced, causing damage to host tissues. Engagement of any of several receptor and nonreceptor types including phagocytosis mediators, chemotactic agents, various cytokines, and microbial products can stimulate phagocytes.13 Resident phagocytes or those recruited to colonic mucosa early in the disease process are considered to augment mechanisms causing uid and electrolyte secretory processes, a so-called amplification process.54,55 Activation of the respiratory burst results in the production and release of large amounts of superoxide anion (O2–) and H2O2.56 In addition to these ROMs, activated phagocytes secrete peroxidase enzyme (myeloperoxidase from neutrophils and eosinophil peroxidase from eosinophils) into the extracellular space. The peroxidases catalyze the oxidation of Cl– by H2O2 to yield HOCl, the active ingredient in household bleach products. The peroxidase-H2O2-halide system is the most cytotoxic system of the phagocytes; HOCl is 100 to 1000 times more toxic than O2– or H2O2. HOCl is a nonspeci c oxidizing and chlorinating agent that reacts rapidly with a variety of biologic

compounds, including deoxyribonucleic acid (DNA), sulfhydryls, nucleotides, amino acids, and other nitrogen-containing compounds. HOCl reacts rapidly with primary amines to produce the cytotoxic N-chloramines. The mechanisms by which these substances damage cells and tissue remain speculative, but possibilities include direct sulfhydryl oxidation, hemoprotein inactivation, protein and amino acid degradation, and inactivation of metabolic co-factors of DNA.57 Luminal perfusion of speci c ROMs increased mucosal permeability, and serosal application caused increases in Cl– secretion in vitro.58 Tissue myeloperoxidase activity, an index of tissue granulocyte in ltration, is used clinically and experimentally to assess degree of intestinal in ammation.59 Myeloperoxidase activity is elevated in acute are-ups of human in ammatory bowel disease and various animal models of acute colitis.59-61 The acute in ammatory response in these conditions is characterized predominantly by neutrophils, the predominant source of myeloperoxidase activity. However, this assay measures total hemoprotein peroxidase, which includes monocyte and eosinophil peroxidase in addition to neutrophils.62 Moreover, levels of peroxidase activity in equine circulating eosinophils are greater than in circulating neutrophils,63 and this may apply to resident tissue eosinophils as well. Arachidonic acid metabolites are thought to play a role in intestinal in ammation in diarrheal disease.22 Elevated levels of these intermediate metabolites have been demonstrated in natural disease and experimental models of colitis and appear to parallel increases in ROMs in in amed intestine.64 Addition of H2O2 or HOCl to rat colonic tissue in Ussing chambers induces PGE2 release and active Cl– secretion.36,65 Prostaglandins can stimulate increases in Cl– secretion in intact intestinal tissue35,65,66 and in isolated colonic T84 cells.34,36 Interactions between ROMs and mesenchymal release of PGE2/PGI2 may be relevant to the mechanisms producing the diarrheic condition. Fibroblasts co-cultured or juxtaposed to colonic T84 cells greatly increased the Cl– secretory response to H2O2 in vitro through the release of PGE2.34 In addition, equine colonic mucosa has an increased sensitivity to endogenously released prostaglandin by exhibiting a signi cant secretory response under in vitro conditions.67

ROLE OF ENDOTOXIN, MALNUTRITION, IMMUNODEFICIENCY, AND INTESTINAL MICROFLORA Endotoxin Endotoxin, the LPS component of the outer cell wall of gram negative bacteria, is present in large quantities in the large intestine of healthy horses.63,67 The intact bowel forms an e ective barrier to the transport of signi cant amounts of these highly antigenic toxins, but the diseased gut absorbs these macromolecules in large amounts, causing the subsequent adverse systemic e ects that are often life threatening.68 A complete review of endotoxemia and systemic inflammatory response syndrome (SIRS) is presented later in this chapter. Endotoxins trigger mucosal immune cells and subsequent release of in ammatory mediators in cases of colitis. The rst report of experimentally induced endotoxemia described clinical signs and hematologic ndings that closely paralleled those reported for severe colitis in horses.69 Studies in which endotoxin was administered intravenously in human beings and laboratory animals caused signi cant doserelated gastrointestinal changes, ranging from mild diarrhea to bloody, watery diarrhea.70,71 In vitro studies on the e ects of endotoxin on intestinal water and electrolyte transport in adult male rats showed a signi cant decrease in net colonic sodium absorption and increased colonic permeability.41 In animal models of protein energy de ciency, endotoxin-induced mortality increased compared with that of wellnourished control animals. Endotoxin depresses lymphocyte responses to speci c mitogens.72 Thus the adverse e ects of malnutrition and endotoxin are mutually aggravating. In horses endotoxin negatively a ects gastrointestinal motility, including in the cecum and right ventral colon, and perfusion.73

Immunodeficiency The importance of a normal immune system to the defense of the mucosal surface of the gastrointestinal tract is evident in the immunosuppressed state. Primary immunode ciencies a ecting the gastrointestinal tract are well documented. Common agammaglobulinemia is the most frequently reported gastrointestinal disease and causes B cell de ciency–associated giardiasis in other species.74 In horses severe combined immunode ciency can result in diarrhea secondary to adenovirus, coronavirus, and Cryptosporidium infection.75,76 Interestingly, selective immunoglobulin A (IgA) deficiency rarely results in intestinal disease because of a speculated increase in mucosal IgM response. However, combined IgA and IgM deficiencies with a higher incidence of intestinal disease occur. A selective deficiency of secretory IgA has been associated with intestinal candidiasis in other species. Certain mucosal pathogens may enhance their pathogenicity by producing IgA proteases.74 Acquired immunode ciency or immunosuppression in adults can result from infectious diseases (particularly viral), nutritional de ciencies, aging phenomenon, and drugs (corticosteroids, azathioprine, cyclophosphamide). Chronic salmonellosis was documented secondary to adult-onset B cell deficiency in a Quarter Horse.77

Nutritional Deficiencies Nutrition is a critical determinant of immunocompetence and risk of illness.78 Impaired systemic and mucosal immunity contributes to an increased frequency and severity of intestinal infections observed in cases of undernourishment. Abnormalities occur in cell-mediated immunity, complement system, phagocytic function, mucosal secretory antibody response, and antibody a nity. Morbidity caused by

diarrheal disease is increased particularly among individuals with stunted growth rate because of malnourishment.79 The critical role of several vitamins and minerals in immunocompetence has been substantiated in animals deprived of one dietary element and ndings in human patients with single-nutrient de ciency. Nutritional de ciency can cause increased colonization of the intestine with microorganisms, alter the symbiotic characteristics of resident intestinal bacterial populations, and impair defenses of the gastrointestinal tract, allowing increased risk of systemic spread of infection and absorption of macromolecules (in particular, endotoxin).

Intestinal Microflorae Indigenous micro orae greatly impede colonization of the gastrointestinal mucosa by pathogenic organisms. The ability of a potential pathogen to initiate an infection depends on its ability to breach the mucosal epithelial barrier. Resident microbes protect against pathogens by producing bacteriocidal substances and competing for available mucosal sites and nutrients.80 In addition, micro orae are important for development of the adaptive immune system.81 Alterations in the gastrointestinal micro orae, or their interaction with the innate immune response, can lead to disorders such as in ammatory bowel disease and the irritable bowel syndrome in humans.82 In fact, hyperresponsiveness to commensal bacteria appears vital to the development of in ammatory bowel disease in humans.83 Whether a similar association is present in horses is currently unknown.

Factors Affecting Motility Disturbances in motility patterns occur during in ammatory diseases of the colon, but the role of motility alterations in the pathogenesis of diarrhea remains unclear. Invasive bacteria cause characteristic motor patterns in the colon consisting of rapid bursts of motor activity that appear to decrease transit time through the large intestine. The result is reduced clearance of bacteria from the large intestine, which may contribute to the virulence of the organism.84 Absorption of endotoxin and the release of in ammatory mediators such as prostaglandins disrupt the motility patterns of the large intestine, resulting in less coordinated contractions, and may contribute to the alterations in motility seen with invasive bacteria. Although the e ect of endotoxin and prostaglandins on transit time is not profound, the disruption of coordinated activity may play a role in causing diarrhea.85 Thorough mixing and prolonged retention time of ingesta are important not only in microbial digestion of nutrients but also in absorption of microbial by-products and uid.23 The ingesta is viscous and therefore must be mixed to bring luminal ingesta in contact with the mucosa for absorption.86 In addition, poor mixing increases the thickness of the unstirred layer, decreasing contact of ingesta with the mucosa and decreasing absorption.23 Progressive motility must be present, however, if a diarrheal state is to occur.23 Ileus may be accompanied by increased uid in the lumen of the large intestine, but without progressive motility the uid is not passed. Frequently, acute colitis causes a period of ileus characterized by scant stool. Diarrhea is apparent only when motility returns and the ingesta is passed. Increased progressive motility has been suggested to cause diarrhea by decreasing transit time and is thought to play a role in irritant catharsis and in the mechanism of action of some laxatives.87 Irritation and distention increase motility and may well decrease transit time, but increased secretion also is thought to contribute to diarrhea caused by these substances.88

PATHOPHYSIOLOGY OF MUCOSAL INJURY AND REPAIR Anthony T. Blikslager

MUCOSAL BARRIER FUNCTION To gain an appreciation of the mechanisms whereby the mucosa is injured and subsequently repaired, it is important to understand how the integrity of the mucosa is physiologically regulated. Regulation of mucosal integrity is referred to as mucosal barrier function, which is vital because it prevents bacteria and associated toxins from gaining access to subepithelial tissues and the circulation. However, the mucosa has two con icting functions: It must serve as a protective barrier while continuing to absorb solutes necessary to maintain well-being of the host. This con ict is most notable at the intercellular (paracellular) space, which allows passage of select solutes and water1-4 but which does not admit large molecules, including bacterial toxins.5 The paracellular space is almost exclusively regulated by the tight junction,6 which is the interepithelial junction at the apical-most aspect of paracellular space. Although these tight junctions were originally viewed as inert cellular adhesion sites, it has become clear in recent years that tight junction permeability is dependent on tissue-speci c molecular structure and regulated by a complex array of intracellular proteins and the cytoskeleton. Tight junctions consist of a group of transmembrane proteins that interdigitate from adjacent cells. Although occludin was originally thought to be the predominant tight junction transmembrane protein, a group of proteins termed claudins appear to “ ne tune” the function of the tight junction. For example, select claudins are responsible for the relative porosity of the barrier to select electrolytes based on their charge within the paracellular space.7 These transmembrane proteins interact with the cytoskeleton via a series of intracellular proteins, including zonula occludens (ZO)-1, ZO-2, ZO-3, cingulin, and others.8 In addition, local regulatory proteins such as the small GTPase Rho are critical to tight junction function. In general, the relative contractile state of the actin cytoskeleton determines the degree to which tight junctions are open or closed, but the complexities of regulation of this process are poorly understood.9,10

The most sensitive measure of mucosal barrier function is transepithelial electrical resistance, which is measured by mounting mucosa in an in vitro system called an Ussing chamber. This measurement is largely a re ection of the permeability of mucosa to ions.11,12 There are two routes ions may follow when traversing epithelium: transcellular and paracellular.5 Since cell membranes have a resistance to passive ow of ions 1.5 to 3 log units greater than that of the epithelium as a whole, measurements of transepithelial resistance largely re ect the resistance of the paracellular space and in particular the tight junctions that regulate paracellular ow of ions.12 Because tight junctions di er in structure from di erent portions of the mucosa,13 measurements of transepithelial resistance re ect the net resistance of epithelium of variable permeability within a given tissue. For example, tight junctions in the intestinal glandular structures called crypts are leakier than those in the surface epithelium because of fewer and less-organized tight junction strands.11,14 Conversely, surface epithelium has a greater number of well-organized tight junction strands that result in epithelium with a relatively high resistance.11 This correlates well with the absorptive function of epithelium located on the mucosal surface and the secretory function of crypt epithelium. Structure of tight junctions also varies with the segment of intestine. For example, tight junctions have more strands in the ileum than the jejunum, which is re ected by a higher transepithelial resistance in the ileum.15 In addition, cells are more closely apposed at the level of the tight junction within the colon. This is in keeping with the largely absorptive role of the colon and is advantageous given the hostile microbial environment of the colon.

Gastric Mucosal Barrier Function There are four regions of the stomach, based on the type of mucosal lining (in an orad to aborad order): nonglandular strati ed squamous epithelium, cardiac epithelium, proper gastric mucosa, and pyloric mucosa.16 Strati ed squamous epithelium has distinct di erences in terms of barrier function compared with the remainder of the gastrointestinal tract. This epithelium has baseline transepithelial resistance measurements of approximately 2000 to 3000 Ω˙cm2, which is an order of magnitude higher than the adjacent cardiac mucosa.17,18 Thus the strati ed squamous mucosa is exceptionally impermeable. This is the only mechanism this mucosa has to defend itself against injury. The strati ed squamous epithelium consists of four layers: the outer stratum corneum, stratum transitionale, stratum spinosum, and the basal stratum germinativum. However, not all layers contribute equally to barrier function, which is largely composed of interepithelial tight junctions in the stratum corneum, and mucosubstances secreted by the stratum spinosum.17,19 The relative impermeability of strati ed squamous mucosa can be demonstrated by the e ects of HCl on this type of epithelium in vitro, which has very little e ect until it reaches a pH of 2.5 or below.18 Thus, although the majority of the literature on equine ulceration pertains to the e ects of HCl and inhibitors of HCl secretion,20-23 other factors may be critical to the development of gastric ulcer disease. The site of HCl secretion (proper gastric mucosa) is also protected from so-called back-di usion of H+ ions by a relatively high transepithelial electrical resistance (compared with cardiac mucosa), but there are also a number of other critical mechanisms to prevent acid injury. The gastric mucosa secretes both mucus and bicarbonate, which together form a HCO3–containing gel that titrates acid before it reaches the lumen.11,24 The mucus layer is principally formed by glycoproteins (mucins) secreted by goblet cells but also includes other gastric secretions and sloughed epithelial cells. Mucins consist of core peptides with a series of densely packed O-linked polysaccharide sidechains that, once secreted, become hydrated and form a viscoelastic gel. However, the mucus layer does not form an absolute barrier to backdi usion of acid. Thus, for acid that does back di use into the gastric mucosa, epithelial Na+/H+ exchangers are capable of expelling H+ once the cell reaches a critical pH.11 Recent studies have renewed interest in the protective mechanisms of mucus because of the discovery of a group of compounds secreted by goblet cells called the trefoil peptides. The name of these peptides is derived from a highly conserved cloverleaf structural motif, which confers substantial resistance to degradation by proteases, including pepsin. There are three known members of this group, pS2, spasmolytic polypeptide (SP), and intestinal trefoil factor (ITF), the latter of which is solely secreted by goblet cells in the small and large intestine. Both pS2 and SP are secreted by goblet cells within the stomach and are believed to intercalate with mucus glycoproteins, possibly contributing to the barrier properties of mucus.25 These peptides also play a critical role in repair of injured mucosa.

ADAPTIVE CYTOPROTECTION An additional mucosal function that serves to reduce the level of injury is adaptive cytoprotection, wherein application of topical irritants to gastric mucosa results in subsequent protection of mucosa in response to repeated exposure to damaging agents. For example, pretreatment with 10% ethanol protected against mucosal damage in response to subsequent application of absolute ethanol, and this e ect was abolished by treatment with the cyclooxygenase (COX) inhibitor indomethacin.26 The cytoprotective e ects of prostaglandins have been demonstrated directly in studies wherein pre-administration of prostaglandins protected gastric mucosa from damage by agents such as concentrated hydrochloric acid and hypertonic saline.27 Prostaglandins appear to be cytoprotective in the stomach at doses lower than those used to inhibit gastric acid secretion, ruling out a simple antacid mechanism.28 Although not fully characterized, cytoprotection has been attributed in part to prostaglandin-stimulated mucus production.29 An associated bene cial e ect of prostaglandins is the increased production of bicarbonate, which is trapped within mucus on the surface of the mucosa.30,31 Interestingly, PGE2 appears to lose its cytoprotective activity in the presence of the mucolytic agent N-acetylcysteine. Attention has also been directed at enhanced mucosal blood ow as a potential mechanism for prostaglandin-mediated cytoprotection. For example, pretreatment with prostaglandin I2 (PGI2) protected against ethanol-

induced mucosal damage as a result of increased mucosal blood ow.32 In addition, PGE2, which is also cytoprotective despite the fact that it does not increase blood ow, prevents vascular stasis associated with irritant-induced vascular damage by inhibiting neutrophil adherence to damaged endothelium.33,34 Sensory nerves distributed throughout gastrointestinal mucosa have also been implicated in cytoprotective mechanisms. As an example of their importance in mucosal cytoprotection, pretreatment of newborn rats with capsaicin (which dose-dependently destroys sensory nerves) renders the mature rats more susceptible to gastric injury.35 Alternatively, use of a low dose of capsaicin, which stimulates rather than destroys sensory nerves, protects gastric mucosa against injurious agents.36,37 Sensory nerves contain neuropeptides such as calcitonin-gene related peptide (CGRP) and substance P, which may play a protective role via vascular mechanisms. For instance, CGRP stimulates increased gastric blood ow, which is theorized to reduce injury in much the same way as prostaglandins. In fact, recent studies suggest that the roles of prostaglandins and CGRP in gastric cytoprotection are intimately intertwined. In particular, PGI2 is believed to sensitize sensory nerves following treatment with a mild irritant, with resultant increases in CGRP release and mucosal ow. Similar studies have shown that antagonists of CGRP inhibit the cytoprotective action of PGE2.38 Another neural mediator, nitric oxide, has also been implicated in adaptive cytoprotection. Interestingly, nitric oxide has a number of actions that are similar to those of prostaglandins, including maintenance of mucosal blood flow.39

INTESTINAL BARRIER FUNCTION Regulation of barrier function in the intestine is not as well characterized as that of the stomach, although mechanisms of barrier function, including secretion of mucus and regulation of mucosal blood ow, are presumed to be similar. The proximal duodenum also has to protect itself from acid damage as it receives gastric contents, and this involves secretion of mucus and bicarbonate in much the same way as the stomach. One other mechanism that helps both the stomach and the intestine to maintain mucosal barrier function is the speed with which the mucosa repairs. Thus, for a defect to develop in the mucosal barrier, injurious factors have to outpace mucosal recovery. Such recovery initially involves epithelial migration across denuded regions of basem*nt membrane (restitution).25 This process is so rapid that epithelial defects may be resurfaced within minutes. For example, in bile salt–injured equine colon, denuded surface mucosa was completely covered by restituting epithelium within 180 minutes.40 In the small intestine intestinal villi greatly amplify the surface area of the mucosal luminal surface, which in turn takes far longer to resurface with restituting epithelium once it has become denuded.41 However, intestinal villi are able to dramatically reduce the denuded surface area by contracting.42

MECHANISMS OF GASTRIC INJURY Stratified Squamous Mucosal Ulceration Although the strati ed squamous epithelium is relatively impermeable to HCl, there are a number of factors that can dramatically enhance the damaging e ects of HCl in this epithelium. In particular, bile salts and short chain fatty acids (SCFAs) are capable of breaking down the squamous epithelial barrier at an acid pH, thereby exposing deep layers to HCl, with subsequent development of ulceration.18,43 Relatively high concentrations of SCFA normally exist within the equine stomach as a result of microbial fermentation.17 These weak acids penetrate squamous mucosa and appear to damage Na+ transport activity principally located in the stratum germinativum. Bile salts also may be present in the proximal stomach as a result of re ux from the duodenum. Although such re ux has a relatively high pH, it appears that bile salts adhere to strati ed squamous epithelium, becoming lipid soluble and triggering damage once the pH falls below 4.44 Diet and management (e.g., periods of fasting) also play crucial roles in the development of conditions conducive to gastric ulceration. Typically, there is a pH gradation in horses from proximal to distal compartments of the stomach, with the lowest pH values in the distal stomach.45 However, during periods of fasting, this strati cation is disrupted such that low pH values may be recorded in the proximal stomach.46 Fasting conditions also increase the concentration of duodenal contents within the proximal stomach, particularly bile.44

Ulceration of Proper Gastric Mucosa Proper gastric mucosa is exposed to injurious agents, including pepsin, bile, and acid. The latter is constantly secreted by parietal cells in the horse as an adaptation to near-continuous intake of roughage,16 but it is tightly regulated by enterochroma n-like (ECL) cells within the proper gastric mucosa and G- and D-cells, which are present within the pyloric mucosa. Acid secretion is ampli ed by ECL-released histamine, which interacts with H2 receptors on parietal cells, and G-cells, which release the prosecretory hormone gastrin. A combination of histamine and gastrin can have a synergistic e ect on parietal cell gastric secretion, because these mediators have distinct receptors and second messengers. On the other hand, D-cells are sensitive to an acidic environment and release somatostatin, which inhibits acid secretion.47 Nonetheless, gastric mucosa may be exposed to acid for prolonged periods of time, particularly in horses that are extensively meal fed and which do not have the benefit of roughage, which tends to buffer stomach contents.44,47 Aside from peptic ulceration, induced by combinations of acid and pepsin, research in the human eld has revealed the tremendous importance of Helicobacter pylori in inducing ulceration. Infection with this organism has the e ect of raising gastric pH because of

disruption of gastric glands, but it also induces an in ammatory reaction that causes damage.48 However, to date there is very little evidence that this organism is involved in gastric ulcers in horses. In the absence of a known role for infectious agents in gastric ulceration in animals, ulceration likely develops from injurious factors similar to those found in the proximal stomach, including gastric acid and bile. However, some factors that are important to induction of squamous epithelial ulceration may not be important in development of proper gastric mucosal ulceration. For example, feed deprivation and intensive training reproducibly induce squamous epithelial ulceration in horses but have little e ect on proper gastric mucosa in horses.49 Gastric acid likely plays a key role, whereas other factors, such as nonsteroidal antiin ammatory drugs (NSAIDs), serve to reduce gastric defense mechanisms. In particular, inhibition of prostaglandin production would reduce mucus and bicarbonate secretion while also reducing gastric mucosal blood ow.50 Some of the NSAIDs also have a topical irritant e ect, although this appears to be of minor signi cance, because the route of administration (oral or parenteral) seems to have little in uence on development of ulceration.51 The source of prostaglandins responsible for gastric protection was originally assumed to be COX-1, because this COX isoform is constitutively expressed in gastric mucosa, whereas COX-2 is not expressed in the stomach unless it is induced by in ammatory mediators. However, mice in which the COX-1 gene has been knocked out fail to develop spontaneous gastric lesions,52 possibly because of compensatory increases in prostaglandin production by COX-2.53 This concept supports recent data indicating that inhibition of both COX isoforms is required to induce gastric ulceration.54 From a clinical perspective this data indicates that drugs selective for either COX-1 or COX-2 may be less ulcerogenic in the horse because they allow the uninhibited COX isoforms to continue to produce protective prostanoids. Because COX-2 elaborates prostaglandins induced by in ammatory stimuli, preferential or selective inhibitors of COX-2 may be particularly useful because of their ability to serve as anti-inflammatory agents that are less ulcerogenic.55

INTESTINAL ISCHEMIA AND REPERFUSION INJURY The most notable cause of intestinal mucosal injury in horses, particularly those su ering from colic, is ischemia. Initially, it seems intuitive that reducing gastrointestinal blood supply would injure the mucosa. However, the anatomy of the gastrointestinal tract and the di ering structure of the intestinal mucosa at various anatomic locations have a signi cant in uence on the extent of mucosal injury. Furthermore, ischemic injury may be induced by several mechanisms, including occlusion of arterial supply by a thrombus, strangulation of intestinal vasculature, and generalized reduction in blood ow associated with various shock states. In addition, a number of seemingly distinct mechanisms of intestinal injury, such as intestinal distention, also trigger mucosal injury via an ischemic mechanism. Finally, reperfusion injury may also in uence the extent of mucosal injury following an ischemic episode, and it has been proposed as a potential site of therapeutic intervention.56,57 Thus it is critical that the mechanisms of ischemia-reperfusion injury be understood in order to develop an understanding of the severity of various clinical conditions and begin to formulate a therapeutic approach to diseases characterized by this devastating form of injury.

Regulation of Intestinal Blood Flow The intestinal circulation is capable of closely regulating blood ow during periods of low systemic perfusion pressure.58,59 Local regulation of resistance vessels within the microvasculature is particularly prominent; metabolic end products of ATP cause continued dilation of resistance vessels despite reductions in systemic arterial pressure. This results in continued perfusion of gastrointestinal tissues during the early stages of shock; other organs, such as skeletal muscle, undergo massive shunting of blood as a result of marked increases in the resistance of resistance vessels. The reasons for these di erences in regulation are not entirely clear but may be related to the relatively high level of energy required to fuel the intestinal mucosa and the serious systemic e ects of breaches in the mucosal barrier. However, as blood ow falls below a critical level, regulatory systems are no longer e ective and oxygen uptake by the gastrointestinal tissue decreases, culminating in tissue damage.58 The villus tip is the most susceptible region a ected by hypoxia in the equine small intestine, largely because of the countercurrent exchange mechanism of blood ow in the small intestinal villus.58 This countercurrent exchange mechanism is attributable to the vascular architecture, which consists of a central arteriole that courses up the core villus, arborizes at the tip, and is drained by venules coursing down the periphery of the villus.60 As oxygenated blood ows into the central arteriole, oxygen tends to di use across to the adjacent venules, which are owing in the opposite direction. This series of events takes place along the length of the villus, resulting in a villus tip that is relatively hypoxic even under normal conditions. Furthermore, when blood ow is reduced, as occurs in hypovolemic or septic shock, the countercurrent exchange of oxygen is enhanced, and the tip becomes absolutely hypoxic.58 This mechanism might explain why the small intestinal mucosa is more susceptible to ischemic injury, compared with the colon, which has no villi. For example, the duration required to produce severe morphologic damage to the equine colon is approximately 25% longer than that required to produce similar damage to the small intestine.61

Ischemic Epithelial Injury Intestinal mucosal epithelium is very susceptible to hypoxia because of the relatively high level of energy required to fuel the Na+/K+-

ATPase that directly or indirectly regulates ion and nutrient ux. The rst biochemical event to occur during hypoxia is a loss of oxidative phosphorylation. The resulting diminished ATP concentration causes failure of the energy-dependent Na+/K+-ATPase, resulting in accumulation of sodium and subsequently intracellular water. The pH of the cytosol drops as lactic acid and inorganic phosphates accumulate from anaerobic glycolysis. The falling pH damages cell membranes, including lysosomal membranes, resulting in the release and activation of lysosomal enzymes into the cytosol, further damaging cellular membranes. Damage to the cell membrane allows the accumulation of high concentrations of calcium in the cytosol, which activates calcium-dependent degradative enzymes.62 These events result in cytoplasmic blebbing of the basal membrane, with subsequent detachment of cells from the underlying basem*nt membrane. Recent studies on epithelial injury during ischemia suggest that the majority of epithelial cells undergo programmed cell death (apoptosis) during ischemia and reperfusion rather than necrosis, allowing retention of reusable components of irreversibly injured cells.63 In one study 80% of detached epithelium during small intestinal ischemia-reperfusion underwent apoptosis.64 Although the most obvious result of apoptosis is loss of surface epithelium, a number of cells on the lower portion of the villus (in the small intestine) and cells within the crypts may also undergo apoptosis that only becomes evident up to 24 hours following reperfusion of ischemic tissue.65 Morphologic changes observed in ischemic-injured small intestinal mucosa follow a similar sequence regardless of whether injury is induced by ischemia alone or ischemia-reperfusion (Table 15-2).66 Initially, epithelium separates from the underlying basem*nt membrane, forming a uid- lled space termed Grüenhagen’s space (Figure 15-3). The mechanism of uid accumulation in this space is not entirely understood but may result from continued epithelial absorption of NaCl and water, before it has fully detached from neighboring epithelial cells. This uid accumulation likely exacerbates epithelial separation from the basem*nt membrane. Subsequently, epithelium progressively sloughs from the tip of the villus toward the crypts, which are the last component of the intestinal mucosa to become injured.67-69 This likely relates to the vascular architecture, because crypts receive a blood supply that is separate from the vasculature involved in the villous countercurrent exchange mechanism. The early morphologic changes observed in the equine large colon during ischemia are somewhat di erent from those described in the equine small intestine because of the lack of intestinal villi. However, as might be expected, the more superficially located surface cells are sloughed before those in crypts61,70 The orderly progression of tissue injury has been used by one group of investigators to accurately predict survival in horses with large colon volvulus. Biopsies were taken from the pelvic exure, which accurately re ects mucosal changes along the length of the colon,71 and histologically examined for the width of the crypts and intercrypt interstitial space. The latter measurements were expressed as an interstitium : crypt width (I:C) ratio. Nonviable colon was de ned as that which had greater than 60% loss of crypt and an I:C ratio greater than 3. Using this methodology, survival was correctly predicted in 94% of horses.72 TABLE 15-2 Grading System for Ischemia-Reperfusion Injury in Small Intestinal Mucosa Grade 1 2 3 4 5

Description Separation of epithelium at the tip of the villus, creating a small space between epithelium and basem*nt membrane called Grüenhagen’s space Loss of epithelium from the tip of the villus Loss of epithelium from the upper third of the villus Complete loss of villus epithelium Injury or loss of epithelium within the crypt in addition to complete loss of villus epithelium

Adapted from Chiu CJ, McArdle AH, Brown R, et al: Intestinal mucosal lesion in low-flow states. 1. A morphological, hemodynamic, and metabolic reappraisal, Arch Surg 101:478–483, 1970.

FIGURE 15-3

Histologic appearance of Grüenhagen’s space in ischemic injured ileal mucosa. Note separation of epithelium at the tip of the villus from its basem*nt membrane, creating a space (arrows). Epithelium subsequently sloughs into the lumen (arrowheads). 1 cm bar = 100 μm

Strangulating Obstruction Since the dramatic decline in Strongylus vulgaris–induced colic, which was frequently associated with infarction of intestinal arterial blood supply,73 the vast majority of ischemic lesions are associated with strangulating obstruction. Therefore it is important to consider mechanisms of ischemic injury in horses with naturally occurring strangulating lesions. The majority of experimental work has assessed either complete ischemia (complete occlusion of the arterial blood supply)61 or low- ow ischemia (reduction of arterial blood ow).74,75 However, during intestinal strangulation a disparity between the degree of occlusion of the veins and arteries occurs whereby veins are occluded before arteries because of di erences in compliance of vascular walls. Thus strangulating lesions are typically hemorrhagic (hemorrhagic strangulating obstruction) because the arteries continue to supply blood to tissues that have little or no venous drainage. This results in ischemic injury, as previously outlined, but also in tremendous congestion of the tissues. Such hemorrhagic congestion has two opposing e ects: It disrupts tissue architecture, including the mucosa and its epithelium, but it continues to provide oxygenated blood to the tissues during much of the ischemic episode. In contrast, when strangulation results in sudden cessation of arterial blood flow (ischemic strangulating obstruction), tissues appear pale, and the mucosa rapidly degenerates because of a complete lack of oxygenated blood.69 From a clinical standpoint, this makes it di cult to assess the degree of mucosal injury in horses with strangulating injuries, because intestine that may look nonviable (dark red) may in fact have less mucosal injury than that of ischemic strangulated intestine.76 An additional consideration in clinical strangulating obstruction is the degree of ischemia that may be induced by intestinal distention. For example, experimental distention (18 cm of H2O for 2 hours) and decompression (2 hours) of jejunum resulted in a signi cant increase in microvascular permeability and a signi cant decrease in tissue oxygenation similar to that which would be expected with low- ow ischemia.77,78 In particular, microscopic evaluation of vasculature revealed capillary endothelial cell damage and local edema formation.79 These data suggest that distended intestine proximal to an obstruction may undergo mucosal injury despite its relatively normal appearance. Indeed, in one study intraluminal pressures greater than 15 cm H2O in naturally occurring cases of colic correlated with a poor prognosis for survival.80

Reperfusion Injury Although it has recently been taken for granted that reperfusion of ischemic tissues results in exacerbation of mucosal injury, it should be remembered that mechanisms underlying intestinal reperfusion injury have been largely de ned in laboratory animals under speci c conditions.81-85 On the other hand, studies on reperfusion injury in horses have had some con icting results.67,75,86 This may be attributable to the way in which the studies have been performed. In particular, the type of ischemia used in most laboratory animal studies has been low- ow ischemia (in which the blood ow is typically reduced to 20% of baseline ow), whereas studies in horses have used a number of di erent ischemic models, including various types of strangulating obstruction. Although strangulating obstruction is of great clinical relevance, this type of ischemic insult is less likely to develop into reperfusion injury.67,87,88 Conversely, low- ow ischemia appears to prime tissues for subsequent injury once the tissue is reperfused, and there is considerable evidence to support the presence of reperfusion injury in horses following low-flow ischemia.74,75,79,89 Nonetheless, low-flow ischemia may not be a common clinical entity. In addition to the type of ischemia, there are other factors involved in priming tissues for reperfusion injury, including species and anatomic-speci c variation in oxidant enzyme and neutrophil levels (Table 15-3). For example, the foal appears to have very low levels of

small intestinal xanthine oxidase, an enzyme that has been shown to play a critical role in triggering reperfusion injury in laboratory animals,83,84,95 whereas adult levels are much greater, particularly in the proximal small intestine.94 In addition, horses appear to have low numbers of resident neutrophils in the intestinal mucosa,96 and it is this population of neutrophils (rather than those recruited from the circulation) that appear to be most critical for induction of reperfusion injury.85 However, studies demonstrating reperfusion injury in the equine colon following low- ow ischemia have shown signi cant accumulation of neutrophils within the mucosa.74 Therefore a complete understanding of mechanisms of neutrophilic infiltration and the mechanisms whereby they damage tissue will require further study. TABLE 15-3 Comparison of Mean Levels of Xanthine Oxidase/Xanthine Dehydrogenase (XO/XDH) and Myeloperoxidase (as an Indication of Granulocyte Numbers) in the Small Intestine of Various Species

Reperfusion injury is initiated during ischemia when the enzyme xanthine dehydrogenase is converted to xanthine oxidase, and its substrate, hypoxanthine, accumulates simultaneously as a result of adenosine triphosphate (ATP) utilization (Figure 15-4).56,97 However, there is little xanthine oxidase activity during ischemia, because oxygen is required as an electron acceptor. During reperfusion xanthine oxidase rapidly degrades hypoxanthine in the presence of oxygen, producing the superoxide radical as a by-product.56 The superoxide radical contributes to oxidative tissue damage and, most important, activates neutrophil chemoattractants.83,84 Thus inhibition of xanthine oxidase in feline studies of intestinal ischemia/-reperfusion injury prevents in ltration of neutrophils and subsequent mucosal injury.82,83 However, inhibition of xanthine oxidase has had no e ect on ischemia/-reperfusion injury in equine small intestine86 and colon,98 suggesting that either reperfusion injury is simply a continuation of injury initiated during ischemia, as suggested in some equine studies,62 or that the classic reperfusion injury pathway is activated by alternate sources of ROMs. The latter has been suggested by studies in feline models of ischemia/-reperfusion injury, in which the source of a signi cant proportion of ROMs is unknown, and independent of xanthine oxidase and neutrophils.82

FIGURE 15-4

Intestinal reperfusion injury cascade. Reperfusion injury is initiated by elaboration of superoxide by metabolism of hypoxanthine by xanthine oxidase and subsequent infiltration of neutrophils.

In a veterinary review of the pathogenesis of intestinal reperfusion injury in the horse, the concept of a therapeutic window wherein treatment of reperfusion injury would be bene cial was suggested.56 The basis of this concept is that there are certain conditions under which ischemic injury is minimal and that tissues are severely damaged during reperfusion.87 Thus, under conditions of low- ow ischemia,

very little injury is demonstrated during 3 hours of ischemia, but remarkable injury occurs during 1 hour of reperfusion.82-84 However, a therapeutic window may not exist under conditions of strangulating obstruction wherein severe injury occurs during ischemia and minimal injury occurs during reperfusion.91 This in turn greatly reduces the clinician’s ability to ameliorate ischemia- reperfusion injury with treatments such as anti-oxidants at the time of reperfusion.

MECHANISMS OF GASTROINTESTINAL MUCOSAL REPAIR Gastric Reparative Mechanisms Mechanisms of gastric repair are highly dependent on the extent of injury. For instance, super cial erosions can be rapidly covered by migration of epithelium adjacent to the wound, a process termed epithelial restitution. However, ulceration (full thickness disruption of mucosa and penetration of the muscularis mucosa) requires repair of submucosal vasculature and ECM. This is initiated by formation of granulation tissue, which supplies connective tissue elements and microvasculature necessary for mucosal reconstruction. Connective tissue elements include proliferating broblasts that accompany newly produced capillaries that form from proliferating endothelium. Recent studies indicate that nitric oxide is critical to both of these processes,39,99 which likely explains the reparative properties of nitric oxide in the stomach.100 Once an adequate granulation bed has been formed, newly proliferated epithelium at the edge of the wound begins to migrate across the wound. In addition, gastric glands at the base of the ulcer begin to bud and migrate across the granulation bed in a tubular fashion.101 Epidermal growth factor is expressed by repairing epithelium and appears to facilitate these processes.102 In addition, these events are facilitated by a mucoid cap, which retains reparative factors and serum adjacent to the wound bed.50 Once the ulcer crater has been lled with granulation tissue and the wound has been re-epithelialized, the subepithelial tissue remodels by altering the type and amount of collagen. Despite the remodeling process, ulcers tend to recur at sites of previous ulceration, and there is the concern that this remodeling can result in excessive deposition of collagen and fibrosis.25

Intestinal Reparative Mechanisms Reparative mechanisms are similar in the intestine, except that in the small intestine mucosal villi contribute to mucosal repair. Once intestinal epithelium is disrupted, there are two events that occur almost immediately to reduce the size of the denuded portion of the villus: contraction of the villus and epithelial restitution (Figure 15-5). For example, in porcine ileum subjected to 2 hours of ischemia, villi were 60% of their former height and 50% of the denuded villous surface area was covered in attened epithelium within 6 hours.41 Villous contraction appears to be regulated by enteric nerves, because inhibition of enteric nerve conduction prevents villous shortening after injury. The contractile component of the villus is a network of myo broblasts distributed throughout the lamina propria of the villus and along the central lacteal. Inhibition of villus contraction results in retarded epithelial repair because of the larger denuded surface that remains to be covered by migrating epithelium compared with similarly injured villi that have contracted.42 Prostaglandin E2 has also been implicated in regulating villous contraction, because application of PGE2 resulted in villus contraction when perfused through normal rat ileum.103 As villi contract, assuming there is an intact basem*nt membrane, epithelium from the margins of the wound migrates in a centripetal direction to resurface toward the tip of the villus.42 The process of restitution is similar in denuded colonic mucosa, except that it may proceed more rapidly because of the lack of villi.41 Epithelial restitution is solely a migratory event that does not depend on provision of new enterocytes by proliferation. Cellular migration is initiated by extension of cellular lamellipodia that receive signals from the basem*nt membrane via integrins. Intracellular signaling converges on the actin cytoskeleton, which is responsible for movement of lamellipodia. Speci c components of the basem*nt membrane appear to be critical to the migratory process. For example, application of antibodies to collagen types III and IV, which are important components of intestinal mucosal basem*nt membrane, impeded epithelial restitution.104,105 Other elements of the basem*nt membrane, including proteoglycans, hyaluronic acid, and noncollagenous proteins such as bronectin and laminin, may also provide important signals.106 These subepithelial matrix components that facilitate restitution may form the basis for clinical treatments designed to speed up the repair process, analogous to administration of matrix components to horses with articular cartilage damage.

FIGURE 15-5

Histologic appearance of repairing intestinal mucosa 6 hours after a 2-hour ischemic episode. Note blunting of the villus, attributable to villous contraction, and evidence of epithelial restitution (arrows). 1 cm bar = 100 μm

Although epithelial restitution results in gross closure of previously denuded regions of gastrointestinal mucosa, closure of interepithelial spaces is ultimately required to restore normal epithelial barrier resistance.107 Because the tight junction is principally responsible for regulating the permeability of the interepithelial space, it is likely that repair and closure of this structure is critical to restore intestinal barrier function. Recent research indicates that prostaglandins play a vital role in recovery of tight junction resistance,108 indicating that administration of nonselective COX inhibitors to horses with colic, particularly those recovering from strangulating obstruction, may be deleterious. Therefore judicious use of NSAIDs is appropriate until more selective drugs that allow continued production of reparative prostaglandins are available for use in horses in this country. Recent studies have shown that NSAIDs preferential for COX-2 allow for optimal repair of injured intestine as compared to traditional nonselective NSAIDs.55 Once the epithelial barrier has been restored, normal mucosal architecture must be re-established to allow normal gut absorptive and digestive function. In porcine ileum subjected to 2 hours of ischemia, the epithelial barrier was restored within 18 hours, but villi were contracted and covered in epithelium with a squamous appearance. Restoration of normal villus architecture required an additional 4 days.41 The attened villus epithelium that characterizes restitution is replaced by newly proliferated crypt epithelium. Under normal circ*mstances, new enterocytes are formed by division of stem cells, of which there are approximately four at the base of each mucosal crypt. Newly divided enterocytes migrate from the crypt onto the villus.108 During migration enterocytes di erentiate and acquire speci c absorptive and digestive functions. Fully di erentiated enterocytes reside on the upper third of the villus for 2 to 3 days and are then sloughed into the intestinal lumen.109 This process is accelerated during mucosal repair, which requires increased proliferative rates. Increased proliferation may be stimulated within 12 to 18 hours by a variety of locally available gut-derived factors, including luminal nutrients, polyamines, and growth factors.41 The return of the normal leaflike shape of the villus occurs subsequent to the appearance of normal columnar epithelium.

MEDIATORS OF REPAIR Prostaglandins Although prostaglandins have been strongly implicated in mucosal cytoprotective function, relatively few studies have assessed their importance in mucosal repair. One study implicated prostaglandins in growth factor–stimulated restitution,110 but a more prominent role of prostaglandins in mucosal repair is their ability to close interepithelial tight junctions.107,111,112 For instance, ischemic-injured small intestine rapidly recovers barrier function (as measured in vitro as transepithelial resistance) in the presence of prostaglandins I2 and E2, despite the fact that these prostanoids have relatively little e ect on villous contraction and epithelial restitution. However, electron-microscopic examination of tissues reveals dilation of tight junctions in tissues treated with NSAIDs,112 whereas those additionally treated with prostaglandins have closely apposed tight junctions (Figure 15-6). Prostaglandins stimulate closure of tight junctions via the second messengers cAMP and Ca2+,105 which, interestingly, were among the rst mediators found to modulate tight junction permeability.113,114 Such tight junction closure is of considerable importance to patients with intestinal injury that are treated with NSAIDs because reduced prostaglandin levels may result in increased intestinal permeability. For example, in a study on ischemic-injured porcine ileum, treatment with the NSAID indomethacin resulted in a signi cant increase in intestinal permeability to inulin and LPS compared with tissues that were additionally treated with PGI2 and PGE2.107

FIGURE 15-6

Ultrastructural appearance of repairing ischemic-injured mucosa. A, Restituting epithelium 2 hours after a 1-hour ischemic episode in the presence of the nonselective COX inhibitor indomethacin. Note dilation of the interepithelial space and the apical tight junction (arrows), which correlates with a “leaky” intestinal barrier B, Similar restituting epithelium that has been additionally treated with PGE2 and PGI2. Note the close apposition of the tight junction (arrows) and the interepithelial space correlated with normalization of intestinal barrier function. 1 cm bar = 6 μm.

Polyamines The process of restitution is absolutely dependent on a group of compounds called polyamines.115,116 The rate-limiting enzyme in the formation of the polyamines spermine, spermidine, and putrescine is ornithine decarboxylase (ODC). In rats with stress-induced duodenal ulcers, systemic administration of the ODC inhibitor α-di uoromethyl ornithine signi cantly reduced polyamine levels and markedly reduced epithelial restitution. Furthermore, intragastric treatment of these same rats with putrescine, spermidine, and spermine prevented the delayed mucosal repair induced by α-di uoromethyl ornithine.115 Interestingly, gastric tissue levels of ODC were increased in rats with stress-induced gastric ulcers, suggesting that polyamine production is enhanced during tissue injury and may contribute to the normal rapid rate of epithelial restitution.117 The mechanisms whereby polyamines stimulate epithelial restitution are not clear. McCormack et al. hypothesized that polyamines increased transglutaminase activity, an enzyme that catalyzes the cross-linking of cytoskeletal and basem*nt membrane proteins.118 Further investigation of the role of polyamines in IEC-6 cell migration showed that depletion of polyamines resulted in disruption of the cytoskeleton and reduced the physical extension of lamellipodia.119 More recent studies have clari ed this pathway. In particular, it has been shown that polyamines regulate cytoskeletal cellular migration via activation of the small GTPase Rho-A by elevating intracellular Ca2+ levels. These elevations in Ca2+ result from polyamine regulation of expression of voltage-gated K+ channels and altered membrane electrical potential.120 Polyamines also play a role in the normal physiologic regulation of crypt cell proliferation and di erentiation.121,122 Polyamines are produced by fully di erentiated enterocytes at the villous tip and may reach the crypt either within sloughed luminal epithelium or via local villous circulation.123 After intestinal injury polyamines appear to stimulate enhanced proliferation by increasing the expression of protooncogenes, which control the cell cycle.124 The mechanism whereby polyamines in uence gene expression likely relates to the cationic nature of these compounds, which may influence the tertiary structure of negatively charged DNA and ribonucleic acid (RNA).115

GROWTH FACTORS Locally produced growth factors, including epidermal growth factor (EGF), transforming growth factor (TGF)-α, TGF-β, and hepatocyte growth factor (HGF), have the ability to modulate mucosal recovery. The most important of these growth factors in early mucosal repair events is TGF-β, which is a potent stimulus of epithelial restitution and modulator of the ECM.25 Neutralization of TGF-β retards epithelial migration in vitro, and it appears that TGF-β may serve as a point of convergence for mediators of restitution, because neutralizing TGF-β also inhibits the e ects of other peptides. However, TGF-β paradoxically inhibits epithelial proliferation, thereby reducing the supply of new enterocytes for mucosal repair. Conversely, EGF, produced by the salivary glands and duodenal Brunner’s glands, and the related TGF-α, produced by small intestinal enterocytes, are potent stimulants of enterocyte proliferation. These growth factors share approximately 30% of their amino acid structure, bind to the same receptor on the basolateral surface of enterocytes, and are not related to TGF-β.125 The physiologic role of EGF is somewhat di cult to discern because it is present in the intestinal lumen, with no apparent access to its basally located receptor.126 However, it has been proposed that EGF acts as a “surveillance agent” that gains access to its receptor during epithelial injury (when the EGF receptor would likely be exposed) to stimulate proliferation.126 TGF-α presumably has a similar role, but it is present

in greater concentrations in the small intestine because it is produced by di erentiated villous enterocytes. The mature peptide is cleaved from the extracellular component of the transmembrane TGF-α precursor and released into the lumen.125

Trefoil Peptides Another group of proreparative peptides that are locally produced within the gastrointestinal tract are the trefoil peptides. Under physiologic conditions trefoil peptides are secreted by mucus-producing cells at distinct anatomic sites. For example, the trefoil peptide pS2 is produced by gastric epithelium, whereas ITF is produced by small and large intestinal mucosa.127 However, any of the trefoil peptides may be upregulated within repairing epithelium regardless of anatomic site.25,128 In addition, trefoil peptides have the ability to induce their own expression, amplifying the level of these reparative factors at sites of mucosal repair.129 Trefoil peptides are the most potent stimulants of epithelial migration in vitro, and their e ects are independent of growth factors, including TGF-β.130 However, recent evidence suggests that EGF receptor activation is required for induction of pS2 and another of the trefoil peptides termed spasmolytic peptide in gastric epithelium in vitro. The importance of trefoil peptides to the mucosal repair response in vivo is illustrated by gene knockout studies, in which mice de cient in ITF have dramatically reduced ability to repair intestinal injury.131 In fact, detergent-induced mucosal injury was lethal because of a lack of restitution as compared with wildtype mice that fully recovered from similar mucosal injury. The fact that restitution was restored by administration of ITF has important therapeutic implications. The mechanism whereby trefoil peptides stimulate epithelial migration is yet to be fully characterized, but it appears to involve translocation of the adherens junction protein E-cadherin, thereby allowing cells to become untethered from neighboring cells.25

Intestinal Nutrients The principal metabolic fuel of enterocytes is glutamine, and for colonocytes it is butyrate. However, recent studies suggest that glutamine and butyrate have more speci c proliferative actions aside from their role as nutrients. For example, in the piglet IPEC-J2 enterocyte cell line, glutamine enhanced gene transcription by increasing mitogen-activated protein kinase activity.132,133 Similarly, butyrate stimulated mucosal growth following colonic infusion in the rat.134 Because of such growth-promoting actions, glutamine was shown to prevent intestinal mucosal atrophy and dysfunction that accompanies starvation135,136 and long-term total parental nutrition.137,138 Additionally, glutamine improves function of transplanted small intestine139,140 and protects intestinal mucosa from injury if administered before chemotherapy141 and radiation.142,143 Intestinal nutrients may also synergize with other proliferative agents. For example, administration of glutamine and TGF-α to porcine ileum that had been subjected to 2 hours of ischemia resulted in a synergistic increase in mitogen-activated protein kinase activity, enterocyte proliferation, and villous surface area.41 Although there has been a concern that such early return to normal surface area may result in dysfunctional mucosal digestive and absorptive function because of resurfacing denuded mucosa with immature epithelium, nutrients and growth factors also appear to promote early di erentiation. In the case of glutamine and TGF-α restoration of postischemic small intestine, rapid recovery of digestive enzymes was also documented.144

GASTROINTESTINAL ILEUS L. Chris Sanchez E ective gastrointestinal motility involves a complex interaction among the enteric nervous system, muscular wall, and luminal contents. Additional factors that in uence the net transit of digesta include gravity, the volume and viscosity of the contents, and pressure gradients created by simultaneous contraction and relaxation of adjacent segments of bowel. Casual use of the term intestinal motility in veterinary medicine often underestimates the complexity of the processes involved in the transit of intestinal contents. This is particularly true when the term is used to describe the frequency and or intensity of intestinal sounds, or borborygmi. The existence of borborygmi does not always equate with progressive movement of intestinal contents. Disruption to normal motility occurs commonly in horses for a variety of reasons. Examples of diseases in which altered motility may be present include grass sickness, gastroduodenal ulceration, intraluminal obstruction or impaction, excessive wall distention, strangulating obstruction, peritonitis, and in ammatory disorders such as duodenitis proximal jejunits and colitis. Ine ective intestinal motility is also a feature of several neonatal diseases, including prematurity, systemic sepsis, and perinatal asphyxia. Certain parasitic infections, electrolyte derangements, and endotoxemia can modify digesta transit in horses of all ages. General anesthesia and speci c sedatives, such as xylazine, romifidine, and detomidine, also disturb motility.

