Lung Biology in Health & Disease Volume 161 High Altitude: An Exploration of Human Adaption - PDF Free Download (2024)

HIGH ALTITUDE An Exploration of Human Adaptation

Edited by

Thomas F. Hornbein Robert B. Schoene University of Washington School of Medicine Seattle, Washington

Marcel Dekker, Inc.

New York • Basel

TM

Copyright © 2001 by Marcel Dekker, Inc. All Rights Reserved.

This cover illustration was featured on the cover of the Journal of Applied Physiology on the April through June issues 2000 (88:4, 88:5, and 88:6). We acknowledge Steve Graepel for the cover design. This illustration is copyrighted by the Mayo Foundation and reproduced with permission.

ISBN: 0-8247-0313-8 This book is printed on acid-free paper. Headquarters Marcel Dekker, Inc. 270 Madison Avenue, New York, NY 10016 tel: 212-696-9000; fax: 212-685-4540 Eastern Hemisphere Distribution Marcel Dekker AG Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland tel: 41-61-261-8482; fax: 41-61-261-8896 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  2001 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. Current printing (last digit): 10 9 8 7 6 5 4 3 2 1 PRINTED IN THE UNITED STATES OF AMERICA

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. I. 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. Fick, 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. Maurer

ADDITIONAL VOLUMES IN PREPARATION

Therapeutic Targets in Airway Inflammation, edited by N. T. Eissa and D. Huston Respiratory Infections in Asthma and Allergy, edited by S. Johnston and N. Papadopoulos Acute Respiratory Distress Syndrome, edited by M. A. Matthay Upper and Lower Respiratory Disease, edited by J. Corren, A. Togias, and J. Bousquet Venous Thromboembolism, edited by J. E. Dalen Acute Exacerbations of Chronic Obstructive Pulmonary Disease, edited by N. Siafakas, N. Anthonisen, and D. Georgopolous Lung Volume Reduction Surgery for Emphysema, edited by H. E. Fessler, J. J. Reilly, Jr., and D. J. Sugarbaker

The opinions expressed in these volumes do not necessarily represent the views of the National Institutes of Health.

INTRODUCTION

Adaptation has always been the subject of considerable interest and debate. Philosophers, students of religion, and scientists of many disciplines have devoted their energies to investigating and explaining where our planet (our world) and we (as human beings) came from, and how we have responded to terrestrial changes. The history of life is indeed a fascinating and mysterious subject. Great names have been associated with it. Among many, two can be singled out: Charles Darwin, who aimed at understanding biological variations and natural selection, and Claude Bernard, who introduced the concept of ‘‘internal environment,’’ which postulates that human beings can achieve a life that is independent of the endless changes in external conditions. To a point, the truth about life remains an open question, but most likely it is an equilibrium between the external, environmental influences and forces and the relative stability of the ‘‘milieu interieur.’’ Perhaps the best illustration of how such an equilibrium has been reached by the human species is found in its adaptation to high altitude, which is the subject of this volume. The reader need only focus on the first chapter to recognize the complexity of the adaptive processes that permit us to ‘‘function’’ at high altitude. It shows that questions about living in thin air have been asked for centuries, even millennia. Because of our great desire to conquer living at high altitude, an understanding of iii

iv

Introduction

how we adapt to that environment has been achieved, but much still remains to be uncovered. Ever since the conquest of the highest Himalayan peaks in the second half of the last century, the biology of adaptation to higher altitude has been of considerable interest. Great scientists from all biological disciplines have devoted years of work to these studies. All have worked in academe, some have explored and conquered peaks never reached before. Many of them have contributed to this book. Their scientific achievements and their first-hand experiences present the reader with an exclusive opportunity to explore a biological/medical area of unique dimension. The Lung Biology in Health and Disease series of monographs has presented many new and challenging topics to its readership, always with the goal of reaching new peaks! Both clinical and basic research subjects have been included: we even explored the human response to the undersea world. High Altitude: An Exploration of Human Adaptation, edited by Thomas F. Hornbein and Robert B. Schoene, truly takes us to one of the highest summits. For this, and for the contributions included in this volume, I am extremely grateful to the authors and editors. Claude Lenfant, M.D. Bethesda, Maryland

PREFACE

Great things are done when men and mountain meet. This is not done by jostling in the street. —William Blake

Maybe it is the proximity to the heavens or perhaps the distance from where most of us live out our lives. The peaks and steppes of high altitude harbor mystery. Something about those regions high above the level of the sea attracts some of us lowlanders to venture and to explore. This book is the creation of just such individuals, those who find in high places not only beauty but also the uncertainty that catalyzes the search for understanding about how humans (and other organisms) adapt in order to survive and even thrive where the pressure of air and hence the partial pressure of its oxygen are diminished. This volume is about how humans respond acutely to, adapt to, and sometimes fail to adapt to high altitude, in particular to the hypoxia of high altitude. We define high altitude as elevations as low as 2000 meters (6500 feet) to those as high as 8850 meters (approximately 29,000 feet). First, this book is not a comprehensive review of the literature about our human affair with thin air. Rather, you will find here a series of essays written by individuals at the forefront of studying the effects of high altitude on the human v

vi

Preface

body. One major goal has been to review critically and selectively, rather than comprehensively, the existing knowledge that is most important to our current understanding. These reviews are freely seasoned with the authors’ (as well as others’) views, hypotheses, and fantasies about how things might work. Our other major intent is to identify some of the black holes in our current knowledge, those areas wanting further exploration. The thinning of the air over our heads as we climb higher above the sea has a number of consequences: the air grows colder and the rays from the sun become more intense. But most germane to the intent of this volume is that the weight of the air above lessens and with it the pressure of oxygen in the air we breathe. At the summit of Mount Everest, the barometric and oxygen pressures are approximately one-third of what we breathe at sea level. The natural laboratory of high altitude enables us to study the impact of exogenous hypoxia on healthy humans to better distinguish the effect of hypoxia from that of cardiopulmonary and other diseases that can also cause a person to be hypoxic. The 25 chapters comprising this volume encompass four domains: 1.

2.

3.

4.

The stage. Chapters 1 to 3 touch on the history of our inquiry into human existence at high altitude, the nature of our atmosphere and how it came to be, and an anthropological perspective on those who dwell permanently at great heights. Organism defense. Chapters 4 and 5 explore how the cells within the body detect that the oxygen they need is in short supply (oxygen sensing) and how they defend themselves against this stress. Oxygen’s journey from air to mitochondrion. The physiological journey to high altitude is expressed by the traverse of molecular oxygen down a gradient from the air we breathe to the mitochondria, where oxygen fuels aerobic metabolism. This path comprises two conductive systems, one that moves gas (the pulmonary system) and one that transports oxygen-containing blood (the circulatory system) and a diffusive component that fills in the gaps (from alveoli to blood, from capillary blood to tissue mitochondria). In addition to lungs, heart, vessels, and blood, many organs play a role in sensing, facilitating, or modulating this oxygen flux. Chapters 6 through 21 explore the role of individual organs as well as their integrated function in enabling physical and mental performance at high altitude. Maladaptation. Chapters 22 to 25 describe failure to thrive at high altitude: acute mountain sickness and high-altitude cerebral and pulmonary edema, chronic mountain sickness, and the challenges faced by lowland dwellers venturing to high altitude who have preexisting medical conditions.

So, where do we go from here? Even during the not-so-brief gestation of this volume, we have watched the nature of high-altitude research evolve, changing in a manner predictable from the directions in which biomedical research in general is unfolding.

Preface

vii

As is apparent from many of the chapters in this book, one inevitable evolution is toward understanding not just what is happening, but how. • How is a change in the partial pressure of oxygen sensed by individual cells in multiple tissues such as the carotid body, renal EPO fibroblasts, pulmonary and other vessels, and a wide array of other organs? • How is this sensed signal genetically expressed (HIF, etc.?) and then transcribed into biochemical messages (VEGF, NO, and myriad more) within cells that tell organs how to defend themselves against hypoxia? • How is the magnitude of the response modulated and why does the response sometimes become inappropriate to the organism’s best interests? Are there genetic antecedents contributing to the diversity—both advantageous and disadvantageous—of response? But the other side of the coin—the what—is still far from complete and much information is needed simply to describe adequately what is going on in acclimatizing lowlanders and permanent high-altitude inhabitants. Geographical, ethnic, gender, and aging effects are incompletely characterized. With increasing numbers of lowlanders traveling to high altitudes for play or work, some with regular intermittent exposures, we need to know how they fare. Studies will require epidemiological approaches abetted by use of genetic markers and applied to larger populations than have been studied thus far. Elucidation of mechanisms at a fundamental level may help connect these explorations of hypoxia and high altitude with the other facets of the molecular machinery that underpins our existence. In the prescient words of John Muir, ‘‘When we try to pick out anything by itself, we find it hitched to everything else in the universe.’’ Thomas F. Hornbein Robert B. Schoene

CONTRIBUTORS

Alberto Arregui, M.D., Ph.D. Professor, Department of Medicine, Cayetano Heredia University, Lima, Peru Peter Ba¨rtsch, M.D. Professor, Division of Sports Medicine, Department of Internal Medicine, University Hospital, Heidelberg, Germany George A. Brooks, Ph.D. Professor, Department of Integrative Biology, University of California, Berkeley, California Gail E. Butterfield, Ph.D., R.D.* Director, Nutritional Studies, Nutritional Research, GRECC, Palo Alto VA Medical Center, Palo Alto, California Neil S. Cherniack, M.D. Professor, UMDNJ–New Jersey Medical School, Newark, New Jersey Kevin P. Davy, Ph.D. Assistant Professor, Department of Health and Exercise Science, Colorado State University, Fort Collins, Colorado * Deceased.

ix

x

Contributors

Jerome A. Dempsey, Ph.D. Professor, The John Rankin Laboratory of Pulmonary Medicine, Department of Preventive Medicine, University of Wisconsin, Madison, Wisconsin Diana Depla Newcastle Upon Tyne, England Henry Gautier, M.D. Professor, Atelier de Physiologie Respiratoire, Faculte´ de Me´decine Saint-Antoine, Paris, France Howard J. Green, Ph.D. Professor, Department of Kinesiology, University of Waterloo, Waterloo, Ontario, Canada Robert F. Grover Professor Emeritus, University of Colorado, Denver, and Arroyo Grande, California Peter H. Hackett, M.D., F.A.C.E.P. President, International Society for Mountain Medicine, Ridgway, Colorado Peter W. Hochachka, O.C., Ph.D., D.Sc., F.R.S.C. Professor, Departments of Zoology, Radiology, and Medicine, University of British Columbia, Vancouver, British Columbia, Canada Thomas F. Hornbein, M.D. Professor and Chairman Emeritus, Department of Anesthesiology, and Professor of Physiology and Biophysics, University of Washington School of Medicine, Seattle, Washington Charles S. Houston, M.D. Professor Emeritus, University of Vermont, Burlington, Vermont Herbert N. Hultgren, M.D.* Professor Emeritus, Department of Medicine, Stanford University School of Medicine, Palo Alto, California Pamela Parker Jones, Ph.D. Research Assistant Professor, Department of Kinesiology and Applied Physiology, University of Colorado, Boulder, Colorado Bengt Kayser, M.D., Ph.D. Faculty of Medicine, University of Geneva, Geneva, Switzerland Sukhamay Lahiri, D.Phil. Professor, Department of Physiology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania

* Deceased.

Contributors

xi

Fabiola Leo´n-Velarde, D.Sc. Professor, Department of Physiology, Cayetano Heredia University, Lima, Peru Benjamin D. Levine, M.D. Director, Institute for Exercise and Environmental Medicine and University of Texas Southwestern Medical Center, Dallas, Texas Richard T. Meehan, M.D., F.A.C.P., F.A.C.R. Department of Medicine, National Jewish Medical Research Center, Denver, Colorado Joseph Milic-Emili, M.D. Professor Emeritus, Departments of Physiology and Medicine, McGill University, Montreal, Quebec, Canada Carlos C. Monge Professor Emeritus, Department of Physiology, Cayetano Heredia University, Lima, Peru Lorna G. Moore, Ph.D. Professor, Department of Anthropology, University of Colorado, and Department of Medicine, University of Colorado Health Sciences Center, Denver, Colorado Susan Niermeyer, M.D. Associate Professor, Department of Pediatrics, University of Colorado Health Sciences Center, Denver, Colorado Marcus E. Raichle, M.D. Professor, Departments of Radiology and Neurology, Washington University Medical Center, St. Louis, Missouri John T. Reeves, M.D. Professor, Departments of Medicine and Pediatrics, University of Colorado Health Sciences Center, Denver, Colorado Jean-Paul Richalet, M.D., D.Sc. Professor, Faculte´ de Medicine de Bobigny, Universite´ Paris 13, Bobigny, France Robert Roach, Ph.D. Research Associate Professor, Department of Life Sciences, New Mexico Highlands University, Las Vegas, New Mexico Clarence F. Sams, Ph.D. Research Biochemist, Life Science Research Laboratories, NASA–Johnson Space Center, Houston, Texas Robert B. Schoene, M.D. Professor, Department of Medicine, University of Washington School of Medicine, Seattle, Washington Douglas R. Seals, Ph.D. Professor, Department of Kinesiology and Applied Physiology, University of Colorado, Boulder, Colorado

xii

Contributors

John W. Severinghaus, M.D., F.R.C.A. Professor Emeritus, Department of Anesthesiology, University of California, San Francisco, California Curtis A. Smith, Ph.D. Senior Scientist, The John Rankin Laboratory of Pulmonary Medicine, Department of Preventive Medicine, University of Wisconsin, Madison, Wisconsin Kurt R. Stenmark, M.D. Professor, Department of Pediatrics, University of Colorado Health Sciences Center, Denver, Colorado John R. Sutton, M.D., F.R.C.P.* Professor, Exercise Research Centre, University of Sydney, Sydney, Australia Erik R. Swenson, M.D. Associate Professor, Department of Medicine, University of Washington School of Medicine, Seattle, Washington Peter N. Uchakin, Ph.D. Lead Research Scientist, Division of Basic Medical Sciences, Mercer University School of Medicine, Macon, Georgia Peter D. Wagner, M.D. Professor, Department of Medicine, University of California, San Diego, La Jolla, California John B. West, M.D., Ph.D., D.Sc., F.R.C.P., F.R.A.C.P. Professor, Department of Medicine, University of California, San Diego, La Jolla, California John V. Weil, M.D. Director, CVP Research Lab, and Professor, Department of Medicine, University of Colorado Health Sciences Center, Denver, Colorado David P. White, M.D. Circadian, Neuroendocrine, and Sleep Disorder Section, Brigham and Women’s Hospital, Boston, Massachusetts Eugene E. Wolfel, M.D. Professor, Division of Cardiology, Department of Medicine, University of Colorado Health Sciences Center, Denver, Colorado Stacy Zamudio, Ph.D. Assistant Professor, Department of Anthropology, University of Colorado, and Department of Anesthesiology, University of Colorado Health Sciences Center, Denver, Colorado

* Deceased.

CONTENTS

Introduction Preface Contributors

Claude Lenfant

1. The Growth of Knowledge About Air, Breathing, and Circulation as They Relate to High Altitude

iii v ix

1

Charles S. Houston I. II. III. IV. V. VI. VII. VIII. IX.

Introduction First Experiences on High Mountains Something in Mountain Air Was Responsible How Did the Atmosphere Relate to Lungs and Heart? What Was the Atmosphere Made Of? Why Breathe—and How? How Does Oxygen Pass from Lungs into Blood? How Is Oxygen Carried in Blood? What Was the Relationship of Oxygen to Mountain Sickness?

1 2 4 5 7 9 10 13 13 xiii

xiv

Contents X. XI.

2.

Finally—The Cause(s) of Mountain Sickness Summary References

The Atmosphere

14 20 21 25

John B. West I. II. III. IV.

3.

Evolution of the Atmosphere Altitude and Barometric Pressure Factors Other Than Barometric Pressure at High Altitude Artificial Atmosphere at High Altitude References

The People

25 28 37 38 41 43

Susan Niermeyer, Stacy Zamudio, and Lorna G. Moore I. II. III. IV.

4.

Introduction History of Human High-Altitude Habitation High-Altitude Adaptation Across the Life Cycle Summary and Conclusions References

Cellular and Molecular Mechanisms of O2 Sensing with Special Reference to the Carotid Body

43 45 50 83 88

101

Sukhamay Lahiri and Neil S. Cherniack I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.

Introduction Physiological Clues to the Identity of O2-Sensing Molecules Tissue Po2 and O2 Sensing The Arterial Chemoreceptors—Carotid and Aortic Bodies Effects of Hypoxia on the Type I Cell Membrane Metabolic Effects of Hypoxia on Type I Cells Models of Chemoreception in the Carotid Body Effects of Hypoxia on Smooth Muscle and Neurons Changes in the Ventilatory Response to Hypoxia with Time Erythropoietin Production During Hypoxia Ventilatory Acclimatization to Hypoxia Blunted Ventilatory and Carotid Chemosensory Function with Prolonged Hypoxia XIII. Effect of Chronic Hypoxia on the Carotid Body and Carotid Body Cells In Vivo and In Vitro XIV. Effects of Hypoxia on Gene Expression References

101 102 103 104 107 109 110 112 113 114 115 116 117 119 122

Contents

xv

5. Molecular/Metabolic Defense and Rescue Mechanisms for Surviving Oxygen Lack: From Genes to Pathways

131

Peter W. Hochachka I. II. III. IV. V.

Cell Level Paradigms for Hypoxia Tolerance Detecting When Oxygen Becomes Limiting ATP Demand Pathway ⫽ ATP Supply Pathway During Hypoxia Coupling Metabolism and Membrane Functions During Hypoxia Hypoxic Sensitivity of Protein Synthesis References

6. Control of Breathing at High Altitude

131 132 133 134 135 136 139

Curtis A. Smith, Jerome A. Dempsey, and Thomas F. Hornbein I. II. III. IV. V. VI. VII.

Introduction Background and Definitions Acute Responses to Hypoxia Short-Term Acclimatization Long-Term Acclimatization Exercise Hyperpnea at High Altitude Breathing and Human Performance at High Altitude References

7. Mechanics of Breathing

139 140 140 149 155 158 160 163 175

Joseph Milic-Emili, Bengt Kayser, and Henry Gautier I. II. III. IV.

Introduction Lung Volumes Mechanical Properties of the Respiratory System Work of Breathing References

8. Gas Exchange

175 176 187 191 194 199

Peter D. Wagner I. II. III. IV.

Gas Exchange at Rest Gas Exchange During Exercise Effects of Acclimatization on Pulmonary Gas Exchange Muscle Tissue Gas Exchange References

201 211 225 227 232

xvi 9.

Contents The Cardiovascular System at High Altitude: Heart and Systemic Circulation

235

Eugene E. Wolfel and Benjamin D. Levine I. II. III. IV. V. VI. VII. VIII. IX. X.

Introduction Acute Hypoxia Sustained Hypoxia Cardiovascular Function During Exercise Acute Hypoxia—Submaximal Exercise Acute Hypoxia—Maximal Exercise Sustained Hypoxia—Submaximal Exercise Sustained Hypoxia—Maximal Exercise High-Altitude Residents and Populations Clinical Correlation References

10. The Pulmonary Circulation at High Altitude

235 237 242 262 262 265 266 269 273 278 283

293

John T. Reeves and Kurt R. Stenmark I. II. III. IV. V. VI.

Introduction History Mechanisms Hemodynamics in Exercising Adult Humans Schema and Unanswered Questions Teleology References

11. Cerebral Circulation at High Altitude

293 294 303 320 332 334 335

343

John W. Severinghaus I. II. III. IV. V. VI. VII. VIII. IX.

Introduction Brain Oxidative Biochemistry and Its Effect on Flow Normal Cerebral Blood Flow Regulation CBF Changes During Acclimatization Cerebral Circulation with Prolonged Hypoxia CBF in AMS and HACE: Cause or Effect? Brain Injuries at Extreme Altitude Personal Comment Regarding Hypoxic Distress Summary References

343 344 344 355 363 364 366 367 367 368

Contents

xvii

12. The High-Altitude Brain

377

Marcus E. Raichle and Thomas F. Hornbein I. II. III. IV. V. VI. VII. VIII. IX. X.

Introduction Effect of Acute Hypoxia on Behavior Effect of More Sustained Hypoxia on Behavior Residual Behavioral Effects of Exposure to Hypoxia Sea Level Forms of Chronic Hypoxemia Normal Brain Function Metabolic Requirements of Cognition Metabolic Requirements of Sleep Neurobiology of the High-Altitude Brain Suggestions for Future Research References

13. Autonomic Nervous System

377 378 381 385 388 390 395 400 401 411 412

425

Douglas R. Seals, Pamela Parker Jones, and Kevin P. Davy I. II. III. IV.

Sympathoadrenal System Parasympathetic Nervous System Influence of Age, Gender, and Physical Activity/Fitness Status Concluding Remarks References

14. The Effects of Altitude on Skeletal Muscle

426 434 437 439 439

443

Howard J. Green and John R. Sutton I. II. III.

Altitude and Skeletal Muscle Adaptations in the Muscle Cell Hypoxia and Muscle Metabolism References

15. Blood

443 450 469 485

493

Robert F. Grover and Peter Ba¨rtsch I. II.

Blood Oxygen Transport Blood Coagulation and Fibrinolysis References

493 509 517

xviii

Contents

16. Renal Function and Fluid Homeostasis

525

Erik R. Swenson I. II. III. IV.

Introduction Effects of Hypoxia on Salt and Water Balance Mechanisms of Changes in Salt and Water Balance at High Altitude Conclusions References

17. Metabolic Response of Lowlanders to High-Altitude Exposure: Malnutrition Versus the Effect of Hypoxia

525 526 532 547 549

569

George A. Brooks and Gail E. Butterfield I. II. III. IV. V. VI. VII.

Introduction Acute Altitude Exposure Acclimatization Malnutrition Studies Uncomplicated by Malnutrition Metabolic Consequences of Altitude Exposure Without Malnutrition Nutrient Recommendations for High Altitude References

18. The Endocrine System

569 570 571 573 574 579 595 595 601

Jean-Paul Richalet I. II. III. IV. V. VI. VII. VIII.

Introduction Hypoxia and Signal Processing in Hormonal Systems Hormones of Fluid Homeostasis Thyroid Hormones Hormones of Calcium and Phosphate Balance Hormones Controlling Blood Glucose Other Hormones Summary and Future Directions References

19. High Altitude and Human Immune Responsiveness

601 602 607 620 622 623 624 632 632 645

Richard T. Meehan, Peter N. Uchakin, and Clarence F. Sams I. II. III. IV.

Introduction Infections at High Altitude Survey of Human High-Altitude Immunology Studies Discussion

645 650 651 654

Contents V.

xix Conclusions References

20. Exercise and Hypoxia: Performance, Limits, and Training

656 658 663

Robert Roach and Bengt Kayser I. II. III. IV. V. VI. VII.

Introduction Work in Hypoxia Anaerobic Metabolism Climbing Everest Without Supplementary Oxygen Altitude Training for Endurance Exercise Effects of Life-Long Acclimatization Where To Now? References

21. Sleep

663 665 678 687 688 689 691 694 707

John V. Weil and David P. White I. II. III. IV. V. VI.

Introduction Physiological Setting Sleep Disturbance at Altitude Mechanisms of Periodic Breathing at Altitude Treatment Functional Significance References

707 707 708 714 723 726 726

22. Acute Mountain Sickness and High-Altitude Cerebral Edema

731

Peter Ba¨rtsch and Robert Roach (with Diana Depla) I. II. III. IV. V. VI. VII.

Introduction AMS High-Altitude Cerebral Edema Pathophysiology of AMS and HACE Retinal Hemorrhages Prevention and Treatment Research Directions References

23. High-Altitude Pulmonary Edema

731 732 740 741 755 758 762 763 777

Robert B. Schoene, Herbert N. Hultgren, and Erik R. Swenson I. II. III. IV.

Introduction History Hypothetical Model Clinical and Physiological Correlation

777 778 778 780

xx

Contents V. VI. VII. VIII. IX. X. XI. XII. XIII.

Pathological Findings Pathophysiology Hemodynamics Site of Leak Alveolar Fluid Inflammation Alveolar Epithelial Fluid Clearance Prevention and Therapy Directions References

24. Chronic Mountain Sickness in Andeans

787 787 788 793 795 798 803 806 806 807 815

Carlos C. Monge, Fabiola Le´on-Velarde, and Alberto Arregui I. II. III. IV. V. VI. VII. VIII. IX.

Introduction Brief Historical Background Clinical and Epidemiological Aspects Mechanisms of CMS Organ Effects Prevention and Treatment Biological Basis of CMS Conclusions Future Directions References

25. High Altitude and Common Medical Conditions

815 816 816 822 826 830 831 831 832 833 839

Peter H. Hackett I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.

Introduction Stresses of the High-Altitude Environment Effects of High Altitude on Common Illnesses Ventilatory Conditions Cardiovascular Problems Hematological Problems Neurological Conditions Diabetes Mellitus Ophthalmological Conditions Altitude and Pregnancy Alcohol at Altitude Future Directions References

Author Index Subject Index

839 840 841 841 848 862 863 866 867 868 873 875 876 886 963

TO THREE WITH WHOM WE SHARED THIS ADVENTURE, WHO FINISHED THEIR OWN JOURNEY TOO SOON: GAIL BUTTERFIELD, scientist, teacher, mentor, and role model, particularly for young women who would follow the path she traveled; HERB HULTGREN, clinician, scholar, gentle guide with impish curiosity, who taught so many of us the joy of exploration; JOHN SUTTON, cauldron of ideas, bubbling energy and enthusiasm, who never understood the phrase ‘‘It can’t be done.’’

1 The Growth of Knowledge About Air, Breathing, and Circulation as They Relate to High Altitude

CHARLES S. HOUSTON University of Vermont Burlington, Vermont

I.

Introduction

Many thousands of years ago, humans realized that an uninterrupted supply of good air was necessary for most life. From accidental as well as deliberate experiments it became clear that humans would die if kept from breathing for only a few minutes. Some philosophers asked why this should be; the answer was many centuries in arriving. Meanwhile, they pondered what happened to air when it was breathed: where did it go, and how did it get there? Soon the ingenious network of lungs and blood vessels was explored and various theories described. Not for many centuries would the wonderful machinery of respiration and circulation begin to be understood. At first only a few imaginative scholars believed that invisible, impalpable air actually had substance and weight, and it took a thousand years to prove this—and to show that air weighed less the higher one climbed. That portion of air that was necessary for life was described and eventually its chemistry defined. From there it was a shorter journey to discover how this vital stuff was transported and eventually used, and its residue discarded. While these voyages of discovery were in process, curious adventurers were 1

2

Houston

pushing across the world, on the oceans, the deserts and jungles, and even onto high mountains, where they described the illnesses that befell them. After all that had gone before, it was only a small leap from these observations to the composition of air and the variability of its weight, and bingo! Thin air, lacking some vital stuff, caused mountain sickness. This connection was possible only because some explorers studied what their forebears had learned of respiration and circulation and the physics of the invisible atmosphere they were penetrating. Clearly, the history of altitude illness is part of the great fabric of knowledge describing the characteristics of air and blood, of transporting and using a necessity of life. To really comprehend altitude sickness, we must understand how we learned about the physiology of life.

II. First Experiences on High Mountains The story of humans and altitude extends far back into the past (1), as our feelings changed from awe, worship, and fear to curiosity and love. Two thousand years ago Chinese poets and artists rapturously described the beauty, the peace, the majesty of mountains. Not until some of the adventuresome went high did they become aware of sickness on high mountains. The ancient Greeks also revered high places: Zeus and his consorts lived on Mt. Olympus; occasionally they might descend for a little earthly dalliance. In the Andes shrines were built on high peaks because the souls of the dead lived there and protected humans, animals, and crops. The remains of stone shelters indicate that people lived for a time and did manual work as high as 22,000 feet, where perfectly preserved mummies have been found (2). In ancient China a high official of the Western Pan Dynasty, Too Kin, wrote Emperor Chung ti (37–32 bc) describing the dangers encountered when crossing the western mountains and deserts to Kashmir (3): ‘‘Next one comes to the Big Headache and Little Headache Mountains as well as the Red Earth and Swelter Hills. They make a man so hot his face turns pale, his head aches, and he begins to vomit. Even the swine react this way.’’ A more explicit description of one form of mountain sickness was written by Buddhist missionary Fa-Hsien (334–420 AD) while crossing a pass some 12,000– 14,000 feet high (4): Fa-Hsien and the two others proceeding southwards, crossed the Little Snowy Mountains. On them the snow lies accumulated both winter and summer. On the north side of the mountains, in the shade, they suddenly encountered a cold wind which made them shiver and unable to speak. Hwuy-Ring could not go any farther. A white froth came from his mouth and he said to Fa-Hsien, ‘‘I cannot live any longer. Do you immediately go away, that we do not all die here’’; and with these words he died.

Hwuy-Ring probably died from a combination of hypothermia and what we now recognize as high altitude pulmonary edema. The party was crossing either the Kilik

History of High Altitude Illness

3

pass or the Ulagh Rabat pass, both higher than 14,000 feet. Three centuries later another Buddhist, Xuan Zang (602–664 ad ), described crossing the mighty Tien Shan, Kun Lun, and Karakoram ranges (5): The journey is arduous and dangerous and the wind dreary and cold. Travellers are often attacked by fierce dragons so that they should neither wear red garments nor carry gourds with them, nor shout loudly. Even the slightest violation of these rules will invite disaster.

The dragon theme recurs in other countries. In Europe many centuries later, reputable scientists took notarized statements from people who had been attacked by an extraordinary variety of dragons (6). From 1559 to 1654 a distinguished naturalist, Ulisse Aldrovandi, and his heirs published 13 volumes of beautiful drawings of these monsters. Whatever they were, dragons seem to have experienced an abrupt extinction less than 300 years ago! Other travelers went high enough to experience mountain sickness. Father Alonzo Ovalde, in an account of his travels in the Andes at the end of the sixteenth century, wrote (7): When we come to ascend the highest point of the mountain, we feel an aire so piercing and subtile that it is with much difficulty we can breathe, which obliges us to fetch our breath quick and strong and to open our mouths wider than ordinary, applying to them likewise our handkerchiefs to protect our mouth and break the extreme coldness of the air and to make it more proportionable to the temperature which the heart requires, not to be suffocated; this I have experienced every time I have passed this mighty mountain.

His better known contemporary, Father Jose Acosta, described his misery while crossing a pass called Pariacaca (15,750 feet) in the Peruvian Andes (8): [W]hen I climbed the Escaleras (de Pariacaca) . . . I felt such a deadly pain I was ready to hurl myself from the horse onto the ground. . . . And almost immediately there followed so much retching and vomiting that I thought I would lose my soul, because after what I ate and the phlegm, there followed bile and more bile both yellow and green so that I brought up blood from the violence I felt in my stomach. . . . [I]t is not only . . . the Paricaca pass which produces this effect but also . . . the entire mountain range . . . and much more for those who ascend from the sea coast to the mountain than for those who return from the mountain to the plains . . . *

In another translation, Acosta is quoted as writing, like Ovalde: ‘‘I therefore perswade myselfe that the element of the aire there is so subtile and delicate as it is not proportionable with the breathing of man’’ (9). In the the third century before the Christian era, Xenophon and Alexander had experienced the synergistic effects of altitude, cold, hunger, and dehydration, although they did not recognize the effects of altitude alone. * Gilbert explains how corrupted translations have altered this, the original text, to imply that it is the first description of acclimatization.

4

Houston

In the fourteenth and fifteenth centuries, the Mongol Hordes rampaged across Central Asia, high Tibet, and into Europe, crossing deserts and high mountains. Mirza Muhammad Haider, one of the Mongol chieftains, described the hazards of altitude in perceptive detail (10): Another peculiarity of Tibet is the dam-giri which the Moghuls call yas and which is common to the whole country, though less prevalent in the region of forts and villages. The symptoms are a feeling of severe sickness (nakhushi) and in every case one’s breath so seizes him that he becomes exhausted, just as if he had run up a steep hill with a heavy burden on his back. On account of the oppression it causes, it is difficult to sleep. Should, however, sleep overtake one, the eyes are hardly closed before one is awake with a start caused by the oppression of the lungs and chest. . . . When overcome by this malady the patient becomes senseless, begins to talk nonsense, and sometimes the power of speech is lost, while the palms of the hands and the soles of the feet become swollen. Often, when this last symptom occurs, the patient dies between dawn and breakfast time; at other times he lingers on for several days. . . . This malady only attacks strangers; the people of Tibet know nothing of it, nor do their doctors know why it attacks strangers. Nobody has ever been able to cure it. The colder the air, the more severe is the form of the malady.

Haider’s account is notable in several respects: he described several important signs and symptoms of mountain sicknesses, he noted that it also affects horses, and he clearly recognized that something protected life-long altitude residents—perhaps the first unequivocal mention of acclimatization. III. Something in Mountain Air Was Responsible Mountain dwellers attributed the sickness in newcomers to emanations from ores bearing antimony or to noxious fumes from a variety of plants and shrubs. But a few later explorers suggested that changes in the atmosphere were responsible. One of these was Horace Benedict de Saussure, a broadly educated philosopher-scientist who made the second ascent of Mont Blanc (15, 771 feet) in 1787. During the next few years, impressed by his sensations on the summit, he studied his pulse, respirations, temperature, and symptoms on that and other mountains. Using a torricellian barometer he found that air weighed less at altitude and attributed his symptoms to the decreased density (11): The sort of weariness which proceeds from the rarity of the air is absolutely insurmountable; when it is at its height, the most imminent peril will not make you move a step faster. . . . Since the air (on the summit of Mont Blanc) had hardly more than half of its usual density, compensation had to be made for the lack of density by the frequency of inspirations. . . . That is the cause of the fatigue that one experiences at great heights. For while the respiration is accelerating, so also is the circulation.

de Saussure may not have heard of the discovery of oxygen 10 years earlier, but he clearly recognized that the difference in air was at least part of the cause of his symptoms.

History of High Altitude Illness

5

Another vivid picture of mountain sickness is that by Friedrich von Tschudi, who described his nasty experience while exploring the Andes in 1838–1842 (12): My panting mule slackened his pace, and seemed unwilling to mount a rather steep ascent which we had now arrived at. To relieve him I dismounted, and began walking at a rapid pace. But I soon felt the influence of the rarefied air, and I experienced an oppressive sensation which I had never known before. I stood still for a few moments to recover myself, and then tried to advance. My heart throbbed audibly; my breathing was short and interrupted. A world’s weight seemed to lie upon my chest; my lips swelled and burst; the capillaries of my eyes gave way, and blood flowed from them. In a few moments my senses began to leave me. I could neither see, hear, nor feel distinctly. A grey mist floated before my eyes, and I felt myself involved in that struggle between life and death which, a short time before, I fancied I could discern on the face of nature. Had all the riches of earth, or the glories of heaven, awaited me a few hundred feet higher, I could not have stretched out my hand toward them. In that half senseless state I lay stretched on the ground until I felt sufficiently recovered to remount my mule.

Such severe symptoms struck some but spared others and was said to be more common in some areas than in others. It was all very puzzling. Pieces of the puzzle had been lying about for centuries and others were being discovered, but it would be necessary to assemble them—together with parts of a greater puzzle—before the picture could become more clear. Ovalde, Acosta, De Saussure, and Tschudi all came close to the truth when they wrote of ‘‘subtle air’’ or its thinness or lack of density. But a key piece was missing. IV. How Did the Atmosphere Relate to Lungs and Heart? The greater puzzle was the intricate relationship between air, breathing, blood, and life. What might there be (or not be) on high mountains that so affected travelers? Was there really a poison in mountaintop air? Or was some stuff lacking up there? The importance of the invisible, impalpable stuff in which we live like fish in water had been recognized by the Sumerians several thousand years before the Christian era. The ancient Chinese taught that proper breathing (lien ch’i) changed air into the soul substance or ‘‘vital essence’’ (1,13), but they did not know what this was or how it entered the body. Although dissection of the human body was forbidden during the First Dynasty (3000 bc), the Egyptians gained an approximate idea of human anatomy from embalming the dead. Gradually they identified the heart as the center of an elaborate system of passages or vessels, some of which carried blood, some air. The Egyptian Ebers Papyrus (1700 bc) clearly describes how air enters the nose, flows to the lungs and from there to the heart (13). Perhaps from such a papyrus came the beginning of ‘‘western’’ cardiopulmonary physiology. During a few incredibly creative centuries shortly before Christ, a brilliant group of philosopher-scientists struggled to understand the relationships between

6

Houston

four elements of which all life consisted. These were the same elements described in the ancient Hindu Vedas, or sacred scripts: air, earth, water, and fire. Of them, only air could not be detected, but not surprisingly the idea of air was commingled with the reality of blood: both were essential for life (13). Anaximenes of Miletus in 455 bc believed that air was formed from water as vapor or mist and was alive. As the ancients had, he suggested that in its invisible form air sustains the soul, which animates life, even though it could not be seen or felt. During a discussion of what filled empty space, Empedocles (490–430 bc) used a clepsydra (a vessel with one small hole in the top and several small holes in the bottom) to show that when the vessel was held under water with the hole in the top closed, water could not enter until the top was opened. Conversely, once filled with the top closed, water would not flow out. This led him to conclude that air had substance; he proposed that it entered the body not only thorough the nose and mouth to the lungs, but also through the skin of the entire body. He also suggested that blood ebbed and flowed through the body, an idea put forward by the Chinese many centuries earlier. Hippocrates accepted Empedocles’ theory of the circulation and believed that the purpose of respiration was to cool the heart with air pumped out of the lungs by the auricles into the right side of the heart via the pulmonary artery and also through the pulmonary vein to the left auricle. Diogenes of Appolonia wrote the first systematic description of the cardiovascular system in which some vessels carried air as well as blood. Plato and his student Aristotle accepted Empedocles’ belief that air was carried from the lungs to the heart in both veins and arteries. Erasistratus was greatly influenced by Strato, who argued that all substance included some empty space and from this hypothesized the existence of a vacuum. Erasistratus concluded that when the heart relaxed and enlarged in diastole, air would rush through the pulmonary artery into the left ventricle to prevent a vacuum. There air was somehow changed or combined with blood into a ‘‘vital spirit,’’ which was distributed in the arteries (14). Written early in the first century bc, the Chinese Huan-ti-nei-ching medical texts seemed to describe the circulation of the blood, (although various translations differ), and portions of the text suggest that air flowed in the ‘‘conduits’’ mixed together with blood, much like what the Greeks later described as pneuma: ‘‘. . . the heart regulates all the blood in the body . . . the blood current flows continuously in a circle and never stops. . . . a circle with no beginning and no end. Blood flows six inches with one respiration making a complete circuit of the body about 50 times in 14 hours . . . ’’ (13,15). Galen (130–199 ad ), arguably the best and most influential physician of the Greco-Roman school, whose dictates dominated medicine for a thousand years, wrongly concluded from animal vivisection that air flowed from lungs into the left side of the heart, where it both fuelled and cooled the ‘‘innate heat’’ which was life itself. There it combined with blood and formed the pneuma or ‘‘vital spirit’’ (14). This was pumped through the arteries, which ‘‘in the whole body . . . communicate

History of High Altitude Illness

7

with the veins and exchange air and blood with them through extremely fine invisible openings.’’ To complete the circuit, Galen had to postulate tiny pores between the two ventricles, an hypothesis that persisted for 10 centuries (16). In mid-thirteenth century Ibn al Nafis (1210–1288) more correctly described the anatomy of the pulmonary circulation and reaffirmed that respiration served to cool the heart and to add air to blood and to remove ‘‘fulginous’’ vapors. This important work in Arabic was probably not widely known (17,18). Michael Servetus broke Galen’s iron grip on orthodox physiology by his accurate description of the whole circulation. In 1553 he was burned at the stake for his heretical disbelief in the Trinity, and because most copies of his revolutionary book were burned with him, it is unlikely that Harvey saw one. Then Vesalius also challenged Galen, writing: ‘‘We are driven to wonder at the handiwork of the Almighty, by means of which blood from the right into the left ventricle sweats through passages which escape human vision.’’ Vesalius too implied that air did not flow in blood vessels, but instead was fixed to blood like the old concept of pneuma, but he never fully grasped the importance of respiration (13). In 1628 William Harvey published De Motu Cordis. His genius lay in combining the works of his many predecessors with some simple arithmetic and some careful observations of living humans. From these he derived an accurate description of the motions of the heart and circulation, lacking only knowledge of the pulmonary capillaries. Forty years later Robert Hooke made a beautiful compound microscope like those already in use in Italy; with this he could have seen the capillaries in the lungs and confirmed Harvey’s hypothesis that such connections had to exist. But he did not. That step was left for Malpighi around 1689.

V.

What Was the Atmosphere Made Of?

The question of what the atmosphere consisted of remained a troublesome mystery. Strato’s concept of a vacuum was forgotten until the seventeenth century, when it was revived by Giovanni Baliani, who respectfully suggested it to Galileo. To support Baliani, a young mathematician named Gaspar Berti around 1638–1640 set up an ungainly water-filled lead pipe 34 feet long, against the wall of his house. With this he demonstrated the vacuum beneath the sealed top, incidentally making the first crude barometer and showing that air had weight! Evangelista Torricelli quickly converted this long pipe to a small mercury-filled instrument with which the weight of the atmosphere could be observed (20). Very soon thereafter in France, Florin Perier took a torricellian barometer up a small mountain (leaving an identical instrument at the base for control observations) and showed that the atmosphere weighed less the higher he went. In Austria Otto von Guericke built the first vacuum pump and dramatically demonstrated how atmospheric pressure held together the two halves of an evacuated sphere. These experiments fitted a major piece into the puzzle (20).

8

Houston

Robert Boyle asked Robert Hooke to make a copy of von Guericke’s ‘‘aire pump’’ in 1658, and together the two showed that mice died in an evacuated bell jar, not because their exhaled air poisoned them, but because animals needed something in air in order to live (21). But what was it? Once more their predecessors seemed to have known! In 756 ad a Chinese scientist Nao Hoa had written that the yin in air was not pure, but could be purified by heating potassium nitrate (which suggests that his yin was oxygen) (13). Leonardo Da Vinci recognized that air consisted of at least two parts and that air in which fire would not burn would not support life either (1). Centuries earlier Pliny the Elder had written in his Natural History that Roman welldiggers would lower a lighted lamp into the well, deciding that if the lamp went out the air was dangerous for them. Alchemists in the sixteenth and seventeenth centuries released oxygen by heating mercuric oxide or potassium nitrate, or, as Boyle did in 1673, by heating lead oxide, almost anticipating the ‘‘discovery’’ of oxygen a century later (1,10). But the alchemists were trying to make gold from dross and failed to appreciate the importance of the gas they had liberated! Although they saw that this gas supported fire and was necessary for life, few asked why. There were false leads. A Danish chemist, Georg Stahl, around 1690 proposed that all combustible materials contained an impalpable gas which he called phlogiston, or ‘‘fire substance.’’ To explain why some materials gained rather than lost weight by burning, Stahl’s followers proposed that phlogiston had a property the opposite of weight—levity—which made the substance heavier when it escaped. This seems rather absurd today, but it captured the minds of many scientists for a century. Sixty years earlier, Jean Rey had interpreted similar observations differently, and he would prove to be right. Puzzled by finding that some matter (he used tin) heated in air gained weight, he wrote: ‘‘This increase in weight comes from air which in the vessel has been rendered denser, heavier, and in some measure adhesive . . . (to the metal) (19). About the same time Ole Borch (aka Borrichius) (1626– 1690) joined those who had isolated oxygen, but he too has been forgotten. Finally, between 1770 and 1773 a Swedish pharmacist Carl Scheele heated silver carbonate and obtained a special gas: ‘‘Since this air is necessarily required for the origination of fire, and makes up about one third of our common air, I shall call it for the sake of shortness ‘‘Fire Air’’ (18). Scheele noted that this new gas supported life as well as combustion. On September 30, 1774, he wrote Antoine Laurent Lavoisier in Paris thanking him for a book and added a description of his experiment, asking Lavoisier to repeat it so that he ‘‘Will see how much air is produced by this reduction and whether a lighted candle can carry on its flame, and animals live in it’’ (18). A British clergyman, Joseph Priestley, who had been studying a gas formed from fermentation, visited Lavoisier the following month and the two may have discussed Scheele’s letter. Lavoisier had been looking at oxidation of inorganic substances, perhaps interested in phlogiston, and in 1772 he had sent a letter to the

History of High Altitude Illness

9

French Academy describing this work but asking that the letter be sealed until he was ready to publish. Both men seemed to sense that something important was close to hand. After their meeting both hurried to isolate the new gas. On Saturday, November 19, 1774, Priestley set up an experiment, but on Sunday he was occupied at church and not until Monday did he actually produce oxygen and take physiology a giant leap forward (22): In this air as I had expected, a candle burned with a valid flame. In this air a mouse lived for a full half hour. My reader will not wonder, that, having ascertained the superior goodness of dephlogisticated air by living mice . . . I should have the curiosity to taste it myself. . . . Who can tell but that, in time, this pure air may become a fashionable article in luxury.

On April 26, 1775 Lavoisier read his paper to the French Academy, 5 weeks after Priestley had read his own to the Royal Society. The existence of oxygen had been proven. Debating who was first to do so is irrelevant. There is ample honor for both, but their unsung predecessors should not be forgotten. One more piece of the puzzle of air, blood, fire, and life was now on the table, but a few more pieces were needed before the relationship of oxygen to altitude sickness would be understood.

VI. Why Breathe—and How? Long ago the Chinese recognized that an uninterrupted supply of ‘‘good’’ air was necessary to support most life. During the age of the Pharoahs, Egyptian physicians described the respiratory tract through which air entered the chest. The GrecoRoman philosopher-scientists, following these observations, believed that the purpose of breathing was to cool the ‘‘innate heat’’ of the heart; they developed various theories about how air entered the blood to form ‘‘pneuma,’’ which could be found in both arteries and veins. Galen’s writings about the mechanics of breathing are sometimes opaque, but basically he believed that air was drawn into the lungs by expansion of the chest, supporting this concept by showing that the lungs collapsed and motion of the chest ceased when the spinal cord was severed (14,17). For the next 1500 years his theory was disputed by those who believed that the lungs expanded actively, sucking in air, thus expanding the thorax. After Gaspar Berti first demonstrated a vacuum in 1640, Torricelli had made the first barometer, proving that the atmosphere had weight. In turn, this revived Galen’s theory of respiratory motion: during inspiration it was the thorax that actively expanded and air was pushed into the lungs by the pressure of the atmosphere. When the thoracic muscles and diaphragm relaxed, the elastic thorax returned to its relaxed position, and air was pressed out. Franciscus Sylvius de la Boe wrote in 1660: ‘‘The lungs do not move naturally of their own motion but they follow the

10

Houston

motion of the thorax and the diaphragm. . . . The lungs are not expanded because they are filled with air, they are filled with air because they are expanded’’ (17). Ten years later Robert Boyle, Richard Lower, and John Mayow agreed, showing that the chest cavity was enlarged by stimulation of the diaphragm and intercostal muscles. Mayow repeated the work of Vesalius a century earlier and found that the lungs did not expand when the chest was opened; he wrote: ‘‘From this we conclude that the lungs are distended by air rushing in and that they do not expand of themselves.’’ But even such work did not persuade the skeptics, and these fundamentals were not fully accepted for another hundred years! But what stimulated the chest to expand and contract? Galen had shown that control of chest and diaphragm motion came through the spinal cord. After his hypothesis of breathing had been resuscitated, it seemed that the stimulus came from the brain itself. Between 1760 and 1837 a number of workers localized the principal control of breathing in a special respiratory center or centers in the medulla of the brain. This control center was responsive to carbon dioxide and later shown to be equally responsive to acidifaction of blood. But the theory did not explain the hyperventilation due to hypoxia that decreased arterial carbon dioxide and made the blood alkaline. Two decades of brilliant experimentation with isolated dog heads perfused with normal blood, and with cross transfusions, made it clear that oxygen also controlled respiration. Gesell, Heymans, and Winterstein showed that this was accomplished by a small collection of cells (the carotid body), richly provided with capillaries direct from the caroid artery in each side of the neck. By 1930 it was clear that respiration was under dual control: by carbon dioxide in the respiratory center and by oxygen in the carotid bodies. This became, incidentally, a nice explanation for Cheyne-Stokes breathing with its fluctuating levels of each gas! Bohr, Henderson, and Hasselbach put the icing on the cake by showing that changes in carbon dioxide significantly altered the affinity of hemoglobin for oxygen.

VII. How Does Oxygen Pass from Lungs into Blood? Once the nature of the atmosphere had been established and the laws that govern gases formulated, oxygen was proven essential for life. The next challenge was to determine how oxygen passed from the atmosphere, through the alveolar structure of the lungs, and into blood. Many had been trying to do this for a long time. Galen and his predecessors thought air somehow entered the heart from the lungs and mixed with blood to form ‘‘pneuma,’’ which penetrated throughout the body. Giovanni Borelli, inspired by Galileo in 1681, applied his imagination and talent as Professor of Mathematics at Pisa to physiology, and wrote perceptively that ‘‘air taken in by breathing is the chief cause of life in animals,’’ although of course the ‘‘vital spirit’’ or oxygen was still to be isolated. Borelli made the new and important observation that air dissolved in water could pass through certain membranes, and thus pioneered the principles of diffusion.

History of High Altitude Illness

11

Soon carbon dioxide would be identified. Jean Baptiste Van Helmont (1577– 1644) collected the gas that pooled in the bottom of a famous cave in Italy where dogs perished but their taller masters survived. He showed that this was the same as the gas formed by adding acid to limestone that would extinguish fire and would not support life (17). But it was still too early to relate this to combustion. A century later Joseph Black found that what he christened ‘‘fixed aire’’ was formed by burning charcoal. Furthermore, in an ingenious experiment, he arranged for the air exhaled by 1500 people during 10 hours inside a church at a religious gathering to pass through a ceiling vent and over rags saturated with limewater. By weighing the calcium carbonates thus formed, he showed that the expired gases contained ‘‘fixed aire’’ or carbon dioxide. Thus, even before oxygen was isolated, the gas exhaled during respiration, presumably generated by the bodily functions, had been identified (17,23). Equally important was the growing recognition that something in air was essential to life. Later, after Borelli had demonstrated that air could pass through a membrane and after oxygen had been proven essential to life, a new question arose: If carbon dioxide was carried in the blood and oxygen inspired into the lung, could the two gases pass the lung membrane in opposite directions? John Dalton, after much study, provided the answer in 1808: ‘‘When a vessel contains a mixture of such elastic fluids (gases), each acts independently on the vessel . . . just as if the other were absent’’ (24). This was the crucial law of partial pressures, which said that the passage of a gas or gases through a membrane was dependent (among other things) on the difference in the pressure of each gas on the opposite sides of the membrane. It seemed probable that Borelli’s observation should apply to the thin membranes of the alveolar walls: oxygen could pass into blood at the same time carbon dioxide was passing out of it. In 1892 Christian Bohr confirmed earlier work by Baptiste Biot in 1807 (and later extended by Moreau) proving that the swim bladders of fish contained a very high concentration of oxygen; this appeared to violate Dalton’s law and could be explained only by active oxygen secretion (25). Haldane and Lorrain Smith, studying how oxygen moved from the lung into blood, visited Bohr in 1894 and were intrigued by his suggestion that similar secretion might occur through the mammalian alveolar walls. To test this hypothesis, Haldane and Smith measured the oxygen pressure in alveolar air, calculated the arterial oxygen pressure after breathing a low concentration of carbon monoxide for 20 minutes, and found the calculated alveolar oxygen pressure always lower than the arterial (26). Bohr’s students Auguste and Marie Krogh devised a more accurate method for estimating arterial blood oxygen by a single breath of carbon monoxide. With this they were unable to confirm their professor’s findings. Reluctantly they published a series of brilliant papers in 1910 and started a famous controversy by their unequivocal statement: ‘‘The passage of (oxygen) and the elimination of carbon dioxide in the lungs takes place by diffusion and by diffusion alone’’ (27). Haldane and Smith’s experiments led them to dispute the diffusion theory. In

12

Houston

1910 at a meeting in Vienna, Haldane met young Yandell Henderson, who suggested that the summit of Pikes Peak in Colorado would be an ideal location to detect oxygen secretion if it took place. The following year Haldane, Henderson, Douglas, and Schneider went to Pikes Peak to study acclimatization, including measurement of oxygen pressure in alveolar air and using an improved carbon monoxide method to estimate oxygen in arterial blood at rest and after exertion. They found the calculated arterial oxygen pressure always higher than the alveolar, which convinced Haldane that active oxygen secretion was a specific function of the lung, which enabled humans to tolerate hypoxia. He proposed that differences between individuals in their susceptibility to mountain sickness were due to different degrees of secretion. His later studies did not shake this belief, even when confronted with contrary evidence collected by his former colleague, Joseph Barcroft (26). Barcroft had begun his scientific career by studying the metabolism of salivary glands, and in 1901, together with Haldane, he developed a method for measuring gases in small amounts of liquid. He found that the partial pressure of oxygen in saliva was higher than in the capillary blood in the salivary gland. At first he could explain this observation only by active oxygen secretion! Several years later, however, his study of factors affecting the shape of hemoglobin dissociation curves and thus the oxygen content of blood led him to reexamine his earlier data and to question oxygen secretion in the salivary gland circulation. Barcroft was called up in World War I to treat pulmonary edema due to chlorine gas, but soon after the Armistice he returned to studies of other causes of hypoxia—like high altitude. Renouncing the attractive secretion theory, he challenged Haldane’s data. To do so in 1920 he spent 10 days in a sealed glass room where the oxygen was gradually decreased to the partial pressure equivalent to 18,000 feet. Using a method developed by Stadie a few years earlier, Barcroft had a large cannula tied into his radial artery (which, he wrote matter of factly, ‘‘of course had to be sacrificed’’). Through this cannula he drew arterial blood and was able to measure oxygen directly. His data showed that his arterial oxygen pressure was lower than the alveolar (28). Haldane countered that Barcroft had studied only himself, and when he was sick at that. Haldane repeated and corrected his own studies and argued that although secretion might not occur during rest or at normal atmospheric pressure, it was one of the means by which humans adjusted or acclimatized to hypoxia. He supported his belief with clinical observations of those who had been sick or well on Pikes Peak, but again, he did not measure arterial oxygen directly (29). The argument raged, of course in gentlemanly terms. Barcroft and a strong team went to Cerro de Pasco in Peru (14,200 feet) and repeated the alveolar-arterial studies along with much other work. Once again the alveolar air always contained a higher oxygen pressure than the arterial blood, even in the well-adapted natives. Haldane was wrong. This ended the oxygen secretion theory of acclimatization.

History of High Altitude Illness

13

VIII. How Is Oxygen Carried in Blood? How, once diffused from the lungs into the blood, is oxygen transported to the tissues that are dependent on an uninterrupted supply? Twenty-one hundred years ago Erasistratus suggested that air somehow combined with blood, and in 1668, long before oxygen was isolated (30), Lower wrote that something in air ‘‘penetrated’’ dark venous blood changing it to the bright scarlet of that in the arteries (19): . . . in the open chest, so long as the lungs were inflated with a bellows, the blood leaving was bright red; when the bellows stopped the blood became dark. . . . This red color is entirely due to the penetration of particles of air into the blood. . . . May we not concede an inward passage of this foodstuff into the blood through similar tiny pores (in the lungs)?

Robert Hooke added experiments that showed that as long as the lungs received fresh air, whether they moved or not, the blood going through them became red. Once again we see the Greco-Roman concept of ‘‘pneuma’’ as a combination of air and blood. In 1910, two millenia after Erasistratus, Christian Bohr described the S-shaped hemoglobin dissociation curve we know today, a more precise map of the curve defined by Bert in 1872. Bohr and others also measured the effect of carbon dioxide and changes in acidity on this dissociation curve and therefore on oxygen transport. They used these data to explain how entry of carbon dioxide into blood in the tissue capillaries ‘‘helped’’ the offloading of oxygen, and how the loss of carbon dioxide into alveolar air enhanced the loading of oxygen in the lungs (17). Such transfer of gases depended on the motions of the chest—the process of breathing.

IX. What Was the Relationship of Oxygen to Mountain Sickness? Twenty years after Priestley’s paper, Thomas Beddoes showed his awareness of the significance of oxygen in mountain sickness when he wrote (31): Now in ascending these rugged heights the muscular exertion must expend a great deal of oxygene which the rarefied atmosphere will supply but scantily. . . . The experiments of Mr. Saussure, Pini, and Reboul, concur in shewing that, independent of its rarefaction, the atmosphere of very elevated mountains contains a far smaller proportion of oxygene than that of lower regions, especially than that of the high vallies of the Alps.

In 1802 Alexander von Humboldt climbed to 18,800 feet on Chimborazo in the Andes and, like Beddoes, suggested that lack of oxygen was responsible for his symptoms (32):

14

Houston [W]e all became by degrees much distressed; a constant desire to vomit, together with vertigo, were the most prominent symptoms, and proved far more trying than the difficulty of breathing which we likewise suffered from. The distress, the weakness, and the desire to vomit certainly came as much from the lack of oxygen in these regions as from the rarity of the air.

There were other theories: Stanhope Speer picked up Haller’s ‘‘hydraulic theory’’ but ignored oxygen in his book written in 1853, although some of his concepts were correct (33): These symptoms (of mountain sickness) may be referred to a threefold source, viz, a gradually increasing congestion of the deeper portions of the circulatory apparatus, increased venosity of the blood, and loss of equilibrium between the pressure of the external air and that of the gases existing within the intestines. . . . The causes of mountain sickness are themselves the result of a change from a given atmospheric pressure and temperature, to one in which both are greatly and suddenly diminished.

Conrad Meyer-Ahrens, a leading Zurich physician, in 1854 came closer to the truth, though he had an odd explanation for the weakness experienced at altitude: ‘‘Mountain sickness is due to (a) decrease in the absolute quantity of oxygen, (b) rapidity of evaporation (c) intensity of light, (d) expansion of intestinal gases and (e) weakening of the coxo-femoral articulation’’ (34). Studies of altitude illness languished for more than 50 years after de Saussure, during which time 57 ascents of Mont Blanc were made, 2 by women. Then, starting in 1852, a London physician, Albert Smith, gave 2000 illustrated lectures describing his adventures on Mont Blanc so vividly that hundreds took up mountaineering and began the Golden Age of alpine climbing. More than half of the budding alpinists had symptoms, piquing medical curiosity and prompting serious studies of mountain physiology. X.

Finally—The Cause(s) of Mountain Sickness

Lower had shown that dark venous blood changed color when exposed to air, but the relationship between this observation and the symptoms of mountain sickness was not recognized for a long time. Then came a man who, as Harvey had done, was able to put together the work of his predecessors to make a coherent whole. After studying law, Paul Bert took his degree in medicine and studied respiration under Claude Bernard. He met a Paris physician, Denis Jourdanet, who had traveled among the mountains of Mexico, and the two became friends and fellowworkers. Jourdanet had built decompression chambers for therapeutic purposes and had formulated the hypothesis that blood contained less oxygen on high mountains because the atmospheric pressure was lower; he called this ‘‘barometric anoxemia’’ (35). Jourdanet provided the means for the expensive equipment and supported Bert in a series of studies to examine the puzzle of mountain sickness.

History of High Altitude Illness

15

Bert used his chamber to decompress animals to different pressures with different oxygen percentages in the chamber and found that there was no consistency in either the pressure or the percentage of oxygen at which they died. When he kept the percentage of oxygen constant but varied the chamber pressure, animals died more rapidly the lower the pressure. Then he took the imaginative step of multiplying the barometric pressure in the chamber by the percentage of oxygen, and this number gave consistent results (35). He had applied Dalton’s law of partial pressure to show that it was the partial pressure of oxygen rather than barometric pressure or percentage alone that was decisive—a major advance in physiology and another piece of the puzzle put in place. As a clincher he had himself rapidly ‘‘taken up’’ in his chamber to 18,000 feet and experienced symptoms we would today call acute hypoxia; then he breathed oxygen and obtained immediate relief. Finally he breathed oxygen while being ‘‘taken up’’ and developed no symptoms. From repeated tests he concluded that mountain sickness was due to lack of oxygen, and that by breathing oxygen even at decreased pressures, he could prevent or alleviate the symptoms. These careful studies would soon be of great importance, when newly developing warplanes struggled toward heights that would endanger pilots without supplementary oxygen, and would be even more relevant much later, when both oxygen and increased pressure would be necessary in very advanced aircraft. Bert also collected anecdotal experiences at altitude from hundreds of travelers and explorers who had been on high mountains all over the world, as Boyle had done in a smaller way. He used instruments which he and Jourdanet designed to measure the amount of oxygen combined with hemoglobin under different conditions. He extracted gases from the blood of animals exposed to decreased pressures and defined a portion of the oxy-hemoglobin dissociation curve we know today. He encouraged his young colleagues to go to high altitude in balloons, until two of three died when they went too high after their meager supply of oxygen was exhausted; this ended Bert’s interest in altitude studies. His La Pression Barometrique, an extraordinary encyclopedia of all that was then known or believed about barometric pressure, high altitude, and mountain sickness, was published in 1878 (36). His altitude studies have overshadowed his pioneering work with hemoglobin and oxygen transport and with increased barometric pressures as in diving. He and Jourdanet had shown beyond a reasonable doubt that lack of oxygen was the main, if not the only cause of the symptoms experienced on high mountains. But one of their contemporaries disagreed (37). Angelo Mosso was already an experienced mountaineer and physiologist when he read the works of German poet-physiologist Albrecht von Haller, who argued in 1761 that the physical effect of decreased atmospheric pressure on the blood vessels was solely responsible for altitude sickness. This idea intrigued Mosso, and he began to study altitude in a decompression chamber and later in a laboratory built for him by Queen Margherita of Italy on one of the summits of Monte Rosa (15,300 feet). He concluded that the decrease in carbon dioxide in the blood that

16

Houston

resulted from overbreathing at altitude, rather than lack of oxygen, caused mountain sickness, and he coined the word ‘‘acapnia’’ for this phenomenon (37). Tom Longstaff, one of the great Himalayan pioneers and a practicing physician seemed to disagree with Mosso in 1906 in his doctoral thesis (38): With the reduced atmospheric pressure there is necessarily a reduction in the amount of CO2 per unit of volume of air breathed. The increased pulmonary ventilation already referred to also tends to lower the alveolar CO2 tension. Nevertheless it must not be forgotten that the main source of CO2 in the alveolar air is not the atmosphere, and Mosso’s theory that the diminished supply of CO2 at altitude is the cause of mountain sickness will not bear close inspection.

In the last three lines of the preceding quotation, Longstaff misquoted Mosso, who believed the loss of CO2 by hyperventilation, not the lower CO2 in the atmosphere, caused mountain sickness. Mosso was mistaken, but his book Life of Man on the High Alps is rich in laboratory experiments and mountain experiences. Other scientists were also turning to high summits. Joseph Vallot was doing botany and geology when he was attracted to Mont Blanc. First he sought to determine whether or not mountain sickness was caused by a decrease in body temperature, at that time a popular theory. In 1887 he spent three nights on the summit of Mont Blanc in a tent and was so impressed that he decided to build a laboratory high on the mountain. A year later he returned to Chamonix with 19 cases containing three duplicate sets of instruments, one to be read in the valley, one halfway up the mountain, and one in a small building a little below the summit. Among the many studies he made was the change in exercise capacity experienced at altitude by squirrels! He taught them to run on a caged wheel and counted the number of turns in Chamonix, during a week in his high laboratory, and again in Chamonix; squirrels lost half of their exercise capacity at altitude and did not fully regain it after descent (39). After many other experiments and observations his building was abandoned and a new one erected in 1898; it has been enlarged and is an active research laboratory today. Soon after Vallot, a distinguished astronomer, Jules Janssen built his own observatory near the summit in August 1891. Believing that exertion was a major cause of the mountain sickness he suffered, he later had himself pulled up the mountain on a sled and felt none of the unpleasant symptoms that affected the 12 men who dragged him (40). Dr. Egli-Sinclair, the leader of Janssen’s party, had to depart, and a young Chamonix doctor, Etienne Henri Jacottet, took his place. Jacottet climbed up rapidly, went on to the summit, and on returning to the laboratory he became ill. As he was dying he wrote a farewell letter to his brother describing his harrowing symptoms (37). An autopsy by Dr. Wizard read in part (42): ‘‘Poumon: couleur violet, gonfle, fonce, congestion bilaterale, oedeme considerable muqueuse bronchique injectee fortement. Le liquide de la coupe est ecumeneux. Congestion egale partout. Foie,

History of High Altitude Illness

17

rate, reins normaux. Pas d’oedeme des jambes’’ (42). This is the first autopsied case of high altitude pulmonary edema. Shortly after his expedition to K2 (second highest mountain in the world) in 1902, mountaineer Guy Knowles wrote: ‘‘Pfannl is really seriously ill. . . . What he has is oedema of one lung which is, of course, very serious’’ (43). The leader of the party, Aleister Crowley, added: ‘‘Pfannl has oedema of the lungs and his brain is gone.’’ In 1893 Hugo Kronecker, consultant for the railroad proposed to reach the 13,645 foot Jungfrau summit, studied the symptoms of mountain sickness not only on mountaineering climbs, but also in a small decompression chamber (a ‘‘bell’’). After studying many cases, he was satisfied that ‘‘mountain sickness does exist’’ and noted that even slight exertion exacerbated symptoms. To decide whether passengers on the train would be made sick by the altitude, he arranged for seven men and women to be carried in chairs up to the pass (11,385 feet) to which the railroad was eventually built. Like Janssen, they did not have symptoms. Kronecker doubted Bert’s evidence that hypoxia was the cause of mountain sickness. Instead, perhaps influenced by Haller’s mistaken hydraulic theory, Kronecker believed that diminished atmospheric pressure was the primary cause. He argued that decreased atmospheric pressure dilated the pulmonary blood vessels, straining the heart and causing edema of the lungs. His description of six cases suggests that these were in fact high altitude pulmonary edema (44). In 1870 Dr. Nathan Zuntz left a country practice for teaching and research in physiology. He made scientific expeditions to Monte Rosa and Pikes Peak and invited Barcroft, Douglas, and Durig to join an expedition to the peak of Teneriffe (12,200 ft) in 1909, specifically to look at Mosso’s acapnia theory. Using equipment he had developed, he studied exercise metabolism at sea level and at altitude in a balloon. Before World War I Zuntz was a pioneer in aviation medicine, going himself to 17,500 feet, breathing oxygen from an apparatus he designed, and planning the first pressurized cabin for use in balloon flights above 35,000 feet. With colleague Adolph Loewy and assistants Mueller and Caspari, he wrote another of the classics in altitude physiology: Hohenklima und Bergwanderung (45). These were the beginnings of a great surge of interest in altitude physiology. Soon leading scientists were making expeditions to mountains all over the world. Half a century earlier, Hennessey, a member of the survey of India, had calculated that Mount Everest was the highest mountain on earth, and in a few decades mountaineers became interested (41). By the end of the century, an attempt to climb the mountain seemed possible as Tibet became more accessible. The summit even seemed attainable as climbers pushed their ceiling higher and higher. In 1892 a distinguished surgeon-mountaineer, Clinton Dent, predicted the summit would be reached (47): Selected men will have to work for a year or more with the one definite object before them. . . . We may agree with Mr Whymper that the effects on respiration will impose

18

Houston limitations on the range of man, but it does not seem conceivable that this limitation is below the level of the highest point on the earth’s crust. . . . It is a tremendous undertaking but a magnificent possibility.

Sixty one years later two men reached the summit, and since then several hundred have done so without supplementary oxygen. But, perhaps influenced by Haller and Meyer-Ahrens, Dent was mistaken about some aspects of altitude illness (47): A far more important factor (than abdominal distention) is the effect of diminished pressure on the portion of the spinal marrow which is concerned with the nutrition of the locomotive agents, the lower limbs; greatly increased pressure also produces much the same symptoms. This effect has no relation to the absence of oxygen.

Everest was one thing, the Alps quite another. During the last half of the nineteenth century hundreds of men and a few women swarmed over the Alps and noted that, even on the same climb under similar conditions, some fell victim to mountain sickness, others did not, and the same individual might be sick one day and not another. This could not easily be explained by acclimatization, and Longstaff, who had climbed many high Himalayan peaks, was not impressed: ‘‘As to the general benefit to mountaineers of ‘acclimatization’ produced by a prolonged residence at an altitude of half an atmosphere or less, I am altogether sceptical, not withstanding that its efficacy has been frequently urged by many writers’’ (38). Instead Longstaff insisted that training and physical fitness were protective, claiming that this was why he and others had made some high altitude climbs without symptoms. Today, despite many careful studies, we have neither proved nor disproved the fitness argument, and the ultimate secrets of acclimatization elude us. Joseph Barcroft, after his Peruvian work with local residents, concluded (28): The acclimatized man is not the man who has attained to bodily and mental powers as great in Cerro de Pasco as he would have in Cambridge (whether that town be situated in Massachusetts or in England). Such a man does not exist. All dwellers at high altitude are persons of impaired physical and mental powers. The acclimatized man is he who is least impaired.’’

The distinguished Peruvian physiologist, Carlos Monge, Sr., violently disagreed, claiming that Barcroft’s observations were flawed because he had himself been affected by the altitude. Both were partly right; we know today that even full acclimatization does not restore sea level ability above 12,000–14,000 feet (48). Following the precept of Claude Bernard that the constancy of the internal environment is the condition of an active life. Barcroft thought acclimatization was the result of changes in many physiological processes. Drawing on this in 1946, Houston and Riley described acclimatization as ‘‘—a series of integrated adaptations which tend to restore the oxygen pressure of the tissues toward normal sea level values despite the lowered PO2 of the atmosphere’’ (49) (note the careful word ‘‘toward’’). The early Himalayan climbers made no physiological observations, and the first serious studies of man’s ability to summit Everest ‘‘without adventitious aids’’

History of High Altitude Illness

19

were those of mountaineer-chemist Alexander Kellas. He himself did no laboratory work, but from others’ data calculated that a well-acclimatized and trained climber could reach the summit of Everest. He was puzzled that the height where Acute Mountain Sickness (AMS) becomes a problem differs between individuals and from place to place. He accepted the belief that facing into the wind increases alveolar oxygen by ‘‘packing air into the lungs’’ (50). Kellas knew that alkalosis resulted from the hyperventilation response to high altitude, but, lacking accurate measurements, he hedged his bet: ‘‘At high altitude . . . the quantity of carbon dioxide in the blood is lowered, but the acidity increases (or rather the alkalosis produced by removal of carbon dioxide is diminished correspondingly) and the respiratory center remains adequately stimulated’’ (51). Kellas died in 1924 en route to Everest with the expedition on which Mallory and Irvine disappeared near the summit and Norton reached 28,000 feet, where the atmospheric oxygen pressure is only a few torr more than on the summit. Not for more than 50 years would Dent’s and Kellas’s prediction be confirmed when that last thousand feet would be climbed breathing only air. During the 1924 expedition Major R. W. G. Hingston, doctor to the party, recorded pulse rates, blood pressure, breath-holding time, and red blood cell counts for eight of the climbers up to 21,500 feet, the first such data collected so high. He noted that resting pulse rate and blood pressure in these acclimatized men did not change greatly from sea level, but breath-holding time decreased and hemoglobin content increased (52). Access to Everest was strictly controlled by the Tibetans, and only a few British parties were allowed, so those who wished to study the effects of altitude turned to other peaks, lower but no less formidable. In the 1930s German mountaineers made several attempts to climb Nanga Parbat, the seventh highest summit in the world. Although their goal was to reach the top, in 1934 Ulrich Luft joined the party as medical officer and collected data up to 19,000 feet, as he did later at the Jungfraujoch laboratory and in Berlin. His definitive thesis, ‘‘Acclimatization to Altitude,’’ was published in 1941, but was only recently translated into English and made generally available (53). Climbing mountains is not the only way to get high in the air: men have always longed to fly like birds. Daedalus and Icarus and Leonardo failed, but others persisted. Laurenco de Guzmao demonstrated a small hot air balloon to the king of Portugal in 1686, and in November 1782 Joseph and Etienne Montgolfier in Paris secretly flew a tethered hot air balloon. After a few more tethered flights, a larger balloon carried a sheep and a rooster over 1000 feet into the sky. In late October 1783 Pilatre de Rozier and the Marquis d’Arlandes entered the wicker basket of a huge balloon and rose above 3000 feet, landing safely a few miles away. The age of flight had begun. More hot air balloon flights, and soon balloons filled with the newly discovered hydrogen, followed (54). A century later in March 1874 two of Paul Bert’s young associates CroceSpinelli and Sivel flew to 24,000 feet in a hydrogen balloon, breathing oxygen from leather bags on Bert’s insistence. Emboldened by this, on the next flight they were

20

Houston

joined by Tissandier, and the trio were lifted even higher. But they ran out of oxygen and when the balloon later descended Croce-Spinelli and Sivel were dead, having succumbed to hypoxia (36). Then came powered flight, and by 1916 aircraft could fly high enough to endanger an aircrew. This prompted urgent studies of how to select pilots who best tolerated altitude (which proved futile), and rudimentary oxygen systems evolved. The research continued after World War I on mountains where laboratory studies could be done. As World War II loomed it was obvious that very high altitude might be decisive in an air war. Programs to increase tolerance for hypoxia by residence in the mountains proved effective but impractical. Instead, better oxygen equipment and soon pressurized breathing equipment enabled men to go as high as their aircraft. Since then oxygen equipment has been rendered moot by pressurization of aircraft. Under the pressure of World War II many studies of altitude physiology opened even more doors to new fields of interest not only to climbers and aviators but to clinicians as well. New equipment, new concepts, and new techniques now build on the work of previous centuries. Although the original puzzle has been assembled, clearly many new pieces must be added to the periphery—and quite probably to the center as well! Not surprisingly, the large decompression chambers, mostly idle as the war ended, were a tempting opportunity to look at how well man might tolerate the thin air on the summit of Mount Everest, if he could get there! An ambitious study, suitably called Operation Everest, was sponsored by the U.S. Navy in 1946. During a 35-day stay in a large decompression chamber, four men were slowly taken to a simulated altitude of 25,000 feet from which, in one day, they made a ‘‘dash for the summit.’’ Two of the four required oxygen, but the other two were able to rest quietly for 20 minutes and to ride a bicycle briefly before descending. They were not fully acclimatized by the relatively short ascent, but the data provided new insights into acclimatization. Most welcome to mountaineers, Operation Everest proved that men could survive breathing only the surrounding air on the highest point on earth (49). Today, as described in this book by others, sophisticated instrumentation makes possible a fantastic array of studies, even on the actual summit of Everest.

XI. Summary In this chapter I have touched on only a few of the advances made in the twentieth century; there have been far too many for such a summary survey. Many of the most important have been described elsewhere (51,57,58). In the last part of our century it has become clear that we have as much to learn today as our predecessors had in centuries past. Only the parameters are different. They were looking at gross anatomy and physiology; we use refined instruments to study minutiae—cells, enzymes, neurotransmitters, and molecular relationships

History of High Altitude Illness

21

as yet unknown. Although we have begun to understand a great deal about the acquisition and utilization of essential oxygen, the ultimate mechanisms elude us still—and may for many years.

Acknowledgment I am very grateful to Dr. Ralph Kellogg for many suggestions—and for saving me from embarrassing mistakes.

References 1. Gilbert DL. Oxygen and Living Processes. An Interdisciplinary Approach. New York: Springer-Verlag, 1981. 2. MacNeish RS. Early man in the Andes. Sci Am 19XX; 38–46. 3. Gilbert DL. The first documented report of mountain sickness: the China or Headache Mountain story. Resp Physiol 1983; 52:315–326. 4. Hsien F. The Travels of Fa-Hsien (399–414 AD) or a Record of the Buddhist Kingdoms. Westport, CT: Greenwood, 1981. 5. Zheng Z, Zhenkai L. Footprints on the Peaks: Mountaineering in China. Seattle: Cloudcap, 1995. 6. Gribble F. The Early Mountaineers. London: T Fisher Unwin, 1899. 7. Pinkerton J. A General Collection of the Best and Most Interesting Voyages and Travels in All Parts of the World. London: Longman, 1808. 8. Gilbert DL. The first documented description of mountain sickness; the Andean or Pariacaca story. Respir Physiol 1983; 52:327–347. 9. Personal communication from Kellogg citing Grimstone. 10. Haider MM, Dughlat. The Tarikh-i-Rashid: History of the Mongols in Central Asia. Lahore: Book Traders, 1894:412–413. 11. de Saussure HB. In: Pinkerton J. A General Collection of the Best and Most interesting Voyages and Travels in all Parts of the World. London: Longman, 1808. 12. Tschudi JJ. Travels in Peru. London: Bogue, 1847. 13. Gordon EL. Medicine Throughout Antiquity. Philadelphia, Davis: 1949. 14. Wilson, LG. Erasistratus, Galen and the pneuma. Bull Hist Med 1959; 33:293–314. 15. Unschuld PU. Medicine in China, a History of Ideas. Berkeley: University of California, 1985. 16. Fleming D. Galen on the motion of the blood in the heart and lungs. Isis 1955; 46:14– 21. 17. Perkins JF. Historical development of respiratory physiology. In: Fenn WO, Rahn R, eds. Handbook of Physiology. Washington, DC: American Physiological Society, 1964. 18. Houston CS. Going Higher: Oxygen, Man and Mountains. Seattle: Mountaineers, 1998. 19. Fulton JF. Selected Readings in the History of Physiology. Springfield, IL: Charles C Thomas, 1966. 20. Middleton WEK. The History of the Barometer. Baltimore: Johns Hopkins, 1964. 21. Hooke R. History of the Royal Society of London. 28:539–540, 1667.

