Lung Biology in Health and Disease Volume 179 Acute Respiratory Distress Syndrome

ACUTE RESPIRATORY DISTRESS SYNDROME

LUNG BIOLOGY IN HEALTH AND DISEASE Executive Editor Claude Lenfant Director, National Heart, Lung, and Blood Institute National Institutes of Health Bethesda, Maryland 1. Immunologic and Infectious Reactions in the Lung, edited by C.H. Kirkpatrick and H.Y.Reynolds 2. The Biochemical Basis of Pulmonary Function, edited by R.G.Crystal 3. Bioengineering Aspects of the Lung, edited by J.B.West 4. Metabolic Functions of the Lung, edited by Y.S.Bakhle and J.R.Vane 5. Respiratory Defense Mechanisms (in two parts), edited by J.D.Brain, D.F.Proctor, and L.M.Reid 6. Development of the Lung, edited by W.A.Hodson 7. Lung Water and Solute Exchange, edited by N.C.Staub 8. Extrapulmonary Manifestations of Respiratory Disease, edited by E.D. Robin 9. Chronic Obstructive Pulmonary Disease, edited by T.L.Petty 10. Pathogenesis and Therapy of Lung Cancer, edited by C.C.Harris 11. Genetic Determinants of Pulmonary Disease, edited by S.D.Litwin 12. The Lung in the Transition Between Health and Disease, edited by P.T. Macklem and S.Permutt 13. Evolution of Respiratory Processes: A Comparative Approach, edited by S.C.Wood and C.Lenfant 14. Pulmonary Vascular Diseases, edited by K.M.Moser 15. Physiology and Pharmacology of the Airways, edited by J.A.Nadel

16. Diagnostic Techniques in Pulmonary Disease (in two parts), edited by M.A.Sackner 17. Regulation of Breathing (in two parts), edited by T.F.Hornbein 18. Occupational Lung Diseases: Research Approaches and Methods, edited by H.Weill and M.Turner-Warwick 19. Immunopharmacology of the Lung, edited by H.H.Newball 20. Sarcoidosis and Other Granulomatous Diseases of the Lung, edited by B.L.Fanburg 21. Sleep and Breathing, edited by N.A.Saunders and C.E.Sullivan 22. Pneumocystis carinii Pneumonia: Pathogenesis, Diagnosis, and Treatment, edited by L.S.Young 23. Pulmonary Nuclear Medicine: Techniques in Diagnosis of Lung Disease, edited by H.L.Atkins 24. Acute Respiratory Failure, edited by W.M.Zapol and K.J.Falke 25. Gas Mixing and Distribution in the Lung, edited by L.A.Engel and M. Paiva 26. High-Frequency Ventilation in Intensive Care and During Surgery, edited by G.Carlon and W.S.Howland 27. Pulmonary Development: Transition from Intrauterine to Extrauterine Life, edited by G.H.Nelson 28. Chronic Obstructive Pulmonary Disease: Second Edition, edited by T.L Petty 29. The Thorax (in two parts), edited by C.Roussos and P.T.Macklem 30. The Pleura in Health and Disease, edited by J.Chrétien, J.Bignon, and A.Hirsch 31. Drug Therapy for Asthma: Research and Clinical Practice, edited by J. W.Jenne and S.Murphy 32. Pulmonary Endothelium in Health and Disease, edited by U.S.Ryan 33. The Airways: Neural Control in Health and Disease, edited by M.A. Kaliner and P.J.Barnes 34. Pathophysiology and Treatment of Inhalation Injuries, edited by J.Loke

35. Respiratory Function of the Upper Airway, edited by O.P.Mathew and G.Sant’Ambrogio 36. Chronic Obstructive Pulmonary Disease: A Behavioral Perspective, edited by A.J.McSweeny and I.Grant 37. Biology of Lung Cancer: Diagnosis and Treatment, edited by S.T. Rosen, J.L.Mulshine, F.Cuttitta, and P.G.Abrams 38. Pulmonary Vascular Physiology and Pathophysiology, edited by E.K. Weir and J.T.Reeves 39. Comparative Pulmonary Physiology: Current Concepts, edited by S.C. Wood 40. Respiratory Physiology: An Analytical Approach, edited by H.K.Chang and M.Paiva 41. Lung Cell Biology, edited by D.Massaro 42. Heart-Lung Interactions in Health and Disease, edited by S.M.Scharf and S.S.Cassidy 43. Clinical Epidemiology of Chronic Obstructive Pulmonary Disease, edited by M.J.Hensley and N.A.Saunders 44. Surgical Pathology of Lung Neoplasms, edited by A.M.Marchevsky 45. The Lung in Rheumatic Diseases, edited by G.W.Cannon and G.A. Zimmerman 46. Diagnostic Imaging of the Lung, edited by C.E.Putman 47. Models of Lung Disease: Microscopy and Structural Methods, edited by J.Gil 48. Electron Microscopy of the Lung, edited by D.E.Schraufnagel 49. Asthma: Its Pathology and Treatment, edited by M.A.Kaliner, P.J. Barnes, and C.G.A.Persson 50. Acute Respiratory Failure: Second Edition, edited by W.M.Zapol and F.Lemaire 51. Lung Disease in the Tropics, edited by O.P.Sharma 52. Exercise: Pulmonary Physiology and Pathophysiology, edited by B.J. Whipp and K.Wasserman 53. Developmental Neurobiology of Breathing, edited by G.G.Haddad and J.P.Farber 54. Mediators of Pulmonary Inflammation, edited by M.A.Bray and W.H. Anderson

55. The Airway Epithelium, edited by S.G.Farmer and D.Hay 56. Physiological Adaptations in Vertebrates: Respiration, Circulation, and Metabolism, edited by S.C.Wood, R.E.Weber, A.R.Hargens, and R. W.Millard 57. The Bronchial Circulation, edited by J.Butler 58. Lung Cancer Differentiation: Implications for Diagnosis and Treatment, edited by S.D.Bernal and P.J.Hesketh 59. Pulmonary Complications of Systemic Disease, edited by J.F.Murray 60. Lung Vascular Injury: Molecular and Cellular Response, edited by A. Johnson and T.J.Ferro 61. Cytokines of the Lung, edited by J.Kelley 62. The Mast Cell in Health and Disease, edited by M.A.Kaliner and D.D. Metcalfe 63. Pulmonary Disease in the Elderly Patient, edited by D.A.Mahler 64. Cystic Fibrosis, edited by P.B.Davis 65. Signal Transduction in Lung Cells, edited by J.S.Brody, D.M.Center, and V.A.Tkachuk 66. Tuberculosis: A Comprehensive International Approach, edited by L.B. Reichman and E.S.Hershfield 67. Pharmacology of the Respiratory Tract: Experimental and Clinical Research, edited by K.F.Chung and P.J.Barnes 68. Prevention of Respiratory Diseases, edited by A.Hirsch, M.Goldberg, J.-P.Martin, and R.Masse 69. Pneumocystis carinii Pneumonia: Second Edition, edited by P.D. Walzer 70. Fluid and Solute Transport in the Airspaces of the Lungs, edited by R. M.Effros and H.K.Chang 71. Sleep and Breathing: Second Edition, edited by N.A.Saunders and C. E.Sullivan 72. Airway Secretion: Physiological Bases for the Control of Mucous Hypersecretion, edited by T.Takishima and S.Shimura 73. Sarcoidosis and Other Granulomatous Disorders, edited by D.G. James

74. Epidemiology of Lung Cancer, edited by J.M.Samet 75. Pulmonary Embolism, edited by M.Morpurgo 76. Sports and Exercise Medicine, edited by S.C.Wood and R.C.Roach 77. Endotoxin and the Lungs, edited by K.L.Brigham 78. The Mesothelial Cell and Mesothelioma, edited by M.-C.Jaurand and J. Bignon 79. Regulation of Breathing: Second Edition, edited by J.A.Dempsey and A.L.Pack 80. Pulmonary Fibrosis, edited by S.Hin. Phan and R.S.Thrall 81. Long-Term Oxygen Therapy: Scientific Basis and Clinical Application, edited by W.J.O’Donohue, Jr. 82. Ventral Brainstem Mechanisms and Control of Respiration and Blood Pressure, edited by C.O.Trouth, R.M.Millis, H.F.Kiwull-Schöne, and M.E.Schläfke 83. A History of Breathing Physiology, edited by D.F.Proctor 84. Surfactant Therapy for Lung Disease, edited by B.Robertson and H.W. Taeusch 85. The Thorax: Second Edition, Revised and Expanded (in three parts), edited by C.Roussos 86. Severe Asthma: Pathogenesis and Clinical Management, edited by S.J. Szefler and D.Y.M.Leung 87. Mycobacterium avium-Complex Infection: Progress in Research and Treatment, edited by J.A.Korvick and C.A.Benson 88. Alpha 1-Antitrypsin Deficiency: Biology • Pathogenesis • Clinical Manifestations • Therapy, edited by R.G.Crystal 89. Adhesion Molecules and the Lung, edited by P.A.Ward and J.C. Fantone 90. Respiratory Sensation, edited by L.Adams and A.Guz 91. Pulmonary Rehabilitation, edited by A.P.Fishman 92. Acute Respiratory Failure in Chronic Obstructive Pulmonary Disease, edited by J.P.Derenne, W.A.Whitelaw, and T.Similowski

93. Environmental Impact on the Airways: From Injury to Repair, edited by J.Chrétien and D.Dusser 94. Inhalation Aerosols: Physical and Biological Basis for Therapy, edited by A.J.Hickey 95. Tissue Oxygen Deprivation: From Molecular to Integrated Function, edited by G.G.Haddad and G.Lister 96. The Genetics of Asthma, edited by S.B.Liggett and D.A.Meyers 97. Inhaled Glucocorticoids in Asthma: Mechanisms and Clinical Actions, edited by R.P.Schleimer, W.W.Busse, and P.M.O’Byrne 98. Nitric Oxide and the Lung, edited by W.M.Zapol and K.D.Bloch 99. Primary Pulmonary Hypertension, edited by L.J.Rubin and S.Rich 100. Lung Growth and Development, edited by J.A.McDonald 101. Parasitic Lung Diseases, edited by A.A.F.Mahmoud 102. Lung Macrophages and Dendritic Cells in Health and Disease, edited by M.F.Lipscomb and S.W.Russell 103. Pulmonary and Cardiac Imaging, edited by C.Chiles and C.E.Putman 104. Gene Therapy for Diseases of the Lung, edited by K.L.Brigham 105. Oxygen, Gene Expression, and Cellular Function, edited by L.Biadasz Clerch and D.J.Massaro 106. Beta2-Agonists in Asthma Treatment, edited by R.Pauwels and P.M. O’Byrne 107. Inhalation Delivery of Therapeutic Peptides and Proteins, edited by A.L. Adjei and P.K.Gupta 108. Asthma in the Elderly, edited by R.A.Barbee and J.W.Bloom 109. Treatment of the Hospitalized Cystic Fibrosis Patient, edited by D.M. Orenstein and R.C.Stern 110. Asthma and Immunological Diseases in Pregnancy and Early Infancy, edited by M.Schatz, R.S.Zeiger, and H.N.Claman 111. Dyspnea, edited by D.A.Mahler

112. Proinflammatory and Antiinflammatory Peptides, edited by S.I.Said 113. Self-Management of Asthma, edited by H.Kotses and A.Harver 114. Eicosanoids, Aspirin, and Asthma, edited by A.Szczeklik, R.J. Gryglewski, and J.R.Vane 115. Fatal Asthma, edited by A.L.Sheffer 116. Pulmonary Edema, edited by M.A.Matthay and D.H.Ingbar 117. Inflammatory Mechanisms in Asthma, edited by S.T.Holgate and W. W.Busse 118. Physiological Basis of Ventilatory Support, edited by J.J.Marini and A. S.Slutsky 119. Human Immunodeficiency Virus and the Lung, edited by M.J.Rosen and J.M.Beck 120. Five-Lipoxygenase Products in Asthma, edited by J.M.Drazen, S.-E. Dahlén, and T.H.Lee 121. Complexity in Structure and Function of the Lung, edited by M.P. Hlastala and H.T.Robertson 122. Biology of Lung Cancer, edited by M.A.Kane and P.A.Bunn, Jr. 123. Rhinitis: Mechanisms and Management, edited by R.M.Naclerio, S.R. Durham, and N.Mygind 124. Lung Tumors: Fundamental Biology and Clinical Management, edited by C.Brambilla and E.Brambilla 125. Interleukin-5: From Molecule to Drug Target for Asthma, edited by C.J. Sanderson 126. Pediatric Asthma, edited by S.Murphy and H.W.Kelly 127. Viral Infections of the Respiratory Tract, edited by R.Dolin and P.F. Wright 128. Air Pollutants and the Respiratory Tract, edited by D.L.Swift and W.M. Foster 129. Gastroesophageal Reflux Disease and Airway Disease, edited by M.R. Stein 130. Exercise-Induced Asthma, edited by E.R.McFadden, Jr. 131. LAM and Other Diseases Characterized by Smooth Muscle Proliferation, edited by J.Moss

132. The Lung at Depth, edited by C.E.G.Lundgren and J.N.Miller 133. Regulation of Sleep and Circadian Rhythms, edited by F.W.Turek and P.C.Zee 134. Anticholinergic Agents in the Upper and Lower Airways, edited by S.L Spector 135. Control of Breathing in Health and Disease, edited by M.D.Altose and Y.Kawakami 136. Immunotherapy in Asthma, edited by J.Bousquet and H.Yssel 137. Chronic Lung Disease in Early Infancy, edited by R.D.Bland and J.J. Coalson 138. Asthma’s Impact on Society: The Social and Economic Burden, edited by K.B.Weiss, A.S.Buist, and S.D.Sullivan 139. New and Exploratory Therapeutic Agents for Asthma, edited by M. Yeadon and Z.Diamant 140. Multimodality Treatment of Lung Cancer, edited by A.T.Skarin 141. Cytokines in Pulmonary Disease: Infection and Inflammation, edited by S.Nelson and T.R.Martin 142. Diagnostic Pulmonary Pathology, edited by P.T.Cagle 143. Particle-Lung Interactions, edited by P.Gehr and J.Heyder 144. Tuberculosis: A Comprehensive International Approach, Second Edition, Revised and Expanded, edited by L.B.Reichman and E.S. Hershfield 145. Combination Therapy for Asthma and Chronic Obstructive Pulmonary Disease, edited by R.J.Martin and M.Kraft 146. Sleep Apnea: Implications in Cardiovascular and Cerebrovascular Disease, edited by T.D.Bradley and J.S.Floras 147. Sleep and Breathing in Children: A Developmental Approach, edited by G.M.Loughlin, J.L.Carroll, and C.L.Marcus 148. Pulmonary and Peripheral Gas Exchange in Health and Disease, edited by J.Roca, R.Rodriguez-Roisen, and P.D.Wagner 149. Lung Surfactants: Basic Science and Clinical Applications, R.H.Notter 150. Nosocomial Pneumonia, edited by W.R.Jarvis

151. Fetal Origins of Cardiovascular and Lung Disease, edited by David J.P. Barker 152. Long-Term Mechanical Ventilation, edited by N.S.Hill 153. Environmental Asthma, edited by R.K.Bush 154. Asthma and Respiratory Infections, edited by D.P.Skoner 155. Airway Remodeling, edited by P.H.Howarth, J.W.Wilson, J.Bousquet, S.Rak, and R.A.Pauwels 156. Genetic Models in Cardiorespiratory Biology, edited by G.G.Haddad and T.Xu 157. Respiratory-Circulatory Interactions in Health and Disease, edited by S. M.Scharf, M.R.Pinsky, and S.Magder 158. Ventilator Management Strategies for Critical Care, edited by N.S.Hill and M.M.Levy 159. Severe Asthma: Pathogenesis and Clinical Management, Second Edition, Revised and Expanded, edited by S.J.Szefler and D.Y.M. Leung 160. Gravity and the Lung: Lessons from Microgravity, edited by G.K.Prisk, M.Paiva, and J.B.West 161. High Altitude: An Exploration of Human Adaptation, edited by T.F. Hornbein and R.B.Schoene 162. Drug Delivery to the Lung, edited by H.Bisgaard, C.O’Callaghan, and G.C.Smaldone 163. Inhaled Steroids in Asthma: Optimizing Effects in the Airways, edited by R.P.Schleimer, P.M.O’Byrne, S.J.Szefler, and R.Brattsand 164. IgE and Anti-IgE Therapy in Asthma and Allergic Disease, edited by R. B.Pick, Jr., and P.M.Jardieu 165. Clinical Management of Chronic Obstructive Pulmonary Disease, edited by T.Similowski, W.A.Whitelaw, and J.-P.Derenne 166. Sleep Apnea: Pathogenesis, Diagnosis, and Treatment, edited by A.I. Pack 167. Biotherapeutic Approaches to Asthma, edited by J.Agosti and A.L. Sheffer 168. Proteoglycans in Lung Disease, edited by H.G.Garg, P.J.Roughley, and C.A.Hales

169. Gene Therapy in Lung Disease, edited by S.M.Albelda 170. Disease Markers in Exhaled Breath, edited by N.Marczin, S.A.Kharitonov, M.H.Yacoub, and P.J.Barnes 171. Sleep-Related Breathing Disorders: Experimental Models and Therapeutic Potential, edited by D.W.Carley and M.Radulovacki 172. Chemokines in the Lung, edited by R.M.Strieter, S.L.Kunkel, and T. J.Standiford 173. Respiratory Control and Disorders in the Newborn, edited by O.P. Mathew 174. The Immunological Basis of Asthma, edited by B.N.Lambrecht, H.C. Hoogsteden, and Z.Diamant 175. Oxygen Sensing: Responses and Adaptation to Hypoxia, edited by S. Lahiri, G.L.Semenza, and N.R.Prabhakar 176. Non-Neoplastic Advanced Lung Disease, edited by J.R.Maurer 177. Therapeutic Targets in Airway Inflammation, edited by N.T.Eissa and D.P.Huston 178. Respiratory Infections in Allergy and Asthma, edited by S.L.Johnston and N.G.Papadopoulos 179. Acute Respiratory Distress Syndrome, edited by M.A.Matthay 180. Venous Thromboembolism, edited by J.E.Dalen 181. Upper and Lower Respiratory Disease, edited by J.Corren, A.Togias, and J.Bousquet ADDITIONAL VOLUMES IN PREPARATION Acute Exacerbations of Chronic Obstructive Pulmonary Disease, edited by N.M.Siafakas, N.R.Anthonisen, and D.Georgopolous Lung Volume Reduction Surgery for Emphysema, edited by H.E. Fessler, J.J.Reilly, Jr., and D.J.Sugarbaker Idiopathic Pulmonary Fibrosis, edited by J.Lynch III Therapy for Mucus-Clearance Disorders, edited by B.K.Rubin andC. P.van der Schans Pleural Disease, edited by D.Bouros

Pharmacotherapy in Chronic Obstructive Pulmonary Disease, edited by B.R.Celli The opinions expressed in these volumes do not necessarily represent the views of the National Institutes of Health.

ACUTE RESPIRATORY DISTRESS SYNDROME Edited by

Michael A.Matthay University of California at San Francisco San Francisco, California, U.S.A.

MARCEL DEKKER, INC. NEW YORK • BASEL

“To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to http://www.ebookstore.tandf.co.uk/.” Although great care has been taken to provide accurate and current information, neither the author(s) nor the publisher, nor anyone else associated with this publication, shall be liable for any loss, damage, or liability directly or indirectly caused or alleged to be caused by this book. The material contained herein is not intended to provide specific advice or recommendations for any specific situation. Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress. ISBN 0-203-91203-9 Master e-book ISBN

ISBN: 0-8247-4076-9 (Print Edition) Headquarters Marcel Dekker, Inc., 270 Madison Avenue, New York, NY 10016, U.S.A. tel: 212–696–9000; fax: 212–685–4540 This edition published in the Taylor & Francis e-Library, 2005. Distribution and Customer Service Marcel Dekker, Inc., Cimarron Road, Monticello, New York 12701, U.S.A. tel: 800–228–1160; fax: 845–796–1772 Eastern Hemisphere Distribution Marcel Dekker AG, Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland tel: 41–61–260–6300; fax: 41–61–260–6333 World Wide Web http://www.dekker.com/ The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special Sales/Professional Marketing at the headquarters address above. Copyright © 2003 by Marcel Dekker, Inc. All Rights Reserved. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the publisher.

INTRODUCTION Acute respiratory distress syndrome (ARDS) first achieved recognition about 35 years ago via a landmark observation of Dr. Thomas Petty. Since that time, there have been many reports about the severity and frequency of this disease, an incidence of 150,000 to 200,000 cases a year is mentioned in many publications, but it can be argued that the actual number is greater or smaller. One of the distinctive features of ARDS is that it may result from many primordial conditions and, if the patient recovers, it may entail significant sequelae. Even the most optimistic among us recognize that ARDS is a complicated and troubling disease. Recent experience has shown that new etiologies may appear in the most unpredictable circumstances. We need only read the penultimate chapter about the SARS story to be convinced that what we know and see today will tomorrow be issues about ARDS. Since Petty’s seminal observations, clinicians and investigators have been fascinated by this disease, but the journey into discovery has been difficult and frustrating. The Lung Biology in Health and Disease series published its first volume about five years after ARDS was initially described, and the first volume devoted to acute respiratory failure appeared just a few years later. Since that time, other volumes have updated the readership on the many advances that have occurred. There is no question that our knowledge base has expanded considerably; however, we must accept the fact that the ARDS problem remains complex and elusive. The solution, of course, is in the research—and, make no mistake about it, this is clinical and risky research! That so many outstanding investigators are devoting their time and talent to researching how to improve the treatment of and recovery from this disease is a tribute to their dedication and a blessing for public health and for the patients. Dr. Matthay and all contributors to this volume are recognized experts in this area: they provide us with an outstanding report of the latest and most challenging finding on ARDS. As the Executive Editor of this series of monographs, I thank them all for the opportunity to introduce and present this new volume to the readership. Claude Lenfant, M.D. Bethesda, Maryland

PREFACE The acute respiratory distress syndrome, also known as acute lung injury, is a major cause of acute respiratory failure in critically ill patients. The syndrome was originally described by Dr. Thomas Petty in 1967 (see Dr. Petty’s overview in Chapter 1). Over the last three decades, clinical and experimental studies have evaluated the pathophysiology, pathogenesis, and treatment for clinical lung injury. During this time, remarkable progress has been made in understanding the molecular, cellular, and physiological basis for the development and resolution of acute lung injury. Furthermore, there is recent evidence that mortality in patients with acute lung injury can be reduced with a lungprotective ventilatory strategy. However, more progress is needed to fill in major gaps in our understanding of how acute lung injury develops, as well as how it resolves. This volume in the Lung Biology in Health and Disease series is designed to integrate both basic science and clinical research in order to provide a comprehensive perspective on the status of research and treatment of the acute respiratory distress syndrome and clinical acute lung injury. In Chapter 1 Dr. Petty provides an overview regarding historical aspects of the acute respiratory distress syndrome. In Chapter 2 there is a comprehensive discussion of clinical characteristics, clinical definitions, as well as the important clinical risk factors for developing acute lung injury. Chapter 3 provides an update on the epidemiology of acute lung injury. Recent data indicate that the incidence of acute lung injury is substantial, similar to the original estimate from the National Heart, Lung, and Blood Institute in 1977, approximately 150,000–200,000 patients per year in the United States alone. Chapter 4 provides an overview of radiographic characteristics, including findings from computerized axial tomography. Chapter 5 is a review of the pathological findings in clinical lung injury, including both ultrastructural and routine histological findings. The pathogenesis of acute lung injury is discussed in Chapter 6 (experimental studies) and Chapter 7 (clinical studies). Chapter 8 considers the role of apoptosis in the pathogenesis and resolution of lung injury. Important advances in understanding ventilator-induced lung injury based on both clinical and experimental studies is discussed in Chapter 9. The role of sepsis in the development of lung injury is reviewed in Chapter 10. New data are available on the potential role of heat-shock proteins in the pathophysiology of acute lung injury, a topic covered in Chapter 11. In some patients fibrosing alveolitis develops during the course of acute lung injury, a topic that is discussed in Chapter 12. A new area in research in clinical acute lung injury and the acute respiratory distress syndrome is the influence of genetic factors in determining which patients are most susceptible to the development of acute lung injury. Chapters 13 and 14 are devoted to this new topic. Chapter 15 focuses on experimental and clinical studies of the resolution of lung injury with a particular emphasis on the resolution of alveolar edema. The remaining chapters in this volume focus on issues that pertain directly to treatment. Chapter 16 considers treatment of sepsis, the most lethal cause of clinical acute

lung injury. Chapter 17 considers how pulmonary hypertension in clinical acute lung injury can be treated. Chapter 18 reviews the clinical studies of glucocorticoid therapy for the acute respiratory distress syndrome. Chapter 19 discusses clinical trials of surfactant replacement in patients with lung injury, and Chapter 20 provides an update on the potential role of prone position for the treatment of patients with clinical lung injury. Chapter 21 provides a combined American and European perspective on lung-protective strategies for patients with clinical lung injury. Chapter 22 provides a brief overview of an important new cause of viral pneumonia, severe acute respiratory syndrome (SARS), an illness that can lead to clinical acute lung injury. The final chapter provides a brief overview of selected areas in which important progress has been made as well as a perspective on potential opportunities for future research and treatment of acute lung injury. I appreciate the hard work of each of the contributors to this edition as well as the excellent editorial supervision provided by Rebecca Cleff, Moraima Suarez, Sandra Beberman, and Dr. Claude Lenfant. Michael A.Matthay

CONTRIBUTORS Kurt H.Albertine, Ph.D. Professor, Departments of Pediatrics, Medicine, and Neurobiology/Anatomy, University of Utah, Salt Lake City, Utah, U.S.A. Antonio Anzueto, M.D. Associate Professor, Department of Medicine, The University of Texas Health Science Center at San Antonio, San Antonio, Texas, U.S.A. Gordon R.Bernard, M.D. Melinda Owen Bass Professor of Medicine and Chief, Division of Allergy, Pulmonary, and Critical Care Medicine, Department of Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee, U.S.A. Laurent J.Brochard, M.D. Professor, Department of Intensive Care Medicine, Université Paris 12, INSERM U 492, and Henri Mondor Hospital, Créteil, France Roy G.Brower, M.D. Professor, Division of Pulmonary and Critical Care Medicine, Department of Medicine, Johns Hopkins University, Baltimore, Maryland, U.S.A. Thilo Busch, Ph.D. Research Scientist, Department of Anesthesiology and Intensive Care Medicine, Charité, Campus Virchow-Klinikum, Humboldt University, Berlin, Germany David C.Christiani, M.D., M.P.H. Professor, Division of Pulmonary and Critical Care Medicine, Department of Medicine, Harvard Medical School, and Massachusetts General Hospital, Boston, Massachusetts, U.S.A. Christine Clerici, M.D., Ph.D. Department of Physiology, Faculté de Medicine de Bobigny, Université Paris 13, Colombes, France Maria Deja, M.D. Department of Anesthesiology and Intensive Care Medicine, Charité, Campus Virchow-Klinikum, Humboldt University, Berlin, Germany Timothy W.Evans, M.D., Ph.D. Professor, Department of Critical Care Medicine, Imperial College School of Medicine, and Royal Brompton Hospital, London, England Konrad J.Falke, M.D., Ph.D. Professor and Chairman, Department of Anesthesiology and Intensive Care Medicine, Charité, Campus Virchow-Klinikum, Humboldt University, Berlin, Germany Xiaohui Fang, M.D. Postdoctoral Research Fellow, Cardiovascular Research Institute, University of California at San Francisco, San Francisco, California, U.S.A. James A.Frank, M.D. Assistant Adjunct Professor, Division of Pulmonary and Critical Care Medicine, and the Cardiovascular Research Institute, Department of Medicine, University of California, San Francisco, San Francisco, California, U.S.A. Luciano Gattinoni, M.D., F.R.C.P. Professor, Department of Anesthesia and Intensive Care Medicine, University of Milan, and Hospital of Milan I.R.C.C.S., Milan, Italy Herwig Gerlach, M.D., Ph.D. Professor and Chairman, Departments of Anesthesiology and Critical Care Medicine, Vivantes-Klinikum Neukoelln, Berlin, Germany Michaela C.Godzich, B.A. Staff Research Associate, Department of Anesthesia, University of California, San Francisco, San Francisco, California, U.S.A.

Michelle Ng Gong, M.D. Instructor, Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Harvard Medical School, and Massachusetts General Hospital, Boston, Massachusetts, U.S.A. Philip C.Goodman, M.D. Department of Radiology, Duke University Medical Center, Durham, North Carolina, U.S.A. Richard B.Goodman, M.D. Associate Professor, Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Washington School of Medicine, Seattle, Washington, U.S.A. Leonard D.Hudson, M.D. Professor and Endowed Chair in Pulmonary Disease Research, Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Washington School of Medicine, Seattle, Washington, U.S.A. Yumiko Imai, M.D. Postdoctoral Research Fellow, Arthur S.Slutsky Research Laboratory, University of Toronto, and Toronto General Hospital, Toronto, Ontario, Canada Udo Kaisers, M.D. Professor and Assistant Medical Director, Department of Anesthesiology and Intensive Care Medicine, Charité, Campus Virchow-Klinikum, Humboldt University, Berlin, Germany Hyon Lee, B.A. Senior Research Associate, Department of Anesthesia, University of California, San Francisco, San Francisco, California, U.S.A. James F.Lewis, M.D., F.R.C.P.(C) Professor, Departments of Medicine and Physiology, Lawson Health Research Institute, University of Western Ontario, and St. Joseph’s Health Center, London, Ontario, Canada Nicholas David Manzo, B.S. Research Technologist, Pulmonary and Critical Care Unit, Massachusetts General Hospital, Boston, Massachusetts, U.S.A. Richard P.Marshall, M.D., M.R.C.P.(UK), Ph.D. Centre for Respiratory Research, University College London, London, England Thomas R. Martin, M.D. Professor, Division of Pulmonary and Critical Care Medicine, Department of Medicine, University of Washington School of Medicine, and VA/Puget Sound Medical Center, Seattle, Washington, U.S.A. Sadis Matalon, Ph.D. Professor, Department of Physiology and Anesthesiology, University of Alabama at Birmingham, Birmingham, Alabama, U.S.A. Michael A.Matthay, M.D. Professor, Departments of Medicine and Anesthesia, and Cardiovascular Research Institute, University of California at San Francisco, San Francisco, California, U.S.A. Gustavo Matute-Bello, M.D. Acting Assistant Professor, Department of Medicine, University of Washington School of Medicine, and VA/Puget Sound Medical Center, Seattle, Washington, U.S.A. Thomas M.McIntyre, Ph.D. Professor of Internal Medicine and Experimental Pathology, Program in Human Molecular Biology and Genetics, University of Utah, Salt Lake City, Utah, U.S.A. Marc Moss, M.D. Associate Professor, Department of Medicine, Emory University School of Medicine, and Grady Memorial Hospital, Atlanta, Georgia, U.S.A. Margaret J.Neff, M.D., M.Sc. Assistant Professor, Division of Pulmonary and Critical Care Medicine, Department of Medicine, Harborview Medical Center, University of Washington, Seattle, Washington, U.S.A.

Mitchell A.Olman, M.D. Associate Professor, Division of Allergy and Critical Care Medicine, Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, U.S.A. Thomas L.Petty, M.D. Professor, Department of Medicine, University of Colorado Health Sciences Center, Denver, Colorado, and Rush-Presbyterian-St. Luke’s Medical Center, Chicago, Illinois, U.S.A. Jean-François Pittet, M.D. Associate Professor in Residence, Departments of Anesthesia and Perioperative Care and Surgery, University of California, San Francisco, San Francisco, California, U.S.A. Desirée M.Quiñones Maymí, M.D. Fellow in Thoracic Radiology, Department of Radiology, Duke University Medical Center, Durham, North Carolina, U.S.A. Gordon D.Rubenfeld, M.D., M.Sc. Associate Professor, Division of Pulmonary and Critical Care Medicine, Department of Medicine, Harborview Medical Center, University of Washington, Seattle, Washington, U.S.A. Tsutomu Sakuma, M.D., Ph.D. Associate Professor, Department of Thoracic Surgery, Kanazawa Medical University, Uchinada, Ishikawa, Japan Arthur S.Slutsky, M.A.Sc., M.D. Professor, Department of Medicine; Director, Interdepartmental Division of Critical Care Medicine, University of Toronto; and Vice President (Research), St. Michael’s Hospital, Toronto, Ontario, Canada Roger G.Spragg, M.D. Professor, Department of Medicine, University of California, San Diego, and San Diego VA Medical Center, San Diego, California, U.S.A. B.Taylor Thompson, M.D. Associate Professor, Department of Medicine, Harvard Medical School, and Director, Medical Intensive Care Unit, Pulmonary and Critical Care Unit, Massachusetts General Hospital, Boston, Massachusetts, U.S.A. Joseph F.Tomashefski, Jr., M.D. Professor, Department of Pathology, Case Western Reserve University School of Medicine, and Chairman, MetroHealth Medical Center, Cleveland, Ohio, U.S.A. Lorraine B.Ware, M.D. Assistant Professor, Division of Allergy, Pulmonary, and Critical Care Medicine, Department of Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee, U.S.A. Aaron B.Waxman, M.D., Ph.D. Assistant Professor, Department of Pulmonary and Critical Care Medicine, Harvard Medical School, and Massachusetts General Hospital, Boston, Massachusetts, U.S.A. Arthur P.Wheeler, M.D. Associate Professor, Department of Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee, U.S.A. Guy A.Zimmerman, M.D. Professor, Department of Internal Medicine, and Director, Program in Human Molecular Biology and Genetics, University of Utah, Salt Lake City, Utah, U.S.A.

CONTENTS Introduction Claude Lenfant Preface Contributors

1. Overview Thomas L.Petty 2. Definitions and Clinical Risk Factors Marc Moss and B.Taylor Thompson 3. Epidemiology of Acute Lung Injury: A Public Health Perspective Gordon D.Rubenfeld and Margaret J.Neff 4. Radiographic Findings of the Acute Respiratory Distress Syndrome Desirée M.Quiñones Maymí and Philip C.Goodman 5. Pulmonary Pathology of the Acute Respiratory Distress Syndrome: Diffuse Alveolar Damage Joseph F.Tomashefski, Jr. 6. Pathogenesis of Acute Lung Injury: Experimental Studies Nicholas David Manzo and Aaron B.Waxman 7. Pathogenesis of Acute Lung Injury: Clinical Studies Lorraine B.Ware and Timothy W.Evans 8. Is Apoptosis Important in the Pathogenesis and Resolution of the Acute Respiratory Distress Syndrome? Gustavo Matute-Bello and Thomas R.Martin 9. Pathogenesis of Ventilator-Induced Lung Injury James A.Frank, Yumiko Imai, and Arthur S.Slutsky 10. Pathogenesis of Sepsis and Septic-Induced Lung Injury Guy A.Zimmerman, Kurt H.Albertine, and Thomas M.McIntyre 11. Heat Shock Response, Heat Shock Proteins, and Acute Lung Injury Hyon Lee, Michaela C.Godzich, and Jean-François Pittet 12. Mechanisms of Fibroproliferation in Acute Lung Injury Mitchell A.Olman 13. Genetic Factors in Acute Lung Injury Richard P.Marshall

xv xvi xviii

1 6 30 44 62

103 129 158

178

217

254

275 311

14. Approach to the Genetic Epidemiology of Acute Lung Injury Michelle Ng Gong and David C.Christiani 15. Resolution of Alveolar Edema: Mechanisms and Relationship to Clinical Acute Lung Injury Michael A.Matthay, Xiaohui Fang, Christine Clerici, Tsutomu Sakuma, and Sadis Matalon 16. Sepsis in the Acute Respiratory Distress Syndrome: Treatment Implications Arthur P.Wheeler and Gordon R.Bernard 17. Modulation of Pulmonary Vascular Tone in the Acute Respiratory Distress Syndrome Udo Kaisers, Thilo Busch, Maria Deja, Herwig Gerlach, and Konrad J.Falke 18. Glucocorticoid Therapy for the Acute Respiratory Distress Syndrome Richard B.Goodman and Leonard D.Hudson 19. Surfactant Therapy in the Acute Respiratory Distress Syndrome Roger G.Spragg and James F.Lewis 20. Prone Position in the Acute Respiratory Distress Syndrome Antonio Anzueto and Luciano Gattinoni 21. Mechanical Ventilation in the Acute Respiratory Distress Syndrome Roy G.Brower and Laurent J.Brochard 22. Severe Acute Respiratory Syndrome Lorraine B.Ware 23. Acute Lung Injury: Recent Progress and Promising Directions for Future Research Michael A.Matthay Index

335 357

381

402

440 459 484 507 544 555

563

ACUTE RESPIRATORY DISTRESS SYNDROME

1 Overview THOMAS L.PETTY University of Colorado Health Sciences Center Denver, Colorado and Rush-Presbyterian-St. Luke’s Medical Center Chicago, Illinois, U.S.A.

The acute respiratory distress syndrome (ARDS) remains a challenge to clinicians and basic scientists alike. Although progress has been made in management, resulting in improved survival (1), the development of effective pharmacological agents to block the basic mechanisms involved in the inflammatory process, which underlies the clinical syndrome, resulting in acute respiratory failure and often multiorgan system failure, remains to be accomplished. This is due in large part to the multiplicity of mechanisms, often redundant, that promote the inflammatory cascade (2). ARDS can be traced back to World War I during which dramatically progressive clinical catastrophes resulting in sudden collapse and respiratory deaths were observed in battlefield casualties (3). Later lung injury, with associated pulmonary edema following trauma of all types, was described as “traumatic wet lung” (4). The pathological state accompanying sudden dramatic development of acute respiratory failure was termed “congestive atelectasis” based on autopsy studies (5). The contribution of the Denver Group appeared in Lancet in 1967 (6). Ashbaugh et al. described 12 patients with a rapidly developing clinical syndrome of acute respiratory failure, characterized by tachypnea-labored breathing, refractory hypoxemia, diffuse bilateral pulmonary infiltrates, and reduced overall compliance of lungs and thorax. Five patients survived with the use of a mechanical ventilator, usually volume-cycled, and the application of positive-end expiratory pressure (7), although the Denver Group had not yet coined the term PEEP (8). The patients who died had congested lungs with collapsed alveoli, cellular debris, and hyaline membranes, which bore a remarkable resemblance to the infantile respiratory distress syndrome. A surfactant abnormality was identified in two autopsy specimens in which these studies were done. Following this report, in 1968 a national conference was held in Washington, D.C., with the Committee on Trauma of the Division of Medical Sciences, National Academy of Sciences, and National Research Council (9). All attendees at this conference presented similar descriptions of the dramatic pulmonary effects of shock and resuscitation in battlefield casualties during the Vietnam war. It was apparent to all in attendance that the observations made by military surgeons were identical to that reported by the Denver Group. Considerable discussions on the role of PEEP permeated this conference (10). Very little was known about the mechanisms involved in lung injury at this juncture. A refinement of the clinical features and factors associated with prognosis

Acute respiratory distress syndrome

2

and more details on principles of management were reported in 1971 (11). Unfortunately, the author used the word “adult” in this and several other reports. This was a mistake. The youngest patient in the series was 11 (age range 11–48 years; mean age 27.3 years). Thus, patients in this series were much younger than those in a more recent series, and certainly there was less comorbidity. In 1973, the 16th Aspen Lung Conference was devoted to ARDS (12). It dealt with a growing body of basic science that was emerging in concert with numerous clinical studies. So far, three Aspen Lung Conferences have dealt with emerging concepts of lung injury and repair in ARDS. The most recent was the 41st Conference in 1998, summarized by Matthay (13). Since that time, a remarkable number of studies have elucidated the mechanisms involved in acute lung injury. Certainly the neutrophil and its products play a role, but the macrophage, an orchestra of proinflammatory cytokines, and the effects of therapy itself on the lung (oxygen high-inflation pressures, etc.) have been added to the complex story of pathogenesis, lung injury, and repair (see Chaps. 6–10). One underlying hypothesis is that ARDS is a result of oxidative stress that is operative in a variety of inflammatory disorders, but this is probably an oversimplication (14). The role of host defenses in the progression of outcome of ARDS has also been considered (15). The involvement of other organ systems, including the kidneys, liver, the hematopoietic system, and the digestive tract, has dominated both clinical and animal model studies during the past 25 years (16). A recent state-of-the-art publication by Matthay et al. made a major contribution in bringing together the diverse concepts of mechanisms of acute lung injury and related organ system dysfunction (2). The epidemiology and risk factors associated with ARDS have been described using different definitions of risk (17, 18) (see Chap. 2). Expanded definitions of ARDS have continued to evolve (19, 20). Costs of caring for ARDS have been described and estimated to be $73,100 per survivor (21). Thus, the economic burden of costs of caring for both survivors and those who die is immense, even if the estimated 150,000 cases in the United States alone is an overestimate (20). A large, multicenter, controlled clinical trial, which showed that low tidal volume ventilation was superior to high tidal volume ventilation, not only in terms of outcome, but in a reduction of multiple organ system damage, was the first major contribution leading to improved survival (22) (see Chap. 21). The ARDS Network has provided the mechanism by which any new therapeutic maneuver, used alone or in combination, can be evaluated to gain evidence concerning the best method of management of ARDS. New approaches to supportive care using nitric oxide or the prone position for mechanical ventilation are aimed at improving oxygen transport across the lung (see Chaps. 17, 20). Preliminary studies have shown increases in oxygenation in some, but not all patients (23, 24). These treatments may be useful when all else fails. It is likely that the surfactant deficiency concept in ARDS has not been adequately addressed. There seems little doubt about surfactant abnormalities in ARDS, which of course cannot be the primary pathogenetic mechanisms but may represent a therapeutic target nonetheless. Figure 1 portrays a simplistic concept, suggesting that damage to surfactant causes alveolar instability and gradual collapse of alveolar units, thus promoting increased flooding of inflammatory pulmonary edema as a result of injury to the air/ blood interface and increased capillary and epithelial permeability (25). Since the

Overview

3

role of surfactant, briefly stated, is to keep alveolar units open, dry, and free of infection, it seems attractive to consider an attack upon the surfactant system to be the “final straw” that unleashes the cascade of events ending in diffuse, yet focal massive alveolar damage (26). Effectively replacing surfactant and restoring its function using products that include surfactant-associated proteins might well be the final defense in the battle between the forces of lung damage and the factors that can thwart a massive attack on the air/blood interface. Preliminary studies suggest that

Figure 1 Hypothesis to explain how mechanisms of capillary/endothelial injury damages or inactivates a surfactant. Alveolar stability and collapse from increased elastic recoil creates hydrostatic forces, favoring further pulmonary edema formation. (From Ref. 25.) surfactant replacement may be beneficial in the early treatment of ARDS (27, 28) (see Chap. 19). The recovery process in ARDS has not been adequately studied. The Denver group reported an encouraging outcome in a small series of survivors (29), as did others (30). More recent series have shown reduction in quality of life and exercise tolerance similar to that experienced by patients with chronic disease (30). In the author’s experience, a

Acute respiratory distress syndrome

4

return to completely normal functioning is not unusual, particularly in younger individuals without comorbidities. The role of corticosteroids in promoting alveolar repair during the fibroproliferative stages of disease is still under study. Preliminary observations comparing corticosteroids with placebo have been encouraging (31). A randomized, prospective, controlled clinical trial is currently underway seeking the definitive answer (15) (see Chap. 18). The chapters in this volume consider all of the concepts presented in this introduction. This should become the new benchmark for our understanding of the pathogenesis and treatment of this common pulmonary catastrophe, which, alas, we may face with increasing frequency as a result of man’s disregard for each other (trauma) and careless life styles.

References 1. Milberg JA, Davis DR, Steinberg KP, Hudson LD. Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983–1993. JAMA 1995; 273:306–309. 2. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342:1334– 1349. 3. Simeone FA. Pulmonary complications of nonthoracic wounds: A historical perspective. J Trauma 1968; 8:625–648. 4. Burford TH, Burbank B. Traumatic wet lung. J Thorac Surg 1945; 14:415–424. 5. Jenkins MT, Jones RF, Wilson B, Moyer CA. Congestive atelectasis—a complication of the intravenous infusion of fluids. Ann Surg 1950; 132:327–347. 6. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967; 2:319–323. 7. Ashbaugh DG, Petty TL, Bigelow DB, Harris TM. Continuous positive-pressure breathing (CPPB) in adult respiratory distress syndrome. J Thorac Cardiovasc Surg 1969; 57:31–41. 8. Petty TL. PEEP. Chest 1972; 61:309–310. 9. Eiseman B. Pulmonary effects of nonthoracic trauma. Introduction to conference. J Trauma 1968; 8:649–650. 10. Grillo HC, Petty TL, Drinker PA. Discussion (Pontoppidan): treatment of respiratory failure in nonthoracic trauma. J Trauma 1968; 8:946–951. 11. Petty TL, Ashbaugh DG. The adult respiratory distress syndrome. Clinical features, factors influencing prognosis and principles of management. Chest 1971; 60:233–239. 12. Petty TL, Hudson LD, Ashbaugh DG, eds. 16th Aspen Lung Conference. Acute pulmonary injury and repair: the adult respiratory distress syndrome. Chest 1974; 65:1S-67S; 66:1S-46S. 13. Matthay MA. 41st Aspen Lung Conference. Acute Lung Injury. Chest 1999; 116:1S-127S. 14. Rahman I, Macnee W. Oxidative stress and regulation of glutathione in lung inflammation. Eur Respir J 2000; 16:534–554. 15. Meduri GU. The role of the host defense response in the progression and outcome of ARDS: pathophysiological correlations and response to glucocorticoid treatment. Eur Respir J 1996; 9:2650–2670. 16. Bell RC, Coalson JJ, Smith JD, Johanson WG. Multiple organ system failure and infection in adult respiratory distress syndrome. Ann Intern Med 1983; 99:293–298. 17. Fowler AA, Hamman RF, Good JT, Benson KN, Baird M, Eberle DJ, Petty TL, Hyers TM. Adult respiratory distress syndrome: risk with common predispositions. Ann Intern Med 1983; 98:593–597. 18. Pepe PE, Potkin RT, Reus DH, Hudson LD, Carrico CJ. Clinical predictors of the adult respiratory distress syndrome. Am J Surg 1982; 144:124–130.

Overview

5

19. Murray JF, Matthay MA, Luce JM, Pick MR. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1988 138:720–723. 20. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall JR, Morris A, Spragg R, and the Consensus Committee for the American-European Conference on ARDS. Definition, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818–824. 21. Valta P, Uusaro A, Nunes S, Ruokonen E, Takala J. Acute respiratory distress syndrome: frequency, clinical course, and costs of care. Crit Care Med 1999; 27:2367–2374. 22. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 23. Staudinger T, Kofler J, Mullner M, Locker GJ, Laczika K, Knapp S, Losert H, Frass M. Comparison of prone positioning and continuous rotation of patients with adult respiratory distress syndrome: results of a pilot study. Crit Care Med 2001; 29:51–56. 24. Dellinger RP. Inhaled nitric oxide versus prone positioning in acute respiratory distress syndrome. Crit Care Med 2000; 28:572–574. 25. Petty TL. The adult respiratory distress syndrome: Historical perspective and definition. Sem Respir Med 1981; 2:99–103. 26. Gattinoni L, Bombino M, Pelosi P, Lissoni A, Pesenti A, Fumagalli R, Tagliabue M. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA 1994; 271:1772–1779. 27. Gregory TJ, Steinberg KP, Spragg R, Gadek JE, Hyers TM, Longmore WJ, Moxley MA, Cai GZ, Hite RD, Smith RM, Hudson LD, Crim C, Newton P, Mitchell BR, Gold AJ. Bovine surfactant therapy for patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 155:1309–1315. 28. Wiswell TE, Smith RM, Katz LB, Mastroianni L, Wong DY, Willms D, Heard S, Wilson M, Hite RD, Anzueto A, Revak SD, Cochrane CG. Bronchopulmonary segmental lavage with Surfaxin (KL4-surfactant) for acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160:1188–1195. 29. Simpson DL, Goodman M, Spector SL, Petty TL. Long-term follow-up and bronchial reactivity testing in survivors of the adult respiratory distress syndrome. Am Rev Respir Dis 1978; 117:449–454. 30. Cooper AB, Ferguson ND, Hanly PJ, Meade MO, Kachura JR, Granton JT, Slutsky AS, Stewart TE. Long-term follow-up of survivors of acute lung injury: lack of effect of a ventilation strategy to prevent barotrauma. Crit Care Med 1999; 27:2616–2621. 31. Meduri GU, Headley AS, Golden E, Carson ST, Umberger RA, Kelso T, Tolley EA. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome: a randomized controlled trial. JAMA 1998; 280:159–165.

2 Definitions and Clinical Risk Factors MARC MOSS Emory University School of Medicine and Grady Memorial Hospital Atlanta, Georgia, U.S.A. B.TAYLOR THOMPSON Harvard Medical School and Massachusetts General Hospital Boston, Massachusetts, U.S.A.

I. Introduction The acute respiratory distress syndrome (ARDS) is characterized by increased permeability of the alveolar capillary membrane, diffuse alveolar damage, and the accumulation of proteinaceous alveolor edema. These pathological changes are accompanied by several physiological alterations including severe hypoxemia and a decrease in pulmonary compliance. ARDS is a relatively uncommon etiology of acute respiratory failure in the intensive care unit. In a 2-week survey of 36 intensive care units in France, ARDS accounted for only 6.9% of all admissions (1). Similarly, ARDS represented 18% of all the patients who required intubation and mechanical ventilation for more than 24 hours in an 8-week survey of 132 intensive care units in Sweden, Denmark, and Iceland (2). However, ARDS patients account for a disproportionately high amount of hospital resources due to the prolonged intensive care unit and length of hospital stays. In one observational study, ARDS patients who required mechanical ventilation for at least 7 days represented only 6% of intensive care unit (ICU) admissions yet comprised 33% of all intensive care unit patient-days and 24% of all hospital charges among intensive care unit patients (3). Because ARDS is a syndrome and not a disease, patients are defined as having ARDS when they meet predetermined diagnostic criteria. The goal of these diagnostic criteria is to identify a homogeneous cohort of patients that represents a unique clinical entity. The specific diagnostic criteria for ARDS have changed and probably improved over time, though the lack of a gold standard or biochemical marker for lung injury remains a limiting factor in determining the sensitivity and specificity of any definition. In this chapter, we will discuss the evolution of the definition of ARDS and the need for additional modifications of the present diagnostic criteria. Over the past two decades sophisticated epidemiological studies have identified a variety of heterogeneous clinical conditions that are associated with a higher likelihood of developing ARDS. These studies have significantly contributed to our improved understanding of the epidemiology

Definitions and clinical risk factors

7

of ARDS. We will also review the clinical and demographic conditions that alter the probability of developing ARDS and impact the likelihood of dying from ARDS.

II. The Definition of ARDS The first official description of ARDS was reported in 1967 by a group of pulmonary and critical care physicians at the University of Colorado (4). This case series of 12 patients described a clinical scenario characterized by the acute onset of dyspnea, tachypnea, severe hypoxemia, chest radiographic abnormalities, and decreased static respiratory system compliance. With the increased availability of pulmonary artery catheterization in intensive care units, ARDS was reported to be a noncardiogenic form of pulmonary edema, characterized by the accumulation of both protein and cells in the alveoli in the presence of normal left ventricular filling pressures. Subsequently, several ARDS definitions were used in the early 1980s that required at least four basic clinical features, three of which are based upon physiological and radiographic criteria that were used in this original case series: hypoxemia (varying severity), decreased respiratory system compliance, and chest radiographs (often of an ill-defined type and degree). The fourth diagnostic criterion is usually the documentation of normal pulmonary artery occlusion pressures using a pulmonary artery catheter (5–10). When the mortality from ARDS did not improve during the 1980s, some investigators raised the possibility that these four strict diagnostic criteria biased the understanding of ARDS and contributed to the negative therapeutic trials for ARDS (11). One concern was that the diagnostic criteria did not have sufficient sensitivity and therefore only identified those patients with severe ARDS and a very poor prognosis. The potential lack of sensitivity was attributed to the necessity for placing a pulmonary artery catheter in order to document a normal pulmonary artery occlusion pressure (10). In a 3-month survey, Rinaldo (11) reported that only 7 of 27 patients with clinical ARDS met all of the strict diagnostic criteria for ARDS. The mortality for these 7 patients was 71% and for the remaining 20 patients only 30%. The requirement for pulmonary artery catheterization to diagnose ARDS may also delay the initiation of therapeutic agents designed to prevent the development of ARDS in at-risk individuals (12). Because approximately 50% of patients develop ARDS within the first 24 hours of meeting an at-risk diagnosis, delaying the administration of a therapy in order to insert a pulmonary artery catheter may diminish the chance of successful therapeutic intervention (13). Finally, the use of a pulmonary artery catheter has been circumstantially linked to adverse clinical outcomes, perhaps from well-intended but potentially harmful pulmonary artery catheter-guided management strategies (14). These initially proposed strict diagnostic criteria of ARDS may also lack specificity. Other pulmonary diseases that represent different inflammatory processes can fulfill all of the diagnostic criteria of ARDS. For example, patients with vasculitis and alveolar hemorrhage meet the diagnostic criteria for ARDS, yet the pathogenesis of this disorder is different from ARDS. Therefore, it is unclear whether patients with alveolar hemorrhage patients should reported as having ARDS (10). In addition, patients with an elevated pulmonary artery wedge pressure are excluded, although these patients may have lung injury in addition to either hypervolemia or congestive heart failure. Patients

Acute respiratory distress syndrome

8

with bilateral pneumonia secondary to Pneumocystis carinii meet criteria for ARDS and are sometimes included in both clinical and epidemiological studies (15, 16). Other reports have excluded patients with AIDS, and therefore individuals with Pneumocystis carinii pneumonia would not be enrolled (12, 17). These discrepancies in the inclusion of patients with bilateral pneumonia in studies of ARDS illustrate the persistent concern of variability in the diagnostic criteria used at different centers. Several investigators postulated that the differences in reported epidemiological data, such as mortality rates, could be attributed to inconsistent cutoff values for the hypoxemia criteria and variations in the interpretation of other diagnostic considerations (8). Finally, these four diagnostic criteria defined ARDS as an all-or-none phenomenon. The presentation of ARDS includes a continuum of radiographic and arterial blood gas abnormalities, and any single cutoff value for the definition of ARDS would be arbitrary. Therefore, the identification of several gradations or categories of acute lung injury that have prognostic implications would be beneficial. A. The Murray Lung Injury Score Based on these concerns, a second phase of ARDS definitions emerged that attempted to improve both the sensitivity and specificity of the diagnostic criteria by including early and limited cases of ARDS, or what some called clinical acute lung injury, while still excluding patients who did not truly have acute lung injury (10, 18). In 1988 Murray and colleagues proposed an expanded definition of ARDS that represented the first attempt to provide a quantitative scoring system to characterize mild, moderate, and more severe forms of lung injury (Table 1) (5). The lung injury score is based upon four components (chest radiograph, hypoxemia, positive end-expiratory pressure, and respiratory system compliance), two of which (chest radiograph and hypoxemia) must be available for all patients. Each component is assigned a score of 1 to 4. The final value is obtained by dividing the aggregate sum by the number of available components, and three categories of lung injury are defined. A final score of zero equals no lung injury, 0.1–2.5 constitutes mild to moderate lung injury, and >2.5 is defined as ARDS. The authors also recommended that this scoring system be used only in patients with specific diagnosis, such as sepsis and trauma. However, some studies that have subsequently used the lung injury score have not always adhered to this recommendation (19, 20). In addition, the placement of a pulmonary artery catheter and measurement of a pulmonary capillary occlusion pressure were not required. This scoring system was immediately praised and used to stratify patients with and at risk for ARDS (21). One study has evaluated the ability of the lung injury score to predict a complicated clinical course in 50 ARDS patients (22). The lung injury score was determined 4 days after the development of ARDS. Using a cutoff value of ≥2.75, the sensitivity and specificity of the lung injury score for a complicated course, defined as death before 14 days or the requirement for mechanical ventilation for longer than 14 days, were 83% and 57%, respectively. B. A Simpler Definition of ARDS In an attempt to diagnose patients earlier in the course of ARDS, Sloane and colleagues proposed a new definition of ARDS based on only two diagnostic criteria: a

Definitions and clinical risk factors

9

ratio of ≤250 in an appropriate clinical setting and bilateral infiltrates on chest radiograph within 7 days of meeting the at risk diagnosis (12). Patients with physical findings or hemodynamic measurements consistent with congestive heart failure, stage III or IV lung

Table 1 The Lung Injury Score Chest Roentgenogram Score No alveolar consolidation

0

Alveolar consolidation in one quadrant

1

Alveolar consolidation in two quadrants

2

Alveolar consolidation in three quadrants

3

Alveolar consolidation in four quadrants

4

Hypoxemia Score 0 1 2 3 4 Respiratory System Compliance Score (when ventilated) (mL/cmH2O) ≥80

0

60–79

1

40–59

2

20–39

3

19

4

Positive End-Expiratory Pressure Score (when ventilated) (cmH2O) ≤5

0

6–8

1

9–11

2

12–14

3

≥15 Final Value

4 a

No lung injury

0

Acute lung injury

0.1–2.5

Severe injury (ARDS)

>2.5

Acute respiratory distress syndrome

a

10

Obtained by dividing aggregate sum by number of components used.

cancer, Pneumocystis carinii pneumonia, organ transplantation, or younger than 15 years of age were excluded for the study. In this 2-year evaluation, 153 patients were diagnosed with ARDS. When compared to a strict definition of ARDS, this liberal definition allowed 13% of the patients to be diagnosed with ARDS 3 days earlier (23). In this study an additional 2% of patients were diagnosed with ARDS who would have not been identified using the strict criteria. Furthermore, the mortality of ARDS patients in this study was still 54%, unchanged from previously reported mortality rates (24). C. The First American-European Consensus Conference In 1994 the recommendations of the American-European Consensus Conference (AECC) on ARDS were published (25). One of the major goals of this conference was to bring “clarity and uniformity to the definition of ARDS.” ARDS was defined as a severe form of “acute lung injury,” defined as “a syndrome of inflammation and increasing permeability that is associated with a constellation of clinical, radiographic, and physiologic abnormalities that cannot be explained by, but may coexist with, left atrial or pulmonary capillary hypertension” (25). The diagnostic criteria for ARDS proposed by this committee were bilateral infiltrates on chest radiograph, and pulmonary artery occlusion pressure ≤18 mmHg when measured or no clinical evidence of left atrial hypertension. The spectrum of disease severity was also expanded to include patients with milder hypoxemia. This definition, which includes patients with ARDS, was called acute lung injury (ALI), and the diagnostic criteria were similar except the ratio was ≤300 (Table 2). The term “non-ARDS ALI” subsequently evolved of 201–300. This definition was to refer to the subgroup of patients with a immediately endorsed by the American Thoracic Society and the European Society of Intensive Care Medicine with the goal of universal acceptance and utilization (25). These three definitions—the Murray lung injury score, the simple criteria defined by Sloane and colleagues, and the AECC definition—were compared to the strict definition of ARDS requiring all of the strict diagnostic criteria in attemp to determine the accuracy of these new definitions (18). All three definitions maintained a high degree of accuracy (>90%) for those ICU patients with a clearly defined at-risk diagnosis for the development of ARDS. Therefore, it is likely that the lung injury score and the AECC definition actually identify a patient population similar to the older strict definitions of ARDS for those patients with clearly defined at-risk diagnoses. D. Update from the Second American-European Consensus Conference Over the next few years the AECC committee met several times with the primary objective of examining the pathophysiological mechanisms of lung damage as they related to mechanical ventilatory strategies and promising agents for the treatment or prevention of acute lung injury and ARDS (26). Though the primary goals of the meeting

Definitions and clinical risk factors

11

were not to examine the diagnostic criteria of ARDS, one subcommittee did recommend that ARDS be identified as the pulmonary manifestation of systemic inflammation. The

Table 2 Recommended Criteria for Acute Lung Injury and Acute Respiratory Distress Syndrome Timing

Oxygenation

Chest radiograph Pulmonary artery wedge pressure

ALI criteria

Acute onset

Bilateral infiltrates ≤18 mmHg when mmHg seen on frontal chest measured or no clinical (regardless of PEEP level) radiograph evidence of left atrial hypertension

ARDS criteria

Acute onset

Bilateral infiltrates ≤18 mmHg when mmHg seen on frontal chest measured or no clinical (regardless of PEEP level) radiograph evidence of left atrial hypertension

Table 3 Stratification System of Acute Lung Injury Letter G

Meaning Gas exchange

Scale

Definition

0 1 2 3

Gas exchange (to be combined with the numeric descriptor)

O

C

A

Organ failure

Cause

Associated diseases

A

Spontaneous breathing, no PEEP

B

Assisted breathing, PEEP 0–5 cmH2O

C

Assisted breathing, PEEP 6–10 cmH2O

D

Assisted breathing, PEEP ≥10 cmH2O

0

Lung only

1

Lung+1 organ

2

Lung+2 organs

3

Lung+≥3 organs

0

Unknown

1

Direct lung injury

2

Indirect lung injury

0

No coexisting diseases that will cause death within 5 years

1

Coexisting disease that will cause death within 5

Acute respiratory distress syndrome

12

years but not within 6 months 2

Coexisting disease that will cause death within 6 months

development of sequential dysfunction of nonpulmonary organ systems such as the kidney and liver or multiple organ dysfunction syndrome (MODS) is one of the common causes of death for ARDS patients (27). Based on this concept, the subcommittee recommended a new stratification system of acute lung injury and ARDS, the GOCA score, based on alterations in gas exchange, organ failure, the cause of acute lung injury, and associated diseases (Table 3). The scoring system was not designed to actually predict outcome, but to succinctly present important clinical information (26). Unlike the diagnostic criteria developed during the first conference, the GOCA score has not been widely accepted or reported in the medical literature.

III. Persistent Problems and Controversies Involving the Definition of ARDS In spite of several alterations in the definition of ARDS or clinical lung injury, fundamental problems still remain with the present diagnostic criteria recommended by the AECC. Variability in the identification of all three components of the AECC definition of ARDS (radiographic, oxygenation, and the exclution of a cardiogenic form of pulmonary edema) still can occur and will continue to impact the validity and reproducibility of epidemiological studies and clinical trials. Some of these issues have recently been reviewed (28). The chest radiographic criteria used to define ARDS remain problematic. In one study, 21 established investigators reviewed 28 randomly selected chest radiographs from intubated patients with a of <300 (29). They were asked to determine whether the radiograph fulfilled the AECC definition of acute lung injury/ARDS of “bilateral infiltrates consistent with pulmonary edema.” The threshold for concluding that a chest radiograph was consistent with the AECC radiographic interpretation ranged from 36 to 71% among the investigators. Similar variability exists when chest radiographs are scored using the 4-point radiographic criteria from the Murray lung injury score. Chest radiographs from patients with ARDS based on clinical criteria were blindly interpreted by radiologists, anesthesiologists, and critical care physicians (30). There was very good agreement in the radiological score when interpreted by two radiologists (kappa 0.9) (18, 30). However, when chest radiographs were interpretation by anesthesiologists or critical care physicians, the interrater reliability was only fair to poor (30). More recently, the impact of formal training on the inadequate consistency of chest radiographic interpretation has been examined (31). Chest films were obtained from patients enrolled in a randomized trial of ventilatory strategy for patients at risk for the development of ARDS. Chest radiographs were scored for the presence or absence of diffuse bilateral infiltrates using the AECC definition of ARDS. A set of 63 films was independently interpreted by one critical care physician and one radiologist. Similar to previous studies, their interrater reliability was only considered moderate. However, after

Definitions and clinical risk factors

13

a standardized session during which the two physicians refined the standards and rules they would apply to radiographic interpretation, their interrater reliability was nearly perfect. These studies demonstrate two important points concerning chest radiographic interpretation as part of the diagnosis of ARDS. Presently, significant variability exists in the radiographic interpretation of chest radiographs that could account for some of the perceived geographic differences in the epidemiology of ARDS. In addition, formal training in radiographic interpretation among investigators is necessary to decrease the heterogeneity of patients enrolled in clinical trials for ARDS and should be considered a mandatory component of the responsibility of principal investigators involved in studies of ARDS patients. A. Oxygenation and PEEP The hypoxemia component of the AECC definition also remains controversial. Individual or institutional variations in mechanical ventilator strategy may impact the incidence of ARDS/ALI in an intensive care unit or medical center. Presently the oxygenation ratio, regardless of the mode of requirement for ARDS is based upon the ventilation. Increasing the level of positive end-expiratory pressure (PEEP) alone may and directly alter the ratio. Therefore, it may be possible improve the that a patient who initially meets the criteria for ARDS or ALI can subsequently improve the classification of their disease simply by altering the PEEP. For example, several investigators have recently advocated the initiation of recruitment maneuvers using a PEEP of 40 cmH2O for a limited period of time in order to successfully recruit previously closed alveoli in attempt to improve oxygenation (15). One case report described the remarkable improvement in oxygenation with recruitment maneuvers that changed the ratio from <100 to 270 (32). The initiation of recruitment maneuvers in this patient would change the diagnosis from ARDS to ALI. Though the incorporation of PEEP as part of the oxygenation criteria may resolve this issue, it would also add a subjective nature to the oxygenation requirement. Physicians who routinely use higher levels of PEEP could alter the classification of their patients simply due to their clinical practice. This problem already exists with the lung injury scoring system as higher point values will occur by increasing the level of PEEP. Other questions concerning the oxygenation criteria also remain. Should the worst degree of hypoxemia over a 24-hour period be used in the definition of ARDS similarly to most severity-of-illness scoring systems (28)? More uniform utilization of mechanical ventilation for patients at risk for or with ARDS may diminish some of the variability in the incidence of ARDS/ALI. Finally, it is uncertain how to properly determine the actual

and therefore the

ratio for spontaneously breathing patients. B. Exclusion of Hydrostatic Pulmonary Edema In order to differentiate ARDS from hydrostatic (cardiogenic/volume overload) pulmonary edema, the AECC consensus definition includes either a pulmonary artery occlusion pressure of ≤18 mmHg with a pulmonary artery catheter or no clinical evidence

Acute respiratory distress syndrome

14

of left atrial hypertension. However, the AECC committee also stated that ARDS may coexist with “left atrial hypertension” (25). Recent studies have demonstrated that some patients who meet clinical criteria for ARDS may have a pulmonary capillary occlusion pressure that is above the selected cutoff value of 18 mmHg. In one study, 11% of ratio criteria initially had a pulmonary patients meeting chest radiograph and artery occlusion pressure of >18 mmHg, which would exclude them from being diagnosed with ARDS (33). However, over the course of their illness, these patients developed a normal pulmonary artery occlusion pressure while still meeting the radiographic and oxygenation requirements for ARDS. A high precentage of these individuals were trauma patients. Applying the diagnosis criteria of the AECC definition of ARDS, these patients would not be classified as having ARDS until several days after the development of their diffuse pulmonary infiltrates and hypoxemia. Another study reported the range of pulmonary artery occlusion pressure measurements in a cohort of patients with or at high risk for ARDS (34). Patients with a history of heart failure, at high risk for cardiac arrythmias or myocardial ischemia, or with a clinical suspicion of cardiogenic pulmonary edema were excluded. Of these patients, 59% had pulmonary artery catheters inserted during their clinical course, and 82% had at least one pulmonary artery occlusion pressure measurement of >18 mmHg. The interrater reliability and accuracy of pulmonary artery occlusion pressures suffer from similar problems previously discussed with chest radiographic interpretations. This potential variability in the pulmonary artery occlusion pressure measurements may also account for the some of the geographic variability in epidemiological data concerning ARDS. Finally, patients with ARDS are often on high levels of PEEP. The increased intrathoracic pressures from the PEEP can often be registered by the pulmonary artery catheter and lead to a false elevation of the actual intravascular occlusion pressure. Therefore, it is possible that a patient who truly has ARDS may be excluded from clinical trials or epidemiological studies because they are severely hypoxemic and require significant levels of PEEP that cause the pulmonary artery occlusion pressure to be >18 mmHg. In regard to the diagnostic criteria related to clinical evidence of left atrial hypertension, the accuracy and reliability of the clinical assessment of this measurement is poor (28, 35–37). In addition, there are no specific recommendations by the AECC committee about what clinical signs are acceptable for estimating left atrial pressure. One group has reported that between 5 and 30% of patients meeting chest radiograph and hypoxemia criteria would be excluded from a formal diagnosis of ARDS depending on the specific diagnostic criteria used to define clinical left atrial hypertension (38). Some investigators have attempted to use simple criteria obtained from the chest radiograph to differentiate cardiogenic/hydrostatic pulmonary edema from ARDS (39, 40). Aberle and colleagues reviewed portable chest radiographs from 45 mechanically ventilated patients with pulmonary edema. The ratio of pulmonary edema fluid protein to plasma protein concentration was utilized to differentiate hydrostatic from increased permeability edema. Based on cardiac size, distribution of edema, presence of either pleural effusions, interstitial changes, or air bronchograms, and the distribution of blood flow and pulmonary blood volume, chest radiographs were classified as having hydrostatic, increased permeability, or a mixed form of pulmonary edema. Overall, 87% of patients with hydrostatic edema but only 60% of patients with increased permeability edema were correctly identified by chest radiographs. The presence of a patchy peripheral distribution

Definitions and clinical risk factors

15

of edema fluid was the most discriminating criterion for increased permeability edema. More recently, conventional portable supine chest radiographs from 33 mechanically ventilated ICU patients were evaluated by three experienced chest radiologists without any information about the clinical course of the patients, their mechanical ventilatory parameters, hemodynamic measurements, or prior chest radiographs (40). All patients had a pulmonary artery catheter in place. The investigators measured the vascular pedicle width (VPW) by dropping a perpendicular line from the point at which the left subclavian artery exits the aortic arch and measuring across to the point at which the superior vena cava crosses the right mainstem bronchus. In addition, the cardiothoracic (CT) ratio was calculated by dividing the widest transverse diameter of the cardiac silhouette by the widest transverse diameter of the thorax above the diaphragm. Patients were classified as having either cardiogenic/hydrostatic pulmonary edema or permeability pulmonary edema based on pulmonary artery occlusion pressure, cardiac index, and clinical diagnosis. The mean accuracy of the radiologists in distinguishing the two types of pulmonary edema was only 41%. However, by including additional information derived from VPW (cutoff value of >63 mm) and CT ratio (cutoff value of >0.52), the accuracy of determining the type of pulmonary edema increased to 73%. Further investigation is necessary to examine whether these simple radiographic measurements should be used to assist with identifying patients with ARDS (see Chap. 4). When ARDS was first described in 1967, the “A” stood for “adult” to differentiate it from the infantile respiratory distress syndrome, which had many similar clinical characteristics. However, with the recognition that ARDS occurs in all age groups, the “A” now stands for “acute.” There are no standard criteria to define over what length of time the syndrome of ARDS can occur. The original AECC investigators did not set a time limit for the word “acute,” but clearly ARDS needs to be differentiated from interstitial lung diseases that develop over weeks to months (25). An epidemiological study from Seattle involving 695 critically ill patients provides insight into this issue (13). ARDS developed during the first 24 hours in 54% and 29% of the patients with sepsis and trauma, respectively. Over 90% of all patients developed ARDS within 5 days of meeting the at risk diagnosis, and all patients developed ARDS by 7 days. Therefore, the length of time for the development of ARDS should be less than 7 days from the time of onset of their critical illness. C. Update from the Third American-European Consensus Conference The America-European Consensus Conference convened for a third time to discuss some of the criticisms and controversies that persist with the present diagnostic criteria for ARDS (41). The committee addressed several important concerns and suggested specific recommendations to improve the consistent usage of the diagnostic criteria. In addition, they acknowledged that the theoretical differentiation of ARDS from ALI based on severity of hypoxemia has not established two separate entities with different clinical associations and prognoses. In regard to the chest radiographic criteria, the committee stated that the “bilateral infiltrates” should be consistent with pulmonary edema, even if mild or patchy in nature.

Acute respiratory distress syndrome

16

Opacities that are not considered appropriate for the radiographic criteria for ALI/ARDS include pleural effusions, pleural thickening, or masses, pulmonary masses or nodules, chronic scarring, volume loss, lobar collapse, platelike atelectasis if the surrounding borders are sharp, extrathoracic opacities, and subcutaneous air. It is unclear if these more explicit descriptions of qualifying radiographic opacities will improve interreader variability. In regard to the timing of the onset of ALI/ARDS, the committee recognized the findings from Seattle and suggested that the diagnostic criteria of lung injury should appear over an interval not exceeding 7 days. The AECC committee also commented on the difficulty in excluding hydrostatic or cardiogenic causes as the sole cause for pulmonary edema. They acknowledged the lack of a perfect cutoff value of the pulmonary artery occlusion pressure that would differentiate the hydrostatic pulmonary edema from permeability pulmonary edema (ARDS). Additionally, the capillary filling pressures clearly fluctuate with volume resuscitation and diuresis, and the pulmonary artery occlusion pressure may occasionally rise above or below the specific cutoff value. However, no alterations in this controversial diagnostic criterion were recommended. Finally, the AECC committee identified a group of patients who spend a considerable length of time being mechanically ventilated. This new category of respiratory failure has been termed “acute lung failure” and includes all patients who require mechanical ratio of ≤300, and show any infiltrate on chest radiograph ventilation, have a (Table 4). Therefore, patients with ALI/ARDS would be a subset of “acute lung

Table 4 Diagnostic Criteria for Various Forms of Acute Pulmonary Dysfunction Chest radiographc Acute lung injurya

0–300 mmHg

Bilateral infiltrates

a

0–200 mmHg

Bilateral infiltrates

Acute lung failure

0–300 mmHg

Any infiltrate

Acute lung injury

a

Acute indicates development of the syndrome within a time period of less than 7 days. Further, both oxygenation and chest radiograph criteria must be simultaneously present within a time window of no greater than 24 hours in order to make the diagnosis. b These levels of abnormality in gas exchange should be reasonably sustained as opposed to transient. c Bilateral opacities seen on chest radiograph or computerized tomography consistent with pulmonary edema. These opacities can be mild, patchy, or asymmetrical.

failure.” Patients with pure ventilatory impairment due to neuromuscular disease, chronic obstructive pulmonary disease (COPD), and acute exacebations of asthma were excluded from this new classification. The demographics, epidemiology, and economic impact of patients with acute lung failure are presently unknown and will require future investigation.

Definitions and clinical risk factors

17

D. Future Directions Even after three consensus conferences and multiple studies, several issues still remain concerning the present diagnostic criteria for ARDS, including the heterogeneity of the patient population, the absence of a diagnostically accurate biochemical marker, and the lack of a reliable noninvasive measure of alveolar-capillary injury or permeability. Presently, ARDS can occur in a heterogeneous group of patients who develop common physiological and radiographic abnormalities. Therefore is ARDS really one unique or several different syndromes? The present definition proposed by the AECC includes a multiplicity of clinical entities ranging from autoimmune disorders, such as lupus pneumonitis, to direct lung injury attributable to causes as diverse as pneumonia or smoke inhalation, to indirect pulmonary injury from bacteremia, trauma, or pancreatitis (42). This issue was first addressed over 20 years ago, when John Murray and Thomas Petty engaged in a published debate concerning ARDS (43, 44). Murray stated that “lumping these disorders together serves no useful purpose and has the disadvantage of detracting from important and distinctive differences in pathogenesis, therapy, and prognosis.” Petty rebutted by observing that ARDS is indeed a specific clinical syndrome manifest by similar clinical, pathophysiological, and morphological features. He blamed some of the difficulties with and misunderstanding of the definition of ARDS on its misuse by clinicians who were unfamiliar with the syndrome. Finally, he concluded that ARDS is a heterogeneous disease not unlike asthma, which is still defined by the pathophysiological response caused by a variety of potential stimuli (44). In support of Murray’s opinion as a “splitter,” the level of certain mediators that are postulated to be involved in the pathogenesis of ARDS vary according to the specific atrisk diagnosis. For example, tumor necrosis factor levels are increased in the blood of patients with sepsis, and in one study the levels were positively associated with the development of ARDS (45). However, several studies have reported the absence of tumor necrosis factor (TNF) levels in trauma patients without clinically significant hypovolemia (23). Similarly, interleukin-1 β (IL-1) has been detected in some patients with sepsis, but circulating IL-1 a levels are undetectable within the first few hours of a traumatic injury (46, 47). Furthermore, there is evidence that total IL-1 production actually is decreased for the first 5 days after traumatic injury (48). Finally, in a cohort of patients with sepsis and trauma, 26% of whom developed ARDS, levels of the soluble adhesion molecules (E-selectin and intercellular adhesion molecule-1) were significantly higher in the septic patients when compared to the trauma patients (49). In addition, the circulating levels of both adhesion molecules in the trauma patients were not elevated above normal controls. These studies suggest that the pathogenesis of ARDS may be different in patients with different at-risk diagnoses or that these specific mediators are not intricately involved or necessary for the development of ARDS. The most recent AECC committee commented of this issue on heterogeneity. The committee acknowledged that the various forms of lung injury may eventually necessitate different therapeutic modalities, and that narrower definitions may be useful in the future. They did not believe that there was sufficient evidence or need to subdivide the ARDS/ALI patients according to the specific underlying process.

Acute respiratory distress syndrome

18

The identification of an accurate diagnostic, predictive, or prognostic marker for ARDS would significantly improve the ability to diagnose patients with ARDS and improve our understanding of this syndrome. The discovery of such a mediator or marker of lung injury would be analogous to the identification of CPK-MB or troponin levels in the diagnostic evaluation of acute myocardial infarctions. As described in several other chapters in this volume, ARDS is associated with the initiation and propagation of an intense inflammatory cascade involving myriad inflammatory cells and mediators that result in injury to the alveolar epithelium. Many of the mediators involved in this cascade have been identified over the last few years, and their specific actions have been carefully explored. However, these mediators require certain characteristics to be a clinically useful marker. The actual collection of the specimen from the patient must be rapid and not expose the patient to any excessive risk (50). In addition, the laboratory assay and its interpretation must be uniform among the various clinical laboratories able to perform the test. Therefore, the most likely source of a biochemical marker for ARDS would be obtained either from the peripheral blood or possibly from endotracheal tube aspirate of edema fluid or urine. Numerous studies have attempted to identify a clinically useful biochemical marker for ARDS. This extensive body of research has been reviewed (51). Although much has been learned about the pathogenesis of ARDS from these studies, there is presently no clearly identified biochemical marker for ARDS. The most fundamental physiological characteristics of ARDS is an increase in permeability to protein across the endothelial and epithelial barrier of the lung (52, 53). Measurements of pulmonary vascular permeability (PVP) are commonly used in experimental models of ARDS as a marker of acute injury. Noninvasive nuclear medicine techniques have been developed that measure this feature in patients. More specifically, positron emission tomography (PET) can measure protein flux between intravascular and extravascular components of the lung. Using this technique, ARDS patients have been reported to have increased PVP when compared to normal controls (53). The ability of PVP measurements to diagnose ARDS has also been examined. In this study, these noninvasive techniques were utilized in an attempt to differentiate patients with ARDS from those with commonly confused diagnoses such as congestive heart failure and pneumonia (54). Though PVP was significantly higher in ARDS patients when compared to those with congestive heart failure, they were not different from measurement of PVP in patients with pneumonia, both in the regions with infilitrate and in radiographically normal areas. However, some investigators consider patients with bilateral pneumonia as meeting the diagnostic criteria for ARDS. Further investigation is needed before these noninvasive measures of PVP can be incorporated into the definition of ARDS.

IV. Clinical Risk Factors for the Development of ARDS Since its initial description, ARDS has been noted to occur in patients with a variety of heterogeneous diagnoses. This heterogeneity was better defined in the 1980s, when several investigators prospectively followed critically ill patients and identified those that eventually developed ARDS (6, 7). Over a 12-month period, Fowler and colleagues at the University of Colorado (6) identified and monitored all patients who required mechanical ventilation with one of eight conditions, including cardiopulmonary bypass, burns,

Definitions and clinical risk factors

19

bacteremia, hypertransfusion, multiple long bone or pelvic fractures, disseminated intravascular coagulation, severe pneumonia, and aspiration of gastric contents (6). Patients were considered to have developed ARDS based on strict criteria including the requirement of a pulmonary artery occlusion pressure of ≤12 mmHg. In total, 88 patients developed ARDS, with an incidence rate ranging from 1.7 per 100 patients with cardiopulmonary bypass to 35.6 per 100 patients with pulmonary aspiration (Table 5). Pepe and colleagues at the University of Washington performed a similar study and followed patients with a variety of diagnoses, including sepsis syndrome, aspiration of gastric contents, pulmonary contusion, hypertransfusion, and multiple fractures (7). A total of 34% of the 136 consecutive patients identified developed ARDS. The incidence of ARDS among the various at-risk diagnoses ranged from 8% for major fractures to 38% with sepsis syndrome (Table 5). Both of these studies also reported that patients with multiple at risk diagnoses were at a markedly increased risk for developing ARDS. Several years later, Hudson and colleagues reported similar incidence rates for the development of ARDS in critically ill patients (13). In general, the majority of these heterogeneous patients who are at increased risk for the development of ARDS can be classified into four common categories: sepsis (pulmonary or nonpulmonary), pneumonia,

Table 5 Incidence of ARDS Among Patients with Specific At-Risk Diagnoses At-risk diagnosis

Incidence rates, 1980– Incidence rates, 1981 (Fowler Series) 1982 (Pepe series)

Sepsis Aspiration of gastric contents

36%

Pulmonary contusion

Incidence rates, 1983–85 (Hudson series)

38%

41%

30%

22%

17%

22%

Hypertransfusion

5%

24%

36%

Multiple fractures

5%

8%

11%

66%

33%

Near-drowning Cardiopulmonary bypass

2%

Burns

2%

major trauma, and aspiration of gastric contents. In some series, these four general categories may account for over 85% of the patients who will develop ARDS (55). Sepsis is usually the most common at-risk diagnosis and may account for approximately 50% of all ARDS cases. Whether these four categories of ARDS represent clinically different syndromes with different epidemiology and pathogenesis is presently unclear. Other investigators have recommended different methods of categorizing the various at-risk diagnoses according to whether the injury to the lung occurs through direct (primary) or

Acute respiratory distress syndrome

20

indirect (secondary) mechanisms (56, 57). There are diverse causes of direct injury, including pneumonia, aspiration of gastric contents, and smoke inhalation. The etiologies of indirect lung injury are even more heterogeneous and include such diagnoses as nonpulmonary sepsis and multiple long bone or pelvic fractures. This differentiation (between direct and indirect injury) is supported by different respiratory mechanics, responsiveness to PEEP, and radiographic manifestations on CT scanning. CT scans from ARDS patients due to direct lung injury have a prevalence of consolidation as opposed to more prevalent edema and alveolar collapse in ARDS patients from an indirect source (57). But clinical response to low tidal volume ventilation does not suggest that there is an actual difference (58). In an analysis of patients enrolled into the ARDSnet trial, the efficacy of ventilatory strategy was similar among patients with different clinical risk factors (58). More recently, one study reported that patients following pulmonary resection surgery are at increased risk for the development of ARDS. In a study of 1139 patients who underwent pulmonary resection surgery, 3.9% of the patients developed ALI or ARDS on average 4 days postoperatively (59). Over half of the operations were performed on patients with lung cancer. Because lung surgery is a fairly common procedure, this relatively low percentage still represented nearly one new ARDS patient every 2 months from pulmonary resection at a single institution. The highest frequency of ARDS (12.9%) was observed in patients who required extensive resections. A. Comorbidity, Demographics, and Incidence of ARDS Not all of the patients with a clearly defined at-risk diagnosis will eventually develop ARDS. In the classic epidemiological studies from the University of Colorado and the University of Washington in Seattle, the overall incidence of ARDS in patients with specific at-risk diagnoses was reported to be only 7% and 34%, respectively (6, 7). No individual diagnoses were associated with an incidence of ARDS of >40%. Therefore, other factors in addition to the specific at-risk diagnosis must play a role in determining which at-risk patients eventually develop ARDS. Some of these secondary demographic or comorbid conditions that alter the probability of developing ARDS have been identified. Alcohol is one of the most commonly used drugs in the world. Over 50% of the U.S. population regularly consumes alcohol (60). However, it was not until recently that an epidemiological association between alcohol abuse and ARDS was identified. Chronic alcohol abuse was reported to increase the risk of complications in trauma patients (61). The risk of respiratory failure, defined as respiratory distress requiring mechanical ventilation, was higher among the trauma patients with evidence of chronic alcohol abuse. In another study, 71% of critically patients with a history of an alcohol-related illness and a low arterial pH developed ARDS compared to only 39% of those patients with no history of alcohol abuse and a normal pH (13). More recently, 351 critically ill patients with one of seven diagnoses associated with the development of ARDS were identified and followed for the development of ARDS (17). Thirty-four percent (121/315) of the patients had a prior history of alcohol abuse. Using a strict definition of ARDS, 43% (52/121) of the alcoholics developed ARDS, as opposed to 22% (50/230) of the nonalcoholics. This effect of chronic alcohol abuse on the development of ARDS

Definitions and clinical risk factors

21

remained significant in a logistic regression model, adjusting for differences in the admission APACHE II scores, at-risk diagnosis, and gender. Recently, two studies have examined the effect of a prior history of cigarette use on the development of ARDS. Using a retrospective cohort study design, 56 ARDS patients were identified from an HMO data base of 121,012 health plan subscribers (62). The risk of ARDS was increased in those patients who were current or former smokers compared to those individuals who had never smoked. A similar association between smoking and the development of ARDS were observed in a large cohort of patients undergoing coronary artery bypass surgery (63). ARDS appears to be more common with increasing age in patients with similar underlying diagnoses (55). In the subset of trauma patients enrolled in a recent study, those individuals older than 70 were twice as likely to develop ARDS when compared to the 18- to 29-year-old patients (13). Increasing age has also been reported to have positive association with the development of ARDS in burn patients (64). Only a few studies have reported a positive association between gender and the development of ARDS. Hudson and colleagues have reported that female trauma patients were more likely to develop ARDS than male trauma patients given a similar severity of illness (13). In Kutlu’s series of patients undergoing pulmonary resection, men had a higher frequency of developing ARDS when compared to women (59). Clearly larger epidemio-logical studies are necessary to determine whether there are gender differences in the pathogenesis of ARDS. Diabetes mellitus is a secondary diagnosis that theoretically alters the incidence of ARDS due to its association with abnormalities in several aspects of the pathogenetic cascade of ARDS. A multitude of data support the theory that proinflammatory signals, and more specifically the activation and recruitment of circulating neutrophils into the lung parenchyma, are involved in the pathogenesis of ARDS. Some investigators hypothesized that the same alterations in the inflammatory cascade that predispose diabetic patients to develop serious infections may be protective for the development of ARDS. One study identified 113 patients with septic shock, of whom 28% had a history of diabetes (65). In this study, nondiabetics were more likely to develop septic shock from a pulmonary source (48%, 39/81) when compared to diabetics who were more likely to develop ARDS from an indirect source of infection such as a wound or urinary tract (25%, 8/32). Overall, 41% (46/113) of the patients with septic shock developed ADRS. The incidence of ARDS was significantly higher in the nondiabetic patients when compared to those with a history of diabetes. In a multivariate logistic regression analysis adjusting for several variables including source of infection, the effect of diabetes on the incidence of ARDS remained significant.

V. Mortality from ARDS Historically, the mortality rate for patients with ARDS has usually exceeded 50%. However, mortality appears to have declined over the past few years. One single institution study reported a steady decline in mortality beginning in 1989 and reaching a low of 36% in 1993 (66). This improvement in mortality occurred in all age groups. In a subsequent study from England, the mortality rate for patients with ARDS was also

Acute respiratory distress syndrome

22

shown to have decreased from 60 to 30% from the early to mid-1990s (67). Most recently, the ARDSnet trial examining ventilatory strategies for ARDS patients reported a mortality rate of 30% in patients who received low tidal volume ventilation (16). However, this large multicenter trial excluded patients with comorbidities associated with a high mortality (such as ARDS with liver disease) and therefore may not reflect the true mortality rate for all ARDS patients. Finally, Rocco and colleagues conducted a 9-year study of 111 trauma and surgical patients with ARDS (68). From 1990 to 1994, the overall annual death rate from ARDS ranged from 67 to 80%. However, from 1995 to 1998 the annual mortality rates were all less than 50%. This improvement in mortality was most significant in those with a primary diagnosis of trauma as opposed to general surgical patients. The exact etiology for this decline in ARDS mortality is presently unclear but is likely related to improvement in general ICU care and possibly changes in ventilatory management (55) (see Chap. 21). It is important to note that the ARDS Network trial demonstrated an approximately 10% absolute mortality reduction (from 40%) and an increase in days free from multiorgan failure with low tidal volume in comparison to conventional tidal volumes (16). This important study suggests that a significant portion of the mortality from ARDS (perhaps 25%) derives from injurious ventilatory strategies per se. A recent evaluation of clinicians’ approaches to mechanical ventilation in the 1990s showed an evolution in tidal volume size from 14 mL/kg of body weight early in the decade to 10 mL/kg of body weight in the late 1990s for patients with ARDS (69). The cause of death for patients with ARDS has been traditionally divided into early causes (within 72 hours) and late causes (after 3 days) (70). Most early deaths are attributed to the original presenting illness or injury. Sepsis, persistent respiratory failure, and the development of multiple organ system dysfunction are the most common causes of death in ARDS patients who survive at least 3 days. Several secondary factors are associated with mortality in patients who develop ARDS. Patients with a primary at-risk diagnosis of sepsis have consistently been reported to have a higher mortality rate when compared to trauma patients. In one study of 423 ICU patients, those developing ARDS after trauma had significantly lower hospital mortality (14%) than did patients with a medical diagnosis (40%), the majority of whom had sepsis (71). Similarly, risk of death from ARDS was only 11 % in those trauma patients enrolled into the ARDSnet study (58). Long-term survival after ARDS also appears to be dependent on the primary at-risk diagnosis. Trauma patients with ARDS who survive to hospital discharge have an excellent prognosis over the following 2 years. However, survivors of sepsis-induced ARDS continue to be at an increased risk of dying after hospital discharge (72). This difference in long-term survival may be due to the presence of more significant and chronic comorbidities in those patients with sepsis-induced ARDS. In addition, patients with a direct cause of lung injury have been reported to have a higher mortality rate when compared to indirect at-risk diagnoses (73). Age is also associated with mortality from ARDS as older patients are more likely to die from ARDS when compared to younger patients. Zilberberg and Epstein demonstrated that age greater than 65 years was an independent predictor of hospital mortality in a cohort of 107 medical patients with ARDS (74). A study of 221 ARDS patients in Scandinavia demonstrated similar effects of age on ARDS and reported a risk ratio for mortality of nearly 2.0 when patients were stratified by age >65 (2). An analysis of the 902 patients enrolled in ARDS Network

Definitions and clinical risk factors

23

studies showed that patients over 70 years old were twice as likely to die even after adjustments for covariates. Older survivors recovered from respiratory failure at similar rates but had greater difficulty weaning from the ventilator (75). Preexisting medical conditions appear to have a dramatic impact on the mortality from ARDS. Although patients with cirrhosis of the liver are predisposed to several of the atrisk diagnoses for ARDS, such as sepsis (76), only a few reports have examined a possible association between cirrhosis and the mortality from ARDS. Matuschak and colleagues retrospectively examined 29 patients with severe liver disease awaiting transplantation and reported that their incidence of ARDS was higher than a random control group of ICU patients (77). Recently, Doyle et al. reported that the mortality from acute lung injury was increased in 26 medical patients with chronic liver disease when compared to acute lung injury patients without liver disease (78). Finally, cirrhosis was identified as the single most important predictive variable of mortality in a cohort of 259 patients with ARDS (73). Similarly, patients with a history of HIV disease, active malignancy, and organ transplantation appear to be at increased of dying form ARDS (74). Larger multi-center epidemiological studies are necessary to determine whether these multiple conditions are truly independent predictors of mortality in ARDS. Since most ARDS patients do not die of persistent respiratory failure, it might not be surprising that initial indexes of oxygenation and ventilation, including the ratio and the lung injury score, do not predict outcome (79). The ARDSnet trial reported that the patients with improved survival due to low tidal volume ventilation actually had a worse

ratio when compared to high tidal volume group. In addition, several

studies have reported that the mortality rate among patients with ALI (

ratio ≤

ratio ≤200) (13, 74, 78). These 300) is similar to those patients with ARDS ( similarities in mortality rates again raise concerns about the usefulness of differentiating between ALI and ARDS patients. In addition, these studies demonstrate that mortality is likely more closely related to the development of multiple organ dysfunction, sepsis, and the preexisting health of the patient. More recently, mean dead-space fraction, measured with a bedside metabolic monitor, was markedly elevated early in the course of ARDS and elevated values were associated with an increased risk of death (79a). Finally, there may be racial and gender differences in ARDS mortality in the United States (80). Using multiple-cause mortality data compiled by the National Center for Health Statistics, more than 333,000 patients who died with ARDS were identified over an 18-year period (1979–1996). Though controversial, ARDS was identified using International Classification of Diseases (ICD) codes for the underlying cause of death and the 20 additional conditions listed on the death certificate. Using these classifications, annual ARDS mortality rates have been continuously higher for males when compared to females and for African Americans when compared to white decedents and decedents of other racial backgrounds. When decedents were stratified by race and gender, African American males had the highest ARDS mortality rates in comparison to all other subgroups (mean annual mortality rate of 12.8 deaths per 100,000 African American males). In addition, a higher percentage of the African American ARDS decedents were reported in the youngest age categories. In those decedents who were less than 35 years of age, 27% were African American, yet only 13% of the U.S. population is African

Acute respiratory distress syndrome

24

American. The exact etiology of these differences is presently unknown. One possibility may be that men and African Americans are simply more likely to develop an at-risk diagnosis associated with the development of ARDS, such as sepsis or trauma. More recently, similar racial and gender differences have been reported in the incidence and mortality of patients with sepsis (80a).

VI. Conclusions Since 1967 the diagnostic criteria used to define ARDS have evolved. However, it remains difficult to answer the question, “What is ARDS?” (81). The ARDS definition is still based on classic physiological and radiographic alterations. These criteria are subject to variable interpretations that likely account for some of the discrepancy in the epidemiology of ARDS from different centers (82). Other questions still remain, such as, “Is ARDS a homogeneous syndrome or the combination of several different disorders loosely bound together by common physiological and radiographic abnormalities?” Eisner and colleagues have reported that low tidal volume ventilation is equally efficacious for patients with a variety of clinical risk factors (58). These findings support the original theory of Petty that ARDS is a specific disorder that responds in a uniform manner to alterations in mechanical ventilatory strategy. However, with the approval of activated protein C for the treatment of severe sepsis, the stratification of patients with ARDS by clinical risk factor (sepsis vs. nonsepsis) may also have therapeutic implications (83). The pulmonary and critical care community should attempt to focus on widespread use and understanding of current diagnostic criteria to ensure reliability and compatibility of epidemiological data from different centers. However, it is unlikely that ARDS will be defined by variations of the present diagnostic criteria in the future. Schuster (84) suggested the following general definition of ARDS: “It is a specific form of lung injury in which structural changes (characterized pathologically as diffuse alveolar damage) and functional abnormalities (principally a breakdown in the alveolar-capillary barrier function) leading first to proteinaceous alveolar edema, and then (as a consequence) to altered respiratory system mechanics and hypoxemia.” We would add that ARDS is an inflammatory disease and that inflammation begets lung injury and subsequent structural change. Presently, there is no clinically useful biochemical marker of inflammation that can be used as part of the diagnostic criteria of ARDS (51). As the pathogenesis of ARDS is more completely understood, certain biochemical tests may become available that remove the inherent subjectivity in the AECC definition of this syndrome. The identification of such an inflammatory mediator that functions as an extremely sensitive and specific marker in diagnosing ARDS would constitute a major advance in this field. Future definitions, based on biochemical or even genetic predisposition to inflammation rather than on physiological and radiographic parameters, are likely to provide more homogeneous groups of patients within the overall population of what is now called ALI (85). This should allow investigators to target subgroups of patients based on their specific pattern of an inflammatory response, opening the door to individualized therapy for ARDS (85).

Definitions and clinical risk factors

25

References 1. Roupie E, Lepage E, Wysocki M, Fagon JY, Chastre J, Dreyfuss D, Mentec H, Carlet J, BrunBuisson C, Lemaire F, Brochard L. Prevalence, etiologies, and outcome of the acute respiratory distress among hypoxemic ventilated patients. Int Care Med 1999; 25:920–929. 2. Luhr OR, Antonsen K, Karlsson M, Aardal S, Thorsteinsson A, Frostell CG, Bonde J. Incidence and mortality after acute respiratory failure and acute respiratory distress syndrome in Sweden, Denmark, and Iceland. Am J Respir Crit Care Med 1999; 159:1849–1861. 3. Rubenfeld GD, Caldwell ES, Steinberg KP, Hudson LD. Persistent lung injury and ICU resource utilization. Am J Respir Crit Care Med 1998; 157:A498. 4. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967; 2:319–323. 5. Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1988; 138:720–723. 6. Fowler AA, Hamman RF, Good JT, Benson KN, Baird M, Eberle DJ, Petty TL, Hyers TM. Adult respiratory distress syndrome: risk with common predispositions. Ann Int Med 1983; 98:593–597. 7. Pepe PE, Potkin RT, Reus DH, Hudson LD, Carrico CJ. Clinical predictors of the adult respiratory distress syndrome. Am J Surg 1982; 144:124–130. 8. Kraus PA, Lipman J, Lee CCJ, Wilson WE, Scribante J, Barr J, Mathivha LR, Brown JM. Acute lung injury at Baragwanath ICU: an eight-month audit and call for consensus for other organ failure in the adult respiratory distress syndrome. Chest 1993; 103:1832–1836. 9. Parsons PE, Giclas PC. The terminal complement complex (sc5b-9) is not specifically associated with the development of the adult respiratory distress syndrome. Am Rev Dis 1990; 141:98– 103. 10. Moss M, Parsons PE. What is acute respiratory distress syndrome? J Intensive Care Med 1998; 13:59–67. 11. Rinaldo JE. The prognosis of the adult respiratory distress syndrome: inappropriate pessimism? Chest 1986; 90:470–471. 12. Sloane PJ, Gee MH, Gottlieb JE, Albertine KH, Peters SP, Rurns JR, Machiedo G, Fish JE. A multicenter registry of patients with acute respiratory distress syndrome: physiology and outcome. Am Rev Respir Dis 1992; 146: 419–426. 13. Hudson LD, Mildberg JA, Anardi D, Maunder RJ. Clinical risk for development of acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:293–301. 14. Connors AF, Speroff T, Dawson NV, Thomas C, Harrell FE, Wagner D, Desbiens N, Goldman L, Wu AW, Califf RM, Fulkerson WJ, Vidaillet H, Broste S, Bellamy P, Lynn J, Knaus WA. The effectiveness of right heart catheterization in the initial care of critically ill patients. JAMA 1996; 276:889–897. 15. Amato MBP, Barbas CSV, Medeiros DM, Magaldi RB, Schettino GP, Lorenzi-Filho G, Kairalla RA, Deheinzelin D, Munoz C, Oliveira R, Takagaki TY, Carvalho CR. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338:347–354. 16. Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 17. Moss M, Bucher B, Moore FA, Moore EE, Parsons PE. The role of chronic alcohol abuse in the development of acute respiratory distress syndrome in adults. JAMA 1996; 275:50–54. 18. Moss M, Goodman PL, Heinig M, Barkin S, Ackerson L, Parsons PE. Establishing the relative accuracy of three new definitions of the adult respiratory distress syndrome. Crit Care Med 1995; 23:1629–1637.

Acute respiratory distress syndrome

26

19. Lewandowski K, Metz J, Deutschmann C, Preib H, Kuhlen R, Artigas A, Falke KJ. Incidence, severity, and mortality of acute respiratory distress in Berlin, Germany. Am J Respir Crit Care Med 1995; 151:1121–1125. 20. Clark JG, Milberg JA, Steinberg KP, Hudson LD. Type III procollagen peptide in the adult respiratory distress syndrome: association of increased petide levels in bronchoalveolar lavage fluid with increased risk for death. Ann Intern Med 1995; 122:17–23. 21. Petty TL. ARDS: refinement of concept and redefinition. Am Rev Respir Dis 1988; 138:724. 22. Heffner JE, Brown LK, Barbieri CA, Harpel KS, DeLeo J. Prospective validation of an acute respiratory distress syndrome predictive score. Am J Respir Crit Care Med 1995; 152:1518– 1526. 23. Bone RC. Toward a theory regarding the pathogenesis of the systemic inflammatory response syndrome: what we do and do not know about cytokine regulation. Crit Care Med 1996; 24:163–172. 24. Suchyta MR, Clemmer TP, Elliot CG, Orme JF, Weaver LK. The adult respiratory distress syndrome: a report of survival and modifying factors. Chest 1992; 101:1074–1079. 25. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall R, Morris A, Spragg R. The American-European consensus conference on ARDS: definitions, mechanisms relevant outcomes, and clinical trial co-ordination. Am J Respir Crit Care Med 1994; 149:818–824. 26. Artigas A, Bernard GR, Carlet J, Dreyfuss D, Gattinoni L, Hudson L, Lamy M, Marini JJ, Matthay MA, Pinsky MR, Spragg R, Suter PM. The American-European consensus conference on ARDS Part 2: ventilatory, pharmacologic, supportive therapy, study design strategies, and issues related to recovery and remodeling. Am J Respir Crit Care Med 1998; 157:1332–1347. 27. Deitch EA. Multiple organ failure: pathophysiology and potential future therapy. Ann Surg 1992; 216:117–134. 28. Neff MJ, Rubenfeld GD. Clinical epidemiology of acute lung injury. Sem Respir Crit Care Med 2001; 22:237–246. 29. Rubenfeld GD, Caldwell E, Granton J, Hudson LD, Matthay MA. Interobserver variability in applying a radiographic definition for ARDS. Chest 1999; 116:1347–1353. 30. Beards SC, Jackson A, Hunt L, Wood A, Frerk CM, Brear G, Edwards JD, Nightingale P. Interobserver variation in the chest radiograph component of the lung injury score. Anaesthesia 1995; 50:928–932. 31. Meade MO, Cook RJ, Guyatt GH, Groll R, Kachura JR, Bedard M, Cook DJ, Slutsky AS, Stewart TE. Interobserver variation in interpreting chest radiographs for the diagnosis of acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 161:85–90. 32. Medoff BD, Harris RS, Kesselman H, Venegas J, Amato MBP, Hess D. Use of recruitment maneuvers and high positive end-expiratory pressure in a patient with acute respiratory distress syndrome. Crit Care Med 2000; 28: 1210–1216. 33. Neff MJ, Rubenfeld GD, Caldwell ES, Hudson LD, Steinberg KP. Exclusion of patients with elevated pulmonary capillary wedge pressure from acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 159:A716. 34. Ferguson ND, Meade MO, Tomlinson G, Stewart TE. Values of the pulmonary artery occlusion pressure (PAOP) in ARDS and acute lung injury (ALI). Am J Respir Crit Care Med 1999; 159:A716. 35. Connors AF, McCaffree DR, Gray BA. Evaluation of right-heart catheterization in the critically ill patient without acute myocardial infarction. N Engl J Med 1983; 308:263–267. 36. Eisenberg PR, Jaffe AS, Schuster DP. Clinical evaluation compared to pulmonary artery catheterization in the hemodynamic assessment of critically ill patients. Crit Care Med 1984; 12:549–553. 37. Cook DJ, Simel DL. The rational clinical examination: does this patient have abnormal central venous pressure? JAMA 1996; 275:630–634.

Definitions and clinical risk factors

27

38. Neff MJ, Caldwell ES, Hudson LD, Rubenfeld GD. The effect of definition of left atrial hypertension (LAH) on identification of patients with acute lung injury (ALI). Am J Respir Crit Care Med 2001; 144:124–130. 39. Aberle DR, Wiener-Kronish JP, Webb WR, Matthay MA. Hydrostatic versus increased permeability pulmonary edema: diagnosis based on radiographic criteria in critically ill patients. Radiology 1988; 168:73–79. 40. Thamason JWW, Ely EW, Chiles C, Ferretti G, Freimans RI, Haponik EF. Appraising pulmonary edema using supine chest roentgenograms in ventilated patients. Am J Respir Crit Care Med 1998; 157:1600–1608. 41. Bernard G. Personal communication. 42. Abraham E, Matthay MA, Dinarello CA, Vincent JL, Cohen J, Opal SM, Glauser M, Parsons P, Fisher CJ, Repine JE. Consensus conference definitions for sepsis, septic shock, acute lung injury, and acute respiratory distress syndrome: time for a reevaluation. Crit Care Med 2000; 28:232–235. 43. Murray JF. The adult respiratory distress syndrome (may it rest in peace). Am Rev Respir Dis 1975; 111:716–718. 44. Petty TL. The adult respiratory distress syndrome (confessions of a lumper). Am Rev Respir Dis. 1975; 111:713–715. 45. Marks JD, Marks CB, Luce JM, Montgomery AB, Turner J, Metz CA, Murray JF. Plasma tumor necrosis factor in patients with septic shock: mortality rate, incidence of adult respiratory distress syndrome and effects of methylprednisolone administration. Am Rev Respir Dis 1990; 141:94–97. 46. Casey LC, Balk RA, Bone RC. Plasma cytokine and endotoxin levels correlate with survival in patients with the sepsis syndrome. Ann Intern Med 1993; 119: 771–778. 47. Hoch RC, Rodriguez R, Manning T, Bishop M, Mead P, Shoemaker WC, Abraham E. Effects of accidental trauma on cytokine and endotoxin production. Crit Care Med 1993; 21:839–845. 48. Rodrick ML, Wood JJ, O’Mahoney JB, Davis CF, Grbic JT, Demling RH, Moss NH, Saporoschetz I, Jordan A, D’Eon P. Mechanisms of immunosuppression associated with severe nonthermal traumatic injuries in man: production of interleukin 1 and 2. J Clin Immunol 1986; 6:310–318. 49. Moss M, Gillespie MK, Ackerson L, Moore FA, Moore EE, Parsons PE. Endothelial cell activity varies in patients at risk for the adult respiratory distress syndrome. Crit Care Med 1996; 24:1782–1786. 50. Parsons PE, Moss M. Circulating markers of sepsis and acute lung injury. In: Fein AM, Abraham EM, Balk RA, Bernard GR, Bone RC, Dantzker DR, Fink MP, eds. Sepsis and Multiorgan Failure. Baltimore: Williams & Wilkins, 1997:277–285. 51. Pittet JF, Mackersie RC, Martin TR, Matthay MA. Biological markers of acute lung injury: prognostic and pathogenetic significance. Am J Respir Crit Care Med 1997; 155:1187–1205. 52. Staub NC. Pulmonary edema. Physiol Rev 1974; 54:678–811. 53. Calandrino FS, Anderson DJ, Mintun MA, Schuster DP. Pulmonary vascular permeability during the adult respiratory distress syndrome: a positive emission tomographic study. Am Rev Respir Dis 1988; 138:421–428. 54. Kaplan JD, Calandrino FS, Schuster DP. A positive emission tomographic comparison of pulmonary vascular permeability during the adult respiratory distress syndrome and pneumonia. Am Rev Respir Dis 1991; 143: 150–154. 55. Steinberg KP, Hudson LD. Acute lung injury and acute respiratory distress syndrome: the clinical syndrome. Clin Chest Med 2000; 21:401–417. 56. Gattinoni L, Pelosi P, Suter PM, Pedoto A, Vercesi P, Lissoni A. Acute respiratory distress syndrome caused by pulmonary and extrapulmonary disease: different syndromes? Am J Respir Crit Care Med 1998; 158:3–11. 57. Pelosi P, Gattinoni L. Acute respiratory distress syndrome of pulmonary and extra-pulmonary origin: fancy or reality? Intensive Care Med 2001; 27:457–460.

Acute respiratory distress syndrome

28

58. Eisner MD, Thompson T, Hudson LD, Luce JM, Hayden D, Schoenfeld D, Matthay MA. Efficacy of low tidal volume ventilation in patients with different clinical risk factors for acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 164:231–236. 59. Kutlu CA, Williams EA, Evans TW, Pastorino U, Goldstraw P. Acute lung injury and acute respiratory distress syndrome after pulmonary resection. Ann Thorac Surg 2000; 69:376–380. 60. Lieber CS. Medical disorders of alcoholism. N Engl J Med 1995; 333:1058–1065. 61. Jurkovich GJ, Rivara FP, Gurney JG, Fligner C, Ries R, Mueller BA, Copass M. The effect of acute alcohol intoxication and chronic alcohol abuse on outcome from trauma. JAMA 1993; 270:51–56. 62. Iribarren C, Jacobs DR, Sidney S, Gross MD, Eisner MD. Cigarette smoking, alcohol consumption, and risk of ARDS: a 15-year cohort study in a managed care setting. Chest 2000; 117:163–168. 63. Kaul TK, Fields BL, Riggins LS, Wyatt DA, Jones CR, Nagle D. Adult respiratory distress sydrome following cardiopulmonary bypass: incidence, prophylaxis, and management. J Cardiovasc Surg 1998; 39:777–781. 64. Dancey DR, Hayes J, Gomez M, Schouten D, Fish J, Peters W, Slutsky AS, Stewart TE. ARDS in patients with thermal injury. Intensive Care Med 1999; 25:1231–1236. 65. Moss M, Guidot DM, Steinberg KP, Duhon GF, Treece P, Wolken R, Hudson LD, Parsons PE. Diabetic patients with septic shock have a decreased incidence of the acute respiratory distress syndrome (ARDS) Crit Care Med 2000; 28:2187–2192. 66. Milberg JA, Davis DR, Steinberg KP, Hudson LD. Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983–1993. JAMA 1995; 273:306–309. 67. Abel SJC, Finney SJ, Brett SJ, Keogh BF, Morgan CJ, Evans TW. Reduced mortality in association with the acute respiratory distress syndrome (ARDS). Thorax 1998; 53:292–292. 68. Rocco TR, Reinert SE, Cioffi W, Harrington D, Buczko D, Simms HH. A 9-year, single institution, retrospective review of death rate and prognostic factors in adult respiratory distress syndrome. Ann Surg 2001; 233:414–422. 69. Thompson BT, Hayden D, Matthay MA, Brower R, Parsons PE. Clinicians’ approaches to mechanical ventilation in acute lung injury and ARDS. Chest 2001; 120:1622–1627. 70. Montgomery AB, Stager MA, Carrico CJ, Hudson LD. Causes of mortality in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1985; 132:485–489. 71. Knaus WA, Sun X, Hakim RB, Wagner DP. Evaluation of definitions for adult respiratory distress syndrome. Am J Respir Crit Care Med 1994; 150:311–317. 72. Davidson TA, Caldwell ES, Hudson LD, Steinberg KP. Long-term mortality following acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 157:A498. 73. Monchi M, Bellenfant F, Cariou A, Joly LM, Thebert D, Laurent I, Dhainaut JF, Brunet F. Early predictive factors of survival in the acute respiratory distress syndrome: a multivariate analysis. Am J Respir Crit Care Med 1998; 158:1076–1081. 74. Zilberberg MD, Epstein SK. Acute lung injury in the medical ICU: comorbid condition, age, etiology, and hospital outcome. Am J Respir Crit Care Med 1998; 157:1159–1164. 75. Ely EW, Wheeler AP, Thompson BT, Ancukiewicz M, Steinberg KP, Bernard GR. Recovery rate and prognosis in older persons who develop acute lung injury and the acute respiratory distress syndrome. Ann Int Med 2002; 136:25–36. 76. Wyke RJ. Problems of bacterial infection in patients with liver disease. Gut 1987; 28:623–641. 77. Matuschak GM, Rinaldo JE, Pinsky MR, Gavaler JS, Van Thiel DH. Effect of end-stage liver disease on the incidence and resolution of the adult respiratory distress syndrome. J Crit Care 1987; 2:162–173. 78. Doyle RL, Szaflarski N, Modin GW, Wiener-Kronish JP, Matthay MA. Identification of patients with acute lung injury: predictors of mortality. Am J Respir Crit Care Med 1995; 152:1818–1824.

Definitions and clinical risk factors

29

79. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342:1334–1349. 79a. Nuckton TJ, Alonso JA, Kallet RH, Daniel BM, Pittet JF, Eisner MD, Matthay MA. Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med 2002; 346:1281–1286. 80. Moss M, Mannino DM. Racial and gender differences in acute respiratory distress syndrome (ARDS) deaths in the United States: Analysis of multiple-cause mortality data (1979–1996). Crit Care Med 2002; 30:1679–1685. 80a. Martin GS, Mannino DM, Eaton S, Moss M. Epidemiology of sepsis in the United States: incidence, demographics, and outcome from 1979–2000. N Engl J Med. In press. 81. Schuster DP. What is acute lung injury? What is ARDS? Chest 1995; 107: 1721–1726. 82. Garber BG, Herbert PC, Yelle JD, Hodder RV, McGowen J. Adult respiratory distress syndrome: a systematic overview of incidence and risk factors. Crit Care Med 1996; 24:687– 695. 83. Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, Lopez-Rodriquez A, Steingrub JS, Garber GE, Helterbrand JD, Ely EW, Fisher CJ. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344:699–709. 84. Schuster D. ARDS and pornography: I know it when I see it. J Intensive Care Med 1998; 13:55–56. 85. Abraham E. Toward new definitions of acute respiratory distress syndrome. Crit Care Med 1999; 27:237–238.

3 Epidemiology of Acute Lung Injury A Public Health Perspective GORDON D.RUBENFELD and MARGARET J.NEFF Harborview Medical Center University of Washington Seattle, Washington, U.S.A.

I. Introduction Critical care clinicians are drawn to practice in the intensive care unit by the physiological nature of critical illness and the application of physiological principles to the care of critically ill patients. We frequently consider the physiological derangements of acute lung injury (ALI): the gas exchange abnormalities, the abnormal thoracic compliance, and the response to positive end expiratory pressure (PEEP). We have come to appreciate the immunological and tissue repair abnormalities seen in patients with acute lung injury. More recently, we have been able to link pathophysiology with cellular mechanisms in the concept of ventilator-induced lung injury and ventilator-induced organ failure. The clinical epidemiology of acute lung injury in terms of its diagnostic criteria, risk factors, and prognostic factors has also evolved (1). However, it is unusual for critical care clinicians and investigators to consider the public health impact of critical illness syndromes in general and, more specifically, acute lung injury. Public health professionals are less interested in the exact pathophysiological mechanism of disease and focus on the disease’s impact on the health of the public and mechanisms for reducing this impact. This is a particularly important perspective on disease, if an unusual one for critical care. Understanding the public health implications of acute lung injury places it in relation to other common diseases and helps to prioritize research and clinical funding. Understanding changes in the burden and outcome of illness tells us whether we are doing a better job at what ultimately matters: improving the health of the public. To address these questions about acute lung injury and critical illness syndromes, answers to some basic epidemiological questions are needed. What is the incidence of acute lung injury? What is the attributable mortality and morbidity of acute lung injury? Are effective treatments or preventive interventions being implemented in the community? Is the incidence, outcome, or use of effective therapies changing over time? These data are available for a variety of diseases. There are a number of populationbased studies on the incidence and outcome of cardiovascular, pulmonary, infectious, and neoplastic diseases. In the United States, the National Center for Health Statistics maintains data on the incidence and mortality of hundreds of diseases (2). Similarly, the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer

Epidemiology of acute lung injury

31

Institute maintains high-quality data on cancer incidence and survival from selected areas across the United States (3). Similar data are not readily available for sepsis, acute lung injury, or multiple organ failure. Studying the epidemiology of ALI is not easy and may explain the lack of data. It is a syndrome that is operationally defined by laboratory, radiological, and physiological criteria that themselves have not been well defined in terms of reliability and validity (4– 7) (see Chap. 2). Even the terminology can be confusing. We will adopt the North American-European Consensus Conference (AECC) nomenclature and use ALI as a comprehensive term for the syndrome and acute respiratory distress syndrome (ARDS) to refer to a specific subset with more severe hypoxemia. There is no diagnostic test for ALI similar to troponin in myocardial infarction or serology in infectious diseases. Discharge diagnostic codes which are used to study the epidemiology of many diseases are extremely inaccurate in ALI. Compared to chronic diseases like cancer or asthma, ALI has a short duration and high mortality rate, which makes the number of prevalent cases available for study at any given time small. Despite these challenges, a growing body of literature exists to allow us to estimate the public health implications of ALI.

II. Incidence The incidence of a disease is defined as the number of new cases divided by the population at risk of developing the disease multiplied by the period of time they were at risk. Prevalence is the number of existing cases at any given time divided by the population at risk for developing the disease. Because ALI is a disease of relatively short duration, incidence and prevalence will approximate each other. It is important to distinguish which population constitutes the denominator in the incidence calculation. For population-based estimates of incidence, this will be the entire community from which patients are admitted to the hospital where ALI might be diagnosed. In epidemiological studies of ALI a more convenient denominator is often used, e.g., the “incidence” of ALI in the intensive care unit (ICU) population, in the population with acute respiratory failure, or in the population of patients with a known risk factor for ALI. These numbers are not useful for estimating the population burden of ALI. Incidence figures for ALI are important because they establish the importance of the disease to justify research and health care funding, allow tracking to explore trends in the disease, and provide data to study the potential explanations of differences in the incidence of the disease. For example, “Respiratory Distress Syndrome, Adult” is listed as a rare disease by the National Organization of Rare Disorders (8). The earliest incidence figure for ARDS is an often quoted 1972 National Heart, Lung, and Blood Institute (NHLBI) report estimate of 150,000 cases per year in the United States or about 75 per 105 person-years (9). Relatively few empirical estimates of the incidence of ALI or ARDS exist (10–13). Available empirical studies place the incidence of ARDS at much lower rates than expected by the NHLBI report: approximately 2–12 per 105 person-years (Table 1). Only one of these studies used the AECC definition for ALI. The others used more restrictive criteria, including more severe hypoxemia, a risk factor for ALI, and reduced thoracic compliance. These studies were also limited by a variety of factors: relatively short observation periods, potential for missed cases due to

Acute respiratory distress syndrome

32

lack of standardized definition, lack of a generalizable study population, and estimate of population incidence made from a small number of hospitals. Several lines of reasoning suggest these incidence estimates are, at least for the United States, as much as an order of magnitude too low. Analysis of the incidence of known associated risk conditions yields incidence rates for ALI that are higher than these published estimates. For example, in all epidemiological series and clinical trials, sepsis and pneumonia are the most common risk factors for ALI. Recent data suggest that the incidence of severe sepsis in patients in ICUs is 150 per 105 person-years (Table 2) (14). Epidemiological studies and clinical trials suggest that 30–43% of patients with severe sepsis develop ARDS (15, 16). Combining these data yields incidence estimates for sepsis associated ARDS of 45–64 cases per 105 person-years. A similar calculation for severe trauma (injury severity score

Table 1 Selected ALI and ARDS Incidence Studies Study location (sample time of study) Grand Canaria (1983–1985)

Utah (12 months, 1989–1990)

Definition

Incidence

1. Risk

1.5 per 105 person-years for

2. or with PEEP 5 and no improvement in 24 h and

3.5 per 105 person-years for

also 3. Bilateral infiltrates 4. No clinical left atrial hypertension

10.6 per 105 person-years for acute respiratory failure

Ref.

51

4.8–8.3 per 105 person-years 52 1. for ARDS 2. Bilateral infiltrates 3. No clinical evidence of left atrial hypertension 4. Static thoracic compliance <50 mL/cm H2O

Berlin (8 weeks in Severe lung injury: Murray-Matthay 1991) score>2.5

3.0 per 105 person-years for 19 severe lung injury 88.6 per 105 person-years for acute respiratory failure

Maryland (1995)

ICD-9 codes

10.5–14.2 per 105 personyears

Sweden, Denmark, Iceland (8 weeks in 1997)

AECC criteria

17.9 per 105 person-years for 18 ALI 13.5 per 105 person-years for ARDS 77.6 per 105 person-years for acute respiratory failure

19a

Epidemiology of acute lung injury

33

Table 2 ARDS Incidence Estimated from Associated Conditions Associated condition

Incidence of associated condition

Patients without risk who develop ARDS

Calculated incidence of ARDS

Severe sepsis

150 per 105 personyearsa

30–43%b

45–64 per 105 personyears

Severe trauma (ISS>15)

44 per 105 person-yearsc

25–40%d

11–18 per 105 personyears

Acute respiratory failure

137 per 105 personyearse

18%f

25 per 105 personyears

a

From Ref. 14. From Refs. 15, 16. c From Ref. 53. d From Refs. 15, 54. e From Ref. 55. f From Ref. 23. b

>15) yields incidence rates for ARDS of 11–18 per 105 person-years associated with trauma alone. These estimates are conservative for the incidence of ALI because they do not include an estimate of the number of patients who meet the less strict hypoxemia criterion for ALI and do not include patients who develop lung injury from causes other than sepsis or trauma, e.g., inhalation injuries, aspiration, and burns. Several recent sources corroborate these higher incidence rates. Moss and colleagues analyzed national death data and, relying on ICD-9 coding, arrived at a figure of 19,460 deaths associated with ARDS in 1993 in the United States (1). Assuming a mortality rate for ARDS of approximately 40%, this yields a case incidence rate of 26 per 105 personyears. Goss and colleagues combined screening log data from the ARDS Network multicenter clinical trial with data on U.S. hospitals to arrive at ALI incidence rates of 45–65 per 105 person-years even assuming that ALI cases only occurred in hospitals with more than 20 ICU beds (17). Finally, two recent studies have examined the incidence of ALI in Scandinavia and Berlin. Lewandoski and colleagues studied acute respiratory failure during a 2-month period in Berlin in 1991 (18–19). They defined acute respiratory failure as intubation and mechanical ventilation of >24 hours. The incidence of acute respiratory failure was 88 per 105 person-years. The authors used a scoring system to categorize the severity of patients’ lung injury, making direct comparison to the AECC criteria difficult. Patients with ALI by AECC criteria could have a score as low as 0.75 (1 point for and 2 points for two quadrants of radiographic opacity without receiving points for PEEP or compliance, which are not in the AECC definition). Using this cutoff and excluding 108 patients in the study with cardiogenic shock or cardiogenic edema leaves an incidence of 48 cases of ALI per 105 person-years. Luhr and colleagues (19) used similar methods to study the incidence of ALI and ARDS in 132 ICUs in Scandinavia over an 8-week period and found incidences of acute respiratory failure, ALI and ARDS of 77.6, 17.9, and 13.5 per 105 person-years, respectively.

Acute respiratory distress syndrome

34

Despite the difficulties in comparing the incidence of ALI across these studies, two striking and consistent findings emerge. The incidence of ALI appears to be significantly higher than the 2–12 per 105 person-years rates previously estimated for ARDS. These studies indicate that all forms of acute respiratory failure have a high mortality rate. The overall mortality rate for acute respiratory failure was 43% in the Berlin study and 41 % in the Scandinavian study. In both studies patients with acute respiratory failure, regardless of etiology, had similar mortality to patients with ALI. There is little support in the available data for a single incidence value for ALI. Variability in incidence rates between studies can reflect chance, true variability, or result simply from methodological differences. None of the existing studies on the population incidence of ALI use comparable methods or definitions: therefore, direct comparison of incidence rates is difficult. The studies used various observation periods, used different definitions for ALI and ARDS, and relied on varying degrees of quality control for case identification and data integrity. Given the evidence of interobserver variability in clinician radiographic interpretation and diagnosis of ALI (5, 6), rigorous protocolized case identification is necessary in epidemiological studies of ALI. True variability in incidence is a potential explanation of the existing studies. No single number reflects the incidence rate for myocardial infarction, colon cancer, or motor vehicle collisions, and we should not expect a single incidence figure for ALI. Potential explanations for this variability include differences in the incidence of risk factors, susceptibility (including genetic variation), and health care utilization. For example, differences in smoking, use of motor vehicles, population density, incidence of respiratory infections, and genetic factors might all influence geographic variability in the incidence of ALI. An interesting and unexplored source of variation is the effect of health care resource use on ALI incidence. Even within the United States there is wide variability in the ratio of hospital beds, ICU beds, emergency medical response time, and other medical resources. These may influence the observed incidence of ALI in two ways. To be diagnosed with ALI, patients must survive long enough to be admitted to an ICU, there must be an ICU bed to be admitted to, and they must have an arterial blood gas and a chest radiograph. Limited access to ICU care, implicit or explicit restriction of intensive care, or differences in emergency medical response time may reduce the number of observed cases of ALI. Similarly, the extent to which a region provides aggressive medical and surgical treatments may also affect the number of cases of ALI. For example, organ and bone marrow transplantation, coronary bypass grafting, and intensive chemotherapy all are associated with ALI, and countries that provide greater access to these treatments may have more cases of lung injury (20–22).

III. Attributable Mortality Attributable mortality is relatively easy to define mathematically: it is the difference in mortality rates between patients with a disease or exposure and those without. Practically, it is much more difficult to attribute a given death to a specific disease. A 68-year-old man who is an alcoholic is severely injured in a motor vehicle crash and develops ALI. After 14 days of progressive organ failure, life-sustaining treatment is withdrawn at the request of the patient’s family. Is the patient’s death attributable to alcoholism? To motor

Epidemiology of acute lung injury

35

vehicle trauma? To ARDS? To multiple organ failure? To the decision to withdraw medical therapy? To address these complexities, it is helpful to think of attributable mortality in two categories: as deaths associated with the disease and as deaths caused by the disease that could be prevented by some therapy or intervention. The former is much easier to calculate, although the latter is more important for public health purposes. A. Attributable Short-Term Mortality Mortality rates attributed to various diseases and reported in cause-of-death tables are calculated based on death records and generally reflect deaths associated with the disease. Attributable mortality associated with ALI can be calculated by multiplying the incidence rate times the mortality rate from the disease. The U.S. adult population (over age 15) in 2000 was 215 million. While the above discussion indicates that the incidence of ALI in the United States has not been described, it is reasonable, based on the studies cited above, to estimate it at between 20 and 50 cases per 105 person-years, or 43,000–107,000 cases per year. Assuming the mortality rate of approximately 40% observed in the recent studies of acute respiratory failure, 17,000–43,000 deaths per year are associated with ALI. Although the figures are arrived at by different methods, it is important to place these numbers into context with other diseases with important public health impact (Table 3).

Table 3 Attributable Mortality for Acute Lung Injury, Acute Respiratory Failure, and Comparison Diseases Disease ALI

Attributable mortality

a

17,000–43,000

Acute respiratory failure

b

60,000–120,000 c

Acute myocardial infarction

199,454

Breast cancerc

41,528

HIV disease Asthma

c

c

14,802 4,657

a

Assumes incidence range 20–50 per 105 person-years, mortality of 40%, and U.S. 2000 census population of 215 million>age 15. b Assumes incidence range 70–140 per 105 person-years and mortality of 40%, and U.S. 2000 census population of 215 million>age 15. c Based on U.S. 1999 death certificate data (56).

One of the surprising observations from recent epidemiological studies in ALI is the similar mortality, approximately 40%, that exists among the following different categories of respiratory failure: (1) patients with ARDS, (2) patients with ALI who meet other criteria for ARDS but with less severe hypoxemia and (3) patients with acute respiratory failure (intubation and mechanical ventilation >24

Acute respiratory distress syndrome

36

hours regardless of etiology, radiograph, or degree of hypoxemia) (23, 24). This observation provides little insight into disease mechanisms in this heterogenous population, but it has significant implications for public health. The incidence of acute respiratory failure is estimated at between 70 and 140 per 105 person-years. If 40% of these patients die, then up to 120,000 adult deaths per year are associated with mechanical ventilation. Even small reductions in the mortality or morbidity associated with mechanical ventilation would have significant implications for the public health. Identifying the independent or causal contribution of ALI to mortality is much more difficult. Two options exist for identifying this figure. By examining observational epidemiological data, one can try to control for other factors associated with mortality and estimate the independent effect of ALI on mortality. This is an important analysis because it is possible that ALI is merely a marker of severity of illness and contributes little on its own to mortality. These are difficult studies to do because they require identifying a cohort of critically ill patients, only a minority of whom will develop lung injury, and following them to compare mortality. The study by Hudson and colleagues attempted to control for this by the epidemiological technique of restriction (15). By comparing patients with ARDS to those at similar risk who did not develop ARDS, they showed that ARDS increased mortality rate in all risk conditions by an average of 3.3fold. This ranged from a relative risk for death attributed to ARDS of 1.4 in sepsis to 4.3 in trauma to 8.6 in drug overdose. The authors further controlled for APACHE in the septic patients and injury severity in the trauma patients without a significant effect on the attributable mortality. More compelling evidence of the attributable and preventable deaths in ALI would be reduction in mortality from an effective intervention to prevent ALI or to prevent death after ALI. No interventions have been shown to prevent ALI; however, recent data from two studies suggest that a ventilator strategy can reduce mortality in ALI by a risk difference of 8.8–33% (25, 26). These data can be analyzed in light of the findings in Table 3 to estimate that 3,800–35,000 deaths annually could be prevented in the United States by implementing lung-protective ventilation in ALI, depending on its incidence and the benefits of lung-protective ventilation beyond the clinical trial population (27, 28). B. Attributable Long-Term Mortality There is growing interest in the effects of critical illness syndromes on long-term outcomes. The methodological challenges are similar to those encountered in establishing attributable short-term mortality rates. Trying to separate the independent and causal effect of ALI on long-term mortality from the effects of the risk factors that cause or are associated with ALI is a challenge. Two studies have documented an effect of sepsis on long-term survival. A study by Quartin et al. showed that, even after controlling for age and comorbidity using ICD-9 diagnostic codes, patients with sepsis have a higher mortality rate than control patients (29). Among patients who survive for a year, those who had an episode of sepsis have an approximately 1.5 times greater rate of death than similar patients without an episode of sepsis. Patients who survive for 30 days after sepsis still have a median survival that is reduced from 6.24 to 2.35 years. Concerns about this study relate to the quality of the ICD-9 coding of comorbidites and the possibility that

Epidemiology of acute lung injury

37

patients who have been admitted to the hospital for a severe illness like sepsis have more comorbidities coded than controls. A number of studies have followed ALI and ARDS patients beyond hospital discharge to explore the long-term survival in patients with lung injury. For example, a study of relatively young (mean age 48) and previously healthy ARDS patients enrolled in a clinical trial of inhaled nitric oxide showed that survivors of ARDS continued to accrue mortality from day 28 after ARDS until about day 180, where the mortality rate stabilized (29a). However, these data cannot be used to assess attributable mortality of ARDS since it only includes patients with disease. Only one study has compared long-term survival in patients who survived to hospital discharge and compared it to controls matched on severity of sepsis or trauma (30). Patients with sepsis had reduced long-term survival compared to patients with trauma, regardless of the presence of ARDS; however, there was no independent effect of ARDS on long-term mortality when the analysis was restricted to patients who survived to hospital discharge. This study was limited by two factors. It was a relatively small study, so important effects of ARDS on long-term mortality may have been missed, and authors could not completely exclude the possibility that the controls had some mild component of ALI. Nevertheless, the best current evidence suggests that ALI does not independently worsen long-term survival in patients who survive to hospital discharge. Importantly, this study found that 80% of all deaths occurred in the hospital, 77% of all deaths occurred by day 30 after the onset of ARDS, and 89% of all deaths occurred by day 100 after the onset of ARDS.

IV. Attributable Morbidity If, as recent clinical evidence suggest, mortality after ALI is declining and, in some subgroups, may be as low as 20%, then the morbidity incurred by survivors of ALI becomes an increasingly significant clinical issue. We can estimate this burden by calculating the number of ALI 5-year survivors in the U.S. health care system. Assuming that there are 107,500 cases of ALI per year (215 million adults×50 cases/100,000 person-year) (Table 3), that 70% of patients with ALI survive their acute illness, and that all patients with trauma associated ALI and 50% of sepsis associated ALI survive for 5 years, then more than 280,000 ALI survivors are alive in the United States. This conservative estimate excludes all survivors whose ALI occurred more than 5 years ago. As we improve our acute care to critically ill patients, we must address the health care sequelae of the large group of ALI survivors we are creating. While information about the late outcomes of critical illness is growing, this is an evolving field with relatively few data particularly regarding mechanisms and treatments. The same methodological limitations apply to identifying the attributable effect of ALI on morbidity as was noted for its effect on mortality. A. Attributable Effect of ALI on Functional Status For the purposes of this discussion, functional status refers to objective and physiological measures of performances after an episode of acute lung injury. This includes pulmonary function, gas exchange, exercise tolerance, and cognitive performance. A number of

Acute respiratory distress syndrome

38

investigators have studied pulmonary function in survivors of ALI. Pulmonary function appears to be severely abnormal within 1 month of ALI onset. The abnormalities are primarily restrictive, although obstructive abnormalities have been reported (31). This is followed by a period of rapid improvement in pulmonary function of over 3–6 months. After approximately 6 months most of the improvement that will occur has occurred. The majority of patients are left with little measurable pulmonary dysfunction except for a reduced DLCO. A small minority have a persistent severe restrictive defect. These physiological data are corroborated by similar changes in radiographic studies (32, 33). Although it seems reasonable to assume that pulmonary function abnormalities are attributable to the parenchyma and vascular pathology of ALI, there are no studies comparing pulmonary function in ALI patients to a similar control group that tests this hypothesis. It is possible that diffusion abnormalities and restrictive disease in ALI survivors is due to the combination of a slowly resolving endothelial injury and critical illness polyneuropathy that are sequelae of systemic inflammation and have nothing to do with lung injury. There is a growing body of literature demonstrating acute cognitive impairment in critically ill patients (34, 35). However, the mechanism, persistence, and relationship to ALI, hypoxemia, or duration of mechanical ventilation is unclear. At one year, the majority of ALI survivors have impaired memory, attention, concentration, and/or decreased mental processing speed (36). The extent to which these cognitive abnormalities are attributable to ALI or to risk conditions is unknown, but they reflect significant morbidity in these patients. B. Attributable Effect of ALI on Psychiatric Outcomes and Quality of Life To an ALI survivor, quality of life is as important as any specific physical or functional parameter. Potential problems were initially appreciated only anecdotally as clinicians saw ALI survivors in follow-up and heard their patients describe depression or difficulty at work or with relationships. Subsequently these outcomes have been more formally studied by means of standardized questionnaires and tools (31, 37). Hopkins and colleagues confirmed results seen by other investigators who interviewed ALI survivors using the Medical Outcomes Study 36-item, short form health survey (SF-36) (36). ALI survivors showed continued poor scores when tested at one year in the categories of role emotional, mental health, bodily pain, and general health. Davidson et al. evaluated quality-of-life measures in ALI survivors as compared to matched critically ill controls who had not developed ALI and found worse results in the domains of physical functioning, general health, and vitality when measured on average 2 years after hospitalization (38). While the degree of impairment was not as profound as for patients with other severe lung diseases, many of these patients still found it difficult to function fully and to return to work. In most studies and clinical reports, patients described feelings of fatigue, memory loss, depression, and fear of relapse. In fact, Weinert and colleagues found that over 75% of survivors had scores on a depression scale that qualified for a diagnosis of depression during the first 15 months after ALI (39). In addition, another study revealed that over 50% of a cohort of critically ill patients transferred to a long-term acute care facility were

Epidemiology of acute lung injury

39

prescribed an antidepressant (40). While historically studied in people who have suffered from trauma or war experiences, posttraumatic stress disorder (PTSD) is a similarly important mental health assessment in critically ill patients. Many clinicians have questioned whether patients suffered from memories of their intensive care unit experience, but except for anecdotes few data have been available. However, Schelling and colleagues have studied this issue using tools such as the SF-36 and the Post Traumatic Stress Syndrome-10 (41). Of the 80 patients studied there was evidence of PTSD in one third of the patient population approximately 4 years post-hospitalization. These remain important outcomes to be incorporated into the future clinical trials of ALI and to be studied among current survivors of ALI. Whether other aspects of the patient’s hospitalization, e.g., hypoxemia or level of sedation, may be associated with the development of depression or PTSD is unknown but is important to explore as we try to optimize the physical and mental well-being of ALI survivors.

V. Effect of Aging Population Age is a complex “exposure” variable. Like gender or race, it is a surrogate marker for a variety of other social and biological exposures. Identifying an association between age and other variables sheds little light on the causal factors associated with age that may actually be driving the relationship. Because age is strongly associated with the decision to admit patients to the ICU and to withdraw life-sustaining treatments in the ICU, the relationships between age and other variables are confounded by these physician decisions. Similarly, associations between age and other variables may not reflect an effect of age, per se, but of other variables that are frequently associated with age. For example, while age is crudely associated with mortality in many studies of critical illness, the effect disappears or is mitigated when comorbidities are accounted for (42). Therefore, the effect of age alone is less than the effect of diabetes, heart failure, and malnutrition which occur more frequently in the elderly than in other populations. These mechanistic issues are of less concern to the public health epidemiologist. Regardless of the mechanism, if older people are at greater risk of developing ALI or at greater risk of mortality and morbidity from ALI, then age is an important factor in the clinical epidemiology of the disease. It is particularly important given the realities of the aging population in the United States. By the year 2050, the population over the age of 65 will increase from 16% of the population to 25% of the population, with approximately 82 million people in this age group (43). There are relatively few data to model the effect of an aging population on incidence and mortality from ALI. We know that the incidence of risk factors for ALI including sepsis and pneumonia increases with age (44). There is a gradual increase in the incidence of sepsis during the adult years (<500 cases per 105/person-year) until approximately age 65, when an abrupt increase in incidence to over 2500 cases of sepsis per 105/person-year in the oldest patients is observed. Older patients are also at greater risk of developing ALI after trauma, even after controlling for severity of injury (45). The effect of age on mortality in patients with ALI has recently been studied. Older patients with ALI are at significantly higher risk of death even after controlling for severity of illness and comorbidity (46). Given these observations, an

Acute respiratory distress syndrome

40

aging population will lead not only to an increased number of patients with ALI but to a group of patients at higher risk of death from ALI.

VI. Pediatrics ALI is not unique to the adult population. A similar clinical picture is seen among pediatric patients but is not well described. In talking about pediatric ALI, a distinction must first be made from neonatal respiratory distress syndrome (RDS) or hyaline membrane disease that is due to surfactant deficiency in neonates. This is in contrast to the mechanism in ALI where the surfactant may be present (or perhaps reduced) but is dysfunctional. ALI in children is felt to represent the same pathophysiological process as is seen in adults, but less is known about its risk factors, outcomes, and incidence. Most studies have represented single site descriptions of their patient populations and outcomes (47–50). True incidence studies, where the total number of children at risk for the disease is truly known, have not been done. Most descriptions of pediatric ALI have defined a minimal age (e.g., 1 month), excluded congenital heart disease, and identified clinical criteria required to be diagnosed with ALI. These criteria have not been uniformly agreed upon but include some combination of radiographic abnormalities and hypoxemia. Risk factors include those for ALI in adults but have not been rigorously studied. It is unknown if children with certain predisposing risks have better outcomes from ARDS. Also of interest to clinicians and investigators is identifying predictors of outcome. Airway pressures, degree of hypoxemia, and oxygen index have been used to try to predict who might benefit from extracorporeal membrane oxygenation (ECMO) and who might be at high risk of dying. Common among most published reports is a mortality rate of approximately 60%, considerably higher than the 30–40% currently quoted for adults with ALI. Unknown for children with ALI is whether their deaths are primarily due to respiratory failure, multiorgan failure, or central nervous system damage. While increasing interest is focused on functional and quality-of-life outcomes in adults, more work is needed to evaluate such outcomes as school performance, learning disabilities, or socialization problems in children. Much remains to be studied in this field, and a good understanding of the epidemiology of the disease including a standard definition will likely help. Particularly important will be risk factor identification in an attempt to impact the frequency of the disease in the same way that successful injury prevention programs (e.g., distribution of bike helmets and booster seats) have reduced unintentional injuries in children.

VII. Conclusions Acute lung injury continues to challenge clinicians and investigators both in understanding the disease itself and in providing care destined to improve survival and long-term outcomes. While focusing on an individual patient’s outcome is critically important, equally meaningful is the continued study of the epidemiology of this disease and, as a consequence, its impact on public health as a whole. Even if conservative estimates of the incidence of ALI in adults are used, this disease process is associated

Epidemiology of acute lung injury

41

with up to 43,000 deaths per year in the United States. While research and clinical trials have led to new approaches to managing ALI patients that have positively impacted mortality, the study of long-term physical and emotional functioning has just begun. As clinicians see more and more survivors of ALI, issues such as pulmonary function, depression, and PTSD will become more prominent. Finally, additional study is also needed to better understand the impact of ALI in special patient populations, such as in children and the elderly.

References 1. TenHoor T, Mannino DM, Moss M. Risk factors for ARDS in the United States: analysis of the 1993 National Mortality Followback Study. Chest 2001; 119:1179–1184. 2. National Vital Statistics Reports, http://www.cdc.gov/nchs/fastats. 2002. 3. National Cancer Institute: Surveillance, Epidemiology, and End Results Program, http://seer.cancer.gov/.2002. 4. American Educational Research Association, American Psychological Association, National Council on Measurement in Education, and Joint Committee on Standards for Educational and Psychological Testing (U.S.). 1999. Standards for educational and psychological testing American Educational Research Association, Washington, DC, ix, 194. 5. Meade MO, Cook RJ, Guyatt GH, Groll R, Kachura JR, Bedard M, Cook DJ, Slutsky AS, Stewart TE. Interobserver variation in interpreting chest radiographs for the diagnosis of acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 161:85–90. 6. Rubenfeld GD, Caldwell E, Granton J, Hudson LD, Matthay MA. Inter-observer variability in applying a radiographic definition for ARDS. Chest 1999; 116:1347–1353. 7. Moss M, Goodman PL, Heinig M, Barkin S, Ackerson L, Parsons PE. Establishing the relative accuracy of three new definitions of the adult respiratory distress syndrome. Crit Care Med 1995; 23:1629–1637. 8. National Organization of Rare Disorders, http://www.rarediseases.org/. 2002. 9. National Heart and Lung Institutes. Task force report on problems, research approaches, needs. Washington, DC: U.S. Government Printing Office. 167–180; 1972. 10. Webster NR, Cohen AT, Nunn JF. Adult respiratory distress syndrome—how many cases in the UK? Anaesthesia 1988; 43:923–926. 11. Zaccardelli DS, Pattishall EN. Clinical diagnostic criteria of the adult respiratory distress syndrome in the intensive care unit. Crit Care Med 1996; 24:247–251. 12. Garber BG, H’Ebert PC, Yelle JD, Hodder RV, McGowan J. Adult respiratory distress syndrome: a systemic overview of incidence and risk factors. Crit Care Med 1996; 24:687–695. 13. Luce JM. Acute lung injury and the acute respiratory distress syndrome. Crit Care Med 1998; 26:369–376. 14. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001; 29:1303–1310. 15. Hudson LD, Milberg JA, Anardi D, Maunder RJ. Clinical risks for development of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:293–301. 16. Bernard GR, Wheeler AP, Russell JA, Schein R, Summer WR, Steinberg KP, Fulkerson WJ, Wright PE, Christman BW, Dupont WD, Higgins SB, Swindell BB. The effects of ibuprofen on the physiology and survival of patients with sepsis. The Ibuprofen in Sepsis Study Group. N Engl J Med 1997; 336:912–918. 17. Goss CH, Rubenfeld GD, Brower RG, Hudson LD. Incidence of acute lung injury and acute respiratory distress syndrome (ALI/ARDS) in the United States. Crit Care Med 2003. In press.

Acute respiratory distress syndrome

42

18. Luhr OR, Antonsen K, Karlsson M, Aardal S, Thorsteinsson A, Frostell CG, Bonde J. Incidence and mortality after acute respiratory failure and acute respiratory distress syndrome in Sweden, Denmark, and Iceland. The ARF Study Group. Am J Respir Crit Care Med 1999; 159:1849–1861. 19. Lewandowski K, Metz J, Deutschmann C, Preiss H, Kuhlen R, Artigas A, Falke KJ. Incidence, severity, and mortality of acute respiratory failure in Berlin, Germany. Am J Respir Crit Care Med 1995; 151:1121–1125. 19a. Reynolds HN, McCunn M, Borg U, Habashi N, Cottingham C, Bar-Lavi Y. Acute respiratory distress syndrome: estimated incidence and mortality rate in a 5 million-person population base. Crit Care (Lond) 1998; 2(1):29–34. 20. Demertzis S, Haverich A. Adult respiratory distress syndrome and lung transplantation. J Heart Lung Transplant 1993; 12:878–879. 21. Kress JP, Christenson J, Pohlman AS, Linkin DR, Hall JB. Outcomes of critically ill cancer patients in a university hospital setting. Am J Respir Crit Care Med 1999; 160:1957–1961. 22. Winer-Muram HT, Gurney JW, Bozeman PM, Krance RA. Pulmonary complications after bone marrow transplantation. Radiol Clin North Am 1996; 34:97–117. 23. Roupie E, Lepage E, Wysocki M, Fagon JY, Chastre J, Dreyfuss D, Mentec H, Carlet J, BrunBruisson C, Lemaire F, Brochard L. Prevalence, etiologies and outcome of the acute respiratory distress syndrome among hypoxemic ventilated patients. SRLF Collaborative Group on Mechanical Ventilation. Societe de Reanimation Langue Francaise. Intensive Care Med 1999; 25:920–929. 24. Milberg JA, Davis DR, Steinberg KP, Hudson LD. Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983–1993. JAMA 1995; 273:306–309. 25. Amato MB, Barbas CS, Medeiros DM, Magaldi RB, Schettino GP, Lorenzi-Filho G, Kairalla RA, Deheinzelin D, Munoz C, Oliveira R, Takagaki TY, Carvalho CR. Effect of a protectiveventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338:347–354. 26. Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342(18):1301–1308. 27. Rothman, KJ. 1998. Modern Epidemiology, 2nd ed. Lippincott-Raven, Philadelphia. 28. Gordis L. 1996. Epidemiology W.B.Saunders Company, Philadelphia. 29. Quartin A A, Schein RM, Kett DH, Peduzzi PN. Magnitude and duration of the effect of sepsison survival. Department of Veterans Affairs Systemic Sepsis Cooperative Studies Group. JAMA 1997; 277:1058–1063. 29a. Angus DC, Musthafa AA, Clermont G, Griffin MF, Linde-Zwirble WT, Dremsizov TT, Pinsky MR. Quality-adjusted survival in the first year after the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163(6):1389–1394. 30. Davidson TA, Rubenfeld GD, Caldwell ES, Hudson LD, Steinberg KP. The effect of acute respiratory distress syndrome on long-term survival. Am J Respir Crit Care Med 1999; 160:1838–1842. 31. McHugh LG, Milberg JA, Whitcomb ME, Schoene RB, Maunder RJ, Hudson LD. Recovery of function in survivors of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1994; 150:90–94. 32. Lakshminarayan S, Stanford RE, Petty TI. Prognosis after recovery from adult respiratory distress syndrome. Am Rev Respir Dis 1976; 113:7–16. 33. Desai SR, Wells AU, Rubens MB, Evans TW, Hansell DM. Acute respiratory distress syndrome: CT abnormalities at long-term follow-up. Radiology 1999; 210:29–35. 34. Ely EW, Inouye SK, Bernard GR, Gordon S, Francis J, May L, Truman B, Speroff T, Gautam S, Margolin R, Hart RP, Dittus R. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA 2001; 286:2703–2710.

Epidemiology of acute lung injury

43

35. Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, Speroff T, Gautam S, Bernard GR, Inouye SK. Evaluation of delirium in critically ill patients: validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit Care Med 2001; 29:1370– 1379. 36. Hopkins RO, Weaver LK, Pope D, Orme JF, Bigler ED, Larson-LOHR V. Neuropsychological sequelae and impaired health status in survivors of severe acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160:50–56. 37. McSweeny AJ, Grant I, Heaton RK, Adams KM, Timms RM. Life quality of patients with chronic obstructive pulmonary disease. Arch Intern Med 1982; 142:473–478. 38. Davidson TA, Caldwell ES, Curtis JR, Hudson LD, Steinberg KP. Reduced quality of life in survivors of acute repiratory distress syndrome compared with critically ill control patients. JAMA 1999; 281:354–360. 39. Weinert CR, Gross CR, Kangas JR, Bury CL, Marinelli WA. Health-related quality of life after acute lung injury. Am J Respir Crit Care Med 1997; 156: 1120–1128. 40. Weinert CR. Epidemiology of psychiatric medication use in patients recovering from critical illness at long-term acute-care facility. Chest 2001; 119:547–553. 41. Schelling G, Stoll C, Haller M, Briegel J, Manert W, Hummel T, Lenhart A, Heyduck M, Polasek J, Meier M, Preuss U, Bullinger M, Schuffel W, Peter K. Health-related quality of life and posttraumatic stress disorder in survivors of the acute respiratory distress syndrome. Crit Care Med 1998; 26:651–659. 42. Cohen IL, Lambrinos J. Investigating the impact of age on outcome of mechanical ventilation using a population of 41,848 patients from a statewide database. Chest 1995; 107:1673–1680. 43. Centers for Disease Control: National Center for Chronic Disease Prevention and Health Promotion, http://www.cdc.gov/nccdphp/aging. 2002. 44. Ely EW, Haponik EF. Pneumonia in the elderly. J Thorac Imaging 1991; 6: 45–61. 45. Sprenkle MD, Caldwell ES, Rudenfeld GD, Hudson LD, Steinberg KP. Mortality following acute respiratory distress syndrome (ARDS) among the elderly. Amm J Respir Crit Care Med 1999; 159:A717. 46. Suchyta MR, Clemmer TP, Elliott CG, Orme JF Jr., Morris AH, Jacobson J, Menlove R. Increased mortality of older patients with acute respiratory distress syndrome. Chest 1997; 111:1334–1339. 47. Lyrene RK, Truog WE. Adult respiratory distress syndrome in a pediatric intensive care unit: predisposing conditions, clinical course, and outcome. Pediatrics 1981; 67:790–795. 48. Timmons OD, Dean JM, Vernon DD. Mortality rates and prognostic variables in children with adult respiratory distress syndrome. J Pediatr 1991; 119:896–899. 49. Davis SL, Furman DP, Costarino AT Jr. Adult respiratory distress syndrome in children: associated disease, clinical course, and predictors of death. J Pediatr 1993; 123:35–45. 50. Paret G, Ziv T, Barzilai A, Ben Abraham R, Vardi A, Manisterski Y, Barzilay Z. Ventilation index and outcome in children with acute respiratory distress syndrome. Pediatr Pulmonol 1998; 26:125–128. 51. Villar J, Slutsky AS. The incidence of the adult respiratory distress syndrome. Am Rev Respir Dis 1989; 140:814–816. 52. Thomsen GE, Morris AH. Incidence of the adult respiratory distress syndrome in the state of Utah. Am J Respir Crit Care Med 1995; 152:965–971. 53. National Center for Injury Prevention and Control: Scientific Data, Surveillance, and Injury Statistics, http://www.cdc.gov/ncipc/osp/data. 2002. 54. Rainer TH, Lam PK, Wong EM, Cocks RA. Derivation of a prediction rule for post-traumatic acute lung injury. Resuscitation 1999; 42:187–196. 55. Behrendt CE. Acute respiratory Failure in the United States: incidence and 31-day survival. Chest 2000;118:1100–1105. 56. Hoyert DL, Arias E, Smith BL, Murphy SL, Kochanek KD. Deaths: Final data for 1999. Natl Vital Stat Rep 2001; 49:1–113.

4 Radiographic Findings of the Acute Respiratory Distress Syndrome DESIRÉE M.QUIÑONES MAY MÍ and PHILIP C.GOODMAN Duke University Medical Center Durham, North Carolina, U.S.A.

I. Introduction In 1967, Ashbaugh and coworkers introduced the phrase acute respiratory distress syndrome of adults, now adult respiratory distress syndrome (ARDS), to describe an illness that resulted in severe respiratory failure in a group of 12 patients with different underlying etiologies (including pneumonia and trauma) (1). Ashbaugh and other investigators described the chest films of several of these patients; the radiographs demonstrated bilateral heterogeneous opacities that rapidly coalesced to form more homogeneous consolidation throughout both lungs. During the last three decades, many articles regarding ARDS have been published. The radiological features of the disease remain an integral part of its definition. This chapter describes the imaging characteristics of ARDS as seen on chest radiographs and other imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET).

II. Definition and Causes of ARDS Several definitions have been proposed for ARDS over the past decades. In 1994, the North American/European Consensus Conference Committee (NAECC) proposed the currently used criteria for the diagnosis of ARDS: (1) acute onset of lung injury, (2) diffuse bilateral infiltrates on chest radiography, (3) refractory hypoxemia, (4) no clinical evidence of congestive heart failure, and (5) decreased lung compliance (see Chap. 2). Thus, imaging abnormalities are both a by-product of this disease and part of its definition. Although the definition recommended by the NAECC includes simple criteria that are easy to apply in the clinical setting, some investigators believe that in a clinical trial setting its lack of specific criteria for radiological findings is a limitation. In a recent study by Rubenfeld and colleagues, the reliability of this consensus radiographic definition for ARDS was evaluated (2). Twenty-one experts, all pulmonologists or critical care physicians, including seven members of the National Institutes of Health ARDS Network, evaluated chest radiographs of critically ill patients with a

Radiographic findings of the acute respiratory distress syndrome

45

ratio of <300 and were asked to decide whether the radiographs fulfilled the NAECC definition for acute lung injury (ALI)-ARDS. There was a high interobserver variability, which could affect the validity of investigators using results of their clinical trials on patients. Radiologists, the authors note, may have had less variability on interpreting the films but were specifically not involved because of the design and intent of the study. Nevertheless, the NAECC 1994 definition has been helpful in understanding more about the disease and as a basis for defining members of a group who are being investigated regarding various treatment options. The NAECC 1998 recommendations revealed no formal changes but did emphasize the importance of recognizing etiological and epidemiological differences between patients when studying ALI-ARDS (3). Major risk factors for ARDS are sepsis, aspiration of gastric contents, trauma (including long bone and pelvic fracture and pulmonary contusion), multiple blood tranfusions, overwhelming pneumonia, and shock (4, 5). Interventions directed at one of these etiologies may not be as effective if other etiologies are causative. Less commonly, near-drowning, drug toxicity, major burns, and toxic inhalation may lead to ARDS. Often more than one etiology is responsible; thus, auxilliary imaging findings in addition to those typically associated with ARDS may be present.

III. Pathophysiology ARDS to some degree is a form of permeability defect with protein-rich pulmonary edema associated with diffuse alveolar damage. This explains the usual diffuse distribution of increased airspace opacities seen on chest films. The disease is characteristically described as having three overlapping phases: exudative, proliferative, and fibrotic (see Chap. 5). The exudative phase, usually seen within hours of the instigating pulmonary insult, is associated with interstitial edema characterized by a high protein content and hemorrhage that rapidly fills the alveoli. Subsequently, hyaline membrane formation occurs. During the proliferative phase, organization of the fibrinous exudate and regeneration of the alveoli occurs. These changes are seen 7–28 days after the onset of disease. Fibrosis, scarring, and formation of subpleural and intrapulmonary cysts characterize the third, fibrotic phase, which may begin within 2 or 3 weeks of the initial insult, and are recognized for weeks to months later.

IV. Chest Radiography Frequently the first chest film of a patient with ARDS demonstrates diffuse symmetrical bilateral heterogeneous or homogeneous opacities, the picture of pulmonary edema (Fig. 1). In a short time, within hours or days, the opacities become significantly more homogeneous while maintaining their symmetric bilateral distribution. This progression of interstitial edema to diffuse airspace disease corresponds to the rapid filling of alveolar spaces by edema and hemorrhage. Characteristically, the opacities on the initial chest radiograph have been described as being located predominantly in the lung periphery, but this should not dissuade one from suggesting the diagnosis when a more central or even

Acute respiratory distress syndrome

46

involvement is observed (Fig 2). On the other hand, in contrast to cardiogenic pulmonary edema, the heart size and vascular pedicle are generally normal in patients with ARDS. Kerley lines (e.g., interstitial edema or lymphatic engorgement) may or may not be seen; they are reported with various frequency. Pleural effusions, common with cardiogenic pulmonary edema, have been thought unusual with ARDS and in the past were assumed to occur late in the disease or suggest the development of acute pneumonia or pulmonary infarct. However, the use of CT in patients with ARDS has revealed an earlier, more common association of pleural fluid in this setting. Of course, this brings into question how often pleural fluid is really a marker of superimposed infection. As edema worsens bilateral ill-defined opacities become more and more homogeneous, involving virtually all of the lungs with severe opacification with or without air bronchograms. Some studies have shown, however, that distinguishing between cardiogenic and permeability pulmonary edema, as occurs in ARDS, is difficult (6). In study by Aberle and coworkers, some features more typical

Figure 1 ARDS: A 27-year-old male developed respiratory failure after being hospitalized for a motor vehicle accident. Portable chest radiograph demonstrates low lung volumes with diffuse bilateral heterogeneous opacities. This is a typical picture of ARDS.

Radiographic findings of the acute respiratory distress syndrome

47

of hydrostatic pulmonary edema such as widened vascular pedicle, pleural effusions, peribronchial cuffs, and septal lines were commonly found in patients with permeability pulmonary edema (7). Nevertheless, chest radiography remains a valuable tool in the evaluation of pulmonary edema. In patients who survive the initial period, radiographs eventually improve slightly but continue to reveal diffuse bilateral homogeneous or heterogeneous opacities. As the disease enters the proliferative phase, characterized by an increasing fibroblast population and deposition of collagen, the chest film stabilizes as the radiographic pattern becomes more coarsely heterogeneous, linear, and reticular. In some patients, complete resolution of abnormalities ultimately occurs, but this may take several months. More commonly, patients develop chronic changes secondary to ARDS. The chest film demonstrates coarse linear, curvilinear, and reticular opacities representing fibrosis. Intrapulmonary cysts, which may form during the acute phase from abscesses or during the proliferative and fibrotic phases from barotrauma or cicatrization, frequently persist. Locu-

Figure 2 ARDS: A 44-year-old female with Pseudomonas pneumonia. Portable chest radiograph demonstrates bilateral homogeneous opacities with a central distribution. The radiographic abnormalities of ARDS may not be peripheral.

Acute respiratory distress syndrome

48

lated pneumothoraces may also become common chronic abnormalities on the chest film of recovering patients.

V. Complications of ARDS A. Support Lines Most individuals with ARDS require days or weeks of endotracheal (ET) intubation with assisted mechanical ventilation. Chest films are frequently obtained to monitor the position of these tubes. The desired placement puts the tip 3–5 cm above the carina with the cervical spine in neutral position. If the patient’s neck at the time the film is taken is flexed, this will drive the ET tube down toward the carina and spuriously misrepresent its position. In extension the tip of the ET tube will appear higher than it is when the patient’s neck is in neutral position. Because of the more vertical course of the right mainstem bronchus, most low malpositions place the ET tube in this bronchus, resulting in collapse of the left lung. If the misplaced tube is not adjusted, positive pressure ventilation delivered to one lung may result in more frequent episodes of barotrauma. If the ET tube is placed too high, adequate delivery of oxygen to the lungs may be inhibited, laryngeal injury may occur, and there is a great possibility that the patient may accidentally become extubated. Too high a placement, certainly above the level of the clavicles, may also lead to mistakes in establishing the proper pressure settings for ventillation. Another important observation to make is the status of the ET tube balloon cuff. This structure, meant to partially seal the airway and prevent back flow of air, is situated near the tip of the ET tube. The balloon should not be overinflated, as this may lead to injury of the tracheal mucosa with subsequent edema and in some instances eventual stenosis. An overinflated ET tube balloon can be recognized as a round or oval lucency near the tip of the tube with expansion of the trachea at this site. In an analysis of 30 survivors of ARDS, Elliot and coworkers concluded that laryngotracheal stenosis is an important cause of exertional dyspnea following treatment of ARDS (8). Etiological factors in this group of patients included difficult orotracheal intubation and high tracheal cuff pressures. Radiographs are usually obtained after placement of catheters within the thoracic deep venous system. The lines are frequently placed via a subclavian or internal jugular vein approach. Pneumothorax may occur in 5–6% of patients. It is also important to document catheter position. Ideally the tip of a central venous catheter should be placed within the central lumen of a large vein such that the tip does not abut the vein or heart wall end-on. Catheter tips can be safely located within the upper right atrium provided they do not abut the atrial wall or pass into the coronary sinus (9). Pulmonary artery catheters are commonly used in critically ill patients for measurement of pulmonary capillary wedge pressure. These catheters are inserted via a subclavian or internal jugular vein and ideally terminate distal to the pulmonic valve with the tip in the right or left pulmonary artery or interlobar arteries. Complications include pulmonary infarction, knotting, and cardiac perforation.

Radiographic findings of the acute respiratory distress syndrome

49

B. Barotrauma Mechanical ventilation is the mainstray of therapy in patients with ARDS. Different strategies of oxygen delivery have been tried since the disease was first described, but the most frequently used regimen remains positive pressure ventilation. Injury to the lungs has been a continuing problem with this therapy, and the incidence of barotrauma can be high. Gammon and colleagues investigated mediastinal emphysema and pneumothorax in a group of patients on mechanical ventilation for a variety of diseases (10). Of 29 patients with ARDS, 66% developed barotrauma, 62% developed mediastinal emphysema, and 60% developed pneumothorax. Recent studies, however, have not reported such a high incidence. A study by Schnapp and colleagues showed an incidence of barotrauma of 13% in 100 patients with ALI (11). In this study, mortality rates were not different in patients with and without barotrauma. Another study by Eisner and colleagues demonstrated an incidence of barotrauma of 13% and that higher positive end-expiratory pressure (PEEP) is associated with an increased risk of barotrauma (12). Patients with decreased lung compliance, one of the pathophysiological consequences of ARDS, are especially subject to this complication. The mechanism is associated with rupture of alveoli. Intense shearing forces develop in the alveolar walls as a result of the high pressures necessary to separate the collapsed walls of surfactant-deficient alveoli during inspiration. After rupture of alveoli, air is introduced into the interstitial space and from there spreads peripherally to the subpleural regions, either adjacent to fissures or along the lateral chest wall; another path takes air centrally into the mediastinum. At the lung periphery, rupture of the visceral pleura results in pneumothorax. Air that has dissected along bronchovascular bundles into the mediastinum (causing pneumodiastinum or pneumopericardium) may migrate into the soft tissues of the neck and chest, resulting in subcutaneous emphysema, or may break through the parietal pleura into the pleural space, resulting in pneumothorax. The earliest radiographic finding of barotrauma is pulmonary interstitial emphysema (PIE). This arises when air first exits the alveoli into the interstitial space. Radiographically PIE is recognized as linear or rounded lucencies radiating from the hilum, along bronchovascular pathways, and sometimes as a mottled or “salt and pepper” pattern throughout the lungs (Fig. 3). Larger collections of air may form radiolucencies in the perihilar or subpleural regions and have been termed air cysts or pneumatoceles. Subpleural air collections have an increased risk of causing pneumothorax. The detection of PIE suggests that the patient has suffered barotrauma and may prompt an effort to lower airway pressures and the patient’s tidal volume, if they are not already maximally reduced. Pneumomediastinum is a sign of barotrauma in mechanically ventilated patients. Chest radiographs show a line or band of lucency outlining the contours of the heart or other mediastinal structures (Fig. 4). The continuous diaphragm sign, another indicator of pneumomediastinum, occurs when air overlying the central portion of the diaphragm under the

Acute respiratory distress syndrome

50

Figure 3 Pulmonary interstitial emphysema (PIE): A 24-year-old male

Radiographic findings of the acute respiratory distress syndrome

51

with disseminated blastomycosis developed ARDS. A portable chest film (A) demonstrates a mottled heterogeneous opacification in the left lung. This is a typical appearance of PIE. A close-up of the left midlung (B) demonstrates PIE. (From Ref. 13a. Reprinted by permission from W.B.Saunders Company.)

Figure 4 Pneumomediastinum: A close-up of the chest demonstrates sharp definition of the complete superior border of the diaphragm. This appearance of a continuous diaphragm is a classic example of pneumomediastinum. (From Ref. 13a. Reprinted by permission from W.B.Saunders Company.) heart allows the top of the diaphragm to be seen as a continuous structure (13). This sign should permit recognition of pneumomediastinum on supine or upright studies. Pneumopericardium or a medially placed pneumothorax may simulate pneumomediastinum. Decubitus views can help differentiate between these air

Acute respiratory distress syndrome

52

collections as air in the mediastinal space rarely moves or moves slowly, whereas air in the pericardium or pleural space moves to nondependent locations within these compartments almost immediately with position changes. Pneumothorax is the most feared form of barotrauma and the only one for which treatment is performed. Prolonged or high-pressure ventilation, especially in patients with decreased lung compliance, may lead to recurrent pneumothoraces or rapid enlargement of a small pneumothorax, occasionally producing tension physiology. A pneumothorax is typically thought of and identified as a thin white curved line (visceral pleura) above the lung apex with absence of pulmonary vessels superiorly or laterally. But these abnormalities are typically seen in patients in upright position. In patients with ARDS, chest films are obtained in the intensive care unit with the patient usually in the supine position. In these cases, air in the pleura collects anteriorly in the most ventral, nondependent portion of the thorax, which is typically at the level of the diaphragm. The radiographic appearances caused by this location of pleural air include the deep sulcus sign, anterior sulcus sign, subpulmonic air collection, and occasionally epicardial fat sign (Fig. 5). The most common of these is the deep sulcus sign, which consists of a deep lateral costophrenic angle on the involved side (14). Subpulmonic pneumothorax, like subpulmonic pleural fluid, is a collection within the pleural space above the diaphragm and can be recognized by a sharp thin curvilinear opacity representing the visceral pleura above and often parallel to the hemidiaphragm (15). Tocino and colleagues examined 88 critically ill patients with 112 pneumothoraces, and they found that the abnormal air collections were located in the anteromedial space in 38% and subpulmonic space in 26% of occurences (16).

Radiographic findings of the acute respiratory distress syndrome

Figure 5 Pneumothorax: This 44-yearold male with apiration pneumonia developed ARDS. Initial portable chest

53

Acute respiratory distress syndrome

54

film (A) demonstrates bilateral heterogeneous opacities. The subsequent film (B) demonstrates lucency overlying the right costophrenic angle. Air collects in this location when patients are supine, creating a deep sulcus representing pneumothorax. Tension physiology resulting from a pneumothorax may be suggested radiographically when complete collapse of the ipsilateral lung and mediastinal shift to the contralateral side are observed (17). However, in patients with ARDS and decreased lung compliance or “stiff lungs,” tension pneumothorax can exist without complete lung collapse. Radiographic findings in these cases may include flattening of the ipsilateral diaphragm, enlarging loculated pneumothorax, and mediastinal shift accompanied by deterioration of clinical status. C. Pneumonia Nosocomial pneumonia is a common complication in patients who are intubated and occurs more often in patients with ARDS than in other ventilated patients. Recently, Chastre and colleagues reported a 55% incidence of nosocomial pneumonia in patients with ARDS, as compared to 28% incidence in those patients without ARDS (18). The same investigators found that the most frequently isolated organisms in these patients were methicillin-resistant Staphylococcus aureus (23%), nonfermenting gram-negative bacilli (Pseudomonas aeruginosa) (21%), and Enterobacteriaceae (21 %). The diagnosis of nosocomial pneumonia in ARDS is difficult as the patient’s chest radiograph is already abnormal. An asymmetrical opacity or more homogeneous opacity than seen elsewhere in the lung may be a radiographic clue of superimposed pneumonia. However, this finding is not sensitive or specific for the diagnosis of pneumonia. WinerMuram and coworkers studied the overall accuracy of chest radiography in diagnosing pneumonia in patients with ARDS and found it to be approximately 50% (19). They found that an increase in false-negative interpretations occured because the diffuse opacities in ARDS obscure the radiographic findings of pneumonia. The same investigators also concluded that the presence of pleural effusion was not a helpful indicator in the diagnosis of super-imposed pneumonia. This is supported by recent work with CT scanning of ARDS, which has revealed an unexpected higher incidence of pleural fluid in this setting even in patients not felt to be infected. Winer-Muram and colleagues also studied the ability of CT to detect pneumonia in patients with ARDS (20). They found that diagnostic accuracy was approximately 65–70% primarily because of a 70% true-negative result. Unfortunately, no individual CT findings were found to reliably differentiate pneumonia from ARDS.

Radiographic findings of the acute respiratory distress syndrome

55

VI. Computed Tomography and ARDS For the last 15 years CT has been used by some investigators to better understand the distribution of abnormalities in the lungs of patients with ARDS. Several studies using this modality have shown that ARDS does not affect the lungs homogeneously as frequently depicted on chest radiography (21–24). During the early stages of disease, CT reveals: (1) normal or near-normal lung frequently located in nondependent regions; (2) ground glass opacities in the anterior and middle portions; and (3) consolidation in the most dependent lung (Fig. 6). It is believed that the weight of adjacent pulmonary parenchyma causes compression of more dependent alveoli, resulting in this distribution of opacities. Furthermore, CT scanning has

Figure 6 ARDS: This is the same patient as in Figure 4. CT study demonstrates ground glass opacities in the anterior and middle portions of the lungs and consolidation in the dependent regions. This is a typical distribution of abnormalities with ARDS. been used to demonstrate a reversal of lung opacification when patients with ARDS are placed in the prone position.

Acute respiratory distress syndrome

56

After approximately one week of initial lung injury, during the organizing phase of ARDS, there is a decrease in overall lung density and the pattern of disease becomes more linear or reticular. During this phase, the lung parenchyma undergoes extensive remodeling with development of some degree of pulmonary fibrosis. Desai and coworkers described the CT findings at long-term follow-up in 27 patients who survived ARDS (25). A coarse reticular pattern with distortion of lung parenchyma was the most frequent pattern in 23 patients (85%). This reticular pattern had an anterior distribution and was associated with CT signs of pulmonary fibrosis including architectural distortion and traction bronchiectasis. The extent of disease was related to the length of time that patients underwent pressure-controlled inverse ratio ventilation. Some feel that the anterior distribution of fibrosis occurs because alveolar overdistention is predominantly seen in this region, that the anterior ground glass opacities seen on CT acutely represent intraalveolar septal inflammation from this overdistention, and that this ultimately results in fibrosis. Complications of ARDS can also be detected using CT (Fig. 7). Some investigators have found that CT scans are more sensitive than chest radiographs in detecting pulmonary interstitial emphysema, the earliest sign of barotrauma (26). The CT findings of PIE in patients with ARDS include air within the interlobular septa, air around the pulmonary veins and bronchi, and variably sized subpleural, air-filled cysts. In regards to other evidence of barotrauma, Tagliabue and coworkers showed that 40% of pneumothoraces and 80% of pneumomediastinums seen on CT scan were not seen on chest radiographs (27). Pleural effusions and lung abscesses were also seen more frequently on CT than on plain chest films. On the other hand, the benefits of observing these abnormalities on patient outcome have not been proved, and some believe that CT for ARDS should be restricted to problem cases (28).

Figure 7 ARDS: This is the same patient as in Figures 4 and 5. Followup CT study demonstrates, in addition

Radiographic findings of the acute respiratory distress syndrome

57

to ground glass and consolidative opacities, an intrapulmonary cyst in the right middle lobe likely secondary to barotrauma. Finally, CT has been used as an investigative tool for studying the pathophysiology of ARDS and novel approaches to treatment of this disease. Pelosi et al. used CT to analyze the effects of prone positioning of patients with ARDS (29). They discovered that some otherwise atelectatic areas of lung near the diaphragm might be recruited while the patient was prone and that this could perhaps lead to improved oxygenation. Gattinoni has presented a number of revelations in regards to pathophysiology of ARDS, including the distribution of disease, differences in CT pattern based on etiology of ARDS, and the relationship of decreased lung compliance versus decreased lung size when volutrauma or barotrauma occurs (28). CT has been helpful in the management of complications due to barotrauma in some instances. Chon and coworkers (30) used CT-guided percutaneous catheter drainage for loculated thoracic air collections (pneumothoraces, pneumatoceles, and tension pneumothorax) in nine patients mechanically ventilated for a clinical diagnosis of ARDS (30). Eight (89%) of the nine patients were successfully treated with CT-guided catheter placement, obviating the need for surgical intervention. Miller and colleagues studied the value of CT scans in patients in the intensive care unit and concluded that, for the majority of patients, there was significant agreement between chest radiographs and CT (31). CT did demonstrate additional findings, but most of these were not clinically important. Those that were found useful were the identification of abscesses or postoperative mediastinal fluid collections and unsuspected pneumonia or pleural effusions. The conclusions of this investigation suggested that CT scanning may be beneficial in selected cases. In another study, Desai and Hansell concluded that CT is best reserved to detect occult complications in patients who are deteriorating or not responding to treatment at the expected rate (32). Moving patients from the ICU to the CT scanner is not a trivial matter as it puts the patient into a less optimal clinical environment and requires additional manpower. Thus, the widespread use of CT in ARDS patients cannot be advocated, but in selected cases CT should be considered. Some medical centers are considering the installation of CT scanners in their intensive care units to facilitate more expeditious and frequent use of this imaging modality.

VII. Other Imaging Modalities A. Nuclear Medicine Several investigators have studied the value of radioactive gallium studies in patients with ARDS. The primary mechanism of uptake of gallium is related to interstitial inflammation and repair of pulmonary tissue as occurs in ARDS (33). In one study, investigators used a dual-radionuclide method (gallium-67 transferrin and Technesium99m red blood cells) to measure pulmonary microvascular permeability in patients with

Acute respiratory distress syndrome

58

pulmonary edema due to ARDS and hydrostatic pulmonary edema (34). Radioactivity measurements over the lungs and blood were used to calculate the pulmonary leak index as a measure of pulmonary microvascular permeability. These investigators concluded that the gallium-67 pulmonary leak index is useful in differentiating ARDS and hydrostatic pulmonary edema. Similarly, PET has been used as a tool in studying the physiology of ARDS. Mintun and colleagues quantified pulmonary vascular permeability with PET and gallium-68labeled transferrin in normal volunteers, in six dogs after oleic acid-induced lung injury, and in patients with ARDS (35). These investigators measured the pulmonary transcapillary escape rate (PTCER) and demonstrated marked differences in PTCER between abnormal and normal lung tissue. B. Magnetic Resonance Imaging There are several studies addressing the imaging of lung parenchyma and pulmonary edema using MRI. However, this imaging modality is not currently used in the evaluation of patients with ARDS. Caruthers and colleagues used three-dimensional MRI to measure pulmonary microvascular barrier permeability in animals after oleic acid-induced lung injury (36). They concluded that 3-D MR imaging of pulmonary edema is sensitive for measuring small changes in lung water. In another study, contrast-enhanced MRI with a macromolecular agent was used to differentiate between hydrostatic pulmonary edema and edema caused by abnormal capillary permeability (37). However, widespread use of these methods has not yet occurred in critically ill patients.

VIII. Conclusions Chest radiographs are essential for the diagnosis and monitoring of ARDS, although better methods are needed to standardize their interpretation. Although new imaging modalities have been developed since the first description of the disease, only CT has some clinical importance in complicated cases. CT and PET scanning have been useful modalities in better understanding the pathophysiology of the disease, particularly in recognizing the heterogeneity of intrapulmonary consolidation. Imaging methods to measure lung vascular permeability or extravascular lung water have not been proven to be sufficiently reproducible or reliable for clinical use. In general, these imaging methods should be limited to clinical research endeavors. Perhaps new technological advances in radiology will lead to improved utilization of these modalities with subsequent better evaluation and treatment of ARDS patients.

References 1. Ashbaugh DG, Bigelow DB, Petty TL, et al. Acute respiratory distress in adults. Lancet 2:319, 1967. 2. Rubenfeld GD, Caldwell E, Granton J, et al. Interobserver variability in applying a radiographic definition for ARDS. Chest 116:1347, 1999. 3. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 342:1334, 2000.

Radiographic findings of the acute respiratory distress syndrome

59

4. TenHoor T, Mannino DM, Moss M. Risk factors for ARDS in the United States: analysis of the 1993 National Mortality Followback Study. Chest 119: 1179, 2001. 5. Doyle RL, Szaflarski N, Modin GW, et al. Identification of patients with acute lung injury: predictors of mortality. Am J Respir Crit Care Med 152:1818, 1995. 6. Thomason JW, Wesley Ely E, Chiles C, et al. Appraising pulmonary edema using supine chest roentgenograms in ventilated patients. Am J Respir Crit Care Med 157:1600, 1998. 7. Aberle DR, Wiener-Kronish JP, Webb WR, et al. Hydrostatic vs increased permeability pulmonary edema: diagnosis based on radiographic criteria in critically ill patients. Radiology 168:73, 1988. 8. Elliot CG, Rasmusson BY, Crapo RO. Upper airway obstruction following adult respiratory distress syndrome. An analysis of 30 survivors. Chest 94:526, 1988. 9. Fletcher SJ, Bodenham AR. Safe placement of central venous catheters: Where should the tip of the catheter lie? Br J Anaesth 85:188, 2000. 10. Gammon RB, Shin MS, Buchalter SE. Pulmonary barotrauma and mechanical ventilation. Chest 102:568, 1992. 11. Schnapp LM, Chin DP, Szaflarski N, Matthay M. Frequency and importance of barotrauma in 100 patients with acute lung injury. Crit Care Med 23:272, 1995. 12. Eisner MD, Thompson BT, Schoenfeld D, et al. Airway pressures and early barotrauma in patients with acute lung injury and acute respiratory distress syndrome. Am J Respir Crit Care Med 165:978, 2002. 13. Bejvan SM, Godwin JD. Pneumomediastinum: old signs and new signs. AJR 166:1041, 1996. 13a. Goodman PC. Radiographic findings in patients with acute respiratory distress syndrome. Clinics in Chest Medicine 21:41, 2000. 14. Gordon R. The deep sulcus sign. Radiology 136:25, 1980. 15. Rhea JT, vanSonnenberg E, McLoud TC. Basilar pneumothorax in the supine adult. Radiology 133:593, 1979. 16. Tocino IM, Miller MH, Fairfax WR. Distribution of pneumothorax in the supine and semirecumbent critically ill adult. AJR 144:901, 1985. 17. Gobien RP, Reiness, HD, Schabel SI. Localized tension pneumothorax: unrecognized form of barotrauma in adult respiratory distress syndrome. Radiology 142:15, 1982. 18. Chastre J, Trouillet JL, Vuagnat A, et al. Nosocomial pneumonia in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 157:1165, 1998. 19. Winer-Muram HT, Rubin SA, Ellis JV, et al. Pneumonia and ARDS in patients receiving mechanical ventilation: diagnostic accuracy of chest radiography. Radiology 188:479, 1993. 20. Winer-Muram HT, Steiner RM, Gurney JW, et al. Ventilator-associated pneumonia in patients with adult respiratory distress syndrome: CT evaluation. Radiology 208:193, 1998. 21. Gattinoni L, Bombino M, Pelosi P, et al. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA 271:1772, 1994. 22. Gattinoni L, Mascheroni D, Torresin A, et al. Morphological response to positive end expiratory pressure in acute respiratory failure: computerized tomography study. Intensive Care Med 12:137, 1986. 23. Gattinoni L, Presenti A, Bombino M, et al. Relationships between lung computed tomographic density, gas exchange, and PEEP in acute respiratory failure. Anesthesiol 69:824, 1988. 24. Gattinoni L, Presenti A, Torresin A, et al. Adult respiratory distress syndrome profiles by computed tomography. J Thorac Imaging 1(3):25, 1986. 25. Desai SR, Wells AU, Rubens MB, et al. Acute respiratory distress syndrome: CT abnormalities at long-term follow up. Radiology 210:29, 1999. 26. Kemper AC, Steinberg KP, Stern EJ. Pulmonary interstitial emphysema: CT findings. AJR 172:1642, 1999. 27. Tagliabue M, Casella TC, Zincone GE, et al. CT and chest radiography in the evaluation of adult respiratory distress syndrome. Acta Radiol 35:230, 1994.

Acute respiratory distress syndrome

60

28. Gattinoni L, Caironi P, Pelosi P, et al. What has computed tomography taught us about the acute respiratory distress syndrome? Am J Respir Crit Care Med 164:1701, 2001. 29. Pelosi P, Crotti S, Brazzi L, Gattinoni L. Computed tomography in adult respiratory distress syndrome: What has it taught us? Eur Respir J 9:1055, 1996. 30. Chon KS, vanSonnenberg E, D’Agostino HB, et al. CT-guided catheter drainage of loculated thoracic air collections in mechanically ventilated patients with acute respiratory distress syndrome. AJR 173:1345, 1999. 31. Miller Jr WT, Tino G, Friedburg JS. Thoracic CT in the intensive care unit: assessment of clinical usefulness. Radiology 209:491, 1998. 32. Desai SR, Hansell DM. Lung imaging in the adult respiratory distress syndrome: current practice and new insights. Intensive Care Med 23:7, 1997. 33. Passamonte, PM, Martinez AJ, Singh A. Pulmonary gallium concentration in the adult respiratory distress syndrome. Chest 85:828, 1984. 34. Raijmakers, PG, Groeneveld AB, Teule GJ, Thijs LG. Diagnostic value of the gallium-67 pulmonary leak index in pulmonary edema. J Nuclear Med 37:1316, 1996. 35. Mintun MA, Dennis DR, Welch MJ, et al. Measurements of pulmonary vascular permeability with PET and gallium-68 transferrin. J Nuclear Med 28:1704, 1987. 36. Caruthers SE, Pachal CB, Pou NA, et al. Regional measurements of pulmonary edema by using magnetic resonance imaging. J Appl Physiol 84:2143, 1998. 37. Berthezene Y, Vexler V, Jerome H, et al. Differentiation of capillary leak and hydrostatic pulmonary edema with a macromolecular MR imaging contrast agent. Radiology 181:773, 1991.

5 Pulmonary Pathology of the Acute Respiratory Distress Syndrome Diffuse Alveolar Damage JOSEPH F.TOMASHEFSKI Jr. Case Western Reserve University School of Medicine and MetroHealth Medical Center Cleveland, Ohio, U.S.A.

I. Introduction The pathological features of the lung in the acute respiratory distress syndrome (ARDS) have been collectively labeled diffuse alveolar damage (DAD) and represent a timedependent, stereotypic response to alveolar injury (1). DAD can be conveniently divided into three sequential and overlapping phases: the exudative phase of edema and hemorrhage, the proliferative phase of organization and repair, and the fibrotic phase of end-stage fibrosis (1–4) (Table 1). The latter two phases are frequently merged and designated as the fibroproliferative phase. For an individual patient, the histological features of DAD correlate more with the duration of injury than its initiating cause. Permeability pulmonary edema, an important consequence of acute alveolar injury, is prominent in the early stages and heralds the onset of rapidly progressive pulmonary fibrosis, which dominates the later course of the disease and is often a limiting factor for survival. In this chapter, each of the phases of DAD will be described separately and morphological features will be correlated with clinical events and proposed mechanisms of injury.

Table 1 Temporal Features of Diffuse Alveolar Damage Exudative phase (days 1–7)a

Proliferative phase (days 7– 21)

Fibrotic phase (>day 21)

Interstitial and intra-alveolar edema

Interstitial myofibroblast reaction Collagenous fibrosis

Hemorrhage

Lumenal organizing fibrosis

Microcystic honeycombing

Leukoagglutination

Chronic inflammation

Traction bronchiectasis

Necrosis

Parenchymal necrosis

Arterial tortuosity

Pulmonary pathology of the acute respiratory distress syndrome

Type 1 pneumocytes

63

Mural fibrosis

Endothelial cells

Type 2 pneumocyte hyperplasia

Hyaline membranes

Obliterative endarteritis

Platelet-fibrin thrombi

Macrothrombi

Medial Hypertrophy

Increased megakaryocytes a

Dates are approximations.

Figure 1 Diffuse alveolar damage, exudative phase. Lung parenchyma is homogeneously dark. Interstitial air dissection is evident along a blood vessel, interlobular septum, and visceral pleura (arrows). Barium sulfate injection medium demonstrates reduced filling of peripheral arteries (scale=0.33 cm). II. Exudative Phase The exudative phase of DAD includes the first 4–7 days after the onset of respiratory failure (1, 4). The lungs of patients who die during this period are heavy, and the visceral pleura is red-blue and focally hemorrhagic. The parenchymal surface is dark red with a diffuse, airless consistency, which, on compression, exudes only minimal blood-tinged fluid (2, 5, 6) (Fig. 1). Close inspection of the formalin-inflated specimen reveals

Acute respiratory distress syndrome

64

punctate airspaces that represent dilated alveolar ducts accentuated by adjacent collapsed or indu-rated alveolar parenchyma. In patients who die of severe respiratory failure, the lungs are usually diffusely involved, but the process may be regionally distributed (7). In the early stages, computed tomography (CT) scans depict accentuated involvement of the postero-dependent lung zones (8). The earliest histological changes are capillary congestion, interstitial and alveolar edema, and intra-alveolar hemorrhage (1, 2, 6, 9). The most distinctive histological feature of the exudative phase, however, is the

Figure 2 Diffuse alveolar damage, exudative phase. Alveolar ducts (AD) appear dilated resulting from collapse and consolidation of adjacent alveoli and layering of hyaline membranes (arrowheads) on alveolar septa. Interlobular septum (IS) is widened by edema (H&E; original magnification×42). (From Ref. 103.) presence of dense, eosinophilic hyaline membranes that usually are most prominent along the alveolar duct, where they adhere to the tips of alveolar septa and extend across the alveolar orifices (3, 9, 10) (Figs. 2, 3). Hyaline membranes are composed of condensed fibrin and serum proteins that have leaked through the injured alveolocapillary membrane and mixed with cellular debris within the alveolar space (Fig. 3). Immunohistochemical and immunofluorescent staining demonstrates immunoglobulin, fibrinogen, sur-factant,

Pulmonary pathology of the acute respiratory distress syndrome

65

and, to a lesser extent, complement within the hyaline membrane and a thin layer of fibronectin on its surface (11) (Fig. 4). Alveolar ducts appear dilated in contrast to the adjacent alveoli, which are collapsed and congested and contain edema and fibrinous exudate (Fig. 2). Immunohistochemical staining for epithelial membrane antigen depicts selective, early loss of alveolar epithelial cells adjacent to hyaline membranes at the tips of alveolar septa along the alveolar duct (12). The reason for the accentuated injury and deposition of hyaline membranes along the alveolar duct is unknown, but relatively high concentrations of oxygen in the alveolar duct are speculated to play a pathogenetic role, as is the drying effect on intra-alveolar edema of

Figure 3 Diffuse alveolar damage, exudative phase. Coarse hyaline membranes line the alveolar duct (AD) and traverse the alveolar orifices. Notice also acute alveolar hemorrhage and alveolar septal edema (H&E; original magnification×160). (From Ref. 84.)

Acute respiratory distress syndrome

66

Figure 4 Direct immunofluorescence staining of hyaline membranes for immunoglobulin (original magnification×167). (From Ref. 103.)