MANIFESTATIONS OF ILEUS The inhibition of propulsive bowel activity usually is referred to as ileus. Ileus is ascribed most frequently to the condition that occurs after laparotomy and is termed simple or uncomplicated postoperative ileus (POI). The term complicated or paralytic ileus describes intestinal motility disturbed for longer periods after surgery. POI in horses is associated most commonly with surgery of the small intestine, particularly after resection and anastomosis,1,2 and can have a negative e ect on short-term postoperative survival.3-6 Motility dysfunction likely is present in all horses after laparotomy, but many are a ected subclinically and require minimal or no speci c intervention. In symptomatic

animals clinical signs are apparent shortly after recovery and include colic, tachycardia, dehydration, decreased borborygmi and fecal output, and sequestration of uid within the stomach. Rectal examination and ultrasound reveal small intestinal distention with rare or absent wall movement. The severity and duration of intestinal stasis varies, lasting from minutes to days. A speci c motility disorder involving the cecum or ileocecocolic region occurs sporadically in horses.7-9 The condition most commonly occurs after general anesthesia and extra-abdominal surgery, particularly orthopedic and upper airway procedures, and therefore often is categorized as a form of POI. Other cases occur spontaneously, often in animals with painful primary conditions such as uveitis or septic tenosynovitis. In a study of 114 horses diagnosed with cecal impaction, 12 were hospitalized for a condition other than colic at the time of diagnosis and 9 others were being treated with phenylbutazone, most for a musculoskeletal injury.10 Eight of the 114 horses had undergone general anesthesia in the 8 days preceding diagnosis. The syndrome is frustrating in that clinical signs are often subtle unless cecal perforation has occurred. In horses with a cecal emptying defect after anesthesia, overt signs are usually apparent 3 to 5 days after the procedure. The earliest detectable signs include depression and a reduction in feed intake and fecal output. Ine ective emptying results in over lling of the cecum with moist contents, which is manifested by signs of mild to moderate colic. If the condition is recognized late or untreated, the cecum may rupture and result in fatal peritonitis.

PHYSIOLOGY Current understanding of motility throughout the equine gastrointestinal tract is remarkably limited, and much of our presumptive knowledge comes from work in other species. The enteric nervous system is involved in all aspects of motility, either directly via neurotransmitters or indirectly via interstitial cells of Cajal (ICC), immune, or endocrine regulation. The inherent rhythmicity of electric activity in the intestine is controlled by the ICC, specialized cells that are electrically coupled to myocytes via gap junctions.11 These cells are responsible for generating and propagating slow wave activity and hence are deemed the pacemaker cells of the intestine. A decrease in ICC density has been observed in horses with obstructive disorders of the large intestine12 and in the ileum and pelvic exure of horses diagnosed with equine grass sickness (dysautonomia),13 although such a decrease was not evident in a horse with dysautonomia that recovered.14 This alteration in ICC infrastructure appears to result in reduced slow wave activity in vitro.15 The enteric nervous system primarily controls and coordinates intestinal contraction. A combination of central and autonomic innervation in uences events, but contraction does not require external neural input. The parasympathetic supply to the gastrointestinal tract is via the vagus and pelvic nerves, and the sympathetic supply is through postganglionic bers of the cranial and caudal mesenteric plexuses. A complex network of interneurons within each plexus integrates and ampli es neural input; the intensity and frequency of resultant smooth muscle contractions are proportional to the amount of sympathetic and parasympathetic input. Additional binding sites for a number of other endogenous chemicals, including dopamine, motilin, and serotonin, exist within the enteric nervous system and on smooth muscle cells.16 Acetylcholine is the dominant excitatory neurotransmitter in the gastrointestinal tract and exerts its action through muscarinic type 2 receptors on smooth muscle cells. Sympathetic bers innervating the gastrointestinal tract are adrenergic, postganglionic bers with cell bodies located in the prevertebral ganglia. Activation of α2-adrenergic receptors on cholinergic neurons within enteric ganglia inhibits the release of acetylcholine and therefore reduces intestinal contraction. β 1-, β 2-, and β-atypical receptors are directly inhibitory to the intestinal smooth muscle.17 Inhibitory nonadrenergic, noncholinergic neurotransmitters include ATP, vasoactive intestinal peptide, and nitric oxide.18,19 These neurotransmitters are critical for mediating descending inhibition during peristalsis and receptive relaxation. Substance P is a nonadrenergic, noncholinergic neurotransmitter that may be involved in contraction of the large colon.20,21 The rate and force of intestinal contractions along the small intestine and large colon of the horse are important determinants of intestinal motility; of even greater importance to the net propulsion of digesta are the cyclical patterns of contractile activity. These patterns are known as the small intestinal and colonic migrating motility (or myoelectric) complexes (MMCs).22,23 The colonic complex usually originates in the right ventral colon and variably traverses the ascending and descending colons. Many of these complexes are related temporally to a specialized motility event of the ileum, the migrating action potential complex.24

PATHOPHYSIOLOGY Inflammation Local in ammation within the intestinal muscularis and inhibitory neural events are important initiators of intestinal ileus.25,26 Intestinal in ammation not only is important in primary intestinal diseases in horses, such as duodenitis-proximal jejunitis and colitis, but also is induced after simple intestinal handling during laparotomy. In rodents simple intestinal manipulation causes a cascade of in ammation within the muscularis that results in leukocyte in ltration and subsequent suppression of muscle contractility. The in ammatory response to bowel manipulation is not limited to the a ected tissue but can also result in global in ammation and ileus throughout the gastrointestinal tract.27 The associated in ammatory events are extremely complex, involving a milieu of pro-in ammatory cytokines, prostaglandins, and leukocytes. Depletion or inactivation of muscularis macrophage function can prevent in ammation associated with intestinal manipulation

and associated decreased contractility.28 Mast cell activation is involved in intestinal manipulation-associated POI in humans.29 In ammation associated with colonic manipulation may involve gut-derived bacterial products.30 Another factor in the development of intestinal stasis after inflammation is the local overproduction of nitric oxide caused by the upregulation of inducible nitric oxide synthase (iNOS) by resident macrophages.31 iNOS upregulation was important for initiation of the in ammatory response and inhibition of motility. Nitric oxide is a key inhibitory neurotransmitter of the nonadrenergic, noncholinergic system.19 In the horse signi cant neutrophilic in ammation is apparent in jejunum from clinical cases necessitating resection and following a period of recovery in jejunum subjected to 1 or 2 hours of ischemia.32 The ischemic tissue also had evidence of leukocyte activation, as demonstrated by calprotectin-positive cells in associated tissue histologically.

Drugs The inhibitory e ects of α2-adrenergic agonists such as xylazine and detomidine on duodenal, cecal, and large colon motility are well described, because these drugs activate presynaptic receptors within the enteric nervous system.33-39 Intravenously administered xylazine inhibits cecal and large colon motility for 20 to 30 minutes without seriously disrupting small intestinal myoelectric activity, and detomidine can reduce large intestinal myoelectric activity for up to 3 hours. Detomidine decreases duodenal motility in a dose-dependent fashion.40 The α2-antagonist yohimbine has a weak but positive effect on cecal emptying in normal ponies, suggesting that normal motility is under constant α2-adrenergic tone.34 Several opioid agonists also have documented inhibitory e ects on equine gastrointestinal motility at both a central and peripheral level. Morphine decreased frequency of defecation and fecal moisture content and increased gastrointestinal transit time in normal horses at a dose of 0.5 mg/kg twice daily for 6 days.41 Single doses of fentanyl or morphine decreased jejunal and colonic MMC activity in ponies, whereas the antagonist naloxone elicited increased propulsive activity in the colon.42 Fentanyl administered as an intravenous constant-rate infusion did not have an apparent deleterious e ect on duodenal motility.43 Butorphanol, an opioid agonist-antagonist, decreases myoelectrical activity in the jejunum but not pelvic exure.44 In another series of experiments, butorphanol alone did not decrease gastric or duodenal motility,45 but administration in combination with xylazine resulted in a synergistic inhibitory e ect that was more pronounced than that obtained by administration of xylazine alone.39 Administration of butorphanol as a constant-rate infusion appears to have minimal to no effect on global gastrointestinal46,47 or duodenal motility.48 N-butylscopolammonium bromide appears to cause a profound but very short-lived negative e ect on duodenal motility, but this e ect was not signi cant between groups.48 Atropine is a postganglionic blocking agent that binds to muscarinic receptors. When administered at 0.04 mg/kg, atropine inhibits individual small intestinal, cecal, and colonic contractions for about 120 minutes but supresses small intestinal and colonic migrating complexes for up to 8 hours.49

Neural Reflexes Neural re exes also may mediate inhibition of motility associated with peritoneal in ammation and abdominal pain.50,51 The a erent segment is composed partly of capsaicin-sensitive visceral a erent C bers that terminate in the dorsal horn of the spinal cord, where they can activate inhibitory sympathetic bers or synapse directly on the sympathetic ganglia. Consequently, the e erent limb of the re ex expresses increased sympathetic out ow, primarily mediated through stimulation of α2-adrenoreceptors, and inhibition of acetylcholine release, which provides the rationale for α2-blockade in treating ileus. Intraluminal infusion of capsaicin before abdominal surgery ameliorated the severity of POI in experimental rats. This nding highlights the importance of visceral a erent bers in the development of POI.52 Decreasing postoperative in ammation and pain decreases the risk of gastrointestinal ileus in many species in both experimental and clinical studies.

Distention Ileus also can occur in association with intestinal obstruction or displacement. Mild to moderate distention of the bowel, such as that occurring in the early stages of an intraluminal obstruction, evokes an increase in local contractile activity.53,54 Excessive distention results in inhibition of motility within the distended segment of bowel. Intestinal stasis is not always detrimental and under certain conditions may be protective. Repeated distention for determination of nociceptive threshold has also resulted in an overall decrease in duodenal motility over time, irrespective of other interventions.43,48

Endotoxin Endotoxemia is a clinical feature of many diseases of the equine gastrointestinal tract, and endotoxins independently can exert a negative e ect on intestinal motility and transit.55 A variety of mediators likely are involved, but activation of α2-adrenoreceptors and production of prostanoids appear to be important because pretreatment with yohimbine or NSAIDs (phenylbutazone or unixin), respectively, ameliorates the inhibitory e ects of experimental endotoxin infusion.56-59 Pretreatment with metocolopromide or cisapride had a similar e ect.60,61

Endotoxin infusion induced an in ammatory response in the intestine of rats that mimicked the response induced by handling during laparotomy.62 The similarity of the responses was highlighted in a recent study demonstrating that prior exposure of the muscularis to endotoxin protected the intestine from the e ects of manipulation.63 In rats colonic manipulation alone causes transference of intraluminal LPS to the muscularis, which likely contributes to the global gastrointestinal in ammatory response and decrease in contractility associated with tissue manipulation.30 In response to endotoxin alone, the in ammatory response within the jejunal muscularis is predominantly monocytic, whereas the response to polymicrobial sepsis is predominantly neutrophilic.64

Other Effects The pathophysiology of cecal emptying defect is not known. This syndrome may best mimic POI in human beings, which is generally considered a large intestinal disorder. An important di erence in horses is that laparotomy is a rare predisposing factor, and most cases occur in horses undergoing routine extra-abdominal surgical procedures. Thus whether it should truly be classi ed as a form of POI is subject to debate. General anesthesia itself is a potent inhibitor of gastrointestinal motility in horses, but these e ects are short-lived and reversible within hours of anesthetic withdrawal.23 The return of normal motility in horses after experimental ileus was most delayed in the cecum, suggesting that this may be a common site of ileus in horses.65 A link between routine postoperative medications, such as phenylbutazone and aminoglycoside antibiotics, has been suspected but not established. An inhibitory e ect of NSAIDs on large colon contractility has been demonstrated using in vitro techniques.66 Primary sympathetic overstimulation could be involved because many of the a ected animals are young male horses or animals with painful diseases. The duration of surgery in uences the development of small intestinal POI but not cecal emptying dysfunction.9,67 Technique may have a weak influence on small intestinal POI after jejunojejunostomy. The duration of intestinal ileus was shorter in animals that received a side-toside stapled anastomosis than those that had a hand sewn end-to-end procedure.3 The duration of ileus after stapled end-to-end anastomosis was not di erent from that after either procedure. Jejunocecostomy more commonly results in POI than other types of small intestinal resection and anastomosis, whether related to diseases necessitating this procedure or the procedure itself.68 Other reported risk factors for the development of POI include age (>10 years), small intestinal resection and anastomosis, breed (Arabians had a greater risk than other breeds), and duration of surgery.67 A prospective study found small intestinal lesion, high packed cell volume and duration of anesthesia to increase the risk of POI, whereas performance of a pelvic exure enterotomy and intraoperative administration of lidocaine may have a modest protective effect against POI.69

DIAGNOSIS The diagnosis of ileus is based on history and physical examination ndings. Case inclusion criteria for clinical studies of POI have varied.67-70 Important tests include determination of heart rate and rhythm, auscultation and percussion of the abdomen, rectal palpation, and passage of a nasogastric tube. A complete blood count with brinogen estimation and cytologic analysis of PF may improve the accuracy of diagnosis. A ected animals have varying signs of abdominal discomfort from mild to severe because of accumulation of uid in the small or large intestine. Decompression of the stomach is important diagnostically and therapeutically in horses with POI after intestinal surgery. Failure to relieve pain with gastric decompression could point toward mechanical obstruction, severe in ammation of the intestine, or peritonitis. Most animals with ileus are depressed and have reduced fecal output and intestinal borborygmi. Intestinal sounds must be interpreted with caution, however, because the presence of borborygmi does not always equate to progressive intestinal motility and may merely reflect local, nonpropagated contractions. Food intake, or the lack thereof, can also significantly alter gastrointestinal sounds.71 Rectal palpation ndings in cases of persistent POI or duodenitis-proximal jejunitis are usually nonspeci c but may include dilated, uidlled loops of small intestine. The clinician occasionally can palpate roughened peritoneal surfaces if peritonitis is present. Cecal distention with digesta can be palpated in horses with advanced cecal dysfunction. Distinguishing functional ileus from mechanical obstruction is important and can be di cult, but horses with mechanical obstruction typically have sustained high volumes of gastric re ux that vary little over time and abdominal pain that is typically not relieved by gastric decompression. Abdominal ultrasound in horses with ileus typically reveals mild to moderately uid- lled hypomotile to amotile small intestine, without alteration in the amount or character of PF or small intestinal wall thickness. Horses with re ux from other causes (peritonitis, mechanical obstruction) will have changes re ective of their disease process. This di erentiation is important for appropriate case management because horses with a mechanical obstruction often need either primary or repeat laparotomy, which should not be delayed.

TREATMENT The management of intestinal ileus depends on the segment of gastrointestinal tract involved. Therapy for ileus of the proximal gastrointestinal tract involves a combination of gastric decompression, uid and electrolyte therapy, analgesic therapy, and anti-in ammatory drug therapy. Electrolyte therapy is critical, particularly for maintaining adequate extracellular concentrations of potassium, calcium, and magnesium. Calculation of the volume of uid to be administered should include maintenance requirements plus an estimate of losses, especially those lost through gastric decompression. The clinician should consider parenteral provision of calories when feed has been

withheld for more than 96 hours, particularly after surgery. Hand walking also may provide some bene t to these animals but is not likely to have a direct effect on intestinal motility. One should either avoid the previously mentioned drugs, which can have an inhibitory e ect on motility, or use them sparingly. Fluid therapy is the key component in managing cecal emptying defect, usually in combination with lubricants or laxatives, such as mineral oil or magnesium sulfate, and with careful use of anti-in ammatory drugs. Horses with primary cecal impaction or impaction caused by an emptying defect may require surgery to prevent fatal rupture. The surgical management of these cases is controversial and may include typhlotomy alone, typhlotomy with a bypass procedure such as ileocolic or jejunocolic anastomosis, or a bypass without typhlotomy.72 Most horses that undergo simple typhlotomy have an uneventful recovery.73 In a large retrospective study 44 of 54 horses treated medically survived to discharge, and 37 of 49 horses treated surgically were allowed to recover, 35 of which survived to discharge.10 Ileocolostomy was performed in only two of the 37 horses treated surgically (one of which survived to discharge), with the remainder receiving typhlotomy without bypass. Survival to 1 year was not statistically di erent between horses treated medically (18 of 19) and those treated surgically (25 of 28), although six horses had a recurrence of cecal impaction.10 Experimental and anecdotal evidence provides a strong rationale for using anti-in ammatory drugs to prevent and treat gastrointestinal ileus, particularly in animals that may have endotoxemia.74 Flunixin meglumine is used widely in equine practice as an analgesic and antiin ammatory agent, and it ameliorates many of the adverse systemic e ects of endotoxin, particularly those on the cardiovascular system. A potential negative e ect of NSAIDs on large intestinal contractility has been suggested. A di erential e ect on contractility between selective and nonselective COX inhibitors is currently unknown. Broad-spectrum antimicrobials are indicated when sepsis is suspected or for the compromised immune system, as in cases of moderate to severe endotoxemia. Theoretical concerns have been raised regarding the use of aminoglycoside antibiotics in animals with ileus. High concentrations of aminoglycoside antimicrobials inhibited intestinal contractions in exposed sections of intestine in vitro, but this inhibitory effect is unlikely to occur at clinically relevant doses.75 Motility-enhancing drugs have been advocated to treat gastrointestinal ileus. Unfortunately, information directly pertinent to horses is limited and must be extrapolated cautiously from that of other species because of the di erences in intestinal anatomy and physiology. Prokinetic drugs potentially can shorten the length of hospitalization, thereby reducing the cost of treatment and the number of potential complications, such as weight loss, thrombophlebitis, and laminitis. Experimental evidence indicates that prokinetic drugs can minimize the development of postoperative abdominal adhesions.76 Most prokinetic drugs require a healthy gut wall to enhance intestinal contraction, and downregulation of motilin receptors has been demonstrated in the in amed equine jejunum.77 Therefore one should not assume that many of these drugs would be e ective in the presence of an in ammatory injury such as that which can occur after intestinal manipulation at surgery or that associated with duodenitis-proximal jejunitis.

Cholinomimetics Bethanechol is a parasympathomimetic agent that acts at the level of the myenteric plexus and directly on intestinal smooth cells through muscarinic receptors. In the horse this e ect is mediated predominantly by the M3 receptor, but the M2 receptor may also play a role.78 Bethanechol is a synthetic ester of acetylcholine and is not degraded by anticholinesterase. Bethanechol has cholinergic side e ects, including abdominal discomfort, sweating, and salivation, although these are minimal when the drug is administered subcutaneously or orally at 0.025 mg/kg body mass. Bethanechol has e cacy in diseases that involve abnormal gastric emptying and delayed small intestinal transit and has been shown to increase gastric contractility and hasten the emptying of liquid and solid phase markers from the stomach of normal horses.79 Bethanechol also increases the strength and duration of wall contractions in the cecum and right ventral colon and consequently speeds up cecal emptying.34 Neostigmine increases receptor concentration of acetylcholine by inhibiting cholinesterase. The drug (0.022 to 0.025 mg/kg intravenously) promotes cecal and colonic contractile activity and hastens the emptying of radiolabeled markers from the cecum.34 Neostigmine has been used to manage small intestinal ileus, but it significantly delayed the emptying of 6-mm beads from the stomach of normal adult horses.80

Benzamides and Dopamine Antagonists Metoclopramide acts principally as a 5-hydroxytryptamine 4-receptor (5HT-4) agonist and 5HT-3 receptor antagonist. In contrast to newer generation benzamides, metoclopramide is also an antagonist at dopamine 1 (DA1) and 2 (DA2) receptors. Antagonism of prejunctional DA2 receptors facilitates acetylcholine release and smooth muscle contraction. Metoclopramide crosses the blood-brain barrier, where its antagonist properties on central DA2 receptors can result in extrapyramidal signs, including seizure. Metoclopramide increased contractility of muscle strips in vitro in the pyloric antrum, proximal duodenum, and midjejunum.81 Those in vitro data support previous work in which metoclopramide administration restored gastroduodenal coordination of motility in a model of POI.82 In another study metoclopramide had no e ect on jejunal or pelvic exure myoelectrical activity.44 Constant intravenous infusion (0.04 mg/kg/hr) of metoclopramide was well tolerated in a population of postoperative horses and signi cantly decreased the volume and duration of gastric re ux over control and intermittent drug infusion groups.83

Cisapride is a second-generation benzamide that acts as a 5HT-4 agonist and 5HT-3 receptor antagonist but is without antidopaminergic action. Stimulation of 5HT-4 receptors within the enteric nervous system enhances release of acetylcholine from the myenteric plexus. Several reports suggest the e cacy of cisapride in managing intestinal disease in horses, including the resolution of persistent large colon impaction, treatment of equine grass sickness, and as a preventative for POI in horses after small intestinal surgery (0.1 mg/kg body mass intramuscularly during the postoperative period).84-87 The horse erratically absorbs tablets administered rectally, but a method for preparing a parenteral form of the drug from tablets has been described.88 Cisapride has the potential to cause adverse cardiac side e ects mediated through blockage of the rapid component of the delayed recti er potassium current that include lengthening of the QT interval and development of torsades de pointes, a potentially fatal arrhythmia.89 These adverse e ects have resulted in withdrawal of the drug in the United States but have not been reported in the horse. Tegaserod, a 5HT-4 agonist, has been shown to increase pelvic exure smooth muscle contractility (0.27 mg/kg PO)90 and hasten gastrointestinal transit time (0.02 mg/kg intravenously) in healthy horses.91 It has not, to the author’s knowledge, yet been objectively evaluated in abnormal horses, but may prove useful. In humans this drug was marketed for women with constipation-predominant or mixed symptom irritable bowel syndrome and demonstrated clear benefits in quality of life and gastrointestinal syptoms, but it is currently available only in a restricted fashion because of its association with ischemic colitis and cardiovascular disease.92 Domperidone acts as a competitive antagonist at peripheral DA2 receptors. The drug is a therapeutic agent (1.1 mg/kg/day) for mares grazing endophyte-infected tall fescue, principally because of drug-enhanced prolactin release. The potential prokinetic e ects of domperidone have not been studied extensively in horses, but a modest e cacy of domperidone (0.2 mg/kg intravenously) has been demonstrated in experimental ileus in ponies.84

Antimicrobials Erythromycin, a macrolide antibiotic, is a direct motilin receptor agonist on smooth muscle cells and also may act within the enteric nervous system to facilitate the release of acetylcholine and motilin. Erythromycin displaces motilin from its receptor in the equine duodenum, jejunum, cecum, and pelvic exure.93 Erythromycin enhances gastric emptying in normal horses but has a more pronounced e ect on the hindgut.79,94 Erythromycin lactobionate (1.0 mg/kg IV) hastens cecal emptying in normal animals and induces colonic MMC-like activity across the colon. Administration often is associated with defecation and abdominal discomfort. The prokinetic e ect of erythromycin apparent in the ileum, cecum, and pelvic exure documented in normal horses was reduced in the immediate postoperative period.95 Similarly, luminal distention and decompression resulted in in ammation and a decreased response to erythromycin.96 A decrease in motilin receptors in response to luminal distention has been documented in the equine jejunum,77 and this may explain the di erence in response between normal and clinically a ected horses. Repeated dosing can cause downregulation of motilin receptors in other species.97 Because erythromycin can induce diarrhea in adults, it should not be administered over many days. Potassium penicillin (20 million IU intravenously to adult horses) can stimulate defecation and increase myoelectrical activity in the cecum and pelvic flexure, and these effects are not produced by an equimolar amount of potassium ion given intravenously as potassium chloride.98

Opioid and α2-Adrenoreceptor Antagonists Naloxone (0.05 mg/kg intravenously) induces contractile activity in the cecum and left colon.42 Defecation commonly follows administration of naloxone within 15 to 20 minutes. N-methylnaltrexone increases jejunal and pelvic exure contractility in vitro99 and has been shown to prevent the negative effects of morphine on fecal output and intestinal transit time when administered concurrently.100 α2-Adrenoreceptor antagonists such as yohimbine or tolazoline counteract increased sympathetic out ow in response to nociceptive stimulation. Yohimbine infusion (75 μg/kg) also may attenuate the negative effects of endotoxin on motility.56,58

Local Anesthetics The use of intravenous lidocaine as a prokinetic has gained tremendous popularity and was reported to be the agent most commonly used by equine surgeons for POI following most gastrointestinal lesions.101 Lidocaine may exert prokinetic e ects by suppressing primary a erent neurons, thereby limiting re ex e erent inhibition of motility.102 Other proposed mechanisms of action include anti-in ammatory properties, potentially through NF-кβ signaling103 or improving mucosal repair.104 Intravenous lidocaine can also exert analgesic e ects, although it was shown to alter somatic but not visceral antinociception in clinically normal horses in one study.105 Lidocaine increased contractile activity in isolated strips of proximal duodenum in vitro.81 The most commonly cited dosage is a 1.3 mg/kg bolus, typically over 15 minutes, followed by 0.05 mg/kg/min constant-rate infusion. This dosage did not alter MMC duration or spiking activity, nor did it reset the MMC in the jejunum in clinically normal horses106 or signi cantly alter a variety of indicators of POI after colic surgery.107 In another study signi cantly more horses with POI stopped re uxing within 30 hours following the institution of lidocaine infusion, relative to saline infusion.108 Lidocaine infusion can be associated with reversible side e ects that include muscle fasciculations, ataxia, and seizure. Consequently, the rate of infusion requires close monitoring. Prolonged infusion of lidocaine in the horse appears safe, although accumulation of the GX metabolite has been documented.109

ENDOTOXEMIA Katharina L. Lohmann, Michelle Henry Barton Endotoxemia is literally de ned as the presence of endotoxin in the bloodstream. Most often, however, the term is used to refer to the associated clinical manifestations caused by an excessive and unbalanced in ammatory reaction. Endotoxemia is not a disease in its own right but rather a complication of many septic and nonseptic disease processes a ecting horses and other animals. Its diagnosis and management must therefore be discussed in the context of underlying primary pathologic conditions. In its pathophysiologic consequences the innate immune response to endotoxin (LPS) is similar to the response to other stimuli (e.g., bacterial infection, viral infection, severe trauma). Moreover, the clinical presentation and clinical pathologic abnormalities of patients su ering from severe systemic in ammation are consistent regardless of the etiologic cause, and outcome may be predicted more accurately by the severity of the in ammatory response than the nature of the inciting insult.1 In 1992, therefore, the term systemic in ammatory response syndrome (SIRS) was introduced in a consensus statement of chest physicians and critical care physicians to account for these similarities and provide a case de nition for clinical and research purposes. Fairly soon thereafter, however, the 1992 SIRS de nition was criticized as being too sensitive while providing little speci city,2 and its usefulness for clinical practice was challenged because it failed to predict outcome and to discriminate patients at a high risk of morbidity and mortality.3 The de nition of SIRS may therefore be valuable from a conceptual standpoint, but it has little value in daily clinical practice. Sepsis was de ned as the systemic in ammatory response to infection, and septic shock as “sepsis-induced hypotension, persisting despite adequate uid resuscitation, along with the presence of hypoperfusion abnormalities or organ dysfunction.”1 According to these de nitions the diagnosis of sepsis requires documentation of infection by culture in addition to two or more of the following ndings: hypothermia or hyperthermia; tachycardia; tachypnea or hypocapnia; and leukocytosis, leukopenia, or an increased proportion of immature leukocyte forms. Organ failure is a common sequela of endotoxic or septic shock, and the term multiple organ dysfunction syndrome describes insu ciency of two or more organ systems, as evident by clinical or clinicopathologic changes. In horses the laminae of the feet should be included in the list of organs susceptible to failure.

ENDOTOXIN German scientist Richard Pfei er (1858–1945), in working with Vibrio cholerae, rst described endotoxin as a toxin “closely attached to, and probably integral of, the bacterial body.”4 He observed this toxin to be distinct from the actively secreted, heat-labile, and proteinaceous bacterial exotoxins. Endotoxin later was found to be a heat-stable LPS structure, and the terms endotoxin and LPS now are often used interchangeably. LPS is a major structural cell wall component of all gram negative bacteria, including noninfectious species (Figure 15-7). With 3 to 4 × 106 molecules per cell, LPS makes up about 75% of the outer layer of the outer cell membrane and is a key functional molecule for the bacterial outer membrane, serving as a permeability barrier against external noxious agents. The LPS molecule consists of four domains, which are essential for the virulence of gram negative bacteria.5 Three of the domains (inner core, outer core and O-speci c chain) represent the hydrophilic polysaccharide portion of the molecule, whereas the lipid A portion represents the hydrophobic lipid portion (Figure 15-8). Combined, these domains confer the overall amphiphilic properties of the molecule that lead to the formation of micellar aggregates in aqueous solutions.

FIGURE 15-7

Lipopolysaccharide.

FIGURE 15-8

Three domains of the lipopolysaccharide molecule.

O-speci c chains (also called O-antigen polysaccharides or O-chains) are characteristic of any given type of LPS and show enormous structural variability among bacterial serotypes.6 O-chains are synthesized by addition of preformed oligosaccharide blocks to a growing polymer chain and therefore have a repetitive structure. O-speci c chains determine part of the immunospeci city of bacterial cells5 and, on interaction with the host immune system, serve as antigens for the production of species-speci c antibodies.7 O-speci c chains are further responsible for the smooth appearance of gram negative bacterial colonies on culture plates,5 and LPS molecules containing an O-chain are termed smooth lipopolysaccharide. The inner (lipid A-proximal) and outer (O-chain-proximal) core oligosaccharide portion is more conserved among di erent strains of gram negative bacteria than the O-speci c chain.6 The core of all LPS molecules contains the unusual sugar KDO (3-deoxy-D-manno-oct-2ulopyranosonic acid), which links the core region to the lipid A molecule. Synthesis of a minimal core is essential for the survival of bacteria,5 and the smallest naturally occurring LPS structure consists of lipid A and KDO.8 In contrast to the S-form colonies, colonies of gram negative bacteria with LPS molecules that lack the O-speci c chain but contain a core region show a rough appearance on culture plates. Rough LPS molecules are denoted further as Ra, Rb, and so on to indicate the length of the core region. In Re-LPS (also called deep rough LPS), the core region is reduced to a KDO residue. Remutants often are used to raise antibodies against the core region in an attempt to provide cross-protection against a variety of bacterial species. The lipid A portion, which serves to anchor the LPS molecule in the bacterial outer membrane, has been identi ed as the toxic principle of LPS,9 and its structure is highly conserved among gram negative bacteria. The common structure shared by lipid A molecules is a 1,4’-bisphosphorylated β1,6-linked D-glucosamine disaccharide backbone (lipid A backbone), which is acylated by up to six fatty acids.6 Figure 15-9 shows the acylation pattern for Escherichia coli LPS. Variation in the lipid A structure between gram negative bacteria a ects the number, length, and position of fatty acids and the backbone structure and the substitution of phosphate by other polar groups.7

FIGURE 15-9

CAUSES OF ENDOTOXEMIA IN HORSES According to its nature as a structural cell wall component, the presence of endotoxin implies the presence of gram negative bacteria as a source. Depending on the nature of the underlying disease, these bacteria may circulate in the bloodstream in their intact form (i.e., bacteremia), may be con ned to a localized infectious process, or may be part of the endogenous bacterial orae colonizing the gastrointestinal tract. In any of these scenarios endotoxin molecules are released as a by-product of bacterial growth and in large numbers on bacterial cell death.10 Common infectious conditions associated with endotoxemia in horses include neonatal gram negative sepsis, bacterial pneumonia and pleuropneumonia, endometritis, peritonitis, and infectious colitis with bacteria such as Salmonella spp. that are not part of the normal intestinal florae. In one study, for example, endotoxin was detectable in plasma of 50% of foals evaluated for presumed sepsis.11 The term translocation describes entry of endogenous bacteria and bacterial products from the gastrointestinal tract into tissues and the systemic circulation.12 The natural intestinal ora of horses consists mainly of gram negative, anaerobic bacteria, and thus large amounts of endotoxin normally exist in the lumen of the equine intestinal tract.13 Even in healthy horse small amounts of endotoxin probably cross the intact mucosal barrier and reach the portal circulation and the liver.14 These molecules are cleared, however, by the mononuclear phagocytic system in the liver and lead only to a localized and restricted activation of the host immune system. For endotoxin translocation to become detrimental, excessive amounts have to cross the intestinal barrier and overwhelm the mononuclear phagocytic system, or the capacity of the liver to detoxify lipopolysaccharide must be compromised. The latter may be a concern in conditions such as hepatitis, cholangiohepatitis, or portosystemic shunting of blood. Permeability of the intestinal mucosal barrier frequently increases in cases of acute gastrointestinal disease. Colic patients are prime candidates to develop endotoxemia, and plasma endotoxin was detectable in 10% to 40% of colic patients on admission.15,16 The studies differed in their inclusion criteria, and the higher percentage of horses testing positive for plasma endotoxin was observed when only patients presenting for surgical intervention were evaluated.16 Aside from gastrointestinal rupture, increased permeability to intact bacteria or free endotoxin molecules is thought to be associated most commonly with ischemic insults such as strangulating obstruction and bowel infarction; severe in ammation, as in proximal enteritis and colitis; bacterial overgrowth; and intraluminal acidosis, which occurs with grain overload.17,18 One study, however, found no di erence in plasma endotoxin detection among disease groups, therefore emphasizing the fact that any disease of the abdominal cavity can induce endotoxemia in horses. In the same study endotoxin was approximately three times more likely to be detected in PF as opposed to plasma samples. Similarly, higher cytokine concentrations have been measured in PF than in plasma. The likely explanation for these ndings is a local in ammatory response in the peritoneal cavity elicited by translocated bacteria or LPS molecules before their absorption into the systemic circulation.15 Although gastrointestinal disease is likely the most important cause of endotoxin translocation in horses, other conditions may also result in translocation of endotoxin and bacteria. In experimental studies using laboratory animals, entry of gut-associated bacteria into the lymphatic system was demonstrated after hypovolemic shock, burn injuries, trauma, malnutrition, and starvation.19-21 Furthermore, endotoxin itself caused bacterial translocation into mesenteric lymph nodes after intraperitoneal administration to mice.22 These ndings have received much attention in the literature concerning human patients because they serve to explain cases of endotoxic shock in the absence of demonstrable bacterial infection. The veterinarian should keep in mind the possibility of translocation when evaluating cases of presumed SIRS in horses, in which bacterial infection or gastrointestinal disease cannot be demonstrated. Endotoxin translocation also may be associated with strenuous exercise, which results in reduced splanchnic blood ow, hypoxemia, and a higher body temperature. In t racehorses a signi cantly increased mean plasma LPS concentration was found after racing, whereas antilipopolysaccharide immunoglobulin G levels were decreased. Fit horses showed signi cantly higher antilipopolysaccharide immunoglobulin G concentrations at rest than sedentary controls, suggesting leakage of small amounts of endotoxin from the intestinal lumen during training and racing.23 A study in endurance horses demonstrated detectable LPS concentration in approximately 50% of horses, and correlation between plasma endotoxin and lactate concentration suggested that plasma endotoxin concentration may be re ective of the vigor of exercise.24 Surprisingly, the latter study failed to show a correlation between plasma endotoxin concentration and levels of in ammatory mediators such as TNF-α and IL-6,24 such that the clinical signi cance of endotoxin translocation during exercise requires further investigation. At least experimentally, however, alterations in innate immune function during and after strenuous exercise in horses have been demonstrated.25

MECHANISMS OF CELLULAR ACTIVATION BY LIPOPOLYSACCHARIDE The initiating event in the pathophysiology of endotoxemia is the activation of LPS-responsive cells by endotoxin, resulting in altered cellular functions and increased expression of in ammatory mediators. Immune cells such as macrophages respond to minute amounts of LPS, which usually allows them to eliminate gram negative bacteria and free LPS molecules e ciently. An important factor in the exquisite sensitivity to LPS is the presence of LPS-binding protein (LBP).6 LBP is an approximately 60-kd plasma glycoprotein26 that is synthesized primarily by

hepatocytes27 and belongs to the family of lipid transfer/LBPs. LBP is an acute phase protein, and in ammatory agents and cytokines such as IL-1 increase LBP plasma concentration tenfold to 100-fold within 24 to 48 hours of an inflammatory stimulus.28,29 The main function of LBP is to transfer LPS to endotoxin-responsive cells, which include mononuclear phagocytes, neutrophils, lymphocytes, and endothelial cells. The importance of a highly sensitive response to LPS for protection against gram negative bacterial infection is demonstrated in experiments using LBP knockout mice— that is, mice that lack the LBP gene and are therefore unable to synthesize LBP. Although these animals are resistant to the e ects of puri ed LPS, they are unable to control infection with viable bacteria and rapidly succumb.30 Despite its crucial importance for an e ective host defense, LBP is not essential for LPS-receptor interaction per se, because high concentrations of LPS can activate cells in the absence of LBP.31 Aside from its role as a catalyst of cellular activation by LPS, LBP has opsonizing activity32 and participates in the phagocytosis of LPS by macrophages and neutrophils.33,34 Although phagocytosis of LPS is receptor dependent, it appears to be uncoupled from intracellular signaling events and occurs in the absence of cell activation.5 LBP further catalyzes transfer of LPS to lipoproteins such as high-density lipoprotein, which neutralizes LPS activity.35 This detoxifying e ect may become important when large amounts of LPS are present. A protective e ect of LBP against LPS challenge and infection has been demonstrated in a murine model of septic shock,36 and further studies investigating potential therapeutic use of LBP are under way. More recently, LBP has also been investigated as a diagnostic and prognostic indicator in human patients, where it may di erentiate between SIRS and sepsis and predict outcome of septic patients.29 One study investigating serum amyloid A and LBP in horses with colic37 found no correlation between serum concentration of LBP with outcome, type of disease process, or affected portion of the gastrointestinal tract. The most important LPS receptors known to date are cluster di erentiation antigen 14 (CD14) 31 and Toll-like receptor 4 (TLR-4).38 Both are classified as pattern recognition receptors,39 which means that they recognize LPS as a structural motif common to all gram negative bacteria. CD14 is a 53-kd protein that in its membrane-bound form (mCD14) is inserted into the cell membrane via a glycosyl-phosphatidylinositol anchor.40 CD14 is expressed primarily on monocytes and tissue macrophages and to a lesser extent on neutrophils.41 CD14 also is found in a soluble form (sCD14)42 that can bind to cell types lacking CD14, such as endothelial cells, and make them LPS-responsive. In addition to this pro-in ammatory e ect, high concentrations of sCD14 can sequester and neutralize LPS.43 The amount of circulating sCD14 greatly increases during inflammation, which makes it a useful marker of acute and chronic inflammation.41 Although CD14 is known to be crucial for cellular activation, it cannot transmit signals to the inside of the cell because it lacks a transmembrane domain. The missing link between CD14 and the cytosolic environment is a Toll-like receptor in association with the molecule MD-2.44 The name Toll-like receptor stems from the hom*ology of the mammalian receptor with a receptor type in Drosophila (Toll) that is important for dorsoventral orientation and immune responses in the y. A number of Toll-like receptors have been identi ed in mammalian species so far, but TLR4 appears to be the receptor subtype most important for LPS signaling.38 The importance of CD14 and TLR4 in the cellular response to LPS has been demonstrated in a number of experiments. Mice deficient in CD14 are incapable of mounting a normal inflammatory response to LPS,43 whereas mutation or deletion of the gene encoding for TLR4 causes LPS hyporesponsiveness.45-47 After binding of LPS to cellular receptors, TLR4 undergoes oligomerization and recruits downstream adaptor proteins in order to activate intracellular signaling pathways.48 Signaling pathways are characterized by sequential phosphorylation and thereby activation of enzymatic activities, and ultimately result in the alterations of cellular metabolism known as cell activation. A typical result of intracellular signaling is the activation of transcription factors—that is, proteins that bind to DNA and promote gene transcription. Translational mechanisms are activated in a similar manner. TLR4-dependent cell signaling pathways can be di erentiated according to the use of adaptor proteins that bind to the intracellular domain of TLR4.48 The MyD88 (myeloid di erentiation primary response gene 88)-dependent pathway results in activation of IKK (Iкβ kinase) and MAPK (mitogen-activated protein kinase) pathways and ultimately in expression of pro-in ammatory cytokine genes. IKK activates the well-described transcription factor NF-кβ by phosphorylating an inhibitor protein complex (IкB) that sequesters and inactivates NF-кβ in the cytoplasm. On phosphorylation IкB is ubiquitinated and degraded, and NF-кβ is translocated to the nucleus, where it unfolds its activity.49 The MyD88-independent pathway, on the other hand, is activated by interaction of TLR4 with the adaptor protein TRIF (Tollinterleukin-1 receptor domain-containing adaptor inducing interferon-β).While the MyD88-independent pathway also activates MAPK and NF-кβ (in addition to a transcription factor called IRF3), this pathway primarily results in activation of type I interferons, which are important for antiviral and antibacterial responses.48 Despite the characterization of seemingly separate and “ordered” pathways, one should recognize that interaction and synergy between pathways is likely to occur. Similarly, inhibitory pathways are required for regulation of the cell response and can target multiple levels of TLR4 signaling.48 The potential manipulation of signaling pathways for therapeutic use in septic patients is under investigation.

INFLAMMATORY MEDIATORS

Although endotoxin can exert some direct e ects, cytokines are a primary mediator of LPS e ects. Cytokines are glycoprotein molecules that regulate in ammatory and immune responses by acting as a signal between cells.50 Cytokines of major interest in the pathogenesis of endotoxemia include TNF-α, the interleukins, chemokines, and growth factors such as granulocyte-monocyte colony-stimulating factor. TNF-α is thought of as the most “proximal” cytokine released in response to LPS. Studies corroborate this by showing that administration of recombinant TNF-α mimics the e ects of LPS51 and that antibodies directed against TNF-α protect against the lethal e ects of endotoxin.52 Increased plasma activity of TNF is associated with increased mortality in equine patients with acute gastrointestinal disease and in septic neonates.15 Despite being a structurally diverse group of proteins, cytokines share several characteristics that allow them to execute their complex functions in the in ammatory response.50 Any individual cytokine generally is produced by several di erent cell types, can act on di erent cell types, and has multiple e ects on any given cell. Furthermore, cytokine e ects are redundant, meaning that di erent cytokines can share the same e ect. In endotoxemia this is particularly true for the e ects of IL-1 and TNF-α.53 Many of the biologic activities of cytokines in vivo result from synergistic or antagonistic actions involving two or more cytokines.54 Within itself the cytokine response is highly regulated: Cytokines induce or suppress synthesis of other cytokines, including their own (feedback regulation); regulate expression of cytokine receptors; and regulate cytokine activities. Additional regulatory mechanisms include the release of speci c cytokine inhibitors such as soluble IL-1 and TNF-α receptors; cytokine receptor antagonists such as IL-1 receptor antagonist; and anti-in ammatory cytokines, including IL-10, IL-4, IL-13, and TGF-β. Glucocorticoids, which are produced increasingly in response to endotoxin, also inhibit the production of cytokines.55 During a “controlled” in ammatory response, therefore, cytokine secretion is a self-limited event, whereas excessive stimulation of cytokine production can lead to the perpetuation of the in ammatory response even after the initial stimulus has been removed. Aberrations from the controlled, self-regulated in ammatory response have been described as predominantly proin ammatory (SIRS), anti-in ammatory or hypoin ammatory (compensatory anti-in ammatory response syndrome or CARS), or combined (mixed antagonist response syndrome) responses.56 That these responses may not represent a continuum in response to infection was demonstrated in a recent study investigating cytokine pro les in an experimental model of sepsis.57 Here, increased plasma concentration of anti-in ammatory mediators such as IL-1 receptor antagonist and IL-10 in the early phase of sepsis predicted early mortality almost as accurately as the typical pro-in ammatory cytokines such as IL-6. Cytokine pro les failed to predict mortality in the later stages of sepsis, thereby suggesting that late outcome is not “preprogrammed” early on in the disease. A progression from pro-in ammatory (SIRS) to antiin ammatory (CARS), as previously proposed, could not be demonstrated.57 The authors concluded that use of biomarkers may be helpful in determining the in ammatory status of clinical patients su ering from sepsis and that the success of treatments aimed at suppressing or stimulating immune responses may depend on the individual patient’s inflammatory profile. Interestingly, tolerance to endotoxin develops after repeated exposure to LPS.19 Tolerance can be demonstrated in vitro and in vivo and encompasses decreased production of cytokines and a diminished clinical response.19,58 Tolerance may be a protective mechanism and was shown to reduce mortality in some experimental models; however, impaired resistance to infectious processes was demonstrated by other investigators.59 Mechanisms that likely are responsible for the development of endotoxin tolerance include receptor downregulation,60 downregulation of intracellular signaling pathways,59,61 and mediators such as glucocorticoids and IL-10.59 The development of endotoxin tolerance in horses has been reported.62,63 More recently, the term “cellular reprogramming” has been introduced to describe altered in ammatory cell functions in septic and SIRS patients. This term better accounts for the observation that cellular LPS-sensing ability appears to be maintained while intracellular signaling and thereby cytokine production is modi ed towards an anti- rather than pro-in ammatory response.59 Aside from cytokines, a number of other molecules function as in ammatory mediators in the pathogenesis of endotoxemia, the synthesis and release of which are stimulated by endotoxin and by cytokines. These mediators include the arachidonic acid metabolites or prostanoids, PAF, oxygen-derived free radicals, nitric oxide, histamine, kinins, and complement components. Table 15-4 summarizes the origins, targets, and e ects of the most important in ammatory mediators involved in the pathogenesis of endotoxemia. Figure 15-10 shows the pathways of arachidonic acid metabolism by COX and lipoxygenase. COX products are the prostaglandins (PGs), prostacyclin (PGI2) and thromboxanes, and the lipoxygenase produces the leukotrienes. TABLE 15-4 Important Mediators of the Systemic Inflammatory Response to Endotoxin Mediator Tumor necrosis factor

Origin Macrophages Monocytes Neutrophils CD4+ T cells Natural killer cells

Effects Induces synthesis of TNF, IL-1, IL-6, and GM-CSF Activates neutrophils Activates fibrinolysis and coagulation Activates contact and complement system Induces a catabolic state

Interleukin-1

Activated macrophages Endothelial cells Fibroblasts Dendritic cells Lymphocytes Keratinocytes

Interleukin-6

Thromboxane A2

Activated macrophages Fibroblasts Keratinocytes T lymphocytes Macrophages Endothelial cells Platelets

Prostaglandin E2

Most nucleated cells

Prostaglandin I2

Vascular endothelial cells

Platelet-activating factor

Macrophages Platelets Neutrophils Mast cells Eosinophils

Prostaglandin F 2α

Most nucleated cells

Leukotriene B4

Most nucleated cells

Leukotrienes C 4, D4, E4

Most nucleated cells

Kinins

Produced from serum precursors

Complement components (C3a, C5a)

—

Oxygen-derived free radicals

Macrophages Neutrophils

Granulocyte- monocyte colony- stimulating factor

—

Interleukin-8

Induces insulin resistance Is a pyrogen (direct action and via IL-1 induction) Induces synthesis of TNF, IL-1, IL-6, PGI2 , PAF, and GM-CSF Activates pyrogen Induces malaise Activates neutrophils and chemotaxis Activates fibrinolysis and coagulation Activates contact and complement system Induces acute phase response Increases activity of lipoprotein lipase Mobilizes amino acids Induces muscle proteolysis Induces acute phase response Induces stress response Is a weak pyrogen Activates neutrophils and chemotaxis Induces vasoconstriction Activates platelet aggregation Induces vasodilation Activates platelet aggregation Induces fever Induces vasodilation Inhibits platelet aggregation Activates platelet aggregation Activates macrophages and neutrophils Induces hypotension Increased vascular permeability Aids recruitment of leukocytes Induces visceral smooth muscle contraction Is a negative inotrope and arrhythmogenic Induces ileus Induces vascoconstriction Activates luteolysis Is a chemoattractant Promotes neutrophil interaction with endothelial cells Increase vascular permeability Induce bronchoconstriction Induce vasoconstriction Increase vascular permeability Induce smooth muscle contraction cause pain Activate neutrophils and chemotaxis Induce smooth muscle constriction Induce mast cell degranulation Induce release of histamine and serotonin Increase vascular permeability Damage cell membranes Inactivate enzymes. Damage tissues Induces rebound neutrophilia

TNF, Tumor necrosis factor; IL, interleukin; GM-CSF, granulocyte-monocyte colony-stimulating factor; PG, prostaglandin; PAF, platelet-activating factor.

FIGURE 15-10

Pathways of arachidonic acid metabolism by cyclooxygenase and lipoxygenase.

PATHOGENESIS The innate immune response to LPS is an e cient defense mechanism that provides maintenance of homeostasis and therefore health in the face of an almost continuous exposure to microorganisms and their products.5 Detrimental consequences of this immune response occur only if excessive and uncontrolled mediator output results in endothelial damage, neutrophil-mediated tissue damage, and uncontrolled activation of the coagulation and brinolytic cascades and the complement system. Ultimately, the combination of these events culminates in cardiovascular instability, impaired hemostasis, organ failure, shock, and death. The following discussion addresses the various pathophysiologic events in the development of endotoxemia and shock and the role of inflammatory mediators.