22

Houston

22. Priestley J. The Discovery of Oxygen. Edinburgh: Livingston, 1961. 23. Guerlac H. Joseph Black and fixed air. A bicentenary retrospective with some new or little known material. Isis 1957; 48:124–151. 24. Dalton J. Experimental enquiry into the proportion of the several gases or elastic fluids, constituting the atmosphere. Manchester: Mem Lit Phil Soc, 1805. 25. Bohr C. The influence of section of the vagus nerve on the disengagement of gases in the swim-bladder of fishes. J Physiol 1894; 15:494–500. 26. Haldane JS, Priestley JG. Oxygen secretion in the lungs. In: West JB, ed. High Altitude Physiology. Stroudsburg: Hutchinson-Ross, 1981. 27. Krogh A. On the mechanism of gas exchange in the lungs. Acta Scand 1910; 23:248– 278. 28. Barcroft J. The Respiratory Function of the Blood; Pt 1: Lessons from High Altitude. London: Cambridge University Press, 1925. 29. West JB, Milledge J. The Haldane-Barcroft debate: Resolved that the lungs secrete oxygen. In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and the Brain. Burlington: Queen City Press, 1995. 30. Wilson LG. The problem of the discovery of the pulmonary circulation. J Hist Med 1962; 16:229–244. 31. Beddoes T. Observations on the Nature and Cure of Calculus, Sea Scurvy, Consumption, Cafarrh, and Fever; Together with Conjectures upon Several Other Subjects of Physiology and Pathology. Edinburgh: Alembic Club, 1902. 32. von Humboldt A. Travels to the equinoctial regions of the new continent during the years 1799–1804. In: Bert P. Barometric Pressure. Bethesda, MD: Undersea Medical Society, 1978. 33. Speer ST. On the physiological basis of mountain sickness. London: Richards, 1853. 34. Meyer-Ahrens C. Die Bergkrankheit. Cited in: Bert P. Barometric Pressure. Hitchcock MA, Hitchcock FA, trans. Columbus: College Book Co, 1943. 35. Kellogg RH. La Pression Barometrique: Paul Bert’s hypoxia theory and its critics. Respir Physiol 1978; 34:1–28. 36. Bert P. Barometric Pressure. Hitchcock MA, Hitchcock FA, trans Columbus: College Book Co, 1943. 37. Mosso A. Life of Man in the High Alps. London: T Fisher Unwin, 1898. 38. Longstaff TG. Mountain Sickness and Its Probable Cause. London: Spottiswoode, 1906. 39. Vallot J, Bayeux, R. Exeriments faites au Mont Blanc, en 1991–93, sur l’activite spontannee aux tres hautes altitudess. Notes presented at Academie des Science, Francais, 1914. 40. Whymper E. A Guide to Chamonix and the Range of Mont Blanc. London: John Murray, 1896. 41. Egli-Sinclair. Sur le Mal de Montagne. Ann Obs Vallot 1893; 1:110–130. 42. Foray J. Personal communication. 43. Knowles G. Unpublished typescript 1902, courtesy Nick Clinch. 44. Kronecker H. Mountain-sickness. Med Mag 1895; 4:651–666. 45. Gunga H-C, Kirsch KA. Nathan Zuntz (1847–1920)—a German pioneer in altitude physiology and aviation medicine, part 1: Biography. Aviat Space Environ Med 1995; 66:168–171. 46. Ward MP. Mount Everest. Cartographic J 1994; 31:33–34. 47. Dent CT. Can Mount Everest Be Ascended? The Nineteenth Century 1892; 32:604– 613.

History of High Altitude Illness

23

48. Monge C. Acclimatization in the Andes. Baltimore: Johns Hopkins University, 1948. 49. Houston CS, Riley RL. Respiratory and circulatory changes during acclimatization to high altitude. Am J Physiol 1947; 149:565–588. 50. West JB. Alexander Kellas and the physiological challenge of Mt. Everest. J Appl Physiol 1987; 63:9–11. 51. West JB. High Life: A History of High Altitude Physiology and Medicine. New York: Oxford, 1998. 52. Norton EF. The Fight for Everest 1924. London: Arnold, 1925:243–261. 53. Luft UC. Acclimatization to Altitude. Berlin 1941. Luft FC, trans. Albuquerque: Lovelace Foundation, 1993. 54. Gillespie CC. The Montgolfier Brothers and the Invention of Aviation. Princeton, NJ: Princeton University Press, 1983. 55. Houston CS, Sutton JR, Cymerman A, Reeves JT. Operation Everest II man at extreme altitude. J Appl Physiol 1987; 63:877–882. 56. Kellogg RH. Altitude acclimatization: a historical introduction emphasizing the regulation of breathing. Physiologist 1968; 11:37–57. 57. Hultgren HN. High Altitude Medicine. Stanford, CA: Hultgren Publications, 1997. 58. Houston CS, Going Higher: Oxygen, Man and Mountains. Seattle: Mountaineers, 1998.

2 The Atmosphere

JOHN B. WEST University of California, San Diego La Jolla, California

I.

Evolution of the Atmosphere

In the beginning God created the heavens and the earth, but only now are the first direct data on the composition of an outer planet atmosphere being transmitted to Earth. It is remarkable that although Earth was formed about 4.5 billion years ago, exploration of the solar system with measurements of the planets’ atmospheres has only occurred in the last 45 years, i.e., in 0.000001% of Earth’s lifetime. Earth and its primitive atmosphere were probably formed by the accretion of material from early solar nebulae by gravitational forces. The material is thought to have had the same composition as the sun, i.e., mostly hydrogen and helium, but with small quantities of heavier elements. Oxides and hydrides of the heavier elements condensed into particles and came together to form the planets. The initial atmosphere mainly consisted of hydrogen and helium, but this atmosphere was lost because of Earth’s weak gravitational field. If we consider the masses and distances of the planets and some of their larger satellites from the sun, it transpires that Mercury is too small to retain an atmosphere, and the same is true, or nearly so, of our moon and Mars. However, the outer planets such as Jupiter and Saturn have abundant atmospheres, and the first data from Galileo’s atmospheric 25

26

West

probe of Jupiter are presently being analyzed in the Jet Propulsion Laboratory in Pasadena, CA. Generally the surface temperature of a planet decreases with increasing distance from the sun, although this is modified by the greenhouse effect of the atmosphere, which is why the surface temperature of Venus exceeds that of Mercury. In addition, Jupiter and Saturn are so massive that they generate and radiate heat. However, only Earth has a surface temperature that allows water to be in the liquid phase, and therefore it is difficult to see how advanced life as we know it could exist elsewhere in the solar system. The primary atmosphere of hydrogen and helium was replaced by a secondary atmosphere formed by degassing of various constituents of the primitive Earth as a result of the increasing temperature (1). The temperature rose mainly because of the kinetic energy of impacting bodies and radioactive decay. Volcanoes and fumaroles also contributed to this secondary atmosphere, and active volcanoes can still be observed elsewhere in the solar system, for example, on Io, a satellite of Jupiter. Liberation of water of crystallization produced large amounts of water vapor, carbonates decomposed to form carbon dioxide, nitrides and ammonium compounds gave rise to nitrogen and ammonia, radioactive decay released hydrogen, helium, and argon, and sulfur compounds gave rise to hydrogen sulfide and sulfur dioxide. This secondary atmosphere was anaerobic, but as Miller (2) showed when he heated a similar gas mixture with water and exposed it to electrical discharges, this primordial soup could produce a remarkable range of organic compounds including many amino acids, nucleotides, lipids, and sugars. Such events were not confined to Earth’s atmosphere. Radioastronomy has detected over 80 different complex molecular species, including a variety of organic molecules, in space between and around distant stars. Some of Earth’s lipids apparently became organized into membranes to encapsulate material as ‘‘protocells.’’ Presumably this was the first stage in the evolution of life on Earth. Although the necessary steps to produce replicating molecules are still poorly understood, some biologists believe that random sequencing of ribonucleic acid (RNA) eventually produced by chance a template for the formation of an enzyme that conferred some biochemical advantage on a ‘‘protocell.’’ This enabled it to compete effectively with its neighbors as well as to replicate its RNA. Later the transmission of the genetic code was taken over by deoxyribonucleic acid (DNA) in organisms with the greatest potential for survival. These primitive forms of life were a far cry from the advanced aerobic life forms that we know today. A dramatic development apparently occurred about 3– 3.5 billion years ago when there was the first utilization of visible light as an energy source with the production of oxygen by photosynthesis. Initially the oxygen was a waste product, and it is not certain how rapidly it accumulated in the atmosphere (Fig. 1). There is some evidence that the major change from a reducing to an oxidizing atmosphere occurred about 2 million years ago (3). However, many biologists believe that the oxygen concentration also rose steeply just before the beginning of the abundant fossil records of the Cambrian period, i.e., about 570 million years ago. It is possible that the oxygen concentration peaked in the late Carboniferous

The Atmosphere

27

Figure 1 Possible scheme for the sequence of events leading to the oxygen concentration in the present atmosphere. Solid lines indicate oxygen-generating processes and broken lines show oxygen-utilizing processes (left-hand scale). The double line represents the resulting accumulation of oxygen in the atmosphere (right-hand scale). Some authorities believe that the net oxygen level rose earlier than depicted here. BY: billion years. PAL: present atmospheric level. (From Ref. 38.)

period, about 290 million years ago, as a result of production of oxygen by plants. At this time there was a dramatic increase in the evolution of both animals and plants with the development of many amphibia and a large vertebrate invasion of the land. Although some scientists argue that the oxygen concentration may have subsequently declined somewhat, there is other evidence that it has remained nearly constant since the Carboniferous period.

28

West

There is evidence that the evolution of life forms has not been a continuous process because of changes in the atmosphere. The extinction of the dinosaurs at the end of Mesozoic era, about 65 million years ago, was probably caused by the impact of an asteroid in the Yucatan, which resulted in an enormous dust cloud that interfered with photosynthesis and the food chain. Apparently an even greater extinction occurred at the end of the Permian era, about 250 million years ago, which decimated marine genera (4). Two other aspects of the evolution of the atmosphere should be mentioned because of their importance today. Oxygen screens out much ultraviolet radiation, but ozone (O3) is much more effective. Recently the ozone layer has been thinned in part as a result of the action of man-made contaminants, particularly chlorofluorocarbons, which are used as propellants and refrigerants. These are now being replaced, but a colorful reminder of the danger is seen in Australia where schoolchildren are required to wear legionnaire’s hats. Another contemporary aspect of atmospheric evolution is the apparent warming of Earth as a result of the greenhouse effect caused in part by man-made carbon dioxide emissions. The extent of this human contribution is controversial, but, in any event, carbon dioxide production is an almost inevitable consequence of an increased standard of living in many countries.

II. Altitude and Barometric Pressure A. History

History is always fun, but in this context it is also important because of recent misunderstandings. Torricelli (5) was probably the first to recognize that the air has weight (‘‘We live submerged at the bottom of an ocean of the element air’’ he wrote to his friend Michelangelo Ricci), and shortly thereafter Pascal (6) persuaded his brother-in-law, Perier, to take a barometer to the top of the Puy de Doˆme (1463 m), thus demonstrating the fall in pressure at increased altitude. A few years later Boyle (7) formulated his well-known law, which states that gas volume and pressure are inversely related (at constant temperature). In the nineteenth century, many travelers went to high altitude and measured barometric pressure. Typical of their results was a table reproduced in Appendix 1 of Paul Bert’s classical La Pression Barome´trique, published in 1878 (8). Bert acknowledged Jourdanet (9) as the source of much of the data. An interesting feature of Bert’s table is that he gives the height and barometric pressure at the summit of Mt. Everest as 8840 m and 248 mmHg, both of which are very nearly correct. Everest had been shown to be the world’s highest mountain in 1852. Zuntz et al. (10) gave the following relationship for determining barometric pressure at any altitude: logb ⫽ logB ⫺

h 72(256⋅4 ⫹ t)

The Atmosphere

29

where h is the altitude difference in meters, t is the mean temperature (°C) of the air column of height h, B is the barometric pressure (mmHg) at the lower altitude, and b is the barometric pressure at the higher altitude. Note that if temperature were constant, logb would be proportional to negative altitude, i.e., the pressure would decrease exponentially as altitude increased. The expression proposed by Zuntz et al. (which itself was derived from other sources) was extensively used by high altitude physiologists. For example, FitzGerald (11) employed it in her study of alveolar Pco2 and hemoglobin concentrations in residents of the Colorado mountains. Kellas (12) used the same expression to predict barometric pressures in the Himalayan ranges, obtaining the correct value of 251 mmHg for the summit of Mt. Everest, assuming a mean temperature of 0°C. The Zuntz et al. formula was also used by Haldane and Priestley (13). All would have been well had it not been for the introduction of the Standard Atmosphere by the aviation industry in the 1920s. It had been recognized for many years that the relationship between pressure and altitude was temperature dependent, as shown by the expression of Zuntz et al. (10) above. This meant that the relationship was also latitude dependent. However, there were clear advantages to the aviation industry in having a model atmosphere that applied approximately to mean conditions over the surface of Earth and could be used for calibrating altimeters and low-pressure chambers and assessing the performance of aircraft under well-defined conditions. This was referred to as the ICAO Standard Atmosphere (14) or the U.S. Standard Atmosphere (15), the two being identical up to altitudes of interest to highaltitude physiologists. The assumptions of the Standard Atmosphere were a sealevel pressure of 760 mmHg (1013 millibars), sea-level temperature of 15°C, and linear decrease in temperature with altitude (lapse rate) of 6.5°C/km up to an altitude of 11 km. It should be emphasized that the model atmosphere was never meant to be used to predict the actual barometric pressure at a particular location. In fact, it was developed by people who clearly recognized that there would be substantial local variations caused by latitude, season, and other factors. Unfortunately the Standard Atmosphere was taken up by many respiratory physiologists in the 1940s and inappropriately used to predict the pressure at various points on Earth’s surface, particularly on high mountains. The most glaring errors were apparent when the Standard Atmosphere was used for the pressure on Mt. Everest (16–19). The barometric pressure calculated in this way for the Everest summit (altitude 8848 m) is 236 mmHg, which is far too low as Bert clearly recognized in 1878 (8). Unfortunately the essentially correct relationships that had been used by Jourdanet (9), Bert (8), Zuntz et al. (10), FitzGerald (11), Kellas (12), and Haldane and Priestley (13) were disregarded. Perhaps the most dramatic consequence of the inappropriate use of the Standard Atmosphere was in 1949 during Operation Everest I. At that time four naval recruits were gradually decompressed to a pressure of 236 mmHg (thought to be the pressure on the summit), and their alveolar Po2 fell to as low as 21 mmHg (18), about 14 mmHg less than that of a well-acclimatized climber on the actual summit (20).

30

West

Figure 2 Relationship between barometric pressure and altitude. The upper curve is calculated from the formula of Zuntz et al. (10) using a mean air column temperature of 15°C. The lower curve follows the Standard Atmosphere (14). The plotted points represent observations made on Mt. Everest and elsewhere, mainly in the Andes. They clearly fit the upper curve better than the lower. (From Ref. 21.)

When expeditions to high altitude resumed after World War II, it soon became apparent that barometric pressures in the Himalayas and Andes exceeded those predicted from the Standard Atmosphere. This was clearly shown by Pugh (21) (Fig. 2) and clinched by the first direct measurement of barometric pressure above 8000 m and on the Everest summit during the course of the 1981 American Medical Research Expedition to Everest (22) (Fig. 3). The measured value of 253 mmHg was 17 mmHg higher than predicted from the Standard Atmosphere and this difference almost certainly made it possible to climb Mt. Everest without supplementary oxygen (23). B. Physical Principles

The reason why barometric pressure decreases with increasing altitude is, as Torricelli recognized, the decrease in the weight of the air above us. If the atmosphere were incompressible, as is nearly the case in a liquid, pressure would decrease linearly with altitude, just as it does in a liquid. However, because the air at lower altitude is compressed, barometric pressure decreases more rapidly with height near Earth’s surface. If the temperature of the gas were constant, the decrease would be

The Atmosphere

31

Figure 3 Barometric pressure-altitude relationships for two model atmospheres. The upper line was drawn using the model atmosphere equation (MAE) as described in the text. It shows how well this model atmosphere predicts the barometric pressure at 13 well-known locations where high altitude studies have been carried out, and the barometric pressures accurately measured. In numerical order the sites are: Collahuasi mine, north Chile; Aucanquilcha mine, north Chile; Vallot Observatory, Mont Blanc; Capanna Margherita, Monte Rosa; base camp, Mt. Everest; Camp 5, Mt. Everest; Summit, Mt. Everest; Cerro de Pasco, Peru; Morococha, Peru; Lhasa, Tibet; Crooked Creek, White Mountain, California; Barcroft Laboratory, White Mountain, California; Pike’s Peak, Colorado; White Mountain Summit, California. The lower line shows the Standard Atmosphere. (Modified from Ref. 24.)

exponential with respect to altitude [see the expression above from Zuntz et al. (10)]. However, because the temperature decreases, the pressure falls more rapidly than the exponential law predicts. The relationship between pressure, volume, and temperature is given by the ideal gas law: PV ⫽ nRT where P is pressure, V is volume, n is the number of gram molecules of the gas, R is the gas constant, and T is the absolute temperature. Both Boyle’s law, pressure times volume is a constant (at constant temperature), and Charles’ law, volume is proportional to absolute temperature (at constant pressure), follow from the ideal gas law.

32

West

Dalton’s law states that each gas in a mixture exerts a pressure according to its own concentration, independently of the other gases present. That is, each component behaves as though it were present alone. This follows from the kinetic theory of gases. The pressure of each gas is referred to as its partial pressure and is given by the total pressure times the fractional concentration of the gas: Px ⫽ PFx where Px is the partial pressure, P is the total pressure, and Fx is the fractional concentration of gas x. The fractional concentration always refers to dry gas. Graham’s law states that the rate of diffusion of a gas in the gas phase is inversely proportional to the square root of its density. This follows from the kinetic theory of gases, which states that the kinetic energy (0.5 mv2 where m is mass and v is velocity) of all gases is the same at a given temperature and pressure. One might therefore expect that very light gases such as hydrogen and helium would be lost from the upper atmosphere, and as we saw earlier, this did occur in the evolution of the atmosphere. However, up to altitudes of about 10 km, which are those of interest to us, convective mixing maintains the composition of the atmosphere constant. As altitude increases, the atmosphere can be divided on the basis of temperature into the troposphere, stratosphere, and the gas above. The troposphere is the region of interest to us and extends to an altitude of about 19 km at the equator but only about 9 km at the poles. In it, the temperature decreases approximately linearly with altitude down to about ⫺60°C. Above the troposphere, the temperature in the stratosphere remains nearly constant at about ⫺60°C for some 10–12 km of altitude. C. Model Atmospheres and Pressures on High Mountains

As explained in the historical summary above, the introduction of the Standard Atmosphere caused great confusion in high-altitude physiology because the pressures it predicts on high mountains are generally too high. However, other model atmospheres have been developed by geophysicists. The critical variable is the change of air temperature with altitude, and therefore model atmospheres have been constructed for different latitudes and seasons of the year. These different models give a large range of pressures at a given altitude. For example, the maximum difference of pressure at an altitude of 9 km is from 206 to 248 torr, i.e., about 20%. However, it has been shown that the mean of the model atmospheres for latitude 15° (in all seasons) and 30° (in the summer) predicts the barometric pressure (PB) at many locations of interest at high altitude very well (24), with predictions being within 1%. This model atmosphere equation (MAE) is PB (torr) ⫽ exp (6.63268 ⫺ 0.1112 h ⫺ 0.00149 h2), where h is the altitude in km. The predictions are good because many high mountain sites are within 30° of the equator, and also many studies are made during the summer. Figure 3 shows the relationship between barometric pressure and altitude for this equation (top line) and demonstrates how

The Atmosphere

33

successfully it predicts the barometric pressure of 13 well-known sites where highaltitude studies have been carried out. Figure 3 also shows the U.S. Standard Atmosphere, which predicts pressures that are too low. The equation is PB (torr) ⫽ exp (6.63266 ⫺ 0.1176 h ⫺ 0.00163 h2). D. Effects of Latitude

Figure 4 shows the barometric pressure at different altitudes plotted against latitude for summer and winter. It can be seen that the pressure at Earth’s surface and at an altitude of about 24 km is essentially independent of altitude. However, in the altitude range of about 6–16 km there is a pronounced bulge in barometric pressure near the equator in both summer and winter. Since the latitude of Mt. Everest is

Figure 4 Increase of barometric pressure near the equator at various altitudes in both summer and winter. Vertical axis shows the pressure increasing upward according to the scale on the right. The numbers on the left show the barometric pressures at the poles for various latitudes. The altitude of Mt. Everest is 8848 m. (From Ref. 25.)

34

West

28°N, the pressure at its summit (8848 m) is considerably higher than would be the case for a hypothetical mountain of the same altitude near one of the poles (see also Fig. 6). The cause of the increase in barometric pressure near the equator is a very large mass of very cold air in the stratosphere above the equator (25). Paradoxically, the coldest air in the atmosphere is above the equator. This is brought about because the heat of the sun near the equator causes expansion of the air and a large upwelling of the atmosphere. For the same reason, the height of the tropopause is much greater (19 km) near the equator than near the poles (9 km), as pointed out previously. E.

Effects of Season of the Year

Summer and winter greatly affect the relationship between barometric pressure and altitude because of the temperature effects. Figure 5 shows the mean monthly pressures for an altitude of 8848 m (Everest summit) as obtained from radio-sonde balloons released from New Delhi, India, over a period of 15 years (Delhi is at about the same latitude as Everest.) Note that the mean pressures in the winter months of January and February (243.0 and 243.7 mmHg, respectively) were substantially lower than in the summer months of July and August (254.5 mmHg for both

Figure 5 Mean monthly pressures for 8848 m altitude as obtained from weather balloons released from New Delhi, India. Note the increase during the summer months. The mean monthly standard deviation (S.D.) is also shown. The barometric pressure measured on the Everest summit on October 24, 1981 (*) was unusually high for that month. (From Ref. 22.)

The Atmosphere

35

˙ o2max near the months). This 11 mmHg difference of pressure will certainly affect V ˙ summit where the slope of the Vo2max-pressure line is about 13 mL/min/torr (23). ˙ o2max at the Everest Therefore, a fall of barometric pressure of 11 mmHg will reduce V summit by about 143 mL/min, i.e., by about 13%. Figure 5 also shows the direct measurement of barometric pressure on the Everest summit made by Christopher Pizzo on October 24, 1981. This was unusually high for the month because Pizzo was fortunate enough to have an exceptionally fine day for his summit climb. The temperature on the summit was measured as ⫺9°C, some 20–30°C higher than expected for that altitude. Another direct measurement of barometric pressure on the Everest summit was made by David Breashears at 7:00 a.m. local time on May 23, 1997, during the course of an expedition sponsored by the television station WGBH. The handheld barometer (Pretel Alti Plus K2) was calibrated in our laboratory after the expedition, and the pressure on the summit was within a mmHg or so of the value of 253 mmHg obtained by Pizzo. Data from radio-sonde balloons were also obtained for the same day and time, and the interpolated pressure at an altitude of 8848 m agreed well with the measured value. An interesting point was that Everest was near the center of a zone of high atmospheric pressure. As Figure 5 shows, the pressures for May and October are expected to be similar. Over 2000 measurements of barometric pressure were made on the South Pole (altitude 7986 m) in 1998 and the results agreed well with the pressure–altitude relationship for the summit (25a). The effects of latitude and season are combined in Figure 6. All the data are from radio-sonde balloons, which are released from meteorological stations all over the world twice a day. The values in Figure 6 are for an altitude of 8848 and are the means from all longitudes (26). The dramatic effects of altitude and season are clearly seen. It is interesting that in midsummer the pressure reaches a maximum near the latitude of Mt. Everest (28°N). Figure 6 also shows that if Everest were at the latitude of Mt. McKinley (63°N), the difficulties of reaching the summit without supplementary oxygen would be enormously increased. It is possible to obtain the relationship between barometric pressure and altitude all over the world on any particular day from the radio-sonde balloon measurements. Details on how to extract these data are given in West (27). The calculations show that when Messner and Habeler made the first ascent of Everest without supplementary oxygen in 1978, the pressure on the summit was 251 mmHg. When Messner made the first solo ascent without supplementary oxygen in August 1980, he was fortunate that the barometric pressure was unusually high at 256 mmHg. The only winter ascent so far was on December 22, 1987, by Ang Rita Sherpa when the pressure was only 247 mmHg. F. Physiological Significance of Barometric Pressure at High Altitude

How important are these differences of barometric pressure in determining physical performance at extreme altitude? An analysis of the factors determining maximal exercise during extreme hypoxia has been carried out using data from the 1981

36

West

Figure 6 Barometric pressure at the altitude of Mt. Everest plotted against latitude in the northern hemisphere for midsummer, midwinter, and a preferred month for climbing in the postmonsoon period (October). Note the considerably lower pressures in the winter. The arrows show the latitudes of Mt. Everest and Mt. McKinley. (Modified from Ref. 22.)

American Medical Research Expedition to Everest as a basis (23). The relationship ˙ o2max and inspired Po2 found on Operation Everest II was very similar between V ˙ o2max at extreme (28). The analysis showed that although many factors influence V altitudes including the level of alveolar ventilation, lung diffusing capacity, cardiac output, hemoglobin concentration, P50 of the blood, and base excess of the blood, the most critical factor is the barometric pressure (Fig. 7). The reason for this is that the pressure determines the inspired Po2. One of the conclusions of the analysis was that it would be impossible to climb the mountain without supplementary oxygen if the pressure on the summit conformed to the Standard Atmosphere value. In addition, the variations of barometric pressure with season shown in Figures 5 and 6 indicate that it would be considerably more difficult to reach the summit without supplementary oxygen in the winter as a result of the reduced inspired Po2, quite apart from the obvious difficulties of very low temperatures. It is significant that,

The Atmosphere

37

Figure 7 Percentage increase in maximal oxygen consumption for a climber on the summit of Mt. Everest when various parameters are increased by 5% leaving all the other variables ˙ o2max is most sensitive to barometric pressure. See original article constant. Note that the V for details. (From Ref. 23.)

although there have been many ascents of Everest without supplementary oxygen in the premonsoon and postmonsoon seasons, only one person has made a winter ascent without supplementary oxygen, and this was on the first day of winter, December 22, 1987. III. Factors Other Than Barometric Pressure at High Altitude Temperature falls with increasing altitude at the rate of about 1°C per 150 m, and this lapse rate is essentially independent of latitude. As a result, the average temperature near the summit of Mt. Everest throughout the year is about ⫺40°C. Of course most climbers choose the pre- and postmonsoon seasons when the temperature is higher. Wind chill factors are even more important. Wind velocities on Himalayan peaks have been estimated to reach in excess of 150 km/h, although there are few direct measurements. Absolute humidity, i.e., the amount of water vapor per unit volume of gas at the prevailing temperature, is extremely low at high altitude because water vapor pressure is reduced at low temperatures. For example, while the water vapor pressure at ⫹20°C is 17 mmHg, it is only 1 mmHg at ⫺20°C. The result is that climbers at

38

West

high altitude are almost always dehydrated. The insensible water loss caused by ventilation is very high because of the dryness of the inspired air and the high levels of ventilation, especially on exercise. Water loss also occurs through sweating, and because of the dryness of the air, sweat rapidly evaporates. Water is difficult to provide because snow must first be melted. Pugh (29) showed that climbers at altitudes over 6000 m require 3–4 L of fluid per day to maintain a urine output of 1.5 L per day. Even so, it is likely that many people at high altitude are in a state of chronic volume depletion. Members of the 1981 American Medical Research Expedition to Everest who were living at an altitude of 6300 m had a significantly increased serum osmolarity compared with sea level in spite of ample fluids being available (30). Solar radiation is extremely intense at high altitude because of the thinner atmosphere and reflection of the radiation from snow. Ionizing radiation is also greater at high altitude because of less absorption by the atmosphere.

IV. Artificial Atmosphere at High Altitude The hypoxia of high altitude results from the low barometric pressure, which, because the oxygen concentration of the air is independent of altitude, causes a low inspired Po2. One way to relieve the hypoxia is to artificially increase the pressure by placing the subject in a lightweight rubberized canvas bag and pressurizing this with a foot pump. Increasing the pressure by about 100 mmHg gives a reduction of equivalent altitude of about 1500 m at altitudes of 4000–5000 m. At least two bags are commercially available: the Gamow from the United States and Certec from France. There have been numerous accounts of their effective use in treating high-altitude pulmonary edema and high-altitude cerebral edema (31). The other potential mechanism for reducing the hypoxia is to raise the oxygen concentration of the atmosphere. There is great deal of interest in this at the present time partly because of the increase in commercial and scientific activities at altitudes of 4000–5500 m. One example is a new mine in north Chile at an altitude of 4500– 4600 m. This mine will employ up to 2000 people, and most of these workers will have their homes at sea level and be bused up to the mine for 7 days and then down to join their families for a further 7 days, with the cycle continuing indefinitely. Another example is a radiotelescope that is planned for an altitude of 5000 m in north Chile. The personnel will live at an altitude of about 2400 m and drive up to the facility every day. These altitudes result in severe hypoxia, with impairment of central nervous system (CNS) function, quality of sleep, and physical work capacity. The use of oxygen-enriched atmospheres in selected parts of these facilities shows great promise (32). Relatively small degrees of oxygen enrichment result in large improvements in inspired Po2. For example, every 1% increase in oxygen concentration (e.g., from 21 to 22%) is equivalent to a reduction of altitude of about 300 m (Fig. 8). Thus, increasing the oxygen concentration of the atmosphere to 26%

The Atmosphere

39

Figure 8 The vertical axis shows the meters of equivalent altitude descent when the oxygen concentration of the atmosphere is increased by 1% at the altitude shown on the horizontal axis. Note that at altitudes up to about 5000 m, each 1% of oxygen enrichment, results in an altitude reduction of more than 300 m. (From Ref. 32.)

at an altitude of 4500 m reduces the equivalent altitude to 3000 m, which is much more easily tolerated. Places where oxygen enrichment has been considered include dormitories, cabins of heavy equipment such as mechanical shovels and trucks, offices, conference rooms, and laboratories. One of the reasons why oxygen enrichment has become more feasible is that large amounts of oxygen-enriched air can now be produced relatively cheaply. Liquid oxygen is being used in pilot studies in north Chile. A less expensive source of large amounts of oxygen is an oxygen concentrator, which uses a molecular sieve. Thousands of these are now in use in the homes of patients with severe lung disease. The sieve consists of a nonflammable ceramic material, for example, synthetic zeolite. Air is pumped through the sieve at high pressure and nitrogen is preferentially adsorbed, with the result that the effluent gas is oxygen-enriched. Periodically the sieve has to be regenerated by pumping air through it at low pressure to wash out the excess nitrogen, and this is easily accomplished by having two sieves, one of which is undergoing regeneration at any one time. These oxygen concentrators have an indefinite life and can easily produce large volumes of 90% oxygen, which is just as useful in the present application as pure oxygen. Recently a pilot study of oxygen enrichment was carried out at the Tambo mine in north Chile, altitude 4300 m. Sixteen dormitory rooms had the oxygen

40

West

concentration raised to 24–25% using a liquid oxygen source and appropriate ducting. The sleep quality of the miners using oxygen enrichment was compared with another group sleeping in ambient air. Measurements showed that the oxygen-enriched group slept better. They had fewer arousals (waking up), respiratory movements were less abnormal with fewer apneas and less total apneic time, and the EEG showed longer periods of deep phase 4 sleep. Psychometric tests performed during the day following the test night showed small but consistent improvement in six psychometric tests, including attention, concentration, and short-term memory. There was also evidence of increased arterial oxygen saturations during the day. These preliminary results are consistent with those obtained by treating patients with obstructive sleep apnea at sea level with continuous positive airway pressure (CPAP). Such patients have improved psychometric function, higher arterial Po2 values, and a change in CO2 sensitivity as a result of treatment (33–35). A gratifying feature of the study was how easy it was to accomplish reliable oxygen enrichment in this relatively large dormitory facility. This pilot study strongly suggests that oxygen enrichment on a large scale in such a commercial facility may well improve the well-being of the workers and therefore the productivity of the mine. Possible disadvantages of oxygen enrichment should also be considered. Some people immediately think of a possible fire hazard, but in fact the flammability of paper and cotton in 26% oxygen at an altitude of 4500 m is less than in air at sea level (36). Flammability of an atmosphere depends on two variables (37): (1) the partial pressure of oxygen, which is much lower in the enriched air at high altitude than at sea level, and (2) the quenching effect of the inert components, i.e., nitrogen, of the atmosphere. This quenching is slightly reduced at high altitude, but the net effect is still a lower flammability. Loss of acclimatization to high altitude is sometimes cited as a possible disadvantage of oxygen enrichment. However, there is no basic difference between entering a room with an oxygen-enriched atmosphere and driving down to a lower altitude. Everybody would sleep at a lower altitude if they could. It is true that frequent exposure to a lower altitude will result in less acclimatization to the higher altitude, other things being equal. However, the ultimate objective is effective working at high altitude, and this can be enhanced using oxygen enrichment. Another argument is that altering the atmosphere in this way might increase the legal liability of the facility if some kind of hypoxia-related illness develops. Actually, the opposite view makes more sense. It is possible that a worker who develops a myocardial infarction while working at high altitude could claim that the altitude was a contributing factor. Any procedure that reduces the equivalent altitude makes altitude-induced illnesses less likely. Oxygen enrichment of room air at high altitude could be a major advance in the development of commercial and scientific facilities at altitudes over 4000 m. Everybody now expects that the ventilation of a room will provide a comfortable temperature and humidity. Control of the oxygen concentration can be regarded as a further logical step in humans’ control of their environment.

The Atmosphere

41

Acknowledgment This work was supported by NIH grant R01 HL 46910. References 1. 2. 3. 4. 5.

6.

7.

8. 9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19.

Oro´ J. Early chemical stages in the origin of life. In: Bengtson S, ed. Early Life on Earth, Nobel Symposium No. 84. New York: Columbia University Press, 1994:48–59. Miller SL. A production of amino acids under possible primitive earth conditions. Science 1953; 117:528–529. Kasting JF. Earth’s early atmosphere. Science 1993; 259:920–926. Chapman CR, Morrison D. Cosmic Catastrophes. London: Plenum Press, 1989:97. Torricelli E. Letter of Torricelli to Michelangelo Ricci. 1644. English translation of relevant pages in: West JB, ed. High Altitude Physiology. Stroudsburg, PA: Hutchinson Ross Publishing Co., 1981:60–63. Pascal B. Story of the great experiment on the equilibrium of fluids. 1648. English translation of relevant pages in: West JB, ed. High Altitude Physiology. Stroudsburg, PA: Hutchinson Ross Publishing Co., 1981:64–69. Boyle R. New Experiments Physico-Mechanical, Touching the Air. 2nd ed. Oxford: Thomas Robinson, 1662. Relevant pages reprinted in: West JB, ed. High Altitude Physiology. Stroudsburg, PA: Hutchinson Ross Publishing Co., 1981:70–75. Bert P. La Pression Barome´trique. Paris: Masson, 1878. English translation by Hitchcock MA, Hitchcock FA. Columbus, OH: College Book Co., 1943. Jourdanet D. Influence de la pression de l’air sur la vie de l’homme. Paris: Masson, 1875. Zuntz N, Loewy A, Muller F, Caspari W. Atmospheric pressure at high altitudes. In: Ho¨henklima und Bergwanderungen in ihrer Wirkung auf den Menschen. Berlin: Bong, 1906:37–39. Translation of relevant pages in: West JB, ed. High Altitude Physiology. Stroudsburg, PA: Hutchinson Ross Publishing Co., 1981:78–80. FitzGerald MP. The changes in the breathing and the blood of various altitudes. Phil Trans R Soc Lond Ser B 1913; 203:351–371. Kellas AM. A consideration of the possibility of ascending the loftier Himalaya. Geogr J 1917; 49:26–47. Haldane JS, Priestley JG. Respiration. 2nd ed. London: Oxford University Press (Clarendon), 1935. ICAO. Manual of the ICAO Standard Atmosphere. 2nd ed. Montreal, Quebec: International Civil Aviation Organization, 1964. NOAA. U.S. Standard Atmosphere. Washington, DC: National Oceanic and Atmospheric Administration, 1976. Houston CS, Riley RL. Respiratory and circulatory changes during acclimatization to high altitude. Am J Physiol 1947; 149:565–588. Houston CS, Sutton JR, Cymerman A, Reeves JT. Operation Everest II: man at extreme altitude. J Appl Physiol 1987; 63:877–882. Riley RL, Houston CS. Composition of alveolar air and volume of pulmonary ventilation during long exposure to high altitude. J Appl Physiol 1951; 3:526–534. Rahn H, Fenn WO. A Graphical Analysis of the Respiratory Gas Exchange. Washington, DC: American Physiology Society, 1955.

42 20.

West

West JB, Hackett PH, Maret KH, Milledge JS, Peters RM, Jr., Pizzo CJ, Winslow RM. Pulmonary gas exchange on the summit of Mt. Everest. J Appl Physiol 1983; 55:678– 687. 21. Pugh LGCE. Resting ventilation and alveolar air on Mount Everest: with remarks on the relation of barometric pressure to altitude in mountains. J Physiol (London) 1957; 135:590–610. 22. West JB, Lahiri S, Maret KH, Peters RM, Jr., Pizzo CJ. Barometric pressures at extreme altitudes on Mt. Everest: physiological significance. J Appl Physiol 1983; 54:1188– 1194. 23. West JB. Climbing Mt. Everest without oxygen: an analysis of maximal exercise during extreme hypoxia. Respir Physiol 1983; 52:265–279. 24. West JB. Prediction of barometric pressures at high altitudes using model atmospheres. J Appl Physiol 1996; 81:1850–1854. 25. Brunt D. Physical and Dynamical Meterology. 2nd ed. Cambridge, UK: Cambridge University Press, 1952:379. 25a. West JB. Barometric pressures on Mt. Everest: new data and physiological significance. J Appl Physiol 1999; 86:1062–1066. 26. Oort AH, Rasmusson EM. Atmospheric Circulation Statistics. Rockville, MD: U.S. Dept. of Commerce, NOAA, 1971:84–85. 27. West JB. Acclimatization and tolerance to extreme altitude. J Wilderness Med 1993; 4:17–26. 28. Sutton JR, Reeves JT, Wagner PD, Groves BM, Cymerman A, Malconian MK, Rock PB, Young PM, Walter SD, Houston CS. Operation Everest II: oxygen transport during exercise at extreme simulated altitude. J Appl Physiol 1988; 64:1309–1321. 29. Pugh LGCE. Animals in high altitudes: man above 5000m—mountain exploration. In: Dill DB, Adolph EF, Wilber CG, eds. Handbook of Physiology, Section 4. Washington, DC: American Physiological Society, 1964:861–868. 30. Blume FD, Boyer SJ, Braverman LE, Cohen A. Impaired osmoregulation at high altitude. J Am Med Assoc 1984; 252:1580–1585. 31. Robertson JA, Shlim DR. Treatment of modern acute mountain sickness with pressurization in a portable hyperbaric (Gamow) bag. J Wilderness Med 1991; 2:268–273. 32. West JB. Oxygen enrichment of room air to relieve the hypoxia of high altitude. Respir Physiol 1995; 99:225–232. 33. Engleman HM, Martin SE, Deary IJ, Douglas NJ. Effect of continuous positive airway pressure treatment on daytime function in sleep apnoea/hypopnoea syndrome. Lancet 1994; 343:572–575. 34. Leech JA, Onal E, Lopata M. Nasal CPAP continues to improve sleep-disordered breathing and daytime oxygenation over long-term follow-up of occlusive sleep apnea syndrome. Chest 1992; 102:1651–1655. 35. Berthon-Jones M, Sullivan CE. Time course of change in ventilatory response to CO2 with long-term CPAP therapy for obstructive sleep apnea. Am Rev Respir Dis 1987; 135:144–147. 36. West JB. Fire hazard in oxygen-enriched atmospheres at low barometric pressures. Aviat Space Environ Med 1997; 68:159–162. 37. Roth EM. Space Cabin Atmospheres: Part II, Fire and Blast Hazards. Washington, DC: NASA Report SP-48, 1964.

3 The People

SUSAN NIERMEYER University of Colorado Health Sciences Center Denver, Colorado

STACY ZAMUDIO and LORNA G. MOORE University of Colorado University of Colorado Health Sciences Center Denver, Colorado

I.

Introduction

Nearly 140 million people reside at high altitude, defined as elevations above 2500 m (8000 ft) (Table 1). While this is a small fraction of the world’s population, it comprises a sizable proportion of the population in certain countries or regions. The numbers of high-altitude residents are likely to increase as the result of population growth and active migration in many of these regions. The fundamental adaptive challenge at high altitude is hypoxia. Hypoxia results from the lower barometric pressure, and hence partial pressure of oxygen, at elevations above sea level. There are other distinctive characteristics of the highaltitude environment, including increased solar radiation, greater diurnal temperature fluctuation, aridity, low biomass, and limitations on energy production. Even though the hypoxia of high altitude cannot be readily modified by culture for at least the majority of persons, cultural factors can exert important influences by conditioning biological responses. Here and throughout, we use ‘‘adaptation’’ to refer to a feature of structure, function, or behavior that enables an organism to live and reproduce in a given environment. These features may be biological or behavioral in nature, genetic or developmental in origin. In most instances, a combination of factors is involved. 43

Table 1

Estimated Numbers of Persons Residing ⬎2500 m in 1995

Region Country Province or state Africa Ethiopia Kenya Rwanda Uganda Zaire Asia Afghanistan Bhutan China Inner Mongolia Qinghai Sichuan Tibet Yunnan Xinjian Uygur Zizhiqu India Himchal Pradesh Jammu and Kashmir Sikkim Utar Pradesh Kazakstan Kirghizistan Nepal Pakistan Tajikistan Central and South America Argentina Bolivia Chile Colombia Ecuador Guatemala Mexico Peru Venezuela North America United States Colorado Utah Total

Annual population growth (%)

Estimated % ⬎ 2500 ma

55,053,000 28,261,00 7,952,000 21,297,000 43,901,000

2.9 2.1 2.6 2.9 2.1

25 10 15 10 10

20,141,000 1,638,000 1,221,462,000 8,960,000 4,120,000 104,070,000 2,150,000 31,920,000 13,368,000 935,744,000 4,269,569 6,981,600 425,000 112,858,019 14,984,100 4,698,000 21,918,000 140,497,000 6,101,000

5.6 2.4 1.0

2.3 2.5 2.5 2.8 2.7

10 45 ⬃2 20 40 5 80 20 40 ⬃3 30 40 40 20 20 10 35 10 30

34,587,000 7,414,000 14,262,000 35,101,000 11,460,000 10,621,000 93,674,000 23,780,000 21,844,000

1.2 2.3 1.4 1.5 2.0 2.8 1.8 1.8 2.0

5 40 10 20 15 10 15 20 5

263,250,000 3,068,000 1,542,000

0.9

⬍1 10 5

Total population

1.8

Estimated no. ⬎ 2500 m 24,301,950 13,763,250 2,826,100 1,192,800 2,129,700 4,390,100 78,677,965 2,014,100 737,100 22,094,700 1,792,000 1,648,000 5,203,500 1,720,000 6,384,000 5,347,200 26,815,115 1,280,871 2,792,640 170,000 22,571,604 2,996,820 469,800 7,671,300 14,049,000 1,830,030 35,821,750 1,729,350 2,965,600 1,426,200 7,020,200 1,719,000 1,062,100 14,051,100 4,756,000 1,092,200 383,900 383,900 306,800 77,100 139,185,565

Population estimates ⬎2500 m for each country, province, or state were made using the total population size, geographic area ⬎2500 m, population size of the largest cities, and population density. Source: Refs. 1,2.

a

The People

45

By ‘‘genetic adaptation’’ we mean a heritable characteristic whose presence reflects the operation of natural selection or other processes modifying gene frequencies over time (e.g., mutation, genetic drift, gene flow). We use ‘‘developmental adaptation’’ to refer to physiological traits that are acquired in response to prolonged exposure to stress(ors). While developmental processes are often considered distinct from genetic ones, the capacity to acquire the trait may be genetically based. ‘‘Acclimatization’’ is used to refer to the time-dependent physiological responses that occur following exposure to high altitude. The distinction between these terms is that adaptations confer evolutionary benefit (enhanced fitness), whereas the physiological responses of acclimatization may or may not be adaptive. A great deal has been learned in recent years about the mechanisms underlying physiological responses to hypoxia. We begin with a consideration of the extent to which the geographical and historical circumstances differ among current highaltitude populations in order to judge whether some high-altitude groups have lived longer at altitude than others. We also review some of these group’s cultural practices that condition their biological responses. The major portion of the chapter is a survey of recent studies of the physiological responses to hypoxia in Himalayan, Andean, and Rocky Mountain high-altitude residents. Our consideration excludes East African highland residents simply because of the lack of recent study in this region. We examine each phase of the life cycle; namely, pregnancy and fetal development, infancy, childhood, adolescence, adulthood, and old age. We end with a consideration of the possible future directions for research on long-term high-altitude residents. In particular, we address recent developments in evolutionary theory and genetic methodology that bear on our ability to detect and interpret genetic differences between populations. Incorporating such perspectives into future studies of high-altitude populations will advance our understanding of the specific ways in which genetic traits interact with physiological and cultural processes.

II. History of Human High-Altitude Habitation A. Duration of High-Altitude Occupation by Region

World high-altitude populations likely vary in their duration of high-altitude residence and degree of genetic admixture with lowland groups. The comparison of these groups is of interest from an evolutionary perspective, since populations that have spent the longest time at high altitude and have the least degree of admixture from lowland populations can be expected to be the best adapted. The Himalayan (Tibetan) Plateau is the largest and most geographically isolated of the high-altitude regions. It is roughly oval in shape, stretching 2400 km (1500 miles) east to west and 1100 km (700 miles) along its north-south axis and encompassing over 200 million hectares (800,000 square miles). The distance to the nearest sea coast, the Bay of Bengal, ranges from 800 to 2400 km (500 to 1500 miles). On the south, it is flanked by the world’s highest mountains and by peaks

46

Niermeyer et al.

reaching 7600 m on the north, west, and east. Only at the northeastern edge does it descend gradually to the low-altitude region drained by the Huang (Yellow) River. Hominids (the taxonomic family to which humans, not apes, belong) have been present in Asia for a million or more years. This is longer than the duration of hominid occupation of North and South America but not as long as in Africa. Paleoliths and microliths consistent with northern Asian tool cultures of the Upper Paleolithic, dating to approximately 25,000–50,000 years ago, have been found at 4500–5200 m sites on the northern Tibetan Plateau (3,4). Even older material from the late Pliocene (approx. 2 million years ago) has been reported from the Tibetan Plateau in northern Pakistan (5). While these data cannot provide assurance of genetic continuity with current inhabitants or when permanent occupation of the Plateau began, they support the possibility that humans have been on the Tibetan Plateau for upward of 50,000 years. Genetic and linguistic studies support a long period of residence for the Tibetan population in its current location. Dental morphology, mitochondrial, and nuclear genetic markers show Tibetans to be related to Korean, Siberian, and Mongolian populations, likely to be descendants of the early inhabitants of Asia, and to differ considerably from Han (‘‘Chinese’’) and other southeast Asian populations (6–10). The Tibetans’ membership in the Tibeto-Burmese language group differentiates them from southeast and north central Asian (including Mongolian) populations and indicates that Tibetans have resided in their south central location long enough for linguistic separation to have occurred. Contact has occurred between Tibetans and other populations via trade and, in the thirteenth and fourteenth centuries, conquest by Mongolians. But the major trading routes (e.g., the Silk Road) avoided the Tibetan Plateau, and the period of Mongolian domination was relatively brief (⬃100 years) and more in the form of patronage than subjugation (11). Therefore, occupation of Tibet by foreigners has been relatively limited until recently in terms of geographic penetration, impact on survival, and well-being. Genetic divergence increases when populations are separated by physical, linguistic, and other cultural barriers, and the western and eastern boundaries of Tibet show such genetic divergence (12,13). Thus, the expectation would be that Tibetans are genetically distinct from adjacent populations except those to the northeast. However, no specific genetic admixture estimates are available between Tibetans and other populations, including the Sherpas and Mongolians. Such information would be useful for determining admixture rates and the genetic history for this region of the world. The Andean Altiplano extends nearly 4800 km (3000 miles) along nearly the whole of South America, averaging 200 km (125 miles) wide and encompassing nearly 100 million hectares (400,000 square miles). The Amazon River basin extends to the east and broad grassland plains flank its southernmost portion. The Pacific coastline parallels the altiplano 75–150 km (50–100 miles) to the west. The region between the altiplano and Pacific coast is generally dry but is punctuated by littoral plains, irrigated valleys, and broad harbors. Native Americans derived from north central Asian populations, who migrated in several waves over the Beringian land bridge which was exposed periodically

The People

47

between approximately 65,000 and 15,000 years ago (14). There is comparatively little mitochondrial DNA (mtDNA) variation among the indigenous populations of the Americas. Existing variation can be accounted for by as few as four mtDNA lineages (15–17). This may reflect a small number of initial migrants (e.g., as few as four women plus some men) or simply limited mtDNA variation among a larger number of persons. Archaeological evidence suggests that humans were present in South America as early as 9,000–12,000 years ago (18–20). Permanent residence of the Andean region required domestication of the potato, quinoa and other grains, camelids (llamas, alpacas), and guinea pigs, which began about 6,000 years ago and was complete by 4,000 years ago (18–20). Somewhat surprisingly, the majority of Quechua and Aymara belong to separate mtDNA lineages (15) and linguistic groups, even though they reside next to each other on the altiplano and have much in common with each other culturally. Such patterns of mtDNA variation may indicate considerable stability and separation of women (maternal lineages) from the two language groups or may be the result of stochastic, genetic drift within small populations (21). The Pacific Coast was the point of contact between the indigenous civilizations of the Andean and European powers in the 1500s. Contact had devastating consequences for the Andean civilizations. Not only were its rulers killed and cities looted, but within 100 years following conquest population size declined from 12 million to 675,000 as the result of pandemics of infectious disease, malnutrition, and forced resettlement (22). This loss of 95% of the indigenous population meant that the surviving fraction underwent an evolutionary ‘‘bottleneck,’’ and, as a result, it is by no means certain that altitude-related traits happened to be preserved. Today, estimates of admixture in Andean populations follow a gradient wherein coastal populations show the most European admixture, sierra (intermediate altitude) populations are intermediate, and altiplano (high altitude) populations show the least admixture. In Ecuador, Peru, and Bolivia the estimates of European admixture vary from 5 to 30% (21,23,24) (R. E. Ferrell, personal communication). There is good correlation between genetic data and surname (23). In groups with Aymara patronyms and matronyms, 89% of the genes surveyed were of Amerindian origin. Spanish-surnamed individuals had fewer Amerindian genes but nonetheless a substantial degree of Indian ancestry, with 67% of their genes being of Amerindian origin. However, given the likelihood that contemporary Andeans are derived from populations whose genetic variation had been severely curtailed, the Amerindian genes present may or may not be of Andean origin and/or contain traits related to high-altitude existence. The Rocky Mountain Plateau encompasses an oval region, approximately 1,200 km (750 miles) long and 400 km (250 miles) wide or nearly 40 million hectares (150,000 square miles) in the western United States in Wyoming, Colorado, Utah, and New Mexico, with the highest segment being in Colorado. Amerindians lived seasonally in this region, but it was not inhabited permanently until 150 years ago. Its current residents are genetically heterogeneous, having descended from lowaltitude European, Amerindian, and Hispanic populations. Thus, the high-altitude

48

Niermeyer et al.

populations of the Rocky Mountain region are, by virtue of their short period of residence, not genetically adapted to high altitude. In summary, it is likely that persons have lived the longest on the Tibetan Plateau, for an intermediate length of time on the Andean altiplano, and for the shortest length of time in the Rocky Mountain region. The greater size, geographic isolation, remoteness from coastal regions, and absence, until recently, of conquest by low-altitude groups also suggests that the residents of the Tibetan Plateau have been subject to less reduction in genetic variation and less genetic admixture with low-altitude groups than the Andean residents. B. Consideration of Cultural Influences

High-altitude residents employ cultural practices that modify environmental stressors and thus condition biological responses. Indigenous cultural practices affecting food sources, energy expenditure, and population movement provide examples of the ways in which such conditioning takes place. Plants and animals genetically adapted to high-altitude environments have been traditionally relied upon as food sources. In Tibet, barley is the staple, with green vegetables being added in the summer and dried vegetables and root crops such as potato and turnip eaten in the winter (25). The staple foods in Andean diets consist of numerous tubers (over 2500 kinds of potato, ulloco, oca, mashua), quinoa, and other grains (kiwicha, tarwi, cannihua) (26). The Quechua practice of freezedrying potatoes (chun˜o) and meats (charqui) preserves them for long periods and reduces their weight, facilitating their transport by groups using widely dispersed resources (27). Other crops (peanuts, beans, fruits, and coca) from low-altitude regions supplement the Andean diet and, in the case of coca, have nutritional benefits as well as narcotic effects (26). Animals indigenous to the high-altitude environment—the llama and guinea pig of the Andes and the yak of the Himalayas—are good sources of food, clothing, and fuel. The llama and yak are particularly so, providing a source of transport, meat, wool, rope, leather, dung for fuel and fertilizer, and, in the case of the yak, milk for butter, cheese, and yogurt as well as labor for pulling the plow. Exchange of resources between altitudes plays an important part in the availability of food in high-altitude regions. In the Andes, an extensive network of Inca roads or wide footpaths links the highland and lowland areas. These have been joined more recently by highways, railroads, and air travel. Animal resources (wool, hides, meat) from the Andean altiplano are exchanged for wheat and other foods grown at lower elevations (27). In Tibet, most exchange of crops and animal products occurs between farmers and pastoralists within a region (25). Between regions, trade routes and relationships between monasteries served historically to move products over larger distances. Until the 1950s, nearly all trade was by foot or pack animal because wheeled vehicles were prohibited by Buddhist beliefs and railroads have yet to penetrate the Tibetan Plateau. Today, highways and air travel link the Tibetan Plateau with lowland regions.

The People

49

Concentrated deposits of minerals with high monetary value are exposed in the mountainous environment and contribute importantly to economic exchanges in the Andean and Rocky Mountain regions. Mining has also been practiced on the Tibetan Plateau but to a lesser extent, at least until recently. The monetary value justifies high levels of expenditure for employing large numbers of persons and purchasing, in the case of Rocky Mountain residents, virtually all the required food stuffs from low-altitude regions. Recreational tourism provides an increasingly important source of income for Rocky Mountain residents. Exchanges within households are important in the production and consumption of resources. In traditional subsistence economies, food is generally produced by adults and adolescents and distributed to the young and older-aged members of the household. This distribution pattern serves to minimize seasonal change in caloric consumption. For example, in rural highland Peru, preharvest household caloric consumption was less than half that present postharvest (28). However, preferential distribution of food to children, the reduction of household consumption by changes to less energy-intensive activities, and the temporary out-migration of adolescent and adult males protected children from seasonal shortages (29). The use of children or adolescents in herding serves to reduce household caloric consumption since a 12-year-old child can complete the herding work of an adult with 30% fewer calories (27). In the Lhasa valley, men and women eat the same kinds and generally the same amounts of food. The relationship of food need to energy expenditure is recognized by Tibetan farmers; those who do the most work eat the most, regardless of sex. Children do not eat as much as adults at meal times but eat whenever hungry between meals (25). Conservation of energy is accomplished by the use of housing and clothing with properties that minimize heat loss and maximize heat gain. Houses made of adobe, thick mortared stone, or sod bricks on the Andean and the Himalayan plateaus offer some insulating value and effectively store radiant heat gained during the almost universally sunny days. Houses of piled stone construction provide little buffering against cold but represent a lesser energy and resource investment for mobile pastoralist families in both locations. Clothing also serves as a primary barrier against cold, creating a warm, portable living environment. Clothing typically consists of multiple layers of insulating fiber (usually wool). The outer layer is often dark, tightly woven, and water-resistant and serves to maximize solar heat absorption and prevent convective heat loss. Covering for the head and face provides shielding and helps maintain a warmed, humidified microenvironment around the face. The limited availability of wood or other fuel for heating houses makes the strategy of fully clothed family members sleeping together an important means of conserving body temperature without increased expenditure of caloric energy or fuel resources (30). Specific cultural practices afford additional protection from environmental stresses around periods of vulnerability in the life cycle. For example, in both Peru and Tibet, the infant or young child sleeps with the mother in the early months of life, is nursed in the warmest location, remains swaddled even while indoors, and

50

Niermeyer et al.

is placed in the sunniest areas when outside. The infant is wrapped in multiple layers—diapering, leggings, an inner garment, and a sweater—and wears a knitted hat (31; S. Niermeyer, unpublished observations). In Tibet, infants are carried inside the traditional outer garment (chuba) where nursing can occur within protective layers of clothing. Quechua and Aymara women in the Andes carry their infants using a carrying cloth (manta) worn across the mother’s back and fully enclosing the infant, who is also encased in a blanket and swaddled by a cloth belt. The insulating value of this manta pouch is sufficient to raise relative humidity and temperature 12°C from the first layer of the pouch to the infant’s skin, but inspired PO2 is 8– 16 mmHg below ambient (32). The adoption of Western-style dress by mothers in the Lhasa valley appears associated with a higher incidence of cold injury in their infants, suggesting that the abandonment of long-standing clothing and carrying practices may be maladaptive (S. Niermeyer, unpublished observations). Another example of a cultural practice thats affords protection from the highaltitude environment is permanent or temporary outmigration, perhaps the ultimate behavioral solution to a biological problem. Such practices in the Andes were recorded by the seventeenth-century historian Antonio de la Calancha, who observed that pregnant women of Spanish origin would descend to give birth at lower altitudes and not return until the child was more than a year old (33). A similar practice occurs today among pregnant Han women in Tibet. They typically descend to their home districts at or near sea level and remain there or leave their infants with extended family until the infant is approximately 2 years of age before returning to high altitude (34). In summary, high-altitude residents engage in cultural practices that modify the effects of hypoxia on energy availability as well as the other attributes of the high-altitude environment. Historical as well as present-day practices of Himalayan and Andean residents rely on well-adapted indigenous plants and animals to produce calorically dense food. By seasonal adjustment in energy expenditure and food consumption within households, resources are distributed in ways that protect infants and children at times of shortage. House construction and clothing aid in the conservation of energy. In the Andes and Rocky Mountains, trade between regions, highly valued mineral resources, and, particularly in North America, recreational opportunities augment the resources available. Close proximity to different climatic zones provided by the vertical layering of highland environment in the relatively equatorial Andean region facilitates such resource exchange.

III. High-Altitude Adaptation Across the Life Cycle The stages of the human life cycle provide a convenient organizing framework for reviewing the physiological responses to hypoxia and demonstrating how these responses influence the ability of one generation to successfully reproduce the next (adaptation). In this section, we consider pregnancy and fetal life; infancy, childhood, and adolescence; adulthood and old age. Mortality risk is unequally distributed

The People

51

across these phases; it is greatest during fetal life, intermediate during infancy and childhood, and lowest in adolescents and adults (except for women during childbearing years) until the oldest ages. Thus, the early phases of the life cycle are of the greatest interest from an evolutionary point of view. At each phase, we compare high-altitude and sea-level residents or high-altitude residents with recently acclimatized newcomers. The underlying question is the extent to which lifelong high-altitude residents of the three world regions reviewed above differ in their physiological responses to hypoxia and whether such differences support the likelihood that some populations are better adapted than others. A. Pregnancy and Fetal Life

From an evolutionary perspective, pregnancy represents a critical overlap between the generations. While the health risk to the fetus and neonate is greater than to the mother, the evolutionary effect of a maternal death is magnified since it results in the loss of two individuals, mother and fetus, and curtails the father’s genetic contribution as well. Fetal Wastage/Pregnancy Loss

Fetal wastage is usually assessed by careful monitoring of completed fertility (total number of births per women aged 15–44 years), spontaneous abortions, or miscarriages. In practice, such information is difficult to obtain. Fertility has been reported to be less in the Andean region than at sea level as judged by smaller completed family size, a shortening of the reproductive span by later menarche and earlier menopause, and an increase in completed fertility in persons who migrated from high to low altitudes (35). Other studies in South America and Nepal do not support an altitude-associated reduction in fertility and, in fact, suggest higher completed fertility at high than at low altitudes (36–38). Delayed menarche and earlier menopause do not limit fertility; higher fertility is achieved in highland Peru and Chile by shorter intervals between births and increased frequency of conception during lactation. Births are likely to be underenumerated in the Andean and Himalayan regions where fewer than half the women give birth in hospitals (39). High neonatal (birth to 28 day) or infant (birth to 1 year) mortality can serve either to increase or decrease fertility by prompting a greater number of births to assure surviving offspring or by reducing the number of infants tallied in periodic censuses. Cultural practices affecting exposure to intercourse (e.g., proportion of the population living as celibate nuns) as well as the contribution and costs of children also influence childbearing patterns. In rural highland Peru, children generate more resources than they consume, making high completed fertility desirable (27). Endocrinological studies suggest alterations in reproductive function at high altitude. Andean highland women of reproductive age showed similar levels of luteinizing hormone but lower prolactin levels compared to low-altitude women (40). Urinary ovarian hormone levels were essentially the same, but serum progesterone was elevated and estrogens (estradiol and estriol) reduced during high- compared

52

Niermeyer et al.

to low-altitude pregnancy in Colorado and Peru (41,42). Measures of testicular function were similar, but testosterone levels were lower at 4340–4500 m than at 150 m in Peru. Sherpa males at high compared with low altitude have lower serum luteinizing hormone and a trend for lower follicle-stimulating hormone levels, suggesting less stimulation of the pituitary gland (43). Thus, there may be some level of impairment of reproductive function at high altitude, but it is not sufficient to limit or even reduce fertility. Intrauterine Growth and Pregnancy Duration

One of the best-documented effects of high altitude is a progressive reduction in birth weight. Birth weights decline an average of 100 g per 1000 m altitude gain in studies conducted over a 40-year period (Fig. 1). In addition to the lower mean birth weight, the percentage of low birth weight babies (⬍2500 g) is fourfold greater at high (⬎2700 m) than low altitude in a U.S. population-based study (44). The reduction in birth weight is due to direct effects of high altitude and not to interactive effects with other risk factors such as maternal age, parity, body size, or prenatal care (45). Similar birth weight reductions under other circumstances of reduced fetal-

Figure 1 Mean birth weights from previously published studies in North America, South America, and Tibet (labeled) with the upper and lower 90% confidence limits (dotted lines). (From Ref. 222.)

The People

53

placental oxygen supply implicate the hypoxia of high altitude as causative. The reduction in birth weight at high altitude has historical significance; the first recognition that fetal growth and length of gestation were separable influences on birth weight was made at high altitude (46). The primary cause of the reduction in birth weight is retardation of intrauterine growth rather than shortened gestation. Babies are both lighter and shorter for gestational age, conforming to a model of growth retardation throughout pregnancy. The greatest absolute reduction in fetal size occurs in the mid to late third trimester, as demonstrated by a progressive reduction in growth rate after 32 weeks at high compared with low altitudes in the Colorado population (47,48). Average gestational ages at high versus low altitude are generally similar (Table 2), but accurate gestational age information is difficult to obtain, particularly in developing countries where standardized reporting of birth weight and gestational age is available for only a small proportion of the population. The magnitude of fetal growth retardation appears to vary in relation to the duration of high-altitude residence, with the longest resident populations experiencing the least decline and the shortest resident groups demonstrating the most reduction in birth weight. Comparing well-matched samples collected by the same investigator at or near sea level and at greater than 3000 m, the reduction in birth weight is greatest in North Americans (⫺352 g, p ⬍ 0.001), intermediate in South Americans (⫺270 g in Peru, ⫺282 g in Bolivia, p ⬍ 0.001), and least in Tibetans (⫺72 g, p ⫽ NS) (50). Comparing women of different ancestry at the same altitude (3600 m), women from long-resident high-altitude populations give birth to heavier weight infants than women of low-altitude population ancestry. This is particularly true in Tibet, where birth weights averaged 294–650 g heavier in Tibetan than Han women (34,55), but also in Aymara women whose infants weighed 143 gm more than babies born to women of European or mestizo ancestry (53). The protection from altitudeassociated reductions in birth weight has also been observed for Nepalese Sherpa (56) but not Ladakhis (49). The Ladakhi women were smaller than the other samples surveyed, complicating the interpretation of the cause of the lower birth weights observed. Thus, several factors appear involved in protecting long-term, native highaltitude residents from intrauterine growth retardation (IUGR). Developmental factors likely contribute to the differences in birth weights between high-altitude newcomers and lifelong residents. Genetic factors are involved in the determination of birth weight; genetic factors account for as much as 70% of the variation in birth weight among offspring of monozygotic twins at sea level (57–59). Thus, the protection from altitude-associated IUGR afforded Tibetans compared to other high-altitude natives may result from the possession of different genetic variants. Nutritional, behavioral, and health-related characteristics during pregnancy are also likely important in each location. The particular maternal physiological characteristics that might be affected by genetic, developmental, or pregnancy-specific factors are described in the section below.

3035 3166 3235 3297

3410 3427 3178 3180 3299

1600 ⬍2130 ⬍2130 ⬍2130 338 600 400 150 150 1200

Weight (g)

39.7

39.0 39.8 39.2

37.0 40.0 39.5

Gest. age (wk)

b

a

Neonatal mortality rate ⫽ deaths within first 28 days/1000 live births. Comparison with low altitude, p ⬍ 0.05. c Infant mortality rate ⫽ deaths within first year/1000 live births.

Rocky Mountains Lichty, 1957 (46) McCullough, 1977 (47) Unger, 1988 (48) Jensen, 1997 (45) Andes Mazess 1965 (54) Beall 1981 (51) Haas, 1980 (53) Gonzales, 1993 (40) Carmen Torres, 1993 (52) Himalayas Zamudio, 1993 (50) Wiley, 1994 (49)

Altitude (m)

Low altitude

7

9

18.2 11.6 11.2

% preterm

3600 3600

3030 3860 3600 4340 4340

28.6a 10.6c

3100 ⬎2740 ⬎2740 ⬎2740

Altitude (m)

23.4a 11.9a 6.0a

Mortality rate

3236 2764

3140b 3165b 2982b 2835b

2655b 2962b 3058b 3056b

Weight (g)

Effect of High Altitude on Birth Weight, Gestational Age, and Neonatal or Infant Mortality

Area (Ref.)

Table 2

39.8 37.8

39.0 38.2* 38.6

39.0 39.5 39.0b

Gest. age (wk)

High altitude

7

12*

19.2 11.5 14.2

preterm %

144a

9.3c

52.8b

41.6a,b 18.5a,b 6.5a

Mortality rate

54 Niermeyer et al.

The People

55

Maternal Oxygen Transport Responses to Pregnancy

Because not all women at high altitude deliver growth-retarded babies, we have conducted a series of studies to test the hypothesis that altitude-associated IUGR is due to maternal oxygen transport insufficient to meet fetal-placental demands. Alternate possibilities are that fetal growth is impaired by limitations of placental diffusing capacity for oxygen, other nutrients, or fetal-specific factors. Early reports indicated that the placenta at high altitude was similar in absolute size and larger in relation to fetal size than at low altitude (41). Recent studies show that the highaltitude placenta is more vascularized and has a greater diffusing capacity than at low altitude (60,61). Thus, impaired placental oxygen diffusing capacity is unlikely to be the source of altitude-associated IUGR. During pregnancy, elevated levels of progesterone and estrogen in combination with increased metabolic rate raise peripheral (carotid body) and central nervous system ventilatory sensitivity to chemosensory stimuli and result in increased resting ventilation (62–64). At low altitude, the increase in ventilation does not appreciably raise arterial oxygen saturation, since it is already nearly maximal, but at high altitude arterial oxygen saturation rises with pregnancy (Table 3). Studies in Peru and Colorado demonstrated that the magnitude of rise in a woman’s ventilation, hypoxic ventilatory sensitivity, and arterial oxygen saturation during pregnancy related positively to the birth weight of her infant (72–74). The rise in maternal arterial oxygen saturation among Tibetans appears diminished relative to that seen in Colorado or Peruvian highlanders (Table 3). However, different samples of Tibetan women were compared in the pregnant versus the nonpregnant condition, whereas the same women were studied while pregnant and postpartum in Colorado and Peru. Hence it is unclear whether interindividual variation or differing effects of pregnancy are involved. These studies suggest that factors that raise maternal arterial oxygenation help preserve fetal growth at high altitude. Since the pregnancy-associated rise in arterial oxygen saturation and higher hemoglobin concentration of high-altitude residents result in levels of arterial oxygen content that are similar to low-altitude values, the question becomes what is responsible for the IUGR observed. One possibility is that the oxygen tension gradient between maternal and fetal blood is more important than maternal arterial oxygen content for deciding fetal-placental oxygen delivery. Maternal arterial PO2 and, to a lesser extent, uterine venous PO2 are reduced in pregnant ewes chronically exposed high altitude (75–77), and hence the maternalfetal oxygen diffusion gradient is likely reduced. Another possibility is that uteroplacental oxygen delivery is reduced as a result of decreased uterine blood flow. In studies conducted at low altitude, we demonstrated that approximately 1 L/min or 15% of the total cardiac output is directed toward the uteroplacental circulation near term (78). The rise in uteroplacental blood flow was accomplished, in approximately equal part, by a doubling of uterine artery diameter by mid-gestation and a rise in uterine artery flow velocity that was progressive throughout pregnancy. The increase in uterine artery diameter in pregnant Colorado women at 3100

nonpreg.

preg.

USA Low Altitude (63,67,70,71a) nonpreg.

3100 4300 11.4 ⫾ 0.3 9.5 ⫾ 0.4 244 ⫾ 5 — 3.3 ⫾ 0.1 — 322 ⫾ 28 23 ⫾ 8 4.3 ⫾ 0.3 0.4 27 ⫾ 1 31 ⫾ 1 92.2 ⫾ 0.2 82.9 ⫾ 1.2 13.8 ⫾ 0.2 14.0 ⫾ 0.4 17.3 ⫾ 0.2 15.9 ⫾ 0.4 69.9 ⫾ 1.9 — 42.9 ⫾ 1.3 — — 5.2 ⫾ 0.5 85 ⫾ 1 — 93 ⫾ 2 77 ⫾ 2 4.6 ⫾ 0.2 — 69 ⫾ 2 — 203 ⫾ 48

preg. 4300 12.0 ⫾ 0.7 — — 87 ⫾ 17 1.4 26 ⫾ 1 87.4 ⫾ 0.4 13.1 ⫾ 0.3 15.6 ⫾ 0.4 — — 5.8 ⫾ 0.1 — 79 ⫾ 2 — —

preg.

Peru High Altitude (69,70)

3658 10.1 ⫾ 0.5 255 ⫾ 11 5.2 ⫾ 0.2 45 ⫾ 8 1.0 ⫾ 0.2 31 ⫾ 1 89.0 ⫾ 0.5 14.9 ⫾ 0.2 18.1 ⫾ 0.3 — — — 76 ⫾ 2 76 ⫾ 2 — —

nonpreg

nonpreg.

3658 10.2 ⫾ 0.5 244 ⫾ 18 4.4 ⫾ 0.3 134 ⫾ 16 2.5 ⫾ 0.4 29 ⫾ 1 89.6 ⫾ 0.5 14.4 ⫾ 0.4 17.3 ⫾ 0.6 — — — 89 ⫾ 4 88 ⫾ 2 2.9 ⫾ 0.4 47 ⫾ 4

preg.

Han newcomers High Altitude (68a)

3658 3658 11.7 ⫾ 0.3 9.1 ⫾ 0.5 267 ⫾ 7 233 ⫾ 18 4.7 ⫾ 0.1 5.0 ⫾ 0.3 134 ⫾ 19 44 ⫾ 11 2.3 ⫾ 0.3 0.8 ⫾ 0.3 27 ⫾ 1 31 ⫾ 1 89.8 ⫾ 0.3 86.7 ⫾ 0.6 12.6 ⫾ 0.3 15.2 ⫾ 0.3 15.5 ⫾ 0.3 17.9 ⫾ 0.2 — — — — — — 85 ⫾ 2 80 ⫾ 2 81 ⫾ 2 81 ⫾ 3 5.5 ⫾ 0.7 — 56 ⫾ 3 —

preg.

Tibetans High Altitude (68,69a)

Mean ⫾ SEM nonpregnant and pregnant values were obtained in the same woman ⱖ3 months postpartum and the third trimester, respectively, in the United States and eru but from different women in Tibet. VE ⫽ Ventilation, 1 BTPS/min; VO2 ⫽ O2 consumption, ml STPD/min; HVR ⫽ hypoxic ventilatory response, shape parameter A; HVR/kg ⫽ HVR/kg body weight; PaCO2 ⫽ arterial or end-tidal PCO2, mm Hg; SaO2 ⫽ arterial O2 saturation, %; Hgb ⫽ g/100 mL blood; CaO2 ⫽ arterial O2 content, mL O2 /100 ml blood; blood vol ⫽ total blood volume, mL/kg; plasma volume, mL/kg; C.O. ⫽ cardiac output, L/min; MAP ⫽ mean arterial blood pressure, mmHg; HR ⫽ heart rate, bpm; UA/CI vel ⫽ ratio of uterine artery/common iliac artery blood flow velocity, cm/sec; UA vel ⫽ uterine artery blood flow velocity, cm/sec; UA flow ⫽ uterine artery blood flow, mL/min. a L.G. Moore, unpublished data.

3100 8.8 ⫾ 0.3 199 ⫾ 5 3.1 ⫾ 0.8 244 ⫾ 20 3.6 ⫾ 0.3 31 ⫾ 1 90.9 ⫾ 0.2 15.1 ⫾ 0.2 18.7 ⫾ 0.2 58.3 ⫾ 1.2 33.7 ⫾ 0.8 — 87 ⫾ 1 81 ⫾ 1 0.9 ⫾ 0.1 10 ⫾ 1 8⫾2

nonpreg.

USA High Altitude (65,66,70a)

Maternal Oxygen Transport During Pregnancy

Altitude, m ⬍2500 ⬍2500 VE 7.1 ⫾ 0.4 9.6 ⫾ 0.4 VO2 196 ⫾ 8 254 ⫾ 9 3.1 ⫾ 0.1 3.5 ⫾ 0.2 VO2 /kg HVR 124 ⫾ 13 237 ⫾ 26 HVR/kg 2.1 3.2 PaCO2 38 32 SaO2 94.6 ⫾ 0.5 95.4 ⫾ 0.4 Hgb 13.9 12.6 CaO2 17.5 14.5 Blood vol 66.6 79.7 Plasma vol 42.6 65.9 C.O. 5.3 6.5 MAP 84 85 HR 74 84 UA/CI vel 1.0 ⫾ 0.2 4.3 ⫾ 0.4 UA vel 9⫾2 61 ⫾ 3 UA flow 6 ⫾2 312 ⫾ 22

Table 3

56 Niermeyer et al.

The People

57

m was only about half as great as that seen at low altitude (1600 m), resulting in one-third lower uterine artery blood flow (Table 3). We also found that pelvic blood flow distribution to the uterine artery was diminished at high versus low altitude (70,79), In the same study, women who developed preeclampsia at high altitude had even less redistribution of pelvic blood flow to favor the uterine circulation and hence no increase in uterine artery flow velocity near term, unlike normal pregnant women. These differences were present before the onset of hypertension, suggesting that lowered uterine artery blood flow may be a cause rather than an effect of preeclampsia (79). The reduction in uterine artery blood flow at high altitude was consistent with the decrease in birth weight observed in these and previous experimental animal studies (Fig. 2). Because babies born to Tibetan women weighed more that those of acclimatized Han newcomers at 3658 m, we asked whether Tibetan preservation of fetal growth could be linked to increased arterial oxygen content or augmented uterine artery blood flow. Pregnancy increased maternal ventilation, but arterial oxygen saturation was unchanged in the Tibetan women. Whereas the increased arterial oxygen saturation and elevated hemoglobin concentration preserved arterial oxygen content at sea-level values in the high-altitude Colorado, Peru, and Han women, the pregnant Tibetan women had lower hemoglobin and lower arterial oxygen content values than the other groups (Table 3). The Tibetans’ lower hemoglobins were likely due

Figure 2 Birth weight, expressed as a percentage of sea-level or low-altitude values, falls as uterine, placental, or uteroplacental blood flow diminishes in experimental animal or human studies. Open circles are previously published values and closed circles are from normotensive or preeclamptic women studied at 3100 m. (Adapted from Ref. 70.)