Figure 5 Early endothelial injury. Endothelial cells (EN) are situated on capillary basement membrane (BM). There is widening of interendothelial

Pulmonary pathology of the acute respiratory distress syndrome

junction (arrow) and focal disruption of endothelial cell membrane (arrowhead). Cellular debris (D) is present in capillary lumen (L) (scale=0.4 µm). (From Ref. 103.)

Figure 6 Early endothelial injury. Portions of two endothelial cells with intact intercellular junction (arrow). The number of pinocytotic vesicles is increased. Strands of fibrin (F) are present in capillary lumen (L) (scale=0.3 ìm). (From Ref. 103.)

67

Acute respiratory distress syndrome

68

Figure 7 In the exudative phase of diffuse alveolar damage, the capillary contains erythrocytes (E), fibrin strands (F), and necrotic cells (NC). The alveolar epithelial surface is denuded (arrowheads). Necrotic remnants of type 1 cells (arrow) have sloughed from the basement membrane (scale=2 µm). (From Ref. 84.) gas delivered at high inspiratory pressures (6, 10). Reduced alveolar volume is considered to be a secondary complication possibly attributable to acquires abnormalities of surfactant and consequent elevation of alveolar surface tension (6, 13, 14). Segmental or lobar absorptive atelectasis is not a regular feature of the exudative phase of DAD but may occur following improper placement of endotracheal tubes resulting in unequal lung ventilation (6). Diffuse compressive atelectasis is a unique complication in severely burned children in whom the burn eschar produces circumferential thoracic constriction (6). In the exudative phase of diffuse alveolar damage, there is ultrastructural evidence of endothelial injury such as endothelial cell swelling, widening of interendothelial junctions, and increased numbers of pinocytotic vesicles (15–18) (Figs. 5, 6). Inconstant

Pulmonary pathology of the acute respiratory distress syndrome

69

features of severe endothelial injury include cellular necrosis and disruption with denudation of the capillary basement membrane and intravascular fibrin accumulation (15, 17, 19) (Fig. 7). Megakaryocytes are often identified in alveolar capillaries, and the number of

Figure 8 Alveolar septum markedly widened by edema (lucent areas), extravasated erythrocytes (E), and fibrin (arrows). The alveolar basement membrane (arrowheads) shows focal loss of epithelium. A necrotic type 2 pneumocyte (NC) is present in the alveolar lumen (L). An intact type 2 cell (P2) rests on the alveolar basement membrane. An alveolar capillary with intact endothelial layer contains a leukocyte (LC) (scale=3.2 µm). (From Ref. 103.) megakaryocytes has been significantly correlated with the presence of endocapillary platelet-fibrin thrombi (20). Intracapillary aggregates of neutrophils are focally prominent, especially in DAD secondary to sepsis or trauma (16, 17, 21). Sparse numbers of neutrophils usually reside within the alveolar space, but when the alveolus contains large numbers of nuetrophils, bacterial pneumonia is also likely to be present. Alveolar

Acute respiratory distress syndrome

70

septa are greatly widened by interstitial edema, fibrin, and extravasated erythrocytes (Fig. 8). Frequently the extent of interstitial edema is disproportionately severe compared to the relatively mild ultrastructural changes in a generally intact endothelial cell layer (15, 16, 22). The subtle morphological changes in endothelial cells may, however, underlie a significant functional loss of integrity of the endothelium, or, as Bachofen and Weibel have suggested, the extraordinary repair capacity of endothelial cells might make it difficult to trace transient lesions (15, 16). In contrast to the inconstant and often slight endothelial changes, the alveolar epithelium usually demonstrates extensive necrosis of type 1 cells, which slough from the alveolar surface, leaving a denuded basement

Figure 9 A hyaline membrane (HM) and cellular debris (D) are apposed to the alveolar surface. Type 1 pneumocytes (P1) are intact, but show focal separation from the alveolar basement membrane (arrow). Capillary endothelial cells (EN) are swollen. Note marked infolding of the alveolar basement membrane (arrowheads). E=erythrocyte (scale=3.2 µm). (From Ref. 103.)

Pulmonary pathology of the acute respiratory distress syndrome

71

membrane to which are adhered coarse hyaline membranes, fibrin, and cellular debris (15–18, 22) (Figs. 7–9). Type 2 cells, which are more resistant to injury, persist and maintain the capacity to differentiate into type 1 cells (15–18, 22, 23). With the loss of the alveolar epithelial barrier, interstitial fluid can escape freely into the alveolar space (16).

III. Proliferative Phase In the proliferative phase of DAD, permeability pulmonary edema is overshadowed by organization, inflammation, and fibrosis (1, 2, 9–11). Type 2 cells proliferate along alveolar septa as early as 3 days following the onset of clinical ARDS, and fibrosis is apparent by the tenth day (1, 10). The identification of type III procollagen in bronchoalveolar lavage (BAL) fluid at 24 hours indicates, however, that the fibroproliferative process is initiated early in ARDS (24, 25). In some patients the rapid conversion of the lung to a noncompliant, end-stage organ is an important limiting factor for survival (26, 27). The lungs of those who die between 1 and 3 weeks after the onset of respiratory failure are heavy (in excess of 2000 g combined weight), and the parenchymal surface is red-gray and glistening with a slippery texture due to

Figure 10 Proliferative phase. Lung parenchyma on right is pale, with a gelatinous appearance. There is effacement of parenchymal airspaces by early connective tissue. Darker parenchyma on left is hemorrhagic. There is extreme reduction of small blood vessels, which contains barium

Acute respiratory distress syndrome

72

sulfate gelatin contrast medium (scale=0.55 cm). an abundance of newly formed connective tissue (Fig. 10). Zones of complete airspace effacement alternate with areas in which dilated airspaces, 1–2 mm in diameter, are accentuated by adjacent fibrous tissue. Histologically and ultrastructurally, epithelial cell regeneration heralds the onset of the proliferative phase, as rows of keratin-bearing type 2 pneumocytes line the alveolar surface (Fig. 11). Many of these cells have cytoplasmic lamellar bodies and surface microvilli and can be stained immunohistochemically for surfactant apoprotein (1, 11, 16, 17, 20, 22, 28–30) (Fig. 12). The cellular density of lamellar bodies, however, is often sparse, and their structure is occasionally abnormally large and complex (17, 30). Tumor necrosis factor (TNF)-α and IL (interleukin)-1β have been detected immunohistochemically in type 2 pneumocytes and alveolar macrophages (31, 32). Squamous differentiation may also be prominent in epithelial cells that stain for keratin but not surfactant apoprotein (4, 11). Cytoplasmic hyaline-like material, a nonspecific marker of cellular injury, is noted in cells expressing ultrastructural evidence of squamous differentiation (4, 11, 33).

Figure 11 Type 2 pneumocytes, immunostained for keratin, diffusely proliferate on the surface of thickened alveolar septa. Flat surface cells represent type 1 pneumocytes (arrows). Focally, the alveolar basement membrane is bare (small arrowheads) (original magnification×250).

Pulmonary pathology of the acute respiratory distress syndrome

73

Figure 12 Type 2 pneumocytes (P2) are characterized by cytoplasmic lamellar bodies and microvilli. The alveolar basement membrane is infolded (arrowheads) and a swollen endothelial cell (EN) occupies the lumen of a capillary. A neutrophil (N) is in the interstitium (scale=1.3 µm). (From Ref. 84.) Epithelial cell nuclei are large, vesicular, and may contain prominent nucleoli. Nuclear atypia is more pronounced when lung injury is induced by cytotoxic agents such as bleomycin, busulfan, and radiation or by viral infection. However, extreme nuclear atypia may occur even in the absence of these factors. Severe epithelial nuclear atypia or florid squamous differentiation can simulate carcinoma on open lung biopsy or cytological smears of lung fluids (34). Extension of columnar epithelium along alveoli adjacent to bronchioles is prominent in ARDS initiated by viral infection (3). Type 2 epithelial cells are mitotically active and capable of regeneration; the origin of metaplastic squamous cells found in the alveolar region is uncertain but may be the residual basal cells in adjacent airways (11, 23). Within the alveolar wall, fibroblasts and myofibroblasts proliferate and subsequently migrate through breaks in the alveolar basement membrane into the fibrinous intra-

Acute respiratory distress syndrome

74

alveolar exudate (28, 35) (Figs. 13, 14). Fibroblasts convert the exudate to cellular granulation tissue and, ultimately, by the deposition of collagen, into sparsely cellular dense fibrous tissue (11, 17). Fibronectin is more prominently seen coating the

Figure 13 Myofibroblast, immunostained for muscle-specific actin, are present in the alveolar wall (arrowheads) and in organizing granulated tissue within the alveolar duct (arrows) (original magnification×400). surface of fibrinous exudate in the proliferative than in the exudative phase (11). Histochemical stains, such as alcian blue or Movat pentachrome for acid mucopolysaccharides, are strongly positive in the organizing granulation tissue. Increased hyaluronic acid has been demonstrated biochemically in lung tissue in experimental bleomycin toxicity and in bronchoalveolar lavage from patients with ARDS (36). Hyaluronic acid may serve as a sponge, drawing and retaining water within the connective tissue of the lung (36). Epithelial cells migrate over the surface of the organizing granulation tissue and thereby transform the intra-alveolar exudate into interstitial tissue. This process of intraluminal fibrosis by accretion occurs from the respiratory bronchiole to the alveolus but is most pronounced in the alveolar duct, where it was deemed “alveolar duct fibrosis” by Pratt and colleagues (3, 10) (Fig. 15). Although alveolar septal fibrosis contributes to lung remodeling, intraluminal fibrosis is emphasized as an essential mechanism of fibroplasia in ARDS (11). Within the alveolar duct, organizing granulation tissue can assume several patterns (4, 10) (Fig. 15).

Pulmonary pathology of the acute respiratory distress syndrome

75

Fibrous tissue can occlude the lumen of the duct or be distributed peripherally within the duct as a fibrous

Figure 14 Massively edematous alveolar wall (AW) is focally lined by cells showing transitional features between type 1 and type 2 pneumocytes (P). A primitive mesenchymal cell (M) has entered the alveolar space (AS) and is in contact with fibrinous exudate (F). Another mesenchymal cell (arrowhead) is aligned parallel to the basement membrane. An alveolar capillary contains prominent endothelial cells. The distance between the capillary and alveolar basement membrane is increased (scale=3.3 µm). (From Ref. 84.) ring. When filled with neutrophils or erythrocytes, ring-like alveolar duct fibrosis can be misinterpreted histologically for microabscesses or vascular malformations (3, 4, 10). Fibrosis by accretion also occurs within the alveolar spaces, producing “alveolar buds” (11, 37, 38). Bronchiolitis obliterans, which results from organized exudate in respiratory

Acute respiratory distress syndrome

76

and terminal bronchioles, is not usually prominent in ARDS but may be an important factor in acute respiratory failure secondary to toxic inhalants (39). After injury to the alveolocapillary unit, alveolar septa collapse and become sealed in apposition by organizing fibrin and proliferating pneumocytes (collapse induration) (4, 30, 38, 40). Collapse induration is a mechanism of alveolar remodeling that is best observed ultrastructurally as deep folds of the alveolar basement membrane traversed and sealed by type 2 pneumocytes (Fig. 16). Distorted remnants of basement membrane material may be observed in the thickened alveolar wall (40). The net result of collapse induration is fewer but larger alveoli and, consequently,

Figure 15 Patterns of alveolar duct fibrosis. Occlusion of alveolar duct lumens by plugs of granulation tissue is seen in longitudinal (1) and cross section (2). Peripheral orientation of fibrous tissue in the alveolar duct produces microcysts (3) (H&E; original magnification×42). (From Ref. 103.) dilated alveolar ducts (38). Dilatation of alveolar ducts and respiratory bronchioles (bronchiolectasis) has also been ascribed to mechanical ventilation and application of positive end-expiratory pressure (PEEP) (41). Morphometric ultrastructural studies of the lung in ARDS due to traumatic or septic shock vividly portray these progressive derangements of lung architecture. An increase in

Pulmonary pathology of the acute respiratory distress syndrome

77

the volume of the interstitial compartment due to interstitial edema is the most striking initial abnormality in the acute (exudative) phase (17, 19). After the onset of respiratory failure, there is an immediate initial reduction in the volume density of the alveolar epithelium, rapidly followed by a two- to threefold increase in the epithelial layer indicative of regeneration and proliferation of type 2 cells (16, 17). Capillary volume density decreases in the exudative phase, and the reduction persists in the proliferative phase, at least for patients with severely abnormal gas exchange (16). Among other cell populations that have been evaluated quantitatively, intracapillary granulocytes show an increased volume density in the early phase, while interstitial fibroblasts increase in the proliferative phase (17). By light microscopic morphometry,

Figure 16 Proliferative phase. Type 2 pneumocytes (P2) partially cover a denuded alveolar basement membrane (arrow). Note marked infolding of the alveolar basement membrane (arrowheads). The alveolar septum is edematous with extravasated erythrocytes (E). Capillaries have an intact endothelial layer and contain rows of erythrocytes. The alveolar lumen (L) is filled with electron dense proteinaceous edema (scale=4 µm). (From Ref. 103.)

Acute respiratory distress syndrome

78

Fukuda et al. demonstrated a gradual increase in the percentage of interstitial tissue with increased duration of ARDS (11). The percentage of intra-alveolar fibrosis was greatly increased beyond 10 days, but by day 35 the distinction between intra-alveolar and interstitial fibrosis was no longer possible because of extensive fibrous remodeling (11). Mast cells increase from the exudative to the proliferative phase, at which point increased mast cells are found in the majority of cases (42) (Fig. 17). The role of mast cells in fibrogenesis is uncertain (42).

IV. Fibrotic Phase and Longstanding Healed ARDS Within 3–4 weeks of the onset of ARDS, the lung may be completely remodeled by sparsely cellular collagenous tissue. The visceral pleura is coarsely nodular while the parenchyma demonstrates diffuse fibrosis or pale, irregular scars alternating with microcystic airspaces, 1 mm or larger in diameter (Fig. 18). The formation of contiguous large cysts, referred to as “adult bronchopulmonary dysplasia,” is unusual (43) (Fig. 19). Healed

Figure 17 Mast cells (arrowheads), immunostained for tryptase, are increased in thickened alveolar walls (original magnification×400). abscesses and chronic pulmonary interstitial emphysema are additional causes of large cyst-like spaces in end-stage ARDS. Peripheral bronchi appear dilated and are abnormally close to the visceral pleural surface (traction bronchiectasis). Fibrocystic changes are most pronounced in the anterior lung zones (8).

Pulmonary pathology of the acute respiratory distress syndrome

79

Microscopically, alveolar septa are widened by collagen. Thick cords or stellate scars are situated within the distorted, dilated remnants of alveolar ducts (Fig. 20). Fibrotic microcysts represent progression and dilatation of ring-shaped alveolar duct fibrosis (Figs. 15, 21). Honeycombing focally resembles that seen in idiopathic pulmonary fibrosis, but the size of the airspaces is generally smaller (Fig. 18). Lung collagen is increased in patients who survive more than 14 days, and the concentration and total amount of collagen have been correlated with the extent of fibrosis assessed histologically (44). Long-term follow-up studies of patients who survive an episode of ARDS indicate that abnormal spirometry persists in the early recovery period, and a subset of survivors (14– 50%) have at least mild airflow limitation one year after the acute episode (45). Severe pulmonary dysfunction affects only a few survivors. Prolonged pulmonary disability has been correlated with the initial severity of lung impairment or the persistence of

Figure 18 In the fibrotic phase of diffuse alveolar damage, areas of diffuse fibrosis (bottom) alternate with microcystic honeycombing. The airways are dilated and abnormally close to the pleural surface (traction bronchiectasis) (scale=2 cm). (From Ref. 84.) lung impairment during the acute episode, but not with histological features seen on open lung biopsy (46, 47). The phenomenon of long-term survivors with moderate or good pulmonary function who during their episode of ARDS demonstrated extensive fibrosis on open lung biopsy has suggested that fibrosis observed early in ARDS is potentially

Acute respiratory distress syndrome

80

reversible (26, 47). However, it is more likely that this discrepancy is due to sampling of regionally distributed DAD at the time of open lung biopsy. There have been few studies of lung morphology in survivors of ARDS. Lakshminarayan and colleagues observed mild interstitial fibrosis, epithelial hyperplasia, and increased alveolar macrophages with interstitial fibrosis, epithelial hyperplasia, and increased alveolar macrophages with interstitial lymphocytes in a patient 9 months after ARDS (46). Pratt found bundles of collagen within alveolar ducts of a patient successfully weaned from the ventilator (48). In a personally studied case, severe obliterative alveolar duct fibrosis persisted in a patient who survived 21 months after having been weaned from the ventilator after ARDS probably caused by a viral infection. A few survivors of ARDS may be prone to repeated episodes of DAD and acute respiratory failure (49). It has been

Figure 19 Adult bronchopulmonary dysplasia. Contiguous large cysts and ectatic bronchi produce a macrocystic honeycomb pattern in the upper lobe. Compare size of cystic airspaces to that of lobar bronchus (arrow). (Courtesy of Andrew Churg, M.D.)

Pulmonary pathology of the acute respiratory distress syndrome

81

suggested that recurrent ARDS is related to narcotic use or to complex drug regimens utilized in therapy (49).

V. Pulmonary Vascular Remodeling Pulmonary vascular injury and pulmonary hypertension are important complications of ARDS (50, 51). Pulmonary vascular lesions follow a temporal sequence that correlates with the duration of alveolar damage. Vascular injection studies using silicone polymer or barium sulfate gelatin

Figure 20 Fibrotic phase. Residual alveolar duct (AD) fibrosis (F) is present as a plug of collagenous tissue tethered to the tips of alveolar septa. Dilated serpignous vessels (arrows) contain postmortem barium injection medium (Elastic van Gieson; original magnification×107). (From Ref. 103.) demonstrate markedly reduced vascular filling of postmortem lung specimens (50, 52, 53) (Figs. 1, 10). Early in ARDS, vasoconstriction, thromboembolism, and interstitial edema may contribute to a raised pulmonary artery pressure (50). After several weeks, fibrous obliteration of the microcirculation and increased arterial muscularization are fixed lesions that contribute to pulmonary hypertension (52).

Acute respiratory distress syndrome

82

In lungs examined postmortem, thromboemboli are the most consistently observed vascular lesion, present in as many as 95% of patients (52). Macrothrombi (in arteries>1 mm in diameter), which correspond to the vascular filling defects demonstrated by bedside balloon occlusion pulmonary angiography in 48% of patients with ARDS, are more prevalent in patients who die in the early phase (52, 54). Thromboemboli detected only by light microscopy (microthrombi) are as prevalent as macrothrombi but tend to be distributed throughout all phases of ARDS (52). Microthrombi are of two types: hyaline platelet-fibrin thrombi in capillaries and arterioles and laminated fibrin clots in preacinar and large intra-acinar arteries (52, 55). Platelet fibrin thrombi confined to small arterioles and capillaries are most numerous in the acute phase of ARDS and are thought to represent localized

Figure 21 Enlarged, fibrous-walled airspaces are arranged in a honeycomb pattern (Movat pentachrome stain; original magnification×160). (From Ref. 84.) or disseminated intravascular coagulation (DIC). Microthrombi contribute to the reduction in peripheral arterial filling on postmortem arteriograms (Fig. 22) Larger clots may be embolic or formed in situ; morphologically it is impossible to tell the difference. In ARDS, it is likely that both mechanisms of clot deposition occur. Critically ill patients with indwelling vascular catheters are predisposed to pulmonary embolism (56). In a study of injured and burned patients, Eeles and Sevitt concluded that

Pulmonary pathology of the acute respiratory distress syndrome

83

pulmonary microthrombi and macrothrombi arose as emboli from systemic venous thrombi (55). Thromboembolism also has been considered to be a major initiating cause of ARDS, especially in patients with traumatic shock (57, 58). On the other hand, pulmonary endothelial injury can cause localized intrapulmonary or disseminated intravascular coagualtion (59, 60). Platelet sequestration has been shown to occur in ARDS, and an injurious role for platelets has been suggested (61, 62). Whether or not they are the primary triggering mechanisms, thromboemboli can contribute to lung injury at any stage, further reducing the pulmonary vascular bed and so causing lung necrosis through ischemia. Classic wedge-shaped hemorrhagic infarcts may be seen in ARDS, but frequently infarcts assume unusual patterns such as subpleural band-like or intermittent lobular necrosis (50, 63). Lung tissue that is adjacent to the visceral pleura is particularly susceptible to the development of ischemic necrosis because of reduced collateral blood flow (50). Postmortem studies using indocyanine green instilled into the airways indicate preferential ventilation of hypoperfused areas (50). These peripheral necrotic regions are thus susceptible to barotrauma and infection, resulting in lung cavitation, pneumothorax, or bronchopleural fistula (50, 64). Communicating pleural arcade vessels that fill from pulmonary artery to pulmonary artery frequently bridge peripheral hypoperfused areas, but the ability of these arcades to provide collateral blood flow is unknown (50, 63) (Fig. 22). In the proliferative and fibrotic stages of DAD, fibrocellular intimal proliferation is a response to endothelial injury in arteries, veins and lymphatics (52) (Fig. 23). Vascular lumens are narrowed by concentrically or eccentrically layered fibrin, proliferating myointimal cells, hyperplastic endothelial cells, and fibromyxoid connective tissue. Obstruction of venous and lymphatic channels potentially further increases intracapillary pressure, contributing to the accumulation of interstitial edema as well as impeding the removal of extra vascular fluid from the lung (52). In the late proliferative and fibrotic phases of ARDS, postmortem arteriograms show narrow preacinar arteries stretched about fibrous-walled cysts and dilated airspaces (52) (Fig. 22). Serpentine arterial branches have thickened fibromuscular walls (Fig. 20). Arterial tortuosity occurs as a result of distortion by irregularly contracting fibrous tissue. Dilated pulmonary capillaries permit the passage of barium-gelatin injection medium, which is normally restricted to vessels greater than 15 µm in diameter, into pulmonary veins. This produces a dense, ground-glass background haze in the postmortem arteriogram (52) (Fig. 22). Capillary proliferation has been ascribed to angiogenesis associated with granulation tissue formation or possibly to a direct effect of oxygen toxicity (65, 66). An increased concentration of blood vessels in patients in the late stages of ARDS probably reflects the combined effects of abnormally dilated vessels, crowding of vessels, and increased profiles of tortuous vessels rather than a restoration or regrowth of normal arteries (50, 52). Hypermuscularization of pulmonary arteries is frequently associated with pulmonary hypertension and has been demonstrated morphometrically in the intermediate and late stages of ARDS (52, 67, 68). With increasing duration of lung injury, the thickness of the media relative to the arterial diameter increases (Fig. 24). The percentage of muscular thickness of

Acute respiratory distress syndrome

84

Figure 22 Postmortem arteriographic patterns in ARDS, Visceral pleura is at bottom. (A) Normal adult lung. (B) Extensive reduction of filled peripheral arteries. Subpleural branches are stretched about dilated airspaces with “picket fence” appearance (16 days

Pulmonary pathology of the acute respiratory distress syndrome

after toxic inhalation). (C) Marked arterial tortuosity, increased background haze, and fine arteries stretched about honeycomb “cysts” (26 days after aspiration). (D) Intense background haze and venous filling (arrow) resulting from capillary dilation (55 days after viral pneumonia). In B, C, and D there are fine arcade vessels in the visceral pleura. (From Ref. 52.)

85

Acute respiratory distress syndrome

86

Pulmonary pathology of the acute respiratory distress syndrome

87

Figure 23 Severe fibrocellular intimal proliferation in a nonmuscular alveolar wall artery. Note cell in mitosis (arrow). The interstitium is widened by collagen, edema, and a mononuclear cell infiltrate. Hyperplastic epithelial cells partially line alveolar surface (toluidine blue; original magnification×400). (From Ref. 52.) preacinar and intraacinar arteries has been shown to be significantly lower in patients surviving fewer than 9 days than in those who survive at least 20 days. The variability in medial thickness of muscular arteries also increases with the duration of ARDS (Fig. 25). Increased medial thickness also correlates with the extent of parenchymal honeycombing and hemorrhage (67). Evidence of peripheral extension of smooth muscle into normally nonmuscularized arteries and arterioles is reflected in the decreased mean external diameter of fully and partially muscular arteries in patients in the later stages of ARDS (52, 68) (Fig. 26). Potential causes of arterial muscularization in ARDS include hypoxia, pulmonary hypertension, and oxygen toxicity (69–74).

VI. Localization of Lesions: Regional Alveolar Damage As originally defined, ARDS is considered to be a generalized lung process characterized roentgenographically by diffuse bilateral alveolar

Acute respiratory distress syndrome

88

Figure 24 Variation in the percentage of medial wall thickness of muscularized pulmonary arteries at different levels with duration of ARDS. The early (exudative phase) group differs significantly from the late fibrotic group at the alveolar duct (p<0.05) and respiratory bronchiolar levels (p<0.01). There is significant difference between the intermediate (proliferative phase) and late groups only at the respiratory bronchiolar level (p<0.05). (From Ref. 52.) infiltrates (19, 75). Generalized lung involvement is also implied in the histopathological term “diffuse alveolar damage” (1, 5, 6, 9, 26, 27). The concept that ARDS-DAD is always a diffuse bilateral process is not, however, consistent with clinical or morphological data. Murray et al. reviewing published studies of patients at risk for developing acute lung injury, concluded that there is a spectrum of clinically evident responses, with most patients developing mild to moderate, rather than severe, lung involvement (76). The diagnosis of ARDS within this continuum is arbitrarily based on the level of hypoxemia (76, 77). Pathologically, diffuse alveolar damage also represents a continuum in its extent of lung involvement. In patients who die of respiratory failure secondary to ARDS, the lung

Pulmonary pathology of the acute respiratory distress syndrome

89

is usually extensively affected, but focal areas may be inexplicably spared. Even in patients who meet the criterion of roentgenographically diffuse lung infiltrates, computed tomog-

Figure 25 Frequency distribution of the percentage of medial wall thickness of muscular pulmonary arteries for individual lungs, arranged according to increasing duration of ARDS. Representative patients have been selected to show the trend of changes. (From Ref. 52.) raphy (CT) scans usually show patchy lung involvement (63, 78, 79). Specific causes of acute lung injury, such as uremia, acid aspiration, radiation, and viral infection, can produce diffuse alveolar damage that remains localized within the lung. Exudative and proliferative alveolar damage may also occur as a localized secondary reaction adjacent to necrotizing lung lesions such as pneumonia or infarcts. Other recognized causes of

Acute respiratory distress syndrome

90

unilateral or localized noncardiogenic pulmonary edema such as rapid lung reexpansion after pneumothorax, thoracic sympathectomy, or pulmonary embolism have not been well studied pathologically but probably do not, in most instances, represent diffuse alveolar damage (80–84).

Figure 26 Variation in the external diameter of three structural types of intraacinar arteries with duration of ARDS (mean±SEM). For partialy muscular arteries, both early (exudative phase) and intermediate (proliferative phase) groups are significantly different from the late (fibrotic phase) group (p<0.05). For muscular arteries the early group differs significantly only from the late group (p<0.01). (From Ref. 52.) In a retrospective autopsy study, Yazdy et al. documented localized alveolar damage, which was not related to any of the above causative factors, in 11 % of all patients with diffuse alveolar damage seen over a 5-year period (7). The lesions of regional alveolar

Pulmonary pathology of the acute respiratory distress syndrome

91

damage tend to be multifocal, often with accentuation in the upper lobes (7) (Fig. 27). Localized hyaline membranes have also been previously described by Cederberg et al. and Sevitt in ventilator-dependent patients breathing moderately increased concentrations of oxygen (85, 86). The underlying risk factors associated with the localized lesion are identical to those of patients with classic ARDS, namely sepsis, shock, pancreatitis, mechanical ventilation, and hyperoxia (7, 85, 86). On the basis of the clinical and morphological similarities between diffuse alveolar damage and regional alveolar damage, it is reasonable to assume that they have a common pathogenesis. The factors responsible for the localization of alveolar injury in the lung are unknown but may relate to

Figure 27 Regional alveolar damage. (A) Regional alveolar damage is confined to right upper lobe that shows

Acute respiratory distress syndrome

92

a pale, diffusely consolidated area with fine airspace enlargement and a distinct border with uninvolved lung, in a 67-year-old woman with metastatic cervical carcinoma and hypotensive shock. Patient was mechanically ventilated with 50–70% inspired oxygen for one day. (B) Photomicrograph from the involved upper lobe showing hyaline membranes situated along the alveolar duct (H&E; original magnification×160). (C) Photomicrograph of lung parenchyma from uninvolved lower lobe showing only mild vascular congestion. Note absence of alveolar damage (H&E, original magnification×160). (From Ref. 7.)

Pulmonary pathology of the acute respiratory distress syndrome

93

the intensity and nature of the inciting stimulus. Sevitt considered localized alveolar damage to be a manifestation of oxygen toxicity (86). Focally distributed lesions may suggest a roentgenographic diagnosis of pneumonia or result in sampling error on open lung biopsy in patients with acute respiratory failure (7, 16, 79). The routine blind sampling of a predesignated site such as the lingula should be discouraged, even in patients with apparent diffuse infiltrates on chest roentgenogram.

Acute respiratory distress syndrome

94

Figure 28 Acute interstitial pheumonia. (Top) Diffuse interstitial remodeling. An organizing thrombus (arrow) occludes a muscular pulmonary artery. Perivascular lymphatics are dilated (arrowhead)

Pulmonary pathology of the acute respiratory distress syndrome

95

(H&E; original magnification×52). (Bottom) Alveolar septa are widened by immature connective tissue. Type 2 pneumocytes line the alveolar surfaces, and macrophages are present in airspaces (H&E; original magnification×206). VII. Idiopathic DAD/Acute Interstitial Pneumonia Acute interstitial pneumonia (AIP) is a rapidly progressive, often fatal form of interstitial pneumonia, which is histologically identical to the proliferative phase of diffuse alveolar damage (30, 87, 88). A histological study that compared cases of AIP to the cases of rapidly progressive interstitial fibrosis originally described by Hamman and Rich showed similar histological features, suggesting that AIP and Hamman-Rich syndrome are the same entity (89). AIP that occurs in the setting of longstanding interstitial fibrosis is considered to represent acute exacerbation of idiopathic pulmonary fibrosis (90). Since the cause of AIP is unknown, AIP has become synonymous with idiopathic DAD (87, 88). Although the majority of patients with AIP fulfill the clinical diagnostic criteria for ARDS, the duration of respiratory symptoms prior to catastrophic respiratory failure tends to be more protracted. On high-resolution CT (HRCT) scan, the features of AIP are similar to those of ARDS, i.e., bilateral ground glass opacities, airspace consolidation, and architectural distortion with traction bronchiectasis (8, 88). The histological features of AIP include diffuse widening of alveolar septa by reactive fibroblasts, modest chronic inflammation, fibromyxoid connective tissue, and type 2 pneumocyte hyperplasia (30) (Fig. 28). Focally distributed, or resorbing, hyaline membranes may be present but are usually not prominent. Organizing thrombi frequently reside in muscular pulmonary arteries and serve as an aid in the histological separation of AIP from other more chronic forms of interstitial pneumonia (30, 88). Unlike ARDS, acute interstitial pneumonia appears to more frequently progress to end-stage fibrosis and/or to recur (49, 87, 88).

VIII. Immunohistochemical Profile and Morphogenesis of Diffuse Alveolar Damage The expanded application of immunohistochemical markers to human lung tissue, coupled with enhanced understanding of biochemical mediators and principles of basic molecular biology, has resulted in new information and concepts of the pathogenesis of diffuse alveolar damage. Matrix metalloproteinases (MMP) and their inhibitors have been identified and implicated in tissue destruction in ARDS (91). Matrix metalloproteinase-2 (MMP-2) showed immunohistochemical co-localization in capillary endothelium and disrupted epithelial basement membrane (collagen type IV), suggesting a role for MMP-2 in collagenolysis (91). Other potential pathogenetic mediators immunohistochemically

Acute respiratory distress syndrome

96

identified in lung tissue in the fibroproliferative stage of diffuse alveolar damage include advanced glycation end products (AGE), heat shock protein 47 (HSP-47), and activin A protein (92–94). AGE are modified proteins that accumulate in aging extracellular matrix and presumably bind to receptor sites on macrophages. Receptors for AGE are thought to stimulate macrophages to release cytokines, which likely play a role in pulmonary fibrosis (93). HSP-47 is a collagen-specific molecular chaperone which is involved in the intracellular processing of the procollagen molecule (94). Overexpression of HSP-47, which has been localized in a smooth muscle-actin-positive interstitial cells, may also contribute to fibrosis in diffuse alveolar damage (94). Activin A, a member of the transforming growth factor (TGF)-β supergene family stimulates the proliferation of fibroblasts and plays a putative role in the induction and differentiation process in early embryogenesis. This potential mediator of lung fibrosis has been identified in metaplastic epithelium, hyperplastic smooth muscle cells, and alveolar macrophages in diffuse alveolar damage (92). AGE, HSP-47 and activin A have also been demonstrated in fibrotic lung diseases other than diffuse alveolar damage, suggesting that there is a common pathway of interstitial fibrosis in conditions of diverse etiology (92–94). An increasing array of cytokines and mediators of inflammation is thought to play an important role in the pathogenesis of ARDS (95, 96) (see Chaps. 7, 10, 12). Many cytokines have been identified in BAL fluid from patients with ARDS; however, there are relatively few studies in which cytokines and other mediators have been localized within lung tissue. Epithelial neutrophil-activating peptide-78 (ENA-78), a signaling molecule for neutrophils, has been identified by immunohistochemical methods in endothelial cells, hyperplastic type 2 pneumocytes, bronchiolar epithelial cells, inflammatory cells, and vascular smooth muscle cells in lung tissue from patients with ARDS due to sepsis (97, 98). IL-8, on the other hand, was observed only in lung inflammatory cells (98). In preliminary studies, Matsubara and colleagues have demonstrated, in pulmonary smooth muscle cells, fibroblasts, and alveolar macrophages, the presence of vascular endothelial growth factor, which increased with the progression of fibrosis in diffuse alveolar damage (99). Thrombospondin-1, an inhibitor of angiogenesis, was seen in all stages of DAD, with the greatest intensity in advanced pulmonary fibrosis (99). The role of apoptosis (programmed cell death) as a molecular basis for the cellular transition that occurs in diffuse alveolar damage was studied by Guinee and colleagues (100). Apoptosis of type 2 pneumocytes was shown to occur in all cases of diffuse alveolar damage examined, whereas apoptosis of interstitial myofibroblasts occurred in approximately 50% of cases (100). P53, a proliferation marker that plays a critical role in apoptosis and G1 cell cycle arrest, and WAF1 (p27), a downstream effector of p53, have also been consistently identified in epithelial cells in diffuse alveolar damage (100). P53, however, was not found and WAF1 showed a much reduced activity in normal control lungs (100). BAX protein, which is induced by p53, and is a promoter of apoptosis, is downregulated through binding with BCL-2. In lung specimens exhibiting diffuse alveolar damage, Guinee et al. also demonstrated BAX to be consistently increased in type 2 pneumocytes and interstitial cells, whereas BCL-2 was identified solely in interstitial cells in only 28% of cases (101). By contrast, in control lungs, BAX is expressed much less intensely in type 2 pneumocytes and is not found in interstitial cells, whereas BCL-2 is only identified in bronchiolar epithelium (101). These results suggest

Pulmonary pathology of the acute respiratory distress syndrome

97

that the induction of BAX in diffuse alveolar damage may enhance the susceptibility of alveolar epithelial cells to apoptosis, while BCL-2 expression contributes to the absence of apoptosis and hence to the promotion of unregulated interstitial cell proliferation (and fibrosis) in some cases (101). Adamson and collegues evaluated c-myc, cyclin-D1, and DNA topoisomerase IIa, markers of cellular proliferation, in lung biopsy specimens showing DAD (102). The DNA topoisomerase IIa proliferation index was significantly higher in DAD than in control lungs and was higher in patients who died of respiratory failure than in those who survived. Expression of c-myc and cyclin-D1 were also enhanced, suggesting that dysregulated cell proliferation plays an important role in the evolution of DAD (102). The complex role of oncogenes, apoptosis, and cellular proliferation in the initiation and evolution of diffuse alveolar damage is an exciting area for future research.

IX. Clinicopathological Correlations The structural alterations serve as a basis for understanding the major clinical and functional manifestations of ARDS. Hypoxemia results from shunting caused by a maldistribution of ventilation and perfusion. Intra-alveolar exudate, fibrous tissue, and microatelectasis all contribute to shunting (19). Lamy et al. observed that patients with a fixed shunt at all fractions of inspired oxygen demonstrated consolidative edema, exudation, and hemorrhage on open lung biopsy, whereas lesser degrees of exudate were seen in patients with high arterial oxygen tensions and improved oxygenation with PEEP (27). Patients with extensive fibrosis on open lung biopsy demonstrated a very slow improvement in oxygenation with increased levels of PEEP and tended to have poorer survival rates (27). The wide range of obliterative pulmonary vascular lesions contribute to pulmonary hypertension and account for an increased physiological dead space due to ventilation of poorly perfused regions of the lung (50, 52). Hypoxia reacts upon the vasculature, causing vasoconstriction and accentuated vascular resistance. In the late stages of ARDS, lung compliance is reduced by extensive interstitial fibrosis, whereas in the early stages compliance is lowered by interstitial edema and inflammatory exudate. The broad widening of alveolar septa by interstitial edema, the separation of the endothelial from the epithelial basement membrane, and the greatly thickened alveolar epithelium composed of type 2 pneumocytes also suggests that impaired diffusion of oxygen may be an important factor limiting respiration (16).

Acknowledgments The author would like to thank Diane Gillihan for expert secretarial assistance, Don Resch for photography services, and the staff of the Brittingham Memorial Library for help with reference material.

Acute respiratory distress syndrome

98

References 1. Katzenstein AA. Bloor CM, Liebow AA. Diffuse alveolar damage—the role of oxygen, shock and related factors. Am J Pathol 1976; 85:210–222. 2. Nash G, Blennerhassett JB, Pontoppidan H: Pulmonary lesions associated with oxygen theraphy and artificial ventilation. N Engl J Med 1967; 276:368–374. 3. Pratt PC, Vollmer RT, Shelburne JD, Crapo JD: Pulmonary morphology in a multihospital collaborative extracorporcal membrane oxygenation project. Light microscopy. Am J Pathol 1979; 95:191–208. 4. Tomashefski JF, Jr: Pulmonary Pathology of Acute Respiratory Distress Syndrome. Clin Chest Med 2000; 21:435–466. 5. Martin AM, Saloway HB, Simmons RL: Pathologic anatomy of the lungs following shock and trauma. J Trauma 1968; 8:687–698. 6. Teplitz C: The core pathobiology and integrated medical science of adult acute respiratory insufficiency. Surg Clin North Am 1976; 56:1091–1133. 7. Yazdy AM, Tomashefski JF Jr, Yagan R, Kleinerman J: Regional alveolar damage (RAD), a localized counterpart of diffuse alveolar damage. Am J Clin Pathol 1989; 92:10–15. 8. Webb WR, Muller NL, Naidich DP: High Resolution CT of the Lung, 3rd ed. Philadelphia, Lippincott, Williams & Wilkins, 2001, pp. 355–420. 9. Orell SR: Lung pathology in respiratory distress following shock in the adult. Acta Pathol Microbiol Scand 1971; 79:65–76. 10. Pratt PC: Pathology of the adult respiratory distress syndrome. In: Thurlbeck WM, Abell MR, eds. The Lung—Structure, Function and Disease. Baltimore, Willimas and Wilkins Co., 1978, pp. 44–57. 11. Fukuda Y, Ishikazi M, Masuda Y, Kimura G, Kawanami O, Masugi Y: The role of intraalveolar fibrosis in the process of pulmonary structural remodeling in patients with diffuse alveolar damage. Am J Pathol 1987; 126:171–182. 12. Kobashi Y, Manabe T: The fibrosing process in so-called organized diffuse alveolar damage. An immunohistochemical study of the change from hyaline membrane to membranous fibrosis. Virch Archiv 1993; 422:47–52. 13. Staub NC: The pathophysiology of pulmonary edema. Hum Pathol 1970; 1:419–432. 14. Pison U, Seeger W, Buchhorn R, Theo J, Brand M, Obertake U, Neuhof H, Schmit-Neuerburg K: Surfactant abnormalities in patients with respiratory failure after multiple trauma. Am Rev Respir Dis 1989; 140:1033–1039. 15. Albertine KH: Ultrastructural abnormalities in increased—permeability pulmonary edema. Clin Chest Med 1985; 6:345–366. 16. Bachofen M, Weibel ER: Alterations of the gas exchange apparatus in adult respiratory insufficiency associated with septicemia. Am Rev Respir Dis 1977; 116:589–615. 17. Riede UN, Joachim H, Hassenstein J, Costabel U, Sandritter W, Augustin P, Mittermayer C: The pulmonary air blood barrier of human shock lungs (a clinical, ultrastructural and morphometric study). Pathol Res Pract 1978; 162:41–72. 18. Schnells G, Voigt WH, Redl H, Schlag G, Glatzl A: Electron microscopic investigation of lung biopsies in patients with post traumatic respiratory insufficiency. Acta Chir Scand (Suppl) 1980; 499:9–20. 19. Balk R, Bone RC: The adult respiratory distress syndrome. Med Clin North Am 1983; 67:685– 700. 20. Wells S, Sissons M, Hasleton PS: Quantitation of pulmonary megakaryocytes and fibrin thrombi in patients dying from burns. Histopathology 1984; 8:517–527. 21. Corrin B: Lung pathology in septic shock. J Clin Pathol 1980; 33:891–894.

Pulmonary pathology of the acute respiratory distress syndrome

99

22. Nash G, Foley FD, Langlinais PL: Pulmonary interstitial edema and hyaline membranes in adult burn patients. Electron microscopic observations. Hum Pathol 1974; 5:149–160. 23. Adamson IYR, Bowden DH: The type 2 cells as progenitor of alveolar eptihelial regeneration: a cytodynamic study in mice after exposure to oxygen. Lab Invest 1974; 30:35–42. 24. Marshall RP, Bellingan G, Webb S, Puddicombe A, Goldsock N, McAnulty RJ, Laurent GJ: Fibroproliferation occurs early in the acute respiratory distress syndrome and impacts on outcome. Am J Respir Crit Care Med 2000, 162:1783–1788. 25. Pugin J, Verghese G, Widmer M-C, Matthay M: The alveolar space is the site of intense inflammatory and profibrotic reactions in the early phase of acute respiratory distress syndrome. Crit Care Med 1999; 27:304–312. 26. Hill JD, Ratliff JL, Parrott JCW, Zlamy M, Fallat RJ, Koeniger E, Yaeger EM, Whitmer G: Pulmonary pathology in acute respiratory insufficiency: lung biopsy as a diagnostic tool. J Thorac CV Surg 1976; 71:64–69. 27. Lamy M, Fallat RJ, Koeniger E, Jarm-Peter D, Ratliff JL, Eberhart RC, Tucker HJ, Hill JD: Pathologic features and mechanisms of hypoxemia in adult respiratory distress syndrome. Am Rev Respir Dis 1976; 114:267–284. 28. Matsubara O, Tamura A, Ohdama S, Mark EJ: Alveolar basement membrane breaks down in diffuse alveolar damage: An immunohistochemical study. Pathol Int 1995; 45:473–482. 29. Sugiyama K, Kawai T: Diffuse alveolar damage and acute interstitial pneumonitis: histochemical evaluation with lectins and monoclonal antibodies against surfactant apoprotein and collagen type IV. Modern Pathol 1993; 6:242–248. 30. Katzenstein AA, Myers JL, Mazur MT: Acute interstitial pneumonia. A clinicopathologic, ultrastructural, and cell kinetic study. Am J Surg Pathol 1986; 10:256–267. 31. Nash JRG, McLaughlin PJ, Hoyle C, Roberts D: Immunolocalization of tumour necrosis factor a in lung tissue from patients dying with adult respiratory distress syndrome. Histopathology 1991; 19:395–402. 32. Pan L, Ohtani H, Yamauchi K, Nagura H: Co-expression of TNF-α and IL-1β. in human acute pulmonary fibrotic diseases: an immunohistochemical analysis. Pathol Int 1996; 46:91–99. 33. Warnock ML, Press M, Churg A: Further observations on cytoplasmic hyaline in the lung. Hum Pathol 1980; 11:49–65. 34. Stanley MW, Henry-Stanley MJ, Gajl-Peczalska K, Bitterman P: Hyperplasia of type II pneumocytes in acute lung injury. Cytologic findings of sequential bronchoalveolar lavage. Am J Clin Pathol 1992; 97:669–677. 35. Pache J, Christakos PG, Gannon DE, Mitchell JJ, Low RB, Leslie KO: Myofibroblasts in diffuse alveolar damage of the lung. Mod Pathol 1998; 11:1064–1070. 36. Hallgren R, Samuelsson T, Laurent TC, Modig J: Accumulation of hyaluronan (hyaluronic acid) in the lung in adult respiratory distress syndrome. Am Rev Respir Dis 1989; 139:682–687. 37. Basset F, Ferrans VJ, Soler P, Takemura T, Fukuda Y, Crystal R: Intraluminal fibrosis in interstitial lung disorders. Am J Pathol 1986; 122:443–461. 38. Burkhardt A: Alvelolitis and collapse in the pathogenesis of pulmonary fibrosis. Am Rev Respir Dis 1989; 140:513–524. 39. Epler GR, Colby TV: The spectrum of bronchiolitis obliterans. Chest 1983; 83:161–162. 40. Katzenstein AA: Pathogenesis of “fibrosis” in interstitial pneumonia: an electron microscopic study. Hum Pathol; 16:1015–1024. 41. Slavin G, Nunn JF, Crow J, Dore CJ: Bronchiolectasis—a complication of artificial ventilation. Br Med J 1982; 285:931–934. 42. Liebler JM, Qu Z, Buckner B, Powers MR, Rosenbaum JT: Fibroproliferation and mast cells in the acute respiratory distress syndrome. Thorax 1998; 53:823–829. 43. Churg A, Golden J, Fligial S, Hogg J: Bronchopulmonary dysplasia in the adult. Am Rev Respir Dis 127:117–120, 1983. 44. Zapol WM, Trelstad RL, Coffey JW, Tsai I, Salvador RA: Pulmonary fibrosis in severe acute respiratory failure. Am Rev Respir Dis 1979; 119: 547–554.

Acute respiratory distress syndrome

100

45. Hert R, Albert RK. Sequelae of the adult respiratory distress syndrome. Thorax 1994; 49:8–13. 46. Lakshminarayan S, Stanford RE, Petty TL: Prognosis after recovery from adult respiratory distress syndrome. Am Rev Respir Dis 1976; 113:7–16. 47. Suchyta R, Elliott CG, Colby T, Rasmussson BY, Morris AH, Jensen RL: Open lung biopsy does not correlate with pulmonary function after the adult respiratory distress syndrome. Chest 1991; 99:1232–1237. 48. Pratt PC: Pathology of adult respiratory distress syndrome: implications regarding therapy. Semin Resp Med 1982; 4:79. 49. Savici D, Katzenstein AA: Diffuse alveolar damage and recurrent respiratory failure: report of 6 cases. Hum Pathol 2001; 32:1398–1402. 50. Jones R, Reid LM, Zapol WM, Tomashefski JF Jr, Kirton OC, Kobayashi K: Pulmonary vascular pathology, human and experimental studies. In Zapol WM and Falke KJ (eds). Acute Respiratory Failure. New York, Marcel Dekker, Inc., 1985, pp. 23–160. 51. Zapol WM, Snider MT: Pulmonary hypertension in severe acute respiratory failure. N Engl J Med 1977; 296:476–480. 52. Tomashefski JF Jr, Davies P, Boggis L, Greene R, Zapol WM, Reid LM: The pulmonary vascular lesions of the adult respiratory distress syndrome. Am J Pathol 1983; 112:112–126. 53. Zapol WM, Kobayashi K, Snider MT, Greene R, Laver MB: Vascular obstruction causes pulmonary hypertension in severe acute respiratory failure. Chest 1977; 71:306S–307S. 54. Greene R, Zapol WM, Snider MT, Reid L, Snow R, O’Connell RS, Novelline RA: Early bedside detection of pulmonary vascular occlusion during acute respiratory failure. Am Rev Respir Dis 1981; 124:593–601. 55. Eeles GH, Sevitt S: Microthrombosis in injured and burned patients. J Pathol Bacteriol 1967; 93:275–293. 56. Moser KM: Pulmonary embolism. Am Rev Respir Dis 1977; 115:829–852. 57. Blaisdell FW, Lim RC, Stallone RJ: The mechanism of pulmonary damage following traumatic shock. Surg Gynecol Obstet 1986; 130:15–22. 58. Saldeen T: The microembolism syndrome. Microvasc Res 1976; 11:227–259. 59. Bone RC, Francis PB, Pierce AK: Intravascular coagulation associated with the adult respiratory distress syndrome. Am J Med 1976; 61:585–589. 60. Carvalho AC: Blood alterations in ARDS. In Zapol WM and Falke KJ (eds). Acute Respiratory Failure. New York, Marcel Dekker, Inc., 1985, pp.303–346. 61. Heffner JE, Sahn SA, Repine JE: The role of platelets in the adult respiratory distress syndrome. Culprits or bystanders? Am Rev Respir Dis 1987; 135:482–492. 62. Schneider RC, Zapol WM, Carvalho AC: Platelet consumption and sequestration in severe acute respiratory failure. Am Rev Respir Dis 1980; 122:445–451. 63. Greene R: Adult respiratory distress syndrome: acute alveolar damage. Radiology 1987; 163:57–66. 64. Redline S, Tomashefski JF Jr, Altose MD: Cavitating lung infarction after bland pulmonary thromboembolism in patients with the adult respiratory distress syndrome. Thorax 1985; 40:915–919. 65. Pratt PC: Pulmonary capillary proliferation induced by oxygen inhalation. Am J Pathol 1958; 34:1033–1049. 66. Kirton OL, Jones R, Zapol WM, Reid L: The development of a model of subacute lung injury after intra-abdominal infection. Surgery 1984, 96:384–92. 67. Earth PJ, Knoch M, Muller E, Sangmeister C, Bittinger A. Morphometry of parenchymal and vascular alterations in ARDS after extracorporeal carbon dioxide removal therapy (ECCO2-Rtherapy)). Path Res Pract 1992; 188:653–656. 68. Snow RL, Davies P, Pontoppidan H, Zapol WM, Reid LM: Pulmonary vascular remodeling in adult respiratory distress syndrome. Am Rev Respir Dis 1982; 126:887–892. 69. Hislop A, Reid L: New findings in pulmonary arteries of rats with hypoxiainduced pulmonary hypertension. Br J Exp Pathol 1976; 57:542–553.

Pulmonary pathology of the acute respiratory distress syndrome

101

70. Meyrick B, Reid L: The effect of continued hypoxia on rat pulmonary artery circulation: an ultrastructural study. Lab Invest 1978; 38:188–200. 71. Naeye RL: Hypoxemia and pulmonary hypertension. Arch Pathol Lab Med 1961; 71:447–452. 72. Jones R, Zapol WM, Reid L: Pulmonary artery remodelling and pulmonary hypertension after exposure to hyperoxia for 7 days: a morphometric and hemodynamic study. Am J Pathol 1984; 117:273–285. 73. Jones R, Zapol WM, Reid L: Oxygen toxicity and restructuring of pulmonary arteries—a morphometric study. The response to 4 weeks’ exposure to hyperoxia and return to breathing air. Am J Pathol 1985; 121:212–223. 74. Hislop A, Haworth SG, Shinebourne EA, Reid L: Quantitative structural analysis of pulmonary vessels in isolated ventricular septal defect in infancy. Br Heart J 1975; 37:1014–1021. 75. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE: Acute respiratory distress in adults. Lancet 1967; 2:319–323. 76. Murray JF, Matthay MA, Luce JM, Flick MR: An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1988; 138:720–723. 77. Bernard GR, Artigas A, Brigham KL: The American-European consensus conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 818–24. 78. Gattinoni L, Mascheroni D, Torresin A, Marcolin R, Fumagalli R, Vesconi S, Rossi GP, Rossi F, Baglioni F, Bassi F, Nastri G, Pesenti A: Morphological response to positive end expiratory pressure in acute respiratory failure. Computerized tomography study. Intensive Care Med 1986; 12:137–142. 79. Maunder RJ, Shuman WP, McHugh JW, Marglin SI, Butler J: Preservation of normal lung regions in the adult respiratory distress syndrome. JAMA 1986; 255:2463–2465. 80. Flick MR, Kantzler GB, Block AJ: Unilateral pulmonary edema with contralateral thoracic sympathectomy in the adult respiratory distress syndrome. Chest 1975; 68:736–739. 81. Hyers TM, Fowler AA, Wicks AB: Focal pulmonary edema after massive pulmonary embolism. Am Rev Respir Dis 1981; 123:232–233. 82. Ketai LH, Godwin JD: A new view of pulmonary edema and acute respiratory distress syndrome. J Thorac Imag 1998; 13:147–171. 83. Sautter RD, Dreher WH, MacIndoe JH, Myers WD, Magnin JE: Fatal pulmonary edema and pneumonitis after re-expansion of chronic pneumothorax. Chest 1971; 60:399–401. 84. Tomashefski JF Jr: Noncardiogenic pulmonary edema and diffuse alveolar damage. In MJ Saldana (ed), Pathology of Pulmonary Disease. Philadephia, JB Lippincott Co., 1994, pp. 125– 138. 85. Cederberg A, Hellsten S, Miorner G: Oxygen treatment and hyaline pulmonary membranes in adults. Acta Pathol Microbiol Scand 1965; 64:450–458. 86. Sevitt S: Diffuse and focal oxygen pneumonitis, a preliminary report on the threshold of oxygen toxicity in man. J Clin Pathol 1974; 27:21–30. 87. Vourlekis JS, Brown KK, Cool CD, Young DA, Cherniak RM, King TE Jr, Schwarz MI: Acute interstitial pneumonitis. Case series and review of the literature. Medicine 2000; 79:369–378. 88. Travis WD, King TE, Jr: American Thoracic Society/European Respiratory Society International Multidisciplinary Consensus Classification for the Idiopathic Pneumonias. Am J Respir Crit Care Med 2002; 165:277–304. 89. Olson J, Colby TV, Elliott CG: Hamman-Rich syndrome revisited. Mayo Clin Proc 1990; 65:1538–1548. 90. Kondoh Y, Taniguhi H, Kawabata Y, Yokoi T, Suzuki K, Takagi K: Acute exacerbation in idiopathic pulmonary fibrosis. Analysis of clinical and pathologic findings in three cases. Chest 1993; 103:1808–1812. 91. Hayashi T, Stettler-Stevenson WG, Fleming MV: Immunohistochemical study of metalloproteinases and their tissue inhibitors in the lungs of patients with diffuse alveolar damage and idiopathic pulmonary fibrosis. Am J Pathol 1996; 149:1241–1256.

Acute respiratory distress syndrome

102

92. Matsuse T, Ikegami A, Ohga EI: Expression of immunoreactive activin A protein in remodeling lesions associated with interstitial pulmonary fibrosis. Am J Pathol 1996; 148:707–713. 93. Matsuse T, Ohga E, Teramoto S: Immunohistochemical localization of advance glycation end products in pulmonary fibrosis. J Clin Pathol 1998; 51:515–519. 94. Razzaque MS, Nazreen A, Taguchi T: Immunolocalizaton of collagen and collagen-binding heat shock protein 47 in fibrotic lung disease. Mod Pathol 1998; 11:1183–1188. 95. Martin TR: Lung cytokines and ARDS. Chest 1999; 116:2S-8S. 96. Hasleton PS, Roberts TE: Adult Respiratory Distress Syndrome—an update. Histopathology 1999; 34:285–294. 97. Zimmerman GA, Albertine KH, Carveth HJ, Gill EA, Grissom CK, Hoidal JR, Imaizumi T, Maloney CG, McIntyre TM, Michael JR, Orme JF, Prescott SM, Tophan MS: Endothelial activation in ARDS. Chest 1999; 116:18S-24S. 98. Albertine KH: Histopathology of pulmonary edema and the acute respiratory distress syndrome, In, MA Matthay and DH Ingbar (eds). Pulmonary Edema. New York: Marcel Dekker, 1998, pp. 37–83. 99. Matsubara O, Imazeki N, Okochi Y, Ozeki Y, Mark EJ: Vascular endothelial growth factor and thrombospondin-1 in diffuse alveolar damage and advanced pulmonary fibrosis. Mod Pathol 2002; 15:324A. 100. Guinee DG Jr, Fleming M, Hayashi T, Woodward M, Zhang J, Walls J, Koss M, Ferrans V, Travis W: Association of p53 and WAF1 expression with apoptosis in diffuse alveolar damage. Am J Pathol 1996; 149:531–538. 101. Guinee DG Jr, Brambilla E, Fleming M, Hayashi T, Rahn M, Koss M, Ferrans V, Travis W: The potential role of BAX and BCL-2 expression in diffuse alveolar damage. Am J Pathol 1997; 151:999–1007. 102. Adamson A, Perkins S, Brambilla E, Tripp S, Holden V, Travis W, Guinee D Jr: Proliferation, C-myc, and cyclin D-1 expression in diffuse alveolar damage: potential roles in pathogenesis and implications for prognosis. Hum Pathol 1999; 30:1050–1057. 103. Tomashefski JF, Jr. Pulmonary pathology of the adult respiratory distress syndrome. Clin Chest Med 1990; 11:593–619.

6 Pathogenesis of Acute Lung Injury Experimental Studies NICHOLAS DAVID MANZO Massachusetts General Hospital Boston, Massachusetts, U.S.A. AARON B.WAXMAN Harvard Medical School and Massachusetts General Hospital Boston, Massachusetts, U.S.A.

I. Introduction Acute lung injury (ALI) and the acute respiratory distress syndrome (ARDS) are devastating illnesses. ARDS is the severe endpoint of acute lung injury and is defined as a syndrome of inflammation and increased permeability in the absence of left heart failure. ARDS is commonly associated with trauma, sepsis, pneumonia, and shock. Anything that can diffusely injury the lung can lead to ARDS. Regardless of the etiology, the pathophysiology and clinical manifestations are the same and can be divided into three phases: the acute or oxidative phase, the proliferative phase, and the fibrotic or reparative phase. While the immediate processes leading to ALI and ARDS are not known, direct injury could occur either to lung endothelium and/or epithelium and trigger an inflammatory response. Alternatively, the inflammatory cascade, if activated systemically, could itself indirectly cause endothelial or epithelial injury through reactive oxidant species, proteases, and cytokines. The result is widespread capillary leak and exudation of protein- and fibrin-rich fluids into the interstitium and the alveolar space. Leakage of protein leads to activation of coagulant pathways and influences a secondary release of growth factors and profibrotic mediators, presumably leading to healing and repair, but occasionally contributing to ongoing inflammation and damage to the lung. This chapter will summarize the current understanding of the contribution and regulation of the different mediators, both biochemical and cellular, that are believed important to the pathogenesis of acute lung injury.

II. Cytokines and Chemokines Any injury to the body, whether pulmonary or extrapulmonary, can lead to the activation and promotion of pulmonary inflammation that is mediated through the actions of signaling molecules or cytokines. Produced from both inflammatory and

Acute respiratory distress syndrome

104

noninflammatory cells, cytokines are extracellular soluble peptides that elicit their biological activity though the binding of cell surface receptors. Cytokines act in both a paracrine and autocrine manner, with the ability to exert their affects at concentrations as low as 10−15 M. Human studies interpreting the role of these cytokines are complicated by the methods of recovering samples as well as countless other constituents within the samples that test the limits of the bioassays used. In recovering alveolar fluid through bronchoalveolar lavage, the samples are diluted with saline (1). In addition, the presence of inhibitors and neutralizing agents in bronchoalveolar lavage (BAL) fluid make functional assays difficult (see Chap. 7). While many inflammatory cytokines have been measured and implicated in the pathogenesis of ALI/ARDS, the early response cytokines tumor necrosis factor alpha (TNF-α) and interleukin-1 beta (IL-1β) are considered the key initial inducers of the inflammatory response. In animal models the production and elaboration of these cytokines is almost immediately increased within the lung after exposure to direct or indirect lung injury (2). Both cytokines are produced primarily by mononuclear phagocytes and elicit similar biological responses despite being biochemically unrelated. In response to lipopolysaccharide (LPS) treatment, both cytokines are elaborated from macrophages. Conversely, β-adrenergic agonists, through increasing intracellular cAMP levels, can attenuate the increased TNF-α production (3, 4). These cytokines act synergistically on the endothelium to induce the upregulation of adhesion molecules such as intracellular adhesion molecule 1 (ICAM-1), P-selectin, and E-selectin (5–7), and promote a pro-coagulant surface through an increase in tissue factor (8), plasminogen activator inhibitor (9, 10), the reduction of surface thrombomodulin (11–13), and the increase production of chemokines such as IL-8 and monocyte chemotactic peptide 1 (MCP-1) (14, 15). The increased chemokine production shows chemotactic preference to neutrophils, aiding their release from bone marrow, as well as priming them for enhanced cytotoxic effects (16–19). Intra-tracheal infusion of IL-1 β or TNF-α increases the recruitment of neutrophils to the lung vasculature and airspaces (20). Interestingly, these pro-inflammatory cytokines have been linked to repair after injury by exerting chemotatotic and mitogenic effects on lung fibroblasts and epithelial cells (21). This exaggerated cytokine response may be a result of increased gene exression (Fig. 1). The signals that lead to increased gene transcription are closely regulated by DNAbinding proteins, such as nuclear factor κB (NFκB), that when activated are required for maximal expression of many cytokines, chemokines, adhesion molecules, and other potent products involved in the pathogenesis of ALI/ARDS. Under normal conditions the binding of inhibitory κB (IκB) proteins neutralizes the activity and mobility to the nucleus of NFκB. Upon activation, IκB is degraded, thus freeing NFκB and allowing it to enter the nucleus, where it binds to specific regions of genes and enhances their transcription. NFκB likely has a pathogenic role in the development of these proinflammatory cytokines. Several studies, of

Pathogenesis of acute lung injury

105

Figure 1 Generalized cellular and biochemical events that occur in the regulation of cytokines and chemokines active in ALI/ARDS. Activation stimuli (direct or indirect injury) result in the increased expression and production of various cytokines and chemokines that are regulated in part by transcription factors, kinases, secondary messengers, and oxygen radicals. both humans and in animal models, have suggested that the production of TNF-α and IL1β is dependent on the presence and activation of NFκB. In patients with established ARDS the presence of activated NFκB was detected in lung macrophages (22). Murine models using cecal ligation and puncture to cause lung injury showed an initial rapid increase in TNF-α mRNA. Activation of NFκB occurred shortly thereafter. Interestingly the upregulation of TNF-α preceded NFκB activation, yet there was no correlation between cytokine or NF-κB levels (23, 181). Similar in vivo models using hemorrhageinduced lung injury investigated the activation of multiple transcription factors (NFκB, CREB, C/EBP, AP-1, Sp1) that may be activated through the course of ALI/ARDS and found that only NFκB and cyclic AMP response- element-binding proteins (CREB) were activated. While the activation of NFκB is dependent on a mechanism that is yet unknown, it appears that the activation of CREB is dependent on xanthine oxidase (XO) (24, 25, 181). Expression of XO, a potent oxidant that has been linked to the generation

Acute respiratory distress syndrome

106

of reactive oxygen species and the pathogenesis of ALI/ARDS, can be upregulated both at the transcriptional and post-transcriptional level by cytokines, LPS, and hypoxia (26, 27, 183). Recent studies have shown that the reduced cytokine protein expression that is achieved by the stress response is due to the reduction of NFκB and the stabilization and increased expression of IκB. The abundance and stabilization of IκB impedes the translocation of NFκB to the nucleus, thus potentially blocking the gene regulatory effects of NFκB, specifically those effects on pro-inflammatory gene expression (28, 181, 184). The activation of NFκB as well as TNF-α and IL-1β mRNA by hemorrhage or endotoxemia-induced lung injury can be reduced in chemically neutropenic mice (29). The inhibition of NFκB can be achieved by a number of different mechanisms, including secretory leukocyte protease inhibitor (SLPI). As its name implies, SLPI is a serine protease inhibitor that is produced by macrophages, neutrophils, and epithelial cells and functions to protect cells from proteolytic degradation by leukocytes. When added exogenously, SLPI can greatly reduce the activity of NFκB by increasing the activity of IκB and attenuating lung injury in an IgG immune complex model of lung injury (30). Exogenous administration of IL-11 to rats showed a significant reduction in many of the inflammatory markers of an IgG immune complex model of lung injury. The reduced neutrophil influx and vascular permeability were shown to be associated with reduced lung activation of NFκB as well as reduced BAL levels of TNF-α (31). The activation and production of the early response cytokines by NFκB act to increase a number of downstream inflammatory products. Interestingly, the liberation of NFκB from its inhibitory IκB can be promoted by stimulation with the early response cytokines such as TNF-α. Thus the activation and production by NFκB acts in an autocrine manner serving to further sustain the production of these cytokines as well as promoting the inflammatory process. While TNF-α and IL-β are important in the pathogenesis of ALI/ ARDS, many other cytokines participate in modulating pulmonary inflammation. Interleukin-6, a product of TNF-α and IL-β-stimulated alveolar macrophages, is found in high concentrations in the BAL fluid of ALI/ ARDS patients and serves to stimulate acute phase reactants (32). Soluble IL-6 receptor is also increased in the BAL fluid of ALI/ARDS patients and functions to promote IL-6 signaling when bound to the gp130 receptor that is ubiquitously present on cell membranes (32). In vivo models using transgenic mice overexpressing transforming growth factor a (TGF-α) showed a significant reduction in IL-6 production associated with protection against polytetrafluoroethylene (PTFE) fumeinduced acute lung injury (33). Treatment with a β-adrenergic agonist showed similar regulation of IL-6 as that of TGF-α, with both the expression and production of IL-6 being reduced despite the presence of endotoxin (34). While the β-adrenergic regulation of TNF-α was due to an increase in cAMP levels, its regulation of IL-6 has not been linked to cAMP levels. Other mechanisms of IL-6 regulation have been shown to center around the signaling cascade leading to a cellular response. IL-6 signaling is mediated, in part, by the mitogen-activated protein kinase (MAPK) signaling pathway, specifically through the phosphorylation of extracellular signal-regulated kinase 1/2 (Erk1/2) and p38. Inhibition of these pathways results in reduced expression of IL-6 mRNA and decreased production of IL-6 and other cytokines (35, 181). Studies of the transcriptional regulation of IL-6 revealed that early activation of NFκB and NF-IL6 correlated with the upregulation of IL-6 mRNA in response to sepsis (23, 181). Several other cytokines have

Pathogenesis of acute lung injury

107

been shown to be active in enhancing and prolonging the inflammation in lungs of patients with ALI/ARDS, such as granulocyte colony stimulating factor (G-CSF) and granulocyte macrophage colony stimulating factor (GM-CSF) (36), macrophage inhibitory factor (MIF) (32, 37–39), and TGF-β (40), while the mechanisms for their activation and regulation are still unclear. While the majority of cytokines found to participate in the pathogenesis of ALI/ARDS are involved in establishing and sustaining pulmonary inflammation, several antiinflammatory cytokines are present and serve to counterregulate pulmonary inflammation, but are expressed at levels so low that their effects are greatly outweighed. IL-10, an anti-inflammatory cytokine, has shown the greatest ability to suppress proinflammatory cytokine production by alveolar macrophages (41, 42). Patients at risk of ARDS have increased BAL levels of IL-10 but with no predictive value for the development or protection against ARDS (43). In comparison patients with established ARDS the levels of IL-10 are relatively low and are associated with poor prognosis (44). The ratio of TNF-α to IL-10 protein in the BAL of ARDS patients and patients at risk for ARDS was 3.52 and 0.85, respectively, suggesting an important relationship between pro-inflammatory and anti-inflammatory cytokines in ARDS patients (45, 46). Animal models, on the other hand, have shown IL-10 to be an antagonist of many proinflammatory mediators. Rabbits with acute necrotizing pancreatitis as a model of lung injury showed a dramatic reversal of inflammation, characterized by a reduction in TNFα, IL-8, adhesion molecules, vascular permeability, and neutrophil recruitment and margination, when treated with the IL-10 agonist IT-9302 (47). While the regulatory mechanisms of IL-10 activation or its suppression in ALI/ARDS are unclear, the mechanism by which IL-10 may operate in reducing pro-inflammatory cytokine expression is possibly due to the destabilization of the mRNA as well as the inhibition of NFκB (48, 184). Other candidate protective cytokines include many of the IL-6-type cytokines. IL-11, cardiotrophin-1 (CT-1) (49), and IL-6 all possess similar effector profiles and share the gp130 receptor subunit. The production of transgenic mice with targeted lung overexpression IL-11 and IL-6 showed a significant survival benefit when exposed to hyperoxia, characterized by reduction of pulmonary edema fluid, cell injury and death, lipid peroxidation, and DNA fragmentation associated with an increase in antiapoptotic Bcl-2 family members (50, 51). While these anti-inflammatory and cytoprotective effects seem to be mediated by a reduction in NFκB activation (31), preliminary data from our laboratory show that these cytokines elicit their cytoprotective effects, at least in part, by a signal transducer and activator of transcription 3 (STAT3)and mitogen activated protein kinase 1 (MEK- Independent signal transduction pathway(s). The generation of transgenic mice that overexpress IL-13 within the lung showed a significant prolonged survival when exposed to hyperoxia. This enhanced survival was associated with and possibly mediated by increased lung vascular endothelial growth factor (VEGF) (52). The intratracheal administration of murine rIL-4, rIL-13, and rIL-12 showed protective effects in an IgG immune complex model of lung injury by reducing the number of lung neutrophils, vascular permeability, and TNF-α concentrations by up to 98, 34, and 97%, respectively, a response similar to that of IL-10 (53). Chemokines are a group of small polypeptides, related to cytokines, which are elaborated from a number of different cell types including endothelial and epithelial cells,

Acute respiratory distress syndrome

108

fibroblasts, monocytes, and neutrophils, via the action of pro-inflammatory mediators such as LPS, TNF-α, and IL-1β. The effects are targeted locally and act in an autocrine and paracrine manner, causing leukocytes (specifically neutrophils in the case of ALI/ARDS) to change shape, follow a chemotactic gradient, promote adhesion to vascular endothelial walls, as well as increase their cytotoxic potential. Several classes of chemokines exist, with nearly all having been implicated in the patho-genesis of ALI/ARDS, and are divided into two groups according to the position of their conserved cysteine residues; the C-C chemokines have their cysteines adjacent to each other, where the C-X-C chemokines have their cysteines separated by some other amino acid (X). IL8, an 8.0–8.4 kDa protein, significantly increases in concetration in the BAL fluids of ALI/ ARDS patients. Increases in IL-8 among ALI/ARDS patients showed a positive correlation with the quantity of neutrophils present in lavage fluid and mortality, strongly implicating IL-8 as a key pathogenic chemoattractant for neutrophils into the airspaces of patients with ALI/ARDS (1, 54, 55). In a rabbit model of ischemia-reperfusion lung injury, production of IL-8 correlated with pulmonary neutrophil recruitment and was later attenuated by the use of anti-IL-8 blocking antibody (56). Similar studies showed that treatment with anti-IL-8 blocking antibody, experimental lung injury can be attenuated despite the model of lung injury (57, 58). Interestingly, IL-8 is a fairly hardy protein that has the ability to resist proteolysis and denaturation while not sacrificing its biological activity. Recent studies of ALI/ARDS patients BAL fluid detected the presence of IL-8 complexed to α2-macroglobulin (alpha 2m). Alpha 2m is a major proteinase inhibitor that is mainly found in plasma and has several functions in the regulation of cytokines, including IL-8. While alpha 2m does not alter IL-8-induced neutrophil chemotaxis and priming, the complexed alpha 2m protected IL-8 from degradation (59, 182). The increased stability of IL-8 could act to further promote the acute inflammation as well as neutrophil recruitment. As noted previously, the activation and increase in IL-8 production is mediated primarily through the stimulating actions of the early response cytokines, TNF-α and IL1β, on specific cells that release IL-8 (60). The exact molecular mechanisms by which IL8 gene expression and production is increased are being elucidated (Fig. 1). When human umbilical vein endothelial cells (HUVECs) are exposed low levels of oxygen, a timedependent release of IL-8 occurs and can be attenuated by the addition of anti-IL-8 antibody. This release of IL-8 is preceded by an increase in IL-8 mRNA levels, which was associated with increased binding of NFκB. Similar in vivo experiments concentrating on IP-10, a murine homologue to IL-8, showed an increase in both mRNA and protein production as well as increased pulmonary neutrophil sequestration in response to hypoxia (61, 62). Activation and release of IL-8 from HUVECs can be achieved through the presence of sublytic concentrations of MAC (membrane attack complex) in a complement-mediated model of lung injury (63). The concomitant gene upregulation and production of IL-8 through the MAC complement cascade occurs through the activation and complete translocation of NFκB to the nucleus. Sublytic concentrations of MAC can increase the DNA-binding capacity shortly after stimulation. This NFκB-mediated overexpression and production of IL-8 can be blocked through the preincubation with SN50. SN50 is an engineered peptide that competes with the nuclear localization sequence of the NFκB p50 subunit (64).

Pathogenesis of acute lung injury

109

Bronchial epithelial cells will undergo apoptosis in response to either TNF-α or fas ligation. During this process these cells will secrete IL-8 (through new gene expression and protein synthesis). The fas ligation appears to regulate the production of IL-8 through the activation of NFκB (65). Once the expression of the IL-8 gene is activated, the actual protein production is alternately controlled by protein kinases, specifically MAPK. Fibroblasts stimulated by bradykinins will produce IL-8 that is mediated through the phosphorylation of ERK1/2 and p38 MAPK. In the presence of PD98509, a specific inhibitor of MEK-1-dependent pathways (ERK1/2), and/or SB203580, a specific p38 MAPK inhibitor, IL-8 production is blocked (35). Additionally TNF-α can activate the p38 MAPK pathway and lead to the production of IL-8 by endothelial cells. Preincubation with an antioxidant that scavenges reactive oxidant species (ROS), such as N-acetyl-L-cysteine (NAC) as well as other structurally different antioxidants such as pyrrolidine dithiocarbamate (PDTC), can attenuate the TNF-α-induced activation of p38 as well as its release of IL-8 from pulmonary endothelial cells (66). This implies the role of ROS, from TNF-α stimulated cells, as a signaling intermediate in the activation of p38 in response to TNF-α as well as its downstream products like IL-8. NAC and similar antioxidants possess the ability to suppress the increased expression of IL-8 that is induced by bleomycin (67) and IL-1β-stimulated bronchial epithelial cells (68). While NAC can exhibit surprising suppression of increased IL-8 gene expression and production, it seems that the inhibitory effects of NAC predominately target the early response cytokines, like TNF-α, and as a consequence their downstream products, like IL-8. Despite the studies strong evidence supporting IL-8 as the predominant polymorphonuclear neutrophil (PMN) chemo-attractant factor in lung fluids, the correlation between IL-8 and the PMN levels in the BAL at the onset of ARDS are not supportive (32), thus possibly implicating other chemokines in the pathogenesis of ALI/ARDS. Such redundancy in chemokine activity helps to further promote and propagate the inflammation stage of ALI/ARDS. Macrophage inflammatory proteins (MIP) are members of the same C-C chemokine family as IL-8. They are primarily secreted from macrophages in response to a number of stimuli. Like other members of the C-C chemokine family, MIPs primarily regulate the chemotaxis of neutrophils and are primarily dependent on the activation and production of upstream products such as TNFα and IL-1β (69). Regardless of the duration of ALI/ARDS, MIP concetrations are considerably increased in BAL fluids (55) and correlate with the degree of lung injury (70). A number of different stimuli increase both the gene activation and production of MIPs, LPS administered to mice induces acute lung inflammation with the hallmark early influx of neutrophils into the alveolar space in conjunction with an increase in the gene expression of MIP-1 and MIP-2 in a time-dependent fashion (71–73). The activation of transcriptional factors like NFκB and CREB also appear in the lungs of mice under similar conditions. The increase in both activation and production of MIP along with these transcriptional regulators is suggestive of a possible mechanism (74, 181). Neutropenic mice, exposed to an endotoxin or hemorrhage model of lung injury showed a diminished activation of NFκB as well as significant reductions in the expression of TNF-α, IL-1β, and MIP-2 (29). Genetic deletion of the mouse CCR1 MIP-1α receptor shows remarkable pulmonary protection from inflammation secondary to acute pancreatitis (75). Despite the stimuli, whether it be LPS, hemorrhage, pacreatitis,

Acute respiratory distress syndrome

110

hyperoxia (76), ozone (77), ventilation (78), or acid aspiration (79), the same pattern is exhibited, which consists of a rapid activation and production of MIPs and other key cytokines (TNF-α, IL-1β) and chemokines (IL-8). This overproduction seems to occur at the level of transcriptional regulatory factors such as NFκB or CREB. The effects of the chemokines also follow a very precise pattern that includes neutrophil recruitment, increased vascular permeability, and increase expression of adhesion molecules. The use of inhibiting antibodies to MIPs greatly attenuates these effects, possibly through the inhibition of transcription factors, the downregulation of upstream early response cytokines, or the upregulation of anti-inflammatory cytokines (80). While several possibilities exist for the mechanisms of MIP activation, it is also possible that the initial pathogenic stimulant first acts on mediators found to be active earlier in the course of ALI/ARDS, such as the early response cytokines, and subsequently activates other downstream products like MIP and IL-8. Other potent chemoattractant molecules such as leukotriene B4 (LTB4) and plateletactivating-factor (PAF) are increased in the BAL fluid of patients suffering from ALI/ARDS (81, 82) and serve to recruit neutrophils into the alveolar space. Both of these mediators share the upstream enzyme know as phospholipase A2 (PLA2). Following cell injury, the phospholipids composing the membranes begin to break down, by action of enzymes like PLA2, into potent inflammatory mediators such as arachidonic acid metabolites (LTB4) and PAF. Elevated plasma levels of PLA2 during septic shock complicated by ALI/ARDS were 20 times greater than that of control and nonALI/ARDS septic shock patients (83, 84, 186). Supportive data have shown there to be a link between the BAL levels of PLA2 and the development of ALI/ARDS in patients suffering from septicemia (85). Several experimental studies have succeeded in demonstrating the role of PLA2 in the pathogenesis of ALI/ARDS. In vivo results of LPS-induced lung injury demonstrated that macrophages are the major source of PLA2 and that its synthesis is TNF-α dependent (86, 186). Because many phospholipases are also Ca2+ dependent, isolated hamster lungs that were exposed to a Ca2+ ionophore developed ultrastructural evidence of lung injury. The levels of free arachondonic acid from lung extracts were also elevated (87). The exogenous administration of PLA2 induces alveolar macrophages to release nitric oxide, partially contributing to the lung injury by damage to endothelial cells. This effect can be blocked by the addition of a PLA2 inhibitor such as quinacrine (88, 186). Other inhibitor studies have shown a dosedependent decrease in BAL PLA2, LTB4 and PAF activity. The decreased activity is associated with diminished lung injury in rabbits exposed to oleic acid (89). Pretreatment with mepacrine, a PLA2 inhibitor, in a rat silica model of ALI/ ARDS resulted in significant reduction in leukocytosis of both neutrophils and alveolar macrophages (90).

III. Neutrophil Trafficking Despite the occurrence of ALI/ARDS in neutropenic patients (91–95), much of the evidence has implicated the neutrophil as being the major effector of cell damage and lung injury (96, 97), by actions of proteases and increased production of reactive oxygen intermediates. Histologically, patients suffering from ALI/ARDS have a dramatic increase in pulmonary neutrophils as they are sequestered within the lung. The physical

Pathogenesis of acute lung injury

111

properties of both neutrophils and the lung vasculature play a significant role in the accumulation of neutrophils in the lung. With nearly 40–60% of pulmonary capillaries being smaller in caliber than neutrophils are in diameter, trafficking through the lung becomes sluggish. Neutrophils must deform and elongate in order to pass through the pulmonary bed (98). The lung acts as a filter slowing neutrophil transit time through the lung to around 26 seconds, as compared to 1.4 seconds for erythrocytes (99, 100). In response to inflammatory stimuli, whether direct or indirect, the pulmonary sequestration of neutrophils, enhanced by chemokines, increases almost 5 times (101). The sludging of neutrophils through the pulmonary circulation is partially explained by the neutrophils becoming less deformable through redistribution of actin from the central (g-actin) to the submembranous areas (f-actin) (102–106), preventing the neutrophil from flattening. A concurrent efflux of immature neutrophils from the bone marrow as well as a relative blood neutropenia occurs as the neutrophils become trapped within the lung (101, 107). The immature neutrophils, which represent nearly all those sequestered, are less deformable, express more L-selectin, are less mobile and responsive to chemotactic signals, and thus become locked within the pulmonary vasculature and contribute to endothelial damage (108). Neutrophils sequestered in the lung arrive in an activated state with enhanced cytotoxic effects in the form of increased oxidant production, increased lysosomal enzyme release, increased expression of adhesion molecules (CD11/18), and reduced Lselectin expression (109, 110, 187). The activation is achieved by G-proteins coupled to serpentine cytokine receptors that transduce the signal to the nucleus when bound by soluble inflammatory mediators and/or with the adherence to the endothelium (Fig. 2). Despite the activation stimuli, the pathways that transmit the signals from the plasma membrane to the nucleus seem to possess a commonality. One such pathway, the mitogen-activated protein kinase (MAPK) pathway, involves the multiple stepwise interacting cascades of serine/threonine phosphorylation of several intermediates. The ERK family of MAPKs shows activation in human neutrophils when stimulated by inflammatory stimuli such as PMA and fMLP and contributes in activating oxidative burst, phagocytosis, and apoptosis (111). In response to hemorrhage or endotexomia, lung neutrophils show an increase in activation of nuclear factors like CREB and NFκB, as well as the increased expression and production of acute phase products

Acute respiratory distress syndrome

112

Figure 2 Transcriptional regulators required for neutrophil and endothelial activation in response to injury or other activation stimuli (acute phase products). Activation pathways of neutrophils and endothelial cells begins at the receptors level, propagated through protein kinases, and transcription factors, resulting in increased expression and production of cytokines, chemokines, and adhesion molecules. like TNF-α and MIP-2. The activation of these nuclear factors was in part due to the effects of the ERK family of MAPKs (112). While the p38 MAPK pathways has been implicated in the modulation of neutrophil activation in response to endotoxemia or hemorrhage, its does not seem to play an active role in the pathogenesis of endotoxin- or hemorrhage-induced ALI/ARDS. Despite the increased activation of p38, its inhibition did not abrogate neutrophil influx and activation, nor did it reduce the production and activation of the pro-inflammatory cytokines and nuclear factors (113). A second possible pathways for activation and sequestration of neutrophils to pulmonary compartments involves transcription factors, such as NFκB. Mice made neutropenic either chemically or with an anti-neutrophil antibody and subsequently exposed to endotoxin-or hemorrhageinduced lung injury showed a significant reduction in lung injury. Control animals showed classic signs of lung injury, characterized by lung edema, neutrophil influx, and

Pathogenesis of acute lung injury

113

NFκB activation. Interestingly the activation of NFκB was diminished in neutropenic mice (29). In mice in a similar model of lung injury, NFκB activation contributed to the recruitment of neutrophils as well the expression of IL-8, TNF-α, and MIP-2 mRNA (74). In addition to enhancing neutrophil expression of chemokine genes that play a role in pulmonary sequestration, NFκB plays a pivotal role in prolonging the survival of activated neutrophils by regulating the expression of apoptosis genes (114). A hallmark characteristic of patients suffering from ARDS is increased activated lung neutrophils, generating reactive oxygen species and pro-inflammatory cytokines. Interestingly, these neutrophils also have decreased rates of apoptosis (36, 115, 116). Experimental models of ALI/ARDS induced by endotoxin of hemorrhage showed that activated lung neutrophils had prolonged or enhanced survival in the presence of increased levels of activated NFκB (116). The combination of activated neutrophils with reduced cell death allows for prolonged cytotoxic effects for an extended period of time. Initially within the pulmonary vascular bed, neutrophils adhere weakly. Subsequently, due to activation of the neutrophils as well as the endothelium, various types of adhesion molecules are expressed, enhancing neutrophil adherence (Figs. 2, 3). The systemic and/or local activation of the endothelial bed can occur independently of, or as a component of cellular injury (117, 118). The expression of adhesion molecules is a key marker of endothelial activation. Selectins, a family of cell-cell adhesion molecules, are expressed on both neutrophils and the endothelial cells and are reactive to specific oligosacharides, resulting in a attachment of neutrophils to the endothelium. Integrins function as transmembrane adhesion molecules that provide a second adhesion complex, thus firmly binding the neutrophil to the endothelium. Integrins also possess the

Figure 3 Events that occur during neutrophil margination and diapedesis to the inflammatory sites within the

Acute respiratory distress syndrome

114

ALI/ARDS lung. Inflammatory events result in the production of acute phase products that activate both the endothelium and neutrophils, aiding neutrophil trafficking to the sites of inflammation in the injured lung. ability to act as signal transducers that, when bound, can activate intra-cellular signaling pathways. Human studies comparing the endothelium from ALI/ARDS patients to that of control patients have shown an increase in expression of various adhesion molecules, such as ICAM-1, vascular cell adhesion molecule 1 (VCAM-1), CD14, tumor necrosis factor receptor II (TNFR II) (119, 185), and P-selectin (120); while the presence of Lselectin is detectable early, it declines as the neutrophils become attached to the endothelium (121, 122). The selective inhibition of P-, L-, or E-selectin results in dosedependent protective effects (123). When P-, L-selectin were inhibited in combination, there was a protective effect that was similar to the effect with inhibition of either L-, or P-selectin, whereas the combined inhibition of P- and E-selectin showed synergistic protection (123). When endothelial cells are treated with LPS, TNF-α, or IL-1β, gene transcription results in de novo protein synthesis and increased expression of E-selectin (124). Thrombin and histamine tretament brings about a sharp increased endothelial expression of P-selectin and PAF on the cell surface (124, 125). Neutrophils themselves can induce and modulate the activation of endothelial cells (126). Like the expression of many cytokines and chemokines, the expression of several genes that encode adhesion molecules is under the regulation of transcriptional factors like NFκB. Endothelial cell expression of adhesion molecules, such as ICAM-1, VCAM-1 and E-selectin, are induced by pro-inflammatory cytokines and regulated by NFκB binding and translocation (127). This NFκB-dependent expression of endothelial adhesion molecules can be downregulated by antioxidants (128). The adherence of the neutrophil to the endothelium shows greater cytotoxic effects, producing up to 1000-fold more oxygen radicals and leading to cell injury, where nonadherent but activated neutrophils do not cause cell injury (111). Blockage of this adherence either with a GD1 1/18 blocking antibody (129) or in patients lacking B2 integrins abrogates this effect (130). Since neither integrins nor selectins possess any enzymatic activity, these adhesion molecules function to transduce the activation signals received from soluble mediators or the endothelium. The cross-linking of integrins and Lselectins activates similar pathways as soluble mediators (131–133) and results in increasing cytokine production (134) and enhanced oxidative burst (135). The regulation of integrin receptor affinity as well as the transduction of signals to the neutrophil cytoskeleton is achieved by signalling molecules, such as CD47 (136), focal adhesion kinase (FAK) (137, 138), and integrin-linked kinase (ILK) (139), which are associated with the intracellular domains of integrins. Once the neutrophils have marginated to the pulmonary vasculature they begin to diapedese across the endothelium and interstitial space and into the alveoli between alveolar epithelial cells (140, 141). The process of neutrophil margination and diapedesis is dependent on adhesion molecules (Fig. 3) and can be blocked through the use of

Pathogenesis of acute lung injury

115

blocking antibodies. Escherichia coli, Pseudomones aeruginosa, LPS, IgG immune complexes, IL-1β, and PMA elicit the use of a CD11/CD18-dependent pathway, upregulating ICAM-1, while Staphylococcus pneumoniae, group B streptococcus, Sterptococcus aureus, HCL, hyperoxia, and C5a allow neutrophil emigration by an unidentified pathway that does not utilize CD 11/CD 18 (142–149) or selectin (150). The induction of ICAM-1 is closely regulated by NFκB itself and is upregulated by such proinflammatory cytokines as TNF-α and IL-1β (151).

IV. Coagulation System Characteristic of patients who suffered from ALI/ARDS is the pathological findings of increased megakaryocytes (platelet precursor) and platelet or fibrin microthrombi found in both the extravascular and intracellular pulmonary compartments (152, 153). The balance of pro-coagulant pathways and fibrin clearance, or fibrinolysis, regulates the formation of microthrombi within the inflamed lung (Fig. 4). These pathways, especially the enhanced pro-coagulation seen in ALI/ARDS, can influence the trafficking of cells through the lung, cell permeability, and activation/propagation of many acute phase products. The presence of coagulation factors such as

Acute respiratory distress syndrome

116

Figure 4 Diagrammatic view of coagulation pathways activated in ALI/ARDS as well as the therapeutic anticoagulation pathways in the treatment of ALI/ARDS. Coagulation factors, denoted by roman numerals, are active when designated with “a.” Solid lines indicate a pro-coagulant effect, while broken lines indicate an anticoagulant effect. A line ending with a flat head notes inhibition.

Pathogenesis of acute lung injury

117

tissue factor and factor VII in a normal lung is abundant, yet the lung still maintains a net fibrinolytic state due to the presence of urokinase plasminogen activator (uPA) that prevents the formation of distal coagulation proteins (154). uPA is a serine protease that targets plasminogen, an inactive serine protease abundant in the blood, cleaving a single bond to yield that active protease plasmin. Unlike uPA, plasmin has a broader specificity, cleaving a variety of coagulation proteins, resulting in lung compartments that are kept clear of fibrin. During pathogenesis of ALI/ARDS this process is overwhelmed, and the result is the formation of hyaline membranes that are rich in fibrin (155). The leakage of pro-coagulant proteins distal to tissue factor/factor VII from capillaries as well as an influx of inhibitors to uPA and plasmin all aid in the deposition of intravascular and extravascular fibrin (156–158) and the formation of thrombi that increase pulmonary vascular resistance (159). The localization of urokinase activity within the pulmonary alveoli is due to the synthesis of urokinase (160) as well as the expression of receptors for urokinase (uPAR) that are produced by epithelial cells (161). Epithelial fibrinolytic activity is demonstrated when in vitro epithelial monolayers lyse plasma-derived fibrin that form over their surface (161). The expression of uPAR, as well as the expression of the leukocyte integrin Mac-1 (CD11b/ CD18) that posses fibrolytic activity, can be induced from epithelial cells by TGFβ. Both uPAR and Mac-1 promote the degradation of fibrin as well as possess adhesive properties with the binding to vitronectin and fibrin, respectively. The attachment of vitronectin to uPAR promotes and enhances the degradation of fibrinogen by Mac-1 and when saturated with exogenous uPA can be inhibited in both Mac-1 fibrin binding and degradation (162). The regulation of uPA has been shown, in part, to be regulated by IL-1β and the availability of intracellular iron. The decrease in intracellular iron was shown to downregulate the expression of uPA as well as upregulate the expression of plasminogen activator inhibitor-1 (PAI-1) in human pulmonary epithelial A549 cells, thus decreasing the cell-surface generation of plasmin (163, 164). The BAL fluid from patients suffering from ALI/ARDS shows decreased fibrinolytic activity, which may be due to damaged epithelium (165). The induction of PAI-1 production from murine alveolar macrophages due to hyaluronic acid fragments has also been shown to disrupt fibrinolytic activity (166). Human BAL samples from ARDS patients have also shown the levels of PAI-1 to be elevated. When challenged by hyperoxia, the development of fibrin was significantly reduced in mice deficient in PAI-1 (167). Similar experiments exposing PAI-1 knockout mice to bleomycin showed a decrease in fibrin deposition in the PAI-1 knockout mice while controls showed PAI-1 to be increased by bleomycin (168). The increases in PAI-1 found in ALI/ARDS BAL samples as well as experimental models alters the normal fibrinolytic activity, promoting the deposition of fibrin and enhancing the formation of thrombi. Hyaluronic acid (HA) is a glycosaminoglycan (GAG) that is part of the extracellular matrix (ECM). Normally in a high molecular weight form, initiation of the inflammatory cascade can make HA break into smaller fragments. These fragments can modulate the functions of macrophages through alterations in chemotaxis and activation of the NFκB transcription pathway (169), as well as modulate the fibrinolytic activity within the lung. In response to lung injury, HA concentrations increased significantly in both the BAL fluid and serum (170). Bleomycin-induced lung injury also increased HA levels in BAL

Acute respiratory distress syndrome

118

fluids. The use of a HA-binding peptide blocks the actions of HA with the reduction of collagen alpha (I) mRNA levels (171). In addition, the generation of thrombin is suspected to be elevated, as BAL samples show an increase fibrinopeptide A. Thrombin acts on soluble fibrinogen in the tissue and plasma to produce insoluble strands of fibrin. Increases in fibrinogen indicate an increase in thrombi-mediated formation of fibrin (172). Thrombin further possesses the ability to stimulate the expression of a number of cytokines such as IL-1β and has proinflammatory effects including the regulation of endothelial cell permeability and contraction and the chemotaxis and sequestrion of inflammatory cells (173). Despite the mounting evidence of increased levels of systematic coagulation and anti-fibrinolytic markers, plasma levels of these markers do not correlate with pulmonary damage in patients suffering from ALI/ARDS (174). The use of anti-coagulant for the treatment of ALI/ARDS has shown some promising results. Anticoagulant therapy with activated protein C (APC) has recently shown to decrease mortality in sepsis (175, 176) (see Chap. 16). Occuring naturally, APC inhibits Va and VIIIa coagulation factors and neutralizes PAI-1. It also has several antiinflammatory effects such as decreasing the production of TNF-α, IL-1β, IL-6, as well as reducing inflammatory adherence to the endothelium (177). Another anticoagulant, tissue factor pathway inhibitor (TFPI), blocks the activity of the tissue factor-VIIa procoagulant complex by blocking tissue factor VIIa. Low levels of TFPI have been detected in the BAL fluid of patients with ALI/ARDS, but these levels are overwhelmed by the presence of other procoagulant factors and thus are insufficient to block fibrin deposition (178, 180). Adult baboons that were treated with TFPI and later exposed to sepsisinduced lung injury showed a dramatic protection from lung injury as compared to controls, characterized by the maintenance of pulmonary hypertension, as well as a reduction in the systemic pro-inflammatory response and fibrin deposition (179).

V. Conclusion The molecular mechanisms involved in the pathogenesis of acute lung injury are complex, interconnected, and redundant. Ongoing studies of various mediators and coagulation factors have revealed that the normal lung operates under a very fine balance of pro- and anti-inflammatory states. With injury, the lung can potentially be thrown out of equilibrium in favor of inflammation. The exggeration of inflammatory and coagulation responses is regulated, for the most part, at the transcriptional level by a highly intricate network of DNA-binding proteins and transcription factors. Advances in our understanding of the molecular mechanisms that control the pathogenesis of ALI/ARDS may lead to the identification of other factors that underlie a patient’s susceptibility to lung injury. A more detailed appreciation of the inflammatory mechanisms does provide a starting point for the design and development of newer pharmacological agents that may provide therapeutic benefit in the future.

References

Pathogenesis of acute lung injury

119

1. Miller EJ, Cohen AB, Nagao S, Griffith D, Maunder RJ, Martin TR, Weiner-Kronish JP, Sticherling M, Christophers E, Matthay MA. Elevated levels of NAP-1/interleukin-8 are present in the airspaces of patients with the adult respiratory distress syndrome and are associated with increased mortality. Am Rev Respir Dis 1992; 146:427–432. 2. Imamura S, Matsukawa A, Ohkawara S, Kagayama M, Yoshinaga M. Involvement of tumor necrosis factor-alpha, interleukin-1 beta, interleukin-8, and interleukin-1 receptor antagonist in acute lung injury caused by local Shwartzman reaction. Pathol Int 1997; 47:16–24. 3. Zhao G, al-Mehdi AB, Fisher AB. Anoxia-reoxygenation versus ischemia in isolated rat lungs. Am J Physiol 1997; 273:L1112–1117. 4. Severn A, Rapson NT, Hunter CA, Liew FY. Regulation of tumor necrosis factor production by adrenaline and beta-adrenergic agonists. J Immunol 1992; 148:3441–3445. 5. Springer TA. Adhesion receptors of the immune system. Nature 1990; 346:425–434. 6. Butcher EC. Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell 1991; 67:1033–1036. 7. Bevilacqua MP, Stengelin S, Gimbrone MA, Jr., Seed B. Endothelial leukocyte adhesion molecule 1: an inducible receptor for neutrophils related to complement regulatory proteins and lectins. Science 1989; 243:1160–1165. 8. Bevilacqua MP, Pober JS, Majeau GR, Fiers W, Cotran RS, Gimbrone MA, Jr. Recombinant tumor necrosis factor induces procoagulant activity in cultured human vascular endothelium: characterization and comparison with the actions of interleukin 1. Proc Natl Acad Sci USA 1986; 83:4533–4537. 9. Stern DM, Kaiser E, Nawroth PP. Regulation of the coagulation system by vascular endothelial cells. Haemostasis 1988; 18:202–214. 10. Crutchley DJ, Conanan LB. Endotoxin induction of an inhibitor of plasminogen activator in bovine pulmonary artery endothelial cells. J Biol Chem 1986; 261:154–159. 11. Esmon CT. The regulation of natural anticoagulant pathways. Science 1987; 235:1348–1352. 12. Clouse LH, Comp PC. The regulation of hemostasis: the protein C system. N Engl J Med 1986; 314:1298–1304. 13. Moore KL, Andreoli SP, Esmon NL, Esmon CT, Bang NU. Endotoxin enhances tissue factor and suppresses thrombomodulin expression of human vascular endothelium in vitro. J Clin Invest 1987; 79:124–130. 14. Stricter RM, Kunkel SL, Showell HJ, Marks RM. Monokine-induced gene expression of a human endothelial cell-derived neutrophil chemotactic factor. Biochem Biophys Res Commun 1988; 156:1340–1345. 15. Stricter RM, Kunkel SL, Showell HJ, Remick DG, Phan SH, Ward PA, Marks RM. Endothelial cell gene expression of a neutrophil chemotactic factor by TNF-alpha, LPS, and IL-1 beta. Science 1989; 243:1467–1469. 16. Larrick JW, Kunkel SL. The role of tumor necrosis factor and interleukin 1 in the immunoinflammatory response. Pharm Res 1988; 5:129–139. 17. Kunkel SL, Remick DG, Stricter RM, Larrick JW. Mechanisms that regulate the production and effects of tumor necrosis factor-alpha. Crit Rev Immunol 1989; 9:93–117. 18. Le J, Vilcek J. Tumor necrosis factor and interleukin 1: cytokines with multiple overlapping biological activities. Lab Invest 1987; 56:234–248. 19. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood 1996; 87:2095–2147. 20. Ulich TR, Watson LR, Yin SM, Guo KZ, Wang P, Thang H, del Castillo J. The intratracheal administration of endotoxin and cytokines. I. Characterization of LPS-induced IL-1 and TNF mRNA expression and the LPS-, IL-1, and TNF-induced inflamatory infiltrate. Am J Pathol 1991; 138:1485–1496. 21. McAnulty RJ, Laurent GJ. Pathogenesis of lung fibrosis and potential new therapeutic strategies. Exp Nephrol 1995; 3:96–107.

Acute respiratory distress syndrome

120

22. Moine P, McIntyre R, Schwartz MD, Kaneko D, Shenkar R, Le Tulzo Y, Moore EE, Abraham E. NF-kappaB regulatory mechanisms in alveolar macrophages from patients with acute respiratory distress syndrome. Shock 2000; 13:85–91. 23. Browder W, Ha T, Chuanfu L, Kalbfleisch JH, Ferguson DA, Jr., Williams DL. Early activation of pulmonary nuclear factor kappaB and nuclear factor interluekin-6 in polymicrobial sepsis. J Trauma-Injury Infect Crit Care 1999; 46:590–596. 24. Shenkar R, Abraham E. Hemorrhage induces rapid in vivo activation of CREB and NF-kappaB in murine intraparenchymal lung mononuclear cells. Am J Respir Cell Mol Biol 1997; 16:145– 152. 25. Shenkar R, Yum HK, Arcaroli J, Kupfner J, Abraham E. Interactions between CBP, NFkappaB, and CREB in the lungs after hemorrhage and endotoxemia. Am J Physiol Lung Cell Mol Physiol 2001; 281:L418–426. 26. Hassoun PM, Yu FS, Cote CG, Zulueta JJ, Sawhney R, Skinner KA, Skinner HB, Parks DA, Lanzillo JJ. Upregulation of xanthine oxidase by lipopolysaccharide, interleukin-1, and hypoxia. Role in acute lung injury. Am J Respir Crit Care Med 1998; 158:299–305. 27. Sanders KA, Huecksteadt T, Xu P, Sturrock AB, Hoidal JR. Regulation of oxidant production in acute lung injury. Chest 1999; 116:56S-61S. 28. Wong HR, Ryan M, Wispe JR. Stress response decreases NF-kappaB nuclear translocation and increases I-kappaBalpha expression in A549 cells. J Clin Invest 1997; 99:2423–2428. 29. Abraham E, Carmody A, Shenkar R, Arcaroli J. Neutrophils as early immunologic effectors in hemorrhage- or endotoxemia-induced acute lung injury. Am J Physiol Lung Cell Mol Physiol 2000; 279:L1137–1145. 30. Lentsch AB, Jordan JA, Czermak BJ, Diehl KM, Younkin EM, Sarma V, Ward PA. Inhibition of NF-kappaB activation and augmentation of IkappaBbeta by secretory leukocyte protease inhibitor during lung inflammation. Am J Pathol 1999; 154:239–247. 31. Lentsch AB, Crouch LD, Jordan JA, Czermak BJ, Yun EC, Guo R, Sarma V, Diehl KM, Ward PA. Regulatory effects of interleukin-11 during acute lung inflammatory injury. J Leukocyte Biol 1999; 66:151–157. 32. Martin TR. Lung cytokines and ARDS: Roger S. Mitchell Lecture. Chest 1999; 116:28–88. 33. Hardie WD, Prows DR, Leikauf GD, Korfhagen TR. Attenuation of acute lung injury in transgenic mice expressing human transforming growth factor-alpha. Am J Physiol 1999; 277:L1045–1050. 34. Dhingra VK, Uusaro A, Holmes CL, Walley KR. Attenuation of lung inflammation by adrenergic agonists in murine acute lung injury. Anesthesiology 2001; 95:947–953. 35. Hayashi R, Yamashita N, Matsui S, Fujita T, Araya J, Sassa K, Arai N, Yoshida Y, Kashii T, Maruyama M, Sugiyama E, Kobayashi M. Bradykinin stimulates IL-6 and IL-8 production by human lung fibroblasts through ERK-and p38 MAPK-dependent mechanisms. Eur Respir J 2000; 16:452–458. 36. Matute-Bello G, Liles WC, Radella F, 2nd, Steinberg KP, Ruzinski JT, Jonas M, Chi EY, Hudson LD, Martin TR. Neutrophil apoptosis in the acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 156:1969–1977. 37. Bucala R. MIF rediscovered: cytokine, pituitary hormone, and glucocorticoid- induced regulator of the immune response. Faseb J 1996; 10:1607–1613. 38. Donnelly SC, Haslett C, Reid PT, Grant IS, Wallace WA, Metz CN, Bruce LJ, Bucala R. Regulatory role for macrophage migration inhibitory factor in acute respiratory distress syndrome [see comments]. Nat Med 1997; 3:320–323. 39. Donnelly SC, Bucala R. Macrophage migration inhibitory factor: a regulator of glucocorticoid activity with a critical role in inflammatory disease. Mol Med Today 1997; 3:502–507. 40. Kumar NM, Rabadi NH, Sigurdson LS, Schunemann HJ, Lwebuga-Mukasa JS, Induction of interleukin-1 and interleukin-8 mRNAs and proteins by TGF beta 1 in rat lung alveolar epithelial cells. J Cell Physiol 1996; 169:186–199.

Pathogenesis of acute lung injury

121

41. Fiorentino DF, Zlotnik A, Mosmann TR, Howard M, O’Garra A. IL-10 inhibits cytokine production by activated macrophages. J Immunol 1991; 147:3815–3822. 42. Ramani M, Ollivier V, Khechai F, Vu T, Ternisien C, Bridey F, de Prost D. Interleukin-10 inhibits endotoxin-induced tissue factor mRNA production by human monocytes. FEBS Lett 1993; 334:114–116. 43. Parsons PE, Moss M, Vannice JL, Moore EE, Moore FA, Repine JE. Circulating IL-1ra and IL10 levels are increased but do not predict the development of acute respiratory distress syndrome in at-risk patients. Am J Respir Crit Care Med 1997; 155:1469–1473. 44. Donnelly SC, Stricter RM, Reid PT, Kunkel SL, Burdick MD, Armstrong I, Mackenzie A, Haslett C. The association between mortality rates and decreased concentrations of interleukin10 and interleukin-1 receptor antagonists in the lung fluids of patients with the adult respiratory distress syndrome. Ann Intern Med 1996; 125:191–196. 45. Armstrong L, Millar AB. Relative production of tumour necrosis factor alpha and interleukin 10 in adult respiratory distress syndrome. Thorax 1997; 52:442–446. 46. Park WY, Goodman RB, Steinberg KP, Ruzinski JT, Radella F, 2nd, Park DR, Pugin J, Skerrett SJ, Hudson LD, Martin TR. Cytokine balance in the lungs of patients with acute respiratory distress syndrome. Am J Respirat Crit Care Med 2001; 164:1896–1903. 47. Osman MO, Jacobsen NO, Kristensen JU, Deleuran B, Gesser B, Larsen CG, Jensen SL. IT 9302, a synthetic interleukin-10 agonist, diminishes acute lung injury in rabbits with acute necrotizing pancreatitis. Surgery 1998; 124:584–592. 48. Shanley TP, Vasi N, Denenberg A. Regulation of chemokine expression by IL-10 in lung inflammation. Cytokine 2000; 12:1054–1064. 49. Pulido EJ, Shames BD, Pennica D, O’Leary R M, Bensard DD, Cain BS, McIntyre RC, Jr. Cardiotrophin-1 attenuates endotoxin-induced acute lung injury. J Surg Res 1999; 84:240–246. 50. Waxman AB, Einarsson O, Seres T, Knickelbein RG, Warshaw JB, Johnston R, Homer RJ, Elias JA. Targeted lung expression of interleukin-11 enhances murine tolerance of 100% oxygen and diminishes hyperoxia-induced DNA fragmentation. J Clin Invest 1998; 101:1970– 1982. 51. Ward NS, Waxman AB, Homer RJ, Mantell LL, Einarsson O, Du Y, Elias JA. Interleukin-6induced protection in hyperoxic acute lung injury. A J Respir Cell Mol Biol 2000; 22:535–542. 52. Corne J, Chupp G, Lee CG, Homer RJ, Zhu Z, Chen Q, Ma B, Du Y, Roux F, McArdle J, Waxman AB, Elias JA. IL-13 stimulates vascular endothelial cell growth factor and protects against hyperoxic acute lung injury. J Clin Invest 2000; 106:783–791. 53. Mulligan MS, Warner RL, Foreback JL, Shanley TP, Ward PA. Protective effects of IL-4, IL10, IL-12, and 11–13 in IgG immune complex-induced lung injury: role of endogenous IL-12. J Immunol 1997; 159:3483–3489. 54. Villard J, Dayer-Pastore F, Hamacher J, Aubert JD, Schlegel-Haueter S, Nicod LP. GRO alpha and interleukin-8 in Pneumocystis carinii or bacterial pneumonia and adult respiratory distress syndrome. Am J Respir Crit Care Med 1995; 152:1549–1554. 55. Goodman RB, Stricter RM, Martin DP, Steinberg KP, Milberg JA, Mauder RJ, Kunkel SL, Walz A, Hudson LD, Martin TR. Inflammatory cytokines in patients with persistence of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154:602–611. 56. Sekido N, Mukaida N, Harada A, Nakanishi I, Watanabe Y, Matsushima K. Prevention of lung reperfusion injury in rabbits by a monoclonal antobody against interleukin-8. Nature 1993; 365:654–657. 57. Matsumoto T, Yokoi K, Mukaida N, Harada A, Yamashita J, Watanabe Y, Matsushima K. Pivotal role of interleukin-8 in the acute respiratory distress syndrome and cerebral reperfusion injury. J Leukoc Biol 1997; 62:581–587. 58. Yokoi K, Mukaida N, Harada A, Watanabe Y, Matsushima K. Prevention of endotoxemiainduced acute respiratory distress syndrome-like lung injury in rabbits by a monoclonal antibody to IL-8. Lab Invest 1997; 76:375–384.

Acute respiratory distress syndrome

122

59. Kurdowska A, Carr FK, Stevens MD, Baughman RP, Martin TR. Studies on the interaction of IL-8 with human plasma alpha 2-macroglobulin: evidence for the presence of IL-8 complexed to alpha 2-macroglobulin in lung fluids of patients with adult respiratory distress syndrome. J Immunol 1997:158:1930–1940. 60. Beck GC, Yard BA, Breedijk AJ, Van Ackern K, Van Der Woude FJ. Release of CXCchemikines by human lung microvascular endothelial cells (LMVEC) compared with macrovascular umbilical vein endothelial cells. Clin Exp Immunol 1999; 118:298–303. 61. Karakurum M, Shreeniwas R, Chen J, Pinsky D, Yan SD, Anderson M, Sunouchi K, Major J, Hamilton T, Kuwabara K, et al. Hypoxic induction of interleukin-8 gene expression in human endothelial cells. J Clin Invest 1994; 93:1564–1570. 62. Osman MO, Kristensen JU, Jacobsen NO, Lautsen SB, Deleuran B, Deleuran M, Gesser B, Matsushima K, Larsen CG, Jensen SL. A monoclonal antiinterleukin 8 antibody (WS-4) inhibits cytokine response and acute lung injury in experimental severe acute necrostising pancreatitis in rabbits. Gut 1998; 43:232–239. 63. Kilgore KS, Flory CM, Miller BF, Evans VM, Warren JS. The membrane attack complex of complement induces interleukin-8 and monocyte chemoattractant protein-1 secretion from human umbilical vein endothelial cells. Am J Pathol 1996; 149:953–961. 64. Kilgore KS, Schmid E, Shanley TP, Flory CM, Maheswari V, Tramontini NL, Cohen H, Ward PA, Friedl HP, Warren JS. Sublytic concentrations of the membrane attack complex of complement induce endothelial interleukin-8 and monocyte chemoattractant protein-1 through nuclear factor-kappa B activation. Am J Pathol 1997; 150:2019–2031. 65. Hagimoto N, Kuwano K, Kawasaki M, Yoshimi M, Kaneko Y, Kunitake R, Maeynama T, Tanaka T, Hara N. Induction of interleukin-8 secretion and apoptosis in bronchiolar epithelial cells by Fas litigation. Am J Respir Cell Mol Biol 1999; 21:436–445. 66. Hashimoto S, Gon Y, Matsumoto K, Takeshita I, Horie T. N-Acetylcysteine attenuates TNFalpha-induced p38 MAP kinase activation and p38 MAP kinase-mediated IL-8 production by human pulmonary vascular endothelial cells. Br J Pharmacol 2001; 132:270–276. 67. Gon Y, Hashimoto S, Nakayama T, Matsumoto K, Koura T, Takeshita I, Horie T. N-acetyl-Lcysteine inhibits bleomycin-induced interleukin-8 secretion by bronchial epithelial cells. Respirology 2000; 5:309–313. 68. Matsumoto K, Hashimoto S, Gon Y, Nakayama T, Takizawa H, Horie T. N-Acetylcysteine inhibits IL-1 alpha-induced IL-8 secretion by bronchial epithelial cells. Respir Med 1998; 92:512–515. 69. Calkins CM, Heimbach JK, Bensard DD, Song Y, Raeburn CD, Meng X, McIntyre RC Jr. TNF receptor I mediates chemokine production and neutrophil accumulation in the lung following systemic lipopolysaccharide. J Surg Res 2001; 101:232–237. 70. Kalyanaraman M, Heidemann SM, Sarnaik AP. Macrophage inflammatory protein-2 predicts acute lung injury in endotoxemia. J Inves Med 1998; 46:275–278. 71. Johnston CJ, Finkelstein JN, Gelein R, Obedorster G. Pulmonary cytokine and chemokine mRNA levels after inhalation of lipopolysaccharide in C57BL/ 6 mice. Toxicol Sci 1998; 46:300–307. 72. Gupta S, Feng L, Yoshimura T, Redick J, Fu SM, Rose CE, Jr. Intra-alveolar macrophageinflammatory peptide 2 induces rapid neutrophil localization in the lung. Am J Respir Cell Mol Biol 1996; 15:656–663. 73. Shanley TP, Schmal H, Friedl HP, Jones ML, Ward PA. Role of macrophage inflammatory protein-1 alpha (MIP)-1 alpha) in acute lung injury in rats. J Immunol 1995; 154:4793–4802. 74. Shenkar R, Abraham E. Mechanisms of lung neutrophil activation after hemorrhage or endotoxemia: roles of reactive oxygen intermediates, NF-kappa B, and cyclic AMP response element binding protein. J Immunol 1999; 163:954—962. 75. Gerard C, Frossard JL, Bhatia M, Saluja A, Gerard NP, Lu B, Steer M. Targeted disruption of the beta-chemokine receptor CCR1 protects againts pancreatitis-associated lung injury. J Clin Invest 1997; 100:2022–2027.

Pathogenesis of acute lung injury

123

76. D’Angio CT, Johnston CJ, Wright TW, Reed CK, Finkelstein JN. Chemokine mRNA alterations in newborn and adult mouse lung during acute hyperoxia. Exp Lung Res 1998; 24:685–702. 77. Zhao Q, Simpson LG, Driscoll KE, Leikauf GD. Chemokine regulation of ozone-induced neutrophil and monocyte inflammation. Am J Physiol 1998; 274:L39–46. 78. Chiumello D, Pristine G, Slutsky AS. Mechanical ventilation affects local and systemic cytokines in an animal model of acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160:109–116. 79. Shanley TP. Davidson BA, Nader ND, Bless N, Vasi N, Ward PA, Johnson KJ, Knight PR. Role of macrophage inflammatory protein-2 in aspiration-induced lung injury. Crit Care Med 2000; 28:2437–2444. 80. Bless NM, Huber-Lang M, Guo RF, Warner RL, Schmal H, Czermak BJ, Shanley TP, Crouch LD, Lentsch AB, Sarma V, Mulligan MS, Friedl HP, Ward PA. Role of CC chemokines (macrophage inflammatory protein-1 beta, monocyte chemoattractant protein-1 RANTES) in acute lung injury in rats. J Immunol 2000; 2650–2659. 81. Nakos G, Kitsiouli EI, Tsangaris I, Lekka ME. Bronchoalveolar lavage fluid characterictics of early intermediate and late phases of ARDS. Alterations in leukocytes, proteins, PAF and surfactant components. Intensive Care Med 1998; 24:296–303. 82. Antonelli M, Bufi M, De Blasi RA, Crimi G, Conti G, Mattia C, Vivino G, Lenti L, Lombardi D, Dotta A, Detection of leukotrienes B4, C4 and of their isomers in artetial, mixed venous blood and bronchoalveolar lavage fluid from ARDS patients. Intensive Care Med 1989; 15:296– 301. 83. Sorensen J, Kald B, Togesson C, Lindahl M. Platelet-activating factor and phospolipase A2 in patients with septic shock and trauma. Intensive Care Med 1994; 20:555–561. 84. Vadas P. Elevated plasma phospholipase A2 levels: correlation with the hemodynamic and pulmonary changes in gram-negative septic shock. J Lab Clin Med 1984; 104:873–881. 85. Vadas P, Pruzanski W. Induction of group II phospholipase A2 expression and pathogenesis of the sepsis syndrome. Circulatory Shock 1993; 39:160–167. 86. Arbibe L, Vial D, Rosinski-Chupin I, Havet N, Huerre M, Vargaftig BB, Touqui L. Endotoxin induces expression of type II phospholipase A2 in macrophages during acute lung injury in guinea pigs: involvement of TNF-alpha in lipopolysaccharide-induced type II phospholipase A2 synthesis. J Immunol 1997; 159:391–400. 87. Rice KL, Duane PG, Mielke G, Sinha AA, Niewoehner DE. Calcium ionophores injure alveolar epithelial cells: relation to phospholipase activity. Am J Physiol 1990; 259:L439–450. 88. Tsukahara Y, Morisaki T, Horita Y, Torisu M, Tanaka M. Phospholipase A2 mediates nitric oxide production by alveolar macrophages and acute lung injury in pancreatitis. Ann Surg 1999; 229:385–392. 89. Furue S, Kuwabara K, Mikawaka K, Nishina K, Shiga M, Maekawa N, Ueno M, Chikazawa Y, Ono T, Hori Y, Matsukawa A, Yoshinaga M, Obara H. Crucial role of group IIA phospholipase A(2) in oleic acid-induced acute lung injury in rabbits. Am J Respir Crit Care Med 1999; 160:1292–1302. 90. Mikes PS, Polomski LL, Gee JB. Mepacrine impairs neutrophil response after acute lung injury in rats. Effects on neutrophil migration. Am Rev Respir Dis 1988; 138:1464–1470. 91. Sivan Y, Mor C, al-Jundi S, Newth CJ. Adult Respiratory distress syndrome in severely neutropenic children. Pediatr Pulmonol 1990; 8:104–108. 92. Vansteenkiste JF, Boogaerts MA. Adult respiratory distress syndrome in neutropenic leukemia patients. Blut 1989; 58:287–290. 93. Ognibene FP, Martin SE, Parker MM, Schlesinger T, Roach P, Burch C, Shelhamer JH, Parillo JE. Adult respiratory distress syndrome in patients with severe neutropenia. N Engl J Med 1986; 315:547–551. 94. Laufe MD, Simon RH, Flint A, Keller JB. Adult respiratory distress syndrome in neutropenic patients. Am J Med 1986; 80:1022–1026.

Acute respiratory distress syndrome

124

95. Maunder RJ, Hackman RC, Riff E, Albert RK, Springmeyer SC. Occurrence of the adult respiratory distress syndrome in neutropenic patients. Am Rev Respir Dis 1986; 133:313–316. 96. Tate RM, Repine JE. Neutrophils and the adult respiratory distress syndrome. Am Rev Respir Dis 1983; 128:552–559. 97. Weiss SJ. Tissue destruction by neutrophils [see comments]. N Engl J Med 1989; 320:365–376. 98. Wiggs BR, English D, Quinlam WM, Doyle NA, Hogg JC, Doerschuk CM. Contributions of capillary pathway size and neutrophil deformability to neutrophil transit through rabitt lungs. J Appl Physiol 1994; 77:463–470. 99. Lien DC, Wagner WW, Jr., Capen RL, Haslett C, Hanson WL, Hofmeister SE, Henson PM, Worthen GS. Physiological neutrophil sequestration in the lung: visual evidence for localization in capillaries. J Appl Physiol 1987; 62:1236–1243. 100. Hogg JC, McLean T, Martin BA, Wiggs B. Erythrocyte transit and neutrophil concentration in the dog lung. J Appl Physiol 1988; 65:1217–1225. 101. Doerschuk CM. The role of CD18-mediated adhesion in neutrophil sequestration induced by infusion of activated plasma in rabbits. Am J Respir Cell Mol Biol 1992; 7:140–148. 102. Inano H, English D, Doerschuk CM. Effect of zymosan-activated plasma on the deformability of rabbit polymorphonuclear leukocytes. J Appl Physiol 1992; 73:1370–1376. 103. Downey GP, Worthen GS. Neurophil retention in model capillaries: deformability, geometry, and hydrodynamic forces. J Appl Physiol 1988; 65:1861–1871. 104. Worthen GS, Schwab Bd, Elson EL, Downey GP. Mechanics of stimulated neutrophils: cell stiffening induces retention in capillaries. Science 1989; 245:183–186. 105. Selby C, Drost E, Wraith PK, McNee W. In vivo neutrophil sequestration within lungs of humans is determined by in vitro “filterability.” J Appl Physiol 1991; 71:1996–2003. 106. Motosugi H, Graham L, Noblitt TW, Doyle NA, Quinlan WM Li Y, Doerschuk CM. Changes in neutrophil actin and shape during sequestration induced by complement fragments in rabbits. Am J Pathol 1996; 149:963–973. 107. Kubo H, Graham L, Doyle NA, Quinlan WM, Hogg JC, Doerschuk CM. Complement fragment-induced release of neutrophils from bone marrow and sequestration within pulmonary capillaries in rabbits. Blood 1998; 92:283–290. 108. van Eeden SF, Kitagawa Y, Sato Y, Hogg JC. Polymorphonuclear leukocytes released form the bone marrow and acute lung injury. Chest 1999; 116:43S–46S. 109. Donnelly SC, MacGregor I, Zamani A, Gordon MW, Robertson CE, Steedman DJ, Little K, Haslett C. Plasma elastase levels and the development of the adult respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:1428–1433. 110. Chollet-Martin S, Jourdain B, Gibert C, Elbim C, Chastre J, Gougerot-Pocidalo MA, Interactions between neutrophils and cytokines in blood and alveolar spaces during ARDS. Am J Respir Crit Care Med 1996; 154:594–601. 111. Downey GP, Dong Q, Kruger J, Delhar S, Cherapanov V. Regulation of neutrophil activation in acute lung injury. Chest 1999; 116:46S–54S. 112. Abraham E, Arcaroli J, Shenkar R. Activation of extracellular signal-regulated kinases, NFkappa B, and cyclic adenosine 5′-monophosphate response element-binding protein in lung neutrophils occurs by differing mechanisms after hemorrhage or endotoxemia. J Immunol 2001; 166:522–530. 113. Arcaroli J, Yum HK, Kupfner J, Park JS, Abraham E. Role of p38MAP kinase in the development of acute lung injury. Clin Immunol 2001; 101:211–219. 114. Wang CY, Mayo MW, Korneluk RG, Goeddel DV, Baldwin AS, Jr. NF-kappaB antiapoptosis: induction of TRAF1 and TRAF2 and c-IAP1 and IAP2 to suppress caspase-8 activation. Science 1998; 281:1680–1683. 115. Parsey MV, Tuder RM, Abraham E, Neutrophils are major contributors to intraparenchymal lung IL-1 beta expression after hemorrhage and endotoxemia. J Immunol 1998; 160:1007–1013.

Pathogenesis of acute lung injury

125

116. Parsey MV, Kaneko D, Shenkar R, Abraham E. Neutrophil apoptosis in the lung after hemorrhage or endotoxemia: apoptosis and migration are independent of interleukin-1 beta. Chest 1999; 116:67S-68S. 117. Cines DB, Pollak ES, Buck CA, Loscalzo J, Zimmerman GA, McEver RP, Pober JS, Wick TM, Konkle BA, Schwartz BS, Barnathan ES, McCrae KR, Hug BA, Schmidt AM, Stern DM. Endothelial cells in physiology and in the pathophysiology of vascular disorders. Blood 1998; 91:3527–3561. 118. Fishman AP, Fishman MC, Freeman BA, Gimbrone MA, Rabinovitch M, Robinson D, Gail DB. Mechanisms of proliferative and obliterate vascular diseases. Insights from the pulmonary and systemic circulations. NHLBI Workshop summary. Am J Respir Crit Care Med 1998; 158:670–674. 119. Grau GE, Mili N, Lou JN, Morel DR, Ricou B, Lucas R Suter PM. Phenotypic and functional analysis of pulmonary microvascular endothelial cells from patients with acute respiratory distress syndrome. Lab Invest 1996; 74:761–770. 120. Zimmerman GA, Albertine KH, Carveth HJ, Gill EA, Grissom CK, Hoidal JR, Imaizumi T, Maloney CG, McIntyre TM, Michael JR, Orme JF, Prescott SM, Topharn MS. Endothelial activation in ARDS. Chest 1999; 116:18S–24S. 121. Donnelly SC, Haslett C, Dransfield I, Robertson CE, Carter DC, Ross JA, Grant IS, Tedder TF. Role of selectins in development of adult respiratory distress syndrome. Lancet 1994; 344:215–219. 122. Stricter RM, Kunkel SL, Acute lung injury: the role of cytokines in the elicitation of neutrophils. J Invest Med 1994; 42:640–651. 123. Mulligan MS, Miyasaka M, Ward PA. Protective effects of combined adhesion molecule blockade in models of acute lung injury. Proc Assoc Am Phys 1996; 108:198–208. 124. Zimmerman GA, Prescott SM, McIntyre TM. Endothelial cell interactions with granulocytes: tethering and signaling molecules. Immunol Today 1992; 13:93–100. 125. Zimmerman GA, McIntyre TM, Prescott SM. Adhesion and signaling in vascular cell-cell interactions. J Clin Invest 1997; 100:S3–5. 126. Modur V, Li Y, Zimmerman GA, Prescott SM, McIntyre TM. Retrograde inflammatory signaling from neutrophils to endothelial cells by soluble interleukin-6 receptor alpha. J Clin Invest 1997; 100:2752–145. 127. Baeuerle PA, Baichwal VR. NF-kappa B as frequent target for immunosuppressive and antiinflammatory molecules. Advances in Immunology 1997; 65:111–137. 128. Chen CC, Manning AM. Transcriptional regulation of endothelial cell adhesion molecules: a dominant role for NF-kappa B. Agents Actions (supp) 1995; 47:135–141. 129. Nathan C, Srimal S, Farber C, Sanchez E, Kabbash L, Asch A, Gailit J, Wright SD. Cytokineinduced respiratory burst of human neutrophils: dependence on extracellular matrix proteins and CD 11/CD 18 integrins. J Cell Biol 1989; 109:1341–1349. 130. Nauseef WM, De Alarcon P, Bale JF, Clark RA. Aberrant activation and regulation of the oxidative burst in neutrophils with Mol glycoprotein deficiency. J Immunol 1986; 137:636–642. 131. Berton G, Fumagalli L, Laudanna C, Sorio C. Beta 2 integrin-dependent protein tyrosine phosphorylation and activation of the FGR tyrosine kinase in human neutrophils. J Cell Biol 1994; 126:1111–1121. 132. Brenner B, Grassme HU, Muller C, Lang F, Speer CP, Gulbins E. L-selectin stimulates the neutral sphingomyelinase and induces release of ceramide. Exp Cell Res 1998; 243:123–128. 133. Brenner B, Weinmann S, Grassme H, Lang F, Linderkamp O, Gulbins E. L-selectin activates JNK via src-like tyrosine kinase and the small G- protein Rac. Immunology 1997; 92:214–219. 134. Laudanna C, Constantin G, Baron P, Scarpini E, Scarlato G, Cabrini G, Dechecchi C, Rossi F, Cassatella MA, Berton G. Sulfatides trigger increase of cytosolic free calcium and enhanced expression of tumor necrosis factor-alpha and interleukin-8 mRNA in human neutrophils. Evidence for a role of L-selectin as a signaling molecule. J Biol Chem 1994; 269:4021–4026.

Acute respiratory distress syndrome

126

135. Waddell TK, Fialkow L, Chan CK, Kishimoto TK, Downey GP. Potentiation of the oxidative burst of human neutrophils. A signaling role for L-selectin. J Biol Chem 1994; 269:18485– 18491. 136. Lindberg FP, Gresham HD, Schwarz E, Brown EJ. Molecular cloning of integrin-associated protein: an immunoglobulin family member with multiple membrane-spanning domains implicated in alpha v beta 3-dependent ligand binding. J Cell Biol 1993; 123:485–496. 137. Lipfert L, Haimovich B, Schaller MD, Cobb BS, Parsons JT, Brugge JS. Integrin-dependent phosphorylation and activation of the protein tyrosine kinase pp125FAK in platelets. J Cell Biol 1992; 119:905–912. 138. Dedhar S, Hannigan GE. Integrin cytoplasmic interactions and bidirectional transmembrane signalling. Curr Opin Cell Biol 1996; 8:657–669. 139. Hannigan GE, Leung-Hagesteijn C, Fitz-Gibbon L, Coppolino MG, Radeva G, Filmus J, Bell JC, Dedhar S. Regulation of cell adhesion and anchorage-dependent growth by a new beta 1integrin-linked protein kinase. Nature 1996; 379:91–96. 140. Walker DC, Behzad AR, Chu F. Neutrophil migration through preexisting holes in the basal laminae of alveolar capillaries and epithelium during streptococcal pneumonia. Microvasc Res 1995; 50:397–416. 141. Behzad AR, Chu F, Walker DC, Fibroblasts are in a position to provide directional information to migrating neutrophils during pneumonia in rabbit lungs. Microvasc Res 1996; 51:303–316. 142. Doerschuk CM, Winn RK, Coxson HO, Harlan JM. CD18-dependent and -independent mechanisms of neutrophil emigration in the pulmonary and systemic microcirculation of rabbits. J Immunol 1990; 144:2327–2333. 143. Hellewell PG, Young SK, Henson PM, Worthen GS, Disparate role of the beta 2-integrin CD 18 in the local accumulation of neutrophils in pulmonary and cutaneous inflammation in the rabbit. Am J Respir Cell Mol Biol 1994; 10:391–398. 144. Qin L, Quinlan WM, Doyle NA, Graham L, Sligh JE, Takei F, Beaudet AL, Doerschuk CM. The roles of CD11/CD18 and ICAM-1 in acute Pseudomonas aeruginosa-induced pneumonia in mice. J Immunol 1996; 157:5016–5021. 145. Ramamoorthy C, Sasaki SS, Su DL, Sharar SR, Harlan JM, Winn RK. CD 18 adhesion blockade decreases bacterial clearance and neutrophil recruitment after intrapulmonary E. coli, but not after S. aureus. J Leukoc Biol 1997; 61:167–172. 146. Motosugi H, Quinlan WM, Bree M, Doerschuk CM. Role of CD1 1b in focal acid-induced pneumonia and contralateral lung injury in rats. Am J Respir Crit Care Med 1998; 157:192–198. 147. Sherman MP, Johnson JT, Rothlein R, Hughes BJ, Smith CW, Anderson DC. Role of pulmonary phagocytes in host defense against group B streptococci in preterm versus term rabbit lung. J Infect Dis 1992; 166:818–826. 148. Keeney SE, Mathews MJ, Haque AK, Rudloff HE, Schmalsteig FC. Oxygen-induced lung injury in the guinea pig proceeds through CD18-independent mechanisms. Am J Respir Crit Care Med 1994; 149:311–319. 149. Mulligan MS, Wilson GP, Todd RF, Smith CW, Anderson DC, Varani J, Issekutz TB, Miyasaka M, Tamatani T, Myasaka M, et al. Role of beta 1, beta 2 integrins and ICAM-1 in lung injury after deposition of IgG and IgA immune complexes [published erratum appears in J Immunol 1993 Jun 1;150(11):5209]. J Immunol 1993; 150:2407–2417. 150. Mizgerd JP, Meek BB, Kutkoski GJ, Bullard DC, Beaudet AL, Doerschuk CM. Selectins and neutrophil traffic: margination and Streptococcus pneumoniae-induced emigration in murine lungs. J Exp Med 1996; 184:639–645. 151. Doerschuk CM, Mizgerd JP, Kubo H, Qin L, Kumasaka T. Adhesion molecules and cellular biomechanical changes in acute lung injury: Giles F. Filley Lecture. Chest 1999; 116:37S-43S. 152. Wells S, Sissons M, Hasleton PS. Quantitation of pulmonary megakaryocytes and fibrin thrombi in patients dying from burns. Histopathology 1984; 8:517–527.

Pathogenesis of acute lung injury

127

153. Tomashefski JF, Jr. Pulmonary pathology of the adult respiratory distress syndrome. Clin Chest Med 1990; 11:593–619. 154. Chapman HA, Allen CL, Stone OL. Abnormalities in pathways of alveolar fibrin turnover among patients with interstitial lung disease. Am Rev Respir Dis 1986; 133:437–443. 155. Bachofen M, Weibel ER. Structural alterations of lung parenchyma in the adult respiratory distress syndrome. Clin Chest Med 1982; 3:35–56. 156. Sitrin RG, Brubaker PG, Fantone JC. Tissue fibrin deposition during acute lung injury in rabbits and its relationship to local expression of procoagulant and fibrinolytic activities. Am Rev Respir Dis 1987; 135:930–936. 157. Idell S, Gonzalez K, Bradford H, MacArthur CK, Fein AM, Maunder RJ, Garcia JG, Griffith DE, Weiland J, Martin TR, et al. Procoagulant activity in bronchoalveolar lavage in the adult respiratory distress syndrome. Contribution of tissue factor associated with factor VII. Am Rev Respir Dis 1987; 136:1466–1474. 158. Idell S, James KK, Levin EG, Schwartz BS, Manchanda N, Maunder RJ, Martin TR, McLarty J, Fair DS. Local abnormalities in coagulation and fibrinolytic pathways predispose to alveolar fibrin deposition in the adult respiratory distress syndrome. J Clin Invest 1989; 84:695–705. 159. Zapol WM, Jones R. Vascular components of ARDS. Clinical pulmonary hemodynamics and morphology. Am Rev Respir Dise 1987; 136:471–474. 160. Gross TJ, Simon RH, Kelly CJ, Sitrin RG. Rat alveolar epithelial cells concomitantly express plasminogen activator inhibitor-1 and urokinase. Am J Physiol 1991; 260;L286–295. 161. Simon RH, Gross TJ, Edwards JA, Sitrin RG. Fibrin degradation by rat pulmonary alveolar epithelial cells. Am J Physiol 1992; 262:L482–488. 162. Simon DI, Rao NK, Xu H, Wei Y, Majdic O, Ronne E, Kobzik L, Chapman HA. Mac-1 (CD11b/CD 18) and the urokinase receptor (CD87) form a functional unit on monocytic cells. Blood 1996; 88:3185–3194. 163. Hasegawa T, Sorensen L, Ooi H, Marshall BC. Decreased intracellular iron availability suppresses epithelial cell surface plasmin generation. Transcriptional and post-transcriptional effects on u-PA and PAI-1 expression. Am J Respir Cell Mol Biol 1999; 21:275–282. 164. Hasegawa T, Sorensen L, Dohi M, Rao NV, Hoidal JR, Marshall BC. Induction of urokinasetype plasminogen activator receptor by IL-1 beta. Am J Respir Cell Mol Biol 1997; 16:683–692. 165. Toews GB. Cellular alternations in fibroproliferative lung disease. Chest 1999; 116:112S116S. 166. Horton MR, Olman MA, Noble PW. Hyaluronan fragments induce plasminogen activator inhibitor-1 and inhibit urokinase activity in mouse alveolar macrophages: a potential mechanism for impaired fibrinolytic activity in acute lung injury. Chest 1999; 116:17S. 167. Barazzone C, Belin D, Piguet PF, Vassalli JD, Sappino AP. Plasminogen activator inhibitor-1 in acute hyperoxic mouse lung injury. J Clin Invest 1996; 98:2666–2673. 168. Eitzman DT, McCoy RD, Zheng X, Fay WP, Shen T, Ginsburg D, Simon RH. Bleomycininduced pulmonary fibrosis in transgenic mice that either lack or overexpress the murine plasminogen activator-1 gene. J Clin Invest 1996; 97:232–237. 169. Noble PW, McKee CM, Cowman M, Shin HS. Hyluronan fragments activate an NF-kappa B/I-kappa B alpha autoregulatory loop in murine macrophages. J Exp Med 1996; 183:2373– 2378. 170. Hallgren R, Samuelsson T, Laurent TC, Modig J. Accumulation of hyaluronan (hyaluronic acid) in the lung in adult respiratory distress syndrome. Am Rev Respir Dis 1989; 139:682–687. 171. Savani RC, Hou G, Liu P, Wang C, Simons E, Grimm PC, Stern R, Greenberg AH, DeLisser HM, Khalil N. A role for hyaluronan in macrophage accumulation and collagen deposition after bleomycin-induced lung injury. Am J Respir Cell Mol Biol 2000; 23:475–484. 172. Idell S, Koenig KB, Fair DS, Martin TR, McLarty J, Maunder RJ. Serial abnormalities of fibrin turnover in evolving adult respiratory distress syndrome. Am J Physiol 1991; 261:L240– 248.

Acute respiratory distress syndrome

128

173. Perez RL, Ritzenthaler JD, Roman J. Transcriptional regulation of the interleukin-1beta promoter via fibrinogen engagement of the CD 18 integrin receptor. Am J Respir Cell Mol Biol 1999; 20:1059–1066. 174. groeneveld AB, Kindt I, Raijmakers PG, Hack CE, Thijs LG. Systemic coagulation and fibrinolysis in patients with or at risk for the adult respiratory distress syndrome. Thrombosis Haemostasis 1997; 78:1444–1449. 175. Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, Lopez-Rodriguez A, Steingrub JS, Garber GE, Helterbrand JD, Ely EW, Fisher CJ, Jr., Recombinant human protien CWEiSSsg. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344:699–709. 176. Bernard GR, Ely EW, Wright TJ, Fraiz J, Stasek JE, Jr., Russell JA, Mayers I, Rosenfeld BA, Morris PE, Yan SB, Helterbrand JD. Safety and dose relationship of recombinant human activated protein C for coagulopathy in severe sepsis. Crit Care Med 2001; 29:2051–2059. 177. Esmon CT. Introduction: are natural anticoagulants candidates for modulating the inflammatory response to endotoxin? Blood 2000; 95:1113–1116. 178. Sabharwal AK, Bajaj SP, Ameri A, Tricomi SM, Hyers TM, Dahms TE, Taylor FB, Jr., Bajaj MS. Tissue factor pathway inhibitor and von Willebrand factor antigen levels in adult respiratory distress syndrome and in a primate model of sepsis. Am J Respir Cri Care Med 1995; 151:758–767. 179. Welty-Wolf KE, Carraway MS, Idell S, Ortel TL, Ezban M, Piantadosi CA. Tissue factor in experimental acute lung injury. Semin Hematol 2001; 38:35–38. 180. Gando S, Kameue T, Morimoto Y, Matsuda N, Hayakawa M, Kemmotsu O. Tissue factor production not balanced by tissue factor pathway inhibitor in sepsis promotes poor prognosis [comment]. Crit Care Med 2002; 30:1729–1734. 181. Jarrar D, Kuebler JF, Rue LW III, Matalon S, Wang P, Bland KI, Chaudry IH. Alveolar macrophage activation after trauma-hemorrhage and sepsis is dependent on NF-kappaB and MAPK/ERK mechanisms. Am J Physiol (Lung Cellular & Molecular Physiology) 2002; 283:L799–L805. 182. Kurdowska AK, Geiser TK, Alden SM, Dziadek BR, Noble JM, Nuckton TJ, Matthay MA. Activity of pulmonary edema fluid interleukin-8 bound to alpha(2)-macroglobulin in patients with acute lung injury. Am J Physiol (Lung Cellular & Molecular Physiology) 2002; 282:L1092–L1098. 183. McDowell SA, Mallakin A, Bachurski CJ, Toney-Earley K, Prows DR, Bruno T, Kaestner KH, Witte DP, Melin-Aldana H, Degen SJ, Leikauf GD, Waltz SE. The role of the receptor tyrosine kinase Ron in nickel-induced acute lung injury. Am J Respir Cell Mol Biol 2002; 26:99–104. 184. Meduri GU, Tolley EA, Chrousos GP, Stentz F. Prolonged methylprednisolone treatment suppresses systemic inflammation in patients with unresolving acute respiratory distress syndrome: evidence for inadequate endogenous glu-cocorticoid secretion and inflammationinduced immune cell resistance to glucocorticoids. Am J Respir Crit Care Med 2002; 165:983– 991. 185. Muller AM, Cronen C, Muller KM, Kirkpatrick CJ. Heterogeneous expression of cell adhesion molecules by endothelial cells in ARDS. J Pathol 2002; 198:270–275. 186. Wu Y, Singer M, Thouron F, Alaoui-El-Azher M, Touqui L. Effect of surfactant on pulmonary expression of type IIA PLA(2) in an animal model of acute lung injury. Am J Physiol (Lung Cellular & Molecular Physiology) 2002; 282:L743–L750. 187. Zhang H, Downey GP, Suter PM, Slutsky AS, Ranieri VM. Conventional mechanical ventilation is associated with bronchoalveolar lavage-induced activation of polymorphonuclear leukocytes: a possible mechanism to explain the systemic consequences of ventilator-induced lung injury in patients with ARDS. Anesthesiology 2002; 97:1426–1433.

7 Pathogenesis of Acute Lung Injury Clinical Studies LORRAINE B.WARE Vanderbilt University School of Medicine Nashville, Tennessee, U.S.A. TIMOTHY W.EVANS Imperial College School of Medicine and Royal Brompton Hospital London, England

I. Introduction Over the past three decades, experimental studies have contributed substantially to an increased understanding of the pathogenesis of acute lung injury (ALI) and its extreme manifestation, the acute respiratory distress syndrome (ARDS) (see Chapter 5). However, many investigators have employed animal models that fail to accurately reproduce the heterogeneity of human ALI/ARDS, and which are difficult to sustain for the protracted periods needed to permit an assessment of the effects of mechanical ventilation and other life support measures upon the evolution of the syndrome. Clinical studies are therefore critical to expanding our understanding of ALI/ARDS, and those that have contributed significantly to progress in this area are summarized herein. Clinical trials of putative treatment strategies for ALI/ARDS are discussed elsewhere.

II. Methodology Used in Clinical Studies The methodology applicable in clinical studies of the pathogenesis of ALI/ ARDS is limited when compared to that which can be applied in animals. Nevertheless, a variety of useful and innovative techniques have been developed (Table 1). A major focus of clinical studies has been the measurement of biological markers of lung inflammation and injury in blood, edema fluid and bronchoalveolar lavage (BAL) fluid (1), which has improved our understanding of the pathophysiology of ALI/ARDS. Measurement of biological markers may have prognostic utility both in patients with established lung injury and in “at-risk” patient populations.

Acute respiratory distress syndrome

130

A. Bronchoalveolar Lavage Bronchoalveolar lavage has been widely used to characterize the cellular content and fluid composition of the alveolar spaces. BAL affords several advantages over other techniques. First, it can be performed at any time point and does not require that pulmonary edema be present for sampling. Thus, BAL has also been carried out in patient populations at risk for ALI/ ARDS in the hope of identifying soluble factors that might predict the development of lung injury. Second, a large volume of lavage is returned, allowing measurement of multiple mediators in one patient. Third, BAL is probably the most reliable technique for sampling the cell population present in alveolar lining fluid. By contrast, BAL also has several disadvantages. First, although BAL appears to be safe in patients with ALI/ARDS, it clearly represents an invasive procedure (2). Second, the method employed introduces, by definition, dilution that is variable and difficult to quantify accurately. The degree of dilution may render undetectable some factors that are present in very low concentrations. B. Sampling Pulmonary Edema Fluid ALI/ARDS is characterized by the presence of high-permeability pulmonary edema. Alveolar fluid can be sampled from patients early in the course of ALI/ARDS using a standard tracheal suction catheter wedged into a distal airway. The application of gentle suction yields a few milliliters of undiluted alveolar edema that can be collected in a suction trap (3, 4). This technique is less invasive than BAL and obviates the problems associated with sample dilution. Concentrations of relevant mediators can be compared directly to those in plasma sampled simultaneously. If serial samples of edema fluid are obtained, their protein content can be used to calculate the rate of alveolar fluid clearance, an indicator of alveolar epithelial fluid transport function (3). Patients with hydrostatic pulmonary edema can serve as appropriate controls (5). However, the technique is usually only successful early in the course of ALI/ARDS when alveolar flooding is present. Indeed, the optimal time to sample pulmonary edema fluid is 15–30 minutes after endotracheal intubation. At this point, serial samples can frequently be obtained for up to 12 hours, but afterwards only patients with severe unremitting alveolar flooding have pulmonary edema fluid that can be aspirated. Further, the technique cannot be used in patients at risk for ALI/ARDS or in those in whom overt pulmonary edema has given way to fibroproliferation. C. Exhaled Gas Collection of exhaled gas from patients with ALI/ARDS is a noninvasive means by which the volatile and aerosolized compounds excreted by the lung may be quantified. Most studies have analyzed exhaled breath condensate by cooling the expired breath as it passes through a Teflon-coated tube. This technique can be used in ventilated or nonventilated subjects. A variety of molecules have been measured in this fashion, including H2O2 (6– 8), markers of lipid peroxidation (9), and, more recently, proteins such as hepatocyte growth factor (10). The feasibility of measuring a variety of cytokines

Pathogenesis of acute lung injury: clinical studies

131

was demonstrated in a study of patients with a variety of lung diseases (11). This technique has several potential advantages over BAL or edema fluid sampling. First, it is noninvasive and can be performed easily on ventilated patients or healthy controls. Second, repeated measures can be made over time. Disadvantages include the lack of evidence for the origin of the aerosolized molecules (e.g., nasopharynx vs. airway versus alveoli), the possible effect of evaporation on the concentration of relevant mediators (12), and the extent to which derived values depend upon variations in pulmonary or bronchial blood flow. Direct analysis of a variety of exhaled gases in patients with ALI/ ARDS has also been reported, including nitric oxide measured by chemiluminescence (13) and gas chromatographic analysis of exhaled gas after charcoal absorption (14). D. Imaging A variety of imaging techniques have been used in patients with ALI/ARDS to enhance understanding of the clinical syndrome. These include chest radiography, computed tomographic (CT) scanning, and positron emission tomography (PET). Although basically descriptive, imaging studies have made important contributions to our understanding of ALI/ARDS, particularly with regard to the spatial distribution of lung injury and establishing anatomical and physiological differences between patients with lung injury attributable to direct and distant (i.e., indirect) pulmonary insults. (A complete discussion of the radiographic manifestations of ALI/ARDS is provided in Chapter 3.)

Table 1 Methodologies Used in Clinical Studies of Acute Lung Injury and the Acute Respiratory Distress Syndrome Methodology Bronchoalveolar lavage

Advantages

Disadvantages

Can be done at any time during the course of ALI/ARDS

Variable dilution factor Invasive

Can be done in at-risk patients

Not feasible if severe hypoxemia is present

Can be done serially in same patient Can sample the alveolar cell population Pulmonary edema fluid sampling

No dilutional factor

Only feasible when alveolar flooding is present

Simultaneous edema fluid and plasma levels of mediators can be compared

Not all patients will have aspirable edema fluid

Relatively noninvasive Can be used to calculate rate of alveolar fluid clearance

Acute respiratory distress syndrome

Blood sampling

Relatively noninvasive

132

Samples only the intravascular compartment

Can be done at any time during the course of ALI/ARDS Completely noninvasive

Exhaled breath condensate

Can be done at any time during the course of ALI/ARDS

Source of exhaled molecules unclear (nasopharynx, airway, alveolus) Evaporation may affect concentrations Relationship to alveolar lining fluid unclear

Exhaled gas analysis

Completely noninvasive

Only useful for volatile compounds May be technically difficult

Imaging Chest radiography

Completely noninvasive Provides information about distribution of disease not available via other modalities

Patient transport may be required Relatively descriptive

Computed tomography PET scanning Extravascular lung water measurements

Provides better quantification of Does not reliably distinguish between degree of pulmonary edema than hydrostatic and increased permeability chest radiograph pulmonary edema Can be used to guide fluid management

Invasive (requires PA catheter) May be confounded by high cardiac output

Lung microvascular permeability

Can quantify protein permeability of the alveolar capillary barrier

May require patient transport Findings differ depending on tracer used

Relatively noninvasive Wasted ventilation

Completely noninvasive

Does not differentiate cause of lung microvascular destruction/obstruction

Quantifies degree of lung microvascular destruction/obstruction Genetic studies

Has potential to identify genetic predispositions for the development of ALI/ARDS

Given the large number of factors that probably determine whether someone develops ALI/ARDS, it may be difficult to pinpoint specific mutations that increase risk

Lung biopsy

Allows histological and

Very invasive

Pathogenesis of acute lung injury: clinical studies

133

ultrastructural analysis at various Rarely done for diagnosis timepoints during ALI/ARDS Autopsy

Allows histological and ultrastructural analysis of entire lung

Samples only sickest patients who die from their illness Not done routinely

ALI/ARDS, acute lung injury and the acute respiratory distress syndrome; PET, positron emission tomography; PA, pulmonary arterial.

E. Physiological Measurements Several physiological measurements have been used in clinical studies of patients with ALI/ARDS. Measurements of extravascular lung water using the thermal-dye double indicator technique have been used to quantify the severity of pulmonary edema formation (15–20a). The technique requires insertion of a pulmonary artery catheter, and results may be confounded by a high cardiac output. Further, it does not reliably distinguish between hydrostatic and permeability pulmonary edema. Nevertheless, a decrease in extra-vascular lung water as ALI/ARDS evolves clinically is associated with better outcome (19–21) and in a recent large retrospective study of 373 critically ill patients, extravascular lung water was significantly higher in nonsurvivors (20a). New techniques are under development to quantify lung water by CT imaging (22). A variety of radionuclide compounds have been used to make noninvasive measurements of lung microvascular permeability (reviewed in Ref. 23). Quantification of lung microvascular permeability to proteins can be achieved using an intravenously injected labeled tracer protein such as 67gallium-transferrin, but immobile detection systems such as positron emission tomography or gamma camera are needed. Portable techniques more suited to mechanically ventilated patients have also been employed, including mobile gamma cameras and microscintillation counter probes (23). The addition of a blood pool marker reduces confounding by changes in pulmonary blood volume (24). Finally, simple lung function studies may be performed in patients with ALI/ARDS (25, 26). Indeed, metabolic measurements of wasted ventilation have been used to quantify physiological dead space (27), which represents a simple, noninvasive quantification of lung microvascular destruction or obstruction. Wasted ventilation was elevated in patients with ALI/ARDS and was an independent predictor of mortality by multivariate analysis. Although this study was not able to determine the etiology of wasted ventilation (capillary destruction versus reversible or irreversible obstruction), it does provide new evidence for the critical importance of capillary injury in the pathogenesis and outcome of ALI/ARDS. F. Genetic Analysis Determination of genetic factors that influence susceptibility to, and outcome from, ALI/ARDS is a promising new area of research that is still in its infancy. Preliminary reports suggest that polymorphisms in the surfactant protein-B gene may affect the risk of developing ALI/ARDS (28, 29). Polymorphisms in the promoter region of the IL-6 gene

Acute respiratory distress syndrome

134

are associated with a poorer outcome in patients with acute respiratory failure associated with the systemic inflammatory response syndrome (30). (For a comprehensive discussion see Chapter 13.) G. Autopsy and Lung Biopsy Classic reports by Bachofen and Weibel defined the histopathological and ultrastructural changes typical in ALI/ARDS (31, 32). However, autopsy studies have inherent limitations in that only the most advanced and severe disease can be studied. Further, autopsies are obtained less frequently today than was the case in past decades. Although lung biopsy is occasionally done for diagnosis in patients with unexplained acute respiratory failure, the procedure is too invasive and too infrequent to be routinely used in ALI/ARDS.

III. The Alveolar Capillary Membrane A. Endothelium Injury and activation of the lung microvascular endothelium is a critical component of ALI/ARDS (Table 2) and was defined in early, ultrastructural studies (31, 32). Subsequently, simultaneous measurements of the protein concentration of pulmonary edema fluid and plasma confirmed that increased endothelial permeability, with an edema fluid-to-plasma protein ratio of >0.75, is present in early ALI/ARDS (3, 4, 33–35). By contrast, patients with hydrostatic pulmonary edema have an edema fluid-to-plasma protein ratio of <0.65 (5, 33, 35). Studies using radiolabeled tracer proteins confirmed that

Table 2 Evidence from Clinical Studies for Endothelial Activation/ Injury in Acute Lung Injury and Acute Respiratory Distress Syndrome Evidence Increased permeability

As determined by: Pulmonary edema fluid sampling Bronchoalveolar lavage protein levels Ultrastructural studies Measurements of lung microvascular permeability

Release of soluble markers of activation/injury

Increased plasma endothelin-1 levels Increased plasma levels of adhesion molecules such as ICAM-1, P-selectin, E-selectin Increased plasma levels of von Willebrand factor antigen

Increased dead space

Bedside measurement of wasted ventilation

Pathogenesis of acute lung injury: clinical studies

135

ICAM-1, intercellular adhesion molecule-1.

lung microvascular permeability is increased in patients with ALI/ARDS compared to those with hydrostatic pulmonary edema (36). Focal, reversible gap formation between endothelial cells is thought to occur (37). A number of markers of endothelial cell injury and activation have been measured in biological samples taken from patients with ALI/ARDS. Endothelin-1 (ET-1) is a vasoconstrictor pep tide released by the injured endothelium (38–40). Several studies have measured increased plasma levels of ET-1 in patients with ALI/ARDS, due both to increased production and decreased metabolism by the pulmonary vasculature (41–43). In autopsy studies of lungs taken from patients who died with ARDS, ET-1 immunostaining was diffusely increased in the vascular endothelium, alveolar macrophages, smooth muscle, and airway epithelium (44). Whether ET-1 is a marker of endothelial injury, has pro-inflammatory action (43), or contributes to increased pulmonary vascular resistance in ALI/ARDS is unclear. The specific endothelin receptor antagonist bosentan, recently shown to have some efficacy in primary pulmonary hypertension (45), may stimulate new studies in this area. The degree of systemic endothelial activation and injury at the onset of acute lung injury may also be an important determinant of outcome in

Figure 1 Boxplot summary of pulmonary edema fluid and plasma levels of VWF versus hospital survival in 51 patients with ALI/ARDS. Plasma VWF was significantly higher in patients that did not survive to hospital

Acute respiratory distress syndrome

136

discharge. Horizontal line represents the median, box encompasses the 25th to 75th percentile and error bars encompass the 10th to 90th percentile. (From Ref. 46.) patients with ALI/ARDS. Von Willebrand factor antigen (VWF) is a high molecular weight antigen produced by endothelial cells and to a lesser extent by platelets. In the setting of endothelial injury or activation, VWF is released from preformed stores into the circulation. In a study of 51 patients with early ALI/ARDS, plasma VWF was significantly higher in those who failed to survive and was an independent predictor of in-hospital mortality by multivariate analysis (Fig. 1) (46). Higher plasma VWF levels were also associated with a shorter duration of unassisted ventilation (Fig. 2). Although VWF levels were higher in patients with sepsis, the predictive value for mortality was preserved in those patients with ALI/ARDS without sepsis. The results of this study complement those of other investigations who have reported that plasma VWF levels may also have some value in predicting which patients are likely to develop ARDS among those at risk (Table 3). Further evidence for the importance of destruction of the lung microvascular bed in ALI/ARDS has recently been provided by a study of

Figure 2 Boxplot summary of plasma levels of VWF versus the duration of unassisted ventilation in 51 patients with ALI/ARDS. Patients were divided into two groups: those that had ≥5 days

Pathogenesis of acute lung injury: clinical studies

137

of unassisted ventilation during the first 28 days after enrollment and those that had <5 days of unassisted ventilation during the first 28 days after enrollment. Plasma VWF levels were significantly higher in patients with <5 days of unassisted ventilation. Horizontal line represents the median, box encompasses the 25th to 75th percentile and error bars encompass the 10th to 90th percentile, (From Ref. 46.) Table 3 Summary of Studies of von Willebrand Factor Antigen in Patients at Risk for or with Established Acute Lung Injury and Acute Respiratory Distress Syndrome Year Patient number 1982 100 ARDS

Description of patients

Plasma VWF (mean)

ALI from all causes ~500%

Summary of findings

VWF was higher in patients with moderate or severe acute lung injury than those with mild acute lung injury or non-lung injury critical illness.

Ref.

135

1989 35 at risk At risk: all causes 10 ARDS except trauma ARDS: all causes except trauma

370% (at risk) 458% (ARDS)

VWF was significantly higher in ARDS 136 compared to at risk. VWF correlated

1990 45 at risk Nonpulmonary sepsis with no evidence for ALI at enrollment

338% (at risk, no ARDS) 588% (at risk, +ARDS)

VWF was significantly higher in at-risk 137 patients who went on to develop ARDS. VWF≥450% was 87% sensitive and 77% specific for development of ARDS. VWF≥450% had an 80% positive predictive value for nonsurvivors.

1995 96 at risk At risk: all causes, many may have already had some ALI

Not reported

VWF was not predictive of 138 development of ARDS. However, some degree of acute lung injury may have been present in many at-risk patients.

VWF did inversely with not predict the development of ARDS. However, most at-risk patients appeared to have some degree of acute lung injury.

Acute respiratory distress syndrome

138

1995 21 at risk 22 ARDS

At risk: all causes, many may have had some degree of ALI ARDS: all causes

1998 25 at risk

All patients 632% (at had sepsis, risk and many also had ARDS) some degree of ALI

VWF was significantly higher in nonsurvivors and was also an independent predictor of mortality and multiorgan dysfunction.

1999 15 at risk 18 ARDS

At risk: all causes, many had some degree of ALI ARDS: all causes

269% (at risk) 425% (ARDS)

Difference in VWF between at-risk and ARDS 141 did not reach statistical significance. VWF did not predict the development of ARDS. However, many at-risk patients had some degree of acute lung injury.

2001 51 ALI/ ARDS

ARDS: all causes

251% (median, ARDS)

VWF was significantly higher in plasma compared to pulmonary edema fluid. VWF was an independent predictor of mortality by multivariate analysis. Plasma VWF≥450% had an 83% positive predictive value for hospital mortality.

297% (at risk) 375% (ARDS)

Difference in VWF between at-risk and ARDS 139 did not reach statistical significance. VWF was not different between survivors and nonsurvivors in either group. Predictive value of VWF for the development of ARDS was not reported. Plasma was taken 1–4 days after onset of illness. 140

46

VWF, von Willebrand factor antigen; ALI, acute lung injury; ARDS, acute respiratory distress syndrome. Source: Adapted from Ref. 46.

wasted ventilation in mechanically ventilated patients with ALI/ARDS (discussed above) (27). B. Epithelium Increasingly, the critical role of injury to the alveolar epithelium in the pathogenesis of ALI/ARDS has been recognized (47), although ultrastructural studies as early as 1977 showed evidence of substantial alveolar epithelial injury and necrosis (31). The alveolar epithelium has multiple functions that can be impaired in ALI/ARDS. First, in the normal lung, the alveolar epithelium forms a barrier to alveolar flooding. This barrier is breached in patients with early ALI/ARDS either due to injury and necrosis of alveolar epithelial cells or as a result of excessive interstitial pressure due to accumulation of fluid in the interstitium from increased endothelial permeability. Widespread alveolar epithelial injury and necrosis is detectable in the lungs of patients dying from ARDS (31, 32). The alveolar epithelium also acts as a barrier to the influx of bacteria from the alveolar space into the circulation. In experimental pneumonia, breakdown of the alveolar epithelial barrier leads to bacteremia and sepsis (48).

Pathogenesis of acute lung injury: clinical studies

139

Figure 3 Plot of hospital mortality of two groups of patients with acute lung injury or the acute respiratory distress syndrome: those with maximal alveolar fluid clearance (≥14%/h) and those with impaired or submaximal alveolar fluid clearance (<14%/h). Columns represent percent hospital mortality in each group. Hospital mortality of patients with maximal alveolar fluid clearance was significantly less (p<0.02). N=number of patients. (From Ref. 3.) Alveolar epithelial type II cells produce both the protein and lipid components of surfactant. Surfactant function is inhibited in patients with ALI/ARDS, which probably contributes to alveolar collapse and hypoxemia, and abnormalities in both the lipid and protein components have been described (49–52). Injury to alveolar epithelial type II cells may contribute to impaired surfactant component production, but metabolism is also altered. Moreover, surfactant components may be inactivated by proteolysis (53) or through the flooding of the alveolar space with serum proteins (49). The alveolar epithelium is responsible for the active transport of fluid and solute out of the alveolar space when alveolar flooding has occurred. Removal of fluid is driven by the active transport of sodium from the alveolar space to the interstitium by alveolar epithelial type II cells. In early ALI/ARDS, clearance of fluid from the alveolar space is impaired in the majority of patients, a finding with adverse prognostic significance (3, 4) (Fig. 3). Similar findings have been reported in patients with reperfusion lung injury after lung transplantation (54). By contrast, the majority of patients with hydrostatic pulmonary edema have intact alveolar fluid clearance, and many have very rapid rates of

Acute respiratory distress syndrome

140

clearance (5) (Fig. 4). The etiology of impaired alveolar fluid clearance in ALI/ARDS has not been definitively determined, but is probably most closely related to the degree of injury to the tight alveolar epithelial barrier and the alveolar epithelial type II cell. (An indepth discussion of both experimental and clinical studies of the resolution of alveolar edema is provided in Chapter 14.) Because of the apparent importance of injury to the alveolar epithelium in determining outcome from ALI/ARDS, a biological marker specific to epithelial injury would undoubtedly be useful. HTI56, an integral apical plasma membrane protein of human alveolar epithelial type I cells has been thus described (55). Both plasma and pulmonary edema fluid levels of HTI56 were higher in patients with ALI/ARDS compared to those with hydrostatic edema (56). Preliminary studies of this marker indicate that levels correlate with alveolar epithelial fluid transport function, another marker of injury (Fig. 5). Finally, a putative marker of injury to alveolar epithelial type II cells has also been proposed recently (57). KL-6 is an epithelial mucin expressed mainly on type II cells. In a preliminary study, concentrations of KL-6 were significantly higher in the lung epithelial lining fluid of patients with ARDS compared to healthy controls (57). However, KL-6 may also be produced by airway epithelial cells and thus lacks anatomical specificity for the alveolar space. Surfactant proteins have been proposed as markers of injury to alveolar epithelial type II cells. In two different studies, BAL fluid (51) and pulmonary edema fluid (52) levels of surfactant protein-D (SP-D) were lower in patients with ALI/ARDS when compared to controls. Moreover,

Figure 4 Comparison of rates of alveolar fluid clearance in two groups of patients: 65 mechanically ventilated patients with severe hydrostatic pulmonary edema and 79 mechanically ventilated patients with acute lung

Pathogenesis of acute lung injury: clinical studies

141

injury or the acute respiratory distress syndrome. Columns represent the percentage of patients in each group with three categories of alveolar fluid clearance: impaired (<3%/h), submaximal (≥3%/h, <14%/h), or maximal (≥14%/h). (Modified from Ref. 3.)

Figure 5 Boxplot summary of levels of HTI56, an apical plasma membrane protein specific for human alveolar epithelial type I cells in 13 patients with acute lung injury or the acute respiratory distress syndrome. Patients were divided into two groups: those with impaired alveolar fluid clearance (<3%/h) and those with intact alveolar fluid clearance (≥3%/h). Horizontal line represents the median, box encompasses the 25th to 75th percentile and error bars encompass the 10th to 90th percentile.

Acute respiratory distress syndrome

142

lower SP-D levels were associated with higher mortality in the ARDS patients. There are several possible explanations for the decline in SP-D in the alveolus when lung injury is severe. Decreased levels might reflect injury to alveolar epithelial type II cells, downregulation of production due to inflammatory mediators or bacterial products, or increased degradation (52). Whether decreased SP-D has pathogenetic consequences or is merely a marker of lung injury is unclear. Decreased SP-D levels might have more of an impact on host defense than on alveolar surface tension since the primary role for SP-D is in host defense rather than stabilization of surfactant (58).

IV. Specific Inflammatory Cells, Pathways, and Mediators A. Neutrophils The neutrophil has long been implicated as a major effector cell in the pathogenesis of ALI/ARDS (59). Thus, neutrophil influx into the airspaces is a histological hallmark of early clinical ALI/ARDS (31, 32). Pulmonary edema fluid and BAL fluid from ALI/ARDS patients also have a predominance of neutrophils (34, 60, 61) and resolution of BAL fluid neutrophilia is a good prognostic indicator (62). Labeled autologous neutrophils localize to the lung when reinfused into patients with ALI/ARDS (63). However, it is worth noting that ARDS can occur in patients with neutropenia (64), suggesting that neutrophils are not absolutely required for the clinical syndrome of acute lung injury. Further, neutrophil accumulation may have important host defense functions. Multiple mechanisms contribute to neutrophil influx into the lung in ALI/ARDS. Due to its large size, the neutrophil must deform to pass through the pulmonary capillary bed (65). A number of cytokines and chemoattractants commonly implicated in ALI/ARDS can render neutrophils stiffer and less able to deform experimentally, hindering their passage through the pulmonary capillary bed (65–68). In a recent study of circulating neutrophils isolated from patients with sepsis, septic shock, or ARDS, both activated and passive neutrophils had decreased deformability as measured by a novel micropipette method compared to those from critically ill patients without sepsis or ARDS (69). Decreased neutrophil deformability was associated with higher circulating levels of proinflammatory cytokines such as TNF-α. Vascular retention of neutrophils is also mediated by the interaction of their cell surface adhesion molecules with those of endothelial cells, a process that is difficult to study clinically. Levels of soluble adhesion molecules such as ICAM-1 (70, 71), E-selectin, and P-selectin (72, 73) are elevated in patients with or at risk for ALI/ARDS, but the pathophysiological significance of these findings is uncertain. Neutrophils may mediate lung injury by a number of mechanisms, including the release of reactive nitrogen and oxygen species (see below), the production of cytokines and growth factors that can amplify the inflammatory response (see next section), and the release of proteolytic enzymes. Neutrophils can produce a variety of proteolytic enzymes potentially injurious to the lung, predominant among which is neutrophil elastase. Plasma levels of neutrophil elastase are elevated in patients at risk for ARDS after trauma (74) and in those with established ARDS after trauma (75). In samples of BAL fluid from patients with ALI/ARDS, variable levels of functional neutrophil elastase have been

Pathogenesis of acute lung injury: clinical studies

143

measured (76–81), and most of that recovered is complexed to endogenous inhibitors such as α2-macroglobulin or α1-antitrypsin. The presence of endogenous inhibitors of elastase activity suggests that the overall degree of proteolytic activity caused by neutrophil activation and degranulation may be attenuated to a great extent by endogenous inhibitors. Other proteases that are elevated in BAL fluid from patients with ALI/ARDS include collagenase (82) and gelatinases A and B (83, 84). Again, however, high levels of endogenous metalloproteinase inhibitors such as TIMP have also been identified (85), indicating that it is the overall protease/antiprotease balance that determines neutrophil-induced proteolytic damage. Recent evidence points to dysregulation of normal neutrophil turn-over as an important proinflammatory mechanism in ALI/ARDS. Apoptosis is thought to play a major role in the resolution of the inflammatory response. Neutrophil apoptosis with subsequent phagocytosis by macrophages provides a way of removing neutrophils without damage to surrounding tissue. BAL fluid from patients with ARDS has been shown to have an antiapoptotic effect on normal human neutrophils, an effect mediated predominantly by G-CSF and GM-CSF, cytokines known to have antiapoptotic activity. Further, neutrophil apoptosis is markedly reduced in patients with ARDS, although levels of apoptosis correlated best with levels of another antiapoptotic cytokine, IL-2 (86, 87). (For an in-depth discussion of neutrophil apoptosis in ARDS, see Chapter 8.) In addition to modulators of neutrophil apoptosis, a variety of natural inhibitors of neutrophil function have been identified in BAL fluid from ARDS patients (88). For example, levels of CC16, an inhibitor of neutrophil chemotaxis, correlate inversely with levels of neutrophil elastase (88). B. Cytokines Levels of a wide variety of cytokines are increased in biological fluids from patients with ALI/ARDS (1). A key concept that has arisen from these studies is that cytokine balance is an important determinant of the duration and degree of the inflammatory response. Although early studies focused on identifying proinflammatory cytokines in biological fluids from patients with ALI/ARDS, levels of most mediators were not clearly prognostic (1). Furthermore, anti-inflammatory strategies targeted at specific cytokines such as IL-1β and TNF-α were unsuccessful clinically (89). It is now clear that the cytokine response in ALI/ARDS is complex and involves both pro-and anti-inflammatory cytokines as well as endogenous inhibitors of pro-inflammatory cytokines. A variety of endogenous inhibitors have been identified, including IL-1 receptor antagonist, soluble IL-1 receptor, soluble TNF receptors I and II, and autoantibodies to IL-8. In a study of relative molar concentrations of various pro-inflammatory cytokines and their endogenous inhibitors in BAL fluid from patients with ALI/ARDS (90), although individual cytokines such as TNF-α and IL1-β increased before and after the onset of lung injury, greater increases in the levels of relevant endogenous inhibitors such as IL1β receptor antagonist and soluble TNF-α receptors were measured. This study provides convincing evidence of a significant anti-inflammatory response early in the course of ALI/ARDS that counteracts the pro-inflammatory response. Measuring levels of proinflammatory and anti-inflammatory molecules using immunological methods provides insufficient information concerning the complex cytokine milieu that characterizes

Acute respiratory distress syndrome

144

ALI/ARDS. For example, relative molar concentrations of pro- and anti-inflammatory cytokines do not necessarily predict biological activity of a given cytokine (89). In a second study, levels of IL-8/anti IL-8 autoantibody complexes correlated better with ARDS outcome than levels of IL-8 alone (91), even though complexing of anti-IL-8 antibody with IL-8 is thought to neutralize its chemotactic activity for neutrophils (92). A better way to assess the net pro- or anti-inflammatory activity of a given cytokine in human samples may be through assays of biological activity. Using cell surface expression of ICAM-1 on alveolar epithelial cell monolayers as a measure of proinflammatory activity, both pulmonary edema fluid (84) and BAL fluid (93) from patients with ALI/ARDS were shown to have potent pro-inflammatory activity, due predominantly to biologically active IL-1. Moreover, pulmonary edema fluid from ALI/ ARDS patients has been shown to stimulate repair in cultured alveolar epithelial monolayers, an effect mediated, in part, by biologically active IL-1 (94). These types of studies and others, such as the assays of neutrophil apoptosis discussed above, allow quantification of the net biological effect of a given cytokine in the presence of endogenous inhibitors. Recently, a great deal of interest has been focused on transcriptional factors that regulate production of proinflammatory cytokines and mediators in ARDS. An example is nuclear factor-κ B (NFκB), a transcription factor that regulates the expression of a number of proinflammatory mol-ecules including ICAM-1, IL-1β, IL-6, IL-8, and TNFα, to name a few (95, 96). Under basal conditions, NFκB resides in the cytoplasm, where it is bound to an inhibitor, IκB. When cells such as alveolar macrophages are stimulated by lipopolysaccharide, TNF-α or IL-1β, IκB is degraded and nuclear localization sites on NFκB are uncovered, allowing NFκB to localize to the nucleus. In the nucleus, NFκB binds to the promoters of various proinflammatory mediators and promotes their transcription, thus generating an amplified acute inflammatory response to the initial stimulus (95). It is postulated that the activation of NFκB may be a key proximal activation signal critical to the initiation and maintenance of the proinflammatory cytokine cascade in ALI/ARDS (96). Although the vast majority of work related to NFκB has been done using cell culture systems and animal models, a few studies have examined the role of NFκB in clinical ALI/ARDS. Human alveolar macrophages isolated from patients with ARDS were compared to those from control patients who were mechanically ventilated but had no evidence of, or risk factors for, lung injury (97). NFκB was activated in the macrophages from ARDS patients compared to the controls. Increased NFκB activation has also been demonstrated in BAL fluid of mixed cell populations in patients with ALI/ARDS (98). However, there was no clear association between levels of NFκB activation and bronchoalveolar lavage neutrophil or IL-8 concentrations, nor was there any association with clinical variables such as mortality (98). In a study of 15 patients with sepsis, binding activity for NFκB in peripheral blood mononuclear cell nuclear extracts was as accurate a predictor of outcome as was APACHE-II score (99). This finding suggests that the overall degree of NFκB activation could be a determinant of outcome from ALI/ARDS, but larger studies are needed.

Pathogenesis of acute lung injury: clinical studies

145

C. Pro-oxidant/Antioxidant Balance There is a severe imbalance between pro-oxidant and antioxidant activity in the lung in patients with ALI/ARDS (100). Under normal conditions, the generation of reactive oxygen or nitrogen species is counteracted by a complex network of antioxidant defense systems (101), which include anti-oxidant enzymes (e.g., superoxide dismutase, catalase), low molecular weight scavengers (e.g., vitamins E and C, glutathione), and mechanisms to remove or repair oxidatively damaged molecules, including DNA and proteins (100). In patients with ALI/ARDS, there appears to be a simultaneous increase in production of pro-oxidants and decrease in antioxidants (Table 4). Despite the growing body of evidence that supports a role for altered prooxidant/antioxidant imbalance in clinical ALI/ARDS, the results of clinical trials of antioxidants have been disappointing (102). Thus, in a large, randomized, double-blind, placebo-controlled, prospective trial comparing N-acetylcysteine to oxothiazolidine to placebo in patients with ARDS, there was no mortality difference between the three groups (103), although duration of lung injury was reduced with treatment. Dietary supplementation with fish oil, γ-linoleic acid, and antioxidants shortened the duration of mechanical ventilation and reduced the number of organ failures in patients with ARDS, but again, there was no mortality benefit (104). Alterations in the levels of endogenous antioxidants may also affect the risk of developing ALI/ARDS. Thus, patients who chronically ingest ethanol have an increased risk of developing ARDS after an inciting event such as severe trauma or aspiration of gastric contents (105). Although the underlying mechanism for the increased risk has not been identified, a recent study (106) reported that glutathione concentrations were approximately sevenfold lower in BAL fluid from chronic alcoholics compared to healthy nonalcoholic controls. Further, the percentage of oxidized glutathione was markedly increased. These findings suggest that an alteration in baseline levels of antioxidants such as glutathione could predispose individuals to the development of ARDS in the appropriate clinical setting. D. Coagulation Pathway In patients with ALI/ARDS, abnormalities in the coagulation cascade are demonstrable (107, 108) as an imbalance between procoagulant and anticoagulant forces in favor of coagulation. The end result is widespread alveolar fibrin deposition, a histological hallmark of ALI/ARDS (32, 109, 110). Procoagulant activity is also enhanced in BAL fluid from patients at risk of (111), and with established (112), ARDS. Thus, levels of tissue factor, a highly thrombogenic mediator in the extrinsic coagulation pathway, are increased in BAL fluid from ARDS patients (113, 114). Procoagulant activity peaks in the first 3 days after onset of ARDS and then declines (115). Anticoagulant activity is decreased simultaneously. Levels of the pro-fibrinolytic plasminogen activator inhibitor1 are diminished in BAL fluid from patients with ARDS (115, 116). Endogenous anticoagulants such as antithrombin III and protein C are diminished in clinical sepsis and lower levels are associated with poorer outcomes (117, 118).

Acute respiratory distress syndrome

146

In the past, coagulation abnormalities were considered to be a byproduct of the release of bacterial endotoxin, microvascular injury, and release of pro-inflammatory cytokines. More recently, they have been recognized as playing a primary role in the generation of the multiorgan failure associated with ALI/ARDS and sepsis (119, 120). Additionally, many of the coagulation abnormalities have pro-inflammatory effects. Thus, in experimental studies, generation of thrombin can induce endothelial neutrophil adhesion (121), the

Table 4 Evidence for Altered Prooxidant/Antioxidant Balance in Clinical Acute Lung Injury and Acute Respiratory Distress Syndrome Comments

Ref.

Evidence of increased oxidants Increased H2O2 in exhaled breath

Breath vapor from normal subjects contains little H2O2.

7, 8

Inactive α1antiproteinase in BAL

Probably damaged by ROS, RNS, or reactive chlorine species.

142

Decreased levels of GSH in lung

Increased levels GSSG found in lung lavage fluid and RBCs.

143

Loss of plasma thiol groups

Nonsurvivors of ARDS often have lower thiol levels.

144

Increased plasma Highly suggestive of oxidative damage by RIS. protein carbonyl groups

144

Presence of catalytic iron in plasma

Present in ARDS patients with multiorgan failure.

145

Presence of catalytic iron in BAL

BAL fluid from normal volunteers contains RIS, as does BAL from ARDS survivors. Nonsurvivors, however, show no RIS in BAL, but high transferrin levels due to leak from the plasma.

146

Increased lipid peroxidation products in plasma

Increased TBA reactivity.

147

Increased 4-hydroxynonenal.

148

Decreased linoleic and arachidonic acid.

149

Increased plasma xanthine oxidase

May be released from injured tissues after oxygenation injury to the lung.

150

Increased plasma hypoxanthine

Indicative of hypoxia and aberrant ATP catabolism during ischemia-reperfusion.

151

Nitrotyrosine formation Immunostaining of lung tissue from ARDS patients revealed nitrotyrosine suggestive of damage by ONOO- or other RNS. HPLC and GC-MS show increased plasma nitrotyrosine in patients

152 151

Pathogenesis of acute lung injury: clinical studies

147

with ARDS compared to control. Nitrotyrosine concentrations increased in BAL from ARDS patients.

153

3-Nitrotyrosine formation increased in BAL from patients with ARDS receiving inhaled NO.

154

Orthotyrosine formation HPLC and GC-MS show increased levels of plasma protein orthotyrosine, suggestive of increased •OH formation.

155

Chlorotyrosine formation

HPLC and GC-MS show increased levels of plasma protein chlorotyrosine which correlated with myeloperoxidase activity in ARDS patients.

155

Isoprostane formation

Isoprostanes (prostanoid compounds formed via primarily lipid peroxidation) are precise markers of in vivo oxidant stress. Levels are elevated in exhaled breath condensates from ARDS patients.

9

Oxidatively modified proteins

Elevated in BAL from ARDS patients.

156

Protein nitration

Surfactant protein A was nitrated in pulmonary edema fluid from patients with ALI/ARDS along with elevated levels of nitrate/nitrite, stable byproducts of NO generation.

157

Plasma from ARDS patients contained nitrated ceruloplasmin, transferrin, alpha(1)-protease inhibitor, alpha(1)antichymotrypsin and beta chain fibrinogen.

158

Low levels of plasma ascorbate

Possibly destroyed by oxidants.

159

Low levels of plasma αtocopherol

Appear to be low when not standardized to lipid content of plasma.

160

Decreased plasma ceruloplasmin ferroxidase activity

Plasma ceruloplasmin protein levels often elevated but ferroxidase activity per unit protein is decreased.

161

Low transferrin level in plasma

Iron-binding antioxidant activity of plasma reduced.

161

Evidence of decreased antioxidants

H2O2, hydrogen peroxide; BAL, bronchoalveolar lavage; ROS, reactive oxygen species; RNS, reactive nitrogen species; RIS, reactive iron species; GSH, glutathione; GSSG, oxidized GSH; RBC, red blood cells; TBA, thiobarbitaric acid; ATP, adenine triphosphate; ONOO−, peroxynitrite; GC-MS, gas chromatography-mass spectrometry: HPLC, high-performance liquid chromatrography; NO, nitric oxide; •OH, hydroxyl radical.

expression of adhesion molecules such as the selectins (122), and the activation of platelet receptors (123). Fibrin deposits may also have a variety of pro-inflammatory effects including increased vascular permeability, activation of endothelial cells, and induction of neutrophil adhesion and margination (108).

Acute respiratory distress syndrome

148

Restoration of pro-coagulant/anticoagulant balance may have an important therapeutic role in ALI/ARDS. In a landmark study of patients with severe sepsis, administration of activated protein C (124) was associated with a drop in mortality from 30.8% to 24.7% compared to patients who received placebo. The major side effect was increased bleeding. As evidence of an anti-inflammatory as well as an anticoagulant effect, plasma interleukin-6 levels were decreased in treated patients. Although no subgroup analysis of ARDS was presented, some of the patients fulfilled the diagnostic criteria for ARDS and the mortality benefit was evident in the subgroup with solitary respiratory dysfunction (124a). Whether activated protein C will be equally effective in ALI/ARDS from all causes is a topic for future study.

V. Repair Mechanisms A. Fibroproliferation The cellular and molecular mechanisms that determine whether a patient with ALI/ARDS will develop pulmonary fibrosis remain unclear. Undoubtedly, altered regulation of fibrin turnover plays a role (discussed in depth in Chapter 12). Levels of procollagen III peptide, a byproduct of collagen synthesis, are elevated in BAL fluid from patients with ALI/ARDS as early as day 3 (125), suggesting that fibroproliferation occurs very early in the clinical course of the syndrome. Elevated levels of procollagen III peptide have been demonstrated in pulmonary edema fluid as early as the first day of ALI/ARDS (126). Nonsurvivors had significantly higher levels. In addition, BAL fluid taken from patients within 24 hours of the onset of ALI/ARDS (127) stimulated human lung fibroblast proliferation. At 7 days, the mitogenic activity remained elevated and was higher in nonsurvivors. Pulmonary edema fluid from patients with ALI/ARDS is also mitogenic for human lung fibroblasts, a finding that depends in part upon bioactive interleukin-1 (128). Thus, pro-inflammatory mechanisms that are set into motion very early in ALI/ARDS are also intimately involved in the initiation of the fibroproliferative process. B. Apoptosis Apoptosis, the process of programmed cell death, permits cellular demise without incitement of an inflammatory response. It seems likely that apoptosis plays a role in the resolution of inflammation and the repair process in patients with ALI/ARDS. For example, the resolution of reactive alveolar epithelial type II cell hyperplasia appears to be mediated via apoptosis (129, 130) along with differentiation to an alveolar epithelial type I cell phenotype. As discussed above, one mechanism of clearance of neutrophils from the airspace is through apoptosis and phagocytosis by macrophages. In patients with ALI/ARDS, soluble Fas ligand protein has been shown to be elevated in BAL fluid before and after the onset of lung injury (131). The concentration of soluble Fas ligand protein in the bronchoalveolar lavage fluid also correlated both with proapoptotic activity in a cell line of distal lung epithelial cells in vitro and decreased patient survival (131). Fas ligand mRNA was also increased in the BAL fluid from patients with early ARDS (132). In another study, both soluble Fas ligand and soluble Fas were

Pathogenesis of acute lung injury: clinical studies

149

increased in the pulmonary edema fluid of patients with ALI/ARDS (133), and extensive Fas and Fas ligand immunostaining was seen in the alveolar and airway epithelium of patients who failed to survive (133). These findings indicate that the Fas/Fas ligand system is activated in the lungs of patients with ALI/ARDS and may play an important role in the initiation of apoptosis.

VI. Future Directions Clinical studies will continue to contribute to our understanding of the pathogenesis of ALI/ARDS. The advent of genomics and proteomics has the potential to greatly expand the amount of information that can be gained from patient-based studies. The formation of large multicenter clinical trials groups such as the NIH ARDS Network also provides the impetus and potential to study pathophysiological mechanisms in large numbers of patients as an adjunct to clinical trials.

References 1. Pittet JF, MacKersie RC, Martin TR, Matthay MA. Biological markers of acute lung injury: prognostic and pathogenetic significance. Am J Respir Crit Care Med 1997; 155:1187–1205. 2. Steinberg K, Mitchell DR, Maunder R, Milberg J, Whitcomb M, Hudson L. Safety of bronchoalveolar lavage in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 148:556–561. 3. Ware LB, Matthay MA. Alveolar fluid clearance is impaired in the majority of patients with acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:1376–1383. 4. Matthay MA, Wiener-Kronish JP. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am Rev Respir Dis 1990; 142:1250–1257. 5. Verghese GM, Ware LB, Matthay BA, Matthay MA. Alveolar epithelial fluid transport and the resolution of clinically severe hydrostatic pulmonary edema. J Appl Physiol 1999; 87:1301– 1312. 6. Heard S, Longtine K, Toth I, Puyana J, Potenza B, Smyrnios N. The influence of lipsomeencapsulated prostaglandin EI on hydrogen peroxide concentrations in the exhaled breath of patients with the acute respiratory distress syndrome. Anesth Analg 1999; 89:353–357. 7. Baldwin S, Grum C, Boxer L, Simon R, Ketai L, Devall L. Oxidant activity in expired breath of patients with adult respiratory distress syndrome. Lancet 1986; 1(8471): 1–14. 8. Sznajder J, Fraiman A, Hall J, Sanders W, Schmidt G, Crawford G, Nahum A, Factor P, Wood L. Increased hydrogen peroxide in the expired breath of patients with acute hypoxemic respiratory failure. Chest 1989; 96:606–612. 9. Carpenter C, Price P, Christman B. Exhaled breath condensate isoprostanes are elevated in patients with acute lung injury or ARDS. Chest 1998; 114: 1653–1659. 10. Nayeri F, Millinger E, Nilsson I, Zetterstrom O, Brudin L, Forsberg P. Exhaled breath condensate and serum levels of hepatocyte growth factor in pneumonia. Respir Med 2002; 96:115–119. 11. Scheideler L, Manke H, Schwulera U, Inacker O, Hammerle H. Detection of nonvolatile macromolecules in breath. A possible diagnostic tool? Am Rev Respir Dis 1993; 148:778–784. 12. Mutlu G, Garey K, Robbins R, Danziger L, Rubinstein I. Collection and analysis of exhaled breath condensate in humans. Am J Respir Crit Care Med 2001; 164:731–737.