Endothelial Dysfunction and Damage Normal endothelium plays an important role in regulating blood pressure and regional tissue perfusion and provides an anticoagulant surface. Endothelial dysfunction and damage result in a decreased responsiveness to vasoactive agents (vasoplegia), increased vascular permeability, and a tendency for clot formation in the microvasculature. If the basem*nt membrane and underlying matrix are compromised, microvascular hemorrhage can occur. Endothelial cell damage is primarily neutrophil-mediated. More speci cally, damage is caused by oxygen-derived free radicals, which are produced within endothelial cells through reactions involving neutrophil-derived elastase and hydrogen peroxide molecules, endothelial cell enzymes such as xanthine oxidase, and endothelial cytosolic iron. The hypochloric anion radical (HO˙) is thought to be responsible most directly for endothelial cell cytotoxicity. Nitric oxide, which is produced by constitutively expressed nitric oxide synthase in endothelial cells and scavenges superoxide radicals in a reaction to form peroxynitrite, may a ord protection from oxygen radical–induced endothelial cell damage. Variations in the ability to produce nitric oxide may explain why vascular beds in di erent organs vary in their susceptibility to neutrophil-mediated damage.64 Excessive production of nitric oxide by an inducible form of nitric oxide synthase (iNOS), however, contributes to tissue damage, and increased peroxynitrite concentrations may be responsible in part for PAF-induced increases in vascular permeability.65 In addition to oxygen-derived free radicals, activated neutrophils release matrix metalloproteinases, which contribute to tissue damage.55 Vascular endothelial cells are further susceptible to direct e ects of various cytokines, most prominently TNF-α and IL-1. These cytokines are thought to cause damage via the induction of COX activity and production of prostanoids and through generation of free radicals. Endothelial cell damage in response to endotoxin infusion and the association with leukocyte attachment have been demonstrated in horses.66

Neutrophil Activation, Margination, and Transmigration Neutrophil activation by LPS and cytokines results in stimulation of phagocytosis and the respiratory burst, release of lysosomal enzymes and in ammatory mediators, and expression of adhesion molecules. Perhaps the single most speci c clinicopathologic indicator of endotoxemia

is pronounced neutropenia,67 which temporally correlates with peak plasma concentrations of TNF.68 Neutropenia is caused primarily by margination of neutrophils in the vasculature, especially of the lungs,69 whereas signi cant loss through active migration into peripheral tissues likely is limited to the presence of a localized source of infection. In fact, neutrophils exposed to endotoxin or inflammatory mediators exhibit reduced capacity to respond to chemotactic stimuli and extravasate.69 Margination is made possible by adhesion molecules on endothelial cells and leukocytes that interact and allow sticking of leukocytes to the endothelial lining of blood vessels. The details of neutrophil margination and transmigration are reviewed in several excellent texts.50,70 Recent evidence demonstrates a role for TLR-4 in neutrophil margination in the lung microvasculature during endotoxemia, and suggests that activation and TLR-4 expression of endothelial cells may be more important than that of circulating neutrophils.69 While neutrophil activation during infections provides the body with an e ective defense system against microbial invaders, margination and reduced migratory ability of neutrophils during endotoxemia results in the accumulation of activated cells at the endothelial surface. These cells are then positioned to e ect endothelial injury, increase vascular permeability, and in some cases cause parenchymal cell death and organ dysfunction.71 In addition, inhibition of transmigration deprives the body of phagocytic cells at sites of infection, resulting in reduced ability to ght bacterial infection. This latter complication of endotoxemia may have signi cant clinical impact, because defective neutrophil recruitment has been demonstrated at low endotoxin doses that do not yet result in tissue damage.71,72 In human patients endotoxemia and impaired neutrophil migration have been associated with infectious complications of surgical procedures.71 Decreased phagocytic function and oxidative burst activity have also been suggested in sick and septic hospitalized foals,73,74 where a bene cial e ect of plasma transfusion on neutrophil function could be demonstrated. The mechanisms of migration inhibition during endotoxemia are incompletely understood but may include occupation of neutrophil chemotactic receptors by cytokines and complement components, resulting in an inability of activated cells to respond to a chemotactic gradient.71 Rebound neutrophilia, which is observed frequently following episodes of endotoxemia, is caused by neutrophil release from the bone marrow reserve pool and by stimulation of myeloid cell proliferation via granulocyte-macrophage colony-stimulating factor and is mediated primarily by TNF and IL-1.53

Coagulopathy and Disseminated Intravascular Coagulation In health coagulation and brinolysis underlie stringent control mechanisms that allow appropriate clot formation and their resolution. Coagulopathies frequently are observed in horses with colic17,75,76 and foals with sepsis11 and are likely attributable to endotoxemia. In human medicine virtually all septic patients are considered to have some degree of coagulopathy, which may range from subclinical abnormalities of the clotting pro le to fulminant disseminated intravascular coagulation (DIC).77,78 DIC results from a widespread activation of the coagulation and brinolytic systems and failure of their control mechanisms. Ultimately, this leads to disseminated brin deposition in the microvasculature, consumption of platelets and clotting factors, and accumulation of brin degradation products (FDPs). Depending on the underlying disease process and the relative impairment of the coagulation and brinolytic systems, DIC can manifest as a di use thrombotic syndrome leading to ischemic organ failure, a fibrinolytic syndrome with uncontrolled hemorrhage, or a combination of both.79 A procoagulant state characterized by clinicopathologic abnormalities of the clotting profile precedes DIC. The intrinsic and extrinsic arms of the coagulation cascade are activated in endotoxemia. The intrinsic pathway is initiated by activation of coagulation factor XII (HF), prekallikrein, and high-molecular-weight kininogen, which compose the contact system.19 Although direct activation of coagulation factor XII by endotoxin has been demonstrated,80 the extrinsic pathway likely is more important for the development of coagulopathy in endotoxemia and sepsis.19 Activation of the extrinsic pathway depends on the interaction of coagulation factor VII with tissue factor, which is the only coagulation factor not constitutively present in blood. Tissue factor is present in subendothelial tissues and is exposed on vascular injury but also is expressed on endothelial cells and mononuclear phagocytes in response to LPS.81,82 Increased expression of monocyte tissue factor (also described as increased procoagulant activity) was found to be associated signi cantly with coagulopathy and poor prognosis in horses with colic.83 Furthermore, LPS-induced tissue factor expression by equine peritoneal macrophages may be associated with the development of intra-abdominal adhesions.62 Regulatory mechanisms of the coagulation cascade include tissue factor pathway inhibitor, antithrombin III (AT III), and the protein C system.19 Protein C acts as an anticoagulant by inactivating clotting factors V and VIII and promotes brinolysis by inactivating plasminogen activator inhibitor (PAI).84 Protein C activation by thrombin-thrombomodulin complexes is important for the anticoagulative properties of normal endothelium,19 and downregulation of endothelial thrombomodulin expression by TNF and IL-1 along with decreased expression of AT III and tissue factor pathway inhibitor by damaged endothelial cells contribute to the procoagulant state in endotoxemia and sepsis.85-87 In addition, activation of vascular endothelial cells leads to a loss of prostacyclin and nitric oxide production and an increased release of thromboxane A2 (TXA2). As a result, platelets are stimulated to aggregate and release TXA2 and PAF, thereby further promoting clot formation.18 The crucial step in the brinolytic cascade is the conversion of plasminogen to plasmin, a brin-degrading enzyme.19 Tissue-type (tPA)

and urokinase-type (uPA) plasminogen activator are the major initiators of brinolysis, whereas PAI and α 2-antiplasmin are the main regulatory components.88,89 TNF and IL-1 have been shown to induce the release of uPA and tPA and the synthesis of PAI.19 Activation of brinolysis leads to consumption of α2-antiplasmin and accumulation of FDPs, which if present in high concentrations can interfere with platelet aggregation, brin polymerization, and thrombin formation and can promote bleeding. Additionally, FDPs mediate an increase in vascular permeability. LPS infusion in rabbits90 and human beings91 resulted in an early increase in plasma tPA activity, followed by a later profound rise in PAI activity and fall in tPA activity. Increased plasma PAI concentrations also were found in horses with colic compared with controls.92,93 Thus although brinolysis may compensate initially for accelerated coagulation, its subsequent inhibition contributes to clot formation. Cross-activation between the in ammatory and coagulation cascades plays an important role in endotoxemia and sepsis, and presence of coagulopathies has been associated with an increased risk of organ failure and poor outcome in septic human patients.77,78 Interestingly, activated protein C, an anticoagulant, is the only new treatment shown to reduce mortality in septic human patients.94 Activated protein C reduces in ammation by inhibiting leukocyte activation and cytokine production and was shown to lower plasma concentration of IL-6.94 Although bleeding complications were increased in treated patients, incidence of severe bleeding was not increased significantly.

Complement Activation Activation of the complement system in endotoxemia occurs via the alternative pathway through interaction with LPS. Increased concentrations of plasmin and kallikrein (caused by activation of the brinolytic and contact system) further promote this pathway by directly activating complement factors C3a and C5a. Aside from being key molecules in the complement cascade, C3a and C5a are anaphylatoxins and cause an increase in vascular permeability via mast cell degranulation. C5a further activates the lipoxygenase pathway in neutrophils and monocytes, acts as a chemotaxin for leukocytes and monocytes, and promotes neutrophil adhesion to endothelial cells.

Acute Phase Response In response to acute in ammation, synthesis and secretion of a number of proteins called the acute phase proteins increases in hepatocytes, whereas synthesis of albumin decreases. The primary function of this acute phase response may be to suppress and contain in ammatory processes.55 IL-6 and IL-1 are the most important cytokines that induce the acute phase response,95 which typically begins within a few hours of the insult and subsides within 24 to 48 hours,50 unless the initiating cause persists. In horses brinogen (the most commonly evaluated acute phase protein), haptoglobin, transferrin, ferritin, ceruloplasmin, coagulation factor VIII:C, serum amyloid A, C-reactive protein, α1-acid glycoprotein, and phospholipase A2 are considered part of the acute phase response.96 Serum amyloid A has been evaluated in several studies and was found to be a sensitive indicator of experimentally induced in ammation.97 Although serum amyloid A is a nonspeci c indicator of in ammation, its measurement may be helpful in the diagnosis of in ammatory gastrointestinal disease,37 neonatal weakness and diarrhea,98 and various infectious diseases. A commercially available turbidometric immunoassay for human serum amyloid A was reliable in measuring equine serum amyloid A in one study.99 The e ect of acute in ammation on the serum concentration of several coagulation factors must be considered when evaluating coagulation pro les. Serum brinogen concentration is determined primarily by the acute phase response, although brinogen is consumed increasingly on activation of the clotting cascade.

Hemodynamic Changes, Development of Shock, and Organ Failure Shock is characterized by a loss of homeostasis attributable to breakdown of hemodynamic control mechanisms, decreases in cardiac output and the e ective circulating volume, and inadequate perfusion of vital organs. Shock caused by endotoxemia is classi ed as distributive shock100 and is largely initiated by vascular dysfunction in the periphery. Peripheral vascular beds are of major importance for the regulation of local tissue perfusion and a ect systemic blood pressure by regulating total peripheral resistance. Normally, vascular smooth muscle tone is regulated by endothelin-1 (vasoconstriction), nitric oxide, and prostacyclin (vasodilation) released from vascular endothelial cells.19 As mentioned before, detrimental e ects of nitric oxide are attributable to overproduction of nitric oxide by iNOS in macrophages and other cell types, rather than endothelial-derived nitric oxide. Peripheral vasomotor e ects of endotoxin manifest as vasodilation and vasoplegia and are mediated by PGI2, nitric oxide, and mediators such as bradykinin. Widespread vasodilation leads to vascular blood pooling, decentralization of blood ow, decreased venous return, and in e ect decreased e ective circulating volume and cardiac output.100 Compensatory responses in the form of an initial hyperdynamic phase include tachycardia, increased cardiac output and central venous pressure, pulmonary hypertension, peripheral vasoconstriction, and increased peripheral vascular resistance.100-102 The early vasoconstrictive phase corresponds to an increased serum concentration of TXA2,18 but additional vasoconstrictors such as arginine vasopressin, angiotensin II, serotonin, endothelin, and norepinephrine likely are implicated in the pathogenesis of shock and organ

failure.55 With progression of disease the animal enters a stage of decompensated shock and progressive systemic hypotension, which corresponds to increased plasma concentrations of prostacyclin, PGE2, and bradykinin.18,55 Inadequate blood ow and oxygen delivery to tissues caused by hypotension are confounded by direct myocardial suppression via nitric oxide,100 increased vascular permeability,18 intravascular microthrombosis, and impaired tissue oxygen extraction100 and results in progressive metabolic acidosis and inhibition of normal cellular metabolism.

CLINICAL SIGNS AND DIAGNOSIS Quanti cation of endotoxin in plasma samples is possible. The Limulus amebocyte lysate assay is an activity assay based on the endotoxinsensitive hemolymph coagulation cascade in the horseshoe crab Limulus polyphemus. In Limulus this reaction is thought to be a defense mechanism against gram negative infection.103 Although frequently used as a research tool, the assay is not convenient enough to become a routine clinical test. The clinician therefore must appreciate the primary disease processes associated with a high risk of endotoxemia and rely on clinical signs and clinicopathologic data to achieve a diagnosis. In a survey of board-certi ed internists and surgeons concerning endotoxemia in horses, colitis and enteritis, small intestinal strangulation and obstruction, retained placenta and metritis in mares, grain overload, and pleuropneumonia were among the most frequently cited conditions associated with endotoxemia.104 In some cases, endotoxemia may be the rst indication of disease or may be the most overt of otherwise subtle clinical manifestations. With colitis or proximal enteritis, for example, one may detect signs of endotoxemia before the development of colic, diarrhea, or gastric reflux, which more speci cally indicate the nature of the primary illness. Presence of neutropenia should always prompt the clinician to investigate causes of endotoxemia, including septic processes. In vivo LPS challenge experiments in horses clearly show that many of the clinical signs associated with acute gastrointestinal disease and sepsis are attributable to endotoxemia. On administration of sublethal doses of LPS, the clinical response can be divided into the early hyperdynamic and the later hypodynamic or shock phases. Clinical signs during the rst phase, which begins within 15 to 45 minutes after LPS administration, include anorexia, yawning, sweating, depression, evidence of abdominal discomfort, muscle fasciculations, and recumbency. Heart and respiratory rates increase, and decreased borborygmi suggest ileus. Hyperemia of the mucous membranes and an accelerated capillary re ll time indicate the hyperdynamic state.67 If large amounts of LPS are administered or if exposure is ongoing, depression worsens progressively, anorexia persists, and feces develop diarrheic character. Signs of colic typically abate after the initial stage. Fever develops as a result of direct action of TNF on the thermoregulatory center and IL-1–induced local production of PGE2 in or near the hypothalamus.105,106 Because of compromised peripheral perfusion, mucous membrane color changes to brick red or purple, a dark “toxic” line appears, and capillary re ll time is prolonged.67 Inadequate peripheral perfusion and compromised organ function nally characterize the hypodynamic shock phase. Body temperature may become subnormal, and the skin, especially on extremities, is cool to the touch. The arterial pulse weakens and venous ll is decreased. Vascular endothelial damage and increased capillary permeability result in a muddy mucous membrane color and di use scleral reddening. Similar changes are evident in horses su ering from endotoxemia associated with natural disease. In the previously mentioned survey,104 tachycardia, fever, abnormal mucous membrane color, and increased capillary re ll time were named by most specialists as indicating endotoxemia. Hemostatic abnormalities can manifest in the form of thrombosis, such as of the jugular vein, or increased bleeding tendency with mucosal petechiation or ecchymoses and prolonged bleeding from venipuncture sites.79 Bleeding also may occur in the form of spontaneous epistaxis or prolonged hemorrhage after nasogastric intubation.18 Additional clinical signs typically re ect the development of organ failure. Renal failure and laminitis104 appear to be common complications of endotoxemia in horses, and endotoxemia was identi ed as the only clinical condition signi cantly associated with the development of acute laminitis in one retrospective case-control study of horses admitted to a referral center.107 Other potential complications include liver failure,79 respiratory failure, colic and ischemia-induced gastrointestinal ulceration,18 cardiac failure, and abortion in pregnant mares.108,109 Renal failure results from ischemic cortical necrosis and acute tubular necrosis caused by coagulopathy-induced a erent arteriolar obstruction. Clinical signs may include oliguria, anuria, or hematuria caused by renal infarction. Laminitis may lead to lameness, increased digital arterial pulsation, increased warmth of the hoof wall, and sensitivity to hoof tester pressure. The exact nature of the association between endotoxemia and laminitis is not understood, and, interestingly, experimental endotoxin infusion does not reliably induce laminitis. Studies have shown, however, that endotoxin administration decreases digital blood ow and laminar perfusion110 coincident with increased plasma concentrations of 5-hydroxytryptamine and thromboxane B2.111 In addition, in vitro vascular reactivity of digital vessels is altered following sublethal endotoxin administration to horses.112 In addition to the response to circulating mediators, LPS exposure also alters the production of vasoactive mediators by digital vascular endothelial cells.113 Alterations in vascular reactivity are therefore likely responsible for development of laminitis in endotoxemia; however, other mechanisms may apply depending on the underlying clinical disease.

CLINICAL PATHOLOGIC TESTING Alterations in the hemogram and serum biochemical profile mainly reflect the underlying disease process and the occurrence of organ failure. Leukopenia caused by neutropenia may be the most speci c indicator of acute bacterial sepsis or endotoxemia.67 In prolonged cases an increased proportion of immature neutrophil forms (bands), and toxic changes are observed. Toxic changes resulting from neutrophil activation include vacuolation, cytoplasmic granulation, basophilic cytoplasm, and Döhle’s bodies. Because neutropenia occurs early in the development of endotoxemia, it also may be a useful parameter for monitoring horses at risk.67 On recovery neutropenia typically is followed by a pronounced rebound neutrophilia. An elevated hematocrit and total serum protein concentration are frequently interpreted as evidence of dehydration; however, splenic contraction caused by increased sympathetic stimulation, increased production of acute phase proteins, or protein losses also in uence these parameters. Hyperproteinemia may be observed as a result of increases in the brinogen or globulin concentration, and determination of protein fractions, including protein electrophoresis, is indicated in hyperproteinemic patients. Hypoproteinemia and hypoalbuminemia can occur because of loss via the gastrointestinal or urinary tract or with pleural or peritoneal cavity e usion. Increased vascular permeability and edema formation contribute to hypoproteinemia. Serum electrolyte abnormalities primarily depend on the nature and duration of underlying disease processes and need to be evaluated individually. Common sources of electrolyte loss are gastrointestinal secretions, urine, and sweat; however, severe e usion into body cavities may contribute. In anorexic patients lack of dietary intake is a confounding factor that warrants consideration. In human patients gram negative sepsis frequently is associated with hypocalcemia, more speci cally a decrease in serum ionized calcium concentration. Endotoxin is thought to be a causative factor, and proposed mechanisms include acquired parathyroid gland insu ciency, dietary vitamin D de ciency, impaired calcium mobilization, and renal 1-hydroxylase insu ciency leading to decreased 1,25-hydroxylation of vitamin D. Hypocalcemia in septic human patients was associated with hypotension and poor outcome.114 In horses with surgically managed gastrointestinal disease, decreased serum ionized calcium concentration was a common nding and was most severe in patients with strangulating or nonstrangulating infarctions. In some horses ionized calcium concentration decreased further throughout surgery. Treatment with calcium gluconate resulted in normalization of serum ionized calcium concentrations in all cases.115 Septic neonatal patients are frequently hypoglycemic, which may be attributable to decreased oral intake, generally increased metabolism, glucose use by the infecting bacteria, inhibition of gluconeogenesis by endotoxin, and insulin-like activity produced by macrophages.18 Interestingly, experimental endotoxin administration results in transient hyperglycemia in adult horses,101 whereas profound hypoglycemia occurs in foals.116 Because of the high incidence of coagulopathies in endotoxemic and septic patients, clinicians should consider monitoring coagulation parameters. The most signi cant changes can be expected with severe in ammatory disease such as colitis,75,76 devitalized intestine as with strangulating obstruction,76,117 and with increased duration of disease. In 30 horses with acute gastrointestinal disease, coagulation pro les were considered normal in only two horses.75 Although coagulation times may be shortened during the procoagulant state, commonly observed abnormalities with developing DIC include an increased concentration of FDPs and soluble brin monomer, prolonged prothrombin time indicative of factor VII consumption, prolonged activated partial thromboplastin time indicative of factor VIII:C and IX consumption, prolonged thrombin time, decreased AT III activity, thrombocytopenia, and decreased protein C and plasminogen activities. Fibrinogen concentration appears to re ect the acute phase response rather than coagulation abnormalities in horses and is frequently increased.83 Some clinicians make a diagnosis of DIC if three or more coagulation parameters (speci cally AT III, FDPs, platelet count, prothrombin time, and activated partial thromboplastin time) are abnormal,117 whereas others require overt clinical signs of hemorrhage and concomitant thrombosis in addition to classic laboratory ndings.76 The prognostic value of coagulation parameters has been evaluated.17,76,93 Overall, persistence or worsening of abnormalities in the face of treatment appears to be more indicative of poor outcome than alterations in any speci c parameter. In one study decreased serum AT III concentration was the parameter most commonly associated with fatal outcome in mature horses with colic.75 Clinical pathologic parameters are also helpful in identifying potential underlying diseases and monitoring organ function. Serum creatinine and urea nitrogen concentration in combination with urinalysis parameters are usually monitored as indicators of renal function, whereas serum activity of liver enzymes, serum indirect and direct bilirubin concentration, serum bile acids, and blood ammonia are useful indicators of liver function. One should evaluate arterial blood gases in patients with primary respiratory disease or with clinical evidence of respiratory failure and in profoundly depressed, recumbent patients, especially neonates. Hypoxemia observed in response to endotoxin infusion is thought to be caused by an increase in ventilation-perfusion mismatch rather than pulmonary edema, as occurs in human patients with acute respiratory distress syndrome. Pulmonary edema may occur in patients with associated sepsis or complications such as DIC.118

MANAGEMENT

The ideal treatment for endotoxemia is prevention. Recognition and close monitoring of patients at risk are crucial because doing so allows institution of timely, possibly proactive treatment, which may reverse the e ects of endotoxin before the in ammatory response has developed a dynamic of its own. Unfortunately, endotoxemia can develop rapidly, and horses are exquisitely sensitive to the e ects of endotoxin; therefore many equine patients are not presented for evaluation until they have reached more severe stages of endotoxemia or shock. Prognosis and patient outcome then frequently depend on the severity of complications associated with endotoxemia.18 Treatment of endotoxemia involves multiple aspects, and the following strategies have been proposed:119 • Inhibition of endotoxin release into the circulation • Scavenging of LPS molecules to prevent direct effects and interaction with inflammatory cells • Inhibition of cellular activation by LPS • Inhibition of mediator synthesis • Interference with the effects of inflammatory mediators • General supportive care In addition, treatment must also address the primary disease process as well as any complications. When evaluating reports concerning the e cacy of any one treatment, the clinician should keep in mind di erences in underlying disease processes and the complexity of the in ammatory cascade. A “one for all” treatment most likely will not be found; similarly, any one treatment can address only a few pathophysiologic aspects of endotoxemia at most. In human septic patients many treatments that showed initial promise for improving patient outcome failed to show bene t in larger populations, although bene t for certain subgroups of patients was sometimes demonstrated. Understanding the rationale for di erent treatment strategies is therefore important to be able to tailor treatment to the needs of the patient. In the following sections discussion will focus on treatments that have been evaluated for use in horses, while mention is also made of studies in other species, where applicable.

Inhibition of Endotoxin Release into the Circulation Inhibition of endotoxin release requires identi cation and removal of its source. Therefore whenever endotoxemia is evident, the clinician should strive to reach a diagnosis of the underlying disease and ascertain whether ischemic or in amed bowel or a gram negative septic process is present. Aside from history, physical examination, and routine laboratory testing, evaluation may include exploratory laparotomy in colic patients, radiographic and ultrasonographic evaluation of the pleural and peritoneal cavity, ultrasonographic evaluation of umbilical remnants in neonatal foals, evaluation of passive transfer and calculation of a sepsis score in foals, and repeated culturing of blood or other specimens. Identi cation of responsible microorganisms and their antimicrobial sensitivity spectrum is a crucial step toward e ective therapy; however, one should not necessarily delay treatment to obtain culture results. Specimen containers with antimicrobial removal devices may be useful in cases for which initiation of treatment precedes specimen collection. Once a diagnosis is reached, appropriate measures to correct the primary disease process must be taken. Examples are removal of devitalized sections of bowel or infected umbilical remnants, drainage of infected pleural or PF, and gastric lavage followed by administration of mineral oil or intestinal adsorbants in cases of grain overload. Di-tri-octahedral smectite (Biosponge™, Platinum Performance Inc., Buellton, California) was shown to remove endotoxin in an in vitro assay120 and may be useful in preventing endotoxemia of intestinal origin. Septic processes must be addressed with appropriate antimicrobial therapy, and principles of antimicrobial therapy should be followed. Regarding endotoxemia speci cally, antimicrobial therapy has been suggested to increase the amount of circulating endotoxin by inducing endotoxin release on cell death of gram negative bacteria. An in vitro study comparing endotoxin release and in ammatory mediator activity among antimicrobials commonly used to treat E. coli septicemia in foals evaluated amikacin, ampicillin, amikacin plus ampicillin, ceftiofur, and imipenem. Although these antimicrobials showed no di erence in the ability to kill bacteria, amikacin and the amikacin-ampicillin combination resulted in the lowest, and ceftiofur in the greatest, release of endotoxin. Endotoxin release appeared to be dose dependent in that lesser amounts were released at higher antimicrobial concentrations.121 On the basis of these results and clinical experience, combining antimicrobial therapy with endotoxinbinding agents such as polymyxin B may be beneficial, especially when using β-lactam antimicrobials.

Scavenging of Lipopolysaccharide Molecules Endotoxin typically has a short plasma half-life and is removed rapidly by mononuclear phagocytes or neutralized by binding to serum proteins and lipoproteins. Many conditions responsible for the development of endotoxemia in horses, however, may be associated with an ongoing release of endotoxin. Examples include severe gastrointestinal in ammation as in proximal enteritis or colitis, grain overload, or

uncontrolled sepsis. Therapy directed against endotoxin itself may be able to interrupt the continuous activation of the in ammatory cascade in these cases. Further bene ts of anti-endotoxin treatment may be derived if large amounts of endotoxin have been released before the inciting cause can be addressed.

IMMUNOTHERAPY An important consideration regarding the e cacy of immunotherapy is the region of the LPS molecule against which antibodies are raised. The O-chain of LPS acts as a potent antigen on infection with gram negative bacteria;7 however, antibodies directed against the O-chain are serotype speci c and cannot a ord signi cant cross-protection against heterologous bacterial strains. The core and lipid A region, both of which show a much higher degree of hom*ology between LPS derived from di erent bacterial strains, o er a more promising target for immunotherapy. Active immunization against endotoxin has been reported for horses. Vaccination with a bacterin-toxoid vaccine prepared from rough mutants of Salmonella typhimurium or S. enteritidis protected horses against hom*ologous and heterologous endotoxin challenge122,123 and carbohydrate overload.132 Despite these encouraging results and the current availability of a vaccine for use in horses (Endovac-Equi, Immvac Inc., Columbus, Missouri), active immunization against endotoxin does not appear to be a common practice. In comparison, passive immunization with antilipopolysaccharide antibodies is used widely. Rough bacterial mutants, most commonly J5 of E. coli O111:B4 and S. minnesota Re595, are used to immunize donor horses and subsequently prepare serum or plasma products. Proposed mechanisms of action after binding of the antibodies to LPS include steric blockade of lipid A interaction with cellular receptors and enhanced bacterial clearance by opsonization.124-126 Studies concerning the e cacy of antibody administration in equine patients vary in their results. Bene cial e ects have been described in experimental models of endotoxemia, acute gastrointestinal disease, and neonates with sepsis,123,127-130 whereas in other studies antibodies failed to protect foals and horses against endotoxin e ects.131-133 Furthermore, administration of a S. typhimurium antiserum to foals was associated with an increased respiratory rate and higher serum activities of IL-6 and TNF.131 Various equine serum and plasma products are commercially available. An antiserum raised against the LPS-core of S. typhimurium (Endoserum®, Immvac Inc., Columbus, Missouri) is available for administration to endotoxemic horses at a recommended dose of 1.5 ml/kg body mass. Diluting the serum tenfold to 20-fold in crystalloid intravenous solutions, administering it slowly over 1 to 2 hours, and monitoring the patient for adverse reactions is advisable. Although the product is marketed for use in foals with failure of passive transfer, adverse effects have been reported,131 and one should use caution when administering it to neonates. Plasma from donors inoculated with J5 (E. coli) and S. typhimurium (Re mutant) is available under a California license (Equiplas J®, Plasvacc USA Inc, Templeton, California). The manufacturer recommends administration of at least 1 to 2 L in cases of endotoxemia. In addition, hyperimmune plasma, which has a guaranteed minimum immunoglobulin G content but does not contain speci c antiendotoxin antibodies (Hi-Gamm Equi, Lake Immunogenics, Inc., Ontario, New York; Equiplas and Equiplas Plus, Plasvacc USA Inc,), is marketed for treatment of failure of passive transfer, and many clinicians use it to treat endotoxemia and sepsis. In addition to antibodies and protein, plasma contains active constituents such as complement components, bronectin, clotting factors, and AT III128 and therefore may be particularly useful in patients with endotoxemia-induced coagulopathy. Volumes of 2 to 10 ml/kg body mass of hyperimmune plasma have been recommended for use in endotoxemic patients.55,134

POLYMYXIN B Polymyxin B is a cationic polypeptide antibiotic that binds to the anionic lipid A portion of LPS and neutralizes its endotoxin capacity.135 At dosages required for antimicrobial activity, polymyxin B carries the risk of respiratory paralysis and ototoxic, nephrotoxic, and neurotoxic side e ects; however, a much lower dose is required for endotoxin-binding activity. The e ects of polymyxin B in horses have been evaluated in di erent experimental models.131,135,136 In an in vivo study in foals, treatment with polymyxin B at a dosage of 6000 U/kg body mass before infusion with S. typhimurium LPS resulted in signi cantly less severe elevations of body temperature, respiratory rate, and serum activities of TNF and IL-6 compared with untreated controls.131 Similarly, polymyxin B treatment of adult horses given endotoxin signi cantly ameliorated clinical signs and decreased plasma TNF activity.137 In the latter study bene ts of treatment were also evident at lower dosages of polymyxin B (1000 and 5000 U/kg body mass) and administration of polymyxin B 1 hour after the start of endotoxin infusion. Conversely, polymyxin B failed to ameliorate clinical signs of endotoxemia or prevent the development of coagulopathy, acidosis, lameness, and shock in experimental carbohydrate overload.138 Side e ects suggestive of neurotoxicity appeared after repeated administration of 5 mg/kg body mass (36,000 U/kg) and in milder form 2.5 mg/kg body mass (18,000 U/kg) polymyxin B. Nephrotoxicity was not observed. Currently, use of polymyxin B in equine patients is recommended at dosages of 1000 to 6000 U/kg body mass every 8 to 12 hours.139,140 Treatment should be initiated as early in the disease process as possible because the bene cial e ects of LPS scavenging may be limited to the rst 24 to 48 hours after the onset of endotoxemia, before endotoxin tolerance develops. Side e ects in the form of neuromuscular blockade and apnea, which necessitate slow infusion of the drug in human patients, have not been observed in horses.

Therefore the entire dose can be administered as a slow bolus. If treating horses with hypovolemia, dehydration, or azotemia, the clinician should attempt to improve peripheral tissue perfusion, minimize the polymyxin B dose, and closely monitor patients for nephrotoxicity. Close monitoring is also important if medications such as aminoglycoside antibiotics, which share a similar spectrum of potential side e ects, are administered concurrently. Azotemic neonates have been reported to be more susceptible to the nephrotoxic e ects of polymyxin B than adult horses.137 In an attempt to decrease the risk for adverse e ects while preserving LPS-neutralizing ability, a conjugate of polymyxin B with dextran has been developed.141 In conjugated form, polymyxin B is prevented from extravasation into tissues, where it exerts toxic e ects by interaction with cell membranes. In addition, conjugation increases the residence time of polymyxin B in the circulation and therefore should prolong the anti-endotoxic effect. The polymyxin B–dextran combination was evaluated at a total dose of 5 mg/kg body mass of polymyxin B in 6.6 g/kg body mass dextran given 15 minutes before administration of endotoxin in horses.142 Treatment was found to block the development of tachycardia, tachypnea, fever, and neutropenia completely and to prevent increases in serum concentrations of TNF, IL-6, TXB2 (a TXA2 metabolite), and the prostacyclin metabolite 6-keto-PGF1α. Although mild adverse e ects in the form of tachypnea, sweating, and increased systolic blood pressure were observed, these were transient and could be prevented by pretreatment with ketoprofen. To the author’s knowledge, the polymyxin B–dextran combination is not commercially available at this time.

NATURAL ENDOTOXIN-BINDING SUBSTANCES Natural endotoxin-binding proteins such as LBP, lipoproteins, and sCD14 have been evaluated experimentally. Results of these studies are somewhat contradictory, and detrimental e ects occurred in some cases.143 A protein receiving much attention regarding potential therapeutic e cacy is the bactericidal permeability-increasing protein (BPI). This protein is structurally similar to LBP but is expressed exclusively in myeloid precursors of polymorphonuclear leukocytes.144 BPI is stored in primary granules of mature neutrophils and during inflammation is expressed on their cell membranes and secreted into the extracellular environment.145 BPI has an even higher affinity for LPS than LBP146 and shows antibacterial activity speci c for gram negative bacteria.5 Binding of BPI to the gram negative bacterial membrane results in growth arrest and is an important factor in the antibacterial activity of intact neutrophils. Furthermore, BPI binding disrupts normal membrane organization and makes bacteria more susceptible to hydrophobic substances, including antimicrobials.147 Experimentally, recombinant BPI protects against the toxic and lethal e ects of isolated LPS and intact gram negative bacteria, and clinical trials in human patients show promising results regarding its therapeutic use.148 The biology and potential use of BPI in horses has not been evaluated. Phospholipid emulsions have recently been evaluated for treatment of experimentally induced endotoxemia in horses. Phospholipid infusion improved clinical parameters, ameliorated neutropenia, and reduced inflammatory mediator production in response to an endotoxin challenge.149,150 Because hemolysis was a complication of phospholipid infusion in some horses in these studies, optimization of dose and time of administration will be necessary before evaluating this treatment for potential clinical use.

Inhibition of Cellular Activation by Lipopolysaccharide Treatments aimed at inhibiting LPS interaction with cells or turning off intracellular signaling pathways are under investigation. Nontoxic LPS or lipid A structures can act as endotoxin antagonists, if they competitively inhibit binding to LBP or cellular receptors or inhibit cellular activation by other mechanisms. Of the potential antagonists that have been evaluated experimentally, LPS and lipid A of the phototrophic bacterium Rhodobacter sphaeroides and the synthetic compounds E5531 and E5564 have been most promising.150-156 Unfortunately, species di erences exist regarding cellular response to these structures, and R. sphaeroides LPS as well as E5531 have been found to have agonist activity in equine cells.157,158 In cell transfection experiments, TLR4 was shown to be responsible for this phenotypic variation with regard to R. sphaeroides LPS.159 Given these results, any future potential LPS antagonists must be evaluated in equine systems.

Inhibition of Mediator Synthesis NONSTEROIDAL ANTI-INFLAMMATORY DRUGS NSAIDs are probably the most commonly used drugs to treat endotoxemia in horses. The rationale for their use is inhibition of COX and thereby inhibition of prostanoid production (see Figure 15-10). Additional bene cial e ects may include scavenging of oxygen-derived free radicals and iron chelation; however, side e ects may occur at dosages required to achieve these e ects.160 Prostanoids have been identi ed as important mediators in the in ammatory response in a number of studies, and inhibition of their synthesis is associated with bene cial e ects. Generally speaking, two COX isoforms are recognized: constitutively expressed COX-1 and inducible COX-2. Upregulation of COX-2 expression results from various pro-in ammatory stimuli, including LPS, TNF α, and IL-1.161 Constitutively expressed COX products are likely important for maintenance of homeostasis, whereas increased production of prostanoids by COX-2 is thought to be responsible for detrimental e ects during in ammation and shock. Research has focused on the development of selective COX-2 inhibitors, and rocoxib

(Equioxx® , Merial Ltd., Duluth, Georgia) has been approved for treatment of osteoarthritis in horses. This drug has not been evaluated critically for treatment of endotoxemia. In horses the most commonly used NSAID to treat endotoxemia is unixin meglumine. Bene cial e ects of unixin meglumine have been described in experimental models of endotoxemia162-164 and in clinical cases. In equine colic patients treatment with unixin meglumine before exploratory surgery resulted in reduced plasma concentrations of TXB2 and PGE2 and had a favorable e ect on cardiovascular parameters.165 Flunixin meglumine was shown further to maintain cardiac output and systemic arterial blood pressure, improve blood ow to vital organs, reduce pulmonary endothelial damage, and improve survival on endotoxin challenge.66,166-168 Conversely, in vitro studies have suggested that unixin meglumine impairs recovery of intestinal barrier function in intestinal segments subjected to ischemia-reperfusion injury and may increase mucosal permeability to LPS.169,170 This e ect may be reduced or eliminated by concurrent administration of continuous-rate infusion of lidocaine. NSAID use in horses carries the risk of side e ects, the most signi cant of which is the development of gastrointestinal ulceration and renal papillary necrosis (renal crest necrosis). Di erences may exist among NSAIDs in their propensity to induce adverse e ects,171 but all NSAIDs must be used cautiously. A concern about NSAID use speci cally in colic patients is the masking of cardiovascular e ects of endotoxin, which are used to determine the necessity of surgical exploration.172 For these reasons a reduced dose of unixin meglumine (0.25 mg/kg body mass thrice daily) has been suggested and is used widely in horses.104 At this dosage unixin meglumine inhibits eicosanoid synthesis e ciently in an in vivo model of endotoxemia.173 Reduction of clinical signs, however, was dose dependent, and lower doses provide minimal, if any, analgesia (see Chapter 4 for further discussion of NSAID therapy in horses). Therefore the veterinarian should choose the appropriate dose of any NSAID after carefully considering the circ*mstances of each case. Some researchers have suggested that ketoprofen o ers superior e ects because of a proposed dual inhibitory e ect on COX and lipoxygenase and may carry a decreased risk of adverse e ects compared with unixin meglumine and phenylbutazone. A comparison of cytokine and eicosanoid production by LPS-stimulated isolated monocytes in vitro, however, showed no signi cant di erence between horses pretreated with flunixin meglumine (1.1 mg/kg body mass) or ketoprofen (2.2 mg/kg body mass), respectively.174 Eltenac has been evaluated in an experimental endotoxemia model in horses.175 Given 15 minutes before LPS infusion, eltenac at a dose of 0.5 mg/kg protected against changes in clinical, hemodynamic, and hematologic parameters and blunted the LPS-induced rise in plasma cytokine concentrations in comparison with controls. Some parameters, however, including heart rate, leukocyte count, lactate concentration, and plasma TNF activity, were not improved. Ibuprofen may have bene cial e ects superior to those of the other NSAIDs because it may be possible to achieve tissue concentrations safely that allow iron chelation to occur. According to a study in healthy foals, dosages of ibuprofen up to 25 mg/kg every 8 hours can be given safely for up to 6 days.160

CORTICOSTEROIDS The use of corticosteroids for anti-in ammatory therapy in sepsis and endotoxemia has been controversial in human and equine patients, and bene cial e ects superior to the ones achieved by NSAIDs have not been demonstrated consistently. Corticosteroids inhibit the activity of phospholipase A2 and the release of arachidonic acid from cell membrane phospholipids, as well as the production of TNF, IL-1, and IL-6 in response to a LPS stimulus. Experimentally, bene cial e ects of dexamethasone in equine endotoxemia have been demonstrated.176,177 To inhibit TNF production by equine peritoneal macrophages, however, the required concentration of dexamethasone was high and corresponded to an in vivo dosage (approximately 3 mg/kg body mass) greatly exceeding current recommendations.176 Although single doses of corticosteroids are unlikely to carry a disproportionate risk of adverse e ects, the clinician should consider the suggested association of laminitis with corticosteroid use in horses. In cases of sepsis immunosuppressive effects could also be detrimental. In human patients with certain types of septic shock, dysfunction of the hypothalamic-pituitary-adrenal axis has been recognized and successfully treated with hydrocortisone replacement therapy.178 Hypothalamic-pituitary-adrenal axis dysfunction has been suggested to occur in septic foals;179 however, the use of low-dose corticosteroids for this indication remains to be investigated in horses.

PENTOXIFYLLINE Pentoxifylline, a methylxanthine derivative and phosphodiesterase inhibitor, has been suggested for use in endotoxemia because of its e ects on neutrophil function and its ability to inhibit the production of various cytokines, interferons, and thromboplastin. Decreased production of TNF, IL-6, TXB2, and thromboplastin in response to endotoxin was shown in an equine ex vivo model.180 In horses given endotoxin followed by treatment with pentoxifylline (7.5 mg/kg body mass followed by continuous infusion of 3 mg/kg/hr for 3 hours), however, only minimal bene cial e ects were observed.181 Treatment signi cantly improved body temperature, respiratory rate, and whole blood recalci cation time, but no e ect was observed regarding heart rate, blood pressure, leukocyte count, plasma brinogen concentration, and serum cytokine concentrations. The conclusion was that bene ts of treatment with pentoxifylline might be restricted to administration of high bolus doses or

continuous infusion early in the pathophysiologic process. In an in vivo endotoxemia model in horses, the combination of pentoxifylline (8 mg/kg body mass) and unixin meglumine (1.1 mg/kg body mass) was found to have greater bene t than each treatment on its own.182 Because of its rheologic properties (i.e., the ability to increase erythrocyte deformability and microvascular blood ow), pentoxifylline has been suggested for use in endotoxemic patients showing evidence of laminitis; however, no e ect on blood ow to the hoof was demonstrated after administration to healthy horses.183 An intravenous preparation of pentoxifylline is not commercially available.

ANTIOXIDANTS Dimethyl sulfoxide (DMSO) is used by some clinicians in an attempt to scavenge oxygen-derived radicals. The treatment may be most appropriate in cases of ischemia-induced intestinal damage and associated reperfusion injury. However, DMSO failed to show bene cial e ects in an experimental model of intestinal ischemia when administered on reperfusion of the ischemic intestine.184 DMSO at a dose of 1 g/kg body mass was shown to increase mucosal loss after ischemia and reperfusion of the large colon,185 and hence a reduced dose of 0.1 g/kg body mass has been proposed for cases of intestinal ischemia. DMSO failed to show signi cant bene t in an experimental model of endotoxemia in horses, although it ameliorated the e ect on fever.186 For intravenous administration, DMSO should be diluted in polyionic solutions to a concentration not exceeding 10%. Oral administration of a 10% to 20% solution via nasogastric intubation is also possible. Aside from DMSO the xanthine oxidase inhibitor allopurinol has been suggested as a treatment to prevent oxygen radical–induced tissue damage. During periods of ischemia tissue xanthine dehydrogenase is converted to xanthine oxidase, which on reperfusion catalyzes the generation of superoxide radicals.187,188 Evaluation in horses showed bene cial e ects of 5 mg allopurinol per kilogram body mass administered 12 hours before endotoxin challenge.189 In another study mucosal damage attributable to oxygen-derived free radicals was not attenuated by allopurinol in an experimental ischemia-reperfusion model.185

LIDOCAINE Lidocaine given intravenously has been suggested as an anti-in ammatory, analgesic, and prokinetic agent, and some clinicians use it to treat colic and laminitis in horses. In an experimental endotoxemia model in rabbits, lidocaine inhibited hemodynamic and cytokine responses to endotoxin profoundly if given immediately after LPS infusion.190 Lidocaine further ameliorated the inhibitory e ects of unixin on recovery of mucosal barrier function following ischemic injury in equine small intestine.191 Use of lidocaine therefore may have merit in endotoxemic patients. A common regimen for lidocaine use in horses is administration of an initial bolus (1.3 mg/kg body mass) followed by continuous infusion at a rate of 0.05 mg/kg/min. The veterinarian should monitor patients for toxic neurologic e ects associated with a lidocaine overdose. ω-3 FATTY ACIDS

High concentrations of -3 fatty acids can alter the phospholipid composition of cellular membranes toward a decreased ratio of -6 to -3 and thereby can a ect membrane functions such as phagocytosis, receptor binding, and activities of membrane-bound enzymes.67 Most important for the treatment of endotoxemia, -3 fatty acid incorporation into cell membranes decreases the availability of arachidonic acid (an -6 fatty acid) for eicosanoid synthesis192 and provides alternative substrates. Metabolism of -3 fatty acids via the COX and lipoxygenase pathway leads to the production of 3-series prostaglandins and 5-series leukotrienes, which have less biologic activity than their 2-series and 4-series counterparts derived from arachidonic acid. Aside from these mechanisms, -3 fatty acids prevent LPS-induced upregulation of CD14 in monocytic cells and therefore may be able to block transmembrane signaling of LPS.193 Cells from horses given linseed oil (high in -3 fatty acids) for 8 weeks before blood collection showed signi cantly decreased expression of procoagulant activity, TXB2, and TNF in response to LPS stimulation.194,195 In an in vivo experimental model of endotoxemia in horses, treatment resulted in prolonged activated partial thromboplastin time and whole blood recalci cation time, suggesting an anticoagulant e ect; however, a signi cant bene cial e ect on clinical response and serum eicosanoid concentrations was not observed.196 Because dietary addition of -3 fatty acids requires several weeks of treatment, intravenous infusion was evaluated and shown to alter the composition of cell membrane phospholipids rapidly.197 Further evaluation of this treatment for use in horses is necessary before speci c dosage recommendations can be made.

Interference with the Effects of Specific Inflammatory Mediators ANTIBODIES DIRECTED AGAINST TUMOR NECROSIS FACTOR Monoclonal and polyclonal antibodies against equine TNF have been evaluated in horses.198-200 Administration of a monoclonal antibody preparation before LPS infusion resulted in signi cantly reduced plasma TNF activity, improved clinical abnormality scores, lower heart rate, and higher leukocyte count compared with controls.199 Furthermore, plasma concentrations of lactate and 6-keto-PGF1α were reduced signi cantly, whereas TXA2 production was not a ected.198 In another study200 administration of a rabbit polyclonal antibody against

recombinant human TNF did not improve clinical and hematologic parameters when given shortly (15 minutes) after LPS infusion, although inhibition of TNF activity was present in vitro.200,201 Findings in horses are in agreement with studies in other species and suggest that bene cial e ects of TNF inhibition may be limited to administration before LPS exposure. Widespread clinical use therefore is unlikely to become feasible. Clinical trials in septic human patients have not shown significant benefits of TNF antibody treatment.202,203

PLATELET-ACTIVATING FACTOR RECEPTOR ANTAGONISTS The e ects of selective PAF receptor antagonists have been evaluated. PAF is implicated in the development of systemic hypotension,204 LPSinduced platelet aggregation,205 ileus,206 and increased vascular permeability207 and may mediate recruitment of leukocytes to in amed tissues.208,209 A study in horses using the PAF receptor antagonist SRI 63-441 before LPS infusion showed signi cant decreases in heart rate and shorter elevation of lactate concentrations in response to the treatment. Although not statistically signi cant, additional bene cial e ects included delayed onset of fever, a shortened period of neutropenia, and reduced maximal platelet aggregation.210

Supportive Care FLUID THERAPY AND CARDIOVASCULAR SUPPORT Fluid therapy is a mainstay of therapy of most endotoxemic patients su ering from the cardiovascular e ects of systemic in ammation. Many endotoxemic equine patients will further require uid therapy for treatment of the underlying disease process and correction of dehydration and electrolyte and acid-base abnormalities. Principles of fluid therapy are discussed elsewhere in this text. Patients with severe hypovolemia and shock present management challenges, especially because increased vascular permeability in endotoxemic patients requires careful consideration of uid therapy plans. A rapid increase in total body uid volume may be detrimental in patients with compromised cardiac and peripheral vasomotor function and may increase the severity of vascular pooling in peripheral organs. In these patients hypertonic solutions or colloids may be more appropriate means of stabilization than large volumes of crystalloid solutions. Hypertonic saline solution (7.5% sodium chloride) is the most commonly used hypertonic solution in horses and has bene cial e ects in endotoxemic patients.211 A dose of 4 ml/kg is recommended, which should be given as a bolus infusion over 10 to 15 minutes, followed by administration of an isotonic solution to restore total body uid volume. The clinician should use hypertonic saline with caution in patients with sodium or chloride derangements and should monitor serum electrolyte concentrations in the case of repeated administration. Improvement of the cardiovascular status in response to uid therapy is indicated by normalization of heart rate, mucous membrane color, and capillary re ll time. Failure of urination to occur despite appropriate uid resuscitation should result in critical evaluation of renal function. In one recent study small volume resuscitation with hypertonic saline plus hetastarch failed to alleviate hemodynamic responses in experimental endotoxin infusion in horses.212 Plasma is an ideal colloid and should be administered to maintain a serum total protein concentration above 4.2 g/dl.134 To raise plasma protein concentration and colloid osmotic pressure signi cantly, however, horses often require large volumes of plasma (7 to 10 L or more in a 450-kg horse), and alternative colloids should be considered. High-molecular-weight polymers are thought to provide superior oncotic e ects in cases of sepsis and endotoxemia, when vascular permeability is increased. Hetastarch, or hydroxyethyl starch (Hespan), is commercially available as a 6% solution in 0.9% sodium chloride. Hetastarch molecules have very high molecular weight, and degradation must occur before renal excretion.213 These properties result in a longer plasma half-life and prolonged oncotic e ects compared with other colloids; persistence of the oncotic e ect for 24 hours was observed in hypoproteinemic horses.214 A dosage of 5 to 15 ml/kg given by slow intravenous infusion along with an equal or greater volume of crystalloid uids has been recommended.213,215 In human patients prolonged activated partial thromboplastin time, decreased factor VIII activity, and decreased serum brinogen concentration have been described in association with hetastarch use.216 In the limited number of equine studies, bleeding times were not a ected;217,218 however, patients treated with hetastarch should be monitored for coagulopathy. Metabolic acidosis in endotoxic shock is attributable to lactic acidemia and inadequate tissue perfusion.219 Acid-base balance often improves considerably after uid resuscitation (preferably with alkalinizing solutions such as lactated Ringer’s solution) alone; however, additional sodium bicarbonate may be required in cases in which serum bicarbonate concentration remains below 15 mEq/L. Foals with sepsis are frequently hypoglycemic, and 5% dextrose solutions are useful as initial resuscitation uids. The clinician should reduce the glucose concentration of intravenous solutions according to the blood glucose concentration to avoid prolonged hyperglycemia. Administration of hyperimmune plasma (20 to 40 ml/kg body mass) is highly recommended in foals with evidence of partial or complete failure of passive transfer. One should consider positive inotropic and vasomotor agents in patients with persistently inadequate tissue perfusion. Lower dosages of dopamine (0.5 to 2 g/kg/min) result in vasodilation of the renal, mesenteric, coronary, and intracerebral vasculature via dopaminergic e ects, whereas higher dosages (up to 10 g/kg/min) also exert stimulation of α1-adrenergic receptors, resulting in increased myocardial

contractility and heart rate.220 Dobutamine is a direct α1-adrenergic agonist and does not appear to have signi cant vasodilator properties. Dosages for dobutamine of 1 to 5 g/kg/min as a continuous intravenous infusion have been recommended for use in horses. In addition, norepinephrine was evaluated in hypotensive critically ill foals that were refractory to the e ects of dopamine and dobutamine.221 At dosages up to 1.5 g/kg/min administered concurrently with dobutamine, six out of seven foals showed an increase in mean arterial pressure, and all foals had increased urine output. Because of the risk of cardiac side effects, close monitoring of heart rate and rhythm should accompany infusion of inotropes. Indirect blood pressure measurements using a tail cuff may be used to monitor the effects of treatment.

MANAGEMENT OF COAGULOPATHY More frequently than overt thrombosis or bleeding attributable to DIC, hemostatic abnormalities occur in the form of alterations in the coagulation pro le. A procoagulant state with shortened bleeding times or prolonged bleeding times caused by consumption of clotting factors may be evident. One should address abnormalities in the coagulation pro le as early as possible but especially if they persist more than 24 hours after initiation of therapy. Because of the complex interactions of coagulation and brinolysis during endotoxemia, it might be necessary to combine anticoagulant therapy with the administration of fresh frozen plasma to replace clotting and brinolytic factors. Heparin acts as an anticoagulant by activation of AT III and subsequent inhibition of thrombin, release of tissue factor pathway inhibitor from endothelial cells, and inhibition of platelet aggregation.222 Because endogenous AT III levels frequently are decreased in patients with coagulopathy, addition of heparin to fresh frozen plasma may be the most e ective route of administration. An initial dose of 100 IU/kg body mass followed by 40 to 80 IU/kg body mass thrice daily has been recommended.134 Anemia caused by erythrocyte agglutination occurs in some patients during therapy with unfractionated heparin223,224 but typically resolves within 96 hours if therapy is discontinued.134 Because of the risk of microthrombosis associated with erythrocyte agglutination, use of low-molecular-weight heparin (50 IU/kg body mass subcutaneously every 24 hours) has been recommended225 but may be cost prohibitive. Aspirin can be given orally (10 to 20 mg/kg body mass, every 48 hours), which irreversibly inhibits platelet COX activity, to inhibit platelet aggregation and microthrombosis. Platelet hyperaggregability has been implicated in the pathogenesis of carbohydrate-induced laminitis,226 and heparin and aspirin have been recommended to prevent development of laminitis. In an in vitro study, however, aspirin did not inhibit endotoxin-induced platelet aggregation.227

OTHER CONSIDERATIONS Luteolysis caused by increased concentrations of PGF2α may lead to pregnancy loss in endotoxemic mares before day 55 of pregnancy (see Chapter 18).228 Daily administration of altrenogest (Regu-Mate, Hoechst-Roussel Agri-Vet, Somerville, New Jersey) at a dose of 44 mg orally consistently prevented fetal loss in mares if administered until day 70 of pregnancy.108 Treatment with unixin meglumine, by blockade of PGF2α release,109 also may contribute to the maintenance of pregnancy in endotoxemic mares. The pathogenesis of fetal loss and abortion caused by endotoxemia, surgery, or systemic disease later in gestation is not understood completely. Proposed mechanisms include direct effects on the fetus, placental function, or placental progesterone production.229 Decreased nitric oxide production by vascular endothelial cells in response to endotoxin has been suggested as a mechanism for vasoconstriction and decreased blood ow leading to laminitis (see Chapter 11);230 however, use of nitric oxide donors remains controversial. Maintenance of adequate peripheral perfusion and anticoagulant and anti-in ammatory therapy may be helpful in preventing and treating laminitis caused by endotoxemia.