58

Niermeyer et al.

to greater plasma volume expansion and may, in turn, have facilitated their directing a larger portion of their pelvic (common iliac) blood flow toward the uterine artery (68). This greater redistribution of lower extremity flow to the uterine artery was associated with heavier birth weights in the Tibetan versus the Han samples. Taken together, these data suggested that the Tibetans employed a strategy that maximized the increase in uterine blood flow, not arterial oxygen content (68). In the aggregate, these data suggest that the ways in which adequate oxygen delivery is maintained to the fetus vary among individuals and populations, with Tibetans emphasizing an increase in uterine blood flow and the other groups protecting arterial oxygen content. Since all the Andean and Tibetan women were born and raised at high altitude and there were no discernible differences between Colorado women born at high altitude versus those moving there as adults, genetic factors appear implicated in the sample differences observed. In addition, because the Tibetan women have oxygen transport characteristics close to sea-level values and give birth to infants with the least reduction in birth weight, those factors preserving oxygen delivery closest to sea-level values appear to be the best adapted. Neonatal, Infant, and Maternal Mortality

Birth weight is a major determinant of infant mortality. Since mortality increases at both the lower and upper extremes of the birth weight distribution, birth weight is considered to be the classic example of the evolutionary process of stabilizing selection. Given the lower birth weights generally present at high altitudes, the expectation would be that neonatal and, by extension, infant mortality would be increased. This is supported by older studies in Colorado in which mortality rates at high altitudes were higher than those observed at lower elevations (Table 2) (46,47). However, current data demonstrate that mortality rates at high altitudes in Colorado have declined to nationwide levels in association with a modest increase in birth weight, fall in percent preterm births, and likely improvements in the detection and management of complicated pregnancies (88,80). Today, as well as for the past 20 years, Peru and Bolivia have had the highest infant mortality rates in South America. In both countries, mortality rises with increasing elevation when all infants or only urban infants are compared (39). The Colorado data have the advantage of complete population enumeration and reasonably accurate data reporting. In South America, neonatal and infant deaths are almost certainly underreported since registration of a birth or death requires payment and only about one third are certified by a physician (39). No reliable infant mortality data, to our knowledge, are available from Tibet. When comparing infant mortality between regions or countries, it is important to recall that factors other than hypoxia may be involved, including political factors (since infant mortality has taken on broader meaning as a yardstick for social and economic development), infectious disease, and other environmental influences. Both growth retardation and preterm delivery affect birth weight and mortality risk. Within the range of values commonly observed, premature delivery has a much greater impact on neonatal survival than a reduction in birth weight alone (81).

The People

59

Accordingly, the increased neonatal mortality observed previously at high altitude in Colorado was due chiefly to deaths among babies who were both growth-retarded and premature (47). In a more recent large U.S. study, there was no difference in birth weight–specific mortality between low and high altitudes when fetal growth retardation and preterm delivery were taken into account (80). In South America, the lack of complete population enumeration and gestational age information in some studies prevents determination of whether birth weight–specific mortality is altered at high versus low altitude. Beall reports a 170 g lower optimal birth weight (i.e., associated with the lowest mortality rate) in a 3860 m compared with 600 m sample from Peru (51). However, since no gestational age information was available, it is possible that the lower optimal birth weight was due to a greater contribution of growth retardation than prematurity to mortality risk at high versus low altitude. Also, the low-altitude optimal birth weight values are considerably higher than in other studies (81,82). A recent study from Ladakh at 3600 m yielded an average birth weight of 2674 g, well below Tibetan and Andean values at similar elevations, which carried with it an even greater birth weight–specific mortality risk (49). Approximately 6% were Tibetans; the remainder were of Indian and other genetic background. It is likely that prematurity, parentage, and other factors such as small maternal body size elevated infant mortality. Thus, in short, ‘‘small’’ is not better at high (or any) altitude. The IUGR and greater frequency of preterm deliveries in some studies at high altitude may be related to an increased frequency of preeclampsia, placental abruptions, and other complications of pregnancy. Preeclampsia is defined clinically as elevation in blood pressure (⬎140/90 mmHg, a systolic rise ⬎30 mmHg or a diastolic rise ⬎15 mmHg) after week 20 accompanied by significant proteinuria (⬎0.3 g/L in a 24-hour collection) in a woman who is normotensive when nonpregnant. Abnormalities of liver function, coagulation, and the central nervous system are sometimes observed as well. Preeclampsia is the leading cause of maternal and fetal mortality in the industrialized world and the third most frequent cause of very low birth weight in the United States (83–85). In three separate studies, we have observed a three- to fourfold increase in the incidence of preeclampsia at high compared to low altitude in Colorado (45,73,86). Data from South America are equivocal as to whether the incidence of preeclampsia is increased (40,88), and no studies have been conducted in the Himalayan region. Interestingly, we found higher blood pressures in Han than in Tibetan pregnant women living at 3658 m in Lhasa (L. G. Moore, unpublished observations). A review of all deliveries in La Oroya, Peru (3750 m), over a 15-year period indicated that placental abruptions were three times more common than at sea level and occurred in 6.8% of women over 40 years and 3.4% with parity greater than 4 (89). An increased incidence of maternal complications during pregnancy may prompt preterm deliveries and raise maternal as well as infant mortality. Maternal mortality in Peru and Bolivia is more than twice the South American average (39) and rises from 13.2 maternal deaths per 10,000 live births at the coast, to 21.5 in the 2000 to 3000 m region, and to 43.1 at elevations above 3000 m in Peru (40). Maternal mortality is not increased at high altitude in

60

Niermeyer et al.

the United States, where there is ready access to medical care. Data from the Himalayan region are lacking. Impaired placentation and uteroplacental ischemia may be a common pathway whereby pregnancy complications are increased and intrauterine growth retarded at high altitude. Uteroplacental ischemia has been associated with IUGR and/or preeclampsia at low altitude (90,91). A growing body of evidence suggests that there is impaired trophoblast invasion of maternal spiral arteries in preeclamptic pregnancies. Trophoblasts are the specialized epithelial cells of the placenta, which ultimately comprise the physical connection between the embryo (fetus) and uterus. Impaired trophoblast invasion results in shallower placentation and maternal spiral arteries that retain their muscular portions in the myometrial (nondecidual) segment, preserve their contractile sensitivity to local and circulating vasopressors and, as a result, reduced uterine blood flow (reviewed in Ref. 92). Trophoblast differentiation and invasion may be oxygen regulated; when grown in the presence of physiological oxygen tensions, trophoblast stem cells differentiate and rapidly invade extracellular matrix. The same cells grown under hypoxic conditions failed to invade extracellular matrix, partially due to inhibition of their transition to an invasive phenotype, marked by switching of integrin cell-extracellular matrix receptors (93). In summary, fertility is not reduced at high altitude but intrauterine growth is retarded, birth weight is lowered, and neonatal and infant survival are impaired except under conditions of accessible medical care. The cause of the IUGR is likely to be insufficient oxygen delivery to meet fetal-placental demands. This is supported by our findings of consistent associations between infant birth weight and maternal ventilation, arterial oxygenation, and uterine blood flow at high altitude as well as by others’ findings that placental diffusing capacity is not impaired relative to lowaltitude values. The degree of IUGR is least in the longest resident high-altitude populations. Whether there is a corresponding reduction in birth weight–specific mortality is unclear and will likely remain so, given the many factors affecting mortality rates. The parallels between the magnitude of IUGR and the extent of maternal oxygen transport responses to pregnancy are suggestive that the adaptive strategies and degrees of success differ between and within regions. That is, Tibetans appear distinguished by lesser reductions in birth weight, lower hemoglobin concentrations, and greater blood flow redistribution to the uterine circulation. The higher birth weights in native Aymara than acclimatized newcomer women may be accompanied by differences in oxygen transport, but these remain, to our knowledge, unstudied. Of interest is that uterine blood flow as well as hemoglobin characteristics of the Tibetan women resemble those of healthy low-altitude Colorado residents, whereas the Han women are more like Colorado high-altitude women with preeclampsia. B. Infant, Childhood, and Adolescent Development

Remarkable changes in oxygen transport take place at birth. At high altitude, these are complicated by the lower ambient oxygen tensions as well as the IUGR and other conditions stemming from prenatal life. Effects of high altitude continue to

The People

61

influence growth and development, exercise performance, and occurrence of illness during infancy and childhood. Differences among regional populations and between groups residing at the same altitude will be considered below to determine the extent to which interpopulation differences exist in infant, childhood, and adolescent development. Oxygen Transport Characteristics

In the transition following birth, the lungs change from fluid- to air-filled, pulmonary blood flow increases dramatically, and vascular shunts reverse in direction and close. Oxygen plays a critical role in the postnatal transition because of its function as a pulmonary vasodilator. At sea level, arterial oxygen saturation is modestly lower during the first week of life than the 94–98% values present during the neonatal period and infancy (94). At 1610 m in Denver, arterial oxygen saturation in healthy, term infants was reduced relative to sea level values, averaging 92–94% from 24–48 hours to age 3 months (95). At 3100 m in Leadville, Colorado, arterial oxygen saturation was lower still, especially during the first week of life (Fig. 3) (96). Unlike the pattern at low altitude, arterial oxygen saturation fell during the first week and then rose gradually to attain near-birth values at 2 and 4 months of age. The fall in arterial oxygen saturation at 1 week was consistent with clinical observations that babies who develop signs of hypoxemia (e.g., cyanosis, irritability, poor feeding, and failure to gain weight) often become symptomatic around 1 week of age. At all ages, values were higher during wakefulness than during active or quiet sleep and intermediate during feeding.

Figure 3 Arterial oxygen saturation in infants during quiet sleep falls with increasing altitude. (Adapted from Refs. 34, 95, and 96.)

62

Niermeyer et al.

At 3658 m in Lhasa, arterial oxygen saturation was higher in Tibetan than in Han infants from birth through 4 months of age (Fig. 3) (34). The neonates were similar in gestational age and Apgar scores; however, the Han had lower birth weights and higher hemoglobin and hematocrit at birth than the Tibetans. In both groups, arterial oxygen saturation was highest in the first 2 days after birth and then declined. Whereas arterial oxygen saturation in the Tibetan infants stabilized at 4 months to values within the range of Leadville infants, arterial oxygen saturation in the Han declined progressively (Fig. 3). Arterial oxygen saturation in 2- to 5-month-old Quechua infants at 3750 m in Peru averaged 88 ⫾ 3% (97), similar to Tibetan values. At 4540 m in Peru, directly measured arterial oxygen saturation ranged from 57 to 75% in newborn infants 30 minutes to 72 hours old (98) and remained in the range of 74–80% throughout infancy (99). The cause of the dramatic differences in arterial oxygen saturation between Tibetan and Han infants residing at the same elevation but differing in their highaltitude ancestry likely reflects differences in ventilation and/or pulmonary blood flow, but these have not been measured. Increased ventilation would serve to decrease the oxygen pressure gradient from the atmosphere to the alveoli and to raise arterial oxygen tensions. The accompanying fall in arterial carbon dioxide tensions may also increase pH and left shift the oxyhemoglobin dissociation curve to raise saturation at a given oxygen pressure, but hemoglobin-oxygen dissociation curve position is unknown. Ventilatory patterns during the neonatal period and infancy change in ways that differ at high compared with low altitude. At sea level, periodic breathing occurs during sleep in the overwhelming majority (78%) of full-term neonates (100). Its prevalence declines perhaps as early as 1 month and definitely by 5–6 months (101) in response to developmental changes in central and peripheral chemoreceptors (101–104). Early studies in healthy infants born in Denver (1610 m) suggested that periodic breathing was ‘‘much more frequent and less transitory . . . than has been previously described,’’ but the prevalence reported (65%) was similar to sea level (105,106). At higher altitudes, the prevalence clearly increases; 100% of Leadville neonates demonstrated a repetitive pattern of 4–6 breaths over 6–7 seconds followed by a pause of equal length (105). Current investigations in Leadville confirm the occurrence of periodic breathing in all infants older than 48 hours and its association with cyclic declines in arterial oxygen saturation (S. Niermeyer, unpublished observations). Perhaps periodic breathing is responsible for the decrease in arterial oxygen saturation at 1 week of age at high altitude and the consistently lower values observed during sleep compared with wakefulness throughout the first 4 months of life. A rapid postnatal fall in pulmonary artery pressure (PPA) occurs in the first days of life at sea level. This is important for achieving first functional and then anatomic closure of the atrial and ductal shunts (107). At 3100 m in Leadville, Colorado, PPA indices obtained by echocardiography in healthy, term infants were normal to moderately elevated during the first week of life and fully normal at 2–

The People

63

4 months (96). However, all neonates received supplemental oxygen during the first 24–48 hours. At very high altitudes without supplemental oxygen, the postnatal fall in PPA is prolonged or fails to occur. Newborns at 4540 m in Peru with an alveolar PO2 of approximately 50 mmHg showed persistence of near-systemic PPA for several days following birth (Fig. 4). Administration of 100% oxygen to three infants at 72 hours resulted in a dramatic fall in PPA to near sea level values. Elevated PPA and pulmonary vascular resistance were confirmed by right heart cardiac catheterization in infants and children under 5 years of age at 4330 m and 4540 m in Peru (99). Further support for delayed or, in some cases, absent postnatal regression of the fetal pulmonary vascular pattern comes from histological observations in South American children dying at high altitudes. These infants and children demonstrated peripheral extension and hypertrophy of the muscular layer of the pulmonary arteries and aretioles (108). Elevated right-sided pressure contributes to an increased prevalence of patent ductus arteriosus (109,110). An increased prevalence of atrial septal defect and patent ductus arteriosus has also been observed in Han and Tibetan infants in Qinghai Province of China (the northern portion of the Tibetan Plateau), rising from near zero at sea level to more than 5% at 4500 m (111). A syndrome of subacute infantile mountain sickness has been described in Lhasa (3658 m) in 15 infants or children who died between 3 and 16 months of age with signs of pulmonary hypertension and right heart failure (112). All were male; 14 were Han and one was Tibetan. All but two were born at low altitude and brought to Lhasa at an average of 2 months before the onset of disease symptoms. Clinical signs and symptoms included dyspnea, cough, cyanosis, sleeplessness, and

Figure 4 The fall in pulmonary arterial pressure (Ppa) in neonates at sea level is greater than at 4540 m in Peru. Supplemental oxygen restores values in high-altitude neonates to sea-level values. (Adapted from Ref. 98.)

64

Niermeyer et al.

irritability, facial edema, hepatomegaly, and oliguria. Histological findings were medial hypertrophy of small pulmonary arteries, muscularization of pulmonary arterioles, and severe right ventricular hypertrophy and dilation. A control group of native Tibetan infants and children who died of noncardiopulmonary causes showed normal thin-walled pulmonary arteries and arterioles after 4 months of age. A similar clinical syndrome was previously reported in five infants and six older children in Leadville, Colorado (3100 m), three of whom underwent cardiac catheterization and demonstrated pulmonary hypertension and one of whom showed medial hypertrophy and intimal thickening at autopsy (113). In summary, the newborn at high altitude experiences a slower transition from fetal to mature patterns of cardiopulmonary function. Arterial oxygen saturation is lower than at sea level, particularly in samples derived from populations that have recently migrated to high altitude. Periodic breathing appears more common and may contribute to lower arterial oxygen saturation during sleep. The postnatal fall in PPA is markedly slower at very high altitudes but only modestly delayed at more intermediate altitudes where supplemental oxygen is used routinely. Whereas fetal cardiovascular patterns persist into childhood in South America and Han migrant populations in Tibet, histological data in native Tibetan infants suggest a more rapid involution of fetal patterns. A possible reason for this difference between Tibetan and Han children is their better-sustained arterial oxygen saturation during early infancy. Growth and Nutrition

Growth at high altitude is a product of genetic and/or developmental factors acting in concert with nutrition, levels of habitual activity, and other socioeconomic and environmental characteristics. The relative contributions of these factors to growth has been actively investigated, particularly in the Andean region. Generally, the decreased growth in utero is sustained postnatally at high altitude. In Andean samples, a consistent reduction in length and weight from birth to 2 years has been observed at high compared with low altitude (114). A comparison of highland and coastal children in Ecuador found that shorter stature in the highlands was due to diminished linear growth velocity within the first 6 months (115). Delayed growth in Andean highlanders continues during childhood and adolescence, resulting in a 1- to 2-year lag in height and a less pronounced adolescent growth spurt. Adolescents grow for about 2 years longer, or until 22 years of age, but adult stature remains shorter than at low altitude (116–121). The growth failure at high altitude in South America is consistent with that seen among impoverished populations worldwide (114), suggesting the involvement of nutritional factors. This is supported by observations that upper socioeconomic status (SES) children in Nun˜oa, Peru, were taller and heavier than lower SES children (28). Over the past two decades, the age at which peak growth velocity was attained and growth completed declined in above-average SES adolescents but re-

The People

65

mained unchanged in lower SES adolescents in the southern Peruvian Andes (122). Comparisons in Bolivia found no difference in anthropometric characteristics between highland and lowland boys when the comparisons were controlled for SES and nutritional status. Boys from a low SES background at high and low altitude were marginally undernourished and showed a growth delay of approximately 2 years (123). The pattern of growth retardation in the Himalayas also suggests an influence of nutritional or other socioeconomic factors (25). There do not appear to be any appreciable differences in growth retardation among Tibetan, Han, and Hui male adolescents at 3200 m and 4300 m in Qinghai Province, the northeastern section of the Tibetan Plateau, but in all groups the average body weights were at or below the National Center for Health Statistics (NCHS) 5th percentile and average height was at or slightly above the 5th percentile (124). Rural Tibetans fell below the NCHS standards in height and weight by 4–6 months in both sexes (25). In the urban Lhasa valley, weight for height was within the normal range, but weight and height for age were at the lower range of the NCHS standards. Weight for height in Lhasa was noted to drop below NCHS standards at two time points: one at around 6 months, perhaps reflecting the increased importance of supplemental foods at this age, and a second time point at 1–2 years, when children begin to walk and to depend more on adult foods (25). High-altitude adolescent Bod girls of Ladakh, Jammu, and Kashmir, India, were significantly lighter and shorter than their lowaltitude counterparts (125). Smaller body sizes were not found in boys of Tibetan ancestry residing at high compared with low altitude in Nepal, but high-altitude girls over the age of 12 were smaller than low-altitude girls (126). Both high- and lowaltitude groups evidenced growth retardation. As mentioned above, Han infants often remain with extended family in lowland China until about 2 years of age, at which time they are brought back to Tibet. Newly arrived Han children showed poor appetite, diarrhea, and intestinal malabsorption, all of which can adversely affect growth (25). The Han children had markedly lower body weights by age 3 than the native Tibetan children, both rural and urban, despite the generally favored economic circumstances of the Han compared to the Tibetan population. Enlarged chest dimensions and accelerated chest development, despite generally smaller body size, has been reported in numerous studies at high altitude. Chest shape appears influenced by altitude in both Andean and Himalayan populations such that chest depth increases more than chest width. The contribution of genetic versus developmental factors has received a great deal of study. Since persons of Quechua ancestry born and raised at low altitude also have enlarged chest dimensions (127,128), larger chest dimensions may be a fixed genetic trait in the Quechua. Chest dimensions have significant heritability (i.e., the extent to which the expression of a trait is due to inherited characteristics in a particular environment) among the Quechua and Aymara, and heritability was greater at high altitude (129). This has been interpreted as being due to natural selection acting to increase the represen-

66

Niermeyer et al.

tation of genes increasing chest dimensions at high altitudes (130). Phenotypic plasticity may also be involved since genes coding for accelerated chest development might be preferentially expressed in the high-altitude environment. It is important to recall that lung volumes, not chest dimensions, have functional significance; greater lung volumes are associated with increased surface area for gas exchange and decreased alveolar-arterial oxygen diffusion differences, which help maintain arterial oxygen saturation during exercise at high altitude. While an equivalence is often assumed, chest dimensions are not the same as lung volumes since the chest contains structures other than the lung and larger lung volumes may be accommodated by a lower diaphragm, not a larger rib cage. Despite the large number of studies of chest dimensions, there is a paucity of complete lung volume measurements at high altitude. However, existing data clearly demonstrate that lung volumes are increased worldwide at high compared with low altitude, whether the high-altitude groups are compared with low-altitude U.S. standards or with samples drawn from newcomers living at the same altitude (Fig. 5). Residual volume increases the most, but vital capacity and total lung capacity are also enlarged at high altitude (Fig. 5). Adolescents of Aymara and mestizo ancestry (a mixture of Quechua, Aymara, and Spanish) have modestly (⬃4%) larger vital capacities when adjusted for differences in body size than similarly aged adolescents of European ancestry born and raised at high altitude (131). Debate has centered on whether lung volumes (and chest dimensions) are equally enlarged in Himalayan and Andean populations. As demonstrated in Figure

Figure 5 Total lung volume (RV ⫹ VC), residual volume (RV), and vital capacity (VC) (all 1, BTPS) in Colorado residents of 3100 m (223,224), Aymara and Quechua South American residents of 3600–4540 m (128,225), and Tibetan residents of 3658 m (226) are greater than sea-level predicted values.

The People

67

5, Tibetan high-altitude residents clearly have enlarged lung volumes. Other studies have shown that lung volumes are increased in young Bod girls at high altitude in Ladakh, Jammu, and Kashmir, India, compared to their lowland counterparts (125) and in boys between 5 and 18 years at high altitude in Himachal Pradesh (132). Rigorous determination as to whether the magnitude of increase varies among the Andean, Himalayan, and Rocky Mountain regions is hampered by substantial interpopulational variation in body size and disagreement regarding the best reference population. The information summarized in Figure 5 demonstrates that if differences are present, they are of modest proportion. Larger lung volumes result from exposure to high altitude during growth and development, as they cannot be acquired in adult animals spending an equivalent period of time at high altitude (133). The increase in each lung volume is present by early adolescence (age 11), with residual volume enlarging progressively in 11to 19-year-old La Paz adolescents (128). A recent, comprehensive review of environmental, developmental, and genetic influences on lung volumes among Bolivian high-altitude residents of varying population ancestry, altitude of birth, and duration of high-altitude residence found that growth and development at high altitude accounts for approximately 25% of the increase in vital capacity and residual volume among males but, interestingly, not among females (134). Genetic factors accounted for an additional 25% of the variability. Occupational characteristics associated with rural high-altitude residence further raised vital capacity but not residual volume. Thus, Frisancho and coworkers concluded that environmental factors (occupation, body composition) exerted greater influence on increasing vital capacity than developmental characteristics, whereas both developmental and genetic factors raised residual volume, at least in males (134). Taken together, available studies indicate that childhood growth at high altitude is characterized by enlarged lung dimensions—which develop universally during infancy, childhood, and adolescence—and by consistent growth retardation compared to sea-level standards. Nutritional stress plays an important role, especially in rural populations. Malnutrition, energy requirements, intercurrent illness, and genetic factors may variably affect growth of regional populations. However, even when nutritional factors are optimized, some degree of growth retardation appears present at high altitude. Exercise Performance

Exercise performance in children at high altitude demonstrates the influences of hypoxia as well as nutritional status, developmental, and genetic factors. Greksa et al. (135) studied 11- to 12-year-old healthy, well-nourished Aymara boys in La Paz. VO2max was slightly lower than that of upper SES European boys in La Paz, but both groups had about a 10% reduction in VO2max when compared with their lowaltitude counterparts. Obert et al. (136) found that VO2max per kg body weight in highland boys averaged 11% lower than lowland boys. At both high and low altitude, the VO2max of boys from a high SES, mestizo background was greater than that

68

Niermeyer et al.

of boys from a low SES, predominantly Quechua and Aymara background. The high SES group was significantly taller, heavier, and fatter, but when VO2max was normalized per kg lean body mass, differences disappeared between the high and low SES groups at a given altitude but remained present between altitudes within a given SES group. Maximal power during a force-velocity test and mean power during a 30-second Wingate test were higher in upper SES than lower SES boys, probably reflecting the presence of mild malnutrition in the latter. The authors concluded that a marginal state of malnutrition did not affect VO2max but led to lower power in prepubertal males at both high and low altitude (136). A similarly designed study of prepubertal Bolivian girls supported these conclusions (137). Girls of low SES had lower VO2max and anaerobic power than high SES girls. Anaerobic power but not VO2max remained lower after normalization by body weight in the low SES girls. Frisancho and coworkers demonstrated that persons born and raised at high altitude (3750 m) in La Paz had greater aerobic capacity than subjects who acclimatized to high altitude during adulthood (138). Younger age of arrival related to greater maximal exercise capacity, particularly when arrival was before age 10. Overall, developmental influences could account for 25% of the variability in maximal exercise capacity. Genetic factors were also influential and could account for an additional 25% of the variability in aerobic capacity among high-altitude residents. In summary, high-altitude exposure reduces maximal exercise capacity in children in a fashion that is relatively independent of nutritional and other SES influences. However, nutrition and other SES-related characteristics are more important than hypoxia in accounting for altitude-associated reductions in maximal and mean anaerobic power. Being born and raised at high altitude as well as genetic factors serve to restore maximal aerobic capacity toward sea-level values. Illnesses

In addition to the effects of hypoxia and nutrition, the presence of childhood diseases can exert important influences on exercise performance and growth. There are several ways in which the environmental conditions associated with high altitude interact with childhood illnesses. Some health problems in children are uniquely related to high altitude. One such problem is reascent high-altitude pulmonary edema (HAPE), which occurs, albeit rarely, in children returning to high altitude after temporary descent to low altitude (139). Reascent HAPE is often preceded by upper respiratory infection (140). Repeated episodes of HAPE can occur in the same child and tend to run in families. Cardiac catheterization in 7 Leadville, Colorado, children (3100 m) after recovery from HAPE showed greater PPA response to hypoxia than in control children and right ventricular hypertrophy by ECG criteria (140). Reascent HAPE has been reported in 4- to 19-year-old residents of Bogota´, Columbia, at 2640 m (141). Data from the Tibetan Plateau suggest an incidence of 1.5% in Han children com-

The People

69

pared to approximately 0.2% in Tibetans (142). The protection afforded the Tibetan children may be due to diminished pulmonary vasoconstrictor response, better oxygenation, less frequent descent to lower altitudes, and/or other factors reducing susceptibility to altitude-related illness. Other health problems in children are influenced by cultural responses to the high-altitude environment. Rickets reach a peak incidence of 21% in the 18- to 35month-old group in the Lhasa valley. Though there are ample hours of sunshine, the cultural responses to the climate (multiple layers of clothing and thick-walled houses with small windows) limit sunlight exposure. Lower prevalence in young children likely reflects the protective effects of breast milk, and older children play outside even during winter. Other cultural factors such as exposure to smoke from cooking fires and crowded living conditions are likely to promote respiratory diseases and exaggerate hypoxemia. Upper respiratory tract infections and pneumonia are more prominent in the colder months, and diarrheal disease is more prevalent in warmer periods (25). Malnutrition and lack of access to health care make diarrheal and respiratory illnesses major contributors to child mortality in high-altitude communities (143). Finally, some child health ‘‘problems’’ may in fact result from inappropriately applied definitions, despite efforts to altitude-adjust the criteria. The World Health Organization’s standard for evaluation of childhood anemia adds 0.57 g of hemoglobin for every 1000 m to the sea-level lower limit of normal (13.03 g/100 mL whole blood) (25). Using these criteria, 49% of Tibetan and 25% of Han children are anemic. It is likely, however, that these criteria overestimate anemia in the Tibetan group. Tibetan hemoglobins are normally distributed in most age groups with mean values being very close to the corresponding sea-level norm. Using sea-level criteria, the overall prevalence of anemia becomes 12% in Tibetan children from 6 months to 7 years of age. This information supports findings in adults and suggests that the Tibetan population does not respond to altitude with as great an increase in hemoglobin as do the Han (25). Consequently, it may be inappropriate to apply the same disease criteria to children from different populations even though they reside at the same altitude. In summary, the effects of hypoxia on infant, childhood, and adolescent development vary among and within high-altitude regions. Common to all regions is lower arterial oxygen saturation, retention of fetal cardiopulmonary characteristics, delayed growth, enlarged lung volume, and diminished exercise capacity. Inadequate nutritional resources exaggerate growth retardation. Tibetan infants have higher arterial oxygen saturations, less muscularized pulmonary arterioles, and more rapid involution of fetal patterns than Han living at the same altitude. Growth retardation may be less marked in Tibetan than Andean highlanders, but definitive interpretation requires information regarding energy expenditure and substrate utilization. Existing data are insufficient to determine whether differences exist between Tibetan and Andean infants in arterial oxygen saturation, cardiopulmonary characteristics, and exercise capacity. Hypoxia prompts diseases such as reascent HAPE and exacerbates

70

Niermeyer et al.

other diseases such as acute respiratory infection and diarrheal diseases, particularly when combined with the aridity, smoke, dust, food shortages, and limited access to health care frequently encountered at high altitude. C. Adulthood and Old Age

During adulthood, members of one generation are confronted with the need to function as the primary producers of food and other resources required for their own, their children’s, and perhaps their parents’ and even grandparents’ survival. This requires work. Hence, exercise performance is often used to evaluate adult highaltitude adaptation. Work capacity of high-altitude residents is diminished relative to low-altitude values. In this section, we examine exercise performance in acclimatized newcomer and lifelong European, Andean, and Himalayan high-altitude residents and compare the parameters of oxygen transport influencing exercise performance in these groups. We conclude with a consideration of chronic mountain sickness and other diseases that can substantially impair an adult’s ability to work and survive at high altitude. Exercise Performance

Maximal exercise capacity is an index of the integrated functioning of the oxygen transport system. It is reduced at high altitude by approximately 11% per 1000 m altitude gain (144). As a result, exercise capacity is lower upon ascent and after acclimatization to high altitude than it is at low altitude (Fig. 6). The reduction in work capacity is accompanied by reduced cardiac output and decreased brain but not muscle oxygen delivery. Despite extensive study, the cause(s) of the reduction in exercise capacity remain elusive (see Chapter 21). Several lines of evidence suggest that maximal exercise capacity is greater in high-altitude natives than in acclimatized newcomers (Fig. 6). Averaging values from published studies, exercise capacity was 8% higher in lifelong natives of European ancestry, 16% higher in native Andean residents, and 12% higher in Himalayan natives than in acclimatized newcomers. The differences between natives and newcomers were most consistent when the two groups were well matched for physical training, age, body size, and lean body mass (145,146). Comparing the change in exercise capacity with ascent versus descent also indicates that native high-altitude residents are less affected than acclimatized newcomers; VO2max decreased 15% with ascent in sea-level residents whereas descent increased VO2max only half as much (8%) in high-altitude natives (147,148). Variability among high-altitude residents in maximal exercise capacity has been related to developmental, genetic, as well as occupational (e.g., training) characteristics (138), which collectively tend to increase exercise capacity. Based on current data, maximal exercise capacity at high altitude appears to be nearer sea-level values in Andean than in European highaltitude natives, with values being intermediate in Himalayan highlanders (Fig. 6). Relatively few studies have been conducted in Himalayan high-altitude natives, and existing data are variable.

The People

71

Figure 6 In studies meeting the criterion for maximal oxygen consumption (VO2max, mL BTPS/min) of a less than 150 mL increase in VO2 with an at least 25 watt increase in workload, exercise capacity was reduced in acclimatized newcomers as well as lifelong highaltitude residents when compared to values from a large series of normal sea level residents (227). Lifelong high-altitude residence appears to confer some recovery in VO2max. Relative to acclimatized newcomer values, exercise capacity was increased 8% in lifelong European high-altitude residents, 16% in Andean residents, and 12% in Himalayan natives. Average maximal heart rates (HRmax , beats/min) are also shown. (Data from lifelong European highaltitude residents of 3100–3830 m from Refs. 153, 201, and 228. Data on Andean residents of 3400–4540 m are taken from Refs. 146, 147, 228–232. Data from Himalayan residents of 3658–4700 m were reported in Refs. 145, 149, and 150.)

Exercise efficiency or the work that can be performed at a given level of oxygen consumption is increased in some but not other studies of Andean and Himalayan high-altitude natives compared with acclimatized newcomers. The amount of work accomplished at a given VO2 was nearly 30% greater in native Quechua highlanders than acclimatized lowlanders (148). Increased exercise efficiency has also been seen among higher-altitude Tibetans in some but not all studies (149– 151). Characteristics of Oxygen Transport

A series of recent studies has addressed the mechanisms by which oxygen transport is restored toward sea-level values in high-altitude natives. The components of the

72

Niermeyer et al.

oxygen transport system are reviewed below to determine whether such mechanisms differ among regional populations and can explain the greater exercise capacity and/ or efficiency of high-altitude natives compared with acclimatized newcomers. Ventilatory acclimatization, commonly defined as an hypoxia-induced rise in ventilation that overcomes the inhibitory effects of hypocapnia, occurs over days among newcomers at high altitude. It is commonly measured as a fall in arterial PCO2 at a given PO2, indicating a rise in effective ventilation (i.e., alveolar ventilation per unit carbon dioxide production) (Fig. 7). Older studies of lifelong Andean (152), Colorado (153,154), and Himalayan (155–157) high-altitude residents found lower effective alveolar ventilation in natives than newcomers. Because infants had similar ventilatory responses at low and high altitude (103) and duration of highaltitude residence correlated with the magnitude of fall in arterial PCO2, it appeared that ventilatory acclimatization was lost after decades of high-altitude residence. However, most of the earlier Andean and Himalayan studies were based on small samples of uncertain parentage. The newcomer groups were also small and, in some cases, had been exposed to a range of altitudes or severity of hypoxic stimuli. More recent studies have compared larger samples of lifelong high-altitude residents with acclimatized newcomers. Some (158–160) but not all (157) demonstrate levels of effective alveolar ventilation in lifelong high-altitude natives equivalent to those of acclimatized newcomers (Fig. 7). In general, a higher proportion of Himalayan studies lie closer to the newcomer ‘‘after-acclimatization’’ curve than did studies of lifelong Andean residents, suggesting that Himalayans have higher effective alveolar

Figure 7 End-tidal PCO2 (PETCO2) falls with decreasing end-tidal PO2 (PETO2) at high altitude in acclimatized newcomers (after acclimatization line) and lifelong Andean and Himalayan high-altitude residents (From Ref. 158.)

The People

73

ventilation than Andeans (158). This conclusion is supported by direct comparisons by the same investigators of large numbers of Andeans and Tibetans living at the same altitude (161). The rise in effective alveolar ventilation is accompanied by an increase in ventilatory sensitivity to hypoxia (162). Conversely, the loss of ventilatory acclimatization has been attributed to a loss of hypoxic ventilatory responsiveness (152,154,163,164). However, it is important to recall that hypoxic ventilatory response is but one factor influencing resting ventilation. For example, Curran et al. (165) showed that Tibetans native to altitudes above 4400 m had levels of resting ventilation similar to Tibetans born and raised at 3658 m, but the higher-altitude Tibetans had lower levels of hypoxic ventilatory response. Likewise, lifelong residents of Leadville, Colorado, with blunted ventilatory drives maintain levels of ventilation equivalent to persons with higher ventilatory responses to hypoxia (154). Consistent with the maintenance of similar levels of effective alveolar ventilation, we found that lifelong Tibetan residents of 3658 m have hypoxic ventilatory responses at least as great as Han acclimatized newcomers and clearly greater than Han who migrated to high altitude as children (158). This agrees with previous reports in which higher or similar levels of hypoxic ventilatory sensitivity were found in Sherpas or Tibetans when compared with acclimatized newcomers (159,160). Comparing all published values, we concluded that Tibetan residents of 3658 m had higher levels of hypoxic ventilatory responsiveness than Andean residents of similar altitudes (158). Recent comparisons in large numbers of Tibetan and Aymara highlanders at the same altitude show that hypoxic ventilatory drives were, on the average, higher and more variable in the Tibetans (161). The higher hypoxic ventilatory drives in Tibetans than Andeans are likely due to innate, possibly genetic factors, although developmental alterations resulting from differences in intrauterine or neonatal oxygenation have not been excluded. Twin studies at low altitude demonstrate that a significant portion of the variation in hypoxic ventilatory response is due to genetic factors (166,167). At high altitude, the studies of Beall, Blangero, and coworkers demonstrate significant heritability for hypoxic ventilatory drives and, interestingly, a higher heritability in Tibetans (46%) than Andeans (12%) (168). The interpretation of heritability is complex, and lower heritabilities do not necessarily mean lesser degrees of genetic involvement. For example, the lower heritability among Andeans may indicate that there is insufficient variation in the genes involved to demonstrate a heritable component or it may mean that the genes influencing hypoxic ventilatory response were lost from the Andean gene pool. Since heritability estimates are derived for a given population, strictly speaking, they should not be compared between populations. In addition to alveolar ventilation, arterial oxygen saturation is influenced by the alveolar-arterial oxygen diffusion difference [(A-a)DO2], hemoglobin-oxygen affinity, and alveolar/end-capillary diffusion disequilibrium. Diffusion limitation is important during heavy levels of exercise at sea level (169). It is also important at high altitude where the oxygen driving pressure is low and arterial oxygen tensions are on the steep part of the dissociation curve, although the lower absolute cardiac

74

Niermeyer et al.

outputs attained at maximal exercise would be expected to lessen diffusion limitation by lengthening transit time. Reduction in (A-a)DO2 permits high-altitude natives versus acclimatized newcomers to better maintain arterial oxygen saturation at a given workload (158) or level of oxygen consumption (Fig. 8). The smaller (Aa)DO2 is likely the result of greater surface area for gas exchange due, in turn, to larger lung volumes (Fig. 5). Of interest is that (A-a)DO2 is equally narrowed in Tibetan, Andean, and Rocky Mountain high-altitude residents (Fig. 8). Hemoglobin-oxygen affinity varies as the result of genetic variation in hemoglobin structure as well as variation in the effectors of hemoglobin-oxygen binding (pH, temperature, red blood cell concentrations of 2,3-diphosphoglycerate and adenosine triphosphate). There are no reports of electrophoretic or DNA sequence hemoglobin variants in human high-altitude populations, but if present, they do not appear to affect hemoglobin-oxygen affinity as measured by the position of the hemoglobinoxygen dissociation curve. Existing Andean and Himalayan data indicate that hemoglobin-oxygen affinity is fully normal by sea-level standards (170,171). Generally, studies have been performed only at rest; in Tibetans at near-maximal exercise, standard P50 does not change and in vivo P50 rises modestly to 28.4 ⫾ 0.5 mmHg as the result of mild acidosis (arterial pH of 7.38 ⫾ 0.01) (L. G. Moore, unpublished observations). Resting arterial oxygen saturation appears influenced by heritable factors in lifelong Tibetan but not in Andean high-altitude residents (172,173). In Tibetans at 4850–5450 m and 3800–4065 m, complex segregation analysis of familial variation in arterial oxygen saturation revealed a pattern of inheritance consistent with a simple, single, autosomal locus with dominance (172). It is not clear what this gene (or

Figure 8 The alveolar-arterial oxygen difference [(A-a)DO2] is lower in high-altitude natives than acclimatized lowlanders. Andean, Rocky Mountain, and Himalayan values are similar in relation to levels of oxygen consumption (VO2). (From Ref. 233.)

The People

75

group of closely linked genes) is. The variation in arterial oxygen saturation was not related to ventilation, ventilatory response to hypoxia, hemoglobin, age, or sex (34,174). It is important to recall, again, that the lack of significant heritability among Andean highlanders does not indicate that genetic influences were absent but may simply be due to insufficient variation to demonstrate a heritable component. Hemoglobin concentration at a given altitude is lower in Himalayan than Andean highlanders. This is evident in large-scale surveys (175) and carefully matched comparisons of Sherpa and Aymara from similar, rural regions (176). Sherpas demonstrate lower and more normally distributed hemoglobin and hematocrit values than the Aymara (Fig. 9). The lower hemoglobin may result from a lesser hypoxic stimulus, due perhaps to better-maintained ventilation, as described above. Alternatively or additionally, higher erythropoietin levels in Aymara than Sherpa residents at the same altitude suggest that Andeans may have greater erythropoietic response to a given level of hypoxic stimulus (176). Other factors may be involved, including developmental regulation of hemoglobin production (177) or variation in red blood cell destruction. Cultural and environmental characteristics are also important; lower hemoglobin concentrations in rural than urban highlanders suggest that dust and other factors increasing the risk of chronic lung disease may magnify the erythropoietic effects of ambient hypoxia (178). Cardiac output rises with progressive exercise, serving to increase oxygen and other nutrient delivery to the exercising muscle. The greater exercise capacities in native highlanders than acclimatized newcomers suggest higher cardiac outputs,

Figure 9 The distribution of hematocrit (%) in adult, rural male residents of equivalent altitudes is shifted to lower values in Nepalese Sherpa (filled bars) than in Chilean Aymara (hatched bars). (From Ref. 176.)

76

Niermeyer et al.

which could be due, in turn, to higher heart rates and/or stroke volume. Existing data do not support higher heart rates at maximal exercise but rather suggest lower values in high-altitude natives than acclimatized newcomers (Fig. 6). Several studies support the concept that natives are more reliant on parasympathetic tone, whereas newcomers depend more on beta-sympathetic stimulation to raise heart rate at maximal effort (179–181). The possibility of increased stroke volume is supported by greater resting heart volumes in young Tibetan men compared with similarly sized and aged, healthy Han residents of Lhasa (604 ⫾ 21 vs. 550 ⫾ 24 cm3 or 377 ⫾ 11 versus 347 ⫾ 12 cm3 /m2 body surface area) (L. G. Moore, unpublished data). A rise in stroke volume can be accomplished by an increase in preload, greater myocardial contractility, or a reduction in afterload. Few studies have measured all three factors. In native Tibetans at 3658 m, we found that preload (right and left heart filling pressures) was low and rose normally with exercise (182). Stroke volume rose with increasing filling pressure in a fashion that was normal by sea-level standards and greater than previously reported for Colorado or Peruvian high-altitude natives (183,184). Furthermore, stroke volume at a given right atrial mean or pulmonary capillary wedge pressure was not elevated by the addition of 100% oxygen to the inspired air, implying no hypoxic depression of myocardial contractility. Pulmonary arterial pressure and resistance were remarkably low and unresponsive to added hypoxia when compared with previous studies in lifelong Rocky Mountain or Andean high-altitude natives (182). Thus, particularly during exercise, Tibetans demonstrated a lower pulmonary arterial pressure gradient across the lung than Andean high-altitude natives (Fig. 10) and a decrease in systemic vascular resistance, enabling the Tibetans to achieve a more than threefold elevation in cardiac output with good cardiac functional reserve. Consistent with the low pulmonary arterial pressures and absence of hypoxic vasoconstriction, Gupta et al. have reported a lack of smooth muscle in the small pulmonary arteries of native Ladakhi men (185). The absence of pulmonary hypoxic vasoconstriction and vascular remodeling in these human studies is similar to that which has been reported in the yak (186,187), snow pig (188), and llama (189)—all species considered to be genetically adapted to high altitude. The Tibetans’ well-maintained myocardial contractility and low vascular resistance in both the pulmonary and systemic circulations would be expected to permit greater cardiac output during exercise than in Andean or Rocky Mountain high-altitude residents. During exercise, the preservation of cerebral blood flow appears to be favored in Tibetan high-altitude natives compared to acclimatized newcomers. Using internal carotid artery blood flow as an index of cerebral blood flow, young Tibetan men maintained higher flow and better preserved cerebral oxygen delivery at peak effort than Han acclimatized newcomers (190). A comparison of Himalayan natives with European elite climbers revealed fewer MRI changes indicative of cortical atrophy and brain damage in the high-altitude natives, despite more frequent exposure to extreme high-altitude without supplemental oxygen (191). As reviewed above, the greater redistribution of lower extremity blood flow to favor the uteroplacental circu-

The People

77

Figure 10 The pressure gradient across the lung (Ppa —Pcw, pulmonary arterial pressure— pulmonary capillary wedge pressure, mmHg) changes little with increasing oxygen consumption (VO2, mL STPD/min) in sea-level residents (open squares) (169,234), sea-level residents after acute exposure to 3000 m (open diamonds) (169,234), Tibetan residents of 3658 m (squares with crosses) (182) but rises markedly in Andean lifelong residents of 3750 m and 4540 m (open triangles) (184,235).

lation in Tibetan than Han pregnant women also indicates good blood flow distribution to organs with high oxygen demand. Tissue oxygen utilization can be increased by an augmentation of arterial oxygen content, blood flow, or tissue oxygen extraction. As a measure of tissue oxygen extraction, we plotted all published data for the differences in arterial-venous oxygen content versus cardiac output in male sea-level residents, high-altitude natives, and after acute or more prolonged exposure to high altitude (Fig. 11). Acute hypoxia lowered the arterial-venous oxygen content per unit blood flow, probably as the result of decreased arterial oxygen saturation. After acclimatization, the slope of the line relating arterial-venous oxygen content to blood flow left-shifted toward sealevel values, probably as the result of a rise in arterial oxygen content. The slopes of the lines in Rocky Mountain, Andean, and Tibetan high-altitude natives were close to that of sea-level natives (Fig. 11), suggesting that further time-dependent changes raised extraction, perhaps by lowering mixed venous oxygen content. Whether regional high-altitude populations differ in terms of oxygen extraction cannot be determined from the available data because only our Tibetan subjects exercised across a sufficient range of exercise loads.

78

Niermeyer et al.

Figure 11 The arterial-venous difference in oxygen content (CaO2 —CvO2, mL O2 /L whole blood) rises with increasing cardiac output during exercise in sea-level residents (open squares) (169,234). The difference is narrowed for a given cardiac output in sea-level residents after acute exposure to 3000 m (open diamonds) (169,234) and in sea-level residents after 14 days acclimatization to 3800 m (open circles) (234). Lifelong high-altitude residents of Tibet (downward pointing triangles) (182), the Andes [squares with crosses (236); diamonds with crosses (235); circles with crosses (184)], and Colorado (upward pointing triangles) (183) more closely resemble sea-level residents.

Tissue oxygen extraction can be influenced by differences in skeletal muscle capillarity, mitochondrial volume density, oxidative enzyme capacity, or reliance on anaerobic metabolism. Capillary density in Sherpa mountaineers was within the range of acclimatized European values. The muscle fiber area perfused by each capillary was greater, not reduced, in Sherpas compared to European climbers (192). The relative or absolute enzyme mitochondrial oxidative capacity was also diminished (144,192). Lower lactate levels at a given workload in lifelong high-altitude natives than acclimatized newcomers (or sea-level residents) suggest that there is not greater reliance on anaerobic metabolism (148,149). Hochachka and coworkers have advanced the idea that closer coupling of ATP supply and demand permitted more efficient mitochondrial oxygen utilization and, in turn, lowered glycolysis and lactate production (144,148). In support of this, NMR studies of muscle metabolism conducted at sea level demonstrated that Quechua highlanders accomplished greater calf muscle work with similar or reduced perturbations in phosphorylation potential, phosphocreatine or ATP concentration than sedentary, endurance, or power-trained athletes. Alternatively, Kayser and coworkers have suggested that shorter oxygen diffusion path, greater oxygen conductance, or increased tissue oxygen stores as the

The People

79

result of elevated myoglobin or other oxygen-binding pigments may be involved (192). Interestingly, the low mitochondrial volume/specific VO2 peak ratio appears to be an inborn feature of Tibetans (193). In summary, maximal exercise capacity appears to be restored toward sea level values after generations among Andean and Himalayan high-altitude residents compared with acclimatized newcomers. Several factors are likely to contribute to this recovery. One factor is the maintenance of increased alveolar ventilation and hypoxic ventilatory response, a significant factor for increasing ventilation during exercise as well as at rest (194). Another is the decreased (A-a)DO2. One of the most decisive differences observed among high-altitude populations is the Tibetans’ low pulmonary arterial pressure (182). The consequent reduction in afterload to the right heart may permit the Tibetans to achieve a greater rise in cardiac output during exercise than acclimatized newcomers or other high-altitude residents. Other tissuerelated characteristics serving to increase tissue oxygen utilization should receive additional investigation. Chronic Mountain Sickness and Other Chronic Diseases

Comparatively little attention has been paid to the occurrence of chronic diseases during adulthood and old age at high altitude with the notable exception of chronic mountain sickness (CMS), also called Monge’s disease (33). CMS occurs in persons who have lost their ventilatory acclimatization to high altitude after seemingly normal adjustment to residence at high altitude. It is characterized by symptoms of clubbing, facial edema, conjunctival congestion, headache, dizziness, paresthesias, fatigue, loss of memory, insomnia, poor appetite, and dyspnea. CMS is usually diagnosed on the basis of hemoglobin (or hematocrit) concentrations above the normal range in the absence of clear evidence of chronic lung or left-sided heart disease (195,196). A symptoms score is sometimes used (197). Several hypotheses have been advanced to account for the occurrence of CMS. One hypothesis is that CMS is due to an age-associated loss of ventilatory sensitivity to hypoxia and consequent reduction in ventilation, arterial oxygen tension, and rise in hemoglobin concentration (198,199). Persons with CMS have lower levels of ventilatory responsiveness to hypoxia, effective alveolar ventilation, and arterial oxygen tensions than healthy persons at the same elevation (200). However, persons with low hypoxic drives live successfully at high altitude for years, and some persons with CMS have hypoxic responses within the normal range, suggesting that an ageassociated decline in hypoxic ventilatory sensitivity is not the only factor. Other abnormalities of ventilatory control may be involved. For example, we have found that ventilatory depression with hypoxia was greater in individuals with CMS than in age-matched controls (196,200). While an age-associated increase in hemoglobin or hematocrit in Leadville has been reported, it is important to recall that there is considerable variation in this age-related pattern (201). Another view is that CMS represents subclinical chronic obstructive lung disease or left-sided heart disease, the symptoms of which are altered by residence at

80

Niermeyer et al.

high altitude (202). In support of this, approximately half the persons with CMS in Leadville, Colorado, demonstrated some impairment of lung function (202). We observed that CMS patients in Lhasa with lower arterial oxygen saturations had smaller vital and total lung capacities than the better-oxygenated CMS patients, suggesting that impaired lung function and widening of the alveolar-arterial oxygen gradient contributed to the severity of the disease (200). Mining is an important economic activity in the North American and many of the South American communities where CMS has been described, creating the possibility that occupational factors increasing the risk of obstructive lung disease contribute to its etiology. In support of this are the generally lower hemoglobin values in rural than in urban regions of South America (178) and the high prevalence of cigarette smoking among persons with CMS (202,203). It has been suggested that persons with CMS have an excessive bone marrow response to a given level of hypoxia, but greater serum immunoreactive erythropoietin levels have not been observed in CMS patients compared with healthy controls (204). Additional investigation of factors influencing erythropoiesis is warranted. CMS may be a manifestation of disordered breathing and brain blood flow regulation during sleep. Kryger and coworkers (205) found sporadic episodes of arterial desaturation in CMS patients in Leadville, Colorado (3100 m). Following treatment with the respiratory stimulant, medroxyprogesterone acetate, episodes of arterial desaturation became less common, hemoglobin fell, and symptoms overall improved. We documented an increased frequency of sleep-disordered breathing (apneas and hypopneas) and prolonged episodes of hypoventilation in CMS patients compared with healthy, age-matched control residents of Lhasa, Tibet (3658 m) (206). While higher hemoglobin concentration prevented lower arterial oxygen saturation from decreasing arterial oxygen content during the day, the nighttime disturbances in breathing reduced arterial oxygen content well below control values. Cerebral oxygen delivery during sleep appeared to be further reduced by a blunted brain blood flow response to hypercapnia and hypoxia; CMS patients had less rise in internal carotid artery blood flow than controls during and immediately following episodes of sleep-disordered breathing (206). The fall in arterial oxygen tensions during the night would be expected to stimulate erythropoiesis and augment pulmonary vasoconstriction. Since metabolic alterations have been implicated in blunting hypoxic ventilatory sensitivity and augmenting hypoxic ventilatory depression, it is possible that similar mechanisms mediate the lack of brain blood flow response and ventilatory characteristics of CMS. Whatever the cause(s), it is likely that the reduction in brain oxygen delivery contributes to the cognitive impairment and neurological symptoms of CMS (206). The prevalence of CMS appears to vary among and within regions as well as between men and women in the same region. Prevalence data, ironically (considering its proximity), are the least complete for the Rocky Mountain region. Kryger et al. remarked that 60 men and 2 women in Leadville, Colorado, at 3100 m were under treatment that would yield a minimal prevalence estimate of only 3% (196). While the mobile nature of the Rocky Mountain population complicates the conduct

The People

81

of surveys, the altitude at which persons sleep may be less variable and of greater physiological importance. A comprehensive prevalence study has been undertaken in 2875 male and 152 female residents of 4300 m in Cerro de Pasco, Peru (197,199,207). Population ancestry was not reported, but most persons in this community are of mestizo parentage. In men, diagnosis of excessive polycythemia was based on hemoglobin concentration greater than 21.3 g/dL in combination with an elevated symptom score. This hemoglobin cutoff was defined as 2 SD above the average value for 20 to 39-year-old men. In women, diagnosis of excessive polycythemia was based on a hematocrit greater than 56%, defined as greater than 2 SD above the average hematocrit for women aged 30–40 years. The prevalence of excessive polycythemia was greater in men than women, occurring in 16% of adult males and 9% of women between the ages 30 and 54 years. Prevalence rose with age for both sexes. There was a uniform age-associated increase in hematocrit in men, whereas in women the relationship was relatively flat until after the age of the menopause (197). Among men, 7% aged 20–29 years versus 34% aged 60–69 years had excessive polycythemia (207). Among the 9% of women with excessive polycythemia, 77% were between 45 and 54 years (197). Prevalence of CMS in the Himalayan region is greater in Han migrants than in Tibetan lifelong residents. In an epidemiological survey of 3201 persons at 3600– 4700 m in Tibet, CMS was much more frequent among Han than Tibetans and more common in men than in women (Table 4). Han had a nearly 10-fold excess of CMS compared with Tibetans in a survey of 25,618 persons living at 2261 to 5225 m in

Table 4 Prevalence (%) of Excessive Polycythemiaa in the Tibet Autonomous Region in 3201 Male (n ⫽ 1749) and Female (n ⫽ 1452) Migrants (Han) and Native Highlanders (Tibetans) More Than 15 Years Old Males

Region (alt, m) Lhasa (3658 m) Gyangze (4040 m) Nagqu (4500–4700 m)

Migrant workers

Native workers

12.97

1.05

31.5 38.4

Females Native farmers, herders

Migrant workers

Native workers

Native farmers, herders

1.64

4.8

1.5

3.8

0.3

14.4

6.6

7.2

6.5

2.7

Excessive polycythemia at 3658 and 4040 m was defined as red blood cell (RBC) counts ⬎ 6.5 ⫻ 106 / µL blood and hemoglobin (hgb) ⬎ 20 g/dL blood, or RBC counts ⬎ 7.15 ⫻ 106 /µL, or hgb ⬎ 22 g/ dL. At 4500–4700 m, excessive polycythemia was considered as RBC ⬎ 7.0 ⫻ 106 /µL and hgb ⬎ 21 g/dL, or RBC ⬎ 7.7 ⫻ 106 /µL, or hgb ⬎ 23 g/dL. Source: Ref. 208. a

82

Niermeyer et al.

Qinghai Province (209). Most cases occurred in cigarette-smoking Han men (none of whom were miners) who emigrated from lowland regions as adults (200,203). Tibetans are not fully protected from CMS, but all the cases we have seen were in persons born in lower altitude regions of Tibet after extended residence at much higher elevations (200,206). It is difficult to compare prevalence estimates among the Himalayan, Andean, and Rocky Mountain regions, given the lack of population ancestry data for Andeans, age-specific prevalence estimates for Himalayans, and complete prevalence information from the Rocky Mountain region. Existing data suggest that the prevalence among the Han is within the range of Andean values and that both the Han and Andean values are greater than the Tibetans’. The Tibetans’ lower prevalence may, in turn, be due to their lower hemoglobin concentrations, higher ventilations, and/or greater hypoxic ventilatory drives. Lower hemoglobin concentrations appear to be present lifelong in Tibetans. Unknown is whether the effects of aging on ventilation and arterial oxygenation, particularly during sleep, differ in Tibetans and other high-altitude residents. We did not observe any differences in the occurrence of sleep-disordered breathing in young Han and Tibetan men at 3658 m (206). Information about other chronic cardiovascular or respiratory diseases in highaltitude residents is limited. The prevalence of systemic hypertension is reported to be low in Peruvian high-altitude communities (210) but increased among lifelong Tibetan residents of high altitude compared with Han newcomers (211). In the Peruvian study, both diastolic and particularly systolic pressures were lower (210), whereas in Tibet higher diastolic and systolic pressures were seen (211). In the Peruvian study, genetic, dietary, and other similarities between the high- and lowaltitude regions led the investigators to postulate that the blood pressure reduction was due to chronic hypoxia acting to lower peripheral vascular resistance and/or cardiac output (210). This is consistent with the decline in blood pressure observed in men after years of high-altitude residence (212), by which time the increase in blood pressure resulting from sympathetic stimulation (213) has presumably been reversed. The high salt intake in the Tibetan diet may be a factor in the high prevalence of hypertension observed. Among Rocky Mountain residents, mortality from emphysema and chronic bronchitis is increased at high compared with low altitudes (214,215). Persons died after a shorter duration of illness and more commonly from right heart failure, suggesting that the hypoxia of high altitude rather than some other factor associated with high-altitude residence worsened survival. There are also fewer elderly, measured as the proportion of the population over the age of 65, at high than at low altitudes in Colorado (216). This is not the result of increased mortality but rather increasing outmigration after the age of 50 years in combination with an influx of youngeraged residents. The most frequent cause for relocation cited by persons who migrated from high altitude was poor health, whereas family-related concerns were the principal cause of migration for persons moving from one low-altitude location to another. The health problems cited were, overwhelmingly, complications of heart and lung disease (216).

The People

83

In summary, CMS is a clear instance of adaptive failure at high altitude. Persons who die from CMS (or its associated conditions, including right heart failure and strokes) are often elderly, but it affects persons in the early years of adulthood as well. Susceptibility to CMS appears to be greater among migrants to high altitude than among natives. Lower prevalence in Andean compared with Tibetan highaltitude residents suggest that generational factors and/or the degree of genetic admixture may be involved. The protection afforded Tibetans may be due to their higher ventilation and hypoxic ventilatory drives as well as to other factors. Further research is needed to provide more consistent, age-specific diagnostic criteria for CMS. Not only altitude but also age and gender should be taken into account for defining cutoff values for what constitutes ‘‘normal’’ hemoglobin. In the case of women, such standards need to take account of other factors influencing hemoglobin concentration, such as menstruation and pregnancy.

IV. Summary and Conclusions Recent studies using an expanded range of techniques have revealed differences between lifelong high-altitude residents and acclimatized newcomers as well as among resident high-altitude populations. These differences are of physiological and anthropological importance insofar as they imply that generational factors influence adaptive processes. Here we try to summarize what has been learned, indicating some of the complexity influencing the interpretation of such studies and suggesting future studies that might be undertaken to resolve at least some of the issues remaining. As reviewed above, geographical and historical circumstances differ among high-altitude populations. The Tibetan Plateau is larger, more geographically remote, and has been occupied by humans for a longer period of time than the Andean Altiplano or Rocky Mountain Plateaus. Given these distinctions, the Tibetan population is likely to have been the longest resident at high altitude. Compared with Andeans, the Tibetan gene pool is less likely to have been constricted by small numbers of initial migrants and/or severe population decline. Tibetans may also have been subject to less genetic admixture with lowland groups, although this may be changing today with the influx of large numbers of Han migrants. Admixture with lowlanders is not an issue for Rocky Mountain inhabitants as they are persons of low-altitude ancestry. While viewed by some investigators as a limitation, the fact that Rocky Mountain residents are of low-altitude ancestry has the advantage of providing a control group with which other high-altitude populations can be compared. Future studies are required to better document the population history, extent of genetic admixture, and relationship of genetic to physiological traits in highaltitude populations. For such admixture and population history studies, the important but often overlooked sex-specific nature of gene flow needs to be taken into account. Population history and admixture studies are performed using mitochon-

84

Niermeyer et al.

drial DNA (mtDNA), y-chromosomal nuclear DNA (yDNA), and/or autosomal nuclear DNA (nDNA) genetic systems. MtDNA and yDNA offer special advantages since they do not undergo recombination (by virtue of being inherited solely through the mother or father, respectively). In addition, mtDNA appears to accumulate variation (mutations) more rapidly than nuclear DNA (nDNA) (217). However, admixture estimates vary, depending on what kind of DNA is employed. For example, in the San Luis Valley, Colorado, intermarriage between Native Americans and SpanishAmericans occurs disproportionately between Native American women and Hispanic men. As a result, 87% of Hispanic persons had mtDNA of Native American origin, whereas in the same individuals the estimate of Native American admixture using nDNA was 48% (15). In an analogous situation, a greater degree of European admixture was inferred in Central American tribes when yDNA rather than mtDNA was evaluated, indicating that the European contribution to Central American gene pools was predominantly male in origin (218). The relevance for high-altitude populations is that admixture with Europeans may not be detected in Andean populations using mtDNA alone. If contact between Tibetan and other populations was sexspecific, admixture estimates would likely vary according to the source of the genetic material. For example, Mongolian conquest and movements of male traders would be expected to yield higher yDNA than mtDNA admixture estimates. The possibility exists that such differences in admixture estimates might be useful for inferring patterns of contact, although such analyses quickly become very complicated. Other cultural practices can also influence mtDNA variation; for example, polygyny (one man with several wives) would be expected to increase mtDNA variation and female infanticide to decrease mtDNA variation. Several differences in adaptive success between natives and newcomers have been identified. Lifelong high-altitude residents of the Andes and/or Himalayas have the following distinctions when compared with acclimatized newcomers: Less intrauterine growth retardation Better neonatal oxygenation and involution of fetal cardiopulmonary characteristics Enlarged lung volumes and decreased alveolar-arterial oxygen diffusion gradients Higher maximal exercise capacity (VO2 max) Better maintained increase in cerebral blood flow during exercise (Tibetans only) Lower hemoglobin concentrations (Tibetans only) Less susceptibility to CMS (Tibetans only) Among high-altitude native populations, several differences in adaptive success and physiological strategy have emerged from recent comparisons. According to an evolutionary hypothesis, the greater duration and genetic isolation of the Tibetan population would be expected to result in better adaptation, as judged by the existence of attributes associated with improved chance of reproductive success. Ordering the comparison in relation to the Tibetan population, Tibetans demonstrate:

The People

85

Less intrauterine growth retardation Greater reliance on redistribution of blood flow than elevated arterial oxygen content to increase uteroplacental oxygen delivery during pregnancy Higher levels of resting ventilation and hypoxic ventilatory responsiveness Less hypoxic vasoconstriction, lower pulmonary arterial pressure and resistance Lower hemoglobin concentration Less susceptibility to CMS Several of the distinctions demonstrated by Tibetans parallel the differences between natives and newcomers, although the degree of protection or adaptive benefit appears enhanced for the Tibetans. However, many of these observations are based on limited information, and the requisite comparative data from other regions are not always available or collected with the same methodologies. For example, direct comparisons of cardiac output and pulmonary arterial pressure are needed in larger numbers of Tibetan and Andean high-altitude residents at similar levels of exercise intensity. Another feature of the above list is that characteristics which are likely to be closely related to mortality risk (e.g., IUGR, uteroplacental blood flow, susceptibility to CMS) demonstrate clearer differences between natives and newcomers or among native populations than other characteristics such as exercise performance. This may be the result of difficulties in accurately measuring exercise performance, given the influences of training, motivation, etc., or it may indicate that differences in exercise performance are less affected by the processes of natural selection operative at high altitudes. The higher frequency of IUGR and preeclampsia at high altitude provides a useful opportunity to generate insights into pathophysiology. Finally, the above list demonstrates that selective pressures differ between the sexes. These differences are particularly apparent during pregnancy. The apparently lower susceptibility to CMS among women than men raises the intriguing possibility that sex differences in adaptive processes are of lifelong significance. There are also several points of similarity. One is the seemingly equal retardation of postnatal growth among highland infants, children, and adolescents when compared with well-validated, low-altitude NCHS standards. Since it is likely that insufficient nutrition and recurrent illnesses contribute to the growth retardation observed, additional studies are warranted that examine the separate and combined effects of hypoxia, nutrition, and illness on growth. Such studies could be usefully conducted in the Rocky Mountain region where the effects of hypoxia alone should be more readily apparent. Such studies of the interactive effects of altitude, poor nutrition, and intercurrent illness are important for regional and health policy planners. Another related issue concerns the impact of IUGR. As reviewed above, information on gestational age along with birth weight is essential for comparing birth weight–specific mortality. This, in turn, requires complete population enumeration or, at least, large and demonstrably representative samples. With the efforts to improve health and heath care in many regions of the world, such information may be forthcoming and will be important for determining the relative influences of pre-

86

Niermeyer et al.

maturity and IUGR on mortality risk at high altitude. Studies in the neonatal period and infancy are required to address the consequences of preterm delivery, IUGR, poor or delayed cardiopulmonary transition, and intercurrent respiratory illness for mortality and morbidity at high altitude. Lung volumes appear universally increased in lifelong high-altitude residents. While subtle variations may be important, the decrease in alveolar-arterial oxygen tension differences appears essentially the same in Rocky Mountain, Andean, and Himalayan highland residents. Additional studies are warranted to determine what kinds of growth signals selectively stimulate the lungs and the mechanisms by which such an increase diminishes alveolar-arterial oxygen tension differences at high altitude. These observations suggest several conclusions regarding convergence and complexity of adaptive events. In terms of convergence, the Himalayan Plateau has not only been the longest occupied and most remote, its occupants demonstrate several physiological advantages that are consistent with their hypothesized evolutionary advantage. Of interest is that the overall pattern of successful adaptation appears to result in establishing the physiological and functional attributes of sealevel residents. For example, Tibetans more closely resemble healthy persons at sea level in terms of their characteristics of intrauterine growth and birth weights, uterine artery blood flow, hemoglobin concentrations, and pulmonary arterial pressures. This supports the view that some of the physiological responses commonly observed in sojourners and North or South American high-altitude residents—e.g., decreased birth weight, polycythemia, diminution of ventilation, moderate elevation in pulmonary arterial pressure—are not adaptive. It also raises the presently unanswered question as to what underlying factor(s) has permitted Tibetans to adapt. In terms of complexity, it is important to recall that not just the directional forces of natural selection but also the chance-driven processes of genetic drift (i.e., the loss or fixation of alleles as a result of small population size), gene flow, and mutation are involved in evolutionary change. These processes are exemplified by the possibly small numbers of initial migrants to the Americas, severe reductions in population size at the time of sixteenth-century European conquest, and presentday gene flow in the Andean and Himalayan regions. In addition, natural selection is not a simple process. One source of complexity is that a change in adaptation may not alter fitness (i.e., the net result of all adaptations affecting differential fertility and mortality). For example, an adaptation that increases fertility may increase susceptibility to a particular disease and therefore be ‘‘canceled out’’ by increased mortality. Fitness, not adaptation, directs evolutionary change. Neither organisms nor environments remain static and hence an individual’s state of adaptation must be evaluated with respect to both historic and present-day environmental conditions. Traits seemingly adaptive today may have arisen under a different set of selective pressures, making the evolutionary scenarios we reconstruct from present-day utility nothing more than ‘‘just so’’ stories (219,220). Other possible explanations of seemingly adaptive traits include close linkage with other traits undergoing selective pressure,

The People

87

random events such as genetic drift, or allometric relationships involved in growth. There may be multiple solutions to a given problem, not a single best adaptive solution. Future progress will likely be achieved in identifying the genes involved in governing physiological responses to hypoxia and determining whether genetic variants (alleles) contribute to differences between natives and newcomers and/or among native populations. The methods of statistical genetics are useful for determining whether the pattern of transmission in a large group of biologically related individuals (a family set) suggests a single gene with multiple alleles, multiple genes, differential penetrance, or non-Mendelian factors (e.g., imprinting, mosaicism and uniparental disomy) (22). Statistical genetic techniques do not stand alone but must be accompanied by linkage analysis to determine the location of the gene(s). This is accomplished by comparing the co-segregation of the trait of interest with traits that have already been genetically mapped, given that the frequency with which they segregate together is determined by their chromosomal location. While a full genomic search is possible, candidate genes are usually chosen on the basis of being plausibly related to the trait of interest. This is facilitated by the observation that traits that are related in function are often located in contiguous gene blocks on the chromosome, perhaps the result of evolution by gene duplication (221). Of interest will be whether a discrete, common set of hypoxia-sensitive genes are identified or whether there are multiple genetic factors involved that vary broadly among populations. Future progress can also be anticipated in achieving a more integrated view of high-altitude adaptation. Part of this integration will derive from an increased understanding of the ways in which levels of biological organization are articulated and influence each other. Assistance will be provided by the increasing availability of new (and often minimally invasive) techniques for evaluating cellular metabolism, energy expenditure, and body composition. Another path toward achieving integration is likely to stem from the inclusion of a broader range of subjects, including women and men, young and old, native and newcomer. The detailed physiology of adaptation and long-term acclimatization that has been worked out in men is being extended to women in ways that consider the influences of the menstrual cycle, hormone replacement, oral contraceptives, menopause, and other gender-related influences. Studies of populations residing at high altitude in the twenty-first century will consider not only historically defined groups native to high-altitude regions but heterogeneous groups who form new settlements at high altitude. References 1. 2. 3. 4.

The Hammond Medallion World Atlas. Maplewood, NJ: Hammond, Inc., 1984. World Health Statistics Annual. Geneva: World Health Organization, 1996. Sensui Z. Uncovering prehistoric Tibet. China Reconstructs 1981; 1:64–65. Zhimin A. Paleoliths and microliths from Shenja and Shuanghu, Northern Tibet Curr Anthropol 1982; 23:493–499.

88

Niermeyer et al.

5. Denell RW, Rendell HM, Hialwood E. Late Pliocene artifacts from northern Pakistan. Curr Anthropol 1988; 29:495–498. 6. Matsumoto H. Characteristics of the Mongloid and neighboring populations on the basis of the genetic markers of immunoglobins. J Anthro Soc Nippon 1987; 95:291– 304. 7. Torroni A, Miller JA, Moore LG, Zamudio S, Zhuang J, Droma T, Wallace DC. Mitochondrial DNA analysis in Tibet: implications for the origin of the Tibetan population and its adaptation to high altitude. Am J Phys Anthropol 1994; 93:189–199. 8. Turner CG. Late Pleistocene and Holocene population history of East Asia based on dental variation. Am J Phys Anthropol 1987; 73:305–321. 9. Lee TD, Zhao TM, Mickey R, Sun YP, Lee G, Song CX, Cheng DZ, Zhou S, Ding SQ, Cheng DX. The polymorphism of HLA antigens in the Chinese. Tissue Antigens 1988; 32:188–208. 10. Zhao TM, Lee TD. Gm and Km allotypes in 74 Chinese populations: a hypothesis of the origin of the Chinese nation. Hum Genet 1989; 83:101–110. 11. Avedon JF. Exile in the Land of Snows. New York: Vintage Books, 1986:3–33. 12. Farabegoli A, Barbujani G. Diversity of some gene frequencies in European and Asian populations. VI. Geographic patterns of PGM and ACP. Human Hered 1990; 40:313– 321. 13. Zei G, Barbujani G, Lisa A, Fiorani O, Menozzi P, Siri E, Cavalli-Sforza LL. Barriers to gene flow estimated by surname distribution in Italy. Ann Hum Genet 1993; 57: 123–140. 14. Neel JV, Biggar RJ, Sukernik RI. Virologic and genetic studies relate Amerind origins to the indigenous people of the Mongolia/Manchuria/southeastern Siberia region. Proc Natl Acad Sci USA 1994; 91:10737–10741. 15. Merriwether DA, Rothhammer F, Ferrell RE. Distribution of the four founding lineage haplotypes in Native Americans suggests a single wave of migration for the New World. Am J Phys Anthropol 1995; 98:411–430. 16. Chen YS, Torroni A, Excoffier L, Santachiara-Benerecetti AS, Wallace DC. Analysis of mtDNA variation in African populations reveals the most ancient of all human continent-specific haplogroups. Am J Hum Genet 1995; 57:133–149. 17. Torroni A, Schurr TG, Cabell MF, Brown MD, Neel JV, Larsen M, Smith DG, Vullo CM, Wallace DC. Asian affinities and continental radiation of the four founding Native American mtDNAs [see comments]. Am J Hum Genet 1993; 53:563–590. 18. MacNeish RS, Berger RP. Megafauna and man from Ayacucho, highland Peru. Science 1970; 166:8975–977. 19. Lynch TF. The paleo-indians. In: Jennings JD, ed. Ancient South Americans. San Francisco: W.H. Freeman and Co., 1978:87–137. 20. Nunez L. Paleoindian and archaic cultural periods in the arid and semiarid regions of northern Chile. Adv World Archaeol 1983; 2:161–203. 21. Blanco R, Chakraborty R. Genetic distance analysis of twenty-two South American Indian populations. Human Hered 1975; 25:177–193. 22. Cook ND. Demographic Collapse, Indian Peru, 1520–1620. New York: Cambridge University Press, 1981. 23. Chakraborty R, Barton SA, Ferrell RE, Schull WJ. Ethnicity determination by names among the Aymara of Chile and Bolivia. Hum Biol 1989; 61:159–177. 24. Post RH, Neel JV, Schull WJ. Tabulation of phenotype and gene frequency for 11 different genetic systems studied in the American Indians. In: Biomedical Challenges

The People

25.

26. 27.

28. 29. 30.

31.

32.

33. 34.

35. 36.

37. 38.

39. 40.

41.

42.

89

Presented by the American Indian. Washington, DC: Pan American Health Organization, 1968:141–185. Kolsteren P, Atkinson S, Maskall K. Nutrition, Health, Water and Sanitation Assessment in the Lhasa Valley. London: Save the Children Fund/Lhasa Health Bureau, 1990. Richardson JB. People of the Andes. Washington, DC: Smithsonian Books, 1994. Thomas RB. Energy flow at high altitude. In: Baker PT, Little MA, eds. Man in the Andes. A Multidisciplinary Study of High Altitude Quechua. Stroudsburg, PA: Dowden, Hutchinson and Ross, Inc., 1976:379–404. Leonard WR, Thomas RB. Biosocial responses to seasonal food stress in highland Peru. Hum Biol 1989; 61:65–85. Leonard WR. Household-level strategies for protecting children from seasonal food scarcity. Soc Sci Med 1991; 33:1127–1133. Hanna JM. Natural exposure to cold. In: Baker PT, Little MA, eds. Man in the Andes. A Multidisciplinary Study of High-Altitude Quechua. Stroudsburg, PA: Dowden, Hutchinson and Ross, Inc., 1976. Baker TS. Child care, child training and environment. In: Baker PT, Little MA, eds. Man in the Andes. Stroudsburg, PA: Dowden, Hutchinson and Ross, Inc., 1976:85– 97. Tronick EZ, Thomas RB, Daltabuit M. The Quechua manta pouch: a caretaking practice for buffering the Peruvian infant against the multiple stressors of high altitude. Child Dev 1994; 65:1005–1013. Monge CM. Acclimatization in the Andes. Baltimore, MD: Johns Hopkins Press, 1948: 36–37. Niermeyer S, Yang P, Shanmina, Drolkar, Zhuang J, Moore LG. Arterial oxygen saturation in Tibetan and Han infants born in Lhasa, Tibet. N Engl J Med 1995; 333:1248– 1252. Abelson AE. Altitude and fertility. Hum Biol 1976; 48:83–92. Gonzales GF. Determinates biomedicos de la fertilidad humana en la altura. In: Gonzales GF, ed. Reproduction Humana en la Altura. Lima, Peru: Consejo Nacional de Ciencia y Tecnologica, 1993:73–87. Goldstein MC, Tsarong P, Beall CM. High altitude hypoxia, culture and human fecundity/fertility: a comparative study. Am Anthropol 1983; 85:28–49. Cruz-Coke R, Crisoffanini A, Aspillaga M, Biancani F. Evolutionary forces in human populations in an environmental gradient in Arica, Chile. Hum Biol 1966; 4:421– 438. Pan American Health Organization. Health Conditions in the Americas. Scientific publication No. 549. Washington, DC: World Health Organization, 1994. Gonzales GF, Guerra-Garcia R. Caracteristicas hormonales y antropometricas del embarazo y del recien nacido en la altura. In: Gonzales GF, ed. Reproduccion Humana en la Altura. Lima, Peru: Consejo Nacional de Ciencia y Tecnologica, 1993:125– 141. Sobrevilla LA, Cassinelli MT, Carcelen A, Malaga JM. Human fetal and maternal oxygen tension and acid-base status during delivery at high altitude. Am J Obstet Gynecol 1971; 111:1111–1118. Zamudio S, Leslie K, White M, Hagerman D, Moore LG. Low serum estradiol and high serum progesterone concentrations characterize hypertensive pregnancies at high altitude. J Soc Gynecol Invest 1994; 1:197–205.