Acute respiratory distress syndrome

150

13. Brett SJ, Evans TW. Measurement of endogenous nitric oxide in the lungs of patients with the acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 157:993–997. 14. Schubert J, Muller W, Benzing A, Geiger K. Application of a new method for analysis of exhaled gas in critically ill patients. Intensive Care Med 1998; 24:415–421. 15. Sturm J, Wisner D, Oestern H-J, Kant C, Tscherne H, Creutzig H. Increased lung capillary permeability after trauma: a prospective clinical study. J Trauma 1986; 326:409–418. 16. Byrne K, Sugerman H. Experimental and clinical assessment of lung injury by measurement of extravascular lung water and transcapillary protein flux in ARDS: a review of current techniques. J Surg Res 1988; 44:185–203. 17. Miniati M, Pistolesi M, Milne M, Giuntini C. Detection of lung edema. Crit Care Med 1987; 15:1146–1155. 18. Sivak E, Wiedemann H. Clinical measurement of extravascular lung water. Crit Care Clin 1986; 2:511–526. 19. Mitchell JP, Schuller D, Calandrino FS, Schuster DP. Improved outcome based on fluid management in critically ill patients requiring pulmonary artery catheterization. Am Rev Respir Dis 1992; 145:990–998. 20. Schuller D, Mitchell J, Calandrino F, Schuster D. Fluid balance during pulmonary edema. Is fluid gain a marker or a cause of poor outcome? Chest 1991; 100:1068–1075. 20a. Sakka SG, Klein M, Reinhart K, Meier-Hellmann A. Prognostic value of extravascular lung water in critically ill patients. Chest 2002; 122:2080–2086. 21. Davey-Quinn A, Gedney J, Whiteley S, Bellamy M. Extravascular lung water and acute respiratory distress syndrome—oxygenation and outcome. Anaesth Intensive Care 1999; 27:357–362. 22. Malbouisson L, Preteux F, Puybasset L, Grenier P, Coriat P, Rouby J-J. Validation of a software designed for computed tomographic (CT) measurement of lung water. Intensive Care Med 2001; 27:602–608. 23. Groeneveld A, Raijmakers P. The 67gallium-transferrin pulmonary leak index in pneumonia and associated adult respiratory distress syndrome. Clin Sci 1997; 93:463–470. 24. Groeneveld A. Radionuclide assessment of pulmonary microvascular permeability. Eur J Nucl Med 1997; 24:449–461. 25. Macnaughton P, Morgan C, Denison D, Evans T. Measurement of carbon monoxide transfer and lung volume in ventilated subjects. Eur Respir J 1993; 6:231–236. 26. Macnaughton P, Evans T. Lung function in adult respiratory distress syndrome. Am J Respir Crit Care Med 1994; 150:770–775. 27. Nuckton T, Alonso J, Kallet R, Daniel B, Eisner M, Pittet J, Matthay M. In early ARDS wasted ventilation predicts mortality. Am J Respir Crit Care Med 2001; 163:A450. 28. Christiani D, Gong M, Xu L, Miller D, Paulauskis J, Thompson B. Polymorphisms in the SP-B gene, gender and risk of ARDS. Am J Respir Crit Care Med 2001; 163:A306. 29. Lin Z, Pearson C, Chinchillli V, Pietschmann S, Luo J, Pison U, Floros J. Polymorphisms of human SP-A, SP-B, and SP-D genes: association of AP-B Thr131Ile with ARDS. Clin Genet 2000; 58:181–191. 30. Marshall R, Webb S, Brull D, Belingan G, McAnulty R, Hill M, Humphries S, Laurent G. Interleukin-6 (—174G > C) polymorphism predicts outcome in acute respiratory failure. Am J Respir Crit Care Med 2001; 163:A305. 31. Bachofen M, Weibel ER. Alterations of the gas exchange apparatus in adult respiratory insufficiency associated with septicemia. Am Rev Respir Dis 1977; 116:589–615. 32. Bachofen M, Weibel ER. Structural alterations of lung parenchyma in the adult respiratory distress syndrome. Clin Chest Med 1982; 3:35–56. 33. Fein A, Grossman RF, Jones JG, et al. The value of edema protein measurements in patients with pulmonary edema. Am J Med 1979; 67:32–39.

Pathogenesis of acute lung injury: clinical studies

151

34. Matthay MA, Eschenbacher WC, Goetzl EJ. Elevated concentrations of leukotriene D4 in pulmonary edema fluid of patients with adult respiratory distress syndrome. J Clin Immunol 1984; 4:479–483. 35. Sprung C, Rackow E, Fein I, Jacob A, Isikoff S. The spectrum of pulmonary edema: differentiation of cardiogenic inermediate and noncardiogenic forms of pulmonary edema. Am Rev Respir Dis 1981; 124:718–722. 36. Raijmakers P, Groeneveld A, Teule G, Thijs L. The diagnostic value of the 67Gallium pulmonary leak index in pulmonary edema. J Nucl Med 1996; 37: 1316–1322. 37. Hurley JV. Types of pulmonary microvascular injury. Ann NY Acad Sci 1982; 384:269–286. 38. Pittet J, Morel D, Hemsen A, Gunning K, Lacroix J, Suter P, Lundberg J. Elevated plasma endothelin-1 concentrations are associated with the severity of illness in patients with sepsis. Ann Surg 1991; 213:261–264. 39. Morel D, Lacroix J, Hemsen A, Steinig D, Pittet J, Lundberg J. Increased plasma and pulmonary lymph levels of endothelin during endotoxin shock. Eur J Pharmacol 1989; 167:427– 428. 40. Miyauchi T, Yanagisawa M, Tomizawa T, Sugishita Y, Suzuki N, Fujino M, Ajisaka R, Goto K, Masaki T. Increased concentrations of endothelin-1 and big endothelin-1 in acute myocardial infarction. Lancet 1989; 2:54–54. 41. Druml W, Steltzer H, Waldhausl W, Lenz K, Hammerle A, Vierhapper H, Gasic S, Wagner O. Endothelin-1 in adult respiratory distress syndrome. Am Rev Respir Dis 1993; 148:1169–1173. 42. Langleben D, Demarchie M, Laporta D, Spanier AH, Schlesinger RD, Stewart DG. Endothelin1 in acute lung injury and the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 148:1646–1650. 43. Sanai L, Haynes W, Mackenzie A, Grant I, Webb D. Endothelin production is sepsis and the adult respiratory distress syndrome. Intensive Care Med 1996; 22:52–56. 44. Albertine K, Wana Z, Michael J. Expression of endothelial nitric oxide synthase, inducible nitric oxide synthase and endothelin-1 in lungs of subjects who died with ARDS. Chest 1999; 104:476–480. 45. Rubin L, Badesch D, Barst R, Galie N, Black C, Keogh A, Pulido T, Frost A, Roux S, Leconte I, Landzberg M, Simonneau G, for the Bosentan Randomized Trial of Endothelin Antagonist Therapy Study Group. Bosentan therapy for pulmonary arterial hypertension. N Engl J Med 2002; 346:896–903. 46. Ware L, Conner E, Matthay M. von Willebrand factor antigen is an independent marker of poor outcome in patients with early acute lung injury. Crit Care Med 2001; 29:2325–2331. 47. Berthiaume Y, Lesur O, Dagenais A. Treatment of acute respiratory distress syndrome: plea for rescue therapy of the alveolar epithelium. Thorax 1999; 54:150–160. 48. Kurahashi K, Kajikawa O, Sawa T, Ohara M, Gropper MA, Frank DW, Martin TR, WienerKronish JP. Pathogenesis of septic shock in Pseudomonas aeruginosa pneumonia. J Clin Invest 1999; 104:743–750. 49. Lewis JF, Jobe AH. Surfactant and the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 147:218–233. 50. Gregory TJ, Steinberg KP, Spragg R, Gadek JE, Hyers TM, Longmore WJ, Moxley MA, Guang-Zuan C, Hite RD, Smith RM, Hudson LD, Crim C, Newton P, Mitchell BR, Gold AJ. Bovine surfactant therapy for patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 155:1309–1315. 51. Greene KE, Wright JR, Steinberg K, Ruzinski JT, Caldwell E, Wong WB, Hull W, Whitsett JA, Akino T, Kuroki Y, Nagae H, Hudson LD, Martin TR. Serial changes in surfactant-associated proteins in lung and serum before and after onset of ARDS. Am J Respir Crit Care Med 1999; 160:1843–1850. 52. Cheng I, Ware L, Greene K, Nuckton T, Eisner M, Matthay M. Prognostic value of surfactant proteins A and D in patients with acute lung injury. Crit Care Med 2003; 31:20–27.

Acute respiratory distress syndrome

152

53. Baker CS, Evans TW, Randle BJ, Haslam PL. Damage to surfactant-specific protein in acute respiratory distress syndrome. Lancet 1999; 353:1232–1237. 54. Ware LB, Golden JA, Finkbeiner WE, Matthay MA. Alveolar epithelial fluid transport capacity in reperfusion lung injury after lung transplantation. Am J Respir Crit Care Med 1999; 159:980– 988. 55. Dobbs L, Gonzalez R, Allen L, Froh D. HT156, an integral membrane protein specific to human alveolar type I cells. J Histochem Cytochem 1999; 47:129–137. 56. Newman V, Gonzalez RF, Matthay MA, Dobbs LG. A novel alveolar type I cell-specific biochemical marker of human acute lung injury. Am J Respir Crit Care Med 2000; 161:990– 995. 57. Ishizaka A, Hasegawa N, Hashimoto S, Matsuda T, Koh H, Kotani T, Kotake Y, Morisaki H, Fujishima S, Takeda J, Yamaguchi K. Pulmonary KL-6 concentration correlates with alveolar septal permeability in early acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:A307. 58. Wright J. Immunomodulatory functions of surfactant. Physiol Rev 1997; 77: 931–962. 59. Lee W, Downey G. Neutrophil activation and acute lung injury. Curr Opin Crit Care 2001; 7:1– 7. 60. Parsons PE, Fowler AA, Hyers T, Henson PM. Chemotactic activity in bronchoalveolar lavage fluid from patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1985; 132:490–493. 61. Steinberg K, Milberg J, Martin T, et al. Evolution of bronchoalveolar cell populations in the adult respiratory distress syndrome. Am J Respir Crit Care Med 1994; 150:113–122. 62. Baughman R, Gunther K, Rashkin M, Keeton D, Pattishall E. Changes in the inflammatory response of the lung during acute respiratory distress syndrome: prognostic indicators. Am J Respir Crit Care Med 1996; 154:76–81. 63. Warshawski F, Sibbald W, Driedger A, Cheung H. Abnormal neutrophil-pulmonary interaction in the adult respiratory distress syndrome: qualitative and quantitative assessment of pulmonaryneutrophil kinetics in humans with in vivo indium-111 neutrophil scintigraphy. Am Rev Respir Dis 1986; 133:792–804. 64. Laufe MD, Simon RH, Flint A, Keller JB. Adult respiratory distress syndrome in neutropenic patients. Am J Med 1986; 80:1022–1026. 65. Worthen GS, Schwab B, Elson EL, Downey GP. Mechanics of stimulated neutrophils: cell stiffening induces retention in capillaries. Science 1989; 245: 183–185. 66. Lavkan A, Astiz M, Rackow E. Effects of proinflammatory cytokines and bacterial toxins on neutrophil rheologic properties. Crit Care Med 1998; 26:1677–1682. 67. Erzurum S, Downey G, Doherty D, et al. Mechanisms of lipopolysaccharide induced neutrophil retention. J Immunol 1992; 149:154–162. 68. Inano H, English D, Doerschuk C. Effect of zymosan activated plasma on deformability of rabbit polymorphonuclear leukocytes. J Appl Physiol 1992; 73:1370–1376. 69. Skoutelis A, Kaleridis V, Athanassiou G, Kokkinis K, Missirlis Y, Bassaris H. Neutrophil deformability in patients with sepsis, septic shock, and adult respiratory distress syndrome. Crit Care Med 2000; 28:2355–2359. 70. Conner E, Ware L, Modin G, Matthay M. Elevated pulmonary edema fluid concentrations of soluble intercellular adhesion molecule-1. Biological and clinical significance. Chest 1999; 116:83S-84S. 71. Sessler C, Windsor A, Schwartz M, Watson L, Fisher B, Sugerman H, Fowler A. Circulating ICAM-1 is increased in septic shock. Am J Respir Crit Care Med 1995; 151:1420–1427. 72. Sakamaki F, Ishizaka A, Handa M, Fujishima S, Urano T, Sayama K, Nakamura H, Kanazawa M, Kawashiro T, Katayama M, Ikeda Y. Soluble form of P-selectin in plasma is elevated in acute lung injury. Am J Respir Crit Care Med 1995; 151:1821–1826.

Pathogenesis of acute lung injury: clinical studies

153

73. Boldt J, Wollbruck M, Kuhn D, Linke L, Hempelmann G. Do plasma levels of circulating soluble adhesion molecules differ between surviving and nonsurviving critically ill patients? Chest 1995; 107:787–792. 74. Donnelly S, MacGregor I, Zamani A, Gordon M, Robertson C, Steedman D, Little K, Haslett C. Plasma elastase levels and the development of the adult respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:1428–1433. 75. Gando S, Kameue T, Nanzaki S, Hayakawa T, Nakanishi Y. Increased neutrophil elastase, persistent intravascular coagulation, and decreased fibrinolytic activity in patients with posttraumatic acute respiratory distress syndrome. J Trauma 1997; 42:1068–1072. 76. Weiland J, Davis B, Holter J, Mohammed J, Dorinsky P, Gadek J. Lung neutrophils in the adult respiratory distress syndrome: clinical and pathophysiologic significance. Am Rev Respir Dis 1986; 133:218–225. 77. Lee C, Fein A, Lipmann M, Holtzmann H, Kimbel P, Weinbaum G. Elastolytic activity in pulmonary lavage fluid from patients with adult respiratory distress syndrome. N Engl J Med 1981; 304:192–196. 78. McGuire WW, Spragg R, Cohen A, Cochrane C. Studies on the pathogenesis of the adult respiratory distress syndrome. J Clin Invest 1982; 69:543–553. 79. Suter P, Suter S, Girardin E, Roux-Lombard P, Grau G, Dayer J. High bronchoalveolar levels of tumor necrosis factor and its inhibitors, interleukin-1, interferon, and elastase in patients with adult respiratory distress syndrome after trauma, shock, or sepsis. Am Rev Respir Dis 1992; 145:1016–1022. 80. Fowler A, Walchak S, Giclas P, Henson P, Hyers T. Characterization of antiprotease activity in the adult respiratory distress syndrome. Chest 1982; 81:50S–51S. 81. Idell S, Kucich U, Fein A, Kueppers F, James H, Walsch P, Weinbaum G, Colman R, Cohen A. Neutrophil elstase releasing factors in bronchoalveolar lavage from patients with adult respiratory distress syndrome. Am Rev Respir Dis 1985; 132:1098–1105. 82. Christner P, Fein A, Goldberg S, Lipmann N, Abrams W, Weinbaum G. Collagenase in the lower respiratory tract of patients with adult respiratory distress syndrome. Am Rev Respir Dis 1985; 131:690–695. 83. Delclaux C, d’Ortho MP, Delacourt C, Lebargy F, Brun-Buisson C, Brochard L, Lemaire F, Lafuna C, Harf A. Gelatinases in epithelial lining fluid of patients with adult respiratory distress syndrome. Am J Physiol 1997; 272:L442–L451. 84. Pugin J, Verghese G, Widmer M-C, Matthay MA. The alveolar space is the site of intense inflammatory and profibrotic reactions in the early phase of ARDS. Crit Care Med 1999; 27:304–312. 85. Ricou B, Nicod L, Lacaraz S, Welgus HG, Suter PM, Dayer J-M. Matrix metalloproteinases and TIMP in acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154:346– 352. 86. Lesur O, Kokis A, Hermans C, Fulop T, Benard A, Lane D. Interleukin-2 involvement in early acute respiratory distress syndrome: relationship with polymorphonuclear neutrophil apoptosis and patient survival. Crit Care Med 2000; 28:3814–3822. 87. Matute-Bello G, Liles WC, Radella II F, Steinberg KP, Ruzinski JT, Jonas M, Chi EY, Hudson LD, Martin TR. Neutrophil apoptosis in the acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 156:1969–1977. 88. Geerts L, Jorens P, Willems J, De Ley M, Slegers H. Natural inhibitors of neutrophil function in acute respiratory distress syndrome. Crit Care Med 2001; 29:1920–1924. 89. Opal S, Cross A. Clinical trials for severe sepsis. Past failures, and future hopes. Infect Dis Clin North Am 1999; 13:285–297. 90. Park WY, Goodman RB, Steinberg KP, Ruzinski JT, Radella II F, Park DR, Pugin J, Skerrett S, Hudson LD, Martin T. Cytokine balance in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 164:1896–1903.

Acute respiratory distress syndrome

154

91. Kurdowska A, Noble JM, Steinberg KP, Ruzinski JT, Hudson LD, Martin TR. Anti-interluekin 8 autoantibody:interleukin 8 complexes in the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:463–468. 92. Kurdowska A, Miller EJ, Noble JM, Baughman RP, Matthay MA, Brelsford WG, Cohen AB. Anti-IL-8 autoantibodies in alveolar fluid from patients with the adult respiratory distress syndrome. J Immunol 1996; 157:2699–2706. 93. Pugin J, Ricou B, Stenberg KP, Suter PM, Martin TR. Proinflammatory activity in bronchoalveolar lavage fluids from patients with ARDS, a prominent role for interleukin-1. Am J Respir Crit Care Med 1996; 153: 1850–1856. 94. Geiser T, Atabai K, Jarreau PH, Ware LB, Pugin J, Matthay MA. Pulmonary edema fluid from patients with acute lung injury augments alveolar epithelial repair by an IL-1β dependent mechanism. Am J Respir Crit Care Med 2001; 163:1384–1388. 95. Fan J, Ye R, Malik A. Transcriptional mechanisms in acute lung injury. Am J Physiol Lung Cell Mol Physiol 2001; 281:L1037–L1050. 96. Christman JW, Sadikot RT, Blackwell T. The role of nuclear factor-κB in pulmonary diseases. Chest 2000; 117:1482–1487. 97. Schwartz M, Moore E, Moore F, Shenkar R, Moine P, Haenel J, Abraham E. Nuclear factor-κB is activated in alveolar macrophages from patients with acute respiratory distress syndrome. Crit Care Med 1996; 24:1285–1292. 98. Christman J, Venkatakrishnan A, Wheeler A, Bernard G, Blackwell T. NF-KB binding activity in mixed BAL cells from patients with SIRS and ARDS. Sepsis 2000; 4:35–41. 99. Bohrer H, Qiu F, Zimmermann T, Zhang Y, Jllmer T, Mannel D, Bottiger B, Stern D, Waldherr R, Saeger H-D, Ziegler R, Bierhaus A, Martin E, Nawroth P. Role of NFκB in the mortality of sepsis. J Clin Invest 1997; 100:972–985. 100. Chabot F, Mitchell JA, Gutteridge JMC, Evans TW. Reactive oxygen species in acute lung injury. Eur Respir J 1998; 11:745–757. 101. Gutteridge J, Halliwell B. Antioxidants in Nutrition, Health and Disease. Oxford: Oxford University Press, 1994. 102. Conner B, Bernard G. Acute respiratory distress syndrome. Potential pharmacologic interventions. Clin Chest Med 2000; 21:563–587. 103. Bernard GR, Wheeler A, Arons M, Morris P, Paz H, Russell J, Wright P, and The Antioxidant in ARDS Study Group. A trial of antioxidants N-acetylcysteine and procysteine in ARDS. Chest 1997; 112:164–172. 104. Gadek J, DeMichele S, Karlstad M, Pacht E, Donahoe M, Albertson T, Van Hoozen C, Wennberg A, Nelson J, Noursalehi M. Effect of enteral feeding with eicosapentanoic acid, gamma-linoleic acid, and antioxidants in patients with acute respiratory distress syndrome. Enteral Nutrition in ARDS Study Group. Crit Care Med 1999; 27:1409–1420. 105. Moss M, Bucher B, Moore FA, Moore EE, Parsons PE. The role of chronic alcohol abuse in the development of acute respiratory distress syndrome in adults. JAMA 1996; 275:50–54. 106. Moss M, Guidot DM, Wong-Lambertina M, Teh Hoor T, Perez R, Brown L. The effects of chronic alcohol abuse on pulmonary glutathione homeostasis. Am J Respir Crit Care Med 2000; 161:414–419. 107. Idell S. Anticoagulants for acute respiratory distress syndrome: Can they work? Am J Respir Crit Care Med 2001; 164:517–520. 108. Abraham E. Coagulation abnormalities in acute lung injury and sepsis. Am J Respir Cell Mol Biol 2000; 22:401–404. 109. McDonald J. The yin and yang of fibrin in the airways. N Engl J Med 1990; 322:929–931. 110. Idell S. Extravascular coagulation and fibrin deposition in acute lung injury. New Horizons 1994; 2:566–574. 111. Seeger W, Hubel J, Klapettek K, Pison U, Obertacke U, Joka T, Roka L. Procoagulant activity in bronchoalveolar lavage of severely traumatized patients-relation to development of acute respiratory distress. Thromb Res 1991; 61:53–64.

Pathogenesis of acute lung injury: clinical studies

155

112. Fuchs-Buder T, deMoerloose P, Ricou B, Reber G, Vifian C, Nicod L, Romand J, Suter P. Time course of procoagulant activity and D dimer in bronchoalveolar fluid of patients at risk for or with acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 153:163–167. 113. Idell S, Gonzalez K, Bradford H, MacArthur C, Fein A, Maunder R, Garcia J, Griffith D, Weiland J, Martin T, et al. Procoagulant activity in bronchoalveolar lavage in the adult respiratory distress syndrome. Am Rev Respir Dis 1987; 136:1466–1474. 114. Idell S, James K, Levin E, Schwartz b, Manchanda N, Maunder R, Martin T, McLarty J, Fair D. Local abnormalities in coagulation and fibrinolytic pathways predispose to alveolar fibrin deposition in the adult respiratory distress syndrome. J Clin Invest 1989; 84:695–705. 115. Idell S, Koenig K, Fair D, Martin T, McLarty J, Maunder R. Serial abnormalities of fibrin turnover in evolving adult respiratory distress syndrome. Am J Physiol 1991; 261:L240–L248. 116. Bertozzi P, Astedt B, Zenzius L, Lynch K, LeMaire F, Zapol W, Chapman H. Depressed bronchoalveolar urokinase activity in patients with adult respiratory distress syndrome. N Engl J Med 1990; 322:890–897. 117. Balk R, Emerson T, Fourrier F, Kruse J, Mammen E, Schuster H, Vinazzer H. Therapeutic use of antithrombin concentrate in sepsis. Sem Thromb Haemostasis 1998; 24:183–194. 118. Fourrier F, Chopin C, Goudemand J, Hendrycx S, Caron C, Rime A, Marey A, Lestavel P. Septic shock, multiple organ failure, and disseminated intravascular coagulation: compared patterns of antithrombin III, protein C, and protein S deficiencies. Chest 1992; 101:816–823. 119. Matthay M. Severe sepsis—a new treatment with both anticoagulant and antiinflammatory properties. N Engl J Med 2001; 344:6750–6752. 120. Vincent J-L. New therapeutic implications of anticoagulation mediator replacement in sepsis and acute respiratory distress syndrome. Crit Care Med 2000; 28:S83–S85. 121. Lo S, Lai L, Ja C, et al. Thrombin-induced generation of neutrophil activating factors in the blood. Am J Pathol 1988; 130:22–32. 122. Kaplanski G, Fabrigoule M, Boulay V, Dinarello C, Bongrand P, Kaplanski S, Farnarier C. Thrombin induces endothelial type II activation in vitro: IL-1 and TNF-alpha-independent IL-8 secretion and E-selectin expression. J Immunol 1997; 158:5435–5441. 123. Coughlin S. Thrombin signalling and protease-activated receptors. Nature 2000; 407:258–264. 124. Bernard G, Vincent J-L, Laterre P-F, La Rosa S, Dhainaut J-F, Lopez-Rodriguez A, Steingrub J, Garber G, Helterbrand J, Ely E. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344:699–709. 124a. Ely EW, Laterre P-F, Angus DC, Helterbrand JD, Levy H, Dhainaut J-F, Vincent J-L, Macias WL, Bernard GR, for the Prowess Investigators. Drotecogin alfa (activated) administration across clinically important subgroups of patients with severe sepsis. Crit Care Med 2003; 31:12–19. 125. Clark JG, Milberg JA, Steinberg KP, Hudson LD. Type III procollagen peptide in the adult respiratory distress syndrome. Ann Intern Med 1995; 122:17–23. 126. Chesnutt AN, Matthay MA, Tibayan FA, Clark JG. Early detection of type III procollagen peptide in acute lung injury. Am J Respir Crit Care Med 1997; 156:840–845. 127. Marshall R, Bellingan G, Webb S, Puddicombe A, Goldsack N, McAnulty R, Laurent G. Fibroproliferation occurs early in the acute respiratory distress syndrome and impacts on outcome. Am J Respir Crit Care Med 2000; 162:1783–1788. 128. Olman M, White K, Ware L, Cross M, Zhu S, Matthay M. Microarray analysis indicates that pulmonary edema fluid from patients with acute lung injury mediates inflammation, mitogen gene expression, and fibroblast proliferation thorugh bioactive interleukin-1. Chest 2002; 121:69S–70S. 129. Fehrenbach H, Kasper M, Tschernig T, Pan T, Schuh D, Shannon JM, Muller M, Mason RJ. Keratinocyte growth factor-induced hyperplasia of rat alveolar type II cells in vivo is resolved by differentiation into type I cells and apoptosis. Eur Respir J 1999; 14:534–544.

Acute respiratory distress syndrome

156

130. Fehrenbach H, Kasper M, Koslowski R, Pan T, Schuh D, Muller M, Mason RJ. Alveolar epithelial type II cell apoptosis in vivo during resolution of keratinocyte growth factor-induced hyperplasia in the rat. Histochem Cell Biol 2000; 114:49–61. 131. Matute-Bello G, Liles WC, Steinberg KP, Kiener PA, Mongovin S, Chi EY, Jonas M, Martin TR. Soluble Fas-ligand induces epithelial cell apoptosis in humans with acute lung injury (ARDS). J Immunol 1999; 163:2217–2225. 132. Hashimoto S, Kobayashi A, Kooguchi K, Kitamura Y, Onodera H, Nakajima H. Upregulation of two death pathways of perforin/granzyme and FasL/Fas in septic acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 161:237–243. 133. Albertine K, Zimmerman G, Matthay MA, Ware LB. Fas and Fas ligand are upregulated in the pulmonary edema and the lung tissue of patients with acute lung injury and the acute respiratory distress syndrome. Am J Pathol 2002; 161:1783–1796. 135. Carvalho ACA, Bellman SM, Saullo VJ, Quinn D, Zapol WM. Altered factor VIII in acute respiratory failure. N Engl J Med 1982; 307:1113–1119. 136. Moalli R, Doyle JM, Tahhan HR, Hasan FM, Braman SS, Saldeen T. Fibrinolysis in critically ill patients. Am Rev Respir Dis 1989; 140:287–293. 137. Rubin DB, Wiener-Kronish JP, Murray JF, Green DR, Turner J, Luce JM, Montgomery AB, Marks JD, Matthay MA. Elevated von Willebrand factor antigen is an early plasma predictor of acute lung injury in nonpulmonary sepsis. J Clin Invest 1990; 86:474–480. 138. Moss M, Ackerson L, Gilespie MK, Moore FA, Moore EE, Parsons PE. Von Willebrand factor antigen levels are not predictive for the adult respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:15–20. 139. Sabharwal AK, Bajaj SP, Ameri S, Tricomi SM, Hyers TM, Dahms TE, Taylor FB, Bajaj MS. Tissue factor pathway inhibitor and von Willebrand factor antigen levels in adult respiratory distress syndrome and in a primate model of sepsis. Am J Respir Crit Care Med 1995; 151:758– 767. 140. Kayal S, Jais J-P, Aguini N, Chaudiere J, Labrousse J. Elevated circulating E-selectin, intercellular adhesion molecule 1 and von Willebrand factor in patients with severe infection. Am J Respir Crit Care Med 1998; 157:776–784. 141. Bajaj MS, Tricomi SM. Plasma levels of the three endothelial-specific proteins von Willebrand factor, tissue factor pathway inhibitor, and thrombomodulin do not predict the development of acute respiratory distress syndrome. Intensive Care Med 1999; 25:1259–1266. 142. Cochrane C, Spragg R, Revak S, Cohen A, McGuire W. The presence of neutrophil elastase and evidence of oxidation activity in bronchoalveolar lavage fluid of patients with adult respiratory distress syndrome. Am Rev Respir Dis 1983; 127:25–27. 143. Pacht ER, Timerman AP, Lykens MG, Merola AJ. Deficiency of alveolar fluid glutathione in patients with sepsis and the adult respiratory distress syndrome. Chest 1991; 100:1397–1404. 144. Quinlan G, Evans T, Gutteridge J. Linoleic acid and protein thiol changes suggestive of oxidative damage in the plasma of patients with adult respiratory distress syndrome (ARDS). Free Rad Res 1994; 20:299–306. 145. Gutteridge JM, Quinlan GJ, Mumby S, Heath A, Evans TW. Primary plasma antioxidants in the adult respiratory distress syndrome patients: changes in iron-oxidizing, iron-binding, free radical-scavenging proteins. J Lab Clin Med 1994; 124:263–273. 146. Gutteridge J, Mumby S, Quinlan G, Chung K, Evans T. Pro-oxidant iron is present in human pulmnary epithelial lining fluid: implications for oxidative stress in the lung. Biochem Biophys Res Commun 1996; 220:1024–1027. 147. Richard C, Lemonnier F, Thibault M, Auzepy P. Vitamin E deficiency and lipid peroxidation during adult respiratory distress syndrome. Crit Care Med 1990; 18:4–9. 148. Quinlan G, Evans T, Gutteridge J. Oxidative damage to plasma proteins in adult respiratory distress syndrome. Free Rad Res 1994; 20:289–298. 149. Grum C, Ragsdale R, Ketai L, Simon R. Plasma xanthine oxidase activity in patients with ARDS. J Crit Care 1987; 2:22–26.

Pathogenesis of acute lung injury: clinical studies

157

150. Quinlan G, Lamb N, Tilley R, Evans T, Gutteridge J. Plasma hypoxanthine levels in ARDS: implications for oxidative stress, morbidity and mortality. Am J Respir Crit Care Med 1997; 155:479–484. 151. Lamb NJ, Gutteridge J, Baker C, Evans T, Quinlan G. Oxidative damage to proteins of bronchoalveolar lavage fluid in patients with acute respiratory distress syndrome: evidence for neutrophil-mediated hydroxylation, nitration and chlorination. Crit Care Med 1999; 27:1738– 1744. 152. Beckman J, Beckman T, Chen J, Marshall P, Freeman B. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci 1990; 87:1620–1624. 153. Sittipoint C, Steinberg K, Ruzinski J, Myles C, Zhu S, Goodman R, Hudson L, Matalon S, Martin T. Nitric oxide and nitrotyrosine in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:503–510. 154. Lamb N, Quinlan G, Westerman S, Gutteridge J, Evans T. Nitration of proteins in bronchoalveolar lavage fluid from patients with ARDS and patients at risk for ARDS. Am J Respir Crit Care Med 1999; 160:1031–1034. 155. Cross C, Forte T, Stocker R, et al. Oxidative stress and abnormal cholesterol metabolism in patients with adult respiratory distress syndrome. J Lab Clin Med 1990; 115:391–404. 156. Lenz A, Jorens P, Meyer B, De Backer W, Van Overveld F, Bossaer Maier K. Oxidatively modified proteins in bronchoalveolar lavage fluid from patients with ARDS and patients at risk for ARDS. Eur Respir Dis 1999; 13:169–174. 157. Zhu S, Ware LB, Geiser T, Matthay MA, Matalon S. Increased levels of nitrate and surfactant protein A nitration in the pulmonary edema fluid of patients with acute lung injury. Am J Respir Crit Care Med 2001; 163:166–172. 158. Gole M, Souza J, Choi I, Hertkorn c, Malcolm S, Foust Rr, Finkel B, Lanker P, Ischiropoulos H. Plasma proteins modified by tyrosine-nitration in acute respiratory distress syndrome. Am J Physiol Lung Cell Mol Physiol 2000; 278:L961–L967. 159. Bertrand Y, Pincemail J, Hanique G, et al. Differences in tocopherol-lipid ratios in ARDS and non-ARDS patients. Int Care Med 1989; 15:87–93. 160. Gutteridge J, Quinlan G, Evans T. Transient iron-overload with bleomycin-detectable iron in the plasma of patients with adult respiratory distress syndrome. Thorax 1994; 49:707–710. 161. Quinlan G, Evans T, Gutteridge J. 4-Hydroxy-2-nonel levels increase in the plasma of patients with adult respiratory distress syndrome as linoleic acid appears to fall. Free Rad Res 1994; 21:95–106.

8 Is Apoptosis Important in the Pathogenesis and Resolution of the Acute Respiratory Distress Syndrome? GUSTAVO MATUTE-BELLO and THOMAS R.MARTIN University of Washington School of Medicine and VA/Puget Sound Medical Center Seattle, Washington, U.S.A.

I. Introduction Apoptosis is a controlled process of cell death, which is triggered by specific stimuli and is carried out by intracellular pathways. The outcome of this process is a decrease in cell size, fragmentation of the DNA, and ultimately uptake of the apoptotic cell by a phagocyte. In contrast, in necrosis the homeostatic mechanisms of a cell are disrupted, leading to swelling and eventually bursting of the necrotic cell with spillage of its contents into the surrounding microenvironment. A key aspect of apoptosis (as opposed to necrosis) is that it is a process controlled at several different levels, some of which are susceptible to therapeutic intervention designed to induce or inhibit the occurrence of cell death. Thus, knowledge of the role of apoptosis in the pathogenesis and/or resolution of a disease may lead to novel therapeutic strategies aimed at modifying the course of that disease by manipulating the apoptotic process. Theoretically, apoptosis could be involved in disease processes by at least three separate mechanisms: inappropriate induction of apoptosis resulting in loss of “desired” cells, inappropriate inhibition of apoptosis allowing persistence of “nondesired” cells, and disruption of the apoptotic clearance process leading to secondary necrosis of apoptotic cells and spillage of their cellular contents. All three of these processes have been implicated in the pathogenesis of acute respiratory distress syndrome (ARDS) and in the repair process that follows lung injury. ARDS is characterized pathologically by neutrophilic alveolitis, microthrombi, interstitial edema, destruction of the alveolar epithelial layer with exposure of the basement membrane, and proteinaceous exudates in the airspaces. The cells involved in this area are neutrophils (PMN), alveolar macrophages, alveolar epithelial cells, and capillary endothelial cells. We will review the evidence suggesting that alterations in the apoptotic process of each of these cells might be involved in the pathogenesis and resolution of ARDS.

Is apoptosis important in the pathogenesis and resolution

159

II. Apoptosis in the Pathogenesis of ARDS A. Neutrophil Apoptosis Apoptosis of PMN has been proposed as a key mechanism of resolution of inflammation (1, 2). After migrating from a blood vessel into tissues, PMN cannot return to the circulation, and thus their fate is either to necrose and disintegrate or to become apoptotic and be phagocytized by macrophages and other phagocytic cells. Clearance of inflammation would occur primarily by induction of PMN apoptosis and subsequent uptake of apoptotic PMN by phagocytes. Thus, inhibition of PMN apoptosis and/or impaired phagocytosis of apoptotic PMN would result in persistence of inflammation and, potentially, increased necrosis, with PMN disintegration and spillage of toxic PMN contents into the surrounding microenvironment. This hypothesis predicts that PMN apoptosis should be inhibited during early ARDS, then increase as inflammation resolves. Presumably, inhibition of PMN apoptosis would be associated with persistence of inflammation and a worse outcome. Studies performed in patients with ARDS showed that the percentage of apoptotic PMN in bronchoalveolar lavage (BAL) fluid from patients with early ARDS is relatively low (3) (Fig. 1). However, this was true throughout ARDS, and the percent of apoptotic PMN did not increase as inflammation resolved. Thus, it would appear that the proportion of PMN that are apoptotic at any given time in the airspaces of patients with ARDS is relatively constant. However, this does not necessarily mean that apoptosis of PMN is unchanged during ARDS. If phagocytosis of apoptotic cells is very rapid there could be major changes in the number of PMN that develop apoptosis, even if the proportion of PMN that are apoptotic remains constant. Additional studies demonstrated that soluble mediators present in BAL fluid and plasma from patients with early ARDS inhibit PMN

Acute respiratory distress syndrome

160

Figure 1 Percentage of apoptotic PMN on cytospin preparations of BAL fluid from patients at risk for ARDS and patients with ARDS, measured by morphological criteria. Each point represents data from one patient. Bars represent medians. The horizontal axis shows the day of illness on which the BAL procedure was performed. (From Ref. 3.) apoptosis (4–6). This inhibition was mediated primarily by the cytokine granulocytemacrophage colony-stimulating factor (GM-CSF) and, to a lesser extent, by the cytokines granulocyte-stimulating factor (G-CSF) and interleukin 2 (IL-2) (Fig. 2). Lung fluids from patients at later stages of ARDS or at risk for ARDS had lower concentrations of GM-CSF (Fig. 3). These BAL fluids did not inhibit PMN apoptosis (Fig. 4). These studies suggest that high concentrations of GM-CSF in BAL fluid and plasma inhibit PMN apoptosis during acute inflammation and that as inflammation resolves and GMCSF concentrations return to normal, PMN apoptosis also returns to baseline levels. However, it is not clear whether GM-CSF-mediated inhibition of PMN apoptosis is involved in the pathogenesis of the disease. Although there are anecdotal reports of patients developing ARDS

Is apoptosis important in the pathogenesis and resolution

161

Figure 2 Effect of immunodepleting BAL of GM-CSF, G-CSF, IL-6, and IFN-γ on PMN apoptosis. BAL fluid from patients with ARDS was immunodepleted of GM-CSF, G-CSF, IL-6, and IFN-γ using blocking monoclonal antibodies. Then normal PMN were incubated in either untreated or depleted BAL for 18 hours, and apoptosis measured by Annexin-V binding. The asterisks show statistical significance (p<0.05) compared to the “no-treatment” group. The dagger shows statistical significance (p<0.05) compared to “normal BALF, no treatment.” (From Ref. 3.) after receiving recombinant GM-CSF (7), in more recent series higher concentrations of GM-CSF in BAL fluid from patients with early ARDS are associated with increased survival (4). In summary, studies in humans with ARDS have confirmed that PMN apoptosis is modulated throughout the course of the disease, but a clear link between PMN apoptosis and pathogenesis has not been established. The role of PMN apoptosis in lung injury has also been investigated in animal studies. Parsey et al. (8) isolated parenchymal PMN from mice subject to either hemorrhage or

Acute respiratory distress syndrome

162

endotoxemia and measured the percent of apoptotic PMN over 48 hours using AnnexinV. At baseline, 18.5±1.9% of the PMN in the lungs were apoptotic (Fig. 5). The percentage of apoptotic PMN decreased significantly after one hour of either hemorrhage or endotoxemia, remained low for 24 hours, and returned to baseline levels at 48 hours. This study is important because it measured apoptosis in lung PMN, thus providing direct information about the apoptotic process in lung PMN, as opposed

Figure 3 Concentration GM-CSF in BALF from normal volunteers, BALF from patients on days 1 and 3 of being identified as being at risk for ARDS, and BALF from patients on days 1, 3, 7, and 14 of established ARDS. Each dot represents a single individual, and the bars show median values. *p<0.05, compared to BALF from normal volunteers (ANOVA with Fischer’s post hoc analysis performed using log 10 transformed data). (From Ref. 4.) to measuring the effects of lung fluids on PMN apoptosis. Parsey’s study confirms the human observations suggesting that PMN apoptosis is inhibited during acute inflammation, but does not clarify the importance of modulation of PMN apoptosis in the pathogenesis of lung injury.

Is apoptosis important in the pathogenesis and resolution

163

The question of biological relevance was addressed in a study by Sookhai et al., (8) who investigated the consequences of inhibiting PMN apoptosis on the development of lung injury following ischemia/reperfusion (I/R) in mice. They found that PMN apoptosis was enhanced by the administration of aerosolized dead opsonized Escherichia coli to the animals (Fig. 6). The improvement in PMN apoptosis was associated with enhanced survival and decreased severity of lung injury following I/R (measured by BAL protein concentrations and MPO activity in lung homogenates). Thus in Sookhai et al.’s study, enhancement of PMN apoptosis was beneficial to the host and resulted in less lung injury and improved survival after ischemia reperfusion.

Figure 4 Percentage of apoptotic PMN, as determined by flow cytometry and annexin-V binding, after incubation of normal PMN for 18 hours in either BALF from normal volunteers, or BALF from patients on days 1, 3, 7, and 14 of ARDS. *p<0.05 compared to normal BALF (ANOVA with Scheffe’s post-hoc analysis). (From Ref. 4.)

Acute respiratory distress syndrome

164

In summary, the role of PMN apoptosis in the pathogenesis of acute lung injury remains unclear. Studies in animals and humans support the hypothesis that PMN apoptosis is inhibited during acute inflammation and suggest that this inhibition is mediated by soluble mediators, in particular GM-CSF. However, in humans with early ARDS, higher concentrations of GM-CSF are associated with improved survival, suggesting that inhibition of PMN apoptosis may be beneficial for the host. In contrast to the human studies, in one animal study enhancement rather than inhibition of PMN apoptosis was beneficial to the host. Additional studies are required to determine the biological importance of PMN apoptosis modulation in the pathogenesis of acute lung injury. B. Epithelial Cell Apoptosis The death of epithelial cells during ARDS is likely to result from both necrosis and apoptosis, but the relative contribution of each of these

Figure 5 Hemorrhage or endotoxemia cause changes in neutrophil apoptosis. Mice were either sham-hemorrhaged, hemorrhaged 30% blood volume, or injected i.p. with 5 mg/kg LPS. At the times indicated, lung intraparenchymal pulmonary neutrophil and

Is apoptosis important in the pathogenesis and resolution

165

mononuclear cells were isolated and annexin V assays performed while gating on the neutrophil population. The rate of neutrophil apoptosis in sham-hemorrhaged controls was the same as in unmanipulated control mice. Results are presented as the mean percentage of apoptotic cells ±SEM. *p<0.01 compared to baseline (control). (From Ref. 8.) processes to epithelial injury remains unknown. Bardales et al. (10) identified features consistent with apoptosis in the alveolar epithelium of patients who died with lung injury. Guinee et al. (11) found that bax, a bcl-2 analog that promotes apoptosis, is upregulated in alveolar pneumocytes of humans with diffuse alveolar damage but not in control lungs. In animals, Fujita et al. (12) found that the administration of lipopolysaccharide (LPS) to the lungs of mice is followed 6 hours later by endothelial and alveolar epithelial apoptosis. Extensive alveolar epithelial cell apoptosis has also been identified in other lung diseases, such as pulmonary fibrosis (13). Although very little is known about the mechanisms that trigger apoptosis in alveolar epithelial cells, recent evidence implicates the Fas/FasL system (14). The Fas/FasL system has an important role in the regulation of cell life and death. This system is comprised of the cell membrane surface

Figure 6 Spontaneous apoptosis in pulmonary neutrophils (PMNs) isolated from the bronchoalveolar

Acute respiratory distress syndrome

166

lavage (BAL) fluid in the ischemiareperfusion+saline, ischemiareperfusion Escherichia coli, and the zVAD-FMK and E. coli groups. PMNs isolated from the BAL fluid of all groups (n=5) were cultured in complete RPMI 1640 medium at 37°C in 5% CO2 for 24 hours. PMN apoptosis was assessed at 0, 6, and 18 hours according to the percentage of cells with hypodiploid DNA by flow cytometry. Data are expressed as mean±SD and are representative of five separate experiments. Significance was assessed with the I/R+saline group (*p< 0.05) and the zVAD-FMK and E. coli-treated group. (From Ref. 9.) receptor Fas (CD95) and its natural ligand (FasL) (15). Fas is a 45 kDa type I membrane protein member of the TNF family of surface receptors (16). The natural ligand of Fas is Fas ligand (FasL), a 37 kDa type II protein (17). FasL exists as membrane-bound and soluble forms (sFasL) (18). Binding of Fas to FasL induces apoptosis of susceptible cells by a mechanism involving activation of the ICE-related proteases caspase-8 and caspase3 (19). Membrane-bound FasL mediates lymphocyte-dependent cytotoxicity, clonal deletion of alloreactive T cells, and activation-induced suicide of T cells (20–22). The soluble form of FasL can induce apoptosis of human and rabbit lung epithelial cells (14, 23). Subsets of alveolar epithelial cells express Fas on their surface and undergo apoptosis in response to Fas ligation (24, 25). Pulmonary alveolar and airway epithelial cells express Fas and FasL on their surface (25–28), and expression of mRNA for Fas and FasL in whole lungs increases following the administration of LPS (29). Fas-mediated apoptosis of alveolar epithelial cells is modulated by at least two important mediators: surfactant protein A (SP-A) and angiotensin II (All). SP-A, the primary protein present in pulmonary surfactant, inhibits type II cell apoptosis in vivo (30, 31). This is important because the concentration of SP-A is decreased in BAL fluid from patients with ARDS, which would favor apoptosis of type II cells (32). Interaction of All with its AT1 receptor on epithelial cells is required for Fas-mediated apoptosis of alveolar epithelial cells in vitro (33). This may be relevant for ARDS because the concentration of angiotensinconverting enzyme (ACE), which catabolizes the conversion of angiotensin I to angiotensin II, is increased in BAL fluid from patients with ARDS (34). In humans with ARDS, the soluble form of FasL (sFasL) can be detected the BAL fluid (14, 35) (Fig. 7). Patients who die have higher concentrations of sFasL in the BAL fluid than patients who survive. The

Is apoptosis important in the pathogenesis and resolution

167

Figure 7 Soluble FasL concentrations measured by immunoassay in BALF from 4 normal volunteers, 20 patients at risk for ARDS, and 45 patients with ARDS. The bars represent median values. The numbers in parenthesis indicate the number of patients studied on each day. (From Ref. 14.) sFasL present in lung fluids from patients with ARDS is biologically active and can induce apoptosis of distal lung epithelial cells (14) (Fig. 8). Thus, the alveolar microenvironment of patients with ARDS appears to favor alveolar epithelial cell apoptosis. Animal studies support the hypothesis that activation of the Fas/ FasL system is involved in the pathogenesis of acute and chronic lung injury. The administration of the Fas-activating mAb Jo2 to the lungs of mice results in alveolar epithelial cell apoptosis, permeability changes, and lung inflammation (36, 37), and instillation of recombinant human sFasL into the lungs of rabbits causes alveolar injury with hemorrhage (23). Fas may also have a role in LPS-induced lung injury. In mice, the admin-

Acute respiratory distress syndrome

168

Figure 8 Effect of ARDS BAL fluid on primary distal lung epithelial cells (DLEC). DLEC were incubated in medium supplemented at a 50% concentration with either BAL fluid from 4 normal volunteers, BAL fluid from 5 patients studied on day 1 of ARDS, or BAL fluid from 4 patients studied on day 1 of becoming at risk for ARDS. The cells were incubated for 18 hours at 37°C, 5% CO2, then stained with acridine orange and examined by fluorescence microscopy. The total number of cells and total number of apoptotic cells were counted on a low power field (160×), and the percentage of apoptotic cells was calculated. *p<0.05 by ANOVA with Fischer’s post hoc analysis, as compared with normal BALF. (From Ref. 14.)

Is apoptosis important in the pathogenesis and resolution

169

istration of LPS either by the intravenously or by intratracheal instillation is followed by increased alveolar epithelial permeability and apoptosis of both alveolar epithelial and endothelial cells and also bronchial epithelial cells (12, 29, 38). Interestingly, very early after LPS instillation significant numbers of alveolar wall cells show DNA fragmentation by in situ nick-end labeling (TUNEL), suggesting that cell death can occur very rapidly after an injurious stimulus. Blockade of apoptosis by the broad caspase inhibitor ZVADfmk improves the histological picture and decreases alveolar epithelial apoptosis (39), whereas blockade of the Fas/FasL system with the anti-Fas antibody P2 attenuates increases in albumin leak and alveolar neutrophilia following intratracheal LPS administration in mice (29). The anti-Fas antibody resulted in a decrease in the number of TUNEL-positive cells, as well as in the measurements of permeability and inflammation, suggesting a role for Fas in LPS-mediated lung injury. Thus, recent evidence suggests that the Fas/FasL pathway is important in the pathogenesis of epithelial injury in acute lung injury in animals and humans. These considerations lead to a Fas-centered hypothesis for the pathogenesis of septic lung injury, according to which soluble FasL released by monocytes, macrophages, and epithelial cells induces apoptosis of the alveolar epithelium (and endothelium), leading to disruption of the alveolar-epithelial barrier and increased permeability. This hypothesis has certain limitations, however. First, it has not been shown that in vivo, macrophages or epithelial cells in the lung release sFasL in response to activation. Second, although apoptotic epithelial cells can be identified in animal models of lung injury, they seem to be sparse, although this apparent paucity could be a result of the techniques used to measure apoptosis. Third, other mediators such as NO have been implicated in the development of apoptosis in epithelial cells (40, 41). Finally, activation of the Fas/FasL system can result in activation of inflammatory pathways in macrophages, further complicating the model. In addition to triggering apoptotic pathways, and independently of its pro-apoptotic function, activation of Fas can also lead to NF-κB translocation and cytokine production (42). Evidence suggests that the Fas/FasL system can modulate host defenses. First, alveolar macrophages, which express Fas on their surface, do not become apoptotic in response to Fas ligation, but instead release inflammatory cytokines such as IL-8 (43). Second, mice deficient in Fas (lpr) show impaired PMN recruitment in response to intratracheal bacteria, although bacterial clearance is not affected (44). Third, animals deficient in Fas (lpr) are prone to systemic dissemination of Pseudomonas aeruginosa pneumonia, suggesting that the Fas/FasL system is involved in host defenses against bacteria (45). Thus, the associations between sFasL and ALI noted above could be due to the pro-inflammatory effects of the Fas/FasL system. In other words, it is unclear whether the changes in alveolar permeability and the alveolar epithelial injury seen in response to Fas ligation in the lungs result primarily from the pro-apoptotic effect of Fas ligation on the alveolar epithelium or instead are secondary to the pro-inflammatory effects of the Fas/FasL system on macrophages and on PMN recruitment. C. Endothelial Cell Apoptosis There is less information about apoptosis of endothelial cells as a possible pathogenic mechanism of endothelial dysfunction in ARDS. Assaly et al. have shown that serum

Acute respiratory distress syndrome

170

from patients with systemic capillary leak syndrome and shock induce apoptosis of dermal microvascular endothelial cells, and this is associated with increases in intracellular ROS and in the bax/bc12 expression ratio (46). However, such a study has not been done in a systematic way using serum or plasma from patients with ARDS. In mice, the systemic administration of LPS is associated with endothelial cell apoptosis in several tissues, including the lung (12, 47, 48). Endothelial cell apoptosis can be detected as early as 6 hours after LPS administration (12) and appears to be mediated by a mechanism involving circulating TNF-α and intracellular ceramide generation (47, 49). Endothelial cell apoptosis is modulated by type II cells, which release a heat-stable peptide that protects endothelial cells from TNF-α-mediated apoptosis (50). Although the evidence supporting a pathogenic role of endothelial apoptosis is limited, the available studies suggest that circulating mediators, in particular TNF-α, may induce apoptosis of alveolar capillary endothelial cells.

III. Apoptosis in the Resolution of ARDS Apoptosis also occurs in alveolar airspaces during the repair phase of acute lung injury. Polunovsky et al. suggested that apoptosis of fibroblasts and endothelial cells is an important method of elimination of granulation tissue during the repair phase of acute lung injury (51). They showed that BAL fluids from patients during lung repair induce apoptosis of fibroblasts and endothelial cells and lung tissue from patients recovering from lung injury shows evidence of apoptosis (51). Hadden et al. have shown that fibronectin peptides induce lung fibroblast death in vitro by anoikis (detachment) by blocking adhesion of fibroblasts to the extracellular matrix (52). Fibroblasts themselves may contribute to apoptosis of lung epithelial cells and disruption of the repair process (53). Foci of apoptotic alveolar

Is apoptosis important in the pathogenesis and resolution

171

Figure 9 Clearance of apoptotic cells during lung inflammation, (a) Apoptotic cells (dark brown) in lungs of wild-type and CD44-deficient mice at day 7 after bleomycin. Magnification×100. (b) Quantitation of apoptotic cells after bleomycin treatment. Fifty fields were counted at a magnification of×400 from four experiments. Results are mean±SE,

Acute respiratory distress syndrome

172

n=4. CD44-deficient (black boxes) and wild-type (white boxes) mice, n=3. (c) Human apoptotic PMNs were instilled into the lungs of LPS-treated mice; the numbers of recovered cells by BAL were counted (p=0.0096) and (d) phagocytotic index was calculated (p=0.0029). Values reported are the mean ±SE from 12 animals in each group, (e) CD44-deficient mice generate decreased concentrations of active TGF-1. TGF-1 was measured in 10-fold concentrated BAL fluids, combined from four different wildtype (white boxes) and CD44-deficient mice (black boxes) at each time point, n=3. Results are mean±SE from three experiments. Active TGF-1 is shown in the left panel and total TGF-1 in the right panel. (From Ref. 56.) epithelium are present near areas of fibroblast proliferation in patients with pulmonary fibrosis (13). This appears to be mediated by release of angiotensinogen by myofibroblasts, subsequent conversion into angiotensin II, and angiotensin-mediated apoptosis of epithelial cells (33, 54, 55). Studies by Teder et al. suggest that failure to clear apoptotic PMN may lead to persistence of inflammation and increased death (56). In a mouse model of bleomycininduced lung injury, they found that mice deficient in the transmembrane adhesion receptor CD44 failed to clear inflammation, as compared to wild-type mice (Fig. 9). This was associated with increased numbers of apoptotic cells in the lungs and a significant failure to clear apoptotic PMN. The effect was reversed by adoptive transfer of normal marrow cells. This is important, because CD44 increases phagocytosis of apoptotic neutrophils by macrophages in vitro (57). However, CD44 has other functions such as clearance of the glycosaminoglycan hyaluronan, which in turn can increase the synthesis of chemokines such as IL-8 (58). A drawback of this study is that it does not show whether the findings were primarily due to the effects of CD44 on the clearance of apoptotic PMN or on clearance of hyaluronan and decreased production of chemokines. In another study using a rat model of oleic acid-induced lung injury, Hussain et al. demonstrated that the resolution phase of lung injury is associated with generalized apoptosis of PMN and uptake of apoptotic PMN by macrophages (59). Thus, the studies of Teder and Hussain support the hypothesis that PMN apoptosis is an important mechanism of resolution of inflammation, but these studies do not clarify whether this is

Is apoptosis important in the pathogenesis and resolution

173

the most important mechanism, or whether the key mechanism is downregulation of chemokine production pathways. Neutrophil apoptosis may also indirectly contribute to resolution of inflammation by modulating the production of pro-inflammatory cytokines by alveolar macrophages. Macrophage phagocytosis of apoptotic PMN inhibits macrophage production of the proinflammatory cytokines IL-1β, IL-8, IL-10, GM-CSF, TNF-α, and increases TGF-β1, PGE2, and platelet-activating factor (PAF), all of which have anti-inflammatory properties (57). Thus an increase in PMN apoptosis leading to increase phagocytosis of apoptotic PMN could “turn down” the inflammatory phenotype in activated alveolar macrophages, favoring resolution of inflammation, while at the same time increasing fibroblast proliferation mediated by TGF-β1.

IV. Summary and Conclusions An important difficulty regarding the interpretation of available studies on apoptosis and ARDS is that the current methods available to study the apoptotic process are limited. The apoptotic process is a dynamic process, involving not only the development or inhibition of apoptosis in a specific cell, but also its removal by nearby phagocytes. An increase in the number of apoptotic cells in tissue may be due to impairment of the clearance process or to an increase in apoptosis. Yet the majority of studies available rely on measurements of apoptotic cells, and no effective method is available to evaluate apoptotic cell clearance in vivo. Another problem is that many of the methods available to detect apoptosis are controversial. This controversy stems from two facts: (1) some of these methods, like the end-nick labeling assay (TUNEL) and also Annexin-V labeling may become positive in necrotic cells; (2) in the absence of effective clearance mechanisms, apoptotic cells will undergo secondary necrosis. No method exists at the present time that can reliably differentiate cells that have died from primary necrosis from cells that are apoptotic and undergoing secondary necrosis. Another limitation of available studies is that most of our knowledge is derived from either animal studies or in vitro studies, rather than human studies, and extrapolating the results from animal studies to human ARDS is difficult because none of the animal models of lung injury currently available accurately reproduces clinical ARDS. Despite these limitations, some conclusions can be drawn from the literature about the role of apoptosis in lung injury. It is likely that PMN apoptosis is inhibited during the early phase of ARDS, compared to normal lungs, and that this inhibition is mediated primarily by GM-CSF. However, the biological consequences of inhibiting PMN apoptosis are unclear and may actually be of benefit to the host. Despite the prevalent paradigm that PMN apoptosis is an important mechanism of resolution of inflammation, this hypothesis has not been proven true in human ARDS. Additional studies are required to clarify these issues. The evidence suggesting a possible role for apoptosis of alveolar epithelial cells in the pathogenesis of the epithelial injury seen in ARDS is stronger, particularly because Fas activation can induce in mice a form of lung injury characterized by PMN infiltration and permeability increases, whereas inhibition of the Fas/FasL system protects against LPS-

Acute respiratory distress syndrome

174

induced lung injury. However, the exact role of alveolar epithelial cell apoptosis (whether mediated by the Fas/FasL system or other mediators) in ARDS is also unclear at this time. Finally, the evidence regarding a role for apoptosis in the repair process and resolution of inflammation is very exciting. An interesting hypothesis can be derived from the available data, suggesting that release of angiotensinogen by fibroblasts and FasL by macrophages could result in apoptosis of the alveolar epithelium; phagocytosis of these apoptotic cells by macrophages would induce release of TGF-β1, and this TGF-β1 would then promote further proliferation of fibroblasts, resulting in abnormal repair. In conclusion, the current evidence suggests that it is highly likely that apoptosis may have a role in the pathogenesis and repair of ARDS, but the evidence remains inconclusive and further studies are needed to determine its true biological relevance.

References 1. Cox G, Gauldie J, Jordana M. Bronchial epithelial-cell derived cytokines (G-CSF and GM-CSF) promote the survival of peripheral blood neutrophils in vitro. Am J Respir Cell Mol Biol 1992; 7:507–513. 2. Grigg JM, Savill JS, Sarraf C, Haslett C, Silverman M. Neutrophil apoptosis and clearance from neonatal lungs. Lancet 1991; 338:720–722. 3. Matute-Bello G, Liles WC, Radella F, Steinberg KP, Ruzinski JT, Jonas M, Chi EY, Hudson LD, Martin TR. Neutrophil apoptosis in the acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 156:1969–1977. 4. Matute-Bello G, Liles WC, Radella II F, Steinberg KP, Ruzinski JT, Hudson LD, Martin TR. Modulation of neutrophil apoptosis by G-CSF and GM-CSF during the course of the acute respiratory distress syndrome (ARDS). Crit Care Med 2000; 28:1–7. 5. Lesur O, Kokis A, Hermans C, Fulop T, Bernard A, Lane D. Interleukin-2 involvement in early acute respiratory distress syndrome: relationship with polymorphonuclear neutrophil apoptosis and patient survival. Crit Care Med 2000; 28:3814–22. 6. Goodman ER, Stricker P, Velavicius M, Fonseca R, Kleinstein E, Lavery R, Deitch EA, Hauser CJ, Simms HH. Role of granulocyte-macrophage colony-stimulating factor and its receptor in the genesis of acute respiratory distress syndrome through an effect on neutrophil apoptosis. Arch Surg 1999; 134:1049–1054. 7. Verhoef G, Boogaerts M. Treatment with granulocyte-macrophage colony stimulating factor and the adult respiratory distress syndrome. Am J Hematol 1991; 36:285–287. 8. Parsey MV, Kaneko D, Shenkar R, Abraham E. Neutrophil apoptosis in the lung after hemorrhage or endotoxemia: apoptosis and migration are independent of IL-1β. Clin Immunol 1999; 91:219–225. 9. Sookhai S, Wang JJ, McCourt M, Kirwan W, Bouchier-Hayes D, Redmond P. A novel therapeutic strategy for attenuating neutrophil-mediated lung injury in vivo. Ann Surg 2002; 235:285–291. 10. Bardales RH, Xie SS, Schaefer RF, Hsu SM. Apoptosis is a major pathway responsible for the resolution of type II pneumocytes in acute lung injury. Am J Pathol 1996; 149:845–852. 11. Guinee D, Jr., Brambilla E, Fleming M, Hayashi T, Rahn M, Koss M, Ferrans V, Travis W. The potential role of BAX and BCL-2 expression in diffuse alveolar damage. Am J Pathol 1997; 151:999–1007. 12. Fujita M, Kuwano K, Kunitake R, Hagimoto N, Miyazaki H, Kaneko Y, Kawasaki M, Maeyama T, Hara N. Endothelial cell apoptosis in lipopolysaccharide-induced lung injury in mice. Int Arch Allergy Immunol 1998; 117: 202–208.

Is apoptosis important in the pathogenesis and resolution

175

13. Uhal BD, Joshi I, Hughes WF, Ramos C, Pardo A, Selman M. Alveolar epithelial cell death adjacent to underlying myofibroblasts in advanced fibrotic human lung. Am J Physiol 1998; 275:L1192–1199. 14. Matute-Bello G, Liles WC, Steinberg KP, Kiener PA, Mongovin S, Chi EY, Jonas M, Martin TR. Soluble Fas ligand induces epithelial cell apoptosis in humans with acute lung injury (ARDS). J Immunol 1999; 163:2217–2225. 15. Nagata S, Golstein P. The Fas death factor. Science 1995; 267:1449–1456. 16. Itoh N, Yonehara S, Ishii A, Yonehara M, Mizushima S, Sameshima M, Hase A, Seto Y, Nagata S. The polypeptide encoded by the cDNA for human cell surface antigen Fas can mediate apoptosis. Cell 1991; 66:233–243. 17. Suda T, Takahashi T, Golstein P, Nagata S. Molecular cloning and expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell 1993; 75:1169–1178. 18. Tanaka M, Suda T, Takahashi T, Nagata S. Expression of the functional soluble form of human Fas ligand in activated lymphocytes. EMBO J 1995; 14:1129–1135. 19. Takahashi A, Hirata H, Yonehara S, Imai Y, Lee KK, Moyer RW, Turner PC, Mesner PW, Okazaki T, Sawai H, Kishi S, Yamamoto K, Okuma M, Sasada M. Affinity labeling displays the stepwise activation of ICE-related proteases by Fas, staurosporine, and CrmA-sensitive caspase-8. Oncogene 1997; 14:2741–2752. 20. Lowin B, Hahne M, Mattman C, Tschopp J. Cytolytic T-cell cytotoxicity is mediated through perform and Fas lytic pathways. Nature 1994; 370:650–652. 21. Kagi D, Vignaux F, Ledermann B, Burki K, Depraetere V, Nagata S, Hengartner H, Golstein P. Fas and perforin pathways as major mechanisms of T-cell mediated cytotoxicity. Science 1994; 265:528–530. 22. Dhein J, Walczak H, Baumler C, Debatin K, Krammer P. Autocrine T-cell suicide mediated by APO-1/(Fas/CD95). Nature (Lond) 1995; 373:438–441. 23. Matute-Bello G, Liles WC, Frevert CW, Nakamura M, Ballman K, Vathanaprida C, Kiener PA, Martin TR. Recombinant human Fas-ligand induces alveolar epithelial cell apoptosis and lung injury in rabbits. Am J Physiol 2001; 281:L328–L335. 24. Wen L, Madani K, Fahrni JA, Duncan SR, Rosen GD. Dexamethasone inhibits lung epithelial cell apoptosis induced by IFN-γ and Fas. Am J Physiol 1997; 273:L921–L929. 25. Fine A, Anderson NL, Rothstein TL, Williams MC, Gochuico BR. Fas expression in pulmonary alveolar type II cells. Am J Physiol 1997; 273:L64– L71. 26. Hamann KJ, Dorscheid DR, Ko FD, Conforti AE, Sperling AI, Rabe KF, White SR. Expression of Fas (CD95) and FasL (CD95L) in human airway epithelium. Am J Respir Cell Mol Biol 1998; 19:537–542. 27. Druilhe A, Wallaert B, Tsicopoulos A, Lapa e Silva JR, Tillie-Leblond I, Tonnel AB, Pretolani M. Apoptosis, proliferation, and expression of Bcl-2, Fas, and Fas ligand in bronchial biopsies from asthmatics. Am J Respir Cell Mol Biol 1998; 19:747–57. 28. Gochuico BR, Miranda KM, Hessel EM, De Bie JJ, Van Oosterhout AJ, Cruikshank WW, Fine A. Airway epithelial Fas ligand expression: potential role in modulating bronchial inflammation. Am J Physiol 1998; 274:L444–449. 29. Kitamura Y, Hashimoto S, Mizuta N, Kobayashi A, Kooguchi K, Fujiwara I, Nakajima H. Fas/FasL-dependent apoptosis of alveolar cells after lipopolysaccharide-induced lung injury in mice. Am J Respir Crit Care Med 2001; 163:762–769. 30. de Lara LV, Becerril C, Montano M, Ramos C, Maldonado V, Melendez J, Phelps DS, Pardo A, Selman M. Surfactant components modulate fibroblast apoptosis and type I collagen and collagenase-1 expression. Am J Physiol Lung Cell Mol Physiol 2000; 279:L950–957. 31. White MK, Baireddy V, Strayer DS. Natural protection from apoptosis by surfactant protein A in type II pneumocytes. Exp Cell Res 2001; 263:183–192. 32. Greene KE, Wright JR, Steinberg KP, Ruzinski JT, Caldwell E, Wong WB, Hull W, Whitsett JA, Akino T, Kuroki Y, Martin TR. Serial changes in surfactant-associated proteins in lung and serum before and after the onset of ARDS. Am J Respir Crit Care Med 1999; 160:1843–1850.

Acute respiratory distress syndrome

176

33. Wang R, Zagariya A, Ang E, Ibarra-Sunga O, Uhal BD. Fas-induced apoptosis of alveolar epithelial cells requires ANG II generation and receptor interaction. Am J Physiol 1999; 277:L1245–1250. 34. Idell S, Kueppers F, Lippmann M, Rosen H, Niederman M, Fein A. Angiotensin converting enzyme in bronchoalveolar lavage in ARDS. Chest 1987; 91: 52–56. 35. Hashimoto S, Kobayashi A, Kooguchi K, Kitamura Y, Onodera H, Nakajima H. Upregulation of two death pathways of perforin/granzyme and FasL/Fas in septic acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 161:237–243. 36. Matute-Bello G, Winn RK, Jonas M, Chi EY, Martin TR, Liles WC. Fas (CD95) induces alveolar epithelial cell apoptosis in vivo: implications for acute pulmonary inflammation. Am J Pathol 2001; 158:153–161. 37. Kuwano K, Hagimoto N, Kawasaki M, Yatumi T, Nakamura N, Nagata S, Suda T, Kunitake R, Maeyama T, Miyazoki H, Hara N. Essential roles of the Fas-Fas ligand pathway in the development of pulmonary fibrosis. J Clin Invest 1999; 104:13–19. 38. Vernooy JHJ, Dentener MA, van Suylen RJ, Buurman WA, Wouters EFM. Intratracheal instillation of lipopolysaccharide in mice induces apoptosis in bronchial epithelial cells. No role for tumor necrosis factor-alpha and infiltrating neutrophils. Am. J. Respir. Cell Mol. Biol. 2001; 24:569–576. 39. Kuwano K, Kunitake R, Maeyama T, Hagimoto N, Kawasaki M, Matsuba T, Yoshimi M, Inoshima I, Yoshida K, Hara N. Attenuation of bleomycin-induced pneumopathy in mice by a caspase inhibitor. Am J Physiol Lung Cell Mol Physiol 2001; 280:L316–325. 40. Janssen YM, Matalon S, Mossman BT. Differential induction of c-fos, c-jun, and apoptosis in lung epithelial cells exposed to ROS or RNS. Am J Physiol 1997; 273:L789–796. 41. Smith JD, McLean SD, Nakayama DK. Nitric oxide causes apoptosis in pulmonary vascular smooth muscle cells. J Surg Res 1998; 79:121–127. 42. Rensing-Ehl A, Hess S, Ziegler-Heitbrock HW, Riethmüller G, Engelmann H. Fas/Apo-1 activates nuclear factor kB and induces interleukin-6 production. J Inflammation 1995; 45:161– 174. 43. Park DR, Thomsen AR, Frevert CW, Skerrett SJ, Martin TR, Kiener PA, Liles WC. Fas ligation activates human monocyte and macrophage inflammatory responses (abstr). Am J Respir Crit Care Med 2000; 161:A901. 44. Matute-Bello G, Frevert CW, Liles WC, Nakamura M, Ruzinski JT, Ballman K, Wong VA, Vathanaprida C, Martin TR. The Fas/FasL system mediates epithelial injury, but not pulmonary host defenses, in response to inhaled bacteria. Infection and immunity 2001; 69:5768–5776. 45. Grassme H, Kirschnek S, Riethmueller J, Riehle A, von Kurthy G, Lang F, Weller M, Gulbins E. CD95/CD95 ligand interactions on epithelial cells in host defense to Pseudomonas aeruginosa. Science 2000; 290:527–530. 46. Assaly R, Olson D, Hammersley J, Fan PS, Liu J, Shapiro JI, Kahaleh MB. Initial evidence of endothelial cell apoptosis as a mechanism of systemic capillary leak syndrome. Chest 2001; 120:1301–1308. 47. Haimovitz-Friedman A, Cordon-Cardo C, Bayoumy S, Garzotto M, McLoughlin M, Gallily R, Edwards CK, 3rd, Schuchman EH, Fuks Z, Kolesnick R. Lipopolysaccharide induces disseminated endothelial apoptosis requiring ceramide generation. J Exp Med 1997; 186:1831– 1841. 48. Kawasaki M, Kuwano K, Hagimoto N, Matsuba T, Kunitake R, Tanaka T, Maeyama T, Hara N. Protection from lethal apoptosis in lipopolysaccharide-induced acute lung injury in mice by a caspase inhibitor. Am J Pathol 2000; 157:597–603. 49. Polunovsky VA, Wendt CH, Ingbar DH, Peterson MS, Bitterman PB. Induction of endothelial cell apoptosis by TNF alpha: modulation by inhibitors of protein synthesis. Exp Cell Res 1994; 214:584–594. 50. Wendt CH, Polunovsky VA, Peterson MS, Bitterman PB, Ingbar DH. Alveolar epithelial cells regulate the induction of endothelial cell apoptosis. Am J Physiol 1994; 267:C893–900.

Is apoptosis important in the pathogenesis and resolution

177

51. Polunovsky VA, Chen B, Henke C, Snover D, Wendt C, Ingbar DH, Bitterman PB. Role of mesenchymal cell death in lung remodeling after injury. J Clin Invest 1993; 92:388–397. 52. Hadden HL, Henke CA. Induction of lung fibroblast apoptosis by soluble fibronectin peptides. Am J Respir Crit Care Med 2000; 162:1553–1560. 53. Uhal B, Joshi I, True A, Mundle S, Raza A, Pardo A, Selman M. Fibroblasts isolated after fibrotic lung injury induce apoptosis of alveolar epithelial cells in vitro. Am. J. Physiol. 1995; 269:L819–L828. 54. Wang R, Ramos C, Joshi I, Zagariya A, Pardo A, Selman M, Uhal BD. Human lung myofibroblast-derived inducers of alveolar epithelial apoptosis identified as angiotensin peptides. Am J Physiol 1999; 277:L1158–1164. 55. Wang R, Zagariya A, Ibarra-Sunga O, Gidea C, Ang E, Deshmukh S, Chaudhary G, Baraboutis J, Filippatos G, Uhal BD. Angiotensin II induces apoptosis in human and rat alveolar epithelial cells. Am J Physiol 1999; 276:L885–889. 56. Teder P, Vandivier RW, Jiang D, Liang J, Cohn L, Pure E, Henson PM, Noble PW. Resolution of lung inflammation by CD44. Science 2002; 296:155–158. 57. Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY, Henson PM. Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J Clin Invest 1998; 101:890–898. 58. Haslinger B, Mandl-Weber S, Sellmayer A, Sitter T. Hyaluronan fragments induce the synthesis of MCP-1 and IL-8 in cultured human peritoneal mesothelial cells. Cell Tissue Res 2001; 305:79–86. 59. Hussain N, Wu F, Zhu L, Thrall RS, Kresch MJ. Neutrophil apoptosis during the development and resolution of oleic acid-induced acute lung injury in the rat. Am J Respir Cell Mol Biol 1998; 19:867–874.

9 Pathogenesis of Ventilator-Induced Lung Injury JAMES A.FRANK University of California, San Francisco San Francisco, California, U.S.A. ARTHUR S.SLUTSKY University of Toronto and St. Michael’s Hospital Toronto, Ontario, Canada YUMIKO IMAI University of Toronto and Toronto General Hospital Toronto, Ontario, Canada

I. Introduction From the time of its inception, the role of mechanical ventilation in acute respiratory failure has been duplicitous—life saving on one hand, while injury promoting on the other. Before the use of positive pressure ventilation became widespread, mortality from acute hypoxemic respiratory failure was nearly 100%. Mortality was still nearly 60% in 1971, when Petty and Ashbaugh (1) first reported on the use of positive pressure ventilation for the treatment of ARDS. At that time, clinicians had already raised concerns about the potential harmful effects of mechanical ventilation. For example, in 1968 Sladen and coworkers (2) reported that prolonged mechanical ventilation resulted in worsening oxygenation, increased lung water, and decreased compliance in patients with ventilatory failure. Similar findings had been reported in animal models as early as the 1940s (3–5). The active role of mechanical ventilation in determining patient outcome was convincingly demonstrated by the recent National Institute of Health (NIH)-sponsored ARDS Network study of 861 patients, which compared low tidal volume ventilation to conventional ventilation. Tidal volume reduction to 6 mL/kg (predicted body weight) from the conventional 12 mL/kg at similar levels of positive end-expiratory (PEEP) resulted in a 9% reduction in absolute mortality (6). This is the only therapeutic intervention that in a large study has convincingly demonstrated a reduction in mortality from ARDS since the initial description of the syndrome; however, the mechanisms of this protective effect are largely unknown.

Pathogenesis of ventilator-induced lung injury

179

The spectrum of ventilator-induced lung injury extends from macroscopic air leaks to ultrastructural changes and molecular/cellular responses. In this chapter we will review the proposed mechanisms for ventilator-induced lung injury, focusing primarily on pathophysiological and molecular/cellular principles.

II. Adverse Effects of Mechanical Ventilation Two general terms are used to describe the injury that is thought to be caused by mechanical ventilation. Ventilator-induced lung injury (VILI) is usually used in the context of animal studies, where the direct effect of mechanical ventilation on lung injury can be definitively ascertained since it is possible to perform appropriate controls. In clinical studies it is often difficult to determine whether the lung injury is due to de novo injury directly attributable to the ventilatory strategy or due to amplification of preexisting injury or simply due to worsening of the underlying disease process that precipitated the acute respiratory failure. Thus, in the clinical context the term that is often used is ventilator-associated lung injury (VALI). For this chapter we will use the term VILI in a general sense to refer to investigations in this area of research and also in the specific context of animal models of injurious ventilation in previously normal lungs. The term VALI will be used to refer to clinical studies and animal models where the direct role of mechanical ventilation on producing the injury is less clear, as is often the case when injurious ventilation is applied to previously injured lungs. Although the focus of this chapter is the cellular and molecular basis of VILI, it is important to acknowledge the other unintended adverse events associated with mechanical ventilation (Table 1). These include large air leaks, which result in the extravasation of air into the pleural space, mediastinum, peritoneum, or subcutaneous tissue. Although such leaks occur in approximately 10% patients with ARDS, the development of an air leak is not a major factor contributing to increased mortality in the vast majority of these patients (7, 8). Endotracheal intubation and mechanical ventilation also increase the risk of aspiration, nosocomial pneumonia, and sinusitis.

Table 1 Complications of Mechanical Ventilation Complication Air leaks

Clinical results Pneumothorax Pneumomediastinum Pneumoperitoneum Subcutaneous emphysema

Ventilator-induced alveolar epithelial and endothelial injury

Increased permeability/pulmonary edema formation Increased plasma and edema fluid levels of inflammatory cytokines Impaired alveolar fluid transport Impaired gas exchange

Acute respiratory distress syndrome

180

Loss of compartmentalization Multisystem organ failure Infectious

Ventilator-associated pneumonia Sinusitis

Traumatic

Tracheal stenosis Unplanned extubation Dental Nasopharyngeal/oral trauma

Other

Hemodynamic compromise Aspiration Respiratory alkalosis/acidosis Oxygen toxicity

How does mechanical ventilation injure the lung? Clearly, mechanical stresses can lead to gross cell death by disruption of the cytoskeleton or plasma membrane. However, at the cellular level, mechanical stresses induce changes in intra- and intercellular signaling cascades. This conversion of a physical stimulus into chemical signals inside the cell is termed mechanotransduction. The extent to which these more subtle changes in cell function influence permeability, ion transport, structural and functional protein expression, inflammation, and cell death in VILI is the focus of ongoing research.

III. Mechanical Forces During Positive-Pressure Ventilation During tidal breathing, the lung is exposed to a variety of mechanical forces. On the most basic level, these include tensile strain, stress, and shear stress (9, 10). Strain, commonly referred to as stretch, is defined as a change in length relative to the original length. Stress is the force per unit area as in compression. Shear stress is the force per unit area in the direction of flow. A. Nature of Mechanical Forces During Ventilation Figure 1 schematically illustrates the forces that have been postulated to act on the pulmonary airways, alveoli, and capillaries. Alveolar inflation results

Pathogenesis of ventilator-induced lung injury

181

Figure 1 Mechanical forces in the lung during ventilation. Alveolar wall tension (Tw) increases with transpulmonary pressure (PTP) and alveolar size (r=radius) according to the law of Laplace (Tw=PTP×r/2). Force from the inflation of adjacent alveoli (λ) also favors alveolar expansion (interdependence). Elastic recoil of the chest wall (Ew) also favors alveolar expansion when lung volume is low. Surface tension (TST) at the air/liquid interface in the alveolus favors alveolar collapse, as does the elastic recoil of the lung (EL). Lung capillaries are also exposed to increasing wall tension as transmural vascular pressure (PTM) increases with lung inflation. Vessels are also exposed to tensile elongation (TEL) during lung inflation. Capillary

Acute respiratory distress syndrome

182

endothelial cells are exposed to shear stress in the direction of blood flow. Airway and alveolar epithelial cells are also exposed to shear stress as air and edema fluid are displaced during reinflation of collapsed lung (not shown). from an increased pressure gradient across the alveolar wall. Alveolar wall tension increases accordingly as per the law of Laplace: T=P×r/2, where T is wall tension, P is the pressure gradient across the alveolar wall, and r is the radius of the alveolus. Intrinsic elasticity of the lung and surface tension at the air/liquid interface in the alveolus oppose expansion, whereas the forces due to adjacent expanded alveoli and elasticity of the chest wall oppose collapse. Surface tension also serves to support the capillary endothelium by opposing endothelial cell bowing into the alveolus induced by vascular hydrostatic pressure (10, 11). The blood vessels of the lung are exposed to additional forces. As lung volume increases, blood vessels are exposed to longitudinal tension. Extra-alveolar blood vessels are potentially exposed to greater wall tension as lung volume increases because perivascular pressure decreases with lung expansion resulting in increased transmural pressure. It is also important to note that gravitational forces affect distribution of blood flow and therefore may affect regional transmural vascular pressure. In addition, shear forces due to blood, pulmonary edema fluid, and air flow affect the lumenal cells of both the capillary and airspaces. Commonly, shear stress refers to the mechanical stress experienced by endothelial cells during blood flow, but in the injured lung, airway and alveolar epithelial cells may also experience shear stress as edema fluid moves through the airways and when collapsed lung units are reinflated (12). Therefore, changes in lung inflation affect both the endothelium and epithelium. This conclusion is supported by the studies of Costello and colleagues (13) in which pulmonary capillary hypertension (33–39 cmH2O) induced both endothelial and alveolar epithelial cell breaks, especially at high lung volumes. VILI is thus the result of a complex interplay among various mechanical forces acting on lung structures during mechanical ventilation. Because of the complexity of the pulmonary organ structure, the variety of cell types, and the variety of mechanical forces to which cells are exposed, possible mechanisms of cellular/molecular responses caused by mechanical ventilation may vary widely. B. Mechanical Forces and Injury In experimental studies of positive-pressure ventilation, lung injury primarily results from excessive lung volume, particularly at end-inspiration, and from insufficient functional residual capacity (FRC) where repetitive opening and collapse of atelectatic lung regions could generate strain and shear force. Under this paradigm, the mechanical forces that precipitate VILI are amplified in the heterogeneously injured lung (Fig. 2). This results in large part from airspace edema formation as edema fills airspaces and inactivates

Pathogenesis of ventilator-induced lung injury

183

Figure 2 Mechanisms of ventilatorassociated lung injury. Left panel shows lung regions at end-expiration. Right panel shows the same lung regions at end-inspiration. (A) Patent alveoli are exposed to increased strain during tidal ventilation due to the uneven distribution of the inflated volume. (B) Some alveoli are damaged by excessive stress caused by the uneven expansion of surrounding lung regions at the margins between atelectatic and aerated alveoli. (C) Small bronchioles and alveoli may be injured by mechanical forces resulting from repeated opening and closing. (From Ref. 175.) surfactant, resulting in increased surface tension and alveolar and small airway collapse (14, 15). Accordingly, heterogeneous lung injury can be modeled as having three

Acute respiratory distress syndrome

184

different compartments: (1) regions of airway and alveolar flooding, in which little lung volume can be recruited during tidal breathing; (2) regions with normal compliance and aeration, appearing to be uninvolved with disease; (3) intermediate regions in which alveolar collapse and flooding are apparent but where aeration can be restored. At the frequencies associated with conventional ventilation, a positive pressure tidal breath will distribute preferentially to the normally aerated areas. As a result, these regions are vulnerable to alveolar overdistention. ARDS is characterized by the formation of protein-rich pulmonary edema, diffuse alveolar damage, and impaired airspace edema clearance in most patients (16). The result is a heterogeneous distribution of injury and edema in the lung. Interdependence refers to the regional traction forces exerted by adjacent lung segments to favor uniform expansion of the lung. These forces arise within the lung because the wall of each alveolar space is also the wall of the neighboring alveolus. In fully recruited normal lung, the alveolardistending force can be thought of as the transpulmonary pressure (alveolar pressure— pleural pressure). However, once heterogeneity is introduced, the local distending forces will differ in a manner so as to oppose the heterogeneity and restore lung expansion. As a result, these regions are vulnerable to alveolar overdistention. Mead et al. (5) postulated that at a transpulmonary pressure of 30 cmH2O, “the pressure tending to expand an atelectatic region surrounded by a fully expanded lung would be approximately 140 cmH2O.” Therefore, in the heterogeneously injured lung, strain may be greater at areas where inflated lung is adjacent to atelectatic or fluid-filled lung due to interdependence (Fig. 2). This may also explain in part why traditional tidal volumes (10–15 mL/kg) can result in alveolar over-distention and subsequent VILI in patients with ARDS and, in part, why ventilatory strategies aimed at maintaining uniform lung recruitment are less injurious than those strategies that permit development of marked regional disparities in lung volumes. At the cellular and molecular levels, the distribution of forces across the fused cytoskeleton-extracellular matrix scaffold of the alveolar-capillary barrier is complex; however, it is reasonable to assume that some of the force applied to one side of the barrier is transmitted to neighboring cells, the extracellular matrix (ECM), and the cytoskeletal structure of the cells on the opposite side of the barrier. As proposed by Ingber et al. (17, 18), the cytoskeleton behaves as a tensegrity structure referring to the architectural principle whereby continuous tension is coupled with compression-resistant supports yielding an efficient structure that yields without breaking. External force applied to this type of structure disrupts the tensional equilibrium and is distributed throughout the entire structure regardless of where the force is applied. In this way, mechanical strain maybe focused to specific regions of the cytoskeleton—for example, at focal adhesion complexes (FACs). FACs are formed by integrin clusters bound to the extracellular ECM and the cytoskeleton, creating a physical link between the ECM and the cytoskeleton. Multiple other intracellular signaling proteins are associated with the FAC, including focal adhesion kinases (FAK), p60src, actin filament-associated proteins, among other proteins, (see Sec. VI). Alternatively, mechanical force may act on cells in a more local fashion. For example, plasma membrane disruption resulting from strain or shear stress and independent of changes in the cytoskeleton may predominate.

Pathogenesis of ventilator-induced lung injury

185

IV. Lung Injury from Overdistention Although all of the forces listed above occur in the lung during the respiratory cycle, it stands to reason that alveolar epithelial cell tensile strain is greatest at end inspiration. Therefore, the larger the lung volume/ transpulmonary pressure at end inspiration, the greater the strain; however, in normal rat lungs, alveolar strain as estimated by the change in basement membrane surface area is not linearly related to lung volume (19). Tschumperlin and Margulies (19) and Bachofen et al. (20) found that, following volume history standardization, epithelial basement membrane surface area increased only 12% as lung volume increased from 24 to 82% of total lung capacity (TLC). When lung volume was increased from 82 to 100% of TLC, the epithelial basement membrane surface area sharply increased an additional 20% (for a total increase of 40% from surface area at 24% TLC). This finding is best explained by an initial unfolding of the alveolar epithelial basement membrane during lung inflation at low lung volumes, followed by stretching at higher lung volumes. A. Excessive Lung Volume in Clinical Studies The most convincing evidence that excessive end-inspiratory lung volume is injurious to patients came from the recently completed ARDS Network study, which compared low tidal volume ventilation (6 mL/kg ideal body weight) and an inspiratory plateau pressure limit of 30 cmH2O to conventional mechanical ventilation with a tidal volume of 12 mL/kg and a similar level of positive end-expiratory pressure (PEEP) (6). In this study of 861 patients with ARDS or ALI, low tidal volume ventilation reduced mortality from 40 to 31% compared with conventional tidal volume ventilation. The protective effect persisted regardless of the underlying etiology of ARDS or the respiratory system compliance (21). As with other studies of ARDS, the most common causes of death were withdrawal of care, sepsis, and multiple organ dysfunction syndrome (MODS), not hypoxemia. In fact, oxygenation was not different between the groups (6). Other studies of protective ventilation incorporating low tidal volumes have yielded conflicting results (22–26). Although the direct measurement of lung volumes in mechanically ventilated patients as a guide for setting PEEP and tidal volume may be an ideal approach, accurate measurement is difficult, and this has not yet been recommended. B. Experimental Models of Lung Overdistention In the years following the first clinical reports of the adverse effects of positive pressure ventilation, researchers began investigating the effects of high-volume ventilation strategies on normal animal lungs. The lung injury induced by mechanical ventilation in these experimental models is characterized by high-protein pulmonary edema, diffuse alveolar damage, and infiltration of inflammatory cells. These findings were also reported in ex vivo and in vivo animal models of ARDS (injured animal lungs). The pathological lesion of VILI in these studies is not specific and closely resembles lung injury resulting from clinical ARDS (27–29). These experimental studies serve as the foundation of our understanding of the potential mechanisms of VILI in ARDS patients who have underlying lung disease.

Acute respiratory distress syndrome

186

Pulmonary Edema and High Tidal Volume Ventilation The first comprehensive report of the harmful effects of high-pressure/high-volume ventilation in normal lungs was that of Webb and Tierney (29). In this classic study, the investigators ventilated rats for up to one hour with one of three tidal volumes (12.5±0.1, 30±0.2, or 43±0.7 mL/kg) without PEEP, or with 10 cmH2O of PEEP and a tidal volume of either 11.6±0.4 or 15.6±0.48 mL/kg (Fig. 3). These volumes were the result of matching peak inspiratory pressures in the two highest tidal volume groups. High tidal volume, zero PEEP ventilation resulted in the development of severe, protein-rich interstitial and airspace edema. Ventilation with a tidal volume of 30 mL/kg resulted in a significant reduction in the severity of edema as measured by wet lung weight and a histological edema score. Further reducing tidal volume to 12.5 mL/kg prevented the development of edema in their model. Combining PEEP with reduced tidal volume ventilation also reduced the severity of pulmonary edema. The rats ventilated with a tidal volume of 11.6 or 15.6 mL/kg and a PEEP level of 10 cmH2O developed significantly less pulmonary edema compared with the rats ventilated with similar peak inspiratory pressures but with large tidal volumes (43 or 30 mL/ kg) and no PEEP. An additional important finding was that ventilation with a tidal volume of approximately 12 mL/kg without PEEP did not induce edema, but the same tidal volume with a PEEP of 10 cmH2O induced mild edema (Fig. 3). These data demonstrate that ventilator-induced pulmonary edema develops when end-inspiratory lung volume is excessive whether the increase in lung volume results from a change in tidal volume alone or from a combination of PEEP and tidal volume (note that end-inspiratory lung volumes were not measured per se in this study). However, the combination of high tidal volume and no PEEP resulted in more severe edema (29).

Figure 3 Ventilator-induced pulmonary edema in a rat model. Pulmonary edema, measured as lung weight/body weight, increased within

Pathogenesis of ventilator-induced lung injury

187

one hour of high tidal volume ventilation without positive endexpiratory pressure (PEEP). A tidal volume of 43 mL/kg [peak inspiratory pressure (PIP) 40 cmH2O] induced alveolar flooding, resulting in mortality in less than one hour. A tidal volume of 30 mL/kg and no PEEP (PIP 30 cmH2O) induced significantly less edema. Ventilation with 12 mL/kg did not increase lung weight compared with unventilated controls. When a PEEP of 10 cmH2O was applied and tidal volume was reduced such that peak inspiratory pressure was matched to 40 cmH2O (tidal volume 15 ml/kg), lung water was significantly lower than with zero PEEP ventilation. When PIP was matched to 30 cmH2O with a PEEP of 10 cmH2O (tidal volume 12 mL/kg), lung water was not significantly different compared with a tidal volume of 30 mL/kg and no PEEP, but was significantly higher compared with the same tidal volume without PEEP. (Adapted from Ref. 29.) Subsequent studies by others (30–32) have confirmed the findings of Webb and Tierney (29) and have found that the change in lung volume rather than airway pressure that may be the critical factor in lung edema formation. In one such study, rats were ventilated with a similar high tidal volume (40 mL/kg), but with either positive pressure or negative pressure ventilation. A comparison group was ventilated with the same peak inspiratory pressure as the group ventilated with the positive pressure, but with approximately half the tidal volume. This was achieved by applying rubber bands to the abdomen and chest. High tidal volumes resulting from either positive or negative pressure ventilation induced severe edema and similar changes in endothelial permeability to albumin. High-pressure, low-volume ventilation did not induce edema or any change in albumin permeability (32). In another study, isolated perfused dog lungs injured by acid instillation were ventilated with a constant tidal volume of 15 mL/kg and increasing levels of PEEP. Although oxygenation initially increased with higher PEEP, the severity of lung edema

Acute respiratory distress syndrome

188

increased with lung volume. This effect on edema formation was modest when PEEP was raised from 0 to 10 cmH2O, but when PEEP was increased to 15 cmH2O, lung water increased over 400% (33). Others have found that ventilation with increasing PEEP, but a constant endinspiratory volume, reduces the rate of edema formation but affects endothelial protein permeability to a smaller degree (32). Therefore, increasing end-inspiratory lung volume increases edema formation and endothelial permeability. Adding PEEP attenuates edema formation, but part of this effect is explained by the changes in hemodynamics. When tidal volume, inspiratory time, and especially PEEP are increased in vivo, intrathoracic and mean alveolar pressures increase. In general, this has negative consequences on cardiac output and pulmonary blood flow in healthy animals. This partly explains the protective effect of PEEP on edema formation in in vivo models (31, 32). For example, Dreyfuss and colleagues (31) have shown that rats submitted to high peak airway pressure ventilation with 10 cmH2O PEEP had more severe edema when the hemodynamic alterations produced by PEEP were corrected with dopamine. Taken together with other studies, the reason why PEEP reduces the amount of edema may be a combination of such hemodynamic alterations, preservation of lung volume, and preservation of surfactant properties (see Sec. V). In contrast, excessive lung volume, even when the result of high PEEP, can promote edema formation by increasing transmural vascular pressures for extra-alveolar vessels (34, 35) and alveolar vessels (36, 37). This partly explains the lack of a protective effect of PEEP on edema formation when the levels of PEEP are above a threshold point. In previously injured lungs, the effects of high tidal volume ventilation on edema formation are more pronounced. Ventilation with high tidal volumes that do not induce injury in normal lungs can result in more severe edema formation during acute lung injury compared with lower tidal volumes (32, 38–41). The increase in edema is largely due to differences in endothelial permeability and injury as assessed by histological and ultra-structural analysis (32, 38, 41). Increased Endothelial Permeability with High Tidal Volume Ventilation Ventilator-induced pulmonary edema is characterized by a high protein concentration indicating increased microvascular permeability. Dreyfuss and colleagues (30) demonstrated this increase in microvascular permeability using a rat model. They found that endothelial permeability to radiolabeled albumin increased as soon as 5 minutes after initiating ventilation with a peak airway pressure of 45 cmH2O (tidal volume 40 mL/kg). This has also been demonstrated in larger animals, including lambs and dogs; however, a longer duration of ventilation was required (28, 42, 43). Although filtration pressure, or the transmural vascular pressure, also increases with high-volume ventilation, it is the increase in permeability resulting in part from increased wall tension that is the most important determinant of edema formation in intact animals (28, 44, 45). As discussed above, changes in cardiac output and blood flow will also influence the rate of edema formation. There are at least two potential mechanisms for the rapid increase in capillary endothelial permeability that occurs during high-volume mechanical ventilation. First, high lung volume increases capillary wall strain (Fig. 2). This potentially results in the

Pathogenesis of ventilator-induced lung injury

189

opening of stretch pores in the endothelium and eventually breaks in endothelial cells and the basement membrane termed capillary stress failure (36, 43, 46). This mechanism is consistent with experimental studies demonstrating edema formation resulting from lung overinflation achieved by either high tidal volume or by the combination of tidal volume and PEEP (31, 33); however, the extent to which stretch pore formation contributes to edema formation has been questioned (47). A second potential mechanism is strain-induced changes in the cytoskeleton and adhesion junctions of endothelial cells that result in a more permeable endothelial barrier. Such changes in endothelial barrier integrity have been well described in vascular endothelial cells in response to shear stress (36). This same process may occur in response to the mechanical forces associated with positive pressure ventilation. Evidence supporting this hypothesis has been reported by Parker and colleagues. These investigators found that the increase in endothelial permeability associated with highvolume ventilation could be prevented by increasing intracellular cAMP or by inhibiting calcium/calmodulin-dependent myosin light chain kinase. Administration of β-adrenergic agonists (48), or inhibitors of phosphodiesterase (49), which increase intracellular cAMP, was found to prevent edema formation in isolated lungs exposed to high-volume ventilation. Gadolinium (an inhibitor of strain-gated calcium channels) also prevented edema formation in this model (50). In addition, the inhibition of tyrosine kinase with either genistein or phenylarsine oxide or the inhibition of myosin light chain kinase (MLCK) and protein kinase C (PKC) activity also attenuated the increase in endothelial permeability in rat lungs (49, 51). Taken together, these data indicate that the calcium/calmodulin-MLCK pathway augments edema formation during high tidal volume ventilation (Fig. 4). This effect is attenuated by increasing intracellular cAMP, inhibiting MLCK and protein kinase C, or the nonspecific inhibition of tyrosine kinases. Accordingly, the increased endothelial permeability in VILI may be regulated by a variety of signaling pathways and not only the result of breaks in endothelial cells and basement membrane. Epithelial Permeability and Injury As with endothelial permeability, alveolar epithelial permeability increases with increasing lung volume. For example, increasing functional residual capacity by the application of PEEP during mechanical ventilation results in increased clearance of inhaled 99mTc-DPTA (MW 393 daltons) in excess of what would be predicted from a change in surface area alone (52, 53). Alveolar epithelial permeability to albumin also increases with increasing lung volume (46, 54). Egan (54) distended isolated rabbit and sheep lung lobes with fluid to a pressure of 40 cmH2O and found that the equivalent pore radius increased from approximately 1 to 5 nm. This translated into mildly increased epithelial permeability; however, large irreversible leaks developed in some lobes. It is important to note that the distending pressure used in these studies resulted in a lung volume in excess of total lung capacity. When entire lungs rather than isolated lobes were tested, the effect was less pronounced in that permeability to small molecules increased, but permeability to albumin was not altered (46). Similar findings were described in a report by Kim and Crandall (55) in which alveolar epithelial permeability in bullfrog lungs was not altered by ventilation within a physiological range. Therefore, lung

Acute respiratory distress syndrome

190

distention near or exceeding the limits of normal physiology results in increased epithelial permeability in uninjured lungs. Ventilation of injured lungs with tidal volumes within a physiological range can also exacerbate epithelial permeability changes. In a rat model of acid-induced acute lung injury, ventilation with 6 mL/kg resulted in less alveolar flooding and less alveolar epithelial injury as measured by plasma levels of a type I cell-specific marker of injury (RTI40) compared with 12 mL/kg and a similar level of PEEP. This finding correlated with histological and ultrastructural differences in airspace edema and epithelial cell injury. When tidal volume was further reduced to 3 mL/kg, epithelial injury and airspace edema improved even more. Reducing PEEP during ventilation with a tidal volume of 12 mL/kg such that end-inspiratory lung volume and mean airway pressures were similar to the 6 mL/kg group did not prevent epithelial injury or edema (41). Similar findings have also been reported following surfactant depletion. In this model, tidal volume reduction prevented airspace edema formation and preserved oxygenation, sug-

Figure 4 Effect of mechanical stresses on endothelial permeability in VILI. Mechanical forces associated with high tidal volume ventilation and shear stress activate strain-gated cation channels, resulting in increased

Pathogenesis of ventilator-induced lung injury

191

intracellular calcium concentration. This activates calcium/calmodulin (Ca/CaM), which phosphorylates myosin light chain kinase (MLCK), resulting in stress fiber formation, cell contraction, and increased paracellular permeability to fluid and leukocytes. Stress-induced plasma membrane disruption (PMD) also increases intracellular calcium and increases expression of the transcription factor cfos. Shear stress also induces phosphorylation and activation of focal adhesion complex tyrosine kinases (TK), including src, and the small GTPase Rho. Rho acts indirectly to inhibit myosin light chain phosphatase, thereby increasing MLCK activity. TK and src activate protein kinase C (PKC) directly and through phospholipase C-γ (PLCγ). PKC induces stress fiber formation and increased permeability by a MLCKindependent mechanism. Tumor necrosis factor receptor (TNFR) also activates PKC, in response to ligand binding, via phosphorylation of the 27 kDa heat shock protein (HSP-27). Increased endothelial permeability is inhibited by cAMP. Intracellular cAMP increases in response to ligand binding to the β2-adrenergic receptor (β2-AR) via G protein-mediated activation of adenylate cyclase (AC). Mechanical stress is also transduced into chemical signals in the cell by extracellular matrix-bound integrins. Integrins bind the RGD amino acid sequence abundant in the extracellular

Acute respiratory distress syndrome

192

matrix. Bound integrins exposed to mechanical stress initiate a variety of signals from the focal adhesion complex. These signals include activation of AC, which increases cAMP, thereby activating protein kinase A (PKA). PKA affects gene transcription through CREB in the nucleus. Focal adhesion kinase (FAK) phosphorylates IKK, which in turn phosphorylates IκB to activate NFκB. Activation of src at the focal adhesion complex also results in the phosphorylation and activation of the mitogen-activated protein kinases (MAPK) ERK and JNK. These act in part to increase expression of the transcription factors Erg-1 and c-fos to modulate gene transcription. gesting preserved epithelial barrier function. Interestingly, when animals were ventilated with high-frequency oscillatory ventilation, edema and histological injury were further reduced (56, 57). Airspace Edema Clearance in Ventilator-Induced Lung Injury The presence of edema fluid in the airspaces is both an effect of lung injury and a potential mechanism by which VILI is amplified. Edema fluid fills alveoli and promotes airspace collapse by inactivating surfactant and filling airways. This loss of lung volume leads to heterogeneity of the lung, resulting in even greater overdistention of the remaining lung units (19). Therefore, if the active sodium transport-dependent clearance of edema fluid from the distal airspaces is reduced, a vicious cycle of airspace edema leading to greater lung overdistention and shear stress will ensue. For example, flooding distal lung units of rats with saline was found to act synergistically with high tidal volume ventilation to increase endothelial permeability to albumin (58). In this study, the authors also found that as respiratory system compliance decreased, permeability to albumin increased, suggesting that as edema worsened, a smaller lung volume was ventilated and greater injury resulted (58). The mechanical strain-induced reduction in energy-dependent sodium transport described by Lecuona and colleagues (59) may further impair airspace edema clearance promoting VILI. Using alveolar type II cells isolated from rats ventilated with a tidal volume of either 30 or 40 mL/kg, these authors reported that sodium-potassium ATPase activity was reduced, but mRNA for the α-1 subunit of this transporter was not reduced

Pathogenesis of ventilator-induced lung injury

193

compared with rats ventilated with lower tidal volume (10 mL/ kg). In another study, airspace edema clearance in lungs isolated from rats ventilated for 40 minutes with a tidal volume of 40 mL/kg was reduced by approximately 50%. Instilling the airspaces of the isolated lungs with the β-adrenergic agonists restored the rate of airspace edema clearance by increasing the activity and quantity of sodium-potassium ATPase in the basolateral membrane. This effect was blocked by disrupting microtubule assembly with colchicine, suggesting that it is the translocation of sodium-potassium ATPase from intracellular pools to the plasma membrane that accounts for much of the effect (60). In a rat model of acute lung injury, tidal volume reduction from 12 to 3 mL/kg resulted in greater preservation of airspace fluid transport. This effect persisted when either endinspiratory lung volume or end-expiratory lung volume was constant (41). High Tidal Volume Ventilation and Extracellular Matrix Constituents In Vivo High-volume ventilation also affects structural protein expression in experimental models of VILI. In a study by Berg and colleagues, open-chested rabbits were ventilated for 4 hours with one lung at a PEEP of 9 cmH2O and the other at a PEEP of 2 cmH2O. Compared with rabbits ventilated with a PEEP of 2 cmH2O on both lungs, the higher lung volume ventilation resulted in increased expression of mRNA for type III and type IV procollagen, fibronectin, basic fibroblast growth factor, and transforming growth factorβ1. Expression of mRNA for type I procollagen and vascular endothelial growth factor was not different between the groups (61). Interestingly, elevated edema fluid levels of procollagen III predict poor outcome in patients with ARDS (62). Other investigators have found that high airway pressure ventilation increases expression of type IV and type I procollagen in isolated rat lungs (63). ECM proteoglycans and expression of enzymes active in matrix remodeling, including matrix metalloproteinases (MMPs) and the extracellular matrix metalloproteinase inducer (EMMPRIN), are also increased following high tidal volume ventilation in rats (64, 65). Pretreatment with a nonspecific inhibitor of MMPs reduced the severity of lung injury following high-volume ventilation (66). These data suggest that alveolar overdistention may initiate an ECM remodeling process that potentially exacerbates lung injury or impairs healing. C. Effects of Mechanical Strain In Vitro An improved understanding of the effects of mechanical ventilation at the cellular level has come from in vitro studies using cell stretch devices, applied to a variety of cells including alveolar epithelial cells, fibroblasts, airway epithelial and smooth muscle cells, pulmonary vascular endothelial and smooth muscle cells. This section will review the effects of stretch on the cytoskeleton, plasma membrane, and ECM constituent expression.

Acute respiratory distress syndrome

194

Disruption of the Cytoskeleton and Plasma Membrane In Vitro Studies of alveolar epithelial type II cells grown on flexible membranes have provided insight into the mechanical properties of these cells. In one study, increasing the duration, amplitude, or frequency of cyclic strain increased cellular uptake of ethidium homodimer-1 (ED-1), a marker of plasma membrane injury and cell death. Most cell injury occurred within 5 minutes of initiating the cyclic strain. If small amplitude deformation was superimposed on basal tonic strain, there was less membrane disruption and cell death compared with large amplitude stain to same peak level. The rate of cellular deformation during a single strain did not affect ED-1 uptake (67). These finding are not surprising if one assumes that mechanical disruption of the cytoskeleton and plasma membrane is the etiology of the injury in this model and that the cytoskeletal structure of cultured cells behaves as most other solids. In a more recent study, plasma membrane disruption (PMD) induced by mechanical strain in vitro was primarily dependent on the rate of plasma membrane trafficking to the cell surface. Inhibition of cytoskeletal remodeling had little impact on the cell injury as measured by PI uptake. Although these data do not exclude strain-induced signaling through the cytoskeleton as an important mechanism of VILI, they support the hypothesis that PMD and lipid trafficking may be a major mechanism (68, 69). Effect of Mechanical Strain on Cell Adhesion In Vitro Cyclic strain also affects cell-cell and cell-ECM attachment. Cavanaugh and colleagues (70) showed that cyclic stretch of alveolar epithelial cells corresponding to lung inflation to total lung capacity (37% increase in surface area) resulted in decreased intercellular attachment and decreased expression and of occludin, a major constituent of tight junctions. Lesser degrees of cyclic stretch did not have these effects. Interestingly, cyclic stretch to near-total lung capacity also resulted in a decrease in intracellular ATP; pharmacological inhibition of glycolytic metabolism also reduced cell-cell attachment and resulted in an abnormal cytoplasmic accumulation of occludin. Although these investigators did not measure permeability, the loss of cell-cell adhesion and disruption of tight junctions potentially contributes to increased alveolar epithelial permeability, as has been shown in endothelial cells (71).

V. Injury from Low Lung Volume Ventilation Although the forces described above occur to some extent at all volumes, additional sources of mechanical force at low lung volume may be impor-tant in VILI. The main additional force is that resulting from the repeated opening and closing of small airways, alveolar ducts, and possibly individual alveoli during tidal ventilation (Fig. 2). This cyclic recruitment of lung units may be partially or completely prevented by the application of PEEP, which likely accounts for some of the protective effect of PEEP in experimental models.

Pathogenesis of ventilator-induced lung injury

195

Airway closure has been observed to occur by one of two mechanisms: development of a fluid meniscus, which occludes an patent bronchiole, or “compliant collapse” of small airways in which opposing walls buckle and become apposed owing to the “adhesive” forces of the airway lining fluid and peribronchiolar pressure. Reopening of the airways requires sufficient force either to rupture the meniscus or to peel the opposing wall and push the fluid interface toward the alveolus. During this process, shear stress sufficient to cause epithelial disruption may be generated. Surface tension forces of airway lining fluid in collapsed or occluded airway create a capillary pressure that must be exceeded for airway opening to occur. In conditions of increased surface tension, the tendency for airways to collapse significantly increases, as do the pressures required to reopen and keep open the airways. Using surfactant-depleted isolated rat lungs, Muscedere and colleagues (72) reported that ventilation with low tidal volume of 7 mL/kg and low or no PEEP resulted in poorer lung compliance and more severe lung injury versus ventilation with the same tidal volume and a PEEP level higher than the lower inflection point on the respiratory system pressure volume curve. Histological injury such as epithelial necrosis, sloughing, and development of hyaline membranes was observed in alveolar ducts as well as respiratory and membranous bronchioles. This may explain, in part, the protective effects of ventilatory strategies that maintain lung volume with PEEP. Repetitive airway opening and collapse may also explain the distribution of dependent lesions, especially in larger species. As pleural pressure is higher in the dependent than in the nondependent lung zones, the airspaces in the dependent zone would be more likely to collapse. In patients with ARDS, the regional cyclic recruitment of lung units can be demonstrated with computed tomographic (CT) imaging. When PEEP is applied at increasing levels, the proportion of lung undergoing cyclic recruitment decreases. This is associated with improved oxygenation and improved quasi-static lung compliance (74). Gattinoni and colleagues (73– 75) demonstrated that the volume of the injured lung exposed to a tidal breath was markedly reduced when low levels of PEEP were used. When PEEP was increased, the proportion of the tidal volume delivered to the nonaerated lung regions gradually increased, becoming more evenly distributed (74). In contrast, Martynowicz and colleagues (14) documented that regional tidal expansion of dependent regions was significantly reduced after oleic acid injury and was restored by application of PEEP. However, oleic acid injury did not lead to the collapse of dependent lung units at FRC, suggesting an alternative mechanism for topographic variability in regional impedance and lung expansion after injury in a model characterized by marked alveolar flooding (14). A. Clinical Studies of Protective Ventilation In prospective studies, ventilation strategies that incorporate relatively high levels of PEEP and low tidal volumes reduce mortality from ARDS and acute lung injury (6, 22, 76). As described above, PEEP could be responsible for minimizing VILI by reducing shear forces created by opening and collapse of alveoli. The best method to select the ideal level of PEEP for a given patient with ARDS is yet to be determined. Although some studies have used the pressure volume curve of the

Acute respiratory distress syndrome

196

respiratory system to set PEEP above the lower inflection point (Pinf), others have used arbitrary scales of PEEP. Both strategies, when combined with low tidal volume ventilation, reduce mortality from ARDS. Amato and colleagues (22) used a lung protective strategy based on (1) a PEEP level greater than the inflection point (14.7±3.9 cmH2O at baseline), (2) recruitment maneuver at start of study, and (3) peak airway pressure<40 cmH2. In this study of 53 patients, there was a reduction in mortality from 71 to 38% using this strategy. Whether the marked decrease in mortality in this study was due to the lung protection due to an excessive mortality in the control group is uncertain. In the ARDS Network study PEEP was set according to a predetermined scale and not the Pinf. A predetermined scale was used because the relationship between the shape of the pressure-volume curve and events at the alveolar level is affected by numerous factors and is the subject of ongoing research and debate (77–80). Interestingly, in a subsequent study by the ARDS Network combining low tidal volume with a scale incorporating even higher PEEP levels, no additional mortality benefit was observed (81). Currently there is renewed interest in high-frequency oscillatory ventilation (HFOV) combined with strategies of lung volume maintenance for adults with ARDS. This strategy would potentially prevent excessive end-inspiratory lung volume and maintain sufficient end-expiratory lung volume to a greater degree than conventional ventilation. Preliminary data suggest that this method of ventilation is safe in adults (82). Ongoing studies are comparing HFOV combined with lung volume maintenance to low tidal volume ventilation in children and adults. Previous studies of HFOV have not always included a protocol for lung volume maintenance. B. Experimental Studies of Low Lung Volume Ventilation The strongest evidence that low end-expiratory lung volume ventilation promotes VALI comes from experimental studies. Following lung injury, low-volume ventilation inactivates surfactant and promotes edema formation in vivo and in isolated lungs (83, 84). In surfactant-depleted isolated rat lungs, ventilation with low tidal volume of 7 mL/kg and a PEEP level below the lower inflection point of the lung pressure volume curve resulted in poorer lung compliance and more severe lung injury. The greatest increase in injury was found to be to the epithelium of membranous bronchioles (72). Although the injured lung is predisposed to cyclic recruitment of lung units with tidal ventilation due to the inactivation of surfactant and the resultant increase in surface tension, D’Angelo and colleagues (85) found that zero PEEP ventilation can induce small airway injury in normal rabbit lungs. Rabbits were ventilated open-chested for 4 hours with a tidal volume of approximately 10 mL/kg, with or without a PEEP level of 2 cmH2O. Ventilation without PEEP resulted in injury to small (<1 mm diameter) airways as assessed by histological analysis. Airway resistance and elastance increased during zero PEEP ventilation. When PEEP was restored, elastance but not airway resistance returned to normal values. These data indicate that ventilation of normal lungs with an end-expiratory lung volume less than FRC induces small airway injury presumably due to the repetitive opening and closing of these airways. In animal models of surfactant-deficient lungs, the preservation of lung volume with PEEP and periodic recruitment maneuvers improves oxygenation, compliance, and

Pathogenesis of ventilator-induced lung injury

197

reduces histological lung injury (86). Combining high-frequency oscillatory ventilation with a strategy to recruit and maintain lung volume has also been found to prevent lung injury in premature baboons and in surfactant-depleted rabbits (87–89). In acid-injured rats, ventilation with low tidal volume and no PEEP resulted in 100% mortality by 2 hours, while rats ventilated with a higher tidal volume or with PEEP survived beyond 4 hours (38).

VI. Mechanotransduction in VILI In the context of VILI, mechanotransduction refers to the mechanism(s) by which mechanical stimuli are converted into chemical signals within the cell. Although our understanding of the precise mechanisms and importance of mechanotransduction in VILI is far from complete, there is increasing evidence that these signaling mechanisms play an important role. Examples of mechanotransduction potentially important in VILI include impaired endothelial barrier function in response to shear stress, integrinmediated signal transduction, mechanosensitive ion channels, and plasma membrane disruption-mediated signaling. A. Endothelial Cell Contraction The vascular endothelium forms the primary barrier between the vessel lumen and the interstitial space of the lung. As discussed earlier, increased permeability of the endothelium to fluid and proteins is a hallmark of VILI and ARDS. Although VILI can induce breaks in the endothelial barrier that contribute to the observed increase in permeability, modifications to the endothelial cytoskeleton in response to physical or inflammatory stimuli also increase permeability. Endothelial cell contraction and stress fiber formation is associated with increased paracellular movement of fluid and leukocytes from the bloodstream into the interstitial compartment and airspaces. In addition, ligand binding and endothelial cell activation result in the disruption of cell-cell adherens and tight junctions (90–92). A well-described mechanism of increased permeability in endothelial cells is myosin light chain kinase (MLCK)-dependent cell contraction. Phosphorylation of calcium/calmodulin-dependent MLCK results in contraction of the actin cytoskeleton, stress fiber formation, and the formation of paracellular gaps (93, 94). MLCK phosphorylation is regulated by intracellular calcium via calcium/calmodulin as well as by tyrosine kinases, including p60src, tyrosine phosphatase activity of the SH2 domain of MLCK, and by the small GTPase Rho, which indirectly inhibits myosin light chain phosphatase (PP1), resulting in increased MLCK phosphorylation (95–97). As discussed above, Parker and colleagues have demonstrated that increased intracellular cAMP or inhibition of the MLCK activation pathway prevents ventilator-induced increases in permeability in isolated lungs (Fig. 4) (48–51). Endothelial cell contraction can also be initiated by MLCK-independent pathways. For example, protein kinase C, TNF-α, LPS, and p38MAPK all induce stress fiber formation and impair endothelial barrier function by mechanisms independent of MLCK (71, 98– 101). Taken together, these data indicate that increased endothelial permeability in VILI

Acute respiratory distress syndrome

198

is regulated by a variety of signaling pathways, including MLCK-dependent cytoskeletal contraction and ligand-dependent disruption of adherens junctions, and is not only the result of breaks in endothelial cells and the basement membrane. B. Integrin-Mediated Signaling Mechanical stresses also influence gene expression through a variety of signaling pathways in both endothelial and epithelial cells. In bovine endothelial cells, shear stress is transduced into increased cAMP via integrin-dependent, G-protein-mediated activation of adenylate cyclase. This increase in cAMP activates PKA, which in turn phosphorylates CREB to mediate gene transcription. This signaling pathway is dependent on integrin binding to RGD, a tripeptide sequence abundant on ECM proteins. Integrin distortion in the absence of binding does not increase intracellular cAMP (102). Shear stress also activates NFκB in endothelial cells by an integrin binding-dependent mechanism (103). Activation and translocation of NFκB to the nucleus is important in signal transduction of various inflammatory stimuli and apoptosis (Fig. 4). Integrin-mediated signaling in response to fluid shear stress also activates the MAPKs extracellular signal-regulated kinase (ERK) and c-Jun NH2-terminal kinase (JNK), in a process dependent on focal adhesion kinase (FAK) activation (104). ERK has been shown to increase expression of the early response transcription factor Egr-1, which transactivates genes involved in inflammation and injury in response to fluid shear stress (105). Integrin binding to ECM also mediates the rapid translation of existing mRNAs and translocation of mRNA and ribosomes to FACs, thereby exerting local control over protein translation at the FAC (106). Although the role of this regulatory mechanism in VILI is not known, cytoskeletal stiffness, measured by magnetic twisting cytometry, increases during static conditions, but decreases in response to cyclic stress in cultured alveolar epithelial cells (107). Using this model, Berrios and colleagues found that stiffness correlated with integrin binding-mediated FAC formation (107). The precise mechanisms of the decrease in stiffness with shear stress are not clear, but changes in protein expression are one potential mechanism. Another potential mechanism for the decrease in cytoskeletal stiffness is the change in phosphorylation state of cytoskeletal scaffolding protein (108). In experimental VILI, high-volume ventilation has been shown to increase endothelial cell tyrosine kinase-mediated phosphorylation of focal adhesion proteins, including FAX, paxillin, JNK, and p38MAPK (109). In fetal lung cells, Liu and colleagues (110) demonstrated that mechanical strain in a three-dimensional culture system induced activation of p60src and its localization to the cytoskeleton. They identified several substrates for src, including the 110 kDa actin filament-associated protein (AFAP) and PLC-gamma. Activation of this pathway affects gene transcription and cell proliferation, but the precise effect on cytoskeletal mechanics and cell adhesion are not known (111). C. Indirect Mechanical Signaling Recent studies have demonstrated that signals initiated by mechanical stimuli can be transmitted between cell types. For example, mechanical stress sensed by airway epithelial cells can be transmitted to unstressed adjacent endothelial cells in the absence

Pathogenesis of ventilator-induced lung injury

199

of any physical connection. In these experiments, cultured bronchial epithelial cells exposed to a tonic pressure for four hours exhibited increased expression of Egr-1, fibronectin, and MMP-9, which degrades type IV collagen. In co-cultured fibroblasts at atmospheric pressure, stress applied to the epithelial cells induced increased expression of type III collagen (112).

VII. Alterations in Surfactant Secretion and Function Surfactant lipid and protein turnover is rapid, and surfactant synthesis and secretion are affected by mechanical factors (113–115). Accordingly, impairment in surfactant metabolism is a potential mechanism of VILI. Data from even the earliest studies of VILI have supported the conclusion that large lung volume ventilation impairs surfactant function, resulting in increased surface tension and reduced lung compliance (29, 116– 118). The reduction in lung compliance and associated lung volume loss resulting from injurious ventilation potentially exacerbates the detrimental forces acting on the lung. In an isolated, nonperfused lung model of injurious ventilation, Veldhuizen and coworkers (119) found that ventilation with a tidal volume of 40 mL/kg and without PEEP for 2 hours resulted in a decrease in surfactant proteins B and C mRNA, but surfactant protein A mRNA was not different from controls. Surfactant proteins are B and C are the major surface active proteins, but surfactant protein A acts in part to prevent the inactivation of other surfactant proteins by plasma proteins (120–122). A subsequent study using a similar model found that the surface activity of large aggregate surfactant from lungs ventilated with an injurious tidal volume was reduced, but the amount of large aggregate surfactant was not changed (123). These authors concluded that the presence of protein-rich edema fluid in the airway combined with reduced synthesis of surfactant proteins B and C contributed to the observed decrease in surface activity. Using models of preterm delivery and acute lung injury, others have found that mechanical ventilation with lung volumes in excess of total lung capacity or without PEEP results in the loss of surfactant function (124–126). In one study, rabbits given Nnitroso-N-methylurethane (NNMU) to induce lung injury were ventilated with a tidal volume of either 5 or 10 mL/kg. Ventilation with the higher tidal volume resulted in more rapid inactivation of large aggregate surfactant compared with the lower tidal volume (126). Altering the level of PEEP from 3.5 to 12 cmH2O with a tidal volume of 5 mL/kg did not alter the rate of surfactant inactivation. These authors concluded that large tidal excursions in lung volume, and therefore large changes in alveolar surface area, were important in surfactant inactivation (i.e., changes in PEEP affected lung volume much less than changes in tidal volume). Others have found that low lung volume also inactivates surfactant (83, 84, 127, 128). For example, in experimental lung injury due to sepsis, the use of PEEP is associated with greater preservation of large aggregate surfactant pools when compared with a similar tidal volume and no PEEP (124). In the injured lung, there are several potential mechanisms for the observed loss of surface active properties. First, because surfactant lipid has little surface activity alone, the loss of surfactant proteins results in a less active surfactant (129). Accordingly, decreased synthesis or secretion of these proteins would result in reduced surfactant function. Second, because plasma proteins bind and inactivate surfactant proteins B and

Acute respiratory distress syndrome

200

C, the presence of protein-rich edema fluid in the airways reduces lung compliance in part by inactivating surfactant (130). Furthermore, airspace edema can act to wash away surfactant from the alveoli (15, 116). Therefore, mechanical ventilation with high tidal volume and frequency or without PEEP can reduce surfactant function directly by affecting surfactant protein synthesis and secretion and indirectly by inducing airspace edema formation.

VIII. Effects of Hypercapnia Another potential mechanism for the protective effect of low tidal volume ventilation is the hypercapnia that may occur with low tidal volume ventilation. Experimental models of ischemia-reperfusion injury have demonstrated reduced tissue injury in association with hypercapnia and acidosis. The presumed mechanisms, possibly mediated by pH changes, include a reduction in reactive oxygen and nitrogen species and an adaptive alteration in oxygen supply and demand kinetics (131, 132). At least one experimental study of VILI demonstrated a similar protective effect of hypercapnia in isolated rabbit lungs (133). Any potential protective effect of hypercapnia in patients has not been clearly demonstrated.

IX. Ventilator-Induced Pulmonary and Systemic Inflammation One of the primary mechanisms of lung injury propagation in models of VALI is increased inflammation as measured by increases in edema fluid and plasma markers, and by an increase in neutrophil infiltration and macrophage activation. This section will review data from clinical, experimental, and in vitro studies supporting the role of increased inflammation in the pathogenesis of VILI. A. Biomarkers of Inflammation in Clinical Studies Elevated levels of proinflammatory mediators have been measured in airspace lavage fluid and in the plasma of patients with ARDS. Ranieri and colleagues (134) measured bronchoalveolar lavage (BAL) and plasma levels of several proinflammatory cytokines in 44 patients with ARDS. At study entry, patients were randomized to receive mechanical ventilation with a conventional strategy (mean tidal volume of 11.1 mL/kg, mean plateau airway pressure 31 cmH2O, and mean PEEP of 6.5 cmH2O), or a reduced tidal volume (7.6 mL/kg) and higher PEEP (14.8 cmH2O) with a mean plateau airway pressure of 24.6 cmH2O. PEEP in the latter group was set above the lower inflection point of the respiratory system pressure volume curve. Baseline measurements of cytokines were made at the time of admission (study entry) and were then measured serially for 2 days. By 36 hours, BAL fluid from patients in the protective ventilation group had significantly fewer polymorphonuclear cells and lower concentrations of TNF-α, IL-1β, IL-6, and IL8. Plasma levels of IL-6 were also significantly lower in the patients receiving protective ventilation (134).

Pathogenesis of ventilator-induced lung injury

201

The NIH ARDS Network study found a similar difference in plasma IL-6 levels at 3 days among patients ventilated with low tidal volume compared with conventional tidal volume (6). In another study of 27 consecutive patients with ARDS, Meduri and colleagues (135) found that persistent elevation of plasma IL-6 or IL-1β was associated with decreased survival although all patients were ventilated with the same strategy. Others have also found that persistent elevation of IL-1β is associated with increased mortality in patients with ARDS (136). Although plasma and BAL levels of cytokines are elevated early in the course of ARDS, the molar ratios of individual cytokines to antiinflammatory soluble receptors and receptor antagonists may be more important in determining the pulmonary and systemic pro-inflammatory effects of these mediators (137). In another study of 39 patients with normal lungs, ventilation for one hour with a high tidal volume (15 mL/kg) and zero PEEP did not affect plasma levels of either IL-6, TNF-α, IL-1 receptor antagonist, or IL-10 (138). In summary, these data indicate that in ARDS patients, higher tidal volume ventilation is associated with poor outcome and higher BAL and plasma levels of pro-inflammatory mediators, especially IL-6. Higher plasma levels of IL-6 are associated with poor outcome independent of the ventilation strategy used. The increase in plasma IL-6 observed in ARDS is not seen in patients with normal lungs ventilated with high tidal volume for short periods of time. B. Mediators of Inflammation in Experimental Studies Under experimental conditions, high tidal volume, low PEEP ventilation induces increased release of pro-inflammatory cytokines into the airspaces and bloodstream, increased neutrophil infiltration into the lung, and activation of alveolar macrophages (139). Although there is some debate about the potential importance of pro-inflammatory cytokines in the development of VILI (140), there is considerable evidence supporting a role for the release of inflammatory mediators in VILI. Tremblay and colleagues (141) found that ventilation of isolated, nonperfused rat lungs with a tidal volume of 40 mL/kg without PEEP for 2 hours resulted in large increases in lavage concentrations of TNF-α, IL-1β, IL-6, and MIP-2 (Fig. 5). Reducing tidal volume to 15 or 7 mL/kg reduced the lavage concentrations of these mediators even if end-inspiratory lung volume was similar. The increase in these cytokines was greater if rats were pretreated with LPS, but the differences among the groups persisted. Northern blot analysis of whole lung homogenates revealed increased expression of c-fos mRNA, a transcription factor important in the early stress response, with both high and moderate tidal volume, zero PEEP ventilation. The increase in c-fos was greater in saline-treated control rats than in LPS-pretreated rats (141). Early response genes, such as c-fos, as well as NFκB contain cis-acting shear stress response elements in their promo tor regions (141, 142). Chiumello and colleagues (38) demonstrated that injurious ventilatory strategies increased release of proinflammatory cytokines [TNF-α and macrophage inflammatory protein-2 (MIP-2)] in the lung as well as systemic circulation in an acid aspiration model in rats. Others have found that in a rat model of cecal ligation-induced sepsis, injurious ventilation with a tidal volume of 20 mL/kg, without PEEP, resulted in higher BAL levels of TNF-α, IL-1β, and IL-6 compared with unventilated controls and rats ventilated with a

Acute respiratory distress syndrome

202

tidal volume of 10 mL/kg and PEEP level of 3 cmH2O. Whole lung mRNA levels for each of these cytokines were also elevated after one hour of injurious ventilation in this study (143). Further evidence of the importance of these inflammatory mediators in the development of VILI comes from experimental studies of the effects of anti-TNF antibody and IL-1 receptor antagonist on lung injury following surfactant depletion. Administration of anti-TNF antibody to surfactant-depleted rabbits prior to the initiation of conventional mechanical ventilation resulted in less severe histological lung injury and preserved oxygenation

Figure 5 Effect of ventilation strategy on absolute lung lavage cytokine concentrations in isolated, nonperfused rat lungs. Two hours of high tidal volume ventilation in association with zero PEEP increased lavage levels of each of these cytokines. The lowest levels were found in the unventilated controls (C). MVHP: moderate volume (15 mL/kg), high PEEP (10 cmH2O). MVZP: moderate tidal volume, zero PEEP. HVZP: high tidal volume (40 mL/kg), zero PEEP. (From Ref. 141.)

Pathogenesis of ventilator-induced lung injury

203

(144). In a similar model, IL-1 receptor antagonist reduced endothelial albumin permeability and neutrophil infiltration at 8 hours (145). Moreover, activation of the heat shock stress response attenuated the increases in TNF-α, IL-1β, and MIP-2 associated with injurious mechanical ventilation of isolated non-perfused rat lungs (146). Activation of the stress response also preserved large aggregate surfactant in this model. These data support the hypothesis that heat shock proteins (HSP) inactivate or prevent the release of pro-inflammatory cytokines in VILI. For example, HSP72 forms a complex with TNF-α, preventing its release from the macrophage (147). Previous studies have demonstrated that induction of the heat shock stress response reduces mortality in experimental acute lung injury (148). Recent data suggest that the signaling pathway by which mechanical distention mediates the release of inflammatory cytokines involves translocation of NFκB to the nucleus. However, initiation of NFκB activation with VILI is independent of TRL-4/LPS receptor and is inhibited by corticosteroids (149). These data raise the possibility that pharmacological inhibition of VILI-mediated NFκB activation could be achieved without the complete inhibition of the potentially adaptive innate immune response. Pro-inflammatory signals can also be transmitted across the ECM from alveolar epithelial cells to endothelial cells. Kuebler and colleagues (150) found that intra-alveolar TNF-α increased intracellular calcium in alveolar epithelial cells and activated phospholipase A2. This in turn induced an increase in intracellular calcium in adjacent endothelial cells that was not dependent on gap junctions. The increase in intracellular calcium in the endothelial cells resulted in increased expression of P-selectin, a neutrophil adhesion molecule. Therefore, intra-alveolar TNF-α induces a proinflammatory response in adjacent endothelial cells in the absence of direct TNFendothelial interaction. C. Strain-Induced Inflammatory Response In Vitro Mechanical stimulation of epithelial cells, macrophages, endothelial cells, fibroblasts, and smooth muscle cells induces a change in protein phosphorylation and alterations in cytoskeleton proteins. In VILI it is the alveolar epithelial cells and lung macrophages that are perhaps most responsible for the transcription and secretion of inflammatory mediators. Vlahakis and colleagues (151) found that in cultured A549 cells, mRNA for IL-8 increased fourfold after 4 hours of cyclic strain sufficient to change the cell surface area by 30%. Continued strain for up to 48 hours resulted in a nearly 50% increase in IL8 secretion compared with nonstrained controls. A lesser degree of strain did not affect IL-8 secretion. This finding was confirmed by Quinn and colleagues (152), who also found that the increase in IL-8 secretion was associated with activation of the JNK family MAPKs. Cyclic strain of 15% for 2 hours induced a 237% increase in phosphorylation of JNK in A549 cells. Phosphorylation of p38MAPK increased by 468%, but phosphorylation of ERK 1/2 was not changed. IL-1β signaling activates ERK, JNK, and p38MAPK by different mechanisms. MacGillivray and colleagues have found that in fibroblasts, the IL-1β-dependent ERK activation depends on the localization of IL-1 receptor associated kinase (IRAK) to the FAC (153). FACs are comprised of cell surface integrins that connect the cytoskeleton to the extracellular matrix. In addition, FACs contain a variety of other signaling proteins,

Acute respiratory distress syndrome

204

including FAK and src tyrosine kinases as discussed above. Interestingly, FACs also colocalize with the IL-1 receptor-1 (153). To identify the source of inflammatory cytokines in VILI, Pugin and colleagues (154) cultured human alveolar macrophages on flexible silastic membranes and exposed the cells to cyclic stretch for up to 32 hours. These authors found that cyclic strain increased secretion of IL-8 and MMP-9 (gelatinase b—a type IV collagenase), but not TNF-α or IL-6. When the macrophages were pretreated with LPS, TNF-α and IL-6 secretion increased to greater extent in strained cells compared with static cultures. Mechanical strain also activated NFκB in macrophages after 30 minutes. In another study, a variety of cell types including macrophages, A549 cells, a bronchial epithelial cell line, two endothelial cell lines, and primary lung fibroblasts were exposed to the same cyclic strain. Of these cell types, only macrophages and A549 cells secreted IL-8 in response to mechanical distention. The relative amount of IL-8 secreted from macrophages was much greater than the amount secreted from A549 cells. In the absence of LPS stimulation, cytokines were not secreted in appreciable amounts from the other cell types (155). Importantly, IL-8 is present in high levels in the edema fluid of ventilated patients with ARDS (156, 157). Furthermore, the activation of NFκB in response to intratracheal instillation of immune complexes is dependent upon the presence of alveolar macrophages (158). Taken together, these data implicate the alveolar macrophage as the initial stretch-responsive cell in the initiation of the inflammatory response observed in VILI. This does not negate a possible role for other cell types in the propagation of early pro-inflammatory signaling in VILI. For example, Grembowicz and colleagues (159) reported that sublethal plasma membrane disruption results in increased expression of cfos, a transcription factor important in cytokine expression and activation of NFκB in vascular endothelial cells and smooth muscle cells. NFκB activation via IKK upregulates the expression of a variety of proinflammatory mediators, including IL-6, IL-8, IL-1β, and TNF-α (158, 160, 161). Tracheal epithelial cells exposed to either magnetic twisting cytometry or to static compressive stress were found to upregulate expression of Egr-1, another early response gene that encodes a transcription factor with binding sites in the promotor regions of genes like TNF-α, PDGF, TGF-β, and PAI-1 (162–164).

X. Biotrauma: Mechanical Ventilation and Multiple Organ Dysfunction Syndrome An interesting observation in most clinical studies of ARDS is that most patients die from MODS rather than respiratory failure. As confirmed in the clinical studies discussed above, injurious mechanical ventilation results in an increase in pro-inflammatory cytokines in the systemic circulation as well as in the airspaces of the lung (6, 38, 165, 166). A. Can Lung Injury/Inflammation Lead to Systemic Inflammation? The lungs are uniquely poised anatomically to affect distant organs. The pulmonary vasculature not only receives the entire cardiac output, but also harbors a large reservoir

Pathogenesis of ventilator-induced lung injury

205

of macrophages and marginated neutrophils. Thus, significant potential exists for the lungs to interact with, and contribute to, the circulating pool of inflammatory cells. In addition, inflammation in the lung caused by mechanical ventilation or injury to the alveolar-capillary interface may cause release or allow efflux of inflammatory mediators from the alveolar space into the circulation. Several investigators have also shown that increased permeability of the alveolar/capillary interface as a result of lung injury leads to the release of mediators into circulation. Von Bethmann and colleagues (166) reported that in an isolated perfused murine lung model, ventilation with higher transpulmonary pressure (25 cmH2O) compared with a normal pressure (10 cmH2O) leads to a significant increase in concentration of both TNF-α and IL-6 in the perfusate. In patients with ARDS, concentrations of TNF-α, IL-1β and IL-6 were higher in the arterial blood (obtained via a wedged pulmonary artery catheter) as compared with mixed venous blood, suggesting that the lungs in these patients were contributing cytokines to the systemic circulation (167). Another mechanism whereby mechanical ventilation may contribute to the development of a systemic inflammatory response is by promoting bacterial translocation from the airspaces into the circulation—analogous to the gut bacterial translocation hypothesis of multiple organ failure (168). Two recent studies evaluated the influence of mechanical ventilation strategy on the translocation of bacteria from the lung into the bloodstream in dogs (169) and rats (170). After intratracheal instillation of bacteria, these animals were ventilated with a high transpulmonary pressure (~30 cmH2O) and minimal (0–3 cmH2O) or 10 cmH2O PEEP. Bacteremia seldom occurred in control animals ventilated with low airway pressure, whereas it was found in nearly all animals ventilated with high tidal volume and a low PEEP. In contrast, ventilation with the same transpulmonary pressure but with 10 cmH2O PEEP resulted in rates of bacteremia as low as in controls. B. Is There Evidence of MODS Secondary to Mechanical Ventilation? The NIH ARDS Network has recently reported the results of a randomized, controlled multicenter study in 861 patients comparing a tidal volume of 12 mL/kg (predicted by body weight) with 6 mL/kg (6). The main finding was a 9% reduction in absolute mortality; plasma levels of IL-6 in the 6 mL/kg tidal volume group were also significantly lower than in the conventional tidal volume group. This was associated with a greater number of organ failure-free days, although this outcome variable may not be independent of mortality. However, it appears plausible that VILI in humans may affect the outcome of ARDS by affecting the systemic inflammatory response leading to MODS. It is important to emphasize that MODS is a complex syndrome, often precipitated and intensified by a series of events rather than a single event. A likely scenario is that there is an ongoing inflammatory response as a result of the persistence of the factors that either initiated or exacerbated the response and/or failure of intrinsic regulatory mechanisms. In an experimental study of acid aspiration, 8 hours of mechanical ventilation

Acute respiratory distress syndrome

206

Figure 6 Potential mechanisms by which mechanical ventilation may contribute to multi-organ system dysfunction (MODS). (Adapted from Ref. 174.) with an injurious strategy led to an increase in pulmonary and systemic production of cytokines, as well as to elevations of biochemical markers indicating organ dysfunction (e.g., creatinine). Furthermore, the injurious ventilatory strategy led to epithelial cell apoptosis in the kidney and small intestine; circulating factors such as soluble FAS ligand may be involved in this process in this model (171). Finally, mechanical ventilation may also affect distal organ function via effects on cardiac output, as well as on the levels of oxygenation and the distribution of blood to the various organ systems (e.g., mesenteric, renal, and hepatic perfusion). For example, although fluid resuscitation of rats ventilated with PEEP returned cardiac output to normal values, mesenteric blood flow remained significantly reduced (172). An increase in distal ileal permeability has also been reported in rats ventilated with larger (20 mL/kg) versus smaller (10 mL/kg) tidal volumes (173). The potential effects of VILI on systemic inflammation and MODS are summarized in Figure 6.

Pathogenesis of ventilator-induced lung injury

207

XI. Summary Recent clinical studies of protective ventilation have confirmed that ventilator-associated lung injury contributes to patient mortality. However, the mechanisms by which low tidal volume ventilation reduces mortality are poorly understood. On the most basic level, our current understanding is that tidal volume reduction prevents excessive end-inspiratory lung volume in the heterogeneously injured and edema-filled lung. This prevents the mechanical disruption of the alveolar-capillary barrier and potentially attenuates straininduced functional changes in alveolar macrophages, epithelial cells, and endothelial cells. These cellular responses to mechanical stimuli include changes in the secretion of pro-inflammatory mediators, alterations in surfactant secretion and function, increased permeability, edema formation, and the loss of lung compartmentalization. Substantial experimental evidence also suggests that excessively low functional residual capacity results in surfactant inactivation, promotes lung inflammation, and results in the loss of compartmentalization. Injurious mechanical ventilation may contribute to patient mortality by exacerbating existing lung injury and thereby prolonging the need for mechanical ventilation and exposing the patient to greater risk of other complications of intensive care. Injurious ventilation may also directly mediate MODS by effecting the release of inflammatory mediators into the systemic blood stream resulting in the dysfunction of other organs. Furthermore, it is likely that low tidal volume strategies of ventilation currently in clinical use do not entirely prevent VALI. An improved understanding of the cellular mechanisms of VALI may further guide the development of ventilation strategies and other therapies for patients with ARDS.

References 1. Petty T, Ashbaugh D. The adult respiratory distress syndrome: clinical features, factors influencing prognosis, and principles of management. Chest 1971; 60:233–239. 2. Sladen A, Laver MB, Pontoppidan H. Pulmonary complications and water retention in prolonged mechanical ventilation. N Engl J Med 1968; 279:448–53. 3. Greenfield L, Ebert P, Benson D. Effect of positive pressure ventilation on surface tension properties of lung extracts. Anesthesiology 1964; 25:312–316. 4. Macklin M, Macklin C. Malignant interstitial emphysema of the lungs and mediastinum as an important occult complication in many respiratory diseases and other conditions: an interpretation of the clinical literature in light of laboratory experiments. Medicine 1944; 23:281–352. 5. Mead J, Takishima T, Leith D. Stress distribution in lungs: a model of pulmonary elasticity. J Appl Physiol 1970; 28:596–608. 6. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000; 342:1301–1308. 7. Schnapp LM, Chin DP, Szaflarski N, Matthay MA. Frequency and importance of barotrauma in 100 patients with acute lung injury. Crit Care Med 1995:23–29.

Acute respiratory distress syndrome

208

8. Weg JG, Anzueto A, Balk RA, et al. The relation of pneumothorax and other air leaks to mortality in the acute respiratory distress syndrome. N Engl J Med 1998:338–347. 9. Wirtz HR, Dobbs LG. The effects of mechanical forces on lung functions. Respir Physiol 2000; 119:1–17. 10. Bachofen H, Schurch S. Alveolar surface forces and lung architecture. Comp Biochem Physiol A Mol Integr Physiol 2001; 129:183–193. 11. West JB. Invited review: pulmonary capillary stress failure. J Appl Physiol 2000; 89:2483– 2489; discussion 2497. 12. Martynowicz MA, Walters BJ, Hubmayr RD. Mechanisms of recruitment in oleic acid-injured lungs. J Appl Physiol 2001; 90:1744–1753. 13. Costello ML, Mathieu-Costello O, West JB. Stress failure of alveolar epithelial cells studied by scanning electron microscopy. Am Rev Respir Dis 1992:145–151. 14. Martynowicz MA, Minor TA, Walters BJ, Hubmayr RD. Regional expansion of oleic acidinjured lungs. Am J Respir Crit Care Med 1999; 160:250–258. 15. Wyszogrodski I, Kyei-Aboagye K, Taeusch HW, Jr., Avery ME. Surfactant inactivation by hyperventilation: conservation by end-expiratory pressure. J Appl Physiol 1975; 38:461–466. 16. Ware L, Matthay M. Alveolar fluid clearance is impaired in the majority of patients with acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:1376–1383. 17. Ingber DE. Tensegrity: the architectural basis of cellular mechanotransduction. Annu Rev Physiol 1997; 59:575–599. 18. Wang N, Naruse K, Stamenovic D, et al. Mechanical behavior in living cells consistent with the tensegrity model. Proc Natl Acad Sci USA 2001; 98: 7765–7770. 19. Tschumperlin DJ, Margulies SS. Alveolar epithelial surface area-volume relationship in isolated rat lungs. J Appl Physiol 1999; 86:2026–2033. 20. Bachofen H, Schurch S, Urbinelli M, Weibel ER. Relations among alveolar surface tension, surface area, volume, and recoil pressure. J Appl Physiol 1987; 62:1878–1887. 21. Eisner MD, Thompson T, Hudson LD, et al. Efficacy of low tidal volume ventilation in patients with different clinical risk factors for acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 164:231–236. 22. Amato MB, Barbas CS, Medeiros DM, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338:347–354. 23. Brochard L, Roudot-Thoraval F, Roupie E, et al. Tidal volume reduction for prevention of ventilator-induced lung injury in acute respiratory distress syndrome. The Multicenter Trail Group on Tidal Volume reduction in ARDS. Am J Respir Crit Care Med 1998; 158:1831–1838. 24. Brower RG, Shanholtz CB, Fessler HE, et al. Prospective, randomized, controlled clinical trial comparing traditional versus reduced tidal volume ventilation in acute respiratory distress syndrome patients. Crit Care Med 1999; 27:1492–1498. 25. Stewart TE, Meade MO, Cook DJ, et al. Evaluation of a ventilation strategy to prevent barotrauma in patients at high risk for acute respiratory distress syndrome. Pressure- and Volume-Limited Ventilation Strategy Group. N Engl J Med 1998:338–342. 26. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342:1334–1349. 27. Bachofen M, Weibel ER. Structural alterations of lung parenchyma in the adult respiratory distress syndrome. Clin Chest Med 1982; 3:35–56. 28. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies. Am J Respir Crit Care Med 1998; 157:294–323. 29. Webb H, Tierney D. Experimental pulmonary edema due to intermittent positive pressure ventilation. Protection by positive end-expiratory pressure. Am Rev Respir Dis 1974; 110:556– 565.

Pathogenesis of ventilator-induced lung injury

209

30. Dreyfuss D, Basset G, Soler P, Saumon G. Intermittent positive-pressure hyperventilation with high inflation pressures produces pulmonary micro-vascular injury in rats. Am Rev Respir Dis 1985:132. 31. Dreyfuss D, Saumon G. Role of tidal volume, FRC, and end-inspiratory volume in the development of pulmonary edema following mechanical ventilation. Am Rev Respir Dis 1993; 148:1194–1203. 32. Dreyfuss D, Soler P, Basset G, Saumon G. High inflation pressure pulmonary edema. Respective effects of high airway pressure, high tidal volume, and positive end-expiratory pressure. Am Rev Respir Dis 1988:137–143. 33. Toung T, Saharia P, Permutt S, Zuidema GD, Cameron JL. Aspiration pneumonia: beneficial and harmful effects of positive end-expiratory pressure. Surgery 1977; 82:279–283. 34. Albert RK, Lakshminarayan S, Charan NB, Kirk W, Butler J. Extra-alveolar vessel contribution to hydrostatic pulmonary edema in in situ dog lungs. J Appl Physiol 1983; 54:1010–1017. 35. Hotchkiss JR, Jr., Blanch L, Naveira A, et al. Relative roles of vascular and airspace pressures in ventilator-induced lung injury. Crit Care Med 2001; 29:1593–1598. 36. West JB, Tsukimoto K, Mathieu-Costello O, Prediletto R. Stress failure in pulmonary capillaries. J Appl Physiol 1991; 70:1731–1742. 37. Clements J. Pulmonary edema and permeability of alveolar membranes. Arch Environ Health 1961; 2:280–283. 38. Chiumello D, Pristine G, Slutsky AS. Mechanical ventilation affects local and systemic cytokines in an animal model of acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160:109–116. 39. Colmenero-Ruiz M, Fernandez-Mondejar E, Fernandez-Sacristan MA, Rivera-Fernandez R, Vasquez-Mata G. PEEP and low tidal volume ventilation reduce lung water in porcine pulmonary edema. Am J Respir Crit Care Med 1997; 155:964–970. 40. Corbridge TC, Wood LD, Crawford GP, Chudoba MJ, Yanos J, Sznajder JL Adverse effects of large tidal volume and low PEEP in canine acid aspiration. Am Rev Respir Dis 1990; 142:311– 315. 41. Frank JA, Gutierrez JA, Jones KD, Allen L, Dobbs L, Matthay MA. Low tidal volume reduces epithelial and endothelial injury in acid-injured rat lungs. Am J Respir Crit Care Med 2002; 165:242–249. 42. Carlton DP, Cummings JJ, Scheerer RG, Poulain FR, Bland RD. Lung overexpansion increases pulmonary microvascular protein permeability in young lambs. J Appl Physiol 1990:69–77. 43. Parker JC, Hernandez LA, Longenecker GL, Peevy K, Johnson W. Lung edema caused by high peak inspiratory pressures in dogs. Role of increased microvascular filtration pressure and permeability. Am Rev Respir Dis 1990; 142:321–328. 44. Hopewell PC. Failure of positive end-expiratory pressure to decrease lung water content in alloxan-induced pulmonary edema. Am Rev Respir Dis 1979; 120:813–819. 45. Hopewell PC, Murray JF. Effects of continuous positive-pressure ventilation in experimental pulmonary edema. J Appl Physiol 1976; 40:568–574. 46. Egan EA. Lung inflation, lung solute permeability, and alveolar edema. J Appl Physiol 1982:53–63. 47. Nicolaysen G, Waaler BA, Aarseth P. On the existence of stretchable pores in the exchange vessels of the isolated rabbit lung preparation. Lymphology 1979:12–14. 48. Parker JC, Ivey CL. Isoproterenol attenuates high vascular pressure-induced permeability increases in isolated rat lungs. J Appl Physiol 1997; 83:1962–1967. 49. Parker JC. Inhibitors of myosin light chain kinase and phosphodiesterase reduce ventilatorinduced lung injury. J Appl Physiol 2000; 89:2241–2248. 50. Parker JC, Ivey CL, Tucker JA. Gadolinium prevents high airway pressure-induced permeability increases in isolated rat lungs. J Appl Physiol 1998; 84:1113–1118. 51. Parker JC, Ivey CL, Tucker A. Phosphotyrosine phosphatase and tyrosine kinase inhibition modulate airway pressure-induced lung injury. J Appl Physiol 1998; 85:1753–1761.

Acute respiratory distress syndrome

210

52. Cooper JA, van der Zee H, Line BR, Malik AB. Relationship of end-expiratory pressure, lung volume, and 99mTc-DTPA clearance. J Appl Physiol 1987; 63:1586–1590. 53. Marks JD, Luce JM, Lazar NM, Wu JN, Lipavsky A, Murray JF. Effect of increases in lung volume on clearance of aerosolized solute from human lungs. J Appl Physiol 1985; 59:1242– 1248. 54. Egan EA. Response of alveolar epithelial solute permeability to changes in lung inflation. J Appl Physiol 1980:49–58. 55. Kim KJ, Crandall ED. Effects of lung inflation on alveolar epithelial solute and water transport properties. J Appl Physiol 1982:52. 56. Imai Y, Nakagawa S, Ito Y, Kawano T, Slutsky AS, Miyasaka K. Comparison of lung protection strategies using conventional and high-frequency oscillatory ventilation. J Appl Physiol 2001; 91:1836–1844. 57. Simma B, Luz G, Trawoger R, et al. Comparison of different modes of high-frequency ventilation in surfactant-deficient rabbits. Pediatr Pulmonol 1996; 22:263–270. 58. Dreyfuss D, Martin-Lefevre L, Saumon G. Hyperinflation-induced lung injury during alveolar flooding in rats: effect of perfluorocarbon instillation. Am J Respir Crit Care Med 1999; 159:1752–1757. 59. Lecuona E, Saldias F, Comellas A, Ridge K, Guerrero C, Sznajder JI. Ventilator-associated lung injury decreases lung ability to clear edema in rats. Am J Respir Crit Care Med 1999; 159:603–609. 60. Saldias FJ, Lecuona E, Comellas AP, Ridge KM, Rutschman DH, Sznajder JI. Beta-adrenergic stimulation restores rat lung ability to clear edema in ventilator-associated lung injury. Am J Respir Crit Care Med 2000; 162:282–287. 61. Berg JT, Fu Z, Breen EC, Tran HC, Mathieu-Costello O, West JB. High lung inflation increases mRNA levels of ECM components and growth factors in lung parenchyma. J Appl Physiol 1997; 83:120–128. 62. Chesnutt AN, Matthay MA, Tibayan FA, Clark JG. Early detection of type III procollagen peptide in acute lung injury. Pathogenetic and prognostic significance. Am J Respir Crit Care Med 1997; 156:840–845. 63. Parker JC, Breen EC, West JB. High vascular and airway pressures increase interstitial protein mRNA expression in isolated rat lungs. J Appl Physiol 1997; 83:1697–1705. 64. Al-Jamal R, Ludwig MS. Changes in proteoglycans and lung tissue mechanics during excessive mechanical ventilation in rats. Am J Physiol Lung Cell Mol Physiol 2001; 281:L1078–1087. 65. Foda HD, Rollo EE, Drews M, et al. Ventilator-induced lung injury upregulates and activates gelatinases and EMMPRIN. Attenuation by the synthetic matrix metalloproteinase inhibitor, prinomastat (ag3340). Am J Respir Cell Mol Biol 2001; 25:717–724. 66. Foda HD, Rollo EE, Drews M, et al. Ventilator-induced lung injury upregulates and activates gelatinases and EMMPRIN: attenuation by the synthetic matrix metalloproteinase inhibitor, prinomastat (AG3340). Am J Respir Cell Mol Biol 2001; 25:717–724. 67. Tschumperlin DJ, Oswari J, Margulies AS. Deformation-induced injury of alveolar epithelial cells. Effect of frequency, duration, and amplitude. Am J Respir Crit Care Med 2000; 162:357– 362. 68. Vlahakis NE, Schroeder MA, Pagano RE, Hubmayr RD. Deformation-induced lipid trafficking in alveolar epithelial cells. Am J Physiol Lung Cell Mol Physiol 2001; 280:L938–946. 69. Vlahakis N, Schroeder MA, Pagano RE, Hubmayr RD. The role of the cytoskeleton in deformation induced alveolar epithelial wounding and repair. Am J Respir Crit Care Med 2002; 165:A379. 70. Cavanaugh KJ, Jr., Oswari J, Margulies SS. Role of stretch on tight junction structure in alveolar epithelial cells. Am J Respir Cell Mol Biol 2001; 25: 584–591. 71. Dudek SM, Garcia JG. Cytoskeletal regulation of pulmonary vascular permeability. J Appl Physiol 2001; 91:1487–1500.

Pathogenesis of ventilator-induced lung injury

211

72. Muscedere JG, Mullen JB, Gan K, Slutsky AS. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med 1994; 149:1327–1334. 73. Gattinoni L, D’Andrea L, Pelosi P, Vitale G, Pesenti A, Fumagalli R. Regional effects and mechanism of positive end-expiratory pressure in early adult respiratory distress syndrome. JAMA 1993; 269:2122–2127. 74. Gattinoni L, Pelosi P, Crotti S, Valenza F. Effects of positive end-expiratory pressure on regional distribution of tidal volume and recruitment in adult respiratory distress syndrome. Am J Respir Crit Care Med 1995:151–160. 75. Crotti S, Mascheroni D, Caironi P, et al. Recruitment and derecruitment during acute respiratory failure: a clinical study. Am J Respir Crit Care Med 2001; 164:131–140. 76. Hickling KG, Walsh J, Henderson S, Jackson R. Low mortality rate in adult respiratory distress syndrome using low-volume, pressure-limited ventilation with permissive hypercapnia: a prospective study. Crit Care Med 1994:22–29. 77. Adams AB, Cakar N, Marini JJ. Static and dynamic pressure-volume curves reflect different aspects of respiratory system mechanics in experimental acute respiratory distress syndrome. Respir Care 2001; 46:686–693. 78. Hickling KG. The pressure-volume curve is greatly modified by recruitment. A mathematical model of ARDS lungs. Am J Respir Crit Care Med 1998; 158:194–202. 79. Jonson B, Richard JC, Straus C, Mancebo J, Lemaire F, Brochard L. Pressure-volume curves and compliance in acute lung injury: evidence of recruitment above the lower inflection point. Am J Respir Crit Care Med 1999; 159:1172–1178. 80. Karason S, Sondergaard S, Lundin S, Wiklund J, Stenqvist O. Evaluation of pressure/volume loops based on intratracheal pressure measurements during dynamic conditions. Acta Anaesthesiol Scand 2000; 44:571–577. 81. ARDSNetwork. Update on the ALVEOLI study (R. Brower for the ARDS Network investigators), American Thoracic Society International Conference, Atlanta, Georgia, 2002. 82. Mehta S, Lapinsky SE, Hallett DC, et al. Prospective trial of high-frequency oscillation in adults with acute respiratory distress syndrome. Crit Care Med 2001; 29:1360–1369. 83. Oyarzun MJ, Stevens P, Clements JA. Effect of lung collapse on alveolar surfactant in rabbits subjected to unilateral pneumothorax. Exp Lung Res 1989; 15:909–924. 84. Albert RK, Lakshminarayan S, Hildebrandt J, Kirk W, Butler J. Increased surface tension favors pulmonary edema formation in anesthetized dogs’ lungs. J Clin Invest 1979; 63:1015– 1018. 85. D’Angelo E, Pecchiari M, Baraggia P, Saetta M, Balestro E, Milic-Emili J. Low-volume ventilation causes peripheral airway injury and increased airway resistance in normal rabbits. J Appl Physiol 2002; 92:949–956. 86. Rimensberger PC, Cox PN, Frndova H, Bryan AC. The open lung during small tidal volume ventilation: concepts of recruitment and “optimal” positive end-expiratory pressure. Crit Care Med 1999; 27:1946–1952. 87. Yoder BA, Siler-Khodr T, Winter VT, Coalson JJ. High-frequency oscillatory ventilation: effects on lung function, mechanics, and airway cytokines in the immature baboon model for neonatal chronic lung disease. Am J Respir Crit Care Med 2000; 162:1867–1876. 88. Sugiura M, McCulloch PR, Wren S, Dawson RH, Froese AB. Ventilator pattern influences neutrophil influx and activation in atelectasis-prone rabbit lung. J Appl Physiol 1994; 77:1355– 1365. 89. McCulloch PR, Forkert PG, Froese AB. Lung volume maintenance prevents lung injury during high frequency oscillatory ventilation in surfactant-deficient rabbits. Am Rev Respir Dis 1988; 137:1185–1192. 90. Del Maschio A, Zanetti A, Corada M, et al. Polymorphonuclear leukocyte adhesion triggers the disorganization of endothelial cell-to-cell adherens junctions. J Cell Biol 1996; 135:497–510.

Acute respiratory distress syndrome

212

91. Ukropec JA, Hollinger MK, Salva SM, Woolkalis MJ. SHP2 association with VE-cadherin complexes in human endothelial cells is regulated by thrombin. J Biol Chem 2000; 275:5983– 5986. 92. Hirata A, Baluk P, Fujiwara T, McDonald DM. Location of focal silver staining at endothelial gaps in inflamed venules examined by scanning electron microscopy. Am J Physiol 1995; 269:L403–418. 93. Tinsley JH, De Lanerolle P, Wilson E, Ma W, Yuan SY. Myosin light chain kinase transference induces myosin light chain activation and endothelial hyperpermeability. Am J Physiol Cell Physiol 2000; 279: C1285–1289. 94. Wysolmerski RB, Lagunoff D. Regulation of permeabilized endothelial cell retraction by myosin phosphorylation. Am J Physiol 1991; 261:C32–40. 95. Garcia JG, Verin AD, Herenyiova M, English D. Adherent neutrophils activate endothelial myosin light chain kinase: role in transendothelial migration. J Appl Physiol 1998; 84:1817– 1821. 96. Shi S, Garcia JG, Roy S, Parinandi NL, Natarajan V. Involvement of c-Src in diperoxovanadate-induced endothelial cell barrier dysfunction. Am J Physiol Lung Cell Mol Physiol 2000; 279:L441–451. 97. Ridley AJ, Hall A. The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 1992; 70:389–399. 98. Petrache I, Verin AD, Crow MT, Birukova A, Liu F, Garcia JG. Differential effect of MLC kinase in TNF-alpha-induced endothelial cell apoptosis and barrier dysfunction. Am J Physiol Lung Cell Mol Physiol 2001; 280:L1168–1178. 99. Rousseau S, Houle F, Kotanides H, et al. Vascular endothelial growth factor (VEGF)-driven actin-based motility is mediated by VEGFR2 and requires concerted activation of stressactivated protein kinase 2 (SAPK2/p38) and geldanamycin-sensitive phosphorylation of focal adhesion kinase. J Biol Chem 2000; 275:10661–10672. 100. Zhao Y, Davis HW. Endotoxin causes phosphorylation of MARCKS in pulmonary vascular endothelial cells. J Cell Biochem 2000; 79:496–505. 101. Schneider GB, Hamano H, Cooper LF. In vivo evaluation of hsp27 as an inhibitor of actin polymerization: hsp27 limits actin stress fiber and focal adhesion formation after heat shock. J Cell Physiol 1998; 177:575–584. 102. Meyer CJ, Alenghat FJ, Rim P, Fong JH, Fabry B, Ingber DE. Mechanical control of cyclic AMP signaling and gene transcription through integrins. Nat Cell Biol 2000; 2:666–668. 103. Bhullar IS, Li YS, Miao H, et al. Fluid shear stress activation of IkappaB kinase is integrindependent. J Biol Chem 1998; 273:30544–30549. 104. Li S, Kim M, Hu YL, et al. Fluid shear stress activation of focal adhesion kinase. Linking to mitogen-activated protein kinases. J Biol Chem 1997; 272:30455–30462. 105. Schwachtgen JL, Houston P, Campbell C, Sukhatme V, Braddock M. Fluid shear stress activation of egr-1 transcription in cultured human endothelial and epithelial cells is mediated via the extracellular signal-related kinase 1/2 mitogen-activated protein kinase pathway. J Clin Invest 1998; 101:2540–2549. 106. Chicurel ME, Singer RH, Meyer CJ, Ingber DE. Integrin binding and mechanical tension induce movement of mRNA and ribosomes to focal adhesions. Nature 1998; 392:730–733. 107. Berrios JC, Schroeder MA, Hubmayr RD. Mechanical properties of alveolar epithelial cells in culture. J Appl Physiol 2001; 91:65–73. 108. Cai S, Pestic-Dragovich L, O’Donnell ME, et al. Regulation of cytoskeletal mechanics and cell growth by myosin light chain phosphorylation. Am J Physiol 1998; 275:C1349–1356. 109. Bhattacharya S, Patel R, Bhattacharya J. High volume ventilation causes tyrosine phosphorylation of endothelial focal adhesion proteins (abst). Am J Respir Crit Care Med 2000; 161:723A. 110. Liu M, Qin Y, Liu J, Tanswell AK, Post M. Mechanical strain induces pp60src activation and translocation to cytoskeleton in fetal rat lung cells. J Biol Chem 1996; 271:7066–7071.

Pathogenesis of ventilator-induced lung injury

213

111. Liu M, Xu J, Liu J, Kraw ME, Tanswell AK, Post M. Mechanical strain-enhanced fetal lung cell proliferation is mediated by phospholipase C and D and protein kinase C. Am J Physiol 1995; 268:L729–738. 112. Swartz MA, Tschumperlin DJ, Kamm RD, Drazen JM. Mechanical stress is communicated between different cell types to elicit matrix remodeling. Proc Natl Acad Sci USA 2001; 98:6180–6185. 113. Gutierrez JA, Ertsey R, Scavo LM, Collins E, Dobbs LG. Mechanical distention modulates alveolar epithelial cell phenotypic expression by transcriptional regulation. Am J Respir Cell Mol Biol 1999; 21:223–229. 114. Nakamura T, Liu M, Mourgeon E, Slutsky A, Post M. Mechanical strain and dexamethasone selectively increase surfactant protein C and tropoelastin gene expression. Am J Physiol Lung Cell Mol Physiol 2000; 278:L974–980. 115. Wirtz HR, Dobbs LG. Calcium mobilization and exocytosis after one mechanical stretch of lung epithelial cells. Science 1990; 250:1266–1269. 116. Faridy EE. Effect of ventilation on movement of surfactant in airways. Respir Physiol 1976; 27:323–334. 117. Faridy EE, Permutt S, Riley RL. Effect of ventilation on surface forces in excised dogs’ lungs. J Appl Physiol 1966; 21:1453–1462. 118. McClenahan JB, Urtnowski A. Effect of ventilation on surfactant, and its turnover rate. J Appl Physiol 1967; 23:215–220. 119. Veldhuizen RA, Tremblay LN, Govindarajan A, van Rozendaal BA, Haagsman HP, Slutsky AS. Pulmonary surfactant is altered during mechanical ventilation of isolated rat lung. Crit Care Med 2000; 28:2545–2551. 120. Yukitake K, Brown CL, Schlueter MA, Clements JA, Hawgood S. Surfactant apoprotein A modifies the inhibitory effect of plasma proteins on surfactant activity in vivo. Pediatr Res 1995; 37:21–25. 121. Veldhuizen RA, Welk B, Harbottle R, et al. Mechanical ventilation of isolated rat lungs changes the structure and biophysical properties of surfactant. J Appl Physiol 2002; 92:1169– 1175. 122. Cockshutt AM, Weitz J, Possmayer F. Pulmonary surfactant-associated protein A enhances the surface activity of lipid extract surfactant and reverses inhibition by blood proteins in vitro. Biochemistry 1990; 29:8424–8429. 123. Veldhuizen RA, Slutsky AS, Joseph M, McCaig L. Effects of mechanical ventilation of isolated mouse lungs on surfactant and inflammatory cytokines. Eur Respir J 2001; 17:488–494. 124. Malloy JL, Veldhuizen RA, Lewis JF. Effects of ventilation on the surfactant system in sepsisinduced lung injury. J Appl Physiol 2000; 88:401–408. 125. Ogawa A, Brown CL, Schlueter MA, Benson BJ, Clements JA, Hawgood S. Lung function, surfactant apoprotein content, and level of PEEP in prematurely delivered rabbits. J Appl Physiol 1994; 77:1840–1849. 126. Ito Y, Veldhuizen RA, Yao LJ, McCaig LA, Bartlett AJ, Lewis JF. Ventilation strategies affect surfactant aggregate conversion in acute lung injury. Am J Respir Crit Care Med 1997:155. 127. Krause M, Olsson T, Law AB, et al. Effect of volume recruitment on response to surfactant treatment in rabbits with lung injury. Am J Respir Crit Care Med 1997; 156:862–866. 128. Krause MF, Jakel C, Haberstroh J, Schulte-Monting J, Leititis JU, Orlowska-Volk M. Alveolar recruitment promotes homogeneous surfactant distribution in a piglet model of lung injury. Pediatr Res 2001; 50:34–43. 129. Clements JA. Lung surfactant: a personal perspective. Annu Rev Physiol 1997; 59:1–21. 130. Baker CS, Evans TW, Randle BJ, Haslam PL. Damage to surfactant-specific protein in acute respiratory distress syndrome. Lancet 1999; 353:1232–1237. 131. Laffey JG, Engelberts D, Kavanagh BP. Buffering hypercapnic acidosis worsens acute lung injury. Am J Respir Crit Care Med 2000; 161:141–146.

Acute respiratory distress syndrome

214

132. Laffey JG, Tanaka M, Engelberts D, et al. Therapeutic hypercapnia reduces pulmonary and systemic injury following in vivo lung reperfusion. Am J Respir Crit Care Med 2000; 162:2287–2294. 133. Broccard AF, Hotchkiss JR, Vannay C, et al. Protective effects of hypercapnic acidosis on ventilator-induced lung injury. Am J Respir Crit Care Med 2001; 164:802–806. 134. Ranieri VM, Suter PM, Tortorella C, et al. Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: a randomized controlled trial. Jama 1999; 282:54–61. 135. Meduri GU, Headley S, Kohler G, et al. Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Plasma IL-1 beta and IL-6 levels are consistent and efficient predictors of outcome over time. Chest 1995; 107:1062–1073. 136. Goodman RB, Stricter RM, Martin DP, et al. Inflammatory cytokines in patients with persistence of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154:602–611. 137. Park WY, Goodman RB, Steinberg KP, et al. Cytokine balance in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 164:1896–1903. 138. Wrigge H, Zinserling J, Stuber F, et al. Effects of mechanical ventilation on release of cytokines into systemic circulation in patients with normal pulmonary function. Anesthesiology 2000; 93:1413–1417. 139. Imanaka H, Shimaoka M, Matsuura N, Nishimura M, Ohta N, Kiyono H. Ventilator-induced lung injury is associated with neutrophil infiltration, macrophage activation, and TGF-beta 1 mRNA upregulation in rat lungs. Anesth Analg 2001; 92:428–436. 140. Dreyfuss D, Saumon G. Production of inflammatory cytokines in ventilator-induced lung injury: a reappraisal. Am J Respir Crit Care Med 2001; 163:1176–1180. 141. Tremblay L, Valenza F, Ribeiro SP, Li J, Slutsky AS. Injurious ventilatory strategies increase cytokines and c-fos m-RNA expression in an isolated rat lung model. J Clin Invest 1997; 99:944–952. 142. Blackwell TS, Christman JW. The role of nuclear factor-kappa B in cytokine gene regulation. Am J Respir Cell Mol Biol 1997; 17:3–9. 143. Nakamura T, Malloy J, McCaig L, et al. Mechanical ventilation of isolated septic rat lungs: effects on surfactant and inflammatory cytokines. J Appl Physiol 2001; 91:811–820. 144. Imai Y, Kawano T, Iwamoto S, Nakagawa S, Takata M, Miyasaka K. Intratracheal anti-tumor necrosis factor-alpha antibody attenuates ventilator-induced lung injury in rabbits. J Appl Physiol 1999; 87:510–515. 145. Narimanbekov IO, Rozycki HJ. Effect of IL-1 blockade on inflammatory manifestations of acute ventilator-induced lung injury in a rabbit model. Exp Lung Res 1995:21–27. 146. Ribeiro SP, Rhee K, Tremblay L, Veldhuizen R, Lewis JF, Slutsky AS. Heat stress attenuates ventilator-induced lung dysfunction in an ex vivo rat lung model. Am J Respir Crit Care Med 2001; 163:1451–1456. 147. Ribeiro SP, Villar J, Downey GP, Edelson JD, Slutsky AS. Effects of the stress response in septic rats and LPS-stimulated alveolar macrophages: evidence for TNF-alpha posttranslational regulation. Am J Respir Crit Care Med 1996; 154:1843–1850. 148. Villar J, Ribeiro SP, Mullen JB, Kuliszewski M, Post M, Slutsky AS. Induction of the heat shock response reduces mortality rate and organ damage in a sepsis-induced acute lung injury model. Crit Care Med 1994; 22:914–921. 149. Held HD, Boettcher S, Hamann L, Uhlig S. Ventilation-induced chemokine and cytokine release is associated with activation of nuclear factor-KB and is blocked by steroids. Am J Respir Crit Care Med 2001; 163:711–716. 150. Kuebler WM, Parthasarathi K, Wang PM, Bhattacharya J. A novel signaling mechanism between gas and blood compartments of the lung. J Clin Invest 2000; 105:905–913. 151. Vlahakis NE, Schroeder MA, Limper AH, Hubmayr RD. Stretch induces cytokine release by alveolar epithelial cells in vitro. Am J Physiol 1999; 277:L167–173.

Pathogenesis of ventilator-induced lung injury

215

152. Quinn D, Tager A, Joseph PM, Bonventre JV, Force T, Hales CA. Stretch-induced mitogenactivated protein kinase activation and interleukin-8 production in type II alveolar cells. Chest 1999; 116:89S–90S. 153. MacGillivray MK, Cruz TF, McCulloch CA. The recruitment of the interleukin-1 (IL-1) receptor-associated kinase (IRAK) into focal adhesion complexes is required for IL-1 betainduced ERK activation. J Biol Chem 2000; 275:23509–23515. 154. Pugin J, Dunn I, Jolliet P, et al. Activation of human macrophages by mechanical ventilation in vitro. Am J Physiol 1998; 275:L1040–1050. 155. Dunn I, Pugin J. Mechanical ventilation of various human lung cells in vitro: identification of the macrophage as the main producer of inflammatory mediators. Chest 1999; 116:95S–97S. 156. Kurdowska A, Miller EJ, Noble JM, et al. Anti-IL-8 autoantibodies in alveolar fluid from patients with the adult respiratory distress syndrome. J Immunol 1996; 157:2699–2706. 157. Miller EJ, Cohen AB, Matthay MA. Increased interleukin-8 concentrations in the pulmonary edema fluid of patients with acute respiratory distress syndrome from sepsis. Crit Care Med 1996; 24:1448–1454. 158. Lentsch AB, Czermak BJ, Bless NM, Van Rooijen N, Ward PA. Essential role of alveolar macrophages in intrapulmonary activation of NF-kappaB. Am J Respir Cell Mol Biol 1999; 20:692–698. 159. Grembowicz KP, Sprague D, McNeil PL. Temporary disruption of the plasma membrane is required for c-fos expression in response to mechanical stress. Mol Biol Cell 1999; 10:1247– 1257. 160. McRitchie DI, Isowa N, Edelson JD, et al. Production of tumour necrosis factor alpha by primary cultured rat alveolar epithelial cells. Cytokine 2000; 12:644–654. 161. Schwartz MD, Moore EE, Moore FA, et al. Nuclear factor-kappa B is activated in alveolar macrophages from patients with acute respiratory distress syndrome. Crit Care Med 1996; 24:1285–1292. 162. Liu C, Yao J, de Belle I, Huang RP, Adamson E, Mercola D. The transcription factor EGR-1 suppresses transformation of human fibrosarcoma HT1080 cells by coordinated induction of transforming growth factor-beta 1, fibronectin, and plasminogen activator inhibitor-1. J Biol Chem 1999; 274:4400–4411. 163. Ressler B, Lee RT, Randell SH, Drazen JM, Kamm RD. Molecular responses of rat tracheal epithelial cells to transmembrane pressure. Am J Physiol Lung Cell Mol Physiol 2000; 278:L1264–1272. 164. Silverman ES, Collins T. Pathways of Egr-1-mediated gene transcription in vascular biology. Am J Pathol 1999; 154:665–670. 165. Murphy DB, Cregg N, Tremblay L, et al. Adverse ventilatory strategy causes pulmonary-tosystemic translocation of endotoxin. Am J Respir Crit Care Med 2000; 162:27–33. 166. von Bethmann AN, Brasch F, Nusing R, et al. Hyperventilation induces release of cytokines from perfused mouse lung. Am J Respir Crit Care Med 1998:157. 167. Douzinas E, Tsidemiadou P, Pitaridis M, et al. The regional production of cytokines and lactate in sepsis-related multiple organ failure. Am. J. Respir. Crit. Care Med. 1997; 155:53–59. 168. Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald WJ. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995; 23:1638–1652. 169. Nahum A, Hoyt J, Schmitz L, Moody J, Shapiro R, Marini JJ. Effect of mechanical ventilation strategy on dissemination of intratracheally instilled Escherichia coli in dogs. Crit Care Med 1997; 25:1733–1743. 170. Verbrugge SJ, Sorm V, van’t Veen A, Mouton JW, Gommers D, Lachmann B. Lung overinflation without positive end-expiratory pressure promotes bacteremia after experimental Klebsiella pneumoniae inoculation. Intensive Care Med 1998; 24:172–177. 171. Imai Y, Kajikawa O, Frevert C, et al. Injurious ventilatory strategies enhance organ apoptosis in rabbits (abst). Am J Respir Crit Care Med 2001; 163: 667A.

Acute respiratory distress syndrome

216

172. Love R, Choe E, Lippton H, Flint L, Steinberg S. Positive end-expiratory pressure decreases mesenteric blood flow despite normalization of cardiac output. J Trauma 1995; 39:195–199. 173. Guery B, Neviere R, Fialdes P, et al. Mechanical ventilation regimen induces intestinal permeability changes in a rat model. Am J Respir Crit Care Med 2002; 165:A505. 174. Slutsky A, Tremblay L. Multiple system organ failure. Is mechanical ventilation a contributing factor? Am J Respir Crit Care Med 1998; 157:1721–1725. 175. Brower RG, Fessler HE. Mechanical ventilation in acute lung injury and acute respiratory distress syndrome. Clin Chest Med 2000; 21:491–510.

10 Pathogenesis of Sepsis and Septic-Induced Lung Injury GUY A.ZIMMERMAN, KURT H.ALBERTINE, and THOMAS M.McINTYRE University of Utah Salt Lake City, Utah, U.S.A.

I. Introduction And here comes in the great tragedy—sepsis everywhere, unavoidable sepsis! Sir William Osler

Sepsis, a systemic condition with several stages and degrees of severity resulting from dysregulated activation of the innate immune and hemostatic systems, is a lethal syndrome and a major cause of acute lung injury (ALI) and its progression to the acute respiratory distress syndrome (ARDS). The concept that sepsis can lead to end organ injury and dysfunction dates to the Greek origin of the term (sepsios) and to investigations of its pathogenesis that predate modern times (1). Although not reported as a cause of lung injury in the 12 patients in the seminal description of ARDS (2), sepsis was later recognized as a major inciting factor by Petty and coworkers (3). In addition, many subsequent reports demonstrated an association between sepsis and ALI, and recent reviews identify it as the leading condition that “triggers” the molecular and cellular cascades that culminate in ALI and ARDS in humans (4–12). Sepsis-induced ALI and ARDS can be caused by infection with gram-positive bacteria, gram-negative organisms, or fungi, although in many series gram-negative bacterial infection dominates (4). Circulating bacterial lipopolysaccharide (LPS) has been associated with development of ARDS induced by diverse predisposing conditions in some, but not all, studies, indicating that it may act in an additive or synergistic fashion with trauma, hemorrhage, and other insults to induce alveolar-capillary membrane injury (13, 14). However, here we will focus for the most part on syndromes in which sepsis is presumed to be the principal or only inciting condition for ALI and ARDS. The incidence of ARDS associated with sepsis and its related syndromes (see below) is variable, but appears to be in the range of 25–40% (4, 7, 12, 15). It is estimated that between 400,000 and 750,000 episodes of sepsis and septic shock occur annually in the United States (potentially as high as 3.0 cases per 1000 population) and that this may increase to as many as 1,100,000 cases by the year 2020 (16, 17). In that event the

Acute respiratory distress syndrome

218

number of sepsis-induced cases of ALI and ARDS might approach or exceed 440,000/year, resulting in mortality of significant proportion based on past and current outcomes of management (7, 12, 18, 19). To this would be added the morbidity and mortality of sepsis-induced ALI and ARDS elsewhere in the world (6, 20).

II. Definitions The most precise definition of sepsis is that it is a clinical syndrome resulting from systemic responses to a localized or disseminated infection (17, 21, 22). There are several related syndromes and/or stages of escalating severity, including severe sepsis with lung or other end organ dysfunction, septic shock, and multiple organ dysfunction syndrome (MODS). Clinical identification of patients in recent years has largely been based on consensus definitions that are widely, but not universally, accepted. The consensus definitions also provide criteria for a systemic inflammatory response syndrome (SIRS) in the absence of documented infection, a condition that can be triggered by other insults and injuries in addition to infection (17, 20, 23). One potential mechanism is entry of LPS and other bacterial toxins into the circulation because of disruption of intestinal integrity (5). More recently, the consensus definitions for sepsis and its various stages and for SIRS have been questioned, and it is suggested that they lack sufficient precision even when an infection is rigorously documented. In part, this is a result of disappointing and confusing negative outcomes of interventional trials (16, 17, 22, 24, 25). One of the reasons that current definitions may not be sufficiently precise is that human sepsis appears to be extremely heterogeneous (24). This makes it difficult to define and rigorously characterize specific phenotypes within the spectrum of sepsis and consequently to identify genetic traits that influence sepsis and its complications and to sort out interaction of these traits with environmental variables (1, 26, 27) (see Chap. 14). Other factors that contribute to imprecision in clinical definition of sepsis include the complexity of pathophysiological responses in sepsis, gaps in fundamental knowledge regarding the cellular and molecular responses to infection, unknown features that influence systemic responses to infection, a deficit in understanding of variables of microbial virulence, and incomplete understanding of mechanisms that contribute to compartmentalization of molecular events triggered by microbes under some conditions yet incite dramatic systemic responses in others (17, 24, 28–30a).

III. Surrogate Models As outlined above, sepsis and its complications are defined by clinical criteria. Many surrogate animal models of sepsis have been developed, some of which have been specifically employed to study lung responses to infectious agents and to mediators that are generated in systemic responses to infection (31–37). Although extremely useful for a number of purposes, no animal model has yet been identified that exactly reproduces the clinical syndrome of human sepsis or the end-organ complications that occur in its various stages and severities (24, 28, 38, 39). A limitation of many animal studies is that they utilize a single bacterial component such as LPS, whereas clinical sepsis involves a

Pathogenesis of sepsis and septic-induced lung injury

219

myriad of bacterial factors; in addition, most use acute and short-term conditions that do not reflect the time frame and comorbidities of clinical sepsis (24, 39). “Natural” infectious models, such as cecal ligation and puncture (CLP), polymicrobial peritonitis induced in other ways, and models involving “multiple hits,” may be more revealing regarding cellular, physiological, and genomic features that mimic human sepsis and sepsis-induced organ failure (36, 37, 40–46). One potential problem, however, is that in many models of CLP in mice, the degree of lung injury and its similarity to human ALI have not been compelling and have been difficult to reproduce from laboratory to laboratory. This likely involves differences in genetic background in the mice used for these studies. Animal models also demonstrate that the lung can be the source of septic infection and a critical modulator of systemic responses in sepsis and SIRS (47, 48). This is consistent with clinical observations (11, 49).

IV. Pathophysiology of Sepsis: An Overview Current concepts of sepsis identify dysregulated or unregulated activation of the innate immune system as a central event in the pathogenesis of sepsis and dysregulated activation of the hemostatic cascade as a second critical mechanism that goes hand-inhand with pathological innate immune responses (39, 50–52a). Additional features, including activation of nitric oxide synthases and cyclooxygenases and altered adrenergic receptor reactivity, mediate the hemodynamic events of septic shock (11, 53, 54). While these hemodynamic mechanisms will not be discussed in detail, there is evidence they can also be influenced by inflammatory and thrombotic effector systems (55–57). The cellular and molecular mechanisms that govern the innate immune system in sepsis and influence its interactions with the hemostatic

Acute respiratory distress syndrome

220

Figure 1 Cellular effectors of sepsis and sepsis-induced lung injury. Tolllike receptors and other receptor systems are involved in the signaling cascades leading to activation of myeloid leukocytes, endothelial cells, and other cellular systems. Dysregulated triggering of hemostatic events, including platelet activation, also occurs. See text for details. system are complex (Fig. 1). Furthermore, new components and pathways continue to be discovered, indicating that we are far from a comprehensive understanding.

Pathogenesis of sepsis and septic-induced lung injury

221

A. Myeloid Leukocytes and Endothelial Cells in Sepsis Several lines of evidence indicate an essential role for host hematopoietic cells in sepsis, including reconstitution experiments in which lethal effects of LPS were restored in C3H/HeJ mice, which are resistant to LPS because of a mutation in the Lps gene (see below), by irradiation and subsequent reconstitution with hematopoietic precursors from an LPS-sensitive mouse strain (1, 58). Myeloid leukocytes are key innate immune/inflammatory effector cells in sepsis (Fig. 1). These immune effector cells are nonclonally derived and capable of responding to pathogens without conditioning by prior exposure or antigenic specificity (1, 17, 59, 60). Myeloid cells with critical roles in sepsis and sepsis-induced ALI include neutrophils (polymorphonuclear leukocytes, PMNs), monocytes, and macrophages. Macrophages recognize bacteria and bacterial products including LPS, modulate systemic pro- and anti-inflammatory responses to infection, and can be effectors of tissue injury as well as of tissue repair (50, 61). Monocytes and neutrophils are also involved in recognition of bacteria and their products and are critical for localized containment of invading microbes (60, 62). In addition, however, PMNs and monocytes are critical effectors of injury in sepsis and its complications and, by virtue of mechanisms such as granular enzyme release and oxygen radical generation, induce vascular leakage and tissue damage (63–67). The phenotypes of monocytes and neutrophils change in response to mediators generated at particular stages in clinical sepsis (66, 68, 69), but the mechanisms involved are largely undefined. Human and rat neutrophils use integrins of both the β1 and β2 classes for adhesion and trafficking in septic conditions, including trafficking into the lungs in experimental sepsis (66, 70). These studies support earlier observations of use of alternative integrins by PMNs (71) in addition to the β2 family (60). Dendritic cells, natural killer cells, and specific populations of gamma delta T cells play roles in sepsis that are not yet completely characterized. These leukocyte subclasses add complexity to contributions by PMNs, monocytes, and macrophages and are also involved in critical interactions between the innate and adaptive immune systems in responses to infectious challenge (50, 72–74a). There is frequently a phase of relative immune refractoriness in later stages after a septic challenge that is likely mediated by CD4+T lymphocytes polarized to the Th2 phenotype and that may predispose to secondary fungal or bacterial infection if the patient survives the initial septic episode (17). Endothelial cells are also critical in host defense and repair mediated by the acute inflammatory system and in dysregulated inflammatory responses in sepsis and sepsisinduced ALI (52, 75–76a) (Figs. 1, 2). Dysregulated interactions between activated or injured endothelial cells and myeloid leukocytes may be particularly important in sepsis (11, 64, 77) and may contribute to intravascular leukocyte trapping in the lung as well as trafficking to alveolar compartments (see below). Endothelial cells may be released into the circulation in septic patients (78) as a result of injury mediated by locally activated neutrophils or monocytes. Endothelial cells express Toll-like receptors (see below) and recognize LPS and other bacterial products (52, 63, 79, 80). In response, stimulation with LPS human endothelial cells in culture express new gene products, including inducible adhesion molecules and chemokines that mediate interactions with leukocytes, and undergo phenotypic and functional alterations (63, 76, 80–83).