SUMMARY Although the innate immune response to endotoxin (LPS) is crucially important for the preservation of homeostasis and health, large amounts of endotoxin can evoke an excessive and uncontrolled in ammatory response and result in a dysfunction of hemostatic and circulatory control mechanisms, loss of vascular integrity, and nally tissue damage and organ failure. Conditions commonly associated with the development of endotoxemia in horses are acute gastrointestinal diseases, especially of ischemic and severe in ammatory nature, and localized or generalized infections. Although measuring endotoxin concentrations in equine plasma is possible, a diagnosis of endotoxemia is typically based on clinical signs and clinicopathologic data. Successful treatment of endotoxemia requires resolution of the primary disease process in addition to neutralization of circulating endotoxin, interference with the activities of in ammatory mediators, and general supportive care. Newer treatments, such as blockade of endotoxin interaction with cells or interruption of cell signaling pathways, are under investigation. Possible sequelae of endotoxemia include DIC, multiple organ failure, circulatory failure, and death. Frequently, the outcome of conditions associated with endotoxemia in horses depends on the severity of associated complications (e.g., renal compromise and laminitis).

ORAL DISEASES∗

L. Chris Sanchez The word mouth is used commonly to signify the rst part of the alimentary canal or the entrance to it.1 The mouth is bounded laterally by the cheeks, dorsally by the palate, and ventrally by the body of the mandible and by the mylohyoideus muscles. The caudal margin is the soft palate. The mouth of the horse is long and cylindric, and when the lips are closed, the contained structures almost ll the cavity. A small space remains between the root of the tongue and the epiglottis and is termed the oropharynx. The cavity of the mouth is subdivided into sections by the teeth. The space external to the teeth and enclosed by the lips is termed the vesicle of the mouth, and in the resting state the lateral margins of the vesicle (i.e., the buccal mucosa) are in close contact with the cheek teeth. Caudally, the external space communicates with the pharynx through the aditus pharyngis. The mucous membrane of the mouth is continuous at the margin of the lips with the skin and during life is chiefly pink but can be more or less pigmented, depending on the skin color and the breed type.

MORPHOLOGY AND FUNCTION The lips are two muscular membranous folds that unite at angles close to the rst cheek teeth. Each lip presents an outer and an inner surface. The upper lip has a shallow median furrow (philtrum); the lower lip has a rounded prominence or chin (mentum). The internal surface is covered with a thick mucous membrane that contains small, pitted surfaces that are the openings of the ducts of the labial glands. Small folds of the mucous membrane called the frenula labii pass from the lips to the gum. The free border of the lip is dense and bears short, sti hairs. The arteries of the mouth are derived from the maxillary, mandibular, labial, and sphenopalatine arteries of the major palatine artery. The veins drain chie y to the lingual facial vein. Sensory nerves originate from the trigeminal nerve (cranial nerve V) and the motor nerves from the facial nerve (VII). The cheeks spread back from the lips and from both sides of the mouth and are attached to the alveolar borders of the bones of the jaws. The cheeks are composed of skin and muscular and glandular layers and then the internal mucous membrane. The skin is thin and pliable. In contrast, the oral mucous membrane is dense and in many areas of the oral cavity is attached rmly to the periosteum so that construction of oral mucosal aps can be achieved only by horizontal division of the periosteal attachment. Such a feature is important in reconstructive techniques applied to the oral cavity. The blood supply to the cheeks comes from the facial and buccal arteries and the sensory nerves from the trigeminal and motor nerves from the facial nerve. The hard palate (palatum durum) is bounded rostrally and laterally by the alveolar arches and is continuous with the soft palate caudally. The hard palate has a central raphe that divides the surface into two equal portions. From the line of the rostral cheek tooth, the hard palate is concave to the line of the caudal cheek tooth. Paired transverse ridges (approximately 18) traverse the concavity and have their free edges directed caudally. The incisive duct is a small tube of mucous membrane that extends obliquely through the palatine ssure. The dorsal component communicates by a slitlike opening in the rostral portion of the ventral nasal meatus, and its palatine end is blind and lies in the submucosa of the palate. When stallions display their ehmen response, watery secretions enter the nose from the glands of the vomeronasal duct. To what extent these secretions aid in pheromone reception is not known.2 That portion of the palatine mucosa immediately behind the incisor teeth frequently is swollen (lampas) during eruption of the permanent teeth. This swelling is physiologic and not pathologic. The tongue is situated on the oor of the mouth between the bodies of the mandible and is supported by the sling formed by the mylohyoideus muscles. The root of the tongue is attached to the hyoid bone, soft palate, and pharynx. The upper surface and the rostral portion of the tongue are free; the body of the tongue has three surfaces. The apex of the tongue is spatulate and has a rounded border. The mucous membrane adheres intimately to the adjacent structure and on the dorsum is dense and thick. From the lower surface of the free part of the tongue, a fold of mucous membrane passes to the oor of the mouth, forming the lingual frenulum. Caudally, a fold passes on each side of the dorsum to join the soft palate, forming the palatoglossal arch. Dorsally from the soft palate, the palatopharyngeal arch attaches and circumvents the aditus laryngis and attaches to the roof of the nasopharynx. The mucous membrane of the tongue presents four kinds of papillae: 1. Filiform papillae are ne threadlike projections across the dorsum of the tongue. They are absent on the root of the tongue and are small on the rostral portion of the tongue. 2. The fungiform papillae are larger and easily seen at the rounded free end. They occur principally on the lateral portion of the tongue. 3. Vallate papillae are usually two or three in number and are found on the caudal portion of the dorsum of the tongue. The free surface bears numerous small, round secondary papillae. 4. Foliate papillae are situated rostral to the palatoglossal arches of the soft palate, where they form a rounded eminence about 2 or 3 cm in

length marked by transverse fissures. Foliate, vallate, and fungiform papillae are covered with taste buds and secondary papillae. The lingual and sublingual arteries supply the tongue from the linguofacial trunk and matching veins. The linguofacial trunk drains into the linguofacial vein. The lingual muscles are innervated by the hypoglossal nerve (XII), and the sensory supply is from the lingual and glossopharyngeal (IX) nerves.

EQUINE DENTITION The formula for the deciduous teeth of the horse is 2 times I3-3 C0-0 P3-3 for a total of 24. The permanent dental formula is 2 times I3-3 C11 P3-3 or P4-3 M3-3 for a total of 40 or 42. In the mare the canine teeth are usually small or do not erupt, hence reducing the number to 36 or 38. The rst premolar tooth (wolf tooth) is often absent and has been reported as occurring in only 20% of the upper dentition of Thoroughbreds.3 The teeth of the horse are complex in shape and are compounded of di erent materials (dentin, cementum, and enamel). They function as grinding blades to masticate and macerate cellulose food in the important rst stage of the digestive process. The cheek teeth in the horse are a well-documented feature of the evolution of Equus caballus.

Deciduous Teeth The rst incisor is present at birth or the rst week of life. The second incisor erupts at 4 to 6 weeks of age; the third incisor, at 6 to 9 months of age; the first and second premolars, at birth to 2 weeks of age; and the third premolar, 3 months of age.

Permanent Teeth The eruption times for the permanent teeth are as follows: rst incisor, 2 ½ years of age; second incisor, 3 ½ years of age; third incisor, 4 ½ years of age; the canine tooth, 4 to 5 years of age; the rst premolar (wolf tooth), 5 to 6 months of age; the second premolar, 2½ years of age; the third premolar, 3 years of age; the fourth premolar, 4 years of age; the rst molar, 10 to 12 months of age; the second molar, 2 years of age; and the third molar, 3½ to 4 years of age. This eruption sequence clearly indicates that the eruption of the second and third permanent premolar teeth gives the potential for dental impaction. The modern horse has six incisor teeth in each jaw that are placed close together so that the labile edges form a semicircle. The occlusal surface has a deep enamel invagin*tion (infundibulum) that is lled only partially with cementum. As the incisor teeth wear, a characteristic pattern forms in which the infundibulum is surrounded by rings of enamel, dentin, enamel, and crown cementum in a concentric pattern. Each incisor tooth tapers from a broad crown to a narrow root so that as the midportion of the incisor is exposed to wear, the cross-sectional diameters are about equal; that is, at 14 years of age, the central incisor tooth of the horse has an occlusal surface that is an equilateral triangle. Observations on the state of eruption, the angles of incidence of the incisor teeth, and the pattern of the occlusal surfaces are used as guides for aging of horses. The canine teeth are simple teeth without complex crowns and are curved. The crown is compressed and is smooth on its labial aspect but carries two ridges on its lingual aspect. No occlusal contact occurs between the upper and lower canine teeth. When erupted, the six cheek teeth of the horse function as a single unit in the mastication of food. Each arcade consists of three premolar and three molar teeth. The maxillary arcade is slightly curved, and the teeth have a square occlusal surface. The occlusal surfaces of the mandibular teeth are more oblong, and each arcade is straighter. The horse is anisognathic; that is, the distance between the mandibular teeth is narrower (one third) than the distance between the upper cheek teeth. This anatomic arrangement a ects the inclination of the dental arcade as the jaws slide across each other in the food preparation process. The unworn upper cheek tooth presents a surface with two undulating and narrow ridges, one of which is lateral and the other medial. On the rostral and lingual side of the medial style is an extra hillock. The central portion of these surfaces is indented by two depressions that are comparable with, but much deeper than, the infundibula of the incisor teeth. When the teeth have been subjected to wear, the enamel that closed the ridges is worn through and the underlying dentin appears on the surface. Thus after a time the chewing surface displays a complicated pattern that may be likened to the outline of an ornate letter B, the upright stroke of the B being on the lingual aspect. Dentin supports the enamel internally, cementum supports the enamel lakes, and the peripheral cementum lls in the spaces between the teeth so that all six teeth may function as a single unit (i.e., the dental arcade). Transverse ridges cross each tooth so that the whole maxillary arcade consists of a serrated edge. The serrations are formed so that a valley is present at the area of contact with adjacent teeth. These serrations match tting serrations on the mandibular arcade. One should note that the mediolateral mandibular motion that occurs while the horse is chewing pellets does not provide full occlusal contact as it does when the horse is chewing hay.4 The true roots of the cheek teeth are short compared with the total length of the tooth. Cheek teeth have three roots: two small lateral roots and one large medial root. By custom, that portion of the crown embedded within the dental alveolus is referred to as the reserve crown, and the term root is con ned to the area of the tooth that is comparatively short and enamel free. Wear on the tooth gradually

exposes the reserve crown, and the roots lengthen. In an adult 1000-lb horse, the maxillary cheek teeth are between 8 and 8.5 cm in length. Dental wear accounts for erosion and loss of tooth substance at a rate of 2 mm annually. The pulp chambers of the teeth are also complex. The incisors and canines have a single pulp chamber. The mandibular cheek teeth have two roots and two separate pulp chambers. The maxillary cheek teeth, although they have three roots, have in fact five pulp chambers. As occlusal wear proceeds, deposition of secondary dentin within the pulp chambers protects the chambers (e.g., the dental star, medial to the infundibulum on the incisor teeth). In the mandibular cheek teeth, the transverse folding of the enamel anlage (during morphogenesis of the tooth) does not take place, and the occlusal surface is a simple surface of central dentin surrounded by enamel. Each tooth then is conformed to a single arcade by the presence of peripheral crown cementum.

MOUTH DISEASES The oral cavity and oropharynx are subject to a variety of diseases. However, many conditions a ecting the rst portion of the alimentary system produce the same clinical signs, regardless of their cause. The clinical signs may include inappetence or reluctance to eat, pain on eating or swallowing, oral swelling, oral discharge, and fetid breath. A ected animals may show some interest in food but hesitate to eat it. Salivation may be excessive and contaminated with purulent exudate or blood. The occurrence of bruxism (i.e., grinding of teeth) can indicate discomfort in other areas of the alimentary tract; for example, bruxism and frothing oral saliva are characteristic features of gastric ulceration in the foal. The clinician should be aware that considerable weight loss can occur rapidly with the inability to feed and swallow. Diseases that result in denervation of the pharynx and inappropriate swallowing can have the complication of aspiration pneumonia.

Examination and Clinical Signs After performing a complete physical examination and ascertaining the history, the clinician should approach examination of the mouth systematically in all cases. One can examine a considerable portion of the mouth and teeth from the outside by palpation of the structures through the folds of the cheek. Most horses allow a cursory oral examination without sedation or the use of an oral speculum. In many cases, however, a detailed oral examination is best achieved by sedation and the use of an oral speculum and a light source. The clinician should irrigate the mouth to wash out retained food material before inspecting and palpating the lips, cheeks, teeth, and gums. The classic signs of dental disease in the horse include difficulty and slowness in feeding, with a progressive unthriftiness and loss of body condition. In some instances the horse may quid (i.e., drop poorly masticated food boluses from the mouth), and halitosis may be obvious. Additional problems reported by owners include bitting and riding problems and headshaking or head shyness. Facial or mandibular swelling may occur. Nasal discharge can result from dental disease associated with maxillary sinus empyema. Mandibular stulae frequently are caused by lower cheek tooth apical infections. Some correlation exists between the age of the animal and clinical signs (Table 15-5). TABLE 15-5 Correlation Between Dental Disease and Dental Therapy Age 2-3 years

Examine for 1. First premolar vestige (wolf teeth)

1. Remove wolf teeth if present.

2. First deciduous premolar (upper and lower)

2 . Remove deciduous teeth if ready; if not, le o corners and points of premolars

3. Hard swelling on ventral surface of mandible beneath first premolar 4. Cuts or abrasions on inside of cheek in region of the second premolars and molars 3-4 years

4-5 years

5 years and older

Necessary Dentistry

3. Obtain X-ray film; extract retained temporary premolar if present 4. Lightly float or dress all molars and premolars if necessary

5. Sharp protuberances on all premolars and molars

5. Rasp protuberances down to level of other teeth in arcade

1. (1), (2), (4), and (5) above

1. (1), (2), (4), and (5) above

2. Second deciduous premolar (upper and lower)

2. Remove if present and ready

1. (1), (4), and (5) above

1. (1), (4), and (5) above

2. Third deciduous premolar

2. Remove if present and ready

1. (1), (4), and (5) above

1. (1), (4), and (5) above

2. Uneven growth and “wavy” arcade

2. Straighten if interfering with mastication

3. Unusually long molars and premolars

3. Unusually long molars and premolars may have to be cut if they cannot be

filed down From Baker GJ: Diseases of the teeth. In Colohan PT, Mayhew IG, Merritt AM, et al, editors: Equine medicine and surgery, ed 4, vol 1, Goleta, Calif, 1991, American Veterinary Publications.

Ancillary Diagnostic Techniques Ancillary aids for a complete examination of the oral cavity of the horse include radiology, endoscopic examination, uoroscopy, biopsy, and culture. The clinician always should take care during endoscopic evaluation of the oral cavity using a exible endoscope. The author recommends sedation and the use of an oral speculum to prevent inadvertent mastication of the endoscope. If general anesthesia is used as part of the diagnostic workup, then endoscopic evaluation of the oral cavity is much easier. In selected cases advanced imaging technologies such as computed tomography, magnetic resonance imaging, or nuclear scintigraphy may be beneficial.

DYSPHAGIA Clinical Signs and Diagnosis The lips of the horse are mobile and prehensile. In many ways they function like the tip of the elephant’s trunk in that they test, manipulate, and sample the environment for potential nutritive value. Consequently, loss of motor function (e.g., facial palsy) a ects the e ciency of the prehensile system. The lips grasp food in grazing or browsing, and the incisor teeth section the food. With mastication and lubrication with saliva, the bolus of food forms and is manipulated from side to side across the mouth, assisted by the tight cheeks of the horse and the palatine ridges. Swallowing begins as the food bolus contacts the base of the tongue and the pharyngeal walls. During swallowing the soft palate elevates to close the nasopharynx, the base of the tongue elevates, and the hyoid bone and the larynx move rostrally following contraction of the hyoid muscles. During this process the rima glottidis closes, and the epiglottis tilts dorsally and caudally to protect the airway so that food is swept through lateral food channels around the sides of the larynx into the laryngoesophagus. Fluoroscopic studies in nursing foals in the dorsoventral view showed that contact occurs between the lateral food channels in the midline so that in outline the food bolus achieves a bow-tie shape.5 Dysphagia is de ned as the di culty or inability to swallow. Anatomic classi cations for dysphagia include prepharyngeal, pharyngeal, and esophageal (postpharyngeal) dysphagias. The site of the cause for dysphagia in uences the clinical signs. Prepharyngeal dysphagia is characterized by dropping food (quidding) or water from the mouth, reluctance to chew, hypersalivation, or abnormalities in prehension. Pharyngeal and esophageal dysphagias are characterized by coughing; nasal discharge containing saliva, water, or food material; gagging; anxiousness; and neck extension during attempts to swallow. The following section describes esophageal dysphagia in more detail. Causes of dysphagia can be divided into four types: painful, muscular, neurologic, or obstructive (Table 15-6). Pain and obstruction cause dysphagia by interfering with the mechanics of prehension, bolus formation and transfer to the pharynx, and deglutition. Muscular and neurologic causes of dysphagia impede prehension and swallowing by a ecting the motor function of the lingual or buccal musculature, muscles of mastication (temporal and masseters), and pharyngeal and cranial esophageal muscles. Sensory loss to the lips, buccal mucous membranes, pharynx, or tongue also may cause dysphagia. Neurologic causes of dysphagia may a ect the forebrain, brainstem, or peripheral nerves that control prehension (cranial nerves Vm, Vs, VII, and XII), transfer of the food bolus to the pharynx (cranial nerves Vs and XII), and swallowing (cranial nerves IX and X). The latter point was classical thinking, but recent evidence suggests that although stimulation of cranial nerve IX stimulates swallowing, bilateral blockade of that nerve is not necessary for normal swallowing of either liquid or solid material.6 Table 15-6 Differential Diagnoses for Dysphagia Class of Dysphagia Painful

Differential Diagnoses Tooth root abscess or periodontal disease Broken teeth Abnormal dentition or wear Stomatitis, glossitis, or pharyngitis Nonsteroidal anti-inflammatory drug toxicity Chemical irritation Thrush (candidiasis) Influenza Streptococcus equi sub sp. equi Vesicular stomatitis virus Actinobacillus lignieresii Buccal, gingival, or glossal trauma (bits or chains)

Foreign bodies Retropharyngeal lymphadenopathy or abscess Mandibular trauma Temposohyoid osteoarthropathy Temporomandibular osteopathy Muscular

Hyperkalemic periodic paralysis Nutritional myopathy (white muscle disease) Polysaccharide storage disease Glycogen branching enzyme deficiency Masseter myositis Hypocalcemia tetany or eclampsia Myotonia Rectus capitis ventralis rupture White snakeroot toxicity Megaesophagus

Obstructive

Retropharyngeal abscess and lymphadenopathy Oral, pharyngeal, retropharyngeal, laryngeal, or esophageal malformations, injury, edema, or neoplasia Pharyngeal or epiglottic cysts Pharyngeal abscess or foreign body Dorsal displacement of the soft palate Rostral displacement of the palatopharyngeal arch Cleft palate

Neurologic forebrain disease; generalized neuropathy; disorders of cranial nerves V, VII, IX, X, or XII

Guttural pouch tympany or empyema Follicular pharyngitis Esophageal obstruction Pharyngeal cicatrix Retropharyngeal abscess or neoplasia Guttural pouch empyema, mycosis, or neoplasia Stylohyoid osteopathy Lead poisoning Petrous temporal bone osteomyelitis or fracture Retropharyngeal abscess Botulism Yellow star thistle toxicity Viral encephalitis Cerebral edema Cerebral or brainstem hemorrhage Intracranial masses (hematoma, neoplasia, abscess) Meningitis

Verminous encephalitis Equine protozoal myeloencephalitis Equine herpesvirus 1 Equine dysautonomia Hepatoencephalopathy Tetanus Polyneuritis equi

Diagnosis of the cause of dysphagia is based on physical examination, including a careful oral examination; neurologic examination; clinical signs; and endoscopy of the pharynx, esophagus, and guttural pouches. Radiology may be useful to assess the bony structures of the head and throat. Ultrasonography is valuable for examining the retropharyngeal space and esophagus to detect and evaluate masses. One may detect pharyngeal or esophageal causes of dysphagia with routine endoscopic examination or with contrast radiography. Although endoscopy can also be used to assess deglutition, it is important to remember that sedation adversely a ects the deglutition mechanism. One may assess deglutition using uoroscopy7 or manometry,8 but these techniques require specialized equipment. Speci c diagnostic procedures for nonalimentary causes of dysphagia are covered elsewhere in this text (see Chapter 3).

Management Speci c treatments aimed at resolving the underlying disorder causing dysphagia are discussed in detail elsewhere. Most horses with dysphagia should not eat roughage with long ber length (hay or grass). Dietary modi cations that promote swallowing, such as feeding slurries made from complete pelleted feeds, may be su cient to manage some cases of partial dysphagia. Care must be taken to prevent or avoid aspiration pneumonia in horses with pharyngeal or esophageal dysphagia. A ected foals can be managed by feeding mare’s milk or a suitable substitute through a nasogastric tube. The clinician also may administer pellet slurries or formulated liquid diets through a nasogastric tube to older horses. Prolonged nutritional management of dysphagic horses may require extra-oral feeding using a tube placed through an esophagostomy.9 Formulated pelleted diets are often easy to administer through a tube as slurry and are balanced to meet the nutritional requirements for healthy horses. Su cient quantities must be fed to deliver adequate calories (16 to 17 Mcal/day for a 500-kg horse). Adjustments may be necessary for horses that are cachectic or have extra metabolic demand (e.g., pregnancy). Adding corn oil to the ration (1 cup every 12 or 24 hours) is a common method of increasing fed calories. Liquid diets also have been used for enteral feeding10 but may not be tolerated as well as pelleted diets. Regardless of the method of nutritional management, the clinician must monitor and replace salivary losses of electrolytes. Saliva contains high concentrations of Na, K, and Cl. A group of ponies with experimental esophagostomies11 and a horse with esophageal squamous cell carcinoma12 were fed a complete pelleted diet through esophagostomy tubes but developed metabolic acidosis, hyponatremia, and hypochloremia, apparently because of salivary losses. Surprisingly, salivary losses of potassium did not result in hypokalemia in these cases, presumably because of replacement in the diet. However, if the diet is de cient in potassium, hypokalemia may result. Electrolyte replacement often can be accomplished by adding NaCl and KCl to the diet. Horses can be maintained for months with frequent feedings through an esophagostomy tube.12 Parenteral nutrition (total or partial) may be useful in the short term but is not often feasible for longterm management.

DENTAL DISEASES Eruption Disorders Tooth eruption is a complex phenomenon involving the interplay of dental morphogenesis and those vascular forces responsible for creating the eruption pathway. These changes are responsible for osteitis and bone remodeling within the maxilla and mandible. Young horses frequently show symmetric bony swelling resulting from these eruption cysts. In some cases additional clinical signs of nasal obstruction with respiratory stridor or nasal discharges may be apparent. Pathologic problems associated with maleruption include a variety of dental diseases.13 Oral trauma can displace or damage erupting teeth or the permanent tooth buds. As a result, teeth may be displaced and erupt in abnormal positions or may have abnormal shapes. Supernumerary teeth, incisors and molars, can develop as well as palatal displacement of impacted teeth (maxillary P3-3, or third cheek tooth). In almost all of these conditions, some form of surgical treatment is necessary, but depending on the number and location, conservative theapy can be successful.14

Signi cant evidence from the location of apical osteitis in diseased teeth con rms that dental impaction is a major cause of dental disease in the horse. In a series of 142 extracted teeth, 63 were P3-3 or P4-4 (cheek tooth 2 or 3, respectively).15 Early observations had indicated that the rst molar (M1, or cheek tooth 4) was the most commonly diseased tooth, and an “open infundibulum” in this tooth has been suggested as the cause.16 In a later study the mandibular cheek teeth 2 and 3 were the most commonly a ected, whereas cheek teeth 2 and 4 were most commonly a ected in the maxillary arcade, which was the more commonly a ected arcade.17 Studies on cementogenesis of the maxillary cheek teeth have shown, however, that in fact most maxillary cheek teeth have a greater or lesser degree of hypoplasia of cementum within the enamel lakes and that this “lesion” rarely expands into the pulp. The central infundibular hole is the site of its vascular supply to the unerupted cement lake. On those occasions in which caries of cementum occurs (i.e., secondary in ammatory disease and acid necrosis of the cementum), apical osteitis may develop.

Dental Decay Pulpitis is key to the pathogenesis of dental decay in the horse. The initiation of in ammatory pulp changes may be a sequela to dental impaction or dental caries or may result from fracture of a tooth. If the onset of the in ammatory process is slow, then formation of secondary dentin within the pulp chambers may protect the pulp and the tooth. Secondary dentin formation occurs from stimulation of odontoblasts within the pulp chamber. Such changes are the normal process of protection during dental wear and attrition as crown substances wear away and the reserve crown comes into wear. In acute disease, however, this defense mechanism is ine ective, and the changes that occur and that are sequelae to pulpitis re ect the location of each a ected tooth. For example, pulpitis and apical osteitis of the third mandibular cheek tooth most commonly results in the development of a mandibular dental stula. Pulpitis of the third maxillary cheek tooth, however, results in an in ammatory disease within the rostral maxillary sinus and in development of chronic maxillary sinus empyema. Oblique radiographs greatly assist the diagnosis of dental decay by demonstrating sinus tract formation, sequestration of bone, mandibular osteitis, hyperplasia of cementum, and new bone formation (so-called alveolar periosteitis).18 Nuclear scintigraphy and computed tomography can aid in an accurate diangosis.19,20 The management of dental decay in the horse usually involves surgical extraction of the diseased tooth. In some cases one can use apicoectomy and retrograde endodontic techniques to save the diseased tooth. The clinician must take care, however, in selection of patients. In most cases of apical osteitis in the horse that result from dental impaction, immature root structures make achieving an apical seal of the exposed pulp difficult.

PERIODONTAL DISEASE Gingival hyperemia and in ammation occur during the eruption of the permanent teeth and are common causes of a sore mouth in young horses (particularly 3-year-olds as the rst dental caps loosen). Such periodontal changes usually resolve as the permanent dental arcade is established. During normal mastication the shearing forces generated by the occlusal contact of the cheek teeth essentially clean the teeth of plaque and e ectively inhibit deposition of dental calculus. Wherever occlusal contact is ine ective, periodontal changes and calculus buildup occur; for example, the deposition of calculus on the canine teeth of mature geldings and stallions is common. Routine dental prophylaxis forms an important component of maintaining normal occlusal contact, and for this reason arcade irregularities that result in enamel point formation on the buccal edges of the maxillary cheek teeth and the lingual edges of the mandibular cheek teeth should be removed. The clinician should remove these edges annually in horses that are at grass and twice yearly in young horses, aged horses, and stabled horses. Horses at grass have been shown to have a greater range of occlusal contact and therefore better periodontal hygiene than stabled horses. In stabled horses the range of occlusal contact is narrower, and the formation of enamel points occurs more frequently with subsequent buccal ulceration and the initiation of a cycle of altered occlusal contact and hence irregular arcade formation. This process leads to severe forms of periodontal disease and wave mouth formation. Periodontal disease occurs with abnormal occlusal contact and initiation of the cycle of irregular wear and abnormal contact. Such changes progress to loss of alveolar bone, gross periodontal sepsis, and loss of tooth support. In this sense periodontal disease truly is the scourge of the equine mouth and results in tooth loss.21

CONGENITAL AND DEVELOPMENTAL ABNORMALITIES Cleft Palate Palatine clefts may result from an inherited defect and are caused by failure of the transverse palatal folds to fuse in the oral cavity. Harelip accompanies few palatine clefts in the horse. The degree of palatine clefting depends on the stage at which interruption in the fusion of the palatopalatal folds occurs. Toxic or teratogenic effects are documented in other species, but little data are available in the horse.

In recent years treatment for repair of uncomplicated palatine defects has been recommended, but the prognosis is generally poor because of the considerable nursing care required and the high incidence of surgical failures. One should emphasize early surgery and the use of mandibular symphysiotomy in a ording surgical exposure. The combination of mandibular symphysiotomy and transhyoid pharyngotomy to approach the caudal margins of the soft palate a ords surgical access, and mucosal aps can be constructed to repair the defects. However, the incidence of surgical breakdown is high, and healing by rst intention is the exception rather than the rule. A recent surgical report documented the successful closure of a median cleft of the lower lip and mandible in a donkey.22

Campylorhinus Lateralis Foals born with a severely deviated premaxilla and palate have a wry nose. Good functional and cosmetic outcome can be achieved with surgical correction.23 Circ*mstantial evidence indicates that such a defect has a genetic cause, and the defect occurs most frequently in the Arabian breed.

Cysts Other developmental abnormalities are subepiglottic cysts resulting from cystic distortion of remnants of the thyroglossal duct, which may cause dyspnea and choking in foals. Surgical removal of these cysts results in normal function.

Parrot Mouth The most signi cant developmental defect of dental origin is a maxilla that is longer than the mandible (i.e., parrot mouth). An overbite of 2 cm in the incisor arcade may be present in a horse with a mismatch of less than 1 cm between the rst upper and lower cheek teeth. Parrot mouth and monkey or sow mouth are thought to be inherited conditions. Some correction of minor incisor malocclusion occurs up to 5 years of age. Recognition and detection of parrot mouth are important in the examination of potential breeding stock. Surgical attempts to inhibit overgrowth of the premaxilla by wiring or by the application of dental bite plate procedures have been documented.24

Oral Wounds As has been indicated, the horse is by nature a curious animal that uses its lips as a means of exploring a variety of objects. Wounds of the lips, incisive bone, and the mandibular incisor area occur commonly in the horse and usually result from the horse getting the lips, jaw, or teeth caught in feeding buckets, in fence posts, or in halters or having a segment of tongue encircled with hair in tail chewing. As the horse panics and pulls away from its oral entrapment, considerable trauma can occur to the lips, teeth, and gums. Most wounds heal satisfactorily, provided one nds them early and observes the basic principles of wound hygiene, excision of necrotic tissue, and wound closure. One must ensure that oral mucosal defects are closed and that e ective oral seals are made before external wounds are closed. In some cases o ering specially constructed diets or even feeding the horse by nasogastric tube or esophagostomy during the healing processes may be necessary.

STOMATITIS AND GLOSSITIS Foreign body penetration of the tongue, cheek, or palate has been reported in grazing and browsing horses, particularly horses that have certain hay sources containing desiccated barley awns or yellow bristle grass.25 Other plant material and grass awns also occasionally may penetrate the tongue, gingiva, or cheek, causing in ammation or abscesses. Metallic foreign bodies have been reported in the tongue, and a history of feeding hay or the use of cable-framed tractor tires was often reported as part of the history.26 Ulcerative stomatitis also results from the toxicity of phenylbutazone therapy.27 Vesicular stomatitis is a highly contagious viral blistering disease described in more detail elsewhere. Treatment of glossitis and stomatitis primarily aims at removing the inciting cause. Actinobacillus lignieresii, the causative agent of actinobacillosis, has been isolated and identi ed from ulcers on the free border of the soft palate and oral and laryngeal granulomata. The bacterium also was reported in a sublingual caruncle in a horse with a greatly swollen tongue.28 Therapy with 150 ml of 20% sodium iodide and 5 g of ampicillin every 8 to 12 hours effected a clinical cure.

SALIVARY GLANDS Function Saliva is important for lubricating and softening food material. The horse has paired parotid, mandibular, and polystomatic sublingual salivary glands. The parotid gland is the largest of the salivary glands in the horse and is situated in the space between the ramus of the mandible and the wing of the atlas. The parotid duct is formed at the ventral part of the gland near the facial crest by the union of three or four smaller ducts. The duct leaves the gland above the linguofacial vein, crosses the tendon of the sternocephalicus muscle, and enters the mouth obliquely in the cheek opposite the third upper cheek tooth. The parotic duct ori ce is small, but some dilation of the duct and a

circular mucous fold (the parotid papillae) exist at this point. The mandibular gland is smaller than the parotid gland and extends from the atlantal fossa to the basihyoid bone. For the most part the mandibular gland is covered by the parotid gland and the lower jaw. The mandibular duct is formed by the union of a number of small duct radicles that emerge along the concave edge of the gland and run rostral to the border of the mouth opposite the canine tooth. The ori ce is at the end of a sublingual caruncle. The mandibular gland possesses serous, mucous, and mixed alveolar glandular components. The parotid gland is a compound alveolar serous gland. The parotid salivary gland can secrete saliva to yield rates of 50 ml/min, and a total daily parotid secretion can be as much as 12 L in a 500-kg horse. Parotid secretion occurs only during mastication, and administration of atropine or anesthesia of the oral mucosa can block secretion. Parotid saliva is hypotonic compared with plasma, but at high rates of flow, concentrations of sodium, chloride, and bicarbonate ions increase.

Salivary Gland Disorders Parotid saliva of the horse has a high concentration of calcium, and occasionally calculi (sialoliths) form within the duct radicles of the parotid salivary gland.29 Congenital parotid duct atresia, acquired stricture from trauma to the duct, or obstruction by plant material (e.g., sticks or foxtails and other seeds) also may occur. The clinical signs of sialolithiasis or other forms of ductule obstruction include a uid swelling in the form of a mucocele proximal to the stone and occasionally in ammation of the parotid gland. Ultrasonography is useful to diagnose salivary mucoceles and to detect foreign bodies or sialoliths. Measurement of electrolyte concentrations in aspirates from suspected mucoceles might be helpful to distinguish them from hematomas. Salivary potassium and calcium concentrations are higher than those in plasma. Treatment may require surgical removal of the stone or plant material in the case of sialolithiasis or foreign body obstructions. Other causes of obstruction may require resection of the affected portion of the duct or chemical ablation of the gland.30 Primary sialoadenitis is unusual but can occur in one or both glands. The condition is painful and may be associated with a fever and anorexia. Secondary sialoadenitis is more common and usually is associated with trauma. Infectious sialoadenitis resulting from Corynebacterium pseudotuberculosis31 or other bacterial pathogens also may occur. Diagnosis is by physical examination and identi cation of an enlarged edematous parotid gland tissue on ultrasonographic examination. Culture and cytologic examination of aspirates may be useful for diagnostic purposes. Treatment is usually palliative, consisting of NSAIDs. Appropriate antibiotic therapy is indicated as directed by culture and sensitivity results. Chemical irritation, glossitis, stomatitis, or other causes of prepharyngeal dysphagia cause ptyalism, or excessive salivation, in horses. Speci c therapy for the ptyalism usually is not required as long as salivary losses are not excessive, resulting in dehydration and electrolyte imbalances. Ingestion of the fungal toxin slaframine also causes hypersalivation in horses.32 The fungus Rhizoctonia leguminicola, which produces slaframine, causes black patch disease in red clover. Slaframine is a parasympathomimetic compound that stimulates exocrine secretion in the parotid gland. Slaframine toxicosis most commonly occurs in the spring or early summer and rarely requires treatment other than removal from the pasture. Mowing removes the source in most cases because regrowth in pastures often has less fungal contamination.33

ESOPHAGEAL DISEASES L. Chris Sanchez

ANATOMY AND FUNCTION The esophagus is a musculomembranous tube that originates from the pharynx dorsal to the larynx and terminates at the cardia of the stomach.1 In adult Thoroughbreds the esophagus is approximately 120 cm long. The cervical portion is approximately 70 cm long; the thoracic portion, approximately 50 cm long; and the short abdominal portion, only approximately 2 cm long. The cervical esophagus generally lies dorsal and to the left of the trachea in the cervical region. In the thorax the esophagus courses through the mediastinum lying dorsal to the trachea and crosses to the right of the aortic arch dorsal to the heart base. The esophagus has no digestive or absorptive functions and serves as a conduit to the stomach for food, water, and salivary secretions. The esophageal mucosa is a keratinized strati ed squamous epithelium.1 The submucosa contains elastic bers that contribute to the longitudinal folds of the esophagus and confer elasticity to the esophageal wall. A transition occurs in the muscle type composing the tunica muscularis from striated skeletal muscle in the proximal two thirds of the esophagus to smooth muscle in the distal third. In the proximal esophagus the skeletal muscle layers spiral across one another at angles. Within the smooth muscle layers of the distal esophagus the outer layer becomes more longitudinal, whereas the inner layer thickens and becomes circular. The wall of the terminal esophagus can be 1 to 2 cm thick. Deep cervical fascia, pleura, and peritoneum contribute to the thin brous tunica adventitia of the esophagus. Motor innervation to the striated skeletal muscle of the esophagus includes the pharyngeal and esophageal branches of the vagus nerve, which originate in the nucleus ambiguus of the medulla oblongata. Parasympathetic bers of the vagus nerve supply the smooth muscle of the distal esophagus.

Sympathetic innervation of the esophagus is minimal. Passage of ingesta through the esophagus can be considered part of the swallowing process, which consists of oral, pharyngeal, and esophageal stages. The oral stage is voluntary and involves transport of the food bolus from the mouth into the oropharynx. During the involuntary pharyngeal stage, the food bolus is forced through the momentarily relaxed upper esophageal sphincter by simultaneous contractions of the pharyngeal muscles. In the esophageal phase of swallowing, the upper esophageal sphincter closes immediately, the lower esophageal sphincter opens, and esophageal peristalsis propels the bolus into the stomach.2 Unlike a food bolus, liquids do not require peristalsis to reach the lower esophageal sphincter and may precede the food bolus during swallowing. The upper esophageal sphincter prevents esophagopharyngeal re ux during swallowing and air distention of the esophagus during inspiration. Upper esophageal pressure increases in response to pressure from a food bolus and to increased intraluminal acidity, as would occur with gastroesophageal re ux. The lower esophageal sphincter is a smooth muscle located at the gastroesophageal junction that is morphologically ill de ned but forms an e ective functional barrier.2 The lower esophageal sphincter is normally closed in response to gastric distention to restrict gastroesophageal re ux. Relaxation of the lower esophageal sphincter permits passage of ingested material from the esophagus to the stomach. Distention of the stomach with ingesta mechanically constricts the lower esophageal sphincter. Gastric distention also triggers a vagal re ex that increases lower esophageal sphincter tone, a safety mechanism against gastroesophageal re ux. The mechanical and vagal mechanisms that promote lower esophageal sphincter tone prevent spontaneous decompression of the stomach, which, along with a lack of a vomiting reflex in the horse, increases the risk of gastric rupture during episodes of severe distention. A wide variety of congenital and acquired disorders of the esophagus have been described in horses. These are summarized in Table 15-7 and discussed in detail in the next sections. TABLE 15-7 Esophageal Disorders of Horses

ESOPHAGEAL OBSTRUCTION Esophageal obstruction has many causes (Table 15-8) and most often is manifested clinically by impaction of food material and resulting esophageal dysphagia. Esophageal obstruction may be caused by primary impactions (simple choke) of roughage, particularly leafy alfalfa hay, coarse grass hay, bedding, and even grass.3 Prior esophageal trauma or poor mastication caused by dental abnormalities may predispose horses to primary esophageal impaction.16 Wol ng or gulping food may precipitate primary impactions, particularly if the horse is exhausted or mildly dehydrated after a long ride or is weakened from chronic debilitation. Impactions also may result from disorders that physically impede the passage of food material and uid by narrowing the luminal diameter, reduce the compliance of the esophageal wall, or alter the conformation of the esophageal wall such that food material accumulates in a pocket or diverticulum. Foreign bodies, intramural or extramural masses, or acquired or congenital anomalies cause these so-called secondary impactions. Intramural causes of esophageal

obstruction include tumors (squamous cell carcinoma), strictures, diverticula, and cysts.6,16,26,34,42,53-56 Mediastinal or cervical masses (tumors or abscesses) may cause extramural obstructions. Congenital anomalies are covered in detail later. TABLE 15-8 Causes of Complete or Partial Esophageal Obstruction in the Horse Category Intraluminal Extramural

Intramural

Functional disorders

Differential Foreign body Feed material Neoplasia Vascular ring anomaly Granuloma Esophageal abscess Granuloma Neoplasia Cysts Diverticula Stenosis Dehydration Exhaustion Pharmacologic Primary megaesophagus Esophagitis Autonomic dysautonomia Vagal neuropathies

Examples Apples, potatoes Squamous cell carcinoma, lymphoma Persistent right aortic arch

Squamous cell carcinoma, leiomyosarcoma Intramural cysts, duplication cysts

Acepromazine, detomidine Congenital ectasia

Clinical Signs and Diagnosis The clinician must perform a thorough physical examination, including a complete oral and neurologic examination, to help rule out causes of dysphagia and nasal discharge other than esophageal obstruction. The clinical signs associated with esophageal obstructions are related to dysphagia.57 Horses with esophageal obstruction are often anxious and stand with their neck extended. They may gag or retch, particularly with acute proximal obstructions. Bilateral frothy nasal discharge containing saliva, water, and food material; coughing; odynophagia; and ptyalism are characteristic clinical signs, the severity of which varies with the degree and location of the obstruction. Distention in the jugular furrow may be evident at the site of obstruction. Other clinical signs related to regurgitation of saliva, water, and food material may be observed, such as dehydration, electrolyte or acid-base imbalances, weight loss, and aspiration pneumonia. In extreme cases pressure necrosis from the impaction or trauma to the esophagus may cause esophageal rupture. If the rupture is in the cervical esophagus, crepitus or cellulitis may be evident along with signs of systemic in ammation. Thoracic auscultation is important to determine whether aspiration pneumonia is present. Intrathoracic esophageal rupture may result in pleuritis and its associated clinical signs. Passage of a nasogastric tube is an e ective way to detect and localize an obstruction but provides little information about the nature of the obstruction or the condition of the esophagus. The most direct method for diagnosis of esophageal obstructions is endoscopic examination. Most cases of esophageal obstruction occur at sites of natural narrowing of the esophageal lumen, such as the cervical esophagus, the thoracic inlet, base of the heart, or the terminal esophagus; thus an endoscope longer than 1 m may be necessary for complete evaluation. Endoscopic evaluation is useful before relief of an impaction to localize the obstruction and to investigate the nature of the impaction if a foreign body is suspected. Foreign bodies may be retrievable through transendoscopic tethering.58 The veterinarian can obtain critical diagnostic and prognostic information after resolution of the impaction. Assessing the a ected esophagus for mucosal ulceration, rupture, masses, strictures, diverticula, and signs of functional abnormalities is important (Figures 15-11 and 15-12).

FIGURE 15-11

Endoscopic view of the cervical esophagus in an adult horse 6 months after an episode of choke that caused circumferential ulceration of the esophageal mucosa. The area of luminal narrowing (stricture) is at the upper right of the image, and the proximal dilation forms an outpouching of the esophageal wall. A contrast esophagram revealed that the outpouching was a pulsion diverticulum.

FIGURE 15-12

Endoscopic view of the proximal esophagus in a yearling filly with recurrent esophageal obstruction. Circumferential mucosal ulceration is evident proximal to incomplete stricture formation.

Ultrasonography of the cervical region is useful not only to con rm a cervical esophageal impaction but also to provide critical information about the location and extent of the impaction and esophageal wall thickness and integrity. Ultrasonography may provide information about the cause.35 Radiographic assessment of the esophagus can con rm the presence of esophageal obstruction when the a ected area cannot be adequately viewed using endoscopy. Impacted food material in the esophagus can be detected by its typical granular pattern, and gas accumulation is often visible proximal to the obstruction. Air or barium contrast radiographic studies are most useful for evaluating the esophagus after relief of the impaction if a stricture is suspected. It is often easier to detect esophageal dilation, diverticula, rupture, functional disorder (megaesophagus), or luminal narrowing caused by extraluminal compression by using contrast radiographic studies instead of endoscopy (Figure 15-13).59-61 One should take care when interpreting radiographic studies in sedated horses, particularly after passage of a nasogastric tube or other esophageal manipulations that may contribute to esophageal dilation.62

FIGURE 15-13

Contrast esophagram in a horse with circumferential esophageal stricture (arrow) and a pulsion diverticulum proximal to the stricture.

Treatment The primary goal of treatment for esophageal impaction is to relieve the obstruction. A variety of approaches have been described, ranging from minimal conservative therapy to aggressive intervention. Parenteral administration of acepromazine (0.05 mg/kg intravenously), xylazine (0.25 to 0.5 mg/kg intravenously) or detomidine (0.01 to 0.02 mg/kg intravenously), oxytocin (0.11 to 0.22 IU/kg intramuscularly), and esophageal instillation of lidocaine (30 to 60 ml of 1% lidocaine) may reduce esophageal spasms caused by pain or may decrease esophageal tone.4,62,63 Some clinicians advocate parasympatholytic drugs such as atropine (0.02 mg/kg intravenously) to reduce salivary secretions and lessen the risk of aspiration. However, undesirable e ects of atropine, including excessive drying of the impaction and inhibition of distal gastrointestinal motility, may preclude its use. In many horses with esophageal obstruction secondary to a feed impaction, the problem may be resolved with conservative management. To facilitate examination, relieve anxiety, and relax the esophagus, the clinician should begin by sedating the horse. Then, the clinician passes a stomach tube to con rm the diagnosis. When an obstruction is encountered, gentle pressure is applied in an attempt to dislodge and move distally the o ending feed material. If the choke is not easily dislodged with gentle pressure, the tube is removed. Appropriate sedatives (e.g., xylazine, detomidine, acepromazine), anti-in ammatory-analgesic (e.g. unixin, butorphanol), and smooth muscle-relaxing drugs (e.g., oxytocin) are administered. The horse is moved to an unbedded stall with absolutely no food or water within reach and is muzzled if necessary. The horse is left alone in the stall for several hours. If there is evidence of dehydration, appropriate intravenous uid therapy is provided. If there is evidence of aspiration pneumonia, appropriate intravenous antimicrobial therapy is provided. Upon re-examination, a stomach tube is passed and gentle pressure is again applied to the area of obstruction. In most horses the impaction will have softened and can be easily dislodged, if it has not resolved already, with minimal pressure from the stomach tube. Some clinicians prefer a more interventional approach to resolution of an impaction, and some impactions are severe enough to require physical dispersal of the material.4 A nasogastric tube can be used to displace the impacted material along with external massage if the obstruction is in the cervical region. Often, carefully lavaging the esophagus with water using an uncu ed or a cu ed nasogastric tube while the head is lowered is necessary to aid in breaking up the impaction. Some clinicians advocate a dual tube method, whereby a tube is placed through each nasal passage into the esophagus for ingress and egress of the lavage uid. Because of the risk of aspiration of water and food material, esophageal lavage sometimes is done under general anesthesia with a cuffed nasotracheal tube. In refractory cases intravenous administration of isotonic uid containing 0.9% NaCl and KCl (10 to 20 mEq/L) for 24 hours at a rate of 50 to 100 ml/kg/day along with esophageal relaxants such as oxytocin may promote hydration and softening of the impaction and help prevent or alleviate any electrolyte or acid-base imbalances resulting from salivary losses of chloride, sodium, and potassium.64 Oxytocin may or may not provide a direct e ect for the resolution of an esophageal obstruction. Oxytocin reduced area under the curve for esophageal smooth muscle strip contractions but had no e ect on skeletal muscle strips in vitro.65 In one in vivo report, oxytocin administration resulted in decreased esophageal tone in the proximal esophagus (aborad to larynx and thoracic inlet).63 However, in another study oxytocin administration did not a ect esophageal manometric recordings.66 Rarely, esophageal obstruction ultimately may require esophagotomy to relieve the impaction. Strict restriction of food and water, including access to bedding material, must be enforced until the obstruction is resolved and the esophagus has regained function. Surgical removal of esophageal foreign bodies can be considered if the size or orientation of the object is such that transendoscopic retrieval is considered unlikely to be successful.6,7 In one report an intraluminal mass composed of exuberant granulation tissue was removed through serial transendoscopic ND:Yag laser ablation.36 Systemic e ects of dysphagia associated with esophageal impaction include dehydration, hyponatremia, hypochloremia, and metabolic alkalosis from prolonged loss of salivary free water and electrolytes.64 If the duration of a complete esophageal obstruction is 48 hours or longer, the clinician should correct dehydration and electrolyte and acid-base imbalances. One can restore uid and electrolyte balance with oral electrolyte solutions if the patient is less than 6% to 7% dehydrated and the esophageal obstruction is resolved. In horses that are greater than 6% to 7% dehydrated or those that have a refractory obstruction or moderate to severe electrolyte imbalances, intravenous uid therapy with solutions containing 0.9% NaCl and KCl (10 to 20 mEq/L) may be required. The clinician should perform esophageal endoscopy after relief of the impaction to determine whether any complications of the impaction have developed or if a primary cause of the obstruction is present. Endoscopic examination is critical to determine the postobstruction treatment plan and for follow-up evaluation of esophageal healing. One should re-evaluate the horse every 2 to 4 weeks after resolution of the impaction if esophageal dilation or mucosal injury is evident. Additional evaluation using radiography may be warranted to assess motility and transit times. Dilation proximal to the site of obstruction, mucosal injury from trauma, stricture formation, formation of a diverticulum, megaesophagus, and esophagitis are sequelae to esophageal obstruction that predispose patients to re-obstruction. Underlying functional or morphologic abnormalities were much more likely in cases of recurrent, relative to first time, obstruction in a retrospective study.3

The rate of re-obstruction may be as high as 37%. Depending on the duration of the obstruction and the degree of trauma or dilation, the risk of re-obstruction is high for 24 to 48 hours or longer; thus food should be withheld for at least 24 to 48 hours after resolution of the obstruction. Sucralfate (20 mg/kg orally every 6 hours) may hasten healing if esophageal ulceration is evident, but the e cacy of sucralfate for this purpose is not established. Some clinicians suggest that administration of an NSAID such as unixin meglumine (1 mg/kg orally or intravenously every 12 hours) or phenylbutazone (1 to 2 mg/kg orally or intravenously every 12 to 24 hours) for 2 to 4 weeks after resolution of the impaction may reduce the development of strictures. Judicious use of NSAIDs is recommended to prevent NSAID-induced worsening of esophageal mucosal injury. Orally administered NSAIDs are not recommended if esophagitis is present. After 48 to 72 hours or when the esophageal mucosa has recovered as assessed by endoscopy, the horse can be fed soft food (e.g., moistened pellets and bran mashes). One can return the patient gradually to a high-quality roughage diet over 7 to 21 days, depending on the degree of esophageal damage induced by the impaction and the nature of any underlying disease. The prognosis for survival is good (78% to 88%), but some horses may require permanent dietary modification if chronic obstruction persists.3,16 Aspiration pneumonia is a potential complication in every case of esophageal obstruction, and perforation is possible with severe or prolonged obstruction. In one report duration of obstruction before presentation was a good predictor of pneumonia, whereas endoscopic evidence of tracheal food contamination was not.3 Thus administration of broad-spectrum antibiotics that are e ective against gram positive and gram negative organisms, including metronidazole for anaerobes, is highly recommended if the duration of obstruction is either unknown or prolonged or if aspiration is at all suspected. A subsequent section describes treatment of esophageal perforation or rupture.