90

Niermeyer et al.

43. Bangham CRM, Hackett PH. Effects of high altitude on endocrine function in the sherpas of Nepal. J Endocrinol 1978; 79:147–148. 44. Yip R. Altitude and birth weight. J Pediatr 1987; 111:869–876. 45. Jensen GM, Moore LG. The effect of high altitude and other risk factors on birthweight: independent or interactive effects? Am J Publ Health 1997. 46. Lichty JL, Ting R, Bruns PD, Dyar E. Studies of babies born at high altitude. Am J Dis Child 1957; 93:666–669. 47. McCullough RE, Reeves JT, Liljegren RL. Fetal growth retardation and increased infant mortality at high altitude. Obstet Gynecol Surv 1977; 32:596–598. 48. Unger C, Weiser JK, McCullough RE, Keefer S, Moore LG. Altitude, low birth weight, and infant mortality in Colorado. J Am Med Assoc 1988; 259:3427–3432. 49. Wiley AS. Neonatal size and infant mortality at high altitude in the western Himalaya. Am J Phys Anthropol 1994; 94:289–305. 50. Zamudio S, Droma T, Norkyel KY, Acharya G, Zamudio JA, Niermeyer SN, Moore LG. Protection from intrauterine growth retardation in Tibetans at high altitude. Am J Phys Anthropol 1993; 91:215–224. 51. Beall CM. Optimal birthweights in Peruvian populations at high and low altitudes. Am J Phys Anthropol 1981; 56:209–216. 52. Carmen-Torres D, Gonzales GF. Edad gestacional al parto a diferentes altitudes del Peru. In: Gonzales GF, ed. Reproduccion Humana en la Altura. Lima, Peru: Consejo Nacional de Ciencia y Tecnologica, 1993:143–151. 53. Haas LD, Frongillo EJ, Stepcik, C., Beard J, Hurtado L. Altitude, ethnic and sex differences in birth weight and length in Bolivia. Hum Biol 1980; 52:459–477. 54. Mazess RB. Neonatal mortality and altitude in Peru. Am J Phys Anthropol 1965; 23: 209–213. 55. Zhoma ZX, Sun SF, Zhang JG, Huang SY, Moore LG. Fetal growth and maternal oxygen supply in Tibetan and Han residents of Lhasa (3658 m). FASEB J 1989; 3: A987. 56. Smith CF. Comparative birthweight data for Sherpa women living at high and low altitudes in Nepal (abstr). Am J Phys Anthropol 1993; 16(suppl):183. 57. Nance WE. Genetic studies of the offspring of identical twins: a model for the analysis of quantitative inheritance in man. Acta Genet Med Gemellol 1976; 25:103–113. 58. Magnus P. Causes of variation in birth weight: a study of offspring of twins. Clin Genet 1984; 25:15–24. 59. Vlietinck R, Derom R, Neale MC, Maes H, van Loon H, Derom C, Thiery M. Genetic and environmental variation in the birth weight of twins. Behav Genet 1989; 19:151– 161. 60. Reshetnikova OS, Burton GJ, Milovanov AP. Effects of hypobaric hypoxia on the fetoplacental unit: the morphometric diffusing capacity of the villous membrane at high altitude. Am J Obstet Gynecol 1994; 171:1560–1565. 61. Mayhew TM, Jackson MR, Haas JD. Oxygen diffusive conductances of human placentae from term pregnancies at low high altitudes. Placenta 1990; 11:493–503. 62. Hannhart B, Pickett CK, Weil JV, Moore LG. Influence of pregnancy on ventilatory and carotid body neural output responsiveness to hypoxia in cats. J Appl Physiol 1989; 67:797–803. 63. Moore LG, McCullough RE, Weil JV. Increased HVR in pregnancy: relationship to hormonal and metabolic changes. J Appl Physiol 1987; 62:158–163. 64. Tatsumi K, Hannhart B, Moore LG. Influences of sex steroids on ventilation and venti-

The People

65.

66.

67. 68. 69.

70.

71.

72.

73.

74.

75. 76. 77.

78.

79. 80. 81. 82.

91

latory control. In: Dempsey JA, Pack AI, eds. Regulation of Breathing. 2d ed. New York: Marcel Dekker, 1995:829–864. Zamudio S, Palmer SK, Dahms TE, Berman JC, Young DA, Moore LG. Alterations in uteroplacental blood flow precede hypertension in preeclampsia at high altitude [see comments]. J Appl Physiol 1995; 79:15–22. Zamudio S, Palmer SK, Dahms TE, Berman JC, McCullough RG, McCullough RE, Moore LG. Blood volume expansion, preeclampsia, and infant birth weight at high altitude. J Appl Physiol 1993; 75:1566–1573. Goodlin RC. Care of the Fetus. Masson Publishing USA, Inc., 1979. Moore LG. Maternal O2 transport and fetal growth in Colorado, Peru and Tibet highaltitude residents. Am J Hum Biol 1990; 2:627–637. Zamudio S, Sun S, Moore LG. Ventilatory characteristics of Peruvian and Tibetan women and men native to high altitude. In: Leon-Velarde F, Arrigui A, eds. Hipoxia: Investgaciones Basicas y Clinicas-Homenaje a Carlos Monge Cassinelli. Lima, Peru: Universidad Perunana Cayetano Heredia, 1993:55–70. Zamudio S, Palmer SK, Stamm E, Coffin C, Moore LG. Uterine blood flow at high altitude. In: Sutton JR, Houston CS, eds. Hypoxia and the Brain. Burlington, VT: Queen City Press, 1995:112–124. Circulatory function during pregnancy: man. In: Altman PL, Dittmer DS, eds. Biological Handbooks, Respiration and Circulation. Bethesda, MD: Federation of American Societies for Experimental Biology, 1971:443–449. Moore LG, Brodeur P, Chumbe O, D’Brot J, Hofmeister S, Monge C. Maternal hypoxic ventilatory response, ventilation, and infant birth weight at 4,300 m. J Appl Physiol 1986; 60:1401–1406. Moore LG, Hershey DW, Jahnigen D, Bowes W Jr. The incidence of pregnancyinduced hypertension is increased among Colorado residents at high altitude. Am J Obstet Gynecol 1982; 144:423–429. Moore LG, Jahnigen D, Rounds SS, Reeves JT, Grover RF. Maternal hyperventilation helps preserve arterial oxygenation during high-altitude pregnancy. J Appl Physiol 1982; 52:690–694. Kitanaka T, Gilbert RD, Longo LD. Maternal responses to long-term hypoxia in sheep. Am J Physiol 1989; 256:R1340–R1347. Jacobs R, Robinson JS, Owens JA, Falconer J, Webster MED. The effect of prolonged hypobaric hypoxia on growth of fetal sheep. J Dev Physiol 1988; 10:97–112. Makowski EL, Battaglia FC, Meschia G, Behrman RE, Schreuffer J, Seeds AE, Bruns PD. Effect of maternal exposure to high altitude upon fetal oxygenation. Am J Obstet Gynecol 1968; 100:852–861. Palmer SK, Zamudio S, Coffin C, Parker S, Stamm E, Moore LG. Quantitative estimation of human uterine artery blood flow and pelvic blood flow redistribution in pregnancy [review]. Obstet Gynecol 1992; 80:1000–1006. Zamudio S, Palmer SK, Droma T, Stamm E, Coffin C, Moore LG. Effect of altitude on uterine artery blood flow during normal pregnancy. J Appl Physiol 1995; 79:7–14. Wilcox AJ. Birth weight and perinatal mortality: the effect of maternal smoking. Am J Epidemiol 1993; 137:1098–1104. Williams RL, Creasy RK, Cunningham GC, Hawes WE, Norris FD, Tashiro M. Fetal growth and perinatal viability in California. Obstet Gynecol 1982; 59:624–632. Karn MN, Penrose LS. Birth weight and gestational time in relation to maternal age, parity and infant survival. Eugenics 1957; 16:147–164.

92

Niermeyer et al.

83. Ales KL, Norton ME, Druzin ML. The clinical significance of predictions based on screening second trimester mean arterial pressure: adverse maternal [corrected] and infant outcomes [published erratum appears in J Clin Epidemiol 1990; 43(5):537]. J Clin Epidemiol 1990; 43:117–124. 84. Ales KL, Frayer W, Hawks G, Auld PM, Druzin ML. Development and validation of a multivariate predictor of mortality in very low birth weight. J Clin Epidemiol 1988; 41:1095–1103. 85. Lehmann DK, Mabie WC, Miller JM, Jr., Pernoll ML. The epidemiology and pathology of maternal mortality: Charity Hospital of Louisiana in New Orleans, 1965–1984. Obstet Gynecol 1987; 69:833–840. 86. Palmer SK, Moore LG, Berman JC, Cregger B, Zamudio S. Altitude and hypertensive complications of pregnancy (abstr). Am J Phys Anthropol 1994; 18:158. 87. Zamudio S, Palmer SK, Regensteiner JG, Moore LG. High altitude and hypertension during pregnancy. Am J Hum Biol 1995; 7:182–192. 88. Lopez-Jaramillo P, y de Felix M. Uso de calcio en la prevencion inducida por el embarazo. Bol Sanit Panam 1991; 110. 89. Quintana D, Briceno G, Axel E. Evaluacion del desprendimento prematuro de placenta en la altura. In: First World Congress of High-Altitude Medicine and Physiology. La Paz, Bolivia: 1994:87. 90. Lunell NO, Sarby B, Lewander R, Nylund L. Comparison of uteroplacental blood flow in normal and in intrauterine growth-retarded pregnancy. Measurements with Indium113m and a computer-linked gammacamera. Gynecol Obstet Invest 1979; 10:106– 118. 91. Lunell NO, Lewander R, Mamoun I, Nylund L, Sarby S, Thornstrom S. Uteroplacental blood flow in pregnancy induced hypertension. Scand J Clin Lab Invest Suppl 1984; 169:28–35. 92. Robertson WB, Khong TY, Brosens I, De Wolf F, Sheppard BL, Bonnar J. The placental bed biopsy: review from three European centers [review]. Am J Obstet Gynecol 1986; 155:401–412. 93. Genbacev O, Joslin R, Damsky C, Polliotti B, Fisher S. Hypoxia alters early gestation human cytotrophoblast differentiation/invasion in vitro and models the placental defects that occur in preeclampsia. J Clin Invest 1996; 97:540–550. 94. Mok JY, McLaughlin FJ, Pintar M, Hak H, Amaro-Galvez R, Levison H. Transcutaneous monitoring of oxygenation: What is normal? J Pediatr 1986; 108:365–371. 95. Thilo EH, Park-Moore B, Berman ER, Carson BS. Oxygen saturation by pulse oximetry in healthy infants at an altitude of 1610 m (5280 ft). What is normal? [see comments]. Am J Dis Child 1991; 145:1137–1140. 96. Niermeyer S, Shaffer EM, Thilo E, Corbin C, Moore LG. Arterial oxygenation and pulmonary arterial pressure in healthy neonates and infants at high altitude. J Pediatr 1993; 123:767–772. 97. Reuland DS, Steinhoff MC, Gilman RH, Bara M, Olivares EG, Jabra A, Finkelstein D. Prevalence and prediction of hypoxemia in children with respiratory infections in the Peruvian Andes. J Pediatr 1991; 119:900–906. 98. Gamboa R, Marticorena E. Pulmonary arterial pressure in newborn infants in high altitude [in Spanish]. Arch Inst Biol Andina 1971; 4:55–66. 99. Sime F, Banchero N, Penaloza D, Gamboa R, Cruz J, Marticorena E. Pulmonary hypertension in children born and living at high altitudes. Am J Cardiol 1963; 11:143– 149.

The People

93

100. Kelly DH, Stellwagen LM, Kaitz E, Shannon DC. Apnea and periodic breathing in normal full-term infants during the first twelve months. Pediatr Pulmonol 1985; 1: 215–219. 101. Glotzbach SF, Ariagno RL. Periodic Breathing. In: Beckerman RC, Brouillette RT, Hunt CE, eds. Respiratory Control Disorders in Infants and Children. Baltimore, MD: Williams & Wilkins, 1992:142–160. 102. Matsuoka T, Mortola JP. Effects of hypoxia and hypercapnia on the Hering-Breuer reflex of the conscious newborn rat. J Appl Physiol 1995; 78:5–11. 103. Lahiri S, Brody JS, Motoyama EK, Velasquez TM. Regulation of breathing in newborns at high altitude. J Appl Physiol 1978; 44:673–678. 104. Barrington KJ, Finer NN, Wilkinson MH. Progressive shortening of the periodic breathing cycle duration in normal infants. Pediatr Res 1987; 21:247–251. 105. Lubchenco LO, Ashby BL, Markarian M. Periodic breathing in newborn infants in Denver and Leadville, Colorado (abstr). Soc Pediatr Res 1964; 50. 106. Deming J, Washburn AH. Respiration in infancy. I. A method of studying rates, volume and character of respiration with preliminary report of results. Am J Dis Child 1935; 49:108–124. 107. Emmanouilides GC, Moss AJ, Duffie ER, Adams FH. Pulmonary artery pressure changes in human newborn infants from birth to 3 days of age. J Pediatr 1964; 65: 327–333. 108. Arias-Stella J, Saldana M. The muscular pulmonary arteries in people living at high altitudes. Fifth Annual Conference on Research in Emphysema, Aspen CO, 1962 (abstract). 109. Alzamora-Castro V, Battilana G, Abugattas R, Sialer S. Patent ductus arteriosus and high altitude. Am J Cardiol 1960; 5:761. 110. Gamboa R, Marticorena E. The ductus arteriosus in the newborn infant at high altitude. Vasa 1972; 1:192–195. 111. Miao CY, Zuberbuhler JS, Zuberbuhler JR. Prevalence of congenital cardiac anomalies at high altitude. J Am Coll Cardiol 1988; 12:224–228. 112. Sui GJ, Liu YH, Cheng XS, Anand IS, Harris E, Harris P, Heath D. Subacute infantile mountain sickness. J Pathol 1988; 155:161–170. 113. Khoury GH, Hawes CR. Primary pulmonary hypertension in children living at high altitude. J Pediatr 1963; 62:177–185. 114. Haas JD. Prenatal and infant growth and development. In: Baker PT, Little MA, eds. Man in the Andes. A Multidisciplinary Study of High-Altitude Quechua. Stroudsburg, PA: Dowden, Hutchinson and Ross, Inc., 1976:161–178. 115. Leonard WR, DeWalt KM, Stansbury JP, McCaston MK. Growth differences between children of highland and coastal Ecuador. Am J Phys Anthropol 1995; 98:47–57. 116. Beall CM, Baker PT, Baker TS, Haas JD. The effects of high altitude on adolescent growth in southern Peruvian Amerindians. Hum Biol 1977; 49:109–124. 117. Beall CM. A comparison of chest morphology in high altitude Asian and Andean populations. Hum Biol 1982; 54:145–163. 118. Frisancho AR, Baker PT. Altitude and growth: a study of the patterns of physical growth of a high altitude Peruvian Quechua population. Am J Phys Anthropol 1970; 32:279–292. 119. Frisancho AR. Human growth and pulmonary function of a high altitude Peruvian Quechua population. Hum Biol 1969; 41:365–379. 120. Mueller WH, Yen F, Rothhammer F, Schull WJ. Multinational Andean genetic and

94

121.

122. 123.

124.

125. 126. 127.

128. 129.

130.

131.

132. 133.

134.

135.

136.

Niermeyer et al. health program: VII. Lung function and physical growth—multivariate analyses in high- and low-altitude populations. Aviat Space Environ Med 1978; 49:1188– 1196. Mueller WH, Schull VN, Schull WJ, Soto P, Rothhammer F. A multinational Andean genetic and health program: growth and development in an hypoxic environment. Ann Hum Biol 1978; 5:329–352. Leatherman TL, Carey JW, Thomas RB. Socioeconomic change and patterns of growth in the Andes. Am J Phys Anthropol 1995; 97:307–321. Obert P, Fellmann N, Falgairette G, Bedu M, Van Praagh E, Kemper H, Post B, Spielvogel H, Tellez V, Qintela A. The importance of socioeconomic and nutritional conditions rather than altitude on the physical growth of prepubertal Andean highland boys. Ann Hum Biol 1994; 21:145–154. Weitz CA, Garruto RM, Zhang Z, Chen JH, Liu RL, Liu JZ, Yuan KF, He X, Chin CT. Comparison of morphological and physiological characteristics of Han, Hui and Tibetans born and raised at high altitude in western China (abstr). Am J Hum Biol 1995; 7:Abs 155. Malik SL, Pandey AK. Respiratory adaptation to high altitude in adolescent Bod girls of the Western Himalayas. Ann Hum Biol 1993; 20:575–581. Pawson IG. Growth characteristics of populations of Tibetan origin in Nepal. Am J Phys Anthropol 1977; 47:473–482. Melton TW. Comparison of growth and development in two Peruvian populations of high altitude ancestry. In: Eckhardt RB, Melton TW, eds. Population Studies on Human Adaptation and Evolution in the Peruvian Andes. 1992:192–241. Greksa LP, Spielvogel H, Caceres E. Total lung capacity in young highlanders of Aymara ancestry. Am J Phys Anthropol 1994; 94:477–486. Kramer AA. Heritability estimates of thoracic skeletal dimensions of a high-altitude Peruvian population. In: Eckhardt RB, Melton TW, eds. Population Studies on Human Adaptation and Evolution in the Peruvian Andes. 1992:25–49. Ahn YI. Heritability estimates of four anthropometric measurements on the thorax in a high altitude Peruvian population. In: Eckhardt RB, Melton TW, eds. Population Studies on Human Adaptation and Evolution in the Peruvian Andes. 1992:113–137. Greksa LP, Spielvogel H, Paz-Zamora M, Caceres E, Paredes-Fernandez L. Effect of altitude on the lung function of high altitude residents of European ancestry. Am J Phys Anthropol 1988; 75:77–85. Malhotra R. Thoracic adaptation to high altitude. Anthropol Anzeig 1986; 44:355– 359. Johnson RL, Jr., Cassidy SS, Grover RF, Schutte JE, Epstein RH. Functional capacities of lungs and thorax in beagles after prolonged residence at 3,100 m. J Appl Physiol 1985; 59:1773–1782. Frisancho AR, Frisancho HG, Albalak R, Villain M, Vargas E, Soria R. Developmental, genetic, and environmental components of lung volumes at high altitude. Am J Hum Biol 1997; 9:191–203. Greksa LP, Spielvogel H, Paredes-Fernandez L. Maximal exercise capacity in adolescent European and Amerindian high-altitude natives. Am J Phys Anthropol 1985; 67: 209–216. Obert P, Bedu M, Fellmann N, Falgairette G, Beaune B, Quintela A, Van Praagh E, Spielvogel H, Kemper H, Post B. Effect of chronic hypoxia and socioeconomic status on VO2max and anaerobic power of Bolivian boys. J Appl Physiol 1993; 74:888–896.

The People

95

137. Blonc S, Fellmann N, Bedu M, Falgairette G, Dejonge R, Obert P, Beaune B, Spielvogel H, Tellez W, Quintela A, Sanmiguel JL, Coudert J. Effect of altitude and socioeconomic status on VO2max and anaerobic power in prepubertal Bolivian girls. J Appl Physiol 1996; 80:2002–2008. 138. Frisancho AR, Frisancho HG, Milotich M, Brutsaert T, Albalak R, Spielvogel H, Villena M, Vargas E, Soria R. Developmental, genetic, and environmental components of aerobic capacity at high altitude. Am J Phys Anthropol 1995; 96:431–442. 139. Scoggin CH, Hyers TM, Reeves JT, Grover RF. High-altitude pulmonary edema in the children and young adults of Leadville, Colorado. N Engl J Med 1977; 297:1269– 1272. 140. Fasules JW, Wiggins JW, Wolfe RR. Increased lung vasoreactivity in children from Leadville, Colorado, after recovery from high-altitude pulmonary edema. Circulation 1985; 72:957–962. 141. Ucros S, Posada E, Llano G. High altitude pulmonary edema in children living in Bogota. In: First World Congress of High-Altitude Medicine and Physiology. La Paz, Boliva: 1994:60. 142. Wu TY, Ki WS, Li Y. High altitude disease on the Tibetan plateau. In: First World Congress of High-Altitude Medicine and Physiology. La Paz, Bolivia, 1994:55. 143. Luerssen JS. Landlessness, health and the failures of reform in the Peruvian highlands. Human Organ 1994; 53:380–387. 144. Matheson GO, Allen PS, Ellinger DC, Hanstock CC, Gheorghiu D, McKenzie DC, Stanley C, Parkhouse WS, Hochachka PW. Skeletal muscle metabolism and work capacity: a 31P-NMR study of Andean natives and lowlanders. J Appl Physiol 1991; 70:1963–1976. 145. Sun SF, Droma TS, Zhang JG, Tao JX, Huang SY, McCullough RG, McCullough RE, Reeves CS, Reeves JT, Moore LG. Greater maximal O2 uptakes and vital capacities in Tibetan than Han residents of Lhasa. Respir Physiol 1990; 79:151–161. 146. Frisancho AR, Martinez C, Velasquez T, Sanchez J, Montoye H. Influence of developmental adaptation on aerobic capacity at high altitude. J Appl Physiol 1973; 34:176– 180. 147. Favier R, Spielvogel H, Desplanches D, Ferretti G, Kayser B, Hoppeler H. Maximal exercise performance in chronic hypoxia and acute normoxia in high-altitude natives. J Appl Physiol 1995; 78:1868–1874. 148. Hochachka PW, Stanley C, Matheson GO, McKenzie DC, Allen PS, Parkhouse WS. Metabolic and work efficiencies during exercise in Andean natives. J Appl Physiol 1991; 70:1720–1730. 149. Ge RL, Chen QH, Wang LH, Gen D, Yang P, Kubo K, Fujimoto K, Matsuzawa Y, Yoshimura K, Takeoka M. Higher exercise performance and lower VO2max in Tibetan than Han residents at 4,700 m altitude. J Appl Physiol 1994; 77:684–691. 150. Curran-Everett L, Zhang JG, Droma TS, Moore LG. Superior work performance in lifelong Tibetan residents of 4,400 m compared with 3,658 m (abstr). Physiologist 1992; 35. 151. Niu W, Youan W, Li B, Chen N, Song S. Effects of long-term acclimatization in lowlanders migrating to high altitude: comparison with high altitude residents. Eur J Appl Physiol 1995; 71:543–548. 152. Severinghaus JW, Bainton CR, Carcelen A. Respiratory insensitivity to hypoxia in chronically hypoxic man. Respir Physiol 1966; 1:308–334. 153. Dempsey JA, Reddan WG, Birnbaum ML, Forster HV, Thoden JS, Grover RF, Rankin

96

154.

155. 156. 157. 158.

159.

160. 161.

162.

163. 164. 165.

166. 167. 168.

169.

170.

171.

Niermeyer et al. J. Effects of acute through life-long hypoxic exposure on exercise pulmonary gas exchange. Respir Physiol 1971; 13:62–89. Weil JV, Byrne-Quinn E, Sodal IE, Filley GF, Grover RF. Acquired attenuation of chemoreceptor function in chronically hypoxic man at high altitude. J Clin Invest 1971; 50:186–195. Lahiri S. Alveolar gas pressures in man with life-time hypoxia. Respir Physiol 1968; 4:373–386. Lahiri S, Milledge JS. Acid-base in Sherpa altitude residents and lowlanders at 4880 m. Respir Physiol 1967; 2:323–334. Santolaya RB, Lahiri S, Alfaro RT, Schoene RB. Respiratory adaptation in the highest inhabitants and highest Sherpa mountaineers. Respir Physiol 1989; 77:253–262. Zhuang J, Droma T, Sun S, Janes C, McCullough RE, McCullough RG, Cymerman A, Huang SY, Reeves JT, Moore LG. Hypoxic ventilatory responsiveness in Tibetan compared with Han residents of 3,658 m. J Appl Physiol 1993; 74:303–311. Huang ZR, Zhu SC, Ba ZF, Hu ST. Ventilatory control in Tibetan highlanders. In: Liu DS, ed. Geological and Ecological Studies of Qinghai-Xizang Plateau. Beijing, PRC: Science Press, 1981:1363–1369. Hackett PH, Reeves JT, Reeves CD, Grover RF, Rennie D. Control of breathing in Sherpas at low and high altitude. J Appl Physiol 1980; 49:374–379. Beall CM, Brittenham GM, Strohl KP, Blangero J, Williams-Blangero S, Almasy LA, Decker MJ, Worthman CM, Goldstein MC, Vargas E, Villena M, Soria R, Alaracon AM, Gonzales C. Ventilation and hypoxic ventilatory response of Tibetan and Aymara high-altitude natives. Am J Phys Anthropol 1997. Huang SY, Alexander JK, Grover RF, Maher JT, McCullough RE, McCullough RG, Moore LG, Sampson JB, Weil JV, Reeves JT. Hypocapnia and sustained hypoxia blunt ventilation on arrival at high altitude. J Appl Physiol 1984; 56:602–606. Lahiri S, Kao FF, Velasquez T, Martinez C, Pezzia W. Irreversible blunted respiratory sensitivity to hypoxia in high altitude natives. Respir Physiol 1969; 6:360–374. Milledge JS, Lahiri S. Respiratory control in lowlanders and Sherpa highlanders at altitude. Respir Physiol 1967; 2:310–322. Curran LS, Zhuang J, Droma T, Land L, Moore LG. Hypoxic ventilatory responses in Tibetan residents of 4400 m compared with 3658 m. Respir Physiol 1995; 100: 223–230. Collins DD, Scoggin CH, Zwillich CW, Weil JV. Hereditary aspects of decreased hypoxic response. J Clin Invest 1978; 62:105–110. Kawakami Y, Yoshikawa T, Shida A, Asanuma Y, Murao M. Control of breathing in young twins. J Appl Physiol 1982; 52:537–542. Strohl K, Beall CM. Ventilatory response to experimental hypoxia in adult male and female natives of the Tibetan and Andean Plateaus. In: Houston C, Coates J, eds. Hypoxia 97: Women at Altitude. Toronto: B.C. Decker, Inc., 1997. Wagner PD, Gale GE, Moon RE, Torre-Bueno JR, Stolp BW, Saltzman HA. Pulmonary gas exchange in humans exercising at sea level and simulated altitude. J Appl Physiol 1986; 61:260–270. Moore LG, Curran-Everett L, Droma TS, Groves BM, McCullough RE, McCullough RG, Sun SF, Sutton JR, Zamudio S, Zhuang JG. Are Tibetans better adapted? Int J Sports Med 1992; 13(suppl 1):S86–S88. Samaja M, Veicsteinas A, Cerretelli P. Oxygen affinity of blood in altitude Sherpas. J Appl Physiol 1979; 47:337–341.

The People

97

172. Beall CM, Blangero J, Williams-Blangero S, Goldstein MC. Major gene for percent of oxygen saturation of arterial hemoglobin in Tibetan highlanders. Am J Phys Anthropol 1994; 95:271–276. 173. Beall CM, Strohl KP, Blangero J, Williams-Blangero S, Brittenham GM, Goldstein MC. Quantitive genetic analysis of arterial oxygen saturation in Tibetan highlanders. Am J Hum Biol 1997; 174. Crausman RS, Lewin JM. Pulmonary arterial hypertension in nonagenarians [letter]. J Am Geriatr Soc 1995; 43:1177. 175. Beall CM, Brittenham GM, Macuaga F, Barragan M. Variation in hemoglobin concentration among samples of high-altitude natives in the Andes and the Himalayas. Am J Hum Biol 1990; 2:639–651. 176. Winslow RM, Chapman KW, Monge CC. Ventilation and the control of erythropoiesis in high-altitude natives of Chile and Nepal. Am J Hum Biol 1990; 2:653–662. 177. Hsu SL, Marks J, Shaw JP, Tam M, Higgs DR, Shen CC, Shen CK. Structure and expression of the human theta 1 globin gene. Nature 1988; 331:94–96. 178. Frisancho AR. Origins of differences in hemoglobin concentration between Himalayan and Andean populations. Respir Physiol 1988; 72:13–18. 179. Zhuang J, Droma T, Sutton JR, McCullough RE, McCullough RG, Groves BM, Rapmund G, Janes C, Sun S, Moore LG. Autonomic regulation of heart rate response to exercise in Tibetan and Han residents of Lhasa (3,658 m). J Appl Physiol 1993; 75: 1968–1973. 180. Savard GK, Kayser B, Binzoni T. Cardiac parasympathetic control in Tibetan newcomers to altitude (5050 m). In: Sutton JR, Houston CS, eds. Hypoxia and the Brain. Burlington, VT: Queen City Press, 1995. 181. Hartley LH, Vogel JA, Cruz JC. Reduction of maximal exercise heart rate at altitude and its reversal with atropine. J Appl Physiol 1974; 36:362–365. 182. Groves BM, Droma T, Sutton JR, McCullough RG, McCullough RE, Zhuang J, Rapmund G, Sun S, Janes C, Moore LG. Minimal hypoxic pulmonary hypertension in normal Tibetans at 3,658 m. J Appl Physiol 1993; 74:312–318. 183. Hartley LH, Alexander JK, Modelski M, Grover RF. Subnormal cardiac output at rest and during exercise in residents at 3,100 m altitude. J Appl Physiol 1967; 23:839– 848. 184. Banchero N, Sime F, Penaloza D, Cruz J, Gamboa R, Marticorena E. Pulmonary pressure, cardiac output and arterial oxygen saturation during exercise at high altitude and at sea level. Circulation 1966; 33:249–262. 185. Gupta ML, Rao KS, Anand IS, Banerjee AK, Boparai MS. Lack of smooth muscle in the small pulmonary arteries of the native Ladakhi. Is the Himalayan highlander adapted? Am Rev Respir Dis 1992; 145:1201–1204. 186. Durmowicz AG, Hofmeister S, Kadyraliev TK, Aldashev AA, Stenmark KR. Functional and structural adaptation of the yak pulmonary circulation to residence at high altitude. J Appl Physiol 1993; 74:2276–2285. 187. Heath D, Williams D, Dickinson J. The pulmonary arteries of the yak. Cardiovasc Res 1984; 18:133–139. 188. Sun SF, Sui GZ, Liu YH, Cheng XS, Anand IS, Harris P, Heath D. The pulmonary circulation of the Tibetan snow pig (marmota himalaya hodgson). J Zoology 1989; 217:85–91. 189. Banchero N, Grover RF, Will JA. High altitude-induced pulmonary arterial hypertension in the Ilama (Lama glama). Am J Physiol 1971; 220:422–427.

98

Niermeyer et al.

190. Huang SY, Sun S, Droma T, Zhuang J, Tao JX, McCullough RG, McCullough RE, Micco AJ, Reeves JT, Moore LG. Internal carotid arterial flow velocity during exercise in Tibetan and Han residents of Lhasa (3,658 m). J Appl Physiol 1992; 73:2638– 2642. 191. Garrido E, Segura R, Capdevilla A, Pujol J, Javierre C, Ventura JL. Are Himalayan Sherpas better protected against brain damage associated with extreme altitude climbs? Clin Sci 1996; 90:81–85. 192. Kayser B, Hoppeler H, Claassen H, Cerretelli P. Muscle structure and performance capacity of Himalayan Sherpas. J Appl Physiol 1991; 70:1938–1942. 193. Kayser B, Hoppeler H, Desplanches D, Marconi C, Broers B, Cerretelli P. Muscle ultrastructure and biochemistry of lowland Tibetans. J Appl Physiol 1996; 81:419– 425. 194. Martin BJ, Weil JV, Sparks KE, McCullough RE, Grover RF. Exercise ventilation correlates positively with ventilatory chemoresponsiveness. J Appl Physiol 1978; 45: 557–564. 195. Monge-C C, Arregui A, Leon-Velarde F. Pathophysiology and epidemiology of chronic mountain sickness. Int J Sports Med 1992; 13(suppl 1):S79–S81. 196. Kryger M, McCullough RE, Collins D, Scoggin CH, Weil JV, Grover RF. Treatment of excessive polycythemia of high altitude with respiratory stimulant drugs. Am Rev Respir Dis 1978; 117:455–464. 197. Leon-Velarde F, Ramos MA, Hernandez JA, Deidiaquez D, Munoz LS, Gaffo A, Cordova S, Durand D, Monge C. The role of menopause in the development of chronic mountain sickness. Am J Physiol 1997; 41:R90–R94. 198. Sime F, Monge C, Whittembury J. Age as a cause of chronic mountain sickness (Monge’s disease). Int J Biometeorol 1975; 19:93–98. 199. Leon-Velarde F, Arregui A, Monge C, Ruiz y Ruiz H. Aging at high altitudes and the risk of chronic mountain sickness. J Wilderness Med 1993; 4:183–188. 200. Sun SF, Huang SY, Zhang JG, Droma TS, Banden G, McCullough RE, McCullough RG, Cymerman A, Reeves JT, Moore LG. Decreased ventilation and hypoxic ventilatory responsiveness are not reversed by naloxone in Lhasa residents with chronic mountain sickness. Am Rev Respir Dis 1990; 142:1294–1300. 201. Grover RF, Reeves JT, Grover EB, Leathers JE. Muscular exercise in young men native to 3,100 m altitude. J Appl Physiol 1967; 22:555–564. 202. Kryger M, McCullough R, Doekel R, Collins D, Weil JV, Grover RF. Excessive polycythemia of high altitude: role of ventilatory drive and lung disease. Am Rev Respir Dis 1978; 118:659–666. 203. Pei SX, Chen XJ, Si Ren BZ, Liu YH, Cheng XS, Harris EM, Anand IS, Harris PC. Chronic mountain sickness in Tibet. Q J Med 1989; 71:555–574. 204. Leon-Velarde F, Monge CC, Vidal A, Carcagno M, Criscuolo M, Bozzini CE. Serum immunoreactive erythropoietin in high altitude natives with and without excessive erythrocytosis. Exp Hematol 1991; 19:257–260. 205. Kryger M, Glas R, Jackson D, McCullough RE, Scoggin C, Grover RF, Weil JV. Impaired oxygenation during sleep in excessive polycythemia of high altitude: improvement with respiratory stimulation. Sleep 1978; 1:3–17. 206. Sun SF, Oliver-Pickett C, Ping Y, Micco AJ, Droma TS, Zamudio S, Zhang JG, Huang SY, McCullough RG, Cymerman A, Moore LG. Breathing and brain blood flow during sleep in patients with chronic mountain sickness. J Appl Physiol 1996; 81:611–618.

The People

99

207. Monge-C C, Leon-Velarde F, Arregui A. Increasing prevalence of excessive erythrocytosis with age among healthy high-altitude miners (letter). N Engl J Med 1989; 321: 1271. 208. Xie CF, Pei SX. Some physiological data of sojourners and native highlanders at three different altitudes in Xizang. In: Liu DS, ed. Geological and Ecological Studies of Qinghai-Xizang Plateau. New York: Gordon and Breach, 1981:1449–1452. 209. Wu TY, Zhang Q, Jin B, Xu F, Cheng Q, Wan X. Chronic mountain sickness (Monge’s Disease): an observation in Qinghai-Tibet plateau. In: Ueda G, ed. High Altitude Medicine. Matsumoto, Japan: Shinshu University Press, 1992:314–324. 210. Ruiz L, Penaloza D. Altitude and hypertension. Mayo Clin Proc 1977; 52:442–445. 211. Sun SF. Epidemiology of hypertension on the Tibetan plateau. Hum Biol 1986; 58: 507–515. 212. Marticorena E, Ruiz L, Severino J, Galvez J, Penaloza D. Systemic blood pressure in white men born at sea level: changes after long residence at high altitudes. Am J Cardiol 1969; 23:364–368. 213. Wolfel EE, Selland MA, Mazzeo RS, Reeves JT. Systemic hypertension at 4,300 m is related to sympathoadrenal activity. J Appl Physiol 1994; 76:1643–1650. 214. Moore LG, Rohr AL, Maisenbach JK, Reeves JT. Emphysema mortality is increased in Colorado residents at high altitude. Am Rev Respir Dis 1982; 126:225–228. 215. Renzetti AD, McClement JH, Litt BD. The Veteran’s Administration Cooperative Study of Pulmonary Function III. Mortality in relation to respiratory function in chronic obstructive pulmonary disease. Am J Med 1966; 41:115–129. 216. Regensteiner JG, Moore LG. Migration of the elderly from high altitudes in Colorado. J Am Med Assoc 1985; 253:3124–3128. 217. Giles RE, Blanc H, Can HM, Wallace DC. Maternal inheritance of human mitochondrial DNA. Proc Natl Acad Sci USA 1980; 77:6715–6719. 218. Torroni A, Chen YS, Semino O, Santachiara-Beneceretti AS, Scott CR, Lott MT, Winter M, Wallace DC. mtDNA and Y-chromosome polymorphisms in four Native American populations from southern Mexico [see comments]. Am J Hum Genet 1994; 54: 303–318. 219. Gould SJ, Lewontin RC. The Spandrels of San Marcos and the Panglossian Paradigm: A Critique of the Adaptationist Program. In: Sober E, ed. Conceptual Issues in Evolutionary Biology. Cambridge, MA: MIT Press, 1979. 220. Lewontin RC. Adaptation. Sci Am 1978; 239:212–8, 220, 222. 221. Weiss KM. Genetic Variation and Human Disease: Principles and Evolutionary Approaches. Cambridge, MA: Cambridge University Press, 1993. 222. Moore LG, Zamudio S, Curran-Everett L, Torroni A, Jorde LB, Shohet RV, Thupten, Drolkar T. Genetic adaptation to high altitude. In: Wood SC, Roach RC, eds. Sports and Exercise Medicine. New York: Marcel Dekker, 1994:225–262. 223. Cerny FC, Dempsey JA, Reddan WG. Pulmonary gas exchange in non-native residents of high altitude. J Clin Invest 1973; 52:2993–2999. 224. DeGraff AC, Grover RF, Johnson RL,Jr., Hammond MW,Jr., Miller JM. Diffusing capacity of the lung in Caucasians native to 3,100 m. J Appl Physiol 1970; 29:71– 76. 225. Hurtado A. Animals in high altitudes: resident man. In: Handbook of Physiology. Section 4: Adaptation to the Environment. Washington, DC: American Physiological Society, 1964:843–860. 226. Droma T, McCullough RG, McCullough RE, Zhuang JG, Cymerman A, Sun SF, Sut-

100

227. 228. 229.

230. 231.

232.

233.

234.

235.

236.

Niermeyer et al. ton JR, Moore LG. Increased vital and total lung capacities in Tibetan compared to Han residents of Lhasa (3,658 m). Am J Phys Anthropol 1991; 86:341–351. Vogel JA, Patton JF, Mello RP, Daniels WL. An analysis of aerobic capacity in a large United States population. J Appl Physiol 1986; 60:494–500. Mazess RB. Exercise performance at high altitude in Peru. Fed Proc 1969; 28:1301– 1306. Kollias J, Buskirk ER, Akers RF, Prokop EK, Baker PT, Picon-Reategui E. Work capacity of long-time residents and newcomers to altitude. J Appl Physiol 1968; 24: 792–799. Vogel JA, Hartley LH, Cruz JC. Cardiac output during exercise in altitude natives at sea level and high altitude. J Appl Physiol 1974; 36:173–176. Elsener RW, Bolstad A, Forno C. Maximum oxygen consumption of Peruvian Indians native to high altitude. In: Weihe WH, ed. Physiological Effects of High Altitude. New York: Macmillan, 1964:217–233. Schoene RB, Roach RC, Lahiri S, Peters RM,Jr., Hackett PH, Santolaya R. Increased diffusion capacity maintains arterial saturation during exercise in the Quechua Indians of Chilean Altiplano. Am J Hum Biol 1990; 2:663–668. Zhuang J, Droma T, Sutton JR, Groves BM, McCullough RE, McCullough RG, Sun S, Moore LG. Smaller alveolar-arterial O2 gradients in Tibetan than Han residents of Lhasa (3658 m). Respir Physiol 1996; 103:75–82. Bebout DE, Story D, Roca J, Hogan MC, Poole DC, Gonzalez-Camarena R, Ueno O, Haab P, Wagner PD. Effects of altitude acclimatization on pulmonary gas exchange during exercise. J Appl Physiol 1989; 67:2286–2295. Lockhart A, Zelter M, Mensch-Dechene J, Antezana G, Paz-Zamora M, Vargas E, Coudert J. Pressure-flow-volume relationships in pulmonary circulation of normal highlanders. J Appl Physiol 1976; 41:449–456. Sime F, Penaloza D, Ruiz L. Bradycardia, increased cardiac output, and reversal of pulmonary hypertension in altitude natives living at sea level. Br Heart J 1971; 33: 647–657.

4 Cellular and Molecular Mechanisms of O2 Sensing with Special Reference to the Carotid Body

SUKHAMAY LAHIRI

NEIL S. CHERNIACK

University of Pennsylvania School of Medicine Philadelphia, Pennsylvania

UMDNJ–New Jersey Medical School Newark, New Jersey

I.

Introduction

A decrease in the O2 in the air brings into play an array of compensatory mechanisms, which include an increase in ventilation, changes in the distribution of blood flow to organs in the body, an increase in the O2-carrying capacity of the blood, and shifts in metabolic pathways. The exact configuration of the response varies with the severity of hypoxia, with its duration, and with the maturity of the human or animal exposed to hypoxia. Even the intensity of a particular response waxes and wanes with time. For example, the initial increase in breathing caused by hypoxic stimulation of arterial chemoreceptors diminishes in a few minutes because of the release of neurotransmitters in the brain that inhibit respiration, but then ventilation increases if hypoxia is prolonged sufficiently, producing the phenomenon of acclimatization seen with adjustment at altitude. The defense against hypoxia occurs at different levels. The arterial chemoreceptors are the first line of defense against hypoxia. They are responsible for the rapidly augmented breathing that occurs even with relatively small decreases in Po2. The rise in Po2 that occurs with greater ventilation feedback helps adjust the arterial chemoreceptor signal to appropriate levels. This ventilatory increase is supported 101

102

Lahiri and Cherniack

by changes in the circulation, constriction of the pulmonary vasculature, and dilation in the systemic circulation. Later the formation of erythropoetin by renal tissue in the adult and by the liver in the immature increases the ability of the blood to transport O2 to the tissues. In addition, all cells that consume oxygen probably have intracellular compensatory mechanisms that are called into play when they suffer from sufficiently severe and/or prolonged hypoxia. Much current research is devoted to identifying the O2-sensing molecules in the arterial chemoreceptor and other tissues and organs that recognize deficiencies in O2 supply and initiate compensatory processes. Studies in the culture of model cell lines derived from tumor tissue suggest that the rapid response to hypoxia involves molecules in either the cell membrane or the mitochondria. In either case a heme compound seems to be involved in signal recognition. Superoxide anions like peroxide are produced by intracellular oxidative processes. These naturally occurring free radicals can damage cells when they are present in excess, but recently it was proposed that they participate in O2 sensing. Ultimately, compensation involves the synthesis of increased amounts of existing or entirely new proteins caused by activation and inactivation of specific genes. The molecular mechanisms responsible for O2 sensing are probably somewhat different in excitable tissues like the carotid body than in nonexcitable tissues like the erythropoietin-producing cells of the kidney. The ability of cells to respond to hypoxia may also vary with age, surroundings, and the level of metabolic activity. Although many exciting insights into the mechanisms of O2 sensing and its cellular consequences have been made, important details are missing. Even more important, findings in cell lines have not as yet been related to the physiological sequence of events that occurs in the whole animal faced with hypoxia. This chapter focuses on the arterial chemoreceptors and reviews data and theories concerning cellular mechanisms of O2 sensing in the carotid body and in erythropoietin-producing tissues and, to round out the picture, in vascular smooth muscle and neurons as well. The complex cellular and systemic changes that occur with prolonged data are described in later sections. Finally, we summarize the new studies on altered gene expression triggered by hypoxia in model cell systems.

II. Physiological Clues to the Identity of O2-Sensing Molecules Although major advances have been made in understanding the biological response to hypoxia, the molecular mechanisms involved in the very first step of O2 sensing remain elusive. Although physiological responses offer important clues, tissues differ in their speed of response to hypoxia. The arterial chemoreceptors respond most quickly, followed by the smooth muscles of the pulmonary arteries, which respond in only a few seconds. Erythropoietin-producing cells require more prolonged hypoxia before reacting. A quick response probably involves only membranes and cytoplasmic pro-

O2 Sensing

103

cesses and does not require the synthesis of new protein. If hypoxia is unrelieved for hours, gene activity can be altered, but in some genes changes can begin in minutes. It would be intellectually quite satisfying if a single molecule or class of molecules was responsible for O2 sensing in all tissues. Heme proteins do appear to be involved in many tissues, and specific heme proteins, such as the cytochromes, have been suggested as the molecular sensors of O2 (1–5). Since different potential O2-sensing molecules (such as cytochrome oxidase) are transformed from fully oxidized to fully reduced states over different ranges of Po2, tissue levels of O2 could be another important clue as to the nature of the actual molecular sensor. In the case of cytochrome oxidase, full oxygenation is achieved at rather low Po2 levels—⬍10 torr (1–5). It is important to remember that the critical Po2 is at the site of the molecule and not arterial or venous Po2. It is possible that there are different molecules involved in O2 sensing which have different Po2 response ranges, perhaps organized in some sort of a hierarchical structure, allowing the organism to respond in a graded fashion to Po2 changes of different duration and severity. Also, the effects of hypoxia may not be direct but rather involve an altered level of some other molecule (e.g., H2O2) whose concentration depends on the Po2 level.

III. Tissue PO2 and O2 Sensing Very low levels of tissue Po2 are probably required to stimulate erythropoietin or cause pulmonary artery vasoconstriction. For erythropoietin production, incubation of hepatic cells (Hep3B) with 7 torr Po2 for 24 hours is necessary (6). A systemic arterial Po2 below 60–80 torr is needed to raise pulmonary arterial pressure, but the tissue Po2 in pulmonary smooth muscle is likely to be far less (7). Tissue Po2 was found to be less than 10 torr when arterial segments were perfused with solutions containing Po2 at 100 torr (8). At lower perfusate O2 tension, tissue levels would be even lower and hence compatible with cytochrome oxidase being the molecular sensor (1–5). However, the tissue Po2 at which carotid body stimulation occurs is controversial. Arterial Po2 of the order of 60–70 torr stimulates carotid body nerve activity (5). Acker et al. (9) found the tissue Po2 in the carotid body to be 7–20 torr at an arterial Po2 of 100. But Whalen and Nair (10) and Buerk et al. (11) reported a value of 70–80 torr in carotid body tissue when the perfusate Po2 was 100–110 torr. At a perfusate Po2 of 60–70, the carotid body tissue Po2 would be near 2–3 Torr according to Acker, whereas according to Whalen and Nair it would be around 30 torr [and perhaps even higher, since in carotid body, O2 consumption decreases with hypoxia (9,10)]. Assuming the Vo2 of the carotid body equals 1.5 mL/min/100 g, flow (Q) equals 1.5 L/min/100 g and the (a-v) O2 difference is 1.0 mL/L, Gonzalez et al.

104

Lahiri and Cherniack

(12) calculated the venous Po2 of the carotid body to be 90 at an arterial Po2 of 100 torr. Since tissue Po2 of the carotid body is about the same whether it is perfused with blood or cell-free fluids (10) and the relationships between Po2 and chemosensory activity were also found to be the same with either type of perfusate (10,13), the influence of red blood cells on tissue Po2 would seem to be minimal. In an in vitro carotid body preparation, Fay (14) found an (a-v) O2 difference of 10 torr when the Po2 of the inflowing perfusate was 100 torr, so that the effluent Po2 was 90 torr. This suggests that tissue Po2 is relatively high even when hypoxia produces obvious ventilatory stimulation (which occurs at an arterial Po2 of about 60 torr). However, Lahiri et al., using phosphorescence quenching by O2, found that the microvascular Po2 in the carotid body was 40–50 torr in vivo when the perfusate Po2 was 100 (5) and 23 torr in vitro when the perfusate Po2 was about 80 torr. At a lower arterial Po2 of 60 torr, the microvascular Po2 would be even less. After a correction for the gradient required for O2 diffusion, the tissue Po2 would be less than 10 torr (a level compatible with cytochrome oxidase being the O2 sensor) (15). The complexity of the regulation of carotid body flow and the geometry of its circulation (15,16) further complicates the assessment in vivo of tissue levels of Po2 in the carotid body and of determining how Po2 is coupled to chemosensory discharge. The microcirculation of the carotid body is very sensitive to various agents. Vasodilators decreased the O2 (a-v) Po2 difference to 5 torr from the normal 10 torr (16). Hemorrhage is also known to excite the chemosensory discharge (17), not by decreasing tissue Po2 (which seems to be independent of red blood cell numbers), but by vasoconstriction.

IV. The Arterial Chemoreceptors—Carotid and Aortic Bodies O2 sensing in higher organisms occurs at the carotid body and aortic bodies, and the information is sent to the brain stem, producing an array of reflex actions, which include breathing. The aortic body and particularly the carotid body also detect changes in pH and Pco2 (18–24), but this aspect is not unique to the chemoreceptors. Other receptors detect changes in Pco2 and acidity (25). Not only are the transmission pathways different, but carotid and aortic bodies, unlike taste buds, have some special mechanism for detecting O2 changes. Carotid and aortic bodies are located near the carotid artery bifurcation and aortic arch, where baroreceptors are also strategically located. The two types of receptors, chemo and baro, together reflexively help coordinate breathing and cardiovascular responses. The carotid body consists of 40,000–60,000 glomus cells in rats and cats, while the aortic bodies are smaller and contain fewer glomus cells (26) and have a weaker response to acid than the carotid body (27). Both have rich networks of fenestrated capillaries, around which two types of cells are clustered (Fig. 1). The major cell type, Type I glomus cells, is organized in groups of 5–10 encased by 2–3 cells of the other type (Type II). Type I cells are innervated and distinguished by the presence of dense-cored vesicles containing various neurotrans-

O2 Sensing

105

Figure 1 Electron micrograph (⫻ 10,000) of chinchilla carotid body. I and II, type I and type II cells; C, capillary; G, Golgi apparatus; GJ, gap junction; M, mitochondria; N, nerve endings; v, dense-cored vesicles.

mitters. Since neural signals are the hallmark of the chemoreceptor cell, it is only natural that these innervated Type I cells have been targeted as chemoreceptor cells (see Fig. 1). Recently, Type I cells of the carotid body (but not as yet the aortic body) have been isolated, cultured, and studied. However, investigation of the cellular responses of Type I cells have been handicapped by the paucity of these cells. Hence, studies of O2 sensing have often used other cells as substitutes. For example, neuroepithelial bodies, which are scattered in the mucosa of the airway, have been isolated and their O2-sensing properties have been examined (28). They secrete serotonin, but their physiological role is not known. The carotid body cells contain

106

Lahiri and Cherniack

tyrosine hydroxylase, which is the rate-limiting enzyme for the synthesis of dopamine and/or epinephrine (29). Tyrosine hydroxylase metabolism is O2 sensitive. The PC-12 cell line established from chromaffin cells of the adrenals (like the Type I cells) produce tyrosine hydroxylase that increases with hypoxia. PC-12 cells have been extensively used to study the effects of hypoxia under the assumption that they behave like Type I cells, but glomus cells differ from chromaffin cells in many respects (29,30). The O2 sensitivity observed in the binding of transcription factors to promoter regions of genes in the PC-12 nucleus (31) may be an example of a generalized mechanism found in many cells, but it may have nothing to do with the specific O2 sensor mechanisms that operate in the carotid body that give rise to increased ventilation. In vivo responses of the carotid body can be influenced by factors that play no role in vitro, e.g., changes in blood flow. Blood flow in the carotid body has been shown to be exquisitely sensitive to O2 and various other agents, such as nitric oxide, a vasodilator (32–36). Tissue Po2 diminishes with constant perfusate Po2, as chemosensory activity is increased by the inhibition of the enzyme nitric oxide synthase. In addition, in vivo changes in Po2 and Pco2 often occur together. Changes in Pco2 affect the activity of the carotid body even without appreciably affecting its circulation (24). There are similarities between the interaction that occurs with respect to CO2 and O2 binding to hemoglobin and the interactive effects of O2 and CO2 on chemoreceptor discharge, which support the hypothesis that some hemoglobin-like molecules in the chemoreceptor receptor cell are responsible for sensing O2 and CO2 /H⫹ (13,20). Single fiber studies have shown that carotid chemoreceptors respond to CO2 linearly, while hypoxia increases the slope and decreases the intercept of the CO2 response line (20). In general, these results conform to those obtained using multifiber preparations in vivo (37) and in vitro (38). The mechanism for the interactive effects of CO2 and hypoxia is not clear. Both are reported to affect K⫹ channels and give rise to increases in intracellular Ca2⫹ although one can abolish O2 sensitivity without appreciably altering the CO2 /H⫹ sensitivity of the carotid body (4). Changes in intracellular pH seem to affect the carotid body response to CO2. The carotid body discharge responds to CO2 with an overshoot, which disappears after inhibition of carbonic anhydrase, an enzyme present in glomus cells (39–41). The carotid body is supplied with sympathetic and parasympathetic fibers, whose activity may be modified by hypoxia and which may alter its activity. Efferent activity, in turn, changes with carotid sinus nerve activity. The targets of these efferent nerve fibers are primarily blood vessels. Petrosal ganglion cells are activated by sensory afferents, and they in turn influence the firing of the ganglion glomerular nerve through cells in sympathetic ganglia and the activity of parasympathetic fibers through cells in the nodose ganglia (34). The petrosal ganglion is not simply a relay station for sensory discharge, since

O2 Sensing

107

different ganglion cells contain different neurotransmitters, which certainly modify the output of the cells. However, exactly what role these nerve cells play is not known. Generally the autonomic innervation of the carotid body exerts an inhibitory effect on its sensory discharge (33,34), although there are reports to the contrary. By affecting blood vessel size, efferent control can alter tissue Po2 in Type I cells. Recent observations reveal that the blood vessels stain positively for nitric oxide synthase (NOS), supporting the idea that nitric oxide is a normal inhibitory transmitter involved in vasomotor control in the carotid body (33–35). Nitric oxide has also been reported to decrease sensory response in experimental situations in which its vascular effect would be minimized (34). During acute hypoxia, nitric oxide production has been found to decrease, which could be in part responsible for the increase of chemosensory discharge. Nitric oxide increases cGMP so that the rise in carotid body activity may be due to a reduction in cGMP (35). Acute hypoxia also suppresses CO production (66), which also lowers cGMP. Another endogenous gas, H2O2, is also believed to cause inhibitory effects on the carotid body and carotid body vasculature.

V.

Effects of Hypoxia on the Type I Cell Membrane

There are two leading theories of O2 sensing by the carotid body: one idea is that hypoxia acts on molecules in the cell membrane, and the other is that it acts on molecules in the mitochondria. Hypoxia depolarizes Type I cells and increases carotid body nerve discharge, perhaps by suppressing K⫹ channels in the cell membrane. The O2-sensing K⫹ channel is believed by some investigators to be the O2sensing molecule. This is based on the following lines of evidence. In the late 1980s, patch clamping began to be used to measure the ionic currents across glomus cells. Some types of K⫹ currents were found to be suppressed by hypoxia (42–50). A single K⫹ channel in an isolated membrane patch could be closed by low Po2 (48), and it was proposed that K⫹ (51) channel O2 interaction in vivo occurs at the cell membrane (48–51). The sequence proposed was that K⫹ current suppression depolarized Type I cells, opened voltage-gated Ca2⫹ channels, and Ca2⫹ entry into the cell triggered neurotransmitter release and increased neural discharge. Gonzalez et al. suggested that the O2 sensor might be a molecule independent of the K⫹ channel but connected to it and not the channel itself (51). However, no experimental verification was offered. Speculation was that the process might involve a heme-linked NADPH oxidase (52,53). Hypoxia could, for example, decrease the amount of H2O2 generated by the oxidase, resulting in increased glutathione levels, and the ratio of GSH to GSSG might decrease cGMP and increase cAMP levels. This might alter membrane proteins, which in turn could alter ion channel potency and membrane ionic conductances (52).

108

Lahiri and Cherniack

There is considerable evidence that CO2 /H⫹ stimulation of carotid chemoreceptors occurs by a mechanism separate from that responsible for O2 sensing (4), but it has been suggested that H⫹ can affect K⫹ channels and that, like hypoxia, K⫹ current suppression is responsible for H⫹ chemoreception (42). Another idea proposed by Gonzalez et al. is that a rise in intracellular [H⫹] exchanges with extracellular Na⫹, which in turn leads to increased Na⫹-Ca2⫹ exchange with Ca2⫹ entry (12). The rise in intracellular [Ca2⫹] is responsible for neurotransmitter release and neural discharge. But blocker of voltage-gated Ca2⫹ channel inhibited the neural response of the carotid body to hypoxia (53a) and hypercapnia (53b), supporting the membrane channel rather than the exchanger hypothesis. There are problems with the idea that O2 acts at the cell membrane (12). For example, voltage-dependent K⫹ current inhibition has a threshold for activation at around ⫺40 mV, whereas the resting potential of glomus cells is around ⫺55 mV. As a result, O2-sensitive K⫹ channels would be already closed in normoxic glomus cells and could not be further closed by lowering Po2. However, if normoxic glomus cells fire spontaneously, then O2-sensitive K⫹ channels can become susceptible to low Po2 inhibition, leading to more Ca2⫹ entry and more neurotransmitter release and nerve firing. Fiber and McCloskey (54) reported that ⫺20 mV or more are needed for glomus cell depolarization, although Buckler and Vaughan-Jones (55) showed that Ca2⫹ entry occurs as a result of smaller depolarization of glomus cells, around ⫺40 mV. Also, as pointed out by Lahiri (13), the range of K⫹ channel sensitivity to hypoxia (80–150 torr) is quite different from the much lower levels of Po2 (⬍ 85) needed to excite the carotid body appreciably (48–49a). Others researchers have expressed doubts as to the precise role of oxygensensitive K⫹ channels in controlling the afferent activity of the carotid body. Wyatt et al. (56) showed that the glomus cells of rats, which have blunted ventilatory responses to hypoxia, do not depolarize in response to hypoxia, although they have O2-sensitive K⫹ channels. Also, Cheng and Donnelly compared carotid sinus nerve responses with the changes in the outward K⫹ current produced by hypoxia in the rat carotid body (57). They found that hypoxia failed to alter the outward current in most cells, even when the carotid body showed a brisk nerve response. K⫹ channel blockade with tetraethylammonium (TEA) does not ablate the carotid sinus nerve response to hypoxia. Charybdotoxin (20 nM), which blocks Ca2⫹sensitive K⫹ channels, inhibited the outward K⫹ current but caused no change in nerve activity (58). Also, charybdotoxin did not alter the hypoxic response of carotid bodies (58). Buckler and Vaughan-Jones (59) found that charybdotoxin (20 nM), tetraethylammonium (10 nM), and 4 aminopyridine (AP) (1 nM) had no effect on intracellular [Ca2⫹] i. Thus, these results are not consistent with the idea that voltage-gated outward currents and nerve response of the carotid body are related. The modulation of O2sensitive K⫹ currents by hypoxia does not appear to be the primary step in initiating carotid body response to hypoxia in the carotid body (57,60–60b), but this conclusion has not been reached by all (61).

O2 Sensing

109

VI. Metabolic Effects of Hypoxia on Type I Cells Hypoxia might be sensed by its metabolic rather than its membrane effects in the carotid body. Hypoxia might interfere with ATP formation at mitochondria. The resulting decrease in cellular phosphate potential ([ATP]/[Pi] [ADP]) could influence Ca2⫹-ATPase activity, leading to intracellular [Ca2⫹] rising and the release of neurotransmitter (30,55,62). Evidence supporting this hypothesis comes from several sources. For example, oligomycin (an inhibitor of oxidative phosphorylation) blocks O2 chemoreception without preventing CO2 chemoreception (4,9,62). Oligomycin, which inhibits the formation of ATP and the energy-linked uptake of calcium by mitochondria, also blocks responses to cyanide, antimycin A, and metabolic uncouplers. When oligomycin is first administered, chemoreceptor discharge rises to a peak (4,9,62) but subsequently decreases to a new steady-state level. This sequence of events can be explained as follows. Upon administration of oligomycin, ATP production rapidly falls so that chemoreceptor activity becomes maximal. Then the sensory discharge decreases to a new steady state as glycolysis increases. Other explanations are possible. It has been shown (2,3) that oligomycin can affect the cell membrane hyperpolarizing glomus cells. Some believe that the effects of oligomycin could result from nonspecific action on K⫹ channels (63–65). Experiments with carbon monoxide also support the metabolic hypothesis. Carbon monoxide is generated in small amounts in vivo by the enzyme hemeoxygenase II and can have many biological effects (66). CO reacts with guanylate cyclase, increasing cGMP, which in turn causes a cascade of reactions such as relaxation in smooth muscle by inhibiting Ca2⫹ entry (66–69). Also, CO at low levels (Pco, 70

Figure 2 High Pco (500 torr) at Po2 of 120 torr diminished the O2 disappearance rate, which was restored by light exposure, compared to control. The O2 disappearance rate was measured by Po microelectrode upon perfusate flow interruption.

110

Lahiri and Cherniack

torr) can bind with membrane hemeprotein and can reverse hypoxia-induced membrane depolarization in carotid body glomus cells (70) and decrease chemosensory discharge (71,72). Thus, low amounts CO take part in high-affinity reactions, binding to heme protein and mimicking the effect of O2. High concentrations of CO (Pco ⬎ 300 torr) can bind with a lower affinity to reduced cytochrome a3, a unit in the cytochrome oxidase complex in mitochondria, displacing O2 from cytochrome oxidase so that the cells will be hypoxic even in the presence of a Po2 of 100 torr (71–73). High Pco (⬎300 torr), even without hypoxia, excites the carotid body in the dark, but not if exposed to white light (Fig. 2). This suppression of excitation by light can be duplicated at wavelengths (73) that match exactly those that affect cytochrome oxidase (74–76). These findings strongly support the idea that cytochrome oxidase is the O2-sensing molecule that initiates carotid body activity. The excitation by CO is associated with a decrease of oxygen consumption by the carotid body, an inhibition that is reversed by light (77,78). Also, CO interacts with CO2 in the chemoreceptor in the dark, imitating the effects of hypoxia on Pco2 (18). This supports the idea that high Pco (⬎300 torr) operates like hypoxia in the dark. Dopamine is released by hypoxia and should be released by high Pco in the dark even in the absence of hypoxia (79–83). The fact that low levels of CO interfere with hypoxic responses, acting like added oxygen, suggests that there is another heme-containing complex with which CO combines in the membrane, which can influence O2 sensing in the carotid body. The importance of this membrane-combining site in the physiological response of the carotid body is unknown.

VII. Models of Chemoreception in the Carotid Body The two major hypotheses explaining O2 sensing in the carotid body are a nonrespiratory membrane hypothesis and a respiratory metabolic hypothesis. According to the membrane hypothesis, O2 sensing begins at the membrane of Type I cells with O2 suppressing directly or indirectly the opening of K⫹ channels, which leads to cell depolarization. The K⫹ channels that are closed by hypoxia may be special O2-sensitive channels or various other types of K⫹ channels, which can be influenced nonspecifically by hypoxia. For example, both hypoxia and acid suppress certain K⫹ and open Ca2⫹ channels in Type I cells (12,42–45,47–51). It may be that patency of K⫹ channels depends on the level of H2O2 and only indirectly on Po2 (85–90). Acker has postulated that O2 radicals are formed by the actions of the ubiquitous membrane protein NADPH oxidase on O2 (9). These radicals are dismutated to H2O2, which then acts as a second messenger according to Acker. The H2O2 can then either act on genes (e.g., to increase erythropoietin in erythropoietin-producing cells) or be scavenged by catalase or glutathione. Catalase scavenging leads to the formation of a complex, which activates heme containing guanylate cyclase and enhances the level of cGMP. Alternatively, glutathione scav-

O2 Sensing

111

enging changes the ratio of GSH to GSSG. In this scheme, hypoxia by altering cGMP and/or GSH GSSG ratio (reducing compounds) closes K⫹ channels and leads to Ca2⫹ influx. This hypothesis provides a mechanism for O2 sensing in both excitable and nonexcitable tissue. It is of interest that sulfhydryl agents that are also reducing agents are known to excite chemoreceptors (91). Drugs that alter H2O2 metabolism might help further evaluate the role of O2 radicals in chemoreception. In addition, the immune deficiency disorder chronic granulomatous disease, in which there is a genetic defect in H2O2 production, offers a unique possibility to evaluate its significance in O2 sensing. Murine models in which portions of NADPH oxidase are knocked out (92) may help determine the role of NADPH in O2 sensing. According to the respiratory metabolic hypothesis, O2 sensing takes place in mitochondria of the cytochrome oxidase complex, reducing energy metabolism and ATP formation. This changes the phosphate potential [ATP]:[ADP] [P1] ratio, allowing intracellular calcium to rise by release of Ca from intracellular stores, which in turn releases neurotransmitters. Figure 3 summarizes the membrane and respiratory metabolic theories of O2 sensing in the carotid body (1,11,15,18,62,77).

Figure 3 Postulated mechanisms of hypoxic chemoreception in type I carotid body cells.

112

Lahiri and Cherniack

There is evidence that Ca2⫹ is released from an internal store by hypoxia (93–95). Thapsigargin, a potent inhibitor for Ca-ATPase, resulted in a persistent elevation of intracellular [Ca2⫹], which increased chemosensory discharge. While it is now agreed that chemoreception in the carotid body takes place in Type I rather than Type II cells, it is still controversial whether respiratory metabolic or the nonrespiratory membrane models best portray O2 sensing. O2 sensing may take place in more than one site in Type I cells. The results of studies of the effects of high and low levels of CO on carotid body activity, which show inhibition with low levels of CO and excitation by higher levels, would be consistent with the idea that at least two different sites (two different molecular mechanisms) are triggered by oxygen lack and produce increasing chemoreceptor activity. Reactions at both sites could give rise to increases in intracellular Ca2⫹ (72). It is of interest that hypercapnia as well as hypoxia (54,58) is followed by a rise in intracellular [Ca2⫹], which corresponds to the increase in chemosensory discharge. In both models, the rise in intracellular Ca2⫹ produces neurotransmitter release. Dopamine seems to be an important neurotransmitter involved in modulating carotid sinus nerve discharge. Tyrosine hydroxylase is the rate-limiting enzyme for dopamine and norepinephrine synthesis and is activated within a few hours of exposure of the carotid body to hypoxia (92,97), but it can also be activated by other stimuli (83). Hypoxia releases dopamine from carotid body glomus cells as sensory discharge increases (79–82). Quantal secretion of dopamine also occurs from single glomus cells and is dependent on Ca2⫹ entry (82). Although both sensory discharge and dopamine release are Ca2⫹ dependent (82), other neurotransmitter may be important as well since dopamine stores can be exhausted in the rat CB and release stopped without a reduction in chemosensory response (83). While dopamine is certainly one neurotransmitter that is released by the carotid body, many other agents have been implicated as mediators of carotid sinus nerve discharge including dopamine acetylcholine and the neurokinins (82,83,96,97).

VIII. Effects of Hypoxia on Smooth Muscle and Neurons Mechanisms of O2 sensing have also been examined in vascular and neural tissues, but no consensus on the molecules or mechanisms involved has emerged. Hypoxia inhibits the contraction of systemic vascular smooth muscles (98), but it initiates contraction of pulmonary arterial smooth muscle cells as in the carotid body (67,85–90). O2 sensing in these tissues may arise from direct effects of hypoxia on ion channels, by an effect on ion channels mediated through some second messenger such as H2O2, or by an effect on oxidative metabolism (98–102). Some experimental observations indicate the difference in behavior of ion channels in the two kinds of smooth muscle. Hypoxia directly suppresses L-type Ca2⫹-channel activity in systemic smooth muscle cells (99). Also, hypoxia reduces potassium currents in cultured rat pulmonary but not mesenteric arterial myocytes (100–102). The K⫹ current depression in pulmonary arterial smooth muscle cells

O2 Sensing

113

depolarizes them, opening voltage-gated Ca2⫹ channels so that Ca2⫹ enters, causing the smooth muscles to contract (100–102). The same sequence of events does not occur in smooth muscle from systemic arteries. Wolin and others (67,85–90) believe that the reduction of superoxide and H2O2 formation by hypoxia lessens the production of cGMP, inhibits K⫹ channels, and causes smooth muscle contraction in pulmonary arteries. Archer et al. (63) have a similar theory. Ulrich and colleagues showed that high Pco in the dark relaxes ileal smooth muscle by activating guanylate cyclase and cGMP production (103,104). It has been proposed that inhibition of CO binding to a cytochrome P450–dependent monooxygenase reduced hemoprotein is the basis for ductus arteriosis smooth muscle relaxation in the lamb (105). Utz and Ulrich found that the photoreversible CO-dependent relaxation occurred with an action spectrum of 422 nm maximum, which was superimposable on that of guanylate cyclase (103). They could not detect any involvement of the respiratory chain. This group also found that CO inhibited the Ca2⫹ rise induced by depolarization (104). Lin and McGrath found that CO decreased [Ca2⫹] in systemic vascular smooth muscle, thereby relaxing and dilating the blood vessel (106). This was accompanied by a rise in cGMP. ATP levels can be modulated by hypoxia in systemic arteries but not in carotid body glomus cells according to Lopez-Barneo et al. (43). Opening of ATP-sensitive K⫹ channels in coronary artery cells by several minutes of severe hypoxia causes hyperpolarization, closure of Ca2⫹ channels, and muscle relaxation (100). Glibenclamide, a potent inhibitor of ATP-sensitive K⫹ channels, caused vasoconstriction, and cromakalim, an opener of K⫹ channels, produced vasodilation (107). Interference with oxidative metabolism by blockade of the respiratory chain to produce a fall in ATP has also been reported to relax the smooth muscle of the ductus arteriosis. The involvement of the respiratory chain in vascular smooth muscle sensing of oxygen lack has been found by one group (98) but denied by others (103–105). O2-sensitive K⫹ channels have also been found in central neurons, although the channels do not all have the same properties. Some of them work through alterations of cytosolic factors (ATP, Ca2⫹, etc.). Other investigators have found a direct mechanism for low Po2 sensing in the membrane, which is not affected by H2O2 generation or by Pco (380 torr) (108).

IX. Changes in the Ventilatory Response to Hypoxia with Time One of the fascinating aspects of the effects of hypoxia are the changes that occur in ventilation and hypoxic sensitivity with the length of hypoxia and the maturity of the responder. Except in a few instances, the cellular basis of these rather complex alterations is poorly understood. Some of the fluctuations in sensitivity probably occur because of the effects if hypoxia on different tissues occurring with different time courses, i.e., the effects

114

Lahiri and Cherniack

of hypoxia on the brain may alter the arterial chemoreceptor signal (efferent control) or modify the interpretation of that signal by central respiratory neurons.

X.

Erythropoietin Production During Hypoxia

Recent studies have considerably advanced our understanding of how hypoxia increases the production of erythropoietin (68,69). The physiological stimulus for erythropoietin production is a decrease in O2 supply (68,69). In fetal life, erythropoietin is produced in the liver, but after birth erythropoetin is formed in renal peritubular and the renal cortex in interstitial cells and perhaps endothelial cells (69). The hormone is a 30.4 kDa glycoprotein which targets hemopoietic tissue, where it binds with receptors on erythroid progenitor cells. These then differentiate into functioning red blood cells (69). Studies in the human hepatoma cell line Hep 3B have shown that hypoxia, cobalt, and nickel chloride stimulate erythropoietin production through a common pathway. On the basis of these studies it was proposed that the O2 sensor for erythropoietin production is a heme protein located outside the mitochondria (109). When the O2 tension is low, the heme protein is in the deoxy form, which initiates the process leading to erythropoietin gene activation. When O2 tension is high, the heme protein is in an inactive oxy form. Cobalt or nickel can substitute for the ferrous ion in the porphyrin ring, locking the heme protein in the deoxy form so that erythropoietin formation is stimulated (109). Red blood cell production also increases in rats inhaling 0.1% CO because the high-affinity CO binding to hemoglobin will inhibit O2 release from hemoglobin, thus lowering the effective O2 supply (69). On the other hand, the induction of erythropoietin in Hep 3B cells following exposure to 1% O2 in vitro is markedly inhibited by the presence of 10% CO presumably because it attaches to the O2 heme sensor locking in the oxy position (109,110). CO does not diminish the induction of erythropoietin by cobalt- or nickel-substituted heme because CO does not combine with these hemes (69). These observations support the hypothesis for O2 sensing by a heme protein (which may deoxygenate cytochrome P450) (69,109). There are similarities in the stimuli that produce erythopoietin and vascular endothelial growth factor. Cobalt also enhances the expression of vascular endothelial growth factor (110). Chronic cobalt administration to rats will give rise to increases in the volume of glomus cells (type I) along with hematocrit. One study found erythropoietin production to be inhibited by H2O2, but glutathione is not involved in erythropoietin formation (112). However, the role of oxygen radicals in erythropoietin production needs further investigation. Studies in Hep 3B cells described later indicate that hypoxia leads to the formation of one or more protein factors, which induce the erythropoietin gene. The activation of the protein factors may depend on the level of reactive O2 intermediates like peroxides (68). In humans there is a single copy of the erythropoietin gene located in the q11

O2 Sensing

115

to q22 region of chromosome 7 (69,110). In some cell types second messenger molecules such as cGMP and cAMP may also participate in gene activation (68). New evidence indicates that O2 sensing activates erythropoietin genes via a protein factor (HIF-1 or hypoxia inducible factor). The same factor also seems to induce vascular endothelial growth factor genes and seems to be participating in metabolic changes that occur with prolonged hypoxia (112a).

XI. Ventilatory Acclimatization to Hypoxia Studies of adaptation to chronic hypoxia demonstrate that a complex interaction of a number of different systems is involved. The cellular basis of this interplay of respiratory cardiovascular and neural factors is the focus of many investigators. The gradual increase in breathing during hypoxia, known as ventilatory acclimatization, has been extensively studied. Mountaineers ascending to high altitudes continue to hyperventilate over and above the acute response (113). Animal experiments have shown that this response is initiated by the carotid body (114) and does not occur in carotid body denervated animals (115) or during systemic hypoxia if the carotid body is vascularly isolated and maintained normoxic (114). It has been demonstrated that continued hypoxia after the initial rise in chemosensory discharge causes a further increase after a delay of 2–3 hours, explaining the increase in Po2 and decrease in CO2 that has been observed in intact animals and (116,117). It is highly likely that the acclimatization is due to an increase in humans carotid body activity. The roll-off of the acute ventilatory response to hypoxia that is observed a few minutes after exposure to hypoxia, however, is not caused by any corresponding change in carotid body activity, but rather seems to involve the release of inhibitor neurotransmitters in the brain (118). Thus the changes in ventilation as hypoxia is prolonged seem to involve a rather complicated interplay of responses to hypoxia in different tissues. The role of various neurotransmitters in ventilatory acclimatization or even the response to acute hypoxia remains controversial, but increases in production of dopamine in PC-12 cells have been used to surrogate for the evaluation of carotid body activity. Acute hypoxia does cause dopamine release in the carotid body and dopamine content to decrease (125–131). But whether dopamine release has an excitatory or inhibitory effect has not been settled (see, e.g., 126, 130). Norepinephrine, another carotid body catecholamine, seems to inhibit chemoreceptor activity (113). Chronic hypoxia stimulates its synthesis, but to a lesser extent than dopamine (127,128). The enlarged carotid bodies in chronic hypoxia contain a high concentration of dopamine. The carotid body is also known to secrete a higher concentration of dopamine as it enlarges (132,133). Despite the uncertainties concerning its effects, a number of studies have focused on dopamine metabolism in the carotid body during hypoxia. In the chronically hypoxic group, dopamine content and its turnover rate in-

116

Lahiri and Cherniack

creased 14- to 15-fold (127,128). Barer et al. (129) observed a blunted hypoxic ventilatory response in rats that lived for more than 2 weeks in an environment of severe hypoxia as compared to normoxic rats. According to Barer et al. (129) and Bee and Pallot (130), increased utilization of dopamine would tend to suppress the ventilatory response to hypoxia. Tatsumi et al. found a similar result in the cat exposed to severe hypoxia for 21 days (126). With less severe hypoxia the chemosensory response is augmented, even though dopamine content and turnover rate are also increased (128,129,131). Thus, a correspondence between dopamine secretion and sensory discharge in chronic hypoxia has not yet been established. XII. Blunted Ventilatory and Carotid Chemosensory Function with Prolonged Hypoxia Studying adult Peruvian high-altitude natives, Severinghaus and colleagues (119) found that the ventilatory response to hypoxia is blunted at altitude and remains blunted for a long time after the return to sea level (120). Similar results were found in the adult natives in the Himalayas (121,122). It seems reasonable to believe that Type I cells may function differently when hypoxia response is blunted. Wyatt et al. (56) showed that glomus cells of rats which developed blunted ventilatory response to hypoxia did not depolarize, even though they had K channels that were closed by low Po2 and K⫹ conductance was suppressed. Progress in this work would be illuminating.