Acute respiratory distress syndrome

222

Figure 2 LPS and other endotoxins activate human endothelial cells: a mechanism of gram-negative sepsis. In addition to activating myeloid leukocytes (see Fig. 1), LPS and other components of gram-negative bacteria bind to Toll-like receptors on human endothelial cells, inducing new gene expression and other functional respones. Similarly, endothelial cells respond to components of grampositive bacteria using TLRs different from those that recognize LPS; thus, endothelial cell activation is also a fundamental mechanism in grampositive sepsis. Similar changes occur in vessels of septic patients, indicating the validity of the model (84). Altered expression of endothelial genes and synthesis of the corresponding protein products in response to LPS may be a central feature of gram-negative sepsis (Fig. 2). However, the pattern of messenger RNAs, including transcripts that code for gene products relevant to systemic manifestations of infection and inflammatory injury, is also altered when human endothelial cells are challenged with bacterial lipoproteins in a fashion that is as robust as when they are challenged with LPS (81). Expression of new gene products by endothelial cells activated by bacterial lipoproteins or stimulated with other microbial products or toxins besides LPS (Fig. 2) may be one of the factors that accounts for failure of antibodies with specificity for LPS to prevent septic complications

Pathogenesis of sepsis and septic-induced lung injury

223

in clinical trials (28, 38). Further, these experiments emphasize that the systemic “endotoxic” response is mediated by microbial products in addition to LPS (39, 63, 81) (Figs. 1, 2). B. Toll-Like Receptors: A Recognition System for LPS and Conserved Motifs Displayed by Pathogens Toll-like receptors (TLRs) are a family of transmembrane receptors expressed by effectors of the innate immune system and related cells. These surface molecules recognize LPS and other microbial molecules, including bacterial lipoproteins, peptidoglycans, and lipoteichoic acid, and transduce signals to the intracellular environment (1, 85–88). The ligands for TLRs have been characterized as displaying pathogen-associated molecular patterns (PAMPs) and contain conserved structural motifs that are critical for microbial survival. Recognition of these motifs is thought to allow the mammalian host to distinguish the invading pathogen from self (89). Mammalian TLRs are related to the Toll surface molecules in Drosophilia, which signal responses to infection and, in addition, are critically involved in development (1, 88). The recent identification and ongoing characterization of TLRs in humans provides new molecular insight into host defense and pathological dysregulation of the innate immune system in sepsis and other diseases (1, 39, 50, 61, 86, 90, 91). Polymorphisms in the gene for TLR4 (see below) are associated with altered inflammatory responses and outcome of septic shock (27, 92). Discovery in the mid-1960s of a strain of inbred mice that are unresponsive to LPS, C3H/HeJ, was central to understanding of TLRs in microbial defense and infectious injury (1). Positional cloning of the LPS locus in C3H/HeJ mice identified the Toll-like receptor 4 (Tlr 4) gene, allowing characterization of the defect in recognition of LPS in C3H/HeJ and identification of a second unresponsive murine strain, C57BL/10ScCr, which is also due to a loss-of-function mutation in Tlr 4 (1, 50, 93). Ten human TLRs have been identified in addition to multiple murine orthologs (1, 61, 88, 90, 91). In humans, as in mice, TLR4 recognizes LPS, although there are species-specific differences in the recognition of partial LPS structures (1). TLR 2 is the receptor for lipopeptides, which include molecules common to gram-negative and gram-positive organisms (39, 81), and for peptidoglycans and lipoteichoic acid of gram-positive bacteria (1, 61, 88). Trace amounts of contaminating bacterial factors in test preparations of LPS or lipopeptides can confuse assays for specific recognition by TLRs (1, 94). TLRs act cooperatively to mediate signaling in response to pathogens (95). TLRs are expressed in several cell types, and their mRNAs are differentially regulated in leukocytes challenged with microbial products (96). TLRs associate with coreceptors in a cell- and tissue-specific fashion (39, 88). In mammalian myeloid leukocytes and other effector cells, presentation of LPS to TLR4 involves binding of LPS to LPS-binding protein (LBP), which is present in soluble form in plasma and other fluids, and binding of the LPS-LBP complex to CD 14, a protein that exists in soluble and glycophosphatidylinositol-anchored forms (39, 83, 88). CD14 then facilitates binding of LPS to TLR4. CD 14 presents LPS to other proteins besides TLR4 (97) and also recognizes lipopeptides and peptidoglycans and presents them to TLR2 (98, 99). Stressinduced increases in soluble CD14 in plasma may bind and transfer LPS to plasma

Acute respiratory distress syndrome

224

lipoproteins, reducing monocyte responses (97). Polymorphisms within the CD14 locus have been associated with susceptibility and outcome in septic shock (27). Recognition mechanisms for LPS have been defined in the lungs of experimental animals and in humans subjected to experimental challenge (29, 100–103). LPS-induced responses are diminished in the lungs of C3H/ HEJ subjected to systemic challenge (101). In humans given low doses of LPS there are differences in pulmonary inflammatory responses when the challenge is via the airway vs. the systemic route (29, 102). It is unknown how these responses compare to those that are induced by multiple bacterial products acting simultaneously and/or by bacterial products in the context of chemokines, cytokines, and other inflammatory mediators in clinical sepsis (see below). The patterns of TLRs and their cellular localizations in the normal human lung are being characterized, but this process is far from complete (92, 104). Furthermore, it is likely that the pattern and level of endogenous expression of TLRs changes dramatically in the inflamed or injured lung, and there is evidence that components of the LBP/CD14 limb of the LPS recognition system are also altered (100, 105– 107). In addition to this molecular recognition system, alveolar surfactant lipids and proteins also bind LPS and modify its activity, providing a lung-specific variable in microbial recognition that can be altered in inflammation and injury (100). C. TLRs in Gene Expression Pathways and Synthesis of Cytokines and Inflammatory Mediators TLRs signal to intracellular pathways that induce altered gene expression, leading to changes in cellular phenotype and synthesis of new proteins that modify the inflammatory milieu. The signaling cascade includes elements shared with those activated via the interleukin-1 (IL-1) receptor family, and IL-1 receptors and TLRs share a consensus motif in the cytoplasm tail (the toll/interleukin-1 receptor motif, TIR) that is critical for signal transduction and triggers multiple additional intracellular events (1, 83, 87, 88). Molecules that mediate these intracellular signaling connections continue to be identified; some transduce signals in both innate and adaptive immune responses (108, 109). The signaling cascade triggered by engagement of TLR4 by LPS activates the conserved NF-κB pathway for transcriptional regulation of inflammatory and immune genes (88, 99). Similarly, TLR2 signals to the NF-κB pathway (88, 99). In human endothelial cells, the TLR2 and TLR4 genes are themselves induced by LPS in an NFκB-dependent fashion (110) and messenger RNAs controlled by NF-κB are expressed in response to engagement of both TLR4 and TLR2 (81). Thus, signaling to NF-κB by TLRs is not limited to macrophages and other myeloid cells, although it is particularly robust in these leukocytes (61). Previously it has been shown that the β2 integrin heterodimers αMβ2 (CD11b/CD 18) and αXβ2 (CD11c/CD18) bind LPS and signal when engaged (83, 111). In human monocytic cells engagement of a β2 integrin induces signaling via components of the TLR pathway (112), suggesting that amplification and signal integration (75) occur as a result of coincident adhesion molecule and TLR engagement when leukocytes encounter bacterial products in the inflammatory milieu (113, 114). In addition to NF-κB, other transcriptional pathways are likely activated

Pathogenesis of sepsis and septic-induced lung injury

225

directly or indirectly by TLRs as well (81, 88, 115, 116; T.M. McIntyre et al. unpublished experiments). Bacterial signaling of monocytes, macrophages, and other innate immune effector cells induces synthesis of gene products with cytokine or chemokine activities that often are transcriptionally regulated by NF-κB. As an example, transcripts for chemokines and cytokines are among the major mRNAs that are induced when monocytes are stimulated with LPS or gram-positive bacterial components (117, 118). Major gene products that are induced by TLR engagement include tumor necrosis factor a (TNFα), the a and β isoforms of interleukin-1 (IL-1), IL-6, IL-8, and others (1, 90). Cytokine synthesis was critical in establishing the biological consequences and cellular basis for signaling via TLR4 (1). Furthermore, a multitude of observations indicate that TNF, IL-1, chemokines, and other pro-inflammatory cytokines are generated in clinical sepsis and septic shock and have central actions in animal models of sepsis; in addition, anti-inflammatory cytokines are also synthesized, and cytokine “imbalance” has been proposed as a central feature of dysregulated inflammation in sepsis (17, 21, 38). Similarly, TNFα, IL-1β, and other proinflammatory cytokines and counter-balancing chemokines are synthesized in septic ALI and ARDS as indicated by assays of clinical samples, and cytokine and chemokine imbalance may be a critical feature of lung damage in these conditions (14, 119–121). Nevertheless, it is unclear if TNF, IL-1, and other cytokines are requisite mediators in human sepsis or its end-organ complications based on clinical and experimental trials in which antibodies and receptor antagonists that specifically block single members of the cytokine network were administered at a particular interval in the continuum of these syndromes (21, 38). There are a number of potential reasons for failure of such clinical trials (17, 24, 122), including the fact that experimental studies demonstrate protective roles for inflammatory cytokines in systemic infection in addition to injurious actions (123, 124). Thus, the outcomes of the clinical trials done to date do not eliminate the importance of cytokines or chemokines and induction of their genes in the pathogenesis of sepsis. Furthermore, association studies of polymorphisms in the TNFα and IL-1 signaling systems suggest that they may contribute to the fabric of genetic predisposition to sepsis and its complications (27). However, additional actions of TLRs besides cytokine induction, such as modulation of apoptosis of specific cell types (88, 89, 125–128), may emerge as equally important biological variables in experimental and clinical sepsis. D. G-Protein-Coupled Receptors in Sepsis: Complement Pathway Signaling, the Platelet-Activating Factor Receptor Signaling System, and Others In addition to TLRs and their modifiers, other signaling systems are involved in complex inflammatory responses to microbial challenge. While each of these cannot be comprehensively reviewed in this chapter, signaling mechanisms involving G-proteincoupled (GPC) receptors have evolved for many biological tasks including responses to infection. They are important to consider and are attractive therapeutic targets. Cells of the innate immune system constitutively express members of the seven membrane-spanning (“serpentine”) GPC receptor class on their surfaces and use them to respond to signaling molecules generated by other cells in the inflammatory milieu (129–

Acute respiratory distress syndrome

226

131). As an example, chemokine receptors are of the GPC class; additional GPC receptors on PMNs and monocytes will be considered below. GPC receptors are critical for activation and de-activation of leukocyte integrins and other events required for their transmigration from the blood to extravascular sites of infection or tissue injury (60, 75, 130–132). Outside-in signaling via GPC receptors also induces multiple additional functional responses and changes in phenotype of myeloid leukocytes that are critical in regulated and dysregulated inflammation, influence mononuclear cell responses and can link inflammation and hemostasis (see below). In addition, GPC mediate diverse responses of other cells in the lung and systemic tissues, including angiogenesis, epithelial activation, and transactivation of growth factor-coupled pathways (129, 130, 133, 134). LPS activates the complement cascade, resulting in enzymatic production of peptide mediators (4). C5a generated by complement activation is recognized by the C5 receptor 1, a GPC receptor present on neutrophils and monocytes that has increased expression in a variety of tissues in experimental sepsis (129, 135). C5a-mediated intravascular aggregation and activation of PMNs was one of the first mechanisms proposed for pulmonary vascular damage and consequent ALI as complications of systemic illness, and this response has been specifically reported in sepsis (4, 14, 136) (see also below). In addition, β2 integrins were increased on blood neutrophils in a C5a-dependent manner in an experimental model of sepsis (70). However, levels of neither C5a nor the downstream terminal attack complex C5b-9 predict ARDS; furthermore, current evidence indicates that complement activation alone is insufficient to induce ALI or ARDS in humans with sepsis, although blocking C5a protected against lung injury in a preclinical primate model (4, 14). It remains possible that activation of PMNs or monocytes via the C5a receptor or other complement-signaling mechanisms induces amplifying responses in subsets of patients with sepsis and sepsis-induced ALI. In addition to the C5a receptor, neutrophils and monocytes bear two GPC receptors, FPR1 and FPRL1, that recognize short formylated bacterial peptides and trigger β2 integrin activation, directed migration, granular enzyme release, oxygen radical generation, and other acute responses (129, 131). Signals delivered by formylated peptide receptors can also induce expression of new gene products by PMNs or monocytes in combination with outside-in signals delivered by adhesion molecule engagement or by “priming” with LPS (137, 138) (T. Mahoney et al., unpublished experiments). The PAF receptor, a third member of the GPC family (129), is constitutively expressed on human neutrophils and monocytes and transmits outside-in signals that induce multiple functional responses when it is specifically ligated by PAF or by structurally-related PAF-like molecules (PAF-like lipids, oxidatively modified phospholipids) that are generated by oxidant attack on membrane phospholipids in pathological conditions (139– 141). The PAF receptor is also expressed on human platelets, where its engagement triggers aggregation, granule exocytosis, and other responses. Expression on both myeloid leukocytes and platelets is an unusual pattern for a GPC receptor that confers the ability to rapidly induce both acute inflammatory and thrombotic events when its ligands are generated, linking these two systems (142). In addition to displaying the PAF receptor and responding to PAF, myeloid leukocytes and platelets synthesize it when stimulated. Similarly, human endothelial cells rapidly synthesize PAF in response to stimulation with a variety of thrombotic and pro-

Pathogenesis of sepsis and septic-induced lung injury

227

inflammatory agonists. Remarkably, the first enzyme in the two-step synthetic pathway, PAF acetyltransferase (139, 141), has not yet been cloned, although its activity has been partially characterized. A group of precise mechanisms that regulate cellular activation via the PAF receptor signaling system has evolved, including tightly controlled enzymatic synthesis of PAF, distribution of the PAF receptor on specific target cells, receptor desensitization, spatially regulated signaling by PAF localized to the surfaces of endothelial cells and certain other cells that produce it (juxtacrine signaling), and rapid degradation of PAF by a group of enzymes termed PAF acetylhydrolases. Breakdown or dysregulation of one or more of these control mechanisms and/or unregulated synthesis of PAF or nonenzymatic generation of PAF-like lipids are potential mech

Figure 3 Signaling via thrombin and PAF receptors induces diverse cellular responses linking inflammation and

Acute respiratory distress syndrome

228

thrombosis, (a) Thrombin induces display of P-selectin and PAF by engaging PAR1 on human endothelial cells; signaling via the PAF receptor then mediates juxtacrine activation of adherent PMNs. The activated PMNs undergo functional changes that can contribute to inflammation and thrombosis in sepsis and other pathologic syndromes, (b) Thrombin activates human platelets, resulting in platelet-platelet and platelet-monocyte aggregates. Thrombin triggers activation and aggregation of human platelets by binding to PAR1 and PAR4. Thrombin-induced activation of platelets also mediates their adhesion to monocytes. Outside-in signaling in cellular aggregates then induces expression of new genes, including monocyte genes that code for chemokines and cytokines, and other functional responses. PAF also induces platelet activation, cellular aggregation, and signaling. Thrombinand PAF-stimulated platelets interact with PMNs in a similar fashion, (c) Thrombin- and PAF-stimulated platelets release IL-1β, which signals new gene expression in endothelial cells. The mechanism involves signaldependent translation of constitutive IL-1β. transcripts in activated platelets. Preformed C-X-C and CC chemokines are also released from aggregated degranulating platelets and can directly activate PMNs and monocytes.

Pathogenesis of sepsis and septic-induced lung injury

229

anisms for pathological inflammation in a number of syndromes and diseases (130, 139– 143). Evidence that the PAF receptor signaling system is involved in sepsis and its complications was recently reviewed (142–145). Activation of human PMNs, monocytes, and platelets in in vitro experimental models induces responses that mediate inflammatory and thrombotic events that are relevant to the pathophysiology of sepsis, including aggregation, adhesion, priming for enhanced granular secretion, oxygen radical generation, and synthesis of chemokines and cytokines by transcriptional and posttranscriptional mechanisms (142, 146) (Fig. 3). Outside-in signals triggered by PAF and soluble forms of P-selectin or E-selectin, which are reported to circulate in humans with sepsis (84, 147), induce enhanced neutrophil responses (148, 149). Innate immune cells from a number of other species also bear the PAF receptor and are activated or undergo phenotypic changes when it is engaged (144, 146). As an example, in a recent study production of TNFα and other mediators by rabbit alveolar macrophages was reported to be dependent on endogenous PAF (150), consistent with the ability of signaling via the PAF receptor to modify TNFα production by human mononuclear leukocytes under some conditions (151). The presence of the PAF signaling system in multiple animal species has been utilized to develop disease models, including several relevant to sepsis. Overexpression of the PAF receptor increased mortality in response to LPS challenge in a mouse strain (152). In a second study, LPS, TNFα, and PAF itself increased expression of the endogenous PAF receptor mRNA in rats with intestinal injury (153). Observations in other animal models are also consistent with dysregulated signaling via the PAF receptor as a mechanism of septic ALI (144). For example, in rats sequential systemic administration of PAF and LPS in doses that alone did not cause lung injury resulted in intravascular neutrophil aggregation and parenchymal accumulation of leukocytes, platelet-fibrin deposition in lung vessels, pulmonary edema, and systemic hypotension, hemoconcentration, and elevated levels of TNFα and thromboxane B2 (154). The “priming” effect of PAF administration on LPS-induced lung injury was inhibited by blockade of the PAF receptor with a competitive antagonist. Subsequent studies aimed at dissecting molecular mechanisms of ALI and mortality in this model identified priming for enhanced production of TNFα by macrophages as a critical feature and interplay of PAF receptor signaling, TNFα signaling, and the complement cascade (144). A second model utilized intraperitoneal LPS administration to induce endogenous PAF synthesis and identified signaling mediated by the PAF receptor as a critical variable in increased pulmonary vascular permeability, oxidative stress, and mortality. In this model there were again separate but amplifying roles for TNFα and PAF as proximal mediators in the inflammatory lung injury (155, 156). More recent animal models support the concept of coordinate cellular activation via the PAF receptor signaling system together with other signaling events in LPS-induced lung injury and in lung responses to interventions or complications that may accompany sepsis and shock, including transfusion and inschemia/ reperfusion (157–159). Clinical observations in patients with sepsis parallel many of the in vitro and in vivo experimental findings. There was evidence for receptor occupancy on circulating platelets from subjects with sepsis in one study, and activity consistent with PAF or oxidatively modified PAF-like lipids has been found in blood and/or tissue in some, but

Acute respiratory distress syndrome

230

not all, samples from patients with sepsis that have been examined (142, 144). A complicating feature of analysis of plasma or bronchoalveolar lavage (BAL) samples is that PAF can act in a juxtacrine fashion at cell surfaces (130, 139, 141) (see above). Therefore, levels measured in solution may have little relationship to its biological effects in mediating cell-cell interactions and activating target cells (142). This ability to mediate intercellular signaling and target cell activation without being released into solution is also shared by TNFα, IL-1β, and certain other factors (130). In addition to difficulties posed by cellular retention, there are a number of technically challenging aspects of measuring PAF and distinguishing it from PAF-like lipids (140) that can complicate assays of clinical samples. Finally, PAF acetylhydrolases in the sample can rapidly degrade both PAF and PAF-like lipids, establishing yet another variable. Clinical trials of PAF receptor antagonists in sepsis have suggested positive effects, although a regimen with clear therapeutic efficacy has not been established (160–163). This may be due to heterogeneity of the patient populations and timing of administration of the drug, as with other interventional studies of sepsis (24, 122), and also to pharmacological characteristics of some of the receptor antagonists that have been utilized to date. These pharmacological features include relatively low efficiency as competitors and short half-lives (142). For example, it takes an approximately 2000-fold excess of the competitive antagonist most extensively studied in human sepsis trials, BN52021 (160, 161), to inhibit the binding of PAF to human neutrophils by 50% (146). These features of receptor antagonists, together with high effective concentrations of PAF at cell surfaces and/or high local concentrations of oxidatively-modified PAF-like lipids, may make an effective competitive strategy difficult to achieve. A recent preliminary trial utilized a noncompetitive strategy by employing recombinant plasma PAF acetylhydrolase, which terminates signaling “upstream” from the PAF receptor by hydrolyzing PAF and PAF-like lipids (139, 141, 142). The study indicated beneficial effects and, in addition, further supports a role for the PAF signaling system in human sepsis and in sepsis-induced ARDS (164). The activity of endogenous plasma PAF acetylhydrolase is decreased in some subjects with sepsis or SIRS (165–168; F.Bozza et al., unpublished data), establishing the potential for pathological imbalance resulting from enhanced generation of agonists recognized by the PAF receptor and depression of a key enzyme that degrades them and terminates their signals (142). The mechanisms of decreased PAF acetylhydrolase activity appear to include LPS-mediated inhibition of its synthesis by macrophages and oxidative inactivation of its activity in the inflammatory milieu (142). These examples illustrate that GPC signaling systems are mechanistically involved in sepsis. The roles of protease-activated receptors (PARs), another important group of GPC receptors that, like the PAF receptor, mediate both inflammatory and thrombotic signaling (169), are outlined below. E. The Complex Biology of Inflammation in Sepsis: New Molecular Players Identification of the TLR family established new biology for sepsis beyond previously known mechanisms (4, 5), and the number of TLRs and the diverse nature of the microbial ligands for these receptors further document its complexity (39, 61). However,

Pathogenesis of sepsis and septic-induced lung injury

231

additional factors that modify the molecular physiology of sepsis, some discovered even more recently than TLRs, highlight this complexity and the redundant and pleiotrophic pathways of the innate immune system (59). These newly discovered mechanisms also underscore that our understanding of the molecular biology of sepsis is far from complete. As an example, NOD2 was recently identified as an intracellular receptor for bacterial components that is restricted to expression by monocytes. NOD2 mutations are associated with dysregulated inflammation and susceptibility to Crohn’s disease (170, 171), but contributions of the NOD2 system to clinical sepsis remains unexplored. As a second example, a factor termed triggering receptor expressed on myeloid cells (TREM1) is selectively displayed on human neutrophils and some monocytes and is upregulated when they are incubated with bacteria, LPS, or lipoteichoic acid (172). TREM-1 was also upregulated on peritoneal neutrophils from mice challenged with LPS (173). Crosslinking of TREM-1 together with administration of LPS enhanced production of TNFα and IL-1β by monocytes. A TREM-1 fusion protein reduced cytokine levels and protected mice from death when administered in sepsis models that included live E. coli and cecal ligation and puncture, in addition to LPS challenge (173). Although the ligands recognized by TREM-1 and the mechanism by which it augments responses to LPS are not yet identified, these observations indicate that it contributes to alterations in myeloid cell function in the early phases of experimental sepsis and suggest that TREM-1 may signal to pathways that interface with TLR-induced gene expression (59). New molecular players are also implicated in later phases of the sepsis syndromes. High mobility group 1 (HMGB1) was recently implicated as a mediator of delayed systemic inflammation and lethality in endotoxemia (174, 175). HMGB1, first identified as an intracellular transcription factor, is released from monocytes and macrophages activated by LPS and is present in the plasma of mice subjected to LPS challenge and in samples from critically ill patients with sepsis (175). HMGB1 is also released by necrotic, but not apoptotic, cells and promotes inflammatory responses by neighboring cells (176). In both in vitro and in vivo experiments, the release of HMGB1 from myeloid leukocytes occurs later than the early peaks of TNFα and IL-1β, suggesting that it is a late mediator of the systemic response to bacterial challenge (175). HMGB1 triggers activation of human mononuclear phagocytes and consequent cytokine synthesis (175). In addition, HMGB1 protein induces lung inflammation and cytokine synthesis when given intratracheally to C3H/HeJ mice (177). Delayed administration of a polyclonal antibody against HMGB1 to wild-type mice challenged with systemic LPS reduced mortality, and administration of HMGB 1 itself was lethal (174). These observations suggest that HMGB1 acts as a mediator of intercellular signaling that influences mortality in murine models of sepsis later than the early effects of TNFα and IL-1β and that it may have similar actions in humans with sepsis and its complications (175, 175a). Macrophage inhibitory factor (MIF) was identified as a cytokine released from the pituitary that potentiates mortality in experimental endotoxemia (178). It is also expressed constitutively by monocytes, macrophages, lymphocyte subsets, and epithelial cells and is further induced by LPS and proinflammatory cytokines (179). More recently, targeted disruption of the MIF gene in mice was reported to confer resistance to LPS, and MIF−/− mice had diminished neutrophil accumulation in the lungs when challenged with Pseudomonas aeuroginosa (180). In addition, murine macrophages rendered deficient in MIF by transfection with an antisense construct were hyporesponsive to LPS and several

Acute respiratory distress syndrome

232

gram-negative bacteria and had reduced NF-κB translocation and release of TNFα when treated with these stimuli; interestingly, this phenotype was attributed to decreased expression of TLR4 (181). MIF is an antagonist of the anti-inflammatory and immunosuppressive effects of glucocorticoids and thus may enhance inflammatory injury in experimental sepsis indirectly by this mechanism as well as by direct proinflammatory actions (179). The participation of MIF in clinical sepsis and sepsis-induced acute lung injury in patients has not yet been dissected, but it has been found in lavage fluid from patients with ARDS (182). F. The Interface Between Inflammation, Coagulation, and Thrombosis in Sepsis and Its Complications Sepsis is perhaps the most dramatic and persuasive example of the intimate relationship between the innate immune and hemostatic systems. Evolutionary and phylogenetic perspectives trace this association (183, 184). In Limulus and other invertebrates, serial activation of serine protease zymogens generates a coagulum that walls off invading microbes. This invertebrate proteolytic system, which has similarity to mammalian clotting and complement cascades, can be triggered by LPS and carbohydrate pattern motifs that are displayed on microbial cell walls (183), suggesting that the primitive coagulation response was part of a common host defense mechanism that subsequently evolved for specialized hemostatic and innate immune functions but with continued molecular links between the two parallel systems. There is substantial evidence from experimental and clinical studies that dysregulated coagulation and thrombosis influence the natural history of sepsis and its end-organ complications. These observations spawned several trials of novel anticoagulants (185). The molecular basis for coagulation, including the generation of thrombin and conversion of fibrinogen to fibrin, have been reviewed recently (186–189) and will not be recapitulated in detail here. Current evidence indicates that the tissue factor pathway is the primary mechanism for thrombin generation in patients with sepsis (185, 190). Tissue factor is synthesized by human monocytes and macrophages in response to LPS, proinflammatory cytokines, and other stimuli (186, 191) and is deposited on cell surfaces or released into the flowing blood in microparticles that may be shed by monocytes and neutrophils (185, 192). Thrombin generated by sequential protease-dependent steps in the tissue factor pathway cleaves fibrinogen to yield fibrin, the ultimate product in coagulation and the central component of the fibrin clot (187). Together with its cleavage of fibrinogen, thrombin activates platelets and induces their aggregation and interaction with leukocytes and, in addition, induces inflammatory responses by endothelial cells and certain other cell types that express the appropriate PARs (169) (Fig. 3). Thus, generation of thrombin and signaling via PARs on target cells are potent mechanisms that link the hemostatic and inflammatory systems. Several examples illustrate this point (Fig. 3). Activation of human endothelial cells by thrombin induces translocation of P-selectin to their surfaces and coincident synthesis of PAF, which results in endothelial cell-dependent adhesion and local activation of PMNs (75, 193). A similar mechanism is induced when immobilized platelets are stimulated to display P-selectin and PAF on their surfaces (142). In addition, P-selectin displayed on the surfaces of platelets activated by thrombin binds to P-selectin glycoprotein ligand 1

Pathogenesis of sepsis and septic-induced lung injury

233

(PSGL-1) on monocytes, resulting in transcellular signaling to transcriptional and translational control pathways that induces new expression of gene products including TNFα, IL-8, MCP-1, and other factors (151, 194, 195). As a third example, human platelets stimulated by thrombin synthesize and release IL-1 β, which can then activate new gene expression in endothelial cells, providing a novel mechanism with the potential to both link and amplify thrombosis and inflammation (196). In addition to these examples, thrombin induces other inflammatory response by both direct and indirect actions on cells of the innate immune and vascular systems (169, 191, 197). Furthermore, additional proteases and molecular complexes generated by the coagulation cascade have proinflammatory properties. The mechanisms are in some cases less well characterized than those activated by thrombin, but are nevertheless relevant to sepsis and other pathological conditions (185, 198). At the same time that the coagulation cascade is triggered and cellular thrombotic systems are activated, endogenous anticoagulant proteins are depleted and the fibrinolytic response is attenuated in dysregulated hemostasis induced by sepsis and tissue injury (185, 190, 198). The major endogenous anticoagulant systems in humans are regulated by protein C, antithrombin, and tissue factor pathway inhibitor (TFPI) (185, 187, 197). Recent observations implicate disruption of the protein C/protein S/thrombomodulin pathway as a pivotal event in patients with sepsis (185, 190, 199). This is likely because it has a major role in preventing thrombosis in the microcirculation and also because it has anti-inflammatory effects (see below), so that its actions are not limited to controlling thrombin formation and modulating fibrinolysis (190, 197). The ability to interrupt proinflammatory as well as procoagulant pathways may be a major feature that contributes to the efficacy of recombinant activated protein C in clinical sepsis (199) (see below). Protein C circulates as an inactive precursor that is rapidly converted to a serine protease, activated protein C (APC), by a complex of thrombin and thrombomodulin (TM). TM is present on the plasma membranes of endothelial cells under basal conditions and binds thrombin when it is generated. This inhibits the procoagulant actions of thrombin while facilitating its cleavage of protein C to APC (188, 189). Protein C activation is also enhanced by the endothelial cell protein C receptor (EPCR) (200). When APC is produced it associates with a plasma factor, protein S, and this complex inactivates factors Va and VIIIa, inhibiting thrombin gener-ation. In addition to providing an endogenous anticoagulant “brake” on thrombin production that is proportional to the thrombotic stimulus under physiological conditions, APC modulates inflammatory responses. Blocking APC or cofactors of the pathway in primates infused with sublethal concentrations of LPS results not only in enhanced thrombosis and mortality but also in enhanced inflammation, as indicated by circulating levels of TNFα, IL-6, and IL-8 (34, 197). APC also reduced blood and tissue levels of TNFα in rodent models (197), including experimental pulmonary vascular injury in rats (201). APC was reported to inhibit LPS-stimulated NF-κB nuclear translocation and other responses in human monocytic cell lines (202, 203). Additional studies in in vitro systems are exploring potential mechanisms by which the APC system blunts inflammatory responses to LPS challenge in in vivo models (190, 191, 204). Plasma levels of protein C and protein S rapidly decrease in septic patients and in subjects with other critical illnesses based on clinical studies (187, 197, 205).

Acute respiratory distress syndrome

234

Furthermore, thrombomodulin is reduced on the human endothelial surface in both in vitro models of LPS challenge and in histological analysis of biopsies from patients with clinical meningococcal sepsis (206, 207), although this has not been seen in all animal models. In parallel, levels of soluble thrombomodulin—which has lower activity than cell surface thrombomodulin—increase in the plasma of patients with sepsis and many other systemic inflammatory diseases (191, 208). Reduction of thrombomodulin on the endothelial surface and shedding into the plasma may dominantly result from its proteolytic cleavage by elastase and other proteases locally released by degranulating neutrophils (191, 208, 209). LPS-stimulated human endothelial cells synthesize several degranulating factors that trigger the release of neutrophil granular proteases, including elastase (79, 210) (Fig. 4). In addition, thrombin-stimulated human endothelial cells “prime” neutrophils for enhanced degranulation by a mechanism involving surface display of P-selectin and PAF and engagement of the PAF receptor (193, 210). Thus, inflammatory signaling of endothelial cells may lead to local neutrophil degranulation and contribute to dysregulated coagulation and subsequent thrombosis by decreasing TM at the endothelial surfaces. Furthermore, neutrophil oxygen radial generation can be both primed and triggered by adhesive interactions with thrombin-stimulated endothelial cells or platelets (211). Oxygen radicals can attack a methionine residue that is important for protein C activation (191, 212), potentially adding another mechanism of dysregulation. Although we emphasize the protein C pathway here, the fibrinolytic system, TFPI, and antithrombin are also critical regulators of the coagulation cascade, and each is altered in sepsis and septic shock. Furthermore,

Figure 4 LPS- and cytokine-activated human endothelial cells synthesize signaling molecules that trigger PMN adhesion, degranulation (“degranulating factors"), and other inflammatory responses. Enzymes and additional factors released from PMN granules contribute to vascular permeability, injury and thrombosis. See text for details.

Pathogenesis of sepsis and septic-induced lung injury

235

each of these regulatory components provides additional points of potential “cross-talk” with the innate immune system (185, 197). In addition to systemic manifestations of sepsis, there is evidence that dysregulated coagulation and thrombosis are involved in septic lung injury (213, 214). Infusions of LPS in experimental animals causes intravascular fibrin deposition in the lungs as well as in other organs involved in sepsis-induced multiple organ failure (213). In a recently reported primate model in which animals were primed with killed E. coli and then challenged 12 hours later with live bacteria, tissue factor generation, pulmonary fibrin deposition, and alveolar neutrophil accumulation occurred and were associated with hypoxemia, decreased lung compliance, lung edema, and pulmonary hypertension and with evidence for concomitant renal failure. These pathological responses were improved by administration of a competitive inhibitor of tissue factor, site-inactivated factor VIIa (FVIIa) and by administration of TFPI (46). Site-inactivated FVIIa also attenuated lung injury and blocked proinflammatory cytokine release in response to intra-tracheal LPS in rats (215). In a rodent “two-hit” model, animals were subjected to hemorrhagic shock, resuscitated, and then challenged with intratracheal LPS to stimulate infection after trauma, a common clinical scenario (43). (It is possible that this model also mimics some of the features of challenge by bacterial products delivered by the airway route, including intubation and airway colonization or infection, etc., after an initial “hit” by systemic septic shock, a situation that also commonly occurs in the intensive care unit.) Hemorrhagic shock potentiated the injury triggered by subsequent administration of low doses of LPS, and the sequential challenge resulted in increased pulmonary tissue factor and plasminogen activator inhibitor expression and a net procoagulant state in the lungs (43). Blocking antibodies and other reagents indicated roles for both TNFα and oxygen radicals generated by alveolar macrophages in the procoagulant response. Each of these observations has relevance to dysregulated coagulation and lung dysfunction in clinical ALI and ARDS (see below) and contributes to the rationale for interrupting the hemostatic cascade in therapeutic strategies (213, 214).

V. Features of Septic ALI and ARDS in Critically III Patients The incidence, natural history, and prognosis of ALI and ARDS precipitated by sepsis were reported in early studies, further described in subsequent clinical series, and have been recently reviewed (3, 7, 11, 12, 216, 217) (also see above). In general, physiological features of ALI and ARDS induced by sepsis have been reported to be similar to those in patients with lung injury caused by other precipitating factors (trauma, aspiration, etc.). However, it is possible that myocardial depression resulting from biochemical sequellae of sepsis and septic shock alters the “physiological phenotype” of individual patients and may complicate the diagnosis of septic-induced ARDS based on the distinction between cardiogenic and noncardiogenic pulmonary edema in some subjects (11, 53). The mortality in ARDS associated with sepsis and septic shock is higher than that in other subsets of ARDS in most, but not all, series, a feature identified in early and more recent clinical reports (3, 12, 49, 216, 218). The mechanisms that account for the lethal nature of septic-induced ARDS remain incompletely defined, although comorbid

Acute respiratory distress syndrome

236

conditions and the pleotrophic molecular events and concurrent systemic injury and multiple organ failure induced by sepsis must contribute. PMNs and monocytes accumulate in the lungs of patients with septic ALI and ARDS. This can be demonstrated using imaging with radiolabeled leukocytes (219–221), as well as by histological analysis and BAL (12, 222) (Fig. 5). Acute accumulation of myeloid leukocytes has similarly been documented in virtually all animal models of sepsis, including those in rodents, ungulates, and primates. Histological evaluation of lung tissue from human subjects with bacteremia or sepsis-induced ARDS has in most cases

Figure 5 Neutrophils and monocytes accumulate in the lungs of patients with septic-induced ARDS. (A) Autologous radio-labeled leukocyte scanning in a patient with gram negative sepsis 24 hours after meeting criteria for ARDS revealed diffuse accumulation of myeloid leukocytes in the lungs (arrows) (220). Biopsy of the lung in patients with septic ALI demonstrates sequestration of myeloid leukocytes in microvessels and accumulation in alveoli (222) and heterotypic leukocyte aggregates in pulmonary arteries and arterioles (B and C). (B) PMNs and monocytes in a heteroytpic cellular aggregate in a small pulmonary artery of a patient dying with ALI secondary to sepsis. Immunocytochemical analysis demonstrated that some of the myeloid

Pathogenesis of sepsis and septic-induced lung injury

237

leukocytes in intravascular aggregates (long arrow) stained positively for IL8, whereas circulating PMNs and monocytes from control subjects do not express this chemokine (137, 138, 227). PMNs and mononuclear leukocytes in alveoli were also positive for IL-8 (short arrow). (From Ref. 226.) (C) PMNs and monocytes in a heterotypic cellular aggregate (arrow) in a small pulmonary arteriole in the lung of a patient dying with ALI precipitated by sepsis. Immunostaining demonstrated expression of IL-8 by myeloid leukocytes in intravascular aggregates, and also by PMNs and monocytes in alveolar spaces. PMNs and monocytes in intravascular aggregates in lung sections from patients with sepsis also stain positively for ENA-78 (227), a second chemokine associated with ALI (not shown). demonstrated intravascular accumulation, or “sequestration,” of PMNs and monocytes in vessels of various sizes (130, 222–226) (Fig. 5). Intravascular cellular aggregates may contribute to the increased pulmonary dead space fraction recently reported in ARDS associated with sepsis and other triggering conditions (218). Intravascular leukocyte aggregation has been thought to be an early pathological event resulting from formation of homotypic and heterotypic aggregates of myeloid leukocytes in the flowing blood in response to circulating inflammatory mediators and to result in endothelial injury because of local release of oxygen radicals, elastase, and other proteases from leukocytes trapped in the lung microcirculation (130, 222). Clinical and experimental studies provide evidence for degranulation and for cytolytic fragmentation of intravascular PMNs that could cause massive local release of granular enzymes (79, 210). Recent observations suggest the possibility that monocytes or neutrophils in mixed cell aggregates in the pulmonary microvasculature may be a source of bloodborne tissue factor that induces and propogates thrombosis (192). In addition, intravascular neutrophils and monocytes in the lungs of septic patients express new gene products, including chemokines (226, 227) (Fig. 5). The relative contributions of circulating factors vs. juxtacrine signaling molecules (chemokines, PAF, etc.) that are locally expressed on the surfaces of pulmonary endotheial cells or entrapped platelets (see below) in causing intravascular

Acute respiratory distress syndrome

238

leukocyte accumulation in sepsis are unknown. There may also be a direct effect of LPS or other endotoxins (228). Early morphometric studies documented expansion of the pulmonary intravascular leukocyte population in humans with sepsis and indicated that there are also increased numbers of platelets (224, 225). Increased intravascular accumulation of platelets may result from platelet-platelet aggregates, local deposition of platelets in platelet-fibrin thrombi, and/or by formation of platelet-monocyte and platelet-PMN aggregates in the systemic blood of septic patients and their deposition in the pulmonary microvasculature. Administration of LPS to human volunteers demonstrated that such platelet-leukocyte aggregates form in the blood (229). Platelet-platelet aggregates also have signaling capacity and can induce endothelial as well as leukocyte gene expression (discussed above). The interaction between platelets, myeloid leukocytes, and endothelial cells in areas of intravascular sequestration appears to be stable over many hours and provides a mechanism for complex intercellular signaling and gene expression (230) via mechanisms that are incompletely characterized. The relative contributions of intravascular, transiting, and intraalveolar leukocytes to alveolar capillary membrane injury in patients with septic ALI and ARDS (224, 231) are not defined. Analysis of histological samples in ongoing studies in our ARDS SCOR indicate that new gene products are expressed by neutrophils and monocytes in each compartment (55, 226, 227) (Fig. 5), in addition to the ability of these leukocytes to release granular enzymes and generate oxygen radicals by innate biochemical mechanisms (see above). While it is possible that PMNs that are most responsive to direct stimulation with LPS are sequestered in the lungs (232), the accumu-lation of PMNs in vessels of other organs and formation of platelet-leukocyte aggregates in blood indicate that they are also activated outside of the pulmonary structures. The route(s) of neutrophil migration to interstitial and alveolar sites in human septic ALI are not completely established, although in mice subjected to inflammatory challenge the pulmonary capillary is the dominant site of PMN emigration (233, 234). Histological observations suggest that venules and arterioles are also routes of myeloid leukocyte trafficking in the inflamed or injured human lung (227; K.Albertine, G.A.Zimmerman, unpublished observations). By whatever route, the accumulation of PMNs in the alveoli of patients with sepsis-induced ALI appears to influence outcome. Neutrophil numbers in BAL samples tended to be higher in the first 2 weeks after onset of septic ALI in subjects that died compared to the subgroup that survived, whereas in trauma patients with ALI the BAL neutrophil counts progressively diminished over time and did not distinguish the patients who died from the survivors (235). This suggests that persistence of the initial neutrophil “alveolitis” contributes to sustained injury and progression rather than resolution of ALI in patients with sepsis and thereby influences mortality. However, this cannot necessarily be generalized to all patients with ALI and ARDS regardless of the inciting condition (14, 235). In addition to neutrophils, monocytes accumulate in the alveolar space and are recovered in BAL samples early in the course of ARDS triggered by sepsis and other causes of inflammatory lung injury; patterns of reduced and sustained monocytic recruitment and differentiation have been described (236). In a series of patients with sepsis-induced ALI an increase in alveolar macrophages in serial lavage samples was associated with survival (235), suggesting that monocyte-to-macrophage differentiation

Pathogenesis of sepsis and septic-induced lung injury

239

may influence the natural history. In a recent study, we found evidence that the gene coding for plasma PAF acetylhydrolase (see above) is induced in alveolar macrophages in patients with ALI compared to control subjects with uninjured lungs, suggesting that this may be one aspect of macrophage expansion that limits inflammatory injury under some conditions (237). However, the factors that alter the inflammatory phenotypes of monocytes and influence their differentiation into macrophages and their expression of specific gene products in pulmonary and systemic compartments of septic patients are largely unknown. The histological features of alveolar-capillary units in patients dying with ARDS caused by sepsis with documented bacteremia are similar to those in other subsets of patients with ALI and ARDS (224). In the acute or “exudative” phase, they include interstitial and alveolar edema, intraalveolar hemorrhage, relative preservation of endothelial morphology and continuity, local destruction of alveolar epithelium with denuded areas covered by fibrin-containing hyaline membranes, and intravascular and extravascular deposition of myeloid leukocytes, platelets, and fibrin as outlined above (224, 225). This is the key constellation of findings in the pathological diagnosis of “diffuse alveolar damage,” the anatomical equivalent of ARDS (231) (see Chap. 5). These features were largely reproduced when primates were infused with LPS or live E. coli bacteria in several early studies (238, 239). While the features of the alveolarcapillary membrane in septic ALI and ARDS and in animal models have been characterized in terms of cellular appearance, we know little about the “molecular phenotypes” of these cells—that is, the patterns of intracellular, surface, and locally released molecules that have been induced and, in parallel, the molecules that have been downregulated. Measurement of soluble factors in BAL and other fluid samples gives an incomplete picture of the molecular phenotypes of inflamed and injured cells. It is likely that we will not completely understand key features of septic ARDS without this information. Histological, genomic, and proteomic studies correlated with in vitro cell models and measurements of soluble factors in BAL are in progress to address this important knowledge gap (55, 226, 227). As with the acute phase, the histological features of the subacute proliferative and fibrotic phases of ARDS caused by documented sepsis appear to be similar to those in subjects with ARDS incited by other triggering conditions (224, 225, 231). Again, we know little about the molecular phenotypes of the lung and inflammatory cells at these later stages or how they compare to those in the earlier acute phase or in subsets of patients with different outcomes (i.e., fibrosis or obliterative arteritis vs. resolution) (231). Pathological fibrin deposition occurs in micro- and macro vessels of the lungs of patients with ALI and ARDS induced by sepsis, as it does in patients with ARDS caused by other inciting conditions (224, 225, 231, 240). This is part of a generalized dysregulation of fibrin generation and fibrinolysis in intravascular and extravascular compartments in acute lung injury and is consistent with findings in preclinical models as outlined above (213, 214, 241). It is unclear if the deranged fibrin deposition in the lungs of septic patients is quantitatively or qualitatively different from that triggered by other mechanisms of injury or to what extent fibrin generation is required for healing and repair of the injured lung (214). It is also not yet clear if therapeutically administered recombinant APC influences deposition of fibrin in the lung and/or inflammatory acute

Acute respiratory distress syndrome

240

injury in sepsis, although its beneficial effects on mortality suggest that it may (199, 214) and preliminary studies indicate that endogenous plasma protein C is decreased in subjects with ALI secondary to sepsis and other triggering conditions (205). A variety of soluble markers have been measured in samples of plasma and alveolar fluid from patients with sepsis, sepsis-induced ALI and ARDS, and ALI and ARDS associated with other triggering conditions (14, 84, 147, 242–245). No single marker or combination of markers has emerged as specific for the etiology of ALI and ARDS or as predictive of occurrence, natural history, or outcome of these syndromes of lung injury. Recently plasma von Willebrand factor, which is released from endothelial cells and platelets and has been examined in a number of previous studies (14), was reported to be elevated in samples from patients with sepsis with two or more organs failing. Together with other findings, the data suggested that the degree of systemic endothelial activation or injury early in the course of ALI is important in its progression (246). This and other studies illustrate the challenges of interpreting single biological markers in complex syndromes. Approaches that may include pattern analysis using proteomic strategies (247) together with genomic interrogation and parallel biological analysis may increase the yield from clinical samples in the future. Such strategies are likely to yield insights regarding critical variables in septic ALI and ARDS that have, to this point, confounded and obscured our understanding of these syndromes (248).

Acknowledgments The authors thank Mary Madsen for preparation of the manuscript and Diana Lim for preparation of the figures. We also thank our colleagues for their invaluable contributions to collaborative work that is cited and for many useful discussions, Steve Opal for sending preprints of unpublished articles, and Jean-Francois Dhainaut for helpful comments. Projects mentioned in this review are supported by the National Institutes of Health (P50 HL50153; HL44525).

References 1. Beutler B. Toll-like receptors: how they work and what they do. Curr Opin Hematol 2002; 9:2– 10. 2. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967; 2:319–323. 3. Kaplan RL, Sahn SA, Petty TL. Incidence and outcome of the respiratory distress syndrome in gram-negative sepsis. Arch Intern Med 1979; 139:867–869. 4. Martin MA, Silverman HJ. Gram-negative sepsis and the adult respiratory distress syndrome. Clin Infect Dis 1992; 14:1213–1228. 5. Welbourn CR, Young Y. Endotoxin, septic shock and acute lung injury: neutrophils, macrophages and inflammatory mediators. Br J Surg 1992; 79:998–1003. 6. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818–824. 7. Hudson LD, Milberg JA, Anardi D, Maunder RJ. Clinical risks for development of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:293–301.

Pathogenesis of sepsis and septic-induced lung injury

241

8. Kollef MH, Schuster DP. The acute respiratory distress syndrome [see comments]. N Engl J Med 1995; 332:27–37. 9. Connelly KG, Repine JE. Markers for predicting the development of acute respiratory distress syndrome. Annu Rev Med 1997; 48:429–445. 10. Hudson LD, Steinberg KP. Epidemiology of acute lung injury and ARDS. Chest 1999; 116:74S-82S. 11. Wheeler AP, Bernard GR. Treating patients with severe sepsis. N Engl J Med 1999; 340:207– 214. 12. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342:1334–1349. 13. Parsons PE, Worthen GS, Moore EE, Tate RM, Henson PM. The association of circulating endotoxin with the development of the adult respiratory distress syndrome. Am Rev Respir Dis 1989; 140:294–301. 14. Pittet JF, Mackersie RC, Martin TR, Matthay MA. Biological markers of acute lung injury: prognostic and pathogenetic significance. Am J Respir Crit Care Med 1997; 155:1187–205. 15. Pepe PE, Potkin RT, Reus DH, Hudson LD, Carrico CJ. Clinical predictors of the adult respiratory distress syndrome. Am J Surg 1982; 144:124–130. 16. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001; 29:1303–1310. 17. Opal S, Huber CE. SEPSIS. Scientific American Medicine 2001 2001; 30:1–17. 18. Zilberberg MD, Epstein SK. Acute lung injury in the medical ICU: comorbid conditions, age, etiology, and hospital outcome. Am J Respir Crit Care Med 1998; 157:1159–1164. 19. Weinacker AB, Vaszar LT. Acute respiratory distress syndrome: physiology and new management strategies. Annu Rev Med 2001; 52:221–237. 20. Rangel-Frausto MS. The epidemiology of bacterial sepsis. Infect Dis Clin North Am 1999; 13:299–312, vii. 21. van der Poll T, van Deventer SJ. Cytokines and anticytokines in the pathogenesis of sepsis. Infect Dis Clin North Am 1999; 13:413–426, ix. 22. Cohen J, Guyatt G, Bernard GR, et al. New strategies for clinical trials in patients with sepsis and septic shock. Crit Care Med 2001; 29:880–886. 23. American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definition for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864–874. 24. Opal SM, Cross AS. Clinical trials for severe sepsis. Past failures, and future hopes. Infect Dis Clin North Am 1999; 13:285–297, vii. 25. Abraham E, Matthay MA, Dinarello CA, et al. Consensus conference definitions for sepsis, septic shock, acute lung injury, and acute respiratory distress syndrome: time for a reevaluation. Crit Care Med 2000; 28:232–235. 26. Stuber F. Human responses to endotoxin: role of the genetic background. In: Brade H, Opal S, Vogel SN, Morrison DC, eds. Endotoxin in Health and Disease. New York: Marcel Dekker, Inc., 1999. 27. Cairou AC, Jean-Daniel; Charpentier, J; Dhainaut, J.F.; Mira, J.P. The era of genomics: impact on sepsis clinical trial design. Crit Care Med 2002; 30:S341–S348. 28. From the bench to the bedside: the future of sepsis research. Executive summary of an American College of Chest Physicians, National Institute of Allergy and Infectious Disease, and National Heart, Lung, and Blood Institute Workshop. Chest 1997; 111:744–753. 29. Blackwell TS, Christman JW. Defining the lung’s response to endotoxin. Am J Respir Crit Care Med 2001; 163:1516–1517. 30. Munford RS, Pugin J. Normal responses to injury prevent systemic inflammation and can be immunosuppressive. Am J Respir Crit Care Med 2001; 163:316–321. 30a. Tracey KJ. The inflammatory reflex. Nature 2002; 420:853–859.

Acute respiratory distress syndrome

242

31. Wiener-Kronish JP, Albertine KH, Matthay MA. Differential responses of the endothelial and epithelial barriers of the lung in sheep to Escherichia coli endotoxin. J Clin Invest 1991; 88:864–875. 32. Gutierrez-Ramos JC, Bluethmann H. Molecules and mechanisms operating in septic shock: lessons from knockout mice. Immunol Today 1997; 18:329–334. 33. Shenkar R, Yum HK, Arcaroli J, Kupfner J, Abraham E. Interactions between CBP, NFkappaB, and CREB in the lungs after hemorrhage and endotoxemia. Am J Physiol Lung Cell Mol Physiol 2001; 281:L418–426. 34. Taylor FB, Jr. Staging of the pathophysiologic responses of the primate microvasculature to Escherichia coli and endotoxin: examination of the elements of the compensated response and their links to the corresponding uncompensated lethal variants. Crit Care Med 2001; 29:S78–89. 35. Welty-Wolf KE, Carraway MS, Huang YC, et al. Antibody to intercellular adhesion molecule 1 (CD54) decreases survival and not lung injury in baboons with sepsis. Am J Respir Crit Care Med 2001; 163:665–673. 36. Tobin MJ. Critical care medicine in AJRCCM 2000. Am J Respir Crit Care Med 2001; 164:1347–1361. 37. Tobin MJ. Critical care medicine in AJRCCM 2001. Am J Respir Crit Care Med 2002; 165:565–583. 38. Freeman BD, Eichacker, PQ, Natanson, C. The role of inflammation in sepsis and septic shock: a meta-analysis of both clinical and preclinical trials of anti-inflammatory therapies. In: Gallin JI, Snyderman R., eds. Inflammation: Basic Principles and Clinical Correlates. Philadelphia: Lippincott Williams & Wilkins, 1999. 39. Ulevitch RJ. New therapeutic targets revealed through investigations of innate immunity. Crit Care Med 2001; 29:S8–12. 40. Villa P, Meazza C, Sironi M, Bianchi M, Ulrich P, Botchkina G, Tracey KJ, Ghezzi P. Protection against lethal polymicrobial sepsis by CNI-1493, an inhibitor of pro-inflammatory cytokine synthesis. J Endotoxin Res 1997; 4: 197–204. 41. Salkowski CA, Detore G, Franks A, Falk MC, Vogel SN. Pulmonary and hepatic gene expression following cecal ligation and puncture: monophosphoryl lipid A prophylaxis attenuates sepsis-induced cytokine and chemokine expression and neutrophil infiltration. Infect Immun 1998; 66:3569–3578. 42. Bosscha K, Nieuwenhuijs VB, Gooszen AW, et al. A standardised and reproducible model of intraabdominal infection and abscess formation in rats. Eur J Surg 2000; 166:963–967. 43. Fan J, Kapus A, Li YH, Rizoli S, Marshall JC, Rotstein OD. Priming for enhanced alveolar fibrin deposition after hemorrhagic shock: role of tumor necrosis factor. Am J Respir Cell Mol Biol 2000; 22:412–421. 44. Chinnaiyan AM, Huber-Lang M, Kumar-Sinha C, et al. Molecular signatures of sepsis: multiorgan gene expression profiles of systemic inflammation. Am J Pathol 2001; 159:1199– 1209. 45. Hollenberg SM, Dumasius A, Easington C, Colilla SA, Neumann A, Parrillo JE. Characterization of a hyperdynamic murine model of resuscitated sepsis using echocardiography. Am J Respir Crit Care Med 2001; 164:891–895. 46. Welty-Wolf KE, Carraway MS, Miller DL, et al. Coagulation blockade prevents sepsis-induced respiratory and renal failure in baboons. Am J Respir Crit Care Med 2001; 164:1988–1996. 47. Kurahashi K, Kajikawa O, Sawa T, et al. Pathogenesis of septic shock in Pseudomonas aeruginosa pneumonia. J Clin Invest 1999; 104:743–750. 48. Coopersmith CM, Stromberg PE, Dunne WM, et al. Inhibition of intestinal epithelial apoptosis and survival in a murine model of pneumonia-induced sepsis. Jama 2002; 287:1716–1721. 49. Brun-Buisson C, Doyon F, Carlet J, et al. Incidence, risk factors, and outcome of severe sepsis and septic shock in adults. A multicenter prospective study in intensive care units. French ICU Group for Severe Sepsis. JAMA 1995; 274: 968–974.

Pathogenesis of sepsis and septic-induced lung injury

243

50. Beutler B, Poltorak A. Sepsis and evolution of the innate immune response. Crit Care Med 2001; 29:S2–6; discussion S6–7. 51. Esmon CT. Sepsis. A myriad of responses. Lancet 2001; 358(suppl):S61. 52. Hack CE, Zeerleder S. The endothelium in sepsis: source of and a target for inflammation. Crit Care Med 2001; 29:S21–27. 52a. Cohen J. The immunopathogenesis of sepsis. Nature 2002; 420:885–891. 53. Parrillo JE. Pathogenetic mechanisms of septic shock. N Engl J Med 1993; 328:1471–1477. 54. Pleiner J, Heere-Ress E, Langenberger H, et al. Adrenoceptor hyporeactivity is responsible for Escherichia coli endotoxin-induced acute vascular dysfunction in humans. Arterioscler Thromb Vasc Biol 2002; 22:95–100. 55. Maloney CG, Kutchera WA, Albertine KH, McIntyre TM, Prescott SM, Zimmerman GA. Inflammatory agonists induce cyclooxygenase type 2 expression by human neutrophils. J Immunol 1998; 160:1402–1410. 56. Eiserich JP, Baldus S, Brennan ML, et al. Myeloperoxidase, a leukocyte-derived vascular NO oxidase. Science 2002; 296:2391–2394. 57. Rocca B, Secchiero P, Ciabattoni G, et al. Cyclooxygenase-2 expression is induced during human megakaryopoiesis and characterizes newly formed platelets. Proc Natl Acad Sci USA 2002; 99:7634–7639. 58. Michalek SM, Moore RN, McGhee JR, Rosenstreich DL, Mergenhagen SE. The primary role of lymphoreticular cells in the mediation of host responses to bacterial endotoxin. J Infect Dis 1980; 141:55–63. 59. Nathan C, Ding A. TREM-1: a new regulator of innate immunity in sepsis syndrome. Nat Med 2001; 7:530–532. 60. Bunting M, Harris ES, McIntyre TM, Prescott SM, Zimmerman GA. Leukocyte adhesion deficiency syndromes: adhesion and tethering defects involving beta 2 integrins and selectin ligands. Curr Opin Hematol 2002; 9:30–35. 61. Aderem A. Role of Toll-like receptors in inflammatory response in macrophages. Crit Care Med 2001; 29:S16–18. 62. Levin BR, Antia R. Why we don’t get sick: the within-host population dynamics of bacterial infections. Science 2001; 292:1112–1115. 63. Gill EA, McIntyre TM, Prescott SM, Zimmerman GA. Mechanisms of vascular injury in the pathogenesis of infectious disease. Curr Opin Infect Dis 1992; 5:381–388. 64. Parent C, Eichacker PQ. Neutrophil and endothelial cell interactions in sepsis. The role of adhesion molecules. Infect Dis Clin North Am 1999; 13:427–447, x. 65. Gautam N, Olofsson AM, Herwald H, et al. Heparin-binding protein (HBP/ CAP37): a missing link in neutrophil-evoked alteration of vascular permeability. Nat Med 2001; 7:1123–1127. 66. Ibbotson GC, Doig C, Kaur J, et al. Functional alpha4-integrin: a newly identified pathway of neutrophil recruitment in critically ill septic patients. Nat Med 2001; 7:465–470. 67. Ley K. Plugging the leaks. Nat Med 2001; 7:1105–1106. 68. Docke WD, Randow F, Syrbe U, et al. Monocyte deactivation in septic patients: restoration by IFN-gamma treatment. Nat Med 1997; 3:678–681. 69. Adib-Conquy M, Adrie C, Moine P, et al. NF-kappaB expression in mono-nuclear cells of patients with sepsis resembles that observed in lipopolysaccharide tolerance. Am J Respir Crit Care Med 2000; 162:1877–1883. 70. Guo RF, Riedemann NC, Laudes IJ, et al. Altered neutrophil trafficking during sepsis. J Immunol 2002; 169:307–314. 71. Bohnsack JF, Akiyama SK, Damsky CH, Knape WA, Zimmerman GA. Human neutrophil adherence to laminin in vitro. Evidence for a distinct neutrophil integrin receptor for laminin. J Exp Med 1990; 171:1221–1237. 72. Wang L, Kamath A, Das H, Das H, Li L, Bukowski JF. Antibacterial effect of human V gamma 2V delta 2 T cells in vivo. J Clin Invest 2001; 108:1349–1357.

Acute respiratory distress syndrome

244

73. Hotchkiss RS, Tinsley KW, Swanson PE, et al. Depletion of dendritic cells, but not macrophages, in patients with sepsis. J Immunol 2002; 168:2493–2500. 74. Oberholzer A, Oberholzer C, Bahjat KS, et al. Increased survival in sepsis by in vivo adenovirus-induced expression of IL-10 in dendritic cells. J Immunol 2002; 168:3412–3418. 74a. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Eng J Med 2003; 348:138–150. 75. Zimmerman GA, McIntyre TM, Prescott SM. Adhesion and signaling in vascular cell-cell interactions. J Clin Invest 1996; 98:1699–1702. 76. Zimmerman GA, Albertine KH, Carveth HJ, et al. Endothelial activation in ARDS. Chest 1999; 116:18S-24S. 76a. McIntyre TM, Prescott SM, Weyrich AS, Zimmerman GA. Cell-cell interactions: leukocyteendothelial interactions. Curr Opin Hematol 2003; 10:150–158. 77. Yipp BG, Andonegui G, Howlett CJ, et al. Profound differences in leukocyte-endothelial cell responses to lipopolysaccharide versus lipoteichoic acid. J Immunol 2002; 168:4650–4658. 78. Mutunga M, Fulton B, Bullock R, et al. Circulating endothelial cells in patients with septic shock. Am J Respir Crit Care Med 2001; 163:195–200. 79. Gill EA, Imaizumi T, Carveth H, et al. Bacterial lipopolysaccharide induces endothelial cells to synthesize a degranulating factor for neutrophils. Faseb J 1998; 12:673–684. 80. Imaizumi T, Aratani S, Nakajima T, et al. Retinoic acid-inducible gene-I is induced in endothelial cells by LPS and regulates expression of COX-2. Biochem Biophys Res Commun 2002; 292:274–279. 81. Neilsen PO, Zimmerman GA, McIntyre TM. Escherichia coli Braun lipoprotein induces a lipopolysaccharide-like endotoxic response from primary human endothelial cells. J Immunol 2001; 167:5231–5239. 82. Zhao B, Bowden RA, Stavchansky SA, Bowman PD. Human endothelial cell response to gramnegative lipopolysaccharide assessed with cDNA micro-arrays. Am J Physiol Cell Physiol 2001; 281:C1587–1595. 83. Henneke P, Golenbock DT. Innate immune recognition of lipopolysaccharide by endothelial cells. Crit Care Med 2002; 30:S207–213. 84. Leone M, Boutiere B, Camoin-Jau L, et al. Systemic endothelial activation is greater in septic than in traumatic-hemorrhagic shock but does not correlate with endothelial activation in skin biopsies. Crit Care Med 2002; 30:808–814. 85. Aderem A, Ulevitch RJ. Toll-like receptors in the induction of the innate immune response. Nature 2000; 406:782–787. 86. Medzhitov R, Janeway C, Jr. Innate immunity. N Engl J Med 2000; 343:338–344. 87. O’Neill LA, Dinarello CA. The IL-1 receptor/toll-like receptor superfamily: crucial receptors for inflammation and host defense. Immunol Today 2000; 21:206–209. 88. Imler JL, Hoffman JA. Toll receptors in innate immunity. Trends Cell Biol 2001; 11:304–311. 89. Medzhitov R, Janeway CA, Jr. Decoding the patterns of self and nonself by the innate immune system. Science 2002; 296:298–300. 90. Medzhitov R, Preston-Hurlburt P, Janeway CA, Jr. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 1997; 388:394–397. 91. Rock FL, Hardiman G, Timans JC, Kastelein RA, Bazan JF. A family of human receptors structurally related to Drosophila Toll. Proc Natl Acad Sci USA 1998; 95:588–593. 92. Arbour NC, Lorenz E, Schutte BC, et al. TLR4 mutations are associated with endotoxin hyporesponsiveness in humans. Nat Genet 2000; 25:187–191. 93. Poltorak A, He X, Smirnova I, et al. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 1998; 282:2085–2088. 94. Lee HK, Lee J, Tobias PS. Two lipoproteins extracted from Escherichia coli K-12 LCD25 lipopolysaccharide are the major components responsible for Toll-like receptor 2-mediated signaling. J Immunol 2002; 168:4012–4017.

Pathogenesis of sepsis and septic-induced lung injury

245

95. Ozinsky A, Underbill DM, Fontenot JD, et al. The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between toll-like receptors. Proc Natl Acad Sci USA 2000; 97:13766–13771. 96. Zarember KA, Godowski PJ. Tissue expression of human Toll-like receptors and differential regulation of Toll-like receptor mRNAs in leukocytes in response to microbes, their products, and cytokines. J Immunol 2002; 168:554–561. 97. Kitchens RL, Thompson PA, Viriyakosol S, O’Keefe GE, Munford RS. Plasma CD 14 decreases monocyte responses to LPS by transferring cell-bound LPS to plasma lipoproteins. J Clin Invest 2001; 108:485–493. 98. Fenton MJ, Golenbock DT. LPS-binding proteins and receptors. J Leukoc Biol 1998; 64:25–32. 99. Zhang G, Ghosh S. Toll-like receptor-mediated NF-kappaB activation: a phylogenetically conserved paradigm in innate immunity. J Clin Invest 2001; 107:13–19. 100. Martin TR. Recognition of bacterial endotoxin in the lungs. Am J Respir Cell Mol Biol 2000; 23:128–132. 101. Held HD, Boettcher S, Hamann L, Uhlig S. Ventilation-induced chemokine and cytokine release is associated with activation of nuclear factor-kappaB and is blocked by steroids. Am J Respir Crit Care Med 2001; 163: 711–716. 102. O’Grady NP, Preas HL, Pugin J, et al. Local inflammatory responses following bronchial endotoxin instillation in humans. Am J Respir Crit Care Med 2001; 163:1591–1598. 103. Koay MA, Gao X, Washington MK, et al. Macrophages are necessary for maximal nuclear factor-kappa B activation in response to endotoxin. Am J Respir Cell Mol Biol 2002; 26:572– 578. 104. Becker MN, Diamond G, Verghese MW, Randell SH. CD14-dependent lipopolysaccharideinduced beta-defensin-2 expression in human tracheobronchial epithelium. J Biol Chem 2000; 275:29731–29736. 105. Martin TR, Rubenfeld GD, Ruzinski JT, et al. Relationship between soluble CD 14, lipopolysaccharide binding protein, and the alveolar inflammatory response in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 155:937–944. 106. Dentener MA, Vreugdenhil AC, Hoet PH, et al. Production of the acute-phase protein lipopolysaccharide-binding protein by respiratory type II epithelial cells: implications for local defense to bacterial endotoxins. Am J Respir Cell Mol Biol 2000; 23:146–153. 107. Fan J, Kapus A, Marsden PA, et al. Regulation of Toll-like receptor 4 expression in the lung following hemorrhagic shock and lipopolysaccharide. J Immunol 2002; 168:5252–5259. 108. Chin AI, Dempsey PW, Bruhn K, Miller JF, Xu Y, Cheng G. Involvement of receptorinteracting protein 2 in innate and adaptive immune responses. Nature 2002; 416:190–194. 109. Kobayashi K, Inohara N, Hernandez LD, et al. RICK/Rip2/CARDIAK mediates signalling for receptors of the innate and adaptive immune systems. Nature 2002; 416:194–199. 110. Faure E, Thomas L, Xu H, Medvedev A, Equils O, Arditi M. Bacterial lipopolysaccharide and IFN-gamma induce Toll-like receptor 2 and Toll-like receptor 4 expression in human endothelial cells: role of NF-kappa B activation. J Immunol 2001; 166:2018–2024. 111. Van Strijp JA, Russell DG, Tuomanen E, Brown EJ, Wright SD. Ligand specificity of purified complement receptor type three (CD1 1b/CD18, alpha m beta 2, Mac-1). Indirect effects of an Arg-Gly-Asp (RGD) sequence. J Immunol 1993; 151:3324–3336. 112. Shi C, Zhang X, Chen Z, Robinson MK, Simon DI. Leukocyte integrin Mac-1 recruits toll/interleukin-1 receptor superfamily signaling intermediates to modulate NF-kappaB activity. Circ Res 2001; 89:859–865. 113. Medvedev AE, Flo T, Ingalls RR, et al. Involvement of CD 14 and complement receptors CR3 and CR4 in nuclear factor-kappaB activation and TNF production induced by lipopolysaccharide and group B streptococcal cell walls. J Immunol 1998; 160:4535–4542. 114. Cuzzola M, Mancuso G, Beninati C, et al. Beta 2 integrins are involved in cytokine responses to whole gram-positive bacteria. J Immunol 2000; 164: 5871–5876.

Acute respiratory distress syndrome

246

115. Kawai T, Takeuchi O, Fujita T, et al. Lipopolysaccharide stimulates the MyD88-independent pathway and results in activation of IFN-regulatory factor 3 and the expression of a subset of lipopolysaccharide-inducible genes. J Immunol 2001; 167:5887–5894. 116. Lentsch AB, Kato A, Davis B, Wang W, Chao C, Edwards MJ. STAT4 and STAT6 regulate systemic inflammation and protect against lethal endotoxemia. J Clin Invest 2001; 108:1475– 1482. 117. Suzuki T, Hashimoto S, Toyoda N, et al. Comprehensive gene expression profile of LPSstimulated human monocytes by SAGE. Blood 2000; 96:2584–2591. 118. Wang ZM, Liu C, Dziarski R. Chemokines are the main proinflammatory mediators in human monocytes activated by Staphylococcus aureus, peptidoglycan, and endotoxin. J Biol Chem 2000; 275:20260–20267. 119. Park WY, Goodman RB, Steinberg KP, et al. Cytokine balance in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 164:1896–1903. 120. Martin TR. Lung cytokines and ARDS: Roger S. Mitchell Lecture. Chest 1999; 116:2S-8S. 121. Stricter RM, Kunkel SL, Keane MP, Standiford TJ. Chemokines in lung injury: Thomas A. Neff Lecture. Chest 1999; 116:1038–110S. 122. Crowther MA, Marshall JC. Continuing challenges of sepsis research. Jama 2001; 286:1894– 1896. 123. Rijneveld AW, Florquin S, Branger J, Speelman P, Van Deventer SJ, van der Poll T. TNFalpha compensates for the impaired host defense of IL-1 type I receptor-deficient mice during pneumococcal pneumonia. J Immunol 2001; 167:5240–5246. 124. Hultgren OH, Svensson L, Tarkowski A. Critical role of signaling through IL-1 receptor for development of arthritis and sepsis during Staphylococcus aureus infection. J Immunol 2002; 168:5207–5212. 125. Hashimoto S, Kobayashi A, Kooguchi K, Kitamura Y, Onodera H, Nakajima H. Upregulation of two death pathways of perforin/granzyme and FasL/Fas in septic acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 161:237–243. 126. Hotchkiss RS, Tinsley KW, Swanson PE, et al. Sepsis-induced apoptosis causes progressive profound depletion of B and CD4+ T lymphocytes in humans. J Immunol 2001; 166:6952– 6963. 127. Kitamura Y, Hashimoto S, Mizuta N, et al. Fas/FasL-dependent apoptosis of alveolar cells after lipopolysaccharide-induced lung injury in mice. Am J Respir Crit Care Med 2001; 163:762–769. 128. Oberholzer C, Oberholzer A, Clare-Salzler M, Moldawer LL. Apoptosis in sepsis: a new target for therapeutic exploration. Faseb J 2001; 15:879–892. 129. Murphy PM. The molecular biology of leukocyte chemoattractant receptors. Annu Rev Immunol 1994; 12:593–633. 130. Zimmerman GA, McIntyre TM, Prescott SM. Cell-to-cell communication. In: Crystal RG, West JB, eds. The Lung: Scientific Foundations. Philadelphia: Lippincott-Raven, 1997:289– 304. 131. Downey GP, Dong Q, Kruger J, Dedhar S, Cherapanov V. Regulation of neutrophil activation in acute lung injury. Chest 1999; 116:46S-54S. 132. Worthylake RA, Burridge K. Leukocyte transendothelial migration: orchestrating the underlying molecular machinery. Curr Opin Cell Biol 2001; 13:569–577. 133. Chvatchko Y, Hoogewerf AJ, Meyer A, et al. A key role for CC chemokine receptor 4 in lipopolysaccharide-induced endotoxic shock. J Exp Med 2000; 191:1755–1764. 134. Lemjabbar H, Basbaum C. Platelet-activating factor receptor and AD AM 10 mediate responses to Staphylococcus aureus in epithelial cells. Nat Med 2002; 8:41–46. 135. Riedemann NC, Guo RF, Neff TA, et al. Increased C5a receptor expression in sepsis. J Clin Invest 2002; 110:101–108.

Pathogenesis of sepsis and septic-induced lung injury

247

136. Hammerschmidt DE, Weaver LJ, Hudson LD, Craddock PR, Jacob HS. Association of complement activation and elevated plasma-C5a with adult respiratory distress syndrome. Pathophysiological relevance and possible prognostic value. Lancet 1980; 1:947–949. 136a. Gerard C. Complement C5a in the sepsis syndrome—too much of a good thing? N Eng J Med 2003; 348:167–169. 137. Stricter RM, Kasahara K, Allen RM, et al. Cytokine-induced neutrophil-derived interleukin-8. Am J Pathol 1992; 141:397–407. 138. Kuhns DB, Nelson EL, Alvord WG, Gallin JI. Fibrinogen induces IL-8 synthesis in human neutrophils stimulated with formyl-methionyl-leucyl-phenylalanine or leukotriene B(4). J Immunol 2001; 167:2869–2878. 139. Prescott SM, McIntyre T.M., Zimmerman, G.A. Platelet-activating factor: a phospholipid mediator of inflammation. In: Gallin JI, Snyderman R., eds. Inflammation: Basic Principles and Clinical Correlates. Philadelphia: Lippincott, Williams & Wilkins, 1999:387–396. 140. McIntyre TM, Zimmerman GA, Prescott SM. Biologically active oxidized phospholipids. J Biol Chem 1999; 274:25189–25192. 141. Prescott SM, Zimmerman GA, Stafforini DM, McIntyre TM. Platelet-activating factor and related lipid mediators. Annu Rev Biochem 2000; 69: 419–445. 142. Zimmerman GA, McIntyre TM, Prescott SM, Stafforini DM. The platelet-activating factor signaling system and its regulators in syndromes of inflammation and thrombosis. Crit Care Med 2002; 30:S294–301. 143. Imaizumi TA, Stafforini DM, Yamada Y, McIntyre TM, Prescott SM, Zimmerman GA. Platelet-activating factor: a mediator for clinicians. J Intern Med 1995; 238:5–20. 144. Mathiak G, Szewczyk D, Abdullah F, Ovadia P, Rabinovici R. Platelet-activating factor (PAF) in experimental and clinical sepsis. Shock 1997; 7: 391–404. 145. Kuijpers TW, T. vdP. The role of PAF in endotoxin-related disease. In: Brade H, Opal S, Vogel SN, Morrison DC, eds. Endotoxin and Disease. New York: Bauer Press, 1999:449–462. 146. Zimmerman G, Prescott S, McIntyre T. Platelet-activating factor: a fluid-phase and cellassociated mediator of inflammation. In: Gallin J, Goldstein I, Snyderman R, eds. Inflammaton. Basic Principles and Clinical Correlates. 2nd ed. New York: Raven Press, 1992:289–304. 147. Parsons PE. Mediators and mechanisms of acute lung injury. Clin Chest Med 2000; 21:467– 476. 148. Lorant DE, Patel KD, McIntyre TM, McEver RP, Prescott SM, Zimmerman GA. Coexpression of GMP-140 and PAF by endothelium stimulated by histamine or thrombin: a juxtacrine system for adhesion and activation of neutrophils. J Cell Biol 1991; 115:223–234. 149. Ruchaud-Sparagano MH, Walker TR, Rossi AG, Haslett C, Dransfield I. Soluble E-selectin acts in synergy with platelet-activating factor to activate neutrophil beta 2-integrins. Role of tyrosine kinases and Ca2+ mobilization. J Biol Chem 2000; 275:15758–15764. 150. Bulger EM, Arbabi S, Garcia I, Maier RV. The macrophage response to endotoxin requires platelet activating factor. Shock 2002; 17:173–179. 151. Weyrich AS, McIntyre TM, McEver RP, Prescott SM, Zimmerman GA. Monocyte tethering by P-selectin regulates monocyte chemotactic protein-1 and tumor necrosis factor-alpha secretion. Signal integration and NF- kappa B translocation. J Clin Invest 1995; 95:2297–2303. 152. Ishii S, Nagase T, Tashiro F, et al. Bronchial hyperreactivity, increased endotoxin lethality and melanocytic tumorigenesis in transgenic mice overexpressing platelet-activating factor receptor. Embo J 1997; 16:133–142. 153. Wang H, Tan X, Chang H, Gonzalez-Crussi F, Remick DG, Hsueh W. Regulation of plateletactivating factor receptor gene expression in vivo by endotoxin, platelet-activating factor and endogenous tumour necrosis factor. Biochem J 1997; 322 (Pt 2):603–608. 154. Rabinovici R, Esser KM, Lysko PG, et al. Priming by platelet-activating factor of endotoxininduced lung injury and cardiovascular shock. Circ Res 1991; 69:12–25.

Acute respiratory distress syndrome

248

155. Chang SW, Feddersen CO, Henson PM, Voelkel NF. Platelet-activating factor mediates hemodynamic changes and lung injury in endotoxin-treated rats. J Clin Invest 1987; 79:1498– 1509. 156. Chang SW. Endotoxin-induced lung vascular injury: role of platelet activating factor, tumor necrosis factor and neutrophils. Clin Res 1992; 40:528–536. 157. Miotla JM, Jeffery PK, Hellewell PG. Platelet-activating factor plays a pivotal role in the induction of experimental lung injury. Am J Respir Cell Mol Biol 1998; 18:197–204. 158. Silliman CC, Voelkel NF, Allard JD, et al. Plasma and lipids from stored packed red blood cells cause acute lung injury in an animal model. J Clin Invest 1998; 101:1458–1467. 159. Kim JD, Baker CJ, Roberts RF, et al. Platelet activating factor acetylhydrolase decreases lung reperfusion injury. Ann Thorac Surg 2000; 70:423–428. 160. Dhainaut JF, Tenaillon A, Le Tulzo Y, et al. Platelet-activating factor receptor antagonist BN 52021 in the treatment of severe sepsis: a randomized, double-blind, placebo-controlled, multicenter clinical trial. BN 52021 Sepsis Study Group. Crit Care Med 1994; 22:1720–1728. 161. Dhainaut JF, Tenaillon A, Hemmer M, et al. Confirmatory platelet-activating factor receptor antagonist trial in patients with severe gram-negative bacterial sepsis: a phase III, randomized, double-blind, placebo-controlled, multicenter trial. BN 52021 Sepsis Investigator Group. Crit Care Med 1998; 26:1963–1971. 162. Poeze M, Froon AH, Ramsay G, Buurman WA, Greve JW. Decreased organ failure in patients with severe SIRS and septic shock treated with the platelet-activating factor antagonist TCV309: a prospective, multicenter, double-blind, randomized phase II trial. TCV-309 Septic Shock Study Group. Shock 2000; 14:421–428. 163. Vincent JL, Spapen H, Bakker J, Webster NR, Curtis L. Phase II multicenter clinical study of the platelet-activating factor receptor antagonist BB-882 in the treatment of sepsis. Crit Care Med 2000; 28:638–642. 164. Group TPIPAPS. Recombinant platelet activating factor acetylhyrolase (PAFASE) decreases the incidence of acute respiratory distress syndrome (ARDS) and 28-day all causes mortality. Intensive Care Med 2000; S321. 165. Graham RM, Stephens CJ, Silvester W, Leong LL, Sturm MJ, Taylor RR. Plasma degradation of platelet-activating factor in severely ill patients with clinical sepsis. Crit Care Med 1994; 22:204–212. 166. Schlame M, Schmid AB, Haupt R, Rustow B, Kox WJ. Study of platelet-activating factor acetylhydrolase in the perioperative period of patients undergoing cardiac surgery. Shock 1998; 9:313–319. 167. Patrick DA, Moore EE, Moore FA, Biffl WL, Barnett CC. Reduced PAF-acetylhydrolase activity is associated with postinjury multiple organ failure. Shock 1997; 7:170–174. 168. Trimoreau F, Francois B, Desachy A, Besse A, Vignon P, Denizot Y. Platelet-activating factor acetylhydrolase and haemophagocytosis in the sepsis syndrome. Mediators Inflamm 2000; 9:197–200. 169. Coughlin SR. Thrombin signalling and protease-activated receptors. Nature 2000; 407:258– 264. 170. Hugot JP, Chamaillard M, Zouali H, et al. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease. Nature 2001; 411:599–603. 171. Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease. Nature 2001; 411:603–660. 172. Bouchon A, Dietrich J, Colonna M. Cutting edge: inflammatory responses can be triggered by TREM-1, a novel receptor expressed on neutrophils and monocytes. J Immunol 2000; 164:4991–4995. 173. Bouchon A, Facchetti F, Weigand MA, Colonna M. TREM-1 amplifies inflammation and is a crucial mediator of septic shock. Nature 2001; 410:1103–1107. 174. Wang H, Bloom O, Zhang M, et al. HMG-1 as a late mediator of endotoxin lethality in mice. Science 1999; 285:248–251.

Pathogenesis of sepsis and septic-induced lung injury

249

175. Wang H, Yang H, Czura CJ, Sama AE, Tracey KJ. HMGB1 as a late mediator of lethal systemic inflammation. Am J Respir Crit Care Med 2001; 164:1768–1773. 175a. Czura C, Tracey KJ. Targeting high mobility group box 1 as a late-acting mediator of inflammation. Crit Care Med 2003; 31 (supplement):S46–S50. 176. Scaffidi P, Misteli T, Bianchi ME. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature 2002; 418:191–195. 177. Abraham E, Arcaroli J, Carmody A, Wang H, Tracey KJ. HMG-1 as a mediator of acute lung inflammation. J Immunol 2000; 165:2950–2954. 178. Bernhagen J, Calandra T, Mitchell RA, et al. MIF is a pituitary-derived cytokine that potentiates lethal endotoxaemia. Nature 1993; 365:756–759. 179. Froidevaux C, Roger T, Martin C, Glauser MP, Calandra T. Macrophage migration inhibitory factor and innate immune responses to bacterial infections. Crit Care Med 2001; 29:S13–15. 180. Bozza M, Satoskar AR, Lin G, et al. Targeted disruption of migration inhibitory factor gene reveals its critical role in sepsis. J Exp Med 1999; 189: 341–346. 181. Roger T, David J, Glauser MP, Calandra T. MIF regulates innate immune responses through modulation of Toll-like receptor 4. Nature 2001; 414:920–924. 182. Donnelly SC, Haslett C, Reid PT, et al. Regulatory role for macrophage migration inhibitory factor in acute respiratory distress syndrome. Nat Med 1997; 3:320–323. 183. Hoffmann JA, Kafatos FC, Janeway CA, Ezekowitz RA. Phylogenetic perspectives in innate immunity. Science 1999; 284:1313–1318. 184. Munford RS, Pugin J. The crucial role of systemic responses in the innate (non-adaptive) host defense. J Endotoxin Res 2001; 7:327–332. 185. Opal SM. Clinical impact of novel anticoagulation strategies in sepsis. Curr Opin Crit Care 2001; 7:347–353. 186. Furie B, Furie BC. Molecular and cellular biology of blood coagulation. N Engl J Med 1992; 326:800–806. 187. Greenberg DL, Davie, E.W. Introduction to hemostasis and the vitamin K-dependent coagulation factors. In: Scriver CR, Beadet A.L., Sly W.S., Valle D., eds. The Metabolic and Molecular Bases of Inherited Disease. Vol. III. New York: McGraw-Hill, 2001:4293–4326. 188. Esmon CT. Regulation of blood coagulation. Biochim Biophys Acta 2000; 1477:349–360. 189. Esmon CT. Anticoagulant protein C/thrombomodulin pathway. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic and Molecular Bases of Inherited Disease. New York: McGraw-Hill, 2001:4327–4343. 190. Esmon CT. Protein C anticoagulant pathway and its role in controlling micro vascular thrombosis and inflammation. Crit Care Med 2001; 29:S48–51; discussion 51–52. 191. Esmon CT. Inflammation and Thrombosis Mutual Regulation by Protein C. The Immunologist. Vol. 6. Hogrefe & Huber, 1998:84–89. 192. Giesen PL, Rauch U, Bohrmann B, et al. Blood-borne tissue factor: another view of thrombosis. Proc Natl Acad Sci USA 1999; 96:2311–2315. 193. Lorant D, McIntyre T, Prescott S, Zimmerman G. Platelet-activating factor: a signaling molecule for leukocyte adhesion. In: Pearson J, ed. Vascular Adhesion Molecules and Inflammation. Basel: Birkhauser, 1999:81–107. 194. Weyrich AS, Elstad MR, McEver RP, et al. Activated platelets signal chemokine synthesis by human monocytes. J Clin Invest 1996; 97:1525–1534. 195. Mahoney TS, Weyrich AS, Dixon DA, McIntyre T, Prescott SM, Zimmerman GA. Cell adhesion regulates gene expression at translational checkpoints in human myeloid leukocytes. Proc Natl Acad Sci USA 2001; 98:10284–10289. 196. Lindemann S, Tolley ND, Dixon DA, et al. Activated platelets mediate inflammatory signaling by regulated interleukin 1beta synthesis. J Cell Biol 2001; 154:485–490. 197. Esmon CT. Role of coagulation inhibitors in inflammation. Thromb Haemost 2001; 86:51–56. 198. Esmon CT. Introduction: are natural anticoagulants candidates for modulating the inflammatory response to endotoxin? Blood 2000; 95:1113–1116.

Acute respiratory distress syndrome

250

199. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344: 699–709. 200. Taylor FB, Jr., Steams-Kurosawa DJ, Kurosawa S, et al. The endothelial cell protein C receptor aids in host defense against Escherichia coli sepsis. Blood 2000; 95:1680–1686. 201. Murakami K, Okajima K, Uchiba M, et al. Activated protein C prevents LPS-induced pulmonary vascular injury by inhibiting cytokine production. Am J Physiol 1997; 272:L197– 202. 202. Grey ST, Tsuchida A, Hau H, Orthner CL, Salem HH, Hancock WW. Selective inhibitory effects of the anticoagulant activated protein C on the responses of human mononuclear phagocytes to LPS, IFN-gamma, or phorbol ester. J Immunol 1994; 153:3664–3672. 203. White B, Schmidt M, Murphy C, et al. Activated protein C inhibits lipopolysaccharideinduced nuclear translocation of nuclear factor kappaB (NF-kappaB) and tumour necrosis factor alpha (TNF-alpha) production in the THP-1 monocytic cell line. Br J Haematol 2000; 110:130– 134. 204. Joyce DE, Gelbert L, Ciaccia A, DeHoff B, Grinnell BW. Gene expression profile of antithrombotic protein c defines new mechanisms modulating inflammation and apoptosis. J Biol Chem 2001; 276:11199–11203. 205. Ware LB, Matthay MA. Lower plasma protein C is associated with worse clinical outcomes in patients with acute lung injury. Am J Respir Crit Care Med 2002; 165:A476. 206. Moore KL, Andreoli SP, Esmon NL, Esmon CT, Bang NU. Endotoxin enhances tissue factor and suppresses thrombomodulin expression of human vascular endothelium in vitro. J Clin Invest 1987; 79:124–130. 207. Faust SN, Levin M, Harrison OB, et al. Dysfunction of endothelial protein C activation in severe meningococcal sepsis. N Engl J Med 2001; 345:408–416. 208. Esmon CT. New Potential Therapeutic Modalities: aPC. Sepsis 1999; 3:161–171. 209. Boehme MW, Deng Y, Raeth U, et al. Release of thrombomodulin from endothelial cells by concerted action of TNF-alpha and neutrophils: in vivo and in vitro studies. Immunology 1996; 87:134–140. 210. Imaizumi T, Topham MK, McIntyre T, Prescott SM, Zimmerman G. Neutrophil degranulation: mechanisms of regulation and dysregulation in interactions with endothelial cells. In: Weir EK, Reeve HL, Reeves JT, eds. Interactions of Blood and the Pulmonary Circulation. New York: Futura Publishing Company, 2002:239–254. 211. Lorant DE, Zimmerman GA, McIntyre TM, Prescott SM. Platelet-activating factor mediates procoagulant activity on the surface of endothelial cells by promoting leukocyte adhesion. Semin Cell Biol 1995; 6:295–303. 212. Glaser CB, Morser J, Clarke JH, et al. Oxidation of a specific methionine in thrombomodulin by activated neutrophil products blocks cofactor activity. A potential rapid mechanism for modulation of coagulation. J Clin Invest 1992; 90:2565–2573. 213. Abraham E. Coagulation abnormalities in acute lung injury and sepsis. Am J Respir Cell Mol Biol 2000; 22:401–404. 214. Idell S. Anticoagulants for acute respiratory distress syndrome: can they work? Am J Respir Crit Care Med 2001; 164:517–20. 215. Miller DL, Welty-Wolf KE, Carraway MS, et al. Extrinsic coagulation blockade attenuates lung injury and pro-inflammatory cytokine release after intratracheal lipopolysaccharide. Am J Respir Cell Mol Biol 2003. In press. 216. Fein AM, Lippmann M, Holtzman H, Eliraz A, Goldberg SK. The risk factors, incidence, and prognosis of ARDS following septicemia. Chest 1983; 83:40–42. 217. McFeely JE, Hudson LD. Sepsis, multiple-organ dysfunction syndrome, and adult respiratory distress syndrome in humans. In: Brigham KL, ed. Endotoxin and the Lungs. New York: Marcel Dekker, Inc., 1994:321–350. 218. Nuckton TJ, Alonso JA, Kallet RH, et al. Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med 2002; 346:1281–1286.

Pathogenesis of sepsis and septic-induced lung injury

251

219. Powe JE, Short A, Sibbald WJ, Driedger AA. Pulmonary accumulation of polymorphonuclear leukocytes in the adult respiratory distress syndrome. Crit Care Med 1982; 10:712–718. 220. Zimmerman GA, Renzetti AD, Hill HR. Granulocyte adherence in pulmonary and systemic arterial blood samples from patients with adult respiratory distress syndrome. Am Rev Respir Dis 1984; 129:798–804. 221. Warshawski FJ, Sibbald WJ, Driedger AA, Cheung H. Abnormal neutrophil-pulmonary interaction in the adult respiratory distress syndrome. Qualitative and quantitative assessment of pulmonary neutrophil kinetics in humans with in vivo 111 indium neutrophil scintigraphy. Am Rev Respir Dis 1986; 133:797–804. 222. Elliott CG, Zimmerman GA, Orme JF, Jr., Morris AH, Mortensen JD. Granulocyte aggregation in adult respiratory distress syndrome (ARDS)-serial histologic and physiologic observations. Am J Med Sci 1985; 289:70–74. 223. Dalldorf FG, Carney CN, Rackley CE, Raney RB, Jr. Pulmonary capillary thrombosis in septicemia due to gram-positive bacteria. JAMA 1968; 206:583–586. 224. Bachofen M, Weibel ER. Alterations of the gas exchange apparatus in adult respiratory insufficiency associated with septicemia. Am Rev Respir Dis 1977; 116:589–615. 225. Bachofen M, Weibel ER. Structural alterations of lung parenchyma in the adult respiratory distress syndrome. Clin Chest Med 1982; 3:35–56. 226. Albertine KH. Histopathology of pulmonary edema and the acute respiratory distress syndrome. In: Matthay M, Ingbar D, eds. Pulmonary Edema. New York: Marcel Dekker Inc, 1998:37–83. 227. Imaizumi T, Albertine KH, Jicha DL, McIntyre TM, Prescott SM, Zimmerman GA. Human endothelial cells synthesize ENA-78: Relationship to IL-8 and to signaling of PMN adhesion. Am J Respir Cell Mol Biol 1997; 17:181–192. 228. Haslett C, Worthen GS, Giclas PC, Morrison DC, Henson JE, Henson PM. The pulmonary vascular sequestration of neutrophils in endotoxemia is initiated by an effect of endotoxin on the neutrophil in the rabbit. Am Rev Respir Dis 1987; 136:9–18. 229. Li N, Soop A, Sollevi A, Hjemdahl P. Multi-cellular activation in vivo by endotoxin in humans—limited protection by adenosine infusion. Thromb Haemost 2000; 84:381–387. 230. Dixon DA, McIntyre T, Prescott SM, Weyrich AS, Zimmerman GA. Molecular mechanisms of juxtacrine cell signalling in micro vascular responses and inflammation. In: SchimidSchoenbein NG, ed. Molecular Mechanisms of Microcirculatory Disorders. Paris: SpringerVerlag, 2003. In press. 231. Tomashefski JF, Jr. Pulmonary pathology of acute respiratory distress syndrome. Clin Chest Med 2000; 21:435–466. 232. Parsons PE, Gillespie MM, Moore EE, Moore FA, Worthen GS. Neutrophil response to endotoxin in the adult respiratory distress syndrome: role of CD14. Am J Respir Cell Mol Biol 1995; 13:152–160. 233. Doerschuk CM. Mechanisms of leukocyte sequestration in inflamed lungs. Microcirculation 2001; 8:71–88. 234. Forlow SB, Rose CE, Ley K. Leukocyte Adhesion and Emigration in the Lung. In: Weir EK, Reeve HL, Reeves JT, eds. Interactions of Blood and the Pulmonary Circulation. Armonk, New York: Futura Publishing Company, 2002:255–275. 235. Steinberg KP, Milberg JA, Martin TR, Maunder RJ, Cockrill BA, Hudson LD. Evolution of bronchoalveolar cell populations in the adult respiratory distress syndrome. Am J Respir Crit Care Med 1994; 150:113–122. 236. Rosseau S, Hammerl P, Maus U, et al. Phenotypic characterization of alveolar monocyte recruitment in acute respiratory distress syndrome. Am J Physiol Lung Cell Mol Physiol 2000; 279:L25–35. 237. Grissom CK, Orme JF, Richer LD, Zimmerman G, McIntyre T, Elstad MR. PAFacetylhydrolase is increased in lung lavage fluid from patients with ARDS. Crit Care Med 2003. In press.

Acute respiratory distress syndrome

252

238. Coalson JJ, Hinshaw LB, Guenter CA. The pulmonary ultrastructure in septic shock. Exp Mol Pathol 1970; 12:84–103. 239. Dormehl IC, Maree M, Cromarty D, et al. Investigation by scintigraphic methods of neutrophil kinetics under normal and septic shock conditions in the experimental baboon model. Eur J Nucl Med 1990; 16:643–647. 240. Zapol WM, Jones R. Vascular components of ARDS. Clinical pulmonary hemodynamics and morphology. Am Rev Respir Dis 1987; 136:471–474. 241. Bertozzi P, Astedt B, Zenzius L, et al. Depressed bronchoalveolar urokinase activity in patients with adult respiratory distress syndrome. N Engl J Med 1990; 322:890–897. 242. Pugin J, Verghese G, Widmer MC, Matthay MA. The alveolar space is the site of intense inflammatory and profibrotic reactions in the early phase of acute respiratory distress syndrome. Crit Care Med 1999; 27:304–312. 243. Harbarth S, Holeckova K, Froidevaux C, et al. Diagnostic value of procalcitonin, interleukin6, and interleukin-8 in critically ill patients admitted with suspected sepsis. Am J Respir Crit Care Med 2001; 164:396–402. 244. Claeys R, Vinken S, Spapen H, et al. Plasma procalcitonin and C-reactive protein in acute septic shock: clinical and biological correlates. Crit Care Med 2002; 30:757–762. 245. Reinhart K, Bayer O, Brunkhorst F, Meisner M. Markers of endothelial damage in organ dysfunction and sepsis. Crit Care Med 2002; 30:S302–312. 246. Ware LB, Conner ER, Matthay MA. von Willebrand factor antigen is an independent marker of poor outcome in patients with early acute lung injury. Crit Care Med 2001; 29:2325–2331. 247. Petricoin EF, Ardekani AM, Hitt BA, et al. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 2002; 359:572–577. 248. Matthay MA, Zimmerman G, Esmon CT, et al. Future Directions in Acute Lung Injury: Summary of a National Heart, Lung, and Blood Institute Working Group, 2003. In press.

11 Heat Shock Response, Heat Shock Proteins, and Acute Lung Injury HYON LEE, MICHAELA C. GODZICH, and JEAN-FRANÇOIS PITTET University of California, San Francisco San Francisco, California, U.S.A.

I. Introduction The heat shock response was discovered in 1962 by Ritossa (1), who reported that Drosophila salivary gland chromosome puffs were induced in response to transient hyperthermia. Since this first observation, a large number of investigators have reported that this pattern was linked to the expression of a specific group of proteins called heat shock proteins. The expression of these proteins in response to hyperthermia was called the “stress response” or “heat shock response,” a ubiquitous and highly conserved defense mechanism in all organisms, from bacteria to animals and humans. The acute respiratory distress syndrome (ARDS) is a devastating syndrome of acute inflammation of the lung that affects both barriers of the lung—the lung endothelium and the alveolar epithelium (2). The early phase of acute lung injury is characterized by the accumulation of inflammatory cells (neutrophils, macrophages) within the alveolar structures that release high levels of oxidant species such as superoxide, H2O2, or reactive nitrogen species (2). One of the most important consequences associated with the induction of the heat shock response is to confer cytoprotection against a variety of stressors, such as oxidant-mediated injury, one of the most important molecular mechanisms of acute lung injury. This chapter will first describe the different families of heat shock proteins, reviewing their intracellular functions as well as the newly discovered extracellular role of some heat shock proteins in the innate immunity. Then, the second objective of this chapter will focus on the mechanisms associated with the induction of the heat shock response. Finally, the last objective of this chapter will be to review the possible cellular mechanisms of cytoprotection associated with the induction of the heat shock response in in vitro and in vivo models of acute lung injury and to discuss potential clinical applications of stress preconditioning to attenuate the severity of this syndrome.

II. Heat Shock Proteins Heat shock proteins (HSPs) were originally identified as a family of proteins whose expression is increased in response to heat shock, although several of these proteins are

Heat shock response, heat shock proteins, and acute lung injury

255

expressed and play an important role in nonstressed cells (3). Heat shock proteins consist of several families of proteins traditionally classified by molecular weight (Table 1). These families of HSPs are characterized by different distributions within the cell, different degrees and kinetics of inducibility, and different tissue specificity. Heat shock factors (HSFs) are transcription factors that regulate HSP expression by interacting with a specific DNA sequence [heat shock element (HSE)] in the promoter of HSPs. Four HSFs have been identified, three of them in mammalian cells (HSF-1, HSF-2, and HSF-4). HSF-1 is involved in the acute heat shock response. In unstressed cells, HSF-1 exists as a monomer bound to heat shock proteins. Activation of the heat shock response releases HSF-1 from HSPs, allowing for the trimerization of HSF-1 and its translocation to the nucleus where it binds to HSE (4). Interestingly, HSF-1 may undergo additional phosphorylation, which will activate or inhibit its transcriptional activity (5, 6). Moreover, the binding of HSF-1 to HSE may be regulated by a constitutively expressed protein known as HSE binding factor (HSE-BF) (7). Thus, the state of phosphorylation of HSF-1 and its ability to bind to HSE by competing with HSE-BF provide additional levels of regulation of HSP expression. A. Intracellular Functions HSPs have three principal functions as intracellular proteins. The first function is chaperonin activity (8, 9). These proteins prevent abnormal protein aggregation and assist with the refolding of denatured proteins.

Table 1 Major Heat Shock Protein Families Family

Chaperone

Location

Function

Ubiquitins Ubiquitin

Nucleus, cytosol

Inducible; target proteins for degradation by 26S proteasome

Small HSPs

Hsp10

Mitochondria

Promotes substrate release in conjunction with HSP60

Hsp27

Nucleus, cytosol, cell surface membrane

Inducible; stabilizes denatured proteins against aggregation

HSP40

HSP60

HSP70

Hsp32 (HO- Cytosol, mitochondria, 1) ER

Inducible; breaks down heme into biliverdinbilirubin, iron, and CO

Hsp40

Nucleus, cytosol

Guides protein folding in conjunction with HSP70

Hsp47

ER

Binds and transports collagen from ER to Golgi

Hsp60

Mitochondria

Refolds and prevents denatured protein aggregation

Grp58

ER

Constitutively expressed

Hsp70

Nucleus, cytosol, mitochondria, ER

Inducible; folds nascent and denatured proteins, involved in interorganelle transport

Acute respiratory distress syndrome

HSP90

HSP110

256

Hsc70

Nucleus, cytosol, peroxisome

Guides protein synthesis and imports for protein degradation

Grp78

ER

Inducible; folds protein in response to low glucose

Hsp90-α

Nucleus, cytosol

Hsp90-β

Nucleus, cytosol

Facilitates conformational maturation of steroid hormone receptors and signal

Grp94

ER

Inducible; binds calcium

Hsp110

Nucleus, cytosol

Inducible; folds denatured protein

Grp170

ER

Constitutively expressed

ER, endoplasmic reticulum; HSP, heat shock proteins.

HSPs also play an important role in the folding of nascent polypeptides during protein synthesis in unstressed cells. The most important HSPs associated with chaperone activity are the members of the HSP40, HSP60, HSP70, and HSP90 families of proteins. The second intracellular function of heat shock proteins is the regulation of cellular redox state. The best known heat shock protein that regulates the intracellular redox state is Hsp32, also known as heme-oxygenase-1 (10). Hsp32 catalyzes the breakdown of heme into biliverdin, carbon monoxide, and free iron. Biliverdin is then converted into bilirubin, a potent antioxidant. Carbon monoxide plays an important role in signal transduction in neural tissue and vascular smooth muscle cells. Free iron is rapidly bound to ferritin. The third intracellular function of heat shock proteins is the regulation of protein degradation. One of these proteins, ubiquitin, is upregulated by heat shock and serves to label proteins for degradation by proteasomes (11). B. Extracellular Functions In addition to their role as molecular chaperones, recent reports suggest an extracellular role for HSPs. HSPs can be released from necrotic cells or expressed at the surface of stressed cells (12). Moreover, Hsp60 and Hsp70 are present in the peripheral circulation of normal individuals, partly bound to autoantibodies (13, 14). Recent experimental evidence indicates that Hsp60 and Hsp70 may play an important role in the modulation of the inflammatory response. Hsp60 and Hsp70 appear to affect the innate immunity because they appear to be endogenous ligands for the Toll-like receptors 2 (important for monocyte activation upon exposure to gram-positive bacteria) and 4 (principal receptor for LPS) (15–17). Indeed, through the binding to these receptors, Hsp60 and Hsp70 induce the release of proinflammatory mediators via the activation of the NF-κB pathway. Thus, because significant levels of Hsp70 can be detected in the pulmonary edema fluid of patients with acute lung injury (J.-F.Pittet et al., unpublished observations), Hsp70 could play an important role in modulating the inflammatory response in the distal airspace of the lung during the early phase of the disease. HSPs (particularly Hsp60 and Hsp70) appear to be a significant element of the immune response to pathogenic microorganisms because a large part of the immune response against these microorganisms is directed toward peptides derived from HSPs

Heat shock response, heat shock proteins, and acute lung injury

257

(18). HSP structure is highly conserved from bacteria to animals and humans, suggesting that HSPs could act as potentially harmful autoantigens. This concept could provide a link between infection and autoimmunity and has been supported by studies involving an immune response to HSPs in arthritis, multiple sclerosis, and diabetes (19– 21). However, there is also some experimental evidence that T-cell reactivity to self-Hsp60 or selfHsp70 could downregulate the inflammatory response, suggesting that the T-cell reactivity to self-HSPs is a part of the regulation of the immune response and could therefore modulate proinflammatory disease processes (reviewed in Ref. 22). In fact, data from experimental animal models of arthritis and from patients with arthritis indicate that the reactivity to self-Hsp60 downregulates the inflammatory response while reactivity to nonself-Hsp60 induces a proinflammatory T-cell phenotype (23). Third, HSPs play an important role in antigen presentation either through the chaperoning of antigenic peptides that are generated within cells and taken by MHC class I molecules or through antigenic peptides released by cell stress or death, taken up by antigen-presenting cells and then presented by their MHC class molecules (reviewed in Ref. 24). In summary, there is a growing body of evidence that in addition to their chaperone activity, HSPs may have an important role as extracellular mediators involved in modulating the inflammatory and immune responses associated with injury.

III. Induction of the Heat Shock Response The heat shock or stress response is a highly conserved cellular defense mechanism characterized by the increased expression and accumulation of heat shock proteins. Because many different stressors lead to an increase in the expression of heat shock proteins, they are also referred as stress proteins. Regardless of their mode of activation, increased levels of the stress proteins by mild hyperthermia are associated with the ability of the cell to withstand a subsequent thermal insult that would otherwise be lethal, a phenomenon referred as “thermotolerance” or “preconditioning.” Nonthermal forms of cellular stress also induce heat shock protein expression and may confer a thermotolerant phenotype (25–27). Induction of heat shock protein expression, by either thermal or nonthermal stimuli, also confers protection against nonthermal cytotoxic stimuli such as oxidants and endotoxin (28). Interestingly, the fate of cells also depends on the sequence of the different stressors. For example, prior heat shock stress can protect cells against inflammatory stress both in vitro and in vivo, while induction of a subsequent heat shock stress in cells exposed to an inflammatory stimulus can precipitate cell death by apoptosis. The phenomenon has been called the “heat shock paradox,” and its mechanism is not fully understood, but could be related to the downstream effect of the induction of NF-κB’s endogenous inhibitor, IκBα, a putative heat shock protein (29). Thus, this phenomenon illustrates how the sequence of different stressors may produce dramatic differences in the cell’s outcome. Thermotolerance is the result of numerous molecular mechanisms of which the induction of HSPs is only one. In fact, the attribution of thermotolerance to the induction of heat shock proteins requires more than correlative evidence. There is a need for data resulting from genetic or direct experimental manipulation in order to establish the functional significance of HSPs. Indeed, the activation of the heat shock response is also

Acute respiratory distress syndrome

258

associated with changes in expression of around 50 different genes that are not considered as HSPs. There are changes in several transcriptional systems other than HSF1: (1) upregulation of the expression of AP-1, egr-1, and C/ EBP-α and -β, but downregulation of c-myc expression (30–32); (2) changes in the activity of transcription factors (i.e., by phosphorylation) (e.g., heat shock-mediated phosphorylation of c-jun by JNK has been shown in mouse 3T3 cells) (33), (3) changes in cellular localization of transcription factors (the best known example is the effect of the heat shock response on the translocation of NF-κB into the nucleus). Moreover, several heat-responsive genes encode for proteins that modulate the MAP kinase pathway. For example, the expression of MAP kinase phosphatases DUSP-1 and -5 is increased after thermal stress (34). This effect may allow the MAP kinase cell signaling pathways to be more responsive to subsequent stressors in thermotolerant cells. Another important effect of the stress response is the arrest of the cell cycle, for which the activation of the P53 gene appears to be critical (35). Finally, as mentioned above, the fate of cells depends not only on the intensity of the heat shock response (from thermotolerance to apoptosis and necrosis), but also on the sequence of the different stressors (29). In summary, the number of genes affected by the stress response is rapidly increasing in tandem with a greater understanding of the cellular response to heat shock, including the induction of HSPs. There is no doubt that new genomic and proteomic techniques will aid in the investigation of the upregulation and/or inhibition of thousand of genes or proteins involved in the cellular response to heat shock.

IV. Protective Effect of the Heat Shock Response in Acute Lung Injury A. In Vitro Studies Several lines of evidence demonstrate that the heat shock response can confer cytoprotection against a large range of lung cell injury. First, the activation of the heat shock response attenuates injury to lung endothelial and smooth muscle cells induced by several stressors. For example, mild hyperthermia or sodium arsenite protected lung endothelial cells against LPS-induced apoptosis (36). Heat shock also inhibited iNOS expression in cultured rat pulmonary artery smooth muscle cells (37). Moreover, heat shock was associated with a decrease in the expression of TNF-α-induced ICAM-1 on lung endothelial cells (38). These results suggest that heat shock may therefore interfere with neutrophil adhesion and migration into the airspace after the onset of acute lung injury. Several studies demonstrated that the induction of thermotolerance protects the alveolar epithelium against oxidative stress, the hallmark of acute lung injury. There is a significant attenuation of the expression of iNOS-dependent NO production and of the chemokine RANTES in alveolar epithelial cells pretreated with mild hyperthermia without affecting the expression of constitutively expressed genes, such as those of surfactant proteins (39). Similarly, prior heat shock significantly decreases the epithelial injury caused by several NO donors (40). Interestingly, not only mild hyperthermia, but also other inducers of the stress response such as geldanamycin, a benzoquinone ansamycin that activates the heat shock response by binding to Hsp90 allowing the

Heat shock response, heat shock proteins, and acute lung injury

259

release and trimerization of HSF-1 (27, 41), may protect the alveolar epithelium against oxidative stress. For example, geldanamycin inhibits NF-κB activation and IL-8 gene expression in cultured human respiratory epithelial cells (42). Taken together, these results indicate that the induction of the stress response is associated with a significant attenuation of the oxidant-mediated injury to the two cell types that are part of the alveolar-capillary barrier, suggesting that thermotolerance may reduce in vivo the severity of oxidant-mediated lung injury. B. In Vivo Studies Villar et al. were the first to demonstrate that the activation of the heat shock response was associated with attenuation of lung damage in experimental animal models of acute lung injury. These investigators reported that prior treatment with mild hyperthermia or a heavy metal, sodium arsenite, decreased lung injury caused by the intratracheal instillation of phospholipase A2 (43) or systemic administration of endotoxin (44). Similar results were obtained when the heat stress was applied after the initiation of endotoxemia (45). The same investigators subsequently reported that the induction of the heat response also conferred protection against acute lung injury induced by cecal ligation and perforation (46). Is the protection associated with the induction of the stress response specific for sepsis-induced acute lung injury? The answer is no because several investigators have reported in vivo protection in experimental models of lung ischemia-reperfusion injury and hemorrhage-induced lung injury. Javadpour et al. (47) investigated the role of the stress response on acute lung injury induced by ischemia-reperfusion. They found a substantial decrease in the amount of lung edema and lung neutrophil infiltration in thermotolerant rats compared to the control animals. In another form of ischemiareperfusion injury, namely lung transplantation, Hiratsuka et al. reported that heat shock protected lung grafts from reperfusion injury after transplantation in rats (48). Interestingly, these investigators found comparable results when Hsp70 was expressed in the lungs by prior administration of an adenovirus encoding for this protein, suggesting that increased expression of this inducible HSP may be involved in the lung protection associated with heat shock (49). These results are particularly interesting because Hsp70 expression appears to be impaired in the lungs of animals with severe sepsis induced by cecal ligation and perforation (50). Finally, recent results from our laboratory provided the first experimental evidence that that stress preconditioning can impact ion transport activities across the alveolar epithelium and thereby help to restore the ability of this barrier to upregulate fluid transport in response to cAMP-dependent stimulation after a severe hemorrhage. In summary, taken all together, stress preconditioning appears to provide protection to the lung against various in vivo forms of acute injury. C. Mechanisms of Protection The mechanisms by which the activation of the stress response confers cytoprotection in the lung are likely to be multifold. Recent experimental evidence indicates that the induction of the stress response is associated with a corresponding attenuation of three important features of acute lung injury, namely the inflammatory response in the lung,

Acute respiratory distress syndrome

260

the migration of neutrophils into the airspace of the lung, and the apoptosis of alveolar epithelial and lung endothelial cells. First, the stress response in the lung is associated with a decrease in the lung inflammatory response. For example, activation of the stress response has been reported to inhibit the endotoxin-mediated expression of TNF-α and IL-1β in cultured mononuclear cells and in vivo in the lung (51). Furthermore, the inhibition of cytokinemediated iNOS expression and the subsequent release of NO in either rat pulmonary smooth muscle cells (37) or in the murine lung epithelium (39) following stress preconditioning has been reported. Recent work from our laboratory has provided in vivo evidence that NO is the proximal mediator of the inflammatory response that limits the rate of alveolar epithelial transport after a prolonged hemorrhagic shock. Increased levels of NO appeared to directly alter the function of membrane proteins involved in βadrenergic receptor-cAMP signaling pathways (52). Consistent with our previous observations, we have recently shown that stress preconditioning, using either whole body hyperthermia or the administration of geldanamycin, was associated with a significant attenuation in the shock-mediated expression of iNOS and the subsequent release of NO in the lung (53). This observation was supported by our in vitro studies examining alveolar macrophages and alveolar epithelial cells, two cells known to be a major source of NO production within the airspaces of the lung (54). Our results showed that stress preconditioning with heat did in fact cause a significant decrease in the production of NO by alveolar macrophages removed from the airspaces of preconditioned and hemorrhaged rats when compared to the control, hemorrhaged animals not provided a preconditioning treatment (53). Similarly, stress preconditioning resulted in a significant decrease in the production of NO by either A549 cells, an alveolar epithelial cell line, or primary cultures of rat ATII cells upon their subsequent exposure to pro-inflammatory cytokines (53). These results are in line with prior studies showing that the induction of the heat shock response inhibited the cytokine-mediated expression of iNOS in pulmonary artery smooth muscle cells (37) and lung epithelial cells (39). Moreover, stress preconditioning has also been reported to inhibit the expression of a variety of other proinflammatory genes including 11–8 (55), RANTES (56), TNF-α (57), ICAM-1 (38), and IL-1β (58). This effect may be rather specific to proinflammatory genes since the expression of other products, such as the surfactant proteins, was largely unaffected (39). Taken all together, activation of the cellular stress response may provide protection in the lung via its attenuation of proinflammatory mediators, explaining (at least in part) why stress preconditioning protects against various forms of acute lung injury. Exactly how stress preconditioning might inhibit cellular inflammatory responses is still under investigation. One study reported that the transcription factor controlling the expression of the heat shock genes (HSF) might act as a represser, binding to regulatory sequences upstream of those genes encoding certain proinflammatory cytokines, such as IL-1β (59). Yet other studies have suggested a more general mechanism, involving the inhibition of the whole NF-κB pathway (60–62) (Fig. 1). For example, it has been reported that the heat shock-mediated inhibition of NF-κB activation after mild hyperthermia was associated with a decrease in the phosphorylation and ubiquitination of IκBα, thereby leading to the stabilization of this NF-κB inhibitor (61, 62). The mechanism by which mild hyperthermia inhibits phosphorylation of IκBα, appears to

Heat shock response, heat shock proteins, and acute lung injury

261

involve a dual effect on I kappa β kinase (IKK) activity and intracellular phosphatase activity (63, 64). This effect may be specific for IKK activation because heat shock did not inhibit the endotoxin-mediated activation of jun kinase (JNK). In addition, activation of the stress response by mild hyperthermia was suggested to result in an increase in the expression of the IκB gene and a corresponding increase in the level of the IκB protein (55, 61, 65). Indeed, in the cultured human respiratory epithelium, heat shock induced a small increase in IκBα gene transcription, as measured by nuclear run-on assays, and a threefold increase in the half-life of IκBα messenger RNA (66). Thus, the heat-shockmediated increase in the expression of Iκ Bα is a complex process involving transcriptional and posttranscriptional mechanisms. The activation of the stress response by chemical inducers, such as geldanamycin, was also associated with an inhibition of NF-κB activation. Interestingly, this effect was independent of IκBα degra-

Acute respiratory distress syndrome

262

Figure 1 Inhibition of the NF-κBdependent cell signaling pathway by the activation of the heat shock response (HSR). dation or p65 nuclear translocation (42). Further experiments demonstrated that the inhibitory effect of geldanamycin was related to its ability to bind DNA and therefore prevent NF-κB binding to DNA (42). It is important to note, however, that these effects of heat on the NF-κB pathway appear to be restricted only to the early time periods following the heating regimen. Thus, how NF-κB activities might be attenuated in cells

Heat shock response, heat shock proteins, and acute lung injury

263

preconditioned via a heat shock treatment 12–24 hours prior to cytokine stimulation is still not clear and therefore represents a major focus for future studies. Activation of the stress response attenuates the oxidative stress in the lung not only by a direct inhibition of the inflammatory response, but also by a decrease in the sequestration of neutrophils in pulmonary capillaries. For example, stress preconditioning with heat attenuated ischemia-reperfusion injury in rat mesenteric postcapillary venules by modulating neutrophil-endothelium interactions in vivo (67). Also, stress preconditioning with heat protected against ischemia-reperfusion-induced lung injury by decreasing neutrophil sequestration in the lung (47). The mechanisms responsible for the decrease in neutrophil sequestration in the lung in thermotolerant animals after oxidant stress, however, are not well understood. They may involve (1) a decrease in the release of neutrophils from the bone marrow in the presence of circulating inflammatory mediators (a recent study has shown that most of the neutrophils sequestered in the lung after an oxidant stress could be accounted for by the number of neutrophils released from the bone marrow) (68), (2) an increase in the rate of neutrophil apoptosis (69) that has been shown to be significantly reduced in the lung after hemorrhage (70), (3) a decrease in the expression of adhesion molecules (i.e., ICAM-1 P-selectin) on the surface of the lung endothelium (71), (4) an attenuation of the chemotactic gradient of neutrophils across the alveolar-capillary barrier by decreasing the release of the chemokines IL-8 and/or MIP-2 within the air-spaces of the lung. Apoptosis of lung parenchymal and inflammatory cells is part of the host response to noxious stimuli during both the acute inflammatory and resolution phases of acute lung injury. There are experimental and clinical data suggesting that abnormalities of the apoptotic process (i.e., inefficient or excessive apoptosis) may increase damage to the lung parenchyma, prevent the resolution of inflammation, and perhaps promote abnormal tissue repair (i.e., lung fibrosis) in acute lung injury (72, 73). Activation of the stress response is associated with the induction of heat shock proteins such as Hsp70 and Hsp27 that can modulate both apoptotic and nonapoptotic death programs (Fig. 2). For example, phosphorylation of Hsp27 by the MAP p38 kinase is important for the survival of cells (74). Hsp27 can interfere with apoptotic signal transduction by reducing cytochrome c release, preventing the activation of procaspases 9 and 3 and by inhibiting the Fasinduced

Acute respiratory distress syndrome

264

Figure 2 Suppression of the apoptotic pathway by heat shock proteins Hsp27 and Hsp72. apoptotic pathway (74). In addition, Hsp70 dramatically reduces the activation of JNK, an effect that appears to be critical for the inhibition of apoptosis in nontransformed cells (75). Inhibition of JNK appears also to be important in the suppression of caspaseindependent death pathways (76). Interestingly, the JNK-inhibiting activity of Hsp70 seems to be sufficient for protection against apoptosis, and the refolding activity of Hsp70 does not appear to be necessary for this particular function of Hsp70 (77). However, the exact role of these two particular HSPs in modulating abnormalities of the

Heat shock response, heat shock proteins, and acute lung injury

265

apoptotic process during the course of acute lung injury remains unknown and deserves further studies. The specific role of HSPs in the protective effect associated with the activation of the stress response remains incompletely understood. In their role as molecular chaperones, stress proteins help stabilize and/or refold proteins damaged as a consequence of the particular stress event. In addition, when present at higher levels the stress proteins collectively facilitate the synthesis and maturation of new proteins needed to replace those irreparably damaged by the earlier metabolic insult. Finally, a specific role for stress-inducible proteins, such as Hsp72 or HO-1 (Hsp32), has been suggested from experiments utilizing molecular means to increase their intracellular levels. For example, the overexpression of Hsp72 was demonstrated to protect cultured lung cells against NO(40), endotoxin- (36), or hyperoxia-mediated cell death (78). Moreover, HO-1 (Hsp 32) is one the best heat shock proteins characterized with respect to its possible protective role in lung biology, although the expression of HO-1 can be upregulated by several inducers independent of stress preconditioning (79, 80). HO-1 catalyzes the first and rate-limiting step in the degradation of heme to yield equimolar quantities of biliverdin, carbon monoxide (CO), and iron (80) (Fig. 3). Three isoforms of HO exist; HO-1 is highly inducible, whereas HO-2 and HO-3 are constitutively expressed (80). Although heme is the major substrate of HO-1, a variety of agents that cause oxidative stress, including hydrogen peroxide, glutathione depleters, ultraviolet irradiation, endotoxin, and hyperoxia, are also strong inducers of HO-1 expression (80). One interpretation of this finding is that HO-1 can serve as a key biological molecule in the defense against oxidative stress. Induction of HO-1 in the lung protects against oxidative stress (79, 81, 82) and mediates the anti-inflammatory effect of IL-10 (83). The mechanisms that explain the protection against oxidative stress associated with increased the expression of HO-1 are incompletely understood. HO-1 could exert its ability to reduce oxidative stress through the biological actions of its reaction products—CO through the activation of the MAP p38 kinase (84, 86) and derepression of the fibrinolytic axis (Fig. 4) (87) and biliverdin (88, 89)—that would prevent oxidant-induced neutrophil sequestration in the lung. Recently, induction of HO-1 has been shown to attenuate iNOS

Acute respiratory distress syndrome

266

Figure 3 Enzymatic reaction catalyzed by heme-oxygenase-1 (HO-1 or Hsp32).

Heat shock response, heat shock proteins, and acute lung injury

267

Figure 4 Induction of heme-oxygenase by ischemic stress or inhaled CO protects against ischemic lung tissue injury. (From Ref. 86.) expression in the kidney (90) and in the lung (J.-F.Pittet et al., unpublished observations). Thus, it is likely that the induction of HO-1 in the lung is one of the mechanisms by which stress preconditioning protects against oxidative stress in acute lung injury. Heat shock factor-1 is a transcription factor that plays a crucial role in the activation of the stress response by regulating the expression of various HSPs. HSF-1 undergoes phosphorylation and trimerization upon stimulation by mild hyperthermia or by other inducers of the stress response, allowing its translocation to the nucleus, where it binds to heat shock elements present on the promoter of various HSPs. Whether HSF-1 plays a role in downregulating the inflammatory response in the lung is not fully understood. HSF-1 has been shown to inhibit the IL-1 promoter (59). Moreover, mice lacking HSF-1

Acute respiratory distress syndrome

268

have an increased proinflammatory response to LPS (91). However, a recent study demonstrated that embryonic fibroblasts from HSF-1-null mutant mice preserved their ability to down-regulate NF-κB translocation to the nucleus secondary to the activation of the heat shock response (92). Whether this observation remains true in an in vivo model including HSF-1-null mutant mice remains to be demonstrated. Taken together, these data indicate that the respective roles of Hsp70, HO-1, and HSF-1 in modulating cellular proinflammatory responses in the lung are beginning to be understood but need to be better defined. It is likely that the role of these HSPs will vary depending on the type of injury and cells involved.

V. Conclusions and Clinical Implications The heat shock or stress response is a highly conserved cellular defense mechanism characterized by the increased expression of stress proteins that allow the cell to withstand a subsequent insult that would otherwise be lethal, a phenomenon referred to as “thermotolerance” or “preconditioning.” Recent experimental work indicates that an important feature of the stress response in the lung is the modulation of the inflammatory response by a transient inhibition of the expression of proinflammatory mediators, thus providing broad organ and tissue protection against the oxidative stress associated with acute lung injury. In particular, stress preconditioning, either induced by whole body hyperthermia or by pharmacological inducers of the stress response, attenuated NOmediated oxidative stress to the alveolar alveolar-capillary barrier in several in vivo and in vitro models of acute lung injury (Fig. 5) (53). This protective effect was in part mediated by an attenuation of the NF-κB pathway and could be secondary to the induction of the expression of several HSPs, such as Hsp70 or HO-1. These results are of potential clinical importance. Indeed, the results of a recent study in critically ill patients revealed an increased production of reactive oxygen-nitrogen intermediates in the distal airspace of patients with acute lung injury. In turn, these changes were associated with an impaired removal of the pulmonary edema fluid from the airspace of the lung (93), an early predictor of a prolonged duration of mechanical ventilation and higher hospital mortality in patients with acute lung injury (94). Future studies should therefore provide (1) a better understanding of the mechanisms of protection against oxidative stress as well as the potential negative consequences associated with the induction of the stress response in the lung, (2) additional information about the mechanisms regulating the heat shock response during acute lung

Heat shock response, heat shock proteins, and acute lung injury

269

Figure 5 Stress preconditioning restores the ability of the alveolar epithelium to respond to catecholamines by upregulating alveolar fluid clearance after hemorrhage. (From Ref. 53.) injury, and (3) additional information about the newly discovered extra-cellular role of some HSPs, such as Hsp60 and Hsp70 in the distal airspace of the lung. An answer to these questions may lead to the development of new therapies directed to the use of the

Acute respiratory distress syndrome

270

heat shock response as a treatment to transiently attenuate the inflammatory response during the early phase of acute lung injury.

Acknowledgments This work was primarily supported by NIH Grant GM 62188 (J.-F. Pittet).

References 1. Ritossa FM. A new puffing pattern induced by a temperature shock and DNP in Drosophila. Experientia 1962; 18:571–573. 2. Ware L, Matthay MA. The acute respiratory distress syndrome. N. Engl. J. Med. 2000; 342:1334–1349. 3. Lindquist S. The heat shock response. Annu. Rev. Biochem. 1986; 55:1151–1191. 4. Sarge KD, Murphy SP, Morimoto RI. Activation of heat shock gene transcription by heat shock factor 1 involves oligomerization, acquisition of DNA binding activity, and nuclear localization and can occur in the absence of stress. Mol, Cell Biol. 1993; 13:1392–1407. 5. Chu B, Zhong R, Soncin F, Stevenson MA, Calderwood SK. Transcriptional activity of heat shock factor 1 at 37 degrees C is repressed through phosphorylation on two distinct serine residues by glycogen synthase kinase 3 and protein kinases Calpha and Czeta. J. Biol. Chem. 1998; 273:18640–18646. 6. Cotto JJ, Kline M, Morimoto RI. Activation of heat shock factor DNA binding precedes stressinduced serine phosphorylation. Evidence for multistep pathway of regulation. J. Biol. Chem. 1996; 271:3355–3358. 7. Kim D, Ouyang H, Li GC. Heat shock protein Hsp70 accelerates the recovery of heat-shocked mammalian cells through its modulation of heat shock transcription factor HSF-1. Proc. Natl. Acad. Sci. USA 1995; 92:2126–2130. 8. Minowada G, Welch WJ. Clinical implications of the stress response. J. Clin. Invest. 1995; 95:3– 12. 9. Morimoto RI, Tissieres A, Georgopoulos C. Stress Proteins in Biology and Medicine. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1990. 10. Otterbein LE, Choi AM. Heme oxygenase: colors of defense against cellular stress. Am. J. Physiol. 2000; 279:L1029–L1037. 11. Parsell DA, Lindquist S. The function of heat shock proteins in stress tolerance: degradation and reactivation of damaged proteins. Annu. Rev. Genet. 1993; 27:437–496. 12. Liao DF, Lin ZG, Bass AS, et al. Purification and identification of secreted oxidative-stressinduced factors from vascular smooth muscle cells. J. Biol. Chem. 2000; 275:189–195. 13. Pockley AG, Sheperd J, Corton J. Detection of heat shock protein 70 (Hsp70) and anti-Hsp70 antibodies in the serum of normal individuals. Immunol. Invest 1998; 27:367–377. 14. Pockley AG, Bulmer J, Hanks BM, Wright BH. Identification of human heat shock protein 60 (Hsp60) and anti-Hsp60 antibodies in the peripheral circulation of normal individuals. Cell Stress Chaperones 1999; 4:29–35. 15. Habich C, Baumgart K, Kolb H, Burkart V. The receptor for heat shock protein 60 on macrophages is saturable, specific, and distinct from receptors from other heat shock proteins. J. Immunol. 2002; 168:569–576. 16. Asea A, Kraeft SK, Kurt-Jones EA, et al. Hsp70 stimulates cytokine production through a CD14-dependent pathway, demonstrating its dual role as a chaperone and cytokine. Nat. Med. 2000; 6:435–442.

Heat shock response, heat shock proteins, and acute lung injury

271

17. Asea A, Rehli M, Kabingu E, et al. Novel signal transduction pathway utilized by extracellular Hsp70: role of TLR2 and TLR4. J. Biol. Chem. 2002; 277:15028–15034. 18. Young RA. Stress proteins and immunology. Annu. Rev. Immunol. 1990; 8:401–420. 19. De Graeff-Meeder ER, Zee Rvd, Rijkers GT, et al. Recognition of human Hsp60 by mononuclear cells from patients with juvenile arthritis. Lancet 1991; 337:1368–1372. 20. Wucherpfennig K, Newcombe J, Li H, Keddy C, Cuzne ML, Hafler DA. Gamma-delta T cell receptor repertoire in acute multiple sclerosis lesions. Proc. Natl. Acad. Sci. USA 1992; 89:4588–4592. 21. Elias D, markovits D, Reshef T, Zee Rvd, Cohen IR. Induction and therapy of autoimmune diabetes in the non-obese diabetic mouse by a 65-kDa heat shock protein. Proc. Natl. Acad. Sci. USA 1990; 87:1576–1580. 22. Pockley AG. Heat shock proteins, inflammation, and cardiovascular disease. Circulation 2002; 105:1012–1017. 23. van Roon JAG, Eden Wv, Roy Jv, Lafeber FJ, Bijlsma JW. Stimulation of suppressive T cell response by human but not bacterial 60 kDa heat shock protein in synovial fluid of patients with rhumatoid arthritis. J. Clin. Invest. 1997; 100:459–463. 24. Li Z, Menoret A, Srivastava P. Roles of heat shock proteins in antigen presentation and crosspresentation. Curr. Opin. Immunol. 2002; 14:45–51. 25. Hedge RS, Zuo J, Voellmy R, Welch WJ. Short-circuiting stress protein expression via tyrosine inhibitor, herbimycin. J. Cell. Physiol. 1995; 165:186–200. 26. Li GC. Induction of thermotolerance and enhanced heat shock protein synthesis in Chinese hamster fibroblasts by sodium arsenite and ethanol. J. Cell Physiol. 1983; 115:116–122. 27. Zou J, Guo Y, Guettouche T, Smith DF, Voellmy R. Repression of heat shock transcription HSF1 activation by Hsp90 (Hsp90 complex) that forms a stress-sensitive complex with HSF1. Cell 1998; 94:471–480. 28. Bellmann K, Wenz A, Radons J, Burkart V, Leemann R, Kolb H. Heat shock induces resistance in rat pancreatic islet cells against nitric oxide, oxygen radicals and streptozotocin. J. Clin. Invest. 1995; 95:2840–2845. 29. DeMeester SL, Buchman TG, Cobb JP. The heart shock paradox; does NF-kappaB determine cell fate? FASEB J. 2001; 15:270–274. 30. Diamond D, Parsian A, Hunt CR, et al. Redox factor-1 mediates the activation of AP-1 in HeLa and NIH 3T3 cells in response to heat shock. J. Biol. Chem. 1999; 274:16959–16964. 31. Wennborg A, Classon M, Klein G, Gabain Av. Down-regulation of c-myc expression after heat shock in human B-cell lines is independent of 5′mRNA sequences. Biol. Chem. Hoppe Seyler 1995; 376:671–680. 32. Lim CP, Jain N, Cao X. Stress-induced immediate-early gene egr-1 involves activation of p38/JNK1. Oncogene 1998; 16:2915–2926. 33. Adler V, Schaffer A, Kim J, Dolan L, Ronai Z. UV irradiation and heat shock mediate JNK activation via alternate pathways. J. Biol. Chem. 1995; 270: 26071–26077. 34. Ishibashi T, Bottaro DP, Michieli P, Kelley CA, Aaronson SA. A novel dual specificity phosphatase induced by serum stimulation and heat shock. J. Biol. Chem. 1994; 269:29897– 29902. 35. Nitta M, Okamura H, Aizawa S, Yamaizumi M. Heat shock induces transient p53-dependent cell cycle arrest at G1/S. Oncogene 1997; 15:561–568. 36. Wong HR, Mannix J, Rusnak JM, et al. The heat shock response attenuates lipopolysaccharidemediated apoptosis in cultured sheep pulmonary artery endothelial cells. Am. J. Respir. Cell Mol. Biol. 1996; 15:745–751. 37. Wong HR, Finder JD, Wasserloos K, Pitt BR. Expression of inducible nitric oxide synthase in cultured rat pulmonary artery smooth cells in inhibited by the heat shock response. Am. J. Physiol. 1995; 269:L843– L848. 38. Kohn G, Wong HR, Bshesh K, et al. Heat shock inhibits TNF-induced ICAM-1 expression in human endothelial cells via I kappa kinase inhibition. Shock 2002; 17:91–97.

Acute respiratory distress syndrome

272

39. Wong HR, Ryan M, Gebb S, Wispe JR. Selective and transient in vitro effects of heat shock on alveolar type II cell gene expression. Am. J. Physiol. 1997; 272: L132–L138. 40. Wong HR, Ryan M, Menendez IY, Denenberg A, Wispe JR. Heat shock protein induction protects human respiratory epithelium against nitric oxide-mediated cytotoxicity. Shock 1997; 8:213–218. 41. Grenert JP, Sullivan WP, Fadden P, et al. The amino-terminal domain of Hsp90 that binds geldanamycin is an ATP/ADP switch domain that regulates Hsp90 conformation. J. Biol. Chem. 1997; 272:23843–23850. 42. Malhotra V, Shanley TP, Pittet JF, Welch WJ, Wong HR. Geldanamycin inhibits NF-kappa B activation and interleukin-8 gene expression in cultured human respiratory epithelium. Am. J. Respir. Cell Mol Biol. 2001; 25: 92–97. 43. Vilar J, Edelson JD, Post M, Mullen JB, Slutsky AS. Induction of heat stress proteins is associated with decreased mortality in an animal model of acute lung injury. Am. J. Respir. Crit. Med. 1993; 147:177–181. 44. Vilar J, Ribeiro SP, J.B.M.Mullen, Kuliszewski M, Post M, Slutsky AS. Induction of heat shock response reduces mortality rate and organ damage in a sepsis-induced acute lung injury model. Crit. Care Med. 1994; 22:914–922. 45. Chu EK, Ribeiro SP, Slutsky AS. Heat stress increases survival in lipopolysaccharidestimulated rats. Crit. Care Med. 1997; 25:1727–1732. 46. Ribeiro SP, Villar J, Downey GP, Edelson JD, Slutsky AS. Sodium arsenite induces heat shock protein 72 kilodalton expression in the lungs and protects rats against sepsis. Crit. Care Med. 1994; 22:922–929. 47. Javadpour M, Kelly CJ, Chen G, Stokes K, Leahy A, Boucher-Hayes DJ. Thermotolerance induces heat shock protein 72 expression and protects against ischemia reperfusion-induced lung injury. Br J. Surg. 1998; 85:943–946. 48. Hiratsuka M, Yano M, Mora BN, Nagahiro I, Cooper JD, Patterson GA. Heat shock pretreatment protects pulmonary isografts from subsequent ischemia-reperfusion injury. J. Heart Lung Transplant. 1998; 17:1238–1246. 49. Hiratsuka M, Mora BN, Yano M, Mohanakumar T, Patterson A. Gene transfer of heat shock protein 70 protects lung grafts from ischemia-reperfusion injury. Ann. Thorac. Surg. 1999; 67:1421–1427. 50. Weiss YG, Bouwman A, Gehan B, Schears G, Raj N, Deutschman CS. Cecal ligation and double puncture impairs heat shock protein 70 expression in the lungs of rats. Shock 2000; 13:19–23. 51. Ribeiro SP, Villar J, Downey GP, Edelson JE, Slutsky AS. Effect of the stress response in septic rats and LPS-stimulated alveolar macrophages: evidence for TNF-alpha post-translational regulation. Am. J. Respir. Crit. Care Med. 1996; 154:1843–1850. 52. Pittet JF, Lu LN, Morris DG, et al. Reactive nitrogen species inhibit alveolar epithelial fluid transport by NF-kB dependent mechanisms after hemorrhagic shock in rats. J. Immunol. 2001; 166:6301–6310. 53. Pittet JF, Lu LN, Geiser T, Lee H, Matthay MA, Welch WJ. Stress preconditioning attenuates oxidative injury to the alveolar epithelium of the lung following haemorrhage in rats. J. Physiol. 2002; 538:583–597. 54. Warner RL, Paine R, Christensen PJ, et al. Lung sources and cytokine requirements for in vivo expression of inducible nitric oxide synthase. Am. J. Respir. Cell Mol. Biol. 1995; 12:649–661. 55. Thomas SC, Ryan MA, Shanley TP, Wong HR. Induction of the stress response with prostaglandin-A increases I-kappa B gene expression. FASEB J. 1998; 12:1371–1378. 56. Ayad O, Stark JM, Fiedler MA, Menendez IY, Ryan MA, Wong HR. The heat shock response inhibits RANTES gene expression on cultured lung epithelium. J. Immunol. 1998; 161:2594– 2599.

Heat shock response, heat shock proteins, and acute lung injury

273

57. Snyder YL, Guthrie L, Evans GF, Zuckerman S. Transcriptional inhibition of endotoxininduced monokine synthesis following heat shock in murine peritoneal macrophages. J. Leuk. Biol. 1992; 51:181–187. 58. Schmidt JA, Abdulla E. Down-regulation of IL-beta biosynthesis by inducers of the heat shock response. J. Immunol. 1988; 141:2027–2034. 59. Cahil CM, Waterman WR, Y.Xie, Auron PE, Calderwood SK. Transcriptional repression of the prointerleukin-1beta gene by heat shock factor 1. J. Biol. Chem. 1996; 271:24874–24879. 60. Feinstein DL, Galea E, Aquino DA, Li GC, Xu H, Reis DJ. Heat shock protein 70 suppresses astroglial-inducible nitric oxide synthase expression by decreasing NF-kappa B expression. J. Biol. Chem. 1996; 271:17224–17232. 61. Wong HR, Ryan M, Wispe JR. Stress response decreases NF-kB nuclear translocation and increases I-kBa expression in A549 cells. J. Clin. Invest. 1997; 99:2423–2428. 62. Pritts TA, Want Q, Sun X, et al. Induction of the stress response in vivo decreases NF-kappa B activity in jejunal mucosa of endotoxemic mice. Arch. Surg. 2000; 135:860–866. 63. Shanley TP, Ryan MA, Eaves-Pyles T, Wong HR. Heat shock inhibits phosphorylation of Ikappa B alpha. Shock 2000; 14:447–450. 64. Keyse SM, Emslie EZ. Oxidative stress and heat shock induce a human gene encoding a protein-tyrosine phosphatase. Nature 1992; 359:644–647. 65. Wong HR, M. Ryan, Menendez I, Wispe JR. Heat shock activates the I-kBa promoter and increases I-kBa mRNA expression. Cell Stress Chaperones 1999; 4:1–7. 66. Wong HR, Dunsmore KE, Denenberg AG. Transcriptional and post-transcriptional regulation of IkappaB gene expression by heat shock (abstr). Crit. Care Med. 2000; 28:A38. 67. Chen G, Kelly C, Stokes K, Wang JH, Leahy A, Boucher-Hayes D. Induction of heat shock protein 72kDa expression is associated with attenuation of ischemia-reperfusion induced microvascular injury. J. Surg. Res. 1996; 69:435–439. 68. Kubo H, Graham L, Doyle NA, Quinlan WM, Hogg JC, Doerschuck CM. Complement fragment-induced release of neutrophils from bone marrow and sequestration with pulmonary capillaries in rabbits. Blood 1998; 92:283–290. 69. Callahan TE, Marins J, Welch WJ, Horn JK. Heat shock attenuates oxidation and accelerates apoptosis in human neutrophils. J. Surg. Res. 1999; 85:317–322. 70. Parsey MV, Kaneko D, Shenkar R, Abraham E. Neutrophil apoptosis in the lung after hemorrhage or endotoxemia: apoptosis and migration are independent of IL-1beta. Clin. Immunol. 1999; 91:219–225. 71. Stokes KY, Abdih HK, Kelly CJ, Redmond HP, Bouchier-Hayes DJ. Thermotolerance attenuates ischemia-reperfusion-induced renal injury and increased expression of ICAM-1. Transplantation 1996; 62:1143–1149. 72. Haslett C. Granulocyte apoptosis and its role in the resolution and control lung inflammation. Am. J. Respir. Crit. Care Med. 1999; 160:S5–S11. 73. Wang R, Zagariya A, Ang E, Ibarra-Sunga O, Uhal BD. Fas-induced apoptosis of alveolar epithelial cells requires ANG II generation and receptor interaction. Am. J. Physiol. 1999; 277:L1245–L1250. 74. Lavoie JN, Lambert H, Hickey E, Weber LA, Landry J. Modulation of thermoresistance and actin filament stability accompanies phosphorylation-induced changes in the oligomeric structure of heat shock protein 27. Mol. Cell. Biol. 1995; 15:505–516. 75. Gabai VL, Meriin AB, Mosser DD, et al. Hsp70 prevents activation of stress kinases. A novel pathway of cellular thermotolerance. J. Biol. Chem. 1997; 272:18033–18037. 76. Gabai VL, Meriin AB, Yaglom JA, Wei JY, Mosser DD, Sherman MY. Suppression of stress kinase JNK is involved in Hsp72-mediated protection of myogenic cells from transient energy deprivation. Hsp72 alleviates the stress-induced inhibition of JNK dephosphorylation. J. Biol. Chem. 2000; 275:38088–38094. 77. Park HS, Lee JS, Huh SH, Seo JS, Choi EJ. Hsp72 functions as a natural inhibitory protein of the c-Jun N-terminal kinase. EMBO J. 2001; 20:446–456.

Acute respiratory distress syndrome

274

78. Wong HR, Menendez IY, Ryan MA, Denenberg AG, Wispe JR. Increased expression of heat shock protein 70 protects A549 cells against hyperoxia. Am. J. Physiol. 1998; 275:L836–L841. 79. Otterbein LE, Alam J, Choi AMK. Hemoglobin provides protection against lethal endotoxemia in rats: the role of heme oxygenase 1. Am. J. Respir. Cell Mol. Biol. 1995; 13:595–601. 80. Choi AMK, Alam J. Heme oxygenase 1: function, regulation and implication of a novel stressinducible protein in oxidant-induced lung injury. Am. J. Respir. Cell Mol. Biol. 1996; 15:9–19. 81. Otterbein LE, Chin BY, Otterbein SL, Lowe JB, Fessler H, Choi AMK. Mechanism of hemoglobin-induced protection against endotoxemia in rat: a ferritin-independent pathway. Am. J. Physiol. 1997; 272:L268–L275. 82. Otterbein LE, Kolls JK, Mantel LL, Cook JL, Alam J, Choi AMK. Exogenous administration of heme oxygenase 1 by gene transfer provides protection against hyperoxia-induced lung injury. J. Clin. Invest. 1999; 103:1047–1054. 83. Lee TS, Chau LY. Heme-oxygenase-1 mediates the anti-inflammatory effect of interleukin-10 in mice. Nat. Med. 2002; 8:240–246. 84. Otterbein LE, Bach FH, Alam J, et al. Carbon monoxide has anti-inflammatory effects involving the mitogen-activated protein kinase pathway. Nat. Med. 2000; 6:422–428. 85. Otterbein LE, Mantell LL, Choi AMK. Carbon monoxide provides protection against hyperoxic lung injury. Am. J. Physiol. 1999; 276:L688–L694. 86. Frank AD, Wagener TG, Silva JLD, et al. Differential effects of heme oxygenase isoforms on heme mediation of endothelial intracellular adhesion molecule 1 expression. J. Pharmacol. Exp. Ther 1999; 291:416–423. 87. Fujita T, Toida K, Yan SH, Naka Y, Yet SF, Pinsky DJ. Paradoxal rescue from ischemic lung injury by inhaled carbon monoxide driven by derepression of fibrinolysis. Nat. Med. 2001; 7:598–604. 88. Llesuy SF, Tomaro ML. Heme oxygenase and oxidative stress. Evidence of involvement of bilirubin as physiological protector against oxidative damage. Biochim. Biophys. Acta 1994; 1223:9–14. 89. Hayashi S, Takamyia R, Yamagushi T, et al. Induction of heme oxygenase 1 suppresses venular leukocyte adhesion elicited by oxidative stress. Circ. Res. 1999; 85:663–671. 90. Datta PK, Koukouritaki SB, Hopp KA, Lianos EA. Heme oxygenase 1 induction attenuates inducible nitric oxide synthase expression and proteinuria in glomerulonephritis. J. Am. Soc. Nephrol. 1999; 10:2540–2550. 91. Xiao X, Zuo X, Davis AA, McMillan DR, Curry BB, Richardson JA, Benjamin IJ. HSF-1 is required for extra-embryonic development, postnatal growth and protection during inflammation. EMBO J. 1999; 18:5943–5952. 92. Malhotra V, Eaves-Pyles T, Odoms K, Quaid G, Shanley TP, Wong HR. Heat shock inhibits activation of NF-kappaB in the absence of heat shock factor-1. Biochem. Biophys. Res. Comm. 2002; 291:453–457. 93. Zhu S, Ware LB, Geiser T, Matthay MA, Matalon S. Increased levels of nitrate and surfactant protein A nitration in the pulmonary edema fluid of patients with acute lung injury. Am. J. Respir. Crit. Care Med. 2001; 163:166–172. 94. Ware LB, Matthay MA. Alveolar epithelial fluid clearance is impaired in the majority of patients with acute lung injury and the acute respiratory distress syndrome. Am. J. Respir. Crit. Care Med. 2001; 163:1376–1383.

12 Mechanisms of Fibroproliferation in Acute Lung Injury MITCHELL A.OLMAN University of Alabama at Birmingham Birmingham, Alabama, U.S.A.

I. Introduction The fibroproliferative phase of acute lung injury (ALI) is characterized by impaired gas exchange and severely reduced lung compliance as a result of progressive accumulation of interstitial and alveolar cells and matrix. The critical feature of this late stage of the injury-repair continuum is the disruption of normal alveolar architecture. This disruption leads to dysfunction in the liquid and protein barrier functions of the lung and in the diffusion properties of the alveolar-capillary structures. The restoration of normal lung architecture and function requires temporal and spatial coordination of alveolar proteolysis, cell migration and proliferation, angiogenesis, matrix synthesis, epithelial cell repopulation, and apoptosis (1–6). These responses are dependent on complex and dynamic interactions between growth factors/cytokines/chemokines, cells of diverse lineage, and their surrounding provisional matrix. It has been assumed that the intact alveolar basement membrane provides a critical roadmap for the remodeling and restoration of normal lung architecture after injury. Unfortunately, following a myriad of intravascular or inhaled insults, a series of temporally and spatially overlapping injury and repair events all too often results in a dysfunctional, fibrotic lung. This chapter will discuss the clinical, biochemical, and cell-matrix interactive events that lead to fibroproliferative repair after lung injury.

II. Clinical Observations of the Fibroproliferative Phase of Lung Injury A. Clinically Observable Features of the Fibroproliferative Phase of Lung Injury Although a plethora of injurious agents can result in acute lung injury, sepsis and major trauma together account for roughly 30–50% of the approximately 100,000 new cases per year in the United States (7). A schematized diagram describes the time course of the major histopathologically observable findings of the fibroproliferative response. Based on these well-characterized findings seen following diffuse alveolar damage, several phases

Acute respiratory distress syndrome

276

of injury/repair have been defined: the acute or exudative phase (1–6 days), the subacute fibroproliferative phase (4–10 days), and the chronic fibrotic phase (after 8–14 days) (Fig. 1). Certain clinical findings point to the development of a severe fibroproliferative response after diffuse alveolar damage (8, 9). Symptoms and signs include the development of a rapid respiratory rate with radiographic abnormalities that change from an alveolar pattern to a interstitial appearance with an increased number of bullae in dependent areas of the lung. The low lung compliance often seen in the exudative phase persists or worsens, and an increase in the fraction of ventilation that goes to poorly perfused lung areas (high V/Q areas) further complicates attempts to maintain carbon dioxide elimination. The low lung compliance, high V/Q areas, and positive pressure mechanical ventilation conspire together to increase the alveolar pressures required to maintain a sustainable rate of carbon dioxide elimination. Fortunately, the alveolarcapillary exchange of oxygen improves, resulting in a reduced requirement for high fractions of inspired oxygen to maintain adequate arterial hemoglobin saturation. In this setting, reductions, not increases, in positive end expiratory pressure, tidal volume, and/or minute ventilation may restore carbon dioxide elimination. Published low tidal volume/permissive hypercapnic ventilation protocols that improve survival and reduce the need for mechanical ventilation in patients with acute lung injury were not designed, and did not specifically assess, the role of ventilator-induced lung injury in the fibroproliferative response (10). It is unclear whether the low tidal volume protocols have their beneficial effect on mortality due to reduced lung injury, inflammation, or improved repair. Nonetheless, these data do

Figure 1 Alveolar injury and remodeling, (a) The normal alveolar

Mechanisms of fibroproliferation in acute lung injury

277

wall (left) is composed of fibroblasts (hatched cells), capillary endothelial cells and flattened type I epithelial cells. During the acute phase of injury (middle), fibroblasts begin to migrate through interruptions in the denuded alveolar basal lamina into a fibrinous provisional matrix. During the subacute fibroproliferative phase (right), the provisional matrix is biochemically remodeled by resident myofibroblasts, contracted and repopulated by proliferating cuboidal type II epithelial cells. Fibrosis in this schema is by accretion, (b) In this schema, the denuded basal lamina folds upon itself and/or adheres to an adjacent basal lamina, which is then repopulated by type II epithelial cells. Fibrosis in this schema is by collapse induration. (From Ref. 36.) support a low tidal volume/permissive hypercapnia approach during the fibroproliferative phase by simplifying management of those with low compliance and high dead space fraction. Patients dying of multiple organ failure/sepsis with acute lung injury have histopathological evidence of fibroproliferation (11, 12). However, definitive data regarding the significance of a fibroproliferative response on mortality are hard to find. In those patients with acute respiratory distress syndrome (ARDS) who die more than 3 days after study entry, respiratory failure was estimated to be a significant contributor to death in 16%, many of which had underlying sepsis syndrome (13). The underlying etiology of the lung injury per se (e.g., sepsis vs. trauma vs. other) has consistently been shown to carry a greater relative risk of mortality than that noted for the respiratory failure due to the fibroproliferative response (7, 13). Confounding comorbidities and the potential for prolonged alveolar remodeling to increase the incidence of sepsis further cloud the issue. Mortality is clearly not the optimal measure of the incidence and/or severity of the fibroproliferative response to injury. Clinical follow-up of ARDS patients who survive to hospital discharge indicates that the development of ARDS engenders a significantly greater reduction in pulmonary disease-specific quality-of-life scores compared with matched trauma and sepsis patients who did not develop ARDS (14, 15). Clinical markers of fibroproliferation in ARDS (prolonged mechanical ventilatory course and poor cumulative lung injury scores)

Acute respiratory distress syndrome

278

correlate with an impaired recovery of pulmonary function and a poorer self-perceived health score at 1 year following injury (16). Taken together, these data indicate that the development of a fibroproliferative response engenders an excess long-term symptomatic and physiological morbidity. B. Cellular and Biochemical Features of the Fibroproliferative Phase of Lung Injury The phases of injury/repair described above represent the predominant histopathological findings (Fig. 1). There is considerable temporal overlap of these phases and considerable histopathological spatial variability within a given patient, which may confound the interpretation of studies using selected lung biopsy specimens (17). Bronchoalveolar lavage (BAL) studies reveal that patients with persistent clinical acute lung injury have cellular and biochemical evidence of sustained inflammatory alveolitis. For example, those patients who require prolonged mechanical ventilation demonstrate sustained elevations (up to 2.1 days) of a number of pro-inflammatory cytokines/chemokines including IL-1β, IL-8, ENA-78, MCP-1, and MIP-1-alpha (18). Furthermore, in those who survive the initial septic insult, the transition from a neutrophilic to a mononuclear alveolitis over 10–14 days portends future survival, as compared with those with sustained alveolar neutrophilia (18). Thus, in clinical acute lung injury the alveolus is repaired/ remodeled in the context of ongoing inflammation. Second, initiation of the fibroproliferative process begins early after injury and is temporally superimposed upon the early acute inflammatory response. Total lung collagen, the predominant matrix protein in lung, is increased by 10 days after clinical onset of ARDS in patients who die (17). Microscopic studies of lung tissue from patients with ARDS reveals an expansion of the interstitial compartment by cells and fibers and interstitial/ alveolar fibrosis as early as 3–5 days after injury (11, 12). Immunohistochemical studies of lung tissue from patients with ARDS reveals early disruption of the alveolar basement membrane and alveolar deposition of interstitial collagens, primarily type III collagen, followed by type I collagen predominance (19). Furthermore, a soluble collagen precursor, type III procollagen peptide (PCP III), is elevated in edema fluid obtained within hours of intubation from patients with ARDS compared to ventilated patients with hydrostatic pulmonary edema (20). Elevated levels of PCP III identifies those patients with ARDS who will have prolonged mechanical ventilation and a poor survival (20, 21). These data demonstrate that the fibroproliferative process is initiated early after lung injury. Markers of collagen synthesis may be better choices to measure the incidence and severity of the fibroproliferation after diffuse alveolar damage. Prolonged (up to 14 days) elevations of PCP III in the BAL independently carries a relative risk of approximately threefold for mortality (21). Positive correlation was also noted at 7 and 14 days between the extent of elevation of the PCP III and a physiological score (based on static compliance, level of PEEP, PaO2 and FIO2) (21). Fibroblast mitogenic activity in BAL fluid at 7 days after injury is higher in nonsurvivors with ARDS compared with ARDS survivors (22). In a smaller study (n=29) all ARDS patients had elevated plasma PCP III on day 1, but only those who did not improve their lung injury scores over the next week had a rise in the plasma levels of PCP III over that time period (23). In consecutive

Mechanisms of fibroproliferation in acute lung injury

279

patients with ARDS (n=25) who underwent transbronchial biopsy 10 days after ARDS onset, histopathological evidence of fibrosis (seen in 14/25) correlated with mortality (24). Finally, elevated PCP III levels in the BAL have been correlated with histological evidence of interstitial and intra-alveolar fibrosis, albeit in only a small number of patients (n=6) with ARDS (25). These data all support the physiological and prognostic significance of fibroproliferation in acute lung injury. Both synthesis and degradation of collagen will lead to increases in PCP III in the BAL. The amino-terminal peptide of procollagen types I and III are removed from procollagen during extracellular processing as a prerequisite for fibrilogenesis. However, a fraction of the mature collagen molecules in the fibril retain their amino-terminal peptide. Thus, both tissue injury and degradation of lung collagen will lead to the generation of soluble fragments containing the amino-terminal peptide, as would aborted fibrilogenesis (25). In fact, analysis of the soluble collagen fragments demonstrates a wide variance in their molecular weights, suggesting that both collagen synthesis and degradation are superimposed. It remains unresolved as to the fraction of total soluble pro-collagen peptide that is generated from de novo synthesis relative to that generated from enzymatic degradation of tissue collagen and/or aborted fibrilogenesis in acute lung injury.

III. Histopathology of the Fibroproliferative Response A. Normal Lung Interstitium At the air/blood interface of the normal alveolus, gas exchange occurs across an anatomical barrier 0.3 µm at its smallest consisting of type I epithelial cells, a fused epithelial-capillary endothelial basement membrane, and a capillary endothelial cell (Fig. 1). Cuboidal type II epithelial cells normally occupy only 5% of the normal alveolar surface area but can replicate and repopulate a basement membrane that has been denuded of the flat type I cells as a consequence of lung injury. The lung interstitium normally occupies the space bounded by the endothelial and epithelial basement membranes in area of the alveolus where they remain separate and is composed of cells and connective tissue macromolecules. Cell types include ontologically mesenchymal (fibroblast, interstitial cells, myofibroblasts, pericytes, and smooth muscle cells) as well as immunomodulatory cells (lymphocytes, macrophages, dendritic cells, and neutrophils). The connective tissue components are organized into endothelial and epithelial basement membranes. The basement membrane is composed largely of collagen types IV and V, proteoglycans, and glycoproteins including laminin and fibronectin. The normal interstitial stroma is composed of the tensile strength-providing fibrillar collagens (type I and III), with elastin being the largest noncollagenous component and that which provides alveolar elasticity. Basement membrane and stromal glycosaminoglycans including hylaronan, heparan sulfate, and chondroitin sulfate, and proteoglycans are present in trace amounts in the uninjured lung. These molecules play a role in maintaining the integrity of the matrix and can modulate matrix homeostasis through their interaction with numerous growth factors (2, 26). Extracellular matrix components interact with each other covalently and noncovalently to form insoluble connective tissue

Acute respiratory distress syndrome

280

that is largely organized into fiber networks that provide structural support and distribute physical strain. Furthermore, intact matrix and its proteolytically derived fragments are active participants in cellular processes of migration, adhesion, attachment, and proliferation, as well as immune-effector functions (27–31). These are accomplished largely through cell surface receptor-mediated signaling events as well as modulation of growth factor and proteolytic activity that occur due to cellular interactions with the matrix (32–34). B. Optimal and Dysfunctional Repair A number of different anatomical patterns of fibrosis can be seen together in a single patient (35). Fibrosis by accretion can be viewed as incorporation of provisional matrix/exudate into thickened alveolar ducts and alveolar walls. This results in complete occlusion of the alveolar ducts and deposition of fibrotic tissue at the periphery of alveoli. Alveoloseptal fibrosis is a consequence of in situ deposition of excess connective tissue within the alveolar/ septal walls. Collapse induration is a consequence of irregular folding and adherence of opposing adjacent basement membranes with complete loss of alveolar spaces. The etiological and pathophysiological significance of the different patterns of fibrosis are unknown, but they demonstrate that a multiplicity of pathways contribute ot the fibroproliferation. In the setting of acute lung injury, the repair response determines whether normal alveolar architecture is restored or an architecturally disrupted, fibrotic process ensues. Within hours of the injury, massive type I cell injury and death is evident, leaving large areas of the epithelial basement membrane denuded and highly permeable (11, 12, 36). This allows free flow of plasma proteins and fosters inflammatory cell transmigration into the alveolar space. Local generation and action of cytokines/growth factors that have overlapping and/or competing immunomodulatory and fibroproliferative effects add to the complexity of the response. The optimal repair process involves coordinated alveolar type II cell replication, migration, and differentiation in order to repopulate the denuded basement membrane along with clearance (diffusive, proteolytic, and phagocytic) of the alveolar exudate and apoptotic cells, with remodeling of the alveolar and interstitial matrix. In contrast, repair that ultimately leads to a fibrotic alveolar space is characterized by the persistence of the alveolar provisional matrix consisting of fibrin, fibronectin, and hyaluronan fragments, as well as the proliferation and migration of interstitium-derived myofibroblasts through interruptions in the basement membrane and a neovascularization (2, 12, 32). This alveolar tissue, somewhat akin to wound granulation tissue, is gradually remodeled through proteolysis and matrix synthesis and replaced by collagen. On the endothelial side of the alveolar capillary barrier, early cellular injury leads to microvascular coagulation and endothelial cell upregulation of leukocyte adhesion molecules, thereby promoting inflammatory cell transmigration (37). With failure of normal parenchymal repair, the alveolar capillaries are lost (38). The larger pulmonary arteries demonstrate intimal fibrosis, medial thickening, and reduced luminal cross-sectional area resulting in progressive pulmonary arterial hypertension (38, 39).

Mechanisms of fibroproliferation in acute lung injury

281

C. Granulation Tissue The hallmark of healing wounds is the formation and remodeling of granulation tissue. Granulation tissue is composed of fibroblasts, myofibroblasts, macrophages, and endothelial cells in a complex milieu of growth factors and matrix glycoproteins, undergoing a dynamic and reciprocal interaction that ultimately results in a collagenous healed wound. In the context of diffuse alveolar damage, the prevailing hypothesis states that restoration of normal alveolar architecture after lung injury requires an intact alveolar basement membrane. Although animal models and injured human lung amply demonstrate cells migrating through disrupted basement membranes and reduplicated, irregular basement membranes in fibrotic lung, this hypothesis remains to be definitively proven. In acute lung injury, alveolar and interstitial granulation tissue is formed and ultimately remodels a provisional matrix, thereby changing its matrix and cellular components. Using observations derived from wound healing in skin and other organs, immunohistochemical staining of human lung injury and experimental model systems, a paradigm of alveolar granulation tissue biology and biochemistry can be deduced (40). Within the first few days after injury, the major insoluble matrix constituents of the alveolar exudate are plasma-derived, cross-linked fibrin, fibronectin, vitronectin, and platelet-derived thrombospondin. With the development of early granulation tissue (days 3–8), alternatively spliced, embryonic forms of fibronectin are synthesized by resident fibroblasts and macrophages, and several antiadhesive proteins appear (Thrombospondin, SPARC-secreted protein acidic and rich in cysteine, and Tenascin). After approximately 5–7 days, collagens type I and III become a more prominent matrix component, as do the proteoglycans versican and decorin and fragments of hyaluronan. During this remodeling process, dynamic and reciprocal cell matrix-growth factor interactions modulate the remodeling.

IV. Interactions Among Growth Factors and the Extracellular Matrix A. Cell Matrix Interactions Modulate Transforming Growth Factor-β Activation A number of cytokines, chemokines, and growth factors have been pathogenically implicated in the fibroproliferative response after lung injury. This topic has been extensively reviewed (41–44). We will focus on the growth factor transforming growth factor-beta (TGF-β), as there are extensive data supporting its role in the fibroproliferative repair response, and its interaction with the matrix illustrates several key growth factor-matrix interactions (45). Because many cell matrix reciprocal interations affect the activity, distribution, and signaling of TGF-β, its biology will be used to illustrate key concepts regarding the complexity of growth factor activity in lung injury. TFG-β is a 30,000 kDa protein with three isoforms (TFG-β1, TFG-β2, and TGF-β3) that largely overlap in function. It is produced in a highly regulated manner by many cell

Acute respiratory distress syndrome

282

types and is pleotropic in its actions (46). It is a member of the TGF-β superfamily of genes that also includes bone morphogenetic protein 2/4, activin, and several Drosophila and C. legans homologs that play important roles in embryonic patterning, organogenesis, immune system regulation, and tissue homeostasis (47, 48). TGF-β is produced in latent form by virtue of its noncovalent association with latency-associated peptide (LAP). Its secretion is enhanced though covalent association of latent TGF-β with latency TGF-β-binding protein (LTBP). LTBP is directly relevant to matrix remodeling as LTBP will target latent TGF-β to the extracellular matrix of fibroblasts (49). Furthermore, release of matrix-bound latent TGF-β is accomplished through proteolytic cleavage by proteases involved in lung injury, including elastase and plasmin. In fact, mice lacking neutrophil elastase have a reduced fibrotic response to bleomycin, possibly due to cleavage of LTBP and loss of matrix-bound TGF-β (50). The pleotrophic nature of TGF-β’s actions further underscores the difficulty in sifting out the precise biological mechanism(s) of its effect on the fibroproliferative response. For example, epithelial cell proliferation is inhibited while fibroblast and mesenchymal cell proliferation may either be enhanced, inhibited, or unchanged by TGF-β. TGF-β is a potent inducer of numerous provisional and terminal matrix proteins, including collagen, fibronectin, laminin, and proteoglycans. It also induces the secretion of serine protease and metalloprotease inhibitors and inhibits collagenase expression, effects that would be predicted to act to block proteolysis of the newly synthesized matrix and lead to net matrix accumulation (51). There is much evidence that TGF-β is an important immunomodulatory molecule. For example, TGF-β null mice experience an autoimmune phenotype and die several weeks after birth. TGF-β can up or downregulate the growth and activity of leukocytes in a manner that is dependent of the cell type, its state of differentiation, and the presence or absence of other growth factors and cytokines (51). The importance of TGF-β in pulmonary fibroproliferation is well documented. TGF-β is highly expressed in actively remodeling areas of lung in human pulmonary fibrosis and bleomycin lung injury in rodents, largely in macrophages (and epithelial cells) and in fibrotic areas (52, 53). Bleomycin-induced lung injury and fibrosis is abrogated in mice by TGF-β-neutralizing antibodies (54). The activation of TGF-β from its latent form is a critical regulatory step in TGF-β in vivo bioactivity. This is amply demonstrable for lung fibrosis (53–57). Chronic adenovirally mediated overexpression of latent TGF-β in rodent lungs was benign, while overexpression of constitutively active TGF-β under the same conditions resulted in a pronounced inflammatory and fibrotic phenotype (58). In vitro, the active conformation of TGF-β is generated through exposure to acidic conditions, through proteolytic cleavage (e.g., by plasmin, calpain, transglutaminase, MMP-2, and MMP-9), interaction with extracelluar matrix proteins including thrombospondin, acidification of cellular microenvironments, reactive oxygen species treatment with retinoids, glucocorticoids, and vitamin D (48, 59–63). The predominant mechanism of in vivo activation of TGF-β during human lung injury remains unknown. Another modulatory step with potential significance to TGF-β activity in human lung injury is its binding to other provisional matrix proteins and/ or soluble proteins that concentrate at sites of lung inflammation. In the context of lung injury a given cytokine may bind to extracellular matrix proteins (soluble or insoluble) and thereby become sequestered at sites of injury/ repair. Its conformation may be altered to become more or less active, more or less susceptible to proteolytic degradation or activation, and/or

Mechanisms of fibroproliferation in acute lung injury

283

cleared more or less rapidly (64). Furthermore, a number of soluble single cytokinebinding proteins/receptors (shed from the cell surface or naturally occurring) are present in the acutely injured alveolus in stochiometric excess over their ligand cytokines (42, 65). The complex of cytokine/receptor may alter clearance and may inhibit (i.e., IL-1) or enhance (i.e., IL-6) biological activity. Several recently described mechanisms are of potential relevance to human lung injury. The epithelial cell integrin αvβ6 is a ligand for latent TGF-β, and this integrin can participate in epithelial cell-specific, spatially restricted TGF-β activation (63). Furthermore, mice lacking the β6 integrin demonstrate a reduced pulmonary fibrotic response to bleomycin (63). These mice also are protected from the TGF-β-dependent early (5 days) pulmonary edema seen in the bleomycin and the bacterial endotoxin models of lung injury (66). The deadhesive protein thrombospondin is a prominent component of granulation tissue matrix. Thrombospondin-1 (TSP-1) binds to latent TGFβ and participates in plasmin and CD36-dependent TGF-β activation on macrophage cell surfaces. Treatment of TSP-1 null mice with a TSP-derived pep tide activates TGF-β and normalizes the phenotype (59– 62). Its role in lung injury-related TGF-β activation remains to be determined, although peptide homologues of the TSP-1 receptor, CD36, will ameliorate bleomycin-induced pulmonary fibrosis in rats (67). Decorin is a 100 kDa proteoglycan with two binding sites for TGF-β (68). While only trace amounts of proteoglycans are seen in normal lung alveolar wall, they localize to sites of inflammation and granulation tissue remodeling in lung injury (69). Immunohistochemical stain shows that versican co-localizes with alpha smooth muscle actin-positive myofibroblasts in areas of organizing alveolar exudate, while hyaluronan and biglycan were weakly expressed, and decorin appeared to stain intracellularly in myofibroblasts and epithelial cells (69). Myofibroblasts (alpha smooth muscle actinpositive) cells in proliferative lesions in ARDS stain intracellularly for decorin (69). Its pathophysiological role in human lung injury is not yet certain, but decorin shows promise as a therapeutic molecule. Adenovirally mediated decorin gene transfer as well as twice-weekly intratracheal instillations of decorin will ameliorate bleomycin-induced hydroxyproline accumulation and neutrophil recruitment in mice and hamsters (70, 71). Another potential TGF-β interaction relevant to human lung injury is with α2macroglobulin (α-2 MG). α2-Macroglobulin is a high molecular weight plasma protein tetramer originally characterized as a broad specificity protease inhibitor that is also produced by fibrolasts and macrophages. As such, the α-2 MG would be predicted to concentrate and exert its effect in lung injury in a spatially and temporally restricted manner that corresponds to sites of impairment of the alveolar-capillary barrier and of inflammation/remodeling (64). α2-Macroglobulin is present in a native form, which does not bind to its clearance receptor, α-2 MG/LRP, and a protease-activated form, which binds to and is cleared by its receptor (64). The consequences of this interaction may be to clear any bound cytokine and/or modulate cytokine-cognate receptor signaling or initiate LRP-dependent signaling (72, 73). TGF-β binds both forms of α-2 MG with dissociation constants in the nanomolar range. TGF-β bound to plasma α-2 MG enhances its clearance, while its inhibitory effect on lung epithelial cells is abrogated upon binding to α-2 MG (74). Several proteases induced in lung injury could potentially regulate TGFβ activity through the activation of α-2 MG, followed by clearance of the TGF-α-2 MG complex (64). Alternatively, acidic microenvironments and/or matrix heparin could

Acute respiratory distress syndrome

284

release the sequestered TGF-β from the α-2 MG and free up TGF-β to initiate cell signaling though its receptors. It is possible, although unstudied, that plasmin could have a triple potentiating action by releasing TGF-β from bound α-2 MG and from matrixbound LTBP, as well as directly activating the free TGF-β (64). The role of α-2 MGTGF-β interactions in human lung injury are potentially relevant but remain unknown. Modulation of other relevant factors including IL-1β, which binds only to the activated form of α-2 MG, and PDGF B-isoforms by α-2 MG may be important in the regulation of fibrogenesis in lung injury (75). Binding of TGF-β to its type II receptor induces serine phosphorylation of the GS domain of the type I receptor by type II receptors, resulting in the initiation of downstream signaling. Both receptors are expressed on many cells from normal and diseased human lungs. The role of modulation of TGF-β signaling at the receptor activation level in lung injury is largely unknown. However; areas of chronic fibrosis in patients with idiopathic pulmonary fibrosis contained cells with relatively fewer type I receptors (76). Type I receptors directly phosphorylate the R-Smads (receptor-regulated Smads 1, 2, 3, 5, and 8) with the carboxy-terminal SSXS motif (48). Smads 2 and 3 specifically transmit the TGF-β signal. The R-Smads, when phosphorylated, form a heteromeric complex with a Co-Smad (common Smad 4, 4B) in the cytoplasm, and the complex translocates into the nucleus (48). Smads 6 and 7 are inhibitory. In fact, adenovirally mediated overexpression of Smad 7 will ameliorate bleomycin-induced pulmonary fibrosis in mice (77). Potential modulation of TGF-β signaling can occur through the ras-MAPK pathway, which phosphorylates Smads 2 and 3, blocking their translocation into the nucleus. Cell surface integrin ligation to the extracellular matrix can, under certain conditions, activate the MAPK pathway, outlining another potential mechanism whereby cell-matrix interactions can modulate TGF-β signaling. Once in the nucleus, transcriptional activation and repression is more efficient upon binding of nuclear cofactors, including AP-1, TFE3, p300/CBP, as well as the TGF-β-inducible interaction of Smad 2 with the homeobox protein TGIF (48). The transcription factor activator protein-1 (AP-1) is thought to play an important role in controlling the expression of many TGF-β-responsive genes, including matrix proteins and proteases/protease inhibitors. Cooperatively between Smads and AP-1 may participate in fibrogenic regulation at multiple levels. AP-1 transcriptional activity is highly responsive to signals derived from ligation of growth factor receptors as well as those derived from matrix protein ligation by integrins. Both c-fos and c-jun bind to Smad 3, with c-fos binding in the MH2 domain. For example, transcriptional activation occurs in a complex of Smad/AP-1 binding to adjacent AP-1 and Smad binding sequences in at least one highly TGF-β responsive promoter (PAI-1) (47). A number of other transcription factors have similarly been shown to partner with Smad and modulate TGF-β signaling, including CREB-binding protein and SP-1, providing evidence for the complexity and multiplicity of pathways that modulate TGF-β action (47). TGF-β pathways cross-talk with other cytokines/chemokines/growth factors relevant to the fibroproliferative response. For example, the Th1 cytokine, interferon-γ (IFN-γ), is antifibrogenic in numerous experimental systems, and administration of IFN-γ shows promise in early clinical trials in human chronic pulmonary fibrosis (78, 79). IFN-γ, among other things, inhibits TGF-β signaling through the STAT-induced transcription of the inhibitory Smad-7, an inhibitor of R-Smad phosphorylation (47, 51). Both TNF-α and

Mechanisms of fibroproliferation in acute lung injury

285

TGF-β neutralization can ameliorate bleomycin-induced pulmonary fibrosis and TNF-α is a prominent cytokine that is increased and partially active in early lung injury (80). Overexpression of active TGF-β in mice lacking the TNF receptors leads to pulmonary fibrosis, indicating that TGF-β is either redundant or downstream but not upstream of TNF-α in the fibrogenic process (81). The use of genetically altered mice led to the discovery of novel and unexpected mechanisms for modulating TGF-β activity in vivo. Much remains to be learned in this interesting and complicated are area that is rife with therapeutic potential. B. Matrix and Cell Surface Proteoglycans and Glycosaminoglycans Modulate Lung Remodeling Processes Several known families of proteoglycans are molecules comprised of a protein core (30– 500 kDa) with one or more attached glycosaminoglycan chains (5–70 kDa). These include the lecticans, which have an N-terminal domain that binds the glycosaminoglycan, hylauronan, and a carboxy-terminal lectin-like region. In the lung, versican is the best documented lectican. Members of the small proteoglycan family include decorin and biglycan, which have leucine-rich repeats as well as an EGF-like domain in their core protein, and chondroitin sulfate/dermatan sulfate or keratin sulfate side chains. Members of this family can bind to TGF-β. Another family of proteoglycans are plasma membrane-associated syndecans and glypicans, while perlecan is associated with the extracellular basement membrane. These proteoglycans have heparan sulfate side chains that are critical in binding to and modulating the function of a number of growth factors. For example, fibroblast growth and contraction after fibronectin assembly may require the interaction of cell surface heparan sulfate proteoglycans as well as an RGD-dependent integrin binding to distinct regions of fibronectin (82, 83). Hylauronan is an N-sulfated glycoasminoglycan polymer (103 kDa) with no core protein. Normal lung blood vessels stain for the proteoglycans versican in the media and decorin, biglycan, and hyalauronan in the adventitia (69). Sub-epithelial connective tissue in the airways stains for versican, decorin, and biglycan, while the normal lung interstitium has only trace amounts of any proteoglycan (69). In acute lung injury, versican is the major component of the organizing alveolar exudate and is found in thickened alveolar walls and colocalizes with alpha smooth muscle actin-positive myofibroblasts (69). Hyaluronan and fragments of hyaluronan and biglycan are also seen in the alveolar exudative areas, while decorin appears intracellularly in myofibrobasts (69, 84). As mentioned above adenovirally mediated decorin gene transfer as well as twiceweekly intratracheal instillations of decorin ameliorate bleomycin-induced pulmonary fibrosis in mice and hamsters (70, 71). The observation that biglycan is not similarly effective suggests that the collagenbinding capacity of decorin plays a role in sequencing TGF-β in the matrix. However, there is in vitro evidence for direct decorin-mediated signaling through ErbB receptors demonstrating a direct cellular effect of decorin. Not all the currently known proteoglycans have been examined in human or experimental lung injury. For example, syndecans are transmembrane proteoglycans that also can be proteolytically shed into the matrix. There is increasing evidence of their role in modulating growth factor signaling, cell adhesion, cell surface assembly of matrix

Acute respiratory distress syndrome

286

proteins, and cell growth (85). The transmembrane core protein may also participate in cell signaling and modulate integrin-initiated signaling (85). While syndecan-1 and syndecan-4 null mice have abnormal skin wound healing, their role in lung injury has yet to be established. The glycosaminoglycan hyaluronan (HA) is increased in association with alveolar macrophages in experimenta lung injury, is increased in the BAL from ARDS patients, and localizes to sites of inflammation/remodeling in human acute lung injury (86). Administration of a hyaluronan-binding peptide to rats reduced alveolar macrophage accumulation in the lung, macrophage motility in vitro, as well as collagen mRNA and hydroxyproline content after bleomycin (86). Low molecular weight fragments of the parent HA molecule also accumulate within alveolar septae and alveolar macrophages at sites of chronic inflammation, including bleomycin-induced lung injury (84). These fragments have unique pro-inflammatory bioactivities relevant to fibroproliferation that are not seen with the parent molecule and, in many cases, not seen with other glycosaminoglycans. These fragments induce a net protease inhibitory state by inducing PAI-1 protein and suppressing urokinase in rodent macrophages cell lines and in freshly isolated macrophages from bleomycin-treated rats (87). They also induce macrophage metalloelastase protein and activity, induce chemokine production, and synergistically modulate the actions of fibrogenically important molecules such as IFN-α (29, 88–90). Thus, local generation of the inflammation-specific HA fragments likely modulates the inflammatory process itself. The clearance receptor for hyaluronan, CD 44, is a proteoglycan with many isoforms. It is present on hematopoetic epithelial cells and fibroblasts and can mediate relevant cell-matrix interactions. Antibodies towards CD 44 inhibit endothelial and fibroblast migration into a three-dimensional fibrin gel and also induce fibroblast apoptosis in a manner that is adhesion-independent (91, 92). Furthermore, bleomycin administration to mice lack-ing CD 44 results in persistent, fatal cellular and biochemical alveolitis associated with persistent accumulation of hyaluronan and its low molecular weight fragments (84). Evidence of an important role for CD 44 in the clearance of apoptotic neutrophils and in TGF-β activation in vivo is convincing (84). These data show that CD44 is a regular of the resolution of inflammation following lung injury (84). The resultant defective apoptosis results in an uncoupling of the inflammatory phase with that fibroproliferative phase of bleomycin-induced lung injury. C. Deadhesive Matricellular Proteins in Modulation of Cell-Matrix Interactions Role in Lung Injury Matrix glycoproteins that localize to sites of morphogenesis, tissue remodeling, and inflammation can be functionally divided into those that directly promote cell adhesion, e.g., collagen, fibronectin, laminin, and vitronectin, and those that modulate cell adhesion, the matricellular proteins. This latter group of multidomain, multifunctional, structurally diverse, but functionally similar proteins includes thrombospondin 1 and 2 (TSP-1, 2), secreted protein rich in cysteine (SPARC), and tenascin-C (93–95). They are present in both soluble and insoluble form and by themselves largely support only initial

Mechanisms of fibroproliferation in acute lung injury

287

and intermediate cell adhesion, namely attachment and spreading. The strong adhesive state, characterized by focal adhesions and cytoskeletal stress fiber formation seen with collagen (and others in this group), is not induced by matricellular proteins (Fig. 2)(96). Furthermore, these proteins will block the adhesive effects of other more adhesive proteins in a mixed substrate and have direct antiadhesive activities when presented in soluble form to cells that are strongly adherent. When presented in soluble form, the matricellular proteins promote a reorganization of the cytoskeleton with loss of actin stress fibers and an uncoupling of vinculin and α-actinin from sites of focal cell-matrix adhesion. Despite this change, the cells remain attached and spread, integrins remain clustered, and integrin-matrix protein interaction remains grossly visually intact (97). Tenascin expression is induced and tenascin protein is deposited in alveolar walls and septal tips at sites of inflammation during the early inflammatory phase of bleomycin lung injury in rodents (98). Furthermore, it is increased in the BAL of patients with pulmonary fibrotic disorders and is present in the subepithelial basement membrane and fibrotic interstitium (99). Tenascin localizes to areas of pleural fibrosis near fibrinous exudate in cells that have markers consistent with myofibroblasts and is induced by TGFβ (100). Mechanistic observations on its role in lung injury are forthcoming.

Figure 2 Influence of matricellular proteins on states of cell adhesion. Progressively stronger cell matrix interactions are characterized by cell attachment, cell spreading, and the formation of focal cell matrix adhesion complexes that connect with cytoskeletal stress fibers (thick black lines). Tenascin, SPARC, and

Acute respiratory distress syndrome

288

thrombospondin can induce the intermediate adhesive state (right to middle arrow) that is characterized by disassembly of stress fibers and focal adhesions in a cell that maintains a spread shape and clustered integrins (clustered lines). The strongly adhesive state (right) is seen in differentiated and quiescent cells with the predicted consequences on their behavior (text below cell). The weakly adhesive state (left) would be seen in cells undergoing apoptosis or cytokenesis. During lung injury and remodeling, the matricellular proteins may induce an intermediate adhesive state and thereby influence cell survival, motility, and phenotype. (From Ref. 96.) SPARC is produced by many cell types, including endothelial cells, macrophages, and fibroblasts, and is present in platelets. SPARC expression is regulated by TGF-β, SPARC can induce focal adhesion disassembly (see below), and it binds to PDGF B chain isoforms that have been implicated in the cell proliferation observed in ARDS (101–103). Two reports show differing effects of SPARC deletion on neutrophil recruitment and collagen accumulation after bleomycin instillation in SPARC null mice (104, 105). TSP-1 is deposited in early granulation tissue upon platelet degranulation, while TSP2 is seen in organizing granulation tissue as a consequence of resident fibroblast production (106). Both TSP-1 and TSP-2 null mice have altered skin wound healing and alterations in angiogenesis (106). Interestingly, TSP-2 has also been shown to have a deadhesive action, perhaps through its interaction with MMP-2 (107). TSP-1 and plasmin cooperate in the activation of TGF-β on the surface of rat alveolar macrophages in vitro (62). TSP-1 null mice show a phenotype similar to that seen with TGF-β null mice, and the phenotype is partially reversible through administration of a TSP-1-derived peptide (106). Furthermore, the fibrotic response to bleomycin-induced lung injury in rats is partially abrogated through administration of a synthetic peptide to a TSP cell surface receptor, CD 36 (67). As with many of the multifunctional proteins that bind several fibrogenic modulating factors, it is likely that the final result of injury/repair is a consequence of complex temporal and spatial interacting events that depend on a coordinated signal between the cells involved, the matrix chemical and physical cues, and the growth factors/cytokines. Working out the details will likely lead to the discovery of novel molecular targets.

Mechanisms of fibroproliferation in acute lung injury

289

Role in Cell Adhesion and Migration The adhesion of cells to the matrix modulates numerous vital cell functions involved in the fibroproliferative response to injury including proliferation/ cell cycle progression, migration, differentiation/phenotype (i.e., differential gene expression of matrix proteins), protease/antiprotease production, and apoptosis (107–111). Two lines of evidence suggest that the matricellular protein-induced changes in cytoskeletal organization would lead to a highly motile state (Fig. 2). Biophysical modeling of cell migration predicts that an intermediate ratio of cytoskeletal contraction force to cell matrix adhesion strength would lead to maximal migration (112, 113). In vitro observations demonstrate that cell migration is absent in cells that are completely round and diminished in those that contain abundant stress fibers and focal adhesions (114). Tenascin-C has been observed to increase endothelial cell motility in a receptor-dependent manner in a monolayer wound assay, and thrombospondin heparin-binding domain exhibits chemotactic and chemokinetic activity (95, 96). In contrast, SPARC inhibits endothelial cell chemotaxis to FGF-2 through a domain that does not induce focal adhesion disassembly. Although the cytoskeletal changes induced by these proteins appear similar, they accomplish these changes through different cell surface receptors and signaling pathways, supporting the development of specific molecular agonists/antagonists (95–97, 103). There are many tantalizing observations that lend significance to the role of these proteins in the fibroproliferative process in human lung injury. The definition of the mechanism of the effects of these multidomain, multifunctional, antiadhesive proteins on cell biology and its extension to the in vivo fibroproliferative process is likely to be a fruitful and exciting area of inquiry and may provide numerous molecular targets for drug development.

V. Fibroblast Proliferation and Differentiation in Acute Lung Injury A. Proliferative Pathways in Acute Lung Injury Histopathological observations of lung tissue from patients with acute lung injury and experimental animal models of lung injury demonstrate that repair and remodeling after injury require the regulated recruitment and proliferation of endothelial, epithelial, and mesenchymal cells. There is ample heterogeneity in fibroblast cell populations in both normal and diseased lungs. Nonetheless, stable phenotypic differences in fibroblasts derived from normal and chronic fibrotic lungs have been well documented. These include alterations in proliferation rates, urokinase receptor number, smooth muscle actin expression, PGE2 production, and expression of cyclooxygenase-2 (COX-2) (115, 116). The stability of these phenotypes over several passages in culture suggests that ongoing signals from the fibrotic microenvironment are no longer required and that such signals have already induced a phenotypic change in the cells. Mesenchymal cells isolated from patients with dying from ARDS (14–28 days after intubation) do manifest a more rapid proliferation rate, as well as differences in their actin filament configuration, compared to those isolated from transplant donor lungs (117). These cells also exhibited peptide growth factor (PDGF, EGF, insulin)-independent proliferation, suggesting a phenotypic change similar to that noted for chronic pulmonary fibrotic fibroblasts (117). It is more

Acute respiratory distress syndrome

290

likely that the observed phenotypic changes are a consequence of oligoclonal expansion of specific cells from a heterogeneous background rather than a stable alteration in phenotype from a homogeneous population. Nonetheless, this phenotype is likely a consequence of cell exposure to factor(s) early in the course of lung injury. The fibroblast-altering factors in human acute lung injury are likely to be multiple, dynamic, and interactive, depending not only on the state of the cell at the time of exposure, but also on its interactions with the matrix. A few clues can be gleaned from the work published to date. Fibroblast mitogenic activity is increased in the BAL of ARDS patients compared with non-ARDS patients (22, 118). Mitogenic activity is greater within the first few days after intubation than after the first week (118). A fraction of the mitogenic activity in BAL 3 days after intubation has been ascribed to plateletderived growth factor (B-chain isoforms) (118). These observations are consistent with the identification of PDGF as one important macrophage-derived fibroblast mitogen (119). The process of bronchoalveolar lavage itself dilutes the alveolar contents 100-fold. This may reduce the concentration of mitogenically active factors below the threshold level of detectability and may even lead to false conclusions about the net activity of a given factor as its soluble inhibitors are diluted out. In contrast, pulmonary edema fluid that is directly aspirated from the lung within hours of intubation may more accurately reflect the actual molecular composition that the fibroblast is exposed to during the initiation of the fibroproliferative process (20, 120). These patient samples demonstrate bioactivity for IL-1β and for TNF-α (80, 120). Interestingly, submitted collaborative data from our laboratory suggests that IL-1β is a key fibroblast, mitogen and modulator of the fibroblast phenotype in edema fluid obtained within hours of intubation in patients with clinical acute lung injury. In acute lung injury edema fluid, IL-1β’s positive proliferative effect appears to act via autocrine induction of other growth factors. Important differences were noted in the mechanism of the IL-1β-mediated proliferative effect in the edema fluid compared with exogenously added IL-1β, indicating the requirement for cofactors that suppress PGE2 synthesis. Differences in the effect of exogenous IL-1β and that seen by inhibiting IL-1β action in edema fluid highlight the importance of interactive effects among several mitogenic factors in acute lung injury. PGE2 concentrates in the lung is the major eicosanoid product of fibroblasts and is a potent inhibitor of fibroblast proliferation and collagen synthesis (115, 116). Fibroblast PGE2 synthesis is inducible by mediators that are present and active in early acute lung injury, including IL-1β and PDGF, and during the subacute phase, including TGF-β (121–123). Furthermore, the regulation of COX-2, the rate-limiting enzyme for prostanoid synthesis by IL-1β, TNF-α, and lipopolysaccharide, is altered in fibrotic lungderived fibroblasts compared with those derived from normal lungs (115, 124). Similarly, TGF-β induction of COX-2 expression is limited in these cells (125). The importance of the PGE2 response to cytokines for their mitogenic action is clear, as these same mediators have opposite effects on fibroblast proliferation in the presence or absence of a PGE2 synthetic response (126). These data are consistent with the hypothesis that fibroblasts with mediator-insensitive COX-2 expression are generated during acute lung injury. Thus, under conditions of persistent mediator expression (e.g., IL-1 or TGF-β), these cells may become proliferative. Supporting this hypothesis is the observation that

Mechanisms of fibroproliferation in acute lung injury

291

mice that lack COX-2 have an augmented inflammatory and fibrotic response to intratracheal bleomycin (125). The importance of COX-2 induction and PGE2 synthesis in promoting normal wound healing in human acute lung injury is a potentially fruitful area of inquiry. In addition to the classical fibroblast cytokine mitogens, a number of other newly identified mitogens have been described with relevance to acute lung injury. For example, the acute phase reactant and key protease inhibitor for neutrophil elastase, α1antitrypsin, is a fibroblast mitogen in physiological concentrations and induces procollagen synthesis in human fibro-blasts (127). Elastase and elastase-α1-antitrypsin complexes and elastin degradation products have long been known to be increased in the BAL of patients with early ARDS (128). Elastase is capable of degrading not only elastin but other structural proteins of the lung, including proteoglycans, fibronectin, and nonfibrillar collagens. Both neutrophils and macrophages generate elastase activity through different enzymes (129). Elastase and α1-antitrypsin complexes can modulate inflammation/fibrosis through multiple mechanisms including generation of bioactive elastin fragments, recruitment and migration of leukocytes, and degradation of structural lung matrix proteins (130, 131). As with other protease, the in vivo mechanisms of action during lung injury are not defined. Recent data in elastase null mice that have reduced hydroxyproline accumulation after bleomycin-induced lung injury suggest that an important operative mechanism is the modulation of TGF-β sequestration in the matrix, perhaps via cleavage of LTBP (50). Angiotensin-converting enzyme is elevated in the BAL of ARDS patients. Angiotensin II is mitogenic for human fibroblasts in vitro through autoinduction of TGFβ (132). Human lung fibroblasts express the angiotensin type I, but not type II receptor, and the angiotensin II effect is blocked by the angiotensin I receptor antagonist Losartan (132). Although other enzymes can generate angiotensin II in the lung (i.e., chymase), these data are consistent with a role for angiotensin peptides in the fibroproliferative response to acute lung injury. B. Importance of Biophysical Cues for Fibroblast Proliferation Growth of fibroblasts in a three-dimensional collagen lattice results in a contraction of the lattice that depends on a force transmission between the cytoskeleton and the extracellular matrix (113). This process induces a quiescent state, a stellate morphology, an upregulation of collagenase, and a downregulation of matrix proteins including collagen and tenascin in an α1β1 and α2β1 integrin-dependent manner (133). In contrast to this quiescent state, fibroblast grown in lattices that are noncontractile where cells generate tension, induce a synthetic phenotype. This phenotype is characterized by an induction of collagen synthesis and repression of collagenase, me-diated in part by a strech-responsive DNA element, GAGACC (133). Various cytokines including TGF-β synthesis are modulated by mechanical strain in fibroblasts. Additional observations have been made with fibroblasts grown on a flexible substrate as a monolayer. When subjected to mechanical strain, fibroblasts proliferate, develop prominent actin stress fibers (as myofibroblasts), and elongate along the plane of the force (133). Furthermore, the upregulation of collagen synthesis seen when the cells were subjected to mechanical strain was dependent on matrix protein under the monolayer (134). Cyclic mechanical

Acute respiratory distress syndrome

292

strain also modulates the cell phenotype. This mechanosignal is known to be transmitted through integrins and involve tension generation in the cytoskeleton and activation of signaling cascades, including MAPK (135, 136). These protean matrix biophysical cues clearly modulate fibroblast function after lung injury in a manner that is currently poorly understood. C. Potential Role of the Myofibroblast in Lung Repair Myofibroblasts represent one band from the broad spectrum of the fibroblast phenotypic heterogeneity noted in normal and diseased lungs. Fractionation of fibroblasts from normal human lungs indicates heterogeneity in morphology, size, and smooth muscle actin expression (137). These cells express alpha smooth muscle actin, are contractile, and localize to sites of organizing exudate in human and experimental lung injury (138– 140). The cell may be derived from contractile interstitial cells and/or from fibroblasts located in adventitial areas of the lung (141). Several lines of evidence suggest that this cell type is important to the fibroproliferative response in acute lung injury. Smooth muscle actin-expressing cells are seen in number of fibrotic disorders (141, 142). These cells express high level of collagen mRNA in situ as well as inflammation/fibrogenesismodulating cytokines/chemokines/ cytokines such as TGF-β and MCP-1 (141, 143). Under certain conditions, these cells can either up- or downregulate the inflammatory response through production of mediators including IL-1, TNF-α, PDGF, TGF-β, MIP-1, IFN-γ, KGF, FGF, Ang II, and PGE2 (140). They also respond to mediators by virtue of their expression of a host of receptors for cytokines and growth factors. Several extracellular matrix molecules, including fibrillar and nonfibrillar collagens, glycoproteins, fibronectin, laminin, and proteoglycans, as well as matrix-modifying proteases and their inhibitors are also produced by these cells (140). In vitro, TGF-β will induce fibroblasts to express alpha smooth muscle actin and acquire a myofibroblast phenotype, while IL-1β will downregulate the phenotype and induce apoptosis by a nitric oxide-mediated mechanism (144). These concordant lines of evidence point to this cell type as one capable of altering lung parenchymal repair on many levels above and beyond their capacity to contract a wound.

VI. Coagulation and Fibrinolytic Systems in Fibroproliferative Phase ARDS The coagulation and fibrinolytic systems regulate intravascular and extra vascular fibrin deposition. Fibrin is a prominent component of the hyaline membranes that comprise the alveolar provisional matrix upon which granulation tissue is established early after lung injury (36, 145). Procoagulant activity is induced in the lung in both human and experimental lung injury as a consequence of induction of tissue factor/factor VII complex (146–149). Turnover of fibrin itself generates bioactive fragments that are fibroblast mitogens and induce protease inhibitor production by fibroblasts (150–154). However, recent observations in mice suggest that the fibroproliferation phase of bleomycin-induced lung injury is fibrin independent (155).