ESOPHAGITIS Esophagitis refers to a clinical syndrome of esophageal in ammation that may or may not be ulcerative. The major protective mechanisms of the esophageal mucosa include salivary and food material bu ers, normal peristaltic motility, and the barrier formed by the gastroesophageal sphincter. Re ux esophagitis is caused by repeated episodes of gastric uid regurgitation into the distal esophagus and subsequent chemical injury to the mucosa.67 Esophageal mucosal ulceration also can occur if the clearance of gastric uid from the esophagus is delayed, such as in functional disorders of the esophagus. Like ulceration of the squamous portion of the stomach in horses, gastric acid and bile salt chemical injury is a major mechanism of esophageal squamous epithelial ulceration.67,68 Re ux esophagitis may occur along with gastric ulcer disease, motility disorders, increased gastric volume from gastric out ow obstructions, gastric paresis, intestinal ileus, or impaired lower esophageal sphincter function.55,67 Other causes of esophagitis in horses include trauma (e.g., foreign bodies, food impactions, nasogastric tubes), infection (e.g., mural abscesses), or chemical injury (e.g., pharmaceuticals, cantharidin) (Figure 15-14).7,8,30,69

FIGURE 15-14 mucosa is notable.

Endoscopic view of the cervical esophagus of a horse that had repeated passages of a sti nasogastric tube. The deep, linear ulceration of the esophageal

Clinical Signs and Diagnosis The clinical signs of esophagitis are nonspeci c and similar to those of esophageal obstruction and gastric ulceration. Gagging or discomfort when swallowing may be evident, along with hypersalivation and bruxism. Esophageal (postpharyngeal) dysphagia may be evident. Partial or complete anorexia may occur, such that horses with chronic esophagitis may have signi cant weight loss. Esophageal hypomotility dysfunction caused by the in ammatory process may result in esophageal impaction. Clinical signs of underlying diseases that predispose to esophagitis may predominate or mask the signs of esophagitis. Horses with gastrointestinal motility disorders such as proximal enteritis or gastric out ow obstruction are at a high risk of developing re ux esophagitis because of the presence of gastric acid and bile salts in the uid

reflux. Foals with obstructive gastroduodenal ulcer disease commonly have reflux esophagitis. Diagnosis requires endoscopic examination of the esophagus. Di use, patchy, linear, or coalescing erosion or ulcerations may be evident (see Figure 15-14), as well as signi cant edema or hyperemia. It is important to determine whether an underlying disease, such as infection, neoplasia, esophageal strictures, or diverticula, is present. In addition, the clinician must examine the stomach to determine whether the esophagitis is associated with gastritis, gastric out ow obstruction, or gastric ulcer disease. Contrast radiography may be helpful to detect esophageal ulceration and is useful to assess esophageal motility and transit time.60

Treatment The principles of therapy for re ux esophagitis include control of gastric acidity, mucosal protection, and correction of any underlying disorder contributing to gastroesophageal re ux. Reduction of gastric acid production with proton pump antagonists such as omeprazole or H2 histamine receptor blockers such as ranitidine is critical for resolution of esophagitis. Some clinicians advocate using sucralfate to promote healing of ulcerated esophageal mucosa. However, the ability of sucralfate to bind ulcerated esophageal mucosa and hasten esophageal ulcer healing has not been established. Horses with re ux esophagitis following delayed gastric out ow caused by gastroduodenal ulcer disease, gastric paresis, or proximal enteritis may bene t from prokinetic drugs that act on the proximal gastrointestinal tract. Metoclopramide (0.02 to 0.1 mg/kg subcutaneously every 4 to 12 hours) reduces gastroesophageal re ux by increasing lower esophageal sphincter tone, gastric emptying, and gastroduodenal coordination. The clinician should exercise caution when giving metoclopramide to horses because they are prone to extrapyramidal neurologic side e ects of the drug. Cholinergic drugs such as bethanechol (0.025 to 0.035 mg/kg subcutaneously every 4 to 24 hours or 0.035 to 0.045 mg/kg orally every 6 to 8 hours) may improve gastric emptying and are e ective for treating re ux esophagitis. The clinician should take care to rule out a physical obstruction before using prokinetic drugs. For esophagitis from trauma or pressure injury after esophageal impaction, judicious use of NSAIDs may be warranted to reduce esophageal inflammation and pain. Dietary modi cation may be necessary for patients with esophagitis, depending on the degree of ulceration or if motility is impaired. Horses with mild esophagitis should be fed frequent small meals of moistened pellets and fresh grass. Severe esophagitis may necessitate withholding food and imposing complete esophageal rest for several days. Although the prognosis for esophagitis is good in the absence of underlying disease, the risk of stricture formation is high if severe circumferential or coalescing ulcerations are present. Esophagitis from severe trauma or infection may lead to stricture formation.

MOTILITY DISORDERS Motility dysfunction of the equine esophagus most commonly is caused by hypomotility resulting in esophageal dilation (ectasia) or megaesophagus. Although megaesophagus in horses usually is acquired, reports indicate that idiopathic megaesophagus in young horses may be congenital.32,33,51,52,70 Acquired megaesophagus in horses may be a consequence of chronic or recurrent esophageal obstruction.16,55 Esophageal impactions of a short duration cause a proximal dilation of the esophagus that is generally reversible.60 However, if the duration of the obstruction is long enough, the motility of the esophagus proximal to the site of obstruction may be impaired permanently. Other causes of acquired megaesophagus include extra-esophageal obstruction by tumors or abscesses, pleuropneumonia, and vascular ring anomalies.16,42 A retrospective report of horses with megaesophagus reveals an overrepresentation of Friesian horses (14 of 18 cases), suggesting the possibility of a predisposition for esophageal disorders in this breed.32 The authors state that preliminary pedigree analyses suggest the possibility of a recessive pattern of inheritance.32 Megaesophagus also may result from neurologic, neuromuscular, and muscular disorders. Neurologic diseases that cause vagal neuropathy (e.g., equine protozoal myeloencephalitis, equine herpesvirus myeloencephalitis, idiopathic vagal neuropathy) have been associated with megaesophagus in horses. Pleuropneumonia may be associated with a vagal neuropathy resulting in megaesophagus. Megaesophagus is an early sign of equine dysautonomia71 and may be observable in patients with botulism. Myasthenia gravis is a wellknown cause of megaesophagus in nonequine species but has not been reported in horses. Also in other species electrolyte disorders, cachexia, primary myopathies, myositis, and Addison’s disease may a ect esophageal motility but have not been associated with megaesophagus in horses. One can induce iatrogenic megaesophagus by the α2-adrenergic agonist detomidine, but this is transient and reversible.62,72 Acepromazine, detomidine, and a combination of xylazine and butorphanol can alter proximal esophageal motility by disrupting coordinated peristalsis and decreasing spontaneous swallowing, but only acepromazine a ected the manometric pro le of the distal esophagus.66 None of these drugs altered contractility of isolated esophageal smooth or striated muscle strips in vitro.65 Nonetheless, the use of these drugs may complicate clinical evaluation of esophageal motility. Esophageal disorders, including megaesophagus, esophageal diverticulum, and esophageal rupture, appear to be more common in young Friesian horses, suggesting the possibility of a genetic predisposition in this breed.

Esophageal in ammation, particularly re ux esophagitis, may a ect motility and cause megaesophagus. However, because esophageal hypomotility a ects the tone and function of the lower esophageal sphincter, re ux esophagitis also may be a complication of a primary functional disorder. Thus assessing esophageal motility in horses with esophagitis that is not responding appropriately to treatment is important.

Clinical Signs and Diagnosis Along with a complete physical examination, the clinician should perform a careful neurologic examination to help rule out primary neurologic causes of megaesophagus. Because esophageal hypomotility is a functional obstruction, the clinical signs of esophageal hypomotility or megaesophagus are similar to those of esophageal obstruction. Unlike mechanical obstruction the onset of clinical signs is insidious rather than acute. The clinical signs include those associated with esophageal dysphagia.33,42,51,52,55,70 The cervical esophagus may be dilated enough to be evident externally. Weight loss is a common sign. Signs attributable to an underlying disease may be evident. Diagnosis of esophageal hypomotility requires transit studies. One can measure the transit time of a bolus from the cervical esophagus to the stomach by uoroscopy or contrast radiography.60,71 Other signs of esophageal hypomotility and megaesophagus include pooling of contrast material and an absence of peristaltic constrictions.55,60,70,71 Endoscopy may reveal a dilated esophagus and an absence of peristaltic waves.55,70 Evidence of underlying disease causing obstruction or esophageal dilation may be apparent.16,55 One should evaluate the esophagus for evidence of esophagitis that is causing esophageal motility dysfunction or is a result of impaired esophageal clearance of gastric uid. Esophageal manometry may be useful to document abnormal postdeglutition contraction pressures, contraction time, and propagation times but is not often available for routine clinical application.70,73 One should perform other diagnostic tests, such as a complete blood count and chemistry, to help determine a possible underlying cause. Cerebral spinal uid analysis may be indicated to rule out neurologic disorders. Specialized testing such as electromyography may also be indicated to detect neuromuscular disorders.

Treatment Treatment of esophageal hypomotility or megaesophagus should aim at treating the underlying cause. Dietary modi cation should aim at improving esophageal transit of food. The horse should be fed slurries of pellets from an elevated position to promote transit by gravity flow. Metoclopramide or bethanechol may bene t patients with re ux esophagitis associated with megaesophagus by increasing lower esophageal tone, gastric emptying, and reducing gastroesophageal re ux. The prognosis depends on the underlying cause and the degree of dilation. Although many cases of megaesophagus associated with re ux esophagitis respond well to treatment, many other forms of megaesophagus, including congenital megaesophagus, have a poor prognosis.

ESOPHAGEAL STRICTURE Strictures most commonly are caused by pressure necrosis from esophageal impactions that induce circumferential erosion or ulceration of the esophageal mucosa, although esophageal injury caused by oral administration of corrosive medicinal agents and trauma to the neck may also result in stricture formation.17 Congenital strictures also have been reported.74 Strictures caused by mucosal and submucosal trauma are termed esophageal webs or rings. Strictures may also originate in the muscular layers and adventitia of the esophagus (mural strictures) or in all of the layers of the esophagus (annular stenosis).18,74 Horses with these lesions have a presentation similar to those with simple obstructions, because strictures result in partial obstruction and impaction of food material in the lumen. Endoscopy can be used to detect esophageal webs or rings (see Figures 15-11 and 15-12), whereas identi cation of mural strictures or annular stenosis may require a doublecontrast esophogram (see Figure 15-13). In a retrospective study of horses with esophageal stricture following simple obstruction, maximal reduction in esophageal lumen diameter occurred within 30 days of the esophageal obstruction. Although surgery has been used to relieve such strictures, initial medical management is warranted because strictures may resolve with conservative therapy, and the esophagus continues to remodel for up to 60 days following ulceration. In one report seven horses with esophageal obstruction-induced stricture were treated conservatively by feeding a slurry diet and administering anti-in ammatory and antimicrobial medications, and ve of seven were clinically normal within 60 days.17 One of the ve successfully treated horses had a 10-cm area of circumferential ulceration, suggesting that the potential exists for extensive mucosal injury to resolve without permanent stricture formation. If resolution of strictures within 60 days is insu cient, the veterinarian should investigate other methods to increase esophageal diameter. Bougienage and balloon dilation have been used successfully in small animal patients and human beings and with variable results in horses.19-21The technique involves passage of a tubular dilatable instrument down the esophagus and stretching of the stricture. One may perform the technique by passing a nasogastric tube with an in atable cu . However, the procedure must be performed frequently to have any success, and horses may not tolerate it well.74 Alternatively, a number of surgical techniques have been used to resolve strictures, including resection and anastomosis,75,76 temporary esophagostomy with fenestration of the stricture,18 esophagomyotomy for strictures of

the muscularis and adventitia,22,77 and patch grafting with local musculature.23 However, such surgeries are fraught with complications, largely because of the propensity of the traumatized esophagus to restricture.16,17 The esophagus lacks a serosal layer and does not rapidly form a brin seal as does the remainder of the intestinal tract, so anastomoses tend to leak.76 In addition, tension on the esophagus during swallowing and movement of the neck impairs healing of anastomoses.18,75 In spite of these di culties, the long-term prognosis for horses with chronic esophageal strictures treated surgically is better than for those treated nonsurgically.16 Two recent reviews describe surgical approaches to the esophagus in detail.24,78

ESOPHAGEAL DIVERTICULA Two types of diverticula are traction (true) diverticula and pulsion (false) diverticula. Traction diverticula result from wounding and subsequent contraction of periesophageal tissues, with resultant tenting of the wall of the esophagus. Pulsion diverticula arise from protrusion of esophageal mucosa through defects in the muscular wall of the esophagus and usually result from trauma or acute changes in intraluminal pressure.74 Traction diverticula appear as a dilation with a broad neck on contrast esophagography, whereas pulsion diverticula typically have a ask shape with a small neck on an esophagram (see Figure 15-13).26,27 Although traction diverticula are usually asymptomatic and of little clinical significance, pulsion diverticula may fill with feed material, ultimately leading to esophageal obstruction and rupture.9,27,28 A movable mass in the midcervical region may be noticeable before onset of complete obstruction.74 Pulsion diverticula may be corrected surgically by inverting or resecting prolapsed mucosa and closing the defect in the wall of the esophagus.26-28 Inversion of excessive mucosa may reduce the diameter of the esophageal lumen and predispose horses to esophageal obstruction and therefore should be reserved for small diverticula.26

CONGENITAL DISORDERS Congenital disorders of the esophagus are rare. Reported congenital abnormalities include congenital stenosis,50 persistent right aortic arch,42-47 other vascular anomalies,48 esophageal duplication cysts,37,39,40 intramural inclusion cysts,41,56 and idiopathic megaesophagus.33,51,52 In the one report of congenital stenosis, double-contrast radiography revealed concentric narrowing of the thoracic esophagus in the absence of any vascular abnormalities at the base of the heart. Successful treatment included having the foal stand with the forelimbs elevated off the ground following each feeding.50 Persistent right aortic arch is a congenital anomaly in which the right fourth aortic arch becomes the de nitive aorta instead of the left aortic arch, which results in constriction of the esophagus by the ligamentum arteriosum as it extends between the anomalous right aorta and the left pulmonary artery.49 Clinical signs may include those associated with esophageal (postpharyngeal) dysphagia, drooling, and distention of the cervical esophagus resulting from partial obstruction of the thoracic esophagus.42,43 Endoscopic examination typically reveals dilation of the esophagus cranial to the obstruction, with evidence of di use esophagitis. Surgical treatment of persistent right aortic arch has been reported in foals, with varying degrees of success.43,45,47 In one report, preoperative computed tomography was used to identify the exact anatomic location of the offending lesion and guide the surgical approach.47 Esophageal duplication cysts and intramural inclusion cysts cause typical signs of esophageal obstruction, including salivation, esophageal dysphagia, and swelling of the cervical esophagus as the cysts enlarge.37,39,41 Such signs can make them di cult to di erentiate from other forms of esophageal obstruction (choke). Endoscopic examination may reveal compression of the esophageal lumen and communication with the esophageal lumen if it exists. Ultrasonographic examination may be the most useful method of antemortem diagnosis if the cyst is in the cervical esophagus. Examination of an aspirate of the mass may aid in the diagnosis by revealing the presence of keratinized squamous cells.39,41 Surgical treatments have included complete surgical resection and surgical marsupialization.37,39,41 The latter appears to be more successful and results in fewer complications.37,41 Complications of surgical resection include laryngeal hemiplegia following surgical trauma to the recurrent laryngeal nerve in the region of the esophagus and esophageal fistula formation.41

Esophageal Perforation Perforation typically occurs in the cervical region in response to external trauma, necrosis of the esophageal wall caused by a food impaction, or rupture of an esophageal lesion such as an impacted diverticulum. The esophagus is particularly vulnerable to external trauma in the distal third of the neck because only a thin layer of muscle covers it at this point.79 Iatrogenic perforation may occur in response to excessive force with a stomach tube against an obstruction or a compromised region of the esophagus.30 Esophageal perforations may be open or closed and tend to cause extensive cellulitis and necrosis of tissues surrounding the wound because of drainage of saliva and feed material within fascial planes. Systemic in ammation resulting from septic cellulitis may occur. Closed perforations of the esophagus are particularly troublesome because food material, water, saliva, and air may migrate to the mediastinum and pleural space via fascial planes.30,79 Because

of the leakage of air into the tissues surrounding the rupture, extensive subcutaneous and fascial emphysema frequently develops and is usually evident clinically and on cervical radiographs. Pneumomediastinum and pneumothorax are potentially fatal complications of esophageal ruptures. If the esophagus ruptures into the mediastinum, horses usually exhibit signs of acute systemic in ammatory response syndrome that may be mistaken for colic. Treatment should include converting closed perforations to open perforations if possible,80 extensive débridement and lavage of a ected tissues, broad spectrum antibiotics, tetanus prophylaxis, and esophageal rest. The clinician may achieve the latter by placing a feeding tube into the esophagus via the wound. Alternatively, one may place a nasogastric tube using a small tube (12-F diameter).30 For open perforations, once the wound has granulated and contracted to a small size, peroral feeding may be attempted.79 Extensive loss of saliva via esophageal wounds may lead to hyponatremia and hypochloremia. In addition, transient metabolic acidosis occurs because of salivary bicarbonate loss, followed by progressive metabolic alkalosis.64 Although reports of esophageal wounds healing well by second intention exist, healing time is prolonged.81 In addition, some perforations never completely heal and form permanent esophagocutaneous stulae that may require surgical correction. The development of esophageal strictures is not common because wounds are usually linear and not circumferential. However, traction diverticula may develop. Other complications of esophageal wounds include Horner’s syndrome and left laryngeal hemiplegia.79 In a retrospective study on esophageal disorders, only 2 of 11 horses with esophageal perforations survived over the long term;16 in a report of esophageal trauma following nasogastric intubation, four of ve horses were euthanized.30 The prognosis is therefore poor in horses with esophageal perforations, largely because of the extent of cellulitis, tissue necrosis, shock, and local wound complications.

DISEASES OF THE STOMACH L. Chris Sanchez The availability of specialized endoscopic equipment allowing visual inspection of the entire adult equine stomach and proximal duodenum is now commonplace for veterinarians in referral and select ambulatory settings. Thus veterinarians, owners, and trainers are increasingly aware of gastric disease in horses.

GASTRODUODENAL ULCERATION Peptic ulcer disease is de ned as erosions or ulcers of any portion of the gastrointestinal tract normally exposed to acid.1 Mucosal damage can include in ammation, erosion (disruption of the super cial mucosa), and ulceration (penetration of the submucosa). In severe cases fullthickness ulceration can occur, resulting in perforation. The proximal (orad) portion of the equine stomach is lined by strati ed squamous mucosa similar to that lining the esophagus. This proximal portion is also commonly termed the nonglandular region of the stomach. The distal (aborad) portion of the stomach is lined with glandular mucosa, and the distinct junction between the two regions is deemed the margo plicatus. Ulceration can occur in either or both gastric regions, although di erent clinical syndromes and pathophysiologic mechanisms apply. As a result, the broad term equine gastric ulcer syndrome (EGUS) has been used to encompass the wide array of associated clinical syndromes.2 EGUS develops in horses of all ages and is arguably the most clinically and economically important disorder of the equine stomach.

Prevalence The prevalence of gastric ulceration has been reported for a variety of breeds and usages; however, most current data involve Thoroughbreds or Standardbreds in race training. The prevalence of squamous ulceration in horses in race training varies between 70% to 95% 3-11 and can be as high as 100% when limited to animals actively racing.7 In recent years horses performing in other disciplines have also been evaluated, including active show horses (58% prevalence),12 endurance horses (67%),13 western performance horses (40%),14 Thoroughbred broodmares (67% pregnant; 77% nonpregnant),15 and nonracing performance horses (17% precompetition; 56% postcompetition).16 In one large retrospective study (3715 adult horses from 1924-1996) evaluating incidence of gastric ulceration identi ed at necropsy, an overall prevalence of 10.3% was found, with the highest prevalence in Thoroughbreds (including Arabians) and Standardbred trotters (19%).17 Horses in a university riding program also demonstrated a low squamous ulceration prevalence (11%).18 Approximately 49% of horses seen in a referral hospital for colic had evidence of gastric ulceration.19 The reported prevalence of gastric ulceration in foals varies between 25% to 57%.20-22 Many earlier studies investigating the prevalence of gastric ulceration do not di erentiate between nonglandular and glandular lesions, and many evaluate only the nonglandular region of the stomach, but this trend is changing. In a recent study of 162 horses in a hospital setting, 58% had antral or pyloric erosions or ulcerations, 58% had squamous mucosal lesions, and 8% had lesions involving the glandular body.23 In other studies 56% of Thoroughbreds in which the pylorus was examined had glandular ulceration,9 and 47% of racehorses (Thoroughbred and Standardbred) in which the pylorus was evaluated had ulceration in that location.11 In the former study9 all horses with glandular lesions also had squamous disease, whereas such an association was not seen in the latter study.11 In endurance horses with a 67% overall lesion prevalence, 57% had squamous ulceration, whereas 27% had glandular disease.13 In a postmortem evaluation, lesions were most commonly located in the squamous mucosa along the margo plicatus, followed by the glandular body, proximal squamous mucosa, and antrum.17

Pathophysiology An imbalance between inciting and protective factors in the mucosal environment can result in ulcer formation.24,25 The major intrinsic factors promoting ulcer formation include hydrochloric acid (HCl), bile acids, and pepsin, with HCl the predominant factor. Various intrinsic factors protect against ulcer formation, such as the mucus-bicarbonate layer, maintenance of adequate mucosal blood ow, mucosal prostaglandin E2 and epidermal growth factor production, and gastroduodenal motility. In humans extrinsic ulcerogenic factors include NSAIDs; Helicobacter pylori; stress; changes in diet; and gastrointestinal disorders, especially those resulting in delayed gastric emptying.1 In human neonates physiologic stress associated with a major primary illness seems to be strongly associated with gastric ulcers.26 Many of the other factors mentioned previously are believed to be important in horses, but clear evidence of an infectious agent has not yet been identi ed in horses or foals with EGUS.27,28 Recently, the possibility of Helicobacter infection in horses has re-emerged with the identi cation of polymerase chain reaction (PCR) products consistent with Helicobacter-like bacteria in horses with and without gastric lesions29,30 and a new species, Helicobacter equorum, identi ed in horse feces.31 Serum antibodies against H. pylori appear common in

postsuckling foals and their dams.32 However, H. equorum did not colonize the equine stomach in an experimental model,33 and a true link between Helicobacter infection and any form of EGUS remains elusive. The speci c factors involved in injury as well as protective mechanisms vary among regions of the proximal gastrointestinal tract, and a thorough review of mucosal injury and repair is presented elsewhere in this chapter. The pathophysiology of squamous mucosal ulceration in the horse appears similar to that seen in gastroesophageal re ux disease (GERD) in human beings and ulceration of the nonglandular mucosa in pigs. Excessive acid exposure is the predominant mechanism responsible for squamous mucosal ulceration, although many details remain unclear.34 Hydrochloric acid is secreted by parietal cells in the gastric glands via a hydrogen-potassium adenosine triphosphatase (H+, K+-ATPase) pump on the luminal side. Horses secrete acid continuously, and measured pH of equine gastric contents is variable from less than 2 to greater than 6, depending on the horse’s dietary state (fed versus fasted).35,36 A protocol of repeated 24-hour periods of fasting and feeding induces squamous erosion and ulceration.37 Because this protocol results in periods of prolonged gastric acidity (pH < 2.0) and concurrent administration of the histamine2 (H2) receptor antagonist ranitidine reduces lesion severity, it supports the role of acid exposure in the pathogenesis of squamous ulcer disease. Several peptides can stimulate or inhibit the secretion of acid by the parietal cell. The predominant stimuli to hydrochloric acid secretion are gastrin, histamine, and acetylcholine via the vagus nerve.1 Gastrin is released by G cells within the antral mucosa, and histamine is released by mast cells and ECL cells in the gastric gland. Histamine binds to type 2 receptors on the parietal cell membrane, causing an increase in cyclic adenosine monophosphate (cAMP), resulting in phosphorylation of enzymes that activate the proton pump. Gastrin and acetylcholine can act via calcium-mediated intracellular pathways and also stimulate histamine release directly.38 Isolated equine parietal cells respond maximally to histamine stimulation and only minimally to carbachol and pentagastrin.39 In vivo, histamine or pentagastrin infusion can stimulate similar maximal acid output.40 Interestingly, pentagastrin stimulation also induces a marked duodenal secretion of a sodium chloride-rich uid, which can re ux back into the stomach under fasting conditions.40 Gastrin release is primarily controlled by gastrin-releasing peptide (GRP), which is stimulated by gastric distention and increased luminal pH, but the interaction between gastrin and histamine has not been fully elucidated in the horse. Gastric acid secretion by parietal cells is primarily inhibited by somatostatin, released by fundic and antral D cells. The inhibitory e ect of somatostatin is primarily paracrine, but plasma levels of somatostatin negatively correlate with gastric luminal acidity.41 Gastric acid secretion also is inhibited by epidermal growth factor (EGF), a peptide produced in saliva.42 Foals can produce signi cant amounts of gastric acid by the second day of life, with consistent periods of acidity (pH < 2.0) in clinically normal animals.43,44 In one study foals tended to have a high gastric pH at day 1 of age,43 but in a study of critically ill foals, some foals demonstrated periods of gastric acidity on the rst day of life.45 Suckling was associated with an immediate rise in gastric pH, whereas periods of rest during which foals did not suck for more than 20 minutes were associated with prolonged periods of acidity.44 Premature human infants are capable of gastric acid production at 28 weeks of gestation.46 In sheep abomasal pH begins to drop at approximately 125 days of gestation but does not reach values normal for adult sheep (2.5) until after birth, which corresponds to the presence of H+/K+/ATPase mRNA, present in the fetal fundus by 125 days of gestation and rising to adult levels until after birth.47 This observed increase in acid secretion corresponds to an increase in bioactive serum gastrin. Only one of seven premature foals demonstrated an acidic pH recording in a study of gastric pH pro les in critically ill foals.45 Although multiple factors were likely involved in those foals, the true ontogeny of gastric acid production in foals is currently unknown. Equine squamous mucosa is very thin at birth but becomes hyperplastic and parakeratotic within days (Figure 15-15).48 The parallel between decreasing pH and proliferation of squamous epithelium correlates with that seen in other species.49 The combination of a relatively thin gastric epithelium with a high acid output may leave neonatal foals susceptible to ulcer formation at a very young age. In addition, one must remember the di erence in normal appearance of the squamous mucosa when interpreting gastric endoscopy in a neonatal population.

FIGURE 15-15

Endoscopic image of the greater curvature of the stomach in a 3-day-old Thoroughbred colt.

In esophageal squamous mucosa, intercellular tight junctions and bicarbonate secretion are the major factors involved in protection against acid injury in other species, although squamous bicarbonate secretion has not been documented in the horse.50-52 The principal barrier is a glycoconjugate substance secreted by cells in the stratum spinosum, with a contribution from the tight junctions in the stratum corneum.52 This barrier function is considered weak at best, and thus a functioning lower esophageal sphincter, normal salivary ow, and salivary mucins contribute to the prevention of acid injury in human GERD. In horses a mechanical barrier like the lower esophageal sphincter is not available to protect the gastric squamous mucosa from acid exposure. The squamous mucosa adjacent to the margo plicatus, especially along the lesser curvature, likely receives regular exposure to acidic gastric contents. Recently, surface mucus demonstrating crossreactivity to a mixture of anti-human MUC5AC (a mucin gene) antibodies has been identi ed in both the glandular and nonglandular regions of healthy equine gastric mucosa.53 Bile salts and pepsin have been implicated as contributing factors to ulcer disease in many species. In rabbit esophageal mucosa, bile salt absorption occurs and is directly correlated with mucosal barrier disruption.54 The unconjugated bile salts cholate and deoxycholate have a pKa of 5 and 5.3, respectively, and therefore cannot remain in solution and cause mucosal damage in the presence of acid. Alternatively, the conjugated bile salt taurocholate (pKa 1.9) can cause mucosal injury in either the ionized salt form at pH 7 or the nonionized acid form at pH 1-2.54 In the pig bile salts or acid alone causes squamous mucosal damage, whereas a combination of the two result in extensive damage in vitro.55 In the horse a similar synergistically damaging e ect was found with the addition of bile salts and acid (pH 2.5) to strati ed squamous mucosa in vitro in one study.56 In addition, the investigators were able to document levels of both bile salts and acid su cient to cause mucosal damage in gastric contents within 14 hours of feed deprivation. This is not surprising, given that duodenogastric re ux occurs normally in the horse.40 In a separate in vitro study of equine squamous mucosa, prolonged exposure to acid alone (pH 1.5) had a damaging e ect, and synergism with exposure to a combination of acid and pepsin or taurocholate was not found.57 The lack of synergism is likely due to the lower pH used in this study and stresses the importance of acid exposure in squamous ulcer disease. The e ect of acid has been consistent among studies. In addition, several volatile fatty acids (acetic, butyric, and proprionic), which had previously been documented in gastric contents of horses fed hay or hay and grain diets, decreased nonglandular mucosal barrier function.58,59 Lactic acid exposure alone does not appear to alter nonglandular mucosal bioelectric properties.60 Pepsinogens are secreted primarily by chief cells, although secretion by neck cells, cardiac glands, and antral pyloric glands also occurs.61 In an acidic environment (pH 85% baseline) is reached by 120 minutes in healthy horses.6,7 The immediate dietary history, gastric emptying rate, intestinal transit, age, and hormonal e ects of the horse in uence glucose peak and curve shape. Prolonged fasting (≥24 hours) results in a delayed and slightly lower peak plasma glucose concentration.8 Higher glucose peaks are recorded from healthy animals eating grass or hay than from those eating concentrates9 and from horses fed a pasture (clover and kikuyu) versus stable (oat hay, complete feed and alfalfa/oat cha ) diet.10 Results from the OGTT can also be a ected by the content of nonstructural carbohydrate and fat in the diet11 and other disorders, such as polysaccharide storage myopathy.12 In one report of 42 mature horses with chronic weight loss, a normal OGTT (peak glucose concentration at 120 minutes >85% baseline) was not associated with abnormal small intestinal morphology in any of the ve horses in which it was documented. When peak glucose

concentration was between 15% and 85% of baseline at 120 minutes (considered partial malabsorption), approximately 72% had small intestinal in ltrative disease, and when peak concentration at 120 minutes was less than 15% above baseline (total malabsorption), all had severe small intestinal in ltrative disease.7 However, other reports have documented horses with at OGTT curves that subsequently showed more normal OGTT responses and resolved clinical condition13 and poor diagnostic sensitivity of carbohydrate absorption tests to detect small intestinal involvement in horses with chronic diarrhea and predominantly large intestinal problems.14 Thus an abnormal absorption test and weight loss can occur in the horse as a transient event and without significant morphologic changes in the small intestine. d-Xylose Absorption Test

The d-xylose absorption test is performed in the same manner as the OGTT, except 0.5 g d-xylose per kilogram of body weight is administered as a 10% solution through a nasogastric tube. The pretest period of fasting and timing of sample collection is identical. Before collecting the samples, the veterinarian should contact the laboratory to ensure that heparinzed plasma is acceptable. Plasma d-xylose should peak between 20 and 25 mg/dl between 60 and 90 minutes after administration, with occasional peaks as late as 120 minutes.15,16 d-xylose absorption testing is not confounded by hormonal e ects or mucosal metabolism, as is glucose, but is altered by diet, length of fasting (which could also be in uenced by recent appetite or degree of cachexia), and age. Horses fed a high-energy (e.g., oat cha , oats, and corn) diet had a lower mean peak plasma d-xylose concentration (14.1 versus 24.9 mg/dl) than those consuming a low-energy (alfalfa cha ) diet.10 Healthy mares fasted for up to 96 hours had atter curves and a slower decrease in plasma xylose than when fasted for 12 to 36 hours.15 Kinetic analysis indicates that prolonged food deprivation does not alter renal or nonrenal excretion of d-xylose; thus the e ect of fasting on the curve is likely related to intestinal transit, small intestinal absorption, or both.17,18 Ponies may have lower peak d-xylose concentrations than horses, although the range is wide and diagnostic discriminatory cutoff points for peak plasma xylose concentrations have not been determined. Foals normally have a higher mean peak xylose concentration at 1 and 2 months of age, but mean peak falls to a level similar to that seen in adults by 3 months of age.19 Gastric emptying rate, intestinal motility, intraluminal bacterial overgrowth, and renal clearance can affect curve shape. d-Xylose, a pentose sugar, is absorbed unchanged by the duodenum and jejunum in other species. The intestinal sugar active transport system has a low a nity for d-xylose in the equine jejunum in vitro; hence d-xylose absorption likely occurs primarily by convection or diffusion.20 Thus an abnormal d-xylose absorption test likely indicates abnormal mucosal surface area or permeability. Such an abnormality, represented by a at curve or delayed absorption, has been observed with most examples of chronic in ammatory bowel disease, parasitism, and idiopathic villous atrophy.21,22 Abnormal absorption curves have been detected in the absence of small intestinal histologic changes,23 and interpretation is clouded further by ndings from small intestinal resection studies in healthy ponies. One study demonstrated a progressive decline in mean peak xylose concentration after 70% distal small intestinal resection in ponies, despite their normal clinical appearance and absence of diarrhea.24 In another study a similar decline in peak xylose concentration following extensive (≥60%) small intestinal resection was accompanied by weight loss, diarrhea, and ill thrift.25 Peak xylose concentrations were much lower in horses with granulomatous enteritis than those with eosinophilic granulomatosis (EG), whereas in EG the absorption curve shifted to the right, with the peak occurring at 240 minutes.26 This is not surprising given the typical lesion distribution with these disorders. As with the OGTT, results of the xylose absorption test can improve after therapy.27

Lactose Absorption Milk intolerance is well documented in children and can result from primary or secondary lactase (neutral β-galactosidase) de ciency. Secondary lactase de ciency can occur when a small intestinal disorder, typically rotavirus, damages the epithelial cells, resulting in decreased brush border disaccharidase activity. Such an association has been reported secondary to clostridial enteritis in the foal.28 In horses lactase activity peaks at birth and declines slowly, such that activity is approximately 50% at 2 years of age and rapidly decelerates after 3 years of age; it is barely detectable by 4 years of age.29 The oral lactose tolerance test is performed in foals after a 4-hour fast, with a standard dose of 1 g/kg lactose monohydrate given through a nasogastric tube as a 20% solution. Blood sampling is performed in the same manner as for the OGTT. Plasma glucose typically peaks 1 hour after lactose administration, with a range from 30 to 90 minutes, with a mean increase of 77 mg/dl and a 35 mg/dl increase covering two standard deviations.30 It should be noted that older foals (12 weeks) had a more significant increase in plasma glucose relative to 1-week-old foals.

Nuclear Medicine Scintigraphy with radiolabeled leukocytes (technetium-99m-labeled hexamethyl-propyleneamine oxime (99mTc-HMPAO) was recently used to evaluate 17 horses with suspected intestinal malabsorption.31 In that report the investigators were able to detect intestinal in ammation in 12 of 17 horses, and increased intestinal uptake was not evident in any of the control horses. Corroborating histopathologic results were not available for all cases; this technique was not able to differentiate between type or location of disease, nor did it indicate prognosis.

DISEASES ASSOCIATED WITH MALABSORPTION AND MALDIGESTION The clinical signs of chronic wasting and poor body condition, although nonspeci c for a diagnosis of malabsorption ante mortem, can be attributed to proliferative or in ammatory intestinal disorders, often collectively referred to as chronic in ammatory bowel diseases (CIBD).22 Summaries of the characteristic clinicopathologic and pathologic ndings for each disease are presented in Tables 15-11 and 15-12, respectively.

Alimentary Lymphosarcoma Alimentary lymphosarcoma of the horse may represent a primary neoplasia of the gut-associated lymphoid tissue with signi cant cellular in ltration of the small intestine and associated lymph nodes with minimal large intestinal or systemic involvement. Case series and pathology reports indicate that young horses 2 to 4 years of age primarily are a ected, although the age range can be broad.32-34 No breed or sex predilection has been documented, and disease prevalence is unknown. Despite the progressive nature of lymphomata, onset of clinical signs can be rapid, and the animal may become acutely ill. As with all adult cases of CIBD, antemortem diagnosis is by a process of exclusion and usually is con rmed post mortem. Frequent abnormalities include anemia, thrombocytopenia, neutrophilia or neutropenia, and hypoalbuminemia with hyperglobulinemia, resulting in either a normal or elevated serum protein. Lymphocytosis is rare. Intra-abdominal masses, mainly enlarged mesenteric lymph nodes, may be rectally palpated. Abdominocentesis and rectal biopsy can provide a diagnosis but are not sensitive indicators of disease. Carbohydrate absorption tests usually reveal partial to total malabsorption indicative of the severely reduced surface area resulting from signi cant villous atrophy and the extensive mucosal or transmural in ltration. Early con rmation of a suspected diagnosis necessitates exploratory laparotomy to obtain multiple intestinal and lymph node biopsies if the results of rectal biopsy or abdominocentesis are normal. Prognosis is poor, especially insofar as most horses are presented in an advanced state of disease. Immunosuppressive drugs or chemotherapy may a ord temporary improvement, but long-term outcome is una ected. A detailed discussion of lymphosarcoma is included in Chapter 14.

Granulomatous Enteritis Granulomatous enteritis was rst described as a chronic wasting condition in 197435; 9 of 10 horses were young Standardbreds. Most a ected horses are 2 to 3 years of age. Case reports from many countries revealed a predominance of Standardbred over Thoroughbreds by three to one.26,36 Some of the Standardbreds were related, implicating a genetic predisposition, but this has not been proved. Prevalence is low. The condition is sporadic and has an insidious onset, and the course can be protracted. Signi cant diagnostic features include anemia, slight increases or decreases in white blood cell counts, hypoalbuminemia, normal serum protein or hypoproteinemia, occasional increases in serum alkaline phosphatase activity, normal serum γ-glutamyltransferase activity, and enlarged mesenteric lymph nodes on rectal palpation. Partial or complete malabsorption is typically documented by carbohydrate absorption testing. The low proportion of horses exhibiting diarrhea can be attributed to the preferential distribution of in ammatory in ltration in the small intestine.37 Rectal biopsy can be a useful aid to diagnosis.2 The cause of granulomatous enteritis is unknown. Several infectious agents have been implicated, including Mycobacterium avium.38 The condition may represent a granulomatous hypersensitivity reaction. Immune-mediated responses to dietary, parasitic, or bacterial antigens may be important initiating factors.22 Six cases purported to represent granulomatous enteritis were linked to environmental contamination with aluminum,39 although problems existed regarding the case definition, data, and interpretation.40 Treatment of horses with granulomatous enteritis with a variety of drugs, particularly corticosteroids, has not a ected the long-term outcome in the majority of cases.41 Prolonged (i.e., 5 months) corticosteroid administration produced clinical remission and a favorable athletic outcome in a 6-year-old Standardbred gelding based on improvement in clinical signs and in d-xylose absorption.27 Surgery may be indicated with localized disease. Two young horses underwent resection of the thickened terminal small intestine; one horse died 4 months after surgery, and the other remained clinically normal for at least 10 years.36

Multisystemic Eosinophilic Epitheliotropic Disease MEED encompasses disorders characterized by a predominant eosinophilic in ltrate in the gastrointestinal tract, associated lymph nodes, liver, pancreas, skin, and other structures and accompanied by some degree of malabsorption and enteric protein loss. The disorders include chronic eosinophilic gastroenteritis,42 eosinophilic granulomatosis,26 chronic eosinophilic dermatitis,43 and probably basophilic enterocolitis.44 Although prevalence is low, MEED appears to be more common than granulomatous enteritis. Most a ected horses are 2 to 4 years of age, and Standardbreds and Thoroughbreds are reported to predominate. The condition is sporadic and has an insidious onset and often a protracted course (i.e., 1 to 10 months). Diarrhea is common. Severe skin lesions with exudative dermatitis and ulcerative coronitis are

prominent and frequently are the principal presenting complaint. Despite extensive tissue eosinophilia, systemic eosinophilia is rare and hematologic values are usually unremarkable. Notable features include hypoalbuminemia and elevations in serum γ-glutamyltransferase and alkaline phosphatase activities. Most reports of carbohydrate absorption test ndings indicate a reduced or normal peak concentration delayed to at least 180 minutes. Morphologic changes are less pronounced in the small intestine than in the large intestine,26 and small intestinal lesions predominate segmentally in the proximal duodenum and distal ileum. Furthermore, signi cant hyperkeratosis of the fundic region may contribute to gastric muscle contractile disruption. Diarrhea can be a consequence of the severe segmental or multifocal granulomatous lesions in the large intestine, with mucosal and transmural thickening and extensive ulceration. Abundant brosis is a feature of all affected tissues. The cause of MEED is unknown and could represent a chronic ongoing immediate hypersensitivity reaction against unde ned antigens ingested or excreted into the lumen from parasitic, bacterial, or dietary sources. Infectious agents have not been identi ed.42,43 Eosinophilia is a feature of parasitism in the equine intestinal tract, although nematodes rarely have been identi ed in any lesions of MEED.42,45 However, failure to detect larval structures in these lesions may be attributable to the chronicity of the disease and destruction of the parasites in tissue.36 Biopsies of the rectal mucosa2 or of the skin, liver, intestinal tract, and lymph nodes may assist in diagnosis. Unlike the other conditions, MEED has definitive liver and pancreatic involvement; thus maldigestion may contribute to the wasting disease. Treatment has been attempted with a variety of drugs, including antibiotics, corticosteroids, and anthelmintics with larvicidal activity. Although some horses improve briefly, the long-term prognosis is poor.

Eosinophilic Enterocolitis Idiopathic eosinophilic enterocolitis a ects segmental lesions in the small or large intestine, inducing signs of colic, and often requires surgical intervention.36,46,47 This problem may not involve evidence of malabsorption and does not have multisystem involvement. Because the problem is often associated with signs of colic and not signs of malabsorption, eosinophilic enterocolitis di ers from the other conditions discussed in this section and is often diagnosed at the time of surgery. It carries a much better prognosis than the other in ammatory bowel diseases.

Lymphocytic-Plasmacytic Enterocolitis The morphologic ndings in lymphocytic-plasmacytic enterocolitis re ect the predominant in ltrative cellular elements of this rarely encountered condition. No speci c clinical or clinicopathologic features di erentiate this condition ante mortem from other in ammatory diseases of adult horses. In a retrospective study of 14 horses, carbohydrate absorption was abnormal or delayed in 9 of 12 horses, consistent with the predominance of small intestinal pathologic changes.3 Rectal biopsies were abnormal in three of seven horses, two of which were reported as having lymphocytic-plasmacytic proctitis. Prognosis is poor. Treatment has been unsuccessful, probably because of the advanced nature of the condition at the beginning of treatment.

Proliferative Enteropathy Proliferative enteropathy (PE) typically a ects weanling foals between 3 to 8 months of age and has been reported in North America, Europe, and Australia, causing disease in individuals or outbreaks of multiple a ected animals on the same premise.48-59 PE has been reported uncommonly in yearlings and adult horses.57,60 The disease a ects many other species, especially swine, and is caused by Lawsonia intracellularis, an obligate intracellular bacterium found in the cytoplasm of proliferative crypt epithelial cells of the jejunum and ileum.51,57,59 The condition in a foal was described rst as intestinal adenomatosis,61 but later work con rmed that intestines from an affected foal contained L. intracellularis.51 Clinical signs include depression, rapid and signi cant weight loss, edema, diarrhea, and colic.57 Poor body condition, a rough hair coat, and potbelly appearance are also reported. It is important to remember that not all clinical signs are present in all cases, and diarrhea has been noted in only approximately half of reported cases. Other problems often were concurrent, including respiratory tract infection, dermatitis, intestinal parasitism, and gastric ulceration. The most signi cant laboratory nding is profound hypoproteinemia, predominantly characterized by hypoalbuminemia, but panhypoproteinemia can also occur.48,52,57 Leukocytosis and hyper brinogenemia are also common, with occasional alterations in electrolytes (hyponatremia, hypokalemia, hypochloremia) and elevated serum creatine kinase concentrations. Abdominal ultrasound commonly reveals increased small intestinal mural thickness.48,52 Colloid oncotic pressure is typically low, if measured.48 One can suspect PE based on clinical signs and severe hypoalbuminemia in a weanling foal with exclusion of common enteric infections. Fecal PCR for bacterial DNA or serum immuno uorescence assay or immunoperoxidase monolayer assay (IMPA) for antibodies against the organism can provide additional support for the diagnosis. The IMPA may be more sensitive, and typically titers greater than or equal to 1:60 are considered positive.60 Submission of both tests is recommended ; although both tests are quite speci c, both also lack sensitivity,

especially early in the course of disease (serology) or with prior antimicrobial therapy (fecal PCR).48,57 Notably, fecal PCR can become negative in a ected foals within 4 days of antimicrobial therapy.58 PE is not typically associated with abnormal carbohydrate absorption test results.57,58 In cases with diarrhea, infectious causes should be ruled out. A de nitive postmortem diagnosis can be con rmed by identifying characteristic mural thickening and intracellular bacteria within the apical cytoplasm of proliferating crypt epithelial cells using silver stains, PCR, or immunohistochemical testing.57 Reported antimicrobial therapy includes eythromycin, alone or with rifampin, azithromycin, clarithromyxin, oxytetracycline, doxycycline, metronidazole, or chloramphenicol.48,57,62 Recent reports appear to favor the use of intravenous oxytetracycline, followed by oral doxycycline, with apparent success.48,62 Duration of therapy is typically 2 to 4 weeks. A ected foals often need supportive therapy, including crystalloid uid and electrolyte replacement and often colloid support with either hydroxyethyl starch or plasma. NSAID therapy can be used as needed for signi cant pyrexia. Corticosteroid therapy is not indicated because in ammation is not a signi cant pathologic nding. Response to therapy has been good, with reported survival rates between 82% and 93%.48,57,62 Rapid improvement in clinical signs, even within 24 hours, preceded the rise in plasma protein concentration. The source of infection is typically not determined. In some epidemiological investigations, close proximity to swine operations was apparent, but in most such an association was not evident.52,57 Comparisons of epidemiologic ndings from the swine disease indicated that overcrowding, feed changes, antibiotic usage, and mixing and transportation were potential risk factors at two of the farms in one study, and recent weaning appears to be a common risk factor.57

Other Conditions Abnormal d-xylose absorption has also been noted in association with AA amyloid-associated gastroenteropathy in an 18-year-old Morgan stallion63 and a horse with a gastric mass and secondary small intestinal villous atrophy.64

MANAGEMENT, THERAPY, AND OUTCOME The chronic wasting horse with suspected malabsorption and probable enteric protein loss has at best a guarded to poor prognosis, with the exception of those su ering from PE. Prognosis may be improved through early and aggressive investigation to achieve a diagnosis. In the short term intravenous infusion of plasma or colloids, with or without uids and electrolytes, may be necessary to stabilize the profoundly hypoalbuminemic patient. Prognosis is much worse for the horse that is inappetent. Prolonged intensive total parenteral nutrition, oral alimentation, or both may not be a realistic course of action. The owner must be cognizant from the start that the outcome may not be altered, even after protracted therapy; only a few case reports of successful responses with long-term follow-up have been documented.

Nutrition Some level of digestive and absorptive capability remains in the diseased small intestine. Interval feeding of small quantities of easily digestible food may be bene cial. Diet should include feeds with a high ber content to favor large intestinal fermentation, including grass hay and access to pasture complemented by high- ber rations based on beet pulp and soybean hulls, which are commercially available in many complete pelleted feeds. Energy intake can be increased through feeding high-energy dense fats that provide 2.25 times more calories than carbohydrates. Most a ected horses should tolerate high fat (5% to 10%) processed feeds containing vegetable oils or rice bran (up to 20% of the concentrate mix, equivalent to 8% vegetable oil) to achieve the higher-fat composition. The transition to a higher-fat concentrate should be gradual. Even in healthy animals that can eat up to 20% added fat, appetite may decrease as the percentage increases, and fecal consistency may change. Clearly, the objective for the horse with suspected malabsorption is to sustain, and preferably increase, dietary intake, value, and efficiency. The owner of an a ected horse must be prepared to experiment with feeds, be patient, and keep records. Exposure to a feed component may contribute to the problem as an allergen eliciting a hypersensitivity reaction. Identifying the potential allergen through immunologic testing or by stepwise removal and outcome assessment over a longer period may be di cult. The clinician should give immunosuppressive drugs early in the process.

Drug Therapy Immunosuppressive agents have produced the most promising responses to ameliorate the e ects of conditions associated with malabsorption, particularly CIBD. Short-duration, and in some cases more prolonged and sustained, improvements in body condition, weight gain, demeanor, energy, and activity levels have occurred following corticosteroid administration. Treatment should begin as early as possible. The clinician should follow initial parenteral (intramuscular or intravenous) loading doses of dexamethasone (sodium phosphate) with a series of depot injections or orally administered prednisolone on a tapered-dose protocol over a period of months. Interval low-dose therapy may be necessary if clinical signs return after treatment ends. The clinician should use the lowest dose that e ectively controls the

clinical signs for alternate-day therapy. Clinical bene ts far outweigh concerns over potential adverse e ects. Chemotherapeutic agents such as vincristine, cytosine, cyclophosphamide, and hydroxyurea have been tried in a few cases of CIBD or lymphosarcoma with no apparent success, probably because of the advanced stage of the disease when treatment was initiated and the dose selected.

Surgery Resection of a segment of intestine that is edematous, hemorrhagic, or constricted is an option in localized forms of CIBD,46,47 particularly if gross changes are not discernible in adjacent or distant parts of the intestinal tract—that is, malabsorption is not a feature. Long-term outcome has been favorable. Removal of a substantial proportion of the diseased small intestine may be indicated in a horse with malabsorption, considering that resection of 70% distal small intestine was performed in healthy animals without inducing adverse e ects.24 However, because pathologic changes may exist in normal-appearing small or large intestine that is not resected or biopsied, the prognosis remains guarded. Two young horses with granulomatous enteritis had the thickened terminal small intestine resected with positive outcomes; one survived 4 months, the other has a follow-up extending more than 10 years.36

INFLAMMATORY DISEASES OF THE GASTROINTESTINAL TRACT CAUSING DIARRHEA Samuel L. Jones Acute diarrhea caused by colitis in adult or young horses is a potentially life-threatening disorder with a variety of possible etiologies (Table 15-13) characterized by hypersecretion of uid, motility disturbances, and an impaired mucosal barrier resulting from direct injury or in ammation. Many of the clinical and clinicopathological features are similar regardless of the underlying cause. Severe dehydration with profound electrolyte abnormalities is common, as is systemic in ammation secondary to absorption of endotoxin or other bacterial products through compromised gastrointestinal mucosa. Severe cases may be complicated by serosal in ammation and mural ischemia and infarction as a direct extension of mucosal in ammation or secondary to coaglopathies. The diagnostic approach for horses with acute diarrhea is aimed at determining the underlying etiology but must be accompanied by clinical and laboratory assessment of hydration, electrolyte and acid base balance, organ function, and evaluation of the degree of systemic in ammation and the integrity of the intestinal wall. The therapeutic approach for horses with colitis, regardless of cause, consists primarily of controlling local and systemic in ammation, maintaining uid and electrolyte balance, and promoting mucosal repair. In addition, some horses with acute colitis require speci c therapy aimed at the underlying etiology. TABLE 15-13 Differentials and Diagnosis of Some Causes of Acute Diarrhea in Adult Horses Category Infectious

Differentials Salmonellosis

Diagnosis Fecal culture (5 consecutive) Fecal PCR

Clostridium perfringens

Quantitative fecal culture Fecal toxin immunoassay or PCR

Clostridium difficile

Fecal culture Fecal toxin immunoassay or PCR

Lawsonia intracellularis

Serology Fecal PCR Small intestinal ultrasound Histopathology

Neoristicii risticii

Fecal or blood PCR Serology

Parasitic

Strongylosis

Fecal egg counts Cranial mesenteric artery palpation

Serum IgG(T) Cyathostomiasis

Fecal egg count Rectal biopsy Cecal or colonic biopsy

Toxic

NSAID

History and clinical signs Right dorsal colon ultrasonography Laparoscopy or laparotomy

Cantharidin

History of exposure Fecal or urine cantharidin concentrations

Arsenic

History of exposure Fecal, blood, urine, or tissue arsenic concentrations

Miscellaneous

Carbohydrate overload

History of inappropriate ingestion of carbohydrate Blood lactate concentration

Sand enteropathy

Auscultation of ventral colon Fecal sand content Abdominal radiography

PCR, Polymerase chain reaction; NSAID, nonsteroidal anti-inflammatory drug.