Figure 4 Effects of lowering end-tidal Po2 on carotid chemoreceptor activity in a cat that was exposed to 100% O2 for 65 hours. (A) Effect of lowering Po2 from hyperoxia (100% O2) to air breathing. (B) Effect of further lowering of Po2. There was no effect of hypoxia on chemoreceptor activity. Traces (top to bottom) the tracheal Po2 (Ptco2); systemic arterial pressure (Psa); tracheal Pco2 (Ptco2); carotid chemoreceptor activity (in impulses per second); and raw impulses.

O2 Sensing

117

Figure 5 Effects of hypercapnia during hyperoxia: effects of sudden hypercapnia during acute hyperoxia on carotid chemoreceptor activity and ventilation in a control cat. Traces (top to bottom) are systemic arterial pressure (Psa); tracheal Pco2 (Ptco2); carotid chemoreceptor activity (in impulses/s); and impulses.

In this context, a blunted carotid chemosensory function has also been found in cats exposed to chronic hyperoxia (123,124) (Fig. 4). The chemosensory response to CO2 was, however, augmented (Fig. 5). This separation of O2 and CO2 response has been seen in other studies. These findings are not compatible with the idea that K⫹ channel opening and closing of Type I cells occurs in both the response to O2 and CO2 by the carotid body (54,55,61). XIII. Effect of Chronic Hypoxia on the Carotid Body and Carotid Body Cells In Vivo and In Vitro Exposure to high altitude or normobaric hypoxia at times causes substantial enlargement of the carotid bodies in humans and in other animal species (114,115,130– 134). The cells of carotid body blood vessels hypertrophy with chronic hypoxia and the vessels become filled high hematocrit blood (increases from 39% to 70%) (132). Bee and Pallot found evidence for hyperplasia of the carotid body cells by light microscopy (130). Quantitative morphometric analysis using stereological techniques at an ultrastructural level showed that the Type I rather than the Type II cells enlarge. The mean volume of Type I cells increased from 320 µm3 in controls to 1212 µm3 in chronically hypoxic rats. Similar observations have been made by Pegingnot et al. (133) and are universally accepted. Studies in cell culture by Nurse and coworkers have provided data on the effects of chronic hypoxia on the transmembrane movement of ions (135,136). Cellular hypertrophy and modifications of channel function occurs in cultured rat glo-

118

Lahiri and Cherniack

mus cells exposed to chronic hypoxia (6% O2 for 1–3 weeks) (135,136). These cells undergo changes in the absence of neural input or input from the blood or cardiovascular system. The mechanisms responsible for initiating and controlling the cellular hypertrophy remain unknown. Changes in cell size were accompanied by changes in membrane currents and probably on channels as well. While inward Na⫹ current increased significantly, because of increased cell size, the K⫹ current density decreased, whereas the inward Ca2⫹ current density remained unaffected. It is impossible to know whether these in vitro changes lead to increases or decreases in O2 sensitivity or chemosensory discharge in vivo. Studies measuring chemosensory discharge show that hypoxia of short-term duration increases (116,118,125,132) and of long-term duration decreases (105) O2 sensitivity. These changes do not seem to be explained by the changes reported in Type I cell size or number. A more recent study was done to compare the type I cells isolated from chronically hypoxic neonatal rats which have blunted hypoxic ventilatory responses of cells from normoxic animals (56). Although the density of O2-sensitive K⫹ channels was decreased in the cells from hypoxic compared to cells from normoxic animals, acute hypoxia caused similar reversible inhibition of K⫹ currents in type I cells from both groups. Type I cells from both normoxic and chronically hypoxic animals possess O2-sensitive K⫹ currents. Resting membrane potentials were also similar in normoxic and hypoxic rats. However, acute hypoxia failed to depolarize the Type I cells from chronically hypoxic rats. Charybodotoxin (20 nM) also failed to suppress the Ca2⫹-activated K⫹ current in hypoxic cells unlike normoxic cells (58). This suggests that absence of charybdotoxin Ca2⫹-sensitive K⫹ channel, rather than a decrease in O2-sensitive K⫹ channel, is responsible for the lack of O2 sensitivity in the carotid body of the chronically hypoxic rats. A major problem in the cultured cell experiments has to do with the range of Po2 sensitivity of O2-sensitive K⫹ channels (K-O2 channel). Excised patches of glomus cell membrane have K⫹-O2 channels (49), which are suppressed at levels of Po2 that do not excite the intact carotid body (13). The maximum inhibition of K⫹ current is obtained at a Po2 above 85 torr, but the carotid body in vivo is maximally excited at much lower levels of Po2 (49). Thus, there seems to be a discrepancy between the lower level of Po2 which stimulates chemosensory discharge and that which inhibits K⫹-O2 channel in the membrane patch. Possible reasons for this difference are as follows: the patch preparations used eliminate the modulation of ionic current by intracellular messengers, which usually occurs in vivo. For example, cAMP is known to increase during acute hypoxia and augment chemosensory response (138); cAMP analogs applied to the whole cell inhibit K⫹ oxygen-sensitive currents (49). These effects of cAMP are lost in the membrane patch experiments (49). Therefore, it has been suggested that some of the apparent differences between in vitro and in vivo studies can be explained by taking into account the decrease in the effectiveness of intracellular messengers such as cAMP. Another possibility is that intracellular O2-sensing mechanisms may be

O2 Sensing

119

coupled to several different types of K⫹ channels (47,55). Confounding the interpretation of these findings are the vast differences between the degree and duration of hypoxic exposure in the studies reported. For example, Nurse et al. incubated cells taken from normoxic rats at a Po2 of 49 torr for 3 weeks, while Peers used 70 torr for 3 days for cells taken from chronically hypoxic rats (61,136).

XIV. Effects of Hypoxia on Gene Expression We now know that changes in O2 tension can initiate changes in the expression of a variety of genes (69). Hepatoma cells (Hep 3B) produce a hypoxia-inducible factor (HIF-1 nucleoprotein) with about the same time course as erythropoietin formation (69). Cobalt chloride, like severe prolonged hypoxia, produces erythropoietin in these cells and also causes HIF-1 formation. HIF-1 formation can be blocked by the transcription inhibitor actinomycin D and by the protein synthesis inhibitor cycloheximide (139). HIF-1 has been shown to bind to the 3′-flanking enhancing regions of the erythropoietin gene (erythropoietin), but this requires phosphorylation to induce gene activity (139–142). HIF-1 has been shown to consist of different subunits, which contact DNA directly. The erythropoietin enhancer has three parts, two of which are essential for induction by hypoxia; the third amplifies the induction signal (140,141). Binding is transient and is quickly and easily reversible with high O2. In kidney cells other factors besides HIF-1 may be needed to produce erythropoietin (69). The increased levels of erythropoietin produced by hypoxia are only in part due to increased gene transcription (69). Erythropoietin mRNA half-life is also markedly prolonged by hypoxia (69). Interestingly, other studies have indicated that HIF-1 is also produced in cell lines that do not produce erythropoietin, suggesting that HIF-1 might activate other genes besides erythropoietin (139,141). Semenza et al. have shown that HIF-1 has a more general regulatory role and that RNAs encoding glycolytic enzymes such as aldolase, phosphoglycerate kinase, and pyruvate kinase were produced when Hep 3B or Hela cells were exposed to hypoxia (140). The formation of these mRNAs did not occur when cycloheximide was used to prevent HIF-1 formation (139,140). Sequences from these genes containing HIF1–binding sites were shown to be activated in transient expression assays using 1% O2 to stimulate HIF-1 formation (140). The studies strongly suggest that HIF1 regulates several glycolytic genes as well as erythropoietin and that HIF-1 functions as an important mediator of adaptation responses to hypoxia. Other gene pathways may be affected by hypoxia (Fig. 6). Hypoxia has been shown to increase the formation of platelet-derived growth factor, endothelin, interleukin-1A, and vascular endothelial growth factor (69,143–146). Exposure of rats to 10% hypoxia for even one hour increases mRNA for tyrosine hydroxylase in Type I carotid body cells (133). Recent studies by Milhorn and coworkers and by Prabhakar and coworkers suggest that the increase of tyrosine hydroxylase probably involves the activation of immediate early response genes of the fos-jun family

120

Lahiri and Cherniack

Figure 6 Effects of hypoxia on IERG expression in PC-12 cells. (A) Northern blot assay showing mRNAs encoding c-fos, junB, junD, and β-actin during control (normoxia; 21% O2) and after 1 hour of hypoxia (5% O2). Average results are summarized in (B). Effects of different durations of hypoxia in c-fos mRNA expression are presented in (C). Results presented are mRNA changes relative to normoxic controls (mRNA are normalized for 28s mRNA). Data presented are mean ⫾ SEM from five experiments in B and three experiments in C.

O2 Sensing

121

(144,145). Immediate early response genes (IERG) are comprised of several different groups of which the fos-jun family have been most extensively studied (146). Exposure of a number of different kinds of cells (PC-12, neuroblastoma, etc) to a variety of stimuli produce fos and jun, the protein products of the c-fos and c-jun genes. These two proteins form heterodimers (AP-1), which then bind to the promoter 3′ region of a number of different genes (146). AP-1 can also be formed by jun-jun homodimers but not by fos-fos homodimers. It has been shown in studies in vitro that AP-1 binding is affected by redox state and is greater under hypoxic conditions, which seems to require the integrity of crucial SH groups present in AP1 (147). In vivo in rats and cats moderate hypoxia (10% O2 for 30 minutes) will induce fos formation in the ventral and dorsal medulla and rather diffusely in other brain regions. A roughly similar distribution is obtained by exposure of the same animals to 15% CO2 in O2. In studies that focused on the nucleus tractus solitarii, fos formation by hypoxia could be blocked by NMDA, a glutamate antagonist, suggesting that glutamate may be needed for fos production in vivo (145,148). Hypoxia produces increased fos and AP-1 binding in PC-12 cells, but not in neuroblastoma cells (145,149), indicating that cells do not respond in a uniform way. Fos, though, can be induced in neuroblastoma cells by other stimuli. Studies of mRNA formation indicate that low levels of O2 in cultures of PC-12 cells (Po2 ⫽ 40 torr) increase transcription of c-fos and c-jun and other genes such as junB and junD (149). It is of interest that nitric oxide exposure of neuroblastoma cells augments fos formation. Both transcription of the tyrosine hydroxylase gene and the stability of tyrosine hydrolase mRNA are augmented by hypoxia (150). The 3′ promoter region of the TH gene contains both AP-1–and HIF-1–binding sites (44). Recent studies show that mutations of the AP-1–binding site drastically reduce the formation of TH mRNA by hypoxia, which suggests the importance of the fos-jun family in the hypoxic induction of tyrosine hydroxylase (44) but does not eliminate HIF-1 as well. It is not clear if the heme protein involved in sensing hypoxia for HIF-1 formation will also activate IERGs. However, Hep3B cells in culture will produce VEGF, a highly specific mitogen for endothelial cells, upon exposure to hypoxia and will increase levels of junB and c-jun (110). In addition, cobalt chloride, which leads to increases in both erythropoietin and VEGF, also increases c-jun and c-fos (109). Adaptations to chronic hypoxia may frequently involve activation of genes and the formation of new or increased amounts of proteins which alter the phenotypic properties of the cells. Some genes may begin to be activated within 6 minutes of hypoxic exposure. Responses to chronic hypoxia seem to involve many organs but may be coordinated by the activation of a relatively small set of genes, which produce protein products that bind to different target genes in different cell types. These gene-mediated changes may occur in the carotid body itself and may alter signaling cascades and affect the sensitivity of Type I cells to chronic hypoxia. Hypoxiainducible factor may be a part of a widespread oxygen-sensing mechanisms (32,140,147).

122

Lahiri and Cherniack

Our understanding at this time is that HIF-1αs are continually degraded under normoxia (151,152). This degradation during normoxia is a remarkable feature that requires iron-dependent von-Hippel-Lindau ligase complexes (153,154). This oxygen-dependent reaction is still unraveling (154,155). Hypoxia reverses this effect on normoxia rapidly within a few minutes so that HIF-1αs then accumulate and combine with the constitutively produced HIF-1β to HIF-1. These are then translocated into the nucleus to set the tone of various genes and mRNA expressions. These reactions may require a few minutes to hours to accomplish (151,152). However, the initial oxygen sensing that involves plasma membrane and cytoplasm may be over within split seconds. Despite all the advances in the area of gene regulation, the molecules that initiate the mechanisms of oxygen sensing still remain elusive. Acknowledgments We are grateful to Mary Pili and Stella Rogers for their secretarial assistance. Supported in part by grants HL-43413-11 and HL-50180-06. References 1.

2. 3. 4. 5. 6. 7. 8.

9.

10.

11.

Wilson DF, Mokashi A, Chugh D, Vinogradov S, Osanai S, Lahiri S. The primary oxygen sensor of the cat carotid body is cytochrome a3 of the mitochondrial respiratory chain. FEBS Lett 1994; 351:370–374. Duchen MR, Biscoe TJ. Mitochondrial function in type I cells isolated from rabbit arterial chemoreceptors. J Physiol 1992; 450:13–31. Duchen MR, Biscoe TJ. Relative mitochondrial membrane potential and [Ca2⫹] I in the type I cells isolated from the rabbit carotid body. J Physiol 1992; 546:33–61. Mulligan E, Lahiri S, Storey BT. Carotid body O2 chemoreception and mitochondrial oxidative phosphorylation. J Appl Physiol 1981; 51:438–446. Lahiri S, Rumsey WL, Wilson DF, Iturriaga R. Contribution of in vivo microvascular Po2 in the cat carotid body chemo transduction. J Appl Physiol 1993; 75:1035–1043. Goldberg, MA, Bunning SP, Bunn, HF. Regulation of the erythropoietin gene: evidence that the oxygen sensor is a hemeprotein. Science 1988; 242:1412–1415. Rotta A, Capena A, Hurado A, Velasquez T, Chavez R. Pulmonary circulation at sea level and at high altitudes. J Appl Physiol 1956; 9:328–336. Fay FS, Nair P, Whalen WJ. Mechanisms of oxygen induced contraction of ductus arteriosus. In: Reivich M, Coburn R, Lahiri S, Britton C, eds. Tissue Hypoxia and Ischemia. New York: Plenum Press, 1977:123–134. Acker H, Delpiano M, Degner F. The meaning of the Po2 field for the chemoreceptive process. In: Acker H, O’Regan RG, eds. Physiology of the Peripheral Arterial Chemoreceptors. Amsterdam: Elsevier, 1983:89–115. Whalen WJ, Nair P. Oxidative metabolism and tissue Po2 of the carotid body. In: Acker H, O’Regan RG, eds. Physiology of the Peripheral Arterial Chemoreceptors. Amsterdam: Elsevier, 1983:117–132. Buerk DG, Iturggiaga R, Lahiri S. Testing the metabolic hypothesis of O2 chemoreception in the cat carotid body in vitro. J Appl Physiol 1994; 76:1317–1323.

O2 Sensing 12.

123

Gonzalez C, Vicario I, Alvarez L, Rigul R. Oxygen sensing in the carotid body. Biol Signals 1995; 4:245–256. 13. Lahiri S. Chemophores in O2 chemoreception: the carotid body model. NIPs 1994; 9: 161–165. 14. Fay FS. Oxygen consumption of the carotid body. Am J Physiol 1970; 218:518– 523. 15. Lahiri S, Buerk DG, Chugh D, Osanai S, Mokashi A. Reciprocal photolabile O2 consumption and chemoreceptor excitation by carbon monoxide in the cat carotid body: evidence for cytochrome a3 as the primary O2 sensor. Brain Res 1995; 684:194–200. 16. Lahiri S. Role of arterial O2 flow in peripheral chemoreceptor excitation. Fed Proc 1980; 39:2648–2652. 16a. Acker H. The involvement of nerve terminals in the NY paraganglionic chemoreceptor system. Ann NY Acad Sci 1987; 519:369–384. 17. Floyd WF, Neil E. Arch Int Pharmocodyn Ther 1952; 91:230–239. 18. Osanai S, Chugh DK, Mokashi A, Lahiri S. Stimulus interaction between CO and CO2 in the cat carotid body chemoreception. Brain Res 1996; 711:56–63. 19. Hornbein TF, Roos A. Specificity of H ion concentration as a carotid body chemoreceptor stimulus. J Appl Physiol 1963; 18:580–584. 20. Lahiri S, DeLaney RG. Stimulus interaction in the response of carotid body chemoreceptor single afferent fibers. Respir Physiol 1975; 24:249–266. 21. Gray BA. Response of the perfused carotid body to changes in pH and Pco2. Respir Physiol 1968; 4:229–245. 22. Black AMS, McCloskey DI, Torrance, RW. The responses of carotid body chemoreceptors in the cat to sudden changes of hypercapnic and hypoxic stimuli. Respir Physiol 1971; 13:36–49. 23. Iturriaga R, Mokashi A, Lahiri, S. Dynamics of carotid body responses in vitro in the presence of CO2-HCO3: role of carbonic anhydrase. J Appl Physiol 1993; 75:1587– 1594. 24. Lahiri S, Itturiaga R, Mokashi A, Botre F, Chugh D, Osanai S. Adaptation to hypercapnia vs intracellular pH in cat carotid body response in vitro. J Appl Physiol 1996; 80: 1090–1099. 25. Beidler LM. Taste receptor stimulation. Prog Biophys Mol Biol 1962; 12:109. 26. McDonald DM. Peripheral chemoreceptors. Structure-function relationships of the carotid body. In: Hornbein TF, ed. Regulation of Breathing. Part I. New York: Marcel Dekker, 1981:105–319. 27. Lahiri S, Mokashi A, Nishino T, Mulligan E. Comparison of aortic and carotid chemoreceptor responses to hypercapnia and hypoxia. J Appl Physiol 1981; 51:55–61. 28. Youngson C, Nurse CA, Yeger H, Cutz E. Oxygen sensing in airway chemoreceptors. Nature 1993; 365:153–155. 29. Czyzyk-Krezeska MR, Bayliss DA, Lawson EE, Millhorn DE. Regulation of tyrosine hydroxylase gene expression in the rat carotid by hypoxia. J Neurochem 1992; 58: 1538–1546. 30. Duchen MR, Biscoe TJ. Relative mitochondrial membrane potential and [Ca2⫹] I in type I cells isolated from the rabbit carotid body. J Physiol 1992; 450:33–61. 31. Czyzyk-Krezeska MR, Dominiski Z, Kole R, Millhorn DE. Hypoxia stimulates binding of the cytoplasmic protein to a pyrimidine-rich sequence in the 3′-untranslated region of rat tyrosine hydroxylase mRNA. J Biol Chem 1994; 269:9940–9945. 32. Fandrey J. Hypoxia-inducible gene expression. Respir Physiol 1995; 101:1–10.

124

Lahiri and Cherniack

33.

Wang ZZ, Bredt DS, Fidone SJ, Stensaas LJ. Neurons synthesizing nitric oxide innervate the mammalian carotid body. J Comp Neurol 1993; 336:429–432. Prabhakar NR, Kumar GK, Chang CH, Agani FH, Haxhiu MA. Nitric oxide in sensory function of the carotid body. Brain Res 1993; 625:16–22. Chugh DK, Katayama M, Mokashi A, Bebout DE, Ray KK, Lahiri S. Nitric oxiderelated inhibition of carotid chemosensory nerve activity in the cat. Respir Physiol 1994; 97:147–156. Grimes PA, Mokashi A, Stone R, Lahiri S. Nitric oxide synthase in autonomic innervation of the cat carotid body. J Auton Nerv Syst 1995; 54:80–88. Lahiri S, Rumsey WL, Wilson DF, Iturriaga R. Contribution of in vivo microvascular Po2 in the cat carotid body chemotransduction. J Appl Physiol 1993; 75:1035–1043. Rumsey WL, Iturriaga R, Spergel D, Lahiri S, Wilson DF. Optical measurements of the dependence of chemoreception on oxygen pressure in the cat carotid body. Am J Physiol 1991; 261:C614–C622. Itturiaga R, Lahiri S, Mokashi A. Carbonic anhydrase and chemoreception in the cat carotid body. Am J Physiol 1991; 261:C565–C573. Black AMS, McCloskey DI, Torrance RW. The responses of carotid body chemoreceptions in the cat to sudden changes of hypercapnic and hypoxic stimuli. Respir Physiol 1971; 13:36–49. Nurse, CA. Carbonic anhydrase and neuronal enzymes in cultures glomus cells of the carotid body of the rat. Cell Tissue Res 1990; 261:55–71. Peers C. Ionic channels in Type I carotid body cells. In: O’Regan RG, Nolan P, McQueen DS, Patterson D, eds. Arterial Chemoreceptors: Cell to System. New York: Plenum Press, 1994:29–40. Lopez-Barneo J, Lopez-Lopez JR, Urena J, Gonzalez C. Chemotransduction in the carotid body: K⫹ currents modulated by Po2 in type I chemoreceptor cells. Science 1988; 241:580–582. Delpiano MA, Hescheler J. Evidence for a Po2-sensitive K⫹ channel in the type-I cell of the rabbit carotid body. FEBS Lett 1989; 249:195–198. Lopez-Lopez J, Gonzalez D, Urena J, Lopez-Barneo J. Low Po2 selectively inhibits K channel activity in chemoreceptor cells of the mammalian carotid body. J Gen Physiol 1989; 93:1001–1015. Hescheler J, Delpiano JA. Ionic currents on carotid body type I cells and the effects of hypoxia and NaCN. In: Eyzaquirre C, Fidone SJ, Fitzgerald RS, Lahiri S, McDonald DM, eds. Arterial Chemoreception. New York: Springer-Verlag, 1990:58–62. Lopez-Lopez JR, De Luis DA, Gonzalez C. Properties of a transient K⫹ current in chemoreceptor cells of rabbit carotid body. J Physiol (London) 1993; 460:15–32. Ganfornina, MD, Lopez-Barneo J. Potassium channel types in arterial chemoreceptor cells and their selective modication by oxygen. J Gen Physiol 1992; 100:401–426. Ganfornina MD, Lopez-Barneo J. Single K⫹ channels in membrane patches of arterial chemoreceptor cells are modulated by O2 tension. Proc Natl Acad Sci USA 1991; 88: 2927–2930. Lopez-Barneo, J. Oxygen-sensitive ion channels: How ubiquitous are they? TINS 1994; 17:133–135. Gonzalez C, Almaraz L, Plesso A, Rignal R. Oxygen and chemoreception in the carotid body chemoreceptors. TINS 1992; 15:146–153. Acker H. Mechanisms and meaning of cellular oxygen sensing in the organism. Respir Physiol 1994; 95:1–10.

34. 35.

36. 37. 38.

39. 40.

41. 42.

43.

44. 45.

46.

47. 48. 49.

50. 51. 52.

O2 Sensing 53. 53a. 53b. 54. 55.

56.

57. 58.

59.

60.

60a.

60b. 61.

62. 63. 64. 65.

66. 67. 68.

125

Acker H, Xue D. Mechanisms of O2 sensing in the carotid body in comparison with other O2-sensing cells. NIPS 1995; 10–24–216. Shirahata M, Fitzgerald RS. Dependency of hypoxic chemotransduction in cat carotid body on voltage-gated calcium channels. J Appl Physiol 1991; 71:1052–1069. Roy A, Rozanov C, Iturriaga R, Mokasi A, Lahiri S. Acid-sensing by carotid body is inhibited by blockers of voltage-sensitive Ca2⫹ channels. Brain Res 1997; 769:396–399. Fieber LA, McCloskey EW. J Neurophysiol 1993; 70:1378–1384. Buckler KJ, Vaughan-Jones RD. Role of intracellular pH and {Ca2⫹} i in acid chemoreception in type–cells of carotid body. In: O’Regan RG, Nolan P, McQueen DS, Paterson D, eds. Arterial Chemoreceptors: Cell to System. New York: Plenum Press, 1994:41–55. Wyatt CN, Wright C, Bee D, Peers C. O2-sensitive K⫹ currents in carotid body chemoreceptor cells from normoxic and chronically hypoxic rats and their roles in hypoxic chemotransduction. Proc Natl Acad Sci 1995; 295–299. Cheng PM, Donnelly DF. Relationship beetween changes of glomus cell current and neural response of rat carotid body. J Neurochem 1995; 74:2077–2086. Osanai S, Buerk DG, Mokashi A, Chugh DK, Lahiri S. Lack of effect of charybdotoxin on cat carotid body chemosensory discharge: a denial of the role of O2-sensitive K⫹ Ca currents. Brain Res (in press). Buckler KJ, Vaughan-Jones RD. Effects of hypoxia on membrane potential and intracellular calcium in rat neonatal carotid body type I cells. J Physiol (London) 1994; 476:423–428. Buckler KJ. Role of potassium channels in mediating the hypoxia evoked depolarization of carotid body type-1 cells. In: Zapta P, Eyzaquirre C, Torrance RW, eds. Frontiers in Arterial Chemoreception. New York: Plenum Press, 1966: Roy A, Rozanov C, BuerkDG, Mokashi A, Lahiri S. Supression of glomus cell K⫹ conductance by 4-aminopyridine is not related to [Ca2⫹], dopamine release and chemosensory discharge from carotid body. Brain Res 1998; 785:228–235. Lahiri S, Roy A, Rpzanov C, Mokashi A. K⫹ current modulated by PO2 in Type cells in rat carotid body is not a chemosensor. Brain Res 1998; 794:162–165. Peers C. Ionic channels in type I carotid body cells. In: O’Reagan R, Nolan P, McQueen DS, Patterson D, eds. Arterial Chemoreceptors: Cell to System. New York: Plenum Press, 1994:29–40. Mulligan E, Lahiri S. The separation of carotid chemoreceptor responses to O2 and CO2 by oligomycin and antimyocin A. Am J Physiol 1982; 242:C200–C206. Archer SL, Huang J, Henry T, Peterson D, Weir KE. A redox-based O2 sensing in rat pulmonary vasculature. Cir Res 1993; 73:1100–1112. Wolin MS. Reactive oxygen species and vascular signal transduction mechanisms. Microcirculation 1995; 3:1–17. Franco-Obregon A, Urena J. Lopez-Barnes J. Oxygen-sensitive calcium channels in vascular smooth muscle and their possible role in hypoxic arterial relaxation. Proc Natl Acad Sci 1995; 92:4715–4719. Prabhakar NR, Dinerman JL, Againi FH, Solomon S. Carbon monoxide: a role in carotid body chemoreception. Proc Natl Acad Sci. 1995:92–1997. Burke TM, Wolin MS. Hydrogen peroxide elicits pulmonary arterial relaxation and guanylate cyclase activation. Am J Physiol 1987; 21:H721–H732. Bunn HF, Payton RO. Oxygen sensing and molecular adaptation to hypoxia. Physiol Rev 1996; 76:839–886.

126 69.

Lahiri and Cherniack

Jelkman W. Erythropoietin: Structure control of production and function. Physiol Rev 1992; 72:449–489. 69a. Fandrey J, Frede S, Ehleben W, Ponwol T, Acker H, Jelkmann W. Cobalt chloride and desferrioxamine antagonize the inhibition of erythropoietin production by reactive oxygen species. Kidney Int 1997; 51:492–496. 70. Lopez-Lopez J, Gonzalez C. Time course of K⫹ current inhibition by low oxygen in chemoreceptor cells of adult rabbit carotid body. Effect of carbon monoxide. FEBS Lett 1992; 299:251–254. 71. Lahiri S, Ray, DK, Chugh D, Iturriaga R, Mokashi A. CO-binding chromophores in oxygen chemoreception in the carotid body. In: O’Regan RG, Nolan P, McQueen DS, Patterson DJ, eds. Arterial Chemoreceptors: Cell to System, New York: Plenum Press, 1994:149–153. 72. Lahiri S. Iturriaga R, Mokashi A, Ray DK, Chugh D. CO reveals dual mechanisms of O2 chemoreception in the carotid body. Respir Physiol 1993; 94:227–240. 73. Wilson DF, Mokashi A, Chugh D, Vinogradove S, Osanai, S, Lahiri S. The primary oxygen sensor of the cat carotid body is cytochrome a3 of the mitochondrial respiratory chain. FEBS Lett 1994; 351:370–374. ¨ ber die photochemische Dissoziation bei inermitterender 74. Warburg O, Negelein E. U Bleichuterung und das absolute Absorpsionsspektrumn des Athmungsferments. Biochem Z 1928; 202:202–228. 75. Melnick JL. The photochemical spectrum of cytochrome oxidase. J Biol Chem 1942; 146:385–390. 76. Castor LN, Chance B. Photochemical action spectra of carbon monoxide-inhibited respiration. J Biol Chem 1955; 217:453–465. 78. Lahiri S. Carotid body O2 chemoreception: respiratory and non-respiratory aspects. Biol Signals 1995; 4:257–262. 79. Fidone SJ, Gonzalez C, Yoshikazi K. A study of the relationship between dopamine release and chemosensory discharge from rabbit carotid body in vitro: preliminary findings. In: Belmonte C, Pallot DJ, Acker H, Fidone S, eds. Arterial chemoreceptors. Leicester: Leicester University Press, 1981:209–219. 80. Donnelly DF. Electrochemical detection of catecholamine release from rat carotid body in vitro. J Appl Physiol 1993; 75:2330–2337. 81. Buerk DG, Lahiri S, Chugh D, Mokashi A. Electrochemical detection of rapid DA release kinetics during hypoxia in perfused-superfused cat CB. J Appl Physiol 1995; 78:830–837. 82. Urena J, Fernandez-Chcosr R, Benot AR, Alvarez de Toledo G, Lopez-Barnes J. Hypoxia induced voltage-dependent Ca2⫹ entry and quantal dopamine secretion in carotid body glomus cells. Proc Natl Acad Sci 1994; 91:10209–10211. 83. Donnelly DF. Does catecholamine secretion mediate the hypoxia-induced increase in nerve activity? Biol Signals 1995; 4:304–309. 84. Gonzalez C, Almaraz L, Obesso A, Rigual R. Carotid body chemoreceptors: from natural stimuli to sensory discharges. Physiol Rev 1994; 74:829–900. 85. Burke-Wolin TM, Wolin MS. H2O2 and cGMP may function as an O2 sensor in the pulmonary artery. J Appl Physiol 1989; 66:167–170. 86. Burke-Wolin TM, Wolin S. Inhibition of cGMP associate pulmonary arterial relaxation to H2O2 and O2 by ethanol. Am J Physiol 1990; 258:H1267–H1270. 87. Cherry PD, Omas HA, Farrell KA, Stuart JS, Wolin MS. Superoxide anion inhibits

O2 Sensing

127

cGMP-associated bovine pulmonary arterial relaxation. Am J Physiol 1990; 239: H1056–H1062. 88. Omar HA, Mohazzab H, Mortelliti MP, Wolin MS. O2 dependent modulation of calf pulmonary artery tone by lactate: role of H2O2 and cGMP. Am J Physiol 1993; 264: L141–L145. 89. Paky A, Michael JR, Burke-Wolin T, Wolin MS, Gurtner GH. Endogenous production of superoxide by rabbit lungs: effects of hypoxia or metabolic inhibitors. J Appl Physiol 1993; 74:2868–2874. 90. Wolin MS, Burke TM. Hydrogen peroxide elicits activation of bovine pulmonary arterial soluble guanylate cyclase by a mechanism associated with its metabolism by catalase. Biochem Biophys Res Commun 1987; 143:20–25. 90a. Lander HM. An essential role for free radicals and derived species in signal transduction. FASEB J 1997; 11:118–124. 91. Lahiri S. Chemical modification of carotid body chemoreception by sulfhydryls. Science 1981; 212:1065–1066. 92. Dinauer MC. A murine model of chronic granulomatous disease generated by targeted disruption of the mouse gphphon gene (abstr.). Blood 1993; 82:328A. 93. Hoth M, Penner R. Calcium release-activated calcium current in rat mast cells. J Physiol 1993; 465:359–386. 94. Putney Jr. JW. A model for receptor-regulated calcium entry. Cell 1986; 7:1–12. 95. Lahiri S, Osanai S, Buerk DG, Mokashi M, Cugh DK. Thepsigargin enhances carotid body chemosensory discharfe in response to hypoxia in zero [Ca2⫹]e : evidence for intracellular Ca2⫹ release. Brain Res 1996; 709:141–144. 96. Fitzgerald RS, Shirahata M. Acetylcholine and carotid body excitation during hypoxia in the cat. J Appl Physiol 1994; 76:1566–1574. 97. Prabhakar NR, Landis SC, Kumar GK, Mullikin-Kilpatrick D, Cherniack NS. Substance P and neurokinin A in the cat carotid body: Localization, exogenous effects and changes in response to arterial Po2. Brain Res 1989; 481:205–214. 98. Katayama T, Coburn RF, Fillers WS, Baron CB. Oxygen sensors in vascular smooth muscle. J Appl Physiol 1994; 77:2088–2092. 99. Franco-Obegon A, Urena J, Lopez-Barneo J. Proc Natl Acad Sci USA 1995; 92:4715– 4719. 100. Weir EK, Archer SL. The mechanism of acute hypoxic pulmonary vasoconstriction: the tale of two channels. FASEB J 1995; 9:183–189. 101. Post JM, Hume JR, Archer SL, Weir EK. Direct role for potassium channel inhibition in hypoxic pulmonary vasoconstriction. Am J Physiol 19092; 262:C882–C890. 102. Yuan XJ, Goldman WF, Tod ML, Rubin LJ, Blaustein MP. Hypoxia reduces potassium currents in cultured rat pulmonary but not mesenteric arterial monocytes. Am J Physiol 1993; 264:L116–L123. 103. Utz J. Ulrich V. Carbon monoxide relaxes ileal smooth muscle through activation of guanylate cyclase. Biochem Pharmacol 1991; 41:1195–1201. 104. Trischmann U, Klockner U, Isenberg G, Utz J, Ulrich V. Carbonmonoxide inhibits depolarization-induced Ca rise and increases cyclic GMP in visceral smooth muscle cells. Biochem Pharmacol 1991; 41:237–241. 105. Coceani F, Hamilton NC, Labdue J, Olley PM. Cytochrome P450-linked monooxygenase: involvement in the lamb ductus arteriosus. Am J Physiol 1984; 246:H640– H643.

128 106.

Lahiri and Cherniack

Lin H, McGrath JJ. Carbon monoxide effects on calcium levels in vascular smooth muscle. Life Sci 1988; 43:1813–1816. 107. Quast U, Cook NS. In vitro and in vivo comparison of two K⫹ channel openers, diazoxide and cromakalim, and their inhibition by glibenclamide. J Pharm Exp Ther 1989; 250:261–271. 108. Jiang C, Haddad GG. A direct mechanism for sensing low oxygen levels by central neurons. Proc Natl Acad Sci 1994; 91:7198–7201. 109. Goldberg M, Dunning S, Bunn HF. Regulation of the erythropoietin genome: evidence that the oxygen sensor is a heme protein. Science 1988; 242:1412–1415. 110. Goldberg MH, Schneider TJ. Similarities between the oxygen-sensing mechanisms regulating the expression of vascular endothelial growth factor and erythropoietin. J Biol Chem 1994; 269:4355–4359. 111. DiGiulio C, Data PG, Lahiri S. Chronic cobalt causes hypertrophy of glomus cells in the carotid body. Am J Physiol 1991; 261:C102–C05. 112. Fandrey J, Frede S, Jelkmann W. Role of hydrogen peroxide in hypoxia-induced erythropoietin production. Biochem J 1994; 303:507–510. 112a. Gassman M, Wenger RH. HIF-1, a mediator of the molecular response to hypoxia. N Physiol Sci 1997; 12:214–218. 113. Bisgard GE, Forster HV. Ventilatory responses to acute and chronic hypoxia. In: Fregley MJ, Blatteis C, eds. Handbook of Physiology: Environmental Physiology. New York: Oxford University Press, 1996; 1207–1240. 114. Busch MA, Bisgard C, Rorster HV. Ventilatory acclimatization to hypoxia is not dependent on arterial hypoxemia. J Appl Physiol, 1985; 58:1874–1880. 115. Lahiri S, Edelman NH, Cherniack NS, Fishman AP. Role of carotid chemoreflex in respiratory acclimatization to hypoxemia in goat and sheep. Respir Physiol 1981; 46: 367–382. 116. Nielsen AM, Bisgard GE, Vidruk EH. Carotid chemoreceptor activity during acute and sustained hypoxia in goats. J Appl Physiol 1988; 65:1796–1802. 117. Howard LSGE, Robbins PA. Alterations in respiratory control during eight hours of isocapnic and poikilocapnic hypoxia in humans. J Appl Physiol (in press). 118. Vizek M, Pickett CK, Weil JV. Increased carotid body hypoxic sensitivity during acclimatization to hypobaric hypoxia. J Appl Physiol 1987; 63:2403–2410. 119. Severinghaus JW, Bainton CK, Carcelen A. Respiratory insensitivity to hypoxia in chronically hypoxic man. Res Physiol 1966; 1:308–334. 120. Lahiri S, Brody JS, Velasquez T, Motoyama EK, Simpser M, Dleaney RG, Polgar G. Pulmonary adaptation to high altitude: genetic vs. environment. Nature 1976; 261: 133–135. 121. Milledge JS, Lahiri S. Respiratory control in lowlanders and Sherpa highlanders at altitude. Respir Physiol 1967; 2:310–322. 122. Curan LS, Zhuan J, Droma T, Moore LG. Hypoxic ventilatory response in Tibetan residents of 4400 m compared with 3658 m. Respir Physiol 1995; 100:223–230. 123. Lahiri S, Mulligan E, Andronikou S, Shirahata M, Mokashi M. Carotid body chemosensory function in prolonged normobaric hyperoxia in the cat. J Appl Physiol 1987; 62:1924–1931. 124. Mokashi A, Lahiri S. Aortic and carotid body chemoreception in prolonged hyperoxia in the cat. Respir Physiol 1991; 86:233–243. 125. Hanbauer I, Hellstrom S. The regulation of dopamine and noradrenaline in rat carotid body and its modification by denervation and hypoxia. J Physiol 1978; 282:21–34.

O2 Sensing 126. 127.

128.

129. 130. 131. 132. 133.

134.

135.

136. 137.

138. 139. 140.

141. 142. 143. 144. 145.

129

Tatsumi K, Pickett CK, Weil JV. Decreased carotid body hypoxic sensitivity in chronic hypoxia: role of dopamine. Respir Physiol 1995; 101:47–57. Haubauer I, Karroum F, Hellstrom S, Lahiri S. Effects of hypoxia lasting up to one month on the catecholamine content in rat carotid body. Neuroscience 1981; 6:81– 86. Pequingnot JM, Collet-Emard JM, Dalmaz Y, Peyrin L. Dopamine and norepinephrine dynamics in rats carotid body during long-term hypoxia. J Autonom Nerv Systm 1987; 21:9–14. Barer GR, Edwards CW, Jolly AI. Changes in the carotid body and the ventilatory response to hypoxia in chronically hypoxic rats. Clin Sci 1976; 50:311–313. Bee D, Pallot DJ. Hypoxic ventilatory response. Carotid body cell division and dopamine content during early hypoxic exposure in rats. J Appl Physiol (in press). Olson EB Jr, Dempsey J. Rat as a model for human-like ventilatory adaptation to chronic hypoxia. J Appl Physiol 1978; 44:763–769. McGregor KH, Gil J, Lahiri S. A morphometric study of the carotid body in chronically hypoxic rats. J Appl Physiol 1984; 57:1430–1438. Pequignot JM, Hellstrom S, Johansson C. Intact and sympathectomized carotid bodies of long-term hypoxic rats: a morphometric ultrastructural study. J Neurochem 1984; 13:481–493. Katz DM, markey KA, Goldstein M, Black IB. Expression of catecholaminergic characteristic by primary sensory neurons in the normal adult rat in vivo. Proc Natl Acad Sci 1981; 80:3526–3530. Stea A, Nurse CA. Whole-cell and perforated patch recording from O2-sensitive rat carotid body cells grown in short- and long-term culture. Pflugers Arch 1991; 418: 93–101. Nurse CA. Carotid body adaptation to hypoxia: cellular and molecular mechanisms in vitro. Biol Signals 1995; 4:286–291. Barnard P, Audronikou S, Pokorski M, Smatresk N, Mokashi A, Lahiri S. Time-dependent effect of hypoxia on carotid body chemosensory function. J Appl Physiol 1987; 63:685–691. Wang WJ, Cheng GF, Yoshizaki K, Dinger B, Fidone S. The role of cyclic AMP in chemoreception in the rabbit carotid body. Brain Res 1991; 540:96–104. Wang IGL, Semenza GL. General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci 1993; 90:4304–4308. Semenza GL, Roth PH, Fang HM, Wang GI. Transcriptional regulation of gene encoding glycolytic enzymes by hypoxia inducible factor-1. J Biol Chem. 1994; 269:23757– 23763. Wang GL, Semenza GL. Characterization of hypoxia-inducible factor-1 and regulation of DNA binding activity by hypoxia. J Biol Chem 1993; 268:21513–21518. Wang GI, Semenza GL. Purification and characterization of hypoxia inducible factor1. J Biol Chem 1995; 270:1230–1237. Millhorn DE, Czyzyk-Krezeska M, Bayliso DA, Lawson EE. Regulation of gene expression by hypoxia. Sleep 1993; 161:S44–48. Norris ML, Millhorn DE. Hypoxia-induced protein binding to O2 responsive sequences in the tyrosine hydroxylase gene. J Biol Chem 1996; 270:23774–23779. Cherniack NS, Prabhakar N, Haxhiu M. Possible genomic mechanisms involved in modeling and control of ventilation. In: Semple SJC, Adams L, Whipp BJ, eds. New York: Plenum Press, 1995:89–94.

130

Lahiri and Cherniack

146.

Morgan JI, Curran T. Stimulus transcription coupling in the nervous system: involvement of the inducible proto-oncogenes fos and jun. Ann Rev Neurosci 1991; 14:421– 451. Xanthoudakis S, Miao G, Wang F, Pon YCE, Curran T. Redox activation of Fos-Jun DNA binding activity is mediated by a DNA repair enzyme. EMBO J 1992; 11:3323– 3335. Haxhiu MA, Stronl KP, Cherniack NS. The N-methyl-D-aspartate receptor pathway is involved in hypoxia-induced c-Fos protein expression in the rat nucleus of the solitary tract. J Autonom Nerv Syst 1995; 55:65–68. Czyzky-Krzesak MF, Funair BA, Millhorn DE. Hypoxia increases rate of transcription and stability of tyrosine hydroxylase mRNA in pheochromocytoma PC-12 cells. J Biol Chem 1994; 269:760–764. Prabhakar NR, Shenoy B, Simonson MS, Cherniack NS. Cell selective induction and transcriptional activation of immediate early genes by hypoxia. Brain Res 1995; 67: 266–270. Salceda S, Caro J. Hypoxia-inducible factor 1α(HIF-1α) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes. J Biol Chem 1997; 272:22642–22647. Semenza GL. Hypoxia-inducible factor 1: master regulator of O2 homeostasis. Curr Opin Genet Dev 1998; 8:588–594. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER, Ratcliffe PJ. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 1999; 399:271– 275. Wenger RH. Mammalian oxygen sensing, signalling and gene regulation. J Exp Biol 2000; 203:1253–1263. Guillemin K, Krasnow MA. The hypoxic response: huffing and hifing. Cell 1997; 89: 9–12.

147.

148.

149.

150.

151.

152. 153.

154. 155.

5 Molecular/Metabolic Defense and Rescue Mechanisms for Surviving Oxygen Lack From Genes to Pathways*

PETER W. HOCHACHKA University of British Columbia Vancouver, British Columbia, Canada

I.

Cell Level Paradigms for Hypoxia Tolerance

It is well known in biology that certain vertebrate species have evolved unusual capabilities for surviving prolonged periods without oxygen or with greatly reduced supplies of oxygen. Studies of such species [some being so anoxia tolerant they are referred to as ‘‘facultative’’ anaerobes (1)] have revealed several seemingly universal strategies of hypoxia adaptation. Two of the most significant of these include severe downregulation of energy turnover (2–6) and upregulation of energetic efficiency of ATP-producing pathways (7). The latter involve stoichiometric efficiencies. In hypoxia adaptation, pathways that maximize the yield of ATP per mole oxygen are favored, while in anoxia adaptation, anaerobic pathways that maximize the yield of ATP per mole of H⫹ formed in the fermentation are favored (8,9). In hypoxia or anoxia adaptation, the ratio of anaerobic/aerobic metabolic potential may well be upregulated, coincident with upregulation of glycogen (fermentable substrate) and of buffering capacities. Even if all these mechanisms are useful in surviv-

* Only literature up to 1996 was surveyed for this chapter.

131

132

Hochachka

ing hypoxia, evaluation of such hypoxia defense strategies having naturally evolved in hypoxia-tolerant animals shows that suppression of energy turnover supplies the greatest protection against, and hence advantage in, hypoxia. The uniquely large advantage of this defense strategy is widely appreciated by many biologists (4,5,10– 12). In one of his last personal communications to the authors, the great comparative physiologist Kjell Johansen referred to this strategy as ‘‘turning down to the pilot light’’; like earlier workers in this field, he was acutely aware of its relative importance. Although recognized as a kind of hallmark of reversible entry into and return from states of severe oxygen deprivation, a number of unexplained problems have remained. First, it has not been clear how cells/tissues ‘‘know’’ when to turn on their hypoxia defense mechanisms. Second, it has not been known which pathways of ATP demand and ATP supply are downregulated or by how much. Additionally, it has been unclear how membrane electrochemical gradients are stabilized and what gene expression and protein expression level adjustments are involved in reorganization of cell structure and function under oxygen limitation. In recent studies of cells and tissues from the aquatic turtle, brain cortical slices were used to probe electrophysiological properties of neurons under anoxia (13–16) while isolated liver hepatocytes were used to probe cell level biochemical responses to anoxia (17–21). In combination with independent lines of research in several other laboratories, these studies supply the raw material for formulating a synthetic cell level model or general hypothesis of hypoxia tolerance, which goes a long way to answering the above unresolved questions.

II. Detecting When Oxygen Becomes Limiting The traditional views of cell level responses to oxygen limitation are formalized in the concept of the Pasteur effect: as ATP generation by oxygen-based mitochondrial metabolism begins to decline due to oxygen lack, the energetic deficit is made up by activation of anaerobic ATP-generating pathways. In turtle liver cells, this kind of oxygen sensing would require activation of anaerobic glycolysis (since this is the only known mechanism for ATP generation without oxygen in these cells). Two telling observations argue potently against this as the mechanims for detecting hypoxia in cells tolerant to oxygen lack. First, hypoxia-tolerant systems rarely activate anaerobic metabolism to make up energy deficits; they favor a reduced energy turnover state instead. Equally importantly, responses to falling oxygen supplies begin at concentrations much higher than the Km(O2 ) for mitochondria (termed oxygen conformity). Indeed, the observation of oxygen conformity first led workers in this area to postulate the occurrence of a mechanism for sensing molecular oxygen as the cell level means for detecting hypoxia (9,10,12,22) (a point to be further discussed below).

Defense and Rescue Mechanisms

133

III. ATP Demand Pathway ⴝ ATP Supply Pathway During Hypoxia In view of the above, it was not surprising to find that when liver hepatocytes encounter oxygen lack, they suppress energy turnover by a factor of almost 10-fold (17,18). In this condition, a mole of ATP could sustain these cells 10 times longer than under normoxic conditions. Without comparable regulation of ATP demand pathways, this mechanism would quickly deplete ATP supplies during hypoxia; thus it is important to know quantitatively which processes are turned down. To this end, we assessed the main energy demand functions under normoxia (energetically balanced by the oxygen consumption rates under these conditions) and compared these to the energy sinks remaining under anoxia. We found that under normoxic conditions, the main energy demand processes (Table 1) are protein synthesis, protein degradation, Na⫹K⫹ pumping, urea biosynthesis, and glucose biosynthesis. The ATP demands of these processes account for pretty well all of the ATP production expected from oxygen consumption (17–21). What happens to the same processes in turtle liver cells under anoxia is most instructive. Under these conditions, the ATP demand of protein turnover drops to less than 10% of normoxic rates; urea biosynthesis drops to essentially zero, as does the biosynthesis of glucose. Although the ATP requirements of the Na⫹K⫹ ATPase are also drastically reduced, the suppression in percentage terms is less than for overall ATP turnover. As a result, under anoxic conditions, the Na⫹ pump becomes the cell’s dominant energy sink, accounting for up to 75% of the ATP demand of the cell (17). Although there may well be other minor energy demand processes remaining under anoxic conditions, the quantitative ATP requirements of the energy demand pathways identified during anoxia fully account for the ATP generated by anaerobic glycolysis (Table 1).

Table 1 The Main ATP Demand Pathways During Normoxia and Anoxia in Turtle Hepatocytes ATP demand (µmol ATP ⫻ g⫺1 ⫻ h⫺1) Pathway Total Na⫹ pump Protein synthesis Protein breakdown Urea synthesis Gluconeogenesis

Normoxia

Anoxia

% Suppression

67.0 19.1 24.4 11.1 2.0 11.4

6.3 4.8 1.6 0.7 0.6 0.0

94 75 93 94 70 100

Source: Modified from Refs. 17, 19, 20.

134

Hochachka IV. Coupling Metabolism and Membrane Functions During Hypoxia

While in relative terms ion pumping (as assessed by the activity of Na⫹K⫹ ATPase) is the single largest ATP sink during anoxia in turtle hepatocytes, its absolute ATPase or pumping activity is only a small fraction (about one fourth) of normoxic levels. Despite this, experimental estimates indicate that the electrochemical potential in anoxic liver cells is essentially the same as in normoxia (17). To account for the large-scale drop in absolute Na⫹K⫹ ATPase activity and the simultaneous maintenance of normal electrochemical gradients would seem to require a similar magnitude decrease in cell membrane permeability [termed generalized ‘‘channel arrest’’ in the literature (10)]. In an earlier synthesis of information in this research field (10), a channel arrest component of a hypoxia tolerance theory postulated (1) that hyoxia-tolerant cells would have an inherent low permeability (either low channel densities or low channel activities) and (2) that they would sustain a further suppression of membrane permeability to ions when exposed to oxygen lack (further channel arrest by either suppression of channel densities or channel activities). Turtle liver cells display both of these characteristics (especially when compared to mammalian homologs); i.e., they fit the classical definition of metabolic and channel arrest as two telling signatures of hypoxia tolerance. In turtle cortical cells, in contrast, only the first criterion is met: a background electrical conductivity that is unusually low and when compared at biological temperatures (15 vs. 37°C) can be as low as 1/25 the conductivity of cell membranes of rat cortical cells (15). When exposed to acute oxygen lack (for up to several hours), there is no further channel arrest and therefore no further decrease in background conductivity of these neuronal cells. This may be one reason why the main energy-saving mechanism in turtle brain (23,24) is downregulation of firing rates or synaptic transmission [termed ‘‘spike arrest’’ by Sick et al. (24)] presumably through adenosine-mediated downregulation of excitatory amino acid (especially glutamate) release with concomitant increase in release of inhibitory amino acids (23,25). It may also explain why the metabolic suppression in brain is substantially less than in liver cells, down to about 1/2 rather than 1/10 of normoxic rates (16). An additional consequence of these results is that the ATP turnover rate of anoxic turtle cortical brain cells is higher than in turtle liver cells. Direct microcalorimetric measurements indicate that heat fluxes in anoxic brain cortical tissue are about threefold higher than in anoxic liver cells (16,18). In vivo, of course, anoxic brain survival depends solely upon anaerobic glycolysis (13,26) fueled by plasma glucose derived ultimately from liver glucose. This appears to indicate that in the turtle brain, hypoxia tolerance equals a more modest metabolic arrest than in liver cells plus the arrest not of ‘‘leakage’’ channels but of channel functions associated with synaptic transmission, in order to remain in energy balance.

Defense and Rescue Mechanisms V.

135

Hypoxic Sensitivity of Protein Synthesis

Protein turnover is the only energy-requiring process in normoxic liver cells that is more costly than ion pumping (17–20); hence evaluating this ATP sink under oxygen limiting conditions is of particular importance. For perspective, it is important to note that one of the first and most dramatic effects of hypoxia on cell metabolic functions is a rapid and large magnitude inhibition of protein synthesis. The decline can be so rapid that its time course is hard to quantify accurately with currently used techniques for measuring protein biosynthesis (27). A hypoxia-induced block in protein synthesis could occur at the level of gene transcription or translation. In animal models, the mechanism for an initial hypoxiainduced block in protein synthesis is unclear, but in plant systems hypoxic suppression of protein synthesis is mediated by a translational block affecting both initiation and elongation. Inhibition of initiation is not yet adequately analyzed, but inhibition ˆ , an elongation factor of elongation appears to depend upon an accumulation of EF1A which at low pH appears to form nonfunctional complexes with polysome-associated mRNA (28). The main role for this elongation factor is to present amino acyl-tRNA to the A site of ribosomes, so it is easy to visualize how failure to dissociate from polysomes would prevent peptidyl synthase and translocation. In hypoxia-sensitive systems, such as rat hepatocytes (27), hypoxia-induced translational arrest and thus blockade of protein synthetic capacity seems to remain general for at least 2 hours. The acute or defense phase of hypoxia tolerance is postulated to blend smoothly into a secondary series of processes; in the literature these are variously termed immediate-early gene responses (40) or acclimatory expression adjustments (2,3). As the combined effect of these processes is to reactivate some mRNA translational capacities and probably to consolidate and stabilize the cell at strikingly reduced ATP turnover rates (at Kjell Johansen’s ‘‘pilot light’’), these combined processes are referred to as a rescue phase for establishing hypoxia tolerance. The rescue phase includes: 1. Heme protein-based, hypoxia-sensing, and signal transduction pathways that activate one-cycle gene expression for the production of key transcription factors (such as HIF1); one-cycle gene regulation may also allow continued production of key elongation factors and so seemingly ‘‘rescue’’ the cell translational capacities for specific mRNAs during continued oxygen limitation. 2. Sets of two-cycle, hypoxia-sensitive genes (probably including genes for glycolytic enzymes) whose expression is upregulated during prolonged oxygen limitation (protein products of these genes are presumably involved in stabilizing cell operations at severely suppressed ATP turnover ˆ could be regulated by such tworates during hypoxia; the gene for EF1A cycle, rather than one-cycle, gene-activation circuits). 3. Sets of two-cycle, hypoxia-regulated genes (possibly including genes for

136

Hochachka

4.

enzymes in gluconeogenesis and the Krebs cycle) whose expression is downregulated during prolonged oxygen limitation (protein products of these genes presumably are not needed, or are not as critical, for survival during prolonged oxygen lack). Sets of more complex, two- or three-cycle, hypoxia-regulated genes (such as cfos and cjun) whose products appear as tertiary messengers in the expression regulation of (probably numerous) other housekeeping genes (involved in restructuring, consolidation, and stabilization of cell functions at the severely suppressed ATP turnover rates requisite for surviving prolonged hypoxia). These latter, more complex regulatory phenomena make it easier to appreciate features of the oxygen-limited cell such as suppressed proteolysis, increased protein stability, and induction of chaperone proteins, features again indicative of extensive reprogramming of the normoxic cell to generate the hypoxia-tolerant cell.

From our currently available data base, it appears that the combination of these molecular processes allows cells and tissues of hypoxia-tolerant species to greatly extend the length of time they are able to survive under hypoxic or even anoxic conditions. However, it is worth emphasizing that summaries only supply key highlights and key processes of hypoxia tolerance. Further studies are required for evaluating how useful these studies of good facultative anaerobes are in explaining hypoxia responses observed in humans under mild or severe hypoxia. Some of the processes (such as downregulation of protein synthesis and adjustments in ATPgenerating metabolic pathways) may well be similar in the two very different kinds of biological systems; others, such as three-cycle gene-mediated cell reprogramming, may occur to a lesser extent (or not at all) in humans in hypobaric hypoxia. Further studies are also required to fill in the details of such generalized frameworks of the nature of hypoxia tolerance. Finally, and perhaps most daunting of all, is the challenge of assessing whether or not these strategic mechanisms can be transferred from hypoxia-tolerant to hypoxia-sensitive cells. This—a kind of Holy Grail for many workers in the field—supplies young researchers with a good target for the next millennium. Acknowledgments This work was supported by an NSERC Research Grant. Special thanks are extended to Dr. Michael Vayda for bringing his work on hypoxia-sensitive protein synthesis to our attention. References 1. Storey KB, Hochachka PW. J Biol Chem 1974; 249:1423–1429. 2. Hochachka PW, Somero GN. Biochemical Adaptation. Princeton, NJ: Princeton University Press, 1984:1–521.

Defense and Rescue Mechanisms

137

3. Hochachka PW, Guppy M. Metabolic Arrest and the Control of Biological Time. Cambridge, MA: Harvard University Press, 1987:1–237. 4. Storey KB, Storey JM. Q Rev Biol 1990; 65:145–193. 5. Hand SC. In: Hochachka PW, Lutz PL, Sick T, Rosenthal M, van den Thillart G, eds. Surviving Hypoxia. Boca Raton, FL: CRC Press, 1993:171–185. 6. Guppy M, Fuery CJ, Flanigan JE. Comp Biochem Physiol 1994; 109B:175–189. 7. Hochachka PW. In: Hochachka PW, Lutz PL, Sick T, Rosenthal M, van den Thillart G, eds. Surviving Hypoxia. Boca Raton, FL: CRC Press, 1993:127–135. 8. Hochachka PW. In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and Molecular Medicine. Queen City Printers, Inc., Burlington, VT: 1993:146–155. 9. Hochachka PW. Muscles as Molecular and Metabolic Machines. Boca Raton, FL: CRC Press, 1994:1–157. 10. Hochachka PW. Science 1986; 231:234–238. 11. Hand SC. In: Gilles R, ed. Advances in Comparative and Environmental Physiology. New York: Springer-Verlag, 1991:1–47. 12. Land SC, Bernier NJ. In: Hochachka PW, Mommsen TP, eds. Biochemistry and Molecular Biology of Fishes. Amsterdam: Elsevier Science, 1995:381–405. 13. Doll CJ. In: Hochachka PW, Lutz PL, Sick T, Rosenthal M, van den Thillart G, eds. Surviving Hypoxia. Boca Raton, FL: CRC Press, 1993:389–400. 14. Doll CJ, Hochachka PW, Reiner PB. Am J Physiol 1991; 261:R1321–R1324. 15. Doll CJ, Hochachka PW, Reiner PB. Am J Physiol 1993; 265:R929–R933. 16. Doll CJ, Hochachka PW, Hand SC. J Exp Biol 1994; 191:141–153. 17. Buck LT, Hochachka PW. Am J Physiol 1993; 265:R1020–R1025. 18. Buck LT, Hochachka PW, Schon A, Gnaiger E. Am J Physiol 1993; 265:R1014–R1019. 19. Land SC, Buck LT, Hochachka PW. Am J Physiol 1993; 265:R41–R48. 20. Land SC, Hochachka PW. Am J Physiol 1994; 266:C1028–C1036. 21. Land SC, Hochachka PW. Proc Natl Acad Sci USA 1995; 92:7505–7509. 22. Thurman RG, Nakagawa Y, Matsumura T, Lemasters JJ, Misra UK, Kauffman FC. In: Hochachka PW, et al., eds. Surviving Hypoxia—Mechanisms of Control and Adaptation. Boca Raton, FL: CRC Press, 1993:329–340. 23. Lutz PL. Ann Rev Physiol 1992; 54:619–637. 24. Sick TJ, Perez-Pinon M, Lutz PL, Rosenthal M. In: Hochachka PW, Lutz PL, Sick T, Rosenthal M, van den Thillart G, eds. Surviving Hypoxia. Boca Raton, FL: CRC Press, 1993:351–363. 25. Nilsson G. In: Hochachka PW, Lutz PL, Sick T, Rosenthal M, van den Thillart G, eds. Surviving Hypoxia. Boca Raton, FL: CRC Press, 1993:401–413. 26. Chih CP, Rosenthal M, Lutz PL, Sick TJ. Am J Physiol 257, R854–R859. 27. Buc-Calderon P, Lefebvre V, van Steenbrugge M. In: Hochachka PW, Lutz PL, Sick T, Rosenthal M, van den Thillart G, eds. Surviving Hypoxia. Boca Raton, FL: CRC Press, 1993:271–280. 28. Vayda ME, Shewmaker CK, Morelli JK. Plant Mol Biol 1995; 28:751–757. 29. Eckardt KU. Nephron 1994; 67:7–23. 30. Firth JD, Ebert BL, Pugh CW, Ratcliffe PJ. Proc Natl Acad Sci USA 1994; 91:6496– 6500. 31. Firth JD, Ebert BL, Ratcliffe PJ. J Biol Chem 1995; 270:21021–21027. 32. Wang GL, Semenza GL. J Biol Chem 1995; 270:1230–1237. 33. Semenza GL, Roth PH, Fang FM, Wang GL. J Biol Chem 1994; 269:23757–23763. 34. Maxwell PH, Pugh CW, Ratcliffe PJ. Proc Natl Acad Sci USA 1993; 90:2423–2427.

138

Hochachka

35. Wang GL, Semenza GL. Proc Natl Acad Sci USA 1993; 90:4304–4308. 36. Keitzmann T, Schmidt H, Probst I, Jungermann K. FEBS Lett 1992; 311:251–255. 37. Keitzmann T, Schmidt H, Unthan-Feschner K, Probst I, Jungermann K. Biochem Biophys Res Comm 1993; 195:792–798. 38. Webster KA, Murphy BJ. Can J Zool 1988; 66:1046–1058. 39. Murphy BJ, Robin ED, Tapper DP, Wong RJ, Clayton DA. Science 1984; 223:707– 709. 40. Webster KA, Discher DJ, Bishopric NH. Circ Res 1994; 74:679–686. 41. Webster KA, Discher DJ, Bishopric NH. J Biol Chem 1993; 268:16852–16858. 42. Prabhakar NB, Chenoy BC, Cherniak NS. FASEB J 1995; 9(4):3769. 43. Murphy BJ, Laderoute KR, Short SM, Sutherland RM. Br J Cancer 1991; 63:69–73. 44. Benjamine IJ, Horie S, Greenberg ML, Alpern RJ, Williams RS. J Clin Invest 1992; 89:1685–1689. 45. Ransome LJ, Verma IM. Ann Rev Cell Biol 1990; 6:539–557. 46. Schlingensiepen KH, Wollnik F, Kunst M, Schlingensiepen R, Herdegen T, Brysch W. Cell Mol Neurobiol 1994; 14:487–505. 47. Bergelson S, Pinkus R, Daniel V. Cancer Res 1994; 54:36–40. 48. Das DK, Engelman RM, Kimura Y. Cardiovasc Res 1993; 27:578–584. 49. Hermes-Lima M, Storey KB. Am J Physiol 1993; 265:R646–R652. 50. Moffat GJ, McLaren AW, Wolff CR. J Biol Chem 1994; 269:16397–16402. 51. Bedoya FJ, Flodstrom M, Eizirik DL. Biochem Biophys Res Comm 1995; 210:816– 821. 52. Rice ME, Lee EJK, Choy Y. J Neurochem 1995; 64:1790–1799. 53. Jackson DC. In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and Molecular Medicine. Burlington, VT: Queen City Printers, Inc., 1993:119–126. 54. Venkatachalam MA, Winberg JM. In: Hochachka PW, Lutz PL, Sick T, Rosenthal M, van den Thillart G, eds. Surviving Hypoxia. Boca Raton, FL: CRC Press, 1993:473– 494.

6 Control of Breathing at High Altitude

CURTIS A. SMITH and JEROME A. DEMPSEY University of Wisconsin Madison, Wisconsin

I.

THOMAS F. HORNBEIN University of Washington School of Medicine Seattle, Washington

Introduction

This chapter will focus on the mechanisms of ventilatory responses to hypoxia from the initial responses of the first seconds to minutes of hypoxic exposure up to many years of residence in hypoxic environments. We will discuss how breathing may influence performance at extreme altitude (see also Chapter 20). Space requires that our approach be a selective one. We will not discuss ventilatory responses to hypoxia comprehensively and/or historically, as a number of excellent reviews of this topic are available (1–6). Rather, we will attempt to highlight what we believe are new, unresolved, or controversial questions in this area that would benefit from further research. We will only touch upon control of breathing during sleep at altitude and genetic/evolutionary considerations of ventilatory control in ethnic groups that have been resident at high altitudes for centuries, as these topics are discussed in Chapters 3 and 21.

139

140

Smith et al. II. Background and Definitions

Upon ascent to altitude or exposure to hypoxia, awake mammals attempt to protect tissue oxygenation. Initially, compensation is primarily via hyperventilation and some selective vasodilation. Later, changes in hematocrit, renal mechanisms, and tissues may also occur. Three distinct phases of the ventilatory response to hypoxia are generally recognized: 1.

2.

3.

Acute hyperventilatory responses, those occurring in seconds to minutes, are thought to be mediated primarily by peripheral chemoreceptors in conjunction with secondary inhibitory feedback from central chemoreceptors. Responses occurring over hours to days are usually termed short-term acclimatization or, simply, acclimatization. Deacclimatization is the timedependent return to eupneic ventilation following a return to normoxic conditions and may be of importance in understanding the acclimatization process. The mechanisms involved in acclimatization and deacclimatization have been long debated and remain controversial to this day but may include both central and peripheral processes. Responses occurring over weeks to years represent long-term acclimatization. These responses are observed in long-term altitude sojourners and those who are native to high altitudes.

There may also be extremely long-term, evolutionary adaptations in populations that have resided at altitude for generations, but this question is beyond the scope of this review (see Chapter 3).

III. Acute Responses to Hypoxia The acute (seconds to minutes) ventilatory responses to hypoxia are the net result of several known and putative feedback mechanisms. During this interval, most studies in unanesthetized humans, goats, and cats found that ventilation increased to a peak (magnitude was dependent on the severity of the hypoxia). This increase in ventilation results chiefly from hypoxic stimulation of the peripheral chemoreceptors, primarily the carotid bodies. Immediately following this increase, ventilation declines somewhat over the subsequent 10–30 minutes, although remaining above control levels under both poikilocapnic and isocapnic conditions (7–13). However, other studies in the dog and the goat found no evidence of a ventilatory decline during this period (14,15). This decline in ventilation (hypoxic ventilatory decline, or HVD, termed ‘‘roll-off’’ by some workers) is somewhat surprising, and its mechanism is the subject of considerable investigation. While the decline is relatively small and short-lived when contrasted with ventilatory acclimatization (Fig. 1), we have chosen to emphasize this topic more than is customary because we felt that

Control of Breathing

141

Figure 1 Idealized time course of ventilatory acclimatization (using Paco 2 or Petco2 as ˙ A) compiled from five different studies in humans. SL ⫽ Sea level; HVD ⫽ indices of V hypoxic ventilatory decline. (Adapted from Refs. 10, 41, 120, 122, 124.)

an understanding of the mechanisms underlying HVD might provide useful insights into the broader topic of ventilatory adaptation to hypoxia.

A. Carotid Body Hypoxia

The only well-established stimulatory mechanism operating in the acute time period is the increased neural output from the peripheral chemoreceptors, chiefly the carotid bodies (see also Sec. III.D). Following a few second overshoot (16), carotid chemoreceptor output remains virtually constant (Fig. 2) during the period of ventilatory decline (17–19); therefore, the decline in ventilation is not due to a decline in sensor output.

B. Hypocapnia

As noted above, hypoxic-induced hyperventilation results in hypocapnia, which may itself diminish breathing by several possible mechanisms.

142

Smith et al.

Figure 2 Comparison of ventilation and carotid sinus nerve activity in one anesthetized cat in response to isocapnic hypoxia. Note the essentially constant nerve activity during hypoxic ventilatory decline. (From Ref. 137.)

C. Central Hypocapnia at the Central Chemoreceptors Central Hypocapnia via Increased Cerebral Blood Flow

In naturally occurring (poikilocapnic) hypoxia, Pao 2 increases and Paco 2 decreases over time. During the hypoxic ventilatory decline period, this trend will reverse for a time, which should tend to stimulate, rather than inhibit, ventilation. Despite these systemic changes, hypocapnia at the level of the central chemoreceptors, enhanced by hypoxic cerebral vasodilatation, could inhibit ventilation (20). The difficulty here, as Neubauer et al. have pointed out, is that the hypocapnia secondary to the cerebral vasodilation reaches a steady state in about 5 minutes and therefore does not correlate with the time course of hypoxic ventilatory decline (21–24). Peripheral Hypocapnia

The hyperventilation that results from natural hypoxic exposure causes systemic hypocapnia, and the major inhibitory effect of this hypocapnia has usually been assumed to be mediated by the central chemoreceptors. However, this hypocapnia is also sensed by the carotid body. Studies of hypoxic ventilatory decline using isocapnic conditions have demonstrated that hypocapnia is not obligatory for its development (8–10,25). Indeed, HVD is more readily demonstrated under isocapnic

Control of Breathing

143

Figure 3 Time course of response of the ventilatory response to specific carotid body hypocapnia or hyperoxia in awake dogs using the vascularly isolated carotid body perfusion technique. Note that the time course and magnitude of the decrease in ventilation is virtually identical whether the inhibition was caused by hypocapnia (about 13 torr less than eupneic value) or hyperoxia. (From Ref. 29.)

conditions. Several studies have produced estimates of 10–50% for the peripheral contribution to the ventilatory response to CO2 (26–28). Although hypocapnia at the level of the carotid body does inhibit ventilation, even in the presence of hypoxia (Fig. 3) (29,30), carotid body hypocapnia is probably not directly related to hypoxic ventilatory decline because (1) the time course of hypoxic ventilatory decline does not correlate with changes in systemic Paco2 and (2) hypoxic ventilatory decline can still be observed under isocapnic conditions. D. A Role for Hypoxic Depression?

Hypoxic depression of neurons involved in ventilatory control is frequently invoked to explain hypoxic ventilatory decline. Depression by hypoxia is seen consistently in anesthetized preparations, usually with CO-induced hypoxemia and with Paco 2 and blood pressure maintained constant (e.g., Refs. 31–33); however, the situation in awake animals is far less clear. In contrast to anesthetized preparations, awake carotid body denervated dogs, rats, ponies, sheep, and goats either showed no change in ventilation or actually hyperventilated when exposed to hypoxia (33–37). This ventilatory stimulation has been shown over a wide range of Pao 2 (30–50 torr) and time (minutes to days). Utilizing extracorporeal circulation to isolate the carotid

144

Smith et al.

bodies of awake goats, Engwall et al. observed that brain hypoxia caused hyperventilation even when the carotid body was maintained normoxic and normocapnic (Fig. 4) (38). Similar results have been obtained recently by Curren et al. (38a) during non-REM sleep in the dog. In the foregoing studies the hyperventilation was due to increases in both Vt and fb. This response clearly differed from the frank tachypnea described in carotid body denervated cats (39) in which hypoxic depression of higher brain centers was hypothesized to disinhibit the respiratory centers in the medulla. In addition, Engwall et al. (38) showed that both inspiratory and expiratory muscle EMGs increased such that the ventilatory response to brain hypoxia was indistinguishable from the ventilatory response to moderate whole-body hypercapnia. Further, the ventilatory response to hypoxic stimulation of the isolated carotid body was the same regardless of whether the brain was hypoxic or normoxic. This latter finding has also been observed in an anesthetized preparation (31) even when hypoxic depression was present. Finally, Dahan et al. (40) have demonstrated in human subjects that peripheral chemoreceptor input seems to be required for the development of HVD. All these observations lead one to doubt that hypoxic depression of the respiratory controller contributes to HVD. Indirect measures of central nervous system (CNS) neuronal function suggest

Figure 4 Hyperventilation in an unanesthetized goat in response to brain hypoxia; carotid bodies were maintained normoxic and normocapnic by means of extracorporeal perfusion of vascularly isolated carotid bodies. Open circle ⫽ brain normoxia; filled square ⫽ brain hypoxia. Note: (1) Increased ventilation in response to brain hypoxia is additive with superimposed carotid body hypoxia. (2) The time-dependent poststimulus hyperpnea indicative of STP was unaffected by brain hypoxia. (Adapted from Ref. 38.)

Control of Breathing

145

that ventilatory acclimatization accompanies a general hyperexcitability (41–43). Recent evidence from tissue slice preparations of medulla and hypothalamus shows that most neurons in these regions were excited by hypoxia (44–49). Presently, our ability to extrapolate these results to the whole animal is limited because of the uncertainty of the exact cell types recorded from and the relatively short durations of hypoxia employed. Another mechanism affected by acute hypoxia is the ventilatory manifestation of short-term potentiation (STP). STP is a property of central neurons in which output increases over time (seconds) in the face of a constant input or stimulus, and this increased output persists for a time (‘‘afterdischarge,’’ manifested in ventilation as a continued but waning hyperpnea) when the stimulus is removed (50–52). In the case of ventilation, it is an excitatory influence that tends to counteract transient inhibitory feedback and stabilize breathing. In sleeping humans, if the duration of hypoxic exposure was ⬍1 minute (53), ventilation returned to normal over the next few breaths, i.e., the usual poststimulus hyperpnea indicative of STP was still present. However, in the same subjects, hypoxic exposures of ⬎5 minutes that were either isocapnic or hypocapnic resulted in diminished or absent poststimulus hyperpnea. In contrast, when awake goats were exposed to systemic hypoxia (including brain) for many minutes and square-wave changes in Po 2 (with normocapnia maintained) were imposed at the isolated, perfused carotid body, there was no effect on the poststimulus hyperpnea unless systemic CO2 was allowed to fall (Fig. 4) (38). Similar observations were made in the anesthetized, normocapnic cat with COinduced hypoxemia (32). While the goat data cited above are the most direct evidence we are aware of suggesting a lack of effect of CNS hypoxia on STP in unanesthetized animals, species differences cannot be ruled out. If STP is affected by hypoxia, it raises some intriguing questions: (1) Are STP and hypoxic ventilatory decline linked to some common mechanism? (2) If so, will STP return over time as acclimatization proceeds? In summary, moderate levels of CNS hypoxia probably have a net stimulatory effect on the respiratory control centers in unanesthetized animals. On the cellular level many CNS neurons have been shown to depolarize in the presence of hypoxia, and there is a plausible mechanism to explain this depolarization. It is not yet clear if STP is affected by CNS hypoxia in unanesthetized animals, but the best available data suggest it is not. In sum, it seems unlikely that a direct depressive effect of hypoxia on the CNS can account for hypoxic ventilatory decline (see Sec. III.E). E. Changes in Central Neuroeffector Levels

A fourth, and perhaps the most likely, mechanism of hypoxic ventilatory decline is the build-up of inhibitory neurotransmitters and/or neuromodulators, and/or the decline of excitatory ones, in the CNS. Recent evidence in support of this idea has been provided by Long et al. (25), who showed that the hypoxic ventilatory decline was immediately apparent when awake cats were reexposed to hypoxia after 5 minutes of normoxia. They suggested that their findings were consistent with a hypoxia-

146

Smith et al.

induced change in the level of some neuroeffector that remained altered for some minutes following return to normoxia. Georgopoulos et al. (54) made similar observations and came to similar conclusions in studies of awake humans made hypoxic (Sao 2 ⫽ 80%) at three different CO2 levels, returned to normoxia for a few minutes, and then reexposed to the same level of isocapnic hypoxia. Several neuroeffector candidates could potentially explain this persistent ventilatory depression with repeated hypoxic exposure. GABA

Gamma-aminobutyric acid (GABA) is a known inhibitory neurotransmitter in the CNS (55). Central GABA levels have been shown to increase in response to hypoxia (56–59). GABA and GABA analogs have been shown to depress ventilation in anesthetized animals (60–63). GABA antagonists have been shown to reverse partially the ventilatory depression of hypoxia in sedated piglets (64) and anesthetized cats (65). Yamada et al. (66) have shown in anesthetized cats that GABA can exert inhibitory effects on ventilation when applied directly to Schlaefke’s area on the ventrolateral medulla. Thus, GABA looks like a good candidate to mediate hypoxic ventilatory decline. However, there are some caveats: rather extreme hypoxia is required to elicit increases in GABA levels. Most studies find a critical Fio 2 of about 0.08 (56,57). One study of neonatal rats (58) reported increases in GABA at milder levels of hypoxia (Fio 2 ⫽ 0.12), but a stimulus-response relationship appeared to be lacking, as a Fio 2 of 0.06 elicited no further increase in GABA levels. There are well-known differences in neonatal vs. adult responses to hypoxia, and this may account for the discrepancy. In general, however, the extreme hypoxia that seems to be required in the adult mammal causes one to question the relevance of GABA in nonpathological situations at terrestrial altitudes. Another problem is that, to our knowledge, no studies of ventilatory responses to GABA agonists or antagonists have been performed in unanesthetized preparations. GABA antagonists such as bicuculline have convulsant properties that make them difficult to use in the unanesthetized preparation. Studies in the unanesthetized animal might help elucidate the role of GABA in hypoxic ventilatory decline. Adenosine

Adenosine is another neuromodulator that could be involved in hypoxic ventilatory decline. Adenosine has been shown to have an inhibitory effect on CNS corticospinal neurons (67,68). In the fetal sheep brain adenosine concentrations have been shown to increase with a time course that is compatible with the development of hypoxic ventilatory decline when the dam breathed 0.09 Fio 2 (69). In anesthetized cats and rats Eldridge et al. (70) and Hedner et al. (71) showed that intracerebroventricular administration of an adenosine agonist inhibited ventilation. Similar observations have been obtained with microdialysis techniques in the piglet (72). In anesthetized preparations, adenosine antagonists partially reduced the hypoxic ventilatory decline (73–75). Similar observations were reported in studies of unanesthetized human subjects (76). However, adenosine had no effect on HVD in awake adult goats (77).