Mechanisms of fibroproliferation in acute lung injury

293

Independent of its effect on fibrin, generation of procoagulant activity in the presence of downstream clotting factors also generates proteases that have mitogenic activity. Factor Xa (and VIIa) is mitogenic for human fibroblasts in vitro at physiological concentrations through binding to effector cell protease receptor and autocrine induction of PDGF A (156). Fur-thermore, thrombin is mitogenic for fibroblasts and stimulates expression of connective tissue growth factor through binding/cleavage of protease activated receptor-1 (157, 158). In fact, a direct thrombin inhibitor (UK-156406) did not affect inflammatory cell recruitment, but did partially abrogate the collagen accumulation and upregulation of connective tissue growth factor expression seen after bleomycin instillation in rats (159). Plasma protein C levels are reduced in sepsis, and exogenous administration of the active anticoagulant protease improves survival in sepsis (160). It is tempting to speculate that one mechanism of its beneficial effect is a consequence of the inhibition of the generation of factor Xa/thrombin-induced lung fibroblast proliferation. The normal alveolus possesses a net profibrinolytic environment (Fig. 3). With injury, human and experimental animal models demonstrate a rapid increase in tissue factordependent procoagulant activity, and rapid suppression of the normal fibrinolytic (plasmin) activity by the combined effect increases in α2-antiplasmin andPAI-1. The consequence of this shift in the coagulation/fibrinolytic balance and the availability of plasma-derived fibrinogen results in the massive deposition of alveolar and interstitial cross-linked fibrin. The leakage of plasma proteins into the alveolar space provides downstream coagulant factors as well as fibrinogen that is enzymatically cleaved and cross-linked into an insoluble alveolar matrix composed of fibrin, fibronectin, vitronectin, surfactant, and other cell- and plasma-derived proteins (145). Despite the large increases in PAI-1, partial breakdown of the alveolar fibrin occurs as evident by the logrithmic increases in fibrin D-dimer in the BAL of patients with ARDS (161). Similar alterations of the alveolar fibrinolytic environment are present in acute infectious pneumonia, but to a lesser degree (162). As with many enzyme cascades, there is a certain amount of redundancy and overlap of function. As such, studies from genetically altered mice have highlighted the importance of specific components of this enzyme cascade in

Acute respiratory distress syndrome

294

Figure 3 General model of cellassociated fibrinolytic system. Cellassociated activation of the zymogen (plasminogen) to the active protease (plasmin) by plasminogen activators (PA) urokinase (u-PA) and/or tissue type plasminogen activator (t-PA) is localized to the cell surface through binding to their respective receptors) (bottom left). Once the broad specificity plasmin is formed, it may degrade fibrin and other matrix glycoproteins and/or release growth factors from the matrix. Active plasminogen activator inhibitor (PAI1) can bind to and inhibit cell surface bound or solution phase PAs. PAI-1 can directly block urokinase receptor (u-PAR)-binding sites on vitronectin and alter cell adhesion (top right). Furthermore, through binding of PAI-1 to receptor-bound u-PAR induced the internalization of the complex by LRP

Mechanisms of fibroproliferation in acute lung injury

295

and modulated its signaling (bottom right). u-PAR-integrin interactions at the cell surface can modulate cell adhesive and migratory phenomena, intracellular signaling, and other integrin functions in a u-PA-dependent manner (top). These pathways provide several potential therapeutic targets designed to alter mesenchymal cell adhesion and migration. vivo and provided clues to support potential mechanisms of their effects in lung injury. In many systems, there is a good deal of functional redundancy between t-PA and u-PA, but knockout of plasminogen or combined u-PA/t-PA deletion results in a severe phenotype with massive intraorgan thromboses and fibrotic reactions in many organs, including the lung (163, 164). Deletion of PAI-1 results in a mild phenotype in a unchallenged animal, but large differences were noted in the amount of fibrin deposition during an inflammatory response where PAI-1 is normally induced (LPS foot pad injection) (165, 166). These mice are protected from the deleterious effects of bleomycin-induced and hyperoxic lung injury (3, 167, 168). It is likely that the mechanism of PAI-1’s effect is related to its effect on the fibrinolytic enzyme cascade, as mice lacking the downstream effector plasminogen but not mice lacking u-PA or u-PAR, show an enhanced fibrogenic response to bleomycin (169). Furthermore, exogenous u-PA and/or adenovirally mediated u-PA expression can reverse fibrosis after bleomycin-induced lung injury if given after its formative phase (170, 171). The long-assumed mechanism for the protective effect of PAI-1 on lung injury relates to its capacity to block plasminogen activation to plasmin and thereby results in the persistence of the alveolar fibrinous provisional matrix that would support enhanced fibroblast proliferation and matrix synthesis. However, hydroxproline accumulation after bleomycin-induced lung injury was similar to wild type in mice lacking fibrinogen, demonstrating that this process is independent of the presence of fibrin and/or bioactive fragments of fibrin (153, 155). Perhaps additional provisional alveolar matrix proteins with redundant functions can equally support pathological repair. Alternate mechanisms of PAI-1’s effect on lung injury may relate to its effect on cell adhesion/migration, although leukocyte recruitment after bleomycin appears to be equivalent to wild-type mice (155, 172). The mechanism of its PAI-1’s protective effect in lung injury may be a consequence of PAI-1’s effect on other cell processes, including LRP-dependent signals that are a consequence of internalization of u-PA-PAI complexes, modulation of growth factor activity or its release from the matrix, or angiogenesis (173, 175). The in vivo mechanism of PAI-1’s modulatory effect on numerous pathological states including atherosclerosis, cancer metastasis, and pulmonary fibrosis is an active area of ongoing investigation.

Acute respiratory distress syndrome

296

VII. Matrix Metalloproteases in Matrix Remodeling in Acute Lung Injury Matrix metalloproteinases are a large class of proteinases with both overlapping and unique specificity for a broad range of matrix proteins relevant to lung fibrosis. They are unique in their ability to cleave fibrillar collagens but can also degrade nonfibrillar collagens, laminin, fibronectin, elastin, and proteoglycans. MMP-related degradation of a number of nonmatrix proteins, including α1-antitrypsin, α-2MG, and the generation of both pro and antiangiogenic molecules, derived from the degradation of plasminogen and collagen XVIII (176, 177). The related protease family, the ADAMs (a disintegrin and metalloproteinase domain) members, can induce cleavage and shedding of functional receptors for TNF, Fas, and IL-6 into the pericellular environment (177). Thus, the potential for this class of proteases to modulate important, and perhaps antagonistic, processes involved in fibroproliferative phase lung injury is large. Family-specific characteristics include metal ion dependence of catalytic activity, secretion as inactive zymogens, and inhibition by tissue inhibitor of metalloproteinases (TIMPs). Many different MMPs and inhibitors (TIMPs) exist. In human acute lung injury, both MMP-2, MMP-9, as well as the inhibitors TIMP-1 and TIMP-2 stain strongly in myofibroblasts and epithelial cells and basement membranes in organizing alveolar buds (178). Interestingly, MMP-2 colocalizes with an area of disrupted alveolar basement membrane, while TIMP-2 colocalizes with Type IV and fibrillar collagens, suggesting a compartmentalization of protease and pro tease-inhibitory activity (178). Collagenase activity and fragments of collagen have been detected in the BAL of patients with ARDS while both TIMP-1 and TIMP-2 proteins and their inhibitory activities are increased in murine bleomycin-induced lung injury (179, 180). Furthermore, intraperitoneal injection of the MMP inhibitor batimastat abrogates the hydroxyproline accumulation and alveolar leukocyte recruitment seen with bleomycin-induced lung injury in mice (181). Given the redundancy of this many members of this class of proteases and inhibitors, that the precursor zymogens require proteolytic activation, in part, by membrane-bound MTMMPs, and that a broad range of potential synergistic or antagonistic molecular mechanisms may be operative, direct validation of their function is essential. The study of mice with genetic alterations in one or more of these enzymes/inhibitors has and will produce much novel and unexpected mechanistic information in this complex and evolving area (176, 182). For example, gelatinase B-deficient mice have defective migration of bronchial cells with clara-cell features into alveoli but manifest similar collagen accumulation after bleomycin-induced lung injury (183). One example that illustrates the dynamic and reciprocal interactions between cells, MMPs, and the matrix is the model system of keratinocytes migrating on a collagen substrate. The expression of the collagen-degrading enzyme that is required for keratinocyte migration on collagen, collagenase-1 (MMP-1), is regulated by the litigation of the α2β1 integrin to matrix collagen (34). In turn, the MMP-1 binds to the I domain of the α2 chain of the integrin, forming a ternary complex of integrin, substrate, and ligand (184). This functionally compartmentalizes MMP-1 activity to sites of cell matrix/substrate contact and participates in cell migration on collagen (185). Reciprocal interactions between epithelial/mesenchymal cells, the matrix, and matrix-degrading

Mechanisms of fibroproliferation in acute lung injury

297

proteases undoubtedly are important for repopulating the denuded basement membrane and reorganizing the alveolar exudate after acute lung injury.

VIII. Apoptosis in Fibroproliferative Phase ARDS Apoptosis enhances the normal resolution of inflammation by removal of cells that have outlived their usefulness without the release of damaging tissue toxins. However, under certain circumstances ligation and clustering of the apoptosis receptor Fas will result in the generation of tissue inflammation and injury. Adding to the confusion are several reports that show a proliferative effect of Fas ligation and activation of downstream caspases (186). The biochemical components and organization of the matrix may also influence cell survival through modulation of anoikis, adhesion-dependent cell survival (187) (Fig. 2). Apoptosis could alter the fibroproliferative process in vivo by its effects on limiting persistent inflammation by removal of apoptotic neutrophils, and/or removal of hyperplastic alveolar epithelial cells, or removal of mesenchymal cells in organizing alveolar matrix (188). There are a number of initiating apoptotic signals. In one pathway, the receptor Fas is expressed on alveolar type II cells, and injection of Fas-activating antibodies will induce apoptosis of Type II cells (189). Apoptotic epithelial cells are seen in acute lung injury in humans (190). In bleomycin lung injury in rodents, Fas expression was upregulated in apoptotic epithelial cells, Fas ligand expression was increased in infiltrating lymphocytes, and inhalation of an activating antibody to Fas induced apoptosis and pulmonary fibrosis in mice (6). In a follow-up study, this same group demonstrated that administration of a soluble form of Fas or anti-Fas ligand antibody prevented apoptosis and fibrosis in the bleomycin model (5). They further demonstrated that in mice with Fas (lpr mice) or Fas ligand (gld mice) mutations are protected from the effects of bleomycin (5). These data suggest in fact that apoptosis is required for the full manifestation of bleomycin pulmonary injury and that epithelial apoptosis induces rather than protects against inflammation and fibrosis. Phagocytosis of apoptotic neutrophils by macrophages induces the release of fibrosismodulating factors including PGE2 and TGF-β from macrophages (188). Furthermore, compared with ingestion of necrotic cells, few pro-inflammatory mediators are released upon macrophage ingestion of apoptotic neutrophils (188). Fibroblasts also demonstrate the capacity to ingest apoptotic cells (188). These data suggest that clearance of apoptotic neutrophils in early lung injury will not only modulate the resolution of inflammation but subsequent fibroproliferative processes. In fact, mice lacking the hyaluronan receptor CD44, have defective clearance of apoptotic neutrophils and develop persistent and fatal pulmonary inflammation, but no fibrosis after bleomycin lung injury (84). Furthermore, during the repair phase (10 days after intubation) of human lung injury, there is morphological evidence of apoptosis in fibroblastic plugs in vivo, and the alveolar fluid has significant pro-apoptotic activity in vitro on proliferating fibroblasts and endothelial cells (191). Thus, apoptosis is a critical, naturally occurring process after lung injury, but its role and consequences appear to be cell type and time restricted. It is possible that acceleration of the normal apoptosis during the repair phase of injury may reduce the number of matrix-synthesizing cells and provide a novel therapeutic modality. Alternatively, apoptosis of myofibroblasts may clear repair-enhancing cells

Acute respiratory distress syndrome

298

and/or induce an immune response. In an attempt to exploit the known anchorage dependence for cell survival, Hadden and Henke were able to induce apoptosis in 60– 87% of primary human lung fibroblasts in a three-dimensional fibrin gel by a combination of three distinct fibronectin-derived pep tides (192). The cell response was characteristic of anoikis as the cells assumed a rounded morphology, cells in suspension did not exhibit this effect, and integrin-dependent signaling was disrupted (192). In a similar vein, the drug lovastatin was shown to induce apoptosis of nontransformed human lung fibroblasts in vitro in a dose- and time-dependent manner that depended on its ability to block HMG-CoA reductase (193). Furthermore, inclusion of lovastatin in subcutaneously implanted, fibrin-containing wound chambers also reduced in vivo granulation tissue formation by 67% (193). These observations demonstrate a proof of principle, but significant drug delivery/toxicity issues must be resolved prior to attempting these approaches in human lung injury.

IX. Current and Future Therapeutics for Amelioration of the Fibroproliferative Response Significant acute and chronic morbidity and mortality are attributable to the development of a fibroproliferative repair response (16). Attributable morbidity includes the prolongation of ventilator dependence with its attendant complications, some of which are fatal. Thus, it can be reasonably estimated that the total attributable mortality is roughly 10,000–20,000 deaths per year in the United States. Fortunately, there are several practical advantages that support the development of therapies designed to abrogate the fibroproliferative repair response after acute lung injury. The earliest signs of the fibroproliferative repair response in sepsis-related acute lung injury appear several days after injury. It is important to keep in mind that it is possible that the early injury process itself predetermines the extent of fibroproliferation the results during the repair phase. Nonetheless, it is likely that therapeutic agents directed at cell-matrix interactions would be effective during the major time period of alveolar remodeling, several days to 3 weeks after presentation. This would allow for time to screen, identify, and characterize potential candidates most likely to obtain benefit for a given therapeutic regimens, as well as modify management to optimize the beneficial effect and minimize potentially harmful interactions with use of a given agent(s). Although elevated procollagen peptide levels in alveolar fluid appear to show promise, the routine use of clinical, cellular, or biochemical markers to specifically identify patients destined to develop a severe fibroproliferative response awaits further study and validation. A number of fundamentally plausible approaches to block fibroproliferation after lung injury can be put forth. Attempts at reducing inflammation in human lung injury have been unsuccessful to date, perhaps due to significant redundancy and cross-talk of cytokine actions and risk of infectious complications. One could consider directly interfering with collagen synthesis and/or processing (194). As another area of approach once could consider modulation of fibroblast proliferation, matrix synthesis, protease/protease inhibitor production, myofibroblast differentiation, and/or other biologically relevant processes such as adhesion and/or migration. The elimination of

Mechanisms of fibroproliferation in acute lung injury

299

fibroblasts by apoptosis/necrosis provides an alternative approach. These approaches would depend on the identification of specific/ selective molecular targets for fibroblast and cell-matrix interactions that control these processes, including growth factors, their receptors, and/or signaling pathways. To date some agents show promise, but there have been no successful trials of pharmacological agents that specifically ameliorate the fibroproliferative repair response and that meet rigorous study design and statistical standards. Administration of methylprednisolone (2 mg/kg/day, n=15) beginning on day 7 in an randomized, placebocontrolled (n=7) partial cross-over trial to a small number of patients with subacute ARDS resulted in an improvement of lung injury scores, multiple organ failure scores, arterial partial pressure of oxygen-to-inspired ratio, successful intubation rate, and mortality (0/16 vs. 5/8) (195). However, the small sample size, the fact that the survival curve in late ARDS is relatively flat, and the biased cross-over scheme support a cautious interpretation of the findings. In uncontrolled observational studies, corticosteroid administration in subacute ARDS correlated with a reduction on inflammatory cytokine in the BAL and a reduction in procollagen peptide levels in the plasma and BAL (23, 196). Corticosteriod use in acute lung injury, even if started during the subacute phase, may predispose to the development of sepsis and may impair would healing and have other untoward effects. These concerns are germane given the background of a lack of efficacy of corticosteroids begun during the acute phase on mortality and reversal of ARDS in several well-designed trials (197). Recommendations on the use of late methylprednisolone therapy for subacute ARDS await the completion of a larger randomized, prospective, blinded, placebo-controlled trial, as sponsored by the National Institutes of Health and performed by the ARDS Network. Given our current knowledge, cytokines/growth factors signaling pathways that appear to be attractive targets include IL-1, TNF-α, TGF-β, PGE2, modulation of Th1 and/or Th2 cytokines and/or angiogenic chemokines, modulation of coagulation/fibrinolysis or other proteases, and modulation of growth factor-cell matrix interactions. These may include modulation of integrin function, integrin interactions with co-receptors, alterations in cell surface and matrix proteoglycans, and protease-dependent activation or release of matrix-bound growth factors. Before promoting or abandoning a given approach, we must better understand the issues of redundancy and cross-talk of pathways, combinatorial effects, and pro- and antifibrotic and cell type-specific responses. Furthermore, if the importance of the heterogeneity of etiologies, timing, and biochemical responses were understood, one could potentially design pathogenically based, individualized therapy. Examination of genetically modified mice in models of lung injury provides tremendous insight into these issues. However, many of these models have the advantage of a specific inciting event/agent of known time and concentration in animals with similar genetic backgrounds. I remain optimistic that, with further understanding, effective, individualized, therapeutic approaches that abrogate the fibroproliferative response will be developed.

Acute respiratory distress syndrome

300

Acknowledgments The work presented in this review was funded by grants from the NIH (HL-58655), VA Merit Review Board, and American Lung Association (MAO). The author is grateful for the scientific and technical work performed by Kimberly White, Stephanie Hallit, and Drs. Sha Zhu and Priya Prabakharan. The author is grateful for the secretarial assistance provided by Amy Kiker.

References 1. Keane MP, Belperio JA, Moore TA, Moore BB, Arenberg DA, Smith RE, et al. Neutralization of the CXC chemokine, macrophage inflammatory protein-2, attenuates bleomycin-induced pulmonary fibrosis. J Immunol 1999; 162(9): 5511–5518. 2. Marinelli WA, Henke CA, Harmon KR, Hertz MI, Bitterman PB. Mechanisms of alveolar fibrosis after acute lung injury. Clin Chest Med 1990; 11(4):657–659. 3. Olman M, Mackman N, Gladson C, Moser K, Loskutoff D. Changes in procoagulant and fibrinolytic gene expression during bleomycin induced lung injury in the mouse. J Clin Invest 1995; 96:1621–1630. 4. Keane MP, Belperio JA, Arenberg DA, Burdick MD, Xu ZJ, Xue YY, et al. IFN-γ-inducible protein-10 attenuates bleomycin-induced pulmonary fibrosis via inhibition of angiogenesis. J Immunol 1999; 163(10):5686–5692. 5. Kuwano K, Hagimoto N, Kawasaki M, Yatomi T, Nakamura N, Nagata S, et al. Essential roles of the Fas-Fas ligand pathway in the development of pulmonary fibrosis. J Clin Invest 1999; 104(1):13–19. 6. Hagimoto N, Kuwano K, Nomoto Y, Kunitake R, Hara N. Apoptosis and expression of Fas/Fas ligand mRNA in bleomycin-induced pulmonary fibrosis in mice. Am J Respir Cell Mol Biol 1997; 16(1):91–101. 7. Ware LB, Matthay MA. Medical progress: the acute respiratory distress syndrome. N Engl J Med 2000; 342(18): 1334–1349. 8. Hall J, Schmidt G, Wood L, eds. Principles of Critical Care. New York: McGraw-Hill, Inc., 1992. 9. Gattinoni L, Bombino M, Pelosi P, Lissoni A, Pesenti A, Fumagalli R, et al. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA 1994; 271(22): 1772–1779. 10. Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342(18):1301–1308. 11. Bachofen A, Weibel ER. Alterations of the gas exchange apparatus in adult respiratory insufficiency associated with septicemia. Am Rev Respir Dis 1977; 116(4):589–615. 12. Fukuda Y, Ishizaki M, Masuda Y, Kimura G, Kawanami O, Masugi Y. The role of intraalveolar fibrosis in the process of pulmonary structural remodeling in patients with diffuse alveolar damage. Am J Pathol 1987; 126(1): 171–182. 13. Montgomery AB, Stager MA, Carrico CJ, Hudson LD. Causes of mortality in patients with adult respiratory distress syndrome. Am Rev Respir Dis 1985; 132:485–489. 14. Davidson TA, Caldwell ES, Curtis JR, Hudson LD, Steinberg KP. Reduced quality of life in survivors of acute respiratory distress syndrome compared with critically ill control patients. JAMA 1999; 281(4):354–360.

Mechanisms of fibroproliferation in acute lung injury

301

15. Davidson TA, Rubenfeld GD, Caldwell ES, Hudson LD, Steinberg KP. The effect of acute respiratory distress syndrome on long-term survival. Am J Respir Crit Care Med 1999; 160(6): 1838–1842. 16. McHugh LG, Milberg JA, Whitcomb ME, Schoene RB, Maunder RJ, Hudson L. Recovery of function in survivors of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1994; 150:90–94. 17. Zapol WM, Trelstad RL, Coffey JW, Tsai I, Salvador RA. Pulmonary fibro-sis in severe acute respiratory failure. Am Rev Respir Dis 1979; 119(4):547—554. 18. Goodman RB, Stricter RM, Martin DP, Steinberg KP, Milberg JA, Maunder RJ, et al. Inflammatory cytokines in patients with persistence of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154(3 pt 1): 602–611. 19. Raghu G, Striker LJ, Hudson LD, Striker GE. Extracellular matrix in normal and fibrotic human lungs. Am Rev Respir Dis 1985; 131(2):281–289. 20. Chesnutt AN, Matthay MA, Tibayan FA, Clark JG. Early detection of type III procollagen peptide in acute lung injury. Am J Respir Crit Care Med 1997; 156:840–845. 21. Clark JG, Milberg JA, Steinberg KP, Hudson LD. Type III procollagen peptide in the adult respiratory distress syndrome. Ann Intern Med 1995; 122(1): 17–23. 22. Marshall RP, Bellingan G, Webb S, Puddicombe A, Goldsack N, McAnulty RJ, et al. Fibroproliferation occurs early in the acute respiratory distress syndrome and impacts on outcome. Am J Respir Crit Care Med 2000; 162(5): 1783–1788. 23. Meduri GU, Tolley EA, Chinn A, Stentz F, Postlethwaite A. Procollagen types I and III aminoterminal propeptide levels during acute respiratory distress syndrome and in response to methylprednisolone treatment. Am J Respir Crit Care Med 1998; 158:1432–1441. 24. Martin C, Papazian L, Payan MJ, Saux P, Gouin F. Pulmonary fibrosis correlates with outcome in adult respiratory distress syndrome. A study in mechanically ventilated patients. Chest 1995; 107(1): 196–200. 25. Farjanel J, Hartmann DJ, Guidet B, Luquel L, Offenstadt G. Four markers of collagen metabolism as possible indicators of disease in the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 147(5): 1091–1099. 26. Klagsbrun M, Shing Y. Heparin affinity of anionic and cationic capillary endothelial cell growth factors: analysis of hypothalamus-derived growth factors and fibroblast growth factors. Proc Natl Acad Sci 1985; 82:805–809. 27. Languino LR, Plescia J, Duperray A, Brian AA, Plow EF, Geltosky JE, et al. Fibrinogen mediates leukocyte adhesion to vascular endothelium through an ICAM-1-dependent pathway. Cell 1993; 73(7): 1423–1434. 28. Languino LR, Duperray A, Joganic KJ, Fornaro M, Thornton GB, Altieri DC. Regulation of leukocyte-endothelium interaction and leukocyte transendothelial migration by intercellular adhesion molecule 1-fibrinogen recognition. Proc Natl Acad Sci 1995; 92(5): 1505–1509. 29. Horton MR, McKee CM, Bao C, Liao F, Farber JM, Hodge-DuFour J, et al. Hyaluronan fragments synergize with interfoeron-γ to induce the C-X-C chemokines mig and interferoninducible protein-10 in mouse macrophages. J Biol Chem 1998; 273(52):35088–35094. 30. Skogen WF, Senior RM, Griffin GL, Wilner GD. Fibrinogen-derived peptide Bβ1–42 is a multidomained neutrophil chemoattractant. Blood 1988; 71: 145–149. 31. Senior RM, Skogen WF, Griffin GL, Wilner GD. Effects of fibrinogen derivatives upon the inflammatory response: studies with human fibrinopeptide B. J Clin Invest 1986; 77:1014– 1019. 32. Sahni A, Sporn LA, Francis CW. Potentiation of endothelial cell proliferation by fibrin(ogen)bound fibroblast growth factor-2. J Biol Chem 1999; 274(21): 14936–14941. 33. Falcone DJ, McCaffrey TA, Haimovitz-Friedman A, Garcia M. Transforming growth factor- β1 stimulates macrophage urokinase expression and release of matrix-bound basic fibroblast growth factor. J Cell Physiol 1993; 155(3):595–605.

Acute respiratory distress syndrome

302

34. Saarialho-Kere UK, Kovacs SO, Pentland AP, Olerud JE, Welgus HG, Parks WC. Cell-matrix interactions modulate interstitial collagenase expression by human keratinocytes actively involved in wound healing. J Clin Invest 1993; 92:2858–2866. 35. Burkhardt A. Alveolitis and collapse in the pathogenesis of pulmonary fibrosis. Am Rev Respir Dis 1989; 140:513–524. 36. Kuhn C III, Boldt J, King TE Jr, Crouch E, Vartio T, McDonald JA. An immunohistochemical study of architectural remodeling and connective tissue synthesis in pulmonary fibrosis. Am Rev Respir Dis 1989; 140:1693–1703. 37. Doerschuk CM, Mizgerd JP, Kubo H, Qin L, Kumasaka T. Adhesion molecules and cellular biomechanical changes in acute lung injury: Giles F. Filley Lecture. Chest 1999; 116(1 suppl):37S–43S. 38. Snow RL, Davies P, Pontoppidan H, Zapol WM, Reid L. Pulmonary vascular remodeling in adult respiratory distress syndrome. Am Rev Respir Dis 1982; 126(5):887–892. 39. Zapol WM, Jones R. Vascular components of ARDS. Clinical pulmonary hemodynamics and morphology. Am Rev Respir Dis 1987; 136(2):471–474. 40. Gailit J, Clark RAF. Wound repair in the context of extracellular matrix. Curr Opin Cell Biol 1994; 6:717–725. 41. Smith RE, Stricter RM, Phan SH, Kunkel SL. C-C chemokines: novel mediators of the profibrotic inflammatory response to bleomycin challenge. Am J Respir Cell Mol Biol 1996; 15(6):693–702. 42. Pittet JF, Mackersie RC, Martin TR, Matthay MA. Biological markers of acute lung injury: prognostic and pathogenetic significance. Am J Respir Crit Care Med 1997; 155(4): 1187–1205. 43. Streiter RM. Mechanisms of pulmonary fibrosis. Chest 2001; 120(1):77S-85S. 44. Lukacs N, Hogaboam C, Chensue SW, Blease K, Kunkel SL. Type 1/type 2 cytokine paradigm and the progression of pulmonary fibrosis. Chest 2001; 120(1):5S–8S. 45. Frolik CA, Dart LL, Meyers CA, Smith DM, Sporn MB. Purification and initial characterization of a type-transforming growth factor from human placenta. Proc Natl Acad Sci 1983; 3676–3680. 46. Nathan C, Sporn M. Cytokines in context. J Cell Physiol 1991; 113:981–986. 47. Zimmerman CM, Padgett RW. Transforming growth factor β signaling mediators and modulators. Gene 2000; 249:17–30. 48. Zhu HJ, Burgess AW. Regulation of transforming growth factor-β signaling. Mol Cell Biol Res Commun 2001; 4:321–330. 49. Taipale J, Miyazono K, Heldin C-H, Keski-Oja J. Latent transforming growth factor-β 1 associates to fibroblast extracellular matrix via latent TGF-β binding protein. J Cell Biol 1994; 124:171–181. 50. Dunsmore SE, Roes J, Chua FJ, Segal AW, Mutsaers SE, Laurent GJ. Evidence that neutrophil elastase-deficient mice are resistant to bleomycin-induced fibrosis. Chest 2001; 120(1 suppl):35S–36S. 51. Yue J, Mulder KM. Transforming growth factor-β signal transduction in epithelial cells. Pharmacol Ther 2001; 91:1–34. 52. Khalil N, Bereznay O, Sporn M, Greenberg AH. Macrophage production of transforming growth factor β and fibroblast collagen synthesis in chronic pulmonary inflammation. J Exp Med 1989; 70:727–737. 53. Broekelmann TJ, Limper AH, Colby TV, McDonald JA. Transforming growth factor β1 is present at sites of extracellular matrix gene expression in human pulmonary fibrosis. Proc Natl Acad Sci 1991; 88:6642–6646. 54. Giri SN, Hyde DM, Hollinger MA. Effect of antibody to transforming growth factor β on bleomycin induced accumulation of lung collagen in mice. Thorax 1993; 48(10):959–966. 55. Roberts CJ, Birkenmeier TM, McQuillan JJ, Akiyama SK, Yamada SS, Chen W-T, et al. Transforming growth factor β stimulates the expression of fibronectin and of both subunits of

Mechanisms of fibroproliferation in acute lung injury

303

the human fibronectin receptor by cultured human lung fibroblasts. J Biol Chem 1988; 263:4586–4592. 56. Border WA, Noble NA. Transforming growth factor β in tissue fibrosis [review]. N Engl J Med 1994; 331(19):1286–1292. 57. Santana A, Saxena B, Noble NA, Gold LI, Marshall BC. Increased expression of transforming growth factor β isoforms (β 1, β 2, β 3) in bleomycin-induced pulmonary fibrosis. Am J Respir Cell Mol Biol 1995; 13(1):34–44. 58. Sime PJ, Xing Z, Graham FL, Csaky KG, Gauldie J. Adenovector-mediated gene transfer of active transforming growth factor-β1 induces prolonged severe fibrosis in rat lung. J Clin Invest 1997; 100(4):768–776. 59. Crawford SE, Stellmach V, Murphy-Ullrich JE, Ribeiro SM, Lawler J, Hynes RO, et al. Thrombospondin-1 is a major activator of TGF-β1 in vivo. Cell 1998; 93(7): 1159–1170. 60. Khalil N, Corne S, Whitman C, Yacyshyn H. Plasmin regulates the activation of cell-associated latent TGFβ1 secreted by rat alveolar macrophages after in vivo bleomycin injury. Am J Respir Cell Mol Biol 1996; 15:252–259. 61. Murphy-Ullrich JE, Schultz-Cherry S, Hook M. Transforming growth factor-β complexes with thrombospondin. Mol Biol Cell 1992; 3:181–188. 62. Yehualaeshet T, O’Connor R, Green-Johnson J, Mai S, Silverstein R, Murphy-Ullrich JE, et al. Activation of rat alveolar macrophage-derived latent transforming growth factor β-1 by plasmin requires interaction with thrombospondin-1 and its cell surface receptor. Am J Pathol 1999; 155(3): 841–851. 63. Mungger JS, Huang X, Kawakatsu H, Griffiths MJ, Dalton SL, Wu J, et al. The integrin a v β 6 binds and activates latent TGF β 1: a mechanism for regulating pulmonary inflammation and fibrosis. Cell 1999; 96(3):319–328. 64. Bonner JC, Brody AR. Cytokine-binding proteins in the lung. Am J Physiol 1995; 268(6 pt 1):L869–L878. 65. Martin TR, Ruzinski JT, Steinberg KP, et al. Cytokine balance in the lungs of patients with acute respiratory distress syndrome (ARDS). Am J Respir Crit Care Med 1998; 157:A679. 66. Pittet JF, Griffiths MJD, Geiser T, Kaminski N, Dalton SL, Huang X, et al. TGF-β is a critical mediator of acute lung injury. J Clin Invest 2001; 107(12): 1537–1544. 67. Yehualaeshet T, O’Connor R, Begleiter A, Murphy-Ullrich JE, Silverstein R, Khalil N. A CD36 synthetic peptide inhibits bleomycin-induced pulmonary inflammation and connective tissue synthesis in the rat. Am J Respir Cell Mol Biol 2000; 23(2):204–212. 68. Border WA, Noble NA, Yamamoto T, Harper JR, Yamaguchi Y, Pierschbacher MD, et al. Natural inhibitor of transforming growth factor-β protects against scarring in experimental kidney disease. Nature 1992; 360(6402):361–364. 69. Bensadoun ES, Burke AK, Hogg JC, Roberts CR. Proteoglycan deposition in pulmonary fibrosis. Am J Respir Crit Care Med 1996; 154(6 pt 1):1819–1828. 70. Giri SN, Hyde DM, Braun RK, Gaarde W, Harper JR. Antifibrotic effect of decorin in a bleomycin hamster model of lung fibrosis. Biochem Pharmacol 1997; 54(11):1205–1216. 71. Kolb M, Margetts PJ, Galt T, Sime PJ, Xing Z, Schmidt M, et al. Transient transgene experssion of decorin in the lung reduces the fibrotic response to bleomycin. Am J Respir Crit Care Med 2001; 163:770–777. 72. LaMarre J, Hayes MA, Wollenberg GK, Hussaini I, Hall SW, Gonias SL. An α2-macroglobulin receptor-dependent mechanism for the plasma clearance of transforming growth factor-β1 in mice. J Clin Invest 1991; 87:39–44. 73. LaMarre J, Wollenberg GK, Gonias SL, Hayes MA. Cytokine binding and clearance properties of proteinase activated α2-macroglobulins. Lab Invest 1991; 65:3–14. 74. Danielpour D, Sporn MB. Differential inhibition of transforming growth facter β 1 and β 2 activity by a 2-macroglobulin. J Biol Chem 1990; 265(12): 6973–6977. 75. Borth W, Luger TA. Identification of α2-macroglobulin as a cytokine binding protein: binding of interleukin-1 β to “F” α2-macroglobulin. J Biol Chem 1989; 264:5818–5825.

Acute respiratory distress syndrome

304

76. Khalil N, Parekh TV, O’Connor R, Antman N, Kepron W, Yehavlaeshet T, Xu YD, Gold LI. Regulation of the effects of TGF-beta 1 by activation of the latent TGF-beta 1 and differential expression of TGF-beta receptors (T beta R-I and T beta R-II) in idiopathic pulmonary fibrosis. Thorax 2002; 56(12):907–915. 77. Nakao A, Fujii M, Matsumura R, Kumano K, Saito Y, Miyazono K, et al. Transient gene transfer and expression of Smad7 prevents bleomycin-induced lung fibrosis in mice. J Clin Invest 1999; 104(1):5–11. 78. Zhao MQ, Amir MK, Rice WR, Enelow RI. Type II pneumocyte-CD8+ T-cell interactions. Relationship between target cell cytotoxicity and activation. Am J Respir Cell Mol Biol 2001; 25(3):362–369. 79. Ziesche R, Hofbauer E, Wittmann K, Petkov V, Block LH. A preliminary study of long-term treatment with interferony γ-1β and low-dose prednisolone in patients with idiopathic pulmonary fibrosis. N Engl J Med 1999; 341(17): 1264–1269. 80. Pugin J, Verghese GM, Widmer MC, Matthay MA. The alveolar space is the site of intense inflammatory and profibrotic reactions in the early phase of acute respiratory distress syndrome. Crit Care Med 1999; 27(2):304–312. 81. Liu JY, Sime PJ, Wu T, Warshamana GS, Pociask D, Tsai SY, et al. Transforming growth factor-β(1) overexpression intumor necrosis factor-α receptor knockout mice induces fibroproliferative lung disease. Am J Respir Cell Mol Biol 2001; 25(1):3–7. 82. Sottile J, Hocking DC, Langenbach KJ. Fibronectin polymerization stimulates cell growth by RGD-dependent and -independent mechanisms. J Cell Sci 2000; 113(pt 23):4287–4299. 83. Hocking DC, Sottile J, Langenbach KJ. Stimulation of integrin-mediated cell contractility by fibronectin polymerization. J Biol Chem 2000; 275(14): 10673–10682. 84. Teder P, Vandivier W, Jiang D, Liang J, Cohn L, Pure E, et al. Resolution of lung inflammation by CD44. Science 2002; 296:155–158. 85. Woods A. Syndecans: transmembrane modulators of adhesion and matrix assembly. J Clin Invest 2001; 107(8):935–941. 86. Savani RC, Hou G, Liu P, Wang C, Simons E, Grimm PC, et al. A role for hyaluronan in macrophage accumulation and collagen deposition after bleomycin-induced lung injury. Am J Respir Cell Mol Biol 2000; 23(4):475–484. 87. Horton MR, Olman MA, Bao C, Rivera KE, Choi AMK, Chin BY, et al. Regulation of plasminogen activator inhibitor-1 and urokinase by hyaluronan fragments in mouse macrophages. Am J Physiol 2000; 279:L207-L215. 88. Horton MR, Shapiro S, Bao C, Lowenstein CJ, Noble PW. Induction and regulation of macrophage metalloelastase by hyaluronan fragments in mouse macrophages. J Immunol 1999; 162(7):4171–4176. 89. McKee CM, Penno MB, Cowman M, Burdick MD, Streiter RM, Bao C, et al. Hyaluronan (HA) fragments induce chemokine gene expression in alveolar macrophages. The role of HA size and CD44. J Clin Invest 1996; 98(10):2403–2413. 90. Noble PW, McKee CM, Cowman M, Shin HS. Hyaluronan fragments activate an NF-κ B/I-κ B α autoregulatory loop in murine macrophages. J Exp Med 1996; 183(5):2373–2378. 91. Henke C, Bitterman P, Roongta U, Ingbar DH, Polunovsky V. Induction of fibroblast apoptosis by anti-CD44 antibody: implications for the treatment of fibroproliferative lung disease. Am J Pathol 1996; 149(5): 1639–1650. 92. Svee K, White J, Vaillant P, Jessurun J, Roongta U, Krumwiede M, et al. Acute lung injury fibroblast migration and invasion of a fibrin matrix is mediated by CD44. J Clin Invest 1996; 98(8): 1713–1727. 93. Bradshaw AD, Sage EH. SPARC, a matricellular protein that functions in cellular differentiation and tissue response to injury. J Clin Invest 2001; 107(9): 1049–1054. 94. Bornstein P, O’Rourke K, Wikstrom K, Wolf FW, Katz R, Li P, et al. A second, expressed thrombospondin gene (Thbs2) exists in the mouse genome. J Biol Chem 1991; 266:12821– 12824.

Mechanisms of fibroproliferation in acute lung injury

305

95. Chung CY, Murphy-Ullrich JE, Erickson HP. Mitogenesis, cell migration, and loss of focal adhesions induced by tenascin-C interacting with its cell surface receptor, annexin II. Mol Biol Cell 1996; 7(6):883–892. 96. Murphy-Ullrich JE. The de-adhesive activity of matricellular proteins: is intermediate cell adhesion an adaptive state. J Clin Invest 2001; 107(7):785–790. 97. Goicoechea S, Orr AW, Pallero MA, Eggleton P, Murphy-Ullrich JE. Thrombospondin mediates focal adhesion disassembly through interactions with cell surface calreticulin. J Biol Chem 2000; 275(46):36358–36368. 98. Zhao Y, Young SL, McIntosh JC. Induction of tenascin in rat lungs undergoing bleomycininduced pulmonary fibrosis. Am J Physiol 1998; 274(6 pt 1): L1049–L1057. 99. Kaarteenaho-Wiik R, Mertaniemi P, Sajanti E, Paakko P. Tenascin is increased in epithelial lining fluid in fibrotic lung disorders. Lung 1998; 176(6): 371–380. 100. Kaarteenaho-Wiik R, Lakari E, Soini Y, Pollanen R, Kinnula VL, Paakko P. Tenascin expression and distribution in pleura inflammatory and fibrotic disease. J Histochem Cytochem 2000; 48(9): 1257–1268. 101. Raines EW, Lane TF, Iruela-Arispe ML, Ross R, Sage EH. The extracellular glycoprotein SPARC interacts with platelet-derived growth factor (PDGF)-AB and -BB and inhibits the binding of PDGF to its receptors. Proc Natl Acad Sci 1992; 89:1281–1285. 102. Sage EH, Bornstein P. Extracellular proteins that modulate cell-matrix interactions. J Biol Chem 1991; 266:14831–14834. 103. Murphy-Ullrich JE, Lane TF, Pallero MA, Sage EH. SPARC mediates focal adhesion disassembly in endothelial cells through a follistatin-like region and the Ca(2+)-binding EFhand. J Cell Biochem 1995; 57(2):341–350. 104. Strandjord TP, Madtes DK, Weiss DJ, Sage EH. Collagen accumulation is decreased in SPARC-null mice with bleomycin-induced pulmonary fibrosis. Am J Physiol 1999; 277(3 pt 1):L628–L635. 105. Savani RC, Zhou Z, Arguiri E, Wang S, Vu D, Howe CC, et al. Bleomycin-induced pulmonary injury in mice deficient in SPARC. Am J Physiol 2000; 279(4):L743–L750. 106. Bornstein P. Thrombospondins as matricellular modulators of cell function. J Clin Invest 2001; 107(8):929. 107. Yang Z, Kyriakides TR, Bornstein P. Matricellular proteins as modulators of cell-matrix interactions: adhesive defect in thrombospondin 2-null fibroblasts is a consequence of increased levels of matrix metalloproteinase-2. Mol Biol Cell 2000; 11(10):3353–3364. 108. Gumbiner BM. Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell 1996; 84(3):345–357. 109. Lelievre SA, Bissell MJ, Pujuguet P. Cell nucleus in context. Crit Rev Eukaryot Gene Expr 2000; 10(1): 13–20. 110. Chen CS, Mrksich M, Huang S, Whitesides GM, Ingber DE. Geometric control of cell life and death. Science 1997; 276(5317): 1425–1428. 111. Hasselaar P, Loskutoff DJ, Sawdey M, Sage EH. SPARC induces the expression of type 1 plasminogen activator inhibitor in cultured bovine aortic endothelial cells. J Biol Chem 1991; 266:13178–13184. 112. DiMilla PA, Barbee K, Kauffenburger DA. Mathematical model for the effects of adhesion and mechanics on cell migration speed. Biophys J 1991; 60(1): 15–37. 113. Ingber DE. Cellular tensegrity: defining new rules of biological design that govern the cytoskeleton. J Cell Sci 1993; 104(Pt 3):613–627. 114. Huttenlocher A, Sandborg RR, Horwitz AF. Adhesion in cell migration. Curr Opin Cell Biol 1995; 7(5):697–706. 115. Wilborn J, Crofford LJ, Burdick MD, Kunkel SL, Stricter RM, Peters-Golden M. Cultured lung fibroblasts isolated from patients with idiopathic pulmonary fibrosis have a diminished capacity to synthesize prostaglandin E2 and to express cyclooxygenase-2. J Clin Invest 1995; 95(4): 1861–1868.

Acute respiratory distress syndrome

306

116. Toews GB. Cellular alterations in fibroproliferative lung disease. Chest 1999; 116(1 suppl):112S–116S. 117. Chen B, Polunovsky V, White J, Blazar B, Nakhleh R, Jessurun J, et al. Mesenchymal cells isolated after acute lung injury manifest an enhanced proliferative phenotype. J Clin Invest 1992; 90(5): 1778–1785. 118. Snyder LS, Hertz MI, Peterson MS, Harmon KR, Marinelli WA, Henke CA. Acute lung injury. Pathogenesis of intraalveolar fibrosis. J Clin Invest 1991; 88(2):663–673. 119. Bitterman PB, Wewers MD, Rennard SI, Adelberg S, Crystal RG. Modulation of alveolar macrophage-driven fibroblast proliferation by alternative macrophage mediators. J Clin Invest 1986; 77:700–708. 120. Geiser T, Atabai K, Jarreau PH, Ware LB, Pugin J, Matthay MA. Pulmonary edema fluid from patients with acute lung injury augments in vitro alveolar epithelial repair by an IL-1β dependent mechanism. Am J Respir Crit Care Med 2001; 163:1384–1388. 121. Diaz A, Varga J, Jimenez SA. Transforming growth factor-β stimulation of lung fibroblast prostagladin E2 production. J Biol Chem 1989; 264:11554–11557. 122. Yucel-Lindberg T, Ahola H, Nilsson S, Carlstedt-Duke J, Modeer T. Interleukin-1β induces expression of cyclooxygenase-2 mRNA in human gingival fibroblasts. Inflammation 1995; 19:549–560. 123. Habenicht AJR, Goerig M, Grulich J, Rothe D, Loth U, Schettler G, et al. Human plateletderived growth factor stimulates prostaglandin synthesis by activation and by rapid de novo synthesis of cyclooxygenase. J Clin Invest 1985; 75:1381–1387. 124. Vancheri C, Sortino MA, Tomaselli V, Mastruzzo C, Condorelli F, Bellistri G, et al. Different expression of TNF-α receptors and prostaglandin E2 production in normal and fibrotic lung fibroblasts. Potential implications for the evolution of the inflammatory process. Am J Respir Cell Mol Biol 2000; 22:628–634. 125. Keerthisingam CB, Jenkins RG, Harrison NK, Hernandez-Rodriguez NA, Booth H, Laurent GJ, et al. Cyclooxygenase-2 deficiency results in a loss of the anti-proliferative response to transforming growth factor- β in human fibrotic lung fibroblasts and promotes bleomycininduced pulmonary fibrosis in mice. Am J Pathol 2001; 158(4):1411–1422. 126. McAnulty RJ, Hernandez-Rodriguez NA, Mutsaers SE, Coker RK, Laurent GJ. Indomethacin suppresses the anti-proliferative effects of transforming growth factor-β isoforms on fibroblast cell cultures. Biochem J 1997; 321:639–643. 127. Dabbagh K, Laurent GJ, Shock A, Leoni P, Papakrivopoulou J, Chambers RC. α-1Antitrypsin stimulates fibroblast proliferation and procollagen production and activates classical MAP kinase signalling pathways. J Cell Physiol 2001; 186(1):73–81. 128. Lee CT, Fein AM, Lippmann M, Holtzmann H, Kimbel P, Wienbaum G. Elastolytic activity in pulmonary lavage fluid from patients with adult respiratory distress syndrome. N Engl J Med 1981; 304:192–196. 129. Chapman H Jr, Stone O, Vavrin Z. Degradation of fibrin and elastin by intact human alveolar macrophages in vitro. J Clin Invest 1984; 73:806–815. 130. Banda MJ, Rice AG, Griffin GL, Senior RM. α1-Proteinase inhibitor is a neutrophil chemoattractant after proteolytic inactivation by macrophage elastase. J Biol Chem 1988; 263:4481–4484. 131. Banda MJ, Rice AG, Griffin GL, Senior RM. The inhibitory coplex of human α-1 proteinase inhibitor and human leukocyte elastase is a neutrophil chemoattractant. J Exp Med 1988; 167:1608–1615. 132. Marshall RP, McAnulty RJ, Laurent GJ. Angiotensin II is mitogenic for human lung fibroblasts via activation of the type 1 receptor. Am J Respir Crit Care Med 2000; 161(6): 1999–2004. 133. Eckes B, Kessler DS, Aumailley M, Krieg T. Interactions of fibroblasts with the extracellular matrix: implications for the understanding of fibrosis. Springer Semin Immunopathol 2000; 21:415–429.

Mechanisms of fibroproliferation in acute lung injury

307

134. Breen EC. Mechanical strain increases type I collagen expression in pulmonary fibroblasts in vitro. J Appl Physiol 1999; 88(1):203–209. 135. Huang S, Ingber DE. The structural and mechanical complexity of cell-growth control. Nat Cell Biol 1999; 1(5):E131–E138. 136. Huang S, Ingber DE. Shape-dependent control of cell growth, differentiation, and apoptosis: switching between attractors in cell regulatory networks. Exp Cell Res 2000; 261(1):91–103. 137. Uhal BD, Ramos C, Joshi I, Bifero A, Pardo A, Selman M. Cell size, cell cycle, and a-smooth muscle actin expression by primary human lung fibroblasts. Am J Physiol 1998; 275:L998– L1005. 138. Zhang HY, Gharaee-Kermani M, Zhang K, Karmiol S, Phan SH. Lung fibroblast α-smooth muscle actin expression and contractile phenotype in bleomycin-induced pulmonary fibrosis. Am J Pathol 1996; 148(2):527–537. 139. Zhang K, Flanders KC, Phan SH. Cellular localization of transforming growth factor β expression in bleomycin-induced pulmonary fibrosis. Am J Pathol 1995; 147:352. 140. Powell DW, Mifflin RC, Valentich JD, Crowe SE, Saada JI, West AB. Myofibroblasts. I. Paracrine cells important in health and disease. J Physiol 1999; 277(1):C1–C19. 141. Phan SH. Role of the myofibroblast in pulmonalry fibrosis. Kidney Int Suppl 1996; 54:S46– S48. 142. Zhang K, Rekhter MD, Gordon D, Phan SH. Myofibroblasts and their role in lung collagen gene expression during pulmonary fibrosis. Am J Pathol 1994; 145:114. 143. Phan SH, Zhang K, Zhang HY, Gharaee-Kermani M. The myofibroblast as an inflammatory cell in pulmonary fibrosis. Curr Top Pathol 1999; 93:173–182. 144. Zhang HY, Gharaee-Kermani M, Phan SH. Regulation of lung fibroblast α-smooth muscle actin expression, contractile phenotype, and apoptosis by IL-β. J Immunol 1997; 158(3):1392– 1399. 145. Olman MA, Simmons WL, Pollman DJ, Loftis AY, Bini A, Miller EJ, et al. Polymerization of fibrinogen in murine bleomycin-induced lung injury. Am J Physiol 1996; 271(15):L519–L526. 146. Idell S, James KK, Levin EG, Schwartz BS, Manchanda N, Maunder RJ, et al. Local abnormalities in coagulation and fibrinolytic pathways predispose to alveolar fibrin deposition in the adult respiratory distress syndrome. J Clin Invest 1989; 84:695–705. 147. Idell S, Koenig KB, Fair DS, Martin TR, Mclarty J, Maunder RJ. Serial abnormalities of fibrin turnover in evolving adult respiratory distress syndrome. Am J Physiol 1991; 261(5):L240– L248. 148. Bertozzi P, Astedt B, Zenzius L, Lynch K, LeMaire F, Zapol W, et al. Depressed bronchoalveolar urokinase activity in patients with adult respiratory distress syndrome. N Engl J Med 1990; 322:890–897. 149. Sitrin RG, Brubaker PG, Fantone JC. Tissue fibrin deposition during acute lung injury in rabbits and its relationship to local expression of procoagulant and fibrinolytic activities. Am Rev Respir Dis 1987; 135:930–938. 150. Gray AJ, Bishop JE, Reeves JT, Laurent GJ. Aα and Bβ chains of fibrinogen stimulate proliferation of human fibroblasts. J Cell Sci 1993; 104(2):409–413. 151. Gray AJ, Bishop JE, Reeves JT, Mecham RP, Laurent GJ. Partially degraded fibrin(ogen) stimulates fibroblast proliferation in vitro. Am J Respir Cell Mol Biol 1995; 12(6): 684–690. 152. Gray AJ, Park PW, Brockelmann TJ, Laurent GJ, Reeves JT, Stenmark KR, et al. The mitogenic effects of the Bβ chain of fibriogen are mediated through cell surface calreticulin. J Biol Chem 1995; 270:26602–26606. 153. Hagood JS, Olman MA, Zeballos JG, Rivera KE, Fuller GM. Regulation of type I plasminogen activator inhibitor by fibrin degradation products in rat lung fibroblasts. Blood 1996; 87:3749–3757. 154. Olman MA, Hagood JS, Simmons WL, Fuller GM, Vinson C, Rivera KE. Fibrin fragment induction of plasminogen activator inhibitor transcription is mediated by activator protein-1 through a highly conserved element. Blood 1999; 94(6):2029–2038.

Acute respiratory distress syndrome

308

155. Hatorri N, Degen J, Sisson T, Lui H, Moore B, Pandrangi R, et al. Bleomycin-induced pulmonary firosis in fibrinogen-null mice. J Clin Invest 2000; 106(12):1441–1443. 156. Blanc-Brude OP, Chambers RC, Leoni P, Dik WA, Laurent GJ. Factor Xa is a fibroblast mitogen via binding to effector-cell protease receptor-1 and autocrine release of PDGF. Am J Physiol 2001; 281(2):C681–C689. 157. Chambers RC, Leoni P, Blanc-Brude OP, Wembridge DE, Laurent GJ. Thrombin is a potent inducer of connective tissue growth factor production via proteolytic activation of proteaseactivated receptor-1. J Biol Chem 2000; 275(45):35584–35591. 158. Hernandez-Rodriguez NA, Cambrey AD, Harrison NK, Chambers RC, Gray AJ, Southcott AM, et al. Role of thrombin in pulmonary fibrosis. Lancet 1995; 346(8982): 1071–1073. 159. Howell DC, Goldsack N, Marshall RP, McAnulty RJ, Starke R, Purdy G, et al. Direct thrombin inhibition reduces lung collagen, accumulation, and connective tissue growth factor mRNA levels in bleomycin-induced pulmo-nary fibrosis. Am J Pathol 2001; 159(4): 1383– 1395. 160. Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, LopezRodriguez A, et al. Efficacy and safety of recombinant human activated C of severe sepsis. N Engl J Med 2001; 344(10):699–709. 161. Fuchs-Buder T, deMoerloose P, Ricou B, Reber G, Vifian C, Nicod L, et al. Time course of procoagulant activity and D dimer in bronchoalveolar fluid of patients at risk for or with acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 153:163–167. 162. Gunther A, Mosavi P, Heinemann S, Ruppert C, Muth H, Markart P, et al. Alveolar fibrin formation caused by enhanced procoagulant and depressed fibrinolytic capacities in severe pneumonia. Am J Respir Crit Care Med 2000; 161:454–462. 163. Carmeliet P, Bouche A, De Clercq C, Janssen S, Pollefeyt S, Wyns S, et al. Biological effects of disruption of the tissue-type plasminogen activator, urokinase-type plasminogen activator, and plasminogen activator inhibitor-1 genes in mice. Ann NY Acad Sci 1995; 748:367–381. 164. Carmeliet P, Collen D. Gene manipulation and transfer of the plasmino-gen and coagulation system in mice. Semin Thromb Hemost 1996; 22(6): 525–542. 165. Carmeliet P, Kieckens L, Schoonjans L, Ream B. Plasminogen activator inhibitor-1 genedeficient mice. I. Generation by homologous recombination and characterization. J Clin Invest 1993; 92(6):2746–2755. 166. Carmeliet P, Stassen JM, Schoonjans L, Ream B. Plasminogen activator inhibitor-1 genedeficient mice. II. Effects on hemostasis, thrombosis, and thrombolysis. J Clin Invest 1993; 92(6):2756–2760. 167. McCoy RD, Eitzman, Ginsburg D, Simon RH. Bleomycin-induced pulmonary fibrosis in transgenic mice that either lack or over-express the murine plasminogen activator inhibitor-1 gene. Am J Respir Crit Care Med 1995; 151(4):A51. 168. Barazzone C, Belin D, Piguet PF, Vassalli JD, Sappinno AP. Plasminogen activator inhibitor1 in acute hyperoxic mouse lung injury. J Clin Invest 1996; 98(12):2666–2673. 169. Swaisgood CM, French EL, Noga C, Simon RH, Ploplis VA. The development of bleomycininduced pulmonary fibrosis in mice deficient for components of the fibrinolytic system. Am J Pathol 2000; 157(1):177–187. 170. Hart DA, Whidden P, Green F, Henkin J, Woods DE. Partial reversal of established bleomycin-induced pulmonary fibrosis by re-urokinase in a rat model. Clin Inves Med 1994; 17(2):69–70. 171. Sisson T, Hattori N, Xu Y, Simon R. Treatment of bleomycin-induced pulmonary fibrosis by transfer of urokinase-type plasminogen activator genes. Hum Gene Ther 1999; 10(14):2315– 2323. 172. Deng G, Curriden SA, Wang S, Rosenberg S, Loskutoff DJ. Is plasminogen activator inhibitor-1 the molecular switch that governs urokinase receptor-mediated cell adhesion and release. J Cell Biol 1996; 134(6): 1563–1571.

Mechanisms of fibroproliferation in acute lung injury

309

173. Romer J, Bugge TH, Pyke C, Lund LR, Flick MJ, Degen JL, et al. Imparied wound healing in mice with a discrupted plasminogen gene. Nat Med 1996; 2(3):287–292. 174. Bugge TH, Kombrinck KW, Flick MJ, Daugherty CC, Danton MJS, Degen JL. Loss of fibrinogen rescues mice from the pleiotropic effects of plasminogen deficiency. Cell 1996; 87:709–719. 175. Chapman HA. Plasminogen activators, integrins, and the coordinated regulation of cell adhesion and migration. Curr Opin Cell Biol 1997; 9(5):714–724. 176. Parks WC, Shapiro SD. Matrix metalloproteinases in lung biology. Respir Res 2001; 2(1):10– 19. 177. Shapiro S, Senior RM. Matrix metalloproteinases matrix degradation and more. Am J Respir Cell Mol Biol 1999; 20:110–1102. 178. Hayashi T, Stetler-Stevenson WG, Fleming MV, Fishback N, Koss MN, Liotta LA, et al. Immunohistochemical study of metalloproteinases and their tissue inhibitors in the lungs of patients with diffuse alveolar damage and idiopathic pulmonary fibrosis. Am J Pathol 1996; 149(4): 1241–1256. 179. Christner P, Fein A, Goldberg S, Lippmann M, Abrams W, Weinbaum G. Collagenase in the lower respiratory tract of patients with adult respiratory distress syndrome. Am Rev Respir Dis 1985; 131:690–695. 180. Madtes DK, Elston AL, Kaback LA, Clark JG. Selective induction of tissue inhibitor of metalloproteinase-1 in bleomycin-induced pulmonary fibrosis. Am J Respir Cell Mol Biol 2001; 24(5):599–607. 181. Corbel M, Caulet-Maugerdre S, Germain N, Molet S, Lagente V, Boichot E. Inhibition of bleomycin-induced pulmonary fibrosis in mice by the matrix metalloproteinase ingibitor batimastat. J Pathol 2001; 193(4):538–545. 182. Hautamaki RD, Kobayashi DK, Senior RM, Shapiro SD. Requirement for macrophage elastase for cigarette smoke-induced emphysema in mice. Science 1997; 277(5334):2002–2004. 183. Betsuyaku T, Fukuda Y, Parks WC, Shipley JM, Senior RM. Gelatinase B is required for alveolar bronchiolization after intratrached bleomycin. Am J Pathol 2000; 157(2):525–535. 184. Stricker TP, Dumin JA, Dickeson SK, Chung L, Nagase H, Parks WC, et al. Structural analysis of the α(2) integrin I domain/procollagenase-1 (matrix metalloproteinase-1) interaction. J Biol Chem 2001; 276(31):29375–29381. 185. Dumin JA, Dickeson SK, Stricker TP, Bhattacharyya-Pakrasi M, Roby JD, Santoro SA, et al. Pro-collagenase-1 (matrix metalloproteinase-1) binds the α(2)β(1) integrin upon release from keratinocytes migrating on type I collagen. J Biol Chem 2001; 276(31):29368–29374. 186. Budd RC. Death receptors couple to both cell proliferation and apoptosis. J Clin Invest 2002; 109(4):437–441. 187. Greenwood JA, Murphy-Ullrich JE. Signaling of de-adhesion in cellular regulation and motility. Microsc Res Tech 1998; 43(5):420–432. 188. Haslett C. Granulocyte apoptosis and its role in the resolution and control of lung inflammation. Am J Respir Crit Care Med 1999; 160:S5–S11. 189. Matute-Bello G, Liles WC, Steinberg KP, Kiener PA, Mongovin S, Chin EY, et al. Soluble fas ligand induces epithelial cell apoptosis in humans with acute lung injury (ARDS). J Immunol 1999; 163(4):2217–2225. 190. Bardales RH, Xie SS, Schaefer RF, Hsu SM. Apoptosis is a major pathway responsible for the resolution of type II pneumocytes in acute lung injury. Am J Pathol 1996; 149(3):845–852. 191. Polunovsky V, Chen B, Henke C, Snover D, Ingbar DH, Ingbar DH, et al. Role of mesenchymal cell death in lung remodeling after injury. J Clin Invest 1993; 92:388–397. 192. Hadden HL, Henke C. Induction of lung fibroblast apoptosis by soluble fibonectin peptides. Am J Respir Crit Care Med 2000; 162:1553–1560. 193. Tan A, Levrey H, Dahm C, Polunovsky V, Rubins J, Bitterman P. Lovastatin induces fibroblast apoptosis in vitro and in vivo. Am J Respir Crit Care Med 1999; 159:220–227.

Acute respiratory distress syndrome

310

194. Greco MJ, Kemnitzer JE, Fox JD, Choe JK, Kohn J, Riley DJ, et al. Polymer of proline analogue with sustained antifibrotic activity in lung fibrosis. Am J Respir Crit Care Med 1997; 155(4):1391–1397. 195. Meduri GU, Headley AS, Golden E, Carson SJ, Umberger RA, Kelso T, et al. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome. JAMA 1998; 280(2):159–165. 196. Meduri GU, Kohler G, Headley S, Tolley EA, Stentz F, Postlethwaite A. Inflammatory cytokines in the BAL of patients with ARDS. Chest 1995; 108(5):1303–1314. 197. Bernard GR, Luce JM, Sprung CL, Rinaldo JE, Tate RM, Sibbald WJ, et al. High-dose corticosteroids in patients with the adult respiratory distress syndrome. N Engl J Med 1987; 317:1565–1570.

13 Genetic Factors in Acute Lung Injury RICHARD P.MARSHALL University College London London, England

I. Introduction: Why Study the Genetics of Acute Lung Injury? Variation within the genetic sequence of individuals is an intrinsic property of the human genome. Such variations may lead to subtle changes in gene expression or protein function that in themselves are not the cause of disease but, together, may produce marked differences in the response to a given injury. That genetics factors influence the pathophysiology of acute lung injury (ALI) is unquestionable. We cannot hope to understand and effectively treat ALI without an intimate understanding of the relationship between genetic variation and pathophysiology. We are increasingly aware of the need to understand and treat the individual patient rather than cohorts or populations. Genetics will be a key component of that process, nowhere more so than in the context of critical illness, where profound differences in disease severity and response to therapy are commonplace. There is no gene for ALI, just as there is no gene for asthma or diabetes or hypertension. These are merely labels that ease the grouping of patients for diagnostic and therapeutic purposes. Any genetic variations associated with these clinical syndromes result from the influence of a protein product on key shared pathophysiological process within these individuals. Associating genes with ALI is a first step. Associating genes with specific biological processes is the ideal. Studying genetics in critically ill patients presents both difficulties and benefits. On the one hand, the somewhat arbitrary criteria for the definition of ALI, acute respiratory distress syndrome (ARDS), and sepsis leads to heterogeneity within patient cohorts and an increase in diagnostic “noise” that might hinder the identification of a genetic association. Similarly, the fact that these individuals are so unwell can render radiological and pathological testing difficult—if not impossible—in some cases. Ultimately, this will prevent accurate phenotyping. On the other hand, however, the critically ill also present many unique opportunities for study. Often, although not always, the time of onset and end of a critical illness (and any predisposing condition or insult) is well defined. This is of enormous benefit when matching a gene to the progression of disease. By contrast, in chronic disease, an individual may present with established pathology that has been developing for an unknown period of time and in whom the precise environmental contributions to phenotype prior to presentation are unknown. Such individuals have already self-selected themselves as those who have, perhaps, responded badly to a given environmental insult,

Acute respiratory distress syndrome

312

as opposed to those in whom the response does not manifest clinically and who recover without seeking medical attention. It is equally important to understand the genetics of people who recover. Genetic studies in the intensive care unit (ICU) have other advantages. We might learn from physicists in this regard, who are often forced to work at the extremes of temperature, pressure, and velocity to understand the fundamental properties of matter. In a similar way, there is much to be learned about human biology at the extremes of pathophysiology. Of course, this is also a potential drawback, in that it may be difficult to detect the influence of specific biological events in specific tissues or cells when the pathology is so complex. Yet, the ability for hemodynamic, respiratory, and biochemical parameters to be monitored and recorded regularly (if not continuously) throughout the course of a critical illness should allow for the more precise matching of genetic data to phenotype. This increases our chances of identifying meaningful genetic associations. Of course, genetic studies in the ICU will not only relate to the influence of genes on disease pathogenesis. As important will be their role in the response to therapy, both pharmacological and nonpharmacological. At present, studies in the critically ill are at an early stage, and therefore before examining the current experimental and clinical evidence for a genetic predisposition to ALI it is useful to consider in some detail the principles and problems of genetic study design as they pertain to the critically ill.

II. The Genetics of Complex Diseases In a “simple” genetic condition, the disease phenotype results from genetic variation in a single gene or a small number of genes. Also implicit in this definition is that the environment will play a relatively minor role in disease pathogenesis. The genetic differences are usually profound and may include gene deletions or major structural changes that profoundly alter gene activity. Rare mutant forms of genes might lead to critical illness (such as those associated with C1-esterase deficiency), yet ALI occurs, largely, in individuals without obvious inherited genetic defects. Yet simple is rarely simple. Even in “monogentic” diseases such as cystic fibrosis and sickle cell anaemia, there are wide variations in clinical presentation and prognosis, no doubt the result of interactions between major disease genes (CFTR in the case of cystic fibrosis), more minor differences in other disease-modifying genes, and the environment. Most diseases are genetically complex, wherein the disease phenotype results from multiple genetic variations interacting with a number of environmental factors. Extensive variation within the human genome exists. The existence of such variation within healthy individuals is vital to the maintenance of biodiversity and the adaptability of the species to a constantly changing environment. Many genetic variations in DNA sequence have no discernible effect on function or phenotype. They might, for instance, occur in a noncoding region of a gene or at a functionally unimportant site in the protein. Some variations will have functional consequences, perhaps resulting in a change in structure or conformation of a protein that might ultimately influence agonist/substrate binding. Alternatively, such variations may alter the transcriptional regulation of a gene by modifying a binding site for a transcriptional regulator to the promoter sequence.

Genetic factors in acute lung injury

313

Variations in the structure of a gene that have less dramatic effects on biological function and that occur commonly within a healthy population (arbitrarily, with an allele frequency of more than 10%) are known as functional polymorphisms. Usually these polymorphisms are not, in themselves, enough to manifest as a specific disease phenotype. But when they occur in combination in an individual and when that individual is exposed to a particular insult, a devastating phenotype may arise.

III. Strategies for Genetics Studies in ARDS When designing genetic studies, a number of issues must be taken into consideration to ensure the genetic and statistical validity of any results obtained. These include (1) the use of appropriate genetic methodology, (2) the acquisition of detailed phenotypic data, (3) recognition of the influence of ethnic diversity and genetic background, (4) the acquisition of study populations of appropriate size for statistical power, (5) the use of appropriate comparison populations, and (6) the replication of results in different populations. These issues are now considered in more detail. A. Genetic Approaches Broadly, two main methodological approaches are used to establish an association between an individual polymorphism and the disease phenotype. Genome-wide approaches require related individuals in whom the phenotype can be accurately established (although incomplete data regarding phenotype can, in part, be compensated for statistically). The advantage of such an approach is that no prior assumptions are made as to the gene(s) involved. The whole genome is considered. Initially, the segregation or movement of a disease phenotype is tracked within a family in relation to commonly occurring noncoding but highly polymorphic sequences (micro-satellite markers, for example) located throughout the genome—a so-called genome scan. If particular polymorphisms appear to segregate or associate with the disease phenotype, this provides evidence that the polymorphism is in linkage disequilibrium with gene(s) responsible for that phenotype, that is, they more commonly occur together than not because of their chromosomal proximity. In this instance, a more detailed study of the chromosomal region in which the marker lies is then made to identify specific gene(s). This is achieved by examining the association with additional microsatellite markers within the region of interest, by studying candidate genes within this region, and by sequencing methods. Collectively, this process is sometimes referred to as positional cloning. It is immediately apparent that such an approach involving related individuals is not feasible for the vast majority of ALI cases. By contrast, a “candidate gene” approach studies known or novel polymorphisms in genes that are purported to play a role in disease pathogenesis and is suitable for unrelated individuals. This is likely to be the main method for identifying genetic factors in ALI. The hypothesis under study is that the frequency of a specific genetic polymorphism in a study population will differ from that in relevant comparison groups. In the case of susceptibility genes, individuals with disease might be compared to those without. In the case of genes that might modify disease severity, genotype and allele

Acute respiratory distress syndrome

314

frequencies are compared in patients with disease in relation to relevant clinical or biochemical markers. It is also important to recognize that within any individual gene, more than one polymorphism may determine function. Similarly, polymorphisms across a number of genes may determine phenotype. Combinations of polymorphisms within or across genes that occur more commonly in association with the disease phenotype are called haplotypes. New high-throughput technologies with which it is possible to screen hundreds, if not thousands, of single nucleotide polymorphisms (SNPs) in a relatively short space of time on a microchip (SNiP chips) (1) lend themselves to haplotyping large numbers of polymorphisms. Ultimately this could lead to a better understanding of complex gene-environment interactions. A number of on-line databases of SNPs exist (for a useful list see http://www.hgmp.mrc.ac.uk/GenomeWeb/human-gen-dbmutation.html). This technology may also mean that the necessity for candidate genes to have been biologically implicated in ALI is somewhat circumvented, such that it becomes a hypothesis-generating tool, highlighting associations with known or unknown genes that implicate novel biological pathways. However, there are potential pitfalls to this approach. For example, the acquisition of suitable phenotypic data that allow a clear association with genotype to be made may not be possible. Clearly, there will be a place—and indeed a need—for both hypothesis-driven and hypothesis-generating genetic studies in ALI patients. This technology also offers the realistic possibility of bedside genetic testing. B. Recruitment of Patients If the hypothesis under study is that a given gene or genes confer a susceptibility to ALI, it becomes vital that individuals are correctly and consistently diagnosed. Given the relatively arbitrary diagnostic criteria for the diagnosis of ARDS/ALI, problems may arise. After all, what is ALI? In a sense, genetic studies may in themselves help to refine current diagnostic criteria by identifying shared pathogenic pathways. But until then it is important that patients fulfill standardized criteria such as the ATS/ ERS Consensus Committee recommendations to allow comparison between studies. In particular, individual centers and collaborative groups must ensure the consistency of chest radiograph interpretation and the correct identification of the predisposing condition, as these are perhaps the most subjective elements of the current criteria. We must also appreciate that alternative explanations for any genetic associations observed with ALI exist. Spurious associations might occur with conditions that predispose to ALI rather than with ALI itself, such as pneumonia, or with other, less specific components of the host response to injury, such as the systemic inflammatory response or disseminated intravascular coagulation (Fig. 1). These associations may be eliminated, in part, by the recruitment of suitable comparison groups that share all of the clinical characteristics of the

Genetic factors in acute lung injury

315

Figure 1 The importance of phenotype. In establishing the true basis for a genetic association, the use of appropriate comparison groups (those above the dotted line) will reduce the possibility of spurious associations. Similarly, the use of more specific clinical and biochemical markers (below the dotted line) will enhance the strength of such associations. study group except the specific phenotype under investigation—in this case, a diagnosis of ALI/ARDS. By identifying all patients admitted to the hospital with pancreatitis, for example, those going on to develop ALI can be compared to the groups in whom ALI does not occur. Similarly, patients admitted to the ICU for non-ALI respiratory failure or critical illness may help to confirm the specificity of any associations made. Better still,

Acute respiratory distress syndrome

316

associations can be made with more specific phenotypes within the ALI population (Fig. 1). A further issue regarding patient recruitment is the issue of when they are recruited. Ideally, patients should be identified as early as possible after the onset/diagnosis of ARDS to avoid survival bias. It is also useful to know how representative recruited patients are of the ARDS population as a whole. It could be, for example, that consent may be more easily obtained from patients who are less unwell, with obvious implications for outcome. Thus, ideally, basic demographic and outcome data should be obtained on all eligible patients admitted to study centers so that severity and outcome can be compared with the study group. Finally, the involvement of patients in interventional or therapeutic clinical studies may lead to differences in clinical course and physiological/ biological parameters. If patients recruited to genetic studies are also involved in clinical trials, it is important that the treatment and control groups are analyzed separately or at least that the distribution of phenotype and genotype observed is similar. C. Phenotype Obtaining DNA from circulating leukocytes or buccal smears is a relatively easy task. Genotyping can be a time-consuming and troublesome process, but is usually achievable. Often more difficult is the ability to match genotype with phenotype. Phenotype is any clinical, physical, or biochemical characteristic that is either directly or ultimately influenced by genotype. Thus, past medical history, height, weight, blood pressure, circulating interleukin (IL)-8 concentration, and mortality are all examples of phenotype. At a basic, but important level, the genetic background and/or ethnicity of a population need to be controlled wherever possible. Dramatic variations in allele frequencies can occur across such ethnic groups. Obviously, the same data need to be obtained for each individual under study. The data must be accurate, repeatable, and should be as detailed as possible. Some elements of phenotype (e.g., histological specimens) may be difficult to obtain. However, it is equally difficult to know which data not to collect. Electronic data-retrieval systems in critical care now enable the recording of minute-to-minute variations in blood pressure, ventilation settings, and drug administration. In the future it may be possible for intelligent software to probe this enormous wealth of data in relation to genotype to search for association. Suitable data-mining techniques are currently being developed for transcriptiomic approaches and are beginning to make sense of the data mountain. At present we may have to be more specific with the data we collect. Ideally, the data should be orientated towards answering a specific hypothesis and be measured and recorded as objectively as possible. Clinical data may, of course, be retrieved in retrospect, if archived manually or electronically. By contrast, there may only be one chance to obtain appropriately timed biological specimens such as plasma and bronchoalveolar lavage (BAL). The regular acquisition of suitable specimens can only be encouraged.

Genetic factors in acute lung injury

317

D. Power The concept of a power in a genetic study is a statistical one. As in any statistical test, the larger the sample size, the more likely it is to be representative of the population as a whole. The null hypothesis (that there is no difference in allele/genotype frequencies between populations) may be rejected because of sampling errors if the cohort under study is small. The probability of making this kind of error represents the significance level of the statistical test of the hypothesis. Conversely, the null hypothesis may be accepted when it is actually false. The power in this instance is the probability that the null hypothesis will be rejected when an alternative hypothesis is true. Imperfect phenotyping, ethnic mix, and incomplete penetration of a gene are other factors that could determine the size of sample required for adequate statistical power. In many instances these factors are unknown or uncontrollable. Another key issue is the frequency of the rare allele. A disease-modifying allele that is uncommon will mean that the total number of alleles (and therefore patients) required to achieve significance is likely to be higher. Predicting the number of genotypes required to establish association conclusively will also depend upon the specificity of the phenotype. For example, establishing an association between a genetic polymorphism in the IL-8 gene and BAL neutrophil count may require fewer patients than for an association with forced vital capacity because neutrophil count is more directly related (biologically) to genotype. However, it may be that polymorphisms in which the rare allele frequency is low have a more profound effect on phenotype. E. Replication The replication of results within the same population and in different populations is, in the long term, vital. Patients from a single, large cohort may be randomized to two separate groups and then analyzed independently, but confirmatory studies in independently recruited populations is the gold standard method of testing the robustness of an association. This fact, coupled with the need for suitably powered studies, highlights the need for collaborative studies. F. Ethical Considerations Assuming that important genetic variants are found that influence the clinical course of patients at risk of ALI or with established ALI, the issue of what to do with this information arises. Underpinning the ethical acceptability of any screening process is the possibility that intervention in susceptible individuals exists. At present this may not be the case for many candidate genes/systems in ALI. Our hope must be that interventions will arise in tandem with genetic discoveries, and indeed be prompted by them. A second, broader issue is whether genetic information obtained in the context of a particular clinical scenario (e.g., ARDS) should be made available to either the patient, other clinicians, or a third party (e.g., an insurer). No doubt the refinement and implementation of suitable ethical codes will continue as important genetic information arises. Currently, genetic data obtained for the purposes of clinical studies should remain

Acute respiratory distress syndrome

318

anonymous—justifiably so pending an improved understanding of the syndrome. In the future, any data used to guide clinical management must be regarded as confidential and should be used only with a patient’s consent. A number of national and international bodies have established working papers in the field of bioethics as they relate to genetics, including The Human Genome Commission in the United Kingdom (www.hgc.gov.uk) and the National Bioethics Advisory Commission (http://bioethics.georgetown.edu/nbac/) in the United States. Finally, there is the issue of how relevant genetic information obtained from population studies will be to an individual patient. Ideally, an assessment of all environmental factors relevant to phenotype will need to be ascertained for each patient and their interaction with genotype understood before fully informed decisions regarding risk can be made. The complexity of such interactions means it is unlikely that we will be able to predict the clinical course with certainty on the basis of such information. Thus, treatment decisions will continue to be made on the basis of acceptable risk. This is analogous to the situation encountered with most therapeutic interventions, whereby a proportion of patients may receive little benefit from a treatment, despite there being an overall population benefit. Genetic studies should, of course, aid the identification of such responders and nonresponders and therefore help improve the risk/benefit ratio for an individual.

IV. Evidence for a Genetic Influence in ARDS A. Introduction That genetic factors influence the course of ALI is suggested by the observation that, despite the apparently sizable population at risk of developing ALI, relatively few do so. Similarly, we cannot yet account for outcome variability with current clinical and physiological measurements. Our clinical experience tell us that, for example, in children of a similar age suffering smoke inhalation in the same room to the same degree, some may live and some die, and we find it hard predict whom. What evidence is there for such a genetic influence? Studies in patients with ALI/ARDS are at an early stage, but both experimental evidence and studies in conditions that share common pathogenic features with ALI, such as sepsis and more chronic forms of interstitial lung disease, should encourage the search for genetic determinants. Certain animal strains show a marked difference in their response to inflammatory and injurious agents, and transgenic animals either lacking or over expressing genes involved in cytokine expression can develop lung injury. The existence of familial forms of acute and chronic interstitial lung disease also implies that genetic abnormalities can lead to the onset of ARDS either directly, perhaps by modifying the host response to an injurious event, or indirectly. Furthermore, there is often a marked variation in response to injurious agents such as ozone, even when the level of exposure is similar (2, 3). The genetic factors involved remain undefined. B. Animal Models of ALI

Genetic factors in acute lung injury

319

A strain-dependent variability in the response to experimental lung injury has been observed in a variety of species. For example, C57BL/6J mice exhibit high sensitivity to the antineoplastic agent bleomycin (4) and to ozone (5), whereas the BALB/cBY strain demonstrates a resistant phenotype. More recent studies using inbred mouse strains confirm the heritable nature of this sensitivity to a number of injurious agents. Wesselkamper and colleagues have examined the sensitivity to NiSO4, ozone, and polytetrafluoroethylene in seven inbred strains, in which the sensitivity patterns of injury were in agreement (6). There was, however, phenotype discordance with regard to lavage fluid protein and neutrophil count. These data suggest that such sensitivities may be controlled by different genes sharing a common mechanism. Genome-wide searches in animal models have identified a number of quantitative trait loci (QTL) that associate with strain susceptibility (Table 1). NiSO4-induced ALI appears to be a complex trait controlled by at least five genes, including aquaporin, transforming growth factor-alpha (TGF-α), and surfactant protein B (7, 8). Similarly, an association between ozone/nickel injury and chromosome 6 has been identified by a number of investigators (Table 1). Candidate genes in this region include the major histocompatability complex locus, which is both biologically plausible and likely, as this region contains a cluster of genes for proinflammatory cytokines as well as HLA genes. Kleeberger and colleagues (9) have identified a significant QTL on chromosome 4 in relation to ozone-induced injury that contains a candidate gene: toll-like receptor 4 (TLR4), which has been implicated in endotoxin

Table 1 Genetic Studies of Mouse Strain Susceptibility to Lung Injury Mouse strain

Injury model

Chromosomal region/ candidate gene

Ref.

C57BL/6J and C3H/HeJ

Ozone

Chromosome 17/ chromosome 6 (TNFα)

80

A/J and C57BL/6J

Ozone

MPO, GLT peroxidase

81

C57BL/6J and C3H/HeJ

Ozone

Chromosome 4/TLR4

9

A/J and C57BL/6J

Nickel

Chromosome 6 (major) chromosomes 1 and 12 (suggested), 9 and 16 (contributory)

8

C57BL/6J and C3H/HeJ

Hyperoxia

Chromosome 2 and 3/NRF2

11

A/J and C57BL/6J

Ozone, PTFE, nickel

Chromosome 6 (major), chromosomes 1, 8, and 12 (suggested), chromosomes 9 and 16 (contributory)

82

A/J and C57BL/6J

Nickel

Chromosome 6 (major), 9 and 16 (contributory)/ SPB, aquaporin 1, and TGF-α

7

PTFE, polytetrafluoroethylene; NFR2, nuclear factor, erythroid derived-2; TNFα, tumor necrosis factor-alpha; TGF-α, transforming growth factor-alpha; SP-B, surfactant protein B; TLR4, toll-like receptor 4; MPO, myeloperoxidase; GLT, glutathione.

Acute respiratory distress syndrome

320

recognition. The QTL explained 70% of the total trait variance in this model. To test the role of TLR4 in ozone-induced hyperpermeability, BAL protein responses to ozone were compared in C3H/HeOuJ (OuJ) and HeJ mice that differ only at a polymorphism in the coding region of TLR4. Significantly greater protein concentrations were found in OuJ mice compared with HeJ mice after exposure to ozone, as well as the differential expression of TLR4 message between the two strains. As an example of how animal studies can generate candidates for human studies, Lorenz and colleagues examined TLR4 genotypes in 91 patients with septic shock as well as 73 healthy blood donor controls for the presence of the TLR4 Asp299Gly and TLR4 Thr399Ile mutations (10). The TLR4 Asp299Gly allele was found exclusively in the patients with septic shock. Furthermore, patients with septic shock with the TLR4 Asp299Gly/ Thr399Ile alleles had a higher prevalence of gram-negative infections. Recently, Cho and colleagues identified NF-erythroid-related factor-2 (NRF2) on chromosome 2 as a possible candidate gene in hyperoxia-induced lung injury based on a genome-wide search in C57BL/6J (susceptible) and C3H/HeJ (resistant) mice (11). NRF2 regulates antioxidant and phase 2 gene expression and is thus of great interest in the context of ALI. In a subsequent study, NRF2-/- mice had an enhanced injury response to hyperoxia (12), suggesting a protective role for this transcription factor in oxidative lung injury. More specific mutations in single genes also occur that may lead to diffuse lung injury. The motheaten gene, for example, is a single recessive mutation that occurs in mice leading to a deficiency of hematopoietic cell phosphatase (HCP). Homozygotes (me/me) die from interstitial lung disease. Histological and BALF findings show evidence of alveolitis and both alveolar and interstitial fibrosis (13). HCP is of particular importance in B-lymphocyte intracellular signaling. However, this enzyme is expressed in many cell types, and, given the ubiquitous role of phosphatases in cellular function, it is not surprising that the phenotype is pleiotropic and does not depend on the presence of T or B lymphocytes, suggesting other cell types are involved (14). Its relationship to ARDS/ALI is perhaps tenuous, but it does highlight the ability of single genetic mutations to promote a diffuse injury response in the lung. These observations may also have relevance to familial forms of diffuse lung disease (see below). The relationship between animal studies and ALI in humans is unclear. Certainly they demonstrate the potential for genetic susceptibility in this condition and also suggest important candidate genes that might be examined in clinical studies. It is also likely that observations of genetic association in humans will prompt experiments in appropriate animal models to help clarify the mechanism by which a particular candidate gene is influencing pathogenesis. C. Familial Interstitial Lung Disease Familial syndromes are of particular interest as they are amenable to study with powerful genome-wide approaches. Although familial ALI has not been described, studies in other forms of acute and chronic lung injury strongly suggest genetic determinants and are worthy of consideration here. Hamman and Rich, published four cases of “acute diffuse fibrosis of the lung” in 1944 (15). These initial cases were acute in onset and produced a rapidly fatal course.

Genetic factors in acute lung injury

321

Although the Hamman-Rich syndrome has been synonymous with idiopathic pulmonary fibrosis (IFF) in the past, it is now referred to as a acute interstitial pneumonia (AIP). Histologically, AIP is characterized by diffuse alveolar damage and is thus pathologically similar to ARDS. Protein and fluid exudation into the lung is accompanied by neutrophilic infiltration, hyaline membrane formation, and progressive fibrosis (16). In the early descriptions of Hamman-Rich syndrome, it was soon recognized that there was a distinct subgroup in which the disease apeared to be familial. It can be difficult to differentiate true inheritance of a disease from the mere clustering of cases within a family exposed to a common environmental factor. However, cases reported so far, such as those occurring in twins and family members separated from an early age (17), strongly suggest the existence of a familial form (18). Such cases appear histologically indistinguishable from the nonfamilial forms, suggesting shared pathogenic pathways. However, one important difference between familial and nonfamilial cases appears to be the age of onset. In familial AIP, cases occur in infancy or early childhood. It is possible that such cases represent a distinct disease entity with a pathogenic mechanism different from that of adult cases and could result from a major genetic determinant. A recent study in one large kindred with usual interstitial pneumonitis has suggested an association with surfactant protein C polymorphism (19), although this has yet to be confirmed by genome-wide analysis. In chronic forms of interstitial lung disease such as IFF (20) and sarcoidosis (21), familial clustering is observed, although no confirmatory genetic studies have been performed to date. A number of other familial syndromes are associated with the development of interstitial lung disease. The Hermansky-Pudlak syndrome is an autosomal recessive disorder characterized by oculocutaneous tyrosinase-positive albinism, platelet dysfunction, and ceroid-like inclusions in the reticuloendothelial system. Progressive pulmonary fibrosis with a poor response to treatment was described in the original case report by Hermansky and Pudlak in 1959 (22). The gene responsible is identified as a novel trans-membrane protein and a component of multiple cytoplasmic organelles (23). An association between pulmonary fibrosis and familial hypercalcaemic hypocalcuria has also been reported (24, 25). This autosomal dominant condition appears to be the result of a mutation in the Ca2+-sensing receptor gene on chromosome 3 (26). As with the animal model studies, the exact relationship between these conditions and ALI is uncertain. They certainly establish the principle of a genetic influence in diffuse lung injury and, again, may suggest candidate genes worthy of study in ALI. D. Incidence of ARDS in At-Risk Patients Perhaps the most compelling evidence for a genetic predisposition to ALI in humans is the sample observation that, despite the large number of patients at risk for ALI, relatively few develop the disease. In a study of 993 non-ICU hospital patients with at least one risk factor for the development of ARDS, the incidence was 5.5 per hundred patients (27). Patients with more than one risk factor were more likely to develop ARDS (incidence: 24.6 per hundred). Similarly, Nelson and colleagues (28) found an incidence of approximately 1–2 per 100 patients hospitalized with severe community-acquired pneumonia. The incidence is higher in ICU populations. Hudson and colleagues prospectively identified 695 patients meeting the criteria for seven clinical risk factors for

Acute respiratory distress syndrome

322

ARDS admitted to the ICU (29). Overall the incidence of ARDS was 26%, being the highest in patients with sepsis syndrome (43%) and those with multiple transfusions (40%). In patients with multiple trauma, 25% developed ARDS, but this increased to 40% if more than one risk factor was present. Increased risk for ARDS was also associated with an elevated Acute Physiologic and Chronic Health Evaluation II (APACHE II) score in patients with sepsis and increased APACHE II and Injury Severity Scores in trauma victims. Pepe and colleagues found similar incidence figures in 136 atrisk ICU patients: sepsis syndrome (38%), gastric aspiration (30%), multiple transfusions (24%), and pulmonary contusion (17%) (30). The overall risk associated with one factor was 25%, increasing to 42% and 85% when two or three factors, respectively, were present. Thus, in these studies the development of ARDS was at least partially related to the severity of injury. But the crucial question remains as to whether “severity” in this instance relates to the initial insult, e.g., bacterial load, or to host response. Is the injury response in patients who go on to develop ARDS proportional to the insult, or is it excessive due to variations in physiology that are genetically regulated? Table 2 summarizes the findings from a number of at-risk populations. Taken together, these data suggest that 10–40% of patients at risk of ARDS may develop the syndrome and that the incidence appears to increase in the presence of multiple risk factors. Of key importance is the observation that the incidence of ARDS increases depending on when the population is sampled. The incidence rises from 0.28% in all hospital admissions (31) to 1–5% in hospital cohorts with risk factors. The incidence rises again to 10–50% for ICU populations. Obviously patients admitted to the ICU have selfselected as individuals with more severe injury/sepsis, etc. This is of crucial importance when one considers when a gene-environment interaction may be occurring. If the hypothesis to be tested is that a gene will influence the progression from communityacquired infection to severe pneumonia, then patients in the community with pneumonia need to be followed from the point of first clinical presentation to hospital admission. Furthermore, the candidate genes in this instance might relate to innate immunity or bacterial invasion (10) (Toll-like receptors are of great interest

Table 2 Incidence of ARDS in At-Risk Populations Setting Non-ICU

ICU

Risk factor All hospital admissions

Number at risk

Incidence of ARDS (%)

Ref.

8100

0.28

31

Several

993

5.5

27

Pneumonia

897

1–2.0

28

Overall

695

26

29

Sepsis

43

Transfusion

40

Trauma

25

Overall Sepsis

136

30 38

Genetic factors in acute lung injury

323

Aspiration

30

Transfusion

24

Pulmonary contusion

17

Thermal injury

126

14

77

Traumatic brain injury

100

20

78

Bleeding esophageal varices

101

14

79

in this respect). However, if the gene is believed to modify the progression to sepsis, then patients with severe pneumonia need to be followed, and so on. Such considerations also influence the selection of suitable control groups. An alteration of gene frequency in a cohort of patients admitted with ARDS might be the result of an association, not with pathophysiological mechanisms specific to ALI, but with the severity of an acute inflammatory response, with pulmonary vascular tone, or with poor ventilatory drive. This brings us back to a fundamental point: that there is no gene for ARDS and, furthermore, that the most powerful genetic studies will match genotype with a phenotype that more specifically represents the biological function of its protein.

V. Clinical Studies in Sepsis/ARDS A. Introduction It is already apparent that for the purposes of genetic study, dividing critically ill patients in to separate diagnostic groups (ARDS, sepsis, pneumonia, multiorgan failure, etc.) may be wholly artificial and misleading. It is likely that there will be considerable overlap in the genetics of these clinical syndromes, the result of common pathophysiological mechanisms. Table 3

Table 3 Genetic Studies of Susceptibility to Sepsis and ARDS Study group (n)

Comparison group (n)

Polymorphism

Association(s)

Ref.

Sepsis (91)

Healthy (73)

TLR4 (Asp299Gly/ Thr399Ile)

Susceptibility

10

Sepsis (326)

Healthy (326)

IL-6 (G-174C)

Mortality

56

Communityacquired pneumonia

N/A

I-CAM (Gly241Arg)

Susceptibility

83

Sepsis (78)

Uncomplicated pneumonia (56) Healthy (130)

IL-1Ra (intron 2)

Mortality

84

Acute respiratory distress syndrome

324

Communityacquired pneumonia (280)

N/A

TNFα (−308) LTβ (+250)

Susceptibility (LTβ+250)

39

Sepsis survivors (50)

First-degree relatives (183)

PAI-1 (4G/5G)

Susceptibility

60

Sepsis (175)

Healthy (226)

PAI-1 (4G/5G)

Susceptibility

59

Sepsis (89)

Healthy (87)

TNFα (−308)

Susceptibility and mortality

38

Trauma (110)

N/A

LTβ (+250)

Susceptibility

85

Sepsis (87)

Healthy (110)

HSP70 (-HOM C/T, No association with −2A/G) either susceptibility or mortality

86

Sepsis (40)

None

TNFα (−308)

Susceptibility

37

ARDS (96)

ICU (88)

ACE (I/D)

Susceptibility

75

ARDS (112)

ICU (88)

IL-6 (−174 G/C)

Susceptibility

75

ARDS (52)

At risk (pneumonia, thoracic trauma, cardiac edema)

SP-B (1580 G/T)

Susceptibility

76

TLR-4, toll-like receptor 4; I-CAM, intracellular adhesion molecule-1; IL-1Ra, interleukin-1 receptor antagonist; TNFα, tumor necrosis factor alpha; PAI-1, plasminogen activator inhibitor-1; LTβ, lymphotoxin beta (TNF-beta); HSP70, heat shock protein 70; ACE, angiotensin-converting enzyme; IL-6, interleukin-6; SP-B, surfactant protein B.

summarizes a number of a key studies performed to date, the majority in sepsis. All the candidates studied so far have a putative role in disease pathogenesis and exhibit polymorphism in which the rare allele frequency is relatively common. This improves the potential for detecting differences in genotype/allele frequencies in relatively small populations, with the caveat that very rare alleles may have a greater effect on phenotype. A more detailed discussion of studies in three key candidates for sepsis highlights some of the issues in study design and data analysis. Not surprisingly, these polymorphisms are mainly in pro-inflammatory cytokine genes, particularly the tumor necrosis factor alpha (TNFα; −308) and lymphotoxin beta (LTβ; +250) loci. B. Tumor Necrosis Factor Alleles of both the TNFα (−308) and LTβ (+250) polymorphisms have been associated with higher TNFα levels (32, 33). Moreover, chromosome 6 (the site of the TNFα locus) has been identified in a number of susceptibility studies to ALI in animals (Table 1). TNFα is a key component of the immunity to invading pathogens that are often the presumed trigger of severe sepsis (34). TNFα drives the production of other proinflammatory cytokines and has been correlated with severity and outcome from sepsis in a number of studies (35, 36).

Genetic factors in acute lung injury

325

Stuber et al. (37) reported no association with the common TNFα (−308) polymorphism and either susceptibility or outcome. This is contradicted by a later report in a French population (38). In both studies healthy controls were used as a comparison group, whereas an at-risk group in whom septic shock did not occur might have been preferable to examine susceptibility. Waterer et al. (39) genotyped and followed 280 patients with community-acquired pneumonia, 31 of whom developed sepsis and 80 type 1 respiratory failure (diagnosed on oxygen criteria alone). In this small group of patients with sepsis, an association was seen with the LTα+250 AA genotype (associated in turn with high TNFα levels) (Table 4). No interaction was detected on linear regression between genotype and clinical factors associated with a poor outcome from pneumonia (e.g., comorbidity or delayed antibiotic therapy), suggesting that the association was with a process intrinsic to the inflammatory response in these individuals. Interestingly, the low secretor genotype (GG) was associated with respiratory

Table 4 Risk of Septic Shock by LT+250:TNF−308 Haplotype in Patients with Community-Acquired Pneumonia Risk (%) Number of haplotypes

2

1

0.5

0

250 A:

0/0

0/8

3/51

28/221

308 A

(0%)

(0%)

(5.9%)

(12.7%)

250 A:

16/84

7/78

3/51

5/67

308 G

(19.0%)

(9.0%)

(5.9%)

(7.5%)

250 G:

3/11

1/21

3/51

24/197

308 A

(27.3%)

(4.8%)

(5.9%)

(12.2%)

250 G:

1/35

8/91

3/51

19/103

308 G

(2.6%)

(8.8%)

(5.9%)

(18.4%)

p-value 0.097

0.014

>0.1

0.011

Source: Ref. 39.

failure. In may be that these individuals had a prolonged or worsening pneumonic process because of the inappropriately low TNF secretion. Too much TNF and sepsis ensues; too little leads to the poor resolution of infection. The authors do not quote the incidence of ALI in these patients, but one might predict that any such cases would have occurred in the sepsis group rather than the respiratory failure group. C. Interleukin-6 IL-6 is another orchestrator of the inflammatory cascade and a candidate gene for both sepsis and ARDS. IL-6 is derived from diverse tissues (40, 41). Inflammatory stimuli themselves (such as smoking) (42, 43) and their associated hormonal or cytokine responses (41, 44) cause IL-6 levels to rise (45), which then drive the acute inflammatory

Acute respiratory distress syndrome

326

response. IL-6 is the only cytokine that can induce all hepatic acute-phase proteins. High levels of cytokines such as IL-6 are predictive of multiple organ failure in patients with community-acquired pneumonia (46). They also predict mortality in septic states (47) and in patients with acute respiratory distress syndrome (48–50). A common functional polymorphism of the IL-6 gene has recently been detected, with a cytosine (C) to guanine (G) switch (51). In vitro at baseline, the polymorphism makes little difference in the level of activity of the gene. However, following stimulation with endotoxin, promoter activity in the presence of the C allele changes little, while with the G allele it increases by up to 360%. The same effect may hold true in vivo, although the results are conflicting. The circulating IL-6 levels in those exposed to the inflammatory stimulus of cigarette smoke are approximately twice as high in individuals with the GG genotype as those homozygous for the C allele (52). Similarly, in 111 patients with multivessel coronary artery disease undergoing elective coronary artery bypass graft surgery, Burzotta and colleagues (53) found that IL-6 levels following surgery (but not fibrinogen, white-blood cell count, or C-reactive protein values) differed significantly according to the -174 genotype, with GG homozygotes exhibiting higher concentrations than carriers of the C allele. GG homozygosity was associated with longer stays in the intensive care unit and in hospital. There was also a trend towards an association with rates of postoperative death in the GG individuals. By contrast, Brull et al. (54) found higher circulating IL-6 levels in patients post-bypass surgery with the -174CC genotype compared with -174G allele carriers. In an additional study of healthy individuals, the 174C allele was associated with higher plasma C-reactive protein levels, although IL-6 levels themselves were not measured directly (55). The reasons for such conflicts are unclear but may relate to study size, the timing of plasma sampling, ethnicity, or other cogenetic factors. It is also possible that the polymorphism may differentially regulate gene expression according to tissue or cell type. Certainly, in the case of IL-6 and critical illness, there are stark contrasts between studies that support this notion In a prospective study of 326 German Caucasian surgical patients, 50 developed sepsis (56). IL-6 -174 genotype distribution and allele frequencies did not differ significantly between patients with or without sepsis and healthy controls. In non-survivors of sepsis, however, significantly fewer GG homozygotes were observed compared with survivors. In the septic patients, IL-6 levels correlated closely with outcome but not with genotype. Heesen et al. (57) found no association between the IL-6 -174 genotypes and the ex vivo, stimulated IL-6 response in severely injured blunt trauma patients. The 25% of these individuals who developed severe sepsis had higher IL-6 concentrations following whole blood stimulation on day 1 compared to those with uncomplicated posttraumatic recovery. The difference was even more significant on day 2. Thus, at present, the role of this polymorphism in sepsis remains unresolved. D. Plasminogen Activator Inhibitor-1 Other candidates that deserve attention in the context of sepsis and ALI include members of the coagulation cascade. The importance of the coagulation cascade in the critically ill has been particularly highlighted by the results of the PROWESS study of activated protein C therapy (58). Plasminogen activator inhbitor-1 (PAI-1) has received particular attention. Hermans et al. (59) were the first to describe an association between 4G/ 4G

Genetic factors in acute lung injury

327

genotype of the PAI-4G/5G polymorphism and mortality in children with meningococcal disease. The 4G allele produces 6 times more RNA than the 5G allele. An association has also been observed by a separate group between the 4G/4G genotype and the development of septic shock in these patients (60). In 61 patients with severe trauma, the 4G/4G genotype was associated with higher plasma PAI-1 levels and mortality (61). Other candidates in the coagulation/fibrinolytic cascade include α-thrombin and fibrinogen (62–64) and are worthy of further study. So far, studies in sepsis have identified a number of potential genes that appear to modify susceptibility and outcome. In the case of PAI-1, the results are highly consistent; those with IL-6 are less so. However, each study is in some way informative and should be scrutinized for subtle differences in patient recruitment, genetic background, and the presence of environmental modifiers that might have influenced eventual phenotype.

VI. Clinical Studies in ARDS To date, genetic studies in patients with ARDS have been limited, although this will not remain the case for long. A number of national and international collaborations are already underway to collect suitable cohorts for genetic study. At the very least, the results of studies performed so far appear to provide suitable encouragement. A. Angiotensin-Converting Enzyme We have examined polymorphisms in a candidate system of interest: the reninangiotensin system (RAS). There is considerable evidence to support the existence of local RAS in a number of human tissues, including the lung. Moreover, an elevation in bronchoalveolar lavage fluid/serum ACE activity and an increase in circulating angiotensin II concentrations have been described in ARDS (65). The relevance of circulating ACE levels in uncertain, given that this may represent mere shedding of the enzyme from the endothelial surface and may not reflect tissue levels. Certainly, activation of a local RAS within the pulmonary circulation and lung parenchyma could influence the pathogenesis of ARDS/ALI via an increase in vascular permeability (66), vascular tone (67), and fibroblast activity (68) and by reducing alveolar epithelial cell survival (69), mechanisms pertinent to the onset and severity of lung injury. Large interindividual differences in plasma ACE levels exist, but levels are similar within families (70), suggesting a strong genetic influence in the control of ACE levels. A restriction fragment length polymorphism (71) in the human ACE gene exists, consisting of the presence (insertion, I) or absence (deletion, D) of a 287-base-pair alu repeat sequence, has been identified (72). Among 80 healthy Caucasians the I/D polymorphism accounted for 47% of the variance in plasma ACE, being highest in those with the DD genotype (71). Tissue ACE levels appear to be similarly influenced. ACE activity in those of DD genotype is thus 75% and 39% higher in plasma and T-lymphocytes, respectively, than in those of II genotype (73). It is believed that this intronic (noncoding) polymorphism may be a marker for another genetic locus/loci with more functional significance, although extensive haplotyping of the region has excluded the promoter region and a number of exons (74).

Acute respiratory distress syndrome

328

In 96 patients with ARDS, the frequency of the DD genotype and D allele was markedly increased in patients with ARDS compared to patients undergoing coronary bypass surgery, patients with acute respiratory failure on the ICU, and a healthy population sample (Table 5) (75). The D allele was also associated with increased mortality in the ARDS group. Given this

Table 5 ACE Genotype and Allele Frequencies in ARDS and Comparison Groups Genotypea II

ID

DD

p (genotype)b

Freq. D allele (95% CI)

ARDS

9 (0.09)

43 (0.45)

44 (0.46)

CABG

40 (0.23)

90 (0.52)

44 (0.25)

0.0009

0.51 (0.46–0.56)

ICU

31 (0.35)

36 (0.41)

21 (0.24)

0.00008

0.44 (0.37–0.52)

459 (0.24)

949 (0.50)

498 (0.26)

0.00004

0.51 (0.49–0.53)

Healthy population a b

0.69 (0.61–0.75)

No. of patients (frequency). Chi-squared analysis for genotype distribution between ARDS and comparison.

result, one might expect the effect of alleles associated with increased angiotensin receptor or angiotensinogen expression to be additive with those of ACE if angiotensin II is the major biological correlate of the association with the D allele. If however, some other action of ACE underlies this association, a relationship with these genotypes may be absent. Thus examining multiple genotypes within a candidate system, such as this, may help to delineate the biological pathways involved. B. Interleukin-6 In the same cohort of patients, the C allele of the IL-6 (-174 G/C) polymorphisms was associated with survival in both patients with ARDS and those with acute respiratory failure. The C allele was also associated with lower IL-6 concentrations in ARDS patients. These data suggest that IL-6 modifies outcome in critically ill patients (certainly ventilated patients), and not just those with ARDS. This could perhaps point to the influence of ventilation itself on the inflammatory response. It also highlights the importance of control groups for the use of specific phenotypes rather than clinical syndromes. This result is contradicted by findings in patients at risk for sepsis in which the GG genotype was associated with a reduced mortality (56). The reasons for such discrepancies are difficult to explain given that both syndromes are markedly inflammatory in nature but, as suggested above, may also reflect differences in the phenotype expressed by particular cells or tissue compartments. Either way, confirmatory studies are required.

Genetic factors in acute lung injury

329

C. Surfactant Protein B In a study of 19 polymorphisms in surfactant protein genes (76), differences in allele frequency for microsatellite markers flanking SP-B and a 1580 C/T polymorphism of the SP-B gene were identified. This polymorphism may determine the presence or absence of an N-linked glycosylation site. Multivariate analysis revealed significant differences only for the C/T (1580) polymorphism, with the C allele being more common in the ARDS group. The odds ratio (0.412; CI 0.179–0.878)) suggests that this allele may be a susceptibility factor for the development of ARDS. On dividing the ARDS population into subgroups such as trauma, sepsis, etc., the C allele was more significantly associated with “idiopathic” ARDS, although this division is not conventional, as it mostly included individuals with pneumonia. Of interest, surfactant protein C has been identified as a QTL in a recent study of familial UIP (19). An examination of this and additional polymorphisms in surfactant proteins is warranted.

VII. Conclusions Given the complexity of the pathogenic mechanisms involved in ALI, there is often difficulty in differentiating the more generalized, stereotypical inflammatory responses from the key, disease-specific processes. Therefore, detecting genotypic variation in patients within ALI could help us to focus research efforts in this regard. There is certainly circumstantial evidence-recently backed up by clinical studies—to support an important role for genetic variation in determining both the susceptibility to and the outcome from ALI. Studies carried out so far highlight the need for the use of modern molecular genetic approaches in both animal models and in patients. Such studies will undoubtedly guide us to targets for pharmaceutical intervention and aid in prognostication and therapeutic targeting. A combination of approaches is required. Large-scale collaborative studies will undoubtedly highlight important haplotypes associated with ALI and its outcome. Equally important, however, will be smaller, more focused studies examining small numbers of alleles in relation to more detailed phenotypic data obtained from tissue, biological fluids, and cells. Similarly, studies in healthy/noncritically ill individuals receiving suitable physiological challenges will be required to understand more precisely the interaction between environment and gene. If there are watchwords to guide us in this process, they might be hypothesis, phenotype, and collaboration.

References 1. Fors L, Lieder KW, Varvra SH, Kwiatkowski RW. Large-scale SNP scoring from unamplified genomic DNA. Pharmacogenomics 2000; 1(2):219–229. 2. Holz O, Jorres RA, Timm P, Mucke M, Richter K, Koschyk S, et al. Ozone-induced airway inflammatory changes differ between individuals and are reproducible. Am J Respir Crit Care Med 1999; 195(3):776–784.

Acute respiratory distress syndrome

330

3. Jorres RA, Holz O, Zachgo W, Timm P, Koschyk S, Muller B, et al. The effect of repeated ozone exposures on inflammatory markers in bronchoalveolar lavage fluid and mucosal biopsies. Am J Respir Crit Care Med 2000; 161(6): 1855–1861. 4. Schrier DJ, Kunkel RG, Phan SH. The role of strain variation in murine bleomycin-induced pulmonary fibrosis. Am Rev Respir Dis 1983; 127(1): 63–66. 5. Ohtsuka Y, Munakata M, Ukita H, Takahashi T, Satoh A, Homma Y, et al. Increased susceptibility to silicosis and TNF-alpha production in C57BL6J mice. Am J Respir Crit Care Med 1995; 152(6 Pt 1):2144–2149. 6. Wesselkamper SC, Prows DR, Biswas P, Willeke K, Bingham E, Leikauf GD. Genetic susceptibility to irritant-induced acute lung injury in mice. Am J Physiol Lung Cell Mol Physiol 2000; 279(3):L575–L582. 7. Leikauf GD, McDowell SA, Wesselkamper SC, Hardie WD, Leikauf JE, Korfhagen TR, et al. Acute lung injury: functional genomics and genetic susceptibility. Chest 2002; 121(3 suppl):70S–75S. 8. Prows DR, Leikauf GD. Quantitative trait analysis of nickel-induced acute lung injury in mice. Am J Respir Cell Mol Biol 2001; 24(6):740–746. 9. Kleeberger SR, Reddy S, Zhang LY, Jedlicka AE. Genetic susceptibility to ozone-induced lung hyperpermeability: role of toll-like receptor 4. Am J Respir Cell Mol Biol 2000; 22(5):620–627. 10. Lorenz E, Mira JP, Cornish KL, Arbour NC, Schwartz DA. A novel polymorphism in the tolllike receptor 2 gene and its potential association with staphylococcal infection. Infect Immun 2000; 68(11):6398–6401. 11. Cho HY, Jedlicka AE, Reddy SP, Zhang LY, Kensler TW, Kleeberger SR. Lingage analysis of susceptibility to hyperoxia. Nrf2 is a candidate gene. Am J Respir Cell Mol Biol 2002; 26(1):42–51. 12. Cho HY, Jedlicka AE, Reddy SP, Kensler TW, Yamamoto M, Zhang LY, et al. Role of NRF2 in protection against hyperoxic lung injury in mice. Am J Respir Cell Mol Biol 2002; 26(2): 175–182. 13. Rossi GA, Hunninghake GW, Kawanami O, Ferrans VJ, Hansen CT, Crystal RG. Motheaten mice—an animal model with an inherited form of interstitial lung disease. Am Rev Respir Dis 1985; 131(1):150–158. 14. Rathbone B, Martin D, Stephens J, Thompson JR, Samani NJ. Helicbacter pylori seropositivity in subjects with acute myuocardial infarction. Heart 1996; 76(4):308–311. 15. Hamman AR, Rich L. Acute diffuse interstitial fibrosis of the lungs. Bull Johns Hopkins Hosp 1944; 74:177–179. 16. Bouros D, Nicholson AC, Polychronopoulos V, du Bois RM. Acute interstitial pneumonia. Eur Respir J 2000; 15(2):412–418. 17. Swaye P, Van Ordstr HS, McCormack LJ, Wolpaw SW. Familial Hamman-Rich syndrome. Report of eight cases. Dis Chest 1969; 55(1):7–12. 18. Marshall RP, McAnulty RJ, Laurent GJ. The pathogenesis of pulmonary fibrosis: is there a fibrosis gene? Int J Biochem Cell Biol 1997; 29(1): 107–120. 19. Thomas AQ, Lane K, Phillips J III, Prince M, Markin C, Speer M, et al. Heterozygosity for a surfactant protein C gene mutation associated with usual interstitial pneumonitis and cellular nonspecific interstitial pneumonitis in one kindred. Am J Respir Crit Care Med 2002; 165(9):1322–1328. 20. Marshall RP, Puddicombe A, Cookson WO, Laurent GJ. Adult familial cryptogenic fibrosing alveolitis in the United Kingdom. Thorax 2000; 55:143–146. 21. Rybicki BA, Iannuzzi MC, Frederick MM, Thompson BW, Rossman MD, Bresnitz EA, et al. Familial aggregation of sarcoidosis. A case-control etiologic study of sarcoidosis (ACCESS). Am J Respir Crit Care Med 2001; 164(11): 2085–2091. 22. Hermansky F, Pudlak P. Albinism associated with hermorrhagic diathesis and unusual pigmented reticular cells in the bone marrow. Report of two cases with histochemical studies. Blood 1959; 14:162.

Genetic factors in acute lung injury

331

23. Oh J, Bailin T, Fukai K, Feng GH, Ho L, Mao JI, et al. Positional cloning of a gene for Hermansky-Pudlak syndrome, a disorder of cytoplasmic organelles. Nat Genet 1996; 14(3):300–306. 24. Auwerx J, Demedts M, Bouillon R, Desmet J. Coexistence of hypocalciuric hypercalcaemia and interstitial lung disease in a family: a cross-sectional study. Eur J Clin Invest 1985; 15(1):6– 14. 25. Tran Van Nhieu J, Misset B, Lebargy F, Carlet J, Bernaudin JF. Expression of tumor necrosis factor-alpha gene in alveolar macrophages from patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 147(6 pt 1):1585–1589. 26. Chou YH, Pollak MR, Brandi ML, Toss G, Arnqvist H, Atkinson AB, et al. Mutations in the human Ca(2+)-sensing-receptor gene that cause familial hypocalciuric hypercalcemia. Am J Hum Genet 1995; 56(5): 1075–1079. 27. Fowler AA, Hamman RF, Good JT, Benson KN, Baird M, Eberle DJ, et al. Adult respiratory distress syndrome: risk with common predispositions. Ann Intern Med 1983; 98(5 Pt 1):593– 597. 28. Nelson S, Belknap SM, Carlson RW, Dale D, DeBoisblanc B, Farkas S, et al. A randomized controlled trial of filgrastim as an adjunct to antibiotics for treatment of hospitalized patients with community-acquired pneumonia. CAP Study Group. J Infect Dis 1998; 178(4): 1075– 1080. 29. Hudson LD, Milberg JA, Anardi D, Maunder RJ. Clinical risks for development of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151(2 Pt 1):293–301. 30. Pepe PE, Potkin RT, Reus DH, Hudson LD, Carrico CJ. Clinical predictors of the adult respiratory distress syndrome. Am J Surg 1982; 144(1):124–130. 31. Goh AY, Chan PW, Lum LC, Roziah M. Incidence of acute respiratory distress syndrome: a comparison of two definitions. Arch Dis Child 1998; 79(3): 256–259. 32. Pociot F, Briant L, Jongeneel CV, Molvig J, Worsaae H, Abbal M, et al. Association of tumor necrosis factor (TNF) and class II major histocom-patibility complex alleles with the secretion of TNF-alpha and TNF-beta by human mononuclear cells: a possible link to insulin-dependent diabetes mellitus. Eur J Immunol 1993; 23(1):224–231. 33. Bouma G, Crusius JB, Oudkerk PM, Kolkman JJ, von Blomberg BM, Kostense PJ, et al. Secretion of tumour necrosis factor alpha and lymphotoxin alpha in relation to polymorphisms in the TNF genes and HLA-DR alleles. Relevance for inflammatory bowel disease. Scand J Immunol 1996; 43(4): 456–463. 34. Bazzoni F, Beutler B. The tumor necrosis factor ligand and receptor families. N Engl J Med 1996; 334(26):1717–1725. 35. Debets JM, Kampmeijer R, van der Linden MP, Buurman WA, van der Linden CJ. Plasma tumor necrosis factor and mortality in critically ill septic patients. Crit Care Med 1989; 17(6):489–494. 36. Waage A, Halstensen A, Shalaby R, Brandtzaeg P, Kierulf P, Espevik T. Local production of tumor necrosis factor alpha, interleukin 1, and interleukin 6 in meningococcal meningitis. Relation to the inflammatory response. J Exp Med 1989; 170(6): 1859–1867. 37. Stuber F, Petersen M, Bokelmann F, Schade U. A genomic polymorphism within the tumor necrosis factor locus influences plasma tumor necrosis factor-alpha concentrations and outcome of patients with severe sepsis [see comments]. Crit Care Med 1996; 24(3):381–384. 38. Mira JP, Cariou A, Grall F, Delclaux C, Losser MR, Heshmati F, et al. Association of TNF2, a TNF-alpha promoter polymorphism, with septic shock susceptibility and mortality: a multicenter study. JAMA 1999; 282(6):561–568. 39. Waterer GW, Quasney MW, Cantor RM, Wunderink RG. Septic shock and respiratory failure in community-acquired pneumonia have different TNF polymorphism associations. Am J Respir Crit Care Med 2001; 163(7): 1599–1604. 40. Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochem J 1990; 265(3):621–636.

Acute respiratory distress syndrome

332

41. Yudkin JS, Yajnik CS, Mohamed-Ali V, Bulmer K. High levels of circulating proinflammatory cytokines and leptin in urban, but not rural, Indians. A potential explanation for increased risk of diabetes and coronary heart disease. Diabetes Care 1999; 22(2):363–364. 42. Jansky L, Pospisilova D, Honova S. immune system of cold-exposed and cold-adapted humans. Eur J Appl Physiol 1996; 72:445–450. 43. Mohler ER3, Sorensen LC, Ghali JK, Schocken DD, Willis PW, Bowers JA, et al. Role of cytokines in the mechanism of action of amlodipine: the PRAISE Heart Failure Trial. Prospective Randomized Amlodipine Survival Evaluation. J Am Coll Cardiol 1997; 30(1):35– 41. 44. Tracy RP, Iemaitre RN, Psaty BM, Ives DG, Evans RW, Cushmas M, et al. Relationship of Creactive protein to risk of cardiovascular disease in the elderly. Results from the Cardiovascular Health Study and the Rural Health Promotion Project. Arterioscler Thromb Vasc Bil 1997; 17(6): 1121–1127. 45. Nieman DC. Immune response to heavy exertion. J Appl Physiol 1997; 82(5): 1385–1394. 46. Takala J, Ruokonen E, Webster NR, Nielsen MS, Zandstra DF, Vundelinckx G, et al. Increased mortality associated with growth hormone treatment in critically ill adults. N Engl J Med 1999; 341(11):785–792. 47. Torpy DJ, Bornstein SR, Chrousos GP. Leptin and interleukin-6 in sepsis. Horm Metab Res 1998; 30(12):726–729. 48. Meduri GU, Headley S, Tolley E, Shelby M, Stentz F, Postlethwaite A. Plasma and BAL cytokine response to corticosteroid rescue treatment in late ARDS. Chest 1995; 108(5):1315– 1325. 49. Meduri GU, Kohler G, Headley S, Tolley E, Stentz F, Postlethwaite A. Inflammatory cytokines in the BAL of patients with ARDS Persistent elevation over time predicts poor outcome. Chest 1995; 108(5):1303–1314. 50. Meduri GU, Headley S, Kohler G, Stentz F, Tolley E, Umberger R, et al. Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS Plasma IL-1 beta and IL-6 levels are consistent and efficient predictors of outcome over time. Chest 1995; 107(4):1062–1073. 51. Fishman D, Faulds G, Jeffery R, Mohamed-Ali V, Yudkin JS, Humphries S, Woo P. The effect of novel polymorphisms in the interleukin-6 (IL-6) gene on IL-6 transcription and plasma IL-6 levels, and an association with systemic-onset juvenile chronic arthritis. J Clin Invest 1998; 102(7):1369–1376. 52. Humphries SE, Faulds G, Mohamed-Ali V, Luong L-A, Woo P, Hamsten A. A common polymorphism in the promoter region of the gene for interleukin-6 (G-176-C) is strongly protective against myocardial infarction at a young age. Eur Heart J 2001; 22:2243–2252. 53. Burzotta F, Iacoviello L, Di Castelnuovo A, Glieca F, Luciani N, Zamparelli R, et al. Relation of the -174 GC polymorphism of interleukin-6 to interleukin-6 plasma levels and to length of hospitalization after surgical coronary revascularization. Am J Cardiol 2001; 88(10):1125–1128. 54. Brull DJ, Montgomery HE, Sanders J, Dhamrait S, Luong L, Rumley A, et al. Interleukin-6 gene-174g>c and -572g>c promoter polymorphisms are strong predictors of plasma interleukin6 levels after coronary artery bypass surgery. Arterioscler Thromb Vasc Biol 2001; 21(9):1458– 1463. 55. Vickers MA, Green FR, Terry C, Mayosi BM, Julier C, Lathrop M, et al. Genotype at a promoter polymorphism of the interleukin-6 gene is associated with baseline levels of plasma C-reactive protein. Cardiovasc Res 2002; 53(4): 1029–1034. 56. Schluter B, aufhake C, Erren M, Schotte H, Kipp F, Rust S, et al. Effect of the interleukin-6 promoter polymorphism (-174 GC) on the incidence and outcome of sepsis. Crit Care Med 2002; 30(1):32–37. 57. Heesen M, Obertacke U, Schade FU, Bloemeke B, Majetschak M. The interleukin-6 G(-174)C polymorphism and the ex vivo interleukin-6 response to endotoxin in severely injured blunt trauma patients. Eur Cytokine Netw 2002; 13(1):72–77.

Genetic factors in acute lung injury

333

58. Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, Lopez-Rodriguez A, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344(10):699–709. 59. Hermans PW, Hibberd ML, Booy R, Daramola O, Hazelzet JA, de Groot R, et al. 4G5G promoter polymorphism in the plasminogen-activator-inhibitor-1 gene and outcome of meningococcal disease. Meningococcal Research Group. Lancet 1999; 354(9178):556–560. 60. Westendorp RG, Hottenga JJ, Slagboom PE. Variation in plasminogen-activator-inhibitor-1 gene and risk of meningococcal septic shock. Lancet 1999; 354(9178):561–563. 61. Menges T, Hermans PW, Little SG, Langefeld T, Boning O, Engel J, et al. Plasminogenactivator-inhibitor-1 4G5G promoter polymorphism and prognosis of severely injured patients. Lancet 2001; 357(9262): 1096–1097. 62. Scarabin P-Y, Bara L, Ricard S, Poirier O, Cambou JP, Arveiler D, et al. Genetic variation in the β-fibrinogen locus in relation to plasma fibrinogen concentrations and risk of myocardial infarction. The ECTIM study. Arteriosclerosis Thrombosis 1993; 13:886–891. 63. SE Humphries, M Cook, M DDubowitz, Y Stirling, TW Meade, Role of genetic variation at fibrinogen locus in determination of plasma fibrinogen concentrations. Lancet 1987; (i): 1452– 1455. 64. Green F, Hamsten A, Blomback M, Humphries S. The role of beta-fibrinogen genotype in determining plasma fibrinogen levels in survivors of myocardial infarction and healthy controls from Sweden. Thromb Haemost 1993; 70:915–920. 65. Wenz M, Steinau R, Gerlach H, Lange M, Kaczmarczyk G. Inhaled nitric oxide does not change transpulmonary angiotensin II formation in patients with acute respiratory distress syndrome. Chest 1997; 112(2):478–483. 66. Yamamoto T, Wang L, Shimakura K, Sanaka M, Koike Y, Mineshita S. Angiotensin II-induced pulmonary edema in a rabbit model. Jpn J Pharmacol 1997; 73(1):33–40. 67. Kiely DG, Cargill RI, Wheeldon NM, Coutie WJ, Lipworth BJ. Haemodynamic and endocrine effects of type 1 angiotensin II receptor blockade in patients with hypoxaemic cor pulmonale. Cardiovasc Res 1997; 33(1):201–208. 68. Marshall RP, McAnulty RJ, Laurent GJ. Angiotensin II is mitogenic for human lung fibroblasts via activation of the type 1 receptor. Am J Respir Crit Care Med 2000; 161 (6): 1999–2004. 69. Wang R, Zagariya A, Ibarra-Sunga O, Gidea C, Ang E, Deshmukh S, et al. Angiotensin II induces apoptosis in human and rat alveolaar epithelial cells. Am J Physiol 1999; 275(5 Pt 1):L885–L889. 70. Cambien F, Alhenc-Gelas F, Herbert B, Andre JL, Rakotovao R, Gonzales MF, et al. Familial resemblance of plasma angiotensin-converting enzyme level: the Nancy Study. Am J Hum Genet 1988; 43:774–780. 71. Rigat B, Hubert C, Alhenc-Gelas F, Cambien F, Corvol P, Soubrier F. An insertiondeletion polymorphism in the angiotensin-1-converting enzyme gene accounting for half the variance of serum enzyme levels. J Clin Invest 1990; 86:1343–1346. 72. Rigat B, Hubert C, Corvol P, Soubrier F. PCR detection of the insertion-deletion polymorphism of the human angiotensin converting enzyme gene (DCP1)(dipeptidyl carboxypeptidase 1). Nucleic Acids Res 1992; 20(6):1433. 73. Costerousse O, Allegrini J, Lopez M, Alhenc-Gelas F. Angiotensin-I converting enzyme in human circulating mononuclear cells: genetic polymorphism of expression in T-lymphocytes. Biochem J 1993; 290:33–40. 74. Keavney B, McKenzie CA, Connell JM, Julier C, Ratcliffe PJ, Sobel E, et al. Measured haplotype analysis of the angiotensin-I converting enzyme gene. Hum Mol Genet 1998; 7(11):1745–1751. 75. Marshall RP, Webb S, Bellingan GJ, Montgomery HE, Chaudhari B, McAnulty RJ, et al. The angiotensin converting enzyme insertion deletion polymorphism is associated with susceptibility and outcome in acute respiratory distress syndrome. Am J Respir Crit Care Med 2002; 166(5):646–650.

Acute respiratory distress syndrome

334

76. Lin Z, Pearson C, Chinchilli V, Pietschmann SM, Luo J, Pison U, et al. Polymorphisms of human SP-A, SP-B, and SP-D genes: association of SP-B Thr131Ile with ARDS. Clin Genet 2000; 58(3):181–191. 77. Dancey DR, Hayes J, Gomez M, Schouten D, Fish J, Peters W, et al. ARDS in patients with thermal injury. Intens Care Med 1999; 25(11):1231–1236. 78. Bratton SL, Davis RL. Acute lung injury in isolated traumatic brain injury. Neurosurgery 1997; 40(4):707–712. 79. Lee H, Hawker FH, Selby W, McWilliam DB, Herkes RG. Intensive care treatment of patients with bleeding esophageal varices: results, predictors of mortality, and predictors of the adult respiratory distress syndrome. Crit Care Med 1992; 20(11):1555–1563. 80. Kleeberger SR, Levitt RC, Zhang LY, et al. Linkage analysis of susceptibility to ozone-induced lung inflammation in inbred mice. Nat Genet 1997; 17(4):475–478. 81. Prows DR, Shertzer HG, Daly MJ, Sidman CL, Leikauf GD. Genetic analysis of ozone-induced acute lung injury in sensitive and resistant strains of mice. Nat Genet 1997; 17(4):471–474. 82. Leikauf GD, McDowell SA, Bachurski CJ, et al. Functional genomics of oxidant-induced lung injury. Adv Exp Med Biol 2001; 500:479–487. 83. Quasney MW, Waterer GW, Dahmer MK, et al. Intracellular adhesion molecule Gly241 Arg polymorphism has no impact on ARDS or septic shock in community-acquired pneumonia. Chest 2002; 121(3 suppl):85S–86S. 84. Arnalich F, Lopez-Maderuelo D, Codoceo R, et al. Interleukin-1 receptor antagonist gene polymorphism and mortality in patients with severe sepsis. Clin Exp Immunol 2002; 127(2):331–336. 85. Majetschack M, Flohe S, Obertacke U, et al. Relation of a TNF gene polymorphism to severe sepsis in trauma patients. Ann Surg 1999; 230(2):207–214. 86. Schroeder S, Reck M, Hoeft A, Stuber F. Analysis of two human leukocyte antigen-linked polymorphic heat shock protein 70 genes in patients with severe sepsis. Crit Care Med 1999; 27(7): 1265–1270.

14 Approach to the Genetic Epidemiology of Acute Lung Injury MICHELLE NG GONG and DAVID C.CHRISTIANI Harvard Medical School and Massachusetts General Hospital Boston, Massachusetts, U.S.A.

Since the initial description of ARDS in 1967(1), much research has been focused on defining the pathogenesis, clinical presentation, course, and outcome of the syndrome. Initially, early studies investigated the role of complement and endotoxin in lung injury (2). In the past decade, research has focused more on the role of pro-inflammatory and anti-inflammatory response in the pathogenesis and course of acute lung injury and acute respiratory distress syndrome (ALI/ARDS) (3). At the same time, clinical studies were conducted evaluating variables that influence the development and outcome of ALI/ARDS. While many of the animal studies have been consistent, many of the in vivo studies have been heterogeneous. Consequently, efforts to find clinical factors or biomarkers to predict, diagnose, or prognosticate outcomes in ARDS have often been disappointing. Our current understanding of why some patients develop and die from ARDS while others do not is incomplete. For example, while major risk factors for ARDS have been identified, the majority of patients with these risk factors do not develop ARDS. Only 3.8% of patients with documented bacteremia, 41.2% of patients with sepsis syndrome, and 11.9% of intensive care unit patients with pneumonia developed ARDS (4, 5). While smoking has been found to be associated with an increased risk of ARDS in one study, more than half of the ARDS patients were never or ex-smokers (6). Other clinical factors in ARDS have mixed results in the literature. While some studies report an increased risk of developing or dying from ARDS with older age, comorbid conditions such as alcohol-related diseases, and severity of illness, other studies did not find the same (5, 7–10). Likewise, the search for biomarkers in ARDS has been equally, if not more, disappointing. Some studies have found increased plasma levels of TNF-α to correlate with the development of ARDS and the severity and mortality in ALI/ARDS (11–13). Other studies from different institutions did not detect the same correlation (14, 15). Variable timing, sample type, and method of measurement can explain some of the conflicting results on cytokines in ARDS (16). Until recently, the role of genetic variability in the development and course of ALI has not been considered seriously. However, the recent explosion in knowledge about the genetic control and regulation of the innate immune defense and inflammatory response has raised the question of whether

Acute respiratory distress syndrome

336

the multiple alleles of genes that encode for cytokines and other mediators of inflammation can result in phenotypic differences in

Figure 1 Role of genetic variability in the development of ALI/ARDS. host inflammatory response. These differences may then account for the heterogeneity in individual susceptibility to, and prognosis in, ARDS (3) (Fig. 1). In this chapter we discuss the evidence for genetic determinants in acute lung injury and the issues related to the genetic epidemiology of complex disorders. Some strategies for deciphering the genetic factors of acute lung injury are discussed. This is not meant to be a comprehensive review of techniques for investigating the genetic make-up of complex diseases. That has been discussed in several recent reviews(17–20). We will instead focus on how different approaches in genetic epidemiology may or may not be amendable to the study of acute lung injury.

I. Evidence for Genetic Determinants in ALI What evidence is there that acute lung injury might be a heritable disease? Heritability is typically addressed in genetic epidemiology with family studies. Studies of twins, adoptees, or related individuals compared to the general population are often used to distinguish the genetic from the environmental contributions to a disease (21). Alternatively, if a large family or pedigree is available, then segregation analysis can be used whereby the pattern of affected and unaffected individuals in the family is fitted to

Approach to the genetic epidemiology of acute lung injury

337

various models to determine the extent to which the condition abides by classic Mendelian inheritance (17, 19). These classical approaches are not as feasible in ARDS. ALI/ARDS was first described in 1967. However, workable criteria for the consistent diagnosis of ARDS were not available until 1988 (22). Our current definition of ARDS and ALI was proposed in 1994 (23). Thus it is likely that ARDS was underdiagnosed in the past. Additionally, the high mortality and older age of onset of ALI/ARDS limits the availability of affected parents or grandparents for study (5, 24). It is not surprising that there are no known family aggregates of ARDS. Multigenerational pedigrees that are needed in segregation analysis, and in some of the study designs discussed later, are not yet possible in ALI. Nevertheless, there are intriguing studies that point to genetic determinants in ARDS. There have been multiple reports of recurrent ARDS in certain individuals (25–29). One report describes six patients with multiple episodes of biopsy-proven diffuse alveolar damage (30). While these patients may have been merely very unfortunate, the discovery of a group of patients prone to repeated lung injury raises the possibility of an idiosyncratic, genetic predisposition to ARDS. Given that mortality in ARDS is high, it is not surprising that these cases of recurrent ARDS are rare. However, with mortality in ARDS improving (24, 31, 32), genetically susceptible patients may now be more likely to survive their first bout of lung injury to develop another episode. While there are no family studies on ARDS, there are intriguing twin and adoption studies on premature deaths. In an epidemiological study of 218 pairs of Danish twins, a monozygotic twin whose co-twin died prematurely before the age of 60 had a significantly increased risk of dying prematurely as well (33). These associations were not found for same sex dizygotic twins who were raised in the same environment. Their life expectancies relative to each other were no different than that between two unrelated random individuals. In another study involving 960 families with adopted children, the premature death of a biological parent before the age of 50 from natural causes correlated with a doubling of the risk of premature death in the adopted individual (34). The relative risk (RR) of dying prematurely from infections in an adopted individual is 5.81 (95% CI 2.47–13.7) if one biological parent died before the age of 50 from infection. In contrast, if the adopted parent were to die prematurely of infection, the RR of dying from infection was 0.73 (95% CI 0.10–5.36) in the adoptee. This is in contrast to the risk of dying prematurely from cancer, where the risk to the adoptee is 1.19 (95% CI 0.16–8.99) versus 5.16 (95% CI 1.20–22.2) if the biological parent versus the adopted parent dies of cancer. The data indicate that premature death from infection has a strong genetic component, while premature death from cancer is influenced more by the family environment than the genetic background of the individual. Genetic contribution to susceptibility to severe infection has important implications for ALI. Sepsis is the leading cause of ALI/ARDS (5) and is associated with worse mortality than ALI secondary to other etiologies (9). Bacterial infection is frequently found in ARDS patients, with one autopsy study indicating a prevalence of 98% (35). Most fatal cases of ARDS die from refractory infection and sepsis, not from respiratory failure (36–38). Recently, variable genetic influences on outcomes have been investigated in sepsis. The most intriguing reports have involved two specific genetic polymorphisms associated

Acute respiratory distress syndrome

338

with increased production of tumor necrosis factor-α (TNF-α) and increased mortality in sepsis. The first is a −308 G to A transition polymorphism in the promoter region of TNF-α gene, which appears to affect its expression. Allele 2 (TNF2) of this biallelic gene has a frequency of 16% and is associated with increased transcription of TNF-α in vitro (39–41). In a large case-control study of Gambian children, homozygosity for TNF2 increased the susceptibility to cerebral malaria as well as the risk of death or other sequelae from cerebral malaria as compared to heterozygotes or homozygotes for TNF1 (40, 42). One study involving 60 patients with sepsis found no correlation between the TNF polymorphism and mortality (43). Another study reported significantly greater frequency of the TNF2 allele in 89 patients with septic shock compared with healthy blood donors and a greater frequency of TNF2 in those who died of sepsis compared to survivors even after adjusting for age and probability of dying based on SAPS-II (44). The other polymorphism known to influence the expression of TNF-α is a NcoI restriction fragment length polymorphism (RFLP) in intron 1 of the TNF-β gene. This polymorphism is biallelic with frequencies of 35% for TNFB1 allele and 65% for TNFB2 allele. Homozygotes for TNFB2 secrete higher amounts of TNF-α in vitro and in vivo (45–47). Septic patients who were homozygous for TNFB2 have greater severity of multiple organ dysfunction including ARDS and increased mortality than heterozygotes and homozygotes for TNFB1 (43, 47). In another study of 110 trauma patients, TNFB2/B2 patients had significantly increased serum TNF-α and were at significantly increased risk of developing severe sepsis 5 days after hospitalization (48). Lastly, a recent case-control study on community-acquired pneumonia found an increased ageadjusted odds ratio of 3.64 (95% CI 1.28–10.66) for patients homozygous for the TNFB2 allele to develop septic shock with their pneumonia compared to carriers of the TNFB1 allele (49). The discovery of polymorphisms in the TNF-α and TNF-β gene that affects the expression of and plasma levels of TNF-α offers yet another reason for conflicting studies on the role of TNF-α in ALI(3). Such findings have led some to question as to whether genetic variation may explain partially the heterogeneous outcomes in therapeutic trials in sepsis (50, 51). As a result of the studies on sepsis, there have been some recent preliminary investigations into the genetic susceptibility in ALI. While promising, these initial studies also demonstrate some of the difficulties involved in designing genetic studies on complex diseases like ALI/ARDS. Polymorphisms in the SP-B gene have been found to be associated with ARDS in some recent studies. In one study, the frequency of an insertion/deletion variant polymorphism in intron 4 was 46.6% among 15 ARDS patients in contrast to 4.3% in the control group of normal blood donors (p<0.05) (52). However, the study was limited in that the controls were not at risk for ARDS and clinical predictors for lung injury were not measured. Another study from the same group found an association between the -1580C/T missense mutation in exon 4 of the SP-B gene and the development of ARDS. This association was due entirely to the increased frequency of the allele in a group of 23 patients with what the authors termed “idiopathic ARDS,” which consisted of mostly lung injury from direct pulmonary insults such as pneumonia (53). While intriguing, this study should be interpreted cautiously as the genotype frequencies in the control group deviated from that predicted by the Hardy-Weinberg equilibrium. The Hardy-Weinberg equilibrium is an important principle in population

Approach to the genetic epidemiology of acute lung injury

339

genetics, which states that one should be able to calculate genotype frequencies from the frequency of homozygous and heterozygous individuals in the population (54). It has been suggested that deviations from the Hardy-Weinberg equilibrium among controls in molecular epidemiology studies should prompt a repeat analysis and investigation for potential complications such as genotyping error and population stratification (55). Recently, our group found an increased association between the variant SP-B polymorphism in intron 4 and the development of ARDS in women at risk for ARDS secondary to sepsis, aspiration, pneumonia, trauma, or massive transfusion (56). The genotype frequencies conformed to the Hardy-Weinberg equilibrium, and demographic and clinical factors such as race, age, comorbid diseases, severity of illness, and alcohol history were controlled for. No such effect was found for the men, indicating possible gender modification of the risk conferred by the variant polymorphism of the SP-B gene. Women with this polymorphism were also more likely to be admitted to the intensive care unit with direct pulmonary injuries such as pneumonia or aspiration as opposed to other risk factors for ARDS. Recently, homozygosity for the deletion polymorphism in the angiotensin-converting enzyme (ACE) gene, which is associated with higher ACE levels and activity, was found in an increased frequency among patients with ARDS as compared to three control groups: patients with non-ARDS respiratory failure, patients undergoing coronary artery bypass surgery, and healthy men in the United Kingdom (57). In addition, the C allele of the -174GC polymorphism in the IL-6 gene, which has been associated with lower IL-6 plasma concentration, was found in a lower frequency in nonsurvivors with and without ARDS. However, the study has several limitations. It is not clear what the biological role of ACE is in ALI. Most studies found ACE levels or activity to be low in ARDS, making it difficult to understand why the DD genotype (associated with high serum ACE levels and activity) would be associated with an increased risk for ALI (58–60). In addition, it is not clear whether the controls were at risk for lung injury initially. Clinical contributors to ARDS and mortality in ARDS, such as severity of illness, alcohol history, and age, were also not controlled for in the study.

II. Acute Lung Injury as a Complex Genetic Disorder While there are likely to be genetic determinants in the development and evolution of acute lung injury, it is also likely that the path to its clarification will not be straightforward since ALI is a complex disorder with complex genetic determinants. Elements of diseases with complex genetic traits are detailed in Table 1. It is now universally accepted that the development and evolution of ALI involves the activation of the inflammatory cascade. However, the innate immune and inflammatory response, like many other mechanisms in physiology, involves a number of integrated biochemical and physiologic systems that respond to and are modulated by environmental stimuli. To ensure stability and adaptability, the inflammatory response has both pro- and antiinflammatory pathways with built-in redundancies, bio-feedback loops for modulation, and counterregulatory mechanisms. While this system is under genetic regulation, it is also likely to be controlled by multiple genes (genetic heterogeneity) with interaction between genes (gene-gene interaction) and between genes and external stimuli such as

Acute respiratory distress syndrome

340

infection, trauma, or other injuries (gene-environment interaction). The redundancies built into the system could create a threshold effect whereby the function of

Table 1 Elements of a Complex Disease in Genetics Genetic heterogeneity

Mutations in a number of different genes can result in the same phenotype or disease.

Environmental phenotypes

The same disease or phenotype can have a genetic component and a purely environmental form with no genetic influence.

“Threshold” inheritance

The number of different mutations or genetic defects must reach a certain threshold before the disease phenotype will be manifested.

Multilocus effects

Multiple genes influence the development of disease.

Locus heterogeneity

Mutations in any number of genes or loci can result in disease independently of each other.

Incomplete penetrance

Some individuals who inherit the susceptibility gene will develop the disease, while others do not, so that the genotype affects not the fate of the individual but the probability of disease.

Incomplete phenocopy

Individuals who did not have the susceptible genotype will develop the disease because of environmental or random causes.

Epitaxis or genegene interaction

Development of the disease depends on the presence of and interaction between multiple genetic mutations.

Gene-environment interaction

Development of disease depends on the inheritance of the susceptible gene and the presence of a required environmental stimulus.

Time-dependent expression

The expression of certain genes may affect phenotype differently depending on the age or development stage at the time.

several genes needs to be affected before lung injury will be manifested. Because of these multiple interactions, any susceptible gene in acute lung injury will have incomplete penetrance (i.e., not all individuals with the gene will develop ALI). Epidemiologically, penetrance can be translated as relative risk where a high penetrant gene corresponds to high relative risk of disease for the individual while a low penetrant gene confers a lower relative risk of the disease. The role of the environment is particularly critical in determining the genetic determinants in a complex disorder like ALI. ARDS is generally thought to be a result of some injury. Although occasionally in cases of “idiopathic ARDS” the exact injury cannot be identified, the inciting risk factor leading to ARDS is usually easily found. In two prospective studies on the development of ARDS, about 78% of the cases of ARDS developed after known conditions such as trauma, aspiration, sepsis, massive transfusion, drug overdose, near-drowning, or pneumonia (4, 5). In one of these studies (5), 28 of the 48 ARDS patients who did not have one of these predefined conditions had “near risks” where clinical risk of sepsis or aspiration was suspected but the patient did not meet the predefined criteria. Only 5 did not have a clearly recognized acute condition leading to ARDS. Thus, a predisposing injury is vital to the development of ALI. In addition, the

Approach to the genetic epidemiology of acute lung injury

341

type of injury affects the incidence of ARDS, the cytokine profile, and the mortality. Forty-three percent of patients with sepsis develop ARDS compared with 26% of patients with aspiration in one study, and sepsis as an etiology for ALI is associated with a higher mortality compared with other etiologies of lung injury (5). Soluble ICAM-1 and Eselectin are significantly higher in patients with sepsis at risk for ARDS than trauma patients at risk for ARDS (61). This difference persisted even after the development of ARDS. Thus, just as it is important to account for tobacco use in genetic studies of lung cancer and chronic obstructive pulmonary disease (COPD), it will be important to account for the injury that predisposes one to lung injury. Because of incomplete penetrance and interactions with other factors such as age, other genes, and the environment, susceptibility genes for complex diseases like acute lung injury will not determine the development and outcome of the disease. Rather, it will affect the probability of the disease. Epidemiology has traditionally been the field of study concerned with the probability of disease. The field of molecular epidemiology is relatively new and has been variably defined. At its core, it combines the use of molecular biological techniques such as modern genetics with epidemiology to identify and characterize disease on a population level in the context of environmental factors. In sorting out the genetic basis for lung injury, the discipline and tools of classical epidemiology will be important in accounting for multiple interactions and environmental influences. In the following discus-sion on approaches to determining genetic factors important in ALI, epidemiological issues of study design and analysis will be discussed.

III. Strategies in Genetic Studies of Acute Lung Injury A. Techniques for Gene Localization and Selection Gene discovery is often done with animal studies involving experimental crosses of mice and rats. Animal studies offer ideal conditions where there are well-defined pedigrees with no genetic heterogeneity and strict control of environmental factors. Animal studies are pivotal in clarifying the underlying molecular and genetic mechanisms, and their importance should not be understated. However, the results are not always applicable to humans. For the purpose of this chapter, we will concentrate on approaches for the genetic analysis of lung injury in humans. One important technique for localizing disease susceptibility genes involves total genome-wide scans of families with the disease (19, 62). Total genomic scanning typically consists of analyzing the highly variable or polymorphic regions (microsatellite markers) of the genome at regularly spaced intervals with no distinction made about the biological plausibility of the gene in the disease. Then multiple comparisons of these markers in the affected families are performed to identify regions of the genome that segregated with the disease. Total genomic scan is appealing since it is not confined to genes that have been previously studied and it can help localize the susceptibility gene in the genome. Rather than search for novel genes across the genome, an alternative approach involves focusing on candidate genes whose products have been well characterized as biologically important in the pathogenesis, progression, or manifestation of the disease

Acute respiratory distress syndrome

342

(19, 62). Unlike total genomic scanning, the candidate gene approach is hypothesis driven and founded upon current knowledge of the disease process. It is easy to perform. Thus there has been widespread use of this approach in the literature. However, with widespread use, there have been accusations of misuse as well (63). As the candidate gene approach is hypothesis driven, the strength of the design rest on the strength of the hypothesis supporting the choice of candidates (Fig. 2). Given the complexity of human physiology and the successes of genetic analysis in identifying polymorphisms, a large number of potential candidates exist. The strongest candidates for investigation are those genes that have been linked to ALI in linkage studies, association studies, or in animal models of the disease (55). If such studies are not yet available, then the biological plausibility of the candidate gene in the pathogenesis of lung injury is important. This requirement is also important if the study

Figure 2 Criteria for strong candidate genes. supporting the first criteria was based upon association studies. Additionally, the candidate polymorphisms should have functional consequences. That is, the

Approach to the genetic epidemiology of acute lung injury

343

polymorphism should result in variable levels, function, or effectiveness of the gene product. Lastly, the results should then be reconciled with the original hypothesis for consistency. B. Study Designs Once genetic markers have been selected by the candidate gene approach or defined by total genomic scan, they must then be analyzed in relation to the phenotype (disease) to determine their importance in disease development or progression. There has been much work toward the appropriate design of studies to determine genetic determinants in disease. Issues such as mode of inheritance, gene penetrance and frequency, disease prevalence, and environmental interaction help determine the optimal study design for a disorder (Table 2). Total genomic scanning is most often used in family-based linkage studies such as linkage analysis or allele-sharing methods. Recently, with the success of the Human Genome Project and the Single Nucleotide Polymorphism (SNP) Consortium, there is growing interest in genome-wide case-control studies (20, 64, 65). The candidate gene approach can be used with any of the following study designs, but it is most commonly used in case-control studies. Linkage Studies Total genomic scanning is most commonly analyzed in relation to the disease phenotype with family-based linkage studies. Linkage occurs when the genetic locus being studied and the disease gene tend to be segregated or co-transmitted together from generation to generation because of the close proximity of the locus to the disease gene. Linkage studies focus only on cases of the disease in pedigrees and seek to ascertain whether there is statistically significant inheritance of the genetic locus of interest among affected family members presumably because of linkage to the true disease alleles. This approach does not actually localize the disease gene. Once the candidate region containing the disease allele and the linked genetic markers have been identified by total genomic scanning and linkage studies, then it can be localized with fine mapping and positional cloning or physical mapping (19). There are two major categories of linkage studies: linkage analysis and allele-sharing methods.

Table 2 Comparison of Study Designs in Genetic Epidemiology and Disease Characteristics Best Suited for Each Design Study design Disease Characteristics

Linkage studies

Case-control

Mode of inheritance

Mendelian

Non-Mendelian

Gene penetrance

High

Low

Disease/disorder prevalence

Rare

Common

Gene frequency

Rare

Common

Acute respiratory distress syndrome

Environmental interaction

Low to none

344

Moderate to high

Linkage analysis is a staple of classical genetics. This approach attempts to find linkage between the genetic marker of interest and the gene responsible for the disease trait, thereby localizing the disease gene to an approximate candidate region on the chromosome that contains both the disease gene and the genetic marker. Linkage is determined by fitting a predetermined inheritance model to a pedigree of affected and unaffected individuals to determine the likelihood of a significant linkage between the locus of interest and the disease-causing gene relative to the likelihood of no linkage. If this ratio (LOD score) exceeds a threshold of 3, then there is significant linkage between the marker and a major disease-causing locus (20, 66). Linkage analysis is extremely powerful when the correct model is chosen. It is the method of choice for Mendelian traits and has a proven track record of success in detecting high penetrant (i.e., high relative risk), rare disease genes such as the BRCA1 (67). However, the wrong choice of model can greatly diminish the power of the study. For ALI/ARDS, a complex disorder with strong environmental determinants likely interacting with low penetrant genes, the optimal model is difficult to determine (68). Thus, the utility of linkage analysis is limited in complex disorders like acute lung injury. Additionally, with the high mortality and late age of onset of ALI/ ARDS, well-defined pedigrees of affected families are not available presently for such analysis. Lastly, the power of linkage techniques can be diminished when the disease susceptible gene is common in the population (17). Linkage analysis focuses on heterozygous individuals and the transmission of the disease allele to affected offspring. A high frequency of the disease allele results in multiple independent copies of the gene in the study family, making it difficult to track transmission (17). This was a research challenge in the study of late-onset Alzheimer’s disease. The initial attempts with linkage analysis revealed a low LOD score for chromosome 19q that was suggestive of but not definitive for the location of the disease allele (69) Subsequently, the apolipoprotein E type 4 allele, which is located on chromosome 19, was implicated in an association study in late-onset Alzheimer’s disease (70, 71). The allele had a relatively high frequency of 16% in the population, which reduced the power of the analysis of detect linkage (17). Thus this strategy is better suited for rare alleles associated with rare diseases or unusual variants of common diseases such as early-onset disease. Alternatively, a model-free or nonparametric strategy, also known as the allele-sharing method, can be used to localize a susceptibility gene (20). The allele-sharing method does not assume a model of inheritance but determines whether diseased members of a family inherit the locus of interest more often than would be expected by chance (17, 20). It can be performed on affected parents and offspring or on affected siblings. If there is significant segregation of the marker among affected family members, then it is presumed that the affected members of the pedigree inherited a common chromosomal region containing the marker and the disease allele from a common ancestor. Allele sharing is better than linkage analysis at localizing susceptibility genes for complex diseases as it is nonparametric and makes no assumption about type of inheritance and degree of penetrance. It is applicable even when the locus of interest has a high prevalence in the population. However, it is still relatively weak at detecting low penetrance genes and it tends to require large number of families with multiple affected

Approach to the genetic epidemiology of acute lung injury

345

siblings (65). Gene-gene and gene-environment interaction could theoretically be examined in this design by stratification, but will further reduce the power because of reduced sample sizes in each stratum (72). The allele-sharing method can be done with affected siblings only, which is important in diseases with late age of onset like ALI/ARDS. However, the use of siblings rather than parents reduces the ability to detect linkage and generally requires larger sample sizes to detect even modest effects (19). That is not to say that linkage studies have no role in the study of ALI in the future. There may be special circumstances in which lung injury may behave in a more Mendelian fashion. By restricting the cases to severe lung injury, family clusters of repeated episodes of ARDS, or early-onset ARDS, one might make the phenotype more homogeneous and, therefore, increase the power of such studies. The successful localization of the BRCA1 gene in breast cancer by linkage analysis was accomplished by restricting the study to family clusters of early-onset breast cancer (67). However, a better national registry of ARDS patients might be needed to determine the characteristics and prevalence of these special clusters of acute lung injury before such studies can be undertaken. In addition, any result will need to be reassessed in the general population as the studied family may not be representative of the general population or the scope of disease. BRCA1 is one example. While it confers a high individual risk (73.5% of carriers develop breast cancer by age 80 years), this gene is rare in the general population and accounts for only 3.0% of breast cancer in the United States (73). Case-Control Studies Lastly, linkage between a gene of interest and the disease can be determined with association studies utilizing the case-control design to determine if the gene of interest occurs at a significantly higher frequency among the cases with disease than among the controls (20). Unlike linkage studies, case-control studies require the delineation of a control group by definition and focus on relative genotype frequency in the general population rather than on inheritance patterns. One of the most important advantages of case-control studies in complex disorders like acute lung injury is the power of the design. Association studies are the most sensitive and powerful of all of the study designs described thus far in detecting common, low penetrant, susceptibility genes in complex diseases (66, 74, 75). For example, the protective association between the alanine allele of the PPARγ Pro12Ala polymorphism and adult-onset diabetes was found in a casecontrol study involving 3000 individuals (76). The authors estimated that it would have required 3 million sibling pairs to discover this by linkage studies. Because of the high prevalence of the polymorphism, the population attributable risk was high at 25%. In other words, 25% of the diabetes cases in the population could be explained in part by the presence of this allele. This is one example of how a low penetrant allele with a large population impact (i.e., population attributable risk) could have been detected only by population-based case control study. In addition, the case-control design in genetic epidemiology has other important benefits. The case-control design is well suited to the study of genetic markers of disease. Genes are stable indicators of disease susceptibility, as they do not change with time or circumstances, unlike biological markers such as plasma or bronchoalveolar lavage (BAI) cytokines (77). As such, the use of genetic markers as the exposure eliminates the recall

Acute respiratory distress syndrome

346

bias that often plagues case-control studies. Also, the case-control design can be used to analyze for gene-gene and gene-environment interaction (77). It is amenable to multivariate modeling, which allows for the simultaneous control of multiple variables, considered essential for the study of multicausal diseases (78). Case-control analyses are not affected by disease alleles that are present in high frequency in the population, which will likely pose less individual risk (i.e., relative risk) but greater population-attributable risk. On the contrary, high-frequency alleles are most amenable to population studies using case-control design. Unlike linkage studies, case-control methods can measure directly the relative risk associated with specific genotypes and measure the degree to which the disease in the general population can be attributed to the genotype (74). Because of the power and versatility of association studies, many believe that the future of deciphering the genetics of complex diseases will involve case-control studies (55, 65, 76). Not surprisingly, this design has become very popular in investigating the genetic contribution to disease. However, with the heavy use of this design comes some misuse as well. Consequently, association studies have been criticized. The most common and troubling criticisms are inconsistency and lack of reproducibility. While some studies show a positive association between a candidate gene and the disease, other studies do not show the same association. This heterogeneity is due to a number of factors. The epidemiological quality of genetic studies that are published is quite variable. In one review of 40 genetic association studies published in 1995 that studied 10 or more patients, 35 (87.5%) of the studies did not pass at least one of 7 basic epidemiological standard including reproducibility of the genetic assay, objectivity with a blinded observer, clear delineation of cases and controls, adequacy of the spectrum of disease among the cases, adequacy of the comparison group, and clear quantitative summary of the results (79). Other factors include the lack of power in some studies (Type II errors) and the lack of control for confounders like population differences or gene-environment interaction (72, 78). Failure to examine the role of environmental exposure can lead to decreased sensitivity in detecting an association between the gene of interest and the disease (77). While troublesome to classical geneticist, the need to confirm studies is common in epidemiology. Any population study needs to be validated for different populations and in larger studies (68). However, as is true in any case-control design in epidemiology, the strength of the study depends entirely on the proper selection of cases and controls and on the appropriate accounting of potential confounders, power, and type I errors (80). The following section will focus on these issues as they pertain to the special circumstances of genetic casecontrol studies in acute lung injury. C. Choice of Cases and Controls As with any case-control study, the choice of cases and controls is pivotal to the design, strength, validity, and generalizability of the study. In contrast, linkage studies are focused on affected family members and need no controls. Choice of phenotype in ALI is problematic, as there is no definitive diagnostic test for lung injury. This problem applies not only to case-control studies. Difficulty with defining cases is especially limiting in linkage studies (19). There is, as yet, no diagnostic test with known sensitivity and specificity for ALI/ARDS. The American-European

Approach to the genetic epidemiology of acute lung injury

347

Consensus criteria serve as an uniformly accepted guideline for defining lung injury, but certain criteria, specifically the radiological criteria, are not always clear and are subject to interobserver variability (81). In addition, the specification of PaO2/FiO2 at 300 mmHg for ALI and 200 mmHg for ARDS was meant to create a cutoff criteria by which the severity of hypoxemic respiratory can be assessed. However, PaO2/FiO2, like the measurement of any continuous physiological parameter, represents a continuum of hypoxemic respiratory failure. Thus, there is likely to be some random misclassification of cases and controls, which will bias the study toward the null hypothesis (80). Nevertheless, the American-European Consensus definition is still the most universally accepted criteria for ALI/ARDS, and random misclassification is inherent in any large epidemiological studies. Therefore, care must be taken to assess the degree to which the cases and controls adhere to the ALI/ARDS criteria, and larger sample size or more powerful studies, such as association studies, might be needed to detect an association. One potential option in ALI is to use an intermediate phenotype, such as septic shock or severe pneumonia, to increase the power to detect a genetic association. This option may be useful in linkage studies as well. Intermediate phenotypes are “closer” to the gene product. There may be less environmental and gene interaction and, therefore, a stronger association with the gene of interest (62) (Fig. 3). Alternatively, using a continuous phenotype such as the actual PaO2/FiO2 or the Lung Injury Score in ALI would be less arbitrary than relying on a dichotomous cut-off while increasing the statistical power (55). The use of intermediate phenotypes has been extensively used in other complex respiratory diseases such as asthma (82). Another option is to limit the case definition to ALI from specific etiologies to decrease the heterogeneity of the cases. Alternatively, the cases could be restricted to phenotypes that might have a stronger genetic influence such as very severe ARDS, early-onset ARDS, or patients with repeated episodes of lung injury (17). The choice of controls in case-control studies is equally important, although often neglected. In case-control studies, controls are not intended to represent the entire nondiseased population but are intended to represent the population that is at risk for the disease and would have been included as a case if they did develop the disease (83). A poor choice of control may result in hidden confounding. There are two common problems in the choice of controls. One is the selection of controls who are not at risk for the disease, making comparisons with the cases difficult. In one review, 30% of the genetic epidemiology studies published in 1995 did not adequately delineate the criteria by which the controls were selected, and in 13.5% of the studies the controls were improperly chosen (79). The controls in many of the studies in sepsis and ALI/ARDS were healthy blood donors or hospitalized patients without a clear prior injury placing them at risk for ALI. Any association between a candidate gene and ALI/ARDS may be due to the greater likelihood of being admitted to the hospital with a risk factor for ALI/ARDS rather than to the development of ALI/ARDS. While this information may still be important, the context of the association and the population studied must be accounted for in the interpretation. The second common problem in the choice of controls in genetic epidemiology studies is the selection of controls from a source population different than the one from which the cases arise. This can lead to the confounder of population stratification, which is

Acute respiratory distress syndrome

348

discussed further below. Basically, if the controls and the cases were inadvertently selected from different populations (e.g., different racial or ethnic history), then any

Figure 3 Role of genetic and environmental influence on intermediate phenotype and disease manifestation. The width of the arrows reflects the effect from the gene. Each plane represents environmental stimuli that may augment or attenuate the direct effect from the gene. The “closer” the phenotype is to the gene product, the greater the effect of any mutation in that gene on the phenotype. Therefore, the power to

Approach to the genetic epidemiology of acute lung injury

349

detect a susceptibility gene in the development of a disease will be improved if one were to examine the intermediate or lower level phenotype. GP, Gene product; LLP, lower level phenotype; IP, intermediate phenotype; HLP, high level phenotype. (Adapted from Ref. 62.) association found may be secondary to the difference in the population and not to their disease state. For this reason, some have advocated using related family members as controls rather than unrelated controls to decrease the risk of confounding from population stratification and genetic heterogeneity, since the cases and the controls are both from the same families (17). In ALI/ ARDS, family controls will likely consist of mostly siblings as the late age of onset of lung injury and the high mortality makes it unlikely that affected parents will be available for study. However, because of the variable risk of ARDS with age, family-based case-control studies may require prospective longitudinal follow-up of the siblings (77). However, in spite of the potential for hidden confounding, unrelated controls will continue to be used commonly in future studies of complex diseases since unrelated controls are easier to collect than family members, allowing for larger sample size. In addition, unrelated controls, unlike family members, can be selected to match the cases in their exposures or risks for disease (such as sepsis in ALI). D. Confounders and Errors in Genetic Case-Control Studies As with any kind of epidemiological study, one must take precautions in evaluating and adjusting for potential confounders. A confounder in genetic epidemiology studies is a factor that is associated with the gene of interest and independently affects the risk of disease. Lack of adequate adjustment for this confounder might lead to the erroneous conclusion that a candidate gene is or is not responsible for the disease when in fact the association was due to an uncontrolled confounder. One potential confounder unique to genetic studies is that of population stratification secondary to differing racial or ethnic mixture among the cases and the controls that is not always obvious (17, 20, 77). In the example of the variant polymorphism in intron 4 of the SP-B gene, the racial mixture of the study group is important. The variant polymorphism in the SP-B gene has a higher frequency in African Americans and Nigerians than among Caucasians(84). To analyze the association between the variant SP-B and ARDS without accounting for race may have biased the estimate toward a null if there were more African Americans in the control group than in the cases or toward finding an association if there were more cases who were African American. While distinguishing between African Americans and Caucasians appears straightforward, determining ethnic and genetic homogeneity among cases and controls is not always obvious.

Acute respiratory distress syndrome

350

Recently, some have argued that for most genetic epidemiology studies, population stratification is not a significant confounder if there is no substantial variation in allele frequency or disease prevalence among different ethnic groups (85). Nevertheless, many are still concerned about population stratification, and care should be taken to make sure that the cases and the controls arise from genetically homogeneous populations. Racial history should be carefully recorded and either matched for or controlled for in the analysis (77). Some have suggested studying an isolated population such as Finland or the Old Order Amish where there is either limited immigration or higher frequency of inbreeding resulting in less genetic heterogeneity(19, 68, 86). Alternatively, others have advocated using internal controls such as family members and analytical tools such as transmission disequilibrium tests (TDT) in a genetic case-control study (17). Lastly, some have suggested comparing multiple genetic markers between the cases and controls to determine and control for genetic stratification (87). Another source of potential confounding is related to linkage disequilibrium between the candidate gene and the true disease gene. If the candidate gene is the true disease allele, then the association between the candidate and the disease will be present consistently regardless of the population studied. However, a positive association between the candidate gene and disease does not imply necessarily that the gene locus is responsible for the disease. Rather, the candidate polymorphism could be located on the same chromosome in close proximity to the actual susceptibility allele (linkage disequilibrium) (17, 20). Linkage disequilibrium can have two effects on analysis. If the linkage between the candidate polymorphism and the true disease allele is not perfect, then association between the candidate gene and the disease phenotype will be weakened toward the null (77). In addition, a positive association between the genetic locus and the disease phenotype in one study is sometimes not reproducible in another study using a different population. The linkage between the candidate marker and the true disease allele may exist in certain young or isolated populations but not in another, older population where there are more racial admixtures and more recombination events (17). This is one reason why genetic association studies should be repeated in different populations. Because of linkage disequilibrium, the finding of an association between a candidate polymorphism and the disease does not implicate that genetic locus as the true disease allele. However, if the candidate gene was chosen with sound scientific rationale, a positive association between the candidate polymorphism and the disease may still localize that candidate gene as important in the development of the disease even if the polymorphism studied is not the direct cause of the disease (74). Type I and Type II errors are also important in genetic case-control studies. A Type I error is the likelihood of a false-positive finding (88). Although a p-value or a Type I error rate of 5% is generally accepted, one will be likely to find an association by chance alone if multiple comparisons of different genetic loci to the development of disease are performed. The problem is particularly important when doing genome-wide association studies where millions of SNPs will be compared. Although it is not entirely clear what the best strategy is, most advocate correcting for multiple comparisons with some variation of statistical methods like the Bonferroni correction, which requires a lower pvalue (e.g., 0.01) for statistical significance (65, 89). The likelihood of a cause-and-effect relationship underlying any genetic association will depend on the reproducibility of

Approach to the genetic epidemiology of acute lung injury

351

well-designed studies in different populations and in the strength of the biological rationale behind the selection of that gene for analysis (77). Type II error refers to the statistical power of the study. Power to detect an association depends upon the size of the effect, the population frequency of the genotype, and the sensitivity of the analysis deployed. While power is always a concern in any epidemiological study, it is especially important in molecular epidemiology (80). Some of the negative studies in molecular epidemiology are due to the lack of adequate power. The association of PPARγ Pro12Ala polymorphism and adult-onset diabetes is one such example. Although two studies indicated a significantly increased odds of Type II diabetes with the more common proline allele (90, 91), other case-control studies did not (92, 93). The controversy was resolved after an analysis of 3000 individuals revealed a significant risk of 1.25 (p=0.002) associated with the proline allele (76).

IV. Conclusion Molecular epidemiological studies represent an exciting and novel approach to the study of acute lung injury. When considering complex genetic traits such as intelligence or social adaptability, the issue of “nature versus nurture” is often raised. In acute lung injury the answer is likely to be both. It is not likely that a single susceptibility gene will be disease producing. Rather, susceptibility genes will be an additional but important risk factor for determining the ultimate probability of disease. While it will be undoubtedly challenging, the quest for genetic determinants in acute lung injury holds great promise in helping to clarify the heterogeneity of the condition. However, the ultimate goal in epidemiology is in the intervention in and prevention of disease on the population level. Unlike other risk factors for ARDS, the stable nature of genetics allows for the prospective determination of individuals at high risk for the development of or mortality from acute lung injury. This knowledge will be important in the design of future preventive and therapeutic trials. In addition, knowledge of genetic risk factors with high population-attributable risk can help focus preventive and therapeutic trials in ways that will have the greatest population impact. Lastly, knowledge of gene-environment interaction allows for the targeting of high-risk individuals and, more importantly, offers them an opportunity to reduce their overall risk by the modulation of their environment.

Acknowledgments The above work is supported by Research Grants HL60710 and ES00002 from the National Institutes of Health. Dr. Gong is supported by K23 HL67197 from NHLBI.

References 1. Ashbaugh DG, Biglelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967; 2:319–323.

Acute respiratory distress syndrome

352

2. Rinaldo J, Cristman J. Mechanisms and mediators of the adult respiratory distress syndrome. Clin Chest Med 1990; 11:621–632. 3. Parson PE. Mediators and mechanisms of acute lung injury. Clin Chest Med 2000; 21:467–476. 4. Fowler A, Hamman R, Zerbe G, et al. Adult respiratory distress syndrome: risk with common predispositions. Ann Intern Med 1985; 99:293–298. 5. Hudson LD, Milberg JA, Anardi D, Maunder RJ. Clinical risks for development of acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:293–301. 6. Iribarren C, Jacobs DR, Sidney S, Gross MD, Eisner MD. Cigarette smoking, alcohol consumption, and risk of ARDS: a 15-year cohort study in a managed care setting. Chest 2000; 117:163–168. 7. Moss M, Bucher B, Moore FA, Moore EE, Parsons PE. The role of chronic alcohol abuse in the development of acute respiratory distress syndrome in adults. JAMA 1996; 275:50–54. 8. Monchi M, Bellenfant F, Cariou A, Joly L-M, Thebert D, Laurent I, Dhainaut J-F, Brunet F. Early predictive factors of survival in acute respiratory distress syndrome: a multivariate analysis. Am J Respir Crit Care Med 1998; 158:1076–1081. 9. Zilberberg MD, Epstein SK. Acute lung injury in the medical ICU: comorbid conditions, age, etiology and hospital outcome. Am J Respir Crit Care Med 1998; 157:1159–1164. 10. Suchyta MR, Clemmer TP, Elliott CG, Orme JF, Morris AH, Jacobson J, Menlove R. Increased mortality of older patients with acute respiratory distress syndrome. Chest 1997; 111:1334– 1339. 11. Headley AS, Tolley E, Meduri GU. Infections and the inflammatory response in acute respiratory distress syndrome. Chest 1997; 111:1306–1321. 12. Mc Kay CJ, Gallagher G, Brooks B, Imrie CW, Baxter JN. Increased monocyte cytokine production in association with systemic complications in acute pancreatitis. Br J Surg 1996; 83:919–923. 13. Meduri GU, Headley AS, Kohler G, Stentz F, Tolley E, Umberger R, Leeper K. Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Chest 1995; 107:1062– 1073. 14. Parsons PE, Moore FA, Moore EE, Ikle DN, Ikle DN, Henson PM, Worthen GS. Studies on the role of tumor necrosis factor in adult respiratory distress syndrome. Am Rev Respir Dis 1992; 146:694–700. 15. Donnelly TJ, Meade P, Jagels M, Cryer HG, Law MM, Hugli TE, Shoemaker WC, Abraham E. Cytokine, complement and endotoxin profiles associated with the development of adult respiratory distress syndrome after severe injury. Crit Care Med 1994; 22:768–776. 16. Pittet JF, Mackersie RC, Martin TR, Matthay MA. Biological markers of acute lung injury: prognostic and pathogenetic significance. Am J Respir Crit Care Med 1997; 155:1187–1205. 17. Lander ES, Schork NJ. Genetic dissection of complex traits. Science 1994; 265:2037–2048. 18. Ellsworth DL, Manolio TA. The emerging importance of genetics in epidemiologic research I: Basic concepts in human genetics and laboratory technology. Ann Epidemiol 1999; 9:1–16. 19. Ellsworth DL, Manolio TA. The emerging importance of genetics in epidemiologic research II: Issues in study design and gene mapping. Ann Epidemiol 1999; 9:75–90. 20. Ellsworth DL, Manolio TA. The emerging importance of genetics in epidemiologic research III: Bioinformatics and statistical genetic methods. Ann Epidemiol 1999; 9:207–224. 21. Kaprio J. Genetic epidemiology. BMJ 2000; 320:1257–1259. 22. Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1989; 138:720–723. 23. Bernard G, Artigas A, Brigham K, Carlet J, Falke K, Hudson L, Lamy M, Legall J, Morris A, Spragg R, and the Consensus Committee. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes and clinical trial coordinations. Am J Respir Crit Care Med 1994; 149:818–824. 24. Milberg JA, Davis DR, Steinberg KP, Hudson LD. Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983–1993. JAMA 1995; 273:306–309.

Approach to the genetic epidemiology of acute lung injury

353

25. Chevrolet J Cl, Guelpa G, Schifferli JA. Recurrent adult respiratory distress-like syndrome associated with propylthiouracil therapy. Eur Respir J 1991; 4:899–901. 26. Yost DA, Michalowski E. Adult respiratory distress syndrome complicating recurrent antepartum pyelonephritis. J Fam Pract 1990; 31:81–82. 27. Gonzolez ER, Cole T, Grimes MM, et al. Recurrent ARDS in a 39 year-old woman with migraine headaches. Chest 1998; 114:919–922. 28. Suarez M, Krieger BP. Bronchoalveolar lavage in recurrent aspirin-induced adult respiratory distress syndrome. Chest 1986; 90:452–453. 29. Brun-Buisson CJL, Bonnet F, Bergeret S, et al. Recurrent high-permeability pulmonary edema associated with diabetic ketoacidosis. Crit Care Med 1985; 13:55–56. 30. Savici D, Katzenstein AL. Diffuse alveolar damage and recurrent respiratory failure: report of 6 cases. Hum Pathol 2001; 32:1398–1402. 31. Rocco TR, Reinert SE, Cioffi W, Harrington D, Buczko G, Simms HH. A 9-year, singleinstitution, retrospective review of death rate and prognostic factors in adult respiratory distress syndrome. Ann Surg 2001; 233:414–422. 32. Jardin F, Fellahi J-L, Beauchet A, Vieillard-Baron A, Loubieres Y, Page B. Improved prognosis of acute respiratory distress syndrome 15 years on. Intens Care Med 1999; 25:936–941. 33. McGue M, Vaupel JW, Holm N, Harvald B. Longevity is moderately heritable in a sample of Danish twins born 1870–1880. J Geronto Biol Sci 1993; 48:B237–244. 34. Sorensen TIA, Nielsen GG, Andersen PK, Teasdale TW. Genetic and environmental influences on premature death in adult adoptees. N Engl J Med 1988; 318:727–732. 35. Bell R, Coalson K, Smith J, et al. Multiple organ system failure and infection in adult respiratory distress syndrome. Ann Intern Med 1983; 99:293–298. 36. Montgomery A, Stager M, Carrico C, et al. Causes of mortality in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1985; 132:485–489. 37. Steinberg KP, McHugh LG, Hudson LD. Causes of mortality in patients with adult respiratory distress syndrome. Am Rev Respir Dis 1993; 147:A347. 38. Ferring M, Vincent JL. Is outcome from ARDS related to the severity of respiratory failure? Eur Respir J 1997; 10:1297–1300. 39. Wilson A, Symons J, McDowell T, McDevitt H, Duff G. Effects of a polymorphism in the human tumor necrosis factor a promoter on transcriptional activation. Proc Natl Acad Sci 1997; 94:3195–3199. 40. Grau G, Taylor T, Molyneux M, Wirima J, Vassalli P, Hommel M, Lambert P. Tumor necrosis factor and disease severity on children with falciparum malaria. N Engl J Med 1989; 320:1586– 1591. 41. Wilson A, de Vries N, Pociot F, di Giovane F, vanderPutte L, Duff G. An allelic polymorphism within the human tumor necrosis factor alpha promoter region is strongly associated with HLA A1, B8 and DR3 alleles. J Exp Med 1993; 177:557–560. 42. Kwiatkowski D, Hill A, Sambou I, Twumasi P, Castracane J, Manogue K, Cerami A, Brewster D, Greenwood B. TNF concentration in fatal cerebral, non-fatal cerebral and uncomplicated Plasmodium falciparum malaria. Lancet 1990; 336:1201–1204. 43. Stuber F, Udalova I A, Book M, Drutskaya LN, Kuprash DV, Turetskaya RL, Schade FU, Nedospasov SA. −308 Tumor necrosis factor polymorphism is not associated with survival in severe sepsis and is unrelated to lipopolysaccharide inducibility of the human TNF promoter. J Inflamm 1996; 46:42–50. 44. Mira JP, Cariou A, Grall F, Delclaux C, Losser MR, Heshmati F, Cheval C, Monchi M, Teboul JL, Riche F, Leleu G, Arbibe L, Mignon A, Delpech M, Dhainaut JF. Association of TNF2, a TNF-α promoter polymorphism, with septic shock susceptibility and mortality. JAMA 1999; 282(6):561–568. 45. Pociot F, Briant L, Jongeneel C, et al. Association of tumor necrosis factor and class II major histocompatibility complexes alleles with the secretion of TNF-α and TNF-β by human

Acute respiratory distress syndrome

354

mononuclear cells: a possible link to insulin-dependent diabetes mellitus. Eur J Immunol 1993; 23:224–231. 46. Molvig J, Pociot F, Baek L, et al. Monocyte function in IDDM patients and healthy individuals. Scand J Immunol 1990; 31:297–306. 47. Stuber F, Petersen M, Bokelmann F, Schade U. A genomic polymorphism within the tumor necrosis factor locus influences plasma tumor necrosis factor-α concentrations and outcomes of patients with severe sepsis. Crit Care Med 1996; 381–384. 48. Majetschak M, Flohe S, Obertacke U, Schroder J, Staubach J, Staubauch K, Nast-Kolb D, Schade U, Stuber F. Relation of a TNF gene polymorphism to severe sepsis in trauma patients. Ann Surg 1999; 230(2):207–214. 49. Waterer GW, Quasney MW, Cantor RM, Wunderlink RG. Septic shock and respiratory failure in community-acquired pneumonia have different TNF polymorphism associations. Am J Respir Crit Care Med 2001; 163:1599–1604. 50. McIntyre I. Importance of underlying mechanism and genotype on outcome of sepsis trials. Crit Care Med 2001; 29:677–679. 51. Dorman T, Faraday N. Do gene variants really explain the heterogeneous outcomes in sepsis? Crit Care Med 2001; 29:684–685. 52. Max M, Pison U, Floros J. Frequency of AP-B and SP-A1 gene polymorphism in the acute respiratory distress syndrome (ARDS). Appl Cardiopulm Pathophysiol 1996; 6:111–118. 53. Lin Z, Oearson C, Chinchilli V, Pietschmann SM, Luo J, Pison U, Floros J. Polymorphisms of the human SP-A, SP-B, and SP-D genes: association of SP-B Thr131Ile with ARDS. Clin Genet 2000; 58:181–191. 54. Li CC. Large random mating population. In: Li CC, ed. Population Genetics. Pacific Grove, CA: Boxwood Press, 1976:1–15. 55. Silverman EK, Palmer LJ. Case-control association studies for the genetics of complex respiratory diseases. Am J Respir Cell Mol Biol 2000; 22:645–648. 56. Unpublished data. 57. Marshall RP, Webb S, Hill MR, Humphries SE, Laurent GJ. Genetic polymorphisms associated with susceptibility and outcome in ARDS. Chest 2002; 121:68S–69S. 58. Fourrier F, Chopin C, Wallaert B, Mazurier C, Mangalaboyi J, Durocher A. Compared evolution of plasma fibronectin and angiotensin-converting en-zyme levels in septic ARDS. Chest 2002; 87:191–195. 59. Johnson AR, Coalson JJ, Ashton J, Larumbide M, Erdos EG. Neutral endopeptidase in serum samples from patients with adult respiratory distress syndrome. Comparison with angiotensinconverting enzyme. Am Rev Respir Dis 1985; 132:1262–1267. 60. Orfanos E, Armaganidis A, Glynos C, Psevdi E, Kalltsas P, Sarafidou P, Catravas JD, Dafni UG, Langleben D, Roussos C. Pulmonary capillary endo-thelium-bound angiotensin-converting enzyme activity in acute lung injury. Circulation 2000; 102:2011–2018. 61. Moss M, Gillespie MK, Ackerson L, Moore FA, Moore EE, Parsons PE. Endothelial cell activity varies in patients at risk for the adult respiratory distress syndrome. Crit Care Med 1996; 24:1782–1786. 62. Schork NL. Genetics of complex disease: approaches, problems, and solution. Am J Respir Crit Care Med 1997; 156:103–109. 63. Gambaro G, Anglani F, D’ Angelo A. Association studies of genetic polymorphisms and complex disease. Lancet 2000; 355:308–311. 64. Zhao LP, Aragaki C, Hsu F, Quiaoit F. Mapping of complex traits by single nucleotide polymorphisms. Am J Hum Genet 1998; 63:225–240. 65. Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science 1996; 273:1516–1517. 66. Rao DC, Keats BJB, Morton NE, Yee S, Lew R. Variability of human linkage data. Am J Hum Genet 1978; 30:516–529.

Approach to the genetic epidemiology of acute lung injury

355

67. Hall JM, Lee MK, Newman B, Morrow JE, Anderson LA, Huey B, King MC. Linkage of earlyonset familial breat cancer to chromosome 17q21. Science 1990; 250:1684–1689. 68. Beaty TH, Khoury MJ. Interface of genetics and epidemiology. Epidemiol Rev 2000; 22:120– 125. 69. Pericak-Vance MA, Bebout JL, Gaskell PC Jr, Yamaoka LH, Hung WY, Alberts MJ, Walker AP, Barlett RJ, Haynes CA, Welsh KA, et al. Linkage studies in familial Alzheimer disease: evidence for chromosome 19 linkage. AM J Hum Genet 1991; 48:1034–1050. 70. Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild J, Salvesem GS, Roses AD. Apolipoprotein E: high-avidity binding to β-amyloid and increased frequency of type 4 allele in late onset familial Alzheimer disease. Proc Natl Acad Sci 1993; 90:1977–1981. 71. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak-Vance MA. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 1993; 261:921–923. 72. Andrieu N, Goldstein AM. Epidemiologic and genetic approaches in the study of geneenvironment interaction: an overview of available methods. Epidemiol Rev 1998; 20:137–147. 73. Whittemore AS, Gong G, Itnyre J. Prevalence and contribution of BRCA1 mutations in breast and ovarian cancer: results from three US population-based case-control studies of ovarian cancer. Am J Hum Genet 1997; 60:496–504. 74. Khoury MJ, Yang Q. The future of genetic studies of complex human diseases: an epidemiologic perspective. Epidemiology 1998; 9:350–354. 75. Greenberg DA. Linkage analysis of necessary disease loci versus susceptibility loci. Am J Hum Genet 1993; 52:135–143. 76. Altshuler D, Hirschhorn JN, Klannemark M, et al. The common PPARγ Pro12Ala polymorphism is associated with decreased risk of type 2 diabetes. Nat Genet 2000; 26:76–80. 77. Khoury MJ, Beaty TH. Applications of the case-control method in genetic epidemiology. Epidemiol Rev 1994; 16:134–150. 78. Vainio H. Promise of molecular epidemiology-epidemiologic reasoning, biological rationale and risk assessment. Scand J Work Environ Health 1999; 25:498–504. 79. Bogardus ST, Concato J, Feinstein AR. Clinical epidemiological quality in molecular genetic research: the need for methodological standards. JAMA 1999; 281:1919–1926. 80. Boffetta P. Molecular epidemiology. J Int Med 2000; 248:447–454. 81. Rubenfeld GD, Caldwell E, Granton J, Hudson LD, Matthay M. Interobserver variability in applying a radiographic definition for ARDS. Chest 1999; 116:1347–1353. 82. Cookson W, Palmer L. Investigating the asthma phenotype. Clin Exp Allergy 1998; 28:88–89. 83. Hennekens CH, Buring JE. Case-control studies. Epidemiology in Medicine. Boston: Little, Brown and Company, 1987:132–152. 84. Veletza SV, Rogan PK, TenHave T, Olowe SA, Floros J. Racial differences in allelic distribution at the human pulmonary surfactant protein B gene locus (SP-B). Exp Lung Res 1996; 22:489–494. 85. Wacholder S, Rothman N, Caporaso N. Population stratification in epidemiologic studies of common genetic variants and cancer: quantification of bias. J Natl Cancer Inst 2000; 1151– 1158. 86. Wright AF, Carothers AD, Pirastu M. Population choice in mapping genes for complex diseases. Nat Genet 1999; 23:397–404. 87. Schork NJ, Fallin D, Thiel B, Xu X, Broeckel U, Jacob HJ, Cohen D. The future of genetic case control studies. Adv Genet 2001; 42:191–212. 88. Hennekens CH, Buring JE. In: Case-Control Studies. Epidemiology in Medicine. Boston: Little, Brown and Company, 1987:132–152. 89. Bailey-Wilson JE, Sorant B, Sorant AJ, Paul CM, Elston RC. Model-free association analysis of a rare disease. Genet Epidemiol 1995; 12:571–575.

Acute respiratory distress syndrome

356

90. Deeb SS, Fajas L, Nemoto M, Pihlajamaki J, Mykkanen L, Kuusisto J, Laakso M, Fujimoto W, Auwerx J. A Pro12Ala substitution in PPARγ2 associated with decreased receptor activity, lower body mass index and improved insulin sensitivity. Nat Genet 1998; 20:284–287. 91. Hara K, Okada T, Tobe K, Yasuda K, Mori Y, Kadowaki H, Hagura R, Akanuma Y, Kimura S, Ito C, Kadowaki T. The Pro12Ala polymorphism in PPARγ2 may confer resistance to type 2 diabetes. Biochem Biophys Res Commun 2000; 271:212–216. 92. Mancini FP, Vaccaro O, Sabatino L, Tufano A, Rivellese AA, Riccardi G, Colanntuoni V. Pro12Ala substitution in peroxisome proliferator-activated receptor-γ2 is not associated with type 2 diabetes. Diabetes 1999; 48:1466–1468. 93. Meirhaeghe A, Fajas L, Hellbecque N, Cottel D, Auwerx J, Deeb SS, Amouyel P. Impact of the peroxisome proliferator-activated receptor-γ2 Pro12Ala polymorphism on adiposity, lipids and non-insulin-dependent diabetes. Int J Obes Relat Metab Disord 2000; 24:195–199.

15 Resolution of Alveolar Edema Mechanisms and Relationship to Clinical Acute Lung Injury MICHAEL A.MATTHAY and XIAOHUI FANG University of California at San Francisco San Francisco, California, U.S.A. TSUTOMU SAKUMA Kanazawa Medical University Uchinada, Ishikawa, Japan CHRISTINE CLERICI Université Paris 13 Colombes, France SADIS MATALON University of Alabama at Birmingham Birmingham, Alabama, U.S.A.

Studies of epithelial fluid transport by the distal pulmonary epithelium have provided important new concepts regarding the resolution of pulmonary edema, a common clinical problem. For many years it was generally believed that differences in hydrostatic and protein osmotic pressures (Starling forces) accounted for the removal of excess fluid from the air spaces of the adult lung. Until the early 1980s, there were no satisfactory adult animal models to study the resolution of alveolar edema, and the isolation and culture of alveolar epithelial type II cells was just becoming a useful experimental method. This chapter discusses the regulation of lung fluid balance by active transport mechanisms across both the alveolar and distal airway epithelium of the mature lung, in both animals with conditions resembling acute lung injury or respiratory distress syndrome and in patients with acute lung injury (ALI), acute respiratory distress syndrome (ARDS), and cardiogenic pulmonary edema. Some of the information and perspectives in this chapter has been included in a recent review (1). A review of fetal and newborn lung fluid balance is beyond the scope of this chapter, although articles by Strang, Olver, Walters, Bland, Folkesson, and other investigators are available (2–4).