INFECTIOUS DISEASES Salmonellosis PATHOGENESIS Salmonella enterica is a species of gram negative facultatively anaerobic bacteria that is a common gastrointestinal pathogen in horses. Many serovars of S. enterica have been reported to infect horses, but those classi ed in group B appear to be more commonly associated with disease than those in other groups. Group B includes S. enterica var. typhimurium and S. enterica var. agona, two of the species most frequently isolated from horses.1-3 S. enterica var. typhimurium is the most pathogenic serotype in horses and is associated with a higher case fatality rate than other serovars of S. enterica1. The number of horses that are inapparently infected with and actively shed S. enterica in their feces has been reported to be as high as 10% to 20%, but actual prevalence of S. enterica shedding in the general horse population is likely to be much lower, less than 2% 4. Horses shedding S. enterica are a potential source of infection to susceptible horses,1,5 as are environmental reservoirs6-8. For these reasons salmonellosis is one of the most common nosocomial diseases in horses. Nosocomial salmonellosis signi cantly a ects morbidity and mortality in hospitalized horses.9 The emergence of multidrug resistance in equine S. enterica isolates has been a cause of concern because of the importance of salmonellosis as a nosocomial disease and because a number of serovars of S. enterica are significant zoonotic pathogens.7,10,13 The virulence of the bacteria varies tremendously with serotype and even among strains of the same serotype. In art, this is due to the important role of host susceptibility in the pathogenicity of particular organisms. The infective dose is generally on the order of millions of organisms inoculated orally, but various environmental and host factors can reduce the infective dose to a few thousand or even hundreds of organisms.14-16 Environmental factors or stresses that increase susceptibility to S. enterica infection are not well de ned, but it is known that high ambient temperature, for example, can greatly increase the prevalence of salmonellosis in horses.6,15,16 Indeed, the peak incidence of salmonellosis in horses occurs in late summer and fall.6,15,16 Other environmental and host factors that increase the risk of or are associated with salmonellosis or shedding of S. enterica organisms in feces include transportation, antibiotic administration before or during hospitalization, gastrointestinal or abdominal surgery, general anesthesia, preexisting gastrointestinal disease (e.g., colic, diarrhea), presence of leukopenia or laminitis during hospitalization, prolonged hospital stay, change in diet, and immunosuppression.1,8,16-18 Interestingly, foals with gastrointestinal disease are more likely to shed S. enterica organisms than are adult horses with gastrointestinal disease 17 Host factors that restrict gastrointestinal colonization and invasion by pathogens include gastric pH, commensal gastrointestinal ora,

gastrointestinal motility, the mucosal barrier, and mucosal immunity.1,19 Gastric acidity is an important defense mechanism for preventing live organisms from reaching the intestine.19 Altering the gastric pH, with histamine H2 receptor antagonists, for example, may increase susceptibility to infection. Gastrointestinal ora inhibit the proliferation and colonization of S. enterica by secreting bacteriocins, short-chain fatty acids (SCFAs), and other substances that are toxic to S. enterica.19 In addition, elements of the normal ora compete for nutrients and space, especially on the mucosa.19 Being predominantly anaerobic, the normal ora maintain a low oxidation-reduction potential in the environment of the large intestine, which inhibits the growth of many bacterial pathogens.20 The importance of normal host gastrointestinal ecology is illustrated by the fact that disturbances of the colonic ora with antibiotics, changes in feed, ileus, or other underlying gastrointestinal disease markedly increases the susceptibility of the host to infection by S. enterica, often resulting in serious disease. The immune status of the host may be one of the most important factors determining not only the susceptibility to S. enterica infections but also the degree of invasion and subsequent outcome of the infection. Local immunity, such as mucosal antibody secretion and enterocytederived cationic peptides, prevents colonization of the mucosa.19,21,22 Opsonizing antibodies and activation of the complement cascade are important in ghting systemic invasion by S. enterica by increasing the e ciency of phagocytosis and by direct bactericidal activity. Humoral immunity, however, is often ine ective in preventing disease and dissemination once invasion occurs and S. enterica has established in its intracellular niche. Following invasion, S. enterica is capable of surviving and multiplying within macrophages, rendering humoral (noncellular) immune systems ine ective.23,24 Speci c cellular immunity may be the most e ective defense mechanism in the host arsenal against dissemination and systemic infection by S. enterica.20,24 Protective immunity in horses and calves may be induced by oral inoculation with small numbers of virulent organisms, but the duration of the immunity is not known.25,26 Oral and parenteral vaccines using killed or attenuated organisms and bacterial products have been promising but are e ective only against hom*ologous organisms and are usually not cross-protective among different serogroups.25-27 In adult horses S. enterica primarily infects the cecum and proximal colon, causing enterocolitis, with limited likelihood of dissemination beyond the intestine. In foals, however, salmonellosis is often associated with septicemia. The ability of S. enterica to cause enterocolitis depends on the ability of the bacteria to invade the gastrointestinal mucosa.19,23 Invasion of the gastrointestinal mucosa occurs preferentially through specialized enterocytes called M cells that overlay intestinal lymphoid tissues such as Peyer’s patches in nonequine species. M cells are exploited by a variety of enteric pathogens during infection of intestinal tissue.28 Invasion of the epithelium occurs by self-induced uptake via the apical membrane of the M cell, often killing the cell in the process.23 S. enterica then invades neighboring cells via the basolateral membrane, eventually spreading the destruction of the epithelium beyond the principal area of attack. Virulent S. enterica have a welldeveloped invasion mechanism that involves the generation of an apparatus called a Type III secretory system that enables virulence gene products to be injected directly into enterocytes.29 Virulence proteins injected by S. enterica into enterocytes engage the cellular machinery and induce the cell to engulf the bacteria by macropinocytosis. S. enterica virulence gene products also induce enterocyte chloride and uid secretion and upregulate enterocyte transcription of in ammatory cytokines (TNF-α and IL-1β) and chemokines that trigger a mucosal inflammatory response.23,29,30 Once S. enterica has invaded the mucosa, the organisms are quickly phagocytosed by macrophages and dendritic cells in the lamina propria and in lymphoid tissues. The ability of S. enterica to disseminate systemically and cause enteric fever is associated with the ability to survive and proliferate in macrophages. Indeed, phagocytes have an important role in dissemination to blood, lymph nodes, liver, and spleen.31 The majority of S. enterica in the blood and tissues of animals infected with a strain of S. enterica that is competent to cause enteric fever are within phagocytic cells.31 In adult horses with salmonellosis, dissemination appears to be limited to the intestine and mesenteric lymph nodes, and S. enterica is rarely cultured from blood. However, in foals and in some adults, S. enterica causes an enteric fever–like disease with dissemination to mesenteric lymph nodes, liver, spleen, and blood. Speci c virulence gene clusters called pathogenicity islands encoded on the chromosome or on plasmids confer the main virulence traits of S. enterica: invasion, enteropathogenesis, intracellular survival, and proliferation.23 Some of the genes encoded within these islands or virulence factors are sensors that signal to the bacteria that it has entered an intracellular environment and turn on other genes required for intracellular survival. Others, such as invasion genes, are transported from the bacteria and injected into macrophage cytosol by a Type III secretory system apparatus to prevent phagosome-lysosome fusion and subvert other essential macrophage-killing mechanisms. Virulent S. enterica may also possess multiple genes that enable adhesion to target cells or confer resistance to reactive oxygen and nitrogen metabolites, perhaps the most lethal antimicrobial mechanisms of macrophages.32 Diarrhea associated with salmonellosis has multiple causes. A S. enterica cytotoxin inhibits protein synthesis in mucosal cells, causing morphologic damage and altered permeability34) Virulent S. enterica also produce an enterotoxin that is similar to the heat-labile (LT) toxin produced by E. coli.34,35 This enterotoxin contributes to, but is not required, in the pathogenesis of diarrhea.36,37 S. enterica enterotoxin increases secretion of chloride and water by colonic mucosal cells in many species, including horses, by increasing intracellular cAMP concentrations.34,35,38

The ability of virulent S. enterica to cause diarrhea appears to be most closely associated with the ability to invade enterocytes and to trigger an inflammatory reaction in the intestinal tissue.23,39 Gene products injected into enterocyte cytosol by the Type III secretory system of invading S. enterica stimulate chloride and uid secretion.29 S. enterica invasion of enterocytes is also a potent activator of in ammatory chemokine and cytokine production, resulting in the recruitment of leukocytes, particularly neutrophils, and activation of resident macrophages and mast cells. Products of these activated leukocytes, including prostaglandins, leukotrienes, reactive oxygen metabolites, and histamine, are potent stimulators of chloride secretion in the colon of many species.19,40-42 The enteric nervous system integrates the diverse processes of pathogen recognition, triggering of the inflammatory response, and induction of enterocyte fluid secretion.42 Many of the in ammatory mediators studied stimulate colonic secretion by prostaglandin-dependent mechanisms, resulting in either increased intracellular cAMP or calcium concentrations, or both, in mucosal cells.40 In addition, these mediators and the enteric nervous system may stimulate secretion by prostaglandin-independent mechanisms, inhibit sodium and water absorption, cause motility disturbances, and potentiate tissue injury, all of which enhance the pathogenicity and dissemination of S. enterica and contribute to the pathogenesis of diarrhea.40,42 Neutrophils recruited to the mucosa by signals generated by the infected enterocytes physically contribute to mucosal injury by producing a variety of products that are lethal to pathogens but are also toxic to host cells.43,44 Moreover, neutrophils attracted to infected epithelial cell accumulate beneath the monolayer, lifting it o the basem*nt membrane in sheets. Neutrophils also migrate across the epithelial monolayer in potentially massive numbers—enough to be detectable in feces as a marker of in ammatory diarrhea. Although the transepithelial migration of neutrophils has a bene t, positioning the host defense cell at the apical membrane to ward o attacks by invading bacteria, the mechanical disruption to the epithelial barrier may be signi cant enough to increase the permeability to macromolecules, bacterial products, and even bacteria.44 Potentially massive losses of electrolytes, water, and protein can occur, depending on bacterial and host factors. Perhaps most devastatingly, mucosal injury and altered permeability allow systemic absorption of bacterial products and dissemination of bacteria, resulting in life-threatening sepsis.

CLINICAL SIGNS AND DIAGNOSIS Four clinical syndromes of S. enterica infection have been documented clinically and reproduced experimentally in horses: (1) inapparent infections with latent or active carrier states; (2) depression, fever, anorexia, and neutropenia without diarrhea or colic; (3) fulminant or peracute enterocolitis with diarrhea; and (4) septicemia (enteric fever) with or without diarrhea.45 Inapparent infections can be activated to clinical disease in compromised horses, such as horses with colic or horses being treated with antibiotics, causing mild to severe enterocolitis. In addition, latent infections (nonshedding) can become active infections (shedding) under certain conditions, such as transportation stress and antibiotic treatment. Horses with depression, anorexia, fever, and neutropenia without diarrhea generally have a good prognosis and recover in several days without speci c treatment.45 The septicemic form is mostly restricted to neonatal foals and is uncommon in adult horses. The focus of this discussion is acute enterocolitis. Acute enterocolitis is characterized by severe brinonecrotic typhlocolitis, with interstitial edema and variable degrees of intramural vascular thrombosis that may progress to infarction.1 Severe ulceration of the large intestinal mucosa may occur, with serosal ecchymoses and congestion. The earliest signs of enterocolitis are usually fever and anorexia.1,16 Signs of colic may be seen early in the course of the disease, especially if ileus is present. Clinical signs of endotoxemia are common, and range from fever, elevated heart and respiratory rates, poor peripheral perfusion, and ileus to fulminant and rapidly progressive signs of endotoxemic shock. Oral mucous membranes are often pale with perigingival hyperemia (a toxic rim) but may be brick red or cyanotic, with prolonged capillary re ll time. Weakness, muscle fasciculations, cold extremities, and other signs suggestive of hypotensive shock; synchronous diaphragmatic utter; abdominal pain; and marked metabolic and electrolyte abnormalities may be noted in severe cases of enterocolitis. Signs of mild dehydration may be observed before diarrhea is seen. Once diarrhea is evident, dehydration may rapidly become severe. Occasionally, horses die peracutely, without developing diarrhea. Diarrhea may not occur for several days, but usually is evident by 24 to 48 hours after the onset of fever.1,16 The duration of diarrhea may be days to weeks. The character of the rst diarrheal feces is usually watery, with particles of roughage, but may rapidly become uid without solid material. Finding frank blood and brin in the feces is unusual. The volume of feces is often large, with frequent defecation. Straining or signs of colic may be observed when the patient is defecating, and rectal prolapse may occasionally occur. Persistent straining and rectal prolapse may be a sign of colonic infarction. Abdominal borborygmi are often absent early in the course of the disease because of ileus but become evident later, usually when diarrhea begins. Fluid and gas sounds are commonly ausculted, but normal progressive motility is less frequently heard than normally. Transrectal palpation may reveal edematous rectal and colonic mucosa and uid- lled colon and cecum. Gastric reflux may be obtained, especially early in the course, when ileus is evident. Hematologic abnormalities early in the course of the disease include moderate to severe neutropenia, lymphopenia, and leukopenia, a mild to moderate left shift, and toxic changes in the neutrophils.1,16 Thrombocytopenia, moderate to severe hemoconcentration, and

hyper brinogenemia are also common. Neutropenia is an early but nonspeci c indicator of salmonellosis, often occurring concurrently with the onset of fever.1 Later in the course of disease, neutrophilic leukocytosis may be seen, indicating recovery. A degenerating left shift, with metamyelocytes and myelocytes seen in the peripheral blood, is a poor prognostic sign. Serum biochemical analysis may reveal azotemia, increases in serum sorbitol dehydrogenase and γ-glutamine aminotransferase activity, and increased blood lactic acid concentration. Azotemia is often prerenal, but acute hemodynamic renal failure may be seen in severely dehydrated, endotoxemic, or septic patients. Indeed, elevation of creatinine concentration is a poor prognostic indicator in horses with acute colitis.46 Hemodynamic renal disease may be complicated by toxic injury caused by administration of nephrotoxic drugs. Hyponatremia may also contribute to prerenal azotemia. Elevations in hepatocellular enzymes are usually mild and re ect damage to the hepatocytes from absorbed toxins such as endotoxin and from poor perfusion resulting from hypotensive shock, dehydration, or both. Lactic acidemia may be present, re ecting poor tissue perfusion. Plasma protein rapidly drops, as protein is lost in the gastrointestinal tract, causing moderate to severe hypoalbuminemia and hypoglobulinemia. Peripheral or organ edema (vascular leak syndrome) may occur if hypoproteinemia is severe, coupled with systemic inflammation-induced increases in endothelial permeability. Hypokalemia, hyponatremia, hypochloridemia, and hypocalcemia are common electrolyte abnormalities in patients with enterocolitis. Metabolic acidosis may also be present and DIC is common. Urinalysis may reveal isosthenuria, proteinuria, hematuria, cylindruria, or glucosuria if hemodynamic or toxic renal injury is present. The number of leukocytes in the feces is usually elevated, and occult blood may be detected. PF is usually normal except when severe mural inflammation or colonic infarction occurs. S. enterica in feces is routinely detected by analyzing ve daily cultures of large samples (10-30 g) of feces using enrichment techniques.1,47,48 However, the sensitivity of fecal culture can be as low as 30% to 50%, even if several fecal samples collected daily are cultured.48 Concurrent culture of rectal biopsy specimens and feces increases the sensitivity of culture techniques to 60% to 75%.48 Currently, the PCR test is the most sensitive and rapid way to detect S. enterica in feces. A single PCR test applied early in the course of disease is a more sensitive test for the presence of S. enterica than repeated fecal cultures, 49,50 with as high as 100% sensitivity and 98% speci city for detection of organisms in some reports.51 Although detection of S. enterica organisms in feces does not prove a diagnosis of salmonellosis, the positive predictive value of either a positive PCR or culture result is high in horses with compatible clinical signs. Culture of peripheral blood may allow isolation of the organism if bacteremia or septicemia is present, but blood cultures are not a sensitive test for salmonellosis in adult horses. However foals are more likely than adults to become septicemic and culture of blood is recommended in all foals with signs of sepsis. Increased numbers of fecal leukocytes suggest an invasive process in the colon but are not specific for salmonellosis. Early in the course of the disease, dehydration, electrolyte and acid-base imbalances, endotoxemia, and sepsis may be life threatening. Aggressive treatment during the acute stages to replace uids lost in the diarrhea and to control sepsis and endotoxemia is often e ective in controlling the primary disease. Weight loss and hypoproteinemia are often severe. Possible complications include multiorgan dysfunction, vascular leak syndrome with peripheral and organ edema, laminitis, acute renal failure, venous thrombosis and septic phlebitis, irreversible protein-losing enteropathy or chronic malabsorption, pulmonary aspergillosis, and gastrointestinal infarction. The reader is referred to Chapter 7 and the section on endotoxemia in this chapter for additional information regarding treatment of horses with severe endotoxemia and SIRS. In many instances horses recover from acute salmonellosis with aggressive treatment, only to succumb to complications of the disease, which partially explains the high fatality rate of equine salmonellosis compared with that of human salmonellosis. Chronic, mild to moderate diarrhea is occasionally seen in horses after a bout of severe salmonellosis, usually with protein-losing enteropathy. If the chronic diarrhea persists beyond 4 to 5 weeks after the onset of signs, the prognosis for recovery is poor.16

Potomac Horse Fever PATHOGENESIS Potomac horse fever is caused by the obligate intracellular rickettsial organism Neorickettsia risticii (formerly called Ehrlichia risticii).52,53-56 The disease is most common from late summer to early fall, with a peak incidence in July and August.53,54 Potomac horse fever was rst described in the Northeastern United States but has since been described now in most areas of the continental United States, with a particularly high prevalence in the Northeast and Midwest. The geographical distribution is characterized by a signi cantly higher percentage of cases found along waterways and rivers.53,54 The disease occurs sporadically, both temporally and geographically, and can a ect any age group of horses. The case fatality rate ranges between 5% to 30%.53 Transmission of N. risticii has been reproduced experimentally by oral, intramuscular, intradermal, subcutaneous, and intravenous routes.53,57 Attempts to transmit the disease experimentally with ticks (Dermacentor variablis) or biting ies (Stomoxys calcitrans) were unsuccessful.58,59 Recently, N. risticii was found to infect virgulate cercariae, larval stages of trematodes that use operculate freshwater snails of the family Pleuroceridae (Juga spp. in California and Elimia spp. in Ohio and Pennsylvania), as intermediate hosts in their life cycle.60-63

Infected virgulate cercariae have been identi ed in aquatic snails collected in other parts of the world as well.64 Although the trematode species infected with N. risticii remain to be de nitively identi ed, at least two species have been identi ed as potential vectors,65 and at least two potential de nitive hosts were identi ed when N. risticii DNA was detected in the blood, liver, or spleen of 23 of 53 little and big brown bats harboring gravid trematodes in their intestinal tracts.66 Aquatic snails release large numbers of infected cercariae into water, where they seek their next intermediate host—any of a variety of aquatic insects.63,67 Successful transmission of N. risticii to horses was accomplished experimentally using trematode stages collected from Juga yrekaensis snails.68 The number of PCR-positive snails in endemic regions corresponds to the seasonal incidence of Potomac horse fever and may be as high as 26%.69 However, preliminary studies suggest that N. risticii may in fact be naturally transmitted to horses through the ingestion of caddisflies and mayflies.63,70 The pathogenesis of N. risticii is not completely understood. The organism infects and survives in monocytes and monocyte-derived leukocytes and can be found in blood monocytes during natural infections, but the sequence of events resulting in enterocolitis remains open to speculation. The organism appears rst to infect blood monocytes in experimentally infected horses, which may be the vehicle of organ infection.55,71 However, it is unclear whether leukocytes of the monocytic lineage or epithelial cells are infected rst in naturally infected horses. The target organ is the gastrointestinal mucosa, with the most severe lesions found in the large intestine.71,72 Infection of human colonic cells in vitro does not cause major cytopathologic e ects for several days.73 Disruption of the microvilli in the region of the plasma membrane where sodium chloride channels are located has been observed in human colonic cell cultures.73 Infection in horses is associated with variable degrees of morphologic damage.71,72 Mild morphologic damage and mononuclear cell in ltration of the lamina propria occur early during the infection, but brinous, necrotizing typhlocolitis with severe mucosal ulceration and in ammation of the lamina propria may occur later in the disease. Vasculitis and intravascular coagulation are consistent features in the large intestine, with perivascular edema.72 N. risticii can be observed in mucosal cells and macrophages and mast cells of the lamina propria.71,72 N. risticii can survive and multiply in macrophages by inhibiting the production of reactive oxygen intermediates and avoiding lysosomal digestion by blocking phagosome-lysosome fusion.74-76 Some researchers have suggested that impaired sodium chloride absorption in the colon contributes to diarrhea in infected horses and may be related to destruction of the enterocyte membrane structure in the region of sodium chloride channels.73,77 Direct injury to the mucosa by N. risticii and colonic in ammation are likely to be prominent features leading to diarrhea, especially later in the disease.72 Loss of fluid, protein, and electrolyte is likely due to mucosal injury and effects on enterocyte fluid secretion caused by the inflammatory response. Like other in ammatory conditions of the colon, systemic in ammation caused by absorption of bacteria and bacterial products is a potential complication of N. risticii infections if mucosal injury is severe, which contributes to the clinical signs seen during the disease.

CLINICAL SIGNS AND DIAGNOSIS N. risticii infection is clinically similar to other forms of enterocolitis and is characterized by anorexia, depression, and fever.53,72,78 Experimental infections produce a biphasic fever wherein the second febrile phase occurs 6 to 7 days after the rst.78,79 Decreased gastrointestinal motility, manifested as reduced borborygmi, occurs during the early stages, before the onset of diarrhea. Diarrhea is seen in 75% of cases, and occurs 2 days after the second fever episode during experimental infections.78,79 The diarrhea can be moderate to severe and dehydrating. Ileus can develop at any stage of the disease and can cause signs of moderate to severe colic. Systemic signs of endotoxemia, shock, and peripheral edema may occur and are similar to those described for salmonellosis. Experimental and natural infection with N. risticii can cause abortion of infected fetuses in pregnant mares.80,81 Laminitis is a complication in 20% to 30% of naturally occurring cases and is often severe.54 Other complications include protein-losing enteropathy, thrombosis, and renal failure, as described for salmonellosis. Hematologic abnormalities re ect endotoxemia, dehydration, and sepsis and are essentially identical to those described for salmonellosis. Neutropenia with a left shift is a consistent feature and occurs concurrently with or soon after the onset of diarrhea.79 Thrombocytopenia is common and often severe.79 Neutrophilic leukocytosis occurs later in the course of the disease. Hyper brinogenemia is usually more pronounced than that seen with salmonellosis. Serum electrolyte, acid-base, and biochemical abnormalities are also similar to those described for salmonellosis. Coagulopathies are commonly seen during N. risticii infection and re ect activation of coagulation pathways. DIC is not uncommon and may be the cause of the high frequency of laminitis associated with N. risticii infection.82 Diagnosis of N. risticii infection cannot be based solely on clinical signs because the disease is clinically similar to other forms of enterocolitis. However, in endemic areas, acute colitis is likely to be caused by N. risticii, and thus the clinical signs of acute in ammatory colitis may in fact have a high predictive value in these areas. Serologic evidence of infection, such as rising antibody titers to N. risticii detected by indirect immuno uorescence (IFA) or enzyme-linked immunosorbent assay (ELISA) in paired serum samples, may be helpful in establishing a diagnosis.54,83 Care should be taken when interpreting the IFA serologic test for N. risticii because the test appears to have a high false-positive rate.84 Culture of the organism from blood is possible but di cult and is generally useful only in the research laboratory.

Recently developed PCR tests for N. risticii DNA are rapid, highly sensitive (as sensitive as culture), and speci c tests for N. risticii infection that can be applied to blood or feces.85-87

Prevention Prevention of the disease by reducing exposure to the etiologic organism is di cult because the mode of transmission is not known. A killed vaccine has been developed that is relatively e ective in preventing clinical illness other than fever in 80% of experimentally challenged horses using the vaccine strain. However, eld studies suggest the vaccine has limited bene t for preventing natural infection or decreasing its severity.88,89 Vaccine failures have been attributed to strain differences in antigenicity or to poor antibody responses to the vaccine.88,89

Equine Intestinal Clostridiosis PATHOGENESIS Clostridiosis is an important cause of acute enterocolitis in foals and adult horses. Clostridium perfringens and Clostridium di cile are most commonly associated with intestinal clostridiosis in horses, but other clostridial species, including Clostridium septicum, Clostridicum cadaveris, and Clostridium sordelli have also been isolated from horses with enterocolitis.90-95 In horses of all ages, clostridial enterocolitis appears to be a common antibiotic-associated and nosocomial cause of enterocolitis.94,96,97 Hemorrhagic enterocolitis caused by C. perfringens in neonatal foals is a distinct clinical entity and will be discussed in more detail elsewhere. This discussion focuses on adult intestinal clostridiosis. The reader is referred to Chapter 21 for information regarding the disease in foals. Clostridium organisms are obligate anaerobic to aerotolerant spore-forming gram positive rods that are ubiquitous in the environment in the spore form.95 They are elements of the normal ora of horses of all ages and are among the rst bacteria acquired after birth. However, Clostridium organisms inhabiting the gastrointestinal tract are normally found in very low numbers and do not produce enterotoxins. Clostridiosis is associated with an increase in the number of a particular species of Clostridia in the gastrointestinal tract and, perhaps most important, exotoxin production. Although the conditions resulting in exotoxin production are not fully understood, several factors increase clostridial numbers in the gastrointestinal tract. Dietary factors a ect the numbers of Clostridium species shed in horse feces.90 Experimental induction of colic has been shown to increase fecal shedding of Clostridium species in the absence of diarrhea.98 Antibiotics, particularly those administered orally or recycled via the enterohepatic system, have been shown to increase the recovery of Clostridia colony-forming units in equine feces and clinical clostridiosis.91,93,99-101 Indeed, clostridiosis associated with C. di cile is likely to be the most important cause of antibiotic-induced enterocolitis in the horse. Clostridium perfringens C. perfringens includes many genetically distinct strains of variable virulence that produce one or more of a large group of exotoxins. The pattern of exotoxin production is used to classify C. perfringens into ve types: A, B, C, D, and E. C. perfringens type A is the most common clostridial isolate from healthy and diarrheic horses of all ages. Clostridium perfringens types A, B, C, and D have all been associated with hemorrhagic enteritis in foals younger than 10 days of age, with type C being the most common cause in North America. The primary toxin produced by C. perfringens type A is α-toxin (phospholipase C), which interferes with glucose uptake and energy production and activates arachidonic acid metabolism and signaling pathways in enterocytes.95 Oral administration of α-toxin does not cause tissue necrosis but causes increased secretion by small intestinal mucosal cells.24,102 The β toxin of types B and C is a cytotoxin that causes enterocyte necrosis, ulceration, and ultimately severe intestinal in ammation and hemorrhage.24,95 A novel toxin designated β 2 may also have a role in C. perfringens enterocolitis.103 The biologic activity of the β 2 toxin is similar to that of β toxin, but β 2 toxin is not related to β toxin in its genetic sequence. The β 2 toxin was prevalent in two groups of horses with acute enterocolitis but not in healthy horses.104 The β 2 toxin is predominantly associated with C. perfringens that would have otherwise been classi ed as type A but that may in fact represent a previously undescribed type. Virulent strains of C. perfringens type A and, to a lesser extent, type C may produce enterotoxin. Enterotoxin is a cytotoxin that inserts into cell membranes to form pores, which alter permeability to water and macromolecules and ultimately lead to cellular necrosis.105 Massive desquamation of the intestinal mucosa that is a result of enterotoxin cytotoxicity triggers an in ammatory response, intestinal edema, mural hemorrhage, and systemic in ammation.106 Enterotoxin also alters tight junction integrity, resulting in increased paracellular permeability by a noncytotoxic mechanism.107 Clostridium difficile C. difficile produces several toxins, only two of which, toxin A and toxin B, have been studied in detail. Toxin B is a potent cytotoxin in vitro, but its role in enterocolitis is less clear than that of toxin A. Toxin B does not induce uid secretion, in ammation, or characteristic

alterations in intestinal morphology. C. di cile toxin A is an enterotoxin that induces an in ammatory response with hypersecretory diarrhea.108 Toxin A induces neutrophil in ux into intestinal tissue, mast cell degranulation, and secretion of prostaglandins, histamine, cytokines, and 5-HT by these activated leukocytes.108-110 The products of neutrophils and mast cells have a significant role in the vasodilatory and secretory responses in the intestine during C. difficile infection. The enteric nervous system (ENS) is central to the induction of intestinal in ammation and mucosal secretion by toxin A. A model for toxin A–induced secretory diarrhea has emerged in which toxin A stimulates substance P–containing a erent sensory nerve bers, which in turn stimulate mast cell degranulation, recruitment and activation of polymorphonuclear leukocyte (PMN) and vasodilation111-113 Toxin A– induced stimulation of enterocyte secretion can occur via secretomotor neuronal stimulation by substance P–containing sensory neurons or products of mast cells and PMN. Mast cell degranulation, PMN in ux, and enterocyte secretion are all abolished by neural blockade or depletion of substance P. How toxin A triggers the sensory component of the ENS remains unknown, but it is likely that toxin A–induced necrosis of enterocytes exposes afferent neurons to the noxious milieu of the intestinal contents.

CLINICAL SIGNS AND DIAGNOSIS Equine intestinal clostridiosis is clinically similar to other forms of acute enterocolitis in horses.90,95 Although the clinical course is usually acute, peracute colitis with rapid death may occur. Occasionally, a milder, more prolonged clinical course is seen. Fever, anorexia, and depression may be observed before the onset of gastrointestinal signs, but the absence of prodromal signs is more common. Signs of endotoxemia and shock may accompany acute signs of colic and severe, dehydrating diarrhea. Diarrhea may not be profuse but is usually dark and foul. Like the clinical signs, hematologic and serum biochemical abnormalities are similar to those associated with other forms of enterocolitis and re ect uid, protein, and electrolyte loss and systemic in ammation resulting from endotoxemia. Neutropenia, leukopenia, and hemoconcentration are common. Hypoproteinemia may be profound. Hyponatremia, hypokalemia, hypochloremia, hypocalcemia, and a mixed prerenal-renal azotemia are often noted, as well as metabolic acidosis and coagulopathies. Serum concentrations of hepatocellular enzymes, such as sorbitol dehydrogenase, may be elevated, and liver function may be reduced. Preliminary diagnosis of equine intestinal clostridiosis caused by C. perfringens is based on the isolation of greater than 100 colonyforming units (CFU) of C. perfringens type A per gram of feces from patients with diarrhea and signs suggestive of toxemia.90,114 Similar criteria are used to screen human patients for C. perfringens type A infection. Normal horses shed less than 100 CFU/g of feces, and usually horses with intestinal clostridiosis shed greater than 106 CFU/g.90,114 However, identi cation of high numbers of Clostridium organisms in the feces does not prove infection. Detection of C. perfringens toxins in feces or intestinal contents in horses with high numbers of fecal CFU and clinical signs of enterocolitis is more conclusive evidence of an enterotoxigenic infection than that based on culture alone.95 Immunoassays are available that are primarily designed to detect C. perfringens enterotoxin.95 However, the reliability (speci city) of some immunoassays for diagnosis of C. perfringens infection has come into question. Recently, PCR multiplex and gene probe assays have been developed for detection of the major lethal toxins in bacterial isolates or fecal samples to determine the pattern of toxin production and are currently the preferred methods of detection.115-117 Like C. perfringens, diagnosis of C. di cile infection depends on culture of the organism from feces and identi cation of toxins in the feces. Bacterial culture of C. difficile may be di cult, and therefore it is an insensitive diagnostic test in horses.118,119 Enrichment techniques and culture of multiple fecal samples may be required.119,120 Detection of toxin A or B (or both) in feces by cell cytotoxicity assay or immunoassay is the preferred test for diagnosis of C. difficile infection in humans.95 These tests are more sensitive than bacterial culture for identifying C. di cile infection in adult horses.118,119 Sensitive PCR methods may also be used to identify genes for toxins A and B in fecal samples from diarrheic horses95.

Proliferative Enteropathy PATHOGENESIS Proliferative enteropathy is a chronic hyperplastic disorder of the small intestine that has been described in a wide variety of mammalian and avian species.121,122 The only etiologic agent identi ed to date that induces PE is the obligate intracellular pathogen Lawsonia intracellularis.122,123 The pig is the most frequently naturally affected species. However, reports of equine PE associated with L. intracellularis have increased in recent years.124-127 The relatedness of the strains of L. intracellularis causing PE in pigs and horses or even among other affected species is not known. There appears to be no host restriction because hamsters and other rodents can be infected with porcine strains of L. intracellularis. Before 2000 there were sporadic reports of PE in the literature describing isolated cases in horses.125-127 Since 2000 outbreaks on breeding farms have been described in Canada and anecdotally reported in the U.S.128 Although the epidemiology of PE is not known speci cally for horses, in other species the mode of transmission is fecal-oral, and large numbers of organisms can be shed in feces of infected animals.122 A ected animals shedding the organism in the feces serve as a source of

infection for herd mates. It is possible that nonequine species serve as reservoirs contributing to outbreaks on horse farms. Factors that increase the risk of PE in pigs include overcrowding, ration changes, transport, and weaning.122,123 Like pigs, horses are a ected as weanlings. Factors associated with weaning and other stresses may a ect immunity and increase susceptibility to infection. The incubation period is 2 to 3 weeks in nonequine species and is presumed to be similar in horses. Experimental L. intracellularis infection produces characteristic pathologic lesions in pigs and hamsters that are identical to lesions in horses with PE.122,123 A profound hyperplasia of the mucosa associated with proliferation of crypt epithelium and crypt hyperplasia is induced locally in infected islands of tissue that eventually extend to the entire distal jejunum and ileum. L. intracellularis preferentially infects proliferating cells, thus the tropism for the crypt epithelium. Infected cells proliferate far more rapidly than uninfected cells, suggesting that L. intracellularis directly induces the proliferative response. However, the molecular basis for enhanced proliferation is not known. L. intracellularis penetrates epithelial cells in a membrane-bound vesicle but eventually escapes the vacuole and is found free in the cytoplasm, concentrated at the apical pole of the cell. The gross pathologic lesions of equine PE are quite characteristic.124-127 Lesions may be segmental and are most commonly found in the ileum and terminal jejunum in horses. However, the duodenum may also be a ected. Severe mucosal hypertrophy is often observed but may wane during the chronic stages of the disease. The mucosa may become corrugated with focal erosions or ulcers. Submucosal edema is often easily identi ed on cut sections of a ected segments. Moderate to severe crypt hyperplasia with atrophy of intestinal villi is a consistent feature. Hyperplastic crypts are branched and may herniate into the submucosa. Necrosis, edema of the submucosal and lamina propria, hemorrhage, mononuclear in ammation, and muscular hypertrophy have been reported in a ected intestinal segments but are not consistent. Special stains such as silver stain are required to detect intracellular organisms. The organisms are curved or comma-shaped rods found clustered in the apical cytoplasm of hyperplastic crypt epithelium. The proliferative response of the intestinal mucosa alters absorption of nutrients and uid secretion by disrupting the architecture of the villi and by altering the maturation of epithelial cells into absorptive cells, accounting for the secretory diarrhea and often severe weight loss.121,123 The combined e ects of the in ammatory response and malabsorption may account for the clinically observed protein-losing enteropathy.

CLINICAL SIGNS AND DIAGNOSIS Weanling foals 4 to 6 months of age are most commonly a ected.124-127 The clinical signs of PE include ill thrift, weight loss, peripheral edema, diarrhea, and colic.124-127 The diarrhea is usually in the form of soft feces but may be profuse and watery. Some foals with mild diarrhea have black, tarry feces. Secondary complications such as gastric ulceration, bronchopneumonia, and parasitism may occur concurrently with the PE. Clinicopathologic features include mild to moderate anemia, moderate to severe hypoalbuminemia (often 0.5 cm) and evidence of colonic edema in horses with RDUC.199,200 Nuclear scintigraphy of horses after infusion with technetium-99–labeled white blood cells can be used to document in ammation of the right dorsal colon.201 Laparotomy or laparoscopic examination of the right dorsal colon may be required for de nitive diagnosis of RDUC. Other causes of enterocolitis, such as salmonellosis, Potomac horse fever, clostridiosis, and antibiotic-associated diarrhea, must be ruled out.

Cantharidin Toxicity PATHOGENESIS Cantharidin is the toxic substance found in beetles of the genus Epicauta, commonly known as blister beetles.202-204 Ingestion of the beetles causes release of the toxin and absorption through the gastrointestinal tract. Transcutaneous absorption may occur but appears to be rare in horses.203 Blister beetles feed on the owers of alfalfa and may be incorporated into processed alfalfa hay if the hay is cut and processed simultaneously, as by crimping.202-204 The beetles often swarm, and large numbers of beetles may be found in relatively small portions of hay. The lethal dose of cantharidin is less than 1 mg/kg, but the concentration of cantharidin varies among species of blister beetles and between sexes.202,203 As many as 100 to as few as six beetles may be lethal. Usually, only one or a few horses fed contaminated hay will ingest beetles because the beetles are concentrated in a small portion of the hay. However, outbreaks involving many horses on a farm have occurred. Most cases have occurred in Texas and Oklahoma, but horses in other states may be a ected as well, especially if hay is imported from states where blister beetles are common. Peak incidence is in late summer and fall.205 The fatality rate may be 50% or greater,202,206 but if the patient survives several days, recovery is probable. Cantharidin is absorbed from the gastrointestinal tract and excreted by the kidneys. Cantharidin is a potent irritant, causing acantholysis and vesicle formation when applied topically.202,204,206 The chemical is thought to disrupt oxidative metabolism in the mitochondria, causing mitochondrial swelling, plasma membrane damage, and changes in membrane permeability.202 The mucosa of the gastrointestinal tract is most commonly a ected in horses because they ingest the toxin. Cell swelling and necrosis occur, resulting in mucosal ulceration. Oral, esophageal, gastric, and small and large intestinal ulceration have been observed in natural and experimental canthariasis.202,204,206 Severe brinous to pseudomembranous in ammation and submucosal edema of the intestine have also been reported. Diarrhea probably results from the severe ulceration and in ammation of the large intestine, causing increased secretion of water, electrolytes, and protein and decreased absorption of uid. Large volumes of uid and protein are lost in the gastrointestinal tract, causing hemoconcentration and profound hypoalbuminemia in some affected horses.202,203,206 Cystitis and myocarditis occur in natural and experimentally induced cases of cantharidin toxicity.202,204,206 The toxin is excreted by the kidneys, and high concentrations of cantharidin in the urine induce cystitis. Occasionally, hemorrhagic cystitis may occur, resulting in hematuria or frank hemorrhage into the bladder.202 The cause of myocarditis and myocardial necrosis is unknown but may be a direct e ect of toxin on the myocardium. Increased plasma creatine kinase activity is often observed and has been postulated to arise from the damaged myocardium.202,203 A ected horses have a characteristically sti gait, but histopathologic evidence of skeletal muscle injury that explains the elevated plasma creatine kinase activity has not been observed.203 The kidneys are often pale, swollen, and moist, with occasional infarcts.204 Hypocalcemia and hypomagnesemia are biochemical features of cantharidin toxicity in horses that have not been explained.202,203,206 Hypocalcemia may result from hypoalbuminemia, but the ionized calcium concentration is often decreased, indicating that hypoalbuminemia is not responsible for the hypocalcemia.203

CLINICAL SIGNS AND DIAGNOSIS

Cantharidin toxicity can cause a range of clinical signs, from mild depression and abdominal discomfort to fulminant signs of toxemia and rapid death, depending on the ingested dose of toxin.202,203,206 Most commonly, clinical signs include depression, sweating, irritability, abdominal pain, elevated heart and respiratory rates, fever, polyuria, polydipsia, and profuse diarrhea.202,203,206 Blood is rarely seen in the feces. Stranguria and pollakiuria are common.202 Signs of hypocalcemia include synchronous diaphragmatic utter and tremors. A sti and stilted gait may be evident. Neurologic signs such as head pressing, swaying, and disorientation may be noted.206 Signs of systemic in ammation resulting from endotoxemia may be seen in severe cases. Some horses develop severe depression and toxemia and may die within hours of ingesting cantharidin without developing diarrhea.202,206 Hematologic abnormalities include hemoconcentration and neutrophilic leukocytosis.202,203 Occasionally, neutropenia and leukopenia may accompany endotoxemia. Serum biochemical analysis usually reveals increased creatine kinase activity, hypocalcemia, and hypoalbuminemia.202,203 Biochemical abnormalities include hypocalcemia (both ionized and total calcium concentrations), hypomagnesemia, and azotemia.202,203,206 Urine specific gravity is characteristically in the hyposthenuric range.202,203 Microscopic hematuria and mild proteinuria may be evident. Fecal occult blood is often present, but hematochezia is unusual. A tentative diagnosis can be made on the basis of clinical signs and identi cation of blister beetles in the hay. Determining the species of the insects may be necessary to estimate the amount of cantharidin ingested. All species of Epicauta contain cantharidin, but some have small amounts. Definitive diagnosis requires the measurement of the cantharidin concentration in gastric or intestinal contents and urine.202,205

Arsenic Toxicosis PATHOGENESIS Arsenic toxicosis is an unusual cause of diarrhea in horses, resulting from ingestion of arsenic-containing herbicides, insecticides, and other pest-control products contaminating water or roughage used as a food source.207 The toxicity of arsenic depends on the valence of the element.207,208 Arsenate may be reduced to arsenite in mammalian systems.208 Arsenite is thought to be more toxic than arsenate and less rapidly excreted in urine.208 Arsenate and arsenite uncouple oxidative phosphorylation, leading to breakdown of energy metabolism in the cells of many tissues.208 Widespread cellular injury and death occur rapidly during acute arsenic toxicosis. Multiorgan failure is usually the result. In fact, cardiomyopathy and pulmonary disease are common causes of death in humans.209 Damage to the large intestine is probably due in part to direct cellular toxicity and corrosion by the compound. However, vasculitis is a hallmark of the disease in humans and horses and is thought to be the most important mechanism of large intestinal disease in humans.207,210 Acute hemorrhagic colitis is a feature of arsenic toxicosis, with severe mural edema and mucosal ulceration.207 Profuse hemorrhagic diarrhea and abdominal pain result. Chronic arsenic toxicity can occur but appears to be rare in horses.

CLINICAL SIGNS AND DIAGNOSIS Depression, weakness, abdominal pain, hemorrhagic diarrhea, and shock are characteristic of acute arsenic toxicosis in horses.207 Death may occur before diarrhea is evident. Initial clinical signs may be di cult to distinguish from other peracute forms of colitis and are related to endotoxic shock, metabolic disturbances, and dehydration. Later, cardiac arrhythmias, pulmonary edema, acute renal failure, and neurologic deficits (ataxia and stupor) may develop.207 Anuria or polyuria may be observed. Hemolytic anemia caused by preferential binding of arsenic compounds to red blood cells is a feature of arsenic poisoning in humans.209 Hematologic abnormalities resulting from injury to bone marrow cells and ongoing hemolysis may be seen after the peracute stage. Leukopenia and thrombocytopenia have been described in human patients.209 Serum biochemical analysis may reveal azotemia, hepatocellular enzyme activities higher than generally attributed to endotoxemia, and increased creatine kinase activity.207 Urine speci c gravity may be in the isosthenuric range, with hematuria, cylindruria, and proteinuria evident by urinalysis. Diagnosis may be possible by measuring blood and urine arsenic concentration, but these tests may not be diagnostic. Postmortem diagnosis is con rmed by measuring arsenic concentrations in liver and kidney samples.207 History of exposure and clinical signs remain the primary means of diagnosis.

MISCELLANEOUS DISORDERS OF THE LARGE INTESTINE Intestinal Anaphylaxis PATHOGENESIS Severe intestinal anaphylaxis is a syndrome in horses characterized by peracute, rapidly fatal colitis.211 The severe syndrome is clinically and pathologically similar to other known causes of peracute colitis, such as salmonellosis, clostridiosis, and antibiotic-associated diarrhea. Some cases are less severe and manifest as mild to moderate diarrhea or colic (or both). The syndrome of intestinal anaphylaxis can be produced

by either an IgE-mediated Type I hypersensitivity or an IgE-independent anaphylactoid reaction.212,213 Intestinal anaphylaxis is usually induced by local gastrointestinal exposure to a food, environmental, drug, or other allergen 212,214 but may also occur with systemic exposure to an allergen.215-217 Massive mast cell degranulation, secretion of in ammatory mediators, and activation of enteric neural re exes in the intestine causes profound alterations in blood flow, increased vascular permeability and interstitial edema, recruitment of neutrophils, altered motility, mucosal injury, absorption of microbial products, and mucosal hypersecretion.218-222 Systemic signs may be caused by the anaphylactic reaction or may be associated with systemic in ammation triggered by microbial products (endotoxin) absorbed through the injured and hyperpermeable mucosa. Intestinal anaphylaxis in horses may be a peracute, fulminant enterocolitis with endotoxemia that may be fatal.211,223 This form is characterized by severe intramural edema and hemorrhagic in ammation of the large intestine, often producing submucosal thickening on the order of many centimeters. Vascular thrombosis may be widespread, with mucosal and serosal petechia and ecchymoses. Less severe forms of intestinal anaphylaxis may manifest as patchy areas of intestinal edema and congestion.215 Diarrhea results from intestinal in ammation initiated by the type I hypersensitivity response. Many of the mediators of type I hypersensitivity, such as histamine and 5-HT, have well-documented stimulatory e ects on mucosal secretory activity, vascular and epithelial permeability, and motility218-220 in the intestine. Systemic in ammation resulting from endotoxemia may be overwhelming once the mucosal barrier breaks down. Infarction of intestinal segments and other organs may result from intravascular coagulation. Ileus, abdominal distention, and moderate to severe abdominal pain may result from motility disturbances and infarction of the large intestine.

CLINICAL SIGNS AND DIAGNOSIS The clinical signs are similar to those described for other forms of peracute colitis. However, the severity may be variable, manifesting as colic or moderate diarrhea. Characteristically, severe shock, signs of systemic in ammation resulting from endotoxemia, and severe metabolic disturbances are observed.211,223 Heart and respiratory rates may be markedly elevated, with other signs of cardiovascular collapse, such as weak and thready peripheral pulses and peripheral vasoconstriction. However, peripheral vasodilation may be seen later in the course of disease. Dark red, muddy, or cyanotic mucous membranes with a prolonged capillary re ll time are consistent with sepsis. Borborygmi are usually absent, and abdominal tympany may be heard on percussion, secondary to ileus. Moderate to severe colic may accompany ileus. Severe diarrhea is possible, but death may occur before diarrhea is evident. Multiorgan failure resulting from DIC is not unusual. The rapid onset of weakness, staggering, and trembling commonly precedes death. The syndrome may cause death in 4 to 24 hours. Hematologic abnormalities include severe neutropenia and leukopenia, thrombocytopenia, and hemoconcentration.211 Serum biochemical alterations include hyponatremia, hypokalemia, hypocalcemia, and severe metabolic acidosis. Blood urea nitrogen and creatinine may be elevated from prerenal or renal azotemia. If acute renal failure occurs, hyperkalemia may result. Hepatocellular enzyme activity may be increased on account of endotoxemia. Severe coagulopathies are common, resulting in prolonged coagulation times, decreased antithrombin III activity, and increased plasma concentration of brin degradation products. Analysis of PF may be valuable because infarction of the large intestine is not unusual. Protein concentration and the white blood cell count may be elevated. Red blood cell counts are less likely to be elevated because infarction, rather than strangulation, of the intestine occurs. Diagnosis is based on clinical signs, postmortem ndings, and exclusion of other causes. Cultures and toxicologic analysis of fecal samples and gastrointestinal tissues fail to demonstrate a clear cause. Other diagnostic tests are also inconclusive. If an antigen is suspected as the trigger of the anaphylaxis, a Prausnitz-Kustner passive cutaneous anaphylaxis sensitization test can con rm the presence of antigen-speci c IgE in the patient serum.215

Carbohydrate Overload PATHOGENESIS Overeating of soluble carbohydrates, especially so-called hot grains such as corn, overwhelms the digestive capability of the small intestine, resulting in a high percentage of soluble carbohydrates entering the large intestine. The amount of soluble carbohydrates that will produce diarrhea varies according to the previous dietary history of the individual. Horses fed diets higher in soluble carbohydrates are more resistant to the deleterious e ects of carbohydrate overload. Gradual accommodation to a diet high in carbohydrates can be accomplished over several weeks. However, horses fed an unusually large amount of grains or other form of soluble carbohydrates often develop diarrhea and may, depending on the amount ingested, develop severe colitis, systemic in ammation resulting from endotoxemia, metabolic acidosis, and laminitis.224-227 The pathogenesis of colitis from carbohydrate overload is primarily due to the toxic e ects on the microbial ora in the large intestine.225 A sudden delivery of soluble carbohydrates to the large intestine causes rapid fermentation by gram positive lactic acid– producing bacteria and a sudden increase in organic acid production.226 The cecal pH level rapidly decreases, and the lactic acid

concentration rapidly increases.226 Rapid organic acid production overwhelms the bu ering capacity of the large intestine, not only by directly depleting the bu ers found in the contents but also by reducing the e ciency of bu er secretion. Bicarbonate secretion is linked to absorption of volatile fatty acids, which are produced in low amounts by fermentation of soluble carbohydrates. The contents of the large intestine become profoundly acidic, resulting in unfavorable conditions for the microbial ora. Lactic acid–producing bacteria ourish, whereas the gram negative bacteria, especially the Enterobacteriaceae, are killed in large numbers by the acids. Large quantities of endotoxin are released from the dying bacteria.226 The osmotic load from lactic acid produced in the large intestine is an important factor in the development of diarrhea because organic acids such as lactic acid are poorly absorbed. Mild cases of carbohydrate overload may result purely from osmotic diarrhea. In more severe cases the acidic contents of the large intestine are toxic to the mucosa, causing necrosis of the mucosal tissues similar to that seen in ruminal acidosis. Mucosal ulceration allows absorption of large quantities of endotoxin and lactic acid produced by the massive die-o of acidintolerant microbes and fermentation of soluble carbohydrates, normally poorly absorbed by intact mucosa.227 Systemic in ammation resulting from endotoxemia may be overwhelming, and profound metabolic acidosis may occur. Secretory diarrhea caused by the direct e ects of acid luminal contents on the mucosa, as well as the e ects of in ammatory mediators on enterocyte secretion, worsens the acidosis and dehydration. Systemic in ammation resulting from endotoxemia, along with intestinal in ammation, adversely a ects intestinal motility, and ileus develops. Ileus and gas production from fermentation of carbohydrates may cause severe distention of the large intestine and signs of abdominal pain. Laminitis is a frequent complication of endotoxemia and lactic acidosis. In fact, carbohydrate overload is used to induce laminitis as an experimental model because of the consistency of the laminitis produced.225-227

CLINICAL SIGNS AND DIAGNOSIS Clinical signs of colitis from carbohydrate overload can vary according to the amount of carbohydrate ingested and accommodation of the ora to a high carbohydrate diet. Mild cases may result in a transient osmotic diarrhea, with no systemic e ects. Severe cases are characterized by signs similar to those described for other forms of colitis, including abdominal pain, moderate to severe diarrhea, and dehydration. Signs of endotoxemia and sepsis are frequently present in severe cases. Elevated heart and respiratory rates are common, with peripheral vasoconstriction early in the disease, followed by peripheral vasodilation as the disease progresses. Depression may be profound as a result of metabolic acidosis and endotoxemia. Abdominal auscultation and percussion may reveal ileus and intestinal tympany. Nasogastric intubation may yield signi cant gastric acidic re ux. Particles of grain may be noted in the gastric re ux and the feces if grain overload is the source of the carbohydrate overload. Laminitis may complicate both mild and severe cases of carbohydrate overload, especially if the animal has had previous bouts of laminitis. Hematologic abnormalities include neutropenia and leukopenia. Severe dehydration may result in profound hemoconcentration. Protein loss later in the course of disease may result in hypoproteinemia. Serum biochemical abnormalities include azotemia, elevated hepatocellular enzyme activity, hyponatremia, and hypokalemia. Severe hypocalcemia and metabolic acidosis are characteristic of the disease. Serum lactate concentrations are elevated in the absence of evidence of intestinal strangulation or infarction. PF analysis often reveals no abnormalities.