Control of Breathing

147

In summary, the bulk of the evidence suggests a role for adenosine in HVD, but additional studies in unanesthetized models would seem to be required before this can be firmly established. Lactic Acid

Brain acidosis is usually thought of as a potent stimulus to ventilation (also see Sec. IV). Several studies, however, have demonstrated that decreased pH in the extracellular fluid around the ventral medullary surface can inhibit ventilation (22,75) and that pretreatment with dichloroacetate (a blocker of lactic acid production) eliminated ventilatory depression until very low oxygen contents were reached (78). Relating these results directly to hypoxic ventilatory decline is problematic for two reasons: (1) the studies cited here generally required rather severe degrees of hypoxemia to elicit decreases in pH and (2) Paco 2 was kept normal in these studies. Hypocapnia and hypoxia are synergistic in the production of lactic acid. Musch et al. (79) showed, in rats exposed to moderate-to-severe hypoxia, that brain tissue lactacidosis could be prevented (moderate hypoxia) or reduced (severe hypoxia) if hypocapnia was prevented. Inasmuch as the normal physiological response to hypoxia is hyperventilation, the lactic acid–mediated inhibition of ventilation might occur even sooner under physiological (i.e., hypocapnic) conditions. More recently, Aaron et al. (80) found that dichloroacetate enhanced the hyperventilatory response when it was administered to unanesthetized goats exposed to moderate hypoxia. These observations are consistent with a depressive role for lactic acid. Severinghaus (81) has put forward a hypothesis to explain the role of lactic acid in HVD. He suggests that the chemosensitive neurons near the ventral medullary surface readily produce lactic acid when exposed to hypoxia (82,83). During the initial hypoxic exposure when ventilation is increased as a result of the hypoxic stimulus at the carotid bodies, the chemosensitive medullary neurons produce lactic acid and thereby increase intracellular [H⫹]. This increased [H⫹] reduces the extracellular fluid (ECF)-to-intracellular fluid (ICF) [H⫹] gradient and tends to hyperpolarize the neurons. The reduced chemosensor output causes a reduction in ventilation that results in an increased systemic Pco 2. Due to the poor buffering of ECF relative to ICF, this increased Pco 2 tends to restore the ECF-ICF [H⫹] gradient and normalize breathing. This is an attractive hypothesis that could account for the pattern of ventilatory stimulation/inhibition/recovery that characterizes HVD. A potential limitation of this hypothesis is that it does not explain why carotid bodies appear to be required for the development of HVD. Endogenous Opioids

Inhibition of ventilation by endogenous opioids has been clearly demonstrated in the neonate (84,85) and is thought to be a vestigial fetal response to the stress of hypoxia (20). In the unanesthetized adult, however, no role for endogenous opioids has been found (11,13,86,87). Accordingly, we will not elaborate on this topic but refer the reader to two excellent reviews on the subject (22,88).

148

Smith et al. Endogenous Benzodiazepines

We are aware of only one study dealing with the role of endogenous benzodiazepines in the ventilatory response to hypoxia (89). These authors found that inhibition of benzodiazepine receptors with a specific antagonist had, if anything, a slight inhibitory effect on ventilation during hypoxia. Thus, these receptors do not appear to be a likely candidate for the mediation of hypoxic ventilatory decline. In summary, there seems little doubt that changes in excitatory and inhibitory neuroeffectors occur. The essential issues are what causes these changes and how significant they are. Possible mechanisms include a direct effect of hypoxia on CNS neurons and/or an effect related to magnitude and duration of input from the peripheral chemoreceptors.

F. Metabolic Rate

A fifth potential cause of hypoxic ventilatory decline is a decrease in metabolic rate secondary to systemic hypoxia. What this means is that ventilation could decrease secondary to a hypoxia-induced fall in metabolic rate without hypoventilation. This effect is known to exist in neonates of most mammalian species and adults of most species smaller than about 5 kg (90,91). The effect decreases progressively as mass ˙ o 2 occurred increases and specific metabolic rate decreases (91). The decrease in V very rapidly (91) and remained stable during the period of hypoxic ventilatory de˙ co 2 may change more slowly, however, for reasons that are not clear but cline. V may be related to CO2 stores. In humans exposed to moderate hypoxia metabolic rate remained normal (e.g., Ref. 92). Similar observations have been obtained in ponies (35) and goats (93). Thus, it is difficult to invoke hypometabolism to explain hypoxic ventilatory decline in species that are as large as human adults. An important caveat here is that in neonates and smaller mammals, hypometabolism and the dis˙ o 2 and V ˙ co 2 may affect at least the parity between the time course of changes in V first few minutes of the response to hypoxia, which could raise significant interpretive problems. In addition, in the rat at least, there is a marked increase in Vd that has reached a plateau by 1 hour of exposure to hypoxia (94). Therefore, some index ˙ a would appear to be essential in studies of hypoxic ventilatory decline, particuof V larly in small mammals. In summary, in the absence of a change in Paco 2 or Pao 2, the carotid body chemoreceptors maintain an essentially constant output during the period of hypoxic ventilatory decline, suggesting that, although the carotid bodies are essential for hypoxic ventilatory decline to occur, the decline must depend on CNS processes. Considerable potential for redundancy and interaction of central mechanisms exists; some or all may play a role in hypoxic ventilatory decline, and the mechanisms that are employed may well change depending on the severity of hypoxia. A major unanswered question here is the significance of hypoxic ventilatory decline in the overall process of ventilatory acclimatization.

Control of Breathing

149

IV. Short-Term Acclimatization Short-term acclimatization is the progressive increase in ventilation that ensues over hours to days of hypoxic exposure (Fig. 1). While the time course of this response varies between species, the essential features are the same in all mammals that have been studied. These are (1) a progressive hyperventilation that reaches a plateau in hours to days, (2) persistent alkalosis in the blood and bulk CSF, and (3) continued hyperventilation upon return to normoxia that diminishes slowly over hours to days (see Sec. IV.C). Despite extensive research since the nineteenth century, the mechanism(s) underlying acclimatization remains poorly understood, although important advances have been made in the past decade. A. Carotid Bodies

It is often assumed that carotid body sensitivity increases over time during hypoxic exposure. This assumption is based on a number of studies of hypoxic ventilatory response in humans before and after several days of hypoxic exposure (95–99). Increased sensitivity could also be inferred simply from the increased ventilation in the face of increasing Pao 2 and decreasing Paco 2 as acclimatization proceeds. Unfortunately, studies of this type do not tell us if the carotid chemosensor’s sensitivity is increasing, if the central integrating mechanism is becoming sensitized, or both. More specific studies are required to elucidate these questions. Considerable evidence has accumulated to show that the carotid bodies are required for acclimatization to hypoxia. Carotid body–denervated, unanesthetized dogs, goats, ponies, and sheep have been shown to not acclimatize to hypoxia (33,35–37,93). Evidence suggests that the carotid bodies could account completely for ventilatory acclimatization. Using the vascularly isolated, perfused carotid body technique in the unanesthetized goat, Busch et al. and Bisgard et al. separated the circulation to the carotid body from the rest of the systemic circulation (including brain and medullary chemosensitive and integrating structures) (100,101). When they rendered the carotid body hypoxic for several hours while maintaining the rest of the body normoxic (either poikilocapnically or isocapnically), they observed essentially normal acclimatization. From these data they concluded that brain hypoxemia or hypocapnia were not required for acclimatization (see Sec. IV.C). Weizhen et al. (102) used identical techniques to extend these observations by maintaining the carotid bodies normoxic and normocapnic while the brain was maintained hypoxic. They showed that prolonged CNS hypoxia alone did not promote acclimatization, i.e., upon return to normoxia Paco 2 returned to its original control value (Fig. 5). Normally, of course, Paco 2 is falling and Pao 2 is increasing over time of hypoxic exposure; possible inhibitory effects of these changing stimuli at the CNS remain to be determined. At the level of the carotid body, recent work in the unanesthetized goat with vascularly isolated and exogenously perfused carotid body (15) has shown that carotid body hypoxic sensitivity to isocapnic hypoxia increases over time of

150

Smith et al.

Figure 5 Lack of acclimatization in response to 4 hours of brain hypoxia with carotid bodies held normoxic and normocapnic using the vascularly isolated carotid body perfusion technique in awake goats. Note the stable ventilation over the period indicating no acclimatization. (From Ref. 102.)

hypoxic exposure, and the level of Pco 2 at the carotid body seems to have no effect on this increase. In summary, carotid body hypoxia alone is capable of causing acclimatization. Is this time-dependent increase in ventilation due to changes at the medullary integrator or at the carotid body? It has been shown that carotid sinus nerve activity increases over time when hypoxia is the stimulus (Fig. 6) (103) but not when CO2 is the stimulus (104). At the cellular and molecular level, chronic hypoxia has been shown to increase excitability in cultured rat glomus cells associated with differential changes in ionic currents (105). In terms of ventilatory response, at the level of the carotid body hypoxia is the only stimulus that will cause acclimatization. It has been demonstrated that long-term hypercapnia at the carotid body (using the isolated carotid body perfusion technique in the unanesthetized goat) stimulates breathing but does not cause acclimatization (106). This observation suggests that the carotid body can differentiate between stimuli and encode this information into afferent carotid sinus nerve traffic. While a discussion of carotid body transduction processes is beyond the scope of this chapter (see Chapter 5 and Ref. 107), we do note that attempts to find a difference in carotid sinus nerve discharge patterns, which might relate to how O2 versus CO2 are encoded in the carotid sinus nerve, have so far been unsuccessful (108). Many neurochemicals could be involved in the carotid body chemoreceptor responses to hypoxia. Space does not permit us to discuss this extensively; an excellent review of this material can be found in Bisgard and Forster (6). A few key points should be noted, however.

Control of Breathing

151

Figure 6 Response of single carotid body chemoreceptor afferents to prolonged isocapnic hypoxia. The control discharge rate is indicated at time 0. Note the progressive increase in activity over time. (From Ref. 136.)

Dopamine and norepinephrine appear to be the predominant neurochemicals in the carotid body and have been the most extensively investigated. Whether or not dopamine or norepinephrine play a role in carotid body chemoreceptor-mediated ventilatory acclimatization to hypoxia is controversial, as there is negative evidence from the goat (summarized in Ref. 6) and positive evidence from the cat (14,109). Regardless of whether there is a role for either of these neurochemicals, there is a clear link between hypoxia and the genetic regulation of catecholamine production. Czyzyk-Krzeska et al. (110) showed that hypoxia regulates the gene that codes for tyrosine hydroxylase, the rate-limiting enzyme in the production of catecholamines, in dopaminergic cells of the mammalian carotid body. Other neurochemicals present in the carotid body could also be involved in the acclimatization process. These include nitric oxide, substance P, neurokinin A, enkephalins, VIP, CGRP, atrial natruretic peptide, and neuropeptide Y (111–114). To date, little is known about the role of these compounds in the carotid body. A number of key questions remain unanswered with respect to these carotid body neurochemicals: Is there a role for these neurochemicals in ventilatory acclimatization, especially the noncatecholamines? Are there species differences in response to these neurochemicals? What are their mechanisms of action at the receptor level? Do different neurochemicals interact, and, if so, how? B. Central Nervous System

The fact that carotid bodies could account for ventilatory acclimatization does not rule out a role for processes in the brain, but it does appear that carotid bodies

152

Smith et al.

are mandatory for manifestation of any central acclimatization processes. Central processes may be more important under nonlaboratory conditions where Pao 2 is increasing and Paco 2 is decreasing over time as acclimatization progresses. Several processes could contribute to ventilatory acclimatization. Sensitization by Peripheral Afferent Inputs

Afferent input from peripheral chemoreceptors may contribute to increased respiratory motor output over time. Dwinell et al. (15) showed in the goat with a vascularly isolated and perfused carotid body that carotid body hypoxic sensitivity was elevated after 4 hours of hypoxic carotid body perfusion. Millhorn et al. (115), in anesthetized, vagotomized, and paralyzed cats, showed that serotonin antagonists blocked a long-term facilitation of phrenic activity that occurred secondary to carotid sinus nerve stimulation. They suggest that the time course of this facilitation is such that it could contribute to ventilatory acclimatization to hypoxia, thus implicating serotonin in this process (see Sec. IV.C). However, Olson (116) has shown that serotonin depletion by parachlorophenylalanine in unanesthetized rats had no effect on ventilatory acclimatization, thus questioning the conclusions of Millhorn et al. Critical comparisons between these two studies are difficult because (1) serotonin levels change over time in hypoxia, (2) serotonin levels may not reflect turnover at the relevant sites, (3) all serotonin pools may not be equally accessible to pharmacological agents (117,118), and (4) species differences and presence of anesthesia may also play a role. Further experiments to clarify the role of serotonin would seem to be required. In the context of enhanced central respiratory activity secondary to increased carotid chemoreceptor input, it is interesting to note that recent evidence has shown that nitric oxide generated by increased carotid sinus nerve input to the nucleus tractus solitarius can act as a retrograde messenger in an l-glutamate–releasing positive feedback system that serves to augment ventilation during hypoxia (119). Could this process contribute to some of the time-dependent increase in ventilation observed during acclimatization (i.e., in addition to direct carotid body afferent inputs)? This novel idea deserves further study. CNS Acid-Base

Central nervous system acid-base changes certainly occur during the period of acclimatization to hypoxia, and it has long been thought that such changes must account for most or all of ventilatory acclimatization. However, whether or not there is in fact a role for central acid-base adjustments in acclimatization, and, if so, the precise mechanisms of action, remain highly controversial. One theory proposes that acclimatization results from regulation of the bulk CSF [H⫹] over time in hypoxia. That is, hypoxia initially results in a carotid body– mediated hyperventilation and respiratory alkalosis. The hypocapnia alkalosis in turn would tend to inhibit ventilation peripherally and, especially, centrally such that ventilatory output would be the net result of ongoing carotid body stimulation

Control of Breathing

153

and the concomitant respiratory alkalosis. In the face of continued carotid body stimulation, CSF [H⫹] would then be regulated progressively back toward normoxic control levels, restoring much of the central ventilatory drive that was lost and thereby increasing ventilation (e.g., Ref. 120). Similarly, deacclimatization (i.e., continued, but reduced, hyperventilation upon return to normoxic conditions; see below) would be accounted for by the respiratory acidosis produced in CSF when the carotid body component of ventilatory drive is abruptly removed by normoxia resulting in reduced ventilation and increased Paco 2. This continued hyperventilation would then diminish over time as bulk CSF [H⫹] was regulated back to normoxic eupneic values (121). Others have demonstrated that bulk CSF [H⫹] changes in the wrong directions over time to account for acclimatization or deacclimatization. In humans (122–124), dogs (25), and ponies (35,126) exposed to various altitudes (3100–4300 m), ventilation and [H⫹]csf were negatively correlated or uncorrelated at all time points; i.e., ventilation was increasing at a time when CSF [H⫹] was falling. Moreover, changes in CSF [H⫹] were compensated to the same extent as in arterial plasma. There are also observations in the anesthetized cat showing that [H⫹] on the medullary surface did not correlate with ventilation even though it increased during hypoxic exposure (74). When acclimatized humans and ponies were abruptly returned to normoxia and followed for up to 13 hours, [H⫹]csf (lumbar in the humans; cisternal in the ponies) again was negatively correlated with ventilation (127), i.e., ventilation continued to decrease in the face of increasing CSF [H⫹]. These findings do not rule out the possibility that the acid-base status of a compartment of the medullary interstitial fluid, which cannot be measured directly, is somehow being regulated such that neurons at some point along a CSF-to-blood [H⫹] gradient would be stimulated appropriately (128,129). However, manipulation of this gradient had no effect on acclimatization in human subjects (130). In addition, Musch et al. (79) have shown that, in rats exposed to hypoxia for up to seven days, changes in medullary intracellular pH and [lactate] did not correlate with ventilatory acclimatization. Jennings (131) has suggested another alternative, namely that the moiety that is sensed by the central chemosensors is the strong ion difference of CSF. The strong ion difference is defined as [⌺ strong cations] ⫺ [⌺ strong anions] (132). The strong ion difference is an especially important variable in CSF because, in an essentially protein-free solution like CSF, the strong ion difference and CO2 determine [H⫹] (132,133). Using this approach, Jennings has shown a correlation between CSF strong ion difference and ventilation for many of the studies cited above. He further suggests that the control of CSF strong ion difference might be mediated by angiotensin II which is known to stimulate ventilation. However, this approach does not appear to hold for the one available study of deacclimatization over periods greater than one hour (127). A re-analysis of these data done by the authors of this chapter shows essentially no correlation between CSF strong ion difference and ventilation during this interval. It should be noted that this does not necessarily invalidate Jennings’ approach but rather may point to some unique characteristics of the deacclimatization process (see Section IV.C).

154

Smith et al. C. Deacclimatization

It was long assumed that deacclimatization, i.e., continued but falling, timedependent hyperventilation following return to normoxia after acclimatization to hypoxia (Fig. 1), was simply the reverse of the acclimatization process. A considerable amount of evidence has accumulated that suggests that this is not the case. Dempsey et al. (127) pointed out that, in deacclimatizing humans and ponies, the measurable chemical stimuli to breathing were changing in the wrong directions to account for the observed continued hyperventilation (also see Sec. IV.B). More recent evidence has shown that, whether in goats acclimatized secondary to specific carotid body hypoxia or whole-body hypoxia and in the presence of either systemic poikilocapnia or isocapnia, ventilatory acclimatization occurred. However, the poikilocapnic goats showed continued hyperventilation upon return to normoxic conditions, but the isocapnic goats did not, even if the latter had also been exposed to systemic hypoxia (100,101,134,135). Carotid body afferent activity in the goat was not elevated upon return to normoxia after ventilatory acclimatization had occurred despite an apparent increase in carotid body sensitivity (136). Further, in the cat deacclimatization persisted after carotid sinus nerve section if the intact animal had first been allowed to acclimatize (137). In humans 24 hours of mild systemic hypocapnia (produced by voluntary hyperventilation) was followed by a persistent hyperventilation, and this after-effect was amplified when hypoxia accompanied the prolonged hypocapnia (138). Taken together, these studies indicate that acclimatization and deacclimatization may, at least in part, utilize different processes. Acclimatization requires carotid bodies and may be largely mediated by them, whereas deacclimatization appears to be an exclusively central process linked to prolonged CNS hypocapnia. Hypoxia may amplify the ventilatory effects but is not required. The mechanism of this central process remains obscure. Recent work in the anesthetized rat (131) showed that repeated 2-minute periods of electrical carotid sinus nerve stimulation or repeated 5-minute periods of isocapnic hypoxia resulted in long-term facilitation; i.e., increased phrenic motor output persisted for up to 30 minutes following stimulation. This long-term facilitation could be linked to increased serotonin levels, as suggested by Millhorn et al. (115) (see Sec. IV.B). However, more recently, Olson (116) has shown that deacclimatization occurred normally in serotonin-depleted rats. We must conclude that, to date, a clear role for serotonin in ventilatory acclimatization or deacclimatization has not been established. In summary, regulation of CNS [H⫹] as a mechanism of ventilatory acclimatization to hypoxia is an elegant hypothesis with considerable logical appeal. However, it has been difficult to establish any major role for CNS [H⫹] in the control of ventilatory acclimatization and deacclimatization, particularly when the entire time course of the process is considered. The apparent dependence of ventilatory deacclimatization on the presence of hypocapnia during the acclimatization period suggests that deacclimatization may have some sort of secondary link to reduced [H⫹], perhaps changes in neuromodulators and/or other sorts of plasticity, but additional study is required to resolve this question.

Control of Breathing V.

155

Long-Term Acclimatization

In some populations indigenous to or resident for many years at high altitude, the ventilatory response to hypoxia may be much diminished compared to that of the acclimatized lowlander (139,140). This ‘‘blunting’’ of the hypoxic response is also evidenced by a resting, eupneic Paco 2 1–3 mmHg higher than that of the acclimatized sojourner at the same altitude. This difference is particularly apparent during moderate to heavy exercise at high altitude, as shown by the highlander’s nearisocapnic hyperpneic response to mild through moderate exercise in contrast to the sojourner’s tachypneic hyperventilation (Fig. 7) (141). While tachypnea and dyspnea are commonly experienced during moderate exercise in the sojourner (Fig. 8) (142), these responses are only rarely experienced in the highlander (141). Furthermore, when the sojourner was suddenly made hyperoxic during exercise, ventilation was reduced precipitously and great relief from dyspnea was experienced, whereas the highlander’s ventilation was not reduced by hyperoxia until heavy work loads were reached, and they were often unaware that their hypoxemia had been eliminated (141). Another manifestation of the blunted hypoxic ventilatory response may be the relative absence of periodic breathing during sleep in hypoxia in Sherpa highlanders (see Chapter 21) (140). Interestingly, blunting appears to be less prevalent in Tibetans indigenous to altitudes ⬎3600 m than in the Andean Quechua and Aymara, suggesting a genetic/ evolutionary adaptation to high altitude (see Chapter 3) (143,144). In other highlanders, the blunting of hypoxic ventilatory response seems to result from many years of prolonged postnatal hypoxic exposure (3). In humans, the blunted response to exercise in hypoxia is achieved in native lowlanders who move to high altitude as young children (3,95). Animals native to high altitudes, such as the llama or yak (144,145), or calves born at high altitudes (146) do not exhibit blunted ventilatory responses to acute hypoxia, although these animals, like human highlanders, tend to show less hyperventilation during resting, air-breathing eupnoea. In the rat, exposure to hypoxia in the early (preweaning) neonatal period was required to achieve a blunted hypoxic response (147), but blunting could be produced after 3–4 weeks of exposure to 5500 m altitude in adult cats (148). What causes this blunted response to hypoxia? Certainly, the carotid body must be implicated. Specific pharmacological stimulation of carotid bodies in highlanders produced a ventilatory response less than in sojourners at the same altitude (149). Additional support for this idea is provided by the observation of increased size and dopamine content of carotid bodies in chronically hypoxic animals (3). Further, cats chronically exposed to severe hypoxia show blunting of both carotid sinus nerve activity and global ventilatory response to acute hypoxia (148). Dopaminergic mechanisms have been implicated (109), but opioids have been ruled out (150) as mediators of this reduced carotid body sensitivity. There are some other clues to the mechanism causing blunting: (1) ventilatory response to CO2 is not abnormal in the highlander, (2) the response to acute hyperoxia is hyperventilation in some highlanders, much as it is in a carotid body-

156

Smith et al.

Figure 7 Pulmonary gas transport during progressive exercise in the sojourner to 3100 m (⫹8 weeks) vs. the second-to-third generation native or long-term resident at 3100 m (highlanders). Note that the level of hyperventilatory response is much less in the highlander, and yet the arterial Po 2 is similar because the alveolar-to-arterial Po 2 difference is much narrower owing to the enhanced diffusion capacity. (Data from Ref. 141.)

Control of Breathing

157

˙ o 2 max) in acute hypoxia (Sao 2 Figure 8 Effects of prolonged heavy exercise (85–90% of V ˙ e that is flow limited ⫽ 80%) and normoxia (Sao 2 ⫽ 95%) on ventilation, the percent of the V during expiration, the force output of the diaphragm per minute, and ratings of dyspneic sensations (on a 10-point scale). (Data from Ref. 142.)

158

Smith et al.

denervated animal (37), and (3) midcollicular decerebration in the chronically hypoxic animal alleviates much of the blunted hypoxic response (151). These data, although incomplete, suggest that the highlander’s altered regulation of breathing is very specific to the hypoxic condition and attributable to both a reduced chemoresponsiveness of the carotid bodies per se, along with a reduced central integration of sensory afferent inputs. VI. Exercise Hyperpnea at High Altitude Any sojourner upon first arriving at high altitude will readily attest to the heightened dyspnea they perceive during exercise. The magnitude of this hyperventilation and the accompanying hypocapnia increases with increasing exercise intensity. Further, after weeks or months at high altitude, the sojourner experiences a substantial tachypnea, hyperventilation, and respiratory alkalosis during sustained exercise of moderately heavy intensity (see Fig. 8). When exercise is combined with hypoxia, the ventilatory response is more than additive (3,152). In fact, hypoxia and other carotid body stimuli, whether physiological (i.e., CO2) or pharmacological (i.e., Doxapram HCl), commonly show synergistic ventilatory responses when combined. In the case of exercise, we speculate that the enhanced sensory input from the hypoxic carotid bodies is multiplied by simultaneous feed-forward–type input to the medullary respiratory controller from locomotor areas of the higher CNS. A number of important consequences derive from this magnified ventilatory response to exercise in the sojourner to high altitudes. On the plus side, the hyperventilatory response increases Pao 2 and hence arterial O2 content during exercise. For example, at 4000 m altitude, the Pao 2 during moderately heavy exercise in the acclimatized sojourner averages about 40 mmHg and Paco 2 about 17 mmHg (152). Without hyperventilation Pao 2 would have been in the 25–30 mmHg range, roughly equivalent to that found in a climber at the summit of Mount Everest (8848 m). Thus, hyperventilation, especially during exercise, has a potent ‘‘altitude-sparing’’ effect. Unfortunately, this sparing of arterial O2 content may come at considerable cost. The work of breathing is 40–60% greater than at sea level at any given exercise intensity (152,153). Although air density is reduced significantly at very high altitudes (i.e., low barometric pressures), the resultant reduction in airway resistance is more than offset by the increasing turbulent flow produced by high flow rates. At very heavy work loads in hypoxia, the increased ventilatory requirement will push the sojourner closer and closer to the limits of his maximum flow-volume loop. Significant expiratory flow limitation will cause additional ventilatory work by increasing flow resistance, distorting the chest wall, and causing end-expiratory lung volume to rise—the latter resulting in shortened inspiratory muscles and increased elastic work of breathing. Furthermore, powerful stimulation of breathing frequency could disrupt coordination of the normal synchronization of locomotion and breathing pattern that one normally accomplishes readily at sea level, particularly during running.

Control of Breathing

159

Hypoxia probably hastens the fatigue of respiratory muscles. This possibility has been demonstrated in humans by showing that heavy endurance exercise to exhaustion caused a reduction in the maximum pressure generated by the diaphragm in response to supramaximal phrenic nerve stimulation (154). This reduced maximal force output or diaphragmatic ‘‘fatigue’’ was similar following hypoxic and normoxic exercise, even though the exercise time to exhaustion was reduced by one half in hypoxia (142). In exercising rats, glycogen depletion and lactic acid accumulation in the diaphragm were markedly enhanced when exercise was conducted in hypoxia (155). The combination of an increased workload placed on the respiratory muscles plus reduced O2 transport to these muscles, along with an increased level of circulating metabolites from hypoxic limb locomotor muscles, are factors that might contribute to enhanced diaphragmatic fatigue during hypoxic exercise (142). ˙ o 2 max and exercise performance are decreased with increasing altitude (see V Chapter 20). Does ventilatory work contribute to this limitation? First, despite the increased work of breathing and possible diaphragm fatigue, the sojourner, even at very high altitudes, is clearly capable of a substantial alveolar hyperventilation dur˙ o 2 max, minute ventilation at high altitude approximates ing heavy exercise. At V ˙ o 2 max is reduced at high altitudes. Accordingly, that at sea level, even though V ventilatory ‘‘failure’’ in terms of the adequacy of alveolar oxygenation and CO2 elimination certainly does not occur during heavy exercise at high altitudes. Thus, despite diaphragmatic fatigue and expiratory flow limitation, the arsenal of ‘‘accessory’’ inspiratory and expiratory muscles appears to be sufficient to generate the appropriate magnitude of pleural pressure to meet the increased requirements for ventilation and ventilatory work during exercise at high altitudes. There are two more likely ways in which the hyperventilation and increased ventilatory work during exercise at high altitudes might contribute to decrements in exercise performance. First, the O2 requirement of exercise hyperpnea per se, particularly under conditions where flow limitations exist throughout a large portion ˙ o 2 max (156). of the breath, has been shown to require as much as 15% of the V These high levels of respiratory muscle work have recently been shown to promote a substantial ‘‘steal’’ of blood flow from limb locomotor muscles during maximum exercise (157) and may, therefore, limit locomotor muscle endurance performance. Second, the perception of dyspnea is extreme during heavy exercise in hypoxia (see Fig. 8). The additional sensory input to the cortex from chemoreceptors and chest wall mechanoreceptors is manifest by sensations that are foreign and highly unpleasant to the exercising sojourner at high altitude. Certainly, this discomfort could contribute to a sense of overall ‘‘effort’’ and whole-body ‘‘fatigue,’’ which may curtail exercise performance. In addition, brain hypoxia leading to cerebral dysfunction may also limit locomotor performance (see Sec. VII and Chapter 13). The hyperventilation of heavy exercise might exacerbate brain hypoxia because of the reduction in cerebral blood flow that has been shown to accompany the hypocapnia of exercise at high altitude (152). Finally, the magnitude of one’s ventilatory responsiveness to hypoxia may have a bearing on exercise performance at high altitude; this possibility is discussed

160

Smith et al.

in Section VII (see Chapter 20). An enhanced ventilatory response to hypoxia would increase alveolar oxygenation even during heavy exercise (see above). In the absence of other more efficient acclimatization mechanisms, this hyperventilatory response is extremely important to the sojourner. However, an excessive ventilatory responsiveness to exercise in hypoxia will also increase the work of breathing with all of the attendant consequences outlined above, especially the steal of blood flow from exercising limbs. The net effect of the opposing influences may be estimated by recalling that O2 transport to working muscle is the product of arterial O2 content and limb blood flow, the former being protected by alveolar hyperventilation and the latter being jeopardized by the attendant increase in respiratory muscle work. On the other hand, highlanders do not appear to be disadvantaged by a lesser ventilatory response to exercise (see Chapter 3). Because their alveolar to arterial O2 difference is much narrower, Pao 2 and Sao 2 remain comparable to that in the hyperventilating sojourner throughout exercise (see Fig. 7). The reason for the smaller A-aDo 2 in exercising highlanders is a much larger pulmonary diffusion capacity, both at rest and during exercise (Fig. 7) (158,159). This highly efficient pulmonary gas exchange probably requires true structural adaptation as a consequence of lifelong hypoxia.

VII. Breathing and Human Performance at High Altitude The final question we wish to explore is: Does how we breathe affect how we fare at high altitude? The more vigorous the ventilatory response to hypoxia (HVR), the higher will be the alveolar Po 2 and hence the arterial Po 2 and arterial oxygen saturation (although as indicated in the prior section, other factors during exercise may serve to widen the alveolar-arterial oxygen tension difference). Individual variability in HVR between high and low responders may span an order of magnitude (160,161). Interindividual variability in HVR appears to have a genetic underpinning (see also Chapter 3) (162,163). Our purpose in this section will be to review how variability of HVR relates to various aspects of human performance at high altitude. Presumably a higher HVR, by elevating Pao 2, should be an asset at high altitude and especially at extreme elevations, i.e., above 7500–8000 m, where, because of the steepness of the oxyhemoglobin dissociation curve, small changes in Pao 2 net large increases in arterial oxygen content. The HVR is often measured by relating changes in moment-to-moment ventilation resulting from breathing progressively more hypoxic inspired gas to concurrent changes in Po 2, either end-tidal or arterial, or to arterial oxygen saturation. Stimulation of breathing by hypoxia causes the Paco 2 to fall, the hypocapnic alkalosis blunting the hypoxic ventilatory response, a poikilocapnic HVR. To more clearly assess the hypoxic responsiveness of the peripheral chemoreceptors, HVR is measured with Paco 2 held constant at the resting level by adding carbon dioxide to the inspired gas, the isocapnic HVR (12). As we have seen with acclimatization, the resting Paco 2 falls adding an element of uncertainty to tracking changes in HVR

Control of Breathing

161

over time; methods have been proposed that attempt to standardize for these changes in resting Pco 2 that take place over the course of time of hypoxia (99,164). As indicated earlier, both measurements of HVR during acclimatization and the progressive increase in ventilation occurring over days to weeks at altitude, even as the chemical stimuli, Po 2 and Pco 2, lessen in potency, indicate that hypoxic gain is increasing (95–99). The magnitude of the HVR in a given individual correlates moderately well with the vigor of exercise hyperpnea and, perhaps to a somewhat lesser extent, with the ventilatory responsiveness to carbon dioxide (HCVR) (165– 169). We can speculate on potential advantages and disadvantages of being gifted with an exceptionally high or low HVR. Endurance athletes commonly possess a low HVR, the benefits of which during performance might well be a lower metabolic cost of ventilation and less dyspnea (160). One can also imagine this same attribute as playing a role in those with chronic obstructive pulmonary disease, contributing to the distinction between those classified as ‘‘pink puffers’’ and ‘‘blue bloaters.’’ Schoene noted that the HVR of 14 mountaineers who had climbed above 7470 m was greater than that of 10 normal controls and significantly so than that of 10 endurance athletes. The putative advantage from a greater HVR at high altitude, particularly at extremely high altitude, is that thanks to the shape of the oxygenhemoglobin dissociation curve, even a small increase in Pao 2 will yield a large increase arterial oxygen content. Although controlled trials on a sufficient number of subjects of capacity to perform work are lacking, observations of climbing performance on several Himalayan expeditions suggest that those possessing high HVR tend to perform better than those with low HVR, even though the latter may be ˙ o 2 max at sea level (170–172). These highly trained aerobic athletes with high V conclusions are based upon such things as maximum climbing and sleeping altitude and subjective appraisal of individual performance and well-being. Inevitably, there are exceptions, one of the more noteworthy being that one of the two who first climbed Everest without supplemental oxygen possessed a quite low HVR at sea level (173). Although HVR clearly increases with acclimatization, what is less well established is how it changes in a given individual over the course of time. In particular, might someone with a low HVR at sea level acquire a significantly more vigorous responsiveness with acclimatization to high altitude? Does HVR influence how quickly and how well lowlanders acclimatize upon ascending to high altitude? Reeves and colleagues noted that a lower resting endtidal Pco 2 in those living at low elevations was associated with a higher sea level HVR, implying first of all that even sea level Pao 2 might influence resting ventilation (174). In addition, both sea level Petco 2 and HVR predicted the magnitude of increase in breathing, as assessed by decrease in Petco 2 after one and 19 days at 4300 meters. Associations of sea level HVR with quality of adaptation have been noted in a number of studies. Individuals with lower HVRs may be more susceptible to acute high-altitude illnesses, including acute mountain sickness (AMS) (175–179) and high altitude pulmonary edema (180–183). No similar information exists with regard to an association between HVR and susceptibility to high-altitude cerebral

162

Smith et al.

edema. The link between a low HVR and AMS has not been universally observed, which is not surprising considering that many other factors dominate, including rate of ascent, altitude gained, and possibly age, gender, and other considerations (see Chapter 22). Acute mountain sickness has been associated with an increased fluid retention and weight gain (184). The peripheral chemoreceptors likely play a role in modulating sodium and water balance with hypoxia (185). Recently Swenson demonstrated in humans that those exhibiting a greater altitude-induced natriuresis and diuresis also evinced a higher HVR (186). No association with plasma concentrations of known salt and water-regulating hormones, renin, aldosterone, atrial natriuretic peptide, or antidiuretic hormone was found (186). These observations indicate a peripheral chemoreceptor influence on renal regulation of water balance but leave us guessing as to what the mechanistic link between the carotid body and kidney may be. For lowlanders acclimatizing to high altitude, a brisk HVR seems to confer an advantage, netting both lower susceptibility to acute illnesses of high altitude and an enhanced capacity to perform physical work at extremely high altitudes. This latter gift may come with strings attached. A number of observers have reported evidence of mild, generally transient brain injury in some individuals who have climbed to extremely high altitudes. In looking for a physiological basis for the variability in susceptibility, Hornbein et al. indeed found an association with HVR, but it was the opposite of the one they had anticipated (187), those with higher HVR appeared to suffer more residual neurobehavioral deficits than those with lower HVR. Yet, as cited above, these same individuals appeared capable of better physical performance while at high altitude (172). One explanation proffered for this unexpected association of higher HVR with greater brain injury was that while a higher HVR increased arterial oxygen content, the greater hypocapnia caused more profound cerebral vasoconstriction; while exercising muscle received more oxygen the brain received less. Because arterial Po 2 and Pco 2 so powerfully influence cerebral blood flow, an association of CBF with HVR seems plausible, but thus far no information exists to indicate whether those with high HVR might possess heightened cerebrovascular reactivity to oxygen and CO2 as well. This association between HVR and brain injury becomes more relevant as increasing numbers attempt to climb Mount Everest, including a small subset who choose to do so without the use of supplemental oxygen. Another possible contributor to brain hypoxia is the greater tendency toward periodic breathing noted in those with higher HVR (140,188). Even though average arterial oxygenation appears to be improved in those with higher HVR, the nadirs of Sao 2 associated with apneic pauses at extreme altitude might result in brief, repeated moments of such severe arterial hypoxemia as to cause brain injury. Hormonal changes have been associated with HVR. For example, Raff et al. noted that hypoxia, acting by way of the carotid bodies in dogs, can stimulate cortisol secretion (189). In premenopausal women, HVR is increased during the luteal phase of the menstrual cycle (169,190,191). Studies in both experimental animals and humans (192) indicate that progesterone increases carotid chemoreceptor output

Control of Breathing

163

while estrogen may have an enhancing effect acting centrally. HVR in premenopausal women exceeds that of men when corrected for differences in body size. Pregnancy enhances HVR further. Whether this gender difference confers a functional advantage at high altitude is a question that has been only partially explored. Premenopausal women appear to be less susceptible to high-altitude pulmonary edema, and those living permanently at high altitude are less likely to experience chronic mountain sickness prior to menopause (see Chapters 23 and 24). Whether a higher HVR of premenopausal women might enable better physical performance at extreme altitude is a question that has not been examined. What happens to HVR in permanent high-altitude populations is discussed elsewhere in this volume (see Chapters 3 and 24). Initial observations that those born at or living for years at high altitude possess a lower HVR than acclimatized lowlanders have been extended by more recent studies indicating that not all highaltitude dwellers experience such blunting. Ethnic differences appear to play a role, with those of Tibetan origin showing greater variability and higher HVRs than the high-altitude Aymara of South America and comparable to the HVRs of acclimatized lowlanders (143).

References 1. 2. 3.

4.

5.

6.

7. 8. 9. 10.

Kellogg RH. Altitude acclimatization, a historical introduction emphasizing the regulation of breathing. Physiologist 1968; 11:37–57. Dempsey JA, et al. Mediation of ventilatory adaptations. Physiol Rev 1982; 62:262– 346. Weil JV. Ventilatory control at high altitude. In: Cherniack NS, Widdicombe JG, eds. Handbook of Physiology, Section 3, The Respiratory System. Vol. 2. Bethesda: American Physiology Society, 1986:703–727. Dempsey JA, Schoene RB. Pulmonary system adaptations to high altitude. In: Bone RC, ed. Pulmonary and Critical Care Medicine, Respiratory Physiology. Vol. 1, Part B6. St. Louis: Mosby Year Book Publishers, 1993:1–22. Neubauer JA, Bisgard GE. Peripheral and central effects of hypoxia. In: Dempsey JA, Pack AI, eds. Regulation of Breathing. 2d ed. New York: Marcel Dekker, 1994:617– 668. Bisgard GE, Forster HV. Ventilatory responses to acute and chronic hypoxia. In: Fregly MJ, Blatteis CM, eds. Handbook of Physiology, Environmental Physiology. New York: Oxford University Press, 1996:1207–1239. Gershan WM, Forster HV, Lowry TF, et al. Effect of metabolic rate on ventilatory roll-off during hypoxia. J Appl Physiol 1994; 76:2310–2314. Long WQ, Giesbrecht GG, Anthonisen NR. Ventilatory response to moderate hypoxia in awake chemodenervated cats. J Appl Physiol 1993; 74:805–810. Weil JV. Ventilatory responses to CO2 and hypoxia after sustained hypoxia in awake cats. J Appl Physiol 1994; 76:2251–2252. Easton PA, Slykerman LJ, Anthonisen NR. Ventilatory response to sustained hypoxia in normal adults. J Appl Physiol 1986; 61:906–911.

164

Smith et al.

11.

Kagawa S, Stafford MJ, Waggener TB, Severinghaus JW. No effect of naloxone and hypoxia-induced ventilatory depression in adults. J Appl Physiol 1982; 52:1030–1034. Weil JV, Zwillich CW. Assessment of ventilatory response to hypoxia: methods and interpretation. Chest 1976; 70:124–128. Freedman A, Scardella AT, Edelman NH, Santiago TV. Hypoxia does not increase CSF or plasma beta-endorphin activity. J Appl Physiol 1988; 64:966–971. Cao K-Y, Zwillich CW, Berthon-Jones M, Sullivan CE. Ventilatory response to sustained eucapnic hypoxia in the adult conscious dog. Respir Physiol 1992; 89:65–73. Dwinell MR, Janssen PL, Bisgard GE. Effects of carotid body hypocapnia during ventilatory acclimatization to hypoxia. J Appl Physiol 1997; 82:118–124. Hornbein TF, Roos A, Griffo AJ. Transient effect of sudden mild hypoxia on respiration. J Appl Physiol 1961; 16:11–14. Barnard P, Andronikou S, Pokorski M, Smatresk N, Mokashi A, Lahiri S. Timedependent effect of hypoxia on carotid body chemosensory function. J Appl Physiol 1987; 63:685–691. Vizek M, Pickett CK, Weil JV. Increased carotid body hypoxic sensitivity during acclimatization to hypobaric hypoxia. J Appl Physiol 1987; 63:2403–2410. Nye PC. Identification of peripheral chemoreceptor stimuli. Med Sci Sports Exerc 1994; 26:311–318. Neubauer JA, Melton JE, Edelman NH. Modulation of respiration during brain hypoxia. J Appl Physiol 1990; 68:441–451. Brown DL, Lawson EE. Brain stem extracellular fluid pH and respiratory drive during hypoxia in newborn pigs. J Appl Physiol 1988; 68:1055–1059. Neubauer JA, Santiago TV, Posner MA, Edelman NH. Ventral medullary pH and ventilatory responses to hyperperfusion and hypoxia. J Appl Physiol 1985; 58:1659–1668. Suzuki A, Nishimura M, Yamamoto H, Miyamoto K, Kishi F, Kowakami Y. No effect of brain blood flow on ventilatory depression during sustained hypoxia. J Appl Physiol 1989; 66:1674–1678. Yoshida K, Meyer JS, Sakamoto K, Honda J. Autoregulation of cerebral blood flow: electromagnetic flow measurements during acute hypertension in the monkey. Circ Res 1966; 19:726–738. Long W, Lobchuk D, Anthonisen NR. Ventilatory responses to CO2 and hypoxia after sustained hypoxia in awake cats. J Appl Physiol 1994; 76:2262–2266. Heeringa J, Berkenbosch A, de Goede J, Olievier CN. Relative contribution of central and peripheral chemoreceptors to the ventilatory response to CO2 during hyperoxia. Respir Physiol 1979; 37:365–379. Berkenbosch A, van Dissel J, Olievier CN, de Goede J, Herringa J. The contribution of the peripheral chemoreceptors to the ventilatory response to CO2 in anaesthetized cats during hyperoxia. Respir Physiol 1979; 37:381–390. DeGoede J, Berkenbosch A, Ward DS, Bellville JW, Olievier CN. Comparison of chemoreflex gains obtained with two different methods in cats. J Appl Physiol 1985; 59:170–179. Smith CA, Saupe KW, Henderson KS, Dempsey JA. Ventilatory effects of specific carotid body hypocapnia in dogs during wakefulness and sleep. J Appl Physiol 1995; 79:689–699. Smith CA, Henderson KS, Harms CA, Zeng Y-J, Dempsey JA. Ventilatory effects of specific carotid body hypocapnia and hypoxia in awake dogs. J Appl Physiol 1997; 82:791–798.

12. 13. 14. 15. 16. 17.

18. 19. 20. 21. 22. 23.

24.

25. 26.

27.

28.

29.

30.

Control of Breathing 31.

165

Melton JE, Neubauer JA, Edelman NH. CO2 sensitivity of cat phrenic neurogram during hypoxic respiratory depression. J Appl Physiol 1988; 65:736–743. 32. Melton JE, Yu QP, Neubauer JA, Edelman NH. Modulation of respiratory responses to carotid sinus nerve stimulation by brain hypoxia. J Appl Physiol 1992; 73:2166– 2171. 33. Bouverot P, Bureau M. Ventilatory acclimatization and csf acid-base balance in carotid chemodenervated dogs at 3550 m. Pflugers Arch 1975; 361:17–23. 34. Olson EB, Jr., Vidruk EH, Dempsey JA. Carotid body excision significantly changes ventilatory control in awake rats. J Appl Physiol 1988; 64:666–671. 35. Forster HV, Bisgard GE, Rasmussen B, Orr JA, Buss DD, Manohar M. Ventilatory control in peripheral chemoreceptor-denervated ponies during chronic hypoxemia. J Appl Physiol 1976; 41:878–885. 36. Lahiri S, Edelman NH, Cherniack NS, Fishman AP. Role of carotid chemoreflex in respiratory acclimatization to hypoxemia in goat and sheep. Respir Physiol 1981; 46: 367–382. 37. Smith CA, Bisgard GE, Nielsen AM, Daristotle L, Kressin NA, Forster HV, Dempsey JA. Carotid bodies are required for ventilatory acclimatization to chronic hypoxia. J Appl Physiol 1986; 60:1003–1010. 38. Engwall MJA, Smith CA, Dempsey JA, Bisgard GE. Ventilatory afterdischarge and central respiratory drive interactions in the awake goat. J Appl Physiol 1994; 76:416– 423. 38a. Curran AK, Rodman JR, Eastwood PR, Henderson KS, Dempsey JA, Smith CA. Ventilatory responses to specific CNS hypoxia in sleeping dogs. J Appl Physiol 2000; 88: 1840–1852. 39. Miller MJ, Tenney SM. Hypoxia-induced tachypnea in carotid-deafferented cats. Respir Physiol 1975; 23:31–39. 40. Dahan A, Ward D, van den Elsen M, Temp J, Berkenbosch A. Influence of reduced carotid body drive during sustained hypoxia on hypoxic depression of ventilation in humans. J Appl Physiol 1996; 81:565–572. 41. Forster HV, Dempsey JA, Chosy LW. Incomplete compensation of CSF [H⫹] in man during acclimatization to high altitude (4,300 m). J Appl Physiol 1975; 38:1067– 1072. 42. Kayser B, Bokenkamp R, Binzoni T. Alpha-motoneuron excitability at high altitude. Eur J Appl Physiol Occup Physiol 1993; 66:1–4. 43. O’Reilly JP, Haddad GG. Chronic hypoxia in vivo renders neocortical neurons more vulnerable to subsequent acute hypoxic stress. Brain Res 1996; 711:203–210. 44. Haddad GG, Donnelly DF. O2 deprivation induces a major depolarization in brain stem neurons in the adult but not in the neonatal rat. J Physiol 1990; 429:411–428. 45. Dillon GH, Waldrop TG. In vitro responses of caudal hypothalamic neurons to hypoxia and hypercapnia. Neuroscience 1992; 51:941–950. 46. Nolan PC, Waldrop TG. In vivo and in vitro responses of neurons in the ventrolateral medulla to hypoxia. Brain Res 1993; 630:101–114. 47. Nolan PC, Waldrop TG. In vitro responses of VLM neurons to hypoxia after normobaric hypoxic acclimatization. Respir Physiol 1996; 105:23–33. 48. Kramer JM, Nolan PC, Waldrop TG. In vitro responses of neurons in the periaqueductal gray to hypoxia and hypercapnia. Brain Res 1999; 835:197–203. 49. Horn EM, Waldrop TG. Oxygen-sensing neurons in the caudal hypothalamus and their role in cardiorespiratory control. Respir Physiol 1997; 110:219–228.

166

Smith et al.

50.

Gesell R, Hamilton MA. Reflexogenic components of breathing. Am J Physiol 1941; 133:694–719. Eldridge FL, Gill-Kumar P. Central neural respiratory drive and afterdischarge. Respir Physiol 1980; 40:49–63. Eldridge FL. Central integration of mechanisms in exercise hyperpnea. Med Sci Sports Exerc 1994; 26:319–27. Badr MS, Skatrud JB, Dempsey JA. Determinants of poststimulus potentiation in humans during NREM sleep. J Appl Physiol 1992; 73:1958–1971. Georgopoulos D, Walker S, Anthonisen NR. Increased chemoreceptor output and ventilatory response to sustained hypoxia. J Appl Physiol 1989; 67:1157–1163. Curtis DR, Duggan AW, Felix D, Johnston GAR. GABA, bicuculline and central inhibition. Nature 1970; 226:1222–1224. Wood JD, Watson WJ, Ducker AJ. The effect of hypoxia on brain γ-aminobutyric acid levels. J Neurochem 1968; 15:603–608. Weyne J, Van Leuven F, Leusen I. Brain amino acids in conscious rats in chronic normocapnic acid levels. Respir Physiol 1977; 31:231–239. Iversen K, Hedner T, Lundborg P. GABA concentrations and turnover in neonatal rat brain during asphyxia and recovery. Acta Physiol Scand 1983:91–94. Hagberg H, Lehman A, Sandberg M, Nystrom B, Jacobson I, Hamberger A. Ischemiainduced shift of inhibitory and excitatory amino acids from intra- to extracellular compartments. J Cerebral Blood Flow Metab 1985; 5:413–419. Hedner T, Hedner J, Bergman B, Iversen K, Jonason J. Effects of GABA and some GABA analogues on respiratory regulation in the preterm rabbit. Biol Neonate 1983; 43:134–145. Hedner J, Hedner T, Wessberg P, Jonason J. An analysis of the mechanism by which γ-aminobutyric acid depresses ventilation in the rat. J Appl Physiol 1984; 56:849– 856. Kneussel MP, Pappagianopoulos P, Hoop B, Kazemi H. Reversible depression of ventilation and cardiovascular function by ventriculocisternal perfusion with gamma aminobutyric acid in dogs. Am Rev Respir Dis 1986; 13:1024–1028. Taveira da Silva AM, Hartley B, Hamosh P, Quest JA, Gillis RA. Respiratory depressant effects of GABA α- and β-receptor agonists in the cat. J Appl Physiol 1987; 62: 2264–2272. Huang J, Suguihara C, Hehre D, Lin J, Bancalari E. Effects of GABA receptor blockage on the respiratory response to hypoxia in sedated newborn piglets. J Appl Physiol 1994; 77:1006–1010. Melton JE, Neubauer JA, Edelman NH. GABA antagonism reverses hypoxic respiratory depression in the cat. J Appl Physiol 1990; 69:1296–301. Yamada KA, Norman WP, Hamosh P, Gillis RA. Medullary ventral surface GABA receptors affect respiratory and cardiovascular function. Brain Res 1982; 248:71– 78. Phillis JW, Kostopoulos GK, Limacher JJ. Depression of corticospinal cells by various purines and pyrimidines. Can J Pharmacol 1974; 52:1226–1229. Phillis JW, Kostopoulos GK, Limacher JJ. A potent depressant action of adenine derivatives on cerebral cortical neurons. Eur J Pharmacol 1975; 30:125–129. Koos BJ, Mason BA, Punla O, Adinolfi AM. Hypoxic inhibition of breathing in fetal sheep: relationship to brain adenosine concentrations. J Appl Physiol 1994; 77:2734– 2739.

51. 52. 53. 54. 55. 56. 57. 58. 59.

60.

61.

62.

63.

64.

65. 66.

67. 68. 69.

Control of Breathing 70. 71. 72.

73. 74. 75. 76. 77. 78. 79.

80. 81. 82.

83. 84.

85. 86. 87. 88. 89. 90.

91.

167

Eldridge FL, Millhorn DE, Kiley JP. Respiratory effects of a long-acting analog of adenosine. Brain Res 1984; 301:273–280. Hedner T, Hedner P, Wessberg P, Jonason J. Regulation of breathing in the rat: indications for a role of central adenosine mechanisms. Neurosci Lett 1982; 33:147–151. Yan S, Laferriere A, Zhang C, Moss IR. Microdialyzed adenosine in nucleus tractus solitarii and ventilatory response to hypoxia in piglets. J Appl Physiol 1995; 79:405– 410. Long WA, Lawson EA. Neurotransmitters and biphasic respiratory response to hypoxia. J Appl Physiol 1984; 57:213–222. Millhorn DE, Eldridge FL, Kiley JP, Waldrop TG. Prolonged inhibition of respiration following acute hypoxia in glomectomized cats. Respir Physiol 1984; 57:331–340. Javaheri S, Teppema LJ, Evers JA. Effects of aminophylline on hypoxemia-induced ventilatory depression in the cat. J Appl Physiol 1988; 64:1837–1843. Easton PA, Anthonisen NR. Ventilatory response to sustained hypoxia after pretreatment with aminophylline. J Appl Physiol 1988; 64:1445–1450. Gershan WM, Forster HV, Lowry TF, Garber AK. Effect of theophylline on ventilatory roll-off during hypoxia in goats. Respir Physiol 1996; 103:157–164. Neubauer JA, Simone A, Edelman NH. Role of brain lactic acidosis in hypoxic depression of respiration. J Appl Physiol 1988; 65:1324–1331. Musch TI, Dempsey JA, Smith CA, Mitchell GS, Bateman NT. Metabolic acids and [H⫹] regulation in brain tissue during acclimatization to chronic hypoxia. J Appl Physiol 1983; 55:1486–1495. Aaron EA, Forster HV, Lowry TF, Korducki MJ, Ohtake PJ. Effect of dichloroacetate on Paco2 responses to hypoxia in awake goats. J Appl Physiol 1996; 80:176–181. Severinghaus JW. Hypoxic ventilatory depression may be due to central chemoreceptor cell hyperpolarization. Adv Exp Med Biol 1995; 393:257–260. Xu FD, Spellman MJ, Jr., Sato M, Baumgartner JE, Ciricillo SF, Severinghaus JW. Anomalous hypoxic acidification of medullary ventral surface. J Appl Physiol 1991; 71:2211–2217. Xu F, Sato M, Spellman MJ, Jr., Mitchell RA, Severinghaus JW. Topography of cat medullary ventral surface hypoxic acidification. J Appl Physiol 1992; 73:2631–2637. DeBoeck C, Van Reempts P, Rigatto H, Chernick V. Naloxone reduces decrease in ventilation induced by hypoxia in newborn infants. J Appl Physiol 1984; 56:1507– 1511. Grunstein MM, Hazinski TA, Schleuter MA. Respiratory control during hypoxia in newborn rabbits: implied action of endorphins. J Appl Physiol 1981; 51:122–130. Schaeffer JJ, Haddad GG. Ventilatory response to moderate and severe hypoxia in adult dogs: role of endorphins. J Appl Physiol 1988; 65:1383–1388. Wheeler MD, Farber JP. Naloxone administration and ventilation in awake cats. Brain Res 1983; 258:343–346. Santiago TV, Edelman NH. Opioids and breathing. J Appl Physiol 1985; 59:1675– 1685. Pokorski M, Paulev P-E, Szereda-Przestaszewska M. Endogenous benzodiazepine system and regulation of respiration in the cat. Respir Physiol 1994; 87:33–45. Mortola J, Gautier H. Interaction between metabolism and ventilation: Effects of respiratory gases and temperature. In: Dempsey JA, Pack AI, eds. Regulation of Breathing. Vol. 79. 2d ed. New York: Marcel Dekker, 1994:1011–1064. Frappell P, Lanthier C, Baudinette RV, Mortola JP. Metabolism and ventilation in

168

92.

93. 94. 95.

96.

97.

98. 99. 100. 101. 102.

103. 104. 105. 106.

107.

108. 109. 110.

Smith et al. acute hypoxia: a comparative analysis in small mammalian species. Am J Physiol 1992; 262:R1040–R1046. Dempsey JA, Forster HV, Birnbaum ML, Reddan WG, Thoden J, Grover RF, Rankin J. Control of exercise hyperpnea under varying durations of exposure to moderate hypoxia. Respir Physiol 1972; 16:213–231. Forster HV, Bisgard GE, Klein JP. Effect of peripheral chemoreceptor denervation on acclimatization of goats during hypoxia. J Appl Physiol 1981; 50:392–398. Olson EB, Jr. Physiological dead space increases during initial hours of chronic hypoxemia with or without hypocapnia. J Appl Physiol 1994; 77:1526–1531. Forster HV, Dempsey JA, Birnbaum ML, Reddan WG, Thoden J, Grover RF, Rankin J. Effect of chronic exposure to hypoxia on ventilatory response to CO2 and hypoxia. J Appl Physiol 1971; 31:586–592. White DP, Gleeson K, Pickett CK, Rannels AM, Cymerman A, Weil JV. Altitude acclimatization: influence on periodic breathing and chemoresponsiveness during sleep. J Appl Physiol 1987; 63:401–412. Schoene RB, Roach RC, Hackett PH, Sutton JR, Cymerman A, Houston CS. Operation Everest II: ventilatory adaptation during gradual decompression to extreme altitude. Med Sci Sports Exerc 1990; 22:804–810. Goldberg SV, Schoene RB, Haynor D, et al. Brain tissue pH and ventilatory acclimatization to high altitude. J Appl Physiol 1992; 72:58–63. Sato M, Severinghaus JW, Powell FL, Xu F-D, Spellman MJ. Augmented hypoxic ventilatory response in men at altitude. J Appl Physiol 1992; 73:101–107. Busch MA, Bisgard GE, Forster HV. Ventilatory acclimatization to hypoxia is not dependent on arterial hypoxemia. J Appl Physiol 1985; 58:1874–1880. Bisgard GE, Busch MA, Forster HV. Ventilatory acclimatization to hypoxia is not dependent on cerebral hypocapnic alkalosis. J Appl Physiol 1986; 60:1011–1015. Weizhen N, Engwall MJA, Daristotle L, Pizarro J, Bisgard GE. Ventilatory effects of prolonged systemic (CNS) hypoxia in awake goats. Respir Physiol 1992; 87:37– 48. Nielsen AM, Bisgard GE, Vidruk EH. Carotid chemoreceptor activity during acute and sustained hypoxia in goats. J Appl Physiol 1988; 65:1796–1802. Engwall MJA, Vidruk EH, Nielsen AM, Bisgard GE. Response of the goat carotid body to acute and prolonged hypercapnia. Respir Physiol 1988; 74:335–344. Nurse C. Carotid body adaptation to hypoxia: cellular and molecular mechanisms in vitro. Biol Signals 1995; 4:286–291. Bisgard GE, Busch MA, Daristotle L, Berssenbrugge AD, Forster HV. Carotid body hypercapnia does not elicit ventilatory acclimatization in goats. Respir Physiol 1986; 65:113–125. Lahiri S, Rozanov C, Cherniack NS. Altered structure and function of the carotid body at high altitude and associated chemoreflexes. High Altitude Med Biol 2000; 1:63– 74. Niu WZ, Engwall MJA, Bisgard GE. Two discharge patterns of carotid body chemoreceptors in the goat. J Appl Physiol 1990; 69:734–739. Tatsumi K, Pickett CK, Weil JV. Decreased carotid body hypoxic sensitivity in chronic hypoxia: role of dopamine. Respir Physiol 1995; 101:47–57. Czyzyk-Krzeska MF, Bayliss DA, Lawson EE, Milhorn DE. Regulation of tyrosine hydroxlyase gene expression in the rat carotid body by hypoxia. J Neurochem 1992; 58:1538–1546.

Control of Breathing 111. 112.

113.

114.

115. 116. 117.

118.

119.

120.

121. 122. 123.

124. 125. 126.

127.

128.

169

Wang ZZ, Bredt DS, Fidone SJ, Stensaas LJ. Neurons synthesizing nitric oxide innervate the mammalian carotid body. J Comp Neurol 1993; 336:419–432. Kummer W. Three types of neurochemically defined autonomic fibers innervate the carotid baroreceptor and chemoreceptor regions in the guinea pig. Anat Embryol 1990; 181:477–489. McQueen DS. Pharmacological aspects of putative transmitters in the carotid body. In: Acker J, O’Regan RG, eds. Physiology of the Peripheral Chemoreceptors. Amsterdam: Elsevier, 1983:149–196. Prabhakar NR, Landis SC, Kumar GK, Mullikin-Lilpatrick D, Cherniack NS. Substance P and neurokinin A in the cat carotid body: localization, exogenous effects and changes in content in response to arterial pO2. Brain Res 1989; 481:205–214. Millhorn DE, Eldridge FL, Waldrop TG. Prolonged stimulation of respiration by endogenous central serotonin. Respir Physiol 1980; 42:171–188. Olson EB, Jr. Ventilatory adaptation to hypoxia occurs in serotonin-depleted rats. Respir Physiol 1987; 69:227–235. Steinman JL, Carlton SM, Haber B, Willis WD. Differential effects of p-chlorphenylalanine on indoleamines in brainstem nuclei and spinal cord of rats. Brain Res 1987; 426:297–309. Carlton SM, Steinman JL, Hillman GR, Willis WD. Differential effects of p-chlorophenylalanine on indoleamines in brainstem nuclei and spinal cord of rats. II. Identification of immunohistochemically stained structures using computer-assisted image enhancement techniques. Brain Res 1987; 426:310–322. Ogawa H, Mizusawa A, Kikuchi Y, Hida W, Miki H, Shirato K. Nitric oxide as a retrograde messenger in the nucleus tractus solitarii of rats during hypoxia. J Physiol 1995; 486:495–504. Severinghaus JW, Mitchell RA, Richardson BW, Singer MM. A respiratory control at high altitude suggesting active transport regulation of CSF pH. J Appl Physiol 1963; 18:1155–1166. Crawford RD, Severinghaus JW. CSF pH and ventilatory acclimatization to altitude. J Appl Physiol 1978; 45:275–283. Dempsey JA, Forster HV, DoPico GA. Ventilatory acclimatization to moderate hypoxemia in man. The role of spinal fluid (H⫹). J Clin Invest 1974; 53:1091–1100. Forster HV, Soto JR, Dempsey JA, Hosko MJ. Effect of sojourn at 4,300 m altitude on electroencephalogram and visual evoked response. J Appl Physiol 1975; 39:109– 113. Weiskopf RB, Gabel RA, Fencl V. Alkaline shift in lumbar and intracranial CSF in man after 5 days at high altitude. J Appl Physiol 1976; 41:93–97. Bureau M, Bouverot P. Blood and CSF acid-base changes, and rate of ventilatory acclimatization of awake dogs to 3,550 m. Respir Physiol 1975; 24:203–216. Orr JA, Bisgard GE, Forster HV, Buss DD, Dempsey JA, Will JA. Cerebrospinal fluid alkalosis during high-altitude sojourn in unanesthetized ponies. Respir Physiol 1975; 25:23–37. Dempsey JA, Forster HV, Bisgard GE, Chosy LW, Hanson PG, Kiorpes AL, Pelligrino DA. Role of cerebrospinal fluid [H⫹] in ventilatory deacclimatization from chronic hypoxia. J Clin Invest 1979; 64:199–205. Fencl V, Miller TB, Pappenheimer JR. Studies on the respiratory response to disturbances of acid-base balance, with deductions concerning ionic composition of cerebral interstitial fluid. Am J Physiol 1966; 210:459–472.

170

Smith et al.

129.

Pappenheimer JR. The ionic composition of cerebral extracellular fluid and its relation to control of breathing. Harvey Lect 1967; 6:71–93. Dempsey JA, Forster HV, Chosy LW, Hanson PG, Reddan WG. Regulation of CSF[HCO3⫺] during long-term hypoxic hypocapnia in man. J Appl Physiol 1978; 44: 175–182. Jennings DB. The physicochemistry of [H⫹] and respiratory control: roles of Pco2, strong ions, and their hormonal regulators. Can J Physiol Pharmacol 1994; 72:1499– 1512. Stewart PA. How to Understand Acid-Base. A Quantitative Acid-Base Primer for Biology and Medicine. New York: Elsevier, 1981. Nattie EE. Ionic mechanisms of cerebrospinal fluid acid-base regulation. J Appl Physiol 1983; 54:3–12. Engwall MJA, Bisgard GE. Ventilatory responses to chemoreceptor stimulation after hypoxic acclimatization in awake goats. J Appl Physiol 1990; 69:1236–1243. Ryan ML, Hedrick MS, Pizarro J, Bisgard GE. Carotid body noradrenergic sensitivity in ventilatory acclimatization to hypoxia. Respir Physiol 1993; 92:77–90. Bisgard GE. The effect of prolonged stimulation of afferent activity of the goat carotid body. In: Acker H, Trzebski A, O’Regan R, eds. Chemoreceptors and Chemoreceptor Reflexes. New York: Plenum Press, 1990:165–170. Vizek M, Pickett CK, Weil JV. Biphasic ventilatory response of adult cats to sustained hypoxia has central origin. J Appl Physiol 1987; 63:1658–1664. Dempsey JA, Thomson JM, Alexander SC, Forster HV, Chosy LW. Respiratory influences on acid-base status and their effects on O2 transport during prolonged muscular work. In: Howald H, Poortmans JR, eds. Adaptations to Prolonged Exercise. Basel: Birkhauser Verlag, 1975:56–64. Milledge JS, Lahiri S. Respiratory control in lowlanders and Sherpa highlanders at altitude. Respir Physiol 1967; 2:310–322. Lahiri S, Maret KH, Sherpa MG, Peters RMJ. Sleep and periodic breathing at high altitude: Sherpa natives vs. sojourners. In: West J, Lahiri S, eds. High Altitude and Man. Bethesda: American Physiological Society, 1984:73–90. Dempsey JA, Reddan WG, Birnbaum ML, Forster HV, Thoden JS, Grover RF, Rankin J. Effects of acute through life-long hypoxic exposure on exercise pulmonary gas exchange. Respir Physiol 1971; 13:62–89. Babcock M, Johnson BD, Pegelow DF, Suman OE, Griffin D, Dempsey JA. Hypoxic effects on exercise-induced diaphragmatic fatigue in normal healthy humans. J Appl Physiol 1995; 78:82–92. Beall CM, Strohl KP, Biangero J, Williams-Blangero S, Almasy LA, Decker MJ, Worthman CM, Goldstein MC, Vargas E, Villena M, Soria R, Alarcon AM, Gonzales C. Ventilation and hypoxic ventilatory response of Tibetan and Aymara high altitude natives. Am J Phys Anthropol 1997; 104:427–447. Brooks JGI, Tenney SM. Ventilatory response of llama to hypoxia at sea level and high altitude. Respir Physiol 1968; 5:269–278. Lahiri S. Unattenuated ventilatory hypoxic drive in ovine and bovine species native to high altitude. J Appl Physiol 1972; 32:95–102. Bisgard GE, Ruiz AV, Grover RF, Will JA. Ventilatory acclimatization to 3400 meters altitude in the Hereford calf. Respir Physiol 1974; 21:271–296. Okubo S, Mortola J. Control of ventilation in adult rats hypoxic in the neonatal period. Am J Physiol 1990; 259:R836–R841.

130.

131.

132. 133. 134. 135. 136.

137. 138.

139. 140.

141.

142.

143.

144. 145. 146. 147.

Control of Breathing 148. 149. 150. 151. 152.

153.

154. 155. 156. 157.

158. 159.

160. 161. 162. 163. 164. 165.

166. 167. 168.

171

Tatsumi K, Pickett CK, Weil JV. Attenuated carotid body hypoxic sensitivity after prolonged hypoxic exposure. J Appl Physiol 1991; 70:748–755. Forster HV, Dempsey JA, Vidruk EH, DoPico G. Evidence of altered regulation of ventilation during exposure to hypoxia. Respir Physiol 1974; 20:379–392. Pokorski M, Lahiri S. Effects of naloxone on carotid body chemoreception and ventilation in the cat. J Appl Physiol 1981; 51:1533–1538. Tenney SM, Ou LC. Hypoxic ventilatory response of cats at high altitude: an interpretation of ‘blunting’. Respir Physiol 1977; 30:185–199. Dempsey JA, Gledhill N, Reddan W, Forster HV, Hanson PG, Claremont AD. Pulmonary adaptation to exercise: effects of exercise type and duration, chronic hypoxia and physical training. Ann NY Acad Sci 1977; 301:243–261. Thoden JS, Dempsey JA, Reddan WG, Birnbaum ML, Forster HV, Grover RF, Rankin J. Ventilatory work during steady-state response to exercise. Fed Proc 1969; 28:1316– 1321. Johnson BD, Babcock M, Suman O, Dempsey JA. Exercise-induced diaphragmatic fatigue in healthy humans. J Physiol 1993; 460:385–405. Fregosi R, Dempsey JA. The effects of exercise in normoxia and acute hypoxia on respiratory muscle metabolites. J Appl Physiol 1986; 60:11274–11283. Aaron EA, Johnson BD, Seow KC, Dempsey JA. Oxygen cost of exercise hyperpnea: measurement. J Appl Physiol 1992; 72:1810–1817. Harms CA, Babcock MA, McClaran SR, Pegelow DF, Nickele GA, Nelson WB, Dempsey JA. Respiratory muscle work compromises leg blood flow during maximal exercise. J Appl Physiol 1997; 82:1573–1583. Cerny FC, Dempsey JA, Reddan WG. Pulmonary gas exchange in nonnative residents of high altitude. J Clin Invest 1973; 52:2993–2999. Johnson RLJ, Cassidy SC, Grover RF, Schutte JE, Epstein RH. Functional capacities of lungs and thorax in beagles after prolonged residence at 3,100 m. J Appl Physiol 1985; 59:1773–1782. Schoene RB. Control of ventilation in climbers to extreme altitude. J Appl Physiol 1982; 53:886–890. Weil JV, Byrne-Quinn E, Sodal IE, et al. Hypoxic ventilatory drive in normal man. J Clin Invest 1970; 49:1061–1072. Collins DD, Scoggin CH, Zwillich CW, Weil JV. Hereditary aspects of decreased hypoxic response. J Clin Invest 1978; 62:105–110. Kawakami Y, Yoshikawa T, Shida A, Asanuma Y, Murao M. Control of breathing in young twins. J Appl Physiol 1982; 52:537–542. Robbins PA. Hypoxic ventilatory decline: site of action. J Appl Physiol 1995; 79:373– 374. Lugliani R, Whipp BJ, Sears C, Wasserman K. Effect of bilateral carotid-body resection on ventilatory control at rest and during exercise in man. N Engl J Med 1971; 285:1105–1111. Martin BJ, Sparks KE, Zwillich CW, Weil JV. Low exercise ventilation in endurance athletes. Med Sci Sport Exerc 1979; 11:181–185. Martin B, Weil JV, Sparks K, McCullough RE, Grover RF. Chemical drives to breathe as determinants of exercise ventilation. Chest 1978; 73:283. Martin BJ, Weil JV, Sparks KE, McCullough RE, Grover RF. Exercise ventilation correlates positively with ventilatory chemoresponsiveness. J Appl Physiol 1978; 45: 557–564.

172

Smith et al.

169.

Schoene RB, Robertson HT, Pierson DJ, Peterson AP. Respiratory drives and exercise in menstrual cycles of athletic and nonathletic women. J Appl Physiol 1981; 50:1300– 1305. Kojima A, Masuyama A, Hasako K, Kouchiyama S, Shinozaki T, Tatsumi K, Kimura K, Houda Y. Japanese high altitude climbers in Mt. Everest Expedition revealed higher hypoxic ventilatory drive than the others (abstr). In: Sutton JR, Coates G, Remmers JE, eds. Hypoxia: The Adaptations. Philadelphia: BC Dekker, 1990:293. Masuyama S, Kimura H, Sugita T, Kuriyama T, Tatsumi K, Kunitomo F, Okita S, Tojima H, Yuguchi Y, Watanabe S, et al. Control of ventilation in extreme-altitude climbers. J Appl Physiol 1986; 61:500–506. Schoene RB, Lahiri S, Hackett PH, Peters RM, Milledge JS, Pizzo CJ, Sarnquist FH, Boyer SJ, Graber DJ, Maret KH, et al. Relationship of hypoxic ventilatory response to exercise performance on Mount Everest. J Appl Physiol 1984; 56:1478–1483. Schoene RB, Hackett PH, Roach RC. Blunted hypoxic chemosensitivity at altitude and sea level in an elite high altitude climber. In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and Cold. New York: Praeger, 1987:532. Reeves JT, McCullough RE, Moore LG, Cymerman A, Weil JV. Sea-level Pco2 relates to ventilatory acclimatization at 4,300 m. J Appl Physiol 1993; 75:1117–1122. Anholm JD, Houston CS, Hyers TM. The relationship between acute mountain sickness and pulmonary ventilation at 2,835 meters (9,300 feet). Chest 1979; 75:33–36. King AB, Robinson SM. Ventilation response to hypoxia and acute mountain sickness. Aerosp Med 1972; 43:419–421. Mathew L, Gopinathan PM, Purkayastha SS, Sen Gupta J. Chemoreceptor sensitivity and maladaptation to high altitude in man. Eur J Appl Physiol 1983; 51:137–144. Matsuzawa Y, Fujimoto K, Kobayashi K, et al. Blunted hypoxic ventilatory drive in subjects susceptible to high-altitude pulmonary edema. J Appl Physiol 1989; 66:1152– 1157. Moore LG, Harrison GL, McCullough RE, McCullough RG, Micco AJ, Tucker A, Weil JV, Reeves JT. Low acute hypoxic ventilatory response and hypoxic depression in acute altitude sickness. J Appl Physiol 1986; 60:1407–1412. Bartsch P, Paul A, McCullough RE, Kucherer H, Hohenhaus E. Hypoxic ventilatory response and hypoxic pulmonary vascular response in HAPE-susceptible subjects. In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and the Brain. Burlington: Queen City Press, 1995:265–270. Hackett PH, Roach RC, Schoene RB, Harrison GL, Mills WJ, Jr. Abnormal control of ventilation in high-altitude pulmonary edema. J Appl Physiol 1988; 64:1268–1272. Hohenhaus E, Paul A, McCullough RE, Kucherer H, Bartsch P. Ventilatory and pulmonary vascular response to hypoxia and susceptibility to high altitude pulmonary oedema. Eur Respir J 1995; 8:1825–1833. Matsuzawa Y, Kobayashi T, Fujimoto K, Kubo K, Sekiguchi M. Hypoxic ventilatory response and pulmonary gas exchange during exposure to high altitude in subjects susceptible to high altitude pulmonary edema (HAPE). In: Sutton JR, Houston CS, Coates G, eds. Hypoxia and the Brain. Burlington: Queen City Press, 1995:331. Hackett PH, Rennie ID. Fluid retention and relative hypoventilation in acute mountain sickness. Respiration 1982; 43:321–329. Honig A. Peripheral arterial chemoreceptors and reflex control of sodium and water homeostasis. Am J Physiol 1989; 257:R1282–R1302. Swenson ER, Duncan TB, Goldberg SV, Ramirez G, Ahmad S, Schoene RB. Diuretic

170.

171.

172.

173.

174. 175. 176. 177. 178.

179.

180.

181. 182.

183.

184. 185. 186.

Control of Breathing

187.

188.

189. 190. 191. 192.