Acute respiratory distress syndrome

358

I. Mechanisms of Alveolar Epithelial Fluid Absorption The general model for transepithelial fluid movement is that active salt transport generates a minisomotic pressure gradient that draws water across a tight epithelial barrier. This paradigm is probably correct for fluid clearance from the distal air spaces of the lung (5, 6). The results of several in vivo studies have demonstrated that changes in hydrostatic or protein osmotic pressures cannot account for the removal of excess fluid from the distal air spaces (7–10). Furthermore, pharmacological inhibitors of sodium transport reduce the rate of fluid clearance in the lungs of several different species, including humans (9, 11–15). In addition, there is good evidence that isolated epithelial cells from lung distal air spaces actively transport sodium and other ions (5). More recent evidence indicates that under some conditions alveolar epithelial cells also secrete chloride both in vivo and in vitro and that movement of chloride significantly impacts on sodium transport. In addition, several functional water channels, aquaporins, have been identified in alveolar epithelial cells; however, in contrast to the kidneys, aquaporins do not seem to play a significant role in fluid movement across the normal or injured alveolar epithelium. A. Structure of the Distal Pulmonary Epithelia Although the large surface area of the alveoli favors the hypothesis that most fluid reabsorption occurs at the alveolar level, active fluid reabsorption may occur across all of the different segments of the pulmonary epithelium of the distal air spaces of the lung. The exact contribution of each of the anatomical segments of the distal airspaces to fluid reabsorption is not firmly established. The human lung consists of a series of highly branched hollow tubes that end blindly in the alveoli, with the conducting airways (the cartilaginous trachea, bronchi, and the membranous bronchioles) occupying the first 16 airway generations. The airways and alveoli, approximately 1.4 m2 and 143 m2 in the adult human lung (16), respectively, constitute the interface between lung parenchyma and the external environment and are lined by a continuous epithelium. The distal airway epithelium is composed of terminal respiratory and bronchiolar units with polarized epithelial cells that have the capacity to transport sodium and chloride, including ciliated Clara cells and nonciliated cuboidal cells. The alveoli are composed of a thin alveolar epithelium (0.1–0.2 µm) that covers 99% of the air space surface area in the lung and contains thin, squamous type I cells and cuboidal type II cells (16, 17). The alveolar type I cell covers 95% of the alveolar surface (16). The close apposition between the alveolar epithelium and the vascular endothelium facilitates efficient exchange of gases, but also forms a tight barrier to movement of liquid and proteins from the interstitial and vascular spaces, thus assisting in maintaining relatively dry alveoli (18). The tight junctions are critical structures for the barrier function of the alveolar epithelium. Ion transporters and other membrane proteins are asymmetrically distributed on opposing cell surfaces, conferring vectorial transport properties to the epithelium. Physiological studies of the barrier properties of tight junctions in the alveolar epithelium indicate that diffusion of water-soluble solutes between alveolar epithelial cells is much

Resolution of alveolar edema

359

slower than through the intercellular junctions of the adjacent lung capillaries (19). Based on tracer fluxes of small water-soluble solutes across the air/blood barrier of the distal air spaces, the effective pore radii were 0.5–0.9 nm in the distal respiratory epithelium and 6.5–7.5 nm in the capillary endothelium. Removal of large quantities of soluble protein from the air spaces appears to occur primarily by restricted diffusion (20, 21), although there is evidence for some endocytosis and transcytosis of albumin across alveolar epithelium (22). Overall, the movement of protein across the normally tight alveolar epithelial barrier is very slow. The alveolar type II cell is responsible for the secretion of surfactant as well as vectorial transport of sodium from the apical to the basolateral surface. Sodium uptake occurs on the apical surface, partly through amiloride-sensitive and amiloride-insensitive channels. Subsequently, sodium is pumped actively from the basolateral surface into the lung interstitium by Na, K-ATPase. An epithelial sodium channel (ENaC) which participates in sodium movement across the cell apical membrane was cloned from the colon of salt deprived rats and characterized in 1994 (23), and work by other investigators has provided new insight into the molecular and biochemical basis for sodium uptake in alveolar epithelial cells (24, 25) (see details below). The role of the alveolar type I cell in vectorial fluid transport in the lung is unknown although several investigators are currently trying to assess the potential contribution of the alveolar type I cell to vectorial fluid transport. On the basis of studies in freshly isolated type I cells, it is known that these cells have a high osmotic permeability to water with expression of aquaporin 5 on the apical surface (26). Immunocytochemical studies in the intact lung by some investigators failed to demonstrate the presence of Na, KATPase in these cells in vivo (27). However, more recent studies have reported the presence of the α1 and α2 subunits of Na, K-ATPase in both type-1 like cells in vitro (28). The presence of the Na, K-ATPase could be consistent with a role for this cell in vectorial fluid transport, although Na, K-ATPase may also be needed to maintain cell volume. Also, recent studies of freshly isolated alveolar type I cells from rats demonstrated that these cells express the Na, K-ATPase α1 and β1 subunit isoforms, but not the α2 subunit (29). In addition, there is evidence for expression of all the subunits of ENaC in freshly isolated alveolar type I cells from two laboratories (29, 30) as well as in situ in the rat lung (29). Finally, there is some evidence that 22Na uptake can be partially inhibited by amiloride in the freshly isolated rat alveolar type I cells (30), although definitive studies of cultured, polarized type I cells have not yet been achieved. Although the inability to study alveolar type I cells in culture has hindered progress in assessing their capacity for ion transport and the role they may play in vectorial fluid transport across the alveolar epithelium, the new evidence provides suggestive evidence that alveolar type I cells may participate in vectorial salt transport in the lung. The alveolar epithelium comprises 99% of the surface area of the lung, a finding that suggests that removal of edema fluid from the lung might primarily occur across the alveolar epithelium. However, it has been demonstrated that the distal airway epithelium actively transports sodium, a process that depends on amiloride inhibitable uptake of sodium on the apical surface and extrusion of sodium through a basolateral Na, KATPase (31). In support of the potential role of distal airway cells is evidence that Clara cells actively absorb sodium and transport from an apical to basal direction (32). Also, there are new data on the possible role of CFTR in upregulating cAMP fluid clearance

Acute respiratory distress syndrome

360

(33). This information provides support for a possible role of distal airway epithelia in fluid clearance since CFTR is expressed abundantly in distal airway epithelial cells of both adult and fetal cells (34). Thus, even though their surface area is limited, a contribution from distal airway epithelia to the overall fluid transport is probable, especially since cells from the distal airway epithelium primarily transport salt from the apical to the basolateral surface. The contribution of distal airway epithelia needs to be better defined. B. Fluid Transport in the Lung Several innovative experimental methods have been used to study fluid and protein transport from the distal airspaces of the intact lung, including isolated perfused lung preparations, in situ lung preparations, surface fluorescence methods, and intact lung preparations in living animals for short time periods (30–240 min) or for extended time periods (24–144 h). The advantages and disadvantages of these preparations have been reviewed in some detail (6, 35). In vivo evidence that active ion transport could account for the removal of alveolar edema fluid across the distal pulmonary epithelium of the mature lung was obtained in studies of anesthetized, ventilated sheep (36). In those studies the critical discovery was that isosmolar fluid clearance of salt and water occurred in the face of a rising concentration of protein in the air spaces of the lung, whether the instilled solution was autologous serum or an isosmolar protein solution. The initial protein concentration of the instilled protein solution was the same as the circulating plasma. After 4 hours, the concentration of the protein had risen from approximately 6.5 to 8.4 g/100 mL, while the plasma protein concentration was unchanged. In longer-term studies in unanesthetized, spontaneously breathing sheep, alveolar protein concentrations increased to very high levels. After 12 and 24 hours, the alveolar protein concentration increased to 10.2 and 12.9 g/ 100 mL, respectively (37). The overall rise in protein concentration was equivalent to an increase in distal air space protein osmotic pressure from 25 to 65 cm H2O. Other studies in the intact lung have supported the hypothesis that removal of alveolar fluid requires active transport processes. For example, elimination of ventilation to one lung did not change the rate of fluid clearance in sheep, thus ruling out changes in transpulmonary airway pressure as a major determinant of fluid clearance, at least in the uninjured lung (38). Furthermore, if active ion transport were responsible for fluid clearance, then fluid clearance should be temperature dependent. In an in situ perfused goat lung preparation, the rate of fluid clearance progressively declined as temperature was lowered from 37 to 18°C (39). Similar results were obtained in perfused rat lungs (40) and ex vivo human lung studies (14) in which hypothermia inhibited sodium and fluid transport. Additional evidence for active ion transport was obtained in intact animals with the use of amiloride, an inhibitor of sodium uptake by the apical membrane of alveolar epithelium and distal airway epithelium. Amiloride inhibited 40–70% of basal fluid clearance in sheep, rabbits, rats, guinea pigs, mice, and in the human lung (1). Amiloride also inhibited sodium uptake in distal airway epithelium from sheep and pigs (41). To further explore the role of active sodium transport, experiments were designed to inhibit

Resolution of alveolar edema

361

Na, K-ATPase. It has been difficult to study the effect of ouabain in intact animals because of cardiac toxicity. However, in the isolated rat lung, ouabain inhibited >90% of fluid clearance (42, 43). Subsequently, following the development of an in situ sheep preparation for measuring fluid clearance in the absence of blood flow, it was reported that ouabain inhibited 90% of fluid clearance over a 4-hour period (38). C. Sodium Transport in Cultured Alveolar Epithelial Cells The success in obtaining high-purity cultures of alveolar epithelial type II cells from rats made it possible to study the transport properties of these cells and relate the results to the findings in the intact lung studies. The initial studies showed that when type II cells were cultured on a nonporous surface such as plastic, they would form a continuous confluent layer of polarized cells after 2–3 days (44, 45). Interestingly, after 3–5 days, small domes of fluid could be appreciated from where the substratum was detached. The domes were thought to result from active ion transport from the apical to the basal surface with water following passively since they were inhibited by the replacement of sodium by another cation or by pharmacological inhibitors of sodium transport, such as amiloride and ouabain (46). More detailed information on the nature of ion transport across alveolar type II cells was obtained by culturing these cells on porous supports and mounting them in Ussing chambers and measuring short circuit current and ion flux under voltage clamp conditions. Further details are available in a recent review (1).

II. Regulation of Epithelial Fluid Transport Until recently, most studies focused primarily on the active transport of sodium as the primary determinant for regulating catecholamine-dependent transport across distal pulmonary epithelium. New evidence indicates that cAMP-stimulated uptake of either chloride or bicarbonate may be an important mechanism for regulating fluid clearance across distal lung epithelium (33, 34, 47–50). Therefore, this section is divided into the effects of catecholamines on fluid transport based primarily on studies that examined the effects of sodium transport inhibitors and the more recent work on the role of chloride. A. Upregulation of Fluid Transport by Sodium-Dependent Mechanisms Studies in newborn animals indicate that endogenous release of catecholamines, particularly epinephrine, may stimulate reabsorption of fetal lung fluid from the air spaces of the lung (4, 11, 51). In most adult mammal species, stimulation of β2-adrenergic receptors by either salmeterol, terbutaline, or epinephrine increases fluid clearance (1). This stimulatory effect occurs rapidly after intravenous administration of epinephrine or instillation of terbutaline in alveolar space and is completely prevented by either a nonspecific β2-receptor antagonist, propranolol, or in rats by a specific β2 antagonist. The increased fluid clearance by β2 agonists can be prevented by amiloride, indicating that the stimulation was related to an increased transepithelial sodium transport. In anesthetized ventilated sheep, terbutaline-induced stimulation of fluid clearance was

Acute respiratory distress syndrome

362

associated with an increase in lung lymph flow, a finding that reflected removal of some of the alveolar fluid volume to the interstitium of the lung (9). Although terbutaline increased pulmonary blood flow, this effect was not important since control studies with nitroprusside, an agent that increased pulmonary blood flow, did not increase fluid clearance. Other studies have demonstrated that β-adrenergic agonists increased fluid clearance in several animal species, as well as human lung (4, 14, 52). Based on studies of the resolution of alveolar edema in humans, it has been difficult to quantify the effect of catecholamines on the rate of fluid clearance (53). However, studies of fluid clearance in the isolated human lung have demonstrated that β-adrenergic agonist therapy increases fluid clearance, and the increased fluid clearance can be inhibited with propranolol or amiloride (14, 52). Subsequent studies showed that long-acting lipidsoluble β-agonists are more potent than hydrophilic β-agonists in the ex vivo human lung (52). The magnitude of the effect is similar to that observed in other species, with a βagonist-dependent doubling of fluid clearance over baseline levels. This data is particularly important because aerosolized β-agonist treatment in some patients with pulmonary edema might accelerate the resolution of alveolar edema. What has been learned about the basic mechanisms the mediate the catecholamine dependent upregulation of sodium transport in the lung? Based on in vitro studies, it was proposed that an increase in intracellular cAMP resulted in increased sodium transport across alveolar type II cells by an independent upregulation of the apical sodiumconductive pathways and the basolateral Na, K-ATPase. Proposed mechanisms for upregulation of sodium transport proteins by cAMP include augmented sodium channel open probability, increases in Na, K-ATPase a subunit phosphorylation, delivery of more ENaC channels to the apical membrane, and more Na, K-ATPases to the basolateral cell membrane (1). B. CFTR in cAMP-Mediated Up regulated Fluid Transport While most experimental studies have attributed a primary role for active sodium transport in the vectorial transport of salt and water from the apical to the basal surface of the alveolar epithelium of the lung, the potential role of chloride, especially in mediating the cAMP-mediated upregulation of fluid clearance across distal lung epithelium, has been the subject of a few recent studies. One older study of cultured alveolar epithelial cells concluded that vectorial transport of chloride across alveolar epithelium occurs by a paracellular route under basal conditions and perhaps by a transcellular route in the presence of cAMP stimulation (54). Another study of cultured alveolar epithelial type II cells suggested that cAMP-mediated apical uptake of sodium might depend of an initial uptake of chloride (47). A more recent study of cultured alveolar type II cells under apical air interface conditions reported that β-adrenergic agonists produced acute activation of apical chloride channels with enhanced sodium absorption (48). However, the results of these studies have been considered to be inconclusive by some investigators (55–57), partly because the data depend on cultured cells of an uncertain phenotype. Furthermore, studies of isolated alveolar epithelial type II cells do not address the possibility that vectorial fluid transport may be mediated by several different epithelial cells including alveolar epithelial type I cells as well as distal airway epithelial cells.

Resolution of alveolar edema

363

In order to define the role of chloride transport in the active transport of salt and water across the distal pulmonary epithelium of the lung, two groups have used in vivo lung studies to define the mechanisms and pathways that regulate chloride transport during the absorption of fluid from the distal air spaces of the lung. This approach may be important because studies in several species, as already discussed, have indicated that distal airway epithelia are capable of ion transport (41, 58, 59) and both ENaC and CFTR are expressed in alveolar and distal airway epithelia (60–64). Both inhibition and ion substitution studies demonstrated that chloride transport was necessary for basal fluid clearance. In the first group of studies (50), rabbits administered Na+ methane sulfonate were observed to have a significantly (p<0.05) decreased AFC compared to the control group (i.e., rabbits that were instilled with NaCl) at 60 and 120 minutes postinstillation. Rabbits instilled with Na+ methane sulfonate and forskolin demonstrated a negative AFC (i.e., fluid secretion into the alveolar space) at 60 minutes postinstillation. These studies provided evidence for the coupled transcellular movement of Na and Cl both under basal and cAMP-activated conditions and indicate that under some conditions (absence of chloride in the alveolar space), chloride secretion may occur. In a later study (34), this group provided evidence that agents that increase cAMP stimulate either chloride or bicarbonate secretion across confluent monolayeers of distal airway cells isolated from the lungs of matured fetal rats. In more definitive studies, the potential role of CFTR under basal and cAMPstimulated conditions was tested using intact lung studies in which CFTR was not functional because of failure in trafficking of CFTR to the cell membrane, the most common human mutation in cystic fibrosis (∆F508 mice). The results supported the hypothesis that CFTR was essential for cAMP-mediated upregulation of isosmolar fluid clearance from the distal airspaces of the lung because fluid clearance could not be increased in the ∆F508 mice with either β-agonists or forskolin, unlike the wild-type control mice (33). Additional studies using pharmacological inhibition of CFTR in both the mouse and human lung with gilbenclamide supported the same conclusion, namely that chloride uptake and CFTR-like transport seemed to be required for cAMP-stimulated fluid clearance from the distal airspaces of the lung (33). Glibenclamide can also inhibit potassium channels so the inhibitory effects are not specific for CFTR, but the ∆F508 mouse studies have provided more direct evidence. Although the absence of CFTR in the upper airways results in enhanced sodium absorption (65), the data in these studies provide additional evidence that the absence of CFTR prevents cAMP-upregulated fluid clearance from the distal air spaces of the lung, a finding that is similar to work on the importance of CFTR in mediating cAMP-stimulated sodium absorption in human sweat ducts (66). Because CFTR is distributed throughout the distal pulmonary epithelium in distal airway epithelium as well at the alveolar level in the human lung (67), the data also suggest that the cAMP-mediated upregulated reabsorption of pulmonary edema fluid may occur across distal airway epithelium as well as at the level of the alveolar epithelium. Finally, additional studies indicated that the lack of CFTR results in a greater accumulation of pulmonary edema in the presence of a hydrostatic stress, thus demonstrating the potential physiological importance of CFTR in upregulating fluid transport from the distal air spaces of the lung (33). C. Catecholamine-Independent Regulation

Acute respiratory distress syndrome

364

In the last few years several interesting catecholamine-independent mechanisms have been identified that can upregulate fluid transport across the distal air spaces of the lung as well as in cultured alveolar type II cells. Hormonal factors, such as glucocorticoids, can upregulate transport by transcriptional mechanisms, while thyroid hormone may work by a post-translational mechanism (68). Some growth factors can work by either a transcriptional or direct membrane effect or by enhancing the number of alveolar type II cells. There is also evidence that a proinflammatory cytokine, tumor necrosis factor (TNF)-α, can rapidly upregulate sodium uptake and fluid transport by a novel mechanism. Finally, serine proteases can regulate the activity of ENaC and potentially increase fluid clearance across the distal airway epithelium. These mechanisms are explored in a recent review in some detail (1).

III. Mechanisms That Impair Vectorial Fluid Transport Several mechanisms have been identified that can impair fluid transport from the distal airspaces of the lung. This section will discuss how hypoxia, anesthetics, and reactive oxygen and nitrogen species may affect the resolution of pulmonary edema under clinically relevant conditions. The next section will review mechanisms that impair fluid transport under specific pathological conditions. A. Hypoxia Hypoxia may occur during residence or recreation at high altitudes and under a variety of pathological conditions associated with acute and chronic respiratory disease. Therefore, it is important to understand the effect of hypoxia on the ion and fluid transport capacity of the lung epithelium. The effect of hypoxia under in vivo conditions has been studied primarily in rats. In anesthetized rats, as well as in isolated perfused lungs, hypoxia decreased alveolar liquid clearance by inhibition of the amiloride-sensitive component (69, 70). In contrast to the in vitro studies, hypoxia increased α-rENaC and β1-Na, K-ATPase mRNA transcripts with little increase or no change in protein amounts, suggesting a posttranslational mechanism such as a direct change of sodium transporter protein activity or protein internalization (70). This latter hypothesis was supported by the normalization of fluid clearance by a cAMP agonist (terbutaline), which is known to increase the trafficking of sodium transporter proteins from the cytoplasm to the membrane (71–72). In part, a recently published article demonstrated that terbutaline increases the activity of ENaC in alveolar epithelial type II cells by increasing the traffic of ENaC subunits to the apical cell membrane under hypoxic conditions (73). B. Anesthetics In alveolar epithelial cells, the halogenated anesthetics affect sodium and fluid transport at the physiological level as well as on a cellular level. In the rat, halothane and isoflurane decrease fluid clearance by inhibition of the amiloride-sensitive component. This effect was rapidly reversible after cessation of halothane exposure (74). Unlike the rat, the

Resolution of alveolar edema

365

ability of the rabbit to clear fluid from the alveolar space through an amiloride-sensitive pathway is not decreased by halothane (75). In vitro, exposure to a low concentration of halothane (1%) for a short time (30 min) induced a reversible decrease in Na, K-ATPase activity and amiloride-sensitive 22Na influx in rat alveolar type II cells (76). The mechanisms whereby halothane induced a decrease in sodium transport protein activity have not been yet elucidated but are not related in vitro to a decrease in intracellular ATP content or to change in cytosolic free calcium. Taken together, these observations suggest that halogenated anesthetics may interfere with the clearance of alveolar edema. Lidocaine is widely used in patients with acute cardiac disorders and has also been recently implicated as a possible cause of pulmonary edema following liposuction. In experimental studies in rats, either intravenous or intra-alveolar lidocaine reduced fluid clearance in rats by 50% (77). Since lidocaine did not inhibit ENaC when expressed in oocytes, it seems that the inhibitory effect on vectorial fluid transport was primarily on the basal surface of alveolar epithelial cells, either through an effect on the activity of Na, K-ATPase or through an indirect effect through blockade of potassium channels, a wellknown property of lidocaine (78). The effect of lidocaine was completely reversible with β2-agonist therapy (77). C. Reactive Oxygen and Nitrogen Species Under several pathological conditions, in response to proinflammatory cytokines, activated neutrophils and macrophages can localize in the lung and migrate into the air spaces of the lung and release reactive oxygen species by the membrane-bound enzyme complex NADPH oxidase and nitric oxide (NO) via the calcium-insensitive iNOS form of NO synthase. NO decreased Isc across cultured rat type II cells without affecting transepithelial resistance. NO also inhibited 60% of amiloride-sensitive Isc across type II cell monolayers following permeabilization of the basolateral membrane with amphotericin B (79). NO reacted with superoxide (·O2) to form peroxynitrite (ONOO-), a potent oxidant and nitrating species that directly oxidizes a wide spectrum of biological molecules, such as DNA constituents, lipids, and proteins (5). Boluses of peroxynitrite (0.5–1 mM) into suspensions of freshly isolated type II cells from rabbits decreased amiloride-inhibitable sodium uptake to 68 and 56% of control values without affecting cell viability (80). Some investigators reported that products of macrophages, including NO, can downregulate sodium transport in fetal distal lung epithelium stimulated with endotoxin (81, 82). Also, another study indicated that a generator of peroxynitrite (3morpholinosydnonime) inhibited the amiloride-sensitive whole cell conductance in Xenopus oocytes expressing the three cloned subunits of ENaC (83). The data indicate that oxidation of critical amino acids residues in ENaC protein is probably responsible for this effect. This evidence matches well with other studies that have shown that protein nitration and oxidation by reactive oxygen and nitrogen species have been associated with diminished function of a variety of important proteins present in the alveolar space, including α1-proteinase inhibitor (84) and surfactant protein A (85–87). Indeed intratracheal instillation of DETANO-NOate, a substance that decomposed to release NO, in the alveolar spaces of rabbits resulted in significant downregulation of amiloridesensitive transport (88). On the other hand, endogenously produced NO may play an

Acute respiratory distress syndrome

366

important role in the regulation of sodium transporters in vivo as shown by the fact that iNOS(–/–) mice lack amiloride-sensitive AFC transport (89).

IV. Alveolar Fluid Transport Under Pathological Conditions Fluid clearance from the distal airspaces of the lung has been measured in mechanically ventilated patients with acute respiratory failure from pulmonary edema as well as in several animal models designed to simulate clinically relevant pathological conditions. A. Clinical Studies Studies of fluid clearance have been done in intubated, ventilated patients by measuring the concentration of total protein in sequential samples of undiluted pulmonary edema fluid aspirated from the distal air spaces of the lung with a standard suction catheter passed through the endotracheal tube into a wedged position in the distal airways of the lung (20, 53, 86, 90–92). This method for measuring fluid clearance in patients was adapted from the method for aspirating fluid from the distal air spaces of the lung in experimental studies in small and large animals (6, 9). The clinical procedure has been validated in patients by demonstrating that there is a relationship between fluid clearance and the improvement in oxygenation and the chest radiograph (53, 90). In patients with severe hydrostatic pulmonary edema, there was net fluid clearance in the majority of the patients during the first 4 hours following endotracheal intubation and the onset of positive pressure ventilation (90). The rate of fluid clearance in these patients varied between maximal (>14%/h) in 38% and submaximal (3–14%/h) in 37%. Overall, 75% of the patients had intact fluid clearance. There was no significant correlation between the levels of fluid clearance and endogenous plasma levels of epinephrine, although twice as many of the patients with intact fluid clearance received aerosolized βadrenergic therapy as those with impaired fluid clearance, but this difference did not reach statistical significance, perhaps because the total number of studied patients was modest. The inability to transport edema fluid from the distal air spaces of the lung in 25% of the patients was not simply related to elevated pulmonary vascular pressures. Several mechanisms could downregulate fluid transport in these patients including elevated levels of atrial natriuretic factor (93, 94) or the presence of ouabain-like substances in the circulation (95–97) or damage to Na transporters by reactive oxygennitrogen intermediates (see above). Experimental studies have provided some insight into the mechanisms that may downregulate fluid transport from the distal air spaces of the lung in the presence of elevated pulmonary vascular hydrostatic pressures (see below). Because hydrostatic pulmonary edema is associated with an uninjured epithelial barrier, the studies of hydrostatic pulmonary edema provide an important comparison group to the patients with pulmonary edema from acute lung injury because some degree of morphological or functional injury to the epithelial barriers probably occurs in most lung injuries (see below). The majority of patients with increased permeability edema and acute lung injury have impaired alveolar epithelial fluid transport, a finding associated with more prolonged respiratory failure and a higher mortality (Fig. 1). In contrast, a minority of patients can

Resolution of alveolar edema

367

remove alveolar edema fluid rapidly, and these patients have a higher survival rate (53, 91, 92). These results indicate that a functional, intact distal lung epithelium is associated with a better prognosis in patients with acute lung injury, thus supporting the hypothesis that the degree of injury to the distal lung epithelium is an important determinant of the outcome in patients with increased permeability pulmonary edema from acute lung injury. What are the mechanisms that may impair fluid clearance from the air spaces of the lung? Some patients have pathological (98) and biochemical (99) evidence of necrotic injury to the alveolar epithelium. There are also some clinical data that a decrease in fluid clearance may be associated

Figure 1 Hospital mortality (y-axis) plotted against two groups of patients with acute lung injury or the acute respiratory distress syndrome: those with maximal fluid clearance (>14%/h) and those with impaired or submaximal fluid clearance (<14%/h). The columns represent percent hospital mortality in each group. Hospital mortality of patients with maximal fluid clearance was significantly less (p<0.02). N=number of patients. (Data from Ref. 92.) with higher levels of nitrate and nitrite in pulmonary edema fluid, a finding that supports the hypothesis that nitration and oxidation of proteins essential to the epithelial fluid transport may occur in some patients with lung injury, depressing their ability to remove alveolar edema fluid (86) (see next section).

Acute respiratory distress syndrome

368

B. Experimental Studies of Hypovolemic Shock Hypovolemic shock from blood loss is an important clinical problem following major trauma. Short-term studies in rats that simulated acute hemorrhagic shock with a 30% loss of blood volume resulted in a sharp rise in endogenous levels of plasma epinephrine, a finding associated with a doubling of fluid clearance from the distal airspaces of the lung (100, 101). The effect was inhibited by propranolol and partially inhibited by amiloride. When hypovolemic shock, however, was prolonged for 4–5 hours in rats, the results were markedly different. Under these conditions, there was no increase in fluid transport from the air spaces of the lung, even when β2-adrenergic agonists were instilled into the distal air spaces. This result prompted a series of experiments to discover the mechanisms that down-regulate fluid clearance after prolonged hemorrhagic shock in rats. The initial studies established that the mechanism was neutrophil dependent (102). Further studies established that the process involves α-adrenergic activity and release of oxidant radicals with interleukin-1 β in the air spaces (103, 104), probably from neutrophils that accumulate in the lung after the onset of hemorrhagic shock. Finally, recent work has indicated that an increase in the expression of iNOS in the lung and release of NO, probably in part from alveolar macrophages, diminishes the capacity of the alveolar epithelium to actively transport fluid from the air spaces after severe hemorrhage. Second, NO inhibits the upregulation of alveolar epithelial fluid transport by cAMPdependent mechanisms by directly affecting the function of the β2-adrenergic receptor and adenylcyclase. Third, shock-mediated release of NO in the air spaces of the lung depends in part on the activation and nuclear translocation of NF-κB (105). However, another study found no difference in AFC between iNOS(+/+) and iNOS(−/−) mice exposed to hyperoxia for 55 hours (89). A potential explanation for this discrepancy is the existence of significant levels of oxidizing and nitrating species in the lungs of iNOS(−/−) mice exposed to oxidant stress (106) generated by the interaction of neutrophil myeloperoxidase with nitrite and hydrogen peroxide. The results of these experimental studies may have important clinical implications for explaining the susceptibility to pulmonary edema in some patients following major trauma. Since, as already discussed, in vitro studies demonstrated that peroxynitrate can directly impair the function of sodium channels (79, 80, 83, 107–109), these in vivo studies fit well with one recent clinical study that reported an inverse relationship between elevated levels of nitrate and nitrite and the rate of fluid clearance in patients with pulmonary edema (86). Also, one prior study identified nitrotyrosine (the stable byproduct of peroxynitrite reactions with tyrosine residues) in the lungs of patients with the acute respiratory distress syndrome (110) as well as in alveolar macrophages isolated from patients with the same cause of acute respiratory failure (111, 112). C. Experimental Studies of Infection The effects of endotoxemia and bacteremia on lung vascular permeability was well described in studies in sheep several years ago (113, 114). However, the impact on the function of the alveolar epithelial barrier was not addressed in those studies. More recent work has indicated that the acute shock produced by severe bacteremia in rats markedly increases plasma epinephrine levels, as in hemorrhagic shock, and the elevated

Resolution of alveolar edema

369

epinephrine levels markedly upregulate the fluid transport capacity of the distal lung epithelium (115). Thus, it is possible that in the short term, upregulation of fluid clearance may protect the air spaces against alveolar flooding when there is an increase in lung vascular permeability and accumulation of interstitial edema. In fact, one study in sheep demonstrated that lung vascular permeability can be augmented markedly with intravenous endotoxin with a rise in protein-rich lung lymph flow, but this effect was not associated with a change in lung epithelial permeability to protein and no change in the capacity of the alveolar epithelium to remove alveolar fluid (116). These studies were done over 4 and 24 hours in sheep, and in some of the studies both intra-alveolar and intravenous endotoxin was administered, but in all cases the epithelial barrier remained intact and capable of transporting alveolar fluid normally. However, when large doses of live bacteria (Pseudomonas aeruginosa) were given to sheep, there was an increase in both lung endothelial and epithelial permeability to protein in the sheep that had developed the most severe shock (117). These sheep had alveolar flooding and their capacity to remove alveolar fluid was impaired, similar to the findings in humans who develop severe permeability pulmonary edema with septic shock (118). The mechanisms for injury to the epithelial barrier probably depend on both neutrophil-dependent release of injurious proteases and reactive oxygen species as well as the bacterial exoproducts (see below). In one study, gram-negative bacteria that produced proteases increased alveolar epithelial barrier permeability to protein by altering basolateral surface permeability while the non-protease-producing strains only increased lung vascular permeability (119). It should be noted that the contribution of neutophils to alveolar epithelial and endothelial injury depends on the originating insult: for example, neutopenic rabbits exposed to hyperoxia have similar levels of AFC as neutrophil-replete ones (120). In sharp contrast to intra-alveolar endotoxin, live bacteria increased alveolar epithelial barrier permeability and decreased fluid transport in sheep (116). Further studies indicated that the products of P. aeruginosa were important in determining the extent of injury. For example, exoenzyme S and phospholipase C mediated injury to the epithelial barrier in rabbits with a decrease in vectorial fluid transport. Subsequent studies indicated that bacterial pneumonia may progress to septic shock when the infecting gram-negative organism generates pro-inflammatory cytokines in the air spaces of the lung that are released into the circulation when bacterial-mediated injury results in sufficient injury to the distal lung epithelial barrier (121). Several experimental studies have indicated that active and passive immunization against P. aeruginosa antigens can prevent epithelial injury in sheep (122) and in mice (123). Recent data also indicate that influenza virus infection (A/PR/8/34) can specifically alter epithelial ion transport by inhibiting amiloride-sensitive sodium current across mouse tracheal epithelium (124). The inhibitory effect of the influenza virus was caused by binding the viral hemagglutinin to a cell-surface receptor, which then activated phospholipase C and protein kinase C. It is well known that PKC can reduce ENaC activity so that influenza infections in the lung may inhibit the function of ENaC (125). Given the importance of sodium channels in vectorial transport of fluid in distal airway epithelia and in the alveoli, these results provide a new mechanism that may explain the accumulation of alveolar edema fluid in patients with viral pneumonia and acute lung injury.

Acute respiratory distress syndrome

370

Interestingly, in the process of studying P. aeruginosa pneumonia in rats, one group of investigators found that the rate of fluid clearance was upregulated in the rats that survived (126). The mechanism for this effect was secondary to release of TNF-α, which was surprising since TNF-α plays an important role in mediating the host inflammatory response to infection (127–129) as well as potentially contributing to the pathogenesis of septic shock (130). It had been previously reported that TNF-α could increase sodium coupled amino acid transport across hepatocytes (131). The capacity of TNF-α to increase fluid clearance was confirmed in a subsequent rat study (132) in which a neutralizing polyclonal anti-TNF-α antibody inhibited the upregulation of fluid clearance induced by intestinal ischemia-reperfusion. Also, the effect of TNF-α is amiloride inhibitable in both rats (132, 133) and isolated A549 cell patch clamp studies (133). Based on the A549 human cell studies, the effect appears to be in part a receptormediated process. It is not clear, however, what signaling pathways are involved since cAMP levels are not elevated by TNF-α (132). D. Experimental Studies of Hyperoxia Several investigators have used hyperoxia as a model to study the effect of acute lung injury on epithelial ion and fluid transport in the lung, in part because the injury develops over 2–5 days in rats and mice, and pathologically resembles clinical acute lung injury with both endothelial and epithelial injury in association with an influx of neutrophils and protein rich pulmonary edema. However, the results of these studies have not been uniform, in part because of variations in the duration of O2 exposure, the exact level of hyperoxia, and the use of rats or mice. For example, exposure of rats or rabbits to 100% O2 at one atmosphere for 3–4 days results in extensive injury to both the endothelial and alveolar epithelial barriers (134–136). On the other hand, exposure of these animals to<85% O2 results in sublethal injury and development of resistance to a subsequent exposure to hyperoxia or any oxidant stress (137, 138). Administration of 85% O2 for 7 days increased the level of αENaC protein, and both inward and outward sodium currents were stimulated in patch clamps of isolated alveolar type II cells (139). Subsequently, another study from the same group showed increased expression and activity of amiloride-inhibitable sodium channels in alveolar type II cells of rats exposed to 85% O2 for 7 days followed by 100% O2 for 4 days. Both the number and the open probability of the L-type sodium channels (25 pS) were increased (25). Another group also studied the effect of 85% O2 for 7 days in rats and found that amiloride-inhibitable sodium uptake was greater than in control rats and that ouabain decreased active sodium transport to a greater percentage in the hyperoxic rats, suggesting an upregulation of Na, K-ATPase activity after subacute hyperoxia (43, 140– 142). In other studies, exposure to 100% O2 for 48 hours produced moderate interstitial lung edema but no impairment of basal or cAMP-stimu-lated fluid transport (143–145). However, when exposure was prolonged to 64 hours, one group of investigators reported decreased transport in rats, an effect that seemed to be related to decreased gene expression of α1 Na, K-ATPase subunits (146). Other studies found that there was rapid upregulation of mRNA for α1 and β1 subunits of Na, K-ATPase as well as antigenic protein shortly after prolonged exposure to >97% O2 in rats for 60 hours (147), suggesting the induction of the Na, K-ATPase could occur as a protective mechanism. In

Resolution of alveolar edema

371

the same vein, intratracheal administration of phenamil, an irreversible blocker of sodium channels, in rats exposed to hyperoxia worsened pulmonary edema as compared to shar instilled animals, indicating that sodium reabsorption plays an important role in limiting edema in oxidant injury (148). Similar findings were reported by Stern et al. (149) in a thiourea model of lung injury in which upregulation of Na, K-ATPase gene expression and protein occurred after initial injury and was associated with recovery from the pulmonary edema. Other studies have reported that hyperoxia does not seem to produce a clear change in sodium and fluid transport during the period of hyperoxia (150–152). Interestingly, one group of investigators reported that pretreatment of rats with aerosolized adenoviral β1 Na, K-ATPase upregulated fluid transport and also made the rats resistant to the lethal effects of hyperoxia (142). This represented the first evidence that gene therapy could potentially be used to produce a sustained upregulation of fluid transport in the lung in the presence of a pathological condition. Furthermore, gene transfer of both α1 and β1 Na, K-ATPase decreases edema formation induced by thiourea in rats (149). An earlier study had shown that administration of the adenoviral β1 Na, KATPase gene, but not the α1 Na, K-ATPase, would increase fluid transport in normal rats (153).

V. Conclusions Several advances have been made in understanding of the reabsorption of edema fluid and the distal epithelia with characterization of sodium and chloride transport and water pathways under both physiological and pathological conditions. However, as discussed in a recent review (1), several fundamental issues require additional study. Alveolar type I and type II cells and distal airway epithelial cells, such as Clara cells, are implicated in sodium and fluid transport but their contribution in both physiological and pathological conditions are not well defined. Innovative approaches are needed to determine the contribution of these cells. Alveolar type I cells cover 95% of the alveolar surface area and the recent demonstration of the presence of water channels and ENaC expression in those cells suggest a role for these cells in net fluid clearance. More studies are needed to assess the differential contribution of alveolar type I and II cells to fluid transport. Another important area of research is the characterization of the sodium transporters involved in sodium and fluid reabsorption and their regulation. Amiloride-sensitive sodium transport is one of the major pathways for sodium entry across distal epithelial cells but several questions remain unsolved. For example, are the molecular and biophysical characteristics of these channels in vitro representative of their in vivo characteristics, and how are these channels regulated during physiological and pathological conditions? The mechanisms that regulate the trafficking of ENaC and Na, K-ATPases between the cytoplasm and the membrane need to be evaluated also in distal lung epithelia. Increased insertion of transport proteins seems to be an important mechanism for increasing sodium and fluid transport under pathological conditions and may potentially contribute to regulating the clearance of edema fluid from distal air spaces of the lung. In addition to amiloride-sensitive sodium transport, a characterization of ion transporters involved in amiloride-insensitive sodium transport needs to be

Acute respiratory distress syndrome

372

defined. Also, the pathways for chloride reabsorption under basal and stimulated conditions need to be determined with particular attention to the role of CFTR under cAMP-stimulated conditions. As discussed in a recent review (1), recent advances have been made with transgenic mice models to define the role of sodium and water channels in the lung fluid balance. Knock-out of the three subunits of ENaC clearly established the preponderant role of a compared to β and γENaC in alveolar transepithelial sodium absorption. Similarly, the knock-out mice for several aquaporin-type water channels have revealed that in the lung these channels are not essential for water transport. However, a genomic disruption of genes that are expressed during development or in multiple tissue types complicates the phenotypic analysis. A solution to this problem may be provided by conditional knockouts. This system permits control of the timing for cell-specific expression of specific proteins, thereby circumventing both embryonic lethality and confounding effects of complex adaptive responses that can occur when the physiological observations follow the gene knock-out events by days or weeks. In this system, gene expression is regulated temporally and spatially using cell-specific promoters, such as SP-C for alveolar type II cells, in combination with a regulatory on-off system. This approach may provide an opportunity to advance the understanding of the role of sodium and fluid transport during pathological conditions during the reabsorption of edema from the distal air spaces of the lung. The rate of edema reabsorption from the distal air spaces of the lung can be estimated in ventilated, critically ill patients with acute pulmonary edema. In conjunction with progress in experimental studies of lung fluid balance under clinically relevant pathological conditions, further studies should be done to test the potential role of catecholamine-dependent and independent therapies that might enhance the resolution of clinical pulmonary edema, both hydrostatic pulmonary edema and acute lung injury edema. Some of the therapies that might be tested for treatment of clinical lung injury, such as beta-2 agonist treatment, might directly improve lung epithelial vectorial ion transport and therefore fluid clearance.

References 1. Matthay MA, Folkesson HG, Clerici C. Lung epithelial fluid transport and the resolution of pulmonary edema. Physiol Rev 2002; 82:569–600. 2. Barker PM, Markiewicz M, Parker KA, Walters DV, Strang LB. Synergistic action of triiodothyronine and hydrocortisone on epinephrine-induced reabsorption of fetal lung liquid. Pediatr Res 1990; 27:588–591. 3. Bland RD. Loss of liquid from the lung lumen in labor: more than a simple “squeeze”. Am J Physiol Lung Cell Mol Physiol 2001; 280:L602–L605. 4. Walters DV, Olver RE. The role of catecholamines in lung liquid absorption at birth. Pediatr Res 1978; 12:239–242. 5. Matalon S, O’Brodovich H. Sodium channels in alveolar epithelial cells: molecular characterization, biophysical properties, and physiological significance. Annu Rev Physiol 1999; 61:627–661. 6. Matthay MA, Folkesson HG, Verkman AS. Salt and water transport across alveolar and distal airway epithelium in the adult lung. Am J Physiol 1996; 270:L487–L503.

Resolution of alveolar edema

373

7. Basset G, Crone C, Saumon G. Fluid absorption by rat lung in situ: pathways for sodium entry in the luminal membrane of alveolar epithelium. J Physiol (London) 1987; 384:325–345. 8. Berthiaume Y, Broaddus VC, Gropper MA, Tanita T, Matthay MA. Alveolar liquid and protein clearance from normal dog lungs. J Appl Physiol 1988; 65: 585–593. 9. Berthiaume Y, Staub NC, Matthay MA. Beta-adrenergic agonists increase lung liquid clearance in anesthetized sheep. J Clin Invest 1987; 79:335–343. 10. Effros RM, Mason GR, Hukkanen J, Silverman P. New evidence for active sodium transport from fluid-filled rat lungs. J Appl Physiol 1989; 66:906–919. 11. Finley N, Norlin A, Baines DL, Folkesson HG. Alveolar epithelial fluid clearance is mediated by endogenous catecholamines at birth in guinea pigs. J Clin Invest 1998; 101:972–981. 12. Norlin A, Finley N, Abedinpour P, Folkesson HG. Alveolar liquid clearance in the anesthetized ventilated guinea pig. Am J Physiol 1998; 274:L235– L243. 13. O’Brodovich H, Hannam V, Seear M, Mullen JB. Amiloride impairs lung water clearance in newborn guinea pigs. J Appl Physiol 1990; 68:1758–1762. 14. Sakuma T, Okaniwa G, Nakada T, Nishimura T, Fujimura S, Matthay MA. Alveolar fluid clearance in the resected human lung. Am J Respir Crit Care Med 1994; 150:305–310. 15. Smedira N, Gates L, Hastings R, Jayr C, Sakuma T, Pittet J-F, Matthay MA. Alveolar and lung liquid clearance in anesthetized rabbits. J Appl Physiol 1991; 70:1827–1835. 16. Weibel ER. Lung morphometry and models in respiratory physiology. In: Chang K, Paiva M, eds. Respiratory Physiology. An Analytical Approach. New York: Marcel Dekker, 1989:1–56. 17. Staub NC, Albertine KH. The structure of the lungs relative to their principal function. In: Murray F, Nadel JA, eds. Textbook of Respiratory Medicine. Philadelphia: W.B. Saunders Company, 1988:12–36. 18. West JB, Mathieu-Costello O. Strength of the pulmonary blood-gas barrier. Respir Physiol 1992; 88:141–148. 19. Schneeberger EE, Karnovsky MJ. The influence of intravascular fluid volume on the permeability of newborn and adult mouse lungs to ultrastructural protein tracers. J Cell Biol 1971; 49:310–334. 20. Hastings RH, Grady M, Sakuma T, Matthay MA. Clearance of differentsized proteins from the alveolar space in humans and rabbits. J Appl Physiol 1992; 73:1310–1316. 21. Hastings RH, Wright JR, Albertine KH, Ciriales R, Matthay MA. Effect of endocytosis inhibitors on alveolar clearance of albumin, immunoglobulin G, and SP-A in rabbits. Am J Physiol 1994; 266:L544–L552. 22. Berthiaume Y, Albertine KH, Grady M, Fick G, Matthay MA. Protein clearance from the air spaces and lungs of unanesthetized sheep over 144 h. J Appl Physiol 1989; 67:1887–1897. 23. Canessa CM, Schild L, Buell G, Thorens B, Gautschi I, Horisberger JD, Rossier BC. Amiloride-sensitive epithelial Na+ channel is made of three homologous subunits [see comments]. Nature 1994; 367:463–467. 24. O’Brodovich H, Canessa C, Ueda J, Rafii B, Rossier BC, Edelson J. Expression of the epithelial Na+ channel in the developing rat lung. Am J Physiol 1993; 264:C491–C496. 25. Yue G, Russell WJ, Benos DJ, Jackson RM, Olman MA, Matalon S. Increased expression and activity of sodium channels in alveolar type II cells of hyperoxic rats. Proc Natl Acad Sci USA 1995; 92:8418–8422. 26. Dobbs LG, Gonzalez R, Matthay MA, Carter EP, Allen L, Verkman AS. Highly water permeable type I alveolar epithelial cells confer high water permeability between the airspace and vasculature in rat lung. Proc Natl Acad Sci USA 1998; 95:2991–2996. 27. Schneeberger EE, McCarthy KM. Cytochemical localization of Na+-K+-ATPase in rat type II pneumocytes. J Appl Physiol 1986; 60:1584–1589. 28. Ridge KM, Rutschman DH, Factor P, Katz AI, Bertorello AM, Sznajder JL. Differential expression of Na-K-ATPase isoforms in rat alveolar epithelial cells. Am J Physiol 1997; 273:L246–L255.

Acute respiratory distress syndrome

374

29. Borok Z, Liebler JM, Lubman RL, Foster MJ, Zhou B, Li X, Zabski SM, Kim KJ, Crandall ED. Na transport proteins are expressed by rat alveolar epithelial type I cells. Am J Physiol Lung Cell Mol Physiol 2002; 282:L599– L608. 30. Johnson MD, Widdicombe JH, Allen L, Barbry P, Dobbs LG. Alveolar epithelial type I cells contain transport proteins and transport sodium, supporting an active role for type I cells in regulation of lung liquid homeostasis. Proc Natl Acad Sci USA 2002; 99:1966–1971. 31. Inglis SK, Corboz MR, Taylor AE, Ballard ST. Regulation of ion transport across porcine distal bronchi. Am J Physiol 1996; 270:L289–L297. 32. Van Scott MR, Chinet TC, Burnette AD, Paradise AM. Purinergic regulation of ion transport across nonciliated bronchiolar epithelial (Clara) cells. Am J Physiol 1995; 269:L30–L37. 33. Fang XH, Fukuda N, Barbry P, Sartori C, Verkman AS, Matthay MA. Novel role for CFTR in fluid absorption from the distal airspaces of the lung. J Gen Physiol 2002; 119:199–208. 34. Lazrak A, Thome U, Myles C, Ware J, Chen L, Venglarik CJ, Matalon S. cAMP regulation of Cl(−) and HCO(−)(3) secretion across rat fetal distal lung epithelial cells. Am J Physiol Lung Cell Mol Physiol 2002; 282:L650–L658. 35. Saumon G, Basset G. Electrolyte and fluid transport across the mature alveolar epithelium. J Appl Physiol 1993; 74:1–15. 36. Matthay MA, Landolt CC, Staub NC. Differential liquid and protein clearance from the alveoli of anesthetized sheep. J Appl Physiol 1982; 53:96–104. 37. Matthay MA, Berthiaume Y, Staub NC. Long-term clearance of liquid and protein from the lungs of unanesthetized sheep. J Appl Physiol 1985; 59:928–934. 38. Sakuma T, Pittet JF, Jayr C, Matthay MA. Alveolar liquid and protein clearance in the absence of blood flow or ventilation in sheep. J Appl Physiol 1993; 74:176–185. 39. Serikov VB, Grady M, Matthay MA. Effect of temperature on alveolar liquid and protein clearance in an in situ perfused goat lung. J Appl Physiol 1993; 75:940–947. 40. Rutschman DH, Olivera W, Sznajder JI. Active transport and passive liquid movement in isolated perfused rat lungs. J Appl Physiol 1993; 75:1574–1580. 41. Ballard ST, Taylor AE. Bioelectric properties of proximal bronchiolar epithelium. Am J Physiol 1994; 267:L79–L84. 42. Basset G, Crone C, Saumon G. Significance of active ion transport in transalveolar water absorption: a study on isolated rat lung. J Physiol (London) 1987; 384:311–324. 43. Olivera W, Ridge K, Wood LD, Sznajder JI. Active sodium transport and alveolar epithelial Na-K-ATPase increase during subacute hyperoxia in rats. Am J Physiol 1994; 266:L577–L584. 44. Goodman BE, Crandall ED. Dome formation in primary cultured monolayers of alveolar epithelial cells. Am J Physiol 1982; 243:C96–C100. 45. Mason RJ, Williams MC, Widdicombe JH, Sanders MJ, Misfeldt DS, Berry LCJ. Transepithelial transport by pulmonary alveolar type II cells in primary culture. Proc Natl Acad Sci USA 1982; 79:6033–6037. 46. Goodman BE, Fleischer RD, Crandall ED. Evidence for active Na+ transport by cultured monolayers of pulmonary alveolar epithelial cells. Am J Physiol 1983; 245:C78–C83. 47. Jiang X, Ingbar DH, O’Grady SM. Adrenergic stimulation of Na+ transport across alveolar epithelial cells involves activation of apical Cl− channels. Am J Physiol 1998; 275:C1610– C1620. 48. Jiang X, Ingbar DH, O’Grady SM. Adrenergic regulation of ion transport across adult alveolar epithelial cells: effects on Cl− channel activation and transport function in cultures with an apical air interface. J Membr Biol 2001; 181:195–204. 49. O’Grady SM, Jiang X, Ingbar DH. Cl-channel activation is necessary for stimulation of Na transport in adult alveolar epithelial cells. Am J Physiol Lung Cell Mol Physiol 2000; 278:L239–L244. 50. Nielsen VG, DuVall MD, Baird MS, Matalon S. cAMP activation of chloride and fluid secretion across the rabbit alveolar epithelium. Am J Physiol 1998; 275:L1127–L1133.

Resolution of alveolar edema

375

51. Brown MJ, Olver RE, Ramsden CA, Strang LB, Walters DV. Effects of adrenaline and of spontaneous labour on the secretion and absorption of lung liquid in the fetal lamb. J Physiol (London) 1983; 344:137–152. 52. Sakuma T, Folkesson HG, Suzuki S, Okaniwa G, Fujimura S, Matthay MA. Beta-adrenergic agonist stimulated alveolar fluid clearance in ex vivo human and rat lungs. Am J Respir Crit Care Med 1997; 155:506–512. 53. Matthay MA, Wiener-Kronish JP. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am Rev Respir Dis 1990; 142:1250–1257. 54. Kim KJ, Cheek JM, Crandall ED. Contribution of active Na+ and Cl− fluxes to net ion transport by alveolar epithelium. Respir Physiol 1991; 85:245–256. 55. Lazrak A, Nielsen VG, Matalon S. Mechanisms of increased Na+ transport in ATII cells by cAMP: we agree to disagree and do more experiments. Am J Physiol Lung Cell Mol Physiol 2000; 278:L233–L238. 56. Widdicombe J. How does cAMP increase active Na absorption across alveolar epithelium? [editorial comment]. Am J Physiol (Lung Cell Mol Physiol) 2000; 278:L231–L232. 57. Widdicombe JH. Yet another role for the cystic fibrosis transmembrane conductance regulator. Am J Respir Cell Mol Biol 2000; 22:11–14. 58. Al-Bazzaz FJ. Regulation of Na and Cl transport in sheep distal airways. Am J Physiol 1994; 267:L193–L198. 59. Ballard ST, Schepens SM, Falcone JC, Meininger GA, Taylor AE. Regional bioelectric properties of porcine airway epithelium. J Appl Physiol 1992; 73: 2021–2027. 60. Farman N, Talbot CR, Boucher R, Fay M, Canessa C, Rossier B, Bonvalet JP. Noncoordinated expression of a-, β-, and γ-subunit mRNAs of epithelial Na+ channel along rat respiratory tract. Am J Physiol 1997; 272:C131–C141. 61. Gaillard D, Hinnrasky J, Coscoy S, Hofman P, Matthay MA, Puchelle E, Barbry P. Early expression of β- and γ-subunits of epithelial sodium channel during human airway development. Am J Physiol Lung Cell Mol Physiol 2000; 278:L177–L184. 62. Kreda SM, Gynn MC, Fenstermacher DA, Boucher RC, Gabriel SE. Expression and localization of epithelial aquaporins in the adult human lung. Am J Respir Cell Mol Biol 2001; 24:224–234. 63. Pitkanen OM, Smith D, O’Brodovich H, Otulakowski G. Expression of alpha-, beta-, and gamma-hENaC mRNA in the human nasal, bronchial, and distal lung epithelium. Am J Respir Crit Care Med 2001; 163:273–276. 64. Renard S, Voilley N, Bassilana F, Lazdunski M, Barbry P. Localization and regulation by steroids of the alpha, beta and gamma subunits of the amiloride-sensitive Na+ channel in colon, lung and kidney. Pflügers Arch 1995; 430:299–307. 65. Stutts MJ, Canessa CM, Olsen JC, Hamrick M, Cohn JA, Rossier BC, Boucher RC. CFTR as a cAMP-dependent regulator of sodium channels [see comments]. Science 1995; 269:847–850. 66. Reddy MM, Light MJ, Quinton PM. Activation of the epithelial Na+ channel (ENaC) requires CFTR Cl− channel function. Nature 1999; 402:301–304. 67. Engelhardt JF, Zepeda M, Cohn J, Yankaskas J, Wilson JM. Expression of the cystic fibrosis gene in adult human lung. J Clin Invest 1994; 93:737–749. 68. Lei J, Mariash C, Ingbar D. T3 stimulation of alveolar epithelial Na, K-ATPase (abstr). FASEB J 2001; 15:A860. 69. Suzuki S, Noda M, Sugita M, Ono S, Koike K, Fujimura S. Impairment of transalveolar fluid transport and lung Na+-K+-ATPase function by hypoxia in rats. J Appl Physiol 1999; 87:962– 968. 70. Vivona ML, Matthay MA, Chabaud M, Friedlander G, Clerici C. Hypoxia reduces alveolar epithelial sodium and fluid transport in rats: reversal by beta-2 adrenergic agonist treatment. Am J Respir Cell Mol Biol 2001; 25:554–561. 71. Butterworth MB, Helman SI, Els WJ. cAMP-sensitive endocytic trafficking in A6 epithelia. Am J Physiol Cell Physiol 2001; 280:C752–C762.

Acute respiratory distress syndrome

376

72. Snyder PM. Liddle’s syndrome mutations disrupt cAMP-mediated trans-location of the epithelial Na+ channel to the cell surface. J Clin Invest 2000; 105:45–53. 73. Planes C, Blor-Chabaud M, Matthay MA, Couette S, Uchida T, Clerici C. Hypoxia and beta-2 agonists regulate cell surface expression of the epithelial sodium channel in native alveolar epithelial cells. J Biol Chem 2002; 277: 47318–47324. 74. Rezaiguia-Delclaux S, Jayr C, Luo DF, Saidi N-E, Meignan M, Duvaldestin P. Halothane and isoflurane decrease alveolar epithelial fluid clearance in rats. Anesthesiology 1998; 88:751–760. 75. Nielsen VG, Baird MS, Geary BT, Matalon S. Halothane does not decrease amiloride-sensitive alveolar fluid clearance in rabbits. Anesth Analg 2000; 90:1445–1449. 76. Molliex S, Crestani B, Dureuil B, Bastin J, Rolland C, Aubier M, Desmonts JM. Effects of halothane on surfactant biosynthesis by rat alveolar type II cells in primary culture. Anesthesiology 1994; 81:668–676. 77. Laffon M, Jayr C, Barbry P, Wang Y, Folkesson HG, Pittet JF, Clerici C, Matthay MA. Lidocaine induces a reversible decrease in alveolar epithelial fluid clearance in rats. Anesthesiology 2002; 96:392–399. 78. Butterworth JFt, Strichartz GR. Molecular mechanisms of local anesthesia: a review. Anesthesiology 1990; 72:711–734. 79. Guo Y, DuVall MD, Crow JP, Matalon S. Nitric oxide inhibits Na+ absorption across cultured alveolar type II monolayers. Am J Physiol 1998; 274:L369– L377. 80. Hu P, Ischiropoulos H, Beckman JS, Matalon S. Peroxynitrite inhibition of oxygen consumption and sodium transport in alveolar type II cells. Am J Physiol 1994; 266:L628– L634. 81. Compeau CG, Rotstein OD, Tohda H, Marunaka Y, Rafii B, Slutsky AS, O’Brodovich H. Endotoxin-stimulated alveolar macrophages impair lung epithelial Na+ transport by an L-Argdependent mechanism. Am J Physiol 1994; 266:C1330–C1341. 82. Ding JW, Dickie J, O’Brodovich H, Shintani Y, Rafii B, Hackam D, Marunaka Y, Rotstein OD. Inhibition of amiloride-sensitive sodium-channel activity in distal lung epithelial cells by nitric oxide. Am J Physiol 1998; 274:L378–L387. 83. DuVall MD, Zhu S, Fuller CM, Matalon S. Peroxynitrite inhibits amiloride-sensitive Na+ currents in Xenopus oocytes expressing αβγ-rENaC. Am J Physiol 1998; 274:C1417–C1423. 84. Moreno JJ, Pryor WA. Inactivation of alpha 1-proteinase inhibitor by peroxynitrite. Chem Res Toxicol 1992; 5:425–431. 85. Zhu S, Basiouny KF, Crow JP, Matalon S. Carbon dioxide enhances nitration of surfactant protein A by activated alveolar macrophages. Am J Physiol Lung Cell Mol Physiol 2000; 278:L1025–L1031. 86. Zhu S, Ware LB, Geiser T, Matthay MA, Matalon S. Increased levels of nitrate and surfactant protein a nitration in the pulmonary edema fluid of patients with acute lung injury. Am J Respir Crit Care Med 2001; 163:166–172. 87. Zhu S, Kachel DL, Martin WJ 2nd, Matalon S. Nitrated SP-A does not enhance adherence of Pneumocystis carinii to alveolar macrophages. Am J Physiol 1998; 275:L1031–L1039. 88. Nielsen VG, Baird MS, Chen L, Matalon S. DETANONOate, a nitric oxide donor, decreases amiloride-sensitive alveolar fluid clearance in rabbits. Am J Respir Crit Care Med 2000; 161:1154–1160. 89. Hardiman KM, Lindsey JR, Matalon S. Lack of amiloride-sensitive transport across alveolar and respiratory epithelium of iNOS(–/–) mice in vivo. Am J Physiol Lung Cell Mol Physiol 2001; 281:L722–L731. 90. Verghese GM, Ware LB, Matthay BA, Matthay MA. Alveolar epithelial fluid transport and the resolution of clinically severe hydrostatic pulmonary edema. J Appl Physiol 1999; 87:1301– 1312. 91. Ware LB, Golden JA, Finkbeiner WE, Matthay MA. Alveolar epithelial fluid transport capacity in reperfusion lung injury after lung transplantation. Am J Respir Crit Care Med 1999; 159:980– 988.

Resolution of alveolar edema

377

92. Ware LB, Matthay MA. Alveolar fluid clearance is impaired in the majority of patients with acute lung injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:1376–1383. 93. Campbell AR, Folkesson HG, Berthiaume Y, Gutkowska J, Suzuki S, Matthay MA. Alveolar fluid clearance persists in the presence of moderate left atrial hypertension in sheep. J Appl Physiol 1999; 86:139–151. 94. Olivera W, Ridge K, Wood LD, Sznajder JI. ANF decreases active sodium transport and increases alveolar epithelial permeability in rats. J Appl Physiol 1993; 75:1581–1586. 95. De Angelis C, Riscazzi M, Salvini R, Piccoli A, Ferri C, Santucci A. Isolation and characterization of a digoxin-like immunoreactive substance from human urine by affinity chromatography. Clin Chem 1997; 43:1416–1420. 96. Ferrandi M, Manunta P, Balzan S, Hamlyn JM, Bianchi G, Ferrari P. Ouabain-like factor quantification in mammalian tissues and plasma: comparison of two independent assays. Hypertension 1997; 30:886–896. 97. Ferri C, Bellini C, Coassin S, De Angelis C, Perrone A, Santucci A. Plasma endogenous digoxin-like substance levels are dependent on blood O2 in man. Clin Sci (Lond) 1994; 87:447– 451. 98. Bachofen M, Weibel ER. Alterations of the gas exchange apparatus in adult respiratory insufficiency associated with septicemia. Am Rev Respir Dis 1977; 116:589–615. 99. Newman V, Gonzalez RF, Matthay MA, Dobbs LG. A novel alveolar type I cell-specific biochemical marker of human acute lung injury. Am J Respir Crit Care Med 2000; 161:990– 995. 100. Modelska K, Matthay MA, McElroy MC, Pittet J-F. Upregulation of alveolar liquid clearance after fluid resuscitation for hemorrhagic shock in rats. Am J Physiol 1997; 273:L305–L314. 101. Pittet J-F, Brenner TJ, Modelska K, Matthay MA. Alveolar liquid clearance is increased by endogenous catecholamines in hemorrhagic shock in rats. J Appl Physiol 1996; 81:830–837. 102. Modelska K, Matthay MA, Brown LAS, Deutch E, Lu LN, Pittet JF. Inhibition of βadrenergic-dependent alveolar epithelial clearance by oxidant mechanisms after hemorrhagic shock. Am J Physiol 1999; 276:L844–L857. 103. Laffon M, Lu LN, Modelska K, Matthay MA, Pittet JF. α-Adrenergic blockade restores normal fluid transport capacity of alveolar epithelium after hemorrhagic shock. Am J Physiol 1999; 277:L760–L768. 104. Le Tulzo Y, Shenkar R, Kaneko D, Moine P, Fantuzzi G, Dinarello CA, Abraham E. Hemorrhage increases cytokine expression in lung mononuclear cells in mice: involvement of catecholamines in nuclear factor-kappaB regulation and cytokine expression. J Clin Invest 1997; 99:1516–1524. 105. Pittet J-F, Lu LN, Morris DG, Matthay MA. Reactive nitrogen species inhibit alveolar epithelial fluid transport after hemorrhagic shock in rats. J Immunol 2001; 166:6301–6310. 106. Hickman-Davis JM, Lindsey JR, Matalon S. Cyclophosphamide decreases nitrotyrosine formation and inhibits nitric oxide production by alveolar macrophages in mycoplasmosis. Infect Immun 2001; 69:6401–6410. 107. Bauer ML, Beckman JS, Bridges RJ, Fuller CM, Matalon S. Peroxynitrite inhibits sodium uptake in rat colonic membrane vesicles. Biochim Biophys Acta 1992; 1104:87–94. 108. Haddad IY, Ischiropoulos H, Holm BA, Beckman JS, Baker JR, Matalon S. Mechanisms of peroxynitrite-induced injury to pulmonary surfactants. Am J Physiol 1993; 265:L555–L564. 109. Matalon S, Hu P, Ischiropoulos H, Beckman JS. Peroxynitrite inhibition of oxygen consumption and ion transport in alveolar type II pneumocytes. Chest 1994; 105:74S. 110. Haddad IY, Pataki G, Hu P, Galliani C, Beckman JS, Matalon S. Quantitation of nitrotyrosine levels in lung sections of patients and animals with acute lung injury. J Clin Invest 1994; 94:2407–2413. 111. Baldus S, Castro L, Eiserich JP, Freeman BA. Is *NO news bad news in acute respiratory distress syndrome? Am J Respir Crit Care Med 2001; 163:308–310.

Acute respiratory distress syndrome

378

112. Sittipunt C, Steinberg KP, Ruzinski JT, Myles C, Zhu S, Goodman RB, Hudson LD, Matalon S, Martin TR. Nitric oxide and nitrotyrosine in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:503–510. 113. Brigham KL, Meyrick B. Endotoxin and lung injury. State of art. Am Rev Respir Dis 1986; 133:913–927. 114. Brigham KL, Woolverton WC, Blake LH, Staub NC. Increased sheep lung vascular permeability caused by Pseudomonas bacteremia. J Clin Invest 1974; 54:792–804. 115. Pittet J-F, Wiener-Kronish JP, McElroy MC, Folkesson HG, Matthay MA. Stimulation of lung epithelial liquid clearance by endogenous release of catecholamines in septic shock in anesthetized rats. J Clin Invest 1994; 94:663–671. 116. Wiener-Kronish JP, Albertine KH, Matthay MA. Differential responses of the endothelial and epithelial barriers of the lung in sheep to Escherichia coli endotoxin. J Clin Invest 1991; 88:864–875. 117. Pittet J-F, Wiener-Kronish JP, Serikov V, Matthay MA. Resistance of the alveolar epithelium to injury from septic shock in sheep. Am J Respir Crit Care Med 1995; 151:1093–1100. 118. Rubin DB, Wiener-Kronish JP, Murray JF, Green DR, Turner J, Luce JM, Montgomery AB, Marks JD, Matthay MA. Elevated von Willebrand factor antigen is an early plasma predictor of acute lung injury in nonpulmonary sepsis syndrome. J Clin Invest 1990; 86:474–480. 119. Pittet J-F, Kudoh I, Wiener-Kronish JP. Endothelial exposure to Pseudomonas aeruginosa proteases increases the vulnerability of the alveolar epithelium to a second injury. Am J Respir Cell Mol Biol 1998; 18:129–135. 120. Laughlin MJ, Wild L, Nickerson PA, Matalon S. Effects of hyperoxia on alveolar permeability of neutropenic rabbits. J Appl Physiol 1986; 61:1126–1131. 121. Kurahashi K, Kajikawa O, Sawa T, Ohara M, Gropper MA, Frank DW, Martin TR, WienerKronish JP. Pathogenesis of septic shock in Pseudomonas aeruginosa pneumonia. J Clin Invest 1999; 104:743–750. 122. Pittet J-F, Matthay MA, Pier G, Grady M, Wiener-Kronish JP. Pseudomonas aeruginosainduced lung and pleural injury in sheep. Differential protective effect of circulating versus alveolar immunoglobulin G antibody. J Clin Invest 1993; 92:1221–1228. 123. Sawa T, Yahr TL, Ohara M, Kurahashi K, Gropper MA, Wiener-Kronish JP, Frank DW. Active and passive immunization with the Pseudomonas V antigen protects against type III intoxication and lung injury. Nat Med 1999; 5:392–398. 124. Kunzelmann K, Beesley AH, King NJ, Karupiah G, Young JA, Cook DL Influenza virus inhibits amiloride-sensitive Na+ channels in respiratory epithelia [see comments]. Proc Natl Acad Sci USA 2000; 97:10282–10287. 125. Guggino WB, Guggino SE. Amiloride-sensitive sodium channels contribute to the woes of the flu. Proc Natl Acad Sci USA 2000; 97:9827–9829. 126. Rezaiguia S, Garat C, Delclaux C, Meignan M, Fleury J, Legrand P, Matthay MA, Jayr C. Acute bacterial pneumonia in rats increases alveolar epithelial fluid clearance by a tumor necrosis factor-alpha-dependent mechanism. J Clin Invest 1997; 99:325–335. 127. Abraham E, Wunderink R, Silverman H, Perl TM, Nasraway S, Levy H, Bone R, Wenzel RP, Balk R, Allred R, et al. Efficacy and safety of monoclonal antibody to human tumor necrosis factor alpha in patients with sepsis syndrome. A randomized, controlled, double-blind, multicenter clinical trial. TNF-alpha MAb Sepsis Study Group. JAMA 1995; 273:934–941. 128. Khimenko PL, Bagby GJ, Fuseler J, Taylor AE. Tumor necrosis factor-α in ischemia and reperfusion injury in rat lungs. J Appl Physiol 1998; 85:2005–2011. 129. Li XY, Donaldson K, Brown D, MacNee W. The role of tumor necrosis factor in increased airspace epithelial permeability in acute lung inflammation. Am J Respir Cell Mol Biol 1995; 13:185–195. 130. Matthay MA. Severe sepsis—a new treatment with both anticoagulant and antiinflammatory properties. N Engl J Med 2001; 344:759–762.

Resolution of alveolar edema

379

131. Pacitti AJ, Inoue Y, Souba WW. Tumor necrosis factor stimulates amino acid transport in plasma membrane vesicles from rat liver. J Clin Invest 1993; 91:474–483. 132. Börjesson A, Norlin A, Wang X, Andersson R, Folkesson HG. TNFα stimulates alveolar liquid clearance during intestinal ischemia-reperfusion in rats. Am J Physiol Lung Cell Mol Physiol 2000; 278:L3–L12. 133. Fukuda N, Jayr C, Lazrak A, Wang Y, Lucas R, Matalon S, Matthay MA. Mechanisms of TNF-alpha stimulation of amiloride-sensitive sodium transport across alveolar epithelium. Am J Physiol Lung Cell Mol Physiol 2001; 280:L1258–L1265. 134. Matalon S, Egan EA. Effects of 100% O2 breathing on permeability of alveolar epithelium to solute. J Appl Physiol 1981; 50:859–863. 135. Matalon S, Cesar MA. Effects of 100% oxygen breathing on the capillary filtration coefficient in rabbit lungs. Microvasc Res 1985; 29:70–80. 136. Freeman BA, Young SL, Crapo JD. Liposome mediated augmentation of superoxide dismutase in endothelial cells prevents oxygen injury. J Biol Chem 1983; 258:12534–12542. 137. Hayatdavoudi G, O’Neil JJ, Barry BE, Freeman BA, Crapo JD. Pulmonary injury in rats following continuous exposure to 60% O2 for 7 days. J Appl Physiol 1981; 51:1220–1231. 138. Baker RR, Holm BA, Panus PC, Matalon S. Development of O2 tolerance in rabbits with no increase in antioxidant enzymes. J Appl Physiol 1989; 66: 1679–1684. 139. Haskell JF, Yue G, Benos DJ, Matalon S. Upregulation of sodium conductive pathways in alveolar type II cells in sublethal hyperoxia. Am J Physiol 1994; 266:L30–L37. 140. Sznajder JI, Olivera WG, Ridge KM, Rutschman DH. Mechanisms of lung liquid clearance during hyperoxia in isolated rat lungs. Am J Respir Crit Care Med 1995; 151:1519–1525. 141. Gonzalez Flecha B, Evelson P, Ridge K, Sznajder JI. Hydrogen peroxide increases Na+/K(+) ATPase function in alveolar type II cells. Biochim Biophys Acta 1996; 1290:46–52. 142. Factor P, Dumasius V, Saldias F, Brown LA, Sznajder JI. Adenovirus mediated transfer of an Na+/K+−ATPase betalsubunit gene improves alveolar fluid clearance and survival in hyperoxic rats. Hum Gene Ther 2000; 11:2231–2242. 143. Garat C, Meignan M, Matthay MA, Luo DF, Jayr C. Alveolar epithelial fluid clearance mechanisms are intact after moderate hyperoxic lung injury in rats. Chest 1997; 111:1381–1388. 144. Lasnier JM, Wangensteen OD, Schmitz LS, Gross CR, Ingbar DH. Terbutaline stimulates alveolar fluid resorption in hyperoxic lung injury. J Appl Physiol 1996; 81:1723–1729. 145. Saldias FJ, Comellas A, Ridge KM, Lecuona E, Sznajder JI. Isoproterenol improves ability of lung to clear edema in rats exposed to hyperoxia. J Appl Physiol 1999; 87:30–35. 146. Olivera WG, Ridge KM, Sznajder JI. Lung liquid clearance and Na, KATPase during acute hyperoxia and recovery in rats. Am J Respir Crit Care Med 1995; 152:1229–1234. 147. Nici L, Dowin R, Gilmore-Hebert M, Jamieson JD, Ingbar DH. Response of rat type II pneumocyte Na, K-ATPase to hyperoxic injury. Chest 1991; 99:31S– 33S. 148. Yue G, Matalon S. Mechanisms and sequelae of increased alveolar fluid clearance in hyperoxic rats. Am J Physiol 1997; 272:L407–L412. 149. Stern M, Ulrich K, Robinson C, Copeland J, Griesenbach U, Masse C, Cheng S, Munkonge F, Geddes D, Berthiaume Y, Alton E. Pretreatment with cationic lipid-mediated transfer of the Na+K+− ATPase pump in a mouse model in vivo augments resolution of high permeability pulmonary oedema. Gene Ther 2000; 7:960–966. 150. Borok Z, Mihyu S, Fernandes VFJ, Zhang X-L, Kim K-J, Lubman RL. KGF prevents hyperoxia-induced reduction of active ion transport in alveolar epithelial cells. Am J Physiol 1999; 276:C1352–C1360. 151. Carter EP, Wangensteen OD, Dunitz J, Ingbar DH. Hyperoxic effects on alveolar sodium resorption and lung Na-K-ATPase. Am J Physiol 1997; 273:L1191–L1202. 152. Carter EP, Wangensteen OD, OGS M, Ingbar DH. Effects of hyperoxia on type II cell Na-KATPase function and expression. Am J Physiol 1997; 272:L542–L551.

Acute respiratory distress syndrome

380

153. Factor P, Saldias F, Ridge K, Dumasius V, Zabner J, Jaffe HA, Blanco G, Barnard M, Mercer R, Perrin R, Sznajder JI. Augmentation of lung liquid clearance via adenovirus-mediated transfer of a Na K-ATPase β1 subunit gene. J Clin Invest 1998; 102:1421–1430

16 Sepsis in the Acute Respiratory Distress Syndrome Treatment Implications ARTHUR P.WHEELER and GORDON R.BERNARD Vanderbilt University School of Medicine Nashville, Tennessee, U.S.A.

I. Relationship Between Sepsis and Acute Lung Injury The interrelationship of sepsis and acute lung injury is complex. The development of signs of systemic inflammation due to infection, sepsis, is the most common cause of acute lung injury (ALI) accounting for 40–50% of all cases (1, 2). The subset of ALI victims with the most severe oxygenation abnormalities are said to have acute respiratory distress syndrome (ARDS). A reciprocal relationship exists in that the lung is the most commonly identified site of infection leading to the development of sepsis (3). When an acute organ system dysfunction occurs in patients with sepsis, the syndrome is then termed “severe sepsis” (4). The incidence of severe sepsis is increasing rapidly, resulting in more cases of acute lung injury (5). Interestingly, acute lung injury is the most common organ dysfunction qualifying patients for the diagnosis of severe sepsis. Thus, ALI is often the result of sepsis and by virtue of their vulnerability to infection, ALI and ARDS victims often develop secondary sepsis (6). Because pneumonia is the infection leading to the development of sepsis in more than half of all cases, it seems natural that the lung would exhibit some degree of malfunction. Clinically, however, the concurrence of pneumonia and acute lung injury introduces uncertainty as to whether the hypoxemia observed represents merely pneumonia or the more diffuse and significant problem of acute lung injury. Patients who develop severe sepsis typically exhibit two or three organ dysfunctions at its onset, but when only one organ fails it is usually the lung (7). Overall, nearly 90% of patients developing severe sepsis will show evidence of pulmonary dysfunction to a degree requiring supplemental oxygen and positive pressure ventilation (2) with up to 42% of patients eventually meeting ARDS criteria (8). Fewer—approximately one third of all patients—will have ARDS at the time that severe sepsis is recognized (9). The frequency of lung dysfunction in severe sepsis appears to be independent of the underlying site of infection. Not only is the lung the most common organ affected by sepsis, it is usually the first organ dysfunction noted (3, 7, 9, 10). Complaints of dyspnea, tachypnea and a declining arterial hemoglobin saturation are the sentinel findings which typically prompt intensive care unit admission. The progression from the first signs of lung dysfunction to maximal severity is typically a few hours to a day (3).

Acute respiratory distress syndrome

382

II. Pathophysiological Similarities Between Sepsis and ALI Sepsis is substantially more complex that just excessive inflammation, involving an exuberant coagulopathic response and impaired fibrinolysis (11–14). These three same pathophysiological components have been identified in acute lung injury. Abundant evidence of excessive lung inflammation in ARDS victims exists, including increased numbers of pulmonary leukocytes and inflammatory cytokines, including tumor necrosis factor and interleukin (IL)-8 (15). Similarly, it is recognized that increased tissue factor expression and fibrin generation lead to pulmonary capillary thrombosis. In addition to the lung’s increased thrombotic tendency, normal fibrinolysis is impaired, as evidenced by expression of elevated plasminogen activator inhibitor levels (16). Long considered to be a problem only of vascular endothelium, it is now clear that ALI and ARDS affect epithelial function as well (17). This epithelial damage or dysfunction may be one factor predisposing to the development of secondary infection in ALI. Why the lung is so commonly injured in sepsis is unclear. In addition to being the most common infection site leading to sepsis, the lung is also the only organ to accept the full cardiac output. The lungs’ extensive vascular capillary network presents a huge surface area for endothelial cell exposure to exogenous and endogenously generated toxins. The lung also has a particularly delicate structure where luxuriantly perfused endothelium and epithelium constantly exposed to the atmosphere are separated by a minimal interstitial component. Furthermore, the lung contains a huge number of mononuclear-macrophages, cells that serve as potent generators of both pro-inflammatory and pro-coagulant activity.

III. Sepsis as a Complication of ALI Because patients with ALI or ARDS usually require a 1- to 2-week intensive care unit (ICU) stay on a mechanical ventilator, they are at significant risk for the development of superimposed infection. A fraction of these infected patients will progress to severe sepsis. Intravenous and urinary catheter-associated infections and ventilator-associated pneumonia are the most common infectious problems. Preexisting lung injury, presence of an endotracheal tube, and use of mechanical ventilation all predispose to the development of nosocomial pneumonia. Mechanical ventilation increases the risk of pneumonia among hospitalized patients between 7- and 21-fold (18, 19). This relative risk increase translates into a 1–3% per day estimated risk for developing nosocomial pneumonia. In older studies the incidence of clinically diagnosed nosocomial pneumonia approached 70% (20). However, when strict bacteriological criteria are required, the diagnostic rate is closer to 20% (21). The presence of ARDS adds additional risk. The relative risk of nosocomial pneumonia development among patients with ARDS is approximately 50% higher than that for mechanically ventilated patients without ARDS (37% vs. 23%), and ventilator-associated pneumonia is associated with an increased mortality rate in this population (18, 19, 22). Like nosocomial pneumonia, urinary tract infection can be difficult to diagnose with certainty. It is often hard to distinguish simple bacteruria representing colonization from true infection (23). Although it is very common to detect bacteruria or pyuria among

Sepsis in the acute respiratory distress syndrome

383

critically ill ALI/ARDS victims with indwelling urinary catheters, it is uncommon for the urinary tract to be a source of severe sepsis (7, 9). Despite the nearly 3% daily incidence of bacteruria, only 10% of those patients will develop bacteremia (24). Furthermore, the treatment of urinary tract infection in the absence of obstruction is relatively simple and highly successful. Perhaps the best strategy to prevent development of severe sepsis from urinary tact infection is to remove all unnecessary catheters immediately and reconsider the need for indwelling catheters on a daily basis. Although vascular catheter-related infections may be less common than lung or urinary tact infections, they carry a substantial risk for the development of bacteremia and sepsis. In fact, nearly three quarters of all episodes of bacteremia in the ICU are now attributed to vascular catheters (25), and up to 70% of infected central venous catheters result in bacteremia (26). The incidence of severe sepsis when catheter-related bacteremia occurs is significant, and the resulting illness is often lethal (27).

IV. Preventing Infection in ALI and ARDS A. Pneumonia Because sepsis induced by ventilator-associated pneumonia is such a frequent complicating factor for patients with ALI/ARDS, approaches to prevention, early diagnosis, and effective treatment make sense. Because ventilator-associated pneumonia risk is proportional to time on the ventilator, perhaps the simplest method of reducing the risk and subsequent sepsis is to reduce the duration of mechanical ventilation. An inexpensive, reproducible method to accomplish this goal is to conduct daily spontaneous breathing trials in all patients meeting simple weaning eligibility criteria (28). Doing so reduces the average time on the ventilator by several days per patient. The most widely accepted sequence for development of pneumonia in ALI/ARDS is rapid microbial colonization of the upper airway with subsequent aspiration and seeding of the lower airway days later (29). Thus, methods to decrease upper airway colonization and reduce the potential for aspiration could be beneficial. Elevation of the head of patients undergoing mechanical ventilation has been shown to be a safe and highly effective method of preventing ventilator-associated pneumonia (30, 31). Use of endotracheal tubes that provide subglottic suctioning, while theoretically attractive, remain unproven and present some practical problems. Reports of substantial (up to 50%) reduction in nosocomial pneumonia rates using these devices are enticing (32, 33). Unfortunately, in clinical use the suction port used for secretion clearance commonly becomes impacted, rendering the tube ineffective. One source of organisms causing nosocomial pneumonia is contaminated ventilator tubing condensate. The inadvertent introduction of a tiny number of organisms into the warm moist circuit can result in proliferation of a huge number of organisms. Thus, efforts to prevent “rain out” of heated water in ventilator tubing with subsequent inoculation and bacterial growth have been pursued. One method of decreasing this problem is through use of heat-moisture exchange devices (HMEs). HMEs are effective at maintaining the humidity of respiratory circuits without using heated circuits but have two significant limitations. First, HMEs introduce dead space into the ventilator circuitry,

Acute respiratory distress syndrome

384

an undesirable effect when ventilating patients with ALI and ARDS. Second, the moisture-exchanging properties of HME devices are less effective as minute ventilation increases, again a common problem among patients with ALI and ARDS (34, 35). A second approach has been to heat the ventilator tubing itself to prevent condensation. This approach lacks the problems of increased dead space and water accumulation in ventilator tubing but has not yet gained wide popularity, perhaps because of the costs of the circuits. To prevent external seeding of organisms into the ventilator circuit, closed suctioning systems are beneficial (36). Ironically, frequent replacement of the ventilator circuitry is associated with a higher, not lower frequency of tubing colonization and subsequent infection (37). This situation is probably analogous to the observation that frequent entry into intravenous catheters is also associated with an increased catheter infection rate. Another simple proven method to reduce the risk of ventilator-associated pneumonia is the avoidance of transnasal tubes (38). By causing irritation and obstructing the sinus ostia, the presence of a nasal tube predisposes to sinus colonization and inflammation. When ventilator-associated pneumonia develops in a patient with a nasal tube, the correlation between organisms found in the sinuses and in the lung is high. Although the evidence is less robust, insertion of any device across the lower esophageal sphincter increases the risk of reflux and aspiration, an additional risk factor for the development of nosocomial pneumonia (39). Antimicrobial therapy to prevent nosocomial pneumonia has been extensively studied. Use of oral antibiotic paste or nonabsorbable gastrointestinal antibiotics to reduce enteral contamination is controversial. Most studies indicate that administration of oral antibiotic paste with or without systemic therapy can reduce risk of infection for a brief period of time, changes the spectrum of organisms, but results in rapid increases in microbial resistance (40, 41). Likewise, although potentially attractive, administration of aerosolized antibiotics has not been proven to decrease colonization rates or incidence of pneumonia. ICU-wide scheduled rotation of antibiotics is a relatively recent strategy designed to decrease the risk of pneumonia, particularly that involving resistant organisms. Unfortunately, at this time the utility of antibiotic rotation remains unproven, and logistical problems in implementing such a strategy in many ICUs are substantial (42, 43). Conflicting data exist regarding the impact of enteral nutrition and modifying gastric pH in patients with ARDS at risk for ventilator-associated pneumonia. Normally the stomach is nearly sterile because of gastric acid. Drugs used to raise gastric pH in an attempt to decrease gastrointestinal bleeding have been shown to increase in the number of microorganisms in the stomach (44). These bacteria (or fungi), if refluxed and aspirated, can be a source of nosocomial pneumonia. Data exist on both sides of this issue to suggest that ventilator-associated pneumonia risk is or is not a significant problem when modifying gastric pH (45, 46). On balance, data favor the use of an acid-modifying drug to decrease the risk of gastrointestinal bleeding coupled with elevation of the head of the bed (46). Although several studies among mostly surgical patients suggest that enteral nutrition reduces the risk of infectious complications among patients on mechanical ventilators, the story is complicated. When benefit has been demonstrated, it has most commonly involved the reduction of nosocomial pneumonia or wound infection rates (47, 48). The

Sepsis in the acute respiratory distress syndrome

385

mechanism by which this protection occurs is unknown. In contrast, at least one study indicates early institution of enteral feeding is associated with an increased risk of aspiration and nosocomial pneumonia (49). Thus, while enteral feeding likely benefits the nutritional status of patients with ALI and ARDS and may reduce infection risks, the optimal timing of administration remains uncertain. A rational even if unproven approach is among patients requiring long-term ventilation to begin enteral feeding at a relatively low rate as soon as hemodynamic stability is achieved. After initiation, the feeding rate is increased as tolerated. Unfortunately, there is no standardized definition of “tolerated,” and local practice often results in substantial underfeeding compared to projected calorie and protein requirements (50). In an attempt to reduce the risks of reflux, aspiration, and pneumonia, some investigators have proposed postpyloric delivery of enteral nutrition. Although postpyloric feeding may decrease the risk of reflux and aspiration, this approach has significant technical limitations. On average it takes longer to achieve acceptable tube placement, resulting in feeding delays, and during feeding the tube is often dislodged from its optimal position (51). Not only are route, timing, and tube position contested, the benefits of specific components of the feeding remain controversial. Enteral preparations enhanced with glutamine, arginine, and omega-3 fatty acids have been associated with reductions in important clinical endpoints, including nosocomial infection (52–54). In randomized but relatively small studies in heterogeneous patient populations, specific nutritional components may offer outcome benefit. A final opinion on the utility of specific nutritional formulations awaits additional large-scale clinical trials (55). B. Intravenous Catheters Due to their duration of insertion, location, and frequency of entry, central venous catheters confer a much greater risk for bacteremia than peripheral intravenous lines (56– 58). Among critically ill patients, up to three quarters of all episodes of bacteremia can be attributed to catheter-related infection (24). When a catheter-related bloodstream infection occurs, the risk of sepsis and death is high and added costs are significant. (59). Although risk factors for development of catheter-related infections are well accepted (60–69), the methods for preventing intravascular catheter-related infections are a subject of controversy. Several basic principles are, however, widely accepted. Central venous catheters have the lowest infection rates if uncomplicated placement is achieved by an experienced operator using complete sterile barrier precautions (65). The insertion site has an influence on the risk of subsequent infection, the femoral site being most likely to become infected, with the internal jugular site the next most likely. Perhaps because of catheter stability and relative freedom from exposure to bacteria-laden body fluids, the subclavian site has usually been reported to have the lowest infection risk (70–73). During insertion, specific skin decontamination procedures further reduce risk of subsequent line infection. Avoiding shaving insertion sites is prudent. When cleansing the skin, chlorhexidine is superior to povidone-iodine combinations but does not enjoy widespread use because of concerns over the remote possibility of toxicity (74). Firmly anchoring the catheter or catheter hub to the patient is important in reducing motion at the skin/catheter interface. Since most line-related infections begin by skin

Acute respiratory distress syndrome

386

colonization followed by extension along the subcutaneous catheter tract, the “pistonlike” movement of an unsecured catheter can act to “inject” organisms along the catheter tract. The use of nonocclusive dressings and minimizing the number of entries into a closed infusion system have also been proven to reduce infection risk (75–79). There is no proven advantage of antibiotic ointments to maintain the insertion site of central venous catheters. Although hotly contested, the value of the routine use of antiseptic- or antibiotic-impregnated or -coated catheters awaits additional evidence (80–83). In contrast, permanent or semi-permanent tunneled catheters are at a lower risk of infection (84). The risk of line colonization and eventual blood stream infection rises exponentially with time after insertion, with a striking inflection point at about 7 days. Therefore, limiting the duration of catheters to 5–7 days is a defensible practice adopted by many. At odds with this common practice, two clinical trials of catheter-changing strategies encompassing a total of approximately 270 patients have not detected benefit to scheduled site changes at intervals of 3 or 7 days (85, 86). Given the many factors influencing catheter-related infection, the decision to replace catheters should balance infection risks and costs against insertion risks and costs in individual patients. At the very least, it is a sound principle to remove central catheters as soon as they are no longer needed. Recent studies have demonstrated impressive results when using a multifaceted approach to catheter insertion and care incorporating all of the best care practices described above (87, 88).

V. Treating Severe Sepsis in ALI and ARDS A. Antibiotics and Infection Source Control When sepsis complicates the course of ALI or ARDS, prompt control of the infection source should be undertaken. All approachable areas of localized infection should be drained (e.g., pleural space, abdominal abscess, urinary tract), necrotic tissue debrided, and infected foreign bodies (e.g., intravenous catheters) removed. Appropriate cultures should be obtained, preferably before institution of antimicrobial therapy because significant differences in the rate of recovery of organisms can result from antibiotic administration (89). Obtaining cultures is a prudent practice, even though a substantial number of patients will remain culture negative from nonbloodstream sources and the vast majority of patients will have sterile blood cultures (9). The knowledge of a specific organism and its sensitivity can permit simplification of antimicrobial therapy, but unfortunately antimicrobial coverage is often not simplified based on culture data. Appropriate antimicrobial therapy is important for the treatment of severe sepsis. Antibiotic selection should be guided by the presumed site of infection, early Gram stain results, suspected or known organisms, resistance patterns in the community, and presence of individual host factors affecting immune status (90). When the source of infection is known, especially if the etiological organism is identified, narrowly targeted anti-microbial therapy should be used. However, when the offending organism is uncertain, the site of infection is unclear, or perhaps even when the patient is desperately

Sepsis in the acute respiratory distress syndrome

387

ill, broader-spectrum antibiotic therapy should be instituted. Even when selecting the best empiric antibiotic combination and giving it promptly, antibiotic therapy has significant limitations. Not all species of microbes will be empirically covered, nor can all possible resistance be anticipated. Thus, there is little room for dogmatism in antibiotic selection. Criticism of antibiotic choices is easy after culture data return, and it is always possible to conceive of an etiological possibility that has not been treated. For example, almost uniformly empiric therapy will not include, coverage for highly resistant gram-negative rods, tuberculosis, or fungi, and thus therapy will be deemed “inadequate.” For most presumed infections there are a large number of safe, acceptable, and cost-efficient drug combinations. Clinicians should have a working knowledge of the common nosocomial organisms and resistance patterns in the hospital in which they practice. By anticipating the likely pathogens infecting the ALI/ARDS patient, the best therapy can be provided. In most settings this will include coverage for both gram-negative and gram-positive organisms. Whether methicillin-resistant staphlycocci or multiply resistant gramnegatives need treatment depends on the frequency of the organisms in the unit at risk. Among patients with proven bacteremia, appropriate antibiotic therapy reduces the risk of shock development and may reduce the risk of death by as much as 11–50% (91, 92). Likewise, among ICU patients with serious infections, failure to provide appropriate antimicrobial therapy within 24–48 hours results in a significantly higher mortality (93). However, antibiotics are not a panacea for severe sepsis and are not sufficient treatment alone, as evidenced by the overall 30–50% mortality that persists (9, 92). In large retrospective surveys or prospective clinical trials, the antibiotics prescribed prove inappropriate for the organism(s) eventually isolated in approximately 10% of patients (7, 93). The speed with which antibiotics must be administered is also controversial. Except perhaps for brief delays to obtain cultures, there is no reason to deliberately postpone antimicrobial therapy. However, data associating improved outcome with more rapid administration of antibiotics for treatment of community-acquired infections do not prove causation (94), and there is not yet a demonstrated advantage to undertaking extraordinary measures to expedite antibiotic administration. Early broad-spectrum antibiotic therapy may be good for individual patients with lifethreatening infections, but prolonged unnecessary therapy is harmful to both that patient and the rest of the community. Hence, it is very important to narrow or discontinue broad-spectrum antibiotic therapy when cultures identify a single causative organism or when all cultures remain negative after a reasonable period (48–72 hours) of observation. Failure to do so exposes patients to adverse drug reactions and the high cost of antibiotics and breeds resistance. B. Circulatory Support Expert consensus advises that circulatory support in ALI/ARDS patients who develop sepsis should begin with replacement of intravascular volume. This recommendation comes from observations that severe sepsis patients typically have intravascular volume deficits resulting from increased venous capacitance and vascular permeability and sweating, vomiting, tachypnea, and bleeding. In addition, the critically ill also usually

Acute respiratory distress syndrome

388

have significant decreases in intake. The intravascular volume deficit is likely to be exacerbated by a reduced plasma oncotic pressure from hypoalbuminemia (95, 96). The basic principle that volume repletion should be sufficient to provide adequate organ perfusion while avoiding volume overload or exacerbating the degree of pulmonary edema is widely accepted. Likewise, clinical indicators of excessive fluid administration including increasing rales, a decrease in oxygenation indices (e.g., ratio), and a worsening chest radiograph are accepted. However, the optimal index for fluid therapy remains uncertain and highly controversial. Collectively, improved mentation, a rising blood pressure, falling heart rate, increased urine output, reversal of skin mottling, and prolonged capillary refill times are typically used as evidence of improvement. These observations usualy correlate with rising central venous pressure and/or pulmonary artery occlusion pressure. Wide variation exists in monitoring practice for patients with ALI/ ARDS. Some physicians choose not to monitor intravascular pressures at all, while others use central venous pressure, and roughly 50% of ALI patients acquire a pulmonary artery catheter (97). Although the pulmonary artery catheter provides additional hemodynamic data (i.e., pulmonary artery occlusion pressure, cardiac output, and mixed venous oxygen), there are significant concerns about the safety and effectiveness of this device among patients with ALI and ARDS (98). In the case-control study of Connors et al. (98), the risk of death among patients treated with a pulmonary artery catheter averaged 20–25% higher than for patients not receiving a catheter. The cause of the apparent mortality increase is unknown. This question is currently being evaluated in a large randomized clinical trial of fluid therapy and monitoring being performed by the ARDS Network (99). Unfortunately, regardless of the method of monitoring, optimal target value(s) for hemodynamic indices are unknown. Whatever target value for filling pressure is selected, the fluid volume required to achieve that goal is surprisingly high. Over the first day 5–10 L of crystalloid or an equivalent amount of colloid is typically required to maintain filling pressure targets (100). Initial resuscitation is usually initiated with rapid infusions of at least 15–20 mL/kg crystalloid (e.g., isotonic saline or Ringer’s solution). Smaller crystalloid boluses are unlikely to alter hemodynamics. Colloids (e.g., albumin) are preferred by some physicians for patients with a hypoalbuminemia, severe hypotension with limited vascular access, or renal failure (99). If albumin is used, sequential doses of 25–50 g are tried. Equivalent vascular volume expansion may be accomplished faster using colloid than crystalloid because of the smaller volume of infusate needed, but the value of this perceived advantage is unknown. For patients deficient in soluble clotting proteins who also have intravascular volume deficiency, fresh frozen plasma provides an alternative but incurs the risks of human blood product exposure. A systematic review of randomized trials of crystalloid versus colloid solutions in critically ill patients found that resuscitation with colloid was associated with a small increased risk of mortality (102). Even if colloid and crystalloid are equal in effect, colloid cost is substantially higher than that of crystalloid. If by whatever measures the clinician chooses, hypotension, oliguria, or a low flow state persists after achieving an acceptable filling pressure, administration of a vasoactive agent is undertaken. Although targets for optimal pressure and flow are unknown, MAP≥60 mmHg, and cardiac index (CI)>2.5 L/min are typically targeted (103).

Sepsis in the acute respiratory distress syndrome

389

Obviously these targets must be individualized: a young, otherwise healthy patient without vascular disease can tolerate a significantly lower pressure than an elderly chronically hypertensive patient with extensive vascular disease. It makes sense to tailor the vasoactive agent to the pathophysiology—for patients with low systemic resistance and normal or high cardiac output agents with predominately αadrenergic activity like norepinephrine are rational. For patients with impaired cardiac output with normal or low systemic vascular resistance, use of a drug possessing βadrenergic activity is rational. The problem with this approach is that it mandates placement of a pulmonary artery catheter with its attendant risks and costs. Furthermore, it presumes that such a strategy has been proven safe and effective. Unfortunately, the major limitation with all vasopressor agents is the they have not been compared head to head and thus the best therapy is unknown. Although experience and personal preference play a role, norepinephrine has as emerged as the preferred vasopressor for most intensivists because it is easily titrated and reliably raises blood pressure even among patients failing substantial dose of dopamine (104–107). Dopamine, long preferred by some practioners because of its perceived salutary effects on renal function, has now been proven to not have significant renoprotective activity (108). Although not renoprotective, one should not conclude that dopamine does not increase blood pressure. Recently vasopressin has been proposed as an effective, economical alternative vasopressor for patients with refractory hypotension (109). Depletion of pituitary vasopressin and low plasma levels have been reported in hypotensive patients with septic shock (110). Replacement with low doses of vasopressin has been successful in raising blood pressure and reducing the need for alternative vasopressors. No data yet exist to show improved outcomes using vasopressin compared to other vasoactive agents (111). The role of strategies to produce supranormal oxygen delivery in severe sepsis patients is uncertain but quite controversial. Among high-risk surgical patients, achieving an arbitrary level of oxygen delivery is associated with a reduction in mortality (112–114). In studies of other patient populations no benefit has been demonstrated or high oxygen delivery strategies were harmful (115–117). Interestingly the studies in which benefit was demonstrated were typically early intervention or prophylaxis studies. This observation has led some to hypothesize that it may not be enough to raise oxygen delivery to a target level but achieving that goal early may be critical. This principle may have been recently illustrated by Rivers et al. using goal-directed therapy for resuscitation of patients with sepsis (118). In this study of early resuscitation, patients with sepsis first had central venous pressure restored with fluid. Then blood pressure was increased above a preset target using with vasoactive agents. If the central venous saturation remained below 70% after these maneuvers, red blood cells and dobutamine were administered. This protocol resulted in significant improvements in hospital and 60-day mortality. The findings of this trial require confirmation before being widely adopted but suggest that early intervention may be essential to achieve optimal results. C. Ventilatory Support Optimal ventilatory support requires an adequate airway and that the patient receives ventilatory support to maintain a PaO2≥60 mmHg or hemoglobin saturation ≥88%. It is

Acute respiratory distress syndrome

390

now clear that maintaining a near-normal PCO2 is a much lower priority than once believed provided the pH is not profoundly depressed or specific contraindications to hypercarbia (e.g., increased intracranial pressure) are present. The traditional approach to mechanical ventilation in patients with ALI/ARDS has been to use tidal volumes of 10– 15 mL/kg of actual body weight. However, excessive alveolar stretch induced by high tidal volumes can produce barotrauma and incite cytokine release contributing to the development of nonpulmonary organ failures. The NIH NHLBI ARDS Network trial of normal tidal volume ventilation compared a traditional breath size (12 mL/kg) with a lower tidal volume (6 mL/kg of predicted body weight) reduced even further, if necessary, to maintain plateau pressures of <30 cmH2O (1). and PEEP were adjusted by protocol to maintain hemoglobin saturations between 88% and 95%. This simple inexpensive strategy resulted in a 22% decrease in hospital mortality, an increase in ventilator-free days, and less nonpulmonary organ dysfunction. A reduced tidal volume strategy designed to manage plateau pressures should be employed in all patients with acute lung injury who lack contraindications. D. Drotrecogin Alfa (Activated) Severe sepsis is a complex combination of excessive inflammation and coagulation and impaired fibrinolysis. The interplay of inflammation and abnormal clotting has been recognized for at least 30 years (119, 120). Disseminated intravascular coagulation, the most extreme clotting disorder of sepsis, occurs in less than 20% of sepsis victims. However, a subtle, often subclinical coagulopathy accompanied by impaired fibrinolysis is nearly universal (121). The most common laboratory findings of this process are elevations in d-dimer levels and depletion of circulating protein C. Along with tissue factor pathway inhibitor and antithrombin III, protein C is one of three natural antithrombotic agents. When converted to its activated form by a thrombinthrombomodulin complex, activated protein C has antithrombotic, anti-inflammatory, and profibrinolytic properties. Activated protein C is a key component in controlling septic coagulopathy, as evidenced by the inverse correlation between the plasma levels of protein C and morbidity and mortality in septic patients (122–125). Activation of protein C is impaired in severe sepsis (126–128). Human recombinant activated protein C is now available as drotrecogin alfa (activated). Numerous Phase I studies of humans, including those heterozygous for protein C, and patients with end-stage renal disease and purpura fulminans demonstrated the safety and biochemical effects of incremental drotrecogin given on one or more occasions. A placebo-controlled Phase II study of drotrecogin alfa (activated) produced dose-proportional reductions in d-dimers and IL-6, key markers of sepsis-associated coagulopathy and inflammation (129). Based on the Phase II data, a large, randomized blinded, placebo-controlled, multicenter Phase III study of severe sepsis was conducted (7). Mortality at 28 days was reduced by a 96-hour infusion of drotrecogin alfa (activated) from 31% to 25%. This represented a 6.1% absolute (19.4% relative) reduction in the risk of death among the 1690 adults enrolled. The salutary effect was remarkably consistent among subgroups, including those stratified by the number of dysfunctional organs, sex, age, site or type of infection, or initial protein C level. This is a significant departure from other

Sepsis in the acute respiratory distress syndrome

391

investigational drugs where benefit was confined to small subgroups (130). As expected, the absolute mortality reduction was largest among the sickest patients, including those with more organ failures, especially shock and respiratory failure, and those with higher modified APACHE II scores at study entry. Drotrecogin’s only recognized toxicity is bleeding. During the 28-day Phase III trial, 30 (3.5%) serious bleeding events occurred among drug-treated patients compared to 17 (2.0%) among placebo recipients. The majority of the serious bleeding occurred during drug infusion. Fortunately the risk of intracranial hemorrhage was well below 1%. Among treated patients bleeding episodes were associated with invasive procedures, platelet counts <30,000/mL, and significant elevations in prothrombin time. For this reason drotrecogin should not be given to patients with active hemorrhage or to patients at high risk of serious bleeding, especially into the neuraxis. To minimize bleeding, infusions should be interrupted 2 hours before invasive procedures and discontinued completely if serious bleeding develops. Infusions may be restarted immediately after minor uncomplicated procedures and 12 hours after surgery provided hemostasis is adequate. There is no antidote for drotrecogin, but the drug is all but completely cleared by plasma enzymes within 2 hours of discontinuing the infusion. Thus, in the event of bleeding, discontinuing the drug rapidly reverses anti-coagulation. Drotrecogin very rarely provokes an immunological response, and neutralizing antibody formation, anaphylactoid reactions, or serum sickness have not occurred after infusion. E. Nutrition and Metabolism Severe sepsis is a catabolic disease, but nutritional support is not usually an immediate concern and can be safely deferred for several days. Indeed, feeding the hemodynamically unstable patient is likely to be unsuccessful, and some practitioners have concerns that feeding may precipitate gut ischemia. Metabolic changes in the patient with severe sepsis include increases in oxygen consumption, rapid catabolism, hyperglycemia with insulin resistance, and negative nitrogen balance. As a consequence of these alterations, nutritional requirements are high and malnutrition often results (131). Absent contraindications, enteral nutrition is now preferred by most practioners. Potential advantages include buffering of gastric acid, maintenance of the gut mucosal barrier, avoidance of parenteral nutrition catheters and their complications, and establishment of physiological enteral hormone secretion (132). For patients with severe sepsis, the ACCP and ASPEN groups have recommended the following: (1) daily caloric intake of 25–30 kcal/kg/usual body weight/day; (2) Protein: 1.3–2.0 g/kg/day; (3) glucose: 30–70% of total nonprotein calories to maintain serum glucose <225 mg/dL; (4) lipids: 15–30% of total nonprotein calories (133, 134). As indicated previously, increasing the percentage of omega-3 and decreasing the percentage of omega-6 fatty acids may have an immune-enhancing effect (53). While the proportion can be reduced, omega-6 fatty acids must be provided in amounts sufficient to avoid essential fatty acids deficiency (generally 1 g/kg/day). The most provocative and interesting nutrition-metabolic development has been the recent description of dramatic reductions in the infection rate and mortality of sepsis associated with tight control of blood glucose in postoperative patients (135). By maintaining blood glucose levels below 110 mg/dL blood stream infections were reduced

Acute respiratory distress syndrome

392

by 42% and hospital mortality was reduced by 32%. Survival benefits persisted to 12 months of follow-up. It remains to be seen if similar findings can be achieved in the general medical ICU population. F. Other Supportive Therapies Gastric stress ulcer prophylaxis to reduce the risk of upper gastrointestinal bleeding is a nearly routine practice in the critically ill (136). Although definitive data are lacking for the severe sepsis population, these patients possess the same risk characteristics as previously studied groups, including prolonged mechanical ventilation, coagulopathy, head injury, burns, and prior ulcer disease. The bleeding risk increases with the number of risk factors present. In perhaps the definitive trial, Cook et al. (45) compared sucralfate to an H2-receptor antagonist for the prevention of upper gastrointestinal bleeding in more than 1000 mechanically ventilated patients. Those receiving H2 blockade had a significantly lower rate of gastrointestinal bleeding than those treated with sucralfate. There is little reason to believe that any particular H2 antagonist or proton pump inhibitor is superior to another, hence drug selection can be driven largely by cost considerations. In contrast, traditional antacids represent an expensive, inconvenient, and perhaps more side effect-prone choice. Because nearly 30% of critically ill patients will develop a deep venous thrombus or pulmonary embolism during their hospital stay without prophylaxis, essentially all critically ill patients should have some form of DVT prophylaxis administered (137). Both chemical anticoagulants (e.g., heparin or low molecular weight heparin) and mechanical devices such as intermittent pneumatic compression devices are effective if applied early. Overall, the relative reduction in thromboembolism risk approaches 50% regardless of which approach is chosen. Selection should probably depend on practical issues such as risk of bleeding and presence of lower extremity injuries. Unfortunately, surveys suggest that perhaps as many as two thirds of critically ill patients do not receive any form of DVT prophylaxis, exposing them to unnecessary thromboembolism risk (138). Renal failure necessitating dialysis develops in <5% of patients with severe sepsis (3). Even though the need for renal replacement therapy is uncommon, acute renal failure is a powerful predictor of mortality. Therefore, steps should be taken to maintain renal function during the disease course. These measures include ensuring prompt, adequate hemodynamic resuscitation and avoiding potentially nephrotoxic medications. Renal replacement therapies for this population include hemodialysis, hemofiltration, isolated ultrafiltration, and peritoneal dialysis (139). Currently the optimal route, timing, and intensity of renal replacement therapy is unknown, but recent data suggest that daily hemodialysis may be superior to alternate-day therapy, even when controlling for total dialysis dose (140). Given the average course of illness, anemia is nearly a universal finding in ALI and ARDS. Phlebotomy, bleeding, and impaired erythropoiesis all contribute. Transfusions should be judiciously used to restore oxygen-carrying capacity in the non-bleeding patient. A more aggressive transfusion strategy is probably more appropriate for the acutely bleeding pa-tient. However, transfusions are not without risk since they can be associated with infection transmission, immunosuppression, transfusion reactions, and

Sepsis in the acute respiratory distress syndrome

393

volume overload. The most appropriate hemoglobin target for transfusion is unknown but probably lies between the traditional value of 10 g/dL and a bare minimum of 3–4 g/dL. With the possible exception of patients with known coronary artery disease, a restrictive transfusion strategy is equivalent and may be superior to a liberal strategy (141). Recombinant human erythropoietin raises hemoglobin values and halves red blood cell transfusion requirements among critically ill patients (142). The optimal dose and dosing schedule are still being determined, but clearly erythropoietin is a useful component of an anemia-management program that includes minimizing phlebotomy and lowering traditional hemoglobin values for transfusion.

VI. Summary Severe sepsis is the most common cause of ALI and ARDS. The onset of lung failure in severe sepsis is typically so rapid as to preclude any meaningful preventative intervention short of averting infection. Preventing the development of secondary sepsis in patients with established ARDS is a more tenable strategy. Measures proven or highly likely to decrease the development of secondary sepsis include closed tracheal suctioning systems, positioning ventilated patients head up, and minimizing the number and duration of central venous catheters. Avoidance of nasal tubes can also decrease the risk of infection. When required, experienced personnel, using full barrier precautions, should insert central venous catheters. After insertion, catheters should be entered sparingly and cared for by a trained catheter team. Reasonable measures to reduce the risk of severe sepsis include minimizing the duration of mechanical ventilation by using a reduced tidal volume strategy with a weaning protocol and removing urinary catheters as rapidly as possible. When sepsis does occur in patients with acute lung injury, source control of the infection, appropriate cultures, and targeted antimicrobial therapy are indicated. Rapid goal-directed fluid and vasopressor therapy probably optimizes survival. Crystalloid infusions followed by norepinephrine now appear to be the most popular therapeutic choices. No consensus exists on the use of pulmonary artery catheters. Nutritional support is indicated, probably best delivered enterally. Unfortunately, at this time the timing, optimal delivery site, and type of feeding are uncertain. It is prudent to provide deep venous thrombosis (DVT) and gastrointestinal bleeding prophylaxis and minimize phlebotomy, thereby allowing minimal transfusion. Finally, activated human recombinant protein C is a break-through life-saving technology for the critically ill patient with severe sepsis not at significant risk of bleeding.

References 1. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 2. Dellinger RP. Current therapy for sepsis. Infect Dis Clin North Am 1999; 13:495–509. 3. Wheeler AP, Bernard GR. Treating patients with severe sepsis. N Engl J Med 1999; 340:207– 214.

Acute respiratory distress syndrome

394

4. Bone RC, Balk RA, Cerra FB, et al. American College of Chest Physicians/ Society of Critical Care Medicine Consensus Conference: Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992; 101:1644–1655. 5. Angus DC, Linde-Zwirble WT, Lidcker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001; 29:1303–1310. 6. Montgomery AB, Stager MA, Carrico CJ, Hudson LD. Causes of mortality in patients with the acute respiratory distress syndrome. Am Rev Respir Dis 1985; 132:485–489. 7. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344: 699–709. 8. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus conference on ARDS, definitions, mechanisms, relevant outcomes and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818–824. 9. Bernard GR, Wheeler AP, Russell JA, Schein R, Summer WR, Steinberg K, et al. The effects of ibuprofen on the physiology and survival of patients with sepsis. N Engl J Med 1997; 336:912– 918. 10. Balk RA. Pathogenesis and management of the multiple organ dysfunction or failure in severe sepsis and septic shock. Crit Care Clin 2000; 16:337–352. 11. Yan SB, Helterbrand JD, Hartman DL, et al. Low levels of protein C are associated with poor outcomes in severe sepsis. Chest 2001; 120:915–922. 12. Vervloet MG, Thijs LG, Hack CE. Derangements of coagulation and fibrinolysis in critically ill patients with sepsis and septic shock. SemiThromb Hemost 1998; 24:33–44. 13. Faust SN, Levin M, Harrison OB, et al. Dysfunction of endothelial protein C activation in severe meningococcal sepsis. N Engl J Med 2001; 345:408–416. 14. Opal SM. Therapeutic rationale for antithrombin III in sepsis. Crit Care Med 2000; 28:S34– S37. 15. Christman JW, Wheeler AP, Bernard GR. Cytokines and sepsis: What are the therapeutic implications? J Crit Care 1991; 6:172–182. 16. Abraham E. Coagulation abnormalities in acute lung injury and sepsis. Am J Respir Mol Cell Biol 2000; 22:401–404. 17. Lesur O, Berthiaume Y, Blaise G, Damas P, Deland E, Guimond JG, Michel RP. Acute respiratory distress syndrome: 30 years later. Can Respir J 1999; 6:71–86. 18. Craven DE, Kunches LM, Kilinsky V, et al. Risk factors for pneumonia and fatality in patient receiving continuous mechanical ventilation. Am Rev Respir Dis 1986; 133:792. 19. Neiderman M, Fein A. The interaction of infection and the adult respiratory distress syndrome. Crit Care Clin 1986; 2:471. 20. Seidenfeld JJ, Pohl, Bell RD, et al. Incidence site and outcome of infections in patients with adult respiratory distress syndrome. Am Rev Respir Dis 1986; 134:12. 21. Sutherland KR, Steinberg KP, Maunder RJ, et al. Pulmonary infection during the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 152:550–556. 22. Kolleff MH. The prevention of ventilator associated pneumonia. N Engl J Med 1999; 340:627– 634. 23. Llewelyn M, Cohen J. Diagnosis of infection in sepsis. Intens Care Med 2001; 27(suppl):S10– S32. 24. Saint S. Clinical and economic consequences of nosocomial catheter related bacteruria. Am J Infect Control 2000; 28:68–75. 25. Gosbell IB, Duggan D, Breust, et al. Infection associated with central venous catheters: a prospective survey. Med J Aust 1995; 162:210–213. 26. Aufwerber E, Ringertz S, Ransjo U. Routine semiquantitative cultures and central venous catheter related bacteremia. APMIS 1991; 99:627–630.

Sepsis in the acute respiratory distress syndrome

395

27. Renaud B, Brun-Buisson C. Outcomes of primary and catheter-related bacteremia. A cohort and case-control study in critically ill patients. Am J Respir Crit Care Med 2001; 163:1584– 1590. 28. Ely EW, Baker AM, Dunagan DP, et al. Effect of the duration of mechanical ventilation of identifying patients capable of breathing spontaneously. N Engl J Med 1996; 335:1864–1869. 29. Cardenosa-Cendrero JA, Sole-Violan J, Bordes Benitez A, et al. Role of different routes of tracheal colonization in the development of pneumonia in patients receiving mechanical ventilation. Chest 1999; 116:462–470. 30. Torres A, Anzar R, Fatell JM, et al. Incidence risk an prognostic factors of nosocomial pneumonia in mechanically ventilated patients. Am Rev Respir Dis 1986; 133:792–796. 31. Drakulovic MB, Torres A, Bauer TT, et al. Supine body position as a risk factor for nosocomial pneumonia in mechanically ventilated patients: a randomized trial. Lancet 1999; 354:1851– 1858. 32. Mahoul PH, Auboyer C, Jospe R, et al. Prevention of nosocomial pneumonia in intubated patients: respective role of mechanical subglottic secretions drainage and stress ulcer prophylaxis. Intens Care Med 1992; 18:20–25. 33. Valles J, Artigas A, Rello J, et al. Continuous aspiration of subglottic secretions in preventing ventilator associated pneumonia. Ann Intern Med 1995; 122:179–186. 34. Ricard JD, Le Miere E, Markowicz P, Lasry S, Saumon G, Djedaini K, Coste F, Dreyfuss D. Efficiency and safety of mechanical ventilation with a heat and moisture exchanger changed only once a week. Am J Respir Crit Care Med 2000; 161:104–109. 35. Iotti GA, Olivei MC, Braschi A. Mechanical effects of heat-moisture exchangers in ventilated patients. Crit Care Med 1999; 3:R77–R82. 36. Littlewood K, Durbin CG Jr. Evidenced-based airway management. Respir Care 2001; 46:1392–1405. 37. Cook D, Ricard JD, Reeve B, Randall J, Wigg M, Brochard L, Dreyfuss D. Ventilator circuit and secretion management strategies: a Franco-Canadian survey. Crit Care Med 2000; 28:3547– 3554. 38. Rouby JJ, Laurent P, Gosnach M, et al. Risk factors and clinical relevance of nosocomial maxillary sinusitis in the critically ill. Am J Respir Crit Care Med 1994; 150:776–783. 39. Cook DJ, De Jonghe B, Brochard L, Brun-Buisson C. Influence of airway management on ventilator associated pneumonia: evidence from randomized trials. JAMA 1998; 279:781–787. 40. Bergmans DC, Bonten MJ, Gaillard CA, Paling JC, van der Geest S, van Tiel FH, Beysens AJ, de Leeuw PW, Stobberingh EE. Prevention of ventilator-associated pneumonia by oral decontamination: a prospective, randomized, double-blind, placebo-controlled study. Am J Respir Crit Care Med 2001; 164: 382–388. 41. Gastinne H, Wolff M, Delatour F, Faurisson F, Chevret S. A controlled trial in intensive care units of selective decontamination of the digestive tract with nonabsorbable antibiotics. N Engl J Med 1992; 326:594–599. 42. Kolleff MH, Vlasnik J, Sharpless L, Pasque C, Murphy D, Fraser V. Scheduled change of antibiotic classes: a strategy to decrease the influence of ventilator associated pneumonia. Am J Respir Crit Care Med 1997; 156: 1040–1048. 43. Gruson D, Hilbert G, Vargas F, Valentino R, Bebear C, Allery A, Bebear C, Gikpi-Benissan G, Cardinaud JP. Rotation and restricted use of antibiotics in a medical intensive care unit. Impact on the incidence of ventilator-associated pneumonia caused by antibiotic-resistant gramnegative bacteria. Am J Respir Crit Care Med 2000; 162:837–843. 44. Tryba M. Risk of acute stress bleeding and nosocomial pneumonia in ventilated intensive care unit patients: sucralafate versus antacids. Am J Med 1987; 83:117–124. 45. Cook DJ, Fuller H, Guyatt GH, et al. Risk factors for gastrointestinal bleeding in critically ill patients. N Engl J Med 1994; 330:377–381.

Acute respiratory distress syndrome

396

46. Cook DJ, Guyatt GH, Marshall J, et al. A comparison of sucralafate and ranitidine for the prevention of upper gastrointestinal bleeding in patients requiring mechanical ventilation. N Engl J Med 1998; 338:791–797. 47. Heyland DK. Nutritional support in the critically ill patient. A critical review of the evidence. Crit Care Clin 1998; 14:423–440. 48. Heys SD, Walker LG, Smith I, Eremin O. Enteral nutritional supplementation with key nutrients in patients with critical illness and cancer: a meta-analysis of randomized controlled clinical trails. Ann Surg 1999; 229:467–477. 49. Ibrahim EH, Mehringer L, Prentice D, Kolleff MH. The effect of timing of enteral feeding on the clinical outcomes of mechanically ventilated patients in the ICU setting. Am J Respir Crit Care Med 2001; 163:A16. 50. Mentec H, Dupont H, Bocchetti M, et al. Upper digestive intolerance during enteral nutrition in critically ill patients: frequency, risk factors, and complications. Crit Care Med 2001; 29:1955– 1961. 51. Heyland DK, Drover JW, MacDonald S, Novak F. Effect of post-pyloric feeding on gastroesophageal regurgitation and pulmonary microaspiration: results of a randomized controlled trial. Crit Care Med 2001; 29:1495–1501. 52. Galban C, Montejo JC, Mesejo A, et al. An immune enhancing enteral diet reduces mortality rate and episode s of bacteremia in septic intensive care unit patients. Crit Care Med 2000; 28:643–648. 53. Gadek J, DeMichele SJ, Karlstad MD, Pacht ER, et al. Effect of enteral feeding with eicosapentaenoic acid, gamma linolenic acid, and antioxidants in patients with acute respiratory distress syndrome et al. Crit Care Med 1999; 27:1409–1420. 54. Beale RJ, Bryg DJ, Bihari DJ. Immunonutrition in the critically ill: a systematic review of clinical outcome. Crit Care Med 1999; 27:2799–2805. 55. Heyland DK, Novak F, Drover JW, et al. Should immunonutrition become routine in critically ill patients? A systematic review of the evidence. JAMA 2001; 289:944–959. 56. National Nosocomial Infections Surveillance (NNIS). System Report, Data Summary from January 1992-June 2001, issued August 2001. Am J Infect Control 2001; 29:404–421. 57. Gil RT, Kruse JA, Thill-Baharozian MC, et al. Triple versus single lumen central venous catheters. A prospective study in a critically ill population. Arch Intern Med 1989; 149:1139– 1143. 58. Raad I, Umphrey J, Kahn A, et al. The duration of placement as a predictor of peripheral and pulmonary arterial catheter infections. J Hosp Infect 1993; 23:17–26. 59. Pittet D, Tarara D, Wenzel RP. Nosocomial bloodstream infection in critically ill patients. Excess length of stay, extra costs, and attributable mortality. JAMA 1994; 271:1598–1601. 60. Cobb DK, High KP, Sawyer RG, et al. A controlled trial of scheduled replacement of central venous and pulmonary-artery catheters. N Engl J Med 1992; 327:1062–1068. 61. Collignon PJ. Intravascular catheter associated sepsis: a common problem. The Australian Study on Intravascular Catheter Associated Sepsis. Med J Aust 1994; 161:374–378. 62. Coopersmith CM, Rebmann TL, Zack JE, et al. Effect of an education program on decreasing catheter-related bloodstream infections in the surgical intensive care unit. Crit Care Med 2002; 30:59–64. 63. Mermel LA, McCormick RD, Springman SR, Maki DG. The pathogenesis and epidemiology of catheter-related infection with pulmonary artery Swan-Ganz catheters: a prospective study utilizing molecular subtyping. Am J Med 1991; 91:197S–205S. 64. Mermel LA. Prevention of intravascular catheter-related infections. Ann Intern Med 2000; 132:391–402. 65. Raad II, Hohn DC, Gilbreath BJ, et al. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during inser-tion. Infect Control Hosp Epidemiol 1994; 15:231. 66. Raad I. Intravascular-catheter-related infections. Lancet 1998; 351:893–898.

Sepsis in the acute respiratory distress syndrome

397

67. Raad II, Hanna HA. Intravascular catheter-related infections: new horizons and recent advances. Arch Intern Med 2002; 162:871–878. 68. Randolph AG, Cook DJ, Gonzales CA, Brun-Buisson C. Tunneling short-term central venous catheters to prevent catheter-related infection: a metaanalysis of randomized, controlled trials. Crit Care Med 1998; 26:1452–1457. 69. Sherertz RJ, Ely EW, Westbrook DM, et al. Education of physicians-in-training can decrease the risk for vascular catheter infection. Ann Intern Med 2000; 132:641–648. 70. Sitzmann JV, Townsend TR, Siler MC, et al. Septic and technical complications of central venous catheterization. A prospective study of 200 consecutive patients. Ann Surg 1985; 202:766–770. 71. McKinley S, Mackenzie A, Finfer S, et al. Incidence and predictors of central venous catheter related infection in intensive care patients. Anaesth Intens Care 1999; 27:164–169. 72. Brun-Buisson C, Abrouk F, Legrand P, et al. Diagnosis of central venous catheter-related sepsis. Critical level of quantitative tip cultures. Arch Intern Med 1987; 147:873–877. 73. Merrer J, De Jonghe B, Golliot F, Lefrant JY, et al. Complications of femoral and subclavian venous catheterization in critically ill patients: a randomized controlled trial. JAMA 2001; 286:700–707. 74. Maki DG, Ringer M, Alvarado CJ. Prospective randomized trial of povidoneiodine, alcohol, and chlorhexidine for prevention of infection associated with central venous and arterial catheters. Lancet 1991; 338:339–343. 75. Maki DG, Stolz SS, Wheeler S, Mermel LA. A prospective, randomized trial of gauze and two polyurethane dressings for site care of pulmonary artery catheters: implications for catheter management. Crit Care Med 1994; 22: 1729–1737. 76. Conly JM, Grieves K, Peters B. A prospective, randomized study comparing transparent and dry gauze dressings for central venous catheters. J Infect Dis 1989; 159:310–319. 77. Maki DG, Botticelli JT, LeRoy ML, Thielke TS. Prospective study of replacing administration sets for intravenous therapy at 48- vs 72-hour intervals. 72 hours is safe and cost-effective. JAMA 1987; 258:1777–1781. 78. Eyer S, Brummitt C, Crossley K, Siegel R, Cerra F. Catheter-related sepsis: prospective, randomized study of three methods of long-term catheter maintenance. Crit Care Med 1990; 18:1073–1079. 79. Snyder RH, Archer FJ, Endy T, et al. Catheter infection. A comparison of two catheter maintenance techniques. Ann Surg 1988; 208:651–653. 80. Darouiche RO, Raad II, Heard SO, et al. A comparison of two anti-microbial impregnated central venous catheters. N Engl J Med 1999; 340:1–8. 81. Maki DG, Stoltz SM, Wheeler S, et al. Prevention of central venous catheter related bloodstream infection by use of an anti-septic impregnated catheter. A randomized controlled trial. Ann Intern Med 1997; 127:257–266. 82. Kamal GD, Pfaller MA, Rempe LE, et al. Reduced intravascular catheter infection by antibiotic bonding. A prospective, randomized, controlled trial. JAMA 1991; 265:2364–2368. 83. Maki DG, Band JD. A comparative study of polyantibiotic and iodophor ointments in prevention of vascular catheter-related infection. Am J Med 1981; 70:739–744. 84. Timsit JF, Sebille V, Farkas JC, et al. Effect of subcutaneous tunneling on internal jugular catheter-related sepsis in critically ill patients: a prospective randomized multicenter study. JAMA 1996; 276:1416–1420. 85. Eyer S, Brummitt C, Crossley K, Siegel R, Cerra F. Catheter related sepsis: prospective randomized study of three methods of long term catheter maintenance. Crit Care Med 1990; 18:1073. 86. Cobb DK, High KP, Sawyer RG, et al. A controlled trial of scheduled replacement of central venous and pulmonary artery catheters. N Engl J Med 192; 327:1062.

Acute respiratory distress syndrome

398

87. Eggimann P, Harbarth S, Constantin MN, Touveneau S, Chevrolet JC, Pittet D. Impact of a prevention strategy targeted at vascular-access care on incidence of infections acquired in intensive care. Lancet 2000; 355:1864–1868. 88. Bijma R, Girbes AR, Kleijer DJ, Zwaveling JH. Preventing central venous catheter-related infection in a surgical intensive-care unit. Infect Control Hosp Epidemiol 1999; 20:618–620. 89. Llewelyn M, Cohen J. Diagnosis of infection in sepsis. Intens Care Med 2001; 27:510–532. 90. Simon D, Trenholme G. Antibiotic selection for patients with septic shock. Crit Care Clin 2000; 16:215–231. 91. Kreger BE, Craven DH, McCabe WR. Gram-negative bacteremia. Reevaluation of clinical features and treatment of 612 patients. Am J Med 1998; 68:332–343. 92. Pittet D, Thjievent B, Wenzel RP, et al. Bedside prediction of mortality from bacteremic sepsis. A dynamic analysis of ICU patients. Am J Respir Crit Care Med 1996; 153:684–693. 93. Kollef MH, Sherman G, Ward S, et al. Inadequate antimicrobial treatment of infections: a risk factor for hospital mortality among critically ill patients. Chest 1999; 115:462–474. 94. Bartlett JG, Dowell SF, Mandell LA, et al. Practice guidelines for the management of community-acquired pneumonia in adults. Clin Infect Dis 2000; 31:347–382. 95. Mangialardi RJ, Martin GR, Bernard GR, Wheeler AP, Christman BW, Dupont WD, Higgins SB, Swindell BB. for the Ibuprofen in Sepsis Study Group. Hypoproteinemia predicts acute respiratory distress syndrome development, weight gain and death in patients with sepsis. Crit Care Med 2000; 28:3137–3145. 96. Martin GS, Mangialardi RJ, Wheeler AP, Dupont WS, Morris JA, Bernard GR. A randomized, controlled clinical trial of albumin and furosemide in hypoproteinemic patients with acute lung injury. Crit Care Med 2002. In press. 97. Carmichael LC, Dorinsky PM, Higgins S, Bernard GR, Dupont W, Swindell B, Wheeler AP. Diagnosis and therapy of adult respiratory distress syndrome: an international survey. J Crit Care 1996; 11:9–18. 98. Connors AF, Speroff T, Dawson NV, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA 1996; 276:889–897. 99. Thompson BT, Connors AF, Harabin AL, Hite D, Wheeler A, Weideman H. for the ARDS Clinical Trials Network. Feasibility and safety of the ARDS Network’s Fluid and Catheter Treatment Trial. Am J Respir Crit Care Med 2002; 165:A697. 100. Rackow EC, Falk JL, Fein IA, et al. Fluid resuscitation in circulatory shock: a comparison of the cardiorespiratory effects of albumin, hetastarch and saline solutions in patients with hypovolemic and septic shock. Crit Care Med 1983; 11:839–850. 101. Waikar SS, Chertow GM. Crystalloid versus colloids for resuscitation in shock. Curr Opin Nephrol Hypertens 2000; 9:501–504. 102. Schierhout G, Roberts I. Fluid resuscitation with colloid or crystalloid solutions in critically ill patients: a systematic review of randomized trials. BMJ 1998; 316:961–964. 103. Vincent JL. Hemodynamic support in septic shock. Intens Care Med 2001; 27(suppl):S80– S92. 104. Martin C, Viviand X, Leone M, et al. Effect of norepinephrine on the outcome of septic shock. Crit Care Med 2000; 28:2758–2765. 105. Desjars P, Pinaud M, Bugnon D, et al. Norepinephrine therapy has no deleterious renal effects in human septic shock. Crit Care Med 1989; 17:426–429. 106. Martin C, Papazian L, Perrin G, et al. Norepinephrine or dopamine for the treatment of hyperdynamic septic shock? Chest 1993; 103:1826–1831. 107. Meadows D, Edwards JD, Wilkins RG, et al. Reversal of intractable septic shock with norepinephrine therapy. Crit Care Med 1988; 16:663–666. 108. Bellomo R, Chapman M, Finfer S, et al. Low-dose dopamine in patients with early renal dysfunction: a placebo-controlled randomized trial. Lancet 2000; 356:2139–2143. 109. Landry DW, Olover JA. The pathogenesis of vasodilatory shock. N Engl J Med 2001; 345:588–595.

Sepsis in the acute respiratory distress syndrome

399

110. Landry DW, Levin HR, Gallant EM, et al. Vasopressin deficiency contributes to the vasodialtion of septic shock. Circulation 1997; 95:1122–1125. 111. Holmes CL, Walley KR, Grieve E, et al. Vasopressin infusion increases MAP and urine output in patients having severe septic shock. Am J Respir Crit Care Med 2000; 161:A879. 112. Shoemaker WC, Appel PL, Kram HB, et al. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest 1988; 94:1176–1186. 113. Tuchschmidt J, Fried J, Astiz M, et al. Elevation output and oxygen delivery improves outcomes in septic shock. Chest 1992; 102:216–220. 114. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high risk surgical patients. JAMA 1993; 270:2699–2707. 115. Yu M, Levy MM, Smith P, et al. Effect of maximizing oxygen delivery on morbidity and mortality rates in critically ill patients: a prospective randomized controlled study. Crit Care Med 1993; 21:830–838. 116. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal oriented hemodynamic therapy in critically ill patients. N Engl J Med 1995; 333:1025–1032. 117. Hayes MA, Timmins AC, Yau EHS, et al. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med 1994; 330: 1717–1722. 118. Rivers E, Nguyen B, Havstad, et al. Early goal directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345:1368–1377. 119. Corrigan JJ Jr, Ray WL, May N. Changes in the blood coagulation system associated with septicemia. N Engl J Med 1968; 279:851–856. 120. Esmon CT. The protein C pathway. Crit Care Med 2000; 28(suppl 9):S44–S48. 121. Vervolet MG, Thijs LG, Hack CE. Derangements of coagulation and fibrinolysis in critically ill patients with sepsis and septic shock. Semin Thromb Hemost 1998; 24:33–44. 122. Fourrier F, Chopin C, Goudemand J, et al. Septic shock, multiple organ failure, and disseminated intravascular coagulation: compared patterns of anti-thrombin III, protein C, and protein S deficiencies. Chest 1992; 101:816–823. 123. Lorente JA, Garcia-Frade LJ, Landin L, et al. Time course of hemostatic abnormalities in sepsis and its relation to outcome. Chest 1993; 103:1536–1542. 124. Powars D, Larsen R, Johnson J, et al. Epidemic meningococcemia and purpura fulminans with induced protein C deficiency. Clin Infect Dis 1993; 17: 254–261. 125. Fisher CJ Jr, Yan SB. Protein C levels as a prognostic indicator of outcome in sepsis and related diseases. Crit Care Med 2000; 28(9 suppl):S49–S56. 126. Taylor FB Jr, Chang A, Esmon CT, et al. Protein C prevents the coagu-lopathic and lethal effects of Escherichia coli infusion in the baboon. J Clin Invest 1987; 79:918–925. 127. Taylor FB Jr, Peer GT, Lockhart MS, et al. Endothelial cell protein C receptor plays an important role in protein C activation in vivo. Blood 2001; 15; 97(6):1685–1688. 128. Taylor FB, Wada H, Kinasewitz G. Description of compensated and uncompensated disseminated intravascular coagulation (DIC) responses (nonovert and overt DIC) in baboon models of intravenous and intraperitoneal Escherichia coli sepsis and in the human model of endotoxemia: toward a better definition of DIC. Crit Care Med 2000; 28(suppl):S12–S19. 129. Bernard GR, Ely EW, Wright TJ, et al. Safety and dose relationship of recombinant human activated protein C for coagulopathy in severe sepsis. Crit Care Med 2001; 29:2051–2059. 130. Natanson C, Esposito CJ, Banks SM. The sirens’ songs of confirmatory sepsis trials: selection bias and sampling error. Crit Care Med 1998; 26:1927–1931. 131. Moriyama S, Okamoto K, Tabria Y, et al. Evaluation of oxygen consumption and resting energy expenditure in critically ill patients with systemic inflammatory response syndrome. Crit Care Med 1999; 27:2133–2136. 132. Kudsk KA, Croce MA, Fabian TC, et al. Enteral vs parenteral feeding. Effects on septic morbidity after blunt and penetrating abdominal trauma. Ann Surg 1992; 215:503–511.

Acute respiratory distress syndrome

400

133. American College of Chest Physicians Consensus Statement. Applied nutrition in ICU patients. Chest 1997; 111:769–778. 134. ASPEN Board of Directors. Guidelines for the use of parenteral and enteral nutrition in adult and pediatric patients. J Parenter Enteral Nutr 1993; 17:1SA–26SA. 135. van den Berghe G, Wouters P, Weekersw F, et al. Intensive insulin therapy the surgical intensive care unit. N Engl J Med 2001; 345:1359–1367. 136. Pérez J, Dellinger RP. Other supportive therapies in sepsis. Intens Care Med 2001; 27:S116– S127. 137. Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest 2001; 119(suppl):132S–175S. 138. Keane MG, Ingenito EP, Goldhaber SZ. Utilization of venous thromboembolism prophylaxis in the medical intensive care unit. Chest 1994; 106:13–14. 139. Meyer MM. Renal replacement therapies. Crit Care Clin 2000; 16:29–58. 140. Schiffl H, Lang SM, Fischer R. Daily hemodialysis and the outcome of acute renal failure. N Engl J Med 2002; 346:305–310. 141. Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med 1999; 340:409–417. 142. Corwin HL, Gettinger A, Rodriguez RM, Pearl RG, Gubler KD, et al. Efficacy of recombinant human erythropoietin in the critically ill patient: a randomized, double-blind, placebo controlled trial. Crit Care Med 1999; 27:2346–2350.

17 Modulation of Pulmonary Vascular Tone in the Acute Respiratory Distress Syndrome UDO KAISERS, THILO BUSCH, MARIA DEJA, and KONRAD J.FALKE Charité, Campus Virchow-Klinikum Humboldt University Berlin, Germany HERWIG GERLACH Vivantes-Klinikum Neukoelln Berlin, Germany

I. Introduction An intriguing feature of the acute respiratory distress syndrome (ARDS) is marked maldistribution of pulmonary perfusion in favor of nonventilated, atelectatic, and edematous areas of the lungs, the cause of pulmonary right-to-left shunting and arterial hypoxemia. This indicates that the physiological response to hypoxia, hypoxic pulmonary vasoconstriction (HPV), an important regulatory mechanism to balance pulmonary perfusion between well- and poorly ventilated areas, is severely impaired. In contrast, vasoconstriction occurs in other, relatively normal parts of the lungs, possibly contributing to an increased shunt perfusion as well as an elevated dead space fraction. In fact, pulmonary vascular resistance (PVR) is usually increased, especially in more progressed phases of ARDS when vascular obliteration takes place, leading to pulmonary arterial hypertension as a predominant feature. Since an increase in pulmonary artery pressure can be a driving force for the development of pulmonary edema and impairs loading conditions of the right ventricle, an important goal in the therapy of ARDS has been to lower pulmonary artery pressure (PAP) and PVR. In contrast to the

Modulation of pulmonary vascular tone in the acute respiratory

Figure 1 Discontinuation of PEEP (10 cmH2O) and inhaled NO (10 ppm) demonstrated similar detrimental effects on arterial oxygenation in acute lung injury. (a) values (means) in 8 patients with ALI submitted to mechanical ventilation (MV) at an of 1.0 with PEEP 10 cmH2O. Upon withdrawal of PEEP (zero endexpiratory pressure, ZEEP) instantly dropped, slowly recovering

403

Acute respiratory distress syndrome

404

after PEEP 10 cmH2O was resumed, indicating recruitment of alveolar space due to MV with PEEP. (b) values in one ALI patient inhaling 10 of 1.0. ppm NO at an Discontinuation of NO instantaneously decreased in a comparable magnitude when compared to withdrawal of PEEP. However, improvement of after restart of inhaled NO is more rapidly. The dynamics of the effects of PEEP and inhaled NO on indicate the potential of both therapies to acutely improve arterial oxygenation. pulmonary and systemic effects of intravenously administered vasodilators, inhalation of the endogenous vasodilator nitric oxide (NO) selectively dilates the pulmonary vasculature (1, 2). When inhaled NO was used in ARDS, it appeared that PAP and PVR could be lowered and arterial oxygenation could be improved, indicating that the maldistribution of pulmonary perfusion was reduced (3, 4). This redistribution of pulmonary blood flow by selectively acting inhaled vasodilators such as NO and prostacyclin (5, 6) represented a new approach to reduce pulmonary right-to-left shunting, an alternative to ventilatory strategies such as positive end-expiratory pressure (PEEP), that are designed to recruit closed alveolar spaces (7). In fact, the effects of are surprisingly similar, but the mechanisms of actions PEEP and inhaled NO on are completely different (Fig. 1). The redistribution of pulmonary blood flow to intact areas of the lungs due to inhaled NO can, however, only take place in the presence of some degree of pulmonary vascular constriction. In sepsis or in certain forms of chronic liver disease, the pulmonary vessels may be maximally dilated and inhaled vasodilators may have no beneficial effect on pulmonary oxygenation. Hence, it should be possible to optimize oxygenation by pharmacological restoration of the balance between pulmonary vasodilatation and constriction. For this purpose, the combination of inhaled NO with intravenous vasoconstrictors has been advocated (8). Furthermore, the effects of inhaled vasodilators can probably be modulated by the inhibition of specific phosphodiesterases that increase cyclic guanosine monophosphate (cGMP) in smooth muscle cells (9). Although a few large multicenter trials have failed to show an overall beneficial effect of inhaled NO in ARDS, a new door has been opened in the management of vascular tone and gas exchange in ARDS. In this context it is also of considerable interest that gaseous NO is formed in the upper airways, in particular in the paranasal sinuses, and is autoinhaled, possibly affecting pulmonary vessels, ciliary function, and non-specific immune defense

Modulation of pulmonary vascular tone in the acute respiratory

405

(10–12). These various aspects of the inhalation of vasodilators and the new possibilities for modulating pulmonary vascular tone in ARDS will be discussed in this chapter.

II. Vasodilators in ARDS ARDS is characterized by pulmonary and endothelial inflammation that induces permeability edema of the lung, loss and dysfunction of surfactant with atelectasis and reduction in pulmonary compliance, ventilation-to-perfusion mismatch associated with intrapulmonary right-to-left shunt causing hypoxemia, and pulmonary arterial hypertension (13). In ARDS, pulmonary arterial hypertension with an increase in PVR often occurs in the setting of lowered SVR, typical for sepsis and the systemic inflammatory response syndrome (SIRS) (14, 15). The pulmonary vascular changes in ARDS are due to diminished hypoxic pulmonary vasoconstriction in shunt areas occurring with vasoconstriction in well-ventilated regions, cytokine-mediated inflammatory response, intravascular coagulation, and consequences of therapeutic interventions such as oxygen toxicity and barotrauma (16). Pulmonary arterial hypertension may increase micro vascular filtration pressure by an increase in pulmonary capillary pressure that aggravates the degree of permeability edema. Moreover, pulmonary arterial hypertension alters right ventricle (RV) loading conditions and may cause RV dysfunction (14, 17). A. Intravenously Administered Systemic Vasodilators Until a decade ago, the usual recommendation was to administer vasodilators (e.g., sodium nitroprusside and prostaglandins) systematically in ARDS patients for the treatment of pulmonary arterial hypertension and for unloading the right ventricle (18), a treatment that induced a significant increase in cardiac output. This approach was reported to significantly improve survival in a single center study with 41 ARDS patients (19), but the results of this study could recently not be confirmed (20). Subsequently, it was shown that the benefit for the pulmonary circulation was at the cost of gas exchange, since the simultaneous dilation of systemic and pulmonary vessels increases cardiac output and intrapulmonary right-to-left shunt (21– 23). The relationship between cardiac output and shunt perfusion was first demonstrated by Dantzker et al. in 1980 (24). The authors had shown that an increase in cardiac output was associated with an increase in shunt perfusion in acute lung injury. B. Nitric Oxide: A Selective Pulmonary Vasodilator in ARDS The shortcomings of intravenously administered vasodilators in ARDS patients prompted the search for a better, more selective pulmonary vasodilator. This agent should selectively decrease pulmonary resistance without influencing systemic circulation and cardiac output. Identification of the endothelial-derived relaxing factor (EDRF) as nitric oxide (NO) in 1987 (25, 26) provided a rational basis for the understanding of the mechanism of action of nitro-vasodilators. In particular, it was suggested that gaseous NO itself might have vasodilation properties. In experimental studies of alveolar hypoxia

Acute respiratory distress syndrome

406

it was demonstrated that inhaled NO selectively vasodilates pulmonary vessels without causing systemic effects (1, 28); this was further validated in awake healthy volunteers (27) and in patients with chronic pulmonary hypertension (2). The effect of inhaled NO is limited to ventilated lung regions since the gas is rapidly bound and inactivated by contact with hemoglobin (29). Administration of inhaled NO in ARDS was first reported by Falke in Berlin in a collaboration with Zapol (3) (Fig. 2), and later by Rossaint et al. in 1993 (4). The authors compared the effects of inhaled NO at 18 ppm with intravenous prostacyclin at 4 ng/kg. Inhalation of NO reduced mean pulmonary artery pressure (MPAP) from 37±3 to 30±2 mmHg without significant changes in mean arterial pressure and cardiac output. Intravenous prostacyclin induced an almost identical reduction in MPAP, but mean systemic arterial pressure was also significantly reduced from 86±4 to 79± 6 mmHg, and cardiac output increased from 7.6±0.9 to 8.8±1.1 L/min, indicating nonselectivity of intravenous prostacyclin. The most striking finding was, however, that inhalation of NO from 152±15 to 199±23 mmHg because of induced a significant increase in arterial a reduction in right-to-left shunt. This effect was explained by a selective vasodilation of ventilated lung regions by inhaled NO, i.e., redistribution of blood flow from nonventilated to ventilated areas of the lung. This report initiated the clinical use of in-

Figure 2 First clinical application of inhaled NO in a patient with ARDS (43-year-old female). Inhaled NO at 18 and 36 ppm induced selective pulmonary vasodilation. In contrast, intravenously administered PGI2

Modulation of pulmonary vascular tone in the acute respiratory

407

demonstrated nonselective vasodilation with a concomitant increase in intrapulmonary right-to-left shunt (Qs/ Qt, measured by SF6 retention). haled NO in ARDS, and subsequently, there have been numerous observational investigations that reported beneficial effects in ARDS patients. In an experimental study, inhaled NO was shown to be equivalently effective in reducing PVR in small arteries, capillaries, and veins (30, 31). Benzing and coworkers (32) found a predominant vasodilating effect of inhaled NO on pulmonary venous vasculature in patients with acute lung injury (ALI), thereby lowering pulmonary capillary pressure and possibly reducing edema formation. In an additional study, Benzing et al. (33) evaluated the effects of inhaled NO and transvascular albumin flux in 9 patients with acute lung injury using a double radioisotope method. They reported that short-term inhalation of NO induced a decrease in pulmonary artery pressure and reduced transvascular albumin flux. In subsequent investigations, inhaled NO reduced pulmonary artery hypertension, unloaded the right ventricle, and improved right ventricle ejection fraction in ARDS patients (34– 36). C. Adjuncts to Inhaled NO in ARDS Inhaled NO is effective in ARDS in conjunction with different therapeutic strategies and numerous clinical studies aimed at evaluating potential additive beneficial effects of combined treatments. Puybasset et al. (37) studied 21 patients with ARDS and investigated the effects of PEEP during inhalation of NO. In patients in whom moderate PEEP induced alveolar recruitment, inhaled NO significantly improved however, in patients who did not respond to PEEP, NO did not improve oxygenation. Using PEEP levels of 5 and 10 cmH2O in ARDS patients receiving inhaled NO at 4 ppm, Okamoto and coworkers reported additive effects on oxygenation for the lower PEEP value and synergistic effects for the higher PEEP level (38). In a single center observational study, Papazian and colleagues (39) evaluated hemodynamic and respiratory effects of inhaled NO and prone position in 14 patients with ARDS. The authors found a significant and and shunt fraction. Various additive effect of NO and prone position on studies have also confirmed an additive effect of inhaled NO and prone positioning of ARDS patients (40–42). A synergistic action of both treatments, however, could not be demonstrated (43). In particular, a recent clinical study revealed no correlation between a response to NO and response to prone positioning (44), suggesting that both interventions should be considered in circumstances when gas exchange is severely impaired. Recently, other therapeutic interventions aiming at alveolar recruitment and optimization of gas exchange have been studied in conjunction with inhaled NO: exogenous surfactant and partial liquid ventilation (PLV) (45– 47). In experimental models of ALI, these studies demonstrated that con-ditioning the lung by these treatments might augment the efficacy of inhaled NO on gas exchange.

Acute respiratory distress syndrome

408

D. Response to Inhaled NO The results of Rossaint and coworkers (4) were confirmed by several small prospective cohort studies in adults and children with ARDS (48–54). These studies were primarily designed to investigate short-term as well as long-term effects of NO in terms of hemodynamics and gas exchange, but were not powered to validate meaningful clinical endpoints such as mortality rate and ventilator-free days. All of these studies revealed beneficial effects of inhaled NO on oxygenation, but they also demonstrated that the response to inhaled NO was not uniform and that clinically relevant improvements were not obtained in a considerable subset of ARDS patients. Two retrospective studies with a larger sample size were carried out to quantify the fraction of NO responders among ARDS patients (55, 56). Since conclusive criteria for a response to NO are not defined, the classification largely depends on the choice of certain thresholds that were considered clinically important by different investigators. In 30 ARDS patients receiving 10–20 ppm NO, Rossaint and coworkers reported response rates of 83, 87, and 63%, when related to improvements in arterial oxygenation gas exchange (∆Qva/Qt≥10%), or a decrease in MPAP (≥3 mmHg), respectively. Manktelow and coworkers (56) summarized efficacy data when administering NO in concentrations between 5 and 80 ppm in 88 ARDS patients. In 59% of the NO trials, they found an improvement of more than 20% in either PVR, or in both parameters. In a subgroup analysis, only 33% of ARDS patients with septic shock responded to inhaled NO compared to a 64% response rate in ARDS patients without septic shock. An understanding of the physiological determinants of the NO response is still incomplete. The situation is further complicated by the fact that the efficacy of inhaled NO depends mainly on the concentration that is used. Dose-response characteristics for gas exchange were different from those for pulmonary arterial pressure and pulmonary vascular resistance (57, 50). In responders to NO, pulmonary arterial pressure decreased steadily but in a nonlinear pattern with increasing NO concentration. In contrast, the maximum effect on arterial oxygenation was obtained in a range between 1 and 10 ppm (57). This possibly indicates a loss of selectivity in respect to ventilated parts of the lungs at higher doses due to an increased diffusion and subsequent vasodilation in nonventilated lung regions (58) (Fig. 3). Only a few studies have been done to investigate the factors that determine NO response. Data from ARDS patients inhaling various doses

Modulation of pulmonary vascular tone in the acute respiratory

409

Figure 3 On-line registration of the (Parairend® time course of combined with DocVue®, Hewlett Packard, Böblingen, Germany) and mean pulmonary artery pressure (MPAP) in a patient with acute lung injury inhaling different doses of NO. Maximum effects on oxygenation and on MPAP were reached at different doses of NO, indicating separate doseresponse characteristics for pulmonary vasodilation and improvements in gas exchange. The computers used to record and MPAP were not synchronized, explaining the apparent time shift of 5 minutes between the curves. suggested that changes in pulmonary artery pressure and pulmonary vascular resistance were related to baseline values of these parameters prior to administration of NO (50, 56). The improvement in arterial oxygenation was directly related to the pretreatment values of pulmonary vascular resistance (52, 56). These findings suggest that the beneficial effects of inhaled NO in ARDS on gas exchange depend on the existence of a certain degree of vasoconstriction in ventilated lung areas and that vasodilation may be limited to these lung regions. These two conditions are necessary but not sufficient. A third requirement is the presence of intrapulmonary right-to-left shunt. Investigations in nonARDS patients with pulmonary hypertension provide arguments for this additional condition. Thus, inhaled NO significantly reduced pulmonary arterial pressure in

Acute respiratory distress syndrome

410

mountaineers susceptible to high-altitude pulmonary edema at an altitude of 4559 m, but arterial oxygenation improved only if edema had developed (59). Similar results were reported in COPD patients characterized by pulmonary hypertension and an impaired gas exchange due to an increased fraction of low Va/Q. Interestingly, in these patients shunt fraction remained low and inhaled NO decreased pulmonary artery pressure but slightly worsened arterial oxygenation (60). Vasoactive mediators that circulate in the blood or are generated directly within the lung tissue are expected to influence pulmonary vascular tone and thus the effects of inhaled NO. The hyporesponsiveness to NO in ARDS patients with septic shock (56) is considered to be a consequence of an increased endogenous NO synthesis. Experimental models of endotoxin-induced sepsis have demonstrated an increase in expression of inducible NO synthase (iNOS), and blockade of iNOS by selective inhibitors restored pulmonary vascular reactivity to inhaled NO to normal (61, 62). A particularly powerful endogenous vasoconstrictor is endothelin-1 (ET-1) (63, 64). The lung has a high capacity for production and clearance of ET-1, and increased plasma levels were measured in ARDS patients (65–67). Immunohistochemical analysis of lung tissue in patients who succumbed to ARDS revealed increased ET-1 levels in lung vascular endothelium, airway epithelium, smooth muscle cells, and alveolar macrophages when compared to the lungs of patients who died without ARDS (68). Using an experimental model of acute lung injury, our group recently demonstrated that improvements in arterial oxygenation induced by inhaled NO were significantly related to the level of the plasma ET-1 concentration (69). Furthermore, in this animal model, inhaled NO reduced ET-1 plasma levels compared to untreated controls. This negative feedback mechanism may indicate that lower NO concentrations can become more effective during long-term treatment with NO. To clarify the dose-response characteristics during long-term inhalation of NO, our group performed a randomized, controlled, single center trial enrolling 40 patients with severe ARDS (70). Patients were randomized to receive long-time inhalation of 10 ppm NO or no continuous inhaled treatment. Dose-response curves ranging from 0.01 to 100 ppm NO registered hemodynamics and gas exchange at regular intervals in both groups. Initially, inhaled NO induced comparable dose-response characteristics in both groups. After 4 days of treatment, however, the NO group showed a left shift in the doseresponse curve with a maximum effect on arterial oxygenation at 1 ppm compared to 10 ppm after 2 days. This effect was not registered in the control group. We found no differences in mortality rate, length of stay, and days on mechanical ventilation between groups. Nevertheless, the requirement of extracorporeal membrane oxygenation (ECMO) therapy was significantly lower in the group continuously inhaling NO. A diminished response to inhaled NO in ARDS patients is not a static phenomenon. Although systematic investigations have not yet been performed, some observations indicate that patients who did not respond initially may become responders during the further course of treatment (50, 54, 55, 71). In some cases this seems to be associated with treatment modalities that provide alveolar recruitment, such as prone positioning (42) and higher levels of PEEP (37). The underlying mechanism that accounts for this increase in NO efficacy is still not determined. It can be hypothesized that the morphological changes due to alveolar recruitment primarily occur at the boundary between ventilated lung regions and non-ventilated, consolidated lung units. The

Modulation of pulmonary vascular tone in the acute respiratory

411

influence of the regional blood flow redistribution along this boundary layer on NO responsiveness remains to be explored. E. Side Effects of Inhaled NO Nitric oxide that is produced in the lung or administered by inhalation can be potentially cytotoxic to lung tissue. Lung macrophages stimulated by bacterial lipopolysaccharides produce significant quantities of NO, which are involved in various antioxidant and antiinflammatory reactions as well as in the evolution of altered pulmonary vascular tone in septic shock (72, 73). Nitric oxide is a radical gas, soluble in both water and lipid (74). The molecule is freely diffusible and forms covalent bonds quite easily (75, 76). The reactivity of NO is dependent on its redox properties, reacting with molecular oxygen to form nitrogen dioxide (NO2) (74, 76, 77). Animals inhaling >10 ppm NO2 developed pulmonary edema and hemorrhage, surfactant dysfunction, and even death (78). In humans, short-term inhalation of 2 ppm NO2 increased alveolar permeability (79). Nitric oxide reacts rapidly with the superoxide anion (O2−) to form peroxynitrite (−OONO), a strongly oxidizing molecule with a long half-life, being substantially more toxic than NO itself (80–83). Under physiological conditions superoxide is naturally captured by specific scavenger systems like superoxide dismutase, rendering formation of peroxynitrite minimal. In disease states with increased amounts of superoxide, the scavenging reaction may well be exhausted, producing significant concentrations of peroxynitrite (84). Peroxynitrite generates hydroxyl radicals in tissue even under physiological conditions and directly causes oxidation and nitration of lipids, proteins, and DNA (85). Recently it was shown that peroxynitrite and adventitious CO2 form nitrosoperoxocarbonate (ONOOCO2), and it was postulated that generation of these species might account for most of the peroxynitrite-mediated nitration reactions (86, 87). Hypercapnia appears to further accelerate this pathway, subsequently causing alveolar epithelial injury (88). Formation of 3-nitrotyrosine (NO2Tyr) by peroxynitrite has been used as a probe for NO mediated oxidation in vivo (89, 90). In patients with ARDS, NO2Tyr content in bronchoalveolar lavage (BAL) was increased, indicating that oxidative stress mediated by NO is involved in the pathogenesis of ALI and ARDS (91). However, this concept is disputed, and it was suggested that other endothelium-derived hyperpolarizing factors, but not NO, might play a pivotal role in the vascular damage seen in ARDS (92). To conclude, generation of reactive oxygen species on the one hand and the capacity of the scavenging systems on the other hand determine whether biologically harmful amounts of peroxynitrite will be generated (93). Conversely, reactive oxygen species (e.g., O2−, H2O2) themselves can aggravate lung injury (94, 95), and binding of these compounds by NO may therefore provide tissue protection. Accordingly, in a model of hyperoxic lung injury, Garat and coworkers (96) demonstrated that inhalation of 10, but not 100 ppm NO prevented oxygen radicalinduced pulmonary edema. At present, the in vivo role of NO as a free radical or antioxidant remains to be determined (97). In plasma, NO is converted to nitrite and S-nitrosothiols, nitrite being rapidly oxidized by hemoglobin and excreted in the urine. Within the erythrocyte, NO is promptly bound to oxyhemoglobin to form nitrosylhemoglobin. Consecutively, methemoglobin (MetHb) is produced by oxidizing the heme iron (Fe2+ to Fe3+) and NO3− is released (98). Exposure

Acute respiratory distress syndrome

412

to NO produces mainly MetHb, and increased concentrations were observed following accidental overdosing (99); however, in the majority of patients, methemoglobinemia has not been problematic. In adults who inhale 40 and 80 ppm NO, few patients with MetHb levels of >5% were reported (100). In addition to inactivation of NO by hemoglobin, it was recently suggested that hemoglobin might even provide a carrier function for NO (101). On the presently available evidence, the above discussed side effects of NO are dose-dependent. Specifically, MetHb concentrations of >5% occurred in patients receiving high doses of inhaled NO, and it is therefore recommended to use doses of ≤10 ppm. Nitric oxide interferes with platelet function (102), and inhaled NO was shown to prolong bleeding time and to reduce platelet aggregation in patients with ARDS (103). Being essential for hemostasis, platelet aggregation may well promote generation of pulmonary microthrombosis and may thus contribute to pulmonary hypertension in ARDS (104, 105). Nevertheless, anticoagulant effects should be considered when inhaled NO is used in patients who have a high risk of bleeding complications, especially premature newborns or postsurgical patients. In patients with ARDS, currently available evidence does not indicate an increase in bleeding complications associated with inhalation of NO. F. Randomized Controlled Clinical Trials on Inhaled NO in Adult Patients with ARDS Numerous studies have consistently shown that inhaled NO improves gas exchange and hemodynamics if patients serve as their own controls. In this section we will discuss the results from randomized controlled trials (RCTs) on inhaled NO in adult patients with ARDS. Available RCTs on NO in neonates with respiratory failure will be discussed below in a separate paragraph. Two unblinded, pilot, randomized controlled single center studies with small sample sizes [n=40 (106) and n=30 (107)] compared inhaled NO treatment and conventional therapy in adult ARDS. Both investigations found only transient improvements in arterial oxygenation during the first 24 hours of treatment, documented either by an increase in (106, 107) or by a decrease in venous admixture (107). There was no effect of inhaled NO on mortality rates within this limited sample size. The results of the randomized double-blind placebo-controlled phase II trial in the United States by Dellinger and coworkers (100) stimulated discussion. The aim of this study was to assess safety issues and physiological effects of various inhaled NO doses, and it was not powered to demonstrate a statistically significant benefit in any outcome parameter. A total number of 177 patients who fulfilled the criteria of early ARDS in accordance with the American European Consensus Conference criteria (108) for less than 72 hours prior to randomization were enrolled in 30 hospitals. Patients with sepsis, severe burns of larger surface area, persistent hypotension, and multisystem organ failure were excluded. This was done by the investigators to reduce the frequency of conditions in which mortality and duration of mechanical ventilation were unlikely to be altered by an improvement in lung function alone. Patients were randomized to receive either inhaled NO at concentrations of 1.25, 5, 20, 40, or 80 ppm (n=120) or placebo gas (n= 57). There were no adverse effects except an increase in methemoglobin concentration

Modulation of pulmonary vascular tone in the acute respiratory

413

after application of 80 and 40 ppm NO, a range usually not required in ARDS. In the treatment group, ARDS patients had a 60% response rate for a change in greater than 20%. Effects on arterial oxygenation and hemodynamics were significant only in the initial phase of NO inhalation. For example, the authors reported transient and a decrease in pulmonary artery pressure during the first 24 improvements in hours, as well as in the oxygenation index during the first 4 days of inhaled NO treatment. Pooled data from all ARDS patients receiving inhaled NO at any of the various doses revealed no significant difference in the number of days alive and off mechanical ventilation at 28 days when compared to the control group. However, according to a post hoc analysis, a significantly higher percentage of patients was alive and off mechanical ventilation in the 5 ppm inhaled NO group as compared to controls. Based on an intention-to-treat analysis, mortality rate was 30% in both the placebo and the inhaled NO group. Preliminary results of the prospective unblinded randomized European multicenter phase III trial on inhaled NO in acute lung injury were reported by Lundin and coworkers in 1997 (109), and the study has meanwhile been published in full (110). The primary endpoint was reversal of acute lung injury. Clinical outcome and safety were assessed as secondary objectives. Originally powered for a sample size of n=600, the study was stopped after inclusion of 286 patients with acute lung injury because of slow recruitment. After initial application of inhaled NO in doses of 2, 10, and 40 ppm for 10 minutes, only responders (63%; n=180) were randomized to either inhale NO at the lowest effective dose (n=93) or to be treated conventionally (n= 87) for 30 days. The study endpoint was reached either with a reversal of acute lung injury or when criteria for severe respiratory failure (correspond-ing to slow and fast ECMO entry criteria) occurred. The response in hemodynamics and arterial oxygenation to NO was only reported for a single time point prior to randomization, and data during the further course of the study were not presented. Reversal of acute lung injury occurred in 61% of the patients receiving inhaled NO and in 54% of the controls. The difference was not significant. However, inhaled NO treatment significantly reduced the frequency of severe respiratory failure (inhaled NO group: 2.2%; controls: 10.3%; p<0.05). In contrast to the U.S. phase II trial, acute renal failure was identified as a possible adverse effect of NO inhalation. This finding is unexplained and did not occur in other clinical studies. Mortality rates did not differ significantly between groups at 30 days (inhaled NO: 45%; controls: 38%; nonresponders: 45%). In our opinion the following conclusions can be drawn from the RCTs on NO treatment: 1. Inhaled NO improves arterial oxygenation and hemodynamics in most ARDS patients in the acute phase and may facilitate the use of a more lung protective mechanical ventilation over several days. 2. Inhaled NO probably does not induce clinically relevant adverse effects. 3. The effects of NO are dose dependent, and lower doses of ≤10 ppm seem to be preferable. 4. The available RCTs do not demonstrate a reduction in mortality by inhaled NO in a population of patients representing the entire range of severity of acute lung injury and early ARDS.

Acute respiratory distress syndrome

414

5. The finding that is identical in controls and the treatment groups that received NO after an initial phase may, in part, be due to inclusion of a considerable percentage of patients in both groups whose lung condition could be improved, regardless of the type of treatment. The published RCTs do not conclusively resolve the question whether a subgroup of severely hypoxemic ARDS patients will respond favorably to NO with an increase in survival rate, more ventilator-free days, or a reduced need for ECMO. 6. In ARDS patients with severe and refractory hypoxemia, inhaled NO is a feasible rescue treatment. This interpretation is essentially supported by a recent meta-analysis of RCTs on inhaled NO therapy reported by Sokol and colleagues (111). G. Clinical Studies in Neonates Persistent pulmonary hypertension of the neonate (PPHN) is a feared disorder characterized by a failure of the pulmonary vascular resistance of appropriately decrease during the transition to extrauterine life leading to myocardial dysfunction and extrapulmonary right-to-left shunting across the foramen ovale and the ductus arteriosus with concomitant hypoxemia. Conventional treatment of this condition has not been very effective (112). In full-term or near-term neonates, selective pulmonary vasodilation by inhaled NO significantly reduced the need for ECMO in hypoxic respiratory failure and PPHN. The Neonatal Inhaled Nitric Oxide Study (113) group carried out a randomized, placebo-controlled trial enrolling 235 full-term or near-term neonates with hypoxic respiratory failure due to a variety of reasons (PPHN, meconium aspiration, pneumonia, sepsis). Infants <14 days of age with an mean of 46 mmHg were randomized to inhale NO at 20 ppm (n=121) or to receive pure oxygen (controls, n=114). Although mortality by 120 days of age was not different (14% and 17%, respectively), significantly fewer neonates in the NO group required ECMO compared to the control group (31% vs. 55%; p<0.0001) (113). These results were repeatedly confirmed, and according to a Cochrane Review of 12 eligible randomized controlled trials, oxygenation improved in 50% of full-term or near-term neonates receiving NO, and the oxygenation index (OI) decreased; however, mortality rates were not reduced (114). The results of the clinical studies in term and near-term infants prompted investigations to evaluate the efficacy of inhaled NO in preterm neonates with PPHN and hypoxemia. A randomized, controlled trial published by the Franco-Belgium Collaborative NO Trial Group including 204 preterm (<33 weeks) and near-term (>33 weeks) hypoxemic neonates (OI 12.5–30.0 and 15–40, respectively) demonstrated that inhalation of 10 ppm NO improved oxygenation in near-term but not preterm infants and decreased the days on mechanical ventilation in near-term neonates who survived (115). Kinsella and coworkers (116) presented a double-blind, multicenter, randomized controlled trial enrolling 80 preterm neonates (<34 weeks) with severe hypoxemia, 48 of whom received inhaled NO. Again, oxygenation was significantly improved after 60 minutes in the NO group compared to controls, and survivors in the NO group spent fewer days on mechanical ventilation. The frequency of adverse events in preterm neonates inhaling NO (e.g., intracranial hemorrhage or chronic lung disease) was not

Modulation of pulmonary vascular tone in the acute respiratory

415

increased compared to the control group. However, in both studies mortality did not significantly differ between treatment groups. Currently, inhaled NO for preterm neonates is still considered an experimental therapy (117). Despite the lack of efficacy in terms of mortality, this treatment option for the most critically ill preterm neonates merits further study. A recent investigation by the NINOS group reporting a comprehensive neurodevelopmental follow-up of infants enrolled in the NINOS trial demonstrated that inhalation of NO was not associated with an increase in neurodevelopmental, behavioral, or medical abnormalities at 2 years of age (118). H. Status of Licensing of Inhaled NO Inhalation of NO significantly reduced the requirement of ECMO in hypoxic neonates with signs of pulmonary hypertension (113). On the evidence presently available, inhalation of NO has been approved for the treatment of term and near-term neonates (>34 weeks) with hypoxic respiratory failure associated with pulmonary hypertension by the U.S. FDA (approval no. 020845, date 12/23/1999) and the European Commission (EU/1/01/194/001, 8/1/2001). In adult ARDS patients, randomized controlled trials are currently lacking power to demonstrate a reduction in the necessity for ECMO therapy by inhaled NO. Therefore, inhaled NO is presently only feasible as a rescue therapy for severe refractory hypoxemia and decompensated pulmonary hypertension in adults with ARDS (“off-label use”).

III. Endogenous Nitric Oxide in the Respiratory Tract The epithelial production and availability of gaseous NO in the respiratory tract is altered in ARDS and may be important in the regulation of pulmonary vascular tone. For this reason, the role of endogenous gaseous NO is considered in this section. A. Nitric Oxide Synthases NO is endogenously produced in a wide variety of cells within the respiratory tract (119). The formation of NO by a five-electron oxidation of the terminal guanidino group of the substrate 1-arginine is catalyzed by a family of enzymes called NO synthases (NOS) (120). Neuronal and endothelial NOS (nNOS and eNOS) are constitutively expressed, and their activity depends on intracellular calcium concentration. These constitutive isoforms were localized in the surface epithelium of the respiratory tract. In particular, eNOS is present in the vascular endothelium and is thought to modulate pulmonary vascular reactivity to vasoactive substances. Accordingly, in healthy volunteers, administration of the NO synthase antagonist NG− monomethyl-1-arginine (L-NMMA) during hypoxia increased pulmonary artery pressure and vascular resistance (121). Lung tissue specimens collected from open lung biopsy, lung transplantation, and autopsy in patients with pulmonary hypertension revealed diminished expression of eNOS (122). The third isoform, termed inducible nitric oxide synthase (iNOS), is also found in the entire respiratory tract epithelium. The abundance of iNOS in the lower respiratory tract of healthy subjects is low, but pro-inflammatory cytokines such as TNF-α and IL-1-β

Acute respiratory distress syndrome

416

have the potential to induce the expression of substantial amounts of iNOS. An increased in iNOS expression is considered as a primary reason for the impaired vascular reactivity in the acute phase of sepsis. Septic patients have increased nitrite and nitrate plasma levels, which cannot be explained by renal dysfunction alone (123). In experimental models of endotoxin-induced sepsis, iNOS-deficient mice had higher survival and better maintenance of systemic blood pressure compared to wild-type mice (124). There is evidence from in vitro and experimental in vivo studies that certain cytokines, such as IL4, IL-6, IL-10, IL-11, IL-13, and in particular transforming growth factor β-1 (TGF-β-1), are able to reduce iNOS expression (125). A decrease in pro-inflammatory cytokines and an increase in anti-inflammatory cytokines in the later phase of sepsis are possibly responsible for returning iNOS expression to normal (126). B. Nitric Oxide Formation in the Lower Respiratory Tract NO is endogenously released into the respiratory tract and can be detected in exhaled gas of various mammals and humans (127, 128). In healthy humans, the main contribution to exhaled NO results from the upper respiratory tract. NO concentrations of several hundred parts per billion (ppb) have been measured within the nasal cavity (10, 129). The NO release from the lower respiratory tract is considerably lower. In intubated patients without airway infection, expired NO concentrations below 5 ppb were reported (10, 130). Similar values were measured in mixed expired gas of spontaneously breathing healthy volunteers when admixture of nasal NO was prevented by different techniques (129, 131). Higher expired NO concentrations result from low flow rates, and it is therefore recommended to use low flow rates in connection with positive airway pressure to close the velum in order to avoid nasal contamination when studying patients who are able to cooperate (132). An attempt to quantify the NO fraction from the pulmonary vessels gave disappointing results (133). Inhalation of the NOS inhibitor L-NMMA in healthy adult subjects reduced NO release from the lower respiratory tract by approximately 40% without causing any systemic effects. In contrast, intravenous administration of L-NMMA significantly increased heart rate and blood pressure, but decreased exhaled NO only by 10%. These results indicate that in healthy subjects, most of the NO release into the lower airways and alveoli result from the surface epithelium rather than from the pulmonary vascular endothelium (133). Intravenous administration of the selective iNOS inhibitor aminoguanidine in healthy subjects did not significantly influence lower respiratory tract NO, demonstrating predominant activity of constitutive NOS (134). Inflammatory processes with consecutive release of cytokines and lipopolysaccharides can induce iNOS expression. Consequently, inflammation that affects the lung interstitium and lower airway epithelium is associated with an increase in iNOS expression. Accordingly, increased levels of exhaled NO have been reported in asthmatics (135), in patients with fibrosing alveolitis (136), bronchiectasis (137), lymphocytic bronchiolitis (138), tuberculosis (139), and pneumonia (140). Exhaled NO concentrations in ARDS patients were significantly lower compared to controls (141), paralleling findings in patients with cystic fibrosis (142). The increased formation of nitrotyrosine in the lung tissue of ARDS patients, even when inhaled NO is not administered, provides evidence for increased NO synthesis (89). It has been suggested

Modulation of pulmonary vascular tone in the acute respiratory

417

that this might result from (1) the absorption of NO in edema fluid and (2) the rapid scavenging reaction with superoxides released from activated alveolar macrophages (141). In addition, the decrease in functional residual capacity due to the collapse of small airways in the ARDS lung has to be considered. Hence, the alveolar surface area of the ARDS lung that participates in gas exchange is decreased and large amounts of NO that may be formed in the inflamed parts of the lungs cannot enter into the airspaces. C. Nitric Oxide Formation in the Upper Respiratory Tract The high nasal NO levels originate from an intense production within the paranasal sinuses from where the gas diffuses away through the sinus ostia (11). NO concentrations within the sinus maxillaris of healthy adults exceed several ppm (11, 143), indicating that local concentrations of this magnitude may be tolerable in healthy subjects. It was an unexpected finding that the paranasal NO production is mainly controlled by an enzymesharing sequence homology to human hepatocyte iNOS (11). Recent studies indicate a possible additional involvement of eNOS but also confirm the presence of iNOS (144, 145). The factors responsible for this considerable local iNOS expression in healthy subjects are currently unknown. The paranasal NO production provides an efficient mechanism for nonspecific immune defense, since the local NO concentrations are sufficient to inhibit the growth of various bacteria and viruses (146–148). The localization of iNOS in apical parts of the epithelial cells indicates a further contribution to local host defense by a direct influence on ciliary beating. It has been demonstrated that the NOS antagonist L-NMMA blocked the increase in ciliary beat frequency (CBF) evoked by the stimulation of muscarinic receptors in ciliated epithelial cells of human tonsils (149). In addition, inhalation of the nebulized NO donor nitroprusside in human volunteers at a dose of 3 mg increased nasal mucociliary activity in vivo by 57% (12). Endogenous NO leads to formation of cGMP, which in turn activates proteinkinase G. The latter, in turn, evokes the effects of the signal transduction pathway via phosphorylation (150). Experimental studies indicate two possible physiological mechanisms: (1) proteinkinase G may induce a calcium influx (151) subsequently leading to an upregulation of dynein ATPase (152); (2) G-kinase phosphorylates a part of the dynein-associated proteins at the axons of the cilia (153, 154). In cytoplasmatic dynein, such subunits influence the dynein-ATPase activity (155, 156). Hence, previous studies showed a ciliary beat frequency of 17 Hz in epithelial cells from human maxillary sinus (157) versus a CBF of 11 Hz in human epithelial cells taken from the concha nasalis inferior (158), where only weak iNOS activity has been reported (11). It might be argued that the high NO production within the paranasal sinuses contributes to maintain CBF on a level that is sufficient for an optimal mucociliary clearance function. Since the sinus ostia are not located at the anatomically lowest level, a continuous, highly efficient transport is required to avoid mucus accumulation within the maxillary sinuses. While iNOS formation in the lower airways is increased in various diseases associated with lower respiratory tract infections, predictions concerning paranasal iNOS expression in response to inflammation and infection seem to be less clear-cut. Increased nasal NO levels have been reported in patients with allergic rhinitis (159) but were not detected in patients with influenza (160). Subnormal nasal NO release has been reported in patients

Acute respiratory distress syndrome

418

with cystic fibrosis (142, 161) and in patients with active Wegener’s granulomatosis (162). In children with acute infectious sinusitis, nasal NO levels were significantly reduced compared to healthy controls and approached normal NO concentrations after the disease was successfully treated with antibiotics (163). Chronic sinusitis is associated in a similar fashion with significantly reduced nasal NO levels and an impaired mucociliary clearance function (164). It has been suggested that the low nasal NO release more likely results from an occlusion of the sinus ostia than from a reduced iNOS expression. This interpretation was supported by further measurements in patients with nasal polyposis, in whom the nasally measured NO concentration was inversely correlated with the number of occluded paranasal sinuses (165). A recent study in septic patients with radiological maxillary sinusitis revealed, however, that NO within the maxillary sinuses is almost absent when measured following fenestration to exclude possible septic focus (143). Further immunohistochemical investigations on biopsy material from sinus epithelium demonstrated a strongly reduced iNOS expression within cilia and microvilli of these patients compared to controls (145). The factors responsible for the paradoxical upregulation of paranasal iNOS in healthy subjects and for the unexpected downregulation in patients with inflammatory and infected sinuses are currently unknown. Future research should aim at deeper insight into these pathophysiological mechanisms that may help develop new therapeutic strategies to avoid nosocomial upper airway infections in mechanically ventilated patients. D. Autoinhalation of Nitric Oxide It has been hypothesized that inhaled nasal NO acts as a selective pulmonary vasodilator, a process that has been termed NO autoinhalation (10). This concept has also been used to provide arguments for the use of inhaled NO in ARDS patients as replacement therapy for nasal NO, whose supply is disrupted due to endotracheal intubation. Although the NO release into the nasal cavity is considerably diluted by the inspiratory gas flow during spontaneous breathing in healthy subjects (131), and the effects are therefore expected to be small, the results of some studies indicate a potentially beneficial effect of autoinhaled NO. Low inhaled NO concentrations in hospital gas, in particular in compressed air, result from environmental pollution. Systematic investigations of the consequences in mechanically ventilated postoperative patients demonstrated a small improvement in arterial oxygenation, an effect that disappeared when the hospital gas was replaced by an NO-free synthetic mixture (166, 167). An improvement in oxygenation in mechanically ventilated patients has been reported following the instillation of nasally aspirated gas into the ventilator circuit (168). Hemodynamic beneficial effects of NO autoinhalation are suggested by a recent study in patients following open heart surgery. In this study, nasal breathing reduced pulmonary vascular resistance index by 10% when compared to oral breathing (169). In ARDS patients, even low doses of NO between 10 and 100 ppb have the potential to improve gas exchange (170). It should, however, be taken into account that the effects of the commonly applied therapeutic doses in the range of 5–10 ppm NO are considerably more pronounced and that the exogenous administration of inhaled NO in ARDS patients should not be simply considered as a substitute for nasal NO.

Modulation of pulmonary vascular tone in the acute respiratory

419

IV. Alternative Means to Selectively Influence the Pulmonary Vasculature in ARDS A. Aerosolized Prostacyclin In 1976 prostacyclin (PgI2) was discovered by Moncada and coworkers (171). PgI2 is endogenously produced through the cyclooxygenase pathway by endothelial cells. It has vasodilatory and anti-inflammatory properties and inhibits platelet aggregation (172, 173). In patients with ARDS, intravenously infused prostacyclin provides a nonselective reduction of pulmonary artery pressure with a concomitant increase in right-to-left shunt and a deterioration in gas exchange (4). Inhalation of aerosolized PgI2 at a dose of 17–50 ng/kg/min was first demonstrated to vasodilate selectively the pulmonary vasculature in a small series of ARDS patients (174). Subsequently, the efficacy of aerosolized PgI2 on hemodynamics and gas exchange was compared to inhaled NO in ARDS patients (5, 6, 175, 176). In these studies, aerosolized PgI2 significantly reduced pulmonary artery pressure and increased oxygenation, similar to the effect of inhaled NO (Fig. 4). However, during inhalation of prostacyclin the selectivity of vasodilation is dependent on the dose: high doses may induce a spillover of PgI2 into the systemic

Figure 4 Inhalation of 10 ppm NO and PGI2 (10 ng/kg/min) in a patient with ARDS. Continuous registration of reveals an immediate off-on response to both treatments. The decrease in during inhalation of both vasodilators reflects improved gas exchange due to a reduction in intrapulmonary right-to-left shunt.

Acute respiratory distress syndrome

420

circulation, reducing the selectivity and potentially reducing systemic blood pressure. In contrast to inhaled NO (the concentration of which can be measured by chemiluminescence or electrochemically), the concentration of PgI2 and the amount absorbed cannot be precisely determined. Thus, the nebulized dose of prostacyclin has to be calculated. In addition, only a much smaller amount of the aerosol fraction of PgI2 is delivered to the distal airspaces, and this quality is dependent on the performance of the nebulizer (177, 178). These technical difficulties are drawbacks to the use of inhaled PgI2 and have not yet been solved, largely precluding explicit dose recommendations in ARDS patients. However, doses as low as 10 ng/kg/min were reported to improve gas exchange in ARDS (6, 175). Another problem with the administration of aerosolized PgI2 is its short half-life, requiring continuous delivery of the drug into the respiratory circuit (179). Recent work with aerosolization of a stable analog of PgI2, iloprost, in patients with pulmonary hypertension resulted in a more sustained reduction of MPAP for up to 120 minutes after discontinuation of the drug (180). At present, the use of long-acting analogues of PgI2 for ambulatory treatment of patients with chronic pulmonary hypertension is under investigation (181). These compounds may facilitate therapeutic application by prolonging the dosing intervals. However, it still has to be determined if this approach will impair efficacy when compared to continuously nebulized PgI2. Considering clinical applicability in ARDS, inhaled prostaglandins yield the substantial risk of inducing prolonged arterial hypotension. Currently no randomized controlled trials of inhaled PgI2 have evaluated outcome parameters in ARDS patients. B. Modulation of Pulmonary Vascular Tone by Vasoconstriction: Effects of Almitrine Combined with NO Almitrine bismesylate was initially described to improve pulmonary gas exchange in COPD patients by an increase in ventilatory drive due to the stimulation of peripheral chemoreceptors (182, 183). Moreover, it has been suggested that almitrine reduces intrapulmonary shunting by enhancement of hypoxic pulmonary vasoconstriction (HPV) (184, 185). Almitrine improved arterial oxygenation in patients with severe ARDS secondary to sepsis and shock (186). The increase in PaO2 was due to a decrease in shunt and was associated with an increase in pulmonary arterial pressure. Using the multiple inert gas elimination technique in ARDS patients, Reyes and coworkers (182) demonstrated that almitrine redistributed pulmonary blood flow from shunt areas to lung units with normal ventilation/perfusion ratios. Therefore, almitrine is considered a selective pulmonary vasoconstrictor in shunt areas. The pulmonary vascular response to almitrine is complex, and the effects vary substantially between studies due to differences in the specific clinical setting and the doses of almitrine applied. In ARDS patients, the majority of investigators have described improvements in pulmonary gas exchange using a high-dose regimen (≥8 µg/kg/min). In contrast, in an animal model of acute lung injury, our group recently reported that high doses of almitrine (>4 µg/kg/min) significantly impaired gas exchange and increased mortality (187). Michard et al. (188) reported harmful effects of short-term administration of high-dose almitrine on pulmonary arterial pressure and right ventricular loading conditions in 9 patients with ARDS and severe hypoxemia. In 30 ARDS patients a maximum increase in PaO2 was reported with almitrine concentrations of <4 µg/kg/min

Modulation of pulmonary vascular tone in the acute respiratory

421

(8). Thus, in clinical ARDS small doses of almitrine (<4 µg/kg/min) appear to be most effective. In 1993 Payen and coworkers (189) were the first to combine inhalation of nitric oxide with intravenous almitrine in two ARDS patients. Co-administration of almitrine and NO induced an additive effect on gas exchange. Increases in pulmonary artery pressure during infusion of higher doses of almitrine were diminished by this combined treatment. These results were later confirmed by several investigators (190, 191), leading to inclusion of the combined use of inhaled nitric oxide and intravenous almitrine as a potential therapeutic strategy for ARDS reported by several groups (8, 192–194). C. Inhibition of Cyclic Nucleotide Phosphodiesterases Nitric oxide activates soluble guanylate cyclase, generating cyclic guanosine 3′5′monophosphate (cGMP), whichin turn vasodilates smooth muscle cells by activation of cGMP-dependent protein G-kinase and calcium-gated potassium channels (195). Various cyclic nucleotide phosphodiesterases (PDE) degrade cGMP, and basal vascular tone is influenced by the balance between cGMP production and PDE-dependent degradation (196–198). In the lung five different PDE isoforms have been identified—PDE 1, 3, 4, 5, and 9 (199–201); among these, PDE 5 is an enzyme that specifically hydrolyzes cGMP. Inhibitors of PDE 5 have been studied because of their ability to increase vascular cGMP levels, to lower pulmonary vascular resistance, and to potentially augment NO-induced pulmonary vasodilation (202, 203). The intravenous PDE 5 inhibitor zaprinast markedly prolonged the pulmonary vasodilation induced by inhaled NO in awake sheep (204). Diypridamol, a PDE 5 inhibitor, has been used in combination with inhaled NO to decrease pulmonary artery pressure in patients with pulmonary hypertension (9, 205). In 11 pediatric patients with severe pulmonary hypertension, Ziegler et al. (205) compared the effects of 20 ppm inhaled NO, 0.6 mg/kg dipyridamol, and the combined treatment on pulmonary and systemic hemodynamics. There was selective pulmonary vasodilation by NO and a nonselective reduction in pulmonary artery pressure mainly due to the increase in cardiac output. The combined treatment augmented pulmonary vasodilation only in 50% of this group of pediatric patients (205). Sildenafil is a newer PDE inhibitor with a higher selectivity for the type 5 isoenzyme than dipyridamol and a predictable gastrointestinal absorption. The drug has been approved for treatment of erectile dysfunction (206). In healthy pigs, oral sildenafil has characteristics of systemic vasodilators, significantly increasing cardiac output and intrapulmonary right-to-left shunt measured by inert gas elimination technique (MIGHT) (207). In awake lambs with pulmonary hypertension, cumulative doses of oral sildenafil (12.5, 25, and 50 mg) induced pulmonary vasodilation with only a moderate reduction in systemic arterial pressure. The combination with inhaled NO did not augment or prolong the ability of NO to vasodilate the pulmonary circulation (208). This animal study suggested that improved selectivity for the pulmonary circulation is possibly achieved by smaller doses of sildenafil. Future clinical studies will need to validate the role of sildenafil as a potential adjunct to inhaled NO in patients with pulmonary hypertension, including nebulized administration. The effects of inhaled PgI2 and iloprost are mediated by cAMP. Its cellular concentrations are directly controlled by the phosphodiesterases 1, 2, 3, and 4. PDEs of

Acute respiratory distress syndrome

422

type 2, 3, and 4 are cAMP-selective. The activity of PDE 2 and PDE 3 is influenced by cGMP and therefore is also indirectly dependent on the presence of the cGMP-specific PDE 5 (209). Schermuly and coworkers (210) investigated the effects of selective inhibitors for PDEs of types 3, 4, and 5 in an experimental rabbit model of pulmonary hypertension. They reported specific doses for these PDE inhibitors that were effective for a nonselective reduction in mean pulmonary artery pressure by about 20%, describing no effects on hemodynamics and gas exchange at a 10-fold lower dose. However, coadministration of PgI2 with the PDE inhibitors in subthreshold doses resulted in a significantly enhanced and prolonged vasodilation. These results were recently confirmed in isolated perfused rabbit lungs when inhibitors of PDE 3 and 4 were administered either intravenously or aerosolized in combination with inhaled PgI2 (211). D. Endothelin Receptor Antagonists in ARDS One year after identification of EDRF as NO (25, 26), Yanagisawa and colleagues (212) isolated endothelin-1, the most powerful endogenous vaso-constrictor yet described. The endothelins are 21-amino-acid peptides of three isoforms: endothelin-1 (ET-1), ET-2, and ET-3 (64). Because of their autocrine, paracrine, and possibly endocrine effects, endothelins contribute greatly to normal homeostasis as well as to numerous disorders, including cardiovascular, renal, and pulmonary diseases. Among the different endothelins, ET-1 is the main subtype. Endothelin-1 is a vasoconstrictor of coronary arteries and causes arterial hypertension in animal models and in humans, and induces pulmonary hypertension and bronchoconstriction (213). Furthermore, ET-1 is a mitogen and is also believed to be involved in pulmonary inflammation (214–216). The endothelin system may therefore contribute to many of the symptoms and pulmonary structural changes in patients with acute lung injury and ARDS. As stated above, ET-1 was found to be increased in lungs of patients with ARDS (68). It is not clear whether endothelins contribute to the pathogenesis of ARDS or whether their presence is just an epiphenomenon that marks lung injury and/or other coexisting diseases, such as sepsis, multiple trauma, burns, disseminated intravascular coagulation (DIC), or eclampsia (214). The cellular sources for the endothelins include the epithelium, endothelium, macrophages, and neuroendocrine cells (217, 218). Since endothelin-producing tissues express specific binding sites, their regulation may be primarily local rather than systemic (219). Endothelins exert their biological effects via two receptor subtypes: the ETA and ETB receptors (220). Stimulation of pulmonary ETA receptors mediate airway and vascular smooth muscle contraction and proliferation, mediator release (leukotrienes, plateletactivating factor, cytokines), edema formation, and enhancement of nerve-induced airway contraction. Activation of endothelial ETB receptors induces the release of vasodilators, such as NO and prostacyclin, while smooth muscle ETB receptors mediate vaso- and bronchoconstriction. While endothelial ETB receptors induce the release of NO and prostacyclin, the stimulation of pulmonary ETA receptors was shown to induce sustained vasoconstriction, revealing the capability of ETA receptor antagonists to act as vasodilators (213, 215). Numerous pep tide and nonpeptide endothelin receptor antagonists have been developed, some selective for A or B receptors and others with

Modulation of pulmonary vascular tone in the acute respiratory

423

nonselective activity for A/ B receptors (220). Endothelin receptor antagonists are usually administered intravenously or orally, and their efficacy has been shown particularly in animal models of airway inflammation (216, 221) and pulmonary hypertension (222– 226). In a recent randomized placebo-controlled trial, Rubin et al. (227) demonstrated that in patients with pulmonary arterial hypertension, oral bosentan (a dual ETA/ETB receptor antagonist) was effective and well tolerated. There are few studies in which ETA receptor antagonists have been inhaled or directly administered into the lungs. Intrapulmonary ap-plied ETA receptor antagonists, for example, markedly inhibited antigen-induced lung inflammation in mice (215, 228). In other studies controversial results were reported as far as the effects of inhaled ETA antagonists on ET-1-induced bronchoconstriction is concerned, partly attributable to differences in bronchopulmonary receptor distribution and density in different species (213, 221, 229). In view of the available data, we hypothesized that in acute lung injury inhalation of an ETA receptor antagonist would induce selective pulmonary vasodilation, thereby reducing intrapulmonary shunt fraction and increasing similar to the effects of mmHg for one inhaled NO. Using a model of surfactant washout ( hour), we studied piglets, randomly assigned to receive either the nebulized ETA receptor antagonist LU-135252 (Knoll AG, Ludwigshafen, Germany; 3 mg/kg) (n=8) over one hour or nebulization of saline (n=8). Inhalation of LU-135252 reduced intrapulmonary right-to-left shunt from 58±8% (mean±SD) to 27 ±12% (p<0.05), and PaO2 increased from 55±12 to 257±148 mmHg (p<0.05). In control animals, gas exchange was not improved during the 6-hour experiment. Animals that inhaled LU-135252 had stable pulmonary artery pressures (26–29 mmHg) during the 6-hour experiment, whereas in controls, pulmonary artery pressure was significantly increased from 28±2 to 41±2 mmHg at the end of the experiment (p<0.05). Inhalation of the ETA receptor antagonist induced a decrease in cardiac output by 31±11% and an increase in SVR of 60±29%, indicating that inhaled LU-135252 did not cause peripheral vasodilation. At the dosage used, inhalation of LU-135252 induced systemic vasoconstriction and reduced cardiac output, possibly due to an alteration in pulmonary clearance of ET-1. The decrease in cardiac output probably contributed to a reduction in the shunt fraction and thus to an improvement in oxygenation (230). Although LU-135252 is characterized a selective ETA antagonist, it is well established that using high doses of intravenous LU-135252 will abbreviate selectivity for the ETA receptor, resulting in an additional blockade of ETB receptors (231). Since pulmonary ETB receptors are considered clearance receptors for endothelins, blockade of the ETB receptors by high doses of the receptor antagonist may well increase ET-1 plasma concentrations (232), resulting in cardiovascular responses similar to those observed in our study (230). In a subsequent study, we tested the following hypothesis: in experimental acute lung injury, inhalation of a much lower dose of LU-135252 (1/10 of our initial dose) would still provide beneficial effects on oxygenation, would not affect mean arterial pressure and cardiac output, and would provide effects similar to that of NO. We carried out a randomized, controlled animal study comparing low-dose inhalation of LU-135252 (0.3

Acute respiratory distress syndrome

424

(mmHg) (a) and Figure 5 intrapulmonary right-to-left-shunt (Qs/Qt, %) (b) (mean±SEM) following inhalation of 0.3 mg/kg LU-135252 for 20 minutes (n= 10, triangles) and during continuously inhaled NO (30 ppm; n=10, squares) and in controls after inhaling saline for 20 minutes (n=10, circles). A model of surfactant washout in piglets was used. Measurements were performed

Modulation of pulmonary vascular tone in the acute respiratory

immediately after induction of ALI and at hourly intervals for 4 hours. *p<0.05; **p<0.01 compared to controls; §p<0.05 compared to ALI. Inhaled LU-135252 and NO both induced a significant and sustained improvement in and Qs/Qt when compared to controls.