Sand Enteropathy Sand enteropathy is described in more detail under the heading of Obstructive Diseases, because acute obstruction is often associated with abnormally large amounts of sand in the large intestine.228 However, chronic sand-induced diarrhea is a distinct syndrome that can occur at any age as a result of the abnormal accumulation of sand in the large intestine.229,230 Chronic diarrhea and signs of colic may be seen without obstruction. The pathogenesis of sand accumulation in individual horses, other than simple ingestion of large quantities, is unclear, as described elsewhere in this chapter. Presumably, the sand causes irritation and may disrupt motility, leading to diarrhea. The diarrhea is usually not severe and dehydrating and may be intermittent. Weight loss is characteristic and can be severe in some cases.229 Complications may occur, such as peritonitis and acute obstruction.229 Diagnosis is usually based on the presence of abnormal amounts of sand in the feces. Because sand-induced chronic diarrhea is primarily associated with sand accumulation in the ventral colon, auscultation of the ventral abdomen immediately behind the xiphoid process may reveal characteristic sand sounds.231 This technique is sensitive only if peristalsis is present. Ultrasonography may also be useful to identify sand in the ventral colon, but it is not useful to quantitate the amount of sand. Occasionally, radiography may be required to detect sand in the colon.229

PRINCIPLES OF THERAPY FOR ACUTE DIARRHEA The principles of therapy for acute diarrhea resulting from colitis are similar regardless of the cause and include replacement of uid and electrolyte losses, control of colonic in ammation, reduction of uid secretion, promotion of mucosal repair, control of endotoxemia and sepsis, and re-establishment of normal ora. The purpose of this section is to review the principles of therapy, with references to speci c therapies for particular etiologies as they arise.

Fluid Replacement and Circulatory Support Replacement of uid and electrolyte losses is of primary concern in treating horses with salmonellosis. Depending on the severity of the disease, uid losses may be minimal or massive. Fluid and electrolytes can be administered orally or intravenously. Some horses with mild to moderate diarrhea may maintain hydration and electrolyte balance by consuming water and electrolytes voluntarily. Fresh water and water containing electrolytes should be available in all cases. In many instances, periodic nasogastric intubation and administration of water and electrolytes through the tube may be su cient to maintain hydration.232 In severe cases indwelling nasogastric tubes can be maintained, and up to 8 L of uid can be administered by the tube every 20 to 30 minutes, if ileus is not evident. However, intravenous administration of fluids is preferred in most cases, requiring significant quantities of fluid to replace and maintain hydration and electrolyte balance.233 It is not unusual for patients with severe diarrhea to require large volumes (50-100 L/day) of intravenous uids to maintain hydration. Frequent monitoring of packed cell volume, serum electrolyte concentration, venous blood gases or total serum carbon dioxide, blood urea nitrogen and creatinine, urine protein and cytology, and body weight is important to monitor hydration, electrolyte and acid-base balance, and renal function. Isotonic sodium chloride or lactated Ringer’s solution is frequently used to restore and maintain uid and electrolyte balance. Potassium chloride can be added to the uids and administered at a rate up to 0.5 to 1.0 mEq/kg/hr. Generally, a rate of less than 0.5 mEq/kg/hr is used. Hypertonic NaCl solutions (1-2 L of 5% or 7.5% NaCl) have been used in horses that are severely hyponatremic (60 beats/min in most cases). In addition, re ux is typically obtained after passage of a stomach tube, and loops of distended small intestine are usually detected on rectal palpation of the abdomen.11 However, these latter ndings are variable, depending on the duration and location of the obstruction. For example, horses with ileal obstructions tend to re ux later in the course of the disease process than horses with jejunal obstructions. Furthermore, a horse that has an entrapment of small intestine in the epiploic foramen or a rent in the proximal small intestinal mesentery may not have palpable loops of small intestine because of the cranial location of these structures.12 Abdominocentesis can provide critical information on the integrity of the intestine and is indicated in

horses with suspected strangulation of the small intestine.11 A horse that has signs compatible with a small intestinal obstruction and also has serosanguinous abdominal uid with an elevated protein level (>2.5mg/dl) is likely to require surgery, although these cases must be di erentiated from proximal enteritis. In general, horses with small intestinal strangulation show continued signs of abdominal pain, whereas horses with proximal enteritis (discussed earlier in this chapter) tend to be depressed after initial episodes of mild abdominal pain.13 In addition, horses with small intestinal strangulation continue to clinically deteriorate despite appropriate medical therapy and begin to show elevated white blood cell counts (>10,000 cells/ l) in the abdominal uid, likely as the duration of strangulation increases. However, there are cases in which small intestinal strangulation and proximal enteritis cannot be not readily distinguished, at which point surgery may be elected rather than prolonging the decision to perform abdominal exploration on a horse with a potential strangulating lesion.13

Prognosis The prognosis for survival in horses with small intestinal strangulating lesions is generally lower than for most forms of colic.14 However, recent studies indicate that more than 80% of horses with small intestinal strangulating lesions will be discharged from the hospital.15 Nonetheless, owners should be warned that the long-term survival rates are substantially reduced to below 70%,16 in part because of longterm complications such as adhesions.18,19 In addition, the prognosis is particularly low for some forms of strangulation, including entrapment of small intestine within a mesenteric rent.19

Epiploic Foramen Entrapment The epiploic foramen is a potential opening (because the walls of the foramen are usually in contact) to the omental bursa located within the right cranial quadrant of the abdomen. It is bounded dorsally by the caudate process of the liver and caudal vena cava and ventrally by the pancreas, the hepatoduodenal ligament, and the portal vein.12 Intestine may enter the foramen from the visceral surface of the liver toward the right body wall or the opposite direction. Studies di er as to which is the most common form. In the case of entrapments that enter the foramen in a left-to-right direction, the omental bursa is ruptured as the intestine migrates through the epiploic foramen, which may contribute to the intra-abdominal hemorrhage often seen with this condition. Clinical signs include acute onset of severe colic with examination ndings compatible with small intestinal obstruction. A recent study has shown that the stereotypic behavior of crib biting is a signi cant risk factor for epiploic foramen entrapment.20 The reason for this is unclear but may be attributable to changes in abdominal pressure as the horse prepares the esophagus to ingest air. Other risk factors include increased height of the horse and previous colic surgery.20 The condition was once believed to be more prevalent in older horses,12 but this has since been refuted by more recent studies.21 However, the disease has also been recognized in foals as young as 4 months of age.22 The diagnosis is de nitively made at surgery, although ultrasonographic ndings of distended loops of edematous small intestine adjacent to the right middle body wall are suggestive of epiploic foramen entrapment.12 In general, thickened, amotile intestine on ultrasonographic examination is highly predictive for small intestinal strangulating obstruction.23 Small intestine entrapped in the epiploic foramen may be limited to a portion of the intestinal wall (parietal hernia),24 and the large colon may become entrapped within the epiploic foramen.25 In treating epiploic foramen entrapment, the epiploic foramen must not be enlarged either by blunt force or with a sharp instrument, because rupture of the vena cava or portal vein and fatal hemorrhage may occur. Prognosis has substantially improved over the last decade, with current short-term survival rates (discharge from the hospital) ranging from 74% 26 to 79%.12 Preoperative abdominocentesis has been consistently found to be the most predictive test of postoperative survival.12,26

Strangulation by Pedunculated Mesenteric Lipoma As horses age, lipomas form between the leaves of the mesentery and develop mesenteric stalks as the weight of the lipoma tugs on the mesentery. The stalk of the lipoma may subsequently wrap around a loop of small intestine or small colon, causing strangulation. Strangulating lipomas should be suspected in aged (>15 years) geldings with acute colic referable to the small intestinal tract.27,28 Ponies also appear to be at risk for developing disease,28 suggesting alterations in fat metabolism may predispose certain horses to development of mesenteric lipomas. The diagnosis is usually made at surgery, although on rare occasions a lipoma can be palpated rectally.29 Treatment involves surgical resection of the lipoma and strangulated bowel, although strangulated intestine is not always nonviable.27 Studies indicate that approximately 50% 28 to 78% 27 of horses are discharged from the hospital after surgical treatment.

Small Intestinal Volvulus A volvulus is a twist along the axis of the mesentery, whereas torsion is a twist along the longitudinal axis of the intestine. Small intestinal volvulus is theoretically initiated by a change in local peristalsis, or the occurrence of a lesion around which the intestine and its mesentery may twist (e.g., an ascarid impaction).11 It is reportedly one of the most commonly diagnosed causes of small intestinal obstruction in foals.30,31 It has been theorized that young foals may be at risk for small intestinal volvulus because of changing feed habits and adaptation to

a bulkier adult diet. Onset of acute, severe colic; a distended abdomen; and radiographic evidence of multiple loops of distended small intestine in a young foal would be suggestive of small intestinal volvulus. However, it is not possible to di erentiate volvulus from other causes of small intestinal obstruction preoperatively. In adult horses volvulus frequently occurs in association with another disease process, during which small intestinal obstruction results in distention and subsequent rotation of the small intestinal around the root of the mesentery. Although any segment of the small intestine may be involved, the distal jejunum and ileum are most frequently a ected because of their relatively longer mesenteries.11 The diagnosis is made at surgery by palpating a twist at the origin of the cranial mesenteric artery. Treatment includes resection of devitalized bowel, which may not be an option because of the extent of small intestinal involvement (similar to large colon volvulus). Prognosis is based on the extent of small intestine involved and its appearance after surgical correction of the lesion. In general, horses in which more than 50% of the small intestine is devitalized are considered to have a grave prognosis.32

Strangulation by Way of Mesenteric or Ligamentous Rents There a number of structures that, when torn, may incarcerate a segment of intestine (typically the small intestine), including intestinal mesentery,19 the gastrosplenic ligament,33 the broad ligament,34 and the cecocolic ligament.35 Horses with such incarcerations exhibit presenting signs typical of a horse with strangulating small intestine, including moderate to severe signs of abdominal pain, endotoxemia, absent gastrointestinal sounds, distended small intestine on rectal palpation, nasogastric re ux, and serosanguinous abdominal uid. However, the prognosis for many of these cases appears to be lower than for horses with other types of small intestinal strangulations. For example, in horses with small intestine entrapped in a mesenteric rent, only 7 of 15 horses were discharged from the hospital, and only 2 of 5 horses for which follow-up information was available survived long term (>5 months).19 Poor outcome may result from the di culty in unentrapping incarcerated intestine, the degree of hemorrhage, and the length of intestine affected.

Inguinal Hernia Inguinal hernias are more common in Standardbred and Tennessee Walking horses that tend to have congenitally large inguinal canals.11 Inguinal hernias may also occur in neonatal foals but di er from hernias in mature horses in that they are typically nonstrangulating. The nature of the hernia (direct versus indirect) is determined on the basis of the integrity of the parietal vagin*l tunic. In horses in which the bowel remains within the parietal vagin*l tunic, the hernia is referred to as indirect because, strictly speaking, the bowel remains within the peritoneal cavity. Direct hernias are those in which strangulated bowel ruptures through the parietal vagin*l tunic and occupies a subcutaneous location. These direct hernias most commonly occur in foals and should be suspected when a congenital inguinal hernia is associated with colic, swelling that extends from the inguinal region of the prepuce, and intestine that may be palpated subcutaneously.36,37 Although most congenital indirect inguinal hernias resolve with repeated manual reduction or application of a diaper, surgical intervention is recommended for congenial direct hernias.36 Historical ndings in horses with strangulating inguinal hernias include acute onset of colic in a stallion that had recently been used for breeding. A cardinal sign of inguinal herniation is a cool, enlarged testicl* on one side of the scrotum (Figure 15-24).38,39 However, inguinal hernias have also been reported in geldings.40 Inguinal hernias can also be detected on rectal palpation, and manipulation of herniated bowel per rectum can be used to reduce a hernia but is not recommended because of the risk of rectal tears. In many cases the short segment of herniated intestine will markedly improve in appearance once it has been reduced and can sometimes be left unresected.41 The a ected testicl* will be congested because of vascular compromise within the spermatic cord, and although it may remain viable, it is generally recommended that it be resected.41 The prognosis in adult horses is good, with up to 75% of horses surviving to 6 months of age.39 Horses that have been treated for inguinal hernias may be used for breeding. In these horses the remaining testicl* will have increased sperm production, although an increased number of sperm abnormalities will be noticed after surgery because of edema and increased temperature of the scrotum.

FIGURE 15-24 inguinal canal.

Inguinal hernia in a horse with colic. Note the enlarged testicl*, which has compromised venous drainage because of herniated small intestine within the

Strangulating Umbilical Hernias Although umbilical hernias are common in foals, strangulation of herniated bowel is rare. In one study 6 of 147 (4%) horses with umbilical hernias had incarcerated intestine.42 Clinical signs include a warm, swollen, rm, and painful hernia sac associated with signs of colic. The a ected segment of bowel is usually small intestine, but herniation of cecum or large colon has also been reported.43 In rare cases a hernia that involves only part of the intestinal wall may be found; this is termed a Richter’s hernia. In foals that have a Richter’s hernia, an enterocutaneous stula may develop.43 In one study 13 of 13 foals with strangulating umbilical hernias survived to discharge, although at least three were lost to long-term complications.43

Intussusceptions An intussusception involves a segment of bowel (intussusceptum) that invagin*tes into an adjacent aboral segment of bowel (intussuscipiens). The reason for such invagin*tion is not always clear, but it may involve a lesion at the leading edge of the intussusception, including small masses, foreign bodies, or parasites. In particular, tapeworms (Anoplocephala perfoliata) have been implicated.44 Ileocecal intussusceptions are the most common intestinal intussusceptions in the horse and typically a ect young animals. In one study evaluating 26 cases of ileocecal intussusception, the median age of the horses was 1 year old.45 Acute ileocecal intussusceptions are those in which the horses has a duration of colic of less than 24 hours and involve variable lengths of intestine, which ranged in one study from 6 cm to 457 cm in length. In acute cases the involved segment of ileum typically has a compromised blood supply. Chronic ileocecal intussusceptions typically involve short segments of ileum (up to 10 cm in length), and the ileal blood supply is frequently intact.45 Abdominocentesis results are variable because strangulated bowel is contained within the adjacent bowel. There is often evidence of obstruction of the small intestine, including nasogastric re ux and multiple distended loops of small intestine on rectal palpation. Horses with chronic ileocecal intussusceptions have mild, intermittent colic, often without evidence of small intestinal obstruction. In one study a mass was palpated in the region of the cecal base in approximately 50% of cases.44 Transabdominal ultrasound may be helpful in discerning the nature of the mass. The intussusception has a characteristic target appearance on cross section.46 Other segments of the small intestine may also be intussuscepted, including the jejunum (Figure 15-25). In one study of 11 jejunojejunal intussusceptions, the length of bowel involved ranged between 0.4 to 9.1 m.47 Attempts at reducing intussusceptions at surgery are usually futile because of intramural swelling of a ected bowel. Jejunojejunal intussusceptions should be resected.

FIGURE 15-25

Jejunojejunal intussusception in a horse with colic. Note the intussusceptum, which has become ischemic as a result of invagin*tion of intestine and its mesenteric blood supply into the intussuscipiens.

For acute ileocecal intussusceptions the small intestine should be transected as far distally as possible, and a jejunocecal anastomosis should be performed. In cases with particularly long intussusceptions (up to 10 m has been reported), an intracecal resection may be attempted.48 For horses with chronic ileocecal intussusceptions, a jejunocecal bypass without small intestinal transection should be performed. The prognosis is good for horses with chronic ileocecal intussusceptions and guarded to poor for horses with acute ileocecal intussusceptions, depending on the length of bowel involved.45

Diaphragmatic Hernias Herniation of intestine through a rent in the diaphragm is rare in the horse. Any segment of bowel may be involved, although small intestine is most frequently herniated.49 Diaphragmatic rents may be congenital or acquired, but acquired hernias are more common.49 Congenital

rents may result from incomplete fusion of any of the four embryonic components of the diaphragm: pleuroperitoneal membranes, transverse septum, body wall, and esophageal mesentery.49 In addition, abdominal compression of the foal at parturition may result in a congenital hernia.49 Acquired hernias are presumed to result from trauma to the chest or a sudden increase in intra-abdominal pressure, such as might occur during parturition, distention of the abdomen, a sudden fall, and strenuous exercise.50 Hernias have been located in a number of di erent locations, although large congenital hernias are typically present at the ventral-most aspect of the diaphragm and most acquired hernias are located at the junction of the muscular and tendinous portions of the diaphragm.49 A peritoneopericardial hernia has been documented in at least one horse.51 The clinical signs are usually associated with intestinal obstruction rather than respiratory distress.50 However, careful auscultation may reveal an area of decreased lung sounds associated with obstructed intestine and increased uid within the chest cavity.52 Such signs may prompt thoracic radiography or ultrasound, both of which can be used to make a diagnosis. Auscultation may also reveal thoracic intestinal sounds, but it is typically not possible to di erentiate these from sounds referred from the abdomen. In one report, two of three horses diagnosed with small intestinal strangulation by diaphragmatic hernia had respiratory acidemia, attributable to decreased ventilation.53 Treatment of horses with diaphragmatic hernia is fraught with complications because of the need to reduce and resect strangulated bowel and the need to repair the defect in the diaphragm.54,55 Because dorsal defects in the diaphragm are among the most common forms of diaphragmatic defect, it may not be possible to close the diaphragmatic hernia by way of the approach used for abdominal exploratory. However, because herniation is likely to recur,54 it is appropriate to schedule a second surgery using an appropriate approach to resolve the diaphragmatic defect.

LARGE COLON VOLVULUS Clinical Signs Horses with large colon volvulus have rapid onset of severe, unrelenting abdominal pain. Postpartum broodmares appear to be at risk for this form of colic.4 Once the large colon is strangulated (>270-degree volvulus), gas distention is marked, leading to gross distention of the abdomen, compromised respiration as the distended bowel presses up against the diaphragm, and visceral pooling of blood as the caudal vena cava is compressed. Horses with this condition are frequently refractory to even the most potent of analgesics. These horses may prefer to lie in dorsal recumbency, presumably to take weight o the strangulated colon. An abbreviated physical examination is warranted in these cases because the time from the onset of strangulation to surgical correction is critical. Under experimental conditions the colon is irreversibly damaged within 3 to 4 hours of a 360-degree volvulus of the entire colon.56 Despite severe pain and hypovolemia, horses may have a paradoxically low heart rate, possibly related to increased vagal tone. In addition, results of abdominocentesis often do not indicate the degree of colonic compromise,4,57 and in many cases it is not worth attempting to obtain abdominal uid because of extreme colonic distention.58 Rectal palpation reveals severe gas distention of the large colon, frequently associated with colonic bands traversing the abdomen. In many cases colonic distention restricts access to the abdomen beyond the pelvic brim. A recent study has shown that plasma lactate levels below 6.0 mmol/L had a sensitivity of 84% and a speci city of 83% in predicting survival in horses with large colon volvulus.59

Surgical Findings At surgery the volvulus is typically located at the mesenteric attachment of the colon to the dorsal body wall, and the most common direction of the twist is dorsomedial when the right ventral colon is used as a reference point.4 However, the colon may twist in the opposite direction, twist greater than 360 degrees (up to 720 degrees has been reported), or twist at the level of the diaphragmatic and sternal exures.4 In all cases the colon should be decompressed as much as possible, and in many cases a colonic evacuation by way of a pelvic exure enterotomy will greatly aid correction of the volvulus. After correction of the volvulus, a determination must be made as to whether the colon has been irreversibly injured. This should be based on mucosal color and bleeding (if an enterotomy has been performed), palpation of a pulse in the colonic arteries, serosal color, and appearance of colonic motility.7 If the colon is judged to be irreversibly damaged, the feasibility of a large colon resection can be considered. Although 95% of the colon can be resected (that part of the colon distal to the level of the cecocolic fold), damage from the volvulus usually exceeds that which can be resected. In these cases surgeons may elect to resect as much damaged bowel as possible or advise euthanasia.7

Prognosis The prognosis is guarded to poor because of the rapid onset of this disease. In one study the survival rate was 35%.57 In a more recent report, the survival rate was 36% for horses with 360-degree volvulus of the large colon compared with 71% for horses with 270-degree volvulus.4 Postoperative complications include hypovolemic and endotoxemic shock, extensive loss of circulating protein, DIC, and laminitis.

In addition, large colon volvulus has a propensity to recur. Although one study documented a recurrence rate of less than 5%,57 some authors believe recurrence may be as high as 50%.7 Therefore methods to prevent recurrence should be considered in patients at risk for recurrence, particularly broodmares that tend to suffer from the disease recurrently during the foaling season.60,61

OTHER CAUSES OF LARGE INTESTINAL ISCHEMIA Intussusceptions The most common intussusceptions of the large intestine are cecocecal and cecocolic intussusceptions (Figure 15-26).62,63 Both are likely attributable to the same disease process, with variable inversion of the cecum. These conditions tend to occur in young horses (63% were younger than 3 years old in one study) and may be associated with intestinal tapeworms.63 Horses are seen with highly variable clinical signs, including acute, severe colic; intermittent pain over a number of days; and chronic weight loss.63 These variable presentations likely relate to the degree to which the cecum has intussuscepted. Initially, the cecal tip inverts, creating a cecocecal intussusception, which does not obstruct the ow of ingesta. As the intussusception progresses, the cecum inverts into the right ventral colon (cecocolic intussusception), which obstructs the ow of ingesta and often causes severe colic. The cause of abdominal pain is often di cult to di erentiate in these cases, although it is sometimes possible to detect a mass on the right side of the abdomen by either rectal palpation or ultrasound examination.62,63 Treatment involves manual surgical reduction by retracting the intussusceptum directly62 or by way of an enterotomy in the right ventral colon.64 However, sometimes the cecum cannot be readily reduced because of severe thickening, and in other cases surgical procedures result in fatal contamination. For example, in one report 8 of 11 horses were euthanized in the perioperative period because of complications,62 and in another report 12 of 30 horses were euthanized either before or during surgery. The latter included all of the horses with chronic disease because of irreversible changes to the cecum.63 However, one recent report on cecocolic intussusceptions indicated that 7 of 8 horses that underwent right ventral colon enterotomy and cecal resection survived long term,64 suggesting that continued improvements in surgical techniques may improve the prognosis.

FIGURE 15-26

Cecocolic intussusception in a horse with colic. An enterotomy has been made in the right ventral colon (short arrow) to reveal an intussuscepted cecum (arrows). Although this picture was taken at necropsy, an enterotomy such as the one shown in this gure can be used to exteriorize and resect the majority of the compromised cecum. Note the ileum adjacent to the colon (double arrow).

Colocolic intussusceptions are exceptionally rare but have reportedly a ected the pelvic exure and the left colons.65-68 Although the condition is reportedly more common in young horses,66-68 older horses may be a ected.65 Clinical ndings may include a palpable mass on the left side of the abdomen.66 Ultrasonography may also be useful. Treatment requires manual reduction of the intussusception at surgery66,68 or resection of a ected bowel.65 Because the left colons can be extensively exteriorized and manipulated at surgery,65-68 the prognosis is fair.

RECTAL PROLAPSE Rectal prolapse may occur secondary to any disease that causes tenesmus, including diarrhea, rectal neoplasia, and parasitism,69 or prolapse can occur secondary to elevations in intra-abdominal pressure during parturition or episodes of coughing.70,71 Rectal prolapses are classi ed into four categories (Table 15-14) depending on the extent of prolapsed tissue and the level of severity.72 Type I rectal prolapse is most common and characterized by a doughnut-shaped prolapse of rectal mucosa and submucosa (Figure 15-27). Type II prolapses involve fullthickness rectal tissue, whereas type III prolapses additionally have invagin*tion of small colon into the rectum. Type IV prolapses involve intussusception of proximal rectum or small colon through the anus in the absence of prolapse of tissue at the mucocutaneous junction at the anus.72 These can be di erentiated from other forms of prolapse by their appearance and a palpable trench between prolapsed tissue and

the anus. TABLE 15-14 Classification of Rectal prolapse Grade I II III IV

Description Prolapse of rectal mucosa Prolapse of full-thickness rectum Grade 2 prolapse with additional protrusion of small colon Intussusception of rectum and small colon through the anus

Prognosis Good Fair Guarded Poor

FIGURE 15-27

Type I rectal prolapse in a horse. Note circumferential protrusion of partial thickness rectal tissue (arrows) that is becoming congested as a result of pressure from the surrounding anus.

In type I prolapses most frequently seen in horses with diarrhea, the rectal mucosa becomes irritated and protrudes intermittently during episodes of tenesmus. If tenesmus persists, rectal mucosa can remain prolapsed. Rectal mucosa rapidly becomes congested and edematous under these conditions, which should be treated with osmotic agents such as glycerin or magnesium sulfate and by massaging and reducing the prolapse.73 A purse-string suture may be necessary to keep the mucosa inside the rectum. Topical application of lidocaine solution or jelly, epidural anesthesia, and sedation may help reduce tenesmus that incites and exacerbates rectal prolapse. Similar treatments can be applied with type II rectal prolapses. However, these more severe prolapses may not be reducible without surgical resection of mucosa and submucosa from the prolapsed bowel.69,73 Type III and type IV rectal prolapses are more serious injuries because the small colon is involved.74 In horses with type III prolapses, an abdominocentesis should be performed to determine if the injury to the small colon has resulted in peritonitis. The small colon component should be reduced manually if possible, whereas prolapsed rectal tissue typically requires mucosal or submucosal resection. Surgical exploration of the abdomen should be performed to determine the status of the small colon, although serial abdominocenteses can be used in lieu of surgery to detect progressive necrosis of bowel. Type IV prolapses are seen most commonly in horses with dystocia.72 These prolapses are almost always fatal because of stretching and tearing of mesenteric vasculature, with subsequent infarction of a ected bowel. Therefore euthanasia is usually warranted on the basis of physical examination ndings. However, con rmation of severe small colonic injury requires abdominal exploration using either a midline approach or laparoscopy.75 It is conceivable that a horse with compromised small colon could undergo a colostomy of the proximal small colon, but the compromised small colon will typically necrose beyond that which can be resected using a midline abdominal approach.73

NONSTRANGULATING INFARCTION Nonstrangulating infarction occurs secondary to cranial mesenteric arteritis caused by migration of Strongylus vulgaris76 and has become a relatively rare disorder since the advent of broad spectrum anthelmintics. Although thromboemboli have been implicated in the pathogenesis of this disease, careful dissection of naturally occurring lesions has not revealed the presence of thrombi at the site of intestinal infarctions in most cases.76 These ndings suggest that vasospasm plays an important role in this disease.11 Clinical signs are highly variable, depending on the extent to which arterial ow is reduced and the segment of intestine a ected. Any segment of intestine supplied by the cranial mesenteric artery or one of its major branches may be a ected, but the distal small intestine and large colon are more commonly involved.76 There are no clinical variables that can be used to reliably di erentiate this disease from strangulating obstruction.76 In some cases massive infarction results in acute, severe colic.76 Occasionally, an abnormal mass and fremitus may be detected on rectal palpation of the root of the cranial

mesenteric artery. This disease should be considered a differential diagnosis in horses with a history of inadequate anthelmintic treatment and the presence of intermittent colic that is di cult to localize. Although fecal parasite egg counts should be performed, they are not indicative of the degree of parasitic infestation. In addition to routine treatment of colic, dehydration, and endotoxemia, medical treatment may include aspirin (20 mg/kg/day) to decrease thrombosis.11 De nitive diagnosis requires surgical exploration. However, these cases are di cult to treat because of the patchy distribution of the lesions and the possibility of lesions extending beyond the limits of surgical resection. In addition, further infarction may occur after surgery. The prognosis is fair for horses with intermittent mild episodes of colic that may be amenable to medical therapy but poor in horses that require surgical intervention.11,76

OBSTRUCTIVE DISORDERS OF THE GASTROINTESTINAL TRACT Anthony T. Blikslager

APPROACH TO THE HORSE WITH COLIC Clinical management of colic is distinctly di erent from management of many other clinical syndromes because the initial focus is not on de ning the de nitive diagnosis but rather on determining whether the horse requires surgical exploration. Therefore the clinician must collect a series of historical and physical examination ndings and decide whether these ndings warrant medical management or whether referral and potential surgical exploration of the abdomen are necessary because of a suspected obstructive or ischemic lesion. For example, a horse may exhibit presenting signs of severe abdominal pain, poor cardiovascular status, and abdominal distention that may be compatible with an extensive list of di erential diagnoses and, more important, may indicate the need for abdominal exploration to minimize the extent of intestinal injury. The speed with which this clinical decision can be made has a tremendous impact on the well-being of the patient,1,2 because delaying surgical exploration of a horse with ongoing intestinal injury exacerbates shock induced largely by microbial toxins such as LPS traversing damaged mucosa, and this in turn correlates with mortality.3

History The initial clinical step in treating horses with colic is taking a thorough history. However, it may be necessary to delay this until after the physical examination and initial treatment because management of abdominal pain takes precedence. The vital components of the history that should be obtained before examination and treatment, if possible, are the duration and severity of colic symptoms, analgesics already administered, and a history of any adverse drug reactions. The two most critical factors from a history that would support a decision to refer and potentially surgically explore a horse with colic are the duration of signs2 and the extent of pain.4 The latter is deduced by asking the owner about the presence and frequency of pawing, looking at the anks, rolling, repeatedly going down and getting back up, and posturing as if to lie down or urinate, in addition to other clinical evidence of pain. Other important components of the history that should be obtained to ascertain the reason for the colic are listed in Table 15-15. TABLE 15-15 History Findings and Their Relevance to Colic and Its Prevention Risk for Colic Feeding

Environment

Exercise

Preventive care

Recent change in feed Coastal Bermuda hay with a high fiber content Feeding round bales Feeding off the ground Excessive concentrate Large infrequent meals Bolting feed Excessive time in stall

Insufficient access to water Exercise-induced exhaustion

Insufficient dental care

Potential Mechanisms Alteration in fluid flux or fermentation in the large colon Obstruction of ileum by fine, fibrous hay Poor-quality hay Horses may ingest sand in some regions of the country Alteration in fluid flux or fermentation in the large colon Alteration in fluid flux or fermentation in the large colon Large boluses of feed entering the esophagus and stomach Insufficient intake of roughage Insufficient exercise Dehydration Dehydration Reduced gastrointestinal motility Poor mastication of feed

Medication Previous medical history

Insufficient anthelmintic treatment Excessive administration of NSAIDs Colic surgery

Large parasite burden Mucosal damage, particularly in the stomach and colon Adhesions Anastomotic obstruction

Physical Examination Findings Just as the clinician may need to postpone the complete history to allow rapid treatment of the colic, so must the clinician be able to alter the extent of the physical examination in order to treat the horse in a timely fashion. The most critical examination nding is the horse’s heart rate, because this has repeatedly been shown to reflect the level of pain and provide an excellent assessment of the cardiovascular status of the horse. The heart rate is likely the single most reliable predictor of the need for surgery and survival.5 Because analgesic treatments can dramatically alter the heart rate, it should be obtained, if at all possible, before administration of analgesics. Other components of the examination are speci cally designed to provide additional information on the cardiopulmonary status of the horse (quality of the pulse, mucous membrane color, capillary re ll time, respiratory rate, and full auscultation of the chest) and the nature of any intestinal obstruction that the horse might have (auscultation of gastrointestinal sounds, per rectal palpation of the abdomen, and presence of nasogastric re ux). Although there are classic presentations for horses with obstructions of either the small or large intestine (Table 15-16), clinical presentations of the various types of intestinal obstructions vary. For example, a horse that has a small intestinal obstruction may have several loops of distended small intestine without any evidence of nasogastric re ux, depending on the site of the obstruction (distal versus proximal), the extent of obstruction, and the duration of obstruction. Other examples include horses with large colon obstruction, which may have re ux because of direct compression of the small intestine by distended colon or because of tension on the duodenocolic ligament. The most useful diagnostic means to determine the type of intestinal obstruction is rectal palpation of the abdomen.4 However, only approximately one third of the abdomen can be reached by way of the rectum, and this may be substantially less in very large or heavily pregnant horses. Nonetheless, it is worthwhile to try to determine the type of obstruction the horse has (i.e., small intestine versus large intestine and simple obstruction versus strangulating obstruction) because this information can be very helpful in determining the prognosis. In one study interns and residents at a veterinary teaching hospital were able to predict the type of lesion with a speci city exceeding 90%.6 This in turn can be very helpful in educating the client about the potential findings in surgery and the likelihood that the horse will survive. TABLE 15-16 Indications for Surgery in Patients with Colic According to Their Clinical Signs Indication Refractory pain

Clinical Signs Repeated episodes of pain despite treatment with analgesics Violent episodes of pain Persistently elevated heart rate (>48 beats/min)

Sepsis in the face of colic

Persistently elevated heart rate Weak peripheral pulse Abnormal mucous membrane color (pale, hyperemic, purple) Delayed capillary refill time (>2 seconds)

Evidence of a refractory small intestinal obstruction

Refractory pain Nasogastric reflux Distended loops of small intestine on rectal palpation

Evidence of a refractory large intestinal obstruction

Refractory pain Abdominal distention Distended large colon on rectal palpation Tight band(s) on rectal palpation

Evidence of devitalized bowel

Sepsis

Abnormal abdominocentesis (TP >2.5 g/dl, TNCC >10,000/μl) TNCC, Total nucleated cell count; TN, total protein

Management of Abdominal Pain Before considering ways to manage signs of colic, the clinician should remember that such signs are very poorly localized. Therefore, although colic is most frequently associated with intestinal disease, dysfunction of other organ systems, including urinary obstruction,7,8 biliary obstruction,9 uterine torsion or tears,10,11 ovarian artery hemorrhage,10 and neurologic disease should be considered as di erential diagnoses.12 However, the duration and severity of colic signs are excellent predictors of whether a horse requires surgical exploration of the abdomen. In fact, refractory pain supersedes all other predictors of the need for surgery in the colic patient. Once signs of colic have been recognized and categorized as to their severity, it is critical to rapidly and e ectively nullify signs of pain, for the well-being of the horse and to reduce the owner’s anxiety. In addition, it is becoming increasingly clear that pain is best managed before it becomes severe.13 There are several classes of analgesics readily available for treatment of horses with colic (Table 15-17), including spasmolytics (N-butylscopolamine), α2 agonists (xylazine, detomidine), opiates (butorphanol), and NSAIDs (e.g., unixin meglumine). Although much of this information is very familiar to most practitioners, a couple of points should be emphasized. The short-duration drugs N-butylscopolamine, xylazine, and butorphanol, which provide analgesia either directly (xylazine and butorphanol) or indirectly (through cessation of intestinal spasm; Nbutylscopolamine) for approximately 30 to 45 minutes, allow the veterinarian to determine if pain is recurrent within the time period of the typical examination. Alternatively, unixin meglumine is not as potent as an analgesic, but its duration is much longer. In fact, the clinician should closely adhere to its treatment interval to prevent deleterious e ects on gastrointestinal mucosa and the kidneys.14,15 The recent discovery of two isoforms of COX, the enzyme inhibited by NSAIDs, has provided drugs that can more selectively inhibit pro-in ammatory COX-2, while permitting continued constitutive production of prostanoids. This may be advantageous in horses with colic, particularly in light of recent evidence of reduced intestinal recovery from an ischemic event when subjects were treated with unixin compared with a drug that is more selective for COX-2.16 The α2-agonist detomidine should be reserved for horses with severe, unrelenting pain because of its tremendous potency in horses.17 In addition, it should be remembered that α2-agonists reduce the heart rate associated with a transient increase in blood pressure,18,19 thereby reducing the predictive value of the heart rate and pulse pressure. TABLE 15-17 Analgesics Commonly Used to Treat Colic in Horses Drug Butylscopolamin Xylazine Detomidine Butorphanol Flunixin

Dosage 0.3 mg/kg 0.3 – 0.5 mg/kg prn 0.01-0.02 mg/kg prn 0.01-0.02 mg/kg prn 0.25-1.1 mg/kg every 8-12 hr∗

Amount for an Adult Horse 150-170 mg 150-250 mg 5-10 mg 5-10 mg 125-500 mg

PRN, As needed. ∗ The longer treatment interval corresponds to the higher dose, whereas lower doses may be given more frequently

Clinical Pathology The most useful clinical pathologic tests are the packed cell volume and total protein because they can be used to substantiate clinical estimates of dehydration and correlate strongly with prognosis.2,20,21 A biochemical pro le is useful for assessing electrolyte imbalances and kidney and liver function, whereas blood gas analysis can be used to assess acid-base status. Horses with colic most frequently show evidence of metabolic acidosis, associated with hypovolemia or endotoxemia, but other abnormalities, such as metabolic alkalosis in association with extensive loss or sequestration of stomach Cl-, may be noted. Metabolic acidosis has been further investigated in horses with colic by measuring blood lactate, which is now an option in many laboratories.22 Lactate levels have also been inferred from measurement of the anion gap, although one study noted that lactate in horses with colic did not account for the entire anion gap.22 Lactate levels and anion gap closely correlate with the level of intestinal injury and the prognosis for survival.20,23,24 Other key components of the clinical pathologic evaluation of horses with colic are the abdominocentesis and complete blood count. Regarding the latter, the total white blood cell and di erential counts can provide crucial evidence of colic attributable to colitis (leukopenia, neutropenia, and a left shift) rather than an obstruction (highly variable complete blood count ndings). PF can be very helpful in determining the integrity of the intestine. Speci cally, as the intestine becomes progressively devitalized, the PF will become serosanguinous

as red blood cells are leaked into the abdomen, followed by an elevation in the total protein (>2.5g/dl) and progressive increases in total nucleated cell count (>10,000 cells/ l). However, these ndings do not always correlate well with the condition of the intestine, particularly in horses with large colon volvulus. For example, in a study of 57 horses with large colon volvulus, the average total protein (2.5 g/dl) and total nucleated cell count (1000 cells/ l) were normal despite the fact that only 36% with a 360-degree volvulus survived.25 This may occur because the development of severe mucosal injury following large colon volvulus is rapid and may not allow su cient time for protein and leukocytes to equilibrate with the abdominal fluid before the horse dies.26 Investigators have attempted to take all the variables routinely assessed during evaluation of horses for colic and develop models that can be used to accurately predict the need for surgery and the prognosis for life.2,27-30 However, none of these models has taken the place of clinical decision making, although these predictive studies have added tremendously to our understanding of the importance of some prognostic factors, particularly those reflecting cardiovascular function.

SMALL INTESTINAL SIMPLE OBSTRUCTION Simple obstruction involves intestinal obstruction of the lumen without obstruction of vascular ow. However, because there is a tremendous volume of uid that enters the small intestinal lumen on a daily basis,31,32 the obstructed intestine tends to become distended, which in turn may cause reduced mural blood ow.33 Ultimately, such distention can result in necrosis of tissues, particularly in the immediate vicinity of the obstruction.34 There are relatively few causes of simple obstruction in the small intestine, and the incidence of these types of obstruction is low (approximately 3% of all referred horses in one large hospital-based study).5 However, in some geographic regions this type of obstruction is prevalent. For example, in the southeastern United States, ileal impactions are relatively common.35,36

Ascarid Impactions Impactions caused by Parascaris equorum typically occur in foals younger than 6 months of age that have been on a poor deworming program and have a heavy parasite burden. Products that cause sudden ascarid death, including organophosphates, ivermectin, and pyrantel pamoate, have been incriminated in triggering acute intestinal obstruction by dead parasites.37 This is a particular problem with ascarids because of the relatively large size of the adult parasite. Clinical signs include acute onset of colic after administration of an anthelmintic and signs compatible with small intestinal obstruction, including nasogastric re ux. Occasionally, dead parasites are present in the re ux. The onset of the disease varies according to the degree of obstruction.37 A tentative diagnosis may be made on the basis of the history and signs referable to small intestinal obstruction. Abdominal radiographs and ultrasound may indicate the presence of multiple loops of distended small intestine but are not required if clinical signs indicate the immediate need for surgery. Initial medical treatment should include treatment of hypovolemic shock resulting from sequestration of uid in the small intestine. Surgical treatment typically involves an enterotomy made over the intraluminal impaction and removal of ascarids. The prognosis is fair in cases that are rapidly addressed but poor in foals with evidence of hypovolemia and septic shock. In a recent study long-term survival of 25 affected horses was 33%.37

Ileal Impaction Ileal impactions occur most commonly in adult horses in the southeastern United States. Although feeding of coastal Bermuda hay has been implicated in this regional distribution, it has been di cult to separate geographical location from regional hay sources as risk factors.38 Nonetheless, it is likely that feeding suboptimal-quality coastal Bermuda hay puts horses at risk for ileal impaction, most likely because when the ber content of this particular type of hay is high, the bers are thin and most likely prematurely swallowed. The relationship between ber content and eating patterns is theoretical and remains to be proved. In addition, sudden changes in feed from an alternative type of hay to coastal Bermuda hay likely put a horse at risk for ileal impaction.38 Studies in the U.K. have revealed tapeworm infection as an important risk factor for ileal impaction. Based on risk analysis, the data suggested that more than 80% of the ileal impaction cases studied were associated with serologic or fecal evidence of tapeworm infection.39 Because of the poor sensitivity of fecal analysis for tapeworms, a serologic test (ELISA) has been developed by Proudman et al. with a sensitivity of approximately 70% and a specificity of 95%.40 Clinical signs of horses with ileal impaction are typical for a horse with small intestinal obstruction, including onset of moderate to severe colic and rectally palpable loops of distended small intestine as the condition progresses. Because the ileum is the distal-most aspect of the small intestinal tract, nasogastric re ux may take a considerable time to develop and is found in approximately 50% of horses requiring surgical correction of impacted ileum.35,41 The diagnosis is usually made at surgery, although an impacted ileum may on occasion be palpated rectally.36 Multiple loops of distended small intestine frequently make the impaction di cult to palpate. Ileal impactions may resolve with medical treatment36 but frequently require surgical intervention (Figure 15-28). At surgery uids can be directly infused into the mass, allowing the surgeon to break down the impaction. Dioctyl sodium sulfosuccinate (DSS) may be included in the infused uid to aid in disruption of the mass. Extensive small intestinal distention and intraoperative manipulation of the ileum may lead to postoperative ileus,42

but recent studies indicate that this complication is less frequent as the duration of disease before admission has decreased.35 However, an enterotomy should be considered to evacuate impacted intestinal contents and reduce manipulation. Recent studies indicate that the prognosis for survival is good.35,36

FIGURE 15-28

Intraoperative view of an ileal impaction. Note the distended appearance of the ileum as it courses toward the cecal base.

Ileal Hypertrophy Ileal hypertrophy is a disorder in which the muscular layers (both circular and longitudinal) of the ileum thicken for unknown reasons (idiopathic) or secondary to an incomplete or functional obstruction. For idiopathic cases proposed mechanisms include parasympathetic neural dysfunction resulting in chronically increased muscle tone and subsequent hypertrophy of the muscular layers of the ileal wall. Such neural dysfunction possibly results from parasite migration.43 Alternative hypotheses include chronic increases in the muscular tone of the ileocecal valve, leading to muscular hypertrophy of the ileum as it contracts against a partially occluded ileocecal valve. The jejunum may also be hypertrophied, either alone or in combination with the ileum.43 Clinical signs include chronic intermittent colic as the ileum hypertrophies and gradually occludes the lumen. In one study partial anorexia and chronic weight loss (1 to 6 months) were documented in 45% of the horses, most likely because of intermittent colic and reduced appetite.43 Because the ileal mucosa is not a ected by this condition, there is no reason to believe that these horses experience malabsorption of nutrients. The diagnosis is usually made at surgery, although the hypertophied ileum may be palpated rectally in some cases.43 For treatment an ileocecal or jejunocecal anastomosis to bypass the hypertrophied ileum is performed. Without surgical bypass intermittent colic persists, and the thickened ileum may ultimately rupture.43 The prognosis is fair with surgical treatment.44 Secondary ileal hypertrophy is most commonly noted in horses that have previously had colic surgery and that may have a partial or functional obstruction at an anastomotic site. For example, in one case report a horse developed ileal hypertrophy after surgical correction of an ileocecal intussusception.45 Ileal hypertrophy was also noted in a horse in which an ileocolic anastomosis was incorrectly orientated during surgical treatment of a cecal impaction.46 Horses are typically re-examined for recurrence of colic in these cases. Surgical therapy is directed at addressing the cause of small intestinal obstruction and resecting hypertrophied intestine.

Meckel’s Diverticulum Meckel’s diverticulum is an embryonic remnant of the vitelloumbilical duct, which fails to completely atrophy and becomes a blind pouch projecting from the antimesenteric border of the ileum.47,48 However, similar diverticulae have also been noted in the jejunum.49 These diverticulae may become impacted, resulting in partial luminal obstruction, or may wrap around an adjacent segment of intestine, causing strangulation.47 Occasionally, an associated mesodiverticular band may course from the diverticulum to the umbilical remnant and serve as a point around which small intestine may become strangulated. Mesodiverticular bands may also originate from the embryonic ventral mesentery and attach to the antimesenteric surface of the bowel, thereby forming a potential space within which intestine may become entrapped.50 Clinical signs range from chronic colic, for an impacted Meckel’s diverticulum, to acute, severe colic if a mesodiverticular band strangulates intestine. The diagnosis is made at surgery, and treatment requires resection of the diverticulum and any associated bands.50 The prognosis is good for horses with simple impaction of a Meckel’s diverticulum and guarded for horses with an associated small intestinal strangulation.50

Adhesions Adhesions of one segment of bowel to another or of a segment of intestine to other organs and the body wall typically occur after abdominal surgery and may be clinically silent, cause chronic colic attributable to partial obstruction, or result in acute obstruction. These di ering

clinical syndromes are attributable to the type of adhesions that develop. For example, a brous adhesion that does not by itself obstruct the intestinal lumen might serve as the pivot point for a volvulus, whereas an adhesion between adjacent segments of the intestinal tract may create a hairpin turn that causes chronic partial obstruction.51 The number of adhesions that develop may also vary dramatically from horse to horse. Some horses may develop a single adhesion adjacent an anastomotic site or a discrete segment of injured intestine, whereas other horses may develop di use adhesions involving multiple segments of intestine, likely because of widespread in ammatory disease at the time of the original surgery. The mechanisms whereby adhesions develop are complex, but they likely involve initial injury to the serosa initiated by intestinal ischemia, reperfusion injury, and luminal distention.52 Such injury involves in ltration of neutrophils into the serosa, accompanied by loss of mesothelial cells (Figure 15-29). In one study assessing the margins of resected small intestine, extensive neutrophil in ltration was documented in the serosa, particularly in the proximal resection margin that had been distended before correction of a variety of strangulating lesions.53 Regions of serosal injury and in ammation subsequently undergo reparative events similar to those of any wound, including local production of brin, de novo synthesis of collagen by in ltrating broblasts, and ultimately maturation and remodeling of brous tissue. Unfortunately, during this process brin may result in injured intestinal surfaces adhering to adjacent injured bowel or an adjacent organ. Once a brinous adhesion has developed, new collagen synthesis may result in a permanent brous adhesion. Alternatively, brinous exudate may be lysed by proteases released by local phagocytes, thereby reversing the adhesive process. Thus formation of adhesions may be viewed as an imbalance of fibrin deposition and fibrinolysis.54

FIGURE 15-29

Histologic appearance of the serosa of equine jejunum. A, Normal intestine, with a clearly distinguishable layer of mesothelial cells. B, Jejunum after 2 hours of ischemia and 18 hours of reperfusion, showing complete loss of mesothelial cells and dense infiltration of neutrophils (arrows).

Prevention of adhesions relies on inhibition of the mechanisms involved in adhesion formation, including reduction of serosal injury with early intervention and good surgical technique, reduction of in ammation by administration of anti-in ammatory medications, physical separation of in amed serosal surfaces (e.g., carboxymethylcellulose, hyaluronan),55-57 and pharmacologic modulation of brinous adhesion formation (e.g., heparin58). In addition, early return of motility in the small intestine after surgery may reduce contact time between in amed surfaces of intestine, thereby reducing the chances of adhesion formation.54 Horses at greatest risk of developing adhesions after colic surgery appear to be those that have small intestinal disease.51,59 For example, in one study of horses undergoing surgical correction of small intestinal obstruction, 22% developed a surgical lesion associated with adhesions. However, foals appear to have an increased incidence of adhesions compared with mature horses, regardless of the nature of the abdominal surgery.51 For example, one study indicated that 17% of foals developed lesions attributable to adhesions regardless of the type of the initial surgery.60 Studies con ict as to whether the degree of surgical intervention in uences adhesion formation,51 but horses that require enterotomy or resection and anastomosis were found to be at greatest risk of developing adhesions in one study.59 As an indication of the importance of postoperative adhesion formation, adhesions were among the most important reasons for repeat laparotomy in postoperative colic patients.59,61 Clinical signs in horses with adhesions are highly variable, depending on whether the adhesion is causing partial obstruction, complete luminal obstruction, or involvement of intestinal vasculature. However, adhesions would be an important di erential for intermittent colic in the postoperative period, particularly if such colic was not relieved by nasogastric decompression of the stomach. Continued intermittent colic should prompt abdominocentesis to determine if there is evidence of septic peritonitis, which may contribute to adhesion formation.

Placement of a large-bore drain and peritoneal lavage (Figure 15-30) will help resolve peritonitis and may reduce adhesion formation by reducing intra-abdominal in ammation. If postoperative colic persists, repeat laparotomy or laparoscopy may be elected. In one study on adhesions, 70% of repeat laparotomies were performed within 60 days, suggesting that surgical colic attributable to adhesions typically occurs within two months of an initial surgical procedure. Unfortunately, the prognosis for horses with colic attributable to adhesions is low, with only 16% of horses surviving adhesion-induced colic in one study.51

FIGURE 15-30

Peritoneal lavage in a horse. Note the use of an intravenous administration set and a large-bore catheter placed in the dependent portion of the abdomen adjacent to the ventral midline incision.