173

effect of acute hypoxia in humans: relationship to hypoxic ventilatory responsiveness and renal hormones. J Appl Physiol 1995; 78:377–383. Hornbein TF, Townes BD, Schoene RB, Sutton JR, Houston CS. The cost to the central nervous system of climbing to extremely high altitude. N Engl J Med 1989; 321:1714– 1719. Masuyama S, Kohchiyama S, Shinozaki T, Okita S, Kunitomo F, Tojima H, Kimura H, Kuriyama T, Honda Y. Periodic breathing at high altitude and ventilatory responses to O2 and CO2. Jpn J Physiol 1989; 52:523–535. Raff H, Tzankoff SP, Fitzgerald RS. Chemoreceptor involvement in cortisol responses to hypoxia in ventilated dogs. J Appl Physiol 1982; 52:1092–1096. Takano N. Changes in ventilation and ventilatory response to hypoxia during the menstrual cycle. Pflugers Arch 1984; 402:312–316. White DP, Douglas NJ, Pickett CK, Weil JV, Zwillich CW. Sexual influence on the control of breathing. J Appl Physiol 1983; 54:874–879. Tatsumi K, Hannhart B, Pickett CK, Weil JV, Moore LG. Influences of gender and sex hormones on hypoxic ventilatory response in cats. J Appl Physiol 1991; 71:1746– 1751.

7 Mechanics of Breathing

JOSEPH MILIC-EMILI

BENGT KAYSER

McGill University Montreal, Quebec, Canada

University of Geneva Geneva, Switzerland

HENRY GAUTIER Atelier de Physiologie Respiratoire Faculte´ de Me´decine Saint-Antoine Paris, France

I.

Introduction

Ventilation is higher at altitude than at sea level (Table 1) (1,2). Reinhold Messner, describing his and Peter Habeler’s approach to the summit of Mount Everest, the first to climb Mount Everest without supplementary bottled oxygen, stated: ‘‘Breathing becomes such a strenuous business that we scarcely have strength to go on,’’ and upon reaching the summit: ‘‘I am nothing more than a single, narrow, gasping lung, floating over the mists and the summits’’ (3). Expressions like these from many other mountaineers venturing to high altitude suggest the possibility that ventilation might limit exercise at least at extremely high altitude. In this chapter we provide an overview of how respiratory mechanics at high altitude may affect the work of breathing and exercise performance (see also Chapter 20). We will focus mainly on research performed on humans. Altitude has both a direct effect on respiratory mechanics because of decreased density of ambient air and an indirect effect linked to hypoxic responses. The lower density should lead to a decrease in airway resistance and dynamic work of breathing together with increased maximal inspiratory and expiratory flows. On the other hand, hypoxia may affect the static lung volumes through increased intrathoracic blood 175

176

Milic-Emili et al.

Table 1 Ventilation at Rest and During Exercise at Different Barometric Pressures (Pb) in a Decompression Chamber (Operation Everest II) Pb (mmHg) 760 429 347 282 252

Ve (l/min)

Altitude (m)

Pio 2 (mmHg)

Rest

0 4,800 6,300 8,100 8,848

149 80 63 49 42

11 15 21 37 42

a

120 watts b 48 72 92 162 184

Pio 2 ⫽ calculated inspired partial pressure of O2; Ve ⫽ minute ventilation corrected to BTPS. a From Ref. 1. b From Ref. 2.

volume and pulmonary edema, and pulmonary dynamics through changes in bronchomotor tone. The latter changes depend on (1) time at altitude—acute exposure (⬍24 hours), short-term acclimatization (days to weeks), and long-term acclimatization (years to generations)—and (2) the population studied—subjects born and living at low altitude (lowlanders, most commonly Caucasians) or subjects born and living at high altitude (highlanders, commonly Indians of South America and Tibetans of the Himalayas). In lowlanders, the changes in respiratory mechanics with altitude have been assessed in field studies carried out on mountains as well as in simulated altitude experiments in which the effect of low barometric pressure (Pb) per se was studied in decompression chambers (hypobaric hypoxia). In order to evaluate the effects of hypoxia per se without possible side effects of reduced Pb, gas mixtures low in oxygen have also been administered at sea level (normobaric hypoxia). Similarly, at reduced Pb (both on mountains and in decompression chambers), the effects of oxygen administration have also been investigated, but only for relatively short periods of O2 administration. In a few acute and short-term experiments highlanders were studied at sea level.

II. Lung Volumes First we will describe the effects of altitude on the static subdivisions of lung volume, and next those on ventilatory capacity.

Mechanics of Breathing

177

A. Static Volumes Acute Exposure (⬍24 Hours) Decompression Chamber

According to Bert (4), the first measurements of the vital capacity (VC) during acute exposure to hypoxia were made by von Vivenot in 1868. In a low-pressure chamber at Pb of 424 mmHg (corresponding to an altitude of 4650 m), the vital capacity of two subjects was reduced by 9 and 13%. Bert found that his own vital capacity was reduced by 32% at Pb of 430 mmHg. In 1932, Schneider (5) reported that 24 subjects exposed to a simulated altitude of 6100 m had an average 8% decrease in VC. He stated that up to about 3000 m of simulated altitude there was no appreciable change in VC, and that at 6100 m the reduction in VC was virtually abolished by pure oxygen breathing. In 1934, Hurtado et al. (6) reported that in three subjects exposed to a simulated altitude of 5,000 m the vital capacity decreased on average by 9%. They also found that the residual volume (RV), measured with the nitrogen washout method, increased by 45%. As a result the total lung capacity (TLC) was essentially unchanged. In all these early studies, the vital capacity values were not corrected to BTPS conditions. Such corrections were made by Rahn and Hammond (7) in a systematic study on 18 subjects exposed to 3050, 4300, and 5500 m. Like Schneider (5), they found no significant change in VC at 3050 m with a small but significant reduction, averaging 2.4 and 3.8% at 4300 and 5500 m, respectively. Without correction for BTPS conditions, the change in VC at 5500 m relative to sea level would be about 5% greater, for a spirometer temperature of 20°C. Rahn and Hammond (7) also acutely exposed four subjects to very low barometric pressures, ranging from 349 to 141 mmHg. Because the subjects were not acclimatized to high altitude, in these experiments the inspired oxygen tension was in all instances kept higher than 94 mmHg. At Pb of 349 mmHg, corresponding to 6100 m, the VC was the same as at sea level, in line with the previous observations of Schneider (5). At lower Pb, however, the VC was less than at sea level, the reduction averaging 4% at Pb of 226 mmHg and 7% at Pb of 141 mmHg. In eight normal subjects breathing pure oxygen exposed to Pb of 380 and 190 mmHg, Finkelstein et al. (8) found a significant reduction of VC averaging 5 and 15%, respectively. These changes were higher than those found by Rahn and Hammond (7) at similar altitudes and were significant at Pb of 380 mmHg. After 4 hours at simulated altitude of 4900 m, Gray et al. (9) found a 7% decrease in vital capacity in five subjects, whereas after 5 hours at 4300 m Coates et al. (10) did not find any significant change in the VC of four subjects but a 78% increase in RV measured with the helium dilution method. As a result, TLC was increased by 21%. They ascribed the rise in RV to increased gas trapping. Indeed, they found a concomitant increase in closing capacity [(phase IV ⫹ RV)/TLC], which was due almost entirely to the increase in residual volume. By contrast, the closing volume (phase IV, % VC) did not change in line with the results of Gray et al. (9), who found no significant changes in either the phase IV volume or the

178

Milic-Emili et al.

closing volume. The increase in TLC observed by Coates et al. (10) was surprisingly large as compared to the 5% increase reported by Saunders et al. (11) during acute normobaric hypoxia simulating an altitude of about 4300 m (see below). Normobaric Hypoxia

To nine of the subjects who had previously served in the decompression chamber experiments (see above), Rahn and Hammond (7) administered 14.2, 11.8, and 9.9% oxygen at sea level. Only with 9.9% oxygen did they find a significant reduction in VC (1.4%), which was smaller than the 3.8% found in the same subjects with the same inspired Po 2 in the decompression chamber. By contrast, with a Pao 2 reduced to 40–50 mmHg, such as found at an altitude of about 4300 m, and a Paco 2 maintained at 38–40 mmHg, Saunders et al. (11), using a body plethysmograph, found no change in VC, whereas the TLC, FRC, and RV increased significantly by 5, 9, and 31%, respectively. All of these changes were reversed within 3 minutes of pure O2 breathing. They suggested that the rise in TLC and FRC could be explained by the concomitant loss of elastic recoil of the lungs. By contrast, when lung volumes were measured by inert gas dilution, with a degree of hypoxia similar to that used by Saunders et al. (11), Goldstein et al. (12) found no significant changes in TLC, VC, closing volume, and closing capacity. They suggested that the changes in TLC reported by Saunders et al. (11) were due to artefacts inherent in plethysmography. It should be noted, however, that by helium dilution method Coates et al. (10) also found a significant increase in TLC at simulated altitude of 4300 m in a decompression chamber. In line with Saunders et al. (11), an average increase in FRC of 14%, also measured with a body plethysmograph, has been also reported by Garfinkel and Fitzgerald (13) in 43 subjects breathing 11% oxygen at sea level. In short, the scanty literature dealing with the effects of acute simulated altitude exposure on RV, FRC, TLC, and closing capacity has yielded conflicting results, and further systematic studies are needed. In contrast, according to most studies there is a small but significant reduction in VC when humans are acutely exposed to equivalent altitudes higher than about 3000 m (Pb ⬍ 525 mmHg). At altitudes between 3000 and 5000 m, this reduction in VC is completely reversed by oxygen inhalation, while at higher altitudes the reversal is only partial. The latter phenomenon may be related in part to expansion of the gases below the diaphragm. Indeed, Rahn and Hammond (7) noted that in many of their subjects the forced expiration at very high simulated altitudes elicited abdominal pain. The observation that, at the same Pb, the reduction in VC is greater with hypobaric hypoxia than with hypobaria alone has been attributed to engorgement of pulmonary blood vessels and decreased respiratory muscle force under hypoxic conditions (7). Short-Term Exposure (Days to Weeks) Decompression Chamber

Only two studies have been concerned with short-term exposure to altitude in a decompression chamber. In the study carried out during Operation Everest II by

Mechanics of Breathing

179

Welsh et al. (14), subjects resided for 40 days with Pb decreasing progressively to 240 mmHg, corresponding to an altitude of 8844 m. The VC began to decrease significantly at Pb lower than 429 mmHg (equivalent altitude of 4572 m), diminishing by 14% at Pb of 240 mmHg (Fig. 1). Unfortunately, in these experiments the effect of oxygen administration on VC was not assessed. However, Ulvedal et al. (15) studied subjects breathing enriched-oxygen mixtures (Pio 2 ⬎ 160 mmHg) at Pb of 380, 260, and 190 mmHg for 30, 14, and 17 days, respectively. They found a significant and sustained 3% decrease in VC at Pb of 380 and 260 mmHg and 7% at Pb of 190 mmHg. These changes were smaller than those reported by Welsh et al. (14). Field Studies

Mosso in 1897 (16) was the first to study the effects of a short-term sojourn on mountains on VC. On Monte Rosa (4560 m), an average decrease of 10% in VC (not corrected to BTPS) was observed in eight subjects. In 1932 on Pike’s Peak (4300 m), Schneider (5) found a similar decrease in VC, which ranged from 7 to 15% (also not corrected to BTPS) on the first day of residence at altitude with a tendency to a return to sea level values in the following days. At the same altitude,

Figure 1 Vital capacity, VC (mean ⫾ SEM), decreased significantly (* p ⬍ 0.05 from sea level) and progressively with chronic altitude exposure (6096–8844 m). There was rapid partial recovery 30 minute after descent (n ⫽ 7 at 0, 429, and 347 torr and for all postaltitude measurements; n ⫽ 6 at 282 torr; n ⫽ 4 at at 240 torr). (From Ref. 14.)

180

Milic-Emili et al.

a transient decrease in VC (3–4% corrected to BTPS) has also been found in several subsequent studies (7,17–20). The decrease was partially reversed when pure oxygen was breathed for 15 minutes (5), whereas when it was administered for only 3 minutes no significant changes were observed (20). In four subjects studied at 4300 m, Tenney et al. (20), using the N2 washout method, found a 10% increase of FRC and RV on days 3–7 of altitude sojourn. The corresponding changes were somewhat greater on days 1–3, but there was marked variability in response between the four subjects. Using the same method, after a 3-day sojourn at the same altitude, Cruz (17) found no significant change in FRC in 6 subjects, while Jaeger et al. (18) found a significant 10% increase in RV in 17 individuals. After a 30-day sojourn at 5366 m, Mansell et al. (21), using the helium dilution method, found a significant increase in TLC (18%), reflecting entirely a large increase in RV (78%), while VC did not change significantly. During short-term sojourns at altitudes lower than 4000 m, the changes in static lung volumes are usually small, if any. Indeed, no significant changes in VC, FRC, and TLC were observed after a 1- to 6-week sojourn at either 3100 m (22,23) or 3660 m (24). Using a body plethysmograph, Gautier et al. (25) made daily measurements of RV, FRC, TLC, and VC in nine subjects during a 6-day sojourn at an altitude of 3460 m. In comparison to sea level values, there was no change in either TLC or RV, while there was a small but significant drop in FRC only on the first day at altitude (Fig. 2). Small but significant decreases in VC were also found on the second and third days at altitude. These changes in VC were reversed by breathing an oxygen-enriched mixture (Pio 2 ⫽ 150–160 mmHg). According to the above results, during short-term altitude exposure, a significant decrease in VC is found only above 4500 m, while during acute exposure it may be observed at lower altitudes. The reason for this discrepancy is not clear because the mechanisms that may contribute to the reduction in VC during shortterm altitude exposure are essentially the same as those pertaining to acute exposure, as elegantly reviewed by Rahn and Hammond (7). These may include decreased respiratory muscle force as well as an increase in thoracic blood volume and interstitial fluid associated with hypoxia (26). The latter may explain the observation that after return from high to low altitude, it takes several days to restore the VC to its sea level value, as shown in Figure 1. At altitudes lower than 4000 m, there are no significant sustained changes in RV, FRC, and TLC. At higher altitudes, however, all of these volumes increase. According to Mansell et al. (21), the increase in RV may reflect an increase in closing capacity due to interstitial lung edema and/or loss of lung recoil. By contrast, the increase in FRC and TLC is probably due to the fact that at altitudes higher than 4000 m the effect of increased thoracic blood volume is more than counterbalanced by the concomitant loss in lung elastic recoil (see Sec. III.A). Long-Term Exposure (Years to Generations)

The first systematic study of subjects born and living at high altitude was made by Hurtado (27), who showed that young Indians born in Morococha (4540 m) had a

Mechanics of Breathing

181

Figure 2 Average values (⫾SEM) of lung volumes at sea level (control) and during a 6day sojourn at 3460 m of nine subjects (filled circles). Breathing ambient air (Pio 2 ⫽ 93 mmHg) (open circles). Breathing a mixture with Pio 2 ⫽ 150–160 mmHg. TLC, total lung capacity; VC, vital capacity; FRC, functional residual capacity; RV, residual volume. Levels of significance between altitude and sea level values: * p ⬍ 0.05; ** p ⬍ 0.01. (From Ref. 25.)

38% larger RV than individuals of the same ethnicity born and living at sea level. The vital capacity was approximately the same in both groups, and, as a result, the TLC was larger in the highland natives. In similar studies carried out at 4350 m, a greater FRC was also found in highland natives (17). Other studies carried out in the Andes (3850 m) have shown that the VC is larger in high-altitude natives than in sea level dwellers of the same ethnic origin (28,29). In Nepalese Sherpas born and living above 3340 m and studied at 4243 m, the vital capacity was 12% higher than that predicted for sea level dwelling Caucasians (30). At altitudes of 3500–4500 m in eastern Kashmir (31) and at 3660 m in Lhasa (32), the VC of the highland natives was found to be higher than that of lowland natives acclimatized to those altitudes. Similarly, a relatively high VC was

182

Milic-Emili et al.

also found in Peruvian and Bolivian native children born and living at high altitude (4270 and 3600 m, respectively) (33,34). From studies carried out in the Andes, it appears that the increased lung volume of highland natives is mainly the result of adaptations to hypoxia occurring during growth. Indeed, after the age of 11 years, the VC of Peruvian boys living at an altitude of 4270 m is higher than in Peruvian boys living at a lower altitude or at sea level (33). Furthermore, Peruvian natives born at sea level but acclimatized to high altitude as adults exhibit smaller VC than Peruvian natives acclimatized during growth or born and living at altitude (35). The above observations are in agreement with results obtained in rats. After prolonged exposure to altitude (3540 m), at a given distending pressure the lungs of young rats were larger than those of control rats kept at sea level. In contrast, under similar experimental conditions adult rats did not show any change in lung volume (36). Interestingly, Sekhon et al. (37) have shown that in growing rats exposed to hypobaric hypoxia, the lung volumes were slightly but significantly larger than in growing rats subjected to normobaric hypoxia. This suggests that decreased Pb per se may have a small growthpromoting effect on lung volume. The effects of descent of high altitude natives to sea level on lung volumes have been investigated in Andean Indians. In Bolivian highlanders born and living above 3500 m, a 1-week sojourn at low altitude (420 m) did not induce any change in TLC or its components (38), a finding replicated in Peruvian natives of 3500– 4500 m who, after 2 and 37 days at sea level, did not change their TLC or its subdivisions, which remained larger than predicted for sea level dwellers (39). In contrast to non-Caucasian natives, Caucasians living at high altitude do not appear to have an increased VC. In Leadville (3100 m), 126 normal Caucasians, aged 18–61 years, were studied to establish the normal predicted values for this altitude. The results indicate that (1) the regression equation for VC was the same for the subjects who were born at altitude and the subjects who had moved to altitude as adults and (2) the predicted values at Leadville were virtually identical to those predicted from sea level regressions (40). In lifelong Leadville residents the VC was also found to be within the range of the predicted values for lowlanders, while the TLC was significantly larger (23). In children of European ancestry born and living at 3600 m, the VC was also found to be within the normal range of sea level Caucasian children (34). B. Dynamic Volumes and Maximal Flows

In 1946, Rahn et al. (41) pointed out that the effective area of the maximal static volume-pressure (V-P) diagram, and hence the maximal potential work available per breathing cycle, decreases with reduced barometric pressure because of compressibility of gas (Fig. 3). This, however, is more than compensated for by a decrease in airway resistance resulting from the lower density of air and possibly also from bronchodilatation (see Sec. III.B), so that the maximal voluntary ventilation (MVV) actually increases at altitude.

Mechanics of Breathing

183

Figure 3 The effect of altitude on the dynamic expiratory and inspiratory pressures that can be exerted starting, respectively, at maximum inspiration and maximum expiration (broken lines). Pi and Pe, maximal static inspiratory and expiratory pressures. Pr, relaxation volume-pressure curve of total respiratory system. (From Ref. 41.)

In 1954, Cotes (42) measured the MVV in a decompression chamber at simulated altitudes of 3050, 5200, and 8200 m (acute exposure) in a group of subjects breathing an oxygen-enriched mixture. Compared to sea level, the MVV increased by 13, 24, and 31%, respectively. In similar acute experiments carried out at equivalent altitudes of 5500 and 10300 m, Finkelstein et al. (8) found an increase in MVV of 15 and 24%, respectively. Ulvedal et al. (15), also in a decompression chamber, measured MVV in subjects breathing oxygen-enriched mixtures during short-term exposure to a simulated altitude of 10,200 m (hypobaric normoxia). Compared to sea level, on days 3 to 17, the MVV rose on average by 37%. The above authors concluded that the main factor influencing the MVV during acute and short-term exposure to altitude is the reduction in air density, not hypoxia per se. In 1958, Pugh (43) found that, in three subjects breathing ambient air during a sojourn on a high mountain (6400 m), the MVV was about 20% higher than at lower altitude (2100 m). The increase in MVV at altitude was confirmed in several

184

Milic-Emili et al.

field studies carried out at various altitudes (3700–4350 m) on a larger number of individuals during sojourns ranging from 3 to 65 days (17,19,24,44). Figure 4 depicts the average increase in MVV, relative to sea level, found at different altitudes in the various field studies as well as in the decompression chamber. There are no consistent differences between the results obtained in the field and decompression chamber studies although hypoxia was present only in the former. Also see in Figure 4 is the substantial scatter of the data. This is not surprising because the changes in MVV with altitude depend on many factors apart from gas density (force and speed of activation of both inspiratory and expiratory muscles, thoracic gas compressibility and mechanical properties of the respiratory system, respiratory frequency, etc). By contrast, the effects of decreased air density on maxi˙ max) can readily be predicted. An increase in maximal expimal expiratory flows (V ratory flows at altitude has been reported in several studies in which a body plethysmograph was not used (8,14,19,21,29). However, because of artefacts due to gas compressibility, the use of a body plethysmograph is mandatory for measurements ˙ max at altitude, and hence the above results are difficult to interpret (41,45). of V Figure 5 shows the maximal expiratory flows obtained by Gautier et al. (25) using a body plethysmograph in nine subjects during a 6-day sojourn at an altitude of 3460 m. Peak flow (PF) increased significantly on the first day at altitude and remained at about the same level thereafter. The maximal flows at 50 and 75% VC were signifi-

Figure 4 Percent change of maximal voluntary ventilation relative to sea level (∆MVV, %) found at different altitudes by various authors in field studies (filled circles) and decompression chamber (open circles). (From Refs. 8,15,17,19,24,42–44.)

Mechanics of Breathing

185

Figure 5 Average values (⫾SEM) of peak expiratory flow (PF) and maximum expiratory ˙ ) at 75, 50, and 25% of vital capacity at altitude, expressed as a fraction of correspondflows (V ing values at sea level (broken lines). Predicted values of PF and V at altitude. For further explanation, see text. (From Ref. 25.)

cantly increased from the first day at altitude but reached a plateau only by the second day. Maximum flow at 25% VC increased until day 2 but then progressively decreased in the following days. Also shown in Figure 5 are the values expected solely on the basis of decreased air density. The latter were predicted according to ˙ max at 25% VC, the data for day 1 closely fit Wood and Bryan (46). Except for V the predicted values. Thereafter the flows exceeded the predicted values on all days ˙ max at 75 and 50% VC and on 3 days for V ˙ max at 25% VC, suggesting that for V there was a bronchodilatation during that period. Since after day 1 at altitude there was a loss in elastic recoil of the lungs (see Sec. III.A), the dilatation of the flowlimiting segments of the airways must have been substantial, as per se the loss of ˙ max (47). Loss in recoil, however, may explain lung recoil should have decreased V ˙ max at 25% VC after day 2 at altitude. the progressive decrease in V To our knowledge, apart from Gautier et al. (25), there are only two other studies of peak expiratory flow made at altitude on lowlanders, namely that of Cruz (17) during the third day at 4350 m and that of Mansell et al. (21) after 9–30 days

186

Milic-Emili et al.

at 5366 m. In the former study the ratio of peak flow at altitude relative to sea level averaged 1.14 compared with a ratio predicted according to Wood and Bryan (46) of 1.23, while in the case of Mansell et al. (21), the corresponding experimental ratio averaged 1.21 compared with a predicted value of 1.32. The results of both studies are in disagreement with those of Gautier et al. (25), which showed a good agreement between predicted and experimental values of peak flow (Fig. 5). It should be noted, however, that peak flow is effort-dependent (46,48), and hence predictions based on changes in gas density alone may be inherently less precise ˙ max. than in the case of V In most studies, a slight increase of the forced expired volume in one second (FEV1) has been observed at altitude (8,14,17,19,21,24,25,29,44). The relative sta˙ max probably mainly bility of FEV1 at altitude in the face of the marked increase in V reflects the fact that as a result of gas compression, the FEV1 is exhaled at lower thoracic gas volumes (and hence lower maximal flows) than at sea level. At 3660 m, Lefranc¸ois et al. (24) measured FEV1 and VC in 18 Bolivian natives born and living at this altitude and in 10 Caucasian lowlanders after a 30day sojourn at that altitude. No significant differences were observed. However, in 126 residents at 3100 m, Kryger et al. (40) found a slightly but consistently greater FEV1 than that predicted for sea level Caucasians while the VC was in the range of predicted. In order to evaluate the role of genetic and environmental factors, Brody et ˙ max at 50 and 25% VC in Peruvian native highlanders (3850 al. (29) measured V m) and lowlanders (800 m), both studied at their respective residence levels. The

Figure 6 Average values of peak inspiratory flow (PF) and maximum inspiratory flows at 75, 50, and 25% vital capacity. Levels of significance between altitude and sea level values as in Figure 2. (From Ref. 25.)

Mechanics of Breathing

187

highlanders had a higher VC than the lowlanders (respectively 116 and 84% of ˙ max were lower in highlanders than in lowlandpredicted). Because the values of V ers, the authors suggested that the airways, which form in fetal life, do not participate in adaptation to altitude, and that the large lungs of highlanders are due to postnatal environmental hypoxic stimulation of their growth. It should be noted, however, that Brody et al. did not measure volume with a body plethysmograph and, therefore, ˙ max 50 and 25% VC were obtained at a lower thoracic gas volume the values of V (TGV) in the highlanders than in the lowlanders because of the different Pb at which the measurements were made. At low ambient Pb, the thoracic gas volume during an FVC maneuver is necessarily lower than at higher Pb, because the gas compress˙ max found in the highlandibility is greater (Fig. 3). As a result, the lower values of V ers may merely reflect artefacts due to gas compressibility. Clearly, further studies ˙ max using body plethysmography are needed to assess the iso-TGV changes in V with altitude. Using body plethysmography, Gautier et al. (25) provided the only available values of maximal inspiratory flows at altitude. As shown in Figure 6, the maximal inspiratory flows increased significantly by the second day of sojourn.

III. Mechanical Properties of the Respiratory System A. Static Volume-Pressure Curves and Compliance

In four subjects after 3 days at 4100 m, Kronenberg et al. (49) measured the static lung compliance (Cst,L) during stepwise lung inflation from FRC to 70–80% TLC and found a significant 20% reduction relative to sea level. Using the same method on four subjects at an altitude of 4300 m, a small (statistically insignificant) increase in Cst,L was found during 2–10 days at altitude (50). In both of these contradictory studies, the lung volume history was similar but different from that used in conventional studies of the elastic properties of the lung (lung deflation from TLC). In 10 subjects after 30 days at 3660 m, Lefranc¸ois et al. (24) measured the static V-P curves of the lung and chest wall during stepwise deflation from TLC. They stated that both curves did not change at altitude. However, inspection of their data reveals a parallel shift to the left of the V-P curve of the lung, reflecting a small loss of lung recoil at all volumes considered. Subsequently, Brody et al. (29) measured the static V-P curve of the lungs in three subjects after 3 days at 3850 m. They stated that lung elastic recoil was unchanged but did not provide the actual data. Mansell et al. (21) measured the static deflation V-P curve of the lungs in three subjects after 9 days and in four subjects after 30 days at an altitude of 5366 m. Except at TLC, at all volumes (% TLC), the elastic recoil of the lung was reduced at altitude, the difference increasing progressively from TLC to FRC. However, Cst,L in the midrange of lung volume (FRC ⫹ 1 L) remained the same at altitude as at sea level. Similar results were obtained by Gautier et al. (25) in nine subjects during a 6-day sojourn at 3460 m, as shown in Figure 7. During altitude sojourn, the static V-P curve of lungs was shifted progressively to the left. The decrease in elastic recoil,

188

Milic-Emili et al.

Figure 7 Average static deflation V-P relationships of lung at sea level and for the first, third, and fifth days at altitude of 3460 m. Volumes are expressed as percent of total lung capacity (% TLC). (From Ref. 25.)

which was almost complete by day 4, averaged about 2 cmH2O and was significant on days 4–6. Cst,L measured from 60 to 70% TLC did not change significantly. Jaeger et al. (18) measured the quasi-static deflation V-P curve of the lungs on 11 soldiers who participated in a 72-hour field exercise at an altitude of 3000– 4300 m. They found a rightward shift of the V-P curve above FRC while below FRC the curve was shifted to the left. In line with Frank et al. (51), they interpreted this change as suggestive of lung congestion and/or interstitial edema. Thus, the discrepancy with the results of Mansell et al. (21) and Gautier et al. (25) may be ascribed to the fact that the soldiers performed strenuous exercise. The loss of elastic recoil of the lung during short-term exposure to altitude may explain the increase in RV and closing capacity. Indeed, an increase in closing capacity at altitude has been reported by Coates et al. (10) and Jaeger et al. (18). The loss of lung recoil at altitude should also per se cause an increase in the elastic equilibrium volume of the respiratory system (Vr) and hence in FRC (25). However, Gautier et al. (25) found that on days 2–6 at an altitude of 3460 m the FRC was the same as at sea level (Fig. 2) in spite of a significant loss of lung recoil (Fig. 7). They attributed this phenomenon to a rightward shift of the static V-P curve of the chest wall due to increased thoracic blood volume. In fact, the changes in FRC at altitude depend on the balance between the loss in recoil of the lung, which should increase Vr, and the opposing effect of loss in recoil of the chest wall due to increased thoracic blood volume (25). In the subjects of Gautier et al., these two effects were balanced on days 2–6, and hence there was no change in FRC relative to sea level. On day 1, however, there was very little change in elastic recoil, and hence

Mechanics of Breathing

189

the FRC actually decreased probably as a result of increased thoracic blood volume. The sustained increase in FRC found at altitudes higher than 4000 m (20,21) probably reflects a predominant effect of loss of elastic recoil of the lungs, which may also explain the concomitant increase in TLC. Brody et al. (29) measured the static deflation characteristics of the lungs in Peruvian native highlanders (3850 m) and lowlanders (800 m), both studied at their respective residence levels. Lung recoil at FRC and TLC, as well as the sizecorrected V-P curves, were similar in the two groups, suggesting that in highlanders there is no loss in elastic recoil. This is in contrast to the loss of lung recoil found in lowlanders during short-term exposure to altitude. Brody et al. (29) also found that, at any given static transpulmonary pressure, the lung volumes of Peruvian highlanders, and hence also Cst,L, was higher than in the Peruvian lowlanders, reflecting their larger lungs. A similar conclusion was reached by Mortola et al. (52), who compared the respiratory system compliance of Bolivian infants born at high altitude (3600 m) with that of Bolivian infants of similar ethnic origin born at 400 m. They found that in the highland infants the respiratory system compliance was 33% higher but concluded that this reflected the larger lungs of highlanders due to their hypoxic environment rather than a genetic characteristic. These results are compatible with the concept that fetal hypoxia at high altitude is more marked than at low altitude and that the developing fetus can recognize environmental hypoxia and respond to it. Several investigators have reported that altitude has no significant effect on dynamic lung compliance (17,21,24,25). B. Resistance

Using the interrupter technique in six lowlanders exposed to 4350 m for 3 days, Cruz (17) found a 7% decrease in resistance versus 17% expected by modification in air density and proposed this as evidence for reduction in large airway size due to bronchoconstriction. At 5366 m, Mansell et al. (21) found a decrease in pulmonary resistance (R L) averaging 29% relative to sea level. In the study of Gautier et al. (25), R L and airway resistance (Raw) decreased by an average of 14 and 17%, respectively, on day 1 at altitude, and a further substantial decrease of RL was found on day 2 (Fig. 8). Thereafter RL remained essentially constant. The drop in resistance caused by the decrease in gas density at altitude can be predicted according to the formula proposed by Vare`ne et al. (53): Rx/Rc ⫽ 0.57 ⫹ 0.44 Pb where Pb is in atmospheres, and Rx/Rc is the ratio of resistance at the altitude corresponding to Pb to sea level resistance. At the altitude of the study of Gautier et al. (25) (Pb ⫽ 0.67 ATA), the Rx/Rc ratio should amount to 0.86. On day 1 at altitude the measured RL , expressed as a fraction of the corresponding value at sea level, was in agreement with the prediction (0.86), indicating that the change in RL could be entirely explained by decreased air density. After day 1, however, the

190

Milic-Emili et al.

Figure 8 Average values (⫾SEM) of airway resistances (Raw, open circles) and pulmonary flow resistances (Rl, filled circles) at sea level and at altitude. Levels of significance between altitude and sea-level values: *p ⬍ 0.05; **p ⬍ 0.01; ***p ⬍ 0.001. (From Ref. 25.)

observed ratios were smaller than predicted, averaging 0.7 over the 2- to 6-day span. This suggests the occurrence of bronchodilatation (see below). Raw was also smaller than predicted during the stay at altitude. These results are consistent with the time course of the changes in maximum inspiratory and expiratory flows observed at altitude (Figs. 5 and 6). C. Mechanisms Affecting Lung Mechanics at Altitude

It has been suggested that engorgement of the pulmonary vascular bed develops in individuals who ascend to high altitude (18,26,31,49,54,55). The available evidence, however, indicates that acute elevation in pulmonary blood volume has relatively little effect on the static V-P curve of the lungs, except at lung volumes lower than FRC, where Pst,L tends to decrease (51). Accordingly, the substantial loss in Pst,L observed over the lung volume range between 60 and 90% TLC (Fig. 7) cannot be ascribed to changes in pulmonary blood volume. Both the magnitude and direction of the changes in static V-P curve of the lung and in resistance observed after day 1 of sojourn at altitude are similar to those found by De Troyer et al. (56) on normal subjects after administration of a β 2-adrenergic agent (fenoterol). They reported that this drug caused a reduction of Raw associated with a shift to the left of the static

Mechanics of Breathing

191

deflation V-P curve of the lungs. In addition, the maximum expiratory flows increased. They postulated that increased activity of the β 2-adrenergic system caused a dilatation not only of the bronchial tree, but also a relaxation of smooth muscle in alveolar ducts or other contractile elements in the lung parenchyma leading to loss of elastic recoil of the lungs. Vincent et al. (57) found analogous results after administration of 1.2 mg atropine sulfate (iv) to normal volunteers. Thus, the decrease of resistance found during sojourns at altitude might be explained in part by a change in activity of the β 2-adrenergic and/or cholinergic systems. Increases in levels of catecholamines at high altitude have been reported (58). In this connection it should be noted that both the hypoxia and the hypocapnia present at altitude may be expected to cause increased resistance (59,60). Such an effect has not been reported except by Cruz (17), but it may become important at very high altitudes when the degree of hypocapnia and hypoxemia becomes more severe. At 4350 m, Cruz (17) measured resistance by the interrupter technique in eight high-altitude natives whose values were similar to those of six low-altitude natives studied after 3 days at the same altitude. Analogous results were found by Lefranc¸ois et al. (24) at 3660 m. However, in the highlanders endowed with large lungs, the non–size-corrected values of resistance would be expected to be lower than in lowlanders.

IV. Work of Breathing In theory, the maximal mechanical work available for a single breathing cycle is given by the area subtended by the curves relating the maximal voluntary static inspiratory and expiratory pressures to lung volume (Fig. 3). At sea level, for young male adults it amounts to 20–30 cal per breath. However, this potential work is never realized during the actual breathing movements. Agostoni and Fenn (48) demonstrated that the maximal inspiratory work that a subject can achieve decreases with increasing flow, in line with the force-velocity relationship of skeletal muscles. Other mechanisms, however, also contribute to limit the maximal work per breath; these include (1) the speed of activation of the respiratory muscles at the beginning of forced inspirations and expirations and (2) gas compressibility (61). As shown in Fig. 3, the effective area of the maximal static V-P diagram, and hence the maximal potential work available per breathing cycle, decreases with reduced barometric pressure because of compressibility of gas. As a result of decreased air density and bronchodilatation, the work of breathing required from the respiratory muscles for a given ventilation should decrease at altitude, particularly at high ventilations. Figure 9 shows the relationship between the mechanical power of breathing and ventilation in a normal adult breathing oxygen at different altitudes. These curves were obtained in a decompression chamber at various simulated altitudes (rapid ascent) ranging from 34 to 7500 m (62). Also shown in Figure 9 is a curve for simulated altitude of 8848 m (corresponding to the top of Mt. Everest), which was obtained by extrapolation. Whereas the respiratory power requirements for a

192

Milic-Emili et al.

Figure 9 Relationships between respiratory mechanical power (cal/min) and ventilation (L/min) at different simulated altitudes in a normal adult. The curve for 8848 m (corresponding to the top of Mt. Everest) was obtained by extrapolation. (From Ref. 62.)

given ventilation decrease with altitude, the opposite is true in terms of the ventilation required for a given O2 consumption during muscular exercise (see Chapter 20). In fact, breathing at extreme altitudes is a particularly interesting situation be˙ o 2 max) is so low whilst the maximum cause the maximum oxygen consumption (V exercise ventilation is so high. For example, during a simulated ascent to Mt. Ever˙ o 2max in three healthy young men averaged 1.12 est, the steady-state (5–8 min) V L/min, while the corresponding ventilation was 183.5 L/min (2). The respiratory mechanical power required for this ventilation and altitude should amount to 260 cal/min (62). Assuming a mechanical efficiency of 0.2 and a caloric equivalent per milliliter of O2 of 4.86 cal, the corresponding oxygen cost of breathing would amount ˙ o 2max (vs. 7% at sea level) (63). In contrast, a mechanical to 0.27 L/min, or 24% of V efficiency of only 0.05 should require a respiratory O2 cost of 1.07 L/min, or 95% ˙ o 2max. In this case, virtually all of the O2 uptake would be required to meet of V the demands of the respiratory muscles with almost nothing left for the rest of the body, let alone for external work! Clearly, a low mechanical efficiency of breathing should severely limit the work tolerance at extreme altitudes. It should be noted that the above analysis was based on results obtained on three individuals. Furthermore, during sojourn at altitude flow resistance may decrease as a result of bronchodilatation (see Sec. III.B). However, even if in the three subjects exercising at simulated altitude corresponding to Mt. Everest (see above)

Mechanics of Breathing

193

the mechanical power requirements of breathing were reduced to one half of those predicted in Figure 9 for that altitude (i.e., 130 instead of 260 cal/min, at efficiency ˙ o 2 available for the body would amount to only 0.59 L/min. This of 0.05), the net V is barely sufficient to satisfy the resting metabolic needs of the body. In this connection, Petit et al. (62) underestimated the mechanical power output of the respiratory muscles because their measurements did not include the work due to gas compressibility (61) and distortion of the chest wall (64), which at altitude may be substantial. In contrast to high altitude, breathing in hyperbaric environments increases the work of breathing (65). Exercise in dense environments is primarily limited by expiratory mechanisms, namely expiratory flow limitation (66), which does not appear to be a problem at high altitude. There are only two reports on changes in mechanical work of breathing from sea level to altitude. In six lowlanders, Cruz (17) reported that at rest the mechanical work on the lung per liter of ventilation was 22% higher at an altitude of 4350 m than at sea level in the same subjects. Surprisingly, this was due to an increase in resistive work. It should be noted, however, that at rest, the work of breathing is very small even at altitude. Cibella et al. (67) measured Wrs during submaximal exercise on a bicycle ergometer in four lowlanders at sea level and after a 1 month sojourn at 5050 m. · In only one subject was Wrs, for a given Ve, consistently lower at altitude than sea level, as expected from decreased air density (62). In contrast, in the other three · individuals, the relationship of Wrs to Ve was the same at altitudes as at sea level. The latter was attributed mainly to bronchoconstriction due to severe hypoxia and hypocapnia. In addition, interstitial pulmonary edema is likely to develop during exercise at very high altitude (68). At 3660 m, Lefranc¸ois et al. (24) measured the respiratory mechanical power at rest and exercise up to 120 watts in five highlanders and three lowlanders after a 30-day sojourn at that altitude. For any given ventilation, the mechanical power was similar in both groups. However, at any given exercise level, the mechanical power requirements were smaller in the highlanders reflecting their lower ventilation. Thus, the lower ventilatory requirements during exercise of highlanders put them at an advantage relative to lowlanders. In this respect, the highlanders with large lungs should also be at an advantage because of smaller respiratory mechanical power requirement for any given ventilation. Unfortunately, Lefranc¸ois et al. (24) did not measure the respiratory mechanical output during maximal exercise. In this connection, the resistive work of breathing depends on the flow waveform, being minimal with constant flow inflation and deflation (69). During heavy exercise at sea level both inspiratory and expiratory flows approach constant flow (70,71). Whether such an optimal pattern is also adopted during exercise at altitude is not known. At sea level, the work per breath during maximal exercise amounts to about 20% of the maximal potential work (69). No such information is available for altitude. However, this percentage should be considerably greater at altitude in view of the decrease in maximal potential work.

194

Milic-Emili et al.

There is only one report on the oxygen cost of breathing at altitude (44). The oxygen cost of breathing was measured during voluntary hyperventilation. In 11 lowlanders after 16 days at 4000 m, the cost of breathing per liter of ventilation was 36% lower than at sea level, probably reflecting in part lower respiratory power due to decreased density (Fig. 9). However, the oxygen cost of breathing during voluntary hyperventilation is usually substantially higher than during spontaneous breathing (69). At sea level, exercise performance is usually not limited by ventilation in normal subjects. In contrast, at high altitudes, the respiratory requirement probably plays a more important role in limiting exercise performance. In spite of this, during maximal exercise at a simulated altitude of 8848 m, the dyspnea ratings were similar and not, as expected, higher than those for leg activity (72). Thus, even at that very high altitude, exercise limitation was attributed to both peripheral and respiratory muscle activity. This surprising finding could be explained by the hypothesis that during exercise at very high altitudes there is preferential blood supply to the respiratory muscles, which should benefit the performance of the respiratory muscles but at the expense of the peripheral muscles. Clearly further studies on mechanical power, energy cost of breathing, blood supply to the respiratory muscles, etc., are needed, particularly at the very high ventilations encountered during exercise at high altitude. In conclusion, while at sea level there is a considerable body of knowledge of respiratory mechanics, as detailed in a recent review (69), for altitude, there is substantial information available only with respect to lung volumes. In contrast, the studies of respiratory mechanics at high altitude are rather scanty. This is particularly true for the work of breathing, which must play an important role in limiting exercise performance at high altitudes.

Acknowledgments The authors thank Ms. M. Gras for typing this manuscript and Ms. J. Chandellier for artwork. They also wish to thank Drs. W. A. Whitelaw and J. P. Mortola for their helpful comments and suggestions.

References 1. Reeves JT, Welsh CH, Wagner PD. The heart and lungs at extreme altitude. Thorax 1994; 49:631–633. 2. Sutton JR, Reeves JT, Wagner PD, Groves BM, Cymerman A, Malconian MK, Rock PB, Young PM, Walter SD, Houston CS. Operation Everest II: oxygen transport during exercise at extreme simulated altitude. J Appl Physiol 1988; 64:1309–1321. 3. Messner R. Everest: Expedition to the Ultimate. London: Kaye and Ward, 1979. 4. Bert P. La Pression Barome´trique. Paris: Masson, 1878.

Mechanics of Breathing

195

5. Schneider EC. The vital capacity of the lungs at low barometric pressure. Am J Physiol 1932; 100:426–432. 6. Hurtado A, Kaltreider N, McCann WS. Respiratory adaptation to anoxemia. Am J Physiol 1934; 109:626–637. 7. Rahn H, Hammond D. Vital capacity at reduced barometric pressure. J Appl Physiol 1952; 4:715–724. 8. Finkelstein S, Tomashefski JF, Shillito FH. Pulmonary mechanics at altitude in normal and obstructive lung disease patients. Aerospace Med 1965; 36:880–884. 9. Gray GW, Rennie IDB, Houston CS, Bryan AC. Phase IV of the single-breath nitrogen washout curve on exposure to altitude. J Appl Physiol 1973; 35:227–230. 10. Coates G, Gray G, Mansell A, Nahmias C, Powles A, Sutton J, Webber C. Changes in lung volume, lung density, and distribution of ventilation during hypobaric decompression. J Appl Physiol 1979; 46:752–755. 11. Saunders NA, Betts MF, Pengelly LD, Rebuck AS. Changes in lung mechanisms induced by acute isocapnic hypoxia. J Appl Physiol 1977; 42:413–419. 12. Goldstein RS, Zamel N, Rebuck AS. Absence of effects of hypoxia on small airway function in humans. J Appl Physiol 1979; 47:251–256. 13. Garfinkel F, Fitzgerald RS. The effect of hyperoxia, hypoxia and hypercapnia on FRC and occlusion pressure in human subjects. Respir Physiol 1978; 33:241–250. 14. Welsh CH, Wagner PD, Reeves JT, Lynch D, Cink TM, Armstrong J, Malconian MK, Rock PB, Houston CS. Operation Everest II: Spirometric and radiographic changes in acclimatized humans at simulated high altitudes. Am Rev Respir Dis 1993; 147:1239– 1244. 15. Ulvedal F, Morgan TE Jr, Cutler RG, Welch BE. Ventilatory capacity during prolonged exposure to simulated altitude without hypoxia. J Appl Physiol 1963; 18:904– 908. 16. Mosso A. Fisiologia dell’uomo sulle Alpi. Milan: Treves, 1897. 17. Cruz JC. Mechanics of breathing in high altitude and sea level subjects. Respir Physiol 1973; 17:146–161. 18. Jaeger JJ, Sylvester JT, Cymerman A, Berberich JJ, Denniston JC, Maher JT. Evidence for increased intrathoracic fluid volume in man at high altitude. J Appl Physiol 1979; 47:670–676. 19. Shields JL, Hannon JP, Harris CW, Platner WS. Effects of altitude acclimatization on pulmonary function in women. J Appl Physiol 1968; 25:606–609. 20. Tenney SM, Rahn H, Stroud RC, Mithoefer JC. Adaptation to high altitude: changes in lung volumes during the first seven days at Mt. Evans, Colorado. J Appl Physiol 1953; 5:607–613. 21. Mansell A, Powles A, Sutton J. Changes in pulmonary PV characteristics of human subjects at an altitude of 5,366 m. J Appl Physiol 1980; 49:79–83. 22. Cerny FC, Dempsey JA, Reddan WG. Pulmonary gas exchange in non-native residents of high altitude. J Clin Invest 1973; 52:2993–2999. 23. Degraff AC Jr, Grover RF, Johnson RL Jr, Hammond JW Jr, Miller JM. Diffusing capacity of the lung in Caucasians native to 3,100 m. J Appl Physiol 1970; 29:71–76. 24. Lefranc¸ois R, Gautier H, Pasquis P. Me´canique ventilatoire chez l’homme a` haute altitude. C R Soc Biol 1969; 163:2037–2042. 25. Gautier H, Peslin R, Grassino A, Milic-Emili J, Hannhart B, Powell E, Miserocchi G, Bonora M, Fischer JT. Mechanical properties of the lungs during acclimatization to altitude. J Appl Physiol 1982; 82:1407–1415.

196

Milic-Emili et al.

26. Roy SB, Guleria JS, Khanna PK, Talwar JR, Manchanda SC, Pande JN, Kaushik VS, Subba PS, Wood JE. Immediate circulatory response to high altitude hypoxia in man. Nature 1968; 217:1177–1178. 27. Hurtado A. Animals in high altitudes: resident man. In: Dill DB, Adolph EF, Wilber CG, eds. Handbook of Physiology. Adaptation to Environment. Washington, DC: American Physiology Society, 1964:843–860. 28. Lahiri S, Delaney RG, Brody JS, Simpser M, Velasquez T, Motoyama EK, Polgar C. Relative role of environmental and genetic factors in respiratory adaptation to high altitude. Nature 1976; 261:133–135. 29. Brody JS, Lahiri S, Simpser M, Motoyama EK, Velasquez T. Lung elasticity and airway dynamics in Peruvian natives to high altitude. J Appl Physiol 1977; 42:245–251. 30. Hackett PH, Reeves JT, Reeves CD, Grover RF, Rennie D. Control of breathing in Sherpas at low and high altitude. J Appl Physiol 1980; 49:374–379. 31. Kamat SR, Rao TL, Sarma BS, Venkataraman C, Raju VRK. Study of cardiopulmonary function on exposure to high altitude. Am Rev Respir Dis 1972; 106:414–431. 32. Sun SF, Droma TS, Zhang JG, Tao JX, Huang SY, McCullough RG, McCullough RE, Reeves CS, Reeves JT, Moore LG. Greater maximal O2 uptakes and vital capacities in Tibetan than Han residents of Lhasa. Respir Physiol 1990; 79:151–162. 33. Frisancho AR. Human growth and pulmonary function of a high altitude Peruvian Quechua population. Human Biol 1969; 41:365–379. 34. Greksa LP, Spielvogel H, Paz-Zamora M, Caceres E, Paredes-Fernandez L. Effect of altitude on the lung function of high altitude residents of European ancestry. Amer J Phys Anthropol 1988; 75:77–85. 35. Frisancho AR. Functional adaptation to high altitude hypoxia. Science 1975; 187:313– 319. 36. Burri PH, Weibel ER. Morphometric evaluation of changes in lung structure due to high altitude. In: Porter R, Knight J, eds. High Altitude Physiology. Edinburgh: Churchill Livingstone, 1971:15–30. 37. Sekhon HS, Wright JL, Thurlbeck WM. Pulmonary function alterations after 3 wk of exposure to hypobaria and/or hypoxia in growing rats. J Appl Physiol 1995; 78:1787– 1792. 38. Paz Zamora M, Coudert J, Ergueta Collao J, Vargas E, Gutierrez N. Respiratory and cardiocirculatory responses of acclimatization of high altitude natives (La Paz, 3500 m) to tropical lowland (Santa Cruz, 420 m). In: Brendel W, Zink RA, eds. High Altitude Physiology and Medicine. New York: Springer-Verlag, 1982:21–27. 39. Jones RL, Man SFP, Matheson GO, Parkhouse WS, Allen PS, McKenzie DC, Hochachka PW. Overall and regional lung function in Andean natives after descent to low altitude. Respir Physiol 1992; 87:11–24. 40. Kryger M, Alrich F, Reeves JT, Grover RF. Diagnosis of airflow obstruction at high altitude. Amer Rev Respir Dis 1978; 117:1055–1058. 41. Rahn H, Otis AB, Chadwick LE, Fenn WO. The pressure-volume diagram of the thorax and lung. Amer J Physiol 1946; 146:161–178. 42. Cotes JE. Ventilatory capacity at altitude and its relation to mask design. Proc R Soc B 1954; 143:32–39. 43. Pugh LGCE. Muscular exercise on Mount Everest. J Physiol Lond 1958; 141:233–261. 44. Mazess RB. The oxygen cost of breathing in man: effects of altitude, training, and race. Am J Phys Anthropol 1968; 29:365–375.

Mechanics of Breathing

197

45. Ingram RH Jr, Schilder DP. Effect of gas compression on pulmonary pressure, flow, and volume relationship. J Appl Physiol 1966; 21:1821–1826. 46. Wood LDH, Bryan AC. Effect of increased ambient pressure on flow-volume curve of the lung. J Appl Physiol 1969; 27:4–8. 47. Hyatt RE. Dynamic lung volumes. In: Fenn WO, Rahn H, eds. Handbook of Physiology. Respiration. Washington, DC: American Physiology Society, 1965:1381–1397. 48. Agostoni E, Fenn WO. Velocity of muscle shortening as a limiting factor in respiratory air flow. J Appl Physiol 1960; 15:349–353. 49. Kronenberg RS, Safar P, Lee J, Wright F, Noble W, Wahrenbrock E, Hickey R, Nemoto E, Severinghaus JW. Pulmonary artery pressure and alveolar gas exchange in man during acclimatization to 12,470 ft. J Clin Invest 1971; 50:827–837. 50. Raymond L, Severinghaus JW. Static pulmonary compliance of man during altitude hypoxia. J Appl Physiol 1971; 31:785–787. 51. Frank NR, Radford EP, Whittenberger JL. Static volume-pressure interrelations of the lungs and pulmonary blood vessels in excised cat lungs. J Appl Physiol 1959; 14:167– 173. 52. Mortola JP, Rezzonico R, Fisher JT, Villena-Cabrera N, Vargas E, Gonzales R, Pena F. Compliance of the respiratory system in infants born at high altitude. Am Rev Respir Dis 1990; 142:43–48. 53. Vare`ne P, Timbal J, Jacquemin C. Effect of different ambient pressures on airway resistance. J Appl Physiol 1967; 22:699–706. 54. Kleiner JP, Nelson WP. High altitude pulmonary edema. J Am Med Assoc 1975; 234: 491–495. 55. Reeves JT, Halpin J, Cohn JE, Daoud F. Increased alveolar-arterial oxygen difference during simulated high-altitude exposure. J Appl Physiol 1969; 27:658–661. 56. De Troyer A, Yernault JC, Rodenstein D. Influence of beta-2 agonist aerosols on pressure-volume characteristics of the lungs. Am Rev Respir Dis 1978; 118:987–995. 57. Vincent NJ, Knudson R, Leith DE, Macklem PT, Mead J. Factors influencing pulmonary resistance. J Appl Physiol 1970; 29:236–243. 58. Wolfel EE, Selland MA, Mazzeo RS, Reeves JT. Systemic hypertension at 4,300 m is related to sympatho adrenal activity. J Appl Physiol 1994; 76:1643–1650. 59. Libby PM, Briscoe WA, King TKC. Relief of hypoxia-related bronchoconstriction by breathing 30 per cent oxygen. Am Rev Respir Dis 1981; 123:171–175. 60. Newhouse MT, Becklake MR, Macklem PT, McGregor M. Effect of alterations in endtidal CO2 tension on flow resistance. J Appl Physiol 1964; 19:745–749. 61. Jaeger MJ, Otis AB. Effects of compressibility of alveolar gas on dynamics and work of breathing. J Appl Physiol 1964; 19:83–91. 62. Petit JM, Milic-Emili G, Troquet J. Travail dynamique pulmonaire et altitude. Rev Med Aeronaut 1963; 2:276–279. 63. Milic-Emili G, Petit JM, Deroanne R. Mechanical work of breathing during exercise in trained and untrained subjects. J Appl Physiol 1962; 17:43–46. 64. Goldman MD, Grimby G, Mead J. Mechanical work of breathing derived from rib cage and abdominal V-P partitioning. J Appl Physiol 1976; 41:752–763. 65. Hesser CM, Linnarsson D, Fragraeus L. Pulmonary mechanics and work of breathing at maximal ventilation and raised air pressure. J Appl Physiol 1981; 50:747–753. 66. Van Liew HD. Mechanical and physical factors in lung function during work in dense environments. Undersea Biomed Res 1983; 10:255–264.

198

Milic-Emili et al.

67. Cibella F, Cuttitta G, Romano S, Grassi B, Bonsignore G, Milic Emili J. Respiratory energetics during exercise at high altitude. J Appl Physiol 1999; 86:1785–1792. 68. West JB. Left ventricular filling pressure during exercise. Chest 1998; 113:1695–1697. 69. Milic-Emili J. Work of breathing. In: Crystal RG, West JB, eds. The Lung. New York: Raven Press, 1991:1065–1075. 70. Proctor DF, Hardy JB. Studies of respiratory airflow. Bull Johns Hopkins Hosp 1949; 85:253–280. 71. Lafortuna CL, Minetti AE, Mognoni P. Modelling the airflow pattern in humans. J Appl Physiol 1984; 57:1111–1119. 72. Sutton J, Balcomb A, Killian K, Green HJ, Young PM, Cymerman A, Reeves JT, Houston CS. Breathlessness at altitude. In: Jones NL, Killian KL, eds. Breathlessness. The Campbell Symposium. Hamilton (ON): Boehringer Ingelheim Canada, 1992:143–148.

8 Gas Exchange

PETER D. WAGNER University of California, San Diego La Jolla, California

The essence of high altitude is hypoxia, reducing arterial Po 2. This chapter deals mainly with how pulmonary gas exchange is modified by altitude. Both rest and exercise are considered, and O2 transport is analyzed in terms of (1) effects of reduced inspired Po 2, (2) diffusion limitation of O2 exchange, and (3) ventilation/ perfusion relationships. Acute altitude exposure and prolonged, acclimatized situations are described separately. Hypoxia is of course the result of barometric pressure (Pb) decreasing with altitude. While air at any altitude remains approximately 21% O2, the virtually exponential fall in Pb with altitude reduces ambient Po 2. When it is further remembered that as soon as air is inhaled into the large airways it becomes warmed to body temperature (37°C usually) and saturated with water vapor [contributing a partial pressure of 47 mmHg (at 37°C) at all altitudes], it is clear that inspired Po 2 in the airways (Pio 2) is rapidly reduced by ascent to altitude. Table 1 indicates pertinent values derived from the U.S. Standard atmosphere relating Po 2 to altitude. The U.S. Standard atmosphere is only a general guide to effects of altitude on inspired Po 2. Geographical, thermal, seasonal, and meteorological factors will cause variation from the figures in Table 1. These variations may be relatively large and biologically substantial. Perhaps the best example is that the U.S. atmosphere 199

200

Wagner

Table 1 Inspired Po2 at Altitude Altitude (m) 0 1000 2000 3000 4000 5000 6000 7000 8000 9000

Altitude (ft.)

Pb (torr)

Pio 2 (saturated, 37°C) (torr)

0 3281 6562 9843 13123 16404 19685 22966 26247 29528

760 674 596 526 463 405 354 308 267 231

149 131 115 100 87 75 64 55 46 39

predicts Pb on the Everest summit to be 236 torr while data obtained during the American Medical Research Expedition to Everest in 1981 (1) indicated a Pb of 253 torr, 17 mm higher. While at sea level, a fall in Pb of 17 torr from 760 to 743 will cause Pio 2 (at 37°C, saturated) to fall by 3.5 torr (0.2093 ⫻ 17), and thus arterial Po 2 to fall similarly in a normal subject, the arterial saturation will fall by about 0.2%, a trivial amount. The 17 torr difference in Pio 2 on the Everest summit translates into the same 3.5 torr drop in Pio 2, but with arterial Po 2 normally at about 30– 35 torr, the fall in arterial saturation would be about 6%. This is not considered trivial under conditions of extreme hypoxia. More discussion of barometric pressure and its effects on inspired Po 2 can be found in Ref. 2. Many excursions to altitude over the years have examined gas exchange, usually focusing on noninvasive measurements (3–5). This has necessarily limited the amount of data collected, especially at extreme altitudes. Two studies in particular (using healthy subjects), however, have resulted in large and complete data sets. One focused on acute exposure to Pb of 429 torr (⬃15,000 ft) on a single day, while the other, called Operation Everest II (OEII) (7), studied subjects gradually decompressed to Pb ⬇ 250 torr over 42 days. Both of these hypobaric chamber studies sampled arterial and pulmonary arterial blood and used the multiple inert gas elimination technique to characterize pulmonary gas exchange. For these reasons, and to promote internal consistency and to present the most complete picture possible, the chapter will be based mostly on the data from these two projects. The philosophy of this chapter is more to describe the physiological responses to altitude exposure from these studies than to present an exhaustive literature survey of necessarily sparse data. In general, the limited data from many other studies are in agreement with those of the two projects featured. With this brief introduction, the effects of altitude on pulmonary gas exchange are now discussed.

Gas Exchange I.

201

Gas Exchange at Rest

A. Consequences of the Fall in Inspired PO 2 with Altitude on Alveolar PO 2

As inspired Po 2 falls, so too must alveolar Po 2 (Pao 2 ). Hypoxemia ensues such that there is ventilatory stimulation via the carotid chemoreceptors, and this helps to mitigate the obligatory fall in alveolar Po 2 that follows a reduction in Pio 2. Before considering the effects of hyperventilation, it is instructive to calculate the hypothetical effects of altitude on alveolar Po 2 in the absence of any ventilatory stimulation. Such a calculation serves to point out the criticality of the ventilatory response to hypoxia. This can be done using the alveolar gas equation (8): Pao 2 ⫽ Pio 2 ⫺ Paco 2 /R ⫹ Paco 2 ⋅ Fio 2 ⋅ (1-R)/R

(1)

In Eq. (1), Paco 2 is alveolar Pco 2 , R is respiratory exchange ratio, and Fio 2 inspired O2 fractional concentration. Typically at rest, alveolar (or arterial, essentially the same) Pco 2 is 40 torr at sea level and R is 0.8, while Fio 2 is 0.2093 independent of altitude. Figure 1 shows how alveolar Po 2 falls essentially linearly with altitude, using Eq. (1) and data from the two above-mentioned studies (6,7). In Figure 1, the closed circles reflect the calculated values of Pao 2 assuming alveolar Pco 2 remained con-

Figure 1 Fall in alveolar Po 2 with increasing altitude. Open circles reflect actual data, and closed circles reflect hypothetical values had progressive compensatory hypocapnia not developed with altitude. The break at Pb ⫽ 430 torr separates different data sets from acute (6) and prolonged (7) hypoxic exposures. Hyperventilation is of increasing adaptive importance the higher the altitude.

202

Wagner

stant at sea level values, and the open circles are actual data from the studies. The highest barometric pressure of 760 torr is that of sea level; the lowest of 253 torr corresponds to the Everest summit. Thus, the criticality of the ventilatory response is seen when Pb falls below about 400 torr, and increasingly so the higher the altitude. Up to common skiing altitudes (about 10,000 ft., Pb about 500 torr or greater) there would be little reduction in Pao 2 from absence of the hypoxic ventilatory response. Due to the shape of the O2-Hb dissociation curve (essentially flat at Po 2 values ⬎60 torr), there is also negligible effect on arterial O2 saturation. However, at 4000 m (Pb 463 torr), failure to hyperventilate would cause an almost 10 torr fall in Pao 2. At about 350 torr, Pao 2 would be some 15 torr lower, at less than 25 torr. At the Everest summit equivalent, alveolar Po 2 is still maintained at or above 30 torr (Fig. 1) whereas in the absence of hyperventilation it would have plummeted to 5 torr. There are many accounts of O2-unaided ascents of Mt. Everest, which must clearly be dependent in large measure on the ability to maintain alveolar and thus arterial Po 2 by hyperventilation. Several studies conducted during mountaineering expeditions measuring alveolar Po 2 and Pco 2 are summarized in Figure 2. Interest is focused on extreme altitude (Pb ⬍ 350 torr) in this figure. There is a progressive, linear fall in both Pao 2 and Paco 2 as altitude is gained. Resting Pao 2 on the Everest summit or equivalent appears to be between 30 and 35 torr; corresponding Paco 2 is between 8 and 11 torr. There is an apparent discrepancy between the data of AMREE (The American Medical Research Expedition to Everest) and those of OEII (closed and open circles, respectively, Fig. 2). The higher Pao 2 and lower Paco 2 of AMREE have been interpreted as reflecting greater ventilatory acclimatization to altitude (2). While the time to reach the summit was only 40 days in OEII (9) and more than 70 in AMREE (10), other explanations of the OEII-AMREE differences are possible. Thus, at about 282 torr, where OEII examined six subjects (11) and AMREE four (12), the data are bunched so closely that there is very little difference in Pao 2 or Paco 2 between them (Fig. 2). This is mirrored by recent data from the British Everest Expedition (BEE) (13). The obvious OEII-AMREE difference is thus seen at the summit altitude. Here, the AMREE data reflect a single subject who, even at lower altitudes, had a generally lower Paco 2 and higher hypoxic ventilatory response than most of the AMREE subjects studied (12). The respiratory gas exchange ratio on the summit for this person was 1.49, probably reflecting considerable acute hyperventilation. Further data will be necessary to determine the reason for these differences, but there is similarity in alveolar Po 2 and Pco 2 average AMREE, BEE, and OEII subjects at the one common altitude (of Pb ⬃ 282 torr) where multiple subjects were available for study. Figure 3 shows the mean arterial Po 2 and Pco 2 in resting subjects from the two studies mentioned [reflecting acute (6) and prolonged (7) altitude exposure]. These data agree with those of others estimated indirectly or measured directly (14– 16). The top panel shows the effects of the remarkable fourfold increase in resting

Gas Exchange

203

Figure 2 Collection of data from several sources depicting alveolar Po 2 and Pco 2 as a function of altitude. Given the small numbers of subjects, intersubject variation, and experimental error, there is remarkable agreement at Pb ⫽ 275–300 torr among AMREE, BEE, and OEII. Differences appear at the summit, but the environmental and technical differences and small numbers of subjects preclude clear interpretation.

204

Wagner

Figure 3 Resting arterial Pco 2 (top) and Po 2 (bottom) for acute hypoxic exposures to Pb ⫽ 429 torr (about 15,000 ft.) and prolonged exposure up to the Everest summit (Pb ⫽ 253 torr). These values mirror those in alveolar gas of Figures 1 and 2 and also reflect a lower alveolar-arterial Po 2 difference with prolonged exposure.

ventilation from sea level to the Everest summit, which correspondingly causes a fourfold reduction in alveolar and thus arterial Pco 2 in these normal subjects. This one-for-one relationship between ventilation and Paco 2 indicates no change in resting metabolic rate with altitude. In the field, additional stresses of anxiety and cold may well impose changes in metabolic rate that would additionally influence the ventilation actually measured at altitude. Figure 3 also shows a discontinuity between the overlapping data from the two studies at the 430 torr Pb point: arterial Pco 2 was lower and Po 2 higher in

Gas Exchange

205

chronic exposure at the same altitude. While this could reflect study population differences, sea level data for the two groups were not different, which suggests that this is not the explanation. The differences at Pb 430 torr more likely reflect the well-known further increase in ventilation (at any altitude) comparing prolonged to acute exposure. These differences are generally thought to be due to removing the braking effect on ventilation of alkalosis in blood and CSF, which are present to a greater extent in acute than chronic exposure and to greater hypoxic sensitivity of the carotid chemoreceptors (see Chapter 6 for further details). B. Pulmonary Gas Exchange Efficiency

There is a small additional difference between acute and chronic altitude exposure that affects arterial Po 2 and Pco 2 and is not explained by differences in ventilation per se. If the alveolar-arterial Po 2 difference (AaPo 2) is computed by subtracting the measured arterial Po 2 from the calculated alveolar Po 2 [Eq. (1)], one sees a small but systematic difference: the AaPo 2 is higher in acute than prolonged altitude exposure (Fig. 4). Reasons for this are discussed later in this chapter, but the smaller AaPo 2 with chronic altitude exposure helps to preserve arterial Po 2 for any given level of ventilation. The efficiency of pulmonary gas exchange is therefore changed, even at rest, with acute or chronic exposure to altitude. Several factors could contribute to this. Changes in air density could affect intrapulmonary gas distribution; increases in ventilation, by changing gas flow rates, could also alter inspired gas distribution.

Figure 4 Resting alveolar-arterial Po 2 difference as a function of altitude (6,7). From normal sea level values of ⬍10 torr, the AaPo 2 progressively falls with altitude to values close to zero. There appears to be a reduction in AaPo 2 with chronic exposure compared to acute exposure at the same altitude. Absolute values of AaPo 2 are generally quite small.

206

Wagner

The sympathetically mediated tachycardia and increase in cardiac output of acute altitude exposure could lead to blood flow distribution changes, and the development of hypoxic pulmonary vasoconstriction could also redistribute blood flow within the lungs, especially if vasoconstriction occurs unevenly as has been suggested (17). Because alveolar Po 2 is reduced with increasing altitude, vulnerability to alveolarcapillary diffusion limitation increases (18). Finally, to the extent that high-altitude pulmonary edema (HAPE) develops, even subclinically, ventilation and/or blood flow distribution could change and shunts might develop. Assessing the combined effects of all of these potential phenomena is com˙ ) in˙ a/Q plex. In particular, for a given (constant) level of ventilation/perfusion (V equality, the AaPo 2 falls with increasing altitude as alveolar Po 2 falls onto the steeper part of the O2-Hb dissociation curve. This is shown in Figure 5 using a computer ˙ mismatching. Log SDQ, an abbreviation ˙ a/Q model having a constant level of V ˙ distribution about its mean, on a log scale, is ˙ a/Q for the second moment of the V ˙ ratios within ˙ a/Q used as an index of the amount of dispersion or heterogeneity of V the lung. Three examples are illustrated: a normal amount of inequality (log SDQ ⫽ 0.4) and two moderately increased values of log SDQ (0.7 and 1.0). There are two contributing factors to this reduction in the AaPo 2. First, as Po 2 falls onto the steep portion of the O2-Hb dissociation curve, the range of Po 2 values from low to ˙ ratios is reduced and this will mathematically bring the alveolar and ˙ a/Q high V arterial Po 2 values closer together. However, as West showed many years ago (19), ˙ ratio of the lung ˙ a/Q a second factor reducing AaPo 2 is the increase in overall V due to hyperventilation. This is also a consequence of the shape and slope of the O2-Hb dissociation curve. Consequently, the fall in AaPo 2 (at rest) with altitude cannot be assumed to ˙ relationships. Figure 4 is a good example of this ˙ a/Q reflect an improvement in V ˙ relationships, ˙ a/Q dilemma: Does the progressive fall in AaPo 2 reflect improving V or is this only a reflection of what is shown in Figure 5? The latter explanation is explored below. Gas exchange at altitude is further affected by behavior of the AaPo 2 when caused by alveolar-capillary diffusion limitation (18). Figure 5 shows how the AaPo 2 due to diffusion limitation would increase with ascent to altitude even if the pulmonary O2 diffusing capacity (Dl O2 ) were to remain constant. These calculations use several values for Dl O2 , each held constant with altitude. None is so low as to produce any AaPo 2 at rest at sea level as the figure shows. Examination of Figure 4 and other data (15) in light of the predictions modeled in Figure 5 suggests that because the AaPo 2 in fact falls with altitude, diffusion limitation at rest does not occur. Indeed, the fall in measured AaPo 2 appears consis˙ mismatch. Figure 6 replots the data ˙ a/Q tent with a constant and normal level of V of Figure 4 with the calculations of Figure 5 superimposed (as a dashed line for the case of log SDQ ⫽ 0.4, a normal value), and the agreement between the actual data and those computed for log SDQ ⫽ 0.4 is generally good. Thus, it is reasonable to postulate that in young normal subjects at altitude there is on average a pattern of gas exchange that simply reflects predictable mathematical effects of reduction in

Gas Exchange

207

Figure 5 Calculated values for the AaPo 2 as a function of altitude in a lung model with a fixed degree of ventilation/perfusion inequality indicated by the dispersion parameter log ˙ inequality falls progressively with ˙ a/Q SDQ (upper panel). The AaPo 2 for any value of V altitude, with values converging as altitude is gained. The lower panel shows predicted AaPo 2 at different altitudes in five lung models, each having a fixed value of oxygen diffusing capacity (Dlo 2 ). Under these resting conditions, AaPo 2 is zero at sea level, but rises progressively with altitude according to the value of Dlo 2, in sharp contrast to the opposite behavior of ˙ inequality. ˙ a/Q the effects of V

208

Wagner

Figure 6 Measured resting alveolar-arterial Po 2 difference in subjects acutely and chronically exposed to hypoxia from sea level to the Everest summit (from Fig. 4). Also plotted ˙ mismatch ˙ a/Q from Figure 5 is the expected fall in AaPo 2 for a constant, normal degree of V (log SDQ ⫽ 0.4), dashed line. Note the scale of the ordinate spanning only 10 torr. No point is more than 2–3 torr from the predicted line. Thus, the measured values are broadly consistent with the expected behavior of a normal lung without diffusion limitation.

˙ relationships or development of diffusion limitation. ˙ a/Q Pio 2 without change in V However, caution is needed in arriving at this conclusion because the AaPo2 is inherently noisy and the curves of Figure 5 tend to converge at low Pb, reducing the ˙ mismatch from the behavior of AaPo 2. ˙ a/Q ability to infer changes in V To better understand how altitude does (or does not) affect resting pulmonary gas exchange, methodology is required that is not only independent of inspired Po 2, ˙ mismatch and diffusion limitation. The multi˙ a/Q but that can distinguish between V

Figure 7 Ventilation/perfusion inequality at rest as a function of altitude in acute (䊊) (6) and prolonged (䊉) (7) chamber exposures. The top panel shows mean data set against the 95% upper confidence limit for log SDQ of 0.6. The two points marked with an asterisk (*) are in the abnormal range, and as described in the text were observed shortly after periods of rapid decompression. The middle panel shows individual subject data from the same studies, showing variability in individual responses. The bottom panel also shows individual data from the chronically exposed subjects with different symbols according to initial sea level resting ventilation/perfusion mismatch. See text for more details.

Gas Exchange

209

210

Wagner

ple inert gas elimination method (20–22) offers both of these possibilities and has ˙ mismatch ˙ a/Q been used in chamber simulations of altitude. Figure 7 shows how V at rest changes with altitude, based on the studies cited earlier (6,7). Figure 7 is complex. Consider first the data for acute altitude exposure (open circles of top and middle panels). The top panel shows mean data; the middle panel ˙ mismatch (i.e., log SDQ) ˙ a/Q all individual data. In both panels, the amount of V that represents, at sea level, the mean plus 2 SD (i.e., approximate 95% upper confidence limit of normal) is shown by the horizontal dashed line and is 0.6. All data for acute exposure (Fig. 7, open circles) lie within the 95% confidence limits. This is true at sea level and at both altitudes (corresponding to about 10,000 and 15,000 ft. above sea level). Acute exposure in this study was ascent to the altitude in question over several minutes in a hypobaric chamber. Altitude over this ˙ inequality. This is in spite of known alterations ˙ a/Q range had no net effect on V in ventilatory and circulatory components that acutely accompany hypoxia. In particular, any hypoxia-induced pulmonary hypertension that might more uniformly distribute blood flow up and down the lungs is too subtle to reduce the very small ˙ mismatch present in the normal lungs at sea level (i.e., log SDQ ˙ a/Q amount of V of 0.3–0.4 as the open circles of Fig. 7 show for most subjects). Also, the very small rise in pulmonary artery pressure at rest in hypoxia in the group of subjects in Figure 7 would not likely cause even interstitial pulmonary edema. Sustained exposure to hypoxia for 2 weeks at an altitude of about 3800 m (at the University of California White Mountain Barcroft Laboratory) revealed no ˙ changes—again, minimal amounts of V ˙ mismatch were ˙ a/Q ˙ a/Q evolution of V noted both before and at the end of the altitude exposure (23), a situation discussed in more detail later in this chapter. This appears not to be the case when decompression to Pb levels lower than those of normal human habitation is examined. Operation Everest II (OEII) addressed this situation (Fig. 7, all three panels, closed circles). The data up to 15,000 ft. (Pb ⫽ 430 torr) are essentially compatible with those of the acute study (open ˙ mismatching at sea ˙ a/Q circles). Most subjects were indistinguishable in their V level and 15,000 ft. as the figure shows. Of potential importance, the subjects who were less altitude-tolerant, i.e., who failed to complete OEII, were those subjects ˙ mismatch at sea level (open squares, Fig. ˙ a/Q with the greatest degree of resting V 7, lower panel). Omitting these subjects makes the two subject groups indistinguish˙ relationships from sea level to 15,000 ft. Statistically, even considering ˙ a/Q able in V all subjects of both groups, they were not different at the altitudes available for direct comparison (Pb ⫽ 760 and 430 torr). On the other hand, at Pb ⫽ 347 (⬃20,000 ˙ mis˙ a/Q ft.) and Pb ⫽ 253 torr (⬃Everest summit) there was clear evidence of V match of moderately severe proportions (Fig. 7). There was great variability among ˙ relationships ˙ a/Q subjects, as the data show. Yet at Pb ⫽ 282 torr, about 25,000 ft., V were again within normal limits. How can these inconsistent results be explained? One possibility is the technical irreproducibility of results, but extensive experience has shown that the coefficient of variation of the log SDQ parameter is only about 6–7% when duplicate

Gas Exchange

211

data are averaged (24). The top panel of Figure 7 shows a factor of 2 separating log SDQ mean values at Pb ⫽ 282 torr from those at both 347 and 253 torr. This variability is therefore not explained by technical factors. A common thread connects the data of Pb ⫽ 347 and 253 torr: Rapid decompression immediately proceeding studies at each of these two altitudes (25). On the other hand, data at Pb ⫽ 282 torr were collected only after subjects had been at that pressure for several days. Combined with independent clinical assessment of ˙ mismatch was greater and shunt was ˙ a/Q rales in more than one subject (when V present) at Pb ⫽ 347 torr (25), a more tenable hypothesis for the abnormal degrees ˙ mismatch at Pb ⫽ 347 and 253 torr is acute high-altitude pulmonary edema ˙ a/Q of V (HAPE). Individual log SDQ values of 1.5–2.0 as noted in Figure 7 at these altitudes are as high as seen in the intensive care setting in patients with respiratory distress syndromes, requiring assisted ventilation (26). An intriguing hypothesis is suggested by the examination of individual resting ˙ mismatch ˙ a/Q data (Fig. 7, lower panel). Those subjects with the most sea level V (open squares) were unable to ‘‘summit,’’ becoming intolerant to altitude. Those subjects with the least sea level mismatch (open triangles) had the smoothest clinical ˙ mismatch, and were ˙ a/Q experience throughout OEII, never developed abnormal V ˙ ˙ a/Q able to reach the summit. Those with intermediate initial sea level values of V mismatch (closed circles) summited successfully, but developed occasionally sub˙ mismatch during periods of rapid ascent. ˙ a/Q stantial V While the small number of subjects requires care not to overinterpret the find˙ relation˙ a/Q ings, the OEII results do suggest a relationship between normoxic V ships at sea level and subsequent development of pulmonary gas exchange problems at altitude. Furthermore, only two of the eight OEII subjects escaped without some ˙ mismatch developing, and this is consistent with the idea that altitude-related ˙ a/Q V gas exchange problems, presumably on the basis of some degree of HAPE, may be more prevalent at extreme altitude than current estimates of HAPE incidence would imply. These possibilities require testing with larger numbers of subjects.