Figure 6 Relative changes (%) in mean pulmonary artery pressure

425

Acute respiratory distress syndrome

426

(MPAP) (a) and mean systemic arterial pressure (MAP) (b) (mean±SEM) following inhalation of 0.3 mg/kg LU135252 for 20 minutes (n=10, triangles) and during continuously inhaled NO (30 ppm; n=10, squares) and in controls after inhaling saline for 20 minutes (n =10, circles). A model of surfactant washout in piglets was used. Measurements were performed immediately after induction of ALI and at hourly intervals for 4 hours. *p <0.05; **p<0.01 compared to controls. Inhalation of LU-135252 and NO exerted similar effects on MPAP when compared to controls. MAP was not different between groups, demonstrating that inhalation of the ETA-receptor antagonist did not induce systemic vasodilation. mg/kg) with continuous inhalation of 30 ppm NO against controls (233). In this study, LU-135252 at a low dose and NO both significantly improved gas exchange and prevented an increase in mean pulmonary artery pressure in a similar fashion without significant systemic effects compared to controls (Figs. 5 and 6). These results provide experimental evidence that the inhaled ETA receptor antagonist LU-135252 at a dosage of 0.3 mg/kg acts as a selective pulmonary vasodilator. Further investigations should focus on: (1) the minimal dose of nebulized ETA receptor antagonist that is effective to improve arterial oxygenation, (2) effects of LU-135252 and inhaled NO on tissue and plasma endothelin concentrations, (3) the ability of LU-135252 to reduce progression of lung tissue damage, and (4) determination of anti-inflammatory and/or antiedematous effects of ETA antagonists on the injured lung. The preclinical studies merit future clinical evaluation of the potential role of endothelin receptor antagonists in the treatment of ARDS patients.

V. Conclusions For decades, interventions to improve arterial oxygenation by means of modifying ventilator support have been the mainstay of ARDS therapy. With the introduction of inhaled NO, it became possible for the first time to improve matching of perfusion to ventilation by redistribution of pulmonary blood flow in favor of ventilated and relatively

Modulation of pulmonary vascular tone in the acute respiratory

427

intact areas of the lungs. Clinical evaluation of this novel concept further demonstrated that combining NO with other interventions yields beneficial and additive effects on arterial oxygenation, rendering inhaled NO a clinically valuable option for the treatment of severe refractory hypoxemia in patients with ARDS. Inhaled NO largely promoted the introduction of this innovative concept of selective pulmonary vasodilation into intensive care practice. Aerosolization of various vasodilators is currently under evaluation in experimental acute lung injury and in human ARDS. Ongoing research is also seeking to augment the efficacy of pulmonary vasodilation by specific inhibitors of PDEs or by the combination with intravenous vasoconstrictors. Hence, the clinician may soon be offered several alternative methods to selectively modulate pulmonary vascular tone in ARDS patients. The efficacy and costs of these novel therapeutic options will determine their future utility.

Acknowledgments This work was supported, in part, by a grant from the Deutsche Forschungsgemeinschaft (U.K., DFG KA 1212/4–1). The authors would like to thank Laraine Visser-Isles for English language editing and Bernd Donaubauer, M.D., for editorial help.

References 1. Frostell C, Fratacci MD, Wain JC, Jones R, Zapol WM. Inhaled nitric oxide. A selective pulmonary vasodilator reversing hypoxic pulmonary vasoconstriction. Circulation 1991; 83:2038–2047. 2. Higenbottam TW, Pepke-Zaba J, Scott J, Woolman P, Coutts C, Wallock J. Inhaled ‘endothelium derived-relaxing factor’ (EDRF) in primary hypertension. Am Rev Respir Dis 1988; 137:A107. 3. Falke K, Rossaint R, Pison U, Slama K, Lopez F, Santak B, Zapol WM. Inhaled nitric oxide selectively reduces pulmonary hypertension in severe ARDS and improves gas exchange as well as right heart ejection fraction—a case report. Am Rev Respir Dis 1991; 143:A248. 4. Rossaint R, Falke KJ, Lopez F, Slama K, Pison U, Zapol WM. Inhaled nitric oxide for the adult respiratory distress syndrome. N Engl J Med 1993; 328:399–405. 5. Pappert D, Busch T, Gerlach H, Lewandowski K, Radermacher P, Rossaint R. Aerosolized prostacyclin versus inhaled nitric oxide in children with severe acute respiratory distress syndrome. Anesthesiology 1995; 82:1507–1511. 6. Walmrath D, Schneider T, Schermuly R, Olschewski H, Grimminger F, Seeger W. Direct comparison of inhaled nitric oxide and aerosolized prostacyclin in acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 153:991–996. 7. Falke KJ, Pontoppidan H, Kumar A, Leith DE, Geffin B, Laver MB. Ventilation with positive end-expiratory pressure in acute lung disease. J Clin Invest 1972; 51:2315–2323. 8. Gallart L, Lu Q, Puybasset L, Umamaheswara Rao GS, Coriat P, Rouby JJ. Intravenous almitrine combined with inhaled nitric oxide for acute respiratory distress syndrome. The NO Almitrine Study Group. Am J Respir Crit Care Med 1998; 158:1770–1777. 9. Kinsella JP, Torielli F, Ziegler JW, Ivy DD, Abman SH. Dipyridamole augmentation of response to nitric oxide. Lancet 1995; 346:647–648. 10. Gerlach H, Rossaint R, Pappert D, Knorr M, Falke KJ. Autoinhalation of nitric oxide after endogenous synthesis in nasopharynx. Lancet 1994; 343:518–519.

Acute respiratory distress syndrome

428

11. Lundberg JON, Farkas-Szallasi T, Weitzberg E, Rinder J, Lidholm J, Änggard A, Hökfelt T, Lundberg JM, Alving K. High nitric oxide production in human paranasal sinuses. Nat Med 1995; 1:370–373. 12. Runer T, Lindberg S. Effects of nitric oxide on blood flow and mucociliary activity in the human nose. Ann Otol Rhinol Laryngol 1998; 107:40–46. 13. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342:1334–1349. 14. Zapol WM, Snider MT. Pulmonary hypertension in severe acute respiratory failure. N Engl J Med 1977; 296:476–480. 15. Wort SJ, Evans TW. The role of the endothelium in modulating vascular control in sepsis and related conditions. Br Med Bull 1999; 55:30–48. 16. Villar J, Blazquez MA, Lubillo S, Quintana J, Manzano JL. Pulmonary hypertension in acute respiratory failure. Crit Care Med 1989; 17:523–526. 17. Steltzer H, Krafft P, Fridrich P, Hiesmayr M, Hammerle AF. Right ventricular function and oxygen transport patterns in patients with acute respiratory distress syndrome. Anaesthesia 1994; 49:1039–1045. 18. Zapol WM, Jones R. Vascular components of ARDS. Clinical pulmonary hemodynamics and morphology. Am Rev Respir Dis 1987; 136:471–474. 19. Holcroft JW, Vassar MJ, Weber CJ. Prostaglandin E1 and survival in patients with the adult respiratory distress syndrome. A prospective trial. Ann Surg 1986; 203:371–378. 20. Vincent JL, Brase R, Santman F, Suter PM, McLuckie A, Dhainaut JF, Park Y, Karmel J. A multi-centre, double-blind, placebo-controlled study of liposomal prostaglandin E1 (TLC C-53) in patients with acute respiratory distress syndrome. Intensive Care Med 2001; 27:1578–1583. 21. Mélot C, Lejeune P, Leeman M, Moraine JJ, Naeije R. Prostaglandin E1 in the adult respiratory distress syndrome. Benefit for pulmonary hypertension and cost for pulmonary gas exchange. Am Rev Respir Dis 1989; 139:106–110. 22. Radermacher P, Santak B, Becker H, Falke KJ. Prostaglandin E1 and nitroglycerin reduce pulmonary capillary pressure but worsen ventilation-perfusion distributions in patients with adult respiratory distress syndrome. Anesthesiology 1989; 70:601–606. 23. Pepke-Zaba J, Higenbottam TW, Dinh-Xuan AT, Stone D, Wallwork J. Inhaled nitric oxide as a cause of selective pulmonary vasodilatation in pulmonary hypertension. Lancet 1991; 338:1173–1174. 24. Dantzker DR, Lynch JP, Weg JG. Depression of cardiac output is a mechanism of shunt reduction in the therapy of acute respiratory failure. Chest 1980; 77:636–642. 25. Palmer RM, Ferrige AG, Moncada S. Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor. Nature 1987; 327: 524–526. 26. Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci USA 1987; 84:9265–9269. 27. Frostell CG, Blomqvist H, Hedenstierna G, Lundberg J, Zapol WM. Inhaled nitric oxide selectively reverses human hypoxic pulmonary vasoconstriction without causing systemic vasodilation. Anesthesiology 1993; 78:427–435. 28. Pison U, Lopez FA, Heidelmeyer CF, Rossaint R, Falke KJ. Inhaled nitric oxide reverses hypoxic pulmonary vasoconstriction without impairing gas exchange. J Appl Physiol 1993; 74:1287–1292. 29. Rimar S, Gillis CN. Selective pulmonary vasodilation by inhaled nitric oxide is due to hemoglobin inactivation. Circulation 1993; 88:2884–2887. 30. Lindeborg DM, Kavanagh BP, Van Meurs K, Pearl RG. Inhaled nitric oxide does not alter the longitudinal distribution of pulmonary vascular resistance. J Appl Physiol 1995; 78:341–348. 31. Shirai M, Shimouchi A, Kawaguchi AT, Sunagawa K, Ninomiya I. Inhaled nitric oxide: diameter response patterns in feline small pulmonary arteries and veins. Am J Physiol 1996; 270:H974-H980.

Modulation of pulmonary vascular tone in the acute respiratory

429

32. Benzing A, Geiger K. Inhaled nitric oxide lowers pulmonary capillary pressure and changes longitudinal distribution of pulmonary vascular resistance in patients with acute lung injury. Acta Anaesthesiol Scand 1994; 38:640–645. 33. Benzing A, Brautigam P, Geiger K, Loop T, Beyer U, Moser E. Inhaled nitric oxide reduces pulmonary transvascular albumin flux in patients with acute lung injury. Anesthesiology 1995; 83:1153–1161. 34. Rossaint R, Slama K, Steudel W, Gerlach H, Pappert D, Veit S, Falke K. Effects of inhaled nitric oxide on right ventricular function in severe acute respiratory distress syndrome. Intens Care Med 1995; 21:197–203. 35. Krafft P, Fridrich P, Fitzgerald RD, Koc D, Steltzer H. Effectiveness of nitric oxide inhalation in septic ARDS. Chest 1996; 109:486–493. 36. Putensen C, Hormann C, Kleinsasser A, Putensen-Himmer G. Cardiopulmonary effects of aerosolized prostaglandin E! and nitric oxide inhalation in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 157:1743–1747. 37. Puybasset L, Rouby JJ, Mourgeon E, Cluzel P, Souhil Z, Law-Koune JD, Stewart T, Devilliers C, Lu Q, Roche S, Kalfon B, Vicaut E, Viars P. Factors influencing cardiopulmonary effects of inhaled nitric oxide in acute respiratory failure. Am J Respir Crit Care Med 1995; 152:318–328. 38. Okamoto K, Kukita I, Hamaguchi M, Kikuta K, Matsuda K, Motoyama T. Combination of inhaled nitric oxide therapy and inverse ratio ventilation in patients with sepsis-associated acute respiratory distress syndrome. Artif Organs 2000; 24:902–908. 39. Papazian L, Bregeon F, Gaillat F, Thirion X, Gainnier M, Gregoire R, Saux P, Gouin F, Jammes Y, Auffray JP. Respective and combined effects of prone position and inhaled nitric oxide in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 157:580–585. 40. Jolliet P, Bulpa P, Ritz M, Ricou B, Lopez J, Chevrolet JC. Additive beneficial effects of the prone position, nitric oxide, and almitrine bismesylate on gas exchange and oxygen transport in acute respiratory distress syndrome. Crit Care Med 1997; 25:786–794. 41. Germann P, Poschl G, Leitner C, Urak G, Ullrich R, Faryniak B, Roder G, Kaider A, Sladen R. Additive effect of nitric oxide inhalation on the oxygenation benefit of the prone position in the adult respiratory distress syndrome. Anesthesiology 1998; 89:1401–1406. 42. Martinez M, Diaz E, Joseph D, Villagra A, Mas A, Fernandez R, Blanch L. Improvement in oxygenation by prone position and nitric oxide in patients with acute respiratory distress syndrome. Intens Care Med 1999; 25:29–36. 43. Borelli M, Lampati L, Vascotto E, Fumagalli R, Pesenti A. Hemodynamic and gas exchange response to inhaled nitric oxide and prone positioning in acute respiratory distress syndrome patients. Crit Care Med 2000; 28:2707–2712. 44. Dupont H, Mentec H, Cheval C, Moine P, Fierobe L, Timsit JF. Short-term effect of inhaled nitric oxide and prone positioning on gas exchange in patients with severe acute respiratory distress syndrome. Crit Care Med 2000; 28:304–308. 45. Wilcox DT, Glick PL, Karamanoukian HL, Morin FC III, Fuhrman BP, Leach C. Partial liquid ventilation and nitric oxide in congenital diaphragmatic hernia. J Pediatr Surg 1997; 32:1211– 1215. 46. Gommers D, Hartog A, van ’t Veen A, Lachmann B. Improved oxygenation by nitric oxide is enhanced by prior lung reaeration with surfactant, rather than positive end-expiratory pressure, in lung-lavaged rabbits. Crit Care Med 1997; 25:1868–1873. 47. Houmes RJM, Hartog A, Verbrugge SJC, Böhm S, Lachmann B. Combining partial liquid ventilation with nitric oxide to improve gas exchange in acute lung injury. Intens Care Med 1997; 23:163–169. 48. Gerlach H, Pappert D, Lewandowski K, Rossaint R, Falke KJ. Long-term inhalation with evaluated low doses of nitric oxide for selective improvement of oxygenation in patients with adult respiratory distress syndrome. Intens Care Med 1993; 19:443–449.

Acute respiratory distress syndrome

430

49. Abman SH, Griebel JL, Parker DK, Schmidt JM, Swanton D, Kinsella JP. Acute effects of inhaled nitric oxide in children with severe hypoxemic respiratory failure. J Pediatr 1994; 124:881–888. 50. Bigatello LM, Hurford WE, Kacmarek RM, Roberts JD, Zapol WM. Prolonged inhalation of low concentrations of nitric oxide in patients with severe adult respiratory distress syndrome. Effects on pulmonary hemodynamics and oxygenation. Anesthesiology 1994; 80:761–770. 51. Puybasset L, Stewart T, Rouby J-J, Cluzel P, Mourgeon E, Belin MF, Arthaud M, Landault C, Viars P. Inhaled nitric oxide reverses the increase in pulmonary vascular resistance induced by permissive hypercapnia in patients with acute respiratory distress syndrome. Anesthesiology 1994; 80:1254–1267. 52. Young JD, Brampton WJ, Knighton JD, Finfer SR. Inhaled nitric oxide in acute respiratory failure in adults. Br J Anaesth 1994; 73:499–502. 53. McIntyre RC Jr, Moore FA, Moore EE, Piedalue F, Haenel JS, Fullerton DA. Inhaled nitric oxide variably improves oxygenation and pulmonary hypertension in patients with acute respiratory distress syndrome. J Trauma 1995; 39:418–425. 54. Lundin S, Nathorst Westfelt U, Stenquist O, Blomqvist H, Lindh A, Berggren L, Arvidsson S, Rudberg U, Frostell CG. Response to nitric oxide inhalation in early acute lung injury. Intens Care Med 1996; 22:728–734. 55. Rossaint R, Gerlach H, Schmidt-Ruhnke H, Pappert D, Lewandowski K, Steudel W, Falke K. Efficacy of inhaled nitric oxide in patients with severe ARDS. Chest 1995; 107:1107–1115. 56. Manktelow C, Bigatello LM, Hess D, Hurford WE. Physiologic determinants of the response to inhaled nitric oxide in patients with acute respiratory distress syndrome. Anesthesiology 1997; 87:297–307. 57. Gerlach H, Rossaint R, Pappert D, Falke KJ. Time-course and dose-response of nitric oxide inhalation for systemic oxygenation and pulmonary hypertension in patients with adult respiratory distress syndrome. Eur J Clin Invest 1993; 23:499–502. 58. Gerlach H, Falke KJ. Low levels of inhaled nitric oxide in acute lung injury. In: Zapol WM, Bloch KD, eds. Lung Biology in Health and Disease. Vol. 98: Nitric Oxide and the Lung. New York: Marcel Dekker, 1997:271–283. 59. Scherrer U, Vollenweider L, Delabays A, Savcic M, Eichenberger U, Kleger G-R, Fikrle A, Ballmer PE, Nicod P, Bärtsch P. Inhaled nitric oxide for high-altitude pulmonary edema. N Engl J Med 1996; 334:624–629. 60. Barberà JA, Roger N, Roca J, Rovira I, Higenbottam TW, Rodriguez-Roisin R. Worsening of pulmonary gas exchange with nitric oxide inhalation in chronic obstructive pulmonary disease. Lancet 1996; 347:436–440. 61. Holzmann A, Bloch KD, Sanchez LS, Filippov G, Zapol WM. Hyporesponsiveness to inhaled nitric oxide in isolated perfused lungs from endotoxin-challenged rats. Am J Physiol 1996; 271:L981–L986. 62. Holzmann A, Manktelow C, Taut FJH, Bloch KD, Zapol WM. Inhibition of nitric oxide synthase prevents hyporesponsiveness to inhaled nitric oxide in lungs from endotoxinchallenged rats. Anesthesiology 1999; 91:215–221. 63. Giaid A, Yanagisawa M, Langleben D, Michel RP, Levy R, Shennib H, Kimura S, Masaki T, Duguid WP, Stewart DJ. Expression of endothelin-1 in the lungs of patients with pulmonary hypertension. N Engl J Med 1993; 328:1732–1739. 64. Gandhi CR, Berkowitz DE, Watkins WD. Endothelins. Biochemistry and pathophysiologic actions. Anesthesiology 1994; 80:892–905. 65. Langleben D, DeMarchie M, Laporta D, Spanier AH, Schlesinger RD, Stewart DJ. Endothelin1 in acute lung injury and the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 148:1646–1650. 66. Asakura H, Jokaji H, Saito M, Uotani C, Kumabashiri I, Morishita E, Yamazaki M, Matsuda T. Role of endothelin in disseminated intravascular coagulation. Am J Hematol 1992; 41:71–75.

Modulation of pulmonary vascular tone in the acute respiratory

431

67. Mitaka C, Hirata Y, Nagura T, Tsunoda Y, Amaha K. Circulating endothelin-1 concentrations in acute respiratory failure. Chest 1993; 104:476–480. 68. Albertine KH, Wang ZM, Michael JR. Expression of endothelial nitric oxide synthase, inducible nitric oxide synthase, and endothelin-1 in lungs of subjects who died with ARDS. Chest 1999; 116:101S-102S. 69. Busch T, Deja M, Boemke W, Wolf S, Donaubauer B, Petersen B, Falke KJ, Kaisers U. Endothelin-1 plasma levels influence efficacy of inhaled nitric oxide in experimental acute lung injury. Am J Respir Crit Care Med 2002; 165:A783. 70. Gerlach H, Busch T, Keh D, Semmerow A, Lewandowski K, Rossaint R, Falke KJ. Dose response characteristics during long-term inhalation of nitric oxide in ARDS: a prospective, randomized study. Am J Respir Crit Care Med 2000; 161:A379. 71. Mira JP, Monchi M, Brunet F, Fierobe L, Dhainaut JF, Dinh-Xuan AT. Lack of efficacy of inhaled nitric oxide in ARDS. Intens Care Med 1994; 20:532. 72. Rubbo H, Radi R, Trujillo M, Telleri R, Kalyanaraman B, Barnes S, Kirk M, Freeman BA. Nitric oxide regulation of superoxide and peroxynitrite-dependent lipid peroxidation. Formation of novel nitrogen-containing oxidized lipid derivatives. J Biol Chem 1994; 269:26066–26075. 73. Li S, Fan SX, McKenna TM. Role of nitric oxide in sepsis-induced hypo-reactivity in isolated rat lungs. Shock 1996; 5:122–129. 74. Fukuto JM. Chemistry of nitric oxide. Biologically relevant aspects. In: Ignarro L, Murad F, eds. Nitric Oxide: Biochemistry, Molecular Biology and Therapeutic Implications. San Diego: Academic Press, 1995:1–13. 75. Misko TP, Moore WM, Kasten TP, Nickols GA, Corbett JA, Tilton RG, McDaniel ML, Williamson JR, Currie MG. Selective inhibition of the inducible nitric oxide synthase by aminoguanidine. Eur J Pharmacol 1993; 233: 119–125. 76. Anggard E. Nitric oxide: mediator, murderer, and medicine. Lancet 1994; 343:1199–1206. 77. Stamler JS, Singel DJ, Loscalzo J. Biochemistry of nitric oxide and its redox-activated forms. Science 1992; 258:1898–1902. 78. Shiel FO. Morbid anatomical changes in the lungs of dogs after inhalation of higher oxides of nitrogen during anaesthesia. Br J Anaesth 1967; 39:413–424. 79. Rasmussen TR, Kjaergaard SK, Tarp U, Pedersen OF. Delayed effects of NO2 exposure on alveolar permeability and glutathione peroxidase in healthy humans. Am Rev Respir Dis 1992; 146:654–659. 80. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990; 87: 1620–1624. 81. Lewis RS, Deen WM. Kinetics of the reaction of nitric oxide with oxygen in aqueous solutions. Chem Res Toxicol 1994; 7:568–574. 82. Szabo C. DNA strand breakage and activation of poly-ADP ribosyltransferase: a cytotoxic pathway triggered by peroxynitrite. Free Radic Biol Med 1996; 21:855–869. 83. Weinberger B, Heck DE, Laskin DL, Laskin JD. Nitric oxide in the lung: therapeutic and cellular mechanisms of action. Pharmacol Ther 1999; 84:401–411. 84. Huie RE, Padmaja S. The reaction of NO with superoxide. Free Radic Res Commun 1993; 18:195–199. 85. Freeman B. Free radical chemistry of nitric oxide. Looking at the dark side. Chest 1994; 105:79S–84S. 86. Denicola A, Freeman BA, Trujillo M, Radi R. Peroxynitrite reaction with carbon dioxide/bicarbonate: kinetics and influence on peroxynitrite-mediated oxidations. Arch Biochem Biophys 1996; 333:49–58. 87. Tien M, Berlett BS, Levine RL, Chock PB, Stadtman ER. Peroxynitrite-mediated modification of proteins at physiological carbon dioxide concentration: pH dependence of carbonyl formation, tyrosine nitration, and methionine oxidation. Proc Natl Acad Sci USA 1999; 96:7809–7814.

Acute respiratory distress syndrome

432

88. Lang JD Jr, Chumley P, Eiserich JP, Estevez A, Bamberg T, Adhami A, Crow J, Freeman BA. Hypercapnia induces injury to alveolar epithelial cells via a nitric oxide-dependent pathway. Am J Physiol Lung Cell Mol Physiol 2000; 279:L994–L1002. 89. Haddad IY, Pataki G, Hu P, Galliani C, Beckman JS, Matalon S. Quantitation of nitrotyrosine levels in lung sections of patients and animals with acute injury. J Clin Invest 1994; 94:2407– 2413. 90. Kooy NW, Royall JA, Ye YZ, Kelly DR, Beckman JS. Evidence for in vivo peroxynitrite production in human acute lung injury. Am J Respir Crit Care Med 1995; 151:1250–1254. 91. Sittipunt C, Steinberg KP, Ruzinski JT, Myles C, Zhu S, Goodman RB, Hudson LD, Matalon S, Martin TR. Nitric oxide and nitrotyrosine in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001; 163:503–510. 92. Stuart-Smith K, Jeremy JY. Microvessel damage in acute respiratory distress syndrome: the answer may not be NO. Br J Anaesth 2001; 87:272–279. 93. Darley-Usmar V, Halliwell B. Blood radicals: reactive nitrogen species, reactive oxygen species, transition metal ions, and the vascular system. Pharm Res 1996; 13:649–662. 94. Gadek JE. Adverse effects of neutrophils on the lung. Am J Med 1992; 92: 27S-31S. 95. Swank DW, Moore SB. Roles of the neutrophil and other mediators in adult respiratory distress syndrome. Mayo Clin Proc 1989; 64:1118–1132. 96. Garat C, Jayr C, Eddahibi S, Laffon M, Meignan M, Adnot S. Effects of inhaled nitric oxide or inhibition of endogenous nitric oxide formation on hyperoxic lung injury. Am J Respir Crit Care Med 1997; 155:1957–1964. 97. Wink DA, Cook JA, Pacelli R, Liebmann J, Krishna MC, Mitchell JB. Nitric oxide (NO) protects against cellular damage by reactive oxygen species. Toxicol Lett 1995; 82–83:221–226. 98. Mansouri A, Lurie AA. Concise review: methemoglobinemia. Am J Hematol 1993; 42:7–12. 99. Heal CA, Spencer SA. Methaemoglobinaemia with high-dose nitric oxide administration. Acta Paediatr 1995; 84:1318–1319. 100. Dellinger RP, Zimmerman JL, Taylor RW, Straube RC, Hauser DL, Criner GJ, Davis K Jr, Hyers TM, Papadakos P. Effects of inhaled nitric oxide in patients with acute respiratory distress syndrome: results of a randomized phase II trial. Inhaled nitric oxide in ARDS Study Group. Crit Care Med 1998; 26:15–23. 101. Jia L, Bonaventura C, Bonaventura J, Stamler JS. S-Nitrosohaemoglobin: a dynamic activity of blood involved in vascular control. Nature 1996; 380:221–226. 102. Mellion BT, Ignarro LJ, Ohlstein EH, Pontecorvo EG, Hyman AL, Kadowitz PJ. Evidence for the inhibitory role of guanosine 3′,5′-monophosphate in ADP-induced human platelet aggregation in the presence of nitric oxide and related vasodilators. Blood 1981; 57:946–955. 103. Samama CM, Diaby M, Fellahi JL, Mdhafar A, Eyraud D, Arock M, Guillosson JJ, Coriat P, Rouby J-J. Inhibition of platelet aggregation by inhaled nitric oxide in patients with acute respiratory distress syndrome. Anesthesiology 1995; 83:56–65. 104. Hasegawa N, Husari AW, Hart WT, Kandra TG, Raffin TA. Role of the coagulation system in ARDS. Chest 1994; 105:268–277. 105. Malik AB. Pulmonary microembolism and lung vascular injury. Eur Respir J Suppl 1990; 11:499S-506S. 106. Michael JR, Barton RG, Saffle JR, Mone M, Markewitz BA, Hillier K, Elstad MR, Campbell EJ, Troyer BE, Whatley RE, Liou TG, Samuelson WM, Carveth HJ, Hinson DM, Morris SE, Davis BL, Day RW. Inhaled nitric oxide versus conventional therapy: effect on oxygenation in ARDS. Am J Respir Crit Care Med 1998; 157:1372–1380. 107. Troncy E, Collet JP, Shapiro S, Guimond JG, Blair L, Ducruet T, Francoeur M, Charbonneau M, Blaise G. Inhaled nitric oxide in acute respiratory distress syndrome: a pilot randomized controlled study. Am J Respir Crit Care Med 1998; 157:1483–1488. 108. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall JR, Morris A, Spragg R. The American-European Consensus Conference on ARDS. Definitions,

Modulation of pulmonary vascular tone in the acute respiratory

433

mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818–824. 109. Lundin S, Mang H, Smithies M, Stenqvist O, Frostell C. European Study Group of Inhaled Nitric Oxide. Inhalation of nitric oxide in acute lung injury: preliminary results of a European multicenter study. Intens Care Med 1997; 23:S2. 110. Lundin S, Mang H, Smithies M, Stenqvist O, Frostell C. Inhalation of nitric oxide in acute lung injury: results of a European multicentre study. The European Study Group of Inhaled Nitric Oxide. Intens Care Med 1999; 25:911–919. 111. Sokol S, Jacobs JE, Bohn D. Inhaled nitric oxide for acute hypoxemic respiratory failure in children and adults (Cochrane Review). In: The Cochrane Library, Issue 3, 2001. Oxford: Update Software. 112. Weinberger B, Weiss K, Heck DE, Laskin DL, Laskin JD. Pharmacologic therapy of persistent pulmonary hypertension of the newborn. Pharmacol Ther 2001; 89:67–79. 113. The Neonatal Inhaled Nitric Oxide Study Group (NINOS). Inhaled nitric oxide and hypoxic respiratory failure in infants with congenital diaphragmatic hernia. Pediatrics 1997; 99:838–845. 114. Finer NN, Barrington KJ. Nitric oxide for respiratory failure in infants born at or near term (Cochrane Review). In: The Cochran Library, Issue 3, 2001. Oxford: Update Software. 115. The Franco-Belgium Collaborative NO Trial Group. Early compared with delayed inhaled nitric oxide in moderately hypoxaemic neonates with respiratory failure: a randomised controlled trial. Lancet 1999; 354:1066–1071. Erratum in: Lancet 1999, 354:1826. 116. Kinsella JP, Walsh WF, Rose CL, Gerstmann DR, Labella JJ, Sardesai S, Walsh-Sukys MC, McCaffrey MJ, Cornfield DN, Bhutani VK, Cutter GR, Baier M, Abman SH. Inhaled nitric oxide in premature neonates with severe hypoxaemic respiratory failure: a randomised controlled trial. Lancet 1999; 354:1061–1065. 117. Saugstad OD. Inhaled nitric oxide for preterm infants—still an experimental therapy. Lancet 1999; 354:1047–1048. 118. Inhaled nitric oxide in term and near-term infants: neurodevelopmental follow-up of the neonatal inhaled nitric oxide study group (NINOS). J Pediatr 2000; 136:611–617. 119. Kobzik L, Bredt DS, Lowenstein CJ, Drazen J, Gaston B, Sugarbaker D, Stamler JS. Nitric oxide synthase in human and rat lung: immunocytochemical and histochemical localization. Am J Respir Cell Mol Biol 1993; 9:371–377. 120. Lamas S, Michel T. Molecular biological features of nitric oxide isoforms. In: Zapol WM, Bloch KD, eds. Lung Biology in Health and Disease, Vol. 98: Nitric Oxide and the Lung. New York: Marcel Dekker, 1987:59–73. 121. Blitzer ML, Loh E, Roddy MA, Stamler JS, Creager MA. Endothelium-derived nitric oxide regulates systemic and pulmonary vascular resistance during acute hypoxia in humans. J Am Coll Cardiol 1996; 28:591–596. 122. Giaid A, Saleh D. Reduced expression of endothelial nitric oxide synthase in the lungs of patients with pulmonary hypertension. N Engl J Med 1995; 333:214–221. 123. Groeneveld ABJ, Hartemink KJ, de Groot MCM, Visser J, Thijs LG. Circulating endothelin and nitrate-nitrite relate to hemodynamic and metabolic variables in human septic shock. Shock 1999; 11:160–166. 124. Hollenberg SM, Broussard M, Osman J, Parrillo JE. Increased microvascular reactivity and improved mortality in septic mice lacking inducible nitric oxide synthase. Circ Res 2000; 86:774–778. 125. Zamora R, Vodovotz Y, Billiar TR. Inducible nitric oxide synthase and inflammatory diseases. Mol Med 2000; 6:347–373. 126. Groeneveld PH, Kwappenberg KM, Langermans JA, Nibbering PH, Curtis L. Relation between pro- and anti-inflammatory cytokines and the production of nitric oxide (NO) in severe sepsis. Cytokine 1997; 9:138–142.

Acute respiratory distress syndrome

434

127. Gustafsson LE, Leone AM, Persson MG, Wiklund NP, Moncada S. Endogenous nitric oxide is present in the exhaled air of rabbits, guinea pigs and humans. Biochem Biphys Res Commun 1991; 181:852–857. 128. Lewandowski K, Busch T, Lewandowski M, Keske U, Gerlach H, Falke KJ. Evidence of nitric oxide in the exhaled gas of Asian elephants (elephas maximus). Respir Physiol 1996; 106:91–98. 129. Alving K, Weitzberg E, Lundberg JM. Increased amount of nitric oxide in exhaled air of asthmatics. Eur Respir J 1993; 6:1368–1370. 130. Dillon WC, Hampl V, Shultz PJ, Rubins JB, Archer SL. Origins of breath nitric oxide in humans. Chest 1996; 110:930–938. 131. Busch T, Kuhlen R, Knorr M, Kelly K, Lewandowski K, Rossaint R, Falke KJ, Gerlach H. Nasal, pulmonary and autoinhaled nitric oxide at rest and during moderate exercise. Intens Care Med 2000; 26:391–399. 132. Kharitonov S, Alving K, Barnes PJ. Exhaled and nasal nitric oxide measurements: recommendations. The European Respiratory Society Task Force. Eur Respir J 1997; 10:1683– 1693. 133. Sartori C, Lepori M, Busch T, Duplain H, Hildebrandt W, Bartsch P, Nicod P, Falke KJ, Scherrer U. Exhaled nitric oxide does not provide a marker of vascular endothelial function in healthy humans. Am J Respir Crit Care Med 1999; 160:879–882. 134. Yates DH, Kharitonov SA, Thomas PS, Barnes PJ. Endogenous nitric oxide is decreased in asthmatic patients by an inhibitor of inducible nitric oxide synthase. Am J Respir Crit Care Med 1996; 154:247–250. 135. Kharitonov SA, Chung KF, Evans D, O’Connor BJ, Barnes PJ. Increased exhaled nitric oxide in asthma is mainly derived from the lower respiratory tract. Am J Respir Crit Care Med 1996; 153:1773–1780. 136. Kharitonov SA, Cailes JB, Black CM, du Bois RM, Barnes PJ. Decreased nitric oxide in the exhaled air of patients with systemic sclerosis with pulmonary hypertension. Thorax 1997; 52:1051–1055. 137. Kharitonov SA, Wells AU, O’Connor BJ, Cole PJ, Hansell DM, Logan-Sinclair RB, Barnes PJ. Elevated levels of exhaled nitric oxide in bronchiectasis. Am J Respir Crit Care Med 1995; 151:1889–1893. 138. Fisher AJ, Gabbay E, Small T, Doig S, Dark JH, Corris PA. Cross sectional study of exhaled nitric oxide levels following lung transplantation. Thorax 1998; 53:454–458. 139. Wang CH, Liu CY, Lin HC, Yu CT, Chung KF, Kuo HP. Increased exhaled nitric oxide in active pulmonary tuberculosis due to inducible NO synthase upregulation in alveolar macrophages. Eur Respir J 1998; 11:809–815. 140. Adrie C, Monchi M, Dinh-Xuan AT, Dall’Ava-Santucci J, Dhainaut JF, Pinsky MR. Exhaled and nasal nitric oxide as a marker of pneumonia in ventilated patients. Am J Respir Crit Care Med 2001; 163:1143–1149. 141. Brett SJ, Evans TW. Measurement of endogenous nitric oxide in the lungs of patients with the acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 157:993–997. 142. Dötsch J, Demirakca S, Terbrack HG, Hüls G, Rascher W, Kühl PG. Airway nitric oxide in asthmatic children and patients with cystic fibrosis. Eur Respir J 1996; 9:2537–2540. 143. Busch T, Deja M, Tegetoff K, Hell B, Bachmann S, Keske U, Falke KJ, Lewandowski K. The release of nitric oxide into the paranasal sinuses in patients with acute sinusitis. Am J Respir Crit Care Med 2000; 161:A323. 144. Kim JW, Min YG, Rhee CS, Lee CH, Koh YY, Rhyoo C, Kwon TY, Park SW. Regulation of mucociliary motility by nitric oxide and expression of nitric oxide synthase in the human sinus epithelial cells. Laryngoscope 2001; 111: 246–250. 145. Busch T, Deja M, Riskowski K, Campean V, Pfeilschifter B, Hell U, Kaisers K, Falke S, Bachmann K. Markedly reduced iNOS protein in sinus epithelium of patients with acute radiologic maxillary sinusitis. Am J Respir Crit Care Med 2002; 165:A325.

Modulation of pulmonary vascular tone in the acute respiratory

435

146. Mancinelli RL, McKay CP. Effects of nitric oxide and nitrogen dioxide on bacterial growth. Appl Environ Microbiol 1983; 46:198–202. 147. Croen KD. Evidence for antiviral effect of nitric oxide. Inhibition of herpes simplex virus type 1 replication. J Clin Invest 1993; 91:2446–2452. 148. Sanders SP, Siekierski ES, Porter JD, Richards SM, Proud D. Nitric oxide inhibits rhinovirusinduced cytokine production and viral replication in a human respiratory epithelial cell line. J Virol 1998; 72:934–942. 149. Yang B, Schlosser RJ, McCaffrey TV. Dual signal transduction mechanisms modulate ciliary beat frequency in upper airway epithelium. Am J Physiol 1996; 270:L745–751. 150. Moncada S, Higgs A. The 1-arginine-nitric oxide pathway. N Engl J Med 1993; 329:2002– 2012. 151. Uzlaner N, Priel Z. Interplay between the NO pathway and elevated [Ca2 +]i enhances ciliary activity in rabbit trachea. J Physiol 1999; 516:179–190. 152. Blum JJ, Hayes A, Jamieson GA Jr, Vanaman TC. Calmodulin confers calcium sensitivity on ciliary dynein ATPase. J Cell Biol 1980; 87:386–397. 153. Bonini NM, Nelson DL. Phosphoproteins associated with cyclic nucleotide stimulation of ciliary motility in Paramecium. J Cell Sci 1990; 95:219–230. 154. Satir P, Barkalow K, Hamasaki T. Ciliary beat frequency is controlled by a dynein light chain phosphorylation. Biophys J 1995; 68:222S. 155. Kini AR, Collins CA. Modulation of cytoplasmatic dynein ATPase activity by the accessory subunits. Cell Motil Cytoskeleton 2001; 48:52–60. 156. Kumar S, Lee IH, Plamann M. Cytoplasmic dynein ATPase activity is regulated by dynactindependent phosphorylation. J Biol Chem 2000; 275:31798–31804. 157. Nuutinen J, Rauch-Toskala E, Saano V, Joki S. Ciliary beating frequency in chronic sinusitis. Arch Otolaryngol Head Neck Surg 1993; 119:645–647. 158. Chilvers MA, O’Callaghan C. Analysis of ciliary beat pattern and beat frequency using digital high speed imaging: comparison with the photomultiplier and photodiode methods. Thorax 2000; 55:314–317. 159. Arnal JF, Didier A, Rami J, M’Rini C, Charlet JP, Serrano E, Besombes JP. Nasal nitric oxide is increased in allergic rhinitis. Clin Exp Allergy 1997; 27:358–362. 160. Murphy AW, Platts-Mills TA, Lobo M, Hayden F. Respiratory nitric oxide levels in experimental human influenza. Chest 1998; 114:452–456. 161. Lundberg JON, Nordvall SL, Weitzberg E, Kollberg H, Alving K. Exhaled nitric oxide in paediatric asthma and cystic fibrosis. Arch Dis Child 1996; 75:323–326. 162. Haubitz M, Busch T, Gerlach M, Schäfer S, Brunkhorst R, Falke K, Koch KM, Gerlach H. Exhaled nitric oxide in patients with Wegener’s granulomatosis. Eur Respir J 1999; 14:113– 117. 163. Baraldi E, Azzolin NM, Biban P, Zacchello F. Effect of antibiotic therapy on nasal nitric oxide concentration in children with acute sinusitis. Am J Respir Crit Care Med 1997; 155:1680– 1683. 164. Lindberg S, Cervin A, Runer T. Nitric oxide (NO) production in the upper airways is decreased in chronic sinusitis. Acta Otolaryngol 1997; 117:113–117. 165. Arnal JF, Flores P, Rami J, Murris-Espin M, Bremont F, Pasto I, Aguilla M, Serrano E, Didier A. Nasal nitric oxide concentration in paranasal sinus inflammatory diseases. Eur Respir J 1999; 13:307–312. 166. Lee KH, Tan PS, Rico P, Delgado E, Kellum JA, Pinsky MR. Low levels of nitric oxide as contaminant in hospital compressed air: physiologic significance? Crit Care Med 1997; 25:1143–1146. 167. Lum LC, Tan PS, Saville A, Venkataraman ST, Pinsky MR. Occult nitric oxide inhalation improves oxygenation in mechanically ventilated children. J Pediatr 1998; 133:613–616.

Acute respiratory distress syndrome

436

168. Lundberg JON, Settergren G, Gelinder S, Lundberg JM, Alving K, Weitzberg E. Inhalation of nasally derived nitric oxide modulates pulmonary function in humans. Acta Physiol Scand 1996; 158:343–347. 169. Settergren G, Angdin M, Astudillo R, Gelinder S, Liska J, Lundberg JON, Weitzberg E. Decreased pulmonary vascular resistance during nasal breathing: modulation by endogenous nitric oxide from the paranasal sinuses. Acta Physiol Scand 1998; 163:235–239. 170. Gerlach H, Pappert D, Lewandowski K, Rossaint R, Falke KJ. Long-term inhalation with evaluated low doses of nitric oxide for selective improvement of oxygenation in patients with adult respiratory distress syndrome. Intens Care Med 1993; 19:443–449. 171. Moncada S, Gryglewski R, Bunting S, Vane JR. An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature 1976; 263:663–665. 172. Vane JR, Anggard EE, Botting RM. Regulatory functions of the vascular endothelium. N Engl J Med 1990; 323:27–36. 173. Tateson JE, Moncada S, Vane JR. Effects of prostacyclin (PGX) on cyclic AMP concentrations in human platelets. Prostaglandins 1977; 13:389–397. 174. Walmrath D, Schneider T, Pilch J, Grimminger F, Seeger W. Aerosolised prostacyclin in adult respiratory distress syndrome. Lancet 1993; 342:961–962. 175. Zwissler B, Kemming G, Habler O, Kleen M, Merkel M, Haller M, Briegel J, Welte M, Peter K. Inhaled prostacyclin (PGI2) versus inhaled nitric oxide in adult respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154: 1671–1677. 176. Van Heerden PV, Blythe D, Webb SA. Inhaled aerosolized prostacyclin and nitric oxide as selective pulmonary vasodilators in ARDS—a pilot study. Anaesth Intens Care 1996; 24:564– 568. 177. Fuller HD, Dolovich MB, Posmituck G, Pack WW, Newhouse MT. Pressurized aerosol versus jet aerosol delivery to mechanically ventilated patients. Comparison of dose to the lungs. Am Rev Respir Dis 1990; 141:440–444. 178. O’Riordan TG, Greco MJ, Perry RJ, Smaldone GC. Nebulizer function during mechanical ventilation. Am Rev Respir Dis 1992; 145:1117–1122. 179. Van Heerden PV, Barden A, Michalopoulos N, Bulsara MK, Roberts BL. Dose-response to inhaled aerosolized prostacyclin for hypoxemia due to ARDS. Chest 2000; 117:819–827. 180. Olschewski H, Walmrath D, Schermuly R, Ghofrani A, Grimminger F, Seeger W. Aerosolized prostacyclin and iloprost in severe pulmonary hypertension. Ann Intern Med 1996; 124:820– 824. 181. Hoeper MM, Olschewski H, Ghofrani HA, Wilkens H, Winkler J, Borst MM, Niedermeyer J, Fabel H, Seeger W. A comparison of the acute hemodynamic effects of inhaled nitric oxide and aerosolized iloprost in primary pulmonary hypertension. German PPH study group. J Am Coll Cardiol 2000; 35:176–182. 182. Reyes A, Roca J, Rodriguez-Roisin R, Torres A, Ussetti P, Wagner PD. Effect of almitrine on ventilation-perfusion distribution in adult respiratory distress syndrome. Am Rev Respir Dis 1988; 137:1062–1067. 183. Melot C, Dechamps P, Hallemans R, Decroly P, Mols P. Enhancement of hypoxic pulmonary vasoconstriction by low dose almitrine bismesylate in normal humans. Am Rev Respir Dis 1989; 139:111–119. 184. Naeije R, Melot C, Mols P, Hallemans R, Naeije N, Cornil A, Sergysels R. Effects of almitrine in decompensated chronic respiratory insufficiency. Bull Eur Physiopathol Respir 1981; 17:153–161. 185. Marshall BE, Hanson CW, Frasch F, Marshall C. Role of hypoxic pulmonary vasoconstriction in pulmonary gas exchange and blood flow distribution. 2. Pathophysiology. Intens Care Med 1994; 20:379–389. 186. Reyes A, Lopez-Messa JB, Alonso P. Almitrine in acute respiratory failure. Effects on pulmonary gas exchange and circulation. Chest 1987; 91:388–393.

Modulation of pulmonary vascular tone in the acute respiratory

437

187. Sommerer A, Kaisers U, Dembinski R, Bubser HP, Falke KJ, Rossaint R. Dose-dependent effects of almitrine on hemodynamics and gas exchange in an animal model of acute lung injury. Intens Care Med 2000; 26:434–441. 188. Michard F, Wolff MA, Herman B, Wysocki M. Right ventricular response to high-dose almitrine infusion in patients with severe hypoxemia related to acute respiratory distress syndrome. Crit Care Med 2001; 29:32–36. 189. Payen DM, Gatecel C, Plaisance P. Almitrine effect on nitric oxide inhalation in adult respiratory distress syndrome. Lancet 1993; 341:1664. 190. Wysocki M, Delclaux C, Roupie E, Langeron O, Liu N, Herman B, Lemaire F, Brochard L. Additive effect on gas exchange of inhaled nitric oxide and intravenous almitrine bismesylate in the adult respiratory distress syndrome. Intens Care Med 1994; 20:254–259. 191. Lu Q, Mourgeon E, Law-Koune JD, Roche S, Vezinet C, Abdennour L, Vicaut E, Puybasset L, Diaby M, Coriat P, Rouby J-J. Dose-response curves of inhaled nitric oxide with and without intravenous almitrine in nitric oxide-responding patients with acute respiratory distress syndrome. Anesthesiology 1995; 83:929–943. 192. Jolliet P, Bulpa P, Ritz M, Ricou B, Lopez J, Chevrolet JC. Additive beneficial effects of the prone position, nitric oxide, and almitrine bismesylate on gas exchange and oxygen transport in acute respiratory distress syndrome. Crit Care Med 1997; 25:786–794. 193. Guinard N, Beloucif S, Gatecel C, Mateo J, Payen D. Interest of a therapeutic optimization strategy in severe ARDS. Chest 1997; 111:1000–1007. 194. Gillart T, Bazin JE, Cosserant B, Guelon D, Aigouy L, Mansoor O, Schoeffler P. Combined nitric oxide inhalation, prone positioning and almitrine infusion improve oxygenation in severe ARDS. Can J Anaesth 1998; 45:402–409. 195. Lincoln TM, Cornwell TL. Intracellular cyclic GMP receptor proteins. FASEB J 1993; 7:328– 338. 196. Martin W, Furchgott RF, Villani GM, Jothianandan D. Phosphodiesterase inhibitors induce endothelium-dependent relaxation of rat and rabbit aorta by potentiating the effects of spontaneously released endothelium-derived relaxing factor. J Pharmacol Exp Ther 1986; 237:539–547. 197. Stamler JS, Loh E, Roddy MA, Currie KE, Creager MA. Nitric oxide regulates basal systemic and pulmonary vascular resistance in healthy humans. Circulation 1994; 89:2035–2040. 198. Celermajer DS, Dollery C, Burch M, Deanfield JE. Role of endothelium in the maintenance of low pulmonary vascular tone in normal children. Circulation 1994; 89:2041–2044. 199. Rabe KF, Tenor H, Dent G, Schudt C, Nakashima M, Magnussen H. Identification of PDE isozymes in human pulmonary artery and effect of selective PDE inhibitors. Am J Physiol 1994; 266:L536–543. 200. Beavo JA. Cyclic nucleotide phosphodiesterases: functional implications of multiple isoforms. Physiol Rev 1995; 75:725–748. 201. Fisher DA, Smith JF, Pillar JS, St Denis SH, Cheng JB. Isolation and characterization of PDE9A, a novel human cGMP-specific phosphodiesterase. J Biol Chem 1998; 273:15559– 15564. 202. Celermajer DS, Cullen S, Deanfield JE. Impairment of endothelium-dependent pulmonary artery relaxation in children with congenital heart disease and abnormal pulmonary hemodynamics. Circulation 1993; 87:440–446. 203. Cohen AH, Hanson K, Morris K, Fouty B, McMurty IF, Clarke W, Rodman DM. Inhibition of cyclic 3′-5′-guanosine monophosphate-specific phosphodiesterase selectively vasodilates the pulmonary circulation in chronically hypoxic rats. J Clin Invest 1996; 97:172–179. 204. Ichinose F, Adrie C, Hurford WE, Zapol WM. Prolonged pulmonary vasodilator action of inhaled nitric oxide by Zaprinast in awake lambs. J Appl Physiol 1995; 78:1288–1295. 205. Ziegler JW, Ivy DD, Wiggins JW, Kinsella JP, Clarke WR, Abman SH. Effects of dipyridamole and inhaled nitric oxide in pediatric patients with pulmonary hypertension. Am J Respir Crit Care Med 1998; 158:1388–1395.

Acute respiratory distress syndrome

438

206. Boolell M, Gepi-Attee S, Gingell JC, Allen MJ. Sildenafil, a novel effective oral therapy for male erectile dysfunction. Br J Urol 1996; 78:257–261. 207. Kleinsasser A, Loeckinger A, Hoermann C, Puehringer F, Mutz N, Bartsch G, Lindner KH. Sildenafil modulates hemodynamics and pulmonary gas exchange. Am J Respir Crit Care Med 2001; 163:339–343. 208. Weimann J, Ullrich R, Hromi J, Fujino Y, Clark MW, Bloch KD, Zapol WM. Sildenafil is a pulmonary vasodilator in awake lambs with acute pulmonary hypertension. Anesthesiology 2000; 92:1702–1712. 209. Hobbs AJ, Ignarro LJ. The nitric oxide-cyclic GMP signal transduction system. In: Zapol WM, Bloch KD, eds. Lung Biology in Health and Disease, Vol. 98: Nitric Oxide and the Lung. New York: Marcel Dekker, 1997:1–57. 210. Schermuly RT, Ghofrani HA, Enke B, Weissmann N, Grimminger F, Seeger W, Schudt C, Walmrath D. Low-dose systemic phosphodiesterase inhibitors amplify the pulmonary vasodilatory response to inhaled prostacyclin in experimental pulmonary hypertension. Am J Respir Crit Care Med 1999; 160: 1500–1506. 211. Schermuly RT, Roehl A, Weissmann N, Ghofrani HA, Schudt C, Tenor H, Grimminger F, Seeger W, Walmrath D. Subthreshold doses of specific phosphodiesterase type 3 and 4 inhibitors enhance the pulmonary vasodilatory response to nebulized prostacyclin with improvement in gas exchange. J Pharmacol Exp Ther 2000; 292:512–520. 212. Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, Yazaki Y, Goto K, Masaki T. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 1988; 332:411–415. 213. Goldie RG. Endothelin receptor subtypes: distribution and function in the lung. Pulm Pharmacol Ther 1998; 11:89–95. 214. Battistini B, Dussault P. Biosynthesis, distribution and metabolism of endothelins in the pulmonary system. Pulm Pharmacol Ther 1998; 11:79–88. 215. Fujitani Y, Trifilieff A, Tsuyuki S, Coyle AJ, Bertrand C. Endothelin receptor antagonists inhibit antigen-induced lung inflammation in mice. Am J Respir Crit Care Med 1997; 155:1890–1894. 216. Finsnes F, Skjonsberg OH, Tonnessen T, Naess O, Lyberg T, Christensen G. Endothelin production and effects of endothelin antagonism during experimental airway inflammation. Am J Respir Crit Care Med 1997; 155:1404–1412. 217. Michael JR, Markewitz BA. Endothelins and the lung. Am J Respir Crit Care Med 1996; 154:555–581. 218. Baraniuk JN, Molet S, Mullol J, Naranch K. Endothelin and the airway mucosa. Pulm Pharmacol Ther 1998; 11:113–123. 219. Benigni A, Remuzzi G. Endothelin antagonists. Lancet 1999; 353:133–138. 220. Battistini B, Dussault P. Blocking of the endothelin system: the development of receptor antagonists. Pulm Pharmacol Ther 1998; 11:97–112. 221. Noguchi K, Ishikawa K, Yano M, Ahmed A, Cortes A, Abraham WM. Endothelin-1 contributes to antigen-induced airway hyperresponsiveness. J Appl Physiol 1995; 79:700–705. 222. Chen SJ, Chen YF, Opgenorth TJ, Wessale JL, Meng QC, Durand J, DiCarlo VS, Oparil S. The orally active nonpeptide endothelin A-receptor antagonist A-127722 prevents and reverses hypoxia-induced pulmonary hypertension and pulmonary vascular remodeling in SpragueDawley rats. J Cardiovasc Pharmacol 1997; 29:713–725. 223. DiCarlo VS, Chen SJ, Meng QC, Durand J, Yano M, Chen YF, Oparil S. ETA-receptor antagonist prevents and reverses chronic hypoxia-induced pulmonary hypertension in rat. Am J Physiol 1995; 269:L690–L697. 224. Holm P, Liska J, Franco-Cereceda A. The ETA receptor antagonist, BMS-182874, reduces acute hypoxic pulmonary hypertension in pigs in vivo. Cardiovasc Res 1998; 37:765–771.

Modulation of pulmonary vascular tone in the acute respiratory

439

225. Oparil S, Chen SJ, Meng QC, Elton TS, Yano M, Chen YF. Endothelin-A receptor antagonist prevents acute hypoxia-induced pulmonary hypertension in the rat. Am J Physiol 1995; 268:L95–L100. 226. Underwood DC, Bochnowicz S, Osborn RR, Louden CS, Hart TK, Ohlstein EH, Hay DW. Chronic hypoxia-induced cardiopulmonary changes in three rat strains: inhibition by the endothelin receptor antagonist SB 217242. J Cardiovasc Pharmacol 1998; 31:S453–S455. 227. Rubin LJ, Badesch DB, Barst RJ, Galie N, Black CM, Keogh A, Pulido T, Frost A, Roux S, Leconte I, Landzberg M, Simonneau G. Bosentan therapy for pulmonary arterial hypertension. N Engl J Med 2002; 346:896–903. 228. Battistini B, Steil AA, Jancar S, Sirois P. Roles of endothelins and their receptors in immune complex-induced/polymorphonuclear-mediated lung injury (reversed passive Arthus reaction) in CD-1 mice. Pulm Pharmacol Ther 1998; 11:165–172. 229. Polakowski JS, Opgenorth TJ, Pollock DM. ETA receptor blockade potentiates the bronchoconstrictor response to ET-1 in the guinea pig airway. Biochem Biophys Res Commun 1996; 225:225–231. 230. Kaisers U, Busch T, Wolf S, Lohbrunner H, Wilkens K, Hocher B, Boemke W. Inhaled endothelin A antagonist improves arterial oxygenation in experimental acute lung injury. Intens Care Med 2000; 26:1334–1342. 231. Münter K, Ehmke H, Kirchengast M. Maintenance of blood pressure in normotensive dogs by endothelin. Am J Physiol 1999; 276:H1022–H1027. 232. Dupuis J, Stewart DJ, Cernacek P, Gosselin G. Human pulmonary circulation is an important site for both clearance and production of endothelin-1. Circulation 1996; 94:1578–1584. 233. Deja M, Wolf S, Busch T, Petersen B, Jaghzies U, Boemke W, Kaisers U. The inhaled ETA receptor antagonist LU-135252 acts as a selective pulmonary vasodilator. Clin Sci 2002; 103:21S-24S.

18 Glucocorticoid Therapy for the Acute Respiratory Distress Syndrome RICHARD B.GOODMAN and LEONARD D.HUDSON University of Washington School of Medicine Seattle, Washington, U.S.A.

I. Introduction Corticosteroids have been considered as a potential treatment for the acute respiratory distress syndrome (ARDS) since its original description in 1967 (1). However, 35 years hence, definitive evidence to support their use is lacking. In this chapter we will review the published clinical trials of glucocorticoids in ARDS. Because ARDS is commonly present in patients with severe sepsis and septic shock, the clinical trials of corticosteroids in sepsis are relevant to this discussion. We will review the published clinical experience with corticosteroids in patients with persistent fibroproliferative-phase ARDS and the more recent experience addressing an evolving concept of corticosteroid replacement therapy in patients with “relative” adrenal insufficiency. The topic has also recently been reviewed by Luce (2). Our goal is to provide the reader with the background and rationale for continued interest in studying corticosteroids as a clinical intervention in ARDS.

II. High-Dose, Short-Course Corticosteroid Trials in Early Sepsis There have been 11 clinical intervention trials using short courses (1–2 days) of very high doses of corticosteroids (≥2 g/day of prednisone-equivalent) studying patients with severe sepsis and/or septic shock, with or without ARDS (Table 1) (3–13). With a single exception (6), all of them studied patients early in their course of disease. Only 4 of these studies focused specifically on treatment of patients with ARDS or prevention of ARDS in patients at risk (8, 10, 12, 13). However, because severe sepsis is the most common risk factor for ARDS (14, 15), all of these studies are relevant and, thus, are included in this review. The first of these early-intervention, high-dose trials was reported by Klastersky et al. in 1971 (3). They randomized 85 patients with disseminated cancer and life-threatening infection to receive placebo or betamethasone 1 mg/kg/day intravenously for 3 days in a double-blinded protocol. The majority of the patients had gram-negative pneumonia. Although there was one 14-year-old patient, the remainder were >30 and most were >60 years of age. Steroid-treated patients defervesced more quickly, but they found neither early nor late mortality benefits. They found no statistically significant difference in

Glucocorticoid therapy for the acute respiratory distress syndrome

441

persistent infection at autopsy in the patients who died, but there was a trend for a higher rate of respiratory bacterial superinfections in the steroid-treated group (28% vs. 6%; 0.1>p>0.05). The authors concluded that there was no mortality benefit from pharmacological doses of betamethasone in these patients. In 1976, Schumer (4) published the only study in this series of trials showing a mortality benefit from corticosteroids. In a prospective, randomized, double-blind, placebo-controlled trial, they compared the efficacy and safety of methylprednisolone (30 mg/kg), dexamethasone (3 mg/kg), and placebo in 172 consecutive patients with septic shock. Drug was administered as a single dose, and some patients received a repeated dose after 4 hours “if necessary,” although the criteria for repeated doses were not described. All patients were treated for only 1 day. They found that corticosteroids substantially reduced mortality (38% vs. 10%; p<0.001) and that there was no difference in efficacy comparing methylprednisolone and dexamethasone. In addition, 328 patients were studied retrospectively, where 168 received corticosteroids and 160 did not. They found a similar reduction in mortality associated with corticosteroid administration (43% vs. 14%; p<0.001). Steroid-related complications in both studies included gastrointestinal bleeding, hyperglycemia, and psychosis occurring at a rate <6%. The author was criticized for not clearly defining the time of vital status determination. In addition, the retrospective study spanned the same time period as the prospective study, thus raising the question of how enrollment determinations were made in the prospective study. Despite published controversy (4, 16, 17), this publication was primarily responsible for the wide use of corticosteroids in patients with septic shock over the next several years. Thompson and colleagues (5) published an abstract in 1976 reporting 60 patients with clinical shock (predominantly septic shock) at two medical centers who were randomly assigned to treatment with methylprednisolone 30 mg/kg or placebo. Repeated doses were administered after 2, 4, and 24 hours if shock persisted. Patients were identified and treated within 9 hours after the onset of shock. They found no effect on blood gases, vital signs, serum solutes, or early (48 hr) or late (hospital discharge) mortality. In 1984 Lucas and Ledgerwood (7) reported 48 patients with clinical septic shock, 23 of whom were treated within 36 minutes of presentation with dexamethasone 3 mg/kg by continuous infusion for 2 days. All patients were intubated and ventilated with average P:F ratios of <300. These investigators found improvement in mean arterial pressure (106 vs. 95 mmHg), diastolic pressure (88 vs. 78 mmHg), and central venous pressure (16 vs. 10 cmH2O) in steroid-treated vs. placebo-treated patients in the first 48 hours (p<0.05 for all), but the differences were not maintained. Further, oxygenation, as measured by shunt fraction and P:F ratio was adversely affected by steroid treatment (p<0.05), although the magnitude of the effects was minimal. There was no difference in mortality between groups. They concluded that massive doses of corticosteroids in patients with sepsis would no longer be a part of their practice. In 1984 Sprung et al. (6) identified 59 patients who were later in their course of severe septic shock (17.5 hr after onset) with prospectively defined diagnostic criteria. Patients were randomly assigned to receive methylprednisolone (30 mg/kg) or dexamethasone (6 mg/kg) or no active drug, with a repeated dose after 4 hours if shock persisted. They found that early after initiation of treatment, corticosteroids resulted in a higher rate of shock reversal, particularly in the subpopulation of patients identified within 4 hours of shock onset. Similarly, mortality in the first 5 days was better in the patients receiving

Acute respiratory distress syndrome

442

corticosteroids. However, neither the shock reversal nor mortality benefits were seen later in the course of disease. They reported no effect on the incidence of ARDS in this study. They found a higher rate of super-infections in patients treated with dexamethasone compared with control (36% vs. 6%; p<0.05). They concluded that corticosteroids do not improve overall survival in severe, late septic shock but may be helpful early in the course in certain subgroups of patients. In 1985 Weigelt et al. (8) studied 81 acutely ill, mechanically ventilated patients at risk for ARDS. They enrolled patients based on their P:F ratio and included patients with a

mmHg on 40%

and a

Table 1 Early Intervention Trials in Sepsis with High-Dose, Short-Course Corticosteroids Trial, location (Ref.)

Design

Klastersky et al., Randomized, double1971, Brussels (3) blind, placebocontrolled, single center

N

Population

85

Disseminated cancer and “lifethreatening infection” (mostly gram-negative pneumonia), ~20% were neutropenic, age 14–79 (most>60)

Schumer, 1976, Chicago (4)

Randomized, doubleblind, placebocontrolled, single center, and a retrospective study

172 (prospective), 328 (retrospective)

Septic shock, all with positive blood cultures. Age 22–84 (mean 50) in prospective study; 20–86 (mean 56) in retrospective study. Gender overall >90% male

Thompson et al., 1978, Cleveland and Baltimore (5)

Randomized, placebocontrolled, two centers

60

Shock (85% septic, 15% cardiogenic), treated within first 9 hr, age and gender not reported

Lucas and Ledgerwood, 1984, Detroit (7)

Randomized, openlabel, single center

48

Septic shock (mostly abdominal source), all intubated with average P: F <300, treated within 36 minutes of shock onset, average age 55

Sprung et al., 1984, Miami (6)

Prospective, randomized, openlabel, controlled, two centers

59

Late (12–24 hr after onset), severe septic shock (clearly defined), average age 54, 78% male

Weigelt et al., 1985, Dallas (8)

Randomized, doubleblind, placebocontrolled, single center

81

SICU patients with early respiratory failure. Age 19–75 (average age ~46), 79% male

Bone et al., 1987, 19 U.S. medical centers (9)

Prospective, randomized, doubleblind, placebocontrolled, multi-center

382

Severe sepsis and septic shock (clearly defined), treated within 2 hr of meeting entry criteria, mean age 54, 62% male

Glucocorticoid therapy for the acute respiratory distress syndrome

443

VA coop, 1987, 10 VA medical centers (11)

Prospective, randomized, doubleblind, placebocontrolled, multi-center

223

VA patients with systemic sepsis (clearly defined) and normal sensorium, treated within 3 hr of meeting entry criteria; mean age 61, predominantly male

Bernard et al., 1987, 7 U.S. medical centers (12)

Prospective, randomized, doubleblind, placebocontrolled, multi-center

99

Patients with early ARDS (P: F <175, bilateral CXR infiltrates, PCW <18 mmHg), enrolled within 12 hr of meeting criteria; mean age 55

Treatment

Betamethasone 1 mg/kg/day ×3 days

Prednisone doseequivalent (day 1)a 420 mg

Placebo vs. treatment mortality

Outcomes

40% vs. 40%

No mortality benefit at 48 hr, 10 days, or 30 days. Trend for higher rate of superinfections in steroid group. Steroid-treated patients defervesced faster.

Dexamethasone 3 mg/kg or ~2 g methylprednisolone 30 mg/kg for 1 day as single dose or repeated once “if necessary”

38% vs. 10% (prospective study), 43% vs. 14% (retrospective study), p<0.001.

Improved mortality (time undefined), no increased complications.

Methylprednisolone 30 mg/kg IV and repeated at 2, 8, and 24 hours as long as shock persisted

10 g (average dose)

78% vs. 79%

No mortality benefit (at hospital discharge). No improvement in gas exchange or vital signs. No overt toxicity.

Dexamethasone 3 mg/kg/day×2 days (continuous infusion)

2.8 g

20% vs. 22%

No 14-day mortality benefit, steroid-treated patients had shortterm improvement in arterial and central venous pressures. Worse gas exchange in steroid-treated group. No increased infectious complications.

Methylprednisolone 30 2.5–5 g mg/kg or dexamethasone 6 mg/kg; dose repeated once after 4 hr if shock persisted

69% vs. 77%

Improved early rate of shock reversal and improved early mortality (at 133 hr), but overall shock reversal rate and mortality at hospital discharge not improved. No difference in ARDS incidence. Superinfections higher with dexamethasone.

Methylprednisolone 30 mg/kg q6hr ×48 hr

31% vs. 46% (p=0.177)

Worse progression to ARDS and higher infectious complications

10 g

Acute respiratory distress syndrome

444

in steroid-treated patients. No mortality benefit at hospital discharge. Methylprednisolone 30 ~2.5 g mg/kg IV in 24 hr (divided q6hr×4 doses)

25% vs. 34% (p=0.06)

No benefit in shock prevention, shock reversal, or mortality (at 14 days). Worse outcomes with corticosteroid in subgroup with renal failure. Higher attributable mortality from secondary infections in steroid group.

Methylprednisolone 30 6–12 g mg/kg IV bolus, followed by continuous infusion of 5 mg/kg/hr ×9 hr (maximum 10 g)

22% vs. 21%

No mortality benefit (at 14 days). Slower resolution of secondary infections.

Methylprednisolone 30 mg/kg q6hr ×24 hr IV

10 g

63% vs. 60%

No mortality benefit (at 45 days). No difference in ARDS reversal or infectious complications.

Trial, location (Ref.)

Design

N

Population

Bone et al., 1987, Prospective, randomized, 304 Subgroup of patients simultaneously enrolled 17 U.S. medical double-blind, placeboat 17 of 19 centers participating in the centers (10) controlled multi-center prospective randomized trial of severe sepsis and septic shock patients (9) Luce et al., 1988, Prospective, randomized, San Francisco double-blind, placebo(13) controlled, single center

75 Patients with septic shock as risk-factor for ARDS, age 18–76 (average age 52), 76% male

a

Calculation based on 70 kg individual, conversions to prednisone-equivalent from Chin et al. (77) and Jantz and Sahn (78).

mmHg on 100% Underlying risk factors for ARDS included sepsis (32%), shock (43%), massive transfusion (36%), lung contusion (21%), alveolar hypoventilation (23%), and aspiration (16%). Multiple risk factors were present in 67% of the of the patients. Patients were treated in a randomized, double-blind, placebocontrolled trial with methylprednisolone 30 mg/kg every 6 hours for 48 hours. The mean on 100% was 270 mmHg (steroid group) and 256 mmHg (placebo qualifying group), and thus, by today’s standards, these patients met the criteria for acute lung injury (ALI) at the time of entry into the study with P:F ratios of <300. These investigators found that the rate of progression to ARDS was higher in steroid-treated patients (64% vs. 33%; p=0.008), although their definition of ARDS was somewhat unconventional by current standards. Steroid-treated patients had a higher rate of infectious complications (77% vs. 43%; p=0.001), had higher blood glucose values, and had a trend toward higher shunt fractions. They concluded that steroids failed to prevent ARDS or to provide a survival advantage and were associated with an increased rate of infectious

Glucocorticoid therapy for the acute respiratory distress syndrome

445

complications. This was one of the few studies with steroid dosing for >24 hours and provided data to raise concerns about the association between infection and duration of treatment. In 1987 Bone et al. (9) published in the New England Journal of Medicine a randomized controlled trial conducted at 19 centers enrolling 382 patients with severe sepsis or septic shock—both explicitly and prospectively defined. Patients were randomly assigned to treatment with methylprednisolone 30 mg/kg given in four divided doses over 24 hours or placebo. Placebo vs. treatment mortality

Outcomes

Treatment

Prednisone doseequivalent (day 1)a

Methylprednisolone 30 mg/kg IV in 24 hr (divided q6hr ×4 doses)

~2.5 g

22% vs. 52% (p=0.004)

Higher mortality (at 14 days), fewer patients with ARDS reversal, and trend to increased ARDS incidence in steroidtreated patients.

Methylprednisolone 30 mg/kg q6hr ×4 doses IV

10 g

54% vs. 58%

No difference in incidence, time to onset, resolution, time to resolution of ARDS, or hospital mortality.

They found no treatment benefit in terms of shock prevention, shock reversal, or mortality. There was a trend toward worse mortality in the steroid treated group (p=0.06). Moreover, in the subpopulation of patients with elevated creatinines at study entry, they found higher mortality with corticosteroids (29% vs. 59%; p<0.01). Finally, they identified a higher attributable mortality from secondary infections in the steroid-treated patients (7% vs. 34%; p<0.015). They concluded that treatment with high-dose corticosteroids provided no benefit in patients with severe sepsis or septic shock. In the same issue of the New England Journal of Medicine in 1987, the VA Systemic Sepsis Cooperative Study Group published their results. Ten VA Medical Centers prospectively studied 223 patients with early systemic sepsis in a randomized, doubleblind, placebo-controlled trial of short-term, high-dose methylprednisolone (11). Treated patients received methylprednisolone 30 mg/kg IV, followed by a constant infusion of 5 mg/kg/hr for 9 hours (maximal dose 10 g). There was no mortality benefit in the whole group or in the subgroups of patients with gram-positive or gram-negative infections. There was slower resolution of secondary infections in the steroid-treated group that did not appear to affect mortality. They concluded that early, high-dose corticosteroids should not be used in this setting. Also in 1987, Bernard and colleagues (12) studied 99 patients who had ARDS at the time of entry in a randomized, double-blind, placebo-controlled trial of methylprednisolone therapy (30 mg/kg every 6 hr for 24 hr), where patients were treated within the first 12 hours of meeting ARDS inclusion criteria. The most common underlying cause for ARDS in this series was sepsis (27%), with other risks including gastric acid aspiration (18%), pancreatitis (4%), or multiple risk factors combined (42%). They found no mortality benefit from steroids, and the rate of ARDS reversal was not

Acute respiratory distress syndrome

446

sig-nificantly different compared to placebo-treated patients. Infectious complications were similar in both groups. Later in 1987, Bone and colleagues (10) published a separate subgroup analysis of 304 of their original 382 septic patients, studying the incidence and resolution of ARDS. They reported a trend toward an increased incidence of ARDS in the steroid-treated group (32% vs. 25%; p=0.10). Of the patients with ARDS, fewer of the steroid-treated patients resolved their lung injury (31% vs. 61%; p=0.005), and the 14-day mortality was higher in the steroid-treated group (52% vs. 22%; p=0.004). Their data indicate that corticosteroids do not prevent development of ARDS in patients with severe sepsis, may impede reversal of ARDS, and could worsen mortality in this population. In 1988 Luce and colleagues (13) prospectively identified 87 patients who were clinically suspected of having septic shock but without evidence of lung injury, 75 of whom were subsequently identified with positive blood cultures. Patients were treated within 4 hours of identification in a randomized, double-blind, placebo-controlled trial of methylprednisolone 30 mg/kg every 6 hours for 4 doses or placebo, and their lung function was followed. They found no difference in incidence, time to onset, resolution or time to resolution of ARDS, or hospital mortality between groups. They concluded that methylprednisolone neither prevents parenchymal lung injury nor improves mortality when administered early to patients with septic shock. In summary, there have been 11 randomized controlled clinical trials investigating the utility of high-dose corticosteroids in several populations of patients with severe sepsis or septic shock, and many included significant numbers of patients with ARDS. Four of these studies specifically focused on patients with ARDS or at risk for ARDS (8, 10, 12, 13), and mortality in these controlled trials has been the topic of two meta-analyses (18, 19). Based on this current body of evidence, high doses of corticosteroids given for 1–2 days in patients who are early in the course of critical illness (severe sepsis or septic shock) are ineffective in preventing the onset of ARDS, hastening resolution of ARDS, or reducing mortality, and may be associated with a higher rate of complications such as secondary infections.

III. Prolonged, Moderate-Dose Corticosteroids in Persistent ARDS Certain patients with ARDS who survive the initial inciting event but do not quickly resolve their lung injury can have a prolonged and expensive course of mechanical ventilation in the ICU (20). These patients have a histological picture that has progressed beyond the diffuse alveolar damage seen early in ARDS to a picture of fibroproliferation in this phase of persistent ARDS (21–25). As pointed out by Ashbaugh and Maier (26), these patients may represent a clinically distinct subpopulation of ARDS with unique histology similar to idiopathic pulmonary fibrosis. Several small case series and retrospective case-control studies have been reported in the last 17 years that showed suprisingly high survival rates in patients with late-phase, “persistent” ARDS (Table 2) (26–33). All of these studies used more moderate doses of corticosteroids compared with the high doses used in the previous studies in patients with sepsis and ARDS. Meduri and colleagues (34) have published the only randomized controlled trial in this patient population, and they reported efficacy of corticosteroids in

Glucocorticoid therapy for the acute respiratory distress syndrome

447

persistent ARDS, although the study has been characterized as a phase II trial (35) and the results have been controversial (36, 37). These reports are the foundation for the continued active interest in studying the use of corticosteroids in patients with late-phase, “persistent” ARDS. Ashbaugh and Maier (26) likened persistent ARDS to idiopathic pulmonary fibrosis (IFF) and first recognized its potential responsiveness to corticosteroids. In their paper, published in 1985, they reported 10 consecutively identified patients with IFF in ARDS at two medical centers in Seattle and Boise and performed lung biopsies in each patient to establish the diagnosis and rule out active infection. The patients’ ages ranged from 6 to 59 years. The underlying risk factors for ARDS were multiple trauma in 8, coronary artery bypass in 1, and kidney transplant removal in 1. At the time of study entry, patients clinically had sepsis syndrome without an identifiable source of infection. Steroids were started 6–22 days (median 10 days) after the onset of ARDS. All patients were treated with methylprednisolone 125 mg IV every 6 hours. When a response in terms of reduced and PEEP requirements was observed, steroids were tapered in 50–100 mg/day increments every 3–5 days. Patients were converted to oral therapy when they were extubated and had achieved a dose as low as 80–100 mg/day. Oral prednisone was then tapered over 3–6 weeks. All patients had improved oxygenation within 3 days. Eight of the 10 patients recovered and were discharged from the hospital. Complications reported included infections, impaired wound healing, and GI bleeding, occurred in 7 of 10 patients, and caused the death of 2 patients (septicemia and intra-abdominal sepsis with hemorrhage). The reported an 80% survival rate. In 1990, and subsequently extended in 1996, Hooper and Kearl (27, 32) reported an uncontrolled series of 26 patients with ARDS which had persisted for 3–40 days who were all given methylprednisolone 125–250 mg every 6 hours with continued treatment and tapering over several weeks. Survival was 81%, and they stated that infection-related morbidity and mortality were surprisingly low.

Table 2 Late phase, “Persistent” ARDS Treatment with Prolonged, Moderate-Dose Corticosteroids Trial, location (Ref.)

Design

N

Population

Ashbaugh Observational 10 Consecutive and treatment patients with Maier, study clinical and 1985, histological Seattle fibroproliferative and Boise phase ARDS, 6– (26) 22 days (median 10 days) after onset of ARDS, age range 6–59, 70% male

Treatment

“Prednisone Outcomes doseequivalent” (day 1)a

Methylprednisolone 625 mg 125 mg q6hr, tapered by 50–100 mg/day when oxygenation improved, converted to oral prednisone postextubation with tapering over 3–6 weeks

80% survival to hospital discharge

Acute respiratory distress syndrome

Hooper and Kearl, 1990 and 1996, Phoenix (27, 32)

Observational 26 ICU patients with treatment established study ARDS, 3–40 days after onset, without evidence of active infection

448

Methylprednisolone 625–1250 125–250 mg/day mg for 3–4 days followed by tapering doses over several weeks

81% survival

Meduri et Observational 25 Patients with late al., 1991 treatment (15 days after and 1994, study onset) Memphis fibroproliferative(28, 30) phase ARDS, mean age ~44 yr, 32% male

Methylprednisolone 175–263 mg 2–3 mg/kg/day in divided doses q6hr until extubation, with subsequent tapering over 6 weeks

72% survival to hospital discharge

Biffl et al., 1995, Denver (31)

Observational treatment study

6 SICU patients with late prolonged ARDS (mean 16 days duration) dying of isolated respiratory failure

Methylprednisolone 1– 350– 83% survival 2 mg/kg q6hr IV with 700 subsequent tapering mg over 3 weeks

Keel et al., 1998, Zurich (33)

Retrospective case-control (13 treated with steroids, 18 untreated)

31 Nontrauma patients with persistent ARDS, 5–44 days after onset (mean 15 days), age 16–82 (mean 50 yr), 48% male

“Equivalent of methylprednisolone” 100–250 mg/day× 3 days, then 80–180 mg/day with subsequent tapering

Meduri et al., 1998, Memphis (34)

Prospective, randomized, double-blind, placebocontrolled trial

24 Patients with severe ARDS failing to improve by day 7, mean age ~49, 38% male

Methylprednisolone 2 175 mg/kg/day, divided mg q6hr with tapering over 32 days

Improved hospital mortality (12% vs. 62%, p=0.03), lung function and multiple organ dysfunction score (MODS)

Varpula et al., 2000, Helsinki (38)

Retrospective case-control (16 treated with steroids, 15 untreated)

31 Patient with acute lung injury, mechanically ventilated for >10 days, mean age 43, 77% male

Methylprednisolone 120 mg/day (divided BID) with tapering over 4 weeks

No mortality benefit (19% vs. 20%), improved gas exchange

500– Trend for 1250 improved mg mortality (38% vs. 67%, p=0.117) and P: F ratio with steroid-treated vs. untreated patients

150 mg

Glucocorticoid therapy for the acute respiratory distress syndrome

449

a

Calculation based on 70 kg individual, conversions to prednisone-equivalent from Chin et al. (77) and Jantz and Sahn (78).

In 1991 Meduri and colleagues (28) reported an uncontrolled study of 8 patients, subsequently expanded to 25 patients (30), who were an average of 15 days after the onset of ARDS. Patients had progressive respiratory failure with worsening lung injury score and no evidence of active infection. Patients had an average APACHE II score of 20 and average lung injury scores of 2.5–3. All patients were treated with methylprednisolone 2–3 mg/kg/day until extubation and subsequently tapered over 6 weeks. They reported a 72% survival to hospital discharge. In 1995 Biffl and colleagues (31) reported 6 surgical patients with late phase ARDS (mean 16 days) who were all treated with methylprednisolone 1–2 mg/kg every 6 hours with subsequent tapering based on clinical response. They noted improvement in oxygenation within 7 days, and 5 of the 6 patients survived to discharge. There were notable complications including gastric necrosis with multiple organ failure, prolonged neuromuscular blockade, and infections. Keel and colleagues (33) reported a retrospective comparative analysis of 31 nontrauma ARDS patients who had been mechanically ventilated for >7 days. Mortality in the steroid-treated patients was 38% (5/13) compared to 67% (12/18) in the untreated group (p=0.117). They also reported an improvement in P: F ratio in the first 48 hours of steroid treatment and did not report any significant steroid-related complications. A second retrospective study was published in 2000 by Varpula et al. (38). They studied 31 patients with acute lung injury from pneumococcal pneumonia (58%), nonpneumococcal pneumonia (29%), or aspiration (13%) who required mechanical ventilation for >10 days. Sixteen patients received methylprednisolone 80 mg in the morning and 40 mg in the evening with gradual tapering over 4 weeks, and the other 15 patients served as untreated controls. They found significant improvements in P:F ratio (p=0.004) and C-reactive protein levels (p<0.001) within 3 days of steroid treatment, although there were no differences in mortality, ICU length of stay, or duration of mechanical ventilation. The only randomized-controlled clinical trial of corticosteroids in patients with persistent ARDS was published in 1998 by Meduri and colleagues (34). They studied 24 prospectively defined patients with ARDS (39) that was unresolving after 7 days of supportive care and randomized them to receive methylprednisolone 2 mg/kg/day in doses divided every 6 hours for 2 weeks, followed by a formal tapering protocol. Patients were randomized 2:1, resulting in 16 patients in the steroid treatment group and 8 in the placebo group. Two patients were withdrawn from the study: one in the placebo group for gastrointestinal bleeding, and one in the treatment group for Candida sepsis. Four patients in the placebo group were crossed over for lack of improvement after 10 days. Data were analyzed on an intention-to-treat basis. They reported that corticosteroids improved lung injury score (p<0.001), P:F ratio (p<0.001), multiorgan dysfunction score (p<0.001), successful extubation (p=0.05), ICU mortality (0% vs. 62%; p=0.002), and hospital mortality (12% vs. 62%; p=0.03). They concluded that prolonged administration of methylprednisolone in patients with unresolving ARDS was associated with improved lung injury and MODS scores and mortality.

Acute respiratory distress syndrome

450

This study had several positive features including the explicit scoring system, which showed that steroid-treated patients had clinical and physiological improvement, and the randomized controlled design. However, in an accompanying editorial by Brun-Buisson and Brochard (36) and in a subsequent letter to the editor by Wheeler and associates (37), concerns were raised about this study. They cited the small sample size, the planned crossovers, a higher incidence of infection, and possible differences in the clinical characteristics between the groups, with the steroid group having lower sepsis, lung injury, and organ dysfunction scores at the time of study entry. Three of the 4 placebo patients who crossed over and received steroids died. When the data were analyzed with these cross-over patients included in the steroid-receiving group, 15 of 20 (75%) steroidtreated patients survived to hospital discharge, as did 2 of 4 (50%) placebo patients (p=0.55). They urged caution in interpreting these data and advocated for a larger randomized trial. Of note, the National Institutes of Health (NIH)-funded ARDS Network of investigators is continuing to enroll patients in a similar study after the Data Safety and Monitoring Board (DSMB) performed an interim analysis. With 120 patients enrolled, they did not meet predetermined criteria for stopping the study for demonstrated efficacy, futility, or harm. Thus, the question of whether corticosteroids are useful in patients with persistent ARDS remains unresolved.

IV. Low-Dose Replacement Therapy for Relative Adrenal Insufficiency in Sepsis Over the past decade there has been evolving interest in the concept of “relative” adrenal insufficiency in patients with sepsis. Although diagnostic criteria for symptomatic outpatients are well established, normal physiological responses to the stress of critical illness remains controversial. Beneficial effects of “adrenal replacement doses” of corticosteroids in recent small clinical studies in patients with septic shock and recent editorials have kindled considerable interest (40–58). In outpatients with suspected adrenal insufficiency, the standard diagnostic approach is well established and includes drawing a morning cortisol level. If the concentration is very low, the diagnosis is established. If the concentration is mid-range, the standard approach is to administer 250 µg of a synthetic ACTH congener such as cosyntropin, which is the biologi-

Table 3 Low-Dose Replacement for Relative Adrenal Insufficiency in Sepsis Trial, location (Ref.) Cooperative Study Group, 1963, Baltimore,

Design

N

Population

Randomized, 194 Septic shock double(poorly blind, defined), age placebo>16 controlled,

Treatment

“Prednisoneequivalent” (day 1)a

Outcomes

Hydrocortisone 300 mg on day 1, then reduced by 50 mg/day×6 days

60 mg

No mortality benefit

Glucocorticoid therapy for the acute respiratory distress syndrome

Boston, Cincinnati, Chicago (63)

multicentered

451

(IV/oral)

McKee and Randomized, Finlay, open-label, 1983, single center Glasgow (55)

18 ICU patients (surgical and medical) with baseline cortisol <13 µg/dL (350 nM). All patients tested (12) also had corticotropininduced cortisol rise of <8 µg/dL (200 nM)

Hydrocortisone 100 mg, then BID doses to maintain serum cortisol >13 µg/dL (350 nM), “preferably” 18–33 µg/dL (500–900 nM) (IV). Treatment duration not specified (continued to “clinical improvement”)

?(>20 mg)

Improved ICU mortality

Bollaert et al., 1998, Nancy (56)

41 Septic shock (ACCP/SCCM) requiring vasopressors >48 hr, all mechanically ventilated, age >18

Hydrocortisone 100 mg q 8 hrs×5 days (IV)

60 mg

Improved time and rate of shock reversal and 28-day mortality; no increased complications

Randomized, doubleblind, placebocontrolled, single center

Briegel et al., 1999, Munich 48

Randomized, double-blind, placebocontrolled, single center

Annane et al., 2002, France (52)

Randomized, double-blind, placebocontrolled, multi-center

a

40 Septic shock (ACCP/SCCM) with high cardiac output, requiring vasopressors, all with acute lung injury, 30/40 with ARDS, age 18–75

300 Refractory septic shock requiring mechanical ventilation with oliguria, lactic acidosis, or acute lung injury

Hydrocortisone 80 Reduced time to loading dose 100 mg cessation of mg over 30 min, vasopressors and then 0.18 mg/kg/hr trend to earlier until shock reversal, resolution of then 0.08 organ failure; no mg/kg/hr×6 days, difference in then tapered at 24 overall shock mg/day (IV) reversal or mortality Hydrocortisone 50 40 Reduced 28-day mg IV every 6 hr mg mortality and time with fludrocortisone on vasopressors in 50 µg orally daily patients who were or 7 days unresponsive to corticotropin; no increased complications

Calculation based on 70 kg individual, conversions to prednisone-equivalent from Chin et al. (77) and Jantz and Sahn (78).

Acute respiratory distress syndrome

452

cally active 1–24 N-terminal peptide of ACTH. Blood is collected 30 and 60 minutes after infusion. Peak cortisol concentrations of <20 µg/dL or incremental increases from baseline that are <5–10 µg/dl or maximally stimulated cortisol concentrations <18 µg/dL are sufficient diagnostic criteria for adrenal insufficiency (59). By contrast, in critically ill patients, baseline cortisol concentrations in excess of 25– 50 µg/dL have been reported (60, 61). Cosyntropin responses in these patients may be blunted, possibly because adrenal responses are maximally stimulated (62). Annane and colleagues (51) found that more than 50% of patients with septic shock have occult adrenal insufficiency defined by a maximal cortisol response to 250 µg of corticotropin of 9 µg/dL or less. Such patients have impaired responsiveness to vasopressors (50) and substantially increased mortality (51). Five randomized controlled trials using low-dose corticosteroids have been published (Table 3). In general, patients were treated for several days, but unlike the previous high-dose steroid trials, patients were not treated early in the course of septic shock. In the classic randomized, double-blind, placebo-controlled corticosteroid trial in sepsis, the Cooperative Study Group, composed of investigators from Baltimore, Boston, Cincinnati, and Chicago, identified 194 patients over age 16 with “life-threatening infection” and randomized them to receive stress replacement doses of corticosteroids (63). Study patients received hydrocortisone 300 mg on day 1, followed by doses that were reduced by 50 mg/day over the next 6 days in a sequential combination of intravenous and oral administration, and control patients received placebo. Although they initially enrolled 135 additional patients under age 16, these patients had a low mortality rate and were excluded from the published analysis. Nearly half the adult patients had meningitis. The authors found no mortality benefit from corticosteroids. In 1983 McKee and Finlay (55) measured serum cortisol levels in 133 critically ill patients over a 1-year period. All 36 patients with morning serum cortisol levels of <13 µg/dL (350 nM) died. Subsequently, they screened an additional 68 patients and found 18 with baseline serum cortisols of <13 µg/dL. Of the 18 patients identified, corticotropin stimulation tests were performed on 12, and none had a serum cortisol rise of >7 µg/dL (200 nM). They randomly assigned half of these 18 patients to receive “replacement doses” of hydrocortisone. Doses were adjusted to maintain morning serum cortisol levels of >13 µg/dL (350 nM). Seven of the 8 treated patients survived, compared with only 1 of the 10 control patients (p< 0.002). They concluded that the incidence of biochemically defined adrenal insufficiency was higher than previously suspected and the mortality benefit from corticosteroid replacement therapy that they observed warranted further investigation. In 1998 Bollaert et al. (56) enrolled 41 patients with septic shock requiring vasopressors for >48 hours in a prospective, randomized, double-blind, placebocontrolled trial. They defined their population using ACCP/ SCCM consensus criteria, and all patients in this study were mechanically ventilated. They further defined shock reversal as a stable arterial pressure of >90 mmHg for >24 hours without catecholamines or fluid infusion. Study patients received hydrocortisone 100 mg every 8 hours for 5 days. They found that steroid-treated patients were more likely to achieve shock reversal at 7 days (68% vs. 21%) for a 47% absolute improvement in rate of shock reversal (p=0.007). The 28-day crude mortality rate was 32% in steroid-treated patients compared to 63% in the placebo group (p=0.09). They concluded that stress-replacement doses of

Glucocorticoid therapy for the acute respiratory distress syndrome

453

corticosteroids in patients with late septic shock who were dependent on vasopressors resulted in a significant improvement in hemodynamics and a beneficial effect on survival. In 1999 Briegel and colleagues (48) in Munich published a randomized, double-blind, placebo-controlled study in 40 patients with septic shock (ACCP/SCCM definition) with high cardiac output and requiring vasopressors. All patients were mechanically ventilated with sepsis-induced acute lung injury. The average P:F ratio was 150, with 30 of 40 patients meeting consensus criteria for ARDS. Patients were treated with a loading dose of 100 mg of hydrocortisone, followed by a continuous infusion of 0.18 mg/kg/ hr until shock was reversed, and then the infusion was reduced to 0.08 mg/kg/ hr for 6 days. The drug was then tapered in steps of 24 mg/day. They found that corticosteroid administration significantly reduced the time on vasopressors, and there was a trend to earlier resolution of organ dysfunction. They were unable to demonstrate a significant improvement in shock reversal or mortality. A recently published abstract further increased interest in stress replacement dose glucocorticoids for septic shock. Chawla and colleagues from Brooklyn, New York (57), presented results from 44 patients with septic shock who required vasopressors for >72 hours who were enrolled in a prospective, randomized, placebo-controlled, double-blind trial. Patients received 100 mg hydrocortisone every 8 hours for 3 days followed by tapering doses for 4 more days. They found that corticosteroids resulted in a reduction in the mean duration of vasopressors (122 vs. 74 h; p<0.005), and that after 3 days of treatment, 70% of steroid-treated patients were free of vasopressors compared with 33% of the placebo group (p=0.001). Mortality outcomes were not reported. Annane and colleagues from 19 intensive care units in France conducted a randomized controlled trial in 300 patients with severe septic shock (52). All patients had hypotension unresponsive to fluid resuscitation and low-dose vasopressor, required mechanical, ventilation, and had inadequate urine output, lactic acidosis, or acute lung injury at study entry. Patients were randomized within 3 hours of the onset of shock. After undergoing a short corticotropin test, patients were randomized to receive hydrocortisone 50 mg IV every 6 hours with fludrocortisone 50 µg orally daily or an IV and oral placebo for 7 days. They found 229 patients (76%) who met criteria for corticotropin unresponsiveness. The definition of relative corticotropin unresponsiveness was based on their prior experience (51) and was specifically defined as <9 µg/dL rise in serum cortisol at their 30 or 60 minutes after receiving 250 µg of corticotropin. In these patients with relative corticotropin unresponsiveness, the authors found that the combination of hydrocortisone and fludrocortisone significantly reduced mortality. They found a 53% mortality (defined at 28 days) in the hydrocortisone/fludrocortisone-treated group compared with a 63% mortality in the placebo group (hazard ratio 0.67; 95% confidence interval 0.47–0.95; p=0.02). They also found that the combination of hydrocortisone and fludrocortisone allowed more rapid withdrawal of vasopressor therapy (7 vs. 10 days). By contrast, in patients who responded to corticotropin (>9 µg/dL rise in serum cortisol), there was no difference between treatment groups. They found similar adverse event rates in the treatment and placebo groups. These data provide strong support for the concept of treatable relative insufficiency in a well-defined population of patients with severe septic shock. Importantly, the biochemical criteria defining this population were distinct from the traditional criteria used to define Addison’s disease, because critically ill patients with

Acute respiratory distress syndrome

454

the worst prognosis had the highest baseline serum cortisol values. Rather, patients who benefited from glucocorticoid therapy were initially characterized by their suboptimal response to corticotropin, suggesting a maximally stimulated adrenal response with limited adrenal reserve. The results of this study also provide a strong rationale for treating patients with glucocorticoids who have severe sepsis, are unresponsive to fluids and vasopressors, and are associated with other organ dysfunction. The authors stress the usefulness of the short corticotropin test for identifying patients that could benefit most from this treatment, suggesting that the treatment should be discontinued in patients who do not meet their criteria for relative adrenal insufficiency. The study was different from prior, smaller studies because of the addition of fludrocortisone, but the necessity for this mineralocorticoid in the treatment regimen will require investigation.

V. Other Trials Patients with Pneumocystis carinii pneumonia (PCP) and respiratory failure meet ACCP consensus conference criteria for ARDS, although this disease entity has not generally been recognized as part of the spectrum of ARDS and such patients have been excluded from the clinical trials detailed here. The reader is referred to the three initial randomized controlled trials (64–66) and the consensus statement from the National Institutes of Health-University of California Expert Panel (67) supporting the use of corticosteroids in patients with PCP. Similarly, the syndrome of acute eosinophilic pneumonia (68) has been reported as a form of acute lung injury that responds to corticosteroids (69–74). Finally, studies have suggested that acute lung injury associated with the fat embolism syndrome is preventable with corticosteroids (75, 76), but because these studies used a minimal clinical definition of the syndrome and did not use standard criteria for identifying acute lung injury, their clinical relevance remain in question. Further investigations of this clinical entity are needed before corticosteroids can be recommended as a therapeutic agent.

VI. Conclusions The role of a brief course of massive doses of corticosteroids given early in the course of sepsis or septic shock has been studied extensively. Although there may be subpopulations of patients, such as those with gram-negative sepsis, that may potentially benefit from further study as discussed in a recent meta-analysis (19), these patients have not been prospectively studied. It is clear from the combined experience of multiple clinical trials that when all comers with early sepsis and septic shock are considered, such massive doses of corticosteroids are ineffective in reducing mortality and may actually increase morbidity (18). By contrast, there is continued interest in investigating more prolonged courses of lower doses of corticosteroids in patients with fibroproliferative-phase (“persistent”) ARDS. The small randomized clinical trial from the University of Eastern Tennessee suggests that there may be benefit in terms of lung function, multiple organ dysfunction score (MODS), and mortality in these patients (34). The Late Steroid Rescue Study

Glucocorticoid therapy for the acute respiratory distress syndrome

455

(LaSRS), funded by the National Heart, Lung, and Blood Institute, is an ongoing prospective randomized controlled multicenter trial that will enroll 200 patients with persistent ARDS to study the efficacy of corticosteroids in this population of patients. Perhaps this trial will provide definitive evidence supporting the use of corticosteroids in patients with persistent ARDS, but until then such practice remains controversial. Finally, the balance of evidence in the literature strongly supports the concept that certain patients with refractory septic shock and organ failure can have relative adrenal insufficiency and can benefit from glucocorticoid therapy. Important in the selection of appropriate patients is evidence of corticotropin hyporesponsiveness suggesting limited adrenal reserve, regardless of the baseline serum cortisol value. The study by Annane and colleages (52) demonstrates that glucocorticoids should be considered in the management of this unique subpopulation of patients with septic shock.

Acknowledgments Supported in part by grants HL69955, HL30542, HR46055, and HL7287 from the National Institutes of Health, and by the Medical Research Service of the Department of Veteran’s Affairs.

References 1. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967; 2:319–323. 2. Luce JM. Corticosteroids in ARDS: an evidence-based review. Crit Care Clin 2002; 18:79–89. 3. Klastersky J, Cappel R, Debusscher L. Effectiveness of betamethasone in management of severe infections: a double-blind study. N Engl J Med 1971; 284:1248–1250. 4. Schumer W. Steroids in the treatment of clinical septic shock. Ann Surg 1976; 184:333–341. 5. Thompson WL, Gurley HT, Lutz BA, Jackson DL, Kvols LK, Morris I A. Inefficacy of glucocorticoids in shock (double-blind study). Clin Res 1976; 24:258A. 6. Sprung CL, Caralis PV, Marcial EH, et al. The effects of high-dose corticosteroids in patients with septic shock. A prospective, controlled study. N Engl J Med 1984; 311:1137–1143. 7. Lucas CE, Ledgerwood AM. The cardiopulmonary response to massive doses of steroids in patients with septic shock. Arch Surg 1984; 119:537–541. 8. Weigelt JA, Norcross JF, Borman KR, Snyder WH. Early steroid therapy for respiratory failure. Arch Surg 1985; 120:536–540. 9. Bone RC, Fisher CJ, Clemmer TP, et al. A controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. N Engl J Med 1987; 317:653–658. 10. Bone RC, Fisher CJ, Clemmer TP, Slotman GJ, Metz CA. Early methyl-prednisolone treatment for septic syndrome and the adult respiratory distress syndrome. Chest 1987; 92:1032–1036. 11. The Veterans Administration Systemic Sepsis Cooperative Study Group. Effect of high-dose glucocorticoid therapy on mortality in patients with clinical signs of systemic sepsis. N Engl J Med 1987; 317:659–665. 12. Bernard GR, Luce JM, Sprung CL, et al. High-dose corticosteroids in patients with the adult respiratory distress syndrome. N Engl J Med 1987; 317:1565–1570.

Acute respiratory distress syndrome

456

13. Luce JM, Montgomery AB, Marks JD, Turner J, Metz CA, Murray JF. Ineffectiveness of highdose methylprednisolone in preventing parenchymal lung injury and improving mortality in patients with septic shock. Am Rev Respir Dis 1988; 138:62–68. 14. Pepe PE, Potkin RT, Reus DH, Hudson LD, Carrico CJ. Clinical predictors of the adult respiratory distress syndrome. Am J Surg 1982; 144:124–130. 15. Hudson LD, Milberg JA, Anardi D, Maunder RJ. Clinical risks for development of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:293–301. 16. Shine KI, Kuhn M, Young LS, Tillisch JH. Aspects of the management of shock. Ann Intern Med 1980; 93:723–734. 17. Sheagren JN. Septic shock and corticosteroids. N Engl J Med 1981; 305:456–458. 18. Cronin L, Cook DJ, Carlet J, et al. Corticosteroid treatment for sepsis: a critical appraisal and meta-analysis of the literature. Crit Care Med 1995; 23: 1430–1439. 19. Lefering R, Neugebauer EAM. Steroid controversy in sepsis and septic shock: a meta-analysis. Crit Care Med 1995; 23:1294–1303. 20. Rubenfeld GD, Caldwell ES, Steinberg KP, Hudson LD. Persistent lung injury and ICU resource utilization [abstr]. Am J Respir Crit Care Med 1998; 157:A498. 21. Lamy M, Fallat RJ, Koeniger E, et al. Pathologic features and mechanisms of hypoxemia in adult respiratory distress syndrome. Am Rev Respir Dis 1976; 114:267–284. 22. Katzenstein AL, Bloor CM, Liebow AA. Diffuse alveolar damage: the role of oxygen, shock, and related factors. Am J Pathol 1976; 85:209–228. 23. Bachofen M, Weibel ER. Structural alterations of lung parenchyma in the adult respiratory distress syndrome. Clin Chest Med 1982; 3:35–56. 24. Tomashefski JF. Pulmonary pathology of the adult respiratory distress syndrome. Clin Chest Med 1990; 11:593–619. 25. Meduri GU, Eltorky M, Winer-Muram HT. The fibroproliferative phase of late adult respiratory distress syndrome. Semin Respir Infect 1995; 10:154–175. 26. Ashbaugh DG, Maier RV. Idiopathic pulmonary fibrosis in adult respiratory distress syndrome. Arch Surg 1985; 120:530–535. 27. Hooper RG, Kearl RA. Established ARDS treated with a sustained course of adrenocortical steroids. Chest 1990; 97:138–143. 28. Meduri GU, Belenchia JM, Estes RJ, Wunderink RG, Torky ME, Leeper KV. Fibroproliferative phase of ARDS: clinical findings and effects of corticosteroids. Chest 1991; 100:943–952. 29. Braude S, Haslam P, Hughes D, MacNaughton P, Evans TW. Chronic adult respiratory distress syndrome—a role for corticosteroids? Crit Care Med 1992; 20:1187–1189. 30. Meduri GU, Chinn AJ, Leeper KV, et al. Corticosteroid rescue treatment of progressive fibroproliferation in late ARDS: patterns of response and predictors of outcome. Chest 1994; 105:1516–1527. 31. Biffl WL, Moore FA, Moore EE, Haenel JB, McIntyre RC, Burch JM. Are corticosteroids salvage therapy for refractory acture respiratory distress syndrome? Am J Surg 1995; 170:591– 596. 32. Hooper RG, Kearl RA. Established adult respiratory distress syndrome successfully treated with corticosteroids. South Med J 1996; 1996:359–451. 33. Keel JB, Hauser M, Stocker R, Baumann PC, Speich R. Established acute respiratory distress syndrome: benefit of corticosteroid rescue therapy. Respiration 1998; 65:258–264. 34. Meduri GU, Headley AS, Golden E, et al. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome. JAMA 1998; 280:159–165. 35. Matthay MA. Conference summary: acute lung injury. Chest 1999; 116(suppl): 119S–126S. 36. Brun-Buisson C, Brochard L. Corticosteroid therapy in acute respiratory distress syndrome: Better late than never? JAMA 1998; 280:182–183. 37. Wheeler A, Bernard GR, Schoenfeld D, Steinberg K. Methylprednisolone for unresolving ARDS [letter]. JAMA 1998; 30:2074.

Glucocorticoid therapy for the acute respiratory distress syndrome

457

38. Varpula T, Pettila V, Rintala E, Takkunen O, Valtonen V. Late steroid therapy in primary acute lung injury. Intens Care Med 2000; 26:526–531. 39. Bernard GR, Artigas A, Brigham KL, et al. The American-European consensus conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818–824. 40. Ihle BU. Adrenocortical response and cortisone replacement in systemic inflammatory response syndrome. Anaesth Intens Care 2001; 29:155–162. 41. Chang SS, Liaw SJ, Bullard MJ, Dhiu TF, Chen JC, Liao HC. Adrenal insufficiency ijn critically ill emergency department patients: a Taiwan preliminary study. Acad Emerg Med 2001; 8:761–764. 42. Oppert M, Reinicke A, Graf KJ, Barckow D, Frei U, Eckardt KU. Plasma cortisol levels before and during “low-dose” hydrocortisone therapy and their relationship to hemodynamic improvement in patients with septic shock. Intens Care Med 2000; 26:1747–1755. 43. Schroeder S, Wichers M, Klingmuller D, et al. The hypothalamic-pituitary-adrenal axis of patients with severe sepsis: altered response to corticotropin-releasing hormone. Crit Care Med 2001; 29:310–316. 44. Woolf PD. Adrenal tea leaves: is the adrenal response to sepsis discernible? Crit Care Med 2001; 29:450–451. 45. Marik PE, Rotello L, Zaloga GP. Secondary adrenal insufficiency is common in critically ill patients [abstr]. Crit Care Med 2000; 29:A163. 46. Zaloga GP. Sepsis-induced adrenal dificiency syndrome. Crit Care Med 2001; 29:688–690. 47. Briegel J, Kellermann W, Forst H, et al. Low-dose hydrocortisone infusion attenuates the systemic inflammatory response syndrome. Clin Invest 1994; 72:782–787. 48. Briegel J, Forst H, Haller M, et al. Stress doses of hydrocortisone reverse hyperdynamic septic shcok: a prospective, randomized, double-blind, single-center study. Crit Care Med 1999; 27:723–732. 49. Briegel J. Hydrocortisone and the reduction of vasopressors in septic shock: therapy or only chart cosmetics? Intens Care Med 2000; 26:1723–1726. 50. Annane D, Bellissant E, Sebille V, et al. Impaired pressor sensitivity to noradrenaline in septic shock patients with and without impaired adrenal function reserve. Br J Clin Pharmacol 1998; 46:589–597. 51. Annane D, Sebille V, Troche G, Raphael J-C, Gajdos P, Bellissant E. A 3-level prognostic classification in septic shock based on cortisol levels and cortisol response to corticotropin. JAMA 2000; 283:1038–1045. 52. Annane D, Sebille V, Charpentier C, Bollaert P-E, Francois B, Korach J-M, Capellier G, Cohen Y, Azoulay E, Troche G, Chaumet-Riffaut P, Bellisant E. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002; 862–871. 53. Annane D. Replacement therapy with hydrocortisone in catecholamine-dependent septic shock. J Endotox Res 2001; 7:305–309. 54. Annane D. Corticosteroids for septic shock. Crit Care Med 2001; 29:S117– S120. 55. McKee JI, Finlay WEI. Cortisol replacement in severely stressed patients. Lancet 1983; 1:484. 56. Bollaert P-E, Charpentier C, Levy B, Debouverie M, Audibert G, Larcan A. Reversal of late septic shock with supraphysiologic doses of hydrocortisone. Crit Care Med 1998; 26:645–650. 57. Chawla K, Kupfer Y, Godman I, Tessler S. Hydrocortisone reverses refractory septic shock (abstr). Crit Care Med 1999; 27:A33. 58. Beale E, Chan LS, Zhu J, Belzberg H. Increased mortality in surgical intensive care unit patients with relative adrenal insuficiency: an introduction to a clinical entity—functional insufficiency of glucocorticoid synthesis (FIGS) [abstr]. Crit Care Med 2001; 29:A164. 59. Shenker Y, Skatrud JB. Adrenal insufficiency in critically ill patients. Am J Respir Crit Care Med 2001; 163:1520–1523.

Acute respiratory distress syndrome

458

60. Braams R, Koppeschaar HPF, van de Pavoordt HDWM, van Vroonhoven TJMV. Adrenocortical function in patients with ruptured aneurysm of the abdominal aorta. Intens Care Med 1998; 24:124–127. 61. Vermes I, Beishuizen A, Hampsink RM, Haanen C. Dissociation of plasma adrenocorticotropin and cortisol levels in critcally ill patients: possible role of endothelin and atrial natriuretic hormone. J Clin Endocrinol Metab 1995; 80:1238–1242. 62. Lamberts SWJ, Bruining HA, De Jong FH. Corticosteroid therapy in severe illness. N Engl J Med 1997; 337:1285–1292. 63. Cooperative Study Group. The effectiveness of hydrocortisone in the management of severe infections: a double-blind study. JAMA 1963; 183:462–465. 64. Montaner JSG, Lawson LM, Levitt N, Belzberg A, Schechter MT, Ruedy J. Corticosteroids prevent early deterioration in patients with moderately severe Pneumocystis carinii pneumonia and the acquired immunodeficiency syndrome (AIDS). Ann Intern Med 1990; 113:14–20. 65. Bozzette SA, Sattler FR, Chiu J, et al. A controlled trial of early adjunctive treatment with corticosteroids for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome: California Collaborative Treatment Group. N Engl J Med 1990; 323:1451–1457. 66. Gagnon S, Boota AM, Fischl MA, ABaier H, Kirksey OW, La Voie L. Corticosteroids as adjunctive therapy for severe Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome: a double-blind, placebo-controlled trial. N Engl J Med 1990; 323:1444–1450. 67. The National Institutes of Health-University of California Expert Panel for Corticosteroids as Adjunctive Therapy for Pneumocystis carinii. Consensus statement on the use of corticosteroids as adjunctive therapy for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. N Engl J Med 1990; 323:1500–1504. 68. Davis WB, Wilson HE, Wall RL. Eosinophilic alveolitis in acute respiratory failure. Chest 1986; 90:7. 69. Allen JN, Pacht ER, Gadek JE, Davis WB. Acute eosinophilic pneumonia as a reversible cause of noninfectious respiratory failure. N Engl J Med 1989; 321: 569–574. 70. Buchheit J, Eid N, Rodgers G, Feger T, Yakoub O. Acute eosinophilic pneumonia with respiratory failure: a new syndrome? Am Rev Respir Dis 1992; 145:716–718. 71. Ogawa H, Fujimura M, Matsuda T, Hakamura H, Kumabashiri I, Kitagawa S. Transient wheeze: eosinophilic bronchiolitis in acute eosinophilic pneumonia. Chest 1993; 104:493–496. 72. Hayakawa H, Sato A, Toyoshima M, Imokawa S, Taniguchi M. A clinical study of idiopathic eosinophilic pneumonia. Chest 1994; 105:1462–1466. 73. Pope-Harman AL, Davis WB, Allen ED, Christoforidis AJ, Allen JN. Acute eosinophilic pneumonia: a summary of 15 cases and review of the literature. Medicine 1996; 75:334–342. 74. Alp H, Daum RS, Abrahams C, Wylam ME. Acute eosinophilic pneumonia: a cause of reversible, severe, noninfectious respiratory failure. J Pediatr 1998; 132:540–543. 75. Ashbaugh DG, Petty TL. The use of corticosteroids in the treatment of respiratory failure associated with massive fat embolism. Surg Gynecol Obstet 1966; 123:493–500. 76. Schonfeld SA, Ploysongsang Y, DiLisio R, et al. Fat embolism prophylaxis with corticosteroids. A prospective study in high-risk patients. Ann Intern Med 1983; 99:438–443. 77. Chin R, Eagerton DC, Salem M. Corticosteroids. In: Chernow B, ed. The Pharmacologic Approach to the Critically 111 Patient. Baltimore: Williams & Wilkins, 1994:715–740. 78. Jantz MA, Sahn SA. Corticosteroids in acute respiratory failure. Am J Respir Crit Care Med 1999; 160:1079–1100.

19 Surfactant Therapy in the Acute Respiratory Distress Syndrome ROGER G.SPRAGG University of California, San Diego and San Diego VA Medical Center San Diego, California, U.S.A. JAMES F.LEWIS University of Western Ontario and St. Joseph’s Health Center London, Ontario, Canada

I. Introduction The acute respiratory distress syndrome (ARDS) may accompany both indirect and direct injury to the lung. An example of the latter is toxic gas inhalation. The poet Wilfred Owen described dramatically the sight of a World War I combatant felled by gas exposure (1):

If in some smothering dream you too could pace Behind the wagon that we flung him in… If you could hear, at every jolt, the blood Come gurgling from the froth corrupted lungs… The alveolar hemorrhage and edema that Owen describes fit well with the first medical description of ARDS provided in 1967 by Ashbaugh et al. (2). These investigators postulated that surfactant function might be diminished in patients with ARDS, with resultant alveolar collapse, edema, increased shunt, and hypoxemia. Indeed, using tools available at the time, they demonstrated a loss of surface-tension lowering capacity in fluid from the lungs of ARDS patients (3). Thus, the froth described by Owen would likely be un-stable and perhaps disappear, much as our ignorance of the pathophysiology of ARDS is, hopefully, fading. A description of the abnormalities of function and lipid composition of surfactant recovered in lavage fluid from ARDS patients was provided by Hallman and coworkers in 1982 (4), and subsequent investigations have provided more detailed descriptions of these abnormalities and of changes in the quantity of surfactant-associated proteins recoverable in lavage fluid (5–8). These reports are generally consistent and detail

Acute respiratory distress syndrome

460

alterations in surfactant composition and overall loss of surfactant mass and function. To determine the clinical relevance of these surfactant changes, it is critical to understand the normal functions of the surfactant system and how loss of these functions may affect patients with ARDS. This knowledge, together with results from preclinical investigations, provides a strong rationale for experimental treatment of ARDS patients using exogenous surfactant. The value of this treatment is yet to be convincingly demonstrated.

II. The Surfactant System of the Mature Lung The surfactant system of the mature lung is a complex mixture of lipids and proteins that has both biophysical and nonbiophysical functions (9). The former include prevention of alveolar collapse at low lung volume, maintenance of patency of small airways, and prevention of alveolar edema. Possible non-biophysical functions include protection from bacterial and viral infection and modulation of the activity of alveolar macrophages, polymorphonuclear leukocytes (pmn), and immunocompetent cells present in the airway. Surfactant is produced in the alveolar type II epithelial cell where the mature complex is found in lamellar bodies, secreted into the alveolar hypophase in the form of tubular myelin, and adsorbed in a thin film to the alveolar air/liquid interface. Compression and expansion of this film during ventilation is believed to result in removal of proteins and nonsaturated lipid components, resulting in a purified film able to reduce surface tension to close to 0 mN/m. Surfactant is cleared from the alveolar space by movement to the central airways, by macrophages, and by reuptake by alveolar type II cells for either degradation or resynthesis into functional components. The composition and function of lung surfactant are described operationally through investigation of material recovered from lung broncho-alveolar lavage fluid (BAL), minced lung, or type II pneumocyte fractions that contain lamellar bodies. Most commonly, crude surfactant preparations are made from cell-free BAL that is subjected to high-speed (approximately 40,000×g) centrifugation. The resultant pellet, which may be further purified by differential density gradient centrifugation, is defined as the largeaggregate lung surfactant fraction, which is composed of large lamella-like structures and tubular myelin. In a normal individual, this fraction represents 80–90% of extracellular surfactant and has excellent surface tension-lowering properties. It is from this pellet that surfactant lipids and proteins are frequently isolated and studies of biophysical function are performed. Small surfactant aggregates are found in the supernatant of the high-speed centrifugation and comprise the remaining 10–20% of extracellular surfactant. These small vesicular forms have limited biophysical function when tested either in vitro or in vivo. The ratio of large to small aggregates (frequently expressed in terms of phosphorus content) is decreased in certain pathological states, including pneumonia and ARDS (7, 8). Of note, surfactant may also be purified directly from BAL by equilibrium buoyant density gradient centrifugation, and the resulting material, while qualitatively similar to that obtained by differential centrifugation, has a somewhat different composition (10). The techniques for acquiring surfactant by lavage and for isolating it from BAL are not standardized, and therefore comparisons between reports from different laboratories must be made carefully.

Surfactant therapy in the acute respiratory distress syndrome

461

The extracellular metabolism of alveolar surfactant, specifically the conversion within the airspace of the surface-active large aggregates into small aggregates, has been investigated using both in vitro and in vivo approaches. In vitro studies involve the endover-end cycling of test tubes partially filled with a suspension of large surfactant aggregates. This cycling causes expansion and contraction of the surface film, which is believed to mimic changes at the alveolar air/liquid interface including formation of nonfunctional, small vesicles that are presumed to result from the preferential removal of surfactant components from the film. In vivo studies in which trace doses of radiolabeled large aggregates are delivered to animals’ lungs show a direct relationship between tidal volume size and the rate of conversion of instilled large aggregates into small aggregates. In normal animals, conversion has no impact on aggregate pool sizes, presumably due to compensatory responses (i.e., uptake of small aggregates), while in animals with preexisting acute lung injury, the increased conversion results in a decreased ratio of large to small aggregates, thereby decreasing the pool of functional surfactant within the airspace. These findings are consistent with clinical observations in patients with ARDS (7, 8). Aggregate conversion, at least in some species, appears to be mediated by serine protease activity. Recently, an enzyme identified as a carboxylesterase and termed “convertase” has been isolated from mouse BAL, purified, and shown to have a target other than dipalmitoylphosphatidylcholine (DPPC), perhaps SP-B (11–14). The biophysical function of lung surfactant may be assessed quantitatively in several ways. Measurements with the bubble surfactometer, which has commonly been used to study the function of surfactant recovered from BAL, require small amounts of material, may be performed rapidly, and are quite reproducible in the hands of experienced investigators (15). This device has been criticized, however, for being an open system from which minute amounts of surfactant may escape, thereby impairing accurate studies of very small amounts of surfactant. This objection is overcome by the captive bubble surfactometer (16). However, measurements with this device are technically demanding, and it has not been used for analysis of large numbers of clinical samples. Measurements made with a Langmuir-Wilhelmy balance have been used for decades but require greater amounts of surfactant and are technically demanding. Improvement in oxygenation and/ or compliance of the lungs of the premature rabbit pup or of the lavaged rat after intratracheal injection of surfactant have also been used to test the biophysical properties of surfactant preparations (17, 18). Despite differences in techniques for recovering and analyzing lung surfactant, analytical results are quite similar. Lung surfactant lipids are predominantly phospholipids (PL), although neutral lipids and cholesterol are also present (Table 1). The most abundant phospholipid is phosphatidylcholine (PC), which comprises approximately 70% of the phospholipids present. Unlike the PC of cell membranes, which is approximately 10% saturated, over 60% of the PC of lung surfactant is disaturated DPPC, and this component is believed to be of critical importance in providing surfactant biophysical function. Indeed, DPPC is an integral component of all exogenous surfactant preparations currently in clinical use. In addition, lung surfactant contains an abundance of phosphatidylglycerol (PG), which may also contribute to surface-tension lowering properties.

Acute respiratory distress syndrome

462

Surfactant contains at least four surfactant associate proteins, all of which have been extensively characterized at the molecular and gene level. SP-A and SP-D are relatively hydrophilic, whereas SP-B and SP-C are markedly hydrophobic. SP-A, the most abundant surfactant protein, is a 32 kDa glycoprotein member of the collectin family with a structure that includes a collagen-like domain and a calcium-dependent lectin domain, also known as a carbohydrate recognition domain (CRD). The CRD is able to bind to type II cells, lipids, and surfaces of microorganisms. Alveolar SP-A has a “bouquet” structure that is an octadecamer composed of 18 SP-A monomers. SP-A facilitates the formation of tubular myelin, enhances the adsorption of surfactant phospholipids at the air/liquid interface, modulates the secretion and uptake of surfactant by type II cells, and participates in innate defense against infection. This last property occurs both by direct interaction with microorganisms, including fungi, viruses, Pneumocystis carinii, and a variety of bacteria, and by modulating macrophage function (19). For example, SP-A is reported to inhibit production of cytokines by stimulated macrophages, and SP-Adeficient mice, relative to wild-type mice, have increased BAL levels of TNFα and MIP2 after exposure to lipopolysaccharide (20). Others have shown increased production of proinflammatory cytokines by stimulated macrophages and peripheral blood monocytes in the presence of SP-A (21). Despite a lack of tubular myelin, SP-A knockout mice have close to normal lung structure and function, but increased susceptibility to a variety of infections (22–24). SP-D (43 kDa) is also a collectin and in mature form in the alveolus is a dodecamer with regions that may bind to bacterial lipopolysaccharide, macrophages, and various lipids (25). SP-D shares many of the antimicrobial properties of SP-A (19), but clearly has additional effects, as mice deficient in SP-D develop emphysema, pulmonary inflammation, and fibrosis and have hypertrophic alveolar macrophages, increased production by macrophages of hydrogen peroxide and metalloproteinases, and increased amounts of saturated PC (26). SP-B, an 8 kDa molecule present as a dimer in the alveolar space, serves to increase markedly the adsorption of surfactant lipids to the air/liquid interface. SP-C, a 4 kDa intensely hydrophobic molecule, is dipalmitoylated and is intimately associated with the surfactant lipid film. SP-B knockout mice, which die of respiratory failure after birth, also have abnormal processing of SP-C precursor, and thus have deficient alveolar levels of both proteins (27). On the other hand, SP-C-deficient mice have normal lung morphology, surfactant synthesis, tubular myelin, and PC pool sizes and have only subtle biophysical abnormalities at low lung volume that suggest a role for SP-C in stabilization of the surfactant film (28). Mice partially deficient in SP-B (SP-B+/−) and fully or partially deficient in SP-C have been used to help elucidate the roles of these two proteins (29). SP-B+/− mice, when stressed with hyperoxia, develop a pulmonary injury characterized by increased BAL levels of total protein, IL-6, IL-1β, and MIP-2; fall in BAL-saturated PC; and decreased compliance. These responses are maximal in the absence of SP-C and markedly attenuated in the presence of that protein. Thus, SP-C may have a physiological role in maintenance of lung function during periods of stress when SP-B levels are diminished. Whether SP-C can fully substitute for SP-B is unknown.

Surfactant therapy in the acute respiratory distress syndrome

463

Table 1 Composition and Function of Surfactant from Patients with ARDS Gri Hall Pison ese, man et al., et al., 1999 1982 (4) 1989 (90) (89)

Patients

Gregory Güther Nakos Bersten Greene et et al., 1991 et al., 1996 et al., et al., 1998 al., 1999 (5) (7) 1997 (92) (6) (91) Sampling technique Meta- Lavage Lavage Lavage Lavage Lavage Lavage Air Lavage anal (1×20 (5×20 mL) (3×50 mL) (10×20 (6×20 (4×20 way (5×30 mL) ysis mL) mL) mL) mL) aspirate No Nor ARDS No ARDS Nor ARDS Nor ARDS Nor AR Nor ARDS Nor ARDS rmal mal rmal mal mal mal DS mal mal day 1

Number of 7 subjects Component PL 0.04 concentration mg/mL (or % normal) PL composition PC (% total) 68.7 73.0 PG (% total) 12.6 12.4 PI (% total) 4.1 2.7 PE(% total) 5.3 2.6 PS (% total) 2.3 3.3 Sph. (% 3.3 3.7 total) LysoPC (% 1 0.4 total) SP-A 4.5 (µg/mL) SP-B 4.9 (µg/mL) SP-C (µg/mL) SP-D 1.1 (µg/mL) Function 0–23 13.9 γmin (mN/m) a Measurement in resuspended pellet.

13

10

17 16–29 34–64

78%

59.5 0.3 3.1 4.3 13.0 17.5

62.8 10.0 8.3 4.8 4.5 7.4

52.8 1.7 13.7 15.8 ND 13.1

76.2 10.7 2.7 2.8 1.8 1.1

1.5 1.3

1.4

0.0

13

16

4.1

6

31%

96%

63.1 83.1 4.0 8.6 6.8 3.2 5.7 1.7 2.2 1.2 4.0 0.8

81.9 68.3 43.0 3.5 8.4 6.8 6.5 3.9 7.2 1.9 3.4 11.0 1.8 4.5 9.2 3.5 4.9 11.0

0.0

0.1

0.3

118.04 6.2 p 1.533 0.849 p 0.66 p 0.05 p 0.867 0.818

21

6

24.0 0–1 15–20

16

15 15– 35

41

45%

68.1 8.3 13.5 4.0 5.7 2.6

57.5 9.3 6.0 12.1 7.1 7.7

0.8

1.1 759 4.8

1.2

7174 0.11

0.04

1.03

1.08

Acute respiratory distress syndrome

464

III. Surfactant Alterations Associated with Acute Lung Injury Over the past two decades, a number of investigators have compared the amount, composition, and function of surfactant in BAL obtained from ARDS patients to values in BAL from patients with normal lungs (Table 1). Because of variability in lavage and analytical technique, comparison among studies can be risky. However, overall, the results are relatively uniform and indicate a loss of total phospholipid and alteration in the composition and function of the remaining phospholipids. Specifically, the fractional contents of PC and PG, the most surface-active lipid components, are reduced significantly, while the fractional contents of other phospholipids show a compensatory increase. Increases in the fractional content of sphingomyelin, however, are likely to be due, in part, to contributions from membranes of necrotic cells in the airway. Detailed examination of the fatty acid profiles of surfactant phospholipids shows that the fatty acids of PC isolated from surfactant of normal individuals are 88% saturated (and composed 80% of palmitate), whereas those from ARDS patients are only 74% saturated (and composed 66% of palmitate) (30). In addition to changes in surfactant lipids, observations on the levels in lavage fluid of surfactant proteins document marked changes (Table 1). There is a significant decrement in the level of SP-A, and most investigators also report decrements in the BAL level of SP-B (5–7, 31). BAL SP-A levels correlate inversely with disease progression, and in the study of Greene et al. (6), no patients with BAL SP-A concentrations greater than 1.2 µg/mL progressed to meet criteria for ARDS. SP-D BAL concentrations, while in the normal range in surviving patients, are very low early in the course of ARDS in those patients who eventually die (6). Reasons for the loss of surfactant components are not well studied, but are likely to be due to injury to type II cells, with decreased production, rather than to increased clearance. As shown in Figure 1, histochemical staining of lung in the early, exudative stage of ARDS discloses almost no SP-A, while lung in the later proliferative stage has evidence of marked production of SP-A by type II pneumocytes. Alterations in plasma levels of surfactant proteins are found in patients developing acute lung injury and in some cases may help predict disease progression. Increases in plasma levels of surfactant proteins are believed to be due to leakage of the protein across the damaged alveolar-capillary membrane into blood. In a study of 54 patients with hypoxemic respiratory failure, plasma samples were collected daily and analyzed by ELISA for SP-A and for SP-B antigen (32). Plasma SP-A circulates as a complex with IgG, while the precise nature of SP-B antigen detected in plasma is uncertain. Initial plasma SP-A and SP-B levels were higher in patients with direct (as opposed to indirect) lung injury, and plasma SP-B was significantly greater in patients who developed ARDS. Plasma values for both SP-B and SP-B/SP-A have been reported to be inversely related to blood oxygenation (33). In a study of 22 patients with respiratory failure and at risk for ARDS, and 41 patients with established ARDS, serum SP-A and SP-D levels increased significantly above normal with onset of the syn-

Surfactant therapy in the acute respiratory distress syndrome

465

Figure 1 Sample of lung tissue from a patient dying in the early, exudative phase of ARDS (left) shows evidence of a reduction in the number of epithelial cells, with a paucity of staining for SP-A (section is representative of tissue from five patients). In contrast, a sample of lung tissue from a patient dying in the later, proliferative phase of ARDS (right) shows evidence of proliferation of type II cells, with a marked increase in the staining for SP-A (section is representative of tissue from five patients). drome and 2 days later were maximal, with a median values approximately 20 and 3 times, respectively, higher than control. Despite the reported correlations with gas exchange, neither serum SP-A nor serum SP-D levels were predictive of survival (6). Few studies have addressed the possibility that genetic abnormalities of surfactant proteins may result in phenotypes that are at increased risk for the development of ARDS. However, a recent study suggests that there may be an important allelic polymorphism (C/T) at nucleotide 1580 within codon 131 (Thr131Ile) of the SP-B gene,

Acute respiratory distress syndrome

466

a polymorphism that has the possibility of determining the presence or absence of a potential N-linked glycosylation site. Lin et al. have reported that the C/C genotype is found with significantly increased frequency in a subset of patients with ARDS (34). Surfactant biophysical function is also markedly altered in patients with ARDS. Minimum surface tension measurements in the bubble sur-factometer show, for normal surfactant, a value that is close to 0 mN/m. Values for surfactant from ARDS patients are most commonly in the range of 20 mN/m—a loss of function that is felt to be consistent with impaired physiological function. This loss of activity is most likely the result of multiple factors, one of which is the altered fatty acid composition of PC. In analyses of surfactants from normal and ARDS patients, Günther et al. (7) found that the minimum surface tension of the surfactant from normal subjects was near zero, while that from ARDS patients was markedly increased and correlated inversely with the palmitate content of PC. In reconstitution experiments, addition of DPPC to surfactant resulted in significant reduction of minimum surface tension. These results underscore the physiological significance of alteration in the biochemical composition of surfactant lipids. Both in vitro and in vivo experiments demonstrate that inhibition of surfactant function by plasma proteins present in alveolar edema fluid of patients with ARDS is also likely to be of marked importance in impairing surfactant biophysical function. Albumin, hemoglobin, fibrinogen, and fibrin monomers are all capable of inhibiting surfactant biophysical function, perhaps through competition for sites at the air/liquid interface (35– 37). Surfactants lacking the hydrophobic proteins B and C are particularly sensitive to fibrinogen inhibition (38), and SP-A also contributes to resistance to protein inhibition (39). Although several components of serum leaking into the airspace can inhibit surfactant function, not all have the same potency or significance. For example the inhibition associated with fibrinogen products may be of particular importance in patients with ARDS, as active fibrin polymerization is likely to occur in the alveoli of these patients as hyaline membranes are formed. In addition, the surfactant-inhibitory capacity of polymerizing fibrin surpasses that of soluble fibrin monomers or fibrinogen by several orders of magnitude, making it the most powerful surfactant inhibitor described (40). Seeger et al. have proposed that surfactant PL and hydrophobic surfactant proteins are incorporated into the fibrin matrix and thus are effectively removed from the air/liquid interface and unable to form a surfactant film. They have also demonstrated that incorporated surfactant components are released by application of fibrinolytic agents, with restoration of surface activity (41, 42). This finding may have future therapeutic application in treatment of patients with ARDS. Fortunately, several in vitro and in vivo studies have shown that protein inhibition of surfactant function can be overcome in the presence of sufficient quantities of added surfactant, a concept that is of critical importance in developing strategies for surfactantreplacement therapy. These observations have provided a rationale for the utilization of relatively high doses of exogenous surfactant in clinical trials (43, 44). Another important mechanism that contributes to loss of surfactant function in ARDS patients is the increased conversion of surfactant from large to small aggregate forms, with a consequent decrease in the functional pool of large surfactant aggregates. Since most patients with ARDS are mechanically ventilated, and most if not all have significant

Surfactant therapy in the acute respiratory distress syndrome

467

quantities of proteases present within their airspaces, it is quite likely that increased conversion of large to small aggregate forms does occur in these patients. Indeed, animal studies specifically evaluating surfactant aggregate conversion in injured lungs show that modes of ventilation that use lower tidal volumes (5 mL/kg) and higher PEEP levels (9 cmH2O), result in less conversion of large aggregates, greater large aggregate pool sizes, and better lung function than found in similar animals ventilated with higher tidal volumes (10 mL/ kg) and lower PEEP levels (5 cmH2O) (45, 46). Recently, strategies involving ventilation of animals with high frequency oscillation (HFO) involving extremely low tidal volumes (e.g., <1 mL/kg) and adequate lung recruitment resulted in minimal large aggregate conversion and superior outcomes, even when compared to the low tidal volume/high PEEP strategies (47). Reactive nitrogen and oxygen radicals are present in the airways of ARDS patients and are able to react with surfactant lipids and proteins (48– 50). In vitro experiments demonstrate that oxidation of lipid components results in loss of biophysical function (50, 51) and that PMN are able to accomplish such oxidation (52). In addition, nitric oxide metabolites, particularly peroxynitrite, which is released from stimulated phagocytes, can modify tyrosine residues present in the CRD of SP-A and impair interaction of that molecule with surfactant lipids and with at least one microorganism—P. carinii (53, 54). Active proteases, including elastase and collagenase, are present as well in the airways of patients with ARDS (55, 56) and are capable of degrading surfactant apoproteins (52, 57). BAL fluid from ARDS patients shows evidence of proteolytic cleavage of SP-A that is consistent with cleavage by neutrophil elastase (58). Phospholipase A2 activity is present in the lungs of ARDS patients (4), and group I secretory phospholipase A2 (sPLA2) is able to hydrolyze PC, thereby producing free fatty acids and lysophosphatidylcholine and also impairing biophysical surfactant function (59).

IV. Preclinical Observations Animal models of acute lung injury have been investigated to determine whether administration of exogenous surfactant may be of benefit and what factors may influence efficacy of this therapy. The variety of injury models utilized for these studies is exemplified in Table 2. The most commonly used

Table 2 Examples of Mature Lung Injury Models in Which Surfactant Application is Effective Model

Species

Surfactant preparation

Delivery method

Improvements

Ref.

Lavage Saline

Guinea pig

Porcine

Bolus

ABG, morphology

93

Saline

Sheep

Survanta, BLES

Bolus, aerosol

ABG, mechanics

61

Saline

Pig

Venticute

Bolus

ABG, mechanics, morphology

73

Acute respiratory distress syndrome

468

Ventilatorinduced Rat

Porcine

Bolus

ABG, mechanics, morphology

94

HCl

Rat

Rabbit

Bolus±lavage

ABG

95

HCl

Rabbit

Synthetic lipids, natural proteins

Bolus

ABG, mechanics, morphology

96

Influenza A

Mice

Bovine

Bolus

mechanics

97

Sendai virus

Rats

Bovine

Bolus

ABG, mechanics, morphology

98

P. carinii

Rats

Bovine

Bolus

ABG, histology

99

Rabbit

Ovine

Bolus

ABG, mechanics, morphology

100

NNNMU

Rat

Survanta

Bolus

ABG

101

NNNMU

Rat

Survanta

Bolus

survival

102

Paraquat

Rat

Survanta

Bolus

ABG, mechanics, morphology

103

Hyperoxia

Baboon

Exosurf

Aerosol

ABG, mechanics, morphology

104

Hyperoxia

Rabbit

CLSE

Bolus

shunt, morphology

105

Xanthine oxidase

Guinea pig

Bovine

Bolus

mechanics

Aspiration

Pneumonia

Neurogenic Cervical vagotomy Toxic

78

model is one in which sequential whole lung lavages are performed. This approach not only removes airway surfactant but also results in an intense inflammatory response, with influx of PMN into the lung interstitium and airways. Almost all experiments using animal models of acute lung injury have been acute, usually not more than 8 hours in duration, and thus long-term outcomes such as mortality have not been investigated. Rather, measures of gas exchange and lung mechanics have served as surrogates to evaluate the presumed beneficial effects of surfactant administration. Preclinical experimentation has revealed some additional unexpected results. For example, it has been demonstrated that delivery of a relatively low dose of surfactant by aerosol can be as effective as delivery of higher doses by intratracheal instillation. An important factor to consider when interpreting these results, however, is that these studies utilized the saline lavage model described above, which is characterized primarily by

Surfactant therapy in the acute respiratory distress syndrome

469

surfactant deficiency and involves a relatively uniform pattern of injury factors that may not adequately represent the more complex situation within the lungs of ARDS patients (60). It is also apparent that not all surfactants have equivalent effects on gas exchange when delivered in a similar fashion. For example, when a bovine lipid extract surfactant (BLES) was compared with a minced lung extract (Survanta) in saline-lavaged sheep, instilled BLES improved gas exchange significantly more than instilled Survanta. However, when these surfactants were administered as aerosols, the opposite response was observed (61). Interactions between the exogenous surfactant and endogenous surfactant proteins may affect the physiologic response. Thus, these puzzling observations might be explained if these interactions vary with different delivery modalities. The effect of mechanical ventilation on endogenous surfactant aggregate conversion has been well demonstrated in studies of preclinical models, and the same effects have been shown for exogenously administered surfactants. Higher tidal volumes convert the predominantly large aggregate exogenous surfactant preparations into poorly functioning small aggregates to a greater extent than ventilation strategies using smaller tidal volumes. Thus, the duration of clinical improvement observed after a particular dose of exogenous surfactant is likely to be influenced by the ventilation strategy employed.

V. Clinical Trials of Surfactant Replacement Reasons for proceeding to clinical trials of surfactant replacement include the observations listed in Table 3. First, a variety of mechanisms may result in

Table 3 Rationale for Clinical Trials of Surfactant Treatment of Acute Lung Injury Patients with ARDS have loss of surfactant function. Loss of function might reasonably contribute to the observed pathophysiology of ARDS. Preclinical studies using animal models of acute lung injury demonstrate consistent benefit to gas exchange. Efficacy in neonatal RDS has been shown, even in the setting of acute lung inflammation.

loss of surfactant function, and it is reasonable to assume that this loss of function contributes to the alveolar collapse, edema, increased resistance to airflow in small airways, shunt, and hypoxemia seen in patients with ARDS. Second, loss of the antiinflammatory and antibacterial functions of lung surfactant might contribute to the inflammation and high risk of infection that are features of ARDS. Third, surfactant administration markedly improves gas exchange in multiple animal models of acute lung injury. Finally, infants with established respiratory distress syndrome (RDS) have markedly improved survival when treated with exogenous surfactant. The cause of surfactant deficiency is, in these infants, due to prematurity, while in ARDS patients it is due both to impaired production and to a variety of mechanisms that inhibit function.

Acute respiratory distress syndrome

470

Nevertheless, in both settings surfactant function is critically lacking and intense pulmonary inflammation is present. That exogenous surfactant is clearly efficacious in the setting of RDS provides substantial hope that it may also be effective in treating patients with ARDS. A. Critical Variables Affecting Studies of Surfactant Treatment of ARDS Patients A number of important variables must be dealt with when planning studies of surfactant treatment of patients with ARDS. These include the choice of surfactant preparation, mode of administration, amount of surfactant to deliver, volume in which to deliver it, frequency and duration of retreatments, and, finally, the ventilation strategy during and after treatment. Surfactant Preparations Surfactants that have seen clinical use may be divided into protein-containing and nonprotein-containing preparations and are summarized in Table 4. The former are either of natural origin (e.g., derived from porcine or bovine lungs or from human amniotic fluid) or are based either on a surfactant protein produced in E. coli by recombinant gene expression or on synthetic

Table 4 Surfactants That Have Been Used Clinically Generic name

Brand name

Constituents

Natural source

Manufacturer

Ref.

Proteincontaining Natural origin Amniotic fluid surfactant

Natural lipids, SP-A, Human SP-B, SP-C, SP-D amniotic fluid

Not available

Natural lipids, SP-B, Minced SP-C bovine lung

Abbott Laboratories (USA)

Bovine lipid BLES extract surfactant

Natural lipids, SP-B, Bovine SP-C lavage

BLES Biochem (Canada)

Calfactant

Natural lipids, SP-B, Bovine calf SP-C lavage

Forest Laboratories (USA)

Beractant

Survanta

Infasurf

HL-10 Poractant alfa

Natural lipids, SP-B, Minced Leo Pharmaceuticals SP-C porcine lung (Denmark) Curosurf

Natural lipids, SP-B, Minced Chiesi SP-C porcine lung Pharmaceuticals (Italy)

106

107 108 109 85 110

Surfactant therapy in the acute respiratory distress syndrome

471

SF-RI 1

Alveofact

Natural lipids, SP-B, Bovine SP-C lavage

Boehringer Ingelheim (Germany)

111

Surfactant-TA

Surfacten

Natural lipids, SP-B, Minced SP-C bovine lung

Mitsubishi Pharma Corp (Japan)

112

Sinapultide

Surfaxin

DPPC, POPG, PA, KL4 peptide

Discovery Laboratories (USA)

113

Lusupultide

Venticute

DPPC, POPG, PA, rhSP-C

BYK Pharm (Germany)

114

115

Synthetic origin

Non-protein-containing Pumactant

ALEC

DPPC, PG

Britannia Pharm. (UK)

Colfosceril

Exosurf

DPPC, hexadecanol,

Glaxo Wellcome (USA)

palmitate

Neonatal

tyloxepol

116

Abbreviations: DPPC, Dipalmitoylphosphatidylcholine; PA, palmitic acid; PG, phosphatidylglycerol; POPG, palmitoyloleoylphosphatidylglycerol; rhSP-C, recombinant human SP-C; SP-A, surfactant protein A; SP-B, surfactant protein B; SP-C, surfactant protein C; SP-D, surfactant protein C.

peptides. Surfactants of animal origin are usually subjected to organic solvent extraction and chromatographic purification, and thus contain neither SP-A nor SP-D. Non-proteincontaining preparations contain lipid (i.e., DPPC and/or PG) and additional constituents such as cetyl alcohol or tyloxapol. There are certain advantages and disadvantages inherent in each class of surfactant. Protein-containing preparations appear to have superior function when analyzed in vitro or in vivo. In addition, protein-containing preparations are more resistant to inhibition by plasma proteins. However, protein-containing preparations of animal origin are frequently inconsistent in composition and have the theoretical possibility of transmitting infectious agents and nonsurfactant lipid-soluble molecules of animal origin such as platelet-activating factor (62). Preparations containing synthetic peptides or recombinant proteins appear to have excellent biophysical function and have the advantage of consistent and defined composition. Their production is, however, technically demanding and expensive. Two non-protein-containing preparations that contain DPPC and PG have been tested in adults with ARDS. Such compounds have inferior biophysical properties and resistance to protein inhibition when compared to protein-containing preparations (63, 64). Mode of Administration Surfactant may be delivered to the airway in a variety of ways. These include intratracheal instillation by bolus or infusion, bronchoscopic delivery to each lobar or segmental bronchus, delivery during lung lavage, or aerosolization.

Acute respiratory distress syndrome

472

Intratracheal bolus instillation has been used in the majority of clinical trials and is usually accomplished through a catheter that is passed via the endotracheal tube to a point just caudal to the carina. Patients may be placed in various positions during instillation of serial surfactant aliquots to facilitate homogeneous distribution. Instillation of aliquots up to 50 mL has been done safely with the ventilator paused and positive end expiratory pressure maintained. Patients are usually transiently heavily sedated or paralyzed to avoid coughing that otherwise frequently occurs during instillation and results in foaming and loss of surfactant. Instillation during active ventilation may also result in foaming of the surfactant in the central airways, with resultant serious increases in airway pressure that result from obstruction by the surfactant foam. Slow continuous intratracheal instillation appears, in preclinical studies, to have inferior effects on gas exchange relative to bolus instillation (65). Advantages of bolus instillation include the rapidity of treatment and the ability to give large amounts over a short period of time. In one study, a dose of 100 mg/kg delivered in 25 mL aliquots could be given in as little as 10–15 minutes (66). Distribution may be affected adversely by gravitational forces, with preferential delivery of the instillate to dependent lung zones. However, both aerated and nonaerated lung zones are likely to receive surfactant if the rapid bolus technique is used and the patient is repositioned during administration. Treatment is limited to intubated patients, preventing study with this technique of patients supported by noninvasive ventilation. Bronchoscopic instillation of surfactant has also been used clinically and is both technically feasible and accompanied by apparent improvement in gas exchange (67). However, despite the theoretical possibility that distribution might be more homogeneous than with bolus instillation, preclinical studies have not shown advantages to this technique (68). In addition, it is somewhat time-consuming as administration of a single dose of 100 mg/kg may take 1–2 hours. Lavage fluid containing suspended surfactant may be tidally administered via a wedged bronchoscope, and this lavage technique has the theoretical advantage of removing cell products that may promote inflammation. Preclinical evaluation supports the concept that surfactant delivery by lavage may be an effective delivery technique (69). Aerosolization of surfactant has been the subject of several preclinical and clinical studies. This modality has the advantages of delivering surfactant continuously, not requiring the presence of a bronchoscopist, avoiding airway obstruction, and, theoretically, providing homogeneous delivery. However, preclinical studies suggest that nonventilated lung units receive little surfactant relative to more compliant ventilated lung units (70). Perhaps the major drawback to aerosol delivery is limitation of the amount of surfactant that can be delivered in a 24-hour period. Using available technology, approximately 112 mg DPPC/day of Exosurf could be aerosolized in one clinical trial. Of this, approximately 5% was assumed be retained in the lung—just a small percentage of the amount predicted by preclinical trials to be required for improved gas exchange. However, recent advances in aerosol technology may allow delivery of significantly greater amounts of surfactant, and approaches that couple instillation and aerosol delivery may be useful. The dose of surfactant necessary to treat patients with ARDS is undefined, as clinical studies have not yet shown convincing evidence of benefit. Furthermore, the optimal dose of surfactant for a particular patient will likely depend on factors such as the severity of the injury at the time of treatment, the specific type of surfactant used, and the method

Surfactant therapy in the acute respiratory distress syndrome

473

used for delivery. Nevertheless, preclinical studies suggest that doses (given by bolus instillation) in the range of 50–100 mg of phospholipids/kg may be required. This amount of surfactant is approximately 25 times that existing in the healthy adult human airway, but such large doses may be required to overcome the inhibition of surfactant biophysical function by components of alveolar edema fluid and to assure treatment of most lung areas despite lack of fully homogeneous delivery (71). Of note, there is some evidence that excessive dosing with surfactant may be harmful. In a study in which a natural surfactant was given by instillation, patients receiving four doses of 100 mg/kg every 6–8 hours appeared acutely to have better gas exchange than patients receiving eight doses (66). Investigators in this study had the subjective impression that patients’ clinical response peaked after approximately four doses, and then waned. The dose and dose volume of administered surfactant should be considered to be independent variables, with concentration a dependent variable. Dose volume is likely to be of considerable importance when delivery is as a bolus into the airway. Investigators have shown that relatively large dose volumes may result in more homogeneous distribution (72). However, as shown in Figure 2, delivery to pigs of 50 or 100 mg rSP-C surfactant/ kg in 1 or 2 mL/kg resulted in a greater improvement in gas exchange than was achieved with the same dose delivered in 4 or 6 mL/kg, despite similar

Figure 2 Pigs with acute lung injury induced by repetitive saline lavage received either 50 or 100 mg rSP-C surfactant phospholipids (PL) in volumes that varied from 1 to 6 mL/kg. The area under the vs. time curve

Acute respiratory distress syndrome

474

for the 4 hours after treatment ( AUC) is displayed as a function of the concentration of the surfactant delivered for each dose. Control animals received saline only (0 mg PL/mL). Values significantly different from control are indicated with an asterisk and indicate that animals receiving either 1 or 2 mL/kg, containing either 50 or 100 mg PL/mL had a significant improvement in gas exchange. (From Ref. 117.) lobar distribution of the instilled surfactant (73). This result may reflect over-dilution of the exogenous surfactant. Certainly there is a limit to the instilled volume that can be administered safely to critically ill patients. The timing of surfactant administration may also impact efficacy. Prophylactic treatment of premature neonates is effective in preventing lung injury (74), and prophylactic treatment of donor canine lungs prevents post-implantation injury (75). While study of the prophylactic treatment of patients at risk of developing ARDS would be attractive, the current power of predicting development of the syndrome is insufficient to support such clinical investigation. Treatment of lung injury early in the course of disease is supported by preclinical studies and by observations that patients who have significant respiratory failure but who have not yet met criteria for ARDS also have impaired surfactant function (5). There is little information that addresses the question of how frequently retreatment should occur. In some recent clinical trials, patients were retreated every 6–8 hours if gas exchange had not improved to predefined levels (66, 76). Bronchoscopic delivery was repeated after 24 hours under similar circumstances (67). In a recent trial of rSP-C surfactant, it was noted that evidence of rSP-C in the lower airway could be detected 24 but not 96 hours after administration of the last dose, suggesting the possibility that treatments might rationally be spaced less frequently than every 6 hours and extend over more than 24 hours. Further clinical investigation of treatment schedules is required. In addition, there are no data that address the question of how late in the course of ARDS surfactant treatment might be of benefit. Finally, choice of the correct ventilation strategy may be critical in achieving successful treatment of ARDS patients with lung surfactant. Fortunately, the rational choice for ventilating these patients is a low-volume strategy (77), and, as noted previously, it is likely that this strategy is also likely to minimize conversion of large functional surfactant aggregates to small poorly functional forms. Thus, one would theorize that use of low ventilating volumes should help preserve exogenous surfactant in the large aggregate form.

Surfactant therapy in the acute respiratory distress syndrome

475

B. Results of Individual Investigations of Surfactant Treatment of ARDS Patients The first suggestion that exogenous surfactant might be of value in treating ARDS patients was offered by Lachmann, who treated a morbidly ill patient with a natural surfactant (78). In the first small trial in nonterminal patients, in which those patients served as their own controls, Curosurf®, a natural surfactant of porcine origin, was delivered bronchoscopically to each lobar bronchus (79). Patients had a small but significant increase in gas exchange, and in several patients the minimum surface tension of surfactant recovered in BAL several hours after treatment was significantly lower than the pretreatment values. Subsequently, additional small studies were performed using Exosurf (80). These provided rationale for a large prospective randomized multi-center phase III trial in which 364 patients received aerosolized Exosurf and 361 received aerosolized 0.45% saline (81). The aerosolized drug contained 13.5 mg DPPC/mL, and approximately 112 mg/kg/day was aerosolized for up to 5 days. Results of this study were conclusively negative. Thirty-day mortality in both the treated and placebo groups was 41%, and no differences were seen when groups were stratified by APACHE III score or etiology. The placebo and treated groups had similar numbers of days of mechanical ventilation and length of ICU stay, and no differences in gas exchange were detected. There are several reasons that the hypothesis that exogenous surfactant might benefit patients with ARDS was not adequately tested in this study. First, the non-protein-containing Exosurf has inferior biophysical activity and is quite susceptible to protein inhibition. Second, it is highly likely that inadequate amounts were able to be delivered by aerosolization. Finally, distribution of the aerosolized material is likely to have been predominately to well-aerated lung units, with resultant failure to recruit poorly ventilated or nonventilated units. In addition to this trial, a small study of the efficacy of a second non-proteincontaining surfactant has had results consistent with the findings of the Exosurf study. Four patients with late-stage ARDS received a single dose of ALEC, a preparation containing only DPPC and POPG, without evidence of therapeutic effect (82). A more promising phase II trial of a natural surfactant, Survanta®, was reported in 1997. In this prospective, randomized, multicenter study, patients received either standard treatment or standard treatment plus instillation of surfactant. Three surfactant-treated groups were studied. Patients received up to eight doses of 50 mg PL/kg, up to 8 doses of 100 mg PL/kg, or up to 4 doses of 100 mg PL/kg. The inspired oxygen concentration was significantly decreased at 120 hours only in the group receiving up to four doses of 100 mg PL/kg. Mortality in that same group was 18.8% as compared to 43.8% mortality in the group receiving standard treatment. Patients were lavaged prior to treatment and 120 hours after initiation of treatment to assess effects of treatment. Significant increases in lavage PL and disaturated PC concentrations were seen in the groups receiving the higher concentrations of surfactant, and biophysical function of surfactant recovered from lavage fluid at 120 hours was significantly (although minimally) decreased in the group receiving 8 doses of 100 mg PL/kg. No significant adverse events were reported. This study was followed by two prospective, randomized, open label, multicenter phase II investigations of the safety and efficacy of Venticute®, a preparation containing

Acute respiratory distress syndrome

476

1 mg rSP-C/mL and 50 mg PL/mL. Based on information from pre-clinical studies, Venticute was delivered intratracheally in a dose volume of 1 mL/kg. In the study performed in North America, patients were prospectively randomized to receive rSP-C surfactant plus standard therapy or standard therapy alone (n=13). Patients received either 100 (n=12) or 200 (n=15) mg PL/kg ideal body weight given in four divided doses over 24 hours. No safety concerns were identified, and the higher dose of rSP-C surfactant was associated with the most ventilator-free days to day 28 (VFD), greatest improvement in gas exchange, and least mortality (76). In the second study, conducted in Europe and South Africa, patients were randomized to standard therapy (n=12) or to receive 200 (n= 14) or 500 mg PL/kg. As in the former study, the group receiving 200 mg PL/ kg had the greatest number of VFD, greatest improvement in gas exchange, and least mortality (83). These promising results were followed by two prospective, randomized, double-blind, multicenter trials designed to examine the efficacy of the 200 mg PL/kg dose of rSP-C surfactant in the treatment of ARDS patients. Parallel trials in North American and in Europe/South Africa included 221 and 227 patients, respectively. The most prevalent predisposing events for ARDS in both studies were sepsis, pneumonia, and trauma or in treated vs. control patients were 34 vs. 15 surgery. The median increases in mmHg (p<0.01) and 23 vs. 11 mmHg (p=0.04) in the North American and European/South African studies, respectively. No differences in survival were detected between groups treated with or without rSP-C surfactant (76). However, post hoc analysis of the subgroup (n=225) with ARDS secondary to primary pulmonary events (pneumonia and/or aspiration) disclosed a trend toward improved survival (84). Most recently, a preliminary report of a prospective randomized multicenter openlabel trial of HL 10 in the treatment of ARDS patients suggests that treatment with up to four doses of 100–200 mg PL/kg was associated with significant improvement in gas exchange and survival (85). In addition to these seven randomized controlled trials, several uncontrolled trials also suggest the possible efficacy of exogenous protein-containing surfactants in the treatment of patients with ARDS. Infasurf, a calf lung surfactant, has been administered to pediatric patients, 11 of whom were between 5 and 16 years of age. The investigators concluded that surfactant administration rapidly improved oxygenation and allowed moderation of ventilator support in the majority of children studied (86). In a second study involving pediatric patients, BLES, a bovine lung surfactant similar to Infasurf, was administered to 13 children with severe ARDS with ages ranging from 2 months to 16 years. Clinical improvements were observed in 8 patients. Retrospective analyses revealed that responders received the surfactant within 2 days of ARDS diagnosis, whereas nonresponders were treated at a mean of 5 days after diagnosis (87). The safety and possible efficacy of surfactant delivered by bolus instillation into each segmental bronchus had been reported by Walmrath et al. (67). In a single-site uncontrolled open-label study, these investigators treated patients with ARDS secondary to sepsis by delivering Alveofact (300 mg PL/kg) in divided doses through a bronchoscope. They observed initial improvement in oxygenation; treatment was repeated after 18–24 hours if this initial improvement was not maintained. Bronchoscopic lavage was used to deliver Surfaxin (KL4-surfactant) in an uncontrolled open label trial in which 12 adults with ARDS were studied; three separate delivery strategies were

Surfactant therapy in the acute respiratory distress syndrome

477

investigated. Overall, values increased marginally, while PEEP and requirements lessened. No serious adverse events were associated with the procedure. However, the lavage procedure did result in temporary deterioration of gas exchange (88). Despite the strong rationale for surfactant therapy of ARDS (Table 3) and the promise of phase II trials, that promise has not been realized in the three larger phase III trials that have used adequate amounts of highly active surfactants. The reasons for this are unclear. One possibility is that, despite similar inclusion and exclusion criteria, the patients enrolled in these phase II and phase III studies are qualitatively different with respect to their disease and that the differences result in substantially different response to surfactant administration. A second possibility is that only an ARDS patient subset, which we are yet unable to define clearly, will benefit from surfactant treatment. Most studies performed to date have included a broad spectrum of patients with ARDS, and the signal from a responding subgroup may not yet be apparent.

VI. Summary Since the first medical description of the acute respiratory distress syndrome, investigators have postulated that loss of lung surfactant function may be of pathophysiological importance. This belief has deepened as multiple studies have confirmed that surfactant function is impaired in patients with ARDS. The hope that restoration of surfactant function may benefit patients with ARDS has been fostered by studies of individual patients and by a limited number of phase II studies. However, phase III studies have failed to demonstrate beneficial effect. Hopefully, further study will identify treatment strategies and patient subgroups that this promising treatment may benefit.

References 1. Owen W. Dulce et Decorum Est. In: Anonymous Poems by Wilfred Owen with an Introduction by Siegfried Sassoon. London: Chatto and Windus, 1921. 2. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967; 2:319–323. 3. Petty TL, Reiss OK, Paul GW, Silvers GW, Elkins ND. Characteristics of pulmonary surfactant in adult respiratory distress syndrome associated with trauma and shock. Am Rev Respir Dis 1977; 115:531–536. 4. Hallman M, Spragg RG, Harrell JH, Moser KM, Gluck L. Evidence of lung surfactant abnormality in respiratory failure: study of bronchoalveolar lavage phospholipids, surface activity, phospholipase activity, and plasma myoinositol. J Clin Invest 1982; 70:673–683. 5. Gregory TJ, Longmore WJ, Moxley MA, Whitsett JA, Reed CR, Fowler AAI, Hudson LD, Maunder RJ, Crim C, Hyers TM. Surfactant chemical composition and biophysical activity in acute respiratory distress syndrome. J Clin Invest 1991; 88:1976–1981. 6. Greene KE, Wright JR, Steinberg KP, Ruzinski JT, Caldwell E, Wong WB, Hull W, Whitsett JA, Akino T, Kuroki Y, Nagae H, Hudson LD, Martin TR. Serial changes in surfactant-

Acute respiratory distress syndrome

478

associated proteins in lung and serum before and after onset of ARDS. Am J Respir Crit Care Med 1999; 160:1843–1850. 7. Günther A, Siebert C, Schmidt R, Ziegler S, Grimminger F, Yabut M, Temmesfeld B, Walmrath D, Morr H, Seeger W. Surfactant alterations in severe pneumonia, acute respiratory distress syndrome, and cardiogenic lung edema. Am J Respir Crit Care Med 1996; 153:176–184. 8. Veldhuizen RA, McCaig LA, Akino T, Lewis JF. Pulmonary surfactant subfractions in patients with the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 152:1867–1871. 9. Robertson B, Van Golde LMG, Batenburg JJ, eds. Pulmonary Surfactant: From Molecular Biology to Clinical Practice. New York: Elsevier, 1992. 10. Gross NJ, Kellam M, Young J, Krishnasamy S, Dhand R. Separation of alveolar surfactant into subtypes. A comparison of methods. Am J Respir Crit Care Med 2000; 162:617–622. 11. Krishnasamy S, Gross NJ, Teng AL, Schultz RM, Dhand R. Lung “surfactant convertase” is a member of the carboxylesterase family. Biochem Biophys Res Commun 1997; 235:180–184. 12. Krishnasamy S, Teng AL, Dhand R, Schultz RM, Gross NJ. Molecular cloning, characterization, and differential expression pattern of mouse lung surfactant convertase. Am J Physiol 1998; 275:L969–L975. 13. Dhand R, Young J, Teng A, Krishnasamy S, Gross NJ. Is dipalmitoylphosphatidylcholine a substrate for convertase? Am J Physiol Lung Cell Mol Physiol 2000; 278:L19–L24. 14. Veldhuizen RA, Inchley K, Hearn SA, Lewis JF, Possmayer F. Degradation of surfactantassociated protein B (SP-B) during in vitro conversion of large to small surfactant aggregates. Biochem J 1993; 295:141–147. 15. Enhorning G. A pulsating bubble technique for evaluating pulmonary surfactant. J Appl Physiol 1977; 43:198–201. 16. Schurch S, Bachofen H, Possmayer F. Surface activity in situ, in vivo, and in the captive bubble surfactometer. Comp Biochem Physiol A Mol Integr Physiol 2001; 129:195–207. 17. Gilliard N, Heldt GP, Loredo J, Gasser H, Redl H, Merritt TA, Spragg RG. Exposure of the hydrophobic components of porcine lung surfactant to oxidant stress alters surface tension properties. J Clin Invest 1994; 93:2608–2615. 18. Gupta M, Hernandez-Juviel JM, Waring AJ, Bruni R, Walther FJ. Comparison of functional efficacy of surfactant protein B analogues in lavaged rats. Eur Respir J 2000; 16:1129–1133. 19. Crouch E, Wright JR. Surfactant proteins A and D and pulmonary host defense. Annu Rev Physiol 2001; 63:521–554. 20. Borron P, McIntosh JC, Korfhagen TR, Whitsett JA, Taylor J, Wright JR. Surfactant-associated protein A inhibits LPS-induced cytokine and nitric oxide production in vivo. Am J Physiol Lung Cell Mol Physiol 2000; 278: L840– L847. 21. Kremlev SG, Phelps DS. Surfactant protein A stimulation of inflammatory cytokine and immunoglobulin production. Am J Physiol 1994; 267:L712– L719. 22. Korfhagen TR, LeVine AM, Whitsett JA. Surfactant protein A (SP-A) gene targeted mice. Biochim Biophys Acta 1998; 1408:296–302. 23. LeVine AM, Hartshorn K, Elliott J, Whitsett J, Korfhagen T. Absence of SP-A modulates innate and adaptive defense responses to pulmonary influenza infection. Am J Physiol Lung Cell Mol Physiol 2002; 282:L563–L572. 24. Linke MJ, Harris CE, Korfhagen TR, McCormack FX, Ashbaugh AD, Steele P, Whitsett JA, Walzer PD. Immunosuppressed surfactant protein A-deficient mice have increased susceptibility to Pneumocystis carinii infection. J Infect Dis 2001; 183:943–952. 25. Hawgood S, Poulain FR. The pulmonary collectins and surfactant metabolism. Annu Rev Physiol 2001; 63:495–519. 26. Wert SE, Yoshida M, LeVine AM, Ikegami M, Jones T, Ross GF, Fisher JH, Korfhagen TR, Whitsett JA. Increased metalloproteinase activity, oxidant production, and emphysema in surfactant protein D gene-inactivated mice. Proc Natl Acad Sci USA 2000; 97:5972–5977.

Surfactant therapy in the acute respiratory distress syndrome

479

27. Clark JC, Wert SE, Bachurski CJ, Stahlman MT, Stripp BR, Weaver TE, Whitsett JA. Targeted disruption of the surfactant protein B gene disrupts surfactant homeostasis, causing respiratory failure in newborn mice. Proc Natl Acad Sci USA 1995; 92:7794–7798. 28. Glasser SW, Burhans MS, Korfhagen TR, Na CL, Sly PD, Ross GF, Ikegami M, Whitsett JA. Altered stability of pulmonary surfactant in SP-C-deficient mice. Proc Natl Acad Sci USA 2001; 98:6366–6371. 29. Ikegami M, Weaver TE, Conkright JJ, Sly PD, Ross GF, Whitsett JA, Glasser SW. Deficiency of SP-B reveals protective role of SP-C during oxygen lung injury. J Appl Physiol 2002; 92:519–526. 30. Schmidt R, Meier U, Yabut-Perez M, Walmrath D, Grimminger F, Seeger W, Gunther A. Alteration of fatty acid profiles in different pulmonary surfactant phospholipids in acute respiratory distress syndrome and severe pneumonia. Am J Respir Crit Care Med 2001; 163:95– 100. 31. Raymondos K, Leuwer M, Haslam PL, Vangerow B, Ensink M, Tschorn H, Schurmann W, Husstedt H, Rueckoldt H, Piepenbrock S. Compositional, structural, and functional alterations in pulmonary surfactant in surgical patients after the early onset of systemic inflammatory response syndrome or sepsis. Crit Care Med 1999; 27:82–89. 32. Bersten AD, Hunt T, Nicholas TE, Doyle IR. Elevated plasma surfactant protein-B predicts development of acute respiratory distress syndrome in patients with acute respiratory failure. Am J Respir Crit Care Med 1998; 164: 648–652. 33. Doyle IR, Bersten AD, Nicholas TE. Surfactant proteins-A and -B are elevated in plasma of patients with acute respiratory failure. Am J Respir Crit Care Med 1997; 156:1217–1229. 34. Lin Z, Pearson C, Chinchilli V, Pietschmann SM, Luo J, Pison U, Floros J. Polymorphisms of human SP-A, SP-B, and SP-D genes: association of SP-B Thr131Ile with ARDS. Clin Genet 2000; 58:181–191. 35. Seeger W, Stöhr G, Wolf HRD, Neuhof H. Alteration of surfactant function due to protein leakage: special interaction with fibrin monomer. J Appl Physiol 1985; 56:326–338. 36. Holm BA, Notter RH. Effects of hemoglobin and cell membrane lipids on pulmonary surfactant activity. J Appl Physiol 1987; 63:1434–1442. 37. Seeger W, Günther A, Thede C. Differential sensitivity to fibrinogen inhibition of SP-C- vs. SP-B-based surfactants. Am J Physiol 1992; 261:L286–L291. 38. Seeger W, Gunther A, Thede C. Differential sensitivity to fibrinogen inhibition of SP-C- vs. SP-B-based surfactants. Am J Physiol 1992; 262:L286–L291. 39. Cockshutt AM, Weitz J, Possmayer F. Pulmonary surfactant-associated protein A enhances the surface activity of lipid extract surfactant and reverses inhibition by blood proteins in vitro. Biochemistry 1990; 29:8424–8429. 40. Seeger W, Elssner A, Günther A, Kramer HJ, Kalinowski HO. Lung surfactant phospholipids associate with polymerizing fibrin: loss of surface activity [published erratum appears in Am J Respir Cell Mol Biol 1993 Oct; 9 (4):following 462]. Am J Respir Cell Mol Biol 1993; 9:213– 220. 41. Günther A, Markart P, Kalinowski M, Ruppert C, Grimminger F, Seeger W. Cleavage of surfactant-incorporating fibrin by different fibrinolytic agents. Kinetics of lysis and rescue of surface activity. Am J Respir Cell Mol Biol 1999; 21:738–745. 42. Schermuly RT, Gunther A, Ermert M, Ermert L, Ghofrani HA, Weissmann N, Grimminger F, Seeger W, Walmrath D. Conebulization of surfactant and urokinase restores gas exchange in perfused lungs with alveolar fibrin formation. Am J Physiol Lung Cell Mol Physiol 2001; 280:L792–L800. 43. Holm BA, Wang Z, Notter RH. Multiple mechanisms of lung surfactant inhibition. Pediatr Res 1999; 46:85–93. 44. Lachmann B, Eijking EP, So KL, Gommers D. In vivo evaluation of the inhibitory capacity of human plasma on exogenous surfactant function. Intensive Care Med 1994; 20:6–11.

Acute respiratory distress syndrome

480

45. Veldhuizen RA, Marcou J, Yao LJ, McCaig L, Ito Y, Lewis JF. Alveolar surfactant aggregate conversion in ventilated normal and injured rabbits. Am J Physiol 1996; 270:L152–L158. 46. Ito Y, Veldhuizen RA, Yao LJ, McCaig LA, Bartlett AJ, Lewis JF. Ventilation strategies affect surfactant aggregate conversion in acute lung injury. Am J Respir Crit Care Med 1997; 155:493–499. 47. Kerr CL, Veldhuizen RA, Lewis JF. Effects of high-frequency oscillation on endogenous surfactant in an acute lung injury model. Am J Respir Crit Care Med 2001; 164:237–242. 48. Haddad IY, Ischiropoulos H, Holm BA, Beckman JS, Baker JR, Matalon S. Mechanisms of peroxynitrite-induced injury to pulmonary surfactants. Am J Physiol 1993; 265:L555–L564. 49. Haddad IY, Crow JP, Hu P, Ye Y, Beckman J, Matalon S. Concurrent generation of nitric oxide and superoxide damages surfactant protein A. Am J Physiol 1994; 267:L242–L249. 50. Gilliard N, Heldt GP, Loredo J, Gasser H, Redl H, Merritt TA, Spragg RG. Exposure of the hydrophobic components of porcine lung surfactant to oxidant stress alters surface tension properties. J Clin Invest 1994; 93:2608–2615. 51. Seeger W, Lepper H, Hellmut RD, Neuhof H. Alteration of alveolar surfactant function after exposure to oxidative stress and to oxygenated and native arachidonic acid in vitro. Biochim Biophys Acta 1985; 835:58–67. 52. Liau DF, Yin NX, Huang J, Ryan SF. Effects of human polymorphonuclear leukocyte elastase upon surfactant proteins in vitro. Biochim Biophys Acta 1996; 1302:117–128. 53. Haddad IY, Crow JP, Hu P, Ye Y, Beckman J, Matalon S. Concurrent generation of nitric oxide and superoxide damages surfactant protein A. Am J Physiol 1994; 267:L242–L249. 54. Zhu S, Kachel DL, Martin WJ, Matalon S. Nitrated SP-A does not enhance adherence of Pneumocystis carinii to alveolar macrophages. Am J Physiol 1998; 275:L1031–L1039. 55. McGuire WW, Spragg RG, Cohen AB, Cochrane CG. Studies on the pathogenesis of the adult respiratory distress syndrome. J Clin Invest 1982; 69: 543–553. 56. Christner P, Fein A, Goldberg S, Lippmann M, Abrams W, Weinbaum G. Collagenase in the lower respiratory tract of patients with adult respiratory distress syndrome. Am Rev Respir Dis 1985; 131:690–695. 57. Pison U, Tam EK, Caughey GH, Hawgood S. Proteolytic inactivation of dog lung surfactantassociated proteins by neutrophil elastase. Biochim Biophys Acta 1989; 992:251–257. 58. Baker CS, Evans TW, Randle BJ, Haslam PL. Damage to surfactant-specific protein in acute respiratory distress syndrome. Lancet 1999; 353:1232–1237. 59. Hite RD, Seeds MC, Jacinto RB, Balasubramanian R, Waite M, Bass D. Hydrolysis of surfactant-associated phosphatidylcholine by mammalian secretory phospholipases A2. Am J Physiol 1998; 275:L740–L747. 60. Lewis JF, Tabor B, Ikegami M, Jobe AH, Joseph M, Absolom D. Lung function and surfactant distribution in saline-lavaged sheep given instilled vs. nebulized surfactant. J Appl Physiol 1993; 74:1256–1264. 61. Lewis JF, Goffin J, Yue P, McCaig LA, Bjarneson D, Veldhuizen RA. Evaluation of exogenous surfactant treatment strategies in an adult model of acute lung injury. J Appl Physiol 1996; 80:1156–1164. 62. Moya FR, Hoffman DR, Zhao B, Johnston JM. Platelet-activating factor in surfactant preparations. Lancet 1993; 341:858–860. 63. Häfner D, Beume R, Kilian U, Krasznai G, Lachmann B. Dose-response comparisons of five lung surfactant factor (LSF) preparations in an animal model of adult respiratory distress syndrome (ARDS). Br J Pharmacol 1995; 115:451–458. 64. Amirkhanian JD, Bruni R, Waring AJ, Taeusch HW. Inhibition of mixtures of surfactant lipids and synthetic sequences of surfactant proteins SP-B and SP-C. Biochim Biophys Acta 1991; 1096:355–360. 65. Segerer H, van Gelder W, Angenent FW, van Woerkens LJ, Curstedt T, Obladen M, Lachmann B. Pulmonary distribution and efficacy of exogenous surfactant in lung-lavaged rabbits are influenced by the instillation technique. Pediatr Res 1993; 34:490–494.

Surfactant therapy in the acute respiratory distress syndrome

481

66. Gregory TJ, Steinberg KP, Spragg R, Gadek JE, Hyers TM, Longmore WJ, Moxley MA, Cai GZ, Hite RD, Smith RM, Hudson LD, Crim C, Newton P, Mitchell BR, Gold AJ. Bovine surfactant therapy for patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1997; 155:1309–1315. 67. Walmrath D, Günther A, Ghofrani HA, Schermuly R, Schneider T, Grimminger F, Seeger W. Bronchoscopic surfactant administration in patients with severe adult respiratory distress syndrome and sepsis. Am J Respir Crit Care Med 1996; 154:57–62. 68. Lewis J, McCaig L, Hafner D, Spragg R, Veldhuizen R, Kerr C. Dosing and delivery of a recombinant surfactant in lung-injured adult sheep. Am J Respir Crit Care Med 1999; 159:741– 747. 69. Schlosser RL, Veldman A, Fischer D, Funk B, Brand J, von L.V. Comparison of effects of perflubron and surfactant lung lavage on pulmonary gas exchange in a piglet model of meconium aspiration. Biol Neonate 2002; 81:126–131. 70. Lewis JF, McCaig L. Aerosolized versus instilled exogenous surfactant in a nonuniform pattern of lung injury. Am Rev Respir Dis 1993; 148:1187–1193. 71. Rebello CM, Jobe AH, Eisele JW, Ikegami M. Alveolar and tissue surfactant pool sizes in humans. Am J Respir Crit Care Med 1996; 154:625–628. 72. Gilliard N, Richman PM, Merritt TA, Spragg RG. Effect of volume and dose on the pulmonary distribution of exogenous surfactant administered to normal rabbits or to rabbits with oleic acid lung injury. Am Rev Respir Dis 1990; 141:743–747. 73. Spragg RG, Smith RM, Harris K, Lewis J, Häfner D, Germann P. Effect of recombinant SP-C surfactant in a porcine lavage model of acute lung injury. J Appl Physiol 1999; 88:674–681. 74. Soll RF, Morley CJ. Prophylactic versus selective use of surfactant in preventing morbidity and mortality in preterm infants [Cochrane Review]. The Cochrane Library, Oxford: Update Software, 2002. 75. Novick RJ, MacDonald J, Veldhuizen RA, Wan F, Duplan J, Denning L, Possmayer F, Gilpin AA, Yao LJ, Bjarneson D, Lewis JF. Evaluation of surfactant treatment strategies after prolonged graft storage in lung transplantation. Am J Respir Crit Care Med 1996; 154:98–104. 76. Spragg RG, Lewis JF, Rathgeb F, Häfner D, Seeger W. Intratracheal instillation of rSP-C surfactant improves oxygenation in patients with ARDS. Am J Respir Crit Care Med. In press. 77. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 78. Lachmann B. Surfactant replacement in acute respiratory failure: animal studies and first clinical trials. In: Lachmann B, ed. Surfactant Replacement Therapy. New York: SpringerVerlag, 1987:212–223. 79. Spragg RG, Gilliard N, Richman P, Smith RM, Hite RD, Pappert D, Robertson B, Curstedt T, Strayer D. Acute effects of a single dose of porcine surfactant on patients with the adult respiratory distress syndrome. Chest 1994; 105:195–202. 80. Weg JG, Balk RA, Tharratt RS, Jenkinson SG, Shah JB, Zaccardelli D, Horton J, Pattishall EN. Safety and potential efficacy of an aerosolized surfactant in human sepsis-induced adult respiratory distress syndrome. J Am Med Assoc 1994; 272:1433–1438. 81. Anzueto A, Baughman R, Guntupalli KK, Weg JG, Wiedemann HP, Raventos AA, Lemaire F, Long W, Zaccardelli DS, Pattishall EN. Aerosolized surfactants in adults with sepsis-induced acute respiratory distress syndrome. N Engl J Med 1996; 334:1417–1421. 82. Haslam PL, Hughes DA, MacNaughton PD, Baker CS, Evans TW. Surfactant replacement therapy in late-stage adult respiratory distress syndrome. Lancet 1994; 343:1009–1011. 83. Günther A, Ruppert C, Schmidt R, Markart P, Grimminger F, Walmrath D, Seeger W. Surfactant alteration and replacement in acute respiratory distress syndrome. Respir Res 2001; 2:353–364. 84. Seeger W, Spragg RG, Taut FJH, Häfner D, Lewis JF. Treatment with rSP-C surfactant reduces mortality in ARDS due to primary pulmonary events. Am J Respir Crit Care Med; 165:A219.

Acute respiratory distress syndrome

482

85. Kesecioglu J, Schultz MJ, Lundberg D, Lauven PM, Lachmann B. Treatment of acute lung injury (ALI/ARDS) with surfactant. Am J Respir Crit Care Med 2001; 165:A819. 86. Willson DF, Jiao JH, Bauman LA, Zaritsky A, Craft H, Dockery K, Conrad D, Dalton H. Calf’s lung surfactant extract in acute hypoxemic respiratory failure in children. Crit Care Med 1996; 24:1316–1322. 87. Lewis JF, Dhillon JS, Singh RN, Johnson CC, Frewen TC. Exogenous surfactant therapy for pediatric patients with the acute respiratory distress syndrome. Can Respir J 1997; 4:21–26. 88. Wiswell TE, Smith RM, Katz LB, Mastroianni L, Wong DY, Willms D, Heard S, Wilson M, Hite RD, Anzueto A, Revak SD, Cochrane CG. Bronchopulmonary segmental lavage with Surfaxin (KL(4)-surfactant) for acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160:1188–1195. 89. Griese M. Pulmonary surfactant in health and human lung diseases: state of the art. Eur Respir J 1999; 13:1455–1476. 90. Pison U, Seeger W, Buchhorn R, Joka T, Brand M, Obertacke U, Neuhof H, Schmit-Neuerburg KP. Surfactant abnormalities in patients with respiratory failure after multiple trauma. Am Rev Respir Dis 1989; 140:1033–1039. 91. Nakos G, Kitsiouli EI, Tsangaris I, Lekka ME. Bronchoalveolar lavage fluid characteristics of early intermediate and late phases of ARDS. Alterations in leukocytes, proteins, PAF and surfactant components. Intensive Care Med 1998; 24:296–303. 92. Bersten AD, Doyle IR, Davidson KG, Barr HA, Nicholas TE, Kermeen F. Surfactant composition reflects lung overinflation and arterial oxygenation in patients with acute lung injury. Eur Respir J 1998; 12:301–308. 93. Berggren P, Lachmann B, Curstedt T, Grossmann G, Robertson B. Gas exchange and lung morphology after surfactant replacement in experimental adult respiratory distress syndrome induced by repeated lung lavage. Acta Anaesthesiol Scand 1986; 30:321–328. 94. Vazquez dAG, Lachmann RA, Gommers D, Verbrugge SJ, Haitsma J, Lachmann B. Treatment of ventilation-induced lung injury with exogenous surfactant. Intensive Care Med 2001; 27:559–565. 95. Eijking EP, Gommers D, So KL, Vergeer M, Lachmann B. Surfactant treatment of respiratory failure induced by hydrochloric acid aspiration in rats. Anesthesiology 1993; 78:1145–1151. 96. Strohmaier W, Redl H, Schlag G. Studies of the potential role of a semi-synthetic surfactant preparation in an experimental aspiration trauma in rabbits. Exp Lung Res 1990; 16:101–110. 97. van Daal GJ, Bos JA, Eijking EP, Gommers D, Hannappel E, Lachmann B. Surfactant replacement therapy improves pulmonary mechanics in end-stage influenza A pneumonia in mice. Am Rev Respir Dis 1992; 145:859–863. 98. van Daal GJ, So KL, Gommers D, Eijking EP, Fievez RB, Sprenger MJ, van Dam DW, Lachmann B. Intratracheal surfactant administration restores gas exchange in experimental adult respiratory distress syndrome associated with viral pneumonia. Anesth Analg 1991; 72:589– 595. 99. Eijking EP, van Daal GJ, Tenbrinck R, Sluiters JF, Hannappel E, Erdmann W, Lachmann B. Improvement of pulmonary gas exchange after surfactant replacement in rats with Pneumocystis carinii pneumonia. Adv Exp Med Biol 1992; 316:293–298. 100. Berry D, Ikegami M, Jobe A. Respiratory distress and surfactant inhibition following vagotomy in rabbits. Am J Physiol 1986; 61:1741–1748. 101. Harris JD, Jackson FJ, Moxley MA, Longmore WJ. Effect of exogenous surfactant instillation on experimental acute lung injury. J Appl Physiol 1989; 66:1846–1851. 102. Anderson BM, Jackson FJ, Moxley MA, Longmore WJ. Effects on experimental acute lung injury 24 hours after exogenous surfactant instillation. Exp Lung Res 1992; 18:191–204. 103. Chen CM, Fang CL, Chang CH. Surfactant and corticosteroid effects on lung function in a rat model of acute lung injury. Crit Care Med 2001; 29:2169–2175.

Surfactant therapy in the acute respiratory distress syndrome

483

104. Huang YC, Sane AC, Simonson SG, Fawcett TA, Moon RE, Fracica PJ, Menache MG, Piantadosi CA, Young SL. Artificial surfactant attenuates hyperoxic lung injury in primates. I Physiology and biochemistry. J Appl Physiol 1995; 78:1816–1822. 105. Loewen GM, Holm BA, Milanowski L, Wild LM, Notter RH, Matalon S. Alveolar hyperoxic injury in rabbits receiving exogenous surfactant. J Appl Physiol 1989; 66:1087–1092. 106. Hallman M, Merritt TA, Schneider H, Epstein BL, Mannino F, Edwards DK, Gluck L. Isolation of human surfactant from amniotic fluid and a pilot study of its efficacy in respiratory distress syndrome. Pediatrics 1983; 71:473–482. 107. Taeusch HW, Keough KM, Williams M, Slavin R, Steele E, Lee AS, Phelps D, Kariel N, Floros J, Avery ME. Characterization of bovine surfactant for infants with respiratory distress syndrome. Pediatrics 1986; 77:572–581. 108. Enhorning G, Shennan A, Possmayer F, Dunn M, Chen CP, Milligan J. Prevention of neonatal respiratory distress syndrome by tracheal instillation of surfactant: a randomized clinical trial. Pediatrics 1985; 76:145–153. 109. Onrust SV, Dooley M, Goa KL. Calfactant: a review of its use in neonatal respiratory distress syndrome. Paediatr Drugs 1999; 1:219–243. 110. Robertson B, Curstedt T, Johansson J, Jornvall H, Kobayashi T. Structural and functional characterization of porcine surfactant isolated by liquid-gel chromatography. In: von Wichert P, Muller B, eds. Basic Research on Lung Surfactant. Progress in Respiration Research. Basel: Karger, 1990:237–246. 111. Disse B, Gortner L, Weller E, Eberhardt H, Ziegler H. Efficacy and standardization of SF-RI 1: a preparation from bovine lung surfactant. In: Lachmann B, ed. Surfactant Replacement Therapy in Neonatal and Adult Respiratory Distress Syndrome. Berlin: Springer, 1999:37–41. 112. Fujiwara T, Chida S, Watabe Y, Maeta H, Morita T, Abe T. Artificial surfactant therapy in hyaline-membrane disease. Lancet 1980; i:55–59. 113. Cochrane CG, Revak SD, Merritt TA, Heldt GP, Hallman M, Cunningham MD, Easa D, Pramanik A, Edwards DK, Alberts MS. The efficacy and safety of KL4-surfactant in preterm infants with respiratory distress syndrome. Am J Respir Crit Care Med 1996; 153:404–410. 114. Häfner D, Germann PG, Hauschke D. Effects of lung surfactant factor (LSF) treatment on gas exchange and histopathological changes in an animal model of adult respiratory distress syndrome (ARDS): comparison of recombinant LSF with bovine LSF. Pulm Pharmacol 1994; 7:319–332. 115. Morley CJ, Bangham AD, Miller N, Davis JA. Dry artificial lung surfactant and its effect on very premature babies. Lancet 1981; 1:64–68. 116. Durand DJ, Clyman RI, Heymann MA, Clements JA, Mauray F, Kitterman J, Ballard P. Effects of a protein-free, synthetic surfactant on survival and pulmonary function in preterm lambs. J Pediatr 1985; 107:775–780. 117. Spragg RG, Smith RM, Harris K, Lewis J, Häfner D, Germann P. Effect of recombinant SP-C surfactant in a porcine lavage model of acute lung injury. J Appl Physiol; 88:674–681.

20 Prone Position in the Acute Respiratory Distress Syndrome ANTONIO ANZUETO The University of Texas Health Science Center at San Antonio San Antonio, Texas, U.S.A. LUCIANO GATTINONI University of Milan and Hospital of Milan I.R.C.C.S. Milan, Italy

I. Introduction Since the description of acute respiratory distress syndrome (ARDS) in 1967 (1), mechanical ventilation has been the mainstay therapy of support to these patients. Furthermore, mechanical ventilation has been shown to result in lung injury; thus therapeutic strategies focused on preventing this injury have been shown to significantly decrease mortality (2). It was almost 10 years after the clinical description of ARDS that Piehl and Brown (3) first described the benefit of positional changes to improve arterial oxygenation in five patients with acute respiratory failure. Douglas et al. (4) confirmed these data in a more extensive study and also described significant improvement in arterial oxygenation in most, but not all, patients who were placed in the prone position. Possible mechanisms to explain the improvement of gas exchange during the prone position included a redistribution of blood flow and/or ventilation, an increase in lung functional and residual capacities, and changes in intrapulmonary pressure gradients. These two studies were based on the theoretical work by Bryan (5), who advocated the prone position in mechanically ventilated patients in order to improve regional inflation of dorsal portions of the lung. Over the last 20 years, there have been extensive reports on the pathophysiology of prone ventilation as well as the clinical application in acute lung injury (ALI) and ARDS. This chapter reviews the factors related to improvement of oxygenation due to prone ventilation including the effect on ventilation and perfusion and how changes in body position affect gas exchange. Finally, there is a summary of clinical data including a recently published multicenter, randomized trial on the use of prone ventilation in patients with ALI/ARDS and other forms of acute lung injury.

Prone position in the acute respiratory distress syndrome

485

II. Pathophysiology in ARDS ARDS is characterized by pulmonary infiltrates, which were previously considered to be rather homogeneously distributed throughout the lung, as evidenced by conventional chest radiographs (6, 7). Considering the complexity of the anatomical changes occurring in ARDS lung structure, regional chest tomography analysis allows the assessment of these parenchymal abnormalities. Chest tomography technology shows that in ARDS, the lung abnormalities are primarily located in the dependent regions, i.e., the dorsal regions (lower) in the supine position (8–10). In contrast, the nondependent regions, i.e., the upper sternal regions in the supine position, seem, at least upon visual inspection, normal. These findings have challenged the commonly held opinion that ALI/ARDS are a generalized lung disease and suggest that lung parenchyma is not affected in a uniform way. In ALI/ARDS patients, infiltrates in lung parenchyma are seen by chest radiographs in up to 70–80% of the lung fields, depending on the severity of respiratory failure. The proportion of lung that can be ventilated is consequently reduced to almost 20–30% of a normal lung and therefore may have the dimensions of a baby lung (11). Thus, the ARDS lung has three compartments: one substantially normal (healthy zone), one fully diseased without any possibility of recruitment (consolidated zone), and one composed of collapsed alveoli, potentially recruitable with maneuvers (recruitable zone) (12). Thus, the overall thoracic volume is not substantially different between ARDS and normal patients, suggesting that in ARDS, the decrease in lung volume does not indicate a decrease in total thoracic volume, but a simple replacement of gas with tissue volume (13). The reduction of respiratory compliance, a typical finding in ARDS, was previously attributed to the severity of the disease and/or to intrinsic mechanical alterations of the lung tissue. Using chest tomography, Gattinoni et al. (12) demonstrated that, contrary to the commonly accepted notion, respiratory compliance was not related to the amount of “disease” tissue, but to the amount of residual inflated lung, indicating that the smaller the portion of lung open to gas, the lower the compliance. In contrast, respiratory compliance was correlated with a normally inflated part of the lung, and the gas exchange impairment was strongly related to the amount of non-inflated tissue mass, i.e., to the extent of the disease (12). These data suggest that the main cause of severe hypoxemia is the perfusion of noninflated lung tissue. Moreover, these investigators hypothesized that the dependent lung regions, where the majority of noninflated tissue is present, may be under-perfused, as evidenced by the discrepancy between the shunt fraction and the noninflated tissue fraction (i.e., 30% of shunt with 60% of noninflated tissue). The underperfusion of this lung region resulted in mechanical compression of the blood vessels and worsening hypoxic pulmonary constriction (13). Further studies, using regional computed tomography analysis of the lung, described that in the supine position homogeneous pattern affected all lung parenchyma with no part of the lung being healthy and that interstitial edema did not exhibit gravitydependent distribution along the vertical gradient (14) (Fig. 1). These findings were in accordance with previous reports obtained in animal models of respiratory failure (15,

Acute respiratory distress syndrome

486

16) and in patients with cardiogenic pulmonary edema (17). These data suggest that the mechanism leading to interstitial edema, at least in ARDS, acts equally in each

Figure 1 Tissue and gas volumes in supine lung levels 1 (ventral) to 10 (dorsal) in normal subjects (first bar in each pair) and in patients with ARDS (second bar in each pair). The total lung volumes were similar in the two groups. In patients with ARDS, the gas volume was lower and tissue volume higher in each level. (From Ref. 14.) part of the lung; thus, edema cannot move freely through the interstitial space. As a consequence, in ARDS the regional superimposed pressure is the major factor explaining the increased regional inflation gradients, since it causes lung deflation and collapse along the vertical gradient due to increased weight (18). Understanding the regional perfusion and ventilation factors is important in order to determine the effect of body position on these variables. In the upright position, perfusion increases from the nondependent to the dependent lung regions, approximately three quarters of the way down the lung and below (19–21). The increase in perfusion gradients has classically been explained by the relationship between the intravascular hydrostatic pressure and the alveolar pressure at various gravitational levels (lung zone 1 and 2) together with distensibility to the pulmonary circulation (zone 3) (19). The decrease in perfusion occurring in more dependent regions (i.e., zone 4) has been

Prone position in the acute respiratory distress syndrome

487

attributed to several factors including extra-alveolar vessel narrowing resulting from low volumes, interstitial edema present in the area (21, 22), and hypoxic vasoconstriction (23). At the same time, the gravitational gradient described in the upright position has also been observed in the supine and lateral decubitus positions. Several investigators have shown that there is a gravitational gradient of regional alveolar volume in the different body positions. In the upright and supine positions, the nondependent alveoli are more expanded than those in the dependent lung regions at functionally residual capacity (FRC) and at all lung volumes above FRC until the lung reaches total lung capacity (TLC) (20, 24, 25). Accordingly, under normal conditions, during tidal breathing from FRC to end expiration, alveolar volume-ventilation increases along the gravitational gradients (26). When ventilation is below FRC, small airways in the dependent lung regions close and the early part of inhalation is directed to less dependent regions. In summary, in patients with ALI/ARDS the changes in perfusion and ventilation to the lung parenchyma are reduced by placing these patients in a supine position, and as a consequence there is usually an improvement in oxygenation.