Postoperative Ileus The de nition of ileus is intestinal obstruction, which includes both physical and functional obstructions. However, in veterinary medicine the term is typically used to designate a lack of progressive aboral propulsion of ingesta resulting in functional obstruction.62 The diagnosis of postoperative ileus is typically based on the presence of excessive nasogastric re ux and may occur after any abdominal exploratory procedure. However, horses undergoing surgery for strangulating small intestinal lesions or small intestinal obstructive lesions such as an ileal impaction are at greatest risk.42 Recently, the syndrome of postoperative ileus in horses has been broadened to include those horses that may have delayed transit of ingesta through the large intestine after surgery. This large intestinal ileus may follow any type of surgery, particularly horses that have had orthopedic surgery, and is characterized by reduced fecal output (14 mg/dl). Hypercalcemia has been associated with alimentary neoplasms such as lymphoma, multiple myleoma, carcinomas, and ameloblastoma.54,65 Hypoglycemia (blood glucose 25% of total Increases Prolonged Decreases Increases Increases Increases Increases Increases Increases Increases

Normal Value 80) anti–Neospora sp. antibody titer was higher in a group of 54 aborted mares than in a randomly chosen group of 121 mares (p < 0.001). N. caninum deoxyribonucleic acid (DNA) was found in 3 of 91 fetal brains, 2 of 77 fetal hearts, and 1 of 1 placenta and present in both brains and hearts of two fetuses.347 The mere presence of the organism in an aborted fetus does not necessarily implicate it as being the cause of the abortion.346 Antibodies to Neospora spp. were detected in the sera from 11.9% of 800 asymptomatic horses in Israel and from 37.5% of mares that aborted.348 Antibodies to Neospora spp. have also been detected in mares in Brazil.343 Further investigations will be needed to determine whether there is an association between equine reproductive failure and Neospora spp. infection.342,343

Mare Reproductive Loss Syndrome In the spring of 2001, the equine breeding industry in central Kentucky was faced with a reproductive crisis that ultimately cost hundreds of millions of dollars, and the impact on the Thoroughbred industry is still being felt. Mares were foaling prematurely with premature separation of the chorioallantois (late-term “red bag” abortions). Many mares that were approximately 45 to 80 days pregnant su ered acute fetal loss, and most a ected mares failed to resume normal cyclic activity until after the o cial breeding season had closed because endometrial cups had been formed before the occurrence of mare reproductive loss syndrome (MRLS).In addition, an unusual number of weak foals were born.349,350 The Appalachian region to the north also reported a high number of similar cases. The author investigated more than 150 late-term abortions in southeast Ohio during the spring of 2001.351 Reports of weather patterns and an unusually high emergence of Eastern tent caterpillars (ETCs; Malacosoma americanum) were identical in both the central Kentucky and southeast Ohio regions.351,352 The epidemic that affected the horse industry in the Ohio Valley in late April and early May of 2001 and 2002 became known as MRLS.349 During the initial outbreak the state diagnostic laboratory was inundated with late-term aborted fetuses, and practitioners identi ed approximately 2000 early fetal losses.353 Many of these were detected during what should have been a routine fetal gender determination. Fortunately, no contagious infectious cause was identi ed, and this newly recognized disease became self-limiting as the breeding season progressed. Veterinarians and scientists involved in the epidemic considered the possibility of mycotoxins, ergot alkaloids, phytoestrogens, and even cyanide from wild cherry tree foliage.354-356 A temporal correlation was established between MRLS and the presence of ETCs was identified.349,350,354,357-369 Mares inadvertently ingest ETCs when large numbers are present on pasture, hay, or in the water supply.368 Abortions consistent with MRLS can be induced by oral administration of whole ETCs or their exoskeletons, which contain hairs (setae).354,366,367 Setae embed into the submucosa of the gastrointestinal tract. Fetal loss occurs when bacteria from the alimentary tract (streptococci, actinobacilli) establish infection where the local immune system is compromised—the fetoplacental unit.367-369 Management strategies for controlling ETC have been proposed to reduce the risk of MRLS.353,362 A similar abortion syndrome in Australia is associated with ingestion of hairy processionary caterpillars (Ochragaster lunifer).370

Viral Abortions Equine arteritis virus (EAV) and equine herpesvirus 1 (EHV1) are the most important causes of viral abortion in horses in the United States. Both viruses are associated with acute respiratory tract infection and are discussed in detail in Chapter 9. EAV is transmitted through inhalation or venereally in the sem*n of asymptomatic (“shedder”) stallions.371-375 Abortion may occur if pregnant mares become infected in the later stage of gestation (5-10 months).373,375-377 Clinical signs are variable but may include fever, conjunctivitis, nasal discharge, and dependent edema that is associated with the vasculitis.374,377-379 Clinical signs are mild or subclinical in many horses and may be clinically indistinguishable from other respiratory infections.380 Viral myometritis with degeneration of myocytes and in ltration of mononuclear cells leads to transplacental infection of the fetus.381 A ected placentae are edematous, and degenerated broblasts may be observed in the subvillous layers.381 Lesions in the fetal tissue include an atrophy of the lymphoid follicles in the spleen

and lymph nodes with degenerated lymphocytes.381 Immuno uorescence can detect EAV antigen in the myometrium and the endometrial glands in the dams, in the subvillous layer of the placentae, and in the aborted fetuses.381 The virus may be recovered from the uterus and fetus, but the placenta is likely to yield the greatest amounts of the virus.381 The disease may be controlled by an e ective vaccination program and screening tests.371 Most abortions associated with herpesvirus are caused by EHV1, but occasional abortions are associated with EHV2, EHV4, and EHV5.276,382 The number of abortions resulting from EHV1 infection has declined over the past 20 years, and isolated abortions, rather than abortion storms, are now a more common feature of this disease.383 This is due to the widespread adoption of stringent vaccination programs in combination with improved management practices on broodmare farms.384,385 Pregnant mares should be vaccinated with an approved vaccine at 5, 7, and 9 months of gestation. Many farms also vaccinate at 3 months. New arrivals should be isolated for 3 weeks, and groups of pregnant mares should be isolated by stage of gestation. It is especially important to segregate pregnant mares from weanlings and other horses.386 EHV1 can infect the fetus if a mare is viremic during pregnancy. The virus causes abortion as a result of the rapid detachment of the placenta.387,388 Endothelial cells in the endometrium and allantochorion are often infected by the virus, with accompanying vascular lesions. The fetus can be infected by way of the chorionic vasculature or by inhalation of infected amniotic uid.389-391 Abortion may occur soon after the mare is infected or several weeks have elapsed. Therefore maternal serology is of little diagnostic value.392 The aborted fetus will be fresh, with copious amounts of pleural and peritoneal uid. The trachea may contain a brin clot. Small necrotic foci may be discernible on the swollen liver. A hyperplastic, necrotizing bronchiolitis may be seen in lung sections, and large intranuclear eosinophilic inclusion bodies are a characteristic histologic lesion. Although vaccination is widely practiced, owners should be aware that the protection is not absolute. If a pregnant mare is exposed to infected animals that have recently been to a show or that are returning from a training facility, it is possible that any protective immunity possibly conferred by the vaccine will be overwhelmed. Abortions have been associated with reactivation of latent virus that was induced by transport stress.393 Therefore a history of regular vaccination of an aborting mare does not eliminate the possibility of a herpesvirus-related abortion. Neutralization and indirect immuno uorescence tests, as well as PCR and virus isolation, are used for EHV1 diagnostics. Antigen detection, in combination with virus isolation and PCR from fetal lungs, gives reliable results.394 More detailed reviews have been published.196,395 A number of other acute viral infections have been associated with abortion in horses, especially if the horse su ers severe systemic illness at the time of initial infection. For example, mares may abort as a result of the systemic e ects of acute infection with equine infectious anemia virus (EIAV), although it is not generally considered an abortigenic virus.396 When widespread vaccination against West Nile virus was adopted in the United States, some suggested that the killed virus vaccine that was used might result in abortions, stillbirths, and deformities in foals. However, subsequent critical review clearly demonstrates that vaccination of pregnant mares during any period of gestation is not associated with an increased incidence of pregnancy loss.397

Noninfectious Causes of Abortion TWINS In North America the incidence of twin abortions has decreased signi cantly because of early intervention after ultrasonographic diagnosis of a multiple pregnancy.84,215,229 Before the widespread adoption of this technology, twin abortions were a major cause of fetal loss.49,50 A large European study (n = 12,648 pregnancies) of Thoroughbred mares revealed a twinning rate of 3.5%, with 443 twins and only two triplets.398 The type of placentation in the mare (di use, microcotyledonary) makes it highly unlikely that a twin pregnancy will be carried to term. Obviously, there is a nite endometrial surface area available for allantochorion attachment. The more common unicornuate twin vesicles are a problem, because one conceptus is inevitably restricted to the proximal aspect of the gravid horn.21,255 Thus the two conceptuses are literally in a deadly competition for adequate nourishment and subsequent placentation. If the twin pregnancy is maintained until the latter part of gestation, the nutrient demands of the rapidly growing fetuses outstrip the placental attachments. Fetal growth is such that nutrient demands may be met until the latter half of gestation, when the marked fetal development in the last trimester usually requires more exchange capability than the smaller placenta provides. The fetus becomes stressed as it becomes progressively emaciated and ultimately dies.215,232 Death of one fetus or both fetuses is followed by abortion, with the characteristic avillous areas on the fetal membranes con rming the amount of placental disruption.50,208,226,229 A ected mares develop premature mammary enlargement and may “run” milk before aborting.50,399 Transabdominal ultrasonographic evaluation may be useful to con rm the diagnosis at this late stage.226 Although the area of apposition of the two chorioallantoic membranes (twin membrane) may be seen, measurement of fetal thoracic diameters and heart rates can confirm the twin diagnosis.226,232 Twin abortions in the last few months of gestation are likely to cause a dystocia. Bicornuate twins are more likely to survive because each membrane can attach to an entire horn and one side of the uterine body, but the resulting foals are likely to be stunted as a result of intrauterine growth retardation (IUGR).4,50,215 The live birth of twin foals is extremely uncommon, and

many of these neonates do not survive, with as few as 14% of surviving foals reaching the second week of neonatal life.50,217 The mares are prone to fetal membrane retention and may be di cult to rebreed. Thus it is not surprising that the equine breeding industry has always tried to prevent twin pregnancies. After such a diagnosis the owner is faced with three choices: attempt elimination of one fetus in utero, manage the pregnancy to term in the remote hope of two viable foals, or induce abortion.226 Management of a twin pregnancy is discussed in more detail in a previous section.

UMBILICAL CORD COMPROMISE On rare occasions, a large ossi ed remnant of the yolk sac may compromise blood ow through the umbilical cord and result in abortion.400 The pathologic condition referred to as umbilical cord torsion may be de ned as excessive twisting of the cord such that there is complete or partial occlusion of the umbilical vessels or urachus.401 The two umbilical veins that return oxygenated blood from the placenta fuse within the distal aspect of the amniotic portion of the cord, and a single vein enters the fetal abdomen.402 Strangulation may occur if the cord becomes tightly wrapped around a portion of the fetus. Pressure sites on or around fetal parts may be due to the e ects of constriction by prolonged tension on a strangulated umbilical cord.403 A ected fetuses are not usually expelled immediately after death, and thus some degree of tissue autolysis should be expected.404 The length of the equine umbilical cord in normal gestation can be quite variable, and factors that a ect cord length and the number of twists present are unknown. Lethal umbilical torsion appears to be a sporadic condition with no apparent increased risk for future problems in mares that abort as a result of fetal complications.405 Whitwell has shown that some mares have have three or more foals with abnormally long cords.49 A Canadian study involving 93 Standardbreds did not demonstrate an e ect of sire on cord length, and much larger studies will be required to ascertain whether genetics are involved.29,406 Although the possibility of a heritable component for cord length cannot be discounted, a more plausible explanation may be that abnormally long cords are associated with the amount of fetal movement.402 Ultrasonographic studies of fetal mobility have helped explain the characteristic twisting that is a feature of the normal equine umbilical cord.4,407-410 As is not the case with ruminants, the equine amnion oats freely within the allantoic uid. Fetal rotation within the amniotic cavity and amniotic sac rotation within the allantoic cavity result in the characteristic twisting of the equine umbilical cord.411,412 The lumen of both uterine horns becomes constricted (presumably by localized circular muscle contractions) between 5 and 7 months, and the allantoic uid along with the fetus is contained within the con nes of the uterine body.4,407-409,412 What Ginther has described as the noncord horn remains closed, whereas the cord horn gradually permits the entry of the hindlimbs between 7 and 9 months. The limbs can enter the horn only when the fetus is in dorsal recumbency because the angle between the horn and body is so acute by this stage of gestation. Thereafter, the hindlimbs remain enclosed within the cord horn and the hooves extend to the horn tip by the tenth month.4,407-409,412 It is interesting that the peak incidence of abortions due to umbilical torsions occurs when the hindlimbs can become permanently enclosed in the uterine horn. It may be that in some instances this prevents the unraveling of those critical few rotations that can lead to circulatory compromise. There is no doubt that a longer umbilical cord predisposes the fetus to this condition. Reports from the United Kingdom suggest that with the decline in twin abortions, excessive twisting of the umbilical cord has become the most frequently made diagnosis in some laboratories. These abortions were associated with excessively long cords, with many over 80 cm in length.276,413 Because spiraling of the equine cord (4 to 5 times over its length) is a normal feature, it is important to ensure that the cause of fetal demise was in fact due to vascular compromise.402,411 Evidence of pathologic twisting with tension and compressive forces on the a ected portion of the cord may include aneurysms, tearing of the intima of vessels, hemorrhage, thrombosis of vessels, blanched constricted areas, local edema, and urachal dilations of varying sizes. Inadequate perfusion can cause intravascular thrombosis in the peripheral tissues of the chorioallantois and possible necrosis of the aspect of the placenta that is most distant from the attachment of the cord (the cervical pole of the body segment).402,404 These ndings are by no means de nitive, and there may be variability in interpretation among laboratories. Other considerations should include general agonal changes and the potential e ects of cord tension when a live fetus is expelled. Whitwell’s morphologic studies have established normal metrics for equine fetal membranes. The average length of the umbilical cord in a Thoroughbred is 55 cm (95% confidence interval; 36 to 83 cm; n = 143).414 Williams reviewed 168 cases of umbilical cord torsion from the University of Kentucky Livestock Disease Diagnostic Center, representing 6% of equine fetus submissions over a 5-year period.415 The gestational age ranged between 5 to 10 months, with a mean of 7.5 months.416 Umbilical cord length varied between 62 to 125 cm, with an average length of 96 cm. The cords tended to be highly twisted, with areas of constriction, edema, hemorrhage, and uid- lled sacculations. The fetuses were slightly to moderately autolyzed, in a manner consistent with fetal death before abortion. Urinary bladder dilation was observed in some cases. The most consistent histopathologic nding was evidence of necrotic changes with secondary deposition of calci ed material in the blood vessels of the chorioallantois.405,417 A less common cause of abortion resulting from cord obstruction is when an excessively long amniotic portion of the cord becomes tightly wrapped around a portion of the fetus.403,414,418 Although it may be more common for a fetal extremity to be involved, the author managed a case in a pony mare in

which the tightly wrapped cord left a deep groove around the lumbar region, with evidence of local edema.403

GENDER DETERMINATION (FETAL SEXING) The advent of fetal sexing has permitted early gender determination to in uence the value of the pregnant mare. Factors that may vary depending on the predicted sex of the foal include choice of state for foaling, appraisals and insurance coverage, sales reserves, bookings for stallion service the next season, and retention or sale of the mare.419 Accurate determination of the sex of the equine fetus can be made using either transrectal or transabdominal ultrasonography.410,419-421 Fetal gender should be certi ed only when the identifying structures have been clearly delineated and the accuracy of the determination is guaranteed. Accurate determination of fetal sex may be di cult or impossible in some cases because of excessive mare or fetal movement or because the fetus is located too deeply to permit adequate imaging. Although tranquilization (e.g., xylazine and butorphanol tartrate) is sometimes used, it may cause the uterus to relax and drop away from the examiner.419 A 5.0-MHz linear array transducer is adequate for transrectal gender determination, but a 3.5-MHz transducer will ensure the depth of penetration that is required to obtain transabdominal images.191 Gender determination is based on ultrasonographic assessment of the relative location of the genital tubercle, an embryologic structure that is initially located between the rear limbs in both sexes. The genital tubercle di erentiates into either a cl*tor*s or a penis and has an ultrasonographically distinctive, hyperechoic, bilobulated appearance in both sexes. Curren reported that the optimal time for gender determination is between days 59 and 68, and Holder concurs that a window between days 60 and 70 is ideal.419,421 A second ideal period for gender determination may be between days 110 and 120, because the genitalia are now well developed and the fetus tends to be more accessible again. After this time the increasing depth of the uterus means that a diagnosis may not be possible if the fetus is in a anterior (cranial) presentation at the time of the examination.419 If the transabdominal approach is used, the optimal window of time in both sexes is between days 100 and 220 of gestation.410,420,421 Thereafter, it may become increasingly di cult to identify the anatomic structures required to make an accurate gender determination. Detailed instructions for fetal sexing by ultrasonography have been described.410,419,420 Transabdominal gender identi cations based on the presence of the penis or prepuce (or both) in males and mammary glands and teats or fetal gonads in females can be quite accurate.421

MONITORING FETAL WELL-BEING Placentitis should be suspected in mares that develop premature mammary enlargement (with or without vulvar discharge). A reduction in plasma relaxin levels may indicate placental compromise because relaxin is produced by the equine placenta.135,140-142,145,313,422-424 Low relaxin levels in late pregnancy have been associated with various causes of placental dysfunction, including fescue toxicosis, oliogohydramnios, and placentitis.315 Measurement of an equine fetal protein and estrone sulfate levels in maternal plasma has not proved useful for early detection of fetal stress associated with medical and surgical colics.99,100,425-428 Estrogens are produced from C19 precursors that are secreted by the gonads of a viable fetus and thus are unlikely to re ect fetal stress per se.83 However, Riddle has reported that measurement of total estrogens can be useful for di erentiating between a normal and a compromised fetus.429 Before day 310 of gestation, total estrogen levels of less than 1000 ng/ml may be indicative of fetal stress, and mares with levels of less than 500 ng/ml are likely to have a severely compromised or dead fetus.429 It is important to remember that the circulating concentrations of hormones in maternal plasma generally represent a small percentage of the levels that are being metabolized by the fetus and uteroplacental tissues.83,89,430 The de nitive studies on mare progestagens (progestins) were performed using gas chromatography–mass spectrometry (GC-MS).83,110,430 It is imperative that clincians appreciate that assays (e.g., RIA, ELISA) using an antibody raised against progesterone will usually cross-react with several of the progestagens but to a variable extent. Thus the value that is typically reported by the laboratory as being progesterone should actually be termed total progestagens. The level of cross-reactivity in a particular assay can cause substantial variation in the reported value.83,430 Although placental pathology has been correlated with increased plasma progestin concentrations in some studies, others have not detected differences in plasma progestin concentrations in mares with impending abortion and mares with normal pregnancies when monthly blood sampling was performed.46,83,110,121,430-433 Therefore it is recommended that serial samples (daily for 3 days, then weekly) be obtained from a mare in which the condition is suspected because doing so may help the clinician identify a clinically useful trend in progestagen concentrations.83,99,259,277,430-434 Progestagen levels increase signi cantly in normal mares as gestation progresses, and concentrations in the maternal plasma may range between 5 to 50 ng/ml or even higher as mares approach term.83,430 It is thought that chronic fetal stress may cause the elevated progesterone and pregnenolone levels (as well as those of several metabolites) that have been observed in mares with placentitis.430,435 Ousey has described three abnormal progestagen patterns that may be clinically useful: a rapid decline, precocious elevation, and failure to increase at term.83,430 A rapid decline is consistent with fetal death or imminent fetal expulsion and may be seen after a uterine torsion or

colic.430,436 Pregnant mares that have experienced colic or uterine torsion and have progestagen levels below 2 ng/ml are at high risk for fetal loss.99 Prolonged elevation of progestagens for several weeks before delivery (especially if earlier than 305 days’ gestation) is consistent with metabolic activity in the fetus and uteroplacental tissues despite the presence of placental pathology.∗ Although typically small with poor skeletal development, foals born after being exposed to chronic fetal stress (placentitis) tend to exhibit precocious maturation—even when delivered several weeks early.83,303,430,439 Analysis of speci c progestagen pro les by GC-MS may di erentiate mares with placentitis from those with villous poverty or placental edema.83,430 If the total progestagen levels in maternal plasma fail to rise at term, fescue toxicosis (ergopeptine alkaloid-producing endophyte) is the most likely cause (see separate discussion).83,118 Transrectal ultrasonography provides an excellent assessment of the current status of the caudal allantochorion, and as such it is an invaluable aid when examining a mare in late pregnancy that exhibits signs of placentitis.434,440,441 An image of the ventral placental tissues in the area adjacent to the cervical star provides the ability to accurately diagnose the early stages of ascending placentitis.259,440,441 Experienced clinicians are able to observe abnormal tissue thickness and even evidence of placental separation with an associated pocket of in ammatory exudate. Early therapeutic intervention may provide the best chance for a successful outcome. In a normal placenta the chorioallantoic membrane and endometrium are intimately connected, making them ultrasonographically indistinguishable from each other. Thus a combined tissue measurement is used, and normal values for the combined thickness of the uterus and placenta (CTUP) have been established.442 The amniotic membrane should not be included, and this should be remembered if the fetal limbs are active at the time of the examination. The area 1 to 2 inches cranial and ventral to the cervix provides the most consistent CTUP measurement in normal mares, and this is the recommended site for any measurements. The ultrasound image should be frozen once the landmark vessel (a branch of the uterine artery) in the ventral uterine wall is located, and caliper measurements are taken from the inner surface of the ventral uterine vessel to the edge of the allantoic uid.259,440 Diagnostic accuracy can be improved by recording the average of at least three measurements. An increased CTUP at any point from midgestation until term indicates placental disruption and pending abortion.431 Transrectal CTUP values indicative of placental pathology and impending abortion are greater than 8 mm (between days 271 and 300), greater than 10 mm (between days 301 and 330), and greater than 12 mm after day 330.259,434,440,443,444 It is important to employ strict hygienic procedures when performing a vagin*l speculum examination on a mare with a high-risk pregnancy because the rst two barriers to the pregnant uterus will be breached (i.e., vulvar lips, vestibular sphincter). Mares at risk for abortion often have a moist, hyperemic, relaxed cervix. Even if vagin*l discharge has not been reported, many of these cases have a purulent cervical discharge if placentitis is present. However, whereas cervical softening and vagin*l discharge may be present if the infection is localized around the cervical star, in the case of nocardioform placentitis the lesion does not involve the cervical star and vagin*l discharge is conspicuously absent.286,291,301 Thus, although transrectal ultrasound is an extremely useful aid in diagnosing ascending placentitis, the site of the nocardioform lesion makes it of limited diagnostic value in these cases.316 Transabdominal examination of the ventral uterus may reveal separation of the chorioallantoic membrane from the uterine wall, often with evidence of in ammatory exudate accumulation between the two surfaces. Placentitis and associated placental edema will result in a thickened uteroplacental image. The uteroplacental thickness (CTUP) on a transabdominal ultrasound image should be between 7.1 ± 1.6 mm and 11.5 ± 2.4 mm.440,445,446 If the uteroplacental thickness exceeds 2.0 cm in late gestation, placental pathology is likely to be present.446,447 Transabdominal ultrasonography of mares in late pregnancy has become a routine diagnostic aid for evaluating fetal wellbeing.399,433,440,446,447 Although the 5.0-MHz linear array transducer is ideal for transrectal reproductive ultrasonography, its shallow depth of penetration (approximately 10 cm) limits its usefulness for transabdominal examinations in mares in late pregnancy. If the mare does not have a pronounced plaque of ventral edema, the 5.0-MHz transducer will often be su cient to image the uteroplacental unit and some of the fetal uids. Either a 3.5- or 2.5-MHz curved linear array or sector scanner transducer is best for transabdominal examinations because these can penetrate to a depth of 20 or 30 cm, respectively.192,440,448 While the 70- to 90-day fetus may be imaged from the ventral abdomen, just cranial to the mammary gland, the late gestation gravid uterus extends along the ventral abdomen to the xiphoid.399,440,445-449 By the ninth month of gestation, the fetus should be in anterior (cranial) presentation and dorsopubic or dorsolateral position.4,408,449 Thus in late gestation the fetal head should be positioned near the mare’s pelvis. An abnormal presentation or the presence of twins is possible if a fetal head is detected along the ventral abdomen during late gestation.448 A more detailed examination is indicated in such cases. The posture of the extremities varies with fetal movement.449 A standardized methodology must be followed when scanning the uterus from the ventral abdomen, starting just cranial to the mare’s mammary gland and moving cranially to locate the fetal thorax. The ribs cause multiple acoustic shadows that delineate the thoracic cavity. A complete examination of the fetus and uterus involves scanning cranially to the xiphoid in multiple parasagittal planes and then scanning from left to right sides of the abdomen in multiple transverse planes.448 Transabdominal ultrasound examination is an important diagnostic tool when attempting to identify the presence of twin fetuses during late gestation. Identi cation of the nongravid horn can be useful to help rule out the possibility of twins. Obvious size discrepancy often serves as con rmation that twins are in fact present. In other cases one

thoracic cavity does not contain a beating heart, confirming that one of the twins has already died. In a normal pregnancy the majority of the fetal uids are within the allantoic cavity. The amnion is imaged as a thin membrane that surrounds the fetus and lies in close contact with the fetus over much of its body. The amniotic membrane divides the imaged fetal uid into two distinct cavities. It is most easily seen around the fetal neck, shoulder, thorax, and foreleg. The largest pocket of amniotic uid is usually imaged at the point where the forelimb and neck meet the thorax.448 The maximum vertical depth of amniotic and allantoic uid and the quality of amniotic and allantoic uid are useful guides to fetal well-being.448 Any measurements of uid depth should be made as perpendicular to the uteroplacental surface as possible.447,448 In the normal equine pregnancy the maximum ventral fetal uid pocket depth for amniotic uid is 8 cm; it is 13 cm for allantoic uid.440,445,448 Extremes in either direction are not normal. Obviously de cient amounts of fetal uid indicate placental dysfunction, and excessive amounts suggest a hydrops condition.445,450 Fetal uid quantities should be considered excessive if the maximal vertical amniotic uid depth exceeds 14.9 cm or the maximum vertical allantoic uid depth exceeds 22.1 cm.448 The quality of the fetal uid is scored from 0 (clear) to 3 (echogenic uid with numerous particles).447,448 It is not unusual for echogenic particles to be noted in the fetal uids, especially during periods of fetal activity. These represent sloughed cells and proteinaceous debris. The fetal skin releases vernix as the pregnancy advances, and these free- oating particles can increase the cloudiness of the amniotic fluid.440,446 Thus an increase in the number of echogenic particles in late gestation may not be abnormal.447 However, if a high-risk pregnancy is being regularly monitored and a sudden increase in uid turbidity (grade 3) is observed, the prognosis is not good.446,447 The clinician should consider the possibility of inflammatory exudates, meconium passage by a compromised fetus, or even hemorrhage. It should be remembered that hippomanes (allantoic calculi) are a normal feature of the equine pregnancy. These structures may be observed on the ventral aspect of the allantoic cavity.448 An equine biophysical pro le has been proposed as a guide to assessing fetal well-being and predicting perinatal morbidity and mortality.440,446,447,451 Although a low score is de nitely indicative of a negative outcome, higher scores do not guarantee the birth of a viable neonate.448 Fetal breathing, heart rate and rhythm, fetal tone, and general activity are useful guides when evaluating fetal health and well-being. Therefore chemical sedation of the mare is not recommended because commonly used drugs are likely to induce fetal bradycardia and suppress normal fetal activity.452,453 Fetal breathing is characterized by movement of the diaphragm between the thorax and abdomen, in conjunction with rib cage expansion—without any other movement by the fetus. Fetal breathing patterns should be monitored for at least 30 seconds.440,448,454 When fetal heart rate and rhythm is monitored, it is not appropriate to scan for only 10 or 15 seconds and then multiply by a correction factor to obtain the number of beats per minute. Beat-to-beat variations and observation of periodic accelerations are important. It is normal for the heart rate accelerations to occur in association with fetal activity. Multiple measurements of fetal heart rate and assessments of fetal heart rhythm should be made over a 30-minute period while evaluating the fetus, fetal uids, and placenta.440,448 Ideally, three measurements should be obtained with the fetus at rest and another three after periods of activity. It is di cult to accurately monitor the heart rate during periods of fetal activity, unless M-mode echocardiography equipment is available.440,448,449 Fetal heart rates vary with the stage of gestation and the amount of fetal activity at the time of the examination.440,455-457 The fetal heart beat is normally regular and will decrease from greater than 120 beats/min in midgestation to between 60 and 90 beats/min in late gestation.440,445-447,449,455-459 Cardiac accelerations (20-40 beats/min above baseline) are normal if they are associated with fetal movement.440,445,446,448 However, persistent tachycardia in the absence of fetal activity indicates fetal stress. A resting heart rate in excess of 104 beats/min indicates fetal stress in a late gestation fetus.447,448 A heart rate of less than 57 beats/min in a fetus that is less than 330 days’ gestation and a rate of less than 50 beats/min in a fetus older than 329 days’ gestation should be considered abnormal.447,448 A fetus su ering from hypoxia will have a slow heart rate, with minimal limb activity or fetal breathing, indicative of central nervous system depression.446-449 However, if the condition is chronic and ischemic conditions are developing, the fetus will become tachycardic despite a lack of fetal activity. This is a prelude to fetal demise. In terminal cases extreme bradycardia ensues just before fetal death.447,454 Whereas failure to observe fetal activity may be due to the stage of the normal rest-activity cycle, con rmation of a regularly beating heart at least con rms that the fetus is alive. This is a major advantage over transrectal fetal ballottement, wherein failure to detect movement can raise unnecessary concerns about fetal health. If Doppler ultrasound equipment is available, the Doppler transducer is placed directly over the site where the best image was detected by the ultrasound scan. Tracings of fetal heart rate and rhythm can be recorded over time—usually intervals of 5 to 10 minutes.460 This makes analysis easier and serves as a permanent record of the fetal status at the time of the recording. If there is some question about the presence of twins after a transabdominal ultrasound examination, fetal electrocardiogram (ECG) tracings may show two distinct fetal patterns.458,459 Features of the ECG tracing that should be noted include fetal heart rate and rhythm, accelerations and decelerations, complex polarity changes, and beat-to-beat variation.440,454 In the last weeks of pregnancy fetal foals usually have a baseline heart rate in the range of 60 to 75 beats per minute. Transient low heart rates of less than 60 beats/min are not uncommon. These troughs warrant concern only if they are not interspersed with accelerations. Likewise, transiently elevated rates in the order of 120 beats/min (occasionally >200

beats/min) are not abnormal provided that they return to baseline.440,454 If the fetal heart rate is found to be less than 60 beats/min or greater than 120 beats/min during an observation period, then more frequent monitoring is justi ed to determine if the fetal is distressed. Beat-to-beat variations are normal, and a nding of no variability is an ominous sign. Maternal medications such as detomidine or butorphanol will reduce fetal heart rate variability transiently.440,454 The fetus is noted to have tone if it is observed to ex and extend the limbs, torso, or neck. Tone is poor or absent if the fetus appears flaccid.440,447,448 Fetal movements include partial to full rotation around the long axis of the fetus as well as less marked activity such as extension and exion of the extremities.440,447,448 Fetal activity is rated on a scale from 0 to 3, with 3 being a very active fetus. A score of 0 indicates that no fetal movement was noted during the examination period.447,448 Long periods without noticeable fetal activity are cause for concern and should be evaluated in conjunction with information about the fetal heart rate and rhythm. The fetus may be distressed, suffering from advanced hypoxia and central nervous system depression.446,447 Fetal aortic diameter is correlated with the weight of the pregnant mare, as well as the nal neonatal foal weight.440,445-447 Thus the pregnant mare’s weight can be used to estimate what the fetal aortic diameter should be, using the regression equation (Y = 0.00912 × pregnant mare’s weight in pounds + 12.46) where Y is the predicted fetal aortic diameter (mm).447,448 The actual diameter of the fetal aorta should then be measured in the thoracic cavity, as close to the fetal heart as possible.440,445,447 A smaller than predicted aortic diameter may indicate a dysmature or growth-retarded fetus (IUGR; intrauterine growth retardation) or twins. The maximal thoracic diameter is measured from the spine to the sternum over the caudal part of the thorax, and in a late gestation fetus it should be 18.4 ± 1.2 cm.447,448 It has been correlated with fetal aortic diameter and neonatal foal weight in high-risk pregnancies.447,448 Foal girth measurements and hip height are also correlated with fetal aortic diameter measurements.446,448,461 Fetal biparietal measurements and orbital diameters have also been used to estimate fetal size.440,448,461 Eye length (sclera to sclera) is a useful predictor of days before parturition in small ponies.462 Decreased blood ow to the placental unit inhibits fetal growth, and some form of chronic placental insu ciency should be suspected when small fetal size is detected. It is important to monitor fetal viability if a high-risk pregnancy is being maintained on altrenogest supplementation. Although most nonviable fetuses will be aborted, there are reports of a mummi ed fetus being retained when the mare was receiving long-term progestagen supplementation.237

COMPLICATIONS IN LATE GESTATION Once con rmed to be at least 45 to 60 days pregnant, most mares can be expected to carry the fetus to term. The incidence of fetal loss after 100 to 120 days of gestation is low and accounts for only a small percentage of total pregnancy wastage. Fetal death and maceration is uncommon in the mare. However, the author has managed a case of macerated twins in a draft breed mare that su ered no ill e ects systemically. The mare was evaluated only after the owner noticed a foul-smelling vagin*l discharge.463 Ventral body wall ruptures and uterine torsions are uncommon, and hydrops of the fetal membranes is an especially rare condition. Accurate diagnosis and appropriate management of these clinical cases can prevent the development of a life-threatening condition. If a ventral body wall rupture or uterine torsion is present, the birth of a viable foal may still be possible provided that the case is managed correctly.464

Hydrops of the Fetal Membranes Hydrops is a condition of the last trimester, with the pregnancy developing normally until somewhere between 7.5 months and term. Hydrallantois and hydramnios are rare conditions that involve a pathologic accumulation of uid within the allantoic and amniotic compartments, respectively. Normal volumes of allantoic uid in mares vary from 8 to 18 L at term. In documented cases of hydrops, the allantoic uid volume ranged from 110 to 230 L.419 Just as with cows, hydrallantois accounts for most dropsical conditions in the mare.465-468 The pathophysiology of hydrallantois in the cow has been related to an abnormality of placentation, whereas hydrops amnion has been associated with a fetal head anomaly that precludes swallowing.195 Dysfunctional placentation may cause an increased production of transudate or disruption of transplacental uid absorption. There does not appear to be any consistent abnormality of the fetus or fetal membranes that is characteristic of the condition in the mare.469 In fact, a case at the University of Florida was con rmed to be hydrops amnion on the basis of analysis of uid obtained through amniocentesis and allantocentesis, yet a viable foal was eventually delivered.470 A mild di use placentitis or endometrial vasculitis has been incriminated in some cases.419 In a report on 15 cases, all of the mares were pluriparous and ranged in age between 6 to 20 years.419 Generally, there is a sudden onset of abdominal distention, and walking becomes di cult.470 The mare will exhibit variable degrees of colic. There is a progressive loss of appetite, and the mare may experience some di culty in defecation. The increasing pressure on the diaphragm causes dyspnea, and the mucous membranes may appear cyanotic, especially when the mare is recumbent.419 On physical examination the rectal temperature is normal, but the heart rate will be elevated. Rectal palpation reveals characteristic ndings. Copious lubrication and extreme caution should be used because passage of the forearm will be impeded by pressure from the large uid- lled uterus. The feces tend to be covered with mucus as a result of the prolonged passage through the lower gastrointestinal tract. The gross distention of the uterus means that the fetus is usually not palpable.470 Failure to detect the fetus by external ballottement further supports the diagnosis. Transabdominal ultrasound may con rm the presence of excessive amounts of hyperechoic uid.470 A thorough examination from both sides of the abdomen should be performed to rule out the possibility of twins.464 Owners should be advised that this is a progressive condition and it is extremely unlikely that the mare will be able to sustain the pregnancy and deliver a live foal—although a recent report documented a case that was successfully managed until a viable foal was delivered at 321 days’ gestation.464,470 In that case the mare was provided with close monitoring of maternal and fetal health, in combination with supportive care that included a supportive belly-wrap, anti-in ammatory medication, and altrenogest. A partial drainage technique is sometimes used in an attempt to manage cases that were diagnosed within 2 to 4 weeks of term.419 A ected mares receive abdominal support (belly band), intravenous uids, broad spectrum antibiotics, and anti-in ammatory medication. The technique of slow, repeated drainage requires a major time commitment and would not be cost e ective in many cases. Fetal death may occur as a result of placental separation. There also appears to be a considerable risk for iatrogenic fetal infection after contamination of the fetal uids, despite attempts to perform the drainage technique in an aseptic manner. Thus, despite heroic attempts in valuable mares, the fetus is likely to be lost in cases of hydrallantois.419 In most cases induction of parturition may be advisable before the mare’s condition deteriorates further. Continued abdominal enlargement will predispose the mare to prepubic tendon rupture,470,471 and uterine rupture has also been reported.472 Induction is not without risk (shock, dystocia), but the prognosis for survival of the mare is good provided that appropriate supportive therapy is instituted.470 The prognosis for the mare’s reproductive future may also be favorable provided that there are no untoward sequelae (cervical lacerations, retained fetal membranes, metritis). Application of PGE to the cervix before induction may facilitate atraumatic fetal extraction.228,473 In one report six of eight mares that had previously developed a hydrops pregnancy subsequently became pregnant and delivered normal healthy foals at term.419 Before a therapeutic induction of parturition, the tail should be wrapped, the perineal area cleansed, and an indwelling intravenous catheter inserted. Large-volume intravenous uid therapy may become necessary if hypovolemic shock develops as the allantoic uid is

discharged.464 In some cases controlled drainage may be bene cial before inducing delivery. An added complication in hydrops cases is that the thickened, edematous chorioallantoic membrane may be di cult to rupture.465 If digital pressure alone is unsuccessful, an endometrial biopsy forceps can be used to bite a piece out of the chorioallantois. Some authors report that the lack of pressure from the atonic uterine wall will result in minimal release of fetal uid from the punctured chorioallantoic sac, but in the few cases that the author has been involved with there was a massive release of uid once the chorioallantoic membrane was ruptured.465,467 If insu cient uid release occurs, a sterile nasogastric tube can be introduced into the uterus to begin controlled siphoning of uid. An alternative technique is to introduce a thoracic trocar catheter through the cervix and employ a sharp puncture of the chorioallantois.419 This approach permits the excess uid to be removed by controlled drainage. Administration of intravenous uids with gradual removal of the excess allantoic uid will permit the mare’s cardiovascular system to adapt. Both oxytocin and PG injections have been employed in an attempt to abort these cases.465-467 Although oxytocin is widely considered to be the most e cacious method for routine induction of parturition, in hydrops cases the distended uterine musculature may not be able to contract e ectively.465,474 This uterine inertia is common, and gentle manual dilation of the cervix, or perhaps prior application of PGE, may be warranted.473 Bain and Wolfsdorf have reported a smooth induction after two doses of cloprostenol administered 30 minutes apart.419 The abdominal musculature may be weakened by stretching, and thus the typical stage 2 abdominal press may be compromised. Malpositioning and malpostures are not uncommon. The fetus may need to be extracted by assisted vagin*l delivery, but care should be taken so as not to traumatize the cervix by overzealous traction.464 The expelled fetus will generally be alive, and humane euthanasia is warranted. In one report at least 50% of foals had some malformation.419 Additional oxytocin (1.0 IU per minute) should be added to the intravenous uids to promote uterine involution. Retention of the fetal membranes should be expected, and appropriate treatment for removal of these membranes and prevention of the metritis-laminitis complex is indicated. Uterine involution should be monitored by transrectal palpation and ultrasonography.

Ventral Body Wall Hernias and Prepubic Tendon Rupture Apart from those with pathologic pregnancies, mares with ventral body wall defects are generally close to term.474,475 Damage to the abdominal sling of the pregnant mare may involve rupture of the transverse abdominis and oblique muscles, the rectus abdominus muscles, and the prepubic tendon.476 In extreme cases the rupture may lead to hemorrhage, shock, and death.477 The prepubic tendon attaches to the cranial border of the pubis, and lordosis occurs if the tendon is ruptured.464,474,475 Although breed (draft-breed mares) and age (older mares) may predispose a mare to development of the condition, in most cases there is no apparent predisposing cause.471,475,476 The extreme abdominal distention associated with the hydrops condition may cause rupture of the ventral musculotendinous support. Defects in the ventrolateral abdominal wall are more common than complete prepubic tendon rupture, and bilateral involvement of the abdominal wall seems to be most common.471,475,476,478,479 In a retrospective study of 13 cases, only three were categorized as being caudal midline (prepubic tendon lesions), and all three had additional involvement of the body wall musculature.475 The most obvious clinical sign of an impending ventral body wall rupture is a thick plaque of ventral edema extending a variable distance cranial to the udder.464,476 However, ventral edema may be a normal consequence of late pregnancy, and it can indicate external trauma. The author managed one case in which a large ventral swelling was associated with a massive hematoma that appeared to have originated from a kick. Mares in late pregnancy often develop a thick plaque of ventral edema that can extend from the udder to between the forelimbs. This is associated with the compressive weight of the gravid uterus on the venous and lymphatic drainage of the ventral abdomen. The presence of a hemorrhagic secretion in the mammary gland supports a diagnosis of tissue trauma rather than pregnancy edema.475 Unilateral edema is more indicative of damage to the ventrolateral body wall, but it may be associated with partial rupture of the prepubic tendon.464,476 The extreme pain that is associated with progressive enlargement of a ventral body wall rupture causes a marked tachycardia that may not be responsive to analgesics. Pregnant mares with a ruptured prepubic tendon or ventral abdominal wall show signs of colic and generally are reluctant to move. If the prepubic tendon is completely ruptured, the pelvis will be tilted such that the tailhead and tuber ischii are elevated, and a lordosis will be present.464,471,475,476 The mammary gland is often displaced craniad and ventrad because of the loss of the caudal attachment to the pelvis. A rent in the abdominal musculature may be complicated by bowel incarceration.464,471,475,476 Con rmation of the tentative diagnosis is sometimes di cult. Because it is not always possible to be certain that a rupture has already occurred, mares with severe ventral edema should be con ned to a stall, with exercise restricted to hand walking. Rectal palpation of the defect is usually not possible owing to the advanced stage of the pregnancy. External palpation is generally unrewarding because of the thickness of the edema, although some crepitation of the ventral abdominal wall may be noted. The mare is generally extremely sensitive and resists palpation of the area.464,475,476 Ultrasonographic examination of the posterior aspect of the ventral abdomen may be useful in some cases and can detect the presence of a bowel segment.448,464,475,477 Repture of the prepubic tendon can be identi ed as disruption of

the tendon bers immediately cranial to the pubis, whereas abdominal wall muscle tears are seen to be discrete tears within muscle bers, often associated with hematomas.475 An accurate assessment of the dimensions of the defect often cannot be made until the fetus and fetal fluids are expelled and the ventral edema has subsided.464,476 Depending on the degree of discomfort and stage of pregnancy, termination of the pregnancy may be the most humane treatment for the mare, given that further tissue damage is likely to occur to some degree until parturition. The possibility of a segment of intestine becoming incarcerated in the defect also should be considered.476 In extreme cases the mare may eventually become recumbent, especially if there is a rapidly expanding body wall defect.464,475 Owners should be aware that unless parturition is imminent, the prognosis for the foal is not good because fetal readiness for birth is di cult to predict.475 Because the present fetus may well be the mare’s last, owners often request that an attempt be made to maintain the pregnancy to term. In these cases the treatment is essentially supportive, and the prognosis for the foal is good.475,476 Anti-in ammatory drugs will help alleviate the mare’s discomfort. An abdominal sling (belly band) made of canvas or padded leather or a snug abdominal bandage help provide support for the ventral abdominal wall.475,476 In the author’s experience, abdominal bandages tend to be purely cosmetic because they soon stretch and thus provide minimal long-term support. If a sling is used, the area over the back must be well padded to prevent pressure necrosis because the purpose of this support is to transfer the weight of the gravid uterus to the vertebral column.464,476 Reducing the bulk of the ration and feeding a mild laxative may help reduce the degree of abdominal exertion associated with defecation.477 Assistance with parturition should be available because the mare may experience di culty in mounting su cient abdominal pressure to expel the fetus.471,475 However, one report suggests that some mares can position the foal and complete delivery unassisted.475,476 Arrangements for an alternate source of colostrum should be made because ventral edema may prevent the foal from suckling. The owner should be informed that although in some cases surgical repair of the defect may be possible by mesh herniorrhaphy, rebreeding the mare may not always be advisable.471,475,476 Some mares with small, unrepaired defects may subsequently foal without assistance, but the possibility that future pregnancies may exacerbate the condition should be considered.477 Embryo transfer o ers a viable alternative if this procedure is condoned by the relevant breed society.475

Uterine Torsion Uterine torsion accounts for 5% to 10% of all complicated obstetric conditions in the mare.480,481 Neither mare age nor parity appear to be significant risk factors.482 The causes of uterine torsion in the mare are not well de ned. The condition is much more common in cattle, and in that species a large term fetus has been implicated as a major risk factor. Most uterine torsions in cows occur at term and are thought to be a direct result of fetal positional changes during late rst-stage and early second-stage labor.483 A striking di erence between the mare and the cow is that more than 50% of uterine torsions in mares occur before the end of gestation.480 In the author’s clinical experience, the vast majority occur before term, and cases may be seen as early as 8 months of gestation.464 In fact, one recent report documented a case as early as 126 days of gestation.484 A multicenter retrospective study of 63 cases reported that 59% of the mares were at less than 320 days of gestation.482 Although Ginther et al. have shown that the fetus is locked into a dorsopubic position during the nal months of gestation, it is still possible for the entire pregnancy (uterus and fetus) to rotate approximately 90 degrees on the lower maternal abdominal wall.4 This occurs because any rotational movement of the caudal half of the fetus (pelvis and hindlimbs) by necessity will involve the close- tting uterus. It seems likely that in extreme cases this rotating action can lead to a clinical uterine torsion.230,419 Owners who work closely with their mares may observe excessive fetal movements in the ank area 1 or 2 days before. In a recent study 80% of term fetuses were in dorsosacral position when the uterine torsion was corrected. This suggests that fetal righting re exes may have played a role in creating the torsion.481 The author believes that vigorous fetal movements during the latter stages of gestation are likely to be a signi cant factor in the etiology of this condition in the mare.419 The clinical signs that attract the owner’s attention are the result of abdominal pain.464,485-488 These include restlessness, sweating, anorexia, frequent urination, sawhorse stance, looking at anks, and kicking at the abdomen. When the veterinarian is rst summoned, the signs may have been present for a period ranging between 2 hours to 3 days or more, especially if signs are intermittent and moderate.464,486 In mares that are close to term, the owner may assume that the signs indicate impending parturition. In more extreme cases the signs are more severe and may be associated with concurrent involvement of the small or large colon.482,484,487 Veterinarians should always consider the possibility of uterine torsion when a mare that is in the last trimester of gestation exhibits a mild, persistent colic. Delay in making a de nitive diagnosis increases the likelihood of fetal compromise.464 In one study of 63 cases, the mean time to admission was more than 20 hours.482 Occasionally, the condition may remain undiagnosed for several weeks.489 In these instances an owner may have attempted treatment with analgesics that were precribed for previous mild colic episodes.464,485 Rectal palpation is essential to determine whether a uterine torsion is present.464,485 The author is of the opinion that all late pregnant mares that display signs of mild to moderate colic warrant a thorough rectal examination to rule out the possibility of uterine torsion.

Although vagin*l involvement in the torsion is very common in the cow, uterine torsions in the mare seldom cause detectable changes in the vagin*.464,483,486,489,490 Thus vagin*l examination is generally not diagnostically useful. On rectal palpation the clinician should aim to carefully advance the forearm while palpating for a taut band on either side of the rectum. The ligament on the side of the torsion tends to be more caudal and is palpable as a tight vertical band. As the arm is advanced further, the opposite ligament will be palpable as it is pulled horizontally across the top of the uterus before being displaced ventrally.464 An accurate examination of the broad ligaments will con rm the diagnosis, determine the direction of the torsion, and give some idea of the severity of the torsion.419 One study (n = 54 mares) reported that 79% of referred cases were rotated no more than than 180 degrees, and 59% of cases involved clockwise torsion.482 A transrectal ultrasound examination is useful to evaluate the condition of the fetal uids and to note whether any placental detachment has occurred. The degree of uterine compromise can be gauged by noting the thickened uterine wall and distended vasculature. Compression of the veins and lymphatics occurs before occlusion of the arterial blood supply. Thus the initial changes will be associated with pooling of uid within the uterine wall.464 The compressive forces of the displaced broad ligaments may cause variable amounts of constriction of the small colon.486,491 Transabdominal ultrasonographic imaging may be used to assess fetal viability (heart rate and rhythm) and to evaluate the condition of the fetal uid. Compression of the uterine blood supply can cause fetal hypoxia and a fetal stress response, especially in more advanced pregnancies. Abdominocentesis may provide prognostic information and guide the clinician in choosing a mode of correction.492 Because it may be di cult to obtain peritoneal uid from a mare in late gestation, transabdominal ultrasonography is sometimes useful in locating a pocket of uid. Uterine rupture can be a complication of uterine torsion in the mare.486,493 In the author’s experience mild uterine torsions, or those of short duration, do not alter the color, cellularity, or total protein content of the peritoneal uid.492 Mares with severe or chronic uterine torsion may develop signi cant uterine compromise that results in changes in the composition of the peritoneal uid. Any alterations in the composition of the peritoneal uid may indicate the presence of a compromised or ruptured uterine wall or concurrent gastrointestinal involvement.493 A ank laparoscopic examination can con rm the condition of the uterine wall.494 This information will facilitate an informed choice of surgical approach or perhaps support a decision for euthanasia if economic considerations preclude surgical intervention. The stage of gestation and the mare’s heart rate are signi cant prognostic indicators for mare survival. A recent study reported that mares that died had signi cantly higher heart rates (mean 74 beats/min) at admission than survivors (mean 59 beats/min).482 Increased fetal size and the weight of the gravid uterus make correction closer to term (>320 days) more di cult, and this may explain the poor mare survival rate (65%) when compared with cases in less advanced pregnancies (10 mmol/L; >400 ppm; >40 mg/dl) before day 310 of gestation.121 Although mammary development is a useful sign of approaching parturition in normal mares, monitoring changes in mammary secretion electrolyte concentrations is the most reliable guide to imminent parturition.534-538 An inversion in the sodium-to-potassium ratio, followed by a rapid rise in calcium (Ca) concentrations in the last 24 to 48 hours, has been correlated with fetal maturity in both mares and jennies.536,539 Exact values may vary with the type of chemistry analyzer used by the diagnostic laboratory. In a normal term pregnancy, the combined mammary secretion levels of Ca (>40 mg/dl), potassium (>30 mEq/ml), and sodium (