III. Effects of Prone Position The gravitational gradients of perfusion described in upright lungs have also been observed in the head-down, supine, and right and left lateral decubitus positions (27, 28). Although some investigators have observed an increased gravitational gradient of perfusion in the prone position, this gradient is markedly reduced compared with that found in the other positions (28, 29). In animal studies, Wiener et al. (30) described a strong gravitational gradient of perfusion in the supine position before and after ALI was produced with oleic acid, but when the animals were turned to the prone position, these perfusion gradients changed very little (Fig. 2). These observations confirm that the “gravitational” gradient in perfusion does exist, but the role played by gravity in determining the gradient is minimal. Several mechanisms have been suggested to explain the more uniform lung perfusion seen in prone positions. Reed and Wood (28) suggested that there might be regional differences in pulmonary vascular resistance resulting from changes in interstitial pressure due to variable regional lung expansion in the two positions. More recently, Beck and Lai-Fook (31) suggested that this explanation is unlikely, particularly over the range of lung volume at which tidal breathing occurs, and changes in pulmonary vascular resistance that occur with tidal breathing are extremely small.

Acute respiratory distress syndrome

488

Figure 2 Effect of position on regional distribution of perfusion after oleic acid-induced acute lung injury. Note that prone position has no effect on regional perfusion distribution. Other mechanisms may be that caudal movement of the ventral portion of the diaphragm might be reduced in a prone position (29), but this hypothesis has not been supported by recent reports (32). Other investigators suggest that the improvement in oxygenation seen with changes in body position is secondary to improvement of ventilation to dependent lung regions. This explanation is supported by both clinical data and animal experiments demonstrating that the prone position reduces arterio-venous shunt. Lung perfusion is preferentially directed to dorsal lung regions regardless of po-sition, such that the prone position causes minimum, if any, alteration in regional perfusion. Thus, if there is no change in perfusion, turning a patient to the prone position most improved regional ventilation. Several investi-gators have studied the effect of the gravitational pleural gradient. The gravitational pleural pressure gradient is more uniform in the prone posi-

Prone position in the acute respiratory distress syndrome

489

tion compared with the supine position, but the pleural pressure in the de-pendent regions becomes positive in the setting of lung edema when measured in the supine position, and turning the patient prone reduces the pleural pressure in the dependent regions (Table 1). Lamm et al. (33) directly measured the regional ventilation/perfusion (VE/Q) ratios with SPECT scanning. In the supine position, the VE/Q ratio in normal lungs was distributed in the skewed fashion with the median VE/Q shifted towards values approximating 0.8 (Fig. 3). The VE/Q ratios increased from the dor-sal (dependent) to the ventral (nondependent) regions. On turning the ani-mals to the prone position, the VE/Q was distributed in a more “gaussian fashion.” The median VE/Q was improved and no gravitational gradient were observed. These investigators found that in injured lungs the VE/Q distribution was markedly altered, and after turning the animal to the prone position the VE/Q distribution shifted considerably. These observations con-firm the premise that the pleural pressure gradient did not simply reverse on turning prone such that pleural pressure in these ventral regions was not sufficiently positive because of airspace collapse or failure of airspace open-

Table 1 Differences in Transpulmonary Pressurea in Prone and Supine Positions Before and After Volume Infusion Before volume infusion

After volume infusion

Supine

Supine

Prone

Prone

Nondependent

−3±0.6

−1.3±0.2*

−2.3±0.2

−0.9±0.2*

Dependent

0.7±0.3

−0.1±0.2*

3±0.5**

0.9±0.3*

a

Ppl cm of water. * p<0.05 supine vs prone position, before and after volume infusion. ** p<0.05 before vs after volume infusion, supine or prone position.

Acute respiratory distress syndrome

490

Figure 3 Distribution of regional ventilation and perfusion in (a) normal dog, supine and prone and (b) injured dogs after oleic acid-induced acute lung injury, supine and prone. (From Ref. 33.) ing (Fig. 4). Table 2 summarizes the effects of the prone position in ventilation and perfusion of patients with ARDS.

IV. Effect of Mediastinal Weight on Lung Parenchyma Above we discussed that reversible airspace closure occurs in the dorsal lung regions when patients with ALI/ARDS are placed in the supine position and

Prone position in the acute respiratory distress syndrome

491

Figure 4 Superimposed pressure in patients with ARDS (open circles) and reference normals (closed circles). The superimposed pressure overlying each lung level from nondependent to dependent regions is markedly increased as compared with normals. (From Ref. 14.) that turning patients to the prone position will alter dorsal lung transpulmonary pressures and reverse airway closure. A number of factors would contribute to the ability of the prone position to alter dorsal lung transpulmonary pressures, including, among others, compressive effects of consolidated lung (34), direct transmission of the weight of abdominal content to the caudal regions of the dorsal lung (5, 35), and direct transmission of the heart weight to the lung regions located beneath it (36, 37). The idea that the heart can have an impact on the regional lung distention and that the lung can be affected by heart transmural pressures dates back to 1947, when Brookhart and Boyd (36) noted that “the dog’s heart produces deformation

Table 2 Ventilation/Perfusion Effect of Prone Position in ARDS Reduction in shunt. Perfusion is preferentially directed to dorsal lung regions regardless of position. Gravitational pleural pressure gradient is more uniform.

Acute respiratory distress syndrome

492

Pleural pressure is reduced in dependent regions. Regional ventilation/perfusion ratio is more uniform and better matched.

of the adjacent lung, raising pressure on the external surface of the heart above the pressure existing between the lung and the wall of the thorax.” These observations were confirmed by other investigators, who suggested that the weight of the heart accounts for regional differences in pleural pressures and compression of lung parenchyma (38–41). The heart-lung interaction and its effect on ventilation were noted by oscillation in expired gas flow (measured at the mouth with a body plethysmograph) that corresponded to the heartbeat (42). These effects are due to a direct local mechanical distortion of the lungs. Other investigators described differences in a single breath oxygen test, which is attributed in part to the weight of the heart (43). Weiner et al. (44) described that left lower lobe ventilation is impaired when patients with cardiomegaly were positioned supine, but not when they were positioned prone. All these observations were later confirmed by Ball et al. (46), who studied the effects of posture on thoracic anatomy using chest tomography and by spatial reconstruction. Hoffman observed marked supineto-prone differences in the regional air content of the lung and found that these differences correlated with a shift in the position of the mediastinal content, including the heart (46). More recently, Albert and Hubmayr (47) measured the relative lung volume that is located directly under the heart both in supine and prone position. The study population consisted of normal subjects (four male and three females) with a mean age of 49±18 years (range 28–73). None of the patients had a history of cardiac disease. Four exhaled tomography sections between the carina and the diaphragm were analyzed (sections 1 through 4). When supine, the percent of the total lung volume located under the heart increased from 7±4% to 42±8% and from 11±4% to 16±40% in sections 1 through 4, in the left and right lungs, respectively (Fig. 5). When patients were studied in prone position, the percent of lung volume located under the heart was less than 1% and 4%, left and right lung, respectively, in all four sections (p<0.05 for each section, supine vs. prone) (Fig. 5). The main finding of this study was that in the supine position, a considerable fraction of both lungs is located underneath the heart and, as such, would be subject to compressive forces resulting from weight of the heart and the blood contained therein. In the prone position, only a very small fraction of either lung was similarly affected. The investigators suggested that the clinical implications of these findings are that these changes will lower the inspiratory and end-expiratory pressures required to obtain maximum air-space recruitment (i.e., alveolar recruitment and/or airway opening) and reduce cyclical airspace opening and closing. If ventilator-induced lung injuries are related to any or all of these factors, then routine use of the prone position could translate into reduction of morbidity and mortality in patients with ALI/ARDS.

Prone position in the acute respiratory distress syndrome

493

Figure 5 Computer chest tomograms—the areas of lung medial to the perpendicular lines were quantified as being under the heart: (a) supine position; (b) prone position. (From Ref. 47.)

Acute respiratory distress syndrome

494

A major limitation of this study is that it was done on normal subjects, and we do not know if the same changes are going to happen in patients with underlying lung disease. The isolated effect of heart weight will be difficult to estimate in patients with ARDS. In this disease state, there are markedly regional variations in lung volume; the alveolar space is fluid-filled to variable degrees; there is soft tissue edema that alters thoracic and abdominal compliance; there are variable abdominal wall edema, ascites, and increased abdominal pressure; and patients may or may not have cardiomegaly. All these parameters may explain the variability in gas exchange improvement that is seen when patients with ARDS are turned from the supine to the prone position.

V. Can Prone Position Decrease Ventilator-Induced Injury? Numerous reports have indicated that ventilating the lungs with excessive volume is deleterious and that ventilator-induced lung injury can occur (48, 49). Furthermore, we now understand that the lung injury is not homogeneous and, as previously described, mechanical ventilation is limited to a small portion of the lung parenchyma (“baby lung”) (11, 15). So, consensus has emerged that tidal volume should be kept at or below the static inflation pressure of 35 cmH2O and PEEP should be titrated to exceed airwayopening pressure (50). The initial studies by Webb and Tierney (51) demonstrated that although high inflation pressure and large volume ventilation produce lung injury and edema, the adverse effects were markedly reduced by adding PEEP. The working hypothesis is that lung injury results from the sheer force associated with repeat airspace opening and closing, and lung injury is not uniform (52). If volutrauma was caused by lung overdistention, this effect will be distributed preferentially to the lung ventral regions where lung distention is greater. The fact that they were not is consistent with the idea that sheer forces associated with airspace opening and closing are responsible for the injury (53). These data suggest that our attention should be directed more towards preventing airspace closure than on avoiding lung overdistention. Accordingly, various ventilatory strategies have been proposed to counter this effect. Amato et al. (54) first demonstrated that such an approach could reduce the mortality of patients with ARDS. This work was confirmed by the study completed by the ARDS Network, which found that using a tidal volume of 6 mL/kg as opposed to 12 mL/kg ideal body weight decreased mortality by 25% (from 40 to 30%) (2). Accordingly, strong arguments can be made today that ventilator-induced injuries are clinically important entities that adversely affect clinical outcome. These studies also confirmed a decrease in the number of multiple organ failure, confirming the prior hypothesis that how mechanical ventilation is applied will have an impact on patient survival, not only by improving oxygenation, but also by decreasing other organ damage (2). It has been suggested by animal experiments that the severity and distribution of ventilator-induced lung injury is altered by the combination of PEEP, prone position, and respiratory frequency (55). The ability of prone ventilator position to generate more uniform lung distention and to reduce the compressive effect of the dorsal lung in the setting of acute lung injury would seem to be of major theoretical benefit and may account for the improvement of oxygenation seen in most patients. Broccard et al. (56) have expanded this theory by confirming that supine ventilation can cause lung injury

Prone position in the acute respiratory distress syndrome

495

and by demonstrating that ventilating animals in the prone position can reduce the extent of lung injury. Albert et al. (57) showed in an oleic acid-injured dog model that early prone position could markedly reduce the lung inflammatory response. In this study, after the animals were injured, they were immediately randomized to either the supine or the prone position and then ventilated for 6 hours on 100% oxygen. Thirty minutes after the start of mechanical ventilation there were significant differences in oxygenation between groups. Furthermore, there were also differences in lung weight to dry ratio, whole lung lavage cell counts, and protein content. The percentage of PMN was lower in animals ventilated in the prone position than in the supine position—49±21 vs. 77±24 (p=0.06), respectively. These data suggested that early use of prone ventilation in dogs not only improved oxygenation but also limited the inflammatory burden. These data have not been confirmed in humans.

VI. The Use of Prone Ventilation in ARDS The importance of body position changes during mechanical ventilation was first demonstrated by Ray et al. (58) in three groups of anesthestized dogs with experimentally induced acute lung injury. A control group was left immobile, a second group was turned from side to side every hour, and a third group was turned every half hour. The values fell sharply in the control group, and there was significant arterial-venous shunting. The second group had some improvement in

values and decreased

values returned to normal in only the third group. The first use of shunting, but the changes in body position in humans was reported by Piehl and Brown (3) in five patients with ARDS. Their data suggest that 180° changes in position resulted in a significant (47 torr), presumably by reduction of alveolar-arterial oxygen gradients increase in and improvement in ventilation-perfusion ratios. These data were later confirmed by Douglas et al. (4), who studied the benefit of prone positioning in six patients—five on the ventilator and one with a nonrebreather oxygen mask. These patients had variable by a periods of prone ventilation, but overall there was a significant increase in mean of 69 mmHg [range 2–178 mmHg using the same tidal volume, fraction of inspired oxygen

and level of PEEP]. The prone position made it possible to reduce the

in four of the five patients who required mechanical ventilation. The investigators did not notice any significant change in other parameters, including respiratory frequency, or static compliance. These studies were done based on the theoretical work postulated by Bryan (5), who advocated a trial of prone position for patients who required mechanical ventilation, suggesting that position changes may enhance the patient’s lung expansion and ventilation of the dorsal areas. Despite these encouraging clinical results, prone ventilation has not become an integral part of the treatment of respiratory failure. Langer et al. (59) demonstrated that prone position resulted in a disappearance of infiltrates in the dorsal regions of the lung visualized by computed tomography scan, and at the same time there was an increase in arterial oxygenation in some, but not all patients. Since then, a large number of research

Acute respiratory distress syndrome

496

studies has been focused on trying to understand the pathophysiology and the reason for improvement in oxygenation previously discussed with prone ventilation in ALI/ARDS. While most of the clinical studies focused on the short-term physiological effect and why prone position improved oxygenation, Fridrich et al. (60) demonstrated a significant long-term benefit on gas exchange in a group of patients with trauma-induced ARDS. These investigators studied 20 patients with ARDS and severe hypoxemia despite the fact they were ventilated using inverse ratio ventilation and high PEEP. These patients were periodically placed in the prone position up to 20 hours per day over several days without any major complications. Patients were reassessed every morning in a supine position, and if the patient’s hypoxemia persisted, they were again placed in the prone position. The oxygen variables improved significantly each time the patient was placed in the prone position. Immediately after the first turn from the supine to prone position, the increased from a mean of 97±4 to 152±15 mmHg investigators observed that the (p≤0.05), with a significant decrease in intrapulmonary shunt. Most of these improvements were lost when the patients were turned supine but could be reproduced when prone position was repeated after a short period (4 hours) in the supine position. Investigators placed patients in supine position in order to allow nursing care, medical evaluation, and intervention such as placement of central lines. No position-dependent changes in systemic hemodynamics were observed. The overall mortality of the study group was 10%. The homogeneity of the patient group in terms of etiology, prestudy ventilation time, and early onset of positioned therapy may be all important factors for the positive response to the prone position seen in these patients. In 1997 Chatte et al. (61) reported the pattern of

and

response when

ratio of <150) were turned patients with severe acute respiratory failure ( from a supine to prone position. Thirty-two consecutive patients with heterogeneous causes of ARDS, including sepsis, aspiration, pneumonia, etc., were studied one hour before, 1 and 4 hours during, and 1 hour after being placed in the prone position. The was 103± 28, 158±62, 159±59, and 128±52, respectively (ANOVA, p<0.01). Seven patients in the study (22%) did not show an improvement in oxygenation; they are referred to as nonresponders. Twenty-three patients (78%) had a significant improvement, and they are referred to as responders. Among the seven non-responders, two did not tolerate the prone position and were returned to the preprone position before the end of the 4-hour trial (Fig. 6). In the other five patients, the

ratio

ratio returned to its remained below 90. In 10 of the 23 responders (43%), the pre-prone value when patients were repositioned—these patients are called partial responders. In 13 of the other 23 patients (57%), improvement in oxygenation persisted in the supine position. These patients did not require repositioning from supine to prone. The investigators collected safety data during 994 episodes of prone positioning. Superficial cutaneous and mucosa

Prone position in the acute respiratory distress syndrome

497

Figure 6 Evolution of before, during, and after the first 4 hours of prone trial. Study groups: RP=responders, persistent; RNP=responders, nonpersistent; NR=no responders; Sbf=1 hr before prone; PH1=1st hour during prone; PH4 = 4th hour during prone; Saft=1 hr after supine. (From Ref. 61.) damage was frequently seen, mainly affecting the anterior chest wall, lips, tongue, or forehead. Subcutaneous edema was seen in all the patients. Two patients had a severe decrease in their oxygen saturation when placed in the prone position. Overall, excluding cutaneous and mucosa damage, these investigators reported a low number of side effects in 6 of 32 (19%) of all patients, but only 6 of 294 during the prone position (2%). The overall mortality of the study group was 56% (18/32), and none were related to the prone position. It is important to point out that in this study, the investigators stated that four attendants helped turn the patients from the supine to the prone position. The tolerance to prone position was systematically monitored during this trial; hemodynamic measures did not show any significant changes in cardiac output, mean arterial pressure, capillary wedge pressure, and pulmonary and/or systemic vascular resistance. This was the first study to identify three different types of response when patients with ARDS are changed from the supine to the prone position: nonresponders, partial responders, and responders. Furthermore, this study identified that patients with the lower highest PEEP are the least likely to improve in the prone position.

ratio and the

Acute respiratory distress syndrome

498

Studies by Jolliet et al. (62) reported a lower percentage of patients responding to the prone position (57%). These investigators reported that patients who improved in the prone position when returned to the supine position were more likely to respond to additional prone positioning (71% of the time), but if the patients were initially nonresponders, they were less likely to respond (25%). Furthermore, these investigators found that the absence of response to the prone position was not accompanied by worsening hypoxemia or hemodynamic instability. They concluded that repeated daily trials of prone position should be considered in the management of ARDS patients with severe hypoxemia. The application of ventilator protective strategies such as low-volume, pressurelimited ventilation in combination with prone position was first reported by Stocker et al. (63). These investigators studied 25 patients with a mean age of 38 years who had trauma or postoperative induced ARDS. Based on the APACHE II score of 24, the predicted mortality rate of the study groups was 35.4%. Seventeen of the 25 patients were treated employing the prone position. Patients were not placed in the prone position if they had severe head injury with elevated intracranial pressure, spine fractures that could not be stabilized, or the presence of focal neurological deficits. The median time interval between ICU admission and first prone position was 10 days (1–26 days). The prone protocol consisted of turning the patients to the prone position and keeping them there until the

ratio improved and/or remained stable. These investigators reported

ratio when they were placed that all the patients had an improvement in their in the prone position. Prone/supine position periods varied considerably from patient to patient (range 2–66 hr). The number of prone positions and maneuvers varied noticeably as well (range 1–9 times). These investigators reported two serious complications during prone positions. In one, the patient became septic after being placed prone, and another suffered an infectious corneal ulceration and needed immediate surgery. The mortality of the overall group was 12%—significantly lower than the predicted mortality. The major contribution of the work of Stocker et al. (63) was that these investigators kept their patients in the prone position as long as gas exchange improved or at least did not deteriorate (up to 48 hr). However, the response in terms of required number and duration of positional maneuvers differed significantly among the patients. Furthermore, they also reported that the clinical response was independent for the patient’s initial lung injury score, and days of ventilator support prior to position changes. Whereas some of the patients showed an immediate improvement, others began to improve after variable time intervals, up to 24 hours. The investigators hypothesized that the improvement in oxygenation was due to a more homogeneous regional inflation distribution and therefore a redistribution of ventilation from dorsal areas, the nondependent lung regions. For the delayed response, the investigators did not have an explanation, and they speculated that edema redistribution, reabsorption, and/or the increase of ventilated lung volume may augment PEEP effects in the nonventilated lung areas. The data suggest that low-volume pressure-limited ventilation combined with permissive hypercapnia and prone positioning may lower the mortality of ARDS. Nakos et al. (64) examined the effect of prone positioning in mechanically ventilated patients with hydrostatic pulmonary edema (HPE), patients with ARDS (early and late disease), and patients with pulmonary fibrosis. All the patients were ventilated with

Prone position in the acute respiratory distress syndrome

499

volume-control ventilation, the tidal volume was set at 6–8 mL/kg body weight, frequency set at 15–25 per minute, and PEEP set just above the value of lower inflation point of the pressure volume curve or 10 cmH2O if they were not able to determine this value. Plateau pressure was kept lower than 35 cmH2O in all the patients. Patients with HPE were turned to the prone position after being mechanically ventilated for at least 6 of more than 60% to achieve an oxygen saturation of 90% or hours, needed an more, and did not respond to recruiting maneuvers. All patients with HPE had a significant improvement in oxygenation when they were placed in the prone position. ratio increased from 72± 16 in the supine position to 208±61 after 6 The hours in the prone position (p< 0.001). This increase in oxygenation was persistent and was not associated with any detrimental effect on hemodynamics. Fifteen of 20 patients ratio with ARDS (75%) improved oxygenation in the prone position. The increased form 83±14 in the supine position to 189±34 after 6 hours of prone position (p<0.01). In contrast, 5 of 20 patients with ARDS (25%) and none of the patients with pulmonary fibrosis showed any improvement in oxygenation during prone position. In this study, not all the patients with ARDS had a significant response. These investigators were able to identify that patients with “early ARDS,” (<36 hours elapsed from the onset of the precipitating factor and diagnosis of ARDS) responded better to prone position that patients with “late ARDS” (>36 hours elapsed from the onset to the diagnosis of ARDS). It is worth pointing out that in “late ARDS” patients the precipitating factor was related to direct insult to the lung. This study showed an improvement in the mortality rate for patients with both HPE and ARDS who were turned prone, compared with their predicted mortality. However, the small number of patients made this comparison powerless to draw definitive conclusions about the influence of position on outcome. In patients with HPE, the oxygenation improvement could be partially due to the relief of the compressive effect of an enlarged heart to the dorsal lung regions. The persistent improvement after turning the patient back to the supine position may be due to a substantial decrease in pulmonary edema, and the relative decrease in heart size, because of treatment. Most of these patients had significant improvement in hemodynamics, mainly manifested as an increase in cardiac index. The increase in cardiac index will improve the V/Q relationship and could be related to the reduction in pulmonary vascular resistance and improved performance of both right and left ventricle. The reduction in pulmonary vascular resistance in both HPE and ARDS patients is probably related to decreased hypoxic pulmonary vasoconstriction or recruitment of atelectatic areas (65). Apart from prone position, factors such as low tidal volume and PEEP could have contributed to the favorable outcomes seen in this study. The major contributions of this study are that the clinical utility of prone ventilation has been shown in other conditions, such as HPE. On the one hand, the existence of pulmonary edema in both early ARDS and HPE can be considered as a predictor of beneficial effect of prone positioning on gas exchange. On the other hand, the presence of lung fibrosis, such as in late ARDS or pulmonary fibrosis, results in no effect of prone positioning. Over the last 30 years, prone ventilation in patients with ARDS has been shown to improve oxygenation in approximately 70% of patients, and most studies suggest that there may be a decrease in mortality. In order to assess the effect of prone ventilation on survival in ARDS, Gattinoni et al. (66) reported the results of an international

Acute respiratory distress syndrome

500

multicenter, randomized trial of patients with ALI and/or ARDS that compared treatment in the supine position with a predefined strategy of placing patients in a prone position for 6 or more hours daily for up to 10 days. These investigators enrolled 304 patients, with 152 patients in each group. Patients were assessed each morning while in supine position and were changed to prone if the

ratio was 200 or less, with a PEEP

ratio of 300 or less with a PEEP of at least 10 of at least 6 cmH2O or a cmH2O. In two patients assigned to the supine group, the prone position was used due to persistent hypoxemia. In the prone group, logistical problems, mainly staffing limitations, resulted in noncompliance in 41 patients, a total of 91 missing periods of prone positioning over a 10-day period. The mortality rate was 23% during the 10-day study period, 49.3% at the time of discharge from the intensive care unit, and 60.5% at 6 months. The relative rate of death in the prone group as compared with the supine group was 0.84 at the end of the study period (95% CI, 0.56–1.27), 1.05 at the time of discharge from the intensive care unit (95% CI, 0.84–1.32), and 1.06 at 6 months (95% CI, 0.88– 1.28). Patients randomized to the prone position remained in this position for an average of 7.0±1.8 hours per day. For all 721 maneuvers, the median change in the ratio was 28 at 1 hour (range −128 to 303) and 44 (range −101 to 319) at the end of the ratio increased period of pronation. In 73% of the pronation procedures, the more than 10, with 70% of the total response observed during the first hour (Fig. 7). A post hoc analysis showed a significant lower 10-day mortality rate in the prone group compared to the supine group in the patients with the

Figure 7 Changes in over 10 days. Data shown are daily measures before prone positioning (circles), after

Prone position in the acute respiratory distress syndrome

501

one hour (squares), at the end period of pronation (triangles), and on the morning of the following day (diamonds). (From Ref. 66.) lowest quartile ratio <88%; 23.1% vs. 47.2% (relative risk of death 0.49, 95% CI, 0.25–0.95). Other parameters identified in post hoc analysis that were associated with decreased mortality associated with prone ventilation included the highest Simplified Acute Physiological Score II (SAPS II) (>49) and the quartile with the highest tidal volume, >12 mL/ kg predicted body weight. The number of new or worsening pressure sores per patient was significantly higher in the prone than in the supine group during the 10-day study period, whereas the number of total days with pressure sores per patient was similar in both groups. As expected, the patients’ weight-bearing areas—thorax, cheekbone, iliac crest, breast, and neck—were significantly more likely to be affected in the prone group. There were no differences in other adverse events, including accidental displacement of tracheal or thoracotomy tube or loss of venous access. The investigators concluded that despite the limitations related to having a smaller number of patients than the originally estimated sample size, the data suggested that prone ventilation of patients with ALI/ARDS did not improve survival. However, these investigators found that prone position improved oxygenation more than 70% of the time in which it was used, with about 30% of the effect occurring during the first hour of pronation. Thus, the ratio increased significantly in the prone patients. These data suggest that prone position may alter the underlying conditions in the respiratory system that leads to worsening of gas exchange in ALI/ARDS. Furthermore, the effect persists beyond the period of pronation. Safety has been one of the major concerns with use of prone ventilation; this study found that in a multicenter trial, there was no increase in complications. Surprisingly, the percentages of patients with new or worsening pressure sores or with displacement of endotracheal tubes, vascular catheters, or thoracotomy tubes were similar in the two groups. It has been speculated that the negative results in survival found by Gattinoni et al. (66) were due to the fact that these investigators studied a more heterogeneous population of patients with ARDS (67) or that there was a need to use combination treatment modalities such as recruitment maneuvers and/or higher levels of PEEP (68). Slutsky (69) stated that these factors are not relevant and underscores the problems involved in performing randomized clinical trials during mechanical ventilation. Thus, the number of patients required to demonstrate the difference in survival will make it necessary to have a heterogeneous patient population with ARDS (e.g., trauma, sepsis, aspiration, pneumonia). Furthermore, it is not clearly defined what the ideal PEEP is in the management of these patients. Several investigators have also suggested the use of high PEEP levels, but this has not been shown in clinical trials. Gattinoni et al. (66) suggested that this study is the beginning of further research collaboration that will address these limitations and evaluate the value of longer periods of prone position and the need for standardization of mechanical ventilation in both prone and supine groups according to predefined lung-protective strategies.

Acute respiratory distress syndrome

502

All these studies have shown that patient selection when studying patients with ARDS caused by primary pulmonary disease or secondary to extrapulmonary conditions has an impact on outcome. Several investigators have suggested that, based on the etiology of ARDS, there may be two separate syndromes (70–73). Differences in respiratory mechanics and response to mechanical ventilation and PEEP are consistent with the presence of consolidation in the lung parenchyma as opposed to edema and alveolar collapse (34, 72). Furthermore, patients with ARDS as their main reason for mechanical ventilation may have a different disease that develops during the course of mechanical ventilation (74). The latter syndrome is associated with increased multiple organ failure and increased mortality (74). Therefore, due to the heterogeity of this syndrome, both precipitating factors and onset of presentation should be considered when designing clinical trials in ARDS. In order to improve long-term outcomes and survival in ARDS, it may be necessary to use a combination of several adjunctive therapies. Papazian et al. (75) evaluated the hemodynamic and respiratory effect of the combination of inhaled nitric oxide and prone position in patients with ARDS. These investigators studied supine and prone position with and without nitric oxide. Inhaled nitric oxide resulted in an increase in ratio that was comparable to the changes seen with prone position alone. The association ratio of nitric oxide and prone position resulted in a further improvement in when compared with baseline, or prone position alone, and/or supine position with nitric oxide. Analysis of variance showed a significant additive effect of nitric oxide to prone ratio (p<0.0001) and shunt fraction (p<0.01). This study position in both showed that there is an additive effect in oxygenation using these two therapies. This study suggested a decrease in mortality when both therapies are used, but it is important to point out that this was not the main endpoint, and it was not statistically powered to show a difference or decreased mortality. These observations have been confirmed by Johannigman et al. (76) in 16 patients with ARDS who were ventilated in the supine position with nitric oxide (1 ppm) or prone position with nitric oxide. The increased by 14% with nitric oxide in 62% of patients (10/16 patients) in the supine position and by 33% in 87% of the patients (14/16) in the prone position. The combination of nitric oxide and prone position resulted in an improvement in 94% of of 59%. The investigators also noted patients (15/16) with a mean increase in that there was a significant reduction in pulmonary vascular resistance during the use of nitric oxide in both the supine and the prone position. There were no significant hemodynamic effects of either therapy. Thus, we can conclude that there may be synergistic effects of these two therapies. Other combinations of therapies used as reported by Varkul et al. (77) include high-frequency oscillatory ventilation, prone positioning, and inhaled nitric oxide in patients with severe hypoxemia. Because patients with ARDS have many complex and dynamic physiological derangements, it may be necessary to consider a combination of multiple therapies for patients with severe disease or persistent hypoxemia.

Prone position in the acute respiratory distress syndrome

503

VII. Conclusion Clinical studies have clearly demonstrated that in patients with ALI/ARDS, prone position results in significant and clinically relevant improvement in oxygenation in most patients (>75%) and that these changes occur quickly. Improvement can persist in patients when they are returned to the supine position. Also, the beneficial response is not limited to patients who are turned early in the course of the disease, although the data suggest that early use may be more beneficial. Patients who fail to respond initially may improve during subsequent attempts of turning. The incidence of complications associated with the use of prone ventilation is small. Animal research and clinical experience suggest that prone ventilation may protect the lung from the potential detrimental effects of mechanical ventilation. One major clinical trial reported no decrease in mortality in ALI/ARDS with prone ventilation (69), but more clinical trials are needed.

References 1. Ashbaugh DG, Bigelow DB, Petty TL, et al. Acute respiratory distress in adults. Lancet 1967; ii:319–323. 2. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 3. Piehl MA, Brown RS. Use of extreme position changes in acute respiratory failure. Crit Care Med 1976; 4:13–14. 4. Douglas WW, Render K, Beynen FM, et al. Improved oxygenation in patients with acute respiratory failure: the prone position. Am Rev Respir Dis 1977; 115:559–566. 5. Bryan AC. Comments of a devil’s advocate. Am Rev Respir Dis 1974; 110:43. 6. Gattinoni L, Pesenti A, Torresin A, et al. Adult respiratory distress syndrome profiles by computed tomography. J Thorac Imag 1986; 1:25–30. 7. Rinaldo JE, Rogers RM. Adult respiratory distress syndrome: changing concepts of lung injury and repair. N Engl J Med 1982; 306:900–909. 8. Gattinoni L, Mascheroni D, Torresin A, et al. Morphological response to positive end-expiratory pressure in acute respiratory failure: computerized tomography study. Intensive Care Med 1986; 56:1091–1130. 9. Rommelscheim K, Lakner K, Westhofen P, et al. Das respiratorische Distress-Syndrom des Erwachsenen (ARDS) im Computer-tomogramn. Anaesth Intensivther Notfallmed 1983; 18:59– 644. 10. Maunder RJ, Shuman WP, McHugh JW, et al. Preservation of normal lung region in adult respiratory distress syndrome: analysis by computed tomography. JAMA 1986; 255:2463–2465. 11. Gattinoni L, Pesenti A. ARDS: the dishomogeneous lung: facts and hypothesis. Intensive Crit Care Digest 1987; 6:1–4. 12. Gattinoni L, Pesenti A, Avalli L, et al. Pressure volume curve of total respiratory system in acute respiratory failure. Am Rev Respir Dis 1987; 36:730–736. 13. Jones R, Reid L, Zapol WM, et al. Pulmonary vascular pathology: human and experimental studies. In: Zapol WM, Falke KJ, eds. Acute Respiratory Failure. New York: Marcel Dekker, 1985:23–160.

Acute respiratory distress syndrome

504

14. Brismar B, Hedenstierna G, Lundquist H, et al. Pulmonary densities during anesthesia with muscular relaxation: a proposal of atelectasis. Anesthesiology 1985; 62:422–428. 15. Gattinoni L, Pelosi P, Pesenti A, et al. CT scan in ARDS: clinical and physiopathological insights. Acta Anesth Scand 1991; 95:87–896. 16. Jones T, Jones HA, Rhodes CG, et al. Distribution of extravascular fluid volumes in isolated perfused lungs measured with H2O. J Clin Invest 1976; 57:706–713. 17. Wollmer P, Rhodes CG, Deanfield J, et al. Regional extravascular density of the lung in patients with acute pulmonary edema. J Appl Physiol 1970; 8:204–229. 18. Pelosi, D’Andrea L, Vitale G, et al. Vertical gradient of regional lung inflation in adult respiratory distress syndrome. Am J Respir Crit Care Med 1994; 149:8–13. 19. West JB, Dollery CT. Distribution of blood flow and ventilation-perfusion ratio in the lung, measured with radioactive CO2. J Appl Physiol 1960; 15: 405–410. 20. Ball WC, Stewart PB, Newsham LG, et al. Regional pulmonary function studied with xenon. J Clin Invest 1962; 41:519–531. 21. Bryan AC, Bentivoglio LG, Beerel F, et al. Factors affecting regional distribution of ventilation and perfusion in the lung. J Appl Physiol 1964; 19:395–402. 22. Hughes JMB, Glazier JB, Maloney JE, et al. Effects of extra-alveolar vessels on distribution of blood flow in the dog lung. J Appl Physiol 1968; 25:701–712. 23. Prefaut C, Engel LA. Vertical distribution of perfusion and inspired gas in supine man. Respir Physiol 1981; 43:209–219. 24. Agostoni E, Hyatt RE. Static behavior of the respiratory system. In: Macklem PT, Mead J, eds. Handbook of Physiology. The Respiratory System. Vol. 3. Mechanics of Breathing. Bethesda, MD: American Physiological Society, 1986:113–130. 25. Milic-Emili J, Henderson JAM, Dolovich MB, et al. Regional distribution of inspired gas in the lung. J Appl Physiol 1966; 21:749–759. 26. Amis TC, Jones HA, Hughes JMB. Effect of posture on inter-regional distribution of pulmonary ventilation in man. Respir Physiol 1984; 56:145–167. 27. West JB, Dollery CT, Naimark A. Distribution of blood flow in isolated lung: relation to vascular and alveolar pressures. J Appl Physiol 1964; 19:713–724. 28. Reed JH, Wood EH. Effect of body position on vertical distribution of pulmonary blood flow. J Appl Physiol 1979; 28:303–311. 29. Orphanidou D, Hughes JMB, Myers MJ, et al. Tomography of regional ventilation and perfusion using krypton 81m in normal subjects and asthmatic patients. Thorax 1986; 41:542– 551. 30. Wiener CM, Kirk W, Albert RL. Prone position reverses gravitational distribution of perfusion in dog lungs with oleic acid-induced injury. J Appl Physiol 1990; 68:1386–1392. 31. Beck KC, Lai-Fook SJ. Pulmonary blood flow vs. gas volume at various perfusion pressures in rabbit lung. J Appl Physiol 1985; 58:2004–2010. 32. Albert RK, Leasa D, Sanderson M, et al. The prone position improves arterial oxygenation and reduces shunt in oleic acid-induced acute lung injury. Am Rev Respir Dis 1987; 135:628–635. 33. Lamm WJE, Graham MM, Albert RK. Mechanism by which the prone position improves oxygenation in acute lung injury. Am J Respir Crit Care Med 1994; 150:184–193. 34. Gattinoni LL, D’Andrea P, Pelos P, et al. Regional effects and mechanisms of positive endexpiratory pressure in early adult respiratory distress syndrome. JAMA 1993; 269:2122–2127. 35. Froese AB, Bryan AC. Effects of anesthesia and paralysis on diaphragmatic mechanics in man. Anesthesiology 1974; 41:242–255. 36. Brookhart JM, Boyd TE. Local differences in intrathoracic pressure and their relation to cardiac filling pressure in the dog. Am J Physiol 1947; 148:434–444. 37. Rutishauser WJN, Banchero AG, Tsakiris AC, et al. Pleural pressures at dorsal and ventral sites in supine and prone body positions. J Appl Physiol 1966; 21: 1500–1510. 38. Wood EH. Some effects of gravitational and inertial forces on the cardiopulmonary system. Aerospace Med 1967; 38:225–233.

Prone position in the acute respiratory distress syndrome

505

39. Hyatt RE, Bar-Yishay E, Abel MD. Influence on the heart on the vertical gradient of transpulmonary pressure in dogs. J Appl Physiol 1985; 58:52–57. 40. Scharf SM, Caldini P, Ingram RH. Cardiovascular effects of increasing airway pressure in the dog. Am J Physiol 1977; 232:H35–H43. 41. Culver BH, Marini JJ, Butler J. Lung volume and pleural pressure effects on ventricular function. J Appl Physiol 1981; 50:630–635. 42. Bosman AR, Gde J, Lee G. The effects of cardiac action upon lung gas volume. Clin Sci 1965; 28:311–324. 43. Cortese DA, Rodarte JR, Rehder K, et al. Effect of posture on the single-breath oxygen test in normal subjects. J Appl Physiol 1976; 41:474–479. 44. Wiener CM, McKenna WJ, Myers MJ, et al. Left lower lobe ventilation is reduced in patients with cardiomegaly in the supine but not in the prone position. Am Rev Respir Dis 1990; 141:150–155. 45. Ball WC, Wicks JD, Mettler FA. Prone-supine change in organ position: CT demonstration. AJR 1980; 135:815–820. 46. Hoffman EA. Effect of body orientation on regional lung expansion: a computed tomographic approach. J Appl Physiol 1985; 59:468–480. 47. Albert RK, Hubmayr RD. The prone position eliminates compression of the lungs by the heart. Am J Respir Crit Care Med 2000; 161:1660–1665. 48. Dreyfuss D, Basset G, Soler P, et al. Intermittent positive-pressure hyperventilation with high inflation pressures produces pulmonary microvascular injury in rats. Am Res Respir Dis 1985; 132:880–884. 49. Dreyfuss D, Saumon G. Role of tidal volume, FRC and end-expiratory volume in the development of pulmonary edema following mechanical ventilation. Am Rev Respir Dis 1993; 148:1194–1203. 50. Slutsky AS. Mechanical ventilation. American College of Chest Physicians’ Consensus Conference. Chest 1993; 104:1833–1859. 51. Webb H, Tierney D. Experimental pulmonary edema due to intermittent positive pressure ventilation with high inflation pressures: protection by positive end-expiratory pressure. Am Rev Respir Dis 1974; 110:556–565. 52. Muscedere JG, Mullen JMB, Gan K, et al. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med 1994; 149:1327–1334. 53. Kolobow T, Moretti MP, Fumagalli R, et al. Severe impairment in lung function induced by high peak airway pressure during mechanical ventilation: an experimental study. Am Rev Respir Dis 1987; 135:312–315. 54. Amato MB, Barbas CS, Medeiros DM, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338:347–354. 55. Sinclair SE, Souders J, Hlastala MP. Severity and distribution of ventilator-induced lung injury (VILI) is altered by PEEP, prone position, and respiratory frequency in normal rabbits. Am J Respir Dis 1998; 157:A107. 56. Broccard A, Shapiro RS, Schmitz LL, et al. Prone position attenuates and redistributes ventilator-induced lung injury in dogs. Crit Care Med 2000; 28: 295–303. 57. Albert RK, Lamm WJE. Ventilator-induced augmentation of acute lung injury is limited by prone ventilation. Am J Respir Crit Care Med 1998; 157: A460. 58. Ray JF III, Yost L, Moallem S, et al. Immobility, hypoxemia, and pulmonary arteriovenous shunting. Arch Surg 1974; 109:537. 59. Langer M, Mascheroni D, Marcolin R, et al. The prone position in ARDS patients. A clinical study. Chest 1988; 94:103–107. 60. Fridrich P, Krafft P, Hochlenthner H, Mauritz W. Effects of long-term prone positioning in patients with trauma-induced adult respiratory distress syndrome. Anesth Analg 1996; 83:1206– 1211.

Acute respiratory distress syndrome

506

61. Chatte G, Sab JM, Dubois JM, et al. Prone position in mechanically ventilated patients with severe acute respiratory failure. Am J Respir Crit Care Med 1997; 1555:473–478. 62. Jolliet P, Bulpa P, Chevrolet J-C. Effects of the prone position on gas exchange and hemodynamics in severe acute respiratory distress syndrome. Crit Care Med 1998; 26:1977– 1985. 63. Stocker R, Neff T, Stein S, et al. Prone positioning and low-volume pressure-limited ventilation improve survival in patients with severe ARDS. Chest 1997; 111:1008–1017. 64. Nakos G, Tsangaris I, Kostamis E, et al. Effect of the prone position on patients with hydrostatic pulmonary edema compared with patients with acute respiratory distress syndrome and pulmonary fibrosis. Am J Respir Crit Care Med 2000; 161:360–368. 65. Benumof JL. Mechanism of decreased blood flow to atelectatic lung. J Appl Physiol 1979; 46:1047–1048. 66. Gattinoni L, Tognoni G, Pesenti A, et al. Effect of prone positioning on the survival of patients with acute respiratory failure. N Engl J Med 2001; 345: 568–573. 67. Thorburn K, Kerr SJ, Raines PB. Prone positioning of patients with acute respiratory failure (correspondence). N Engl J Med 2002; 346(4):295–297. 68. Malhotra A, Ayas N, Kacmarek R. Prone positioning of patients with acute respiratory failure (correspondence). N Engl J Med 2002; 346(4):295–297. 69. Slutsky AS. Prone positioning of patients with acute respiratory failure (correspondence). N Engl J Med 2002; 346(4):295–297. 70. Gattinoni L, Bombino M, Pelosi P, et al. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA 1994; 271: 1772–1779. 71. Gattinoni L, Pelosi P, Suter PM, et al. Acute respiratory distress syndrome caused by pulmonary and extrapulmonary disease. Am J Respir Crit Care Med 1998; 158:3–11. 72. Rocker GM. Acute respiratory distress syndrome: different syndromes, different therapies? Crit Care Med 2001; 29:210–211. 73. Croce MA, Fabian TC, Davis KA, et al. Early and late acute respiratory distress syndrome: two distinct clinical entities. J Trauma, Injury, Infection, Crit Care 1999; 46:361–368. 74. Anzueto A, Esteban A, Alia, et al. ARDS before and after start of mechanical ventilation. Am J Respir Crit Care Med 2000; 161:A-382. 75. Papazian L, Bregeon F, Gaillat F, et al. Respective and combined effects of prone position and inhaled nitric oxide in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 157:580–585. 76. Johannigman JA, Davis K Jr, Miller SL, et al. Prone positioning and inhaled nitric oxide: synergistic therapies for acute respiratory distress syndrome. J Trauma, Injury, Infection, Crit Care 2001; 50:589–596. 77. Varkul MD, Stewart TE, Lapinsky SE, et al. Successful use of combined high-frequency oscillatory ventilation, inhaled nitric oxide, and prone positioning in the acute respiratory distress syndrome. Anesthesiology 2001; 95:797–799.

21 Mechanical Ventilation in the Acute Respiratory Distress Syndrome ROY G.BROWER Johns Hopkins University Baltimore, Maryland, U.S.A. LAURENT J.BROCHARD Université Paris 12, INSERM U 492 and Henri Mondor Hospital Créteil, France

I. Introduction In patients with acute respiratory distress syndrome (ARDS), inflammation of the pulmonary circulation causes increased vascular permeability. Protein-rich fluid leaks from the blood into the pulmonary interstitium and alveolar airspaces (1). Surfactant production by type II pneumocytes is reduced, and existing surfactant is inactivated by extravasated plasma proteins (Chap. XX). Increased weight of the edematous lungs causes compression atelectasis, mainly in dependent areas. Surface tension rises at air/fluid interfaces, causing atelectasis of substantial portions of the pulmonary parenchyma (2), (Chap. XX). Intrapulmonary shunt caused by noncardiogenic edema and diffuse atelectasis may exceed 40% (3), causing life-threatening impairments of arterial oxygenation (hypoxemic respiratory failure). The reduction in aerated lung volume and increased surface tension cause lung compliance to decrease. Airway resistance may also be increased by edema of bronchial mucosa and excessive airway secretions (4). Lower lung compliance and higher airway resistance increase the work of breathing per unit of ventilation. Some relatively unaffected lung units become overventilated because ventilation is preferentially distributed to the aerated zones, whereas the more diseased lung units are less accessible for tidal ventilation. Other lung units have excessive ventilation relative to perfusion because small blood vessels become occluded by inflammation and microthrombi or compressed by high airway pressures. These “high VQ” units contribute further to increased physiological deadspace, which may exceed 60% of total ventilation (3, 5). Increased metabolic demands from acute illness further increase requirements for ventilation. In many patients, the increased work of breathing and requirements for ventilation cannot be sustained with spontaneous respiratory efforts. Without mechanical support for respiration, many ARDS patients would die within hours to days from acute hypoxemic and hypercarbic respiratory failure. With mechanical ventilation, survival can usually be ensured for days to weeks. This provides time to administer therapies specific to the cause of ARDS, such as antibiotics in patients with

Acute respiratory distress syndrome

508

pneumonia or sepsis, for the host’s immune system to fight infections, and for natural healing processes to occur. Providing time for healing without causing complications or further damage to the lungs is the primary objective of mechanical ventilation.

II. Traditional Approaches to Mechanical Ventilation in ARDS Recognition of the syndrome of “acute respiratory distress in adults” (6) occurred at an important time in the development of mechanical ventilation as therapy for respiratory failure of various causes. In the eighteenth and nineteenth centuries, techniques were developed in Europe for resuscitation of victims of apparent drowning by applying intermittent positive pressure to the airway (7, 8). This approach was subsequently discouraged, however, because of concerns for barotrauma (9). “Negative pressure ventilation” was used frequently in the mid-twentieth century when large numbers of patients experienced respiratory failure from polio-induced neuromuscular disease (10– 12). Positive pressure ventilation was reintroduced during the first half of the twentieth century, primarily for support of patients requiring general anesthesia for surgery, especially for thoracic procedures. When the earliest case series of ARDS patients were reported in the late 1960s (6, 13), positive pressure ventilation was used with increasing frequency for nonsurgical patients with acute respiratory failure from various causes, including obstructive airways disease and severe pneumonia. Because these techniques were developed primarily for patients requiring general anesthesia, initial approaches to mechanical ventilation in ARDS resembled those used for years in anesthesiology practices. There was no clearly recognized “standard of care” for mechanical ventilation of ARDS patients. However, many clinicians adopted approaches similar to the approach outlined in Table 1, which may be considered “traditional.” A. Mode and Tidal Volume A volume-cycled mode was used because pressure-cycled modes frequently resulted in relatively small or highly variable tidal volumes in patients with severe parenchymal lung disease. Control of tidal volume was an important part of the traditional approach, in which an important clinical goal was to achieve near-normal and pH. Tidal volumes of 10–15 mL/kg were recommended (14, 15), and volumes up to 20 or 25 mL/kg were frequently used for the most severe cases (16). These were several times the size of normal resting tidal volumes (17). Such generous tidal volumes were useful for achieving the and pH goals, especially in ARDS patients with high deadspace fractions. Moreover, earlier studies demonstrated that shunt from atelectasis could be prevented with generous tidal volumes in patients receiving general anesthesia for surgery (18). Studies in animal models of acute lung injury confirmed that shunt could be reduced if generous tidal volumes were used (19, 20). However, dangerous consequences of this traditional approach to setting tidal volume and minute ventilation have been demonstrated in numerous animal models. Moreover, the traditional approach to setting tidal volume and minute ventilation has been challenged in recent clinical trials, as detailed later in this chapter.

Mechanical ventilation in the acute respiratory distress syndrome

509

B. Support for Arterial Oxygenation Arterial oxygenation can be supported in patients with hypoxemia and acute respiratory failure by increasing the fraction of inspired oxygen

This

Table 1 Traditional Approach to Mechanical Ventilation in ARDS Endotracheal intubation Intermittent positive pressure applied to the airway Ventilator mode:

Volume-cycled (controlled, assist/control, intermittent mandatory ventilation)

Tidal volume:

10–15 mL/kg measured or dry body weight

Respiratory rate:

~10–25/min. Adjust to achieve near-normal 0.5–1.0. Adjust to achieve normal or supranormal

PEEP: I:E:

~5–10 cmH2O. Higher if

needed to achieve

goal.

~1:1–1:3

=fraction of inspired oxygen; I: E=ratio of the duration of inspiration to the duration of expiration; PEEP=positive end-expiratory pressure.

can correct the modest degree of hypoxemia caused by lung units with low ventilationperfusion ratios. However, increases in

are not effective enough when shunt

to high levels may fractions are high, as in ARDS. Moreover, sustained increases in cause oxidant-induced acute lung injury (21, 25). There is very little information from clinical studies to define safe limits for Relying on information from animal experiments and limited uncontrolled observations in humans, many clinicians’ practices as safe (26, 27). accept prolonged ventilation with Positive end-expiratory pressure (PEEP) reduces intrapulmonary shunt and improves arterial oxygenation. These beneficial effects are related to the increase in functional residual capacity that occurs (28) as PEEP reverses or prevents atelectasis of some unstable lung units (29, 30) and redistributes fluid from alveolar to interstitial compartments (31). However, adverse effects of PEEP include circulatory depression (32–35). Although decreased cardiac output from PEEP could also contribute to improvement in oxygenation (36), the net decrease in oxygen delivery to systemic tissues could cause dysfunction of other organs and systems. One approach to optimizing PEEP is to empirically identify the level associated with maximal oxygen delivery (37). Many clinicians are attracted to this approach (26), but it could require pulmonary arterial catheterization for measurements of cardiac output. In their initial description of this approach, Suter et al. suggested that oxygen delivery would be maximal when respiratory system compliance was maximal. If so, then

Acute respiratory distress syndrome

510

measurements of respiratory system compliance (tidal volume divided by inspiratory plateau pressure minus PEEP) could be used as a surrogate for measurements of oxygen delivery (37). Compliance measurements, however, are influenced by several factors, reducing the reliability of this approach. These factors include the nonlinear nature of the pressure-volume relationship and differences in immediate versus time-dependent recruitment (38, 39). Another shortcoming of this approach is that other effects of PEEP, such as higher risk of barotrauma or lower risk of oxygen toxicity, were not considered in the method for identifying “best PEEP.” Clinicians must balance potentially beneficial when designing strategies for and deleterious effects of both PEEP and increased achieving acceptable arterial oxygenation. Unfortunately, there is very little information from clinical studies to guide clinicians in this difficult task. In an early case series, arterial oxygenation goals could be achieved in most patients with modest PEEP levels (7–10 cmH2O) and with partial pressures of inspired oxygen ≤500 mmHg (equivalent to at sea level) (13). This approach to supporting arterial oxygenation was adopted by many. Higher PEEP levels were generally not used unless high levels of were necessary to achieve acceptable arterial oxygenation (26, 27). The physiological goal of PEEP is to open (recruit) alveoli that were previously closed or filled and to keep these alveoli open during subsequent ventilation. When PEEP is applied, some alveoli may open quickly, and some previously aerated alveoli may expand further. When the lungs are ventilated for some time at a given end-expiratory pressure, additional recruitment may occur gradually over several breaths (40). This additional recruitment can be measured by constructing a series of pressure-volume curves on a common volume axis (39, 41–45). This technique has shown that the shape and position of pressure-volume curves can provide qualitative information about levels of recruitment but cannot provide clinicians certainty about optimal levels of PEEP (44, 46). Arterial oxygenation remains the best, though imperfect, clinically available tool to estimate the effects of PEEP on alveolar recruitment (41, 44).

III. Complications of Mechanical Ventilation A. Ventilation-Associated Lung Injury Several observations and studies strongly suggested that traditional approaches to mechanical ventilation may cause additional acute lung injury in patients with ARDS. Appreciation for the potential for ventilation-associated lung injury (VALI) was heightened in the 1980s, when some of the first computerized tomographic images of ARDS lungs were viewed. Some ARDS lung regions have high density and radiographic textures consistent with dense consolidation, edema, and atelectasis, while other regions have radiographic characteristics of normal lung (2, 47). This contrasted with the typical appearance of “diffuse infiltrates” on standard chest radiographs. Pressure-volume relationships of the lung in ARDS patients were normal when adjusted for the volume of aerated tissue (48). These findings complemented earlier observations on histological sections of lung tissue from ARDS patients that showed some areas that appeared to be normal in the same microscopic field as other areas in which alveoli were atelectatic,

Mechanical ventilation in the acute respiratory distress syndrome

511

hemorrhagic, or filled with edema fluid, cellular debris, and hyaline membranes (Chap. XX) (49). The recognition of the heterogeneous involvement of the lung parenchyma in ARDS suggested two potential mechanisms by which ventilation could cause acute lung injury: (1) injurious mechanical forces could occur if excessive gas volume were forced into some lung regions (high-volume/high-pressure injury), and (2) injurious mechanical forces could occur from ventilation with atelectasis or alveolar flooding at end-expiration (low-volume/ low-pressure injury). VALI from High-Volume/High-Pressure Ventilation The radiographic, physiological, and histological observations summarized above suggested that the aerated lung in ARDS was normal except that its volume was much smaller than normal (2, 48). Delivery of tidal volumes of 10–15 mL/kg required inspiratory airway pressures that were considerably higher than those that occurred while ventilating patients with normal lungs. This suggested that the aerated lung of some ARDS patients was subjected to high levels of distention or stretch during inspiration. Numerous studies in laboratory animals were conducted to understand the effects of high inspiratory volumes on normal lung (reviewed in Chap. XX). High-volume/high-pressure ventilation caused increased vascular permeability, endo- and epithelial disruption, neutrophil infiltration, alveolar hemorrhage and atelectasis, hyaline membranes, infiltrates on chest radiographs, and hypoxemia from elevated intrapulmonary shunt tissue (50–54). This suggested that the traditional generous tidal volume approach to mechanical ventilation could be a second cause of acute lung injury in some patients whose lung disease was caused initially by sepsis, trauma, or pneumonia. Additional studies suggested that there could be synergy between the injurious effects of highvolume/high-pressure ventilation and of other causes of acute lung injury (55). Perhaps some patients could have recovered from the initial cause of ARDS, but acute lung injury was exacerbated or perpetuated by high-volume/high-pressure ventilation. Distention of the aerated lung parenchyma during inspiration can be reduced by ventilating with smaller tidal volumes and lower inspiratory pressures (pressure-andvolume limited ventilation). This approach, however, requires a different set of priorities with respect to clinical objectives. The pressure-and-volume limited approach gives lower priority to the traditional goals of near-normal and acid-base balance. Moreover, shunt from atelectasis may increase when tidal volumes and inspiratory and PEEP may increase. Many pressures are decreased (18, 20); requirements for patients are more dyspneic while receiving smaller tidal volumes, especially when rises and pH falls. Requirements for sedation may increase with this strategy. VALI from Low-Volume/Low-Pressure Ventilation In several experimental models, VALI occurred when lungs were ventilated with low volumes and airway pressures during expiration (50, 56–59). In some studies, lowvolume/low-pressure injury was apparent macroscopically (Fig. 1) and on histological examination (50, 58). Indirect evidence of low-volume/low-pressure injury included local

Acute respiratory distress syndrome

512

release of inflammatory (and anti-inflammatory) mediators (59). In some experiments, bacteria inoculated into the lungs passed into the systemic circulation when low PEEP levels were used during mechanical ventilation but not when higher PEEP was applied (60, 61).

Figure 1 Effects of mechanical ventilation on rat lungs. (Left) Lung of a normal rat that had not received mechanical ventilation. (Right) Lung of a normal rat that had received mechanical ventilation with a large tidal volume, with peak inspiratory pressure=45 cmH2O. Histological examination demonstrated hemorrhage, edema, and inflammation. (Middle) Lung of a normal rat that had received mechanical ventilation with a large tidal volume with peak inspiratory pressure=45 cmH2O and with positive end-expiratory pressure=10 cmH2O. Histological examination demonstrated near-normal lung tissue. (From Ref. 50.)

Mechanical ventilation in the acute respiratory distress syndrome

513

The mechanism of low-volume/low-pressure VALI is not clear. If there is atelectasis at end-expiration, some small bronchioles and alveoli may open and close with each breath. Mechanical forces associated with tidal opening and closing could be injurious (58). Surfactant function could deteriorate from repeated large changes in surface area in alveoli that open and close with each breath (62). There could be excessive stress at margins between aerated and atelectatic lung units (63). Many studies in experimental models (reviewed in Chap. XX) demonstrated that lowvolume/low-pressure VALI could be attenuated by applying some level of PEEP during mechanical ventilation (50, 57–59). This beneficial effect of PEEP could be from a reduction in cyclic opening and closing of alveoli (58, 64) or from a reduction in stress between aerated and atelectatic lung units (63). This explanation has been challenged in other models (65). More lung units are aerated, and the distribution of ventilation is more homogeneous when PEEP is applied in ARDS patients (29, 66). Thus, PEEP could reduce regional hyperinflation. On the other hand, higher PEEP levels could contribute to hyperinflation if tidal volumes are not reduced to limit peak volumes and pressures (67). As with all experimental models, the observations in the animal models of acute lung injury must be interpreted cautiously with respect to their applicability to the clinical experience. Some experiments were conducted using ex vivo lungs (47, 58, 59) in which the end-expiratory position of the lung is altered. Use of negative end-expiratory pressures may exaggerate effects of low-volume/low-pressure ventilation (68). The lung lavage model, while useful for studying surfactant deficiency, may not accurately represent clinical acute lung injury with respect to effects of PEEP (58). It is also difficult to interpret some studies of effects of PEEP on VALI in which tidal volume was held constant when different PEEP levels were applied. In these experimental conditions, higher PEEP raises volume and pressure at end-inspiration as well as end-expiration. Potentially beneficial effects of PEEP on recruitment may be obscured by potentially deleterious effects of ventilating at high volume/high pressure. Another important interaction between ventilator settings occurs between the size of the tidal volume and the level of PEEP. The amount of aerated lung volume at end-expiration depends not only on the PEEP level but also on the end-inspiratory lung volume and of the tidal volume used (46, 69). Despite the limitations of the studies in experimental animal models, there is substantial evidence that some level of PEEP can prevent low-volume/ low-pressure VALI. Unfortunately, the studies in experimental models do not provide enough information to guide clinicians who must select PEEP in individual patients. As discussed earlier, some investigators (70) have suggested that PEEP levels should be prescribed according to the shape and position of pressure-volume curves of the respiratory system (Fig. 2). In many patients, slope (compliance) of the lower portion of the inspiratory curve increases as volume and pressure rise, presumably reflecting recruitment of alveoli that are atelectatic at lower volumes and pressures. By setting PEEP ~2 cmH2O higher than the middle of this part of the curve (the “lower inflection point”), higher levels of recruitment can be achieved and, perhaps, low-volume/low-pressure VALI can be avoided. However, in some patients the increasing-slope portion of the respiratory system pressure-volume curve is attributable to characteristics of the chest wall, not the lung (71). Some studies also strongly suggest that lung recruitment continues as pressure and

Acute respiratory distress syndrome

514

volume rise to levels that are much higher than those at the lower inflection point (39, 72).

Figure 2 Relationships of airway pressure to changes in respiratory system volume in four patients with acute respiratory distress syndrome. (From Ref. 187.) Another limitation of the pressure-volume curve approach to setting PEEP is that higher levels of PEEP, while potentially useful to reduce low-volume/low-pressure VALI, may also have adverse effects. As reviewed earlier, higher PEEP may reduce cardiac output (33, 34, 36, 73). Higher PEEP may also increase end-inspiratory lung volumes and inspiratory pressures, increasing the risk of high-volume/high-pressure VALI and the risk

Mechanical ventilation in the acute respiratory distress syndrome

515

of barotrauma (67, 74). The pressure-volume curve approach to setting PEEP does not account for these potentially adverse effects, which may counteract the beneficial effects of the higher levels of PEEP that may be identified from pressure-volume curves. Clinical benefits of relatively high levels of PEEP in ARDS patients were strongly suggested by the results of the study by Amato et al. (75), described in more detail later in this chapter. The NIH ARDS Network trial of volume-and-pressure limited ventilation did not use higher PEEP in the lung-protective group (76). However, respiratory rates were more rapid in the volume-and-pressure limited group of the NIH trial, which could have caused air-trapping with auto-PEEP. Recent studies in which strategies similar to the NIH volume-and-pressure limited ventilation protocol were applied demonstrated end-expiratory alveolar pressures that were 3–6 cmH2O above the PEEP settings when respiratory rates were ~30 breaths/min (77, 78). Thus, some patients in the NIH trial volume-and-pressure limited group could have received some lung-protective effects from auto-PEEP-induced lung recruitment. B. Barotrauma Pneumothorax and pneumomediastinum are common forms of barotrauma (Fig. 3). Other manifestations include pneumatoceles, subcutaneous emphysema, and pulmonary interstitial air. These air-leak injuries are classically attributed to mechanical stresses in small bronchioles and alveoli during mechanical ventilation at high airway pressures and volumes (79). In patients with ARDS, barotrauma has been reported to occur in 0–76% of patients (74, 80–85). Barotrauma is associated with increased risk of death and a greater duration of mechanical ventilation (83, 86). Recent modifications to traditional mechanical ventilation strategies were designed to reduce high-volume/high-pressure VALI in ARDS by reducing mechanical stresses in the lung parenchyma during inspiration. In a randomized trial by Amato et al., the incidence of barotrauma was markedly lower in the study group that received volume-and-pressure limited ventilation (75). However, there was no difference in the incidence of barotrauma in four other studies in which patients were randomized to either traditional or volumeand-pressure limited ventilation strategies (76, 87–89). Inspiratory pressures in the traditional ventilation group were higher in the study of Amato et al. than in the other four studies (~37 vs. ~29–33 cmH2O). This suggests that a threshold level of inspiratory stress must be exceeded to contribute to the incidence of barotrauma. Another explanation for the difference between these studies is that the lung-protective ventilation group in the study by Amato et al. received higher levels of PEEP in addition to

Acute respiratory distress syndrome

516

Figure 3 Severe barotrauma in a patient with acute respiratory distress syndrome receiving mechanical ventilation. Bilateral pneumothoraces are present. The left hemi-diaphragm is depressed and the mediastinum is shifted to the right from elevated pressure in the left pleural space. There is pneumomediastinum and extensive subcutaneous air. lower tidal volumes with lower inspiratory pressures. Thus, the “driving pressure” (tidal swing between plateau pressure and PEEP), which was much lower in the lung-protective ventilation group in the study by Amato et al., may be an important determinant of the incidence of barotrauma. A recent analysis of the NIH ARDS Network study of volume-and-pressure limited ventilation suggested that PEEP (but not inspiratory pressures) was a significant risk were controlled by protocol factor for barotrauma (74). However, both PEEP and rules to achieve acceptable levels of arterial oxygenation. Because higher requirements are markers of severe lung injury, the apparent relationship between for PEEP and PEEP and barotrauma could simply reflect a relationship between lung injury severity

Mechanical ventilation in the acute respiratory distress syndrome

517

and barotrauma. In a different large study in ARDS patients (90), PEEP and were not controlled by protocol rules. Therefore, some variations in PEEP settings in this study probably reflected physicians’ practice styles or biases rather than severity of lung injury. Analysis of the database from this study did not demonstrate a relationship between ventilator settings and the frequency of air leaks (84). Similar findings were reported in another study that included an analysis of plateau pressures and driving pressures (85). In a literature review, the incidence of barotrauma was apparently higher in patients with inspiratory plateau pressures greater than 35 cmH2O and in whom respiratory system compliance was below 30 mL/cmH2O (85). In studies in which inspiratory airway pressures were lower, the incidence of barotrauma was lower (~10%), and there was no apparent relationship of barotrauma to ventilator parameters. This suggests that air leaks are frequently related to the acute lung injury process or, perhaps, to inadvertent lung injuries during invasive procedures. If so, then the term “barotrauma” may be a misnomer when it is applied to many cases of air leak. It is very likely, however, that ventilation with high pressures and volumes, as used frequently in ARDS patients from the 1960s to the mid-1980s (Table 2), caused far more frequent episodes of barotrauma (91). C. Circulatory Complications of Mechanical Ventilation Positive pressure ventilation frequently causes cardiac output to decrease (32, 73, 92). Since cardiac output is an important determinant of oxygen delivery to systemic tissues, beneficial effects of mechanical ventilation on arterial oxygenation may be counteracted by the adverse circulatory effects.

Table 2 Examples of Tidal Volume Settings Used Over 20 Years in ARDS Tidal volume (mL/kg)

Ref.

10–15

178

8–14

179

11–13

180

10–15

15

9–24

181

10–16

182

12–18

183

14–21

184

15

37

15

185

12–20

34

12–15

186

Acute respiratory distress syndrome

518

In some patients, substantial reductions in cardiac output could cause dysfunction or failure in other organs and systems. Some ARDS patients experience acute cor pulmonale, as demonstrated in studies in which echocardiographic images were used to assess left- and right-cardiac function (33, 34, 93, 94). Several factors contribute to elevated right ventricular impedance in ARDS patients. Pulmonary vascular resistance may be elevated from hypoxic vasoconstriction, vasoactive mediators such as thromboxane, partial vascular obstruction, and from respiratory and metabolic acidosis. Application of PEEP may compress alveolar vessels, raising pulmonary vascular resistance. Alveolar pressure rises further during inspiration, causing increased right ventricular impedance and decreased right ventricular stroke volume (93, 94). This effect of ventilation was especially prominent when relatively high PEEP levels were applied (94), as in some recent studies (75, 96). One of the manifestations of acute cor pulmonale is a leftward shift of the interventricular septum (33, 34), which could limit left ventricular filling and stroke volume. Another manifestation of acute cor pulmonale is elevated right atrial pressure, causing venous return and cardiac output to fall. Acute cor pulmonale may occur frequently in ARDS. A retrospective analysis suggests that a substantial fraction of all episodes of cor pulmonale in ARDS patients was caused by excessive lung volumes and airway pressures during ventilation (97). Right atrial pressure may also rise from the effects of positive pressure ventilation on pleural pressure. Pleural pressure is normally approximately −5 cmH2O at end-expiration and decreases by several cmH2O with spontaneous inspiratory efforts. When mechanical ventilation with PEEP is used, pleural pressure at end-expiration may increase by several cmH2O. Moreover, if there are no spontaneous inspiratory efforts, pleural pressure may rise further by several cmH2O during positive pressure inspiration. Because pleural pressure surrounds the heart, right atrial pressure also rises from these effects of PEEP and positive pressure inspiration. This further impedes venous return, causing cardiac output to fall. E. Oxygen Toxicity Traditional approaches to mechanical ventilation in ARDS use moderate to high

to

support arterial oxygenation (13, 26). However, there may be toxic effects of high (23). Increased pulmonary vascular permeability, increased lung water, and pleural effusions occurred when normal rats breathed spontaneously for 60 hours with (22). Neutrophil chemoattractant and inflammatory cell concentrations in lung lavage fluids increased within 48 hours of breathing hyperoxic gas mixtures. Inflammation and edema were present on histological examination of lung tissue within 72 hours, and most rats died within 96 hours while breathing (98). Hyperoxia also caused lung inflammation, increased vascular permeability, and increased lung water in dogs (21) and baboons (99). In normal human volunteers, spontaneous breathing at for 6–17 hours caused tracheitis, retrosternal discomfort, decreased mucociliary function, and increased release of inflammatory mediators by alveolar macrophages (24, 100, 101). There were significant impairments in

Mechanical ventilation in the acute respiratory distress syndrome

519

vital capacity, diffusing capacity, and alveolar-arterial oxygen gradients in normal humans who breathed for up to 3 days (25). These and other studies strongly suggest that hyperoxia can cause or contribute to acute lung injury. There may also be synergistic effects of hyperoxia, mechanical ventilation, and acute respiratory disease (102). However, injurious effects of hyperoxia may be attenuated by some clinical circumstances that are common in ARDS patients. For example, when rats were exposed to moderate levels of hyperoxia, concentrations of inducible antioxidant enzymes was tolerated much better than in rats that increased. Subsequent exposure to high were not conditioned first to moderate hyperoxia (103). Exposure to sublethal doses of endotoxin protected the sheep and rats from lung injury and death from hyperoxia. In humans with ARDS, antioxidant activity of bronchoalveolar fluid is increased (104). This could have occurred from previous exposures to moderate endotoxin, or the antioxidant properties of plasma proteins that leak into the alveolar space.

IV. Modern Approaches to Lung-Protective Ventilation Using Conventional Mechanical Ventilators The studies in experimental animal models reviewed in Chapter XX strongly suggested that traditional mechanical ventilation approaches could cause VALI and that modified approaches could protect the lungs from VALI. However, the modifications could also have adverse effects. Therefore, clinical studies were needed to assess the effects of lungprotective ventilation strategies on important clinical outcomes such as mortality and time on mechanical ventilation. Several clinical studies were conducted between 1985 and 2002 (Table 3). A. Clinical Studies of Volume-and-Pressure Limited Ventilation In a retrospective review, Hickling et al. reported physiological and outcome data on 50 severe ARDS patients who received a volume-and-pressure limited mechanical ventilation approach designed to reduce high-volume/high-pressure VALI (105). Tidal volume was decreased in steps until peak inspiratory

Table 3 Tidal Volumes Used in Clinical Trials of Lung-Protective Ventilation Strategies in ALI/ARDS Tidal volumes as reported Traditional a

~12

11.8

b c

10.3

Mortality (%)

Lower

Traditional

Lower

Ref.

b

71

38

75

b

40

31

76

c

38

47

87

≤6

6.2

7.1

Acute respiratory distress syndrome

10.8 10.2

b

520

7.2d

47

50

89

b

46

50

88

7.3

a

Tidal volumes expressed in mL/kg measured body weight. Tidal volumes expressed in mL/kg predicted body weight (PBW): Male PBW (kg)=50+2.3 [(height in inches)−60] Female PBW (kg)=45.5+2.3 [(height in inches)−60]. c Tidal volumes expressed in mL/kg dry body weight (measured weight minus estimated weight gain from water and salt retention). d Tidal volumes expressed in mL/kg ideal body weight (IBW): IBW=25×[(height in meters)2]. b

pressure was <30 cmH2O or until tidal volume was approximately 5 mL/kg. As expected, increased and pH decreased in many patients, but mortality was surprisingly low (16%). Favorable experiences were also reported in a subsequent prospectively identified cohort of severe ARDS patients who received a similar volume-and-pressure limited approach (106). Neither of these studies included a control group treated concurrently by the same clinicians using a traditional mechanical ventilation approach. Five clinical trials were subsequently conducted in which ARDS patients were randomized to either a lung-protective mechanical ventilation strategy or a strategy that resembled the traditional approach outlined in Table 1. Essential elements of these trials and the mortality rates for traditional and lung protective ventilation groups are summarized in Table 3. There were many similarities among these studies, but the important outcomes were quite different. Some methodological differences between the trials may account for the different outcomes. The lung-protective ventilation strategy in the study by Amato et al. (75) was designed to prevent VALI from both high-volume/high-pressure and low-volume/low-pressure ventilation. Tidal volumes and inspiratory pressures were decreased substantially from those used in the traditional ventilation group. Pressure-volume curves of the respiratory system were constructed on each patient in the lung-protective ventilation group, and initial PEEP was set to 2 cmH2O greater than the pressure at the midpoint of the portion of the curve with increasing slope. The mean PEEP in the lung-protective ventilation group during the first 36 hours was 16.4 cmH2O, which was 7.7 cmH2O higher than the mean PEEP in the traditional ventilation group. Recruitment maneuvers (continuous positive airway pressure of 40 cmH2O for 40 seconds) were conducted after disconnections from the ventilator circuit and other circumstances in which hypoxemia may have worsened from atelectasis. The higher PEEP and recruitment maneuvers were associated with substantial improvements in arterial oxygenation, allowing reductions in needed to maintain acceptable arterial oxygenation. The protective ventilation the approach was associated with striking improvements in 28-day survival, rates of weaning, and frequency of barotrauma. This was a relatively small study, with 24 and 29 patients in the traditional and lung-protective ventilation groups, respectively. Mortality in the traditional study group was higher than in several other recent studies, and several of these patients died within 5 days of randomization. This suggested that unrecognized imbalances between the study groups could have favored the lung-protective ventilation group. Moreover, it was not clear which aspects of the lung-protective ventilation strategy contributed to the beneficial effects.

Mechanical ventilation in the acute respiratory distress syndrome

521

In four of the randomized trials outlined in Table 2, the lung-protective ventilation strategy was designed only to reduce VALI from high-volume/ high-pressure ventilation. There was no attempt to separate the study groups with respect to end-expiratory pressure or volume. Mortality was significantly lower in one of the trials but not in the others (75). There are several possible explanations for this difference. Separation in Tidal Volumes Between the Study Groups Direct comparisons of tidal volumes between studies is impossible because of differences in definitions and protocol rules by which tidal volumes were set. In one of the studies, tidal volumes were set according to “dry body weight,” which was defined as measured body weight minus weight attributable to abnormal fluid retention (87). In another study, tidal volumes were set according to “ideal body weight,” which was calculated from height (89). In two of the trials, tidal volumes were set according to a lean predicted body weight calculated from gender and height (76, 88). Comparisons of tidal volumes in the studies are also difficult because only one of the studies required adjustments to the set tidal volumes to correct for volume lost from ventilator circuit tube expansion and gas compression during inspiration (76). This adjustment was not required in the other studies, but some of the ventilators used in these studies made automatic adjustments. A comparison of tidal volumes used in two of the trials, after adjustments for methodological differences, is shown in Table 4. It appears that the

Table 4 Tidal Volumes Used in Two Clinical Trials of Lung-Protective Ventilation Strategies in ALI/ARDS Adjusted for Methodological Differences Tidal volumes (mL/kg PBW) Traditional

Lower

Ref.

ARDS Network

11.8

6.2

76

a

~10.8

~7.1

87

Brochard et al.

PBW (predicted body weight)=50+0.91[Ht (cm)−152] (males) or 45.5+0.91[Ht (cm)−152] (females) a Tidal volumes of 10.3 and 7.1 mL/kg dry body weight were reported (not adjusted for gas compression/tube expansion). To convert to tidal volumes according to PBW, it is assumed: 1) Dry body weight=1.1 PBW 2) Mean tidal volume lost from gas compression/tube expansion= 2.0 * [(24)−PEEP] 3) Mean PBW=60 kg 4) 70% of ventilators in use did not make automatic adjustments

between-group separation in tidal volumes was greater in the NIH study because of differences in both the traditional and volume-and-pressure limited study groups. The traditional group tidal volumes were higher than in the study by Brochard et al. (87). Thus, there could have been more high-volume/high-pressure injury in the NIH traditional study group than in the study by Brochard et al. The volume-and-pressure

Acute respiratory distress syndrome

522

limited tidal volumes were lower in the NIH study than in the traditional study group in the study by Brochard et al. Thus, there could have been less high-volume/high-pressure injury in the low-volume/low-pressure group of the NIH study than in the study by Brochard et al. There may be a threshold of lung volume/pressure that must be exceeded to cause high-volume/high-pressure VALI. This is suggested by the comparison of the five trials summarized in Table 3. In the two trials in which there was a difference in mortality between traditional and lung-protective ventilation groups, the mean plateau pressure in the traditional study groups exceeded 32 cmH2O (107). Traditional group plateau pressures on days 1 and 2 after enrollment in the trials by Brochard et al. (87) and by the NIH were ~31.5 and ~33 cmH2O, respectively. However, methods for measuring plateau pressures varied among some of the studies. For example, Brochard et al. used an inspiratory pause of 2 seconds for plateau pressure measurements. In the NIH trial, plateau pressures were measured with 0.5-second inspiratory pause. Part of the modest difference in plateau pressures between the studies could have been caused by this difference in the duration of the inspiratory pause. Power It was necessary to randomize over 800 patients to demonstrate the 9% difference in mortality between the traditional and the volume-and-pressure limited strategies in the NIH trial. Because the between-group separations in tidal volumes and plateau pressures were probably less in three of the studies summarized in Table 3 (87–89), effects on mortality of the volume-and-pressure limited strategies, if any, were probably smaller. To demonstrate smaller effects on mortality would have required randomization of more patients. However, the combined enrollment of these three trials was 288 patients. Thus, small beneficial effects of volume-and-pressure limited ventilation may have been missed because sample sizes were too small. Beneficial effects were demonstrated in the study by Amato et al., which included only 53 patients (75). Perhaps this was possible because of the large separation between study groups in tidal volumes and inspiratory pressures and also because additional lung protective measures were included in the volume-andpressure study group. Management of Acidosis Recent studies in experimental animal models strongly suggest that hypercapnia and acidosis may prove beneficial by reducing oxidant-induced acute lung injury (108, 109). However, there could be other deleterious effects of hypercapnia and acidosis on various organs and systems. The NIH volume-and-pressure limited protocol required reciprocal increases in ventilator rate with initial tidal volume reductions to limit the hypercapnia from tidal volume reduction. Ventilator rate was increased further (to a maximum of 35) if arterial pH was <7.30. Infusions of bicarbonate were allowed (neither encouraged nor discouraged) if pH was <7.30 with ventilator rate=35. Management of acidosis was not controlled by protocol rules in two of the other trials of volume-and-pressure limited strategies, and there was more hypercapnia and acidosis in the volume-and-pressure

Mechanical ventilation in the acute respiratory distress syndrome

523

limited groups. This could have counteracted beneficial effects of the lung protective strategies in these studies. Protection from High-Pressure/High-Volume Versus Low-Volume/LowPressure Ventilation All of the lung-protective strategies used in the studies outlined in Table 3 reduced tidal volume and inspiratory pressures to reduce VALI from high volume/high pressure. Only the strategy used by Amato et al. (75) included measures to reduce VALI from lowvolume/low-pressure VALI, and these were combined with volume-and-pressure limitation. It was not clear which aspects of this combined strategy contributed to the beneficial effects. The more substantial improvements in the clinical outcomes in this study suggest that the beneficial effects of the various lung-protective measures were additive. However, alternative explanations such as imbalances between study groups may have contributed as well. Levels of PEEP were higher by ~1 cmH2O in the volume-and-pressure limited group of the NIH study. Moreover, there could have been more auto-PEEP in this group than in the traditional group because respiratory rates were higher (77, 78). If end-expiratory alveolar pressures were higher, then beneficial effects of the volume-and-pressure limited strategy could have been from reduced atelectasis at end-expiration. However, ratios were lower in the volume-and-pressure limitation group, as they were in animal models of acute lung injury and in normal humans during prolonged general anesthesia (18, 19, 110). This suggests that the smaller tidal swings in volume and pressure allowed additional atelectasis to occur. If this occurred, then the volume-andpressure limited approach would have been better despite more atelectasis, not because there was less atelectasis. Interpretation of oxygenation alone is difficult, however, because it can be influenced by hemodynamic effects (36). Also, a recent study showed that the alveolar-arterial difference was larger when intrinsic PEEP replaced external PEEP (77). Results from several animal experiments strongly suggest that ventilating with higher levels of PEEP can reduce VALI, presumably by increasing lung recruitment during expiration. However, other experiments showed increased edema formation (111, 112) and decreased venous return and cardiac output (73, 92) when higher PEEPs were used. Ventilation with higher PEEP will also tend to raise airway pressures and lung volumes during inspiration, which could increase high-pressure/high-volume VALI. If tidal volume is decreased further to control inspiratory volumes and pressures when PEEP is raised, then hypercapnia and acidosis will worsen. It is not yet clear that ventilator strategies designed to increase lung recruitment will improve clinical outcomes in ARDS patients receiving volume-and-pressure limited ventilation. The NIH ARDS Network recently concluded a trial in which 550 ALI/ARDS patients were randomized to receive either a traditional or a higher PEEP mechanical ventilation strategy (112a). In both study groups, tidal volumes and inspiratory plateau pressures were limited as in the previous NIH trial of volume-and-pressure limited ventilation. The difference in PEEP between the study groups was about 5 cmH2O. Mortality in both study groups was similar to the mortality in the study group that received the tidal volume-and-pressure limited approach in the previous NIH ARDS Network study, confirming the value of this approach.

Acute respiratory distress syndrome

524

However, there was no significant difference in mortality between the traditional and higher PEEP study groups of the more recent trial. At least two other clinical trials of higher PEEP ventilation strategies are currently enrolling patients in Canada and France. The results of these studies, especially in comparison with the results of other trials, will enhance understanding of the relative values of lung protection from high-volume/ highpressure versus low-volume/low-pressure ventilation and may allow design of better lung-protective ventilation strategies.

V. Noninvasive Ventilation With traditional approaches to mechanical ventilation, an endotracheal tube provides the interface or connection between the patient and a positive pressure ventilator. There are several potential complications of insertion and maintenance of endotracheal tubes. These include injury to the upper airway during intubation, aspiration of gastric or oral contents during or after intubation, and complications from use of the sedative and neuro-muscular blocking medications that are frequently required in patients in whom an endotracheal tube has been placed. These complications include neuropsychiatric and neuromuscular sequelae, ileus, and prolonged ventilator dependence. With noninvasive ventilation (NIV), a face mask is used instead of an endotracheal tube to connect the patient to the ventilator. This approach may avoid many complications of endotracheal intubation. Many patients require no sedation during NIV. Moreover, many patients require only intermittent ventilator support, which can be accomplished easily by removing and replacing the NIV face mask. During respites from NIV, some patients can eat, speak, and attend to hygiene. However, it is not possible to provide the same high levels of ventilator support through the face mask interface, and several technical challenges must be overcome before sufficient support can be achieved (113–115). The most convincing successes with NIV were obtained in patients with acute respiratory acidosis in whom hypoxemia was not the main reason for respiratory failure (116–118). These encouraging results were extended to some patients with hypoxemic respiratory failure, including selected patients with acute lung injury and ARDS (119– 123). In an earlier randomized study in patients with hypoxemic respiratory failure but no history of chronic lung disease, beneficial effects of NIV were apparent only in a subgroup of patients with acute hypercapnia (124). More recently, however, several studies have shown that, in selected patients with primarily hypoxemic respiratory failure, NIV may reduce the need for intubation and improve important clinical outcomes (120–122). In a recent study, patients with hypoxemic respiratory failure but without chronic obstructive pulmonary disease, hemodynamic instability, or neurological impairment were randomized to receive NIV with pressure support and PEEP or mechanical ventilation with an endotracheal tube when protocol-defined criteria for intubation were identified. Marked beneficial effects occurred in patients randomized to receive NIV (120). Improvements in oxygenation were similar with the noninvasive and invasive approaches. Despite a 30% failure rate, patients treated with NIV had shorter durations of mechanical ventilation and intensive care unit (ICU) stay and experienced fewer complications. Mortality was very high in patients who were intubated after failing NIV, suggesting that delayed intubation may adversely affect outcomes. Earlier initiation

Mechanical ventilation in the acute respiratory distress syndrome

525

of NIV (before criteria for intubation are identified) may reduce the frequency of NIV failure and reduce the associated risks of delayed intubation. In a recent study, patients with acute hypoxemic respiratory failure (most from acute lung injury) were randomized to receive NIV with oxygen and continuous positive airway pressure (CPAP) versus oxygen alone (125). There were early favorable responses in the CPAP group in oxygenation and in patient comfort, but there were no differences in the proportion of patients who required intubation, in-hospital mortality, or length of ICU stay. The patients treated with CPAP experienced significantly more complications, including cardiac arrest at the time of intubation, suggesting that intubation was performed at a late stage. These results suggest that NIV with pressure support and PEEP may be more effective than NIV with CPAP alone in patients with acute lung injury. Also, patients who receive NIV must be carefully monitored to minimize risks associated with NIV failure. Important beneficial effects of NIV include a reduction in infectious complications (126). Avoiding endotracheal intubation markedly reduces the risk of nosocomial pneumonia. Interestingly, other infectious complications are also reduced, perhaps because NIV allows a less “aggressive” approach to management of several comorbid problems, with less frequent use of sedative medications and of vascular and bladder catheters (121, 126). Patients with acute lung injury and high risk for nosocomial infection may be most likely to benefit from NIV. Several recent trials showed substantial beneficial effects of NIV when used as a preventive measure during episodes of acute hypoxemic respiratory failure in solid organ-transplant patients and in patients with severe immunosuppression, as in hematological malignancies and neutropenia (121, 123, 127). The rates of intubation and of infectious complications, length of stay, and mortality were significantly reduced in patients who received NIV. A study in 42 intensive care units in France demonstrated how often and in which clinical circumstances NIV was used among all patients requiring mechanical ventilator support (121). Patients treated with NIV contributed 16% of all patients requiring ventilator support and 35% of patients requiring ventilatory support without previous endotracheal intubation. NIV was used in fewer than 20% of all patients with hypoxemic acute respiratory failure, in one fourth of patients with pulmonary edema, in half of the patients with hypercapnic respiratory distress, and never in comatose patients. Forty percent of patients treated with NIV subsequently required endotracheal intubation. These results suggest that NIV can be used in only a small proportion of patients with hypoxemic respiratory failure. Selection of patients to receive NIV must be based on the severity and the presence of associated organ dysfunctions. Presence of hemodynamic or neurological dysfunction usually contraindicates NIV. In the remaining patients, it seems that NIV may help in reducing the need for endotracheal intubation by approximately 50%, but more studies are needed to better define groups of patients that will not benefit from such trials.

VI. High-Frequency Ventilation High-frequency ventilation (HFV) uses very small tidal volumes and very high respiratory rates (128–130). It is an attractive lung-protective strategy in ARDS because

Acute respiratory distress syndrome

526

it offers the potential to achieve high levels of lung recruitment (to reduce VALI from low-volume/low-pressure ventilation) while also avoiding injurious levels of volume and pressure during inspiration (to reduce VALI from high-volume/high-pressure ventilation) (131). With very high respiratory rates (3–10 Hz), can usually be maintained at near-normal levels, even with tidal volumes that are smaller than traditional estimations of physiological deadspace. Several different mechanical approaches have been used to deliver HFV (131–134). One of the major challenges has been to provide adequate airway humidification. Initial experiences with high-frequency jet ventilation in ARDS did not show beneficial effects (135, 136), and interest in HFV in adult intensive care units waned. There is renewed interest now because of the increasing concern for VALI in ARDS and because of the recent studies that demonstrated improved clinical outcomes with modifications to the traditional mechanical ventilation approach in ARDS (75, 76). However, further improvements with conventional ventilators may be limited by the ability to maintain acceptable gas exchange. High-frequency oscillatory ventilation (HFOV) has been tested extensively in babies with respiratory distress from surfactant deficiency and meconium aspiration. An early controlled trial did not demonstrate improvements in any important clinical outcome variables (137). The HFOV technique was then modified to achieve higher levels of lung recruitment. Several subsequent trials demonstrated lower incidence of chronic lung disease, air leak syndrome, and requirements for surfactant dosing (138– 141). HFOV is now used commonly in clinical practice in neonatal intensive care units. However, a recent multicenter study in neonates did not show beneficial effects, and some concerns were expressed regarding adverse effects (142). Moreover, HFOV machines used for babies are not powerful enough to effectively ventilate most adults with ARDS. More powerful HFOV machines were recently introduced to clinical practice for adults with ARDS. Two case series demonstrated that gas exchange could be maintained or improved with HFOV in many severe ARDS patients (132, 143). A recent multicenter trial compared clinical outcomes among patients randomized to either a conventional ventilator-based strategy or HFOV (Derdak, In Press AJRCCM). In the conventional ventilation group, tidal volumes were lower than those used in traditional approaches, but they were not as low as in the volume-and-pressure limited group of the NIH trial. PEEP levels in the conventional ventilation study group were higher than those used in traditional ventilation approaches, but not as high as those used for lung protection in trial by Amato et al. (75). Mortality at 30 days showed an encouraging trend in the HFOV group (37 versus 52%; p=0.10). This was a small study that did not have sufficient power to demonstrate a realistic treatment effect of HFV on mortality. Moreover, since it was an unblinded study, the results could be confounded by unintended differences in the quality or intensity of care between the study groups. Several questions should be answered before HFOV can be considered for routine use in ARDS patients. 1. Which ARDS patients should receive HFOV? Some have suggested that this mode should be reserved for patients with mean airway pressures >24 cmH2O. This criterion is not met in many ARDS patients until several days after onset of illness. Some may

Mechanical ventilation in the acute respiratory distress syndrome

527

deteriorate during this time because of VALI. Perhaps earlier intervention with HFOV would prevent deterioration in these patients. 2. How should HFOV be initiated, monitored, and adjusted? Most current recommendations are designed to achieve high levels of lung recruitment. In some patients, this requires raising mean airway pressure (MAP) to 35–45 cmH2O. As with high levels of PEEP, this could exacerbate pulmonary edema formation and cause circulatory depression. More information is needed from studies in animal models and in humans with ARDS to address the question of “optimal MAP” in HFOV. Near-normal levels can be achieved in many ALI/ARDS patients with HFV. This may require higher HFV tidal volumes than would be necessary if some hypercapnia were accepted. When combined with high mean airway pressure and very rapid respiratory rates, VALI from high volume/high pressure could occur. 3. Can HFOV techniques be developed that will allow patients to continue spontaneous respiratory efforts? Most of the ARDS patients in the two recent case series received neuromuscular blocking agents for several days (132, 143) because spontaneous respiratory efforts interfered with effective HFOV. This could increase the frequency and severity of neuromuscular complications, prolonging the course of ventilator dependency and rehabilitation after weaning from mechanical ventilation. In summary, HFOV is an interesting potential solution to the challenge of providing good respiratory support while avoiding VALI from high-pressure/high-volume and lowvolume/low-pressure ventilation. However, much experience with this new technology is needed to refine HFOV techniques. HFOV must then be tested in randomized trials designed to compare important clinical outcomes among patients who receive the best HFOV strategy with those who receive the best conventional ventilator-based lungprotective ventilation strategy.

VII. Inverse Ratio Ventilation and Airway Pressure Release Inverse ratio ventilation (IRV) uses long periods of inspiration and short periods of exhalation to raise mean airway pressure, which is usually associated with improved gas exchange (19, 144–146). If arterial oxygenation can be improved by raising mean airway pressure without raising the volumes and pressures at end-expiration and end-inspiration, then IRV could be used to reduce levels of PEEP and needed to support arterial oxygenation. However, when IRV was carefully tested in ARDS patients (147–149), there was little or no improvement in arterial oxygenation unless the short periods of exhalation caused some auto-PEEP (150). Studies in animal models and in humans demonstrated decreased physiological dead-space and with IRV. This may occur from improved ventilation-perfusion matching, enhanced collateral ventilation, or because exhalation occurs during the IRV respiratory cycle when alveolar CO2 concentration is near its maximal level rather than near its nadir, as occurs during ventilation when the duration of inspiration is shorter than the duration of expiration

Acute respiratory distress syndrome

528

(151, 152). Improved CO2 clearance could provide some relief from hypercapnia and acidosis, but the decreases in that occurred in clinical studies were quite modest (144, 145, 149, 153). A disadvantage of IRV is that it is a very uncomfortable mode of breathing. Most patients receiving IRV require heavy sedation and neuromuscular blockade. There are no controlled trials comparing important clinical outcomes among patients who received IRV versus other modes of ventilation. Airway pressure release ventilation (APRV) has some similarities to IRV but also some differences. As in IRV, the ventilator cycles between lower and higher pressures, with more time during each cycle at the higher pressure. This is designed to protect the lungs from VALI from both low volume/low pressure and high volume/high pressure while ensuring good gas exchange. The initial rationale for APRV was to improve clinical tolerance to high levels of CPAP, but it is now proposed as a way to preserve spontaneous breathing during pressure-controlled ventilation. This last feature, which is different from IRV and requires an exhalation valve that is highly responsive to patient efforts, may be useful for two reasons. First, atelectasis frequently occurs in dependent lung regions when respiratory muscles are inactive (154–158). With spontaneous ventilatory efforts, these lung regions are more likely to remain ventilated, reducing requirements for oxygenation support and also the potential for low-volume/low-pressure VALI (154–156, 158). Second, the brief decreases in pleural pressure that occur during spontaneous inspiratory efforts could increase venous return/cardiac output. This could partially counteract the decrease in venous return/cardiac output that may occur from ventilating with high positive airway pressures (32, 73). APRV applies a high airway pressure (e.g., 25 cmH2O) for most of the ventilator respiratory cycle to promote high levels of lung recruitment and reduce potential for lowvolume/low-pressure VALI. Some of the work of ventilation occurs from the patient’s spontaneous efforts, which typically generate tidal volumes of ~200 mL. With these small tidal volumes, the increase in lung distention is only modestly greater than the level induced by the high ventilator airway pressure. However, many patients cannot sustain normal ventilation with these spontaneous efforts. Therefore, the ventilator contributes to ventilation by periodically releasing airway pressure to a low level (0–10 cmH2O) and then raising it back to the high level. The resulting ventilator tidal volume may be substantial in size, but it does not cause additional distention above the level induced by the high airway pressure. However, these airway pressure release breaths could cause VALI from low-volume/low-pressure ventilation. Gas exchange can be supported effectively in many ALI/ARDS patients with APRV (157, 159). In a recent study, 30 patients at risk for ARDS from trauma were randomized to receive either APRV or IRV (157). Patients who received APRV had significantly ratios, cardiac indices, and oxygen delivery, supporting the value in higher APRV of preserving spontaneous respiratory efforts. Patients who received IRV received more medication for sedation and neuromuscular blockade. This may also support the value of preserving spontaneous respiratory muscle efforts. However, this finding resulted in part from the different protocols used for adjusting these medications in the PCIRV vs. APRV study groups; by study design, deep sedation and paralysis was used only in the PCIRV group. There were encouraging reductions in length of ventilatory support, duration of intubation, and ICU length of stay in the APRV group.

Mechanical ventilation in the acute respiratory distress syndrome

529

Patients’ respiratory efforts have been compared during APRV and other modes of assisted ventilation, such as pressure support. Patient work of breathing was usually much higher with APRV. Thus, care must be exercised when using this modality (160, 161). There are no randomized trials comparing important clinical outcomes among patients who receive APRV vs. traditional or other lung-protective ventilation approaches.

VIII. Pressure Control Ventilation There is no clinically useful way to directly measure distention of the lung parenchyma during inspiration. A commonly used surrogate marker for end-inspiratory distention is the inspiratory “plateau” pressure, which reflects mean intra-alveolar pressure at endinspiration (162, 163). The traditional approach to mechanical ventilation in ARDS uses volume control modes, in which the ventilator delivers tidal volumes prescribed by the clinician. With this mode, end-inspiratory alveolar pressure is a dependent variable that is determined primarily by tidal volume in addition to PEEP and respiratory system compliance. Because of the concern for VALI from high volume/high pressure, many investigators and clinicians have adopted pressure control (PC) modes of ventilation for ARDS patients, in which the ventilator raises airway pressure to a level prescribed by the clinician. This allows the clinician to limit alveolar pressure during inspiration to a level that is thought to be safe. With this approach, tidal volume is a dependent variable, determined primarily by the PC inspiratory pressure, PEEP, and respiratory system compliance. If PEEP and respiratory system compliance are constant, a tidal volume and an endinspiratory alveolar pressure are tightly linked. For any tidal volume delivered with a volume control mode, there will be an end-inspiratory alveolar pressure which, if prescribed as a PC inspiratory pressure, would result in a comparable tidal volume if inspiratory time is long enough to allow inspiratory flow to become nil at the end of inspiration. And for any inspiratory pressure delivered with a PC mode, there will be a resulting tidal volume which, if prescribed as a volume control tidal volume, would result in the same inspiratory alveolar pressure. Thus, the same levels of tidal volume/ inspiratory alveolar pressure can be achieved with either volume or pressure control mode: volume control tidal volume can be adjusted to achieve a desired plateau pressure; or PC inspiratory pressure can be adjusted to achieve a desired tidal volume. Spontaneous respiratory muscle activity alters the relationship of tidal volume and inspiratory alveolar pressure. If there are spontaneous inspiratory efforts during the period of PC inspiration, tidal volume and the resulting level of distention may rise substantially while the PC inspiratory pressure remains at its assumed safe level. Spontaneous respiratory muscle efforts have little effect on tidal volumes or levels of inspiratory distention when volume control modes are used. However, some patients have active expiratory efforts when plateau pressures are measured. This causes false elevations of plateau pressures. Pressure control provides inspiratory flow rates in the initial portion of inspiration that are higher than those that are usually delivered with volume control modes. Theoretically, rapid inspiration could improve gas exchange by allowing more time for a tidal volume to

Acute respiratory distress syndrome

530

distribute evenly among lung units with different airway resistances. However, in ARDS and did not improve when ventilator mode was changed from patients, volume control to PC with tidal volume and PEEP held constant (148, 164). To search for a possible time-dependent effect, gas exchange was compared during PC and volume control ventilation over 6-hour periods: no beneficial effects of PC on oxygenation were identified during this time (165). In some studies, PCIRV was associated with significant hemodynamic compromise (164, 165).

IX. Extracorporeal Gas Exchange Extracorporeal membrane oxygenation (ECMO) is a technique in which blood is temporarily removed from the systemic circulation and pumped across a gas exchange membrane. Most of a patient’s requirements for oxygen are provided through the membrane to the blood, and most of the patient’s CO2 production is simultaneously removed. This approach was promising because it offered an opportunity to provide “lung rest,” to reduce the amount of ventilation required to clear CO2, and to reduce risks from oxygen toxicity. However, the technique entailed certain risks because it required placement of large-bore catheters in a large systemic artery and vein and also system anticoagulation. In a randomized, controlled clinical trial in patients with severe ARDS, mortality was not reduced in the study group that received ECMO (166). Extracorporeal CO2 removal (ECCO2-R) is a modified extracorporeal gas exchange technique designed to avoid some of the technical problems with ECMO. A smaller portion of systemic blood flow is removed from a systemic vein, pumped through a smaller gas exchange device, and returned to a systemic vein. Most of a patient’s CO2 production can be removed, but only a fraction of the oxygen requirements can be provided through the extra-corporeal circuit. Most of the requirements for oxygenation are provided by low-frequency positive pressure ventilation (LFPPV). In several case series, mortality was lower in severe ARDS patients who received LFPPV-ECCO2-R than in earlier severe ARDS patients who did not receive extracorporeal gas exchange (167–169). However, a randomized controlled clinical trial did not demonstrate improved mortality with LFFPV-ECCO2-R (170). Patients who received LFPPV-ECCO2-R experienced more bleeding complications and required more blood transfusions than those who did not receive extracorporeal support. Thus, it is possible that there were beneficial effects of LFFPPV-ECCO2-R that were counteracted by adverse effects. Despite these disappointing results, some centers have continued to improve extracorporeal gas exchange techniques to reduce complications and improve selection criteria to identify patients most likely to benefit (171). Extracorporeal gas exchange may be a useful technique for some patients in the future, but additional studies are needed before this approach can be recommended for more widespread use.

X. Recommendations for Mechanical Ventilation in ARDS Clinical studies in the past 15 years have provided much useful information to guide clinicians to modify traditional approaches to mechanical ventilation in ARDS. Improved

Mechanical ventilation in the acute respiratory distress syndrome

531

clinical outcomes, including fewer hospital-acquired infections and lower in-hospital mortality, can be expected with these modifications. A. Noninvasive Ventilation In patients without hemodynamic compromise, NIV should be used early in the course of ARDS, before gas exchange becomes severely compromised and before a patient’s sensorium becomes impaired. With early use of NIV, some patients may be spared from complications of endotracheal intubation, such as nosocomial pneumonia and adverse effects of heavy sedation and neuro-muscular blockade. This will probably decrease mortality and shorten duration of intensive care. B. Ventilator Mode When it is necessary to ventilate with an endotracheal tube, either a volume or pressure control mode can be used. The same lung protective ventilation objectives can be achieved with both. There are no clinical studies supporting the superiority of either mode. The choice of mode should be made primarily on the basis of familiarity of the clinical staff. However, with all modes it is crucial that the user understands the physiological relationship between volume and the different pressures available in the ventilator. Regardless of mode, both inspiratory pressures and tidal volumes should be monitored and adjusted to achieve specific volume- and pressure-limited ventilation goals. This monitoring must be specific to either the volume-or pressure-based approach. When volume control modes are used, monitoring of plateau pressure is of utmost importance. When pressure control modes are used, expired volume must be monitored carefully. C. Tidal Volumes and Inspiratory Pressures Smaller tidal volumes (lower inspiratory pressures) should be used than those used with the traditional approach. If a volume-control mode is used, there is consensus that tidal volume should be decreased to achieve plateau pressures of ≤30 cmH2O. If a pressurecontrol mode is used, there is consensus that the ventilator inspiratory pressure should be ≤30 cmH2O. Some investigators and clinicians have suggested that plateau pressures or PC inspiratory pressures of 30 cmH2O are safe; that it is not necessary to reduce tidal volume further if plateau pressure is ≤30 cmH2O (or to use PC inspiratory pressures <30 cmH2O). This suggestion is based on several observations: 1. In some animal experiments, the relationship between ventilator pressures and lung injury appeared to be curvilinear, suggesting that a threshold of distension or pressure must be exceeded to cause lung injury (55, 172). However, some animal experiments were brief in duration (minutes-hours). Effects of mechanical ventilation with moderate inspiratory airway pressures for days, as is usually necessary in ARDS patients, were not assessed. Moreover, many of the animal experiments were conducted with uninjured lungs. Safe inspiratory pressures and volumes in uninjured lungs may be higher than when there is another cause of acute lung injury.

Acute respiratory distress syndrome

532

2. In normal humans, there are no apparent ill effects from occasional deep inspirations to total lung capacity, at which transpulmonary pressure is approximately 30 cmH2O (173). However, effects of hundreds to thousands of deep inspirations over hours to days have not been assessed. 3. In some randomized studies, there were no apparent beneficial effects of volume-andpressure limited strategies when the mean plateau pressures in traditional strategy study groups were <32 cmH2O (87–89). (Other explanations for these results were discussed earlier in this chapter.) Is there a beneficial effect of decreasing tidal volumes to achieve plateau pressures lower than 30 cmH2O (or to prescribing PC inspiratory pressures lower than 30 cmH2O)? In the NIH trial of volume-and-pressure limited ventilation, many patients in the traditional tidal volume group had plateau pressures <30 cmH2O (76). An analysis of subsets in this study suggests that mortality of these patients would have been lower if they received tidal volumes of 6 mL/kg with plateau pressures considerably lower than 30 cmH2O. There may be circumstances in which deleterious effects of tidal volume reduction may exceed beneficial effects. For example, in some patients with acute disorders of the central nervous system, hypercapnia from tidal volume reduction could cause increased intracranial pressure. Some patients become severely dyspneic when tidal volume is reduced. In volume-cycled modes, some severely dyspneic patients trigger second tidal volumes before exhalation of initial tidal volumes, effectively increasing the delivered tidal volume to twice the intended volume. A modest increase in tidal volume, for example, from 6 to 7 or 8 mL/kg, may relieve the desperate sensation of dyspnea in these patients and prevent the double-breaths. In some very dyspneic patients, work of breathing on small tidal volumes may be excessive. It may be prudent to allow modestly higher tidal volumes or inspiratory pressures in these patients to avoid complications from excessive sedative or neuromuscular blocking medications. Several practical issues should be considered when using small tidal volumes in patients with severe ARDS. Some ventilators can automatically adjust to compensate for the portion of tidal volume lost due to ventilator circuit tube expansion and gas compression during inspiration. This automatic adjustment is not available on many ventilators and must be made manually instead. Without this adjustment, the tidal volume delivered to the patient may be substantially less than intended. Ventilator circuit deadspace should also be reduced as much as is practical. Heat- and moisture-exchanging filters with internal volumes close to 100 mL, as frequently used in Europe (174), may substantially reduce alveolar ventilation, leading to worse hypercapnia (175, 176). D. PEEP and FiO2 The approach used in the NIH trial of volume-and-pressure limited ventilation represented a consensus of how PEEP and FiO2 were used by the NIH investigators and clinical colleagues in 1995. The consensus evolved with the knowledge that both PEEP and supplemental oxygen can support arterial oxygenation, but that both can also have adverse effects (177). An alternative approach, which is based on similar concerns, is to adjust PEEP to achieve acceptable arterial oxygenation while maintaining This approach, which was used in one of the trials of volume-and-

Mechanical ventilation in the acute respiratory distress syndrome

533

pressure limited ventilation (89), gives higher priority to preventing oxygen toxicity in on the NIH scheme, but it gives lower priority to patients who would require high preventing complications from higher PEEP. A third approach is to use even higher PEEP levels to prevent low-volume/low-pressure VILI (75). are often delayed and very difficult to Adverse effects of both PEEP and high distinguish from respiratory and hemodynamic effects of ARDS, sepsis, and related conditions. There is very little clinically useful information to guide clinicians who must decide how to achieve immediate goals for arterial oxygenation while avoiding oxygen toxicity and adverse effects from both low and high PEEP levels. There appears to be consensus that PEEP levels should be at least as high as those used in the NIH scheme. Additional information from current clinical trials of traditional versus higher PEEP strategies will provide to the foundation for more evidence-based approaches to using PEEP and

to optimize important clinical outcomes.

Acknowledgments The authors thank Catherine Weaver for assistance with preparation of the manuscript. Supported by NIH NHLBI Contract NO1-HR-46063.

References 1. Rinaldo JE, Rogers RM. Adult respiratory-distress syndrome. N Engl J Med 1982; 306:900–909. 2. Gattinoni L, Present! A, Torresin A. Adult respiratory distress syndrome profiles by computed tomography. J Thorac Imaging 1986; 1:25–30. 3. Dantzker DR, Brook CJ, DeHart P, et al. Ventilation-perfusion distributions in the adult respiratory distress syndrome. Am Rev Respir Dis 1979; 120:1039–1052. 4. Broseghini C, Brandolese R, Poggi R, Polese G, Manzin E, Milic-Emili J, Rossi A. Respiratory mechanics during the first day of mechanical ventilation in patients with pulmonary edema and chronic airway obstruction. Am Rev Respir Dis 1988; 138:355–361. 5. Nuckton TJ, Alonso JA, Kallet RH, Daniel BM, Pittet JF, Eisner MD, et al. Pulmonary deadspace fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med 2002; 346:1281–1286. 6. Ashbaugh DB, Bigelow DB, Petty TL. Acute respiratory distress in adults. Lancet 1967; 2:319– 323. 7. Baker AB. Artificial respiration, the history of an idea. Med Hist 1971; 15:336–351. 8. Morch ET. History of mechanical ventilation. In: Kirby RR, Smith RA, Desautels DA, eds. Mechanical Ventilation. New York: Churchill Living-stone, 1985:1–58. 9. Colice GL. Historical perspective on the development of mechanical ventilation. In: Tobin M, ed. Principles and Practice of Mechanical Ventilation. New York: McGraw-Hill, Inc., 1994:1– 35. 10. Drinker PA, Shaw LA. An apparatus for the prolonged administration of artificial respiration. J Clin Invest 1929; 7:229–247. 11. Drinker PA, McKhann CF. The use of a new apparatus for the prolonged administration of artificial respiration. JAMA 1929; 92:1658–1660. 12. Emerson JH. The Evolution of Iron Lungs. Cambridge, MA: J.H. Emerson, 1978.

Acute respiratory distress syndrome

534

13. Ashbaugh DG, Petty TL, Bigelow DB, Harris TR. Continuous positive-pressure breathing (CPPB) in adult respiratory distress syndrome. J Thorac Cardiovasc Surg 1969; 57:31–41. 14. Bendixen HH, Egbert LD, Hedley-Whyte J, Laver MD, Pontoppidan H, eds. Respiratory Care. 149th ed. St. Louis: CV Mosby Co., 1965. 15. Pontoppidan H, Geffin B, Lowenstein E. Acute Respiratory Failure in the Adult. Boston: Little, Brown, & Co., 1973. 16. Barber RE, Lee J, Hamilton WK. Oxygen toxicity in man. N Engl J Med 1970; 283(27): 1478– 1484. 17. Tobin M, Chadha TS, Jenouri G, Birch SJ, Gazeroglu HB, Sackner MA. Breathing patterns. Chest 1983; 84(2):202–206. 18. Benedixen HH, Hedley-Whyte J, Laver MB. Impaired oxygenation in surgical patients during general anesthesia with controlled ventilation. N Engl J Med 1963; 269:991–997. 19. Cheney FW, Burnham SC. Effect of ventilatory pattern on oxygenation in pulmonary edema. J Appl Physiol 1971; 31(6):909–912. 20. Hedley-Whyte J, Laver MB, Benedixen HH. Effect of changes in tidal ventilation on physiologic shunting. Am J Physiol 1964; 206:891–897. 21. Royer F, Martin DJ, Benchetrit G, Grimbert FA. Increase in pulmonary capillary permeability in dogs exposed to 100% O2. J Appl Physiol 1988; 65(3): 1140–1146. 22. Royston BD, Webster NR, Nunn JF. Time course of changes in lung permeability and edema in the rat exposed to 100% oxygen. J Appl Physiol 1990; 69(4): 1532–1537. 23. Lodato RF. Oxygen toxicity. In: Tobin MJ, ed. Principles and Practice of Mechanical Ventilation. New York: McGraw-Hill, Inc., 1994:837. 24. Davis WB, Rennard SI, Bitterman PB, Crystal RG. Pulmonary oxygen toxicity. N Engl J Med 1983; 32(7):878–883. 25. Caldwell PRB, Lee WL, Schildkraut HS, Archibald ER. Changes in lung volume, diffusing capacity, and blood gases in men breathing oxygen. J Appl Physiol 1966; 21(5):1477–1483. 26. Carmichael LC, Dorinsky PM, Higgins SB, Bernard GR, Dupont WD, Swindell B, et al. Diagnosis and therapy of acute respiratory distress syndrome in adults: an international survey. J Crit Care 1996; 11(1):9–18. 27. Thompson BT, Hayden D, Matthay MA, Brower R, Parsons PE, for the NIH ARDS Network. Clinicians’ approaches to mechanical ventilation in acute lung injury and ARDS. Chest 2001; 120:1622–1627. 28. Falke K, Pontoppidan H, Kumar A, Leith D, Geffin B, Laver HB. Ventilation with positive end-expiratory pressure in acute lung disease. J Clin Invest 1972; 51:2315–2323. 29. Gattinoni L, Pelosi P, Crotti S, Valenza F. Effects of positive end-expiratory pressure on regional distribution of tidal volume and recruitment in adult respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:1807–1814. 30. Shapiro BA, Cane RD, Harrison RA. Positive end-expiratory pressure therapy in adults with special reference to acute lung injury: A review of the literature and suggested clinical correlations. Crit Care Med 1984; 12:127–141. 31. Malo J, Ali J, Wood LDH. How does positive end-expiratory pressure reduce intrapulmonary shunt in canine pulmonary edema? J Appl Physiol 1984; 57: 1002–1010. 32. Scharf SM. Mechanical cardiopulmonary interactions in critical care. In: Dantzker DR, Scharf SM, eds. Cardiopulmonary Critical Care. Philadelphia: W.B. Saunders Company, 1998:75–91. 33. Jardin F, Delorme G, Hardy A, Auvert B, Beauchet A, Bourdarias J-P. Reevaluation of hemodynamic consequences of positive pressure ventilation emphasis on cyclic right ventricular afterloading by mechanical lung inflation. Anesthesiology 1990; 72:966–970. 34. Jardin R, Farcot JC, Boisante L, Curien N, Margairaz A, Bourdarias J-P. Influence of positive end-expiratory pressure on left ventricular performance. N Engl J Med 1981; 304:387–392. 35. Pinsky MR. The effects of mechanical ventilation on the cardiovascular system. Crit Care Clin 1990; 6:663–678.

Mechanical ventilation in the acute respiratory distress syndrome

535

36. Dantzker DR, Lynch JP, Weg JG. Depression of cardiac output is a mechanism of shunt reduction in the therapy of acute respiratory failure. Chest 1980; 77:636–642. 37. Suter P, Fairley HB, Isenberg MD. Optimum end-expiratory airway pressure in patients with acute pulmonary failure. N Engl J Med 1975; 292:284–289. 38. Suter P, Fairley HB, Isenberg MD. Effect of tidal volume and positive end-expiratory pressure on compliance during mechanical ventilation. Chest 1978; 73:158–162. 39. Jonson B, Richard JC, Straus C, Mancebo J, Lemaire F, Brochard L. Pressure-volume curves and compliance in acute lung injury: evidence of recruitment above the lower inflection point. Am J Respir Crit Care Med 1999; 159:1172–1178. 40. Katz JA, Ozanne GM, Zinn SE, Fairley HB. Time course and mechanisms of lung volume increase with PEEP in acute pulmonary failure. Anesthesiology 1981; 54:9–16. 41. Ranieri VM, Eissa NT, Corbeil C, Chasse M, Braidy J, Matar N, et al. Effects of positive endexpiratory pressure on alveolar recruitment and gas exchange in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1991; 144:544–551. 42. Ranieri VM, Mascia L, Fiore T, Bruno F, Brienza A, Giuliani R. Cardio-respiratory effects of positive end-expiratory pressure during progressive tidal volume reduction (permissive hypercapnia) in patients with acute respiratory distress syndrome. Anesthesiology 1995; 83(4):710–720. 43. Naureckas ET, Dawson CA, Gerber BS, Gaver DP, Gerber HL, Linehan JH, et al. Airway reopening pressure in isolated rat lungs. J Appl Physiol 1994; 76(3):1372–1377. 44. Maggiore SM, Jonson B, Richard JC, Jaber S, Lemaire F. Alveolar derecruitment at decremental positive end-expiratory pressure levels in acute lung injury: comparison with the lower inflection point, oxygenation, and compliance. Am J Respir Crit Care Med 2001; 164:795–801. 45. Brochard L. Respiratory pressure-volume curves. In: Tobin M, ed. Principles and Practice of Intensive Care Monitoring. New York: McGraw-Hill, 1998: 597–616. 46. Crotti S, Mascheroni D, Caironi P, Pelosi P, Ronzoni G, Mondino M, et al. Recruitment and derecruitment during acute respiratory failure. A clinical study. Am J Respir Crit Care Med 2001; 164:131–140. 47. Maunder RJ, Shuman WP, McHugh JW, Marglin SI, Butler M. Preservation of normal lung regions in the adult respiratory distress syndrome. JAMA 1986; 255:2463–2465. 48. Gattinoni L, Pesenti A, Avalli L, Ross F, Bomino M. Pressure-volume curve of total respiratory system in acute respiratory failure: computed tomographic scan study. Am Rev Respir Dis 1987; 136:730–736. 49. Lamy M, Fallat RJ, Koeniger E, Dietrich HP, Ratliff JL, Eberhart RC, et al. Pathologic features and mechanisms of hypoxemia in adult respiratory distress syndrome. Am Rev Respir Dis 1976; 114:267–284. 50. Webb HH, Tierney DF. Experimental pulmonary edema due to intermittent positive pressure ventilation with high pressures. Am Rev Respir Dis 1974; 110:556. 51. Parker JC, Hernandez LA, Longenecker GL, Peevy K, Johnson W. Lung edema caused by high peak inspiratory pressures in dogs. Am Rev Respir Dis 1990; 142:321–328. 52. Dreyfuss D, Soler P, Basset G, Saumon G. High inflation pressure pulmonary edema. Am Rev Respir Dis 1988; 137:1159–1164. 53. Tsuno K, Miura K, Takeya M, Kolobow T, Morioka T. Histopathologic pulmonary changes from mechanical ventilation at high peak airway pressures. Am Rev Respir Dis 1991; 143:1115–1120. 54. Dreyfuss D, Saumon G. State of the art: ventilator-induced lung injury; lessons from experimental studies. Am J Respir Crit Care Med 1998; 157:294–323. 55. Dreyfuss D, Soler P, Saumon G. Mechanical ventilation-induced pulmonary edema: interaction with previous lung alterations. Am J Respir Crit Care Med 1995; 151:1568–1575. 56. Dreyfuss D, Saumon G. Role of tidal volume, FRC, end-inspiratory volume in the development of pulmonary edema following mechanical ventilation. Am Rev Respir Dis 1993; 136:730–736.

Acute respiratory distress syndrome

536

57. Corbridge TC, Wood LDH, Crawford GP, Chudoba MJ, Yanos J, Sznajder JI. Adverse effects of large tidal volume and low PEEP in canine acid aspiration. Am Rev Respir Dis 1990; 142:311–315. 58. Muscedere JG, Mullen JBM, Gan K, Slutsky AS. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med 1994; 149:1327–1334. 59. Tremblay L, Valenza F, Ribeiro SP, Li J, Slutsky AS. Injurious ventilatory strategies increase cytokines and c-fos m-RNA expression in an isolated rat lung model. J Clin Invest 1997; 99(5):944–952. 60. Verbrugge S, Sorm V, van’t Veen A, Mouton JW, Gommers D, Lachman B. Lung overinflation without positive end-expiratory pressure promotes bacteremia after experimental Klebsiella pneumoniae inoculation. Intens Care Med 1998; 24(2):172–177. 61. Nahum A, Ravenscraft SA, Nakos G, Adams AB, Burke WC, Marini JJ. Effect of catheter flow direction on CO2 removal during tracheal gas insufflation in dogs. J Appl Physiol 1993; 75:1238–1246. 62. Faridy EE, Permutt S, Riley RL. Effect of ventilation on surface forces in excised dogs’ lungs. J Appl Physiol 1996; 21:1453–1462. 63. Mead J, Takishima T, Leith D. Stress distribution in lungs: a model of pulmonary elasticity. J Appl Physiol 1970; 28:596–608. 64. Slutsky AS, Tremblay LN. Multiple system organ failure; is mechanical ventilation a contributing factor? Am J Respir Crit Care Med 1998; 157:1721–1725. 65. Martynowicz MA, Walters BJ, Hubmayr RD. Mechanisms of recruitment in oleic acid-injured lungs. J Appl Physiol 2001; 90(5): 1744–1753. 66. Gattinoni L, D’Andrea L, Pelosi P, Vitale G, Pesenti A, Fumagalli R. Regional effects and mechanisms of positive end-expiratory pressure in early adult respiratory distress syndrome. JAMA 1993; 269:2122–2135. 67. Dambrosio M, Roupie E, Mollet JJ, Anglade MC, Vasile N, Lemaire F, et al. Effects of PEEP and different tidal volumes on alveolar recruitment and hyperinflation. Anesthesiology 1997; 87:495–503. 68. Tasker V, John J, Evander E, Robertson B, Jonson B. Surfactant dysfunction makes lungs vulnerable to repetitive collapse and reexpansion. Am J Respir Crit Care Med 1997; 155:313– 320. 69. Richard JC, Maggiore SM, Johnson B, Mancebo J, Lemaire F, Brochard L. Influence of tidal volume on alveolar recruitment: respective role of PEEP and a recruitment maneuver. Am J Respir Crit Care Med 2001; 163:1609–1613. 70. Amato MBP, Barbas CSV, Medeiros DM, Schettino GPP, Filho GL, Kairallo RA, et al. Beneficial effects of the “open lung approach” with low distending pressures in acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 152:1835–1846. 71. Mergoni M, Martelli A, Volpi A, Primavera S, Zuccoli P, Rossi A. Impact of positive endexpiratory pressure on chest wall and lung pressure-volume curve in acute respiratory failure. Am J Respir Crit Care Med 1997; 156:846–854. 72. Hickling KG. The pressure-volume curve is greatly modified by recruitment; a mathematical model of ARDS lungs. Am J Respir Crit Care Med 1998; 158:194–202. 73. Pinsky MR. Heart-lung interactions. In: Pinsky MR, Dhainaut JF, eds. Pathophysilogic Foundations of Critical Care. Baltimore: Williams & Wilkins, 1993:472–490. 74. Eisner MD, Thompson BT, Schoenfeld DA, Anzueto A, Matthay MA. Acute Respiratory Distress Syndorme Network. Airway pressures and early barotrauma in patients with acute lung injury and acute respiratory distress syndrome. Am J Respir Crit Care Med 2002; 165:978–982. 75. Amato MBP, Barbas CSV, Medeiros DM, Magaldi RB, Shettino GPP, Lorenzi-Filho G, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338(6):347–354.

Mechanical ventilation in the acute respiratory distress syndrome

537

76. Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 77. Richard JC, Brochard L, Bria W, Aboab J, Vandelet P, Tamion F, et al. Influence of respiratory rate on gas trapping during low volume ventilation of patients with acute lung injury. Intens Care Med 2002; 28:1078–1083. 78. De Durante G, Del Turco M, Rustichini L, Cosimini P, Giunta F, Hudson L, et al. Low tidal volume/high respiratory rate ventilation (“ARDSnet strategy”) may generate intrinsic PEEP in patients with ARDS. Am J Respir Crit Care Med 2002; 165:1271–1274. 79. Macklin MT, Macklin CC. Malignant interstitial emphysema of the lungs and mediastinum as an important occult complication in many respiratory diseases and other conditions: an interpretation of the clinical literature in the light of laboratory experiment. Medicine 1944; 23:281–352. 80. DiRusso SM, Nelson LD, Safcsak K, Miller RS. Survival in patients with severe adult respiratory distress syndrome treated with high-level positive end-expiratory pressure. Crit Care Med 1995; 23:1485–1496. 81. Gammon RB, Shin MS, Groves RH, Hardin JM, Hsu C, Buchalter SE. Clinical risk factors for pulmonary barotrauma: a multivariate analysis. Am J Respir Crit Care Med 1995; 152:1235– 1240. 82. Gammon RB, Shin MS, Buchalter SE. Pulmonary barotrauma in mechanical ventilation. Patterns and risk factors. Chest 1992; 102:568–572. 83. Schnapp LM, Chin DP, Szaflarski N, Matthay MA. Frequency and impor-tance of barotrauma in 100 patients with acute lung injury. Crit Care Med 1995; 23:272–278. 84. Weg JG, Anzueto A, Balk RA, Wiedemann HP, Pattishall EN, Schork MA, et al. The relation of pneumothorax and other air leaks to mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338(6):341–346. 85. Boussarsar M, Thierry G, Jaber S, Roudot-Thoraval F, Lemaire F, Brochard L. Relationship between ventilatory settings and barotrauma in the acute respiratory distress syndrome. Intens Care Med 2002; 28:406–413. 86. Esteban A, Anzueto A, Frutos F, Alia I, Brochard L, Stewart TE, et al. A 28-day international study of the characteristics and outcomes in patients receiving mechanical ventilation. JAMA 2002; 287:345–355. 87. Brochard L, Roudot-Thoraval F, Roupie E, Delclaux C, Chastre J, Fernandez-Mondejar E, et al. Tidal volume reduction for prevention of ventilator-induced lung injury in the acute respiratory distress syndrome. Am J Respir Crit Care Med 1998; 158:1831–1838. 88. Brower RG, Shanholtz CB, Fessler HE, Shade DM, White P, Wiener CM, et al. Prospective randomized, controlled clinical trial comparing traditional vs. reduced tidal volume ventilation in ARDS patients. Crit Care Med 1999; 27:1492–1498. 89. Stewart TE, Meade MO, Cook DJ, Granton JT, Hodder RV, Lapinsky SE, et al. Evaluation of a ventilation strategy to prevent barotrauma in patients at high risk for acute respiratory distress syndrome. N Engl J Med 1998; 338(6):355–361. 90. Anzueto A, Baughman RP, Guntupalli KK, Weg JG, Wiedemann HP, Raventos AA, et al. Aerosolized surfactant in adults with sepsis-induced acute respiratory distress syndrome. N Engl J Med 1996; 334(22): 1417–1421. 91. Jardin F, Fellahi L, Beauchet A, Viellard-Baron A, Loubieres Y, Page B. Improved prognosis of acute respiratory distress syndrome 15 years on. Intens Care Med 1999; 25:936–941. 92. Cournand A, Motley HL, Werko L, Richards DW. Physiological studies of the effects of intermittent positive pressure breathing on cardiac output in man. Am J Physiol 1948; 152:162– 174. 93. Viellard-Baron A, Loubieres Y, Schmitt J-M, Page B, Dubourg O, Jardin F. Cyclic changes in right ventricular output impedance during mechanical ventilation. J Appl Physiol 1999; 87(5): 1644–1650.

Acute respiratory distress syndrome

538

94. Schmitt J-M, Viellard-Baron A, Augarde R, Prin S, Page B, Jardin F. Positive end-expiratory pressure titration in acute respiratory distress syndrome patients: impact on right ventricular outflow impedance evaluated by pulmonary artery Doppler flow velocity measurements. Crit Care Med 2001; 29(6): 1154–1158. 95. Quigley MJ, Fraser RS. Pulmonary pneumatocele: pathology and pathogenesis. AJR 1988; 150:1275–1277. 96. Ranieri VM, Suter P, Tortorella C, De Tullio R, Dayer J, Brienza A, et al. Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome. JAMA 1999; 282(1):54–61. 97. Viellard-Baron A, Schmitt J-M, Augarde R, Fellahi L, Prin S, Page B, et al. Acute cor pulmonale in acute respiratory distress syndrome submitted to protective ventilation: incidence, clinical implications, and prognosis. Crit Care Med 2001; 29:1551–1555. 98. Fox RB, Hoidal JR, Brown DM, Repine JE. Pulmonary inflammation due to oxygen toxicity: involvement of chemotactic factors and polymorphonuclear leukocytes. Am Rev Respir Dis 1981; 123:521–523. 99. Fracica PJ, Knapp MJ, Piantadosi A, Takeda K, Fulkerson WJ, Coleman RE, et al. Responses of baboons to prolonged hyperoxia: physiology and qualitative pathology. J Appl Physiol 1991; 71(6):2352–2362. 100. Sackner MA, Landa J, Hirsch J, Zapata A. Pulmonary effects of oxygen breathing. Ann Int Med 1975; 82:40–43. 101. Montgomery AB, Luce JM, Murray JF. Retrosternal pain is an early indicator of oxygen toxicity. Am Rev Respir Dis 1989; 139:1548–1550. 102. Katzenstein AL, Bloor CM, Leibow AA. Diffuse alveolar damage—the role of oxygen, shock, and related factors. Am J Path 1976; 85:210–222. 103. Frank L, Roberts R J. Endotoxin protection against oxygen-induced acute and chronic lung injury. J Appl Physiol Respir Environ Exercise Physiol 1979; 47(3):577–581. 104. Lykens MG, Davis WB, Pacht ER. Antioxidant activity of bronchoalveolar lavage fluid in the adult respiratory distress syndrome. Am J Physiol 1992; 262(6):L169–L175. 105. Hickling KG, Henderson SJ, Jackson R. Low mortality associated with low volume pressure limited ventilation with permissive hypercapnia in severe adult respiratory distress syndrome. Intens Care Med 1990; 16:372–377. 106. Hickling KG, Walsh J, Henderson S, Jackson R. Low mortality rate in adult respiratory distress syndrome using low-volume, pressure-limited ventilation with permissive hypercapnia: a prospective study. Crit Care Med 1994; 22(10): 1568–1578. 107. Tobin M. Culmination of an era in research on the acute respiratory distress syndrome. N Engl J Med 2000; 342:1360–1361. 108. Laffey JG, Tanaka M, Engelberts D, Luo X, Yuan S, Tanswell AK, et al. Therapeutic hypercapnia reduces pulmonary and systemic injury following in vivo lung reperfusion. Am J Respir Crit Care Med 2000; 162:2287–2294. 109. Shibata K, Cregg N, Engelberts D, Takeuchi A, Fedorko L, Kavanaugh BP. Hypercapnic acidosis may attenuate acute lung injury by inhibition of endogenous xanthine oxidase. Am J Respir Crit Care Med 1998; 158(5 pt 1):1578–1584. 110. Hedley-Whyte J, Pontoppidan H, Morris MJ. The response of patients with respiratory failure and cardiopulmonary disease to different levels of constant volume ventilation. J Clin Invest 1966; 45(10): 1543–1554. 111. Toung T, Saharia P, Permutt S, Zuidema GD, Cameron JL. Aspiration pneumonia: beneficial and harmful effects of positive end-expiratory pressure. Surgery 1977; 82(2):279–283. 112. Toung T, Saharia P, Mitzner W, Permutt S, Cameron JL. The beneficial and harmful effects of positive end expiratory pressure. Surg Gynecol Obstet 1978; 147:518–524. 112a. Brower R. Ventilation with traditional versus higher end-expiratory pressures in ALI/ARDS. Presented at the annual scientific meeting of the American Thoracic Society Atlanta, 2002.

Mechanical ventilation in the acute respiratory distress syndrome

539

113. Mehta S. Noninvasive positive pressure ventilation in acute respiratory failure. Intens Care Med 1998; 24:1113–1114. 114. Brochard L. Noninvasive ventilation in acute respiratory failure. Respir Care 1996; 41:456– 465. 115. Mehta S, Hill NS. Noninvasive ventilation. Am J Respir Crit Care Med 2001; 163:540–577. 116. Brochard L, Mancebo J, Wysocki M, Lofaso F, Conti G, Rauss A, et al. Noninvasive ventilation for acute exacerbations of chronic obstructive pulmonary disease. N Engl J Med 1995; 333:817–822. 117. Brochard L. Non-invasive ventilation for acute exacerbations of COPD: a new standard of care. Thorax 2000; 55:817–818. 118. Kramer N, Meyer TJ, Meharg J, Cece RD, Hill NS. Randomized, prospective trial of noninvasive positive pressure ventilation in acute respiratory failure. Am J Respir Crit Care Med 1995; 151:1799–1806. 119. Abou-Shala N, Meduri U. Non-invasive mechanical ventilation in patients with acute respiratory failure. Crit Care Med 1996; 24:705–715. 120. Antonelli M, Conti G, Rocco M, Maurizio B, De Blasi RA, Vivino G, et al. A comparison of noninvasive positive-pressure ventilation and conventional mechanical ventilation in patients with acute respiratory failure. N Engl J Med 1998; 339:429–435. 121. Antonelli M, Contin G, Bufi M, Costa MG, Lappa A, Rocco M, et al. Noninvasive ventilation for treatment of acute respiratory failure in patients undergoing solid organ transplantation. JAMA 2000; 283:235–241. 122. Martin TJ, Hovis JD, Costantino JP, Bierman MI, Donahoe MP, Rogers RM, et al. A randomized prospective evaluation of non-invasive ventilation for acute respiratory failure. Am J Respir Crit Care Med 2000; 161:807–813. 123. Hilbert G, Gruson D, Vargas F, Valentino R, Gbikpi-Benissan G, Dupon M, et al. Noninvasive ventilation in immunosuppressed patients with pulmonary infiltrates, fever, and acute respiratory failure. N Engl J Med 2001; 344(7):481. 124. Wysocki M. Noninvasive ventilation in acute cardiogenic pulmonary edema: better than continuous positive airway pressure? Intens Care Med 1999; 25:1–2. 125. Delclaux C, L’Her E, Alberti C, Mancebo J, Abroug F, Conti G, et al. Treatment of acute hypoxemic nohypercanic respiratory insufficiency with continuous positive airway pressure delivered by a face mask. JAMA 2000; 284:2352–2360. 126. Girou E, Schortgen F, Delclaux C, Brun-Buisson C, Blot F, Lefort Y. Association of noninvasive ventilation with nosocomial infections and survival in critically ill patients. JAMA 2000; 284:2361–2367. 127. Azoulay E, Alberti C, Bornstain C, Leleu G, Moreau D, Recher C, et al. Improved survival in cancer patients requiring mechanical ventilatory support: impact of non-invasive mechanical ventilatory support. Crit Care Med 2001; 29:519–525. 128. Chang HK. Mechanisms of gas transport during ventilation by high frequency oscillation. J Appl Physiol 1984; 56:553–563. 129. Slutsky AS, Drazen JM, Kamm RD. Alveolar ventilation at high frequencies using tidal volumes less than the anatomic dead space. In: Engel LA, Paiva M, Lenfant CE, eds. Lung Biology in Health and Disease. New York: Marcel Dekker, 1984:137–176. 130. Drazen JM, Kamm RD, Slutsky AS. High frequency ventilation. Physiol Rev 1984; 64:505– 543. 131. Krishnan J, Brower R. High-frequency ventilation for acute lung injury and ARDS. Chest 2000; 118:795–807. 132. Fort P, Farmer C, Westerman J, Johanningman J, Beninati W, Dolan S, et al. High-frequency oscillatory ventilation for adult respiratory distress syndrome—a pilot study. Crit Care Med 1997; 25:937–947. 133. Gluck E, Heard S, Patel C. Use of ultrahigh frequency ventilation in patients with ARDS: a preliminary report. Chest 1993; 103:1413–1420.

Acute respiratory distress syndrome

540

134. Carlon GC, Miodownik S, Ray C, Kahn RC. Technical aspects and clinical implications of high frequency jet ventilation with a solenoid valve. Crit Care Med 1981; 9:47–50. 135. Carlon GC, Howland WS, Ray C, Miodownik S, Griffin JP, Groeger JS. High-frequency jet ventilation: prospective, randomized evaluation. Chest 1983; 84(5):551–559. 136. Hurst JM, Branson RD, Davis K, Barrette RR, Adams KS. Comparison of conventional mechanical ventilation and high frequency ventilation. Ann Surg 1990; 211(4):486–491. 137. HIFI Study Group. High-frequency oscillatory ventilation compared with conventional mechanical ventilation in the treatment of respiratory failure in preterm infants. N Engl J Med 1989; 320:88–93. 138. Gerstmann DR, Minton SD, Stodard RA, Meredith KS, Monaco F, Bertrand JM, et al. The Provo multicenter early high-frequency oscillatory ventilation trial: improved pulmonary and clinical outcomes in respiratory distress syndrome. Pediatrics 1996; 98(6): 1044–1057. 139. HiFO Study Group. Randomized study of high-frequency oscillatory ventilation in infants with severe respiratory distress syndrome. J Pediatr 1993; 122:609–619. 140. Clark RH, Gerstmann DR, Null DM, de Lemos RA. Prospective randomized comparison of high frequency oscillatory and conventional ventilation in respiratory distress syndrome. Pediatrics 1992; 89:5–12. 141. Arnold JH, Truog RD, Thompson JE, Fackler JC. High-frequency oscillatory ventilation in pediatric respiratory failure. Crit Care Med 1993; 21(2): 272–278. 142. Moriette G, Paris-Llado J, Walti H, Escande B, Magny JF, Cambonie G, et al. Prospective randomized multicenter comparison of high-frequency oscillatory ventilation and conventional ventilation in preterm infants of less than 30 weeks with respiratory distress syndrome. Pediatrics 2001; 107(2):363–372. 143. Mehta S, Lapinsky SE, Hallett D, Merker D, Groll RJ, Cooper AB, et al. Prospective trial of high-frequency oscillation in adults with acute respiratory distress. Crit Care Med 2001; 29:1360–1369. 144. Gurevitch MJ, Van Dyke J, Young ES, Jackson K. Improved oxygenation and lower peak airway pressure in severe adult respiratory distress syndrome: treatment with inverse ratio ventilation. Chest 1986; 89:211–213. 145. Tharratt RS, Allen RP, Albertson TE. Pressure controlled inverse ratio ventilation in severe adult respiratory failure. Chest 1988; 94(4):755–762. 146. Al-Saady N, Bennett ED. Decelerating inspiratory flow waveform improves lung mechanics and gas exchange in patients on intermittent positive-pressure ventilation. Intens Care Med 1985; 11:68–75. 147. Lessard MR, Guerot E, Lorino H, Lemaire F, Brochard L. Effects of pressure-controlled with different I/E ratios versus volume-controlled ventilation on respiratory mechanics, gas exchange, and hemodynamics in patient with adult respiratory distress syndrome. Anesthesiology 1994; 80:983–991. 148. Mercat A, Graiini L, Teboul J-L, Lenique F, Richard C. Cardiorespiratory effects of pressurecontrolled ventilation with and without inverse ratio in the adult respiratory distress syndrome. Chest 1993; 104:871–875. 149. Cole AGH, Weller SF, Sykes MK. Inverse ratio ventilation compared with PEEP in adult respiratory failure. Intens Care Med 1984; 10:227–232. 150. Pepe PE, Marini JJ. Occult positive end-expiratory pressure in mechanically ventilated patients with airflow obstruction—the auto-PEEP effect. Am Rev Respir Dis 1982; 126:166– 170. 151. Marcy TW, Marini JJ. Inverse ratio ventilation in ARDS. Chest 1991; 100: 494–504. 152. Knelson JH, Howatt WF, DeMuth GR. Effect of respiratory pattern on alveolar gas exchange. J Appl Physiol 1970; 29:328–331. 153. Lain DC, BiBenedetto R, Morris SL, Van Nguyen AV, Saulters R, Causey D. Pressure control inverse ratio ventilation as a method to reduce peak inspiratory pressure and provide adequate ventilation and oxygenation. Chest 1989; 95:1081–1088.

Mechanical ventilation in the acute respiratory distress syndrome

541

154. Putensen C, Rasanen J, Lopez F, Downs JB. Effect of interfacing between spontaneous breathing and mechanical cycles on the ventilation-perfusion distribution in canine lung injury. Anesthesiology 1994; 81:921–930. 155. Putensen C, Rasanen J, Lopez F. Ventilation-perfusion distribution during mechanical ventilation with superimposed spontaneous breathing in canine lung injury. Am J Respir Crit Care Med 1994; 150:101–108. 156. Putensen C, Mutz N, Putensen-Himmer G, Zinserling J. Spontaneous breathing during ventilatory support improves ventilation-perfusion distributions in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 159:1241–1248. 157. Putensen C, Zech S, Wrigge H, Zinserling J, Stuber F, Von Spiegel T, et al. Long-term effects of spontaneious breathing during ventilatory support in patients with acute lung injury. Am J Respir Crit Care Med 2001; 164:43–49. 158. Froese AB, Bryan AC. Effects of anesthesia and paralysis on diaphragmatic mechanics in man. Anesthesiology 1974; 41:242–255. 159. Rasanen J. Airway pressure release ventilation. In: Tobin M, ed. Principles and Practice of Mechanical Ventilation. May wood, IL: McGraw-Hill, Inc., 1994:341–348. 160. Calzia E, Lindner KH, Witt S, Schirmer U, Lange H, Stenz R, et al. Pressure-time product and work of breathing during biphasic continuous positive airway pressure and assisted spontaneous breathing. Am J Respir Crit Care Med 1994; 150:904–910. 161. Katz-Papatheophilou E, Heindl W, Gelbmann H, Hollaus P, Neumann M. Effects of biphasic positive airway pressure in patients with chronic obstructive pulmonary disease. Eur Respir J 2000; 15:498–504. 162. Hubmayr RD, Abel MD, Rehder K. Physiologic approach to mechanical ventilation. Crit Care Med 1990; 18:103–113. 163. Kreit JW, Eschenbacher WL. The physiology of spontaneous and mechanical ventilation. Clin Chest Med 1988; 9:11–21. 164. Lessard MR, Guerot E, Lorino H, Lemaire F, Brochard L. Effects of pressure-controlled ventilation with different I:E ratios versus volume-controlled ventilation on respiratory mechanics, gas exchange, and hemodynamics in patients with adult respiratory distress syndrome. Anesthesiology 1994; 80:983–991. 165. Mercat A, Titiriga M, Anguel N, Richard C, Teboul J-L. Inverse ratio ventilation (I/E=2/1) in acute respiratory distress syndrome, a six-hour controlled study. Am J Respir Crit Care Med 1997; 155:1637–1642. 166. Zapol WM. Volotrauma and the intravenous oxygenator in patients with adult respiratory distress syndrome. Anesthesiology 1992; 77:847–849. 167. Gattinoni L, Pesenti A, Mascheroni D, et al. Low-frequency positive-pressure ventilation with extracorporeal CO2 removal in severe acute respiratory failure. JAMA 1986; 256:881–886. 168. Brunet F, Belghith M, Mira J-P, Lanore JJ, Vazelaire JF, Santucci JD, et al. Extracorporeal carbon dioxide removal and low-frequency positive-pressure ventilation. Chest 1993; 104:889– 898. 169. Brunet F, Mira J-P, Belghith M, et al. Extracorporeal carbon dioxide removal technique improves oxygenation without causing overinflation. Am J Respir Crit Care Med 1994; 149:1557–1562. 170. Morris AH, Wallace CJ, Menlove RL, Clemmer TP, Orme JF, Weaver LK, et al. Randomized clinical trial of pressure-controlled inverse ratio ventilation and extracorporeal CO2 removal for adult respiratory distress syndrome. Am J Respir Crit Care Med 1994; 149:295–305. 171. Bartlett RH. Extracorporeal life support in the management of severe respiratory failure. Clin Chest Med 2000; 21:555–561. 172. Parker JC, Townsley MI, Rippe B, Taylor AE, Thigpen J. Increased micro-vascular permeability in dog lungs due to high peak airway pressures. J Appl Physiol 1984; 57:1809– 1816. 173. Otis AB, Fenn W, Rahn H. Mechanics of breathing in man. J Appl Physiol 1950; 2:592–607.

Acute respiratory distress syndrome

542

174. Cook D, Ricard JD, Reeve B, Randall J, Wigg M, Brochard L, et al. Ventilator circuit and secretion management strategies: a Franco-Canadian survey. Crit Care Med 2000; 28:3547– 3554. 175. Le Bourdelles G, Mier L, Fiquet B, Djedaini K, Saumon G, Coste F, et al. Comparison of the effects of heat and moisture exchangers and heated humidifiers on ventilation and gas exchange during weaning trials from mechanical ventilation. Chest 1996; 110:1294–1298. 176. Pelosi P, Solca M, Ravagnan I, Tubiolo D, Ferrario L, Gattinoni L. Effects of heat and moisture exchangers on minute ventilation, ventilatory drive, and work of breathing during pressure-support ventilation in acute respiratory failure. Crit Care Med 1996; 24:1184–1188. 177. Almog A, Brower R. Complications of mechanical ventilation. In: Matthay MA, Schwartz DE, eds. Complications in the Intensive Care Unit. New York: Chapman & Hall, 1997:50–61. 178. Sykes MK. The place of measurement in the management of acutely ill patients. Monitoring in patients with respiratory failure. Proc R Soc Med 1965; 58(10):775–777. 179. McIntyre RW, Laws AK, Ramachandran PR. Positive expiratory pressure plateau: improved gas exchange during mechanical ventilation. Can Anaesth Soc J 1969; 16(6):477–486. 180. Kumar A, Falke K, Geffin B, Aldredge CF, Laver MB, Lowenstein E, et al. Continuous positive-pressure ventilation in acute respiratory failure. N Engl J Med 1970; 283(26):1430– 1436. 181. Falke K, Pontoppidan H, Kumar A, Leith D, Geffin B, Laver MB. Ventilation with endexpiratory pressure in acute lung disease. J Clin Invest 1972; 51(9):2315–2323. 182. Lutch JS, Murray JF. Continuous positive-pressure ventilation: effects on systemic oxygen transport and tissue oxygenation. Ann Int Med 1972; 76(2): 193–202. 183. Kumar A, Pontoppidan H, Falke KJ, Wilson RS, Laver MB. Pulmonary barotrauma during mechanical ventilation. Crit Care Med 1973; 1(4):181–186. 184. Steier M, Ching N, Roberts EB, Nealon TF. Pneumothorax complicating continuous ventilatory support. J Thorac Cardiovasc Surg 1974; 67:17–23. 185. Hemmer M, Suter P. Treatment of cardiac and renal effects of PEEP with dopamine in patients with acute respiratory failure. Anesthesiology 1979; 50(5):399–403. 186. Mathru M, Rao TL, Venus B. Ventilator induced barotrauma in controlled mechanical ventilation versus itermittent mandatory ventilation. Crit Care Med 1983; 11(5):359–361. 187. Matamis D, Lemaire F, Hart A, Brun-Buisson C, Ausquer JC, Atlan G. Total respiratory pressure-volume curves in the adult respiratory syndrome. Chest 1984; 86:58–66.

22 Severe Acute Respiratory Syndrome LORRAINE B.WARE Vanderbilt University School of Medicine Nashville, Tennesee, U.S.A.

I. Introduction At the time that this book was going to press, a multicountry outbreak of a new viral respiratory illness was occurring. The term severe acute respiratory syndrome (SARS) was chosen by the World Health Organization (WHO) to describe this highly contagious viral pneumonia. The resemblance to the acute respiratory distress syndrome (ARDS) is not in name alone. SARS frequently progresses to acute respiratory failure or acute lung injury (ALI), ARDS, and death. In this chapter, the epidemiology, clinical features, diagnosis, and treatment for SARS will be discussed. Because new information on SARS is being discovered and disseminated on an almost daily basis, the reviewer is referred to the World Health Organization (http://www.who.int/) and Centers for Disease Control (http://www.cdc.gov/) websites to access the most current information.

II. Epidemiology SARS first came to international attention in March 2003 when outbreaks were reported in Vietnam, Canada, Singapore, and Hong Kong. However, in retrospect, the first cases likely occurred in Guangdong Province of China, where an outbreak of a highly contagious and severe atypical pneumonia of unknown cause began in the fall of 2002 (1). The outbreak was not initially publicized (2), and the Guangdong cases only came to public attention after the spread of SARS to other areas (3). On March 12, 2003, the World Health Organization issued a global alert (4) and instituted worldwide surveillance for the disease. At the time of this writing, 7761 probable cases of SARS had been reported to the World Health Organization (5) in 29 countries, with a total of 623 deaths for a cumulative case fatality rate of 8.0%. The majority of these cases were in China, including Hong Kong, Beijing, Guangdong, and Shanxi provinces and Taiwan. Other countries with significant numbers of probable cases included Canada, Singapore, Vietnam, and the United States. The popularity and accessibility of international airline travel has facilitated the rapid spread of SARS worldwide. In addition, SARS appears to be highly contagious with numerous documented cases of secondary and tertiary transmission. Of particular concern is the large number of health care workers who have been secondarily infected. In the largest case series reported to date, 85 of the 138 cases were health care workers or

Severe acute respiratory syndrome

545

medical students, all of whom were infected in a Hong Kong hospital by one highly contagious patient (6). Some health care workers have died from the disease, including Dr. Carlo Urbani, a World Health Organization physician who was instrumental in recognizing the emergence of this new disease through his work in the Vietnam outbreak (6a). In addition to directly infecting many health care workers, SARS has had a generalized impact on health care provision in areas of outbreaks. For example, in Toronto, elective surgical procedures were canceled in an effort to contain the outbreak (7). SARS appears to be spread through close contact. Transmission by both droplet and fomite contact is suspected. In addition, the causative agent has been isolated from the stool of some patients (8). Apparently, even minor contact can be contagious as evidenced by infection in a hotel guest who stayed in the same Hong Kong hotel as one of the index cases but had no known contact with the index individual (1). The mode of transmission in some cases may be due to environmental contamination. For example, in Hong Kong 320 residents of an apartment complex were infected. The source of that outbreak is still under investigation but preliminary findings suggest a faulty sewage system (8a). In Toronto, infection of several health care workers occurred, apparently despite strict respiratory and contact isolation procedures (9). Despite uncertainty as to all the possible modes of transmission, strict infection control procedures seem to be largely effective. On April 28, 2003, the World Health Organization declared that the outbreak in Vietnam was completely contained after no new cases were re-ported for 20 days (10). By May 14, 2003, the outbreak in Toronto also appeared to be contained.

III. Causative Agent In order to facilitate rapid identification of the causative agent for SARS, the World Health Organization organized an unprecedented collaborative effort involving 11 international laboratories (11, 12). As a result, less than 2 months after SARS was recognized as a new global epidemic, the causative agent was tentatively identified as a novel coronavirus (8, 13, 14). This virus has been simultaneously isolated from SARS patients (but not healthy controls) on several continents using a variety of techniques. In addition, acute and convalescent serum from many patients shows a rise in antibody titer specific for this novel virus (8, 13, 14). The virus has now been shown in preliminary studies to induce a SARS-like illness in macaques (15). Coronaviruses are enveloped RNA viruses that can infect humans and animals (16). In humans they are most commonly associated with the common cold (17). Coronaviruses have also occasionally been reported to cause more serious illness, including pneumonia in military recruits, the elderly, the immunocompromised, and neonates (17–19). Coronavirus strains also can cause severe disease in multiple organ systems in animals including rats, mice, chickens, turkeys, calves, dogs, cats, rabbits, and pigs (20). However, the novel coronavirus that apparently causes SARS has only modest homology to any known strain of coronavirus and the source of the outbreak is currently unknown. Interestingly, Coronaviruses are known to undergo heterologous recombination, a process that may have led to the emergence of this new pathogen (16, 19a).

Acute respiratory distress syndrome

546

IV. Clinical Features and Diagnosis The most common features at the time of presentation are fever, rigors, cough, myalgia, and malaise (1, 6, 8, 21). A summary of the presenting symptoms in the first 208 patients that have been reported is shown in Table 1. Interestingly, although lower respiratory symptoms are almost universal, upper respiratory symptoms are distinctly less common, as is diarrhea. The most common findings on physical examination at presentation are fever, basilar rales, and hypoxemia. Neither rash nor adenopathy has been reported. Laboratory findings have been variable, and no finding is pathognomonic. The most consistently reported findings included leukopenia, lymphopenia, elevated aspartate aminotransferase, elevated lactate dehy-

Table 1 Frequency of Symptoms at Time of Initial Presentation in Patients with SARS Symptom

Percent of patients with symptom

Fever

100

Rigors or chills

75

Cough

62

Myalgia

57

Malaise

56

Dyspnea

49

Headache

47

Dizziness

36

Sputum production

28

Coryza

22

Sore throat

22

Diarrhea

19

Source: Based on initial symptoms reported in 208 patients described in Refs. 1, 6, 8, 21.

drogenase, and elevated creatine kinase (1, 6, 8, 21). Coagulation abnormal-ities have also been reported, including thrombocytopenia and elevated par-tial thromboplastin time (6). Chest radiographs typically show air space opacification that is unilateral or bilateral and has a predilection for the lower lobes. However, a wide variety of radiographic patterns have been described (1, 6, 8, 21). Focal infiltrates are common (Fig. 1a), but may pro-gress to multilobar involvement (Fig. 1b) or bilateral infiltrates indistin-guishable from ARDS (Fig. 2). On computed tomographic scans the infiltrates are typically patchy and peripheral with a ground glass appear-ance (6). At the time of this writing, no diagnostic test with specificity for SARS is yet clinically available. With the tentative identification of the causative agent, it is hoped that a diagnostic test will rapidly become available. Pos-sible candidates include an

Severe acute respiratory syndrome

547

antibody titer to the virus or identification of the virus itself using RT-PCR. Because of the similarity of SARS symptoms to

Figure 1 (a) Frontal chest radiograph at presentation from a 42-year-old man with SARS showing focal right upper

Acute respiratory distress syndrome

548

lobe infiltrate, (b) Frontal chest radiograph from a 48-year-old woman with SARS several days into her course. The patient initially presented with a unilobar infiltrate but progressed to bilateral multilobar infiltrates over the next few days. (Radiographs courtesy of Harry Shulman, M.D., Sunnybrook & Women’s College Health Sciences Center, Toronto, Ontario.)

Severe acute respiratory syndrome

549

Figure 2 Frontal chest radiographs from a 62-year-old man with SARS at the time of presentation (a) and 2 days later (b). Note the progression from

Acute respiratory distress syndrome

550

dense left upper lobe infiltrate to diffuse bilateral infiltrates with interval placement of endotracheal tube. (Radiographs courtesy of Harry Shulman, M.D., Sunnybrook & Women’s College Health Sciences Center, Toronto, Ontario.) Table 2 Case Definitions of Severe Acute Respiratory Syndrome Suspect case Temperature >100.4 AND One or more clinical findings of respiratory illness (e.g., cough, shortness of breath, difficulty breathing or hypoxia) AND Travel within 10 days of onset of symptoms to an area with documented or suspected community transmission of SARSa OR Close contactb within 10 days of onset of symptoms with a person known to be a suspect SARS case Probable case

A suspect case with one of the following: Radiographic evidence of pneumonia or respiratory distress syndrome Autopsy findings consistent with respiratory distress syndrome without an identifiable cause

Source: Adapted from Ref. 28. a Travel includes transit in an airport in an area with documented or suspected community transmission of SARS. b Close contact is defined as having cared for, having lived with, or having direct contact with respiratory secretions and/or body fluids of a patient known to be a suspect SARS case.

those of common illnesses including influenza and other viral pneumonias as well as community-acquired bacterial pneumonias, diagnosis of a probable case is highly dependent on contact or exposure history. In the United States, definitions were put forward by the Centers for Disease Control and Prevention (Table 2) that include symptoms of atypical pneumonia as well as travel to an area where SARS is endemic, or close contact with a known probable case of SARS. The time between exposure and development of clinically apparent disease is variable. In the largest case series reported to date, the incubation time was 2–16 days with a median of 6 days (6). Reports of the histopathological findings in the lung are limited at present. Only a few of the sickest patients have undergone lung biopsy, and autopsy data are also limited. However, in the majority of cases reported, the pathology is consistent with early or organizing diffuse alveolar damage, varying from mild to severe (1, 6, 13). Macrophage infiltration and multi-nucleate giant cells have also been reported (21a). Viral inclusions are not a prominent feature. Thus, the available data suggest that the most severe cases of SARS are histologically indistinguishable from other causes of ARDS. SARS has a variable clinical course that is still being fully characterized. Some patients are only mildly affected, with a self-limited febrile illness and minimal pulmonary symptoms. Others develop hypoxemic respiratory failure. Overall, in the two

Severe acute respiratory syndrome

551

largest case series, 51 of 188 patients (27%) required ICU admission, predominantly for respiratory failure (6, 8). In the report by Lee et al. of 138 patients, 19 required invasive mechanical ventilation (14%) (6). The percentage of patients with SARS that meet the definition of ALI/ARDS is unknown, but it is likely similar to the number requiring invasive mechanical ventilation (14% in one series). As noted above, the current case fatality rate is 8.0%. Advanced age or underlying medical conditions such as chronic hepatitis B seem to predispose to fatal outcome (6, 8a, 21b). In an epidemiological study of the first 1425 cases in Hong Kong, age greater than 60 was associated with a 43.3% case fatality rate (21b).

V. Treatment No specific treatment has been proven to be efficacious in SARS. Currently, the U.S. Army Medical Research Institute of Infectious Diseases is testing a large panel of antiviral drugs for activity against the virus and plans to test all FDA-approved medications as well (11). In Hong Kong, a large number of patients have been treated with oral intravenous ribavirin and high-dose corticosteroids (6, 8, 8a). However, to date there is no in vitro evidence that the SARS coronavirus is sensitive to ribavarin. Furthermore, given the relatively low case fatality rate and the lack of controlled trials, it is impossible to judge whether this treatment regimen is of any clinical benefit (22). The reports of clinical worsening at 7–9 days in the majority of a group of 75 patients treated with corticosteroids and ribavirin (8a) raises additional concerns that corticosteroids may prolong or worsen the course of SARS. Therefore, it is currently recommended that patients with SARS be treated with standard supportive therapy. In addition, patients should receive empirical antibiotic therapy directed against typical and atypical causes of community-acquired pneumonia (23). If patients progress to ALI/ ARDS, then mechanical ventilation with a low tidal volume, lung-protective strategy should be instituted (24). The reader is referred to Chapter 21 for a complete discussion of mechanical ventilation in ALI/ARDS. For complete reviews of standard supportive therapy for ALI/ARDS, please see Refs. 25 and 26. In addition to appropriate therapy, careful attention to isolation procedures is of utmost priority for any patient with suspected or probable SARS. Current Centers for Disease Control and Prevention guidelines recommend standard, contact, and airborne precautions (27). Standard precautions include hand washing and the use of eye protection for all patient contact. Contact precautions include the use of gown and gloves for all contact with patients or their environment. Airborne precautions include the use of a negative pressure room and of N-95 filtering disposable respirators for persons entering the room.

VI. Conclusions At present there is more about SARS that is unknown than known. Although the likely causative agent has been identified, exact modes of transmission remain unclear and no proven therapy has been identified. Whether the current multicountry epidemic will be

Acute respiratory distress syndrome

552

contained or will continue to grow is also unclear. One thing that appears certain is that the media interest and widespread fear surrounding this outbreak are not likely to abate soon. Despite the current case count of less than 5000 cases worldwide, SARS is already having a global impact on the economy, the airline and travel industries, and access to healthcare and education in affected cities. This has occurred despite the fact that other potentially fatal, highly contagious viral illnesses such as influenza are much more common and, in absolute numbers, far more deadly. The emergence of SARS as a new cause of ARDS reminds us of the many unresolved questions in ARDS pathogenesis and treatment. As evidenced by the content of this book, we have made good progress in understanding the pathogenesis of ARDS in the past 20 years. However, as is the case with SARS, there is still more that is unknown than is known. In particular, the interplay between pathogens (including both viruses and bacteria) and the lung is poorly understood and may be of fundamental significance in the pathogenesis of ARDS. Perhaps the advent of SARS will lead to new insights into the pathogenesis and treatment of ARDS, particularly ARDS that results from acute bacterial and viral infection.

References 1. Tsang K, Ho P, Ooi G, Yee W, Wang T, Chan-Yeung M, Lam W, Seto W, Yam L, Cheung T, Wong P, Lam B, Ip M, Chan J, Yuen K, Lai K. A cluster of cases of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003; 348:1977–1985. 2. Emerging stronger from the China crisis. Lancet 2003; 361:1311. 3. World Health Organization. Severe acute respiratory syndrome (SARS). Wkly Epidemiol Rec 2003; 78:86. 4. World Health Organization. Cases of severe respiratory illness may spread to hospital staff. Accessed 4/23/03 at http://www.who.int/mediacentre/releases/2003/%20pr22/en/. 5. World Health Organization. Cumulative number of reported probable cases of Severe Acute Respiratory Syndrome (SARS). Accessed 5/19/03 at http://%20www.%20who.int/csr/sarscountry/2003_05_l7/en. 6. Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt G, Ahuja A, Yung M, Leung C, To K, Lui S, Szeto C, Chung S, Sung J. A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003; 348: 1986–1994. 6a. Reilley B, Van Herp M, Sermand D, Dentico N. SARS and Carlo Urbani. N Engl J Med 2003; 348:1951–1952. 7. Alphonso C. Hospitals scramble to protect SARS staff. The Globe and Mail. Toronto, 2003: accessed 4/23/03 at http://www.globeandmail.com/servlet/%20story/RTGAM.20030422.usars0422/BNStory/Nation al/?query=SARS. 8. Peiris J, Lai S, Poon L, Guan Y, Yam L, Nicholls J, Yee W, Yan W, Cheung M, Cheng V, Chan K, Tsang D, Yung R, Ng T, Yuen K. Members of the SARS Study Group. Coronavirus as a possible cause of severe acute respiratory syndrome. Lancet 2003; 361:1319–1325. 8a. Peiris JSM, Chu CM, Cheng VCC, Chan KS, Hung IFN, Poon LLM, Law KI, Tang BSF, Hon TYW, Chan CS, Chan KH, Ng JSC, Zheng BJ, Ng WL, Lai RWM, Guan Y, Yuen KY, and members of the HKU/UCH SARS Study Group. Clinical progression and viral load in a community outbreak of coronavirus-associated SARS pneumonia: a prospective study. Lancet 2003, published online 5/9/03 at http://image.thelancet.com/extras/03art4432web.pdf. 9. Perkins T, US experts eye Toronto regimen. The Globe and Mail. Toronto, 2003; 4/23/03; A5.

Severe acute respiratory syndrome

553

10. World Health Organization. Viet Nam SARS-free. Accessed online 4/28/03 at http://www.who.int/mediacentre/releases/2003/pr_sars/en/. 11. Enserink M, Vogel G. Deferring competition, global net closes in on SARS. Science 2003; 300:224–225. 12. Gerberding J. Faster…but fast enough? N Engl J Med, 2003; 348:2030–2031. 13. Ksiazek T, Erdman D, Goldsmith C, Zaki S, Peret T, Emery S, Tong S, Urbani C, Comer J, Lim W, Rollin P, Dowell S, Ling A-E, Humphrey C, Shieh W-J, Guarner J, Paddock C, Rota P, Fields B, DeRisi J, Yang J-Y, Cox N, Hughes J, LeDuc J, Bellini W, Anderson L, the SARS Working Group. A novel coronavirus associated with severe acute respiratory syndrome. N Engl J Med 2003; 348:1947–1958. 14. Drosten C, Gunther S, Preiser W, van der Werf S, Brodt H-R, Becker S, Rabenau H, Panning M, Kolesnikova L, Fouchier R, Berger A, Burguiere A-M, Cinatl J, Eickmann M, Escriou N, Grywna K, Kramme S, Manuguerra J-C, Muller S, Rickerts V, Sturmer M, Vieth S, Klenk H-D, Osterhaus A, Schmitz H, Doerr H. Identification of a novel coronavirus in patients with severe acute respiratory syndrome. N Engl J Med, 2003; 348:1907–1976. 15. Fouchier RAM, Kuiken T, Schutten M, et al. Koch’s postulates fulfilled for SARS virus. Nature 2003; 423:240. 16. Holmes K. Coronaviruses. In: Knipe D, Howley P, eds. Fields Virology. Philadelphia: Lippincott Williams and Wilkins, 2001:1187–1203. 17. El-Sahyly H, Atnar R, Glezen W, Greenberg S. Spectrum of clinical illness in hospitalized patients with “common cold” virus infections. Clin Infect Dis 2000; 31:96–100. 18. Wenzel R, Hendley J, Davies J, Gwaltney JJ. Coronavirus infections in military recruits: threeyear-study with coronavirus strains OC43 and 229E. Am Rev Respir Dis 1974; 109:621–624. 19. Fotz R, Elkordy M. Coronavirus pneumonia following autologous bone marrow transplantation for breast cancer. Chest 1999; 115:901–905. 19a. Holmes KV. SARS-associated coronavirus. New Engl J Med 2003; 348:1948–1950. 20. McIntosh K. Coronaviruses. In: Mandell, Bennett, Dolin, eds. Principles and practice of infectious diseases. New York: Churchill Livingstone Inc., 2000: 1767–1769. 21. Poutanen S, Low D, Henry B, Finkelstein S, Rose D, Green K, Tellier R, Draker R, Adachi D, Ayers M, Chan A, Skowronski D, Salit I, Simor A, Slutsky A, Doyle P, Krajden M, Petric M, Brunham R, McGeer A, for the National Microbiology Laboratory Canada, and the Canadian Severe Acute Respiratory Syndrome Study Team. Identification of severe acute respiratory syndrome in Canada. N Engl J M, 2003; 348:1995–2005. 21a. Nicholls JM, Poon LLM, Lee KC, et al. Lung pathology of fatal severe acute respiratory syndrome. Lancet 2003; published online 5/16/03 at http://%20image.thelancet.com/extras/03art4347web.pdf. 21b. Donnelly CA, Ghani AC, Leung GM, et al. Epidemiological determinants of spread of causal agent of severe acute respiratory syndrome in Hong Kong. Lancet 2003; published online 5/7/03 at http://image.thelancet.com/extras/%2003art4453web.pdf. 21c. So LK-Y, Lau ACW, Yam LYC, et al. Development of a standard treatment protocol for severe acute respiratory syndrome. Lancet 2003; 361:1615–1617. 22. Falsey A, Walsh E. Novel coronavirus and severe acute respiratory syndrome. Lancet, 2003; 361:1312–1313. 23. American Thoracic Society. Guidelines for the management of adults with community-acquired pneumonia. Am J Respir Crit Care Med 2001; 163:1730–1754. 24. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 25. Brower RG, Ware LB, Berthiaume Y, Matthay MA. Treatment of ARDS. Chest 2001; 120:1347–1367. 26. Ware LB, Matthay MA. Medical progress: the acute respiratory distress syndrome. N Engl J Med 2000; 342:1334–1349.

Acute respiratory distress syndrome

554

27. Centers for Disease Control and Prevention. Updated interim domestic infection control guidance in the health-care and community setting for patients with suspected SARS. Accessed 4/23/03 at http://www.cdc.gov/ncidod/sars/%20infectioncontrol.htm. 28. Centers for Disease Control and Prevention. Updated interim U.S. case definition of severe acute respiratory syndrome (SARS). Accessed April 21, 2003, at http://www.cdc.gov/ncidod/sars/pdf/sars-casedefinition.PDF.

23 Acute Lung Injury Recent Progress and Promising Directions for Future Research MICHAEL A.MATTHAY University of California at San Francisco San Francisco, California, U.S.A.

Acute lung injury (ALI) and the acute respiratory distress syndrome (ARDS) are syndromes of acute respiratory failure that account for considerable morbidity and mortality in critically ill patients. The syndrome was originally described by Dr. Petty in 1967 (see Chap. 1). The support of the Lung Division of the National Heart, Lung, and Blood Institute (NHLBI) for basic research, clinical research, as well as for clinical trials has had an enormously positive impact in this field. The purpose of this final chapter will be to highlight areas where particular progress has been made both in clinical and basic research as well as to emphasize some potentially promising areas for future research.

I. Clinical Definitions Because the initial definition of the acute respiratory distress syndrome lacked specific criteria that could be used to identify patients systematically, there was controversy over the incidence and natural history of the syndrome and the mortality associated with it (1). In 1988 an expanded definition was proposed to quantify the physiological respiratory impairment through the use of a four-point lung injury scoring system that was based on the level of positive end-expiratory pressure (PEEP), the ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen, the static respiratory compliance, and the degree of infiltration on the chest radiograph (2). Other factors included in the assessment were the inciting clinical disorder and the presence or absence of nonpulmonary organ dysfunction (see Chap. 2). The lung injury scoring system has been widely used to quantify the severity of acute lung injury in both clinical research and clinical trials. In 1994, a new definition was recommended by the North American-European Consensus Conference Committee (see Chap. 2) (3). The consensus definition has two advantages. First, it recognizes that the severity of clinical lung injury varies: patients with less severe hypoxemia (as defined by ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen of 300 or less) are considered to have acute lung injury, and those with more severe hypoxemia (as defined by a ratio of 200 or less) are considered to have ARDS. The recognition of patients with ALI has facilitated early

Acute respiratory distress syndrome

556

enrollment of affected patients in clinical trials (4–6). Second, the definition is simple to apply in the clinical setting. However, it should be appreciated that this simplicity also has a disadvantage, since factors that influence the outcome, such as the underlying clinical disorder and whether other organ systems are affected, do not need to be assessed (7–9). In addition, the criterion for the presence of bilateral infiltrates on the chest radiograph consistent with the presence of pulmonary edema has been insufficiently specific for it to be applied consistently by experienced clinicians (10, 11). Nevertheless, the wide-spread acceptance of the 1994 consensus definition and the 1988 scoring system has improved the standardization of clinical research and trials.

II. Clinical and Biochemical Predictors of Outcome The search for biochemical and clinical predictors of markers of ALI has been complicated by a number of factors, including (1) the lack of correlation between the clinical diagnosis and pathogenesis, (2) the recognition that patients at risk for and with ALI are heterogeneous, (3) the ongoing discovery of new mediators and modulators of inflammation, (4) the recognition that the development of ALI is likely the result of a balance of mediators and modulators of inflammation, and (5) an increasing awareness of the importance of correlating biochemical markers with physiological variables. Patients with a higher severity of illness score are more likely to develop ALI/ARDS and to die from lung injury. This conclusion has been substantiated in several epidemiological studies (7, 8, 12, 13). However, the search for reliable pulmonary specific variables that are independently associated with mortality has been difficult. Indices of arterial hypoxemia have not been predictive of clinical outcome in most studies (7, 8, 13). However, one recent study established that an elevated pulmonary deadspace fraction has an independent predictive power for identifying patients with ARDS who are more likely to die (13). This finding will need to be explored in conjunction with biochemical studies as well as other methods to determine the pathological basis for the high deadspace fraction in early ARDS. It is possible that activation of procoagulant mechanisms in the circulation and in the distal airspaces of the lung in patients with early lung injury contributes to high ventilation to perfusion lung units, which account in part for the markedly elevated deadspace. The same study also established that a decrease in quasistatic respiratory compliance also has independent predictive value for mortality. Patients who eventually died with ARDS had a significantly lower respiratory compliance (27±9 mL/cmH2O) than those who survived (32±12 mL/cmH2O) during the first 24 hours after the onset of ARDS. The reduced respiratory compliance in nonsurvivors probably indicates that these patients had more pulmonary edema and less functional surface-active material early in the course of their respiratory failure, two physiological hallmarks of the severity of initial lung injury. Biochemical markers of endothelial injury, such as elevated levels of von Willebrand factor antigen, may be useful. Von Willebrand factor antigen reflects activation and injury to pulmonary and systemic endothelium (14). Several other biochemical markers have been associated with a poor outcome, and several recent clinical studies have evaluated these biological markers (see Chap. 7). For example, because of the growing evidence that supports a contribution of procoagulant mechanisms in the pathogenesis of

Acute lung injury

557

lung injury (see Chap. 10), it is important to test the predictive value of plasma levels of protein C, thrombomodulin, and plasminogen activator-1 for their relationship to endothelial injury and clinical outcomes.

III. Pathogenesis Considerable progress has been made in understanding the pathogenesis of acute lung injury. Much of the progress has been made in animal studies that model human acute lung injury. Progress has also been made in a variety of clinical studies. The results have been summarized in several chapters in this volume (see Chaps. 6–12). One of the areas of major progress has been recognizing the importance of ventilatorinduced lung injury and ventilator-associated lung injury. The initial studies that suggested a potential role for adverse ventilatory strategies were done in animal models (1). Subsequently, phase II and phase III clinical studies were conducted which demonstrated the clinical value of lung-protective ventilatory strategies in reducing mortality in patients with acute lung injury (5, 15). Chapter 21 in this volume summarizes in detail the results and significance of the clinical studies and the importance of future clinical trials to test additional methods to further improve outcome with novel lung protective ventilatory strategies. A lung-protective ventilatory strategy is the only proven therapy for ALI/ARDS that has actually reduced mortality. The favorably results of the initial NHLBI-sponsored clinical trial showed that mortality was reduced from 40 to 31% (5). This finding has been reproduced in a recently completed NHLBI trial in which the low tidal volume, plateau pressure-limited strategy resulted in a 26% mortality in both arms of a clinical trial that test different levels of PEEP (data presented in preliminary form at the American Thoracic Society meeting in May 2002). Although the higher levels of PEEP did not improve outcomes compared to the levels of PEEP used in the first NHLBI clinical trial, the reduction in mortality to 26% validates the results of the first large NHLBI clinical trial because it shows that a low tidal volume plateau pressure strategy, regardless of the level of PEEP, substantially reduces mortality (5).

IV. Future Directions: Research Although considerable progress has been made in understanding the pathogenesis and pathophysiology of acute lung injury, there are significant knowledge gaps that must be addressed in order to facilitate further progress in understanding ALI/ARDS and for designing new strategies for detection, prevention, management, and therapy. A recent NIH acute lung injury workshop that addressed several of these issues was summarized in a recent report (16). Although much has been learned about the cell biology of the alveolar capillary membrane since ALI/ARDS was described, the functional responses, biochemical pathways, patterns of gene expression, molecular mechanisms of interaction with other cells, and responses to injury remain incompletely characterized, and our understanding of these processes is fragmentary. New and evolving technologies in analytical methods may provide an opportunity to advance our knowledge. For example, the important role of the alveolar epithelial barrier in the pathogenesis and resolution of

Acute respiratory distress syndrome

558

ALI/ARDS has been increasingly appreciated, including the role it plays in the production of surface-active material to maintain alveolar stability as well as providing salt and water transport pathways for the resolution of alveolar edema. However, our understanding of the responses of the alveolar epithelium to injury remains incomplete. Disruption of the alveolar epithelium integrity is a major contributor of increased permeability, and alveolar flooding with protein-rich edema are hallmarks of ALI/ARDS. A better understanding of the molecular signaling pathways and mechanism cell-cell interaction in alveolar epithelial cells is needed. We need to know more about the biology of alveolar epithelial type I and type II cells and how the epithelial barrier is restored once it has been injured. The degree of endothelial injury probably has considerable impact on the severity of injury to the alveolar epitheliar barrier as well. More basic knowledge is required to understand how endothelial injury occurs as well as how injury evolves throughout different segments of the endothelial barrier. There is considerable diversity among lung endothelial cells, and we currently have an incomplete understanding of the molecular mechanisms that govern the responses of lung endothelial cells and their interactions with the nearby alveolar epithelium. A better understanding of the mechanisms of apoptosis and necrosis in the initial injury and repair of lung epithelial and endothelial cells is also needed. The mechanisms that govern apoptosis and necrosis in epithelial and endothelial cells in the key phases of ALI/ARDS need better definition. Chapter 8 in this volume considers this issue in depth. It is also important to increase our understanding of how the cells of the innate immune system (neutrophils, macrophages, natural killer cells, and other cells) and cells involved in hemostasis and thrombosis (platelets and endothelial cells, together with interacting leukocytes) are dysregulated in ALI/ARDS (see Chaps. 6–10). Although it was originally thought that increased permeability pulmonary edema and other features of ALI/ARDS could be explained by an increase in endothelial and epithelial permeability, it is now clear that acute inflammation, thrombosis, and activation of the coagulation system are probably also involved. There is evidence that dysregulation of innate immune and thrombotic cascades contribute to the initiation and progression of ALI/ARDS, especially in sepsis (see Chap. 10). Additional knowledge regarding the mechanisms that link acute inflammatory responses to thrombosis and to activation of the coagulation cascade is also important since there is substantial evidence that microvascular thrombosis and dysregulated intracellular and extracellular fibrin deposition are early events in ALI/ ARDS, in both experimental and clinical lung injury (see Chaps. 5 and 7). The recent approval of recombinant activated protein C for treatment of severe sepsis indicates that studies of molecular mechanisms that link acute inflammation in the hemostatic system are important since this agent interrupts both the inflammatory and thrombotic response (see Chap. 10). The new out break of a severe community acquired viral pneumonia termed severe acute respiratory syndrome (SARS) emphasizes the need for both new prevention and treatment strategies for potentially fatal pneumonias that can cause ALI/ARDS (see Chap. 22).

Acute lung injury

559

V. Genomics and Proteomics in Acute Lung Injury Research Genomic approaches using microarrays and other methods that display multiple DNA sequences have become major tools in biological investigation and are being increasingly applied in critical care medicine. Microarray analysis can be applied to isolated cells and normal and diseased tissues. Laser capture micro-dissection enhances the power of this technology. Genomic approaches are increasingly being applied to questions in pulmonary critical care medicine and seems to have considerable potential to address many unanswered questions related to changes in the phenotypes of critical cells in ALI/ARDS. Many of these studies can be done on an experimental level in animal models of acute lung injury. It is important to remember that some of the early events in ALI/ARDS do not involve new gene expression or changes in the level of transcripts (16). In addition, evolving proteomic technologies have the potential to identify each protein expressed in isolated cells or in complex tissues such as the injured lung and also to determine posttranslational modifications, protein-protein interactions, and other critical features that define phenotype and regulate cell function. Proteome approaches can also provide physiological relevence that is not inherent in measuring transcript levels and profiles. Proteomic analysis of clinical samples may be useful in establishing molecular patterns that are characteristic of individual diseases or the stages of ALI/ARDS (16). There is a growing awareness that genetic susceptibilities to ALI/ARDS or to the clinical disorders that predispose to ALI/ARDS may be useful in providing new insight into the pathogenesis of ALI/ARDS. Polymorphisms in genes related to inflammatory markers ( surfactant proteins and IL-6) and pathogen receptors (CD14, TLRs) have some correlation to the incidence and outcomes in ALI/ARDS (see Chaps. 13 and 14). These observations have primarily utilized case-control association studies. Although association studies pose significant logistic challenges in critically ill patient populations and have important limitations, they may be the most practical approach to studying genetic predispositions in patients with ALI/ARDS (17). One of the challenges in using classical genetics approaches to assess susceptiblities in ALI/ARDS is the heterogeneity of phenotypes in patients with ALI/ARDS. For example, the incidence, natural history, and outcome of ALI/ARDS induced by sepsis is quite different from the ALI/ARDS that occurs from trauma. Also, patients develop pneumonia from several different etiologies, including gram-positive and gram-negative bacteria, viruses, fungi, and opportunistic infections. The signaling pathways that regulate the inflammatory response to any particular organism depend on the initial molecular interactions between the host cells and the organism. Therefore, a better understanding of bacterial genetics and the specific virulence factors of infecting bacteria should provide new approaches to understanding susceptibility to pneumonia both as a cause of injury and as a cause of nosocomial pneumonia in patients with ALI/ARDS. As discussed in a recent NIH conference (16), a better understanding of why some patients with sepsis develop lung injury is also needed. Is there a host susceptibility to sepsis of the patients

Acute respiratory distress syndrome

560

who are more likely to upregulate gene expression for pro-inflammatory cytokines, making them more likely to develop acute lung injury?

VI. Clinical Trials The NIH-supported ARDS clinical trials network has provided a critical mechanism for testing new therapies for patients with ALI/ARDS. The network has been expanded from the initial 10 university centers to its current 19 university centers, including one center in Canada. This network has been successful in testing a lung-protective ventilatory strategy that resulted in a reduction in mortality from 40% to 31%, the first treatment modality that has been proven to reduce mortality in ALI/ARDS using a large number of patients in a multicenter trial (5). The ARDS Network is also in the process of studying another important question, namely the utility of the central venous versus the pulmonary arterial catheter for providing either a fluid liberal or a fluid conservative strategy in patients with ALI/ ARDS. There is considerable equipoise regarding which strategies may be superior. There is also uncertainty regarding the use of either the central venous or pulmonary arterial catheter. The prospective clinical study being carried out by the NIH ARDS Network will provide important answers to these long-standing questions regarding the best supportive therapy for patients with ALI/ARDS. The NIH-supported ARDS Network also has proven to be an excellent vehicle for testing potential new pharmacological therapies for ALI/ARDS (4, 6). It has now been recognized that multicenter studies must be carried out in order to have adequate power to generate convincing results in clinical trials of ALI/ARDS, particularly since mortality in these trials has declined to less than 30%. The clinical network also represents a valuable means for carrying out clinical research studies on patients who are enrolled in the clinical trials. For example, the association of several potentially useful plasma biological markers of lung or systemic injury with clinical outcomes and treatment strategies can be evaluated in these trials. Also, genetic susceptibility to ALI/ARDS can be assessed in these multicenter trials. Recently, there has been some debate regarding the best mechanisms for designing and executing clinical trials in patients with ALI/ARDS. The details have been recently summarized (18). There is an extensive process for evaluating the design and safety of NIH-sponsored clinical trials with several levels of review by the NIH and individual university constituted institutional review boards (19). A recent blue-ribbon panel of five independent experts concluded that the NHLBI sponsored ARDS network trials were safe and well designed (19).

References 1. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342:1334– 1349. 2. Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1988; 138:720–723.

Acute lung injury

561

3. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference of ARDS: definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818–824. 4. The ARDS Network. Ketoconazole for early treatment of acute lung injury and acute respiratory distress syndrome: a randomized controlled trial. JAMA 2000; 283:1995–2002. 5. The ARDS Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342:1301–1308. 6. The ARDS Network. A randomized placebo controlled trial of lisofylline for early treatment of acute lung injury and acute respiratory distress syndrome. Am J Respir Crit Care Med 2000. Submitted. 7. Doyle RL, Szaflarski N, Modin GW, Wiener-Kronish JP, Matthay MA. Identification of patients with acute lung injury. Predictors of mortality. Am J Respir Crit Care Med 1995; 152:1818– 1824. 8. Zilberberg MD, Epstein SK. Acute lung injury in the medical ICU: comorbid conditions, age, etiology, and hospital outcome. Am J Respir Crit Care Med 1998; 157:1159–1164. 9. Abraham E, Matthay MA, Dinarello CA, et al. Consensus conference definitions for sepsis, septic shock, acute lung injury, and acute respiratory distress syndrome: time for a reevaluation. Crit Care Med 2000; 28:232–235. 10. Rubenfeld GD, Caldwell E, Granton J, Hudson LD, Matthay MA. Inter-observer variability in applying a radiographic definition for ARDS. Chest 1999; 116:1347–1353. 11. Meade MO, Cook RJ, Guyatt GH, et al. Interobserver variation in interpreting chest radiographs for the diagnosis of acute respiratory distress syndrome. Am J Respir Crit Care Med 2000; 161:85–90. 12. Monchi M, Bellenfant F, Cariou A, et al. Early predictive factors of survival in the acute respiratory distress syndrome. A multivariate analysis. Am J Respir Crit Care Med 1998; 158:1076–1081. 13. Nuckton TJ, Alonso JA, Kallet RH, et al. Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med 2002; 346:1281–1286. 14. Ware LB, Conner E, Matthay MA. VWF antigen, a marker of endothelial activation and injury, is associated with higher mortality in ALI/ARDS. Am J Respir Crit Care Med 2001; 163:A620. 15. Amato MB, Barbas CS, Medeiros DM, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338:347–354. 16. Matthay M, Zimmerman G, Esmon C, et al. Future research directions in acute lung injury: summary of a NHLBI working group. Am J Resp Crit Care Med 2003; 167:1027–1035. 17. Marshall RP, Webb S, Bellingan GJ, et al. Angiotensin converting enzyme insertion/deletion polymorphism is associated with susceptibility and outcome in acute respiratory distress syndrome. Am J Respir Crit Care Med 2002; 166:646–650. 18. Steinbrook R. How best to ventilate? Trial design and patient safety in studies of the acute respiratory distress syndrome. N Engl J Med 2003; 348:1393–1401. 19. Drazen JM. Controlling research trials. N Engl J Med 2003; 348:1377–1380.

.

INDEX A Active transport, alveolar epithelium, 410–417 Acute lung injury (ALI) (see also Acute respiratory distress syndrome) age, 48–49 clinical manifestations of, 1–2 cytokines, 162–164 definitions, 8–20 diagnostic criteria for, 8–20 epidemiology, 37–50 functional status, 46–48 genetic susceptibility in, 46–48 genomic polymorphisms in, 371–376 high-frequency ventilation for, 610–612 incidence of, 38–43 inhaled nitric oxide and (see also Nitric oxide) interleukins, 116–124 inverse ratio ventilation for, 612–614 mechanical ventilation in, 589–619 mediators in, 116–132 mortality related to, 26–29, 43–46 neutrophils, 124–128 noninvasive ventilation, 608–610 oxygen toxicity, 601–602 pediatric, 49–50 pressure control ventilation for, 614–615 prone position, 563–583 risk factors for, 22–26 surfactant alterations, 538–542 surfactant replacement therapy in, 544–553 prevention of, lung protective strategies in, 589–619 ventilator-associated, 593–600 Acute respiratory distress syndrome (ARDS) (see also Acute lung injury) definitions, 8–20 historical perspective, 1–3 network, clinical trials, 3 oxidants, 164–165, pathology of, 75–107 surfactant deficiency, 3–4 Age, as factor in ARDS, 48–49 ALI (see Acute lung injury)

Index

564

Alveolar edema fluid clearance, 158–161 sampling, 148–149 Apoptosis and ARDS, 106–107, 168–169, 181–196, 338–339 ARDS (see Acute respiratory distress syndrome) B Bronchoalveolar lavage (BAL) in ARDS, 148 C Coagulation abnormalities in ALI and ARDS, 128–132, 165–168, 262–266, 333–336 Corticosteroids treatment for ARDS, 4 D Diffuse alveolar damage (DAD), 75–89 F Fibrosis, lung, 85–92 Fibroproliferation, lung injury, 313–333 G Genetics in acute lung injury, 335–369, 385–402 Genomic polymorphisms in acute lung injury, 371–376 Glucocorticoids treatment for ARDS, 509–528 H Heat shock proteins, 289–305 High-frequency ventilation (HFV) in acute lung injury, 610–612 I Increased-permeability pulmonary edema (see also Acute respiratory distress syndrome) Inflammatory cytokines in acute lung injury, 162–164 Interlukin(s) in acute lung injury, 162–164 M Mechanical ventilation in acute lung injury (see also Acute lung injury, mechanical ventilation in) Multiple organ system failure, 230–232 N Nitric oxide, as gas ARDS, 468–471, 475–477 response in acute lung injury, 471–474 side effects, 474–475 O Oxygenation, prone ventilation effects on, in ARDS, 563–583

Index

565

Oxygen toxicity in ARDS, 601–602 P Pathology, lung, 75–107 Pneumonia in ARDS patients, 442–444 Prone ventilation for ARDS, 563–583 radiograph, 57–61 sampling, 148–149 Pulmonary hypertension in ARDS, 465–467 Pulmonary vascular remodeling, in ARDS, 83–105 Q Quality of life in ARDS survivors, 46–48 R Radiography, chest, in ARDS patients, 57–66 Resolution of alveolar edema in acute lung injury, 417–426 S Severe acute respiratory syndrome (SARS), 633–643 Sepsis acute lung injury, 266–271, 439–450 clinical features, 245–247, 439–441 cytokines, 253–254 definitions, 246–247 pathophysiology, 247–266 platelet activating factors, 254–256 acetylhydrolase, 256–260 thrombosis, 262–266 treatment, 446–454, 521–526 Surfactant, exogenous, ARDS treatment in patients with, 544–553 T Transforming growth factor, 320–325 V Ventilator-induced lung injury, 201–229 von Willebrand’s factor (vWF) antigen, 154–158