II. Gas Exchange During Exercise A. Overview

Physical activity is so much a part of ascent to altitude that there continues to be great interest in pulmonary gas exchange during exercise at altitude, both in terms of overall ventilatory and cardiovascular responses and with respect to pulmonary exchange efficiency, which is addressed in this section. Overall ventilatory and cardiovascular responses are covered in detail in other chapters of this volume. In acute altitude exposure, both total ventilation and cardiac output increase at a greater rate (with exercise intensity) than at sea level over the same work range (2,27). However, maximal values of both ventilation and cardiac output are similar to sea level values (Fig. 8), reached of course at lower workloads than at sea level. The mechanisms of these heightened responses are discussed in Chapters 6 and 9.

212

Wagner

Figure 8 Effects of acute altitude exposure on ventilation and cardiac output at rest and during exercise. Both are higher at altitude than at sea level at a given oxygen uptake, but maximal values are essentially independent of altitude (6).

With chronic altitude exposure, the ventilatory response to exercise is even more exaggerated than with acute exposure, but conversely, the cardiac output response declines and closely follows that for sea level (28,29) (Fig. 9). These differences are likely explained by (1) enhanced ventilatory responses to chronic hypoxia as the blood and CSF alkalosis is gradually compensated by renal bicarbonate excretion and carotid body sensitivity is increased and (2) reduced heart rate response to

Gas Exchange

213

Figure 9 Ventilation and cardiac output at rest and during exercise in prolonged altitude exposure (7). In contrast to acute exposure (Fig. 8), ventilatory responses are even more enhanced than with acute exposure while the cardiac output response essentially reverts to that seen at sea level. Moreover, maximal cardiac output falls progressively with altitude (in contrast to acute exposure).

214

Wagner

hypoxia with chronic exposure possibly related to sympathetic receptor downregulation (see Chapters 6 and 9 for much fuller discussion). Stroke volume is also found to be slightly reduced. Such global responses in ventilation and cardiac output affect alveolar Po 2 and Pco 2 as well as mixed venous Po 2 and Pco 2. Gas exchange efficiency in the lungs suffers during exercise at altitude. Several factors are responsible, but their relative importance changes with altitude. Even at sea level, gas exchange is compromised during exercise with the development of ˙ mismatch at moderate and heavy levels of work. During ˙ a/Q a modest amount of V maximal effort, alveolar-capillary diffusion limitation can be observed at sea level, and there is evidence that the more elite the athlete, the more diffusion limitation can be detected (30–34). ˙ mismatch and diffu˙ a/Q At intermediate altitudes (10,000–15,000 ft.) both V sion limitation are more in evidence during exercise than at sea level (31,35). At ˙ mismatch and more diffusion limitation ˙ a/Q extreme altitudes, there is still more V as well (7). Diffusion limitation is expected as alveolar Po 2 falls onto the steep portion of the O2-Hb dissociation curve (18) but appears for the most part not to reflect any pathological reduction in diffusing capacity of the lung. In contrast, ventilation/perfusion mismatch develops as the result of some abnormal process (or processes) created by altitude and exercise in combination. Indirect evidence points to transient pulmonary edema as the basic mechanism (7,36,37). ˙ inequality and diffusion limitation are more apparent the ˙ a/Q While both V higher the altitude, the implications of Figure 5 must be kept in mind when assessing the effects of these two phenomena on O2 exchange. Thus, in spite of worsening ˙ relationships with altitude, the effects on arterial Po 2 are minimal (Fig. 5), ˙ a/Q V ˙ a/ while diffusion limitation increasingly affects arterial Po 2 (Fig. 5). On average, V ˙ inequality and diffusion limitation are of similar importance to the AaPo 2 during Q sea level exercise, but diffusion limitation becomes much more of a factor at high altitude. Figure 10 shows this relationship for the acutely exposed subjects as well ˙ mismatch at altitude ˙ a/Q as those of OEII. Despite sometimes major degrees of V (see below), diffusion limitation remains the dominant contributor to the AaPo 2, reducing the arterial O2 concentration by some 4 mL ⋅ dL⫺1 at peak exercise on the Everest ‘‘summit’’ as emulated in OEII. B. Ventilation/Perfusion Relationships During Exercise at Altitude

˙ mismatch worsens at any exercise level at altitude. ˙ a/Q As mentioned above, V Figure 11 shows this for both acute (6) and chronic (7) altitude exposures, using the same studies as discussed for resting conditions above. Because altitude reduces ˙ mismatch (Fig. 4), the only reliable data ˙ a/Q the component of the AaPo 2 due to V ˙ inequality come from the multiple inert gas elimination technique. The ˙ a/Q on V ˙ inequality of Figure 11 is not likely due to hypobaria per se. Sea ˙ a/Q increase in V level studies reducing Fio 2 but keeping Pb ⫽ 760 torr have produced the same exercise-induced worsening of mismatch as seen in hypobaric exercise breathing room air (35).

Gas Exchange

215

Figure 10 Components of the alveolar-arterial Po 2 difference (AaPo 2 ) as a function of altitude. Data reflect peak exercise levels at each altitude. Top panels reflect acute altitude exposure (6), bottom panels prolonged exposure (7). Total AaPo 2 declines progressively with altitude. The fraction due to diffusion limitation increases progressively, while that due to ˙ inequality decreases. Differences between acute and chronic hypoxic groups at sea ˙ a/Q V ˙ relationships. ˙ a/Q level presumably reflect differences in pulmonary diffusing capacity and V

˙ mismatch observed to date during ˙ a/Q Table 2 summarizes the changes in V exercise in acute hypoxia produced either by hypobaric or normobaric reduction in Fio 2. The changes are all in the same direction, but the magnitude is small and therefore of little importance to O2 transport. Chronically hypoxic subjects (Fig. 11) exhibit more variability in response than those studied during acute hypoxia. Three subject responses from OEII make this point (Fig. 12). Subject 3 was essentially insensitive to exercise and altitude in

216

Wagner

Figure 11 Amount of ventilation/perfusion mismatch (characterized by log SDQ values) in acute (upper panel) and prolonged (lower panel) altitude exposures at rest and during exer˙ inequality increases ˙ a/Q cise. Note the different ordinate scales. Under both conditions, V with exercise and is greater at a given oxygen uptake at altitude than at sea level. There is considerable individual variability, and most subjects show highly abnormal patterns with chronic hypoxia. Log SDQ values above about 1 equate to considerable, clinically significant disturbances of gas exchange.

Gas Exchange

217

˙ Mismatch During Acute Hypoxia: Literature Summary ˙ a/Q Table 2 Exercise-Induced V (mean ⫾ SE) Ref.

Pb (torr)

6 6 6 35 35

760 523 429 760 760

Fio 2

Pio 2 (torr)

Log SDQ at rest

0.21 0.21 0.21 0.21 0.12

149 100 80 149 86

0.35 0.32 0.33 0.38 0.41

⫾ ⫾ ⫾ ⫾ ⫾

0.04 0.02 0.02 0.02 0.03

Log SDQ heavy exercise

˙ o 2 (L ⋅ min⫺1) V heavy exercise

⫾ ⫾ ⫾ ⫾ ⫾

3.72 3.15 2.27 2.74 2.31

0.40 0.52 0.50 0.48 0.62

0.03 0.04 0.02 0.02 0.09

˙ inequality (Fig. 12, top panel). Only at Pb ⫽ 347 torr ˙ a/Q terms of developing V at the heaviest level was there an increase in mismatch to above normal as defined by the 95% upper confidence limit at sea level of the dispersion parameter log SDQ. In sharp contrast, subject 6 developed unusually large degrees of inequality on exercise (log SDQ exceeding 1.5) at several altitudes. Subject 9 showed little or no mismatch at sea level or moderate altitude but developed severe inequality at Pb ⫽ 347 torr, the lowest pressure at which data could be obtained from him. Recall that normally, log SDQ ranges from about 0.3 to 0.6 at rest at sea level and that the most severely ill patients with respiratory failure in the intensive care setting typically have log SDQ values of 2.0–2.5. Subject 9 clearly became ill at Pb ⫽ 347 and developed rales on clinical exam consistent with pulmonary edema. Such pathological signs were not seen in subject 6 (middle panel). Why there were marked intersubject differences is not known. Figures 11 and 12 indicate that altitude increases the degree of exercise˙ mismatch. Figure 13 shows that a given level of exercise causes ˙ a/Q induced V ˙ mismatch the higher the altitude, both in acute and chronic altitude ˙ a/Q greater V exposure. Subjects with a prior history of high-altitude pulmonary edema (HAPE) display a number of physiological differences from HAPE-free subjects: In prior HAPE victims, the hypoxic pulmonary vasoconstriction response is exaggerated (38), recent evidence suggests a higher wedge pressure during exercise as well (39), implying greater pulmonary capillary pressures than in normal subjects. Exercise˙ mismatch is greater in prior HAPE victims than in normal subjects ˙ a/Q induced V ˙ mismatch was related closely ˙ a/Q (40), and in these studies it was also found that V to pulmonary artery pressure (Fig. 14) but not to total ventilation or cardiac output. ˙ ˙ a/Q Thus, as pulmonary artery pressure rose with increasing exercise, so too did V ˙ relationships ˙ a/Q inequality. It is therefore clear that exercise at altitude worsens V in normal subjects and more so in prior HAPE victims. ˙ relationships to deteriorate ˙ a/Q The key question, of course, is what causes V during exercise, especially so at altitude? Some possible reasons are discussed in the following paragraphs.

218

Wagner

Gas Exchange

219

˙ mismatch expressed as the increase ˙ a/Q Figure 13 Summary of the effects of altitude on V ˙ inequality (log SDQ) per liter per minute increase in oxygen uptake during exercise. ˙ a/Q in V ˙ o 2 is progressively In both acute and prolonged exposures, the increase in log SDQ per liter V greater at altitude than sea level, while prolonged exposure caused greater changes.

One possibility is alterations in ventilation distribution due to altered gas density, magnified by the high gas flow rates of exercise. Exercise- or cold, dry air– induced bronchoconstriction could be a factor as well. However, despite extreme ˙ mismatch observed in OEII, spirometric tests performed immediately after ˙ a/Q V exercise failed to provide any evidence at all of airways obstruction, central or pe˙ inequality persists beyond the time for essentially ˙ a/Q ripheral (41). Moreover, V complete return of ventilation to resting values (36), suggesting that high rates of ˙ mismatch of exercise. Finally, prior HAPE victims ˙ a/Q airflow do not cause the V ˙ mismatch but lower minute ventilation during exer˙ a/Q were found to have more V cise than normal subjects (40), supporting this conclusion. For all of these reasons, factors related to airway mechanics and/or total ventilation are thought not to be ˙ mismatch on exercise, at sea level or at altitude. ˙ a/Q causative of greater V

Figure 12 Individual responses of ventilation/perfusion relationships to exercise and altitude in three subjects from Operation Everest II (prolonged exposure). Subject 3 (top) was essentially invulnerable to the combined effects of exercise and altitude, log SDQ staying within the normal range for virtually the entire study. Subject 6 shows data that, while normal at rest (except on the summit), increased to above normal values at each altitude as exercise was undertaken. Subject 9 (bottom) showed little response to exercise at altitudes up to 15,000 ft. (Pb ⫽ 429) but became clinically ill at the next altitude studied (Pb ⫽ 347 torr). Chest rales indicated acute high altitude pulmonary edema, and log SDQ values reached levels equivalent to those seen in patients in intensive care with severe lung disease.

220

Wagner

˙ inequality (log SDQ) to pulmonary vascular pressures in ˙ a/Q Figure 14 Relationship of V normal control subjects (open circles) and subjects with a prior history of high altitude pulmonary edema (HAPE-S, closed circles). Normoxic exercise data are shown in the upper panel, those during acute hypoxia in the lower panel. Each data point is the average for all subjects ˙ ˙ a/Q at a given exercise level in each group. There is a close linear relationship between V inequality and pulmonary vascular pressures (whether gauged by wedge or pulmonary artery values) in both normoxia and hypoxia. Prior HAPE subjects develop both higher pressures ˙ inequality but the relationship between log SDQ and pressure is the same ˙ a/Q and more V for both groups.

Gas Exchange

221

˙ mismatch is the result of transient, mild ˙ a/Q A second possibility is that V pulmonary edema. Except for extreme conditions (e.g., in OEII at ⱖ20,000 ft.), this is probably interstitial rather than alveolar and is fairly mild. It is well known that transcapillary fluid flux is greatly increased during exercise due to increases in intravascular pressure (42), and it seems reasonable to postulate that, if lymph clearance capacity is transiently exceeded during heavy exercise, interstitial edema may develop. This should be related to pulmonary hypertension and should therefore become more apparent with altitude (due to hypoxic vasoconstriction). The available ˙ mismatch beyond ˙ a/Q observations support these expectations. The persistence of V the postexercise period required for ventilation and cardiac output to return to resting ˙ worsen˙ a/Q levels is further consistent with this idea (36), and the elimination of V ing by breathing 100% O2 during exercise at altitude is further evidence supporting ˙ mismatch ˙ a/Q the relationship between hypoxia, pulmonary artery pressure, and V (43). Unfortunately, direct confirmation of this hypothesis in humans is lacking because the tools needed to identify minor degrees of interstitial edema do not exist. Exercise-induced perivascular cuffing has been observed in the pig (37), supporting ˙ inequality caused by edema. Some reports of radiographic ˙ a/Q the hypothesis of V techniques support the development of mild edema but remain somewhat equivocal (44). In OEII in at least one subject, acute pulmonary edema developed during exer˙ mismatch (see Fig. 12) ˙ a/Q cise at Pb ⫽ 347 torr. There were rales (25), gross V and a large shunt of over 30% of the cardiac output, similar to values seen in many intensive care unit (ICU) patients with respiratory failure (26). Whether the more common, milder degrees of exercise-induced mismatch differ only by degree or rather by basic mechanism from the HAPE observed in the above subject cannot be resolved, but Figure 14 suggests the former, since there is a smooth, single rela˙ mismatch during exercise. ˙ a/Q tionship between pulmonary artery pressure and V C. Pulmonary Diffusion Limitation of O2 Exchange During Exercise at Altitude

Diffusion of O2 in the lungs is usually thought of (a) as responsible for mixing of inspired and alveolar gas and (b) as the process of transfer of O2 from alveolar gas to capillary blood. Diffusion limitation becomes more and more apparent with increasing altitude at a given workload. The problem is O2 diffusion between alveolar gas and capillary blood, rather than any diffusive limitation to gas mixing in the alveolar spaces. The latter is barely detectable as a gas exchange defect (45) while the former becomes a major factor limiting exercise at altitude. Why should diffusion limitation become a problem at altitude? The answer is deduced best by examining Fick’s first law of diffusion: FLUX ⫽ A ⋅ k ⋅ α ⋅ [Pao 2 ⫺ Pcapo 2]/(√MW ⋅ T)

(2)

where the diffusive FLUX is a function of the alveolar surface area available (A), the physical properties of the alveolar wall (represented by a diffusion constant k),

222

Wagner

the solubility (α) of O2 in the Hb-free water of the blood-gas barrier, the molecular weight (MW) of O2 , the thickness of the alveolar blood-gas barrier (T), and the Po 2 difference between alveolar gas (Pao 2) and the pulmonary capillary (Pcapo 2) at any instant. For a given capillary blood volume (Vc), the FLUX of O2 is equivalent to the rate of change of blood O2 concentration [O2]. This in turn is equal to the product of the rate of change of blood O2 partial pressure, Pcapo 2 , and the instantaneous slope of the O2-Hb dissociation curve, β. Thus: FLUX ⫽ d[O2 ]/dt ⫻ Vc/100 ⫽ β ⋅ dPcapo 2 /dt ⋅ Vc/100

(3)

Combining Eqs. (2) and (3) gives: Pcapo 2 /dt ⫽ (A/(T ⋅ Vc)) ⋅ (100k) ⋅ (α/(β√MW)) ⋅ (Pao 2 ⫺ PcapO2 )

(4)

The key point of Eq. (4) is that the rate of equilibration (i.e., the rate of rise of Pcapo 2 ) is dependent on the gas-specific ratio α/(β√MW). Between sea level and altitude, there is likely no major change in pulmonary microvascular structure [i.e., A, T, Vc, or k of Eq. (4)], but the ratio α/β is very dependent on altitude. While α (O2 water solubility) is a constant independent of altitude, β (the slope of the O2 dissociation curve) increases with altitude, as shown in Figure 15. Since β is in the denominator of Eq. (4), the rate of diffusion equilibration falls substantially as altitude is gained. During exercise at sea level, β is about 0.2 mL O2 ⋅ dL⫺1 ⋅ torr⫺1, while on the Everest summit, OEII data show that β is about 0.7. Even the probably longer capillary transit time associated with a reduced maximal cardiac output cannot overcome this effect of increase in β.

Figure 15 The oxyhemoglobin dissociation curve showing measured arterial and pulmonary arterial oxygen concentrations as a function of Po 2 at peak exercise during Operation Everest II (prolonged exposure). Straight lines join pairs of arterial and venous points, showing that the average slope of the oxyhemoglobin dissociation curve increases with altitude. This increase in average slope retards diffusion equilibration in the pulmonary capillaries as explained in the text.

Gas Exchange

223

It is this mathematical consequence of the O2-Hb curve shape that underlies the susceptibility of the lungs to diffusion limitation at altitude. It is not necessary to invoke pathological changes in O2-diffusing capacity per se. In fact, OEII data ˙ o 2 actually indicate the converse—pulmonary O2 diffusion capacity at a given V increases with altitude (7), probably due to the associated increase in pulmonary artery pressure and/or blood hemoglobin level, both of which would, if anything, increase the effective surface area due to more red cells in more perfused capillaries. The assessment of O2-diffusing capacity (Dlo 2 ) at altitude is complicated by the ˙ inequality and its small but not insignificant (Fig. 10) contri˙ a/Q development of V butions to the total AaPo 2. However, using the multiple inert gas elimination technique, Dlo 2 can be estimated (46). The portion of the total AaPo 2 due to measured ˙ inequality is subtracted from the actual AaPo 2 and the Dlo 2 value necessary ˙ a/Q V to account for the remainder of the AaPo 2 calculated. A particular assumption is ˙ ) are correlated such that Dlo 2 / that throughout the lungs Dlo 2 and blood flow (Q ˙ ratios are uniform. While this is an untestable hypothesis, it results in the smallest Q overall value for Dlo 2 required to explain the data and is a useful parameter in a comparative sense (to determine if Dlo 2 is affected by altitude, for example). Calculations of Dlo 2 in this manner at sea level are not always reliable. This is seen ˙ inequality. ˙ a/Q when the AaPo 2 is due mostly to V Using this approach, Dlo 2 was assessed during exercise in acute altitude exposure to Pb ⫽ 523 and 429 (6) and also during the prolonged exposure of OEII (7). Figure 16 shows the effects of progressive altitude on Dlo 2 at peak exercise in these

Figure 16 Calculated diffusing capacity of the lungs for oxygen (Dlo 2 ) during peak exer˙ o 2 clearly falls progressively with cise in both prolonged and acute hypoxia. While peak V altitude (indicated by barometric pressure, Pb) peak Dlo 2 in each group is unaffected by altitude.

224

Wagner

Figure 17 Progressive increase in lung oxygen diffusion capacity (Dlo 2 ) from rest to exercise in acute and in prolonged hypoxic exposures. The relationship is not measurably dependent upon altitude in acute hypoxia (upper panel), but in the lower panel, at an oxygen uptake of 1 L ⋅ min⫺1, Dlo 2 increases progressively with altitude.

Gas Exchange

225

two scenarios (sea level data not computed). Neither acute nor chronic exposure has any effect on Dlo 2 at peak exercise. More data need be collected to be confident of these results. If one examines how estimated Dlo 2 varies with exercise intensity from the same studies, one obtains the results depicted in Figure 17. Dlo 2 increases with exercise at any altitude in both acute and chronic exposure, consistent with the older reports using carbon monoxide to measure diffusing capacity (47). In acute hypoxia, the relationship is not measurably different between the altitudes studied (Fig. 17, top), but with prolonged hypoxia Dlo 2 clearly increased with altitude at submaxi˙ o 2 (e.g.,⬃1 L ⋅ min⫺1, Fig. 17, bottom). mal, constant V ˙ o 2 ⬃ 1 L ⋅ min⫺1 ) to Pb using the data Figure 18 relates Dlo 2 at 60 watts (V of Figure 17 and demonstrates this increase with altitude. One can imagine at least two explanations: (1) greater numbers of perfused capillaries at altitude due to pulmonary hypertension or (2) greater Dlo 2 due to the increase in [Hb] with altitude. If one uses standard correction equations (48) for the effect of the rise in [Hb], from about 15 to about 18 g dL⫺1 on diffusing capacity, seen in OEII (Fig. 18, lower panel), there is still a large increase in Dlo 2 that is not accounted by increases in [Hb]. Figure 18, middle panel, shows for the OEII subjects a relationship between mean pulmonary artery pressure and Dlo 2. The vascular pressure data at 429 torr were determined by interpolation using measured values (at the same work rate) at Pb ⫽ 760, 347, 282, and 253 torr. The conclusion from Figure 18 is that Dlo 2 at ˙ o 2 increases substantially with altitude probably due to greater perfusion constant V of the capillary network. This is speculative for several reasons: it depends on data from small numbers of subjects. Second, [Hb] in the OEII subjects did not increase as much as expected due to very frequent blood sampling for the large number of research projects, so that the importance of [Hb] to Dlo 2 may have been underesti˙ inequality described above ˙ a/Q mated. Third, the moderately extensive degrees of V at high altitude may have in fact reduced Dlo 2 if alveolar or interstitial edema developed, since edema could increase blood-gas barrier thickness. If so, Dlo 2 would have been even higher at high altitude had such problems not developed. Finally, the reader is again referred to Figure 10 for a summary of the relative ˙ mismatch to the AaPo 2 during both ˙ a/Q contributions of diffusion limitation and V acute and chronic altitude exposure. This figure is consistent with Figure 5, and the anticipated effects of these two processes on gas exchange as altitude is gained.

III. Effects of Acclimatization on Pulmonary Gas Exchange To this point, this chapter has dealt with either acute hypoxic exposures over the course of minutes or prolonged exposures with steady continuing altitude ascent over several weeks. A totally different issue arises when subjects acutely ascend to a given altitude and then remain at that altitude for several days to weeks, achieving acclimatization. Does acclimatization at a given altitude affect pulmonary gas exchange, either at rest or during exercise? Very little work has explored this question

226

Wagner

Gas Exchange

227

because field and chamber studies alike have rarely opted to spend several days at any one altitude, let alone permit serial studies at any one altitude. This issue was addressed at the White Mountain Research Station in California, where pulmonary gas exchange was assessed before and after 2 weeks of residence at 3800 m (Pio 2 ⫽ 90 torr). Data were collected both at rest and during exercise (23). Acclimatization led to an increase in ventilation as expected, with concomitant increases in arterial Po 2 and decreases in arterial Pco 2. While these changes were anticipated, there was an unexpected small fall in the alveolar-arterial Po 2 difference (AaPo 2) at any level of exercise. This was consistent with a prior report of increasing arterial O2 saturation after acclimatization (49). Using the multiple inert gas elimination technique, it was shown that of the possible reasons for a lower AaPo 2 (less ˙ mismatch, higher lung O2-diffusing capacity, or greater diffusion equilibration ˙ a/Q V due to higher ventilation and lower cardiac output after acclimatization), only the latter was found to occur. In other words, improved gas exchange was simply the consequence of a longer pulmonary capillary red cell transit time due to reduction in cardiac output and higher alveolar Po 2 from increased ventilation (increasing the alveolar-capillary Po 2 difference and thus the rate of diffusion). It is of interest to note that systemic O2 delivery (the product of arterial O2 concentration and cardiac output) was unaffected by acclimatization—the reduced cardiac output was balanced by increased blood [O2]. This observation is consistent with the known lack of effect ˙ o 2 (50). of such short-term acclimatization on maximal V IV. Muscle Tissue Gas Exchange This chapter has focused on pulmonary gas exchange, since major changes occur in the lungs and cardiovascular systems as described. However, the processes allowing O2 to be extracted from muscle microcirculatory blood and to reach the mitochondria are also subject to effects of altitude and may also contribute to exercise limitation. It has long been known that even in healthy subjects exercising maximally at sea level, venous blood from muscle is never fully O2-depleted (51). This failure to completely extract O2 could reflect a transport limitation for O2 in muscle (due to

Figure 18 Analysis of potential causes of the increase in Dlo 2 (constant work rate, 60 watts) with increasing altitude. Top panel shows the significant increase in Dlo 2 with altitude. Middle panel relates Dlo 2 to mean pulmonary artery pressure at these altitudes, with an almost proportional, linear relationship between the two. When Dlo 2 is plotted against blood hemoglobin concentration (lower panel), there is also a positive correlation, but correction of Dlo 2 for the increase in hemoglobin concentration according to standard formulas shows that hemoglobin concentration is a minor contributor to the increase in Dlo 2 with altitude. See text for more details.

228

Wagner

˙ o 2 heterogeneity, shunts, or diffusion limitation), or it could reflect blood flow/V maximal mitochondrial O2 utilization rates such that the residual venous O2 reflects unusable O2 that mitochondria could not burn due to their being in a state of maximal ATP turnover. To distinguish between these two alternatives, one can increase arterial O2 availability by increasing Pio 2, muscle blood flow, or [Hb] and determine ˙ o 2 is increased or not. Each of these manipulations has been whether maximal V ˙ o 2 , allowing the conclusion that the residual studied, and each increases maximal V venous O2 represents an O2 transport limitation in the muscle (see Ref. 52 for re˙ o 2 with altitude and view). These findings are consistent with the fall in maximal V ˙ o 2max (completely in acute hypoxia and almost comthe immediate restoration of V pletely in prolonged hypoxia) when the hypoxia is eliminated by breathing 100% O2. As normal subjects exercise maximally at increasing altitudes, both arterial ˙ o 2 (53). Figure 19 shows and muscle venous O2 levels fall in parallel with maximal V how the principal components of the O2 transport chain respond to maximal exercise in subjects of OEII at each altitude. Mixed venous O2 levels are still considerably greater than zero, even on the summit. More interestingly, there are correlations ˙ o 2 max), and mean ˙ o 2 and total O2 transport, venous Po 2 (at V between maximal V capillary Po 2, as Figure 20 shows from the acute hypoxic exposure (6) and OEII ˙ o 2max to venous and data (11), respectively. Note that in OEII the correlations of V mean capillary Po 2 fail below Pb ⫽ 347 torr. The existence of these relationships is postulated to shed light on the role of ˙ o 2max at altitude. The O2 transport in muscle in contributing to the reduction in V data of Figure 20 (acute hypoxia) are compatible with the notion of a constant conductance for O2 in muscle between the red cell and the mitochondria as Pb is varied. ˙ o 2 to the mean Po 2 difference between the Conductance is the ratio of maximal V capillaries and the mitochondria and is the slope of the line in Figure 20 (lower ˙ o 2max is negligibly small (54). If mean capillary panel) if mitochondrial Po 2 at V Po 2 falls in proportion to muscle venous Po 2 as altitude is gained, this conductance ˙ o 2max to venous Po 2 at the various will be proportional to the slope of the plot of V altitudes (Fig. 20, middle). Figure 20 does indicate proportionality and thus a constant conductance as altitude is changed. This conductance is clearly finite since complete O2 extraction does not occur, but whether it is limited by diffusive properties of the intramuscular O2 transport pathway or rather by inhomogeneity of blood flow with respect to metabolic rate or by muscle shunts cannot be deduced from the data presented. Animal studies in which mean capillary Po 2, but not convective O2 delivery, is changed by altering ˙ o 2 max changing in proportion to mean capillary Po 2 at constant O2 Hb P50 show V delivery (55). Such an outcome is not compatible with inhomogeneity but is the expected consequence of diffusion limitation. The residual O2 in venous blood is thus hypothesized to result from diffusion limitation of intramuscular O2 transport. Yet other studies have pointed to the importance of the amount of capillary structure in muscle as the key determinant of the total muscle conductance (56), and in addition, theoretical calculations suggest that muscle diffusive conductance is a more

Gas Exchange

229

Figure 19 Principal systemic oxygen transport variables in the subjects of Operation Everest II at altitudes from sea level to the Everest summit. Systemic oxygen transport falls progressively with altitude because both cardiac output and arterial oxygen saturation fall with altitude exposure. Peripheral oxygen extraction also falls, which appears to buffer Po 2 of mixed venous blood. For further interpretation, see text.

230

Wagner

Gas Exchange

231

important determinant of mitochondrial O2 supply at altitude than at sea level (57). The muscle biopsy data of Oelz and coworkers (58) in elite climbers reaching the summit of Everest without supplemental O2 showed increased muscle capillarity, and it is tempting to speculate that a key element of climbing success is muscle capillarity facilitating diffusive unloading of O2. Much work needs to be done in this area to better understand the factors that are important to maximal O2 transport to, and extraction by, exercising muscle at altitude. The data on prolonged hypoxia (Fig. 20) are not so easy to interpret as those ˙ o 2 remains closely tied to total O2 for acute hypoxia (Fig. 20), since while peak V ˙ o 2 and mean capillary Po 2 transport (top panel), the relationship between peak V (lower panel) loses proportionality at extreme altitude (Pb ⫽ 282 and 253 torr). Between sea level and Pb ⫽ 347 torr, the results are compatible with previous data in fit subjects during acute hypoxic exposure. They indicate a muscle O2 conductive ˙ o 2max at altitude by limit that contributes to a predictable, declining ceiling on V preventing complete O2 extraction. At higher altitudes, the apparent conductance ˙ o 2 is lower at a given capillary Po 2.) This fall in has fallen some 20%. (Peak V conductance has occurred despite a reduction in average muscle fiber diameter that reduces average diffusion distances within muscle (59) and which should therefore enhance O2 conductance. One possibility is that loss of muscle mass at extreme altitude reduces peak O2 consumption and thus total conductance. Another is that because of the considerable reduction in maximal cardiac output at extreme altitude (Fig. 19), perfusion of muscle capillaries is more nonhomogeneous than at lower altitudes. One possibility has been excluded—loss of muscle capillaries with severe hypoxia: biopsy data suggests that muscle capillarity was not reduced in OEII (59). However, since the venous blood sampled for the analysis of maximal O2 transport is mixed venous and not muscle venous in origin, another possibility is increasing contamination of muscle venous blood with venous blood from nonexercising tissues (kidney, brain, gut). This contamination would become evident as a higher than anticipated mixed venous Po 2 since muscle venous Po 2 is less than mixed venous (53). This could play a greater role at extreme altitude compared to sea level because of the steadily falling maximal cardiac output, which implies a greater relative contribution to mixed venous blood by nonexercising tissue blood

Figure 20 Relationship between peak oxygen uptake and total oxygen transport (upper panel), mixed venous Po 2 (middle panel) and calculated mean capillary Po 2 (lower panel) in acute (䊊) and prolonged (䊉) hypoxic exposures. In all three panels there is an essentially linear and proportional relationship for acute hypoxia, indicating a finite peripheral muscle oxygen conductance for oxygen that interacts with systemic oxygen transport to limit peak ˙ o 2. In prolonged hypoxia, there is a closely proportional relationship between V ˙ o 2 and sysV ˙ o 2 and venous or mean capillary Po 2 is proportional temic oxygen transport, but that between V ˙ o 2 is less than expected for calculated only to modest altitude. At extreme altitudes, peak V mean capillary Po 2, suggesting a fall in muscle oxygen conductance. See text for further interpretation.

232

Wagner

at high altitude. Direct femoral venous blood sampling offers a better chance of choosing among these various explanations, especially because muscle blood flow rather than cardiac output can be measured using the same catheters. References 1. West JB, Lahiri S, Maret KH, et al. Barometric pressures at extreme altitudes on Mt. Everest: physiological significance. J Appl Physiol 1983; 54:1188. 2. Ward MP, Milledge JS, West JB. In: Ward MP, Milledge JS, West JB, eds. High Altitude Medicine and Physiology. London: Chapman and Hall Ltd., 1989:27–44. 3. Greene R. Observations on the composition of alveolar air on Everest, 1933. J Physiol Lond 1934; 32:481. 4. Pugh LGCE. Resting ventilation and alveolar air on Mount Everest: with remarks on the relation of barometric pressure to altitude in mountains. J Physiol (Lond) 1957; 135: 590. 5. Gill MB, Milledge JS, Pugh LGCE, et al. Alveolar gas composition at 21,000 to 25,000 feet (8,400–7,830 m). J Physiol (Lond) 1962; 163:373. 6. Wagner PD, Gale GE, Moon RE, et al. Pulmonary gas exchange in humans exercising at sea level and simulated altitude. J Appl Physiol 1986; 61(1):260. 7. Wagner PD, Sutton JR, Reeves JT, et al. Operation Everest II: pulmonary gas exchange during a simulated ascent of Mt. Everest. J Appl Physiol 1987; 63(6):2348. 8. Rahn H, Fenn WO. A Graphical Analysis of the Respiratory Gas Exchange. Washington, DC: American Physiological Society, 1955. 9. Houston CS. Acclimatization to hypoxia: Operations Everest I and II. Ann Sports Med 1988; 4(4):171. 10. West JB. Everest—the Testing Place. New York: McGraw-Hill Book Co., 1985. 11. Sutton JR, Reeves JT, Wagner PD, et al. Operation Everest II: oxygen transport during exercise at extreme simulated altitude. J Appl Physiol 1988; 64:1309. 12. West JB, Hackett PH, Maret KH, et al. Pulmonary gas exchange on the summit of Mount Everest. J Appl Physiol 1983; 55:678. 13. Peacock AJ, Jones PL. Gas exchange at extreme altitude: results from the British 40th Anniversary Everest Expedition. Eur Respir J 1997; 14. West JB, Lahiri S, Gill MB, et al. Arterial oxygen saturation during exercise at high altitude. J Appl Physiol 1962; 17:617. 15. Hurtado A. Animals in high altitudes: resident man. In: Dill ed. Adaptation to the Environment. Washington, DC: American Physiological Society, 1964:843. 16. Kreuzer F, Tenney SM, Mithoeffer JC, et al. Alveolar-arterial oxygen gradient in Andean natives at high altitude. J Appl Physiol 1964; 19:13. 17. Hultgren HN. High altitude pulmonary edema. In: Hegnauer AH, ed. Biomedicine Problems of High Terrestrial Altitude. New York: Springer-Verlag, 1969:131. 18. West JB, Wagner PD. Predicted gas exchange on the summit of Mt. Everest. Respir Physiol 1980; 42:1. 19. West JB. Ventilation/perfusion inequality and overall gas exchange in computer models of the lung. Respir Physiol 1969; 7:88. ˙ distributions from analysis of experimental ˙ A/Q 20. Evans JW, Wagner PD. Limits on V inert gas elimination. J Appl Physiol 1977; 42(6):889.

Gas Exchange

233

21. Wagner PD, Naumann PF, Laravuso RB. Simultaneous measurement of eight foreign gases in blood by gas chromatography. J Appl Physiol 1974; 36(5):600. 22. Wagner PD, Laravuso RB, Uhl RR, et al. Continuous distributions of ventilationperfusion ratios in normal subjects breathing air and 100% O2. J Clin Invest 1974; 54(1): 54. 23. Bebout DE, Story D, Roca J, et al. Effects of altitude acclimatization on pulmonary gas exchange during exercise. J Appl Physiol 1989; 67:2286. 24. Wagner PD, Hedenstierna G, Bylin G, et al. Reproducibility of the multiple inert gas elimination technique. J Appl Physiol 1987; 62:1740. 25. Houston CS, Sutton JR, Cymerman A, et al. Operation Everest II: man at extreme altitude. J Appl Physiol 1987; 63(2):877. 26. Dantzker DR, Brook CJ, DeHart P, et al. Ventilation/perfusion distributions in the adult respiratory distress syndrome. Am Rev Respir Dis 1979; 120:1039. 27. Cerretelli P. Gas exchange at high altitude. In: West JB, ed. Organism and Environment. New York: Academic Press, 1980:97. 28. Pugh LGCE, Gill MB, Lahiri S, et al. Muscular exercise at great altitudes. J Appl Physiol 1964; 19:431. 29. Pugh LGCE. Cardiac output in muscular exercise at 5,800 m (19,000 ft). J Appl Physiol 1964; 19:441. 30. Gledhill N, Froese AB, Dempsey JA. Ventilation to perfusion distribution during exercise in health. In: Dempsey JA, Reed CE, eds. Muscular Exercise and the Lung. Madison: University of Wisconsin Press, 1977:325. 31. Torre-Bueno JR, Wagner PD, Saltzman HA, et al. Diffusion limitation in normal humans during exercise at sea level and simulated altitude. J Appl Physiol 1985; 58(3): 989. 32. Dempsey JA, Reddan WG, Birnbaum ML, et al. Effects of acute through life-long hypoxic exposure on exercise pulmonary gas exchange. Respir Physiol 1971; 13:62. 33. Dempsey JA, Hanson PG, Henderson KS. Exercise-induced arterial hypoxemia in healthy human subjects at sea level. J Physiol (Lond) 1984; 355:161. 34. Powers SK, Lawler J, Dempsey J, et al. Effects of incomplete pulmonary gas exchange ˙ o 2max. J Appl Physiol 1989; 66:2491. on V 35. Hammond MD, Gale GE, Kapitan KS, et al. Pulmonary gas exchange in humans during normobaric hypoxic exercise. J Appl Physiol 1986; 61(5):1749. ˙ distribution during heavy exercise ˙ A/Q 36. Schaffartzik W, Poole DC, Derion T, et al. V and recovery in humans: implications for pulmonary edema. J Appl Physiol 1992; 72(5): 1657. 37. Schaffartzik W, Arcos J, Tsukimoto K, et al. Pulmonary interstitial edema in the pig after heavy exercise. J Appl Physiol 1993; 75(6):2535. 38. Fasules JW, Wiggins JW, Wolfe RR. Increased lung vasoreactivity in children from Leadville, Colorado, after recovery from high-altitude pulmonary edema. Circulation 1985; 72(5):957. 39. Eldridge MW, Podolsky A, Richardson RS, et al. Pulmonary hemodynamics response to exercise in subjects with prior high-altitude pulmonary edema. J Appl Physiol 1996; 81(2):911. ˙ inequality ˙ A/Q 40. Podolsky A, Eldridge MW, Richardson RS, et al. Exercise-induced V in subjects with prior high altitude pulmonary edema. J Appl Physiol 1996; 81(2): 922.

234

Wagner

41. Welsh CH, Wagner PD, Reeves JT, et al. Operation Everest II: spirometric and radiographic changes in acclimatized humans as simulated high altitudes. Am Rev Respir Dis 1993; 147:1239. 42. Coates G, O’Brodovich H, Jefferies AL, et al. Effects of exercise on lung lymph flow in sheep and goats during normoxia and hypoxia. J Clin Invest 1984; 74:133. 43. Gale GE, Torre-Bueno JR, Moon RE, et al. Ventilation/perfusion inequality in normal humans during exercise at sea level and simulated altitude. J Appl Physiol 1985; 58(3): 978. 44. Anholm JD, Bourne JC, Pai RG, et al. Hypoxic pulmonary vasoconstriction is not necessary for development of radiographic evidence of pulmonary edema following exercise at moderate altitude. Hypoxia Brain 1995:312. 45. Hlastala MP, Scheid P, Piiper J. Interpretation of inert gas retention and excretion in the presence of stratified inhomogeneity. Respir Physiol 1981; 46:247. 46. Hammond MD, Hempleman SC. Oxygen diffusing capacity estimates derived from ˙ distributions in man. Respir Physiol 1987; 69:129. ˙ a/Q measured V 47. Johnson RL Jr, Spicer WS, Bishop JM, et al. Pulmonary capillary blood volume, flow and diffusing capacity during exercise. J Appl Physiol 1960; 15:893. 48. Cotes JE, Dabbs JM, Elwood PC, et al. Iron-deficiency anemia: its effect on transfer factor for the lung, diffusing capacity and ventilation and cardiac frequency during submaximal exercise. Clin Sci 1972; 42:325. 49. Bender PB, McCullough RE, McCullough RG, et al. Increased exercise Sao 2 independent of ventilatory acclimatization at 4300 m. J Appl Physiol 1989; 66:2733. 50. Bender PR, Groves BM, McCullough RE, et al. Oxygen transport in exercising leg in chronic hypoxia. J Appl Physiol 1988; 65:2592. 51. Pirnay F, Lamy M, Dujardin J, et al. Analysis of femoral venous blood during maximum exercise. J Appl Physiol 1972; 33:289. 52. Wagner PD. Determinants of maximal oxygen transport and utilization. In: Massaro D, ed. Annual Reviews of Physiology. vol. 58. Palo Alto, CA: Annual Reviews, 1996:21. ˙ o 2max 53. Roca J, Hogan MC, Story D, et al. Evidence for tissue diffusion limitation of V in normal humans. J Appl Physiol 1989; 67:291. 54. Honig CR, Gayeski TEJ, Federspiel WJ, et al. Muscle O2 gradients from hemoglobin to cytochrome: new concepts, new complexities. Adv Exp Med Biol 1984; 169:23. ˙ o 2 max at 55. Hogan MC, Bebout DE, Wagner PD. Effect of increased Hb-O2 affinity on V constant O2 delivery in dog muscle in situ. Med Sci Sports Exerc 1991; 23S36. 56. Mathieu-Costello O. Comparative aspects of muscle capillary supply. Ann Rev Physiol 1993; 55:503. ˙ o 2 max. Respir Physiol 1993; 57. Wagner PD. Algebraic analysis of the determinants of V 93:221. 58. Oelz O, Howald H, di Prampero PE, et al. Physiological profile of world-class highaltitude climbers. J Appl Physiol 1986; 60:1734. 59. Green HJ, Sutton JR, Cymerman A, et al. Operation Everest II: adaptations in human skeletal muscle. J Appl Physiol 1989; 66(5):2454.

9 The Cardiovascular System at High Altitude Heart and Systemic Circulation

EUGENE E. WOLFEL

BENJAMIN D. LEVINE

University of Colorado Health Sciences Center Denver, Colorado

Institute for Exercise and Environmental Medicine University of Texas Southwestern Medical Center Dallas, Texas

I.

Introduction

We climbed a hundred meters, mouths open, trying to suck in air, resting after a few steps, then again a few more. —Reinhold Messner, May 1978

The cardiovascular system has been a major focus of interest in high-altitude physiology and medicine since it was first observed that symptoms in normal individuals at altitude were similar to those experienced by patients with circulatory insufficiency at sea level (1). Breathlessness, excessive fatigue, and tachycardia in healthy climbers during exertion or in patients with high-altitude illness at rest suggested a potential cardiovascular abnormality. Some of the key historical figures in hypoxia research—Douglas and Haldane on Pikes Peak; Barcroft in the Peruvian Andes— created an early controversy by making directionally opposite observations regarding an increase (2) or decrease (3) in cardiac output after acclimatization to altitude. Although the magnitude and temporal progression of this adaptation has been more precisely worked out over the past 50 years (4), controversy still exists regarding its mechanisms and physiological consequences. The major role of the cardiovascular system is to transport oxygen and sub235

236

Wolfel and Levine

Figure 1 Schematic of the cardiovascular hemodynamic changes that occur with acute and chronic hypoxia compared to sea level. Heart rate, cardiac output, muscle flow, and stroke volume initially rise with acute hypoxia but decrease to levels at or below sea level with sustained exposure. Blood pressure and vascular resistance initially fall with acute hypoxia but rise progressive with prolonged hypoxia.

Cardiovascular System at High Altitude

237

strate to support the metabolic demands of the body. Moreover, during increases in physical activity, the cardiovascular system must not only maintain oxygen transport to the vital organs, but also augment peripheral oxygen delivery to skeletal muscle by as much as 10-fold during maximal exercise. Both the systemic (cardiac output) and regional mechanisms by which the cardiovascular system matches oxygen supply with oxygen demand are challenged by a hypoxic environment. The response appears to be a central activation via the sympathetic nervous system coupled with local vascular control mechanisms directing regional distribution of flow. Thus gross changes in the pump function of the heart (affecting heart rate and stroke volume) are ‘‘fine-tuned’’ at a local level to ensure that organ-specific metabolic demands are met. In addition, the heart is not only a pump but a neurohumoral organ as well, both initiating and responding to changes in autonomic (sympathetic and parasympathetic) tone and circulating neurohormones. Moreover, recent insights regarding the peripheral vasculature as a sensor and modulator of circulatory control via both endothelium-dependent and -independent mechanisms have improved understanding of the precision by which the cardiovascular system operates to serve its primary function of supporting oxygen transport. Finally, acute responses are modified by chronic adaptations that restore circulatory function towards normoxic levels over time periods that may range from a few days or weeks for the sea level sojourner to years for the high-altitude native. A qualitative summary of the acute and chronic effects of hypoxia on the cardiovascular system is presented in Figure 1. The data supporting this summary will be reviewed, and areas of controversy, with specific attention to the most important issues for future research, will be highlighted. II. Acute Hypoxia Acute hypoxia decreases Pao 2 and arterial oxygen content, which activates the sympathetic nervous system primarily via peripheral chemoreceptors. The carotid bodies appear to respond to partial pressure of oxygen, at least with respect to ventilatory signals, while local vascular beds seem to be regulated by arterial oxygen content. This activation is the key acute adaptive response of the cardiovascular system. However, peripheral vasoconstriction does not occur due to release of local vasodilatory substances and ‘‘functional sympatholysis.’’ Thus, heart rate and cardiac output increase, but systemic vascular resistance and blood pressure decrease transiently. With acute exposure to normobaric or hypobaric hypoxia, the partial pressure of oxygen in the air, alveoli, and blood is reduced, and oxyhemoglobin desaturates, reducing arterial oxygen content. Although the specific stimulus to the cardiovascular system is surprisingly unclear, the ultimate circulatory response is an increase in systemic blood flow in order to maintain tissue oxygen delivery. Classic studies by Grollman on Pikes Peak (4300 m) demonstrated that there is an approximately 40% increase in resting cardiac output within the first few days of ascent to high altitude (4). Similar observations have been made under more rigorously controlled

238

Wolfel and Levine

conditions in the laboratory, with equivalent degrees of normobaric (5,6) or hypobaric (7–9) hypoxia. The magnitude of the increase in cardiac output appears proportional to the hypoxia, with no apparent increase at altitudes below 700 m (7), reaching as much as 75% increase at altitudes approaching 5000 m (7–9). This response appears to be independent of gender (7,10) or age (11). A. Components of Cardiac Output—Acute Hypoxia Heart Rate

The increase in cardiac output at rest with acute high-altitude exposure is for the most part due to an elevation in heart rate (7–9,12). Resting heart rate is a function of both the intrinsic heart rate and the balance of sympathetic and parasympathetic neural activity, which may vary widely among individuals. Intrinsic heart rate does not change with acute hypoxia [i.e., no change in the face of combined adrenergic and vagal blockade (13)]; therefore, autonomic mechanisms must be invoked to explain the increase in heart rate. The evidence for sympathetic activation is compelling (see Chapter 13). Indirect measures such as plasma and urinary catecholamines (14–16) have shown a consistent elevation in both norepinephrine and epinephrine concentrations with acute altitude exposure, with epinephrine increasing to a greater degree (15). However, such data must be interpreted cautiously as norepinephrine clearance also increases prominently with acute hypoxia (17). Frequency analysis of heart rate variability has also been used as a means of analyzing changes in autonomic function, with high-frequency variability (0.15–0.40 H) reflecting modulation of parasympathetic activity and low-frequency variability (0.04–0.15 H) modulated by both sympathetic and parasympathetic activity. However, this growing body of data attempting to use dynamic analysis of heart rate variability as an index of autonomic function at altitude (18–26) suffers from the frequent failure to quantify many of the variables that markedly influence heart rate variability, such as ventilatory volume and respiratory rate (27), cardiac volume and mechanical function (28), arterial stiffness and pulse amplification, arterial baroreflex function (29), and sinus node responsiveness (30), all of which are altered by altitude exposure. Thus, it may be difficult to unravel true changes in sympathetic or parasympathetic function at altitude (or any other condition) from analysis of heart rate variability alone (31). Fortunately, recent direct measurements of efferent postganglionic sympathetic activity to skeletal muscle have confirmed a proportional relationship between the magnitude of hypoxia (32) or high altitude (33–35) and sympathetic activation. It is this increase in sympathetic activity that is likely responsible for the global stimulation of the circulatory system in the face of acute hypoxia. However, sympathetic activation by itself cannot explain all of the increase in heart rate or cardiac output. For example, beta-blockade alone does not completely abolish the heart rate response to acute hypoxia (12,13). In one study, blockade reduced the hypoxia-induced increase in cardiac output and heart rate by only 50% (12), suggesting an important component of vagal withdrawal in addition to sympa-

Cardiovascular System at High Altitude

239

thetic activation. Further evidence for vagal withdrawal during hypoxia was presented by Eckberg et al., who noted the rapid onset of shortening of the R-R interval, indicative of vagal withdrawal, after brief periods of hypoxia deemed to be too transient to fully activate the sympathetic nervous system (36). This vagal contribution was confirmed in a study by Koller et al., in which vagal blockade with atropine in combination with beta-blockade prevented any further increase in resting heart rate with hypoxia (13). These observations are at odds with the traditional concept that stimulation of peripheral chemoreceptors by hypoxia leads to an increase rather than a decrease in efferent vagal activity (37). However, cardiac vagal motoneurons appear to be inhibited by central inspiratory neuronal activity as well as by afferents from thoracic stretch receptors (38). Thus, bradycardia, rather than tachycardia, is present during apnea (37) when these inhibitory influences are silent. Moreover, R-R shortening was noted almost exclusively during inspiration in the study by Eckberg et al. cited above (36). These findings support the role of vagal withdrawal in the tachycardia associated with acute hypoxia. In summary, it appears that hypoxia causes sympathetic activation directly via stimulation of peripheral chemoreceptors (39) and vagal withdrawal indirectly via increases in ventilation, resulting ultimately in tachycardia and an increase in cardiac output. The recent observation that afferent fibers from both baroreflex and chemoreflex receptor populations synapse very closely together in the nucleus tractus solitarius (40) raises the intriguing possibility that regulation of the cardiovascular and ventilatory responses to acute hypoxia may be neurally connected. Preliminary observations by Asano et al. (41) demonstrating a relationship between urinary norepinephrine excretion and ventilation with altitude acclimatization support this notion and suggest an important area for future research. Stroke Volume

Changes in stroke volume play a minor role in the acute cardiovascular response to hypoxia. Although myocardial contraction velocity is augmented, consistent with sympathetic activation (42), and end-systolic volume is slightly reduced, this increase in contractility may be offset by a reduction in end-diastolic volume resulting in only a small increase or no change in stroke volume (4,7–9), particularly when the measurements are performed within 1–2 hours after exposure. Most importantly, even severe acute hypoxia is not associated with depression of myocardial contractile function in normal hearts, although sustained hypoxia equivalent to 10,000 m altitude will result in heart failure in dogs if adrenergic compensatory mechanisms are inhibited (43). Changes in stroke volume, however, become much more important during acclimatization and will be discussed in more detail below. Changes in Peripheral Vascular Resistance

Activation of the sympathetic nervous system under a variety of conditions results in peripheral vasoconstriction. However, with the sympathetic activation of acute hypoxia, this vasoconstriction is absent. In fact, in virtually every study published, acute hypoxia results in vasodilation in all vascular beds except the lung (7,44–47)

240

Wolfel and Levine

directly in proportion to the reduction in arterial oxygen content. An example of the dramatic ability of hypoxia to override sympathetically mediated vasoconstriction at the local level is shown in Figure 2 (44). In this study, lower body negative pressure was used to pool blood in the lower extremities, thus reducing central blood volume and unloading cardiopulmonary and arterial baroreceptors (27,48). This technique, when applied at relatively low levels of suction in healthy subjects under normoxic conditions, always leads to a prominent increase in sympathetic nerve activity and increase in limb vascular resistance, with maintenance of arterial pressure. In contrast, when hypoxia, induced by breathing 10% oxygen, is superimposed on lower body negative pressure, forearm vascular resistance fails to rise, and the ability to maintain blood pressure is compromised (see Fig. 2). This compromise in blood pressure control during orthostatic stress may explain the increased incidence of orthostatic hypotension and syncope recently reported in some otherwise healthy individuals early after ascent to high altitude (49). The capacity of local, metabolic control mechanisms to override sympathetic vasoconstriction has been well described during exercise and has been termed ‘‘functional sympatholysis’’ (50,51). The combination of global activation of the sympathetic nervous system with a resultant increase in heart rate and cardiac output, accompanied by simultaneous local vasodilation resulting in distribution of flow to vascular beds with the greatest metabolic demand, thus is a physiological strategy

Figure 2 Hypoxia prevents the rise in forearm vascular resistance with lower body negative pressure (LBNP) and the ability to maintain blood pressure is compromised. (Data from Ref. 44.)

Cardiovascular System at High Altitude

241

common to both exercise and hypoxia. During exercise, the mechanism of functional sympatholysis appears to be muscular contraction–induced opening of ATPsensitive potassium channels by endothelium-derived hyperpolarizing factor (52) mediated, at least in part, by nitric oxide (53). Whether a similar mechanism is operative during hypoxia is not known. In the coronary arteries, acute hypoxia causes neurally mediated cholinergic vasodilation (54). However, there is no evidence that such a mechanism exists in the peripheral circulation. Furthermore, beta-adrenergic blockade in the forearm does not alter the vasodilator response to acute hypoxia (12), providing evidence against direct adrenergic vasodilation. The increment in leg blood flow associated with hypoxia, both at rest as well as during exercise, precisely matches the decrease in arterial oxygen content, keeping O2 delivery constant (45,46). Any putative mechanism must explain this remarkably tight regulation of peripheral blood flow to oxygen availability and metabolic demand. Recently an elegant study by Ellsworth et al. suggested that the red cell itself might be a potential regulator of the vascular resistance, thus providing a unifying hypothesis for the mechanism of ‘‘sensing’’ of arterial oxygen content (55). They demonstrated that the release of ATP from red blood cells, as a direct function of hypoxemia and acidosis, could bind to P2y receptors on the vascular endothelium, resulting in a propagated vasodilation both proximally and distally. Whether these ATP-sensitive receptors are the same ones mediating exercise-induced local vasodilation (52) remains to be determined. Whether sympathetic vasoconstriction or hypoxic vasodilation prevails ultimately determines what happens to arterial blood pressure; the outcome for any given individual probably depends on the absolute altitude achieved, the rapidity of early acclimatization responses, which may quickly restore arterial oxygen content despite persistent hypoxia, and the timing of the blood pressure measurement. When blood pressure measurements are made very early—within the first hour of exposure—most studies show a decrease in total peripheral resistance and a small but measurable reduction in blood pressure (7,11,44,56). However, when measurements are made even a few hours after exposure, the acute increase in hemoglobin concentration from reduction in plasma volume may tip the balance toward sympathetic vasoconstriction, and modest hypertension has been observed under chronic conditions as discussed below (57). With the increase in hemoglobin concentration there is a local increase in arterial oxygen content, which may activate some peripheral sensor that decreases flow by vasoconstriction in order to match oxygen supply with demand. A key question arises: What is the hypoxic stimulus to both peripheral chemoreceptors and peripheral vasculature—partial pressure of oxygen or arterial oxygen content? It is clear that the receptors in the carotid body that mediate ventilation respond directly to Pao 2 (58). Whether the receptors that mediate sympathetic activation respond similarly is unknown, but likely. However, as discussed above, the peripheral circulation, as well as other organs such as the kidney, respond to arterial oxygen content such that anemia results in vasodilation or synthesis and release of erythropoietin even with a normal Pao 2. This distinction may be critically

242

Wolfel and Levine

important in understanding the chronic response to hypoxia and is an important area of future research. In summary, acute hypoxia stimulates peripheral chemoreceptors to increase sympathetic nerve activity, resulting in an increase in heart rate and systemic blood flow (cardiac output). Local vasodilating mechanisms, responding to reductions in arterial oxygen content result in a reduction in peripheral resistance and a modest transient reduction in arterial pressure. Key Unanswered Questions—Acute Hypoxia

What are the biological signals derived from oxygen that are being sensed by the cardiovascular system, and what are the pathway and feedback loops which regulate them? Are the cardiovascular and ventilatory systems linked together via neural connections in the nucleus tractus solitarius (NTS) facilitating matching of ventilation and oxygen transport during hypoxia? Is the sympathetic system necessary for successful acclimatization, or can it be potentially harmful both to healthy individuals and to those with highaltitude illness or underlying disease? Is endothelial function the important factor in the modulation of functional sympatholysis? III. Sustained Hypoxia Acclimatization occurs predominantly in other organ systems, with the cardiovascular system responding to, rather than initiating changes in oxygen availability. Thus, as Cao2 increases from first hemoconcentration and ventilatory acclimatization, and later increased red cell mass, the hyperdynamic state of the circulation diminishes and both cardiac output and peripheral blood flow return toward normal. This decrease in blood flow may be adaptive by allowing more diffusion time for extraction of oxygen, both in the lung as well as in skeletal muscle and other organs. Sympathetic activation persists, particularly at higher altitudes, and thus, resting heart rate remains elevated. This chronic sympathetic hyperactivity leads to downregulation of beta receptors, which contributes to a reduction in rest and exercise heart rate. Blood pressure gradually rises as the peripheral mechanisms responsible for sympatholysis abate, but sympathetic activation actually increases, possibly as a function of increased chemosensitivity. Changes in stroke volume play a much greater role in the response to sustained hypoxia, with a consistently reported decrease at all altitudes studied. However, ventricular contractile function remains normal, and the reduction in stroke volume is entirely a consequence of ventricular filling and the Starling mechanism. In contrast to the acute response to hypoxia, in which the cardiovascular system plays a primary role in restoring oxygen transport toward normal, the response of the cardiovascular system to sustained hypoxia depends more on adaptation by

Cardiovascular System at High Altitude

243

other organ systems. There is an ongoing interplay between the ventilatory, metabolic, and hematological adaptations, which improve O2 delivery and result in secondary changes in heart rate, stroke volume, cardiac output, blood pressure, and peripheral vascular resistance over time at high altitude. Other factors, such as severity and duration of hypoxia, level of activity, nutritional status, and fluid balance, may also play a role. Ventilatory acclimatization, the key early adaptive response, is characterized by increases in alveolar ventilation, which maximize Pao 2 and shift the oxyhemoglobin dissociation curve to the left. Moreover, within the first few hours to days of altitude exposure, there is a prominent decrease in plasma volume that increases the hemoglobin concentration. Together, these adaptations substantially increase the Cao 2 within the first few days of sustained altitude exposure (see Chapters 6 and 15). A. Cardiac Adaptations—Sustained Hypoxia Heart Rate

In the context of increased chronic sympathetic activation with sustained hypoxia, resting heart rate has been shown to be consistently elevated at altitudes greater than 3000 m. The degree of elevation in resting heart rate is dependent on the duration and severity of sustained hypoxia (Fig. 3). The greater the altitude, the higher the resting heart rate, with little evidence for a decrease toward sea level values over 2–3 weeks. Studies examining the effects of progressively increasing altitude (i.e.,

Figure 3 Mean resting heart rates with varying degrees of exposure to high altitude. Numbers identify each individual study listed in the references. There appears to be a linear relationship between progressive altitude and resting heart rate.

244

Wolfel and Levine

superimposing a more severe hypoxic stimulus upon prior acclimatization) have shown further increases in resting heart rate with each increment in altitude. For example, in the American Medical Research Expedition to Mount Everest, resting heart rate increased 20 bpm from sea level to 6300 m (60). Similar data have been obtained in a hypobaric chamber study, confirming that hypobaric hypoxia per se, rather than the stresses of mountaineering, is responsible for this tachycardia (61,62). Thus, even in the setting of adequate acclimatization, the ability to respond to a further increase in hypoxia with cardioacceleration remains present. Although beta blockade prior to ascent prevents most of the increase in resting heart rate with sustained altitude exposure (63,64), confirming the essential role of hypoxia-induced sympathetic activation in this response, when beta blockade (65) or supplemental oxygen (62,66) has been administered acutely after acclimatization has occurred, there is a persistent elevation of heart rate above sea level baseline values. The mechanism for this difference is uncertain, but it may well be that more prolonged administration of supplemental oxygen or sustained beta blockade at altitude would restore heart rate to the sea level value. Indeed, this explanation is likely to be true, as after a few days at sea level heart rate does return to normal (60,67). The decrease in plasma volume during the first few weeks of altitude exposure may unload both cardiopulmonary and arterial baroreceptors such that sympathetic activity remains above baseline even in the absence of chemoreflex activation. Moreover, a heightened chemoreflex associated with acclimatization could well yield an increased basal firing rate even under normoxic conditions. Finally, such hypovolemia would also be expected to cause vagal withdrawal so that the tachycardia with sustained hypoxia may not be entirely dependent on sympathetic activation. Indeed, chronic sympathetic hyperactivity usually leads to a downregulation of cardiac beta receptors (68), reducing the tachycardia response to catecholamines. Such a downregulation has been well described in animals and humans after acclimatization to high altitude (69–73). Thus, the heart rate response to acute altitude exposure may be dominated by sympathetic influences, while with more prolonged hypoxia exposure, some attenuation of this response occurs. Effects of the parasympathetic nervous system, along with other reflex interactions, appear to have an increasingly important influence on the resting heart rate. Despite this chronic increase in resting heart rate, cardiac output is consistently depressed after acclimatization to high altitude, often to below baseline sea level values. Resting hemodynamic data from a large number of studies are summarized in Table 1. Although these results are consistent in studies performed at moderate high altitude with the subjects remaining at a given altitude over the course of acclimatization (Table 1), important differences are observed if hypoxia is incremental and more extreme, as seen in a typical climbing expedition. For example, in the studies performed on Pikes Peak (Table 2), although arterial oxygen content increased over the course of acclimatization, cardiac output actually continued to decrease so that conductive O2 transport was reduced compared to sea level and did not change with acclimatization (63,74). In contrast, in Operation Everest II (Table 3) (62,75), resting cardiac output actually increased above 7000 m but remained

50 5 8 4 8 4 12

3658 3800 4300 4350 3100 4300 4300

21–34 20–73 18–24 20–22 23–32 20–21 21–28

Age (yr) 10 21–28 21 10 10 14 21

imped. CO2 rb dye dilut. dye dilut. Fick dye dilut. dye dilut

Method

SV ⫺36* ⫺11* ⫺20* ⫺22* nc ⫺23* ⫺30*

CO ⫺26* ⫺4 ⫺8 ⫺21* ⫺7 ⫹28* ⫺17*

Responses (%)a

⫹18* ⫹9* ⫹18* ⫹15* nc ⫹31* ⫹11*

HR

59 78 127 126 80 128 63,74

Ref.

% changes in responses from sea level. rb ⫽ Rebreathing; dye dilut. ⫽ indocyanine green dye method; imped. ⫽ impedance; nc ⫽ no change; CO ⫽ cardiac output, SV ⫽ stroke volume, HR ⫽ heart rate. * p ⬍ 0.05 vs sea level.

a

n

Duration (days)

Effects of Sustained Hypoxia on Resting Cardiac Hemodynamics

Altitude (m)

Table 1

Cardiovascular System at High Altitude 245

246

Wolfel and Levine

Table 2 Resting Hemodynamics in Healthy Men at 4300 m—Pikes Peak, Colorado (n ⫽ 12) Heart rate (bpm) Cardiac output (L/min) Stroke volume (mL) Vo 2 (mL/min) a-v o 2 (vol%) CaO 2 (vol%) O2 delivery (L O2 /min) MAP (mmHg) SVR (dynes-sec-cm⫺5)

Sea level a

Acute hypoxia b

Sustained hypoxia c

72 ⫾ 5 6.6 ⫾ 0.6 91 ⫾ 6 290 ⫾ 17 4.4 ⫾ 0.5 18.4 ⫾ 0.3 1230 ⫾ 120 88 ⫾ 3 1176 ⫾ 116

80 ⫾ 4* 6.7 ⫾ 0.6 84 ⫾ 6 316 ⫾ 14 4.7 ⫾ 0.8 15.3 ⫾ 0.3* 1010 ⫾ 90* 86 ⫾ 3 1162 ⫾ 126

80 ⫾ 4* 5.5 ⫾ 0.6* 70 ⫾ 7*† 376 ⫾ 14* 6.8 ⫾ 0.8*† 18.3 ⫾ 0.4† 1015 ⫾ 96* 107 ⫾ 3*† 1772 ⫾ 204*†

a

Sea level studies performed in Palo Alto, California. Acute hypoxia data obtained within 4 hours of arrival at 4300 m, Pikes Peak. c Prolonged hypoxia studies performed at 21 days on Pikes Peak. MAP ⫽ mean arterial pressure; SVR ⫽ systemic vascular resistance. * p ⬍ 0.05 versus sea level; † p ⬍ 0.05 versus acute hypoxia. The increase in a-v o 2 with sustained hypoxia was related to the decrease in cardiac output (80%) and the increase in Vo 2 (20%). Source: Refs. 63, 74. b

constant for a given oxygen uptake during exercise. However, conductive O2 delivery was nevertheless decreased, similar to the Pikes Peak studies due to the progressive increase in altitude and superimposition of acute on chronic hypoxia. In both groups of studies, metabolic demand was defended by an increase in peripheral oxygen extraction, which appears to have resulted in a relative ‘‘sparing’’ effect on cardiac output and systemic blood flow. When diffusion gradients for O2 transfer

Table 3 Resting Hemodynamic Measurements with Progressive High Altitude— Operation Everest II

Heart rate (bpm) Cardiac output (L/min) Stroke volume (mL) CaO 2 (vol%) a-v o 2 (vol%) O2 delivery (mL O2 /min) Vo 2 (mL/min) MAP (mmHg) SVR (dynes-sec-cm⫺5)

Sea level

6100 m

7260 m

8848 m

64 ⫾ 4 6.3 ⫾ 1.2 107 ⫾ 8 17.9 ⫾ 1.2 5.7 ⫾ 0.5 1125 ⫾ 144 360 ⫾ 20 96 ⫾ 3 1219 ⫾ 200

86 ⫾ 6* 5.0 ⫾ 1.1 72 ⫾ 6* 15.7 ⫾ 2.0* 6.4 ⫾ 1.1 786 ⫾ 220* 306 ⫾ 25 96 ⫾ 3 1536 ⫾ 48*

95 ⫾ 6* 7.3 ⫾ 1.5 69 ⫾ 10* 13.6 ⫾ 1.7* 5.7 ⫾ 1.0 992 ⫾ 155 406 ⫾ 16 90 ⫾ 2 986 ⫾ 107*

99 ⫾ 6* 8.6 ⫾ 0.8 81 ⫾ 11* 11.8 ⫾ 1.9* 4.6 ⫾ 0.2 1018 ⫾ 152 386 ⫾ 17 96 ⫾ 9 893 ⫾ 300*

Simulated altitude where Pio 2 ⫽ 63 torr (6100 m); 49 torr (7620 m); 43 torr (8840 m). * p ⬍ 0.05 versus sea level. Source: Refs. 62, 106, 138.

Cardiovascular System at High Altitude

247

are reduced at very low Po 2, a more prolonged capillary transit time would allow more time for oxygen transfer. Thus, a higher cardiac output might trade off a higher conductive O2 delivery for a reduced diffusional oxygen transfer such that actual net O2 flux could actually be decreased. In this context, Wagner presented a model suggesting that even at maximal exercise after acclimatization to high altitude, when maximal cardiac output is markedly reduced compared to sea level (see below), if one could further increase cardiac output, it would not lead to an increase in oxygen uptake (76). At present, a mechanism for the sensing and effecting of such a matching of diffusion time to diffusion gradients is unknown. Stroke Volume

Since heart rate is consistently increased with sustained altitude exposure, the reduction in cardiac output must be mediated by a decrease in stroke volume, and such a decrease has been uniformly observed. With sustained hypoxic exposure, reductions in stroke volume have been reported at rest (Table 1), with upright tilt (77), and at all levels of submaximal and at maximal exercise (Tables 4,5). Reductions in stroke volume can be seen as early as a few hours after exposure to hypobaric hypoxia during submaximal exercise (8). However, consistent reductions at rest and at all levels of exercise including maximal exercise are usually seen only after 2 days at high altitude (78) with some continued decline over the first week, after which stroke volume stabilizes for a given altitude (79). With progressively increasing altitude there appears to be further reductions in stroke volume (62), although differences in techniques make precise determination of the effect of progressive prolonged hypoxia on stroke volume difficult. Since stroke volume is end-diastolic volume minus end-systolic volume, it will be reduced only if end-systolic volume is increased due to impaired contractile function, or if end-diastolic volume is reduced due to impaired left ventricular filling. Each of these possibilities will be considered in turn. Cardiac Contractile (Systolic) Function

Early studies evaluating the hemodynamic responses to sustained hypoxia suggested that decreased cardiac contractile function contributed to the reductions in stroke volume (80–82). However, the measures used in these studies were extremely load dependent and misleading in the setting of sustained hypoxia associated with a prominent reduction in plasma volume and LV end-diastolic volume. More recently, Fowles and Hultgren normalized LV ejection indices to left ventricular end-diastolic dimensions and confirmed that contractile function was actually enhanced rather than impaired for a given preload (83), similar to observations made after diuretic treatment (84). The most convincing evidence that cardiac function is well preserved during sustained hypoxia comes from the Operation Everest II project. In this study, twodimensional echocardiograms were performed at progressively greater simulated altitudes up to and including that equivalent to the summit of Mount Everest, both

5 8 8 4 4 6 12 4 6 6 6

3800 4300 3100 4300 4350 3100 4300 5800 6100 7620 8840

3–4 wks 3 wks 10 days 2 wks 10 days 2–3 wks 3 wks 2–3 mos 3–4 wks 1–2 wks 20 min

Duration CO2 rb dye dilut. Fick dye dilut. dye dilut. N2O dye dilut. C2 H2 rb thermo/Fick thermo/Fick thermo/Fick

Method 1.88 1.70 1.60 1.57 1.43 1.62 1.77 1.48 1.49 1.45 1.18

(23.3) (23.0) (20.5) (20.3) (19.6) (21.4) (24.7) (23.2) (19.8) (19.1) (15.5)

SV ⫺14* ⫺16* ⫺11* ⫺15* ⫺32* ⫺8 ⫺26* ⫺26* ⫺28* ⫺30* ⫺26*

CO ⫺7* nc ⫺15* nc ⫺23* nc ⫺19* nc ⫺13 ⫺7 ⫹8

Responses (%) a

⫹9* ⫹6* ⫹4 ⫹8* ⫹24* ⫹13* ⫹8* ⫹14* ⫹17* ⫹16* ⫹25*

HR

78 127 80 127 126 139,140 63,74 132 62 62 62

Ref.

% changes in responses from sea level; changes relate to same absolute oxygen uptake at sea level and altitude. rb ⫽ Rebreathing; dye dilut. ⫽ indocyanine green dye method; C2H2 ⫽ acetylene; thermo ⫽ thermodilution; nc ⫽ no change; CO ⫽ cardiac output, SV ⫽ stroke volume, HR ⫽ heart rate. * p ⬍ 0.05 vs. sea level.

a

n

Oxygen uptake at altitude, L/min (mL/kg/min)

Sustained Hypoxia—Submaximal Exercise Hemodynamic Changes

Altitude (m)

Table 4

248 Wolfel and Levine

8 4 4 6 4 6 6 6

4300 4300 4350 5000 5800 6100 7620 8848 d

3 wks 2 wks 10 days 10 wks 2–3 mos 3–4 wks 1–2 wks 20 min

Duration dye dilut. dye dilut. dye dilut. CO2 rbc C2H2 rb thermo/Fick thermo/Fick thermo/Fick

b

Method 2.42 (32.7) 2.61 (33.9) 2.41 (33.0) not available 2.15 (33.1) 2.10 (27.8) 1.45 (19.2) 1.12 (14.8)

SV ⫹14* ⫺19* ⫺24* ⫺24* ⫺5 ⫺4 ⫺30* ⫺14†

CO ⫹4 ⫺22* ⫺29* ⫺23* ⫺29* ⫺16* ⫺31* ⫺30*

Responses (%) a

⫺4 ⫺3 ⫺7* nc ⫺25* ⫺14* ⫺23* ⫺26*

HR

127 128 126 141 132 62 62 62

Ref.

b

% changes from sea level. Potential concerns about methodology of dye dilution (see Ref. 128). c Studies done immediately after descent under more normoxic conditions. d Only 3 subjects studied at simulated 8848 m altitude. rb ⫽ Rebreathing, dye dilut. ⫽ indocyanine green dye method; thermo ⫽ thermodilution; C2H2 ⫽ acetylene; nc ⫽ no change; CO ⫽ cardiac output; SV ⫽ stroke volume; HR ⫽ heart rate. * p ⬍ 0.05 vs sea level.

a

n

Peak oxygen uptake at altitude, L/min (mL/kg/min)

Sustained Hypoxia—Maximal Exercise Hemodynamic Changes

Altitude (m)

Table 5

Cardiovascular System at High Altitude 249

250

Wolfel and Levine

at rest and during submaximal exercise (85). As noted above and in Table 3, stroke volume was decreased with chronic altitude exposure, even at 8848 m. However, at each altitude, end-systolic volume was smaller than at sea level. Virtually all indices of left ventricular systolic function, including left ventricular ejection fraction, mean normalized systolic ejection rate, and peak systolic pressure/end-systolic volume, were either unchanged or improved even at the most extreme altitudes. Invasive measurements of cardiac filling pressure allowed the construction of Starling ventricular function curves and demonstrated that they were superimposable among sea level and altitude studies (62). In other words, stroke volume was always appropriate for the LV filling pressure, which was reduced at altitude. There was no increase in stroke volume with acute administration of 100% O2, arguing against any hypoxic depression of contractile function. These data confirm that in normal individuals LV contractile function is not diminished during sustained exposure to high altitude, and thus an increase in LV end-systolic volume cannot explain the reduction in stroke volume observed at altitude, even under the most extreme hypoxic conditions. Although it is now clear that myocardial contractile function is preserved in the setting of prolonged hypoxia, the mechanisms responsible for myocardial protection from severe hypoxia are not well understood. Various candidates for the development of a ‘‘hypoxia-tolerant’’ myocardium include heightened sympathetic activity (86), a shift in myocardial metabolism to glucose and lactate, with increased ATP yield per molecule of oxygen (87), transcriptional changes in mitochondrial and nuclear genes (88,89), and the induction of various cellular regulatory factors that preserve mitochondrial function such as hypoxia-inducible factor (89), nuclear factor (NF)-κB (90) and heat shock proteins (91,92), which appear to be activated by sustained hypoxia in various animal and tissue culture models. In addition, sustained hypoxia exposure may result in downregulation of some cellular functions, such as the inducible nitric oxide synthase (iNOS) transcriptional response to circulating cytokines, which could provide protection against myocardial and vascular endothelial damage (90). Although none of these mechanisms have been definitely shown to occur in humans at high altitude, these studies present compelling data that changes on the cellular level do occur with prolonged exposure to hypoxia. A combination of these molecular and cellular adaptations could result in a more energyefficient, normally functioning contractile state of the myocardium in the setting of prolonged hypoxia. Some of these genetic adaptations may also explain the tolerance of high-altitude populations as demonstrated by nearly normoxic cardiovascular function despite life-long exposure to significant hypoxia. The reduction in stroke volume with sustained altitude exposure therefore must be due to a reduction in LV end-diastolic volume (LVEDV), which has been universally observed in echocardiographic studies at altitude (79,83,85,93). For any given level of left ventricular contractility or afterload, stroke volume is determined to a large extent by left ventricular filling as a function of the Frank-Starling mechanism (94). Left ventricular end-diastolic pressure (LVEDP) is determined by

Cardiovascular System at High Altitude

251

LVEDV, and this relationship is influenced by cardiac chamber stiffness. Because of the curvilinear nature of the left ventricular pressure-volume relationship, cardiac stiffness is dynamic with the instantaneous stiffness, defined as dP/dV, dependent on the specific value of left ventricular volume (95). Cardiac stiffness, LVEDV and consequently SV may therefore be altered by either (1) shifting up (hydrated state) or down (dehydrated state) on any individual pressure-volume curve or (2) changing the underlying pressure-volume relationship. These changes may occur with either an alteration in intravascular volume or a sudden change in extracardiac influences (pericardial or pulmonary mechanical restraint) or via a specific cardiac adaptation such as bedrest deconditioning (96). These concepts are illustrated in Figure 4. One obvious mechanism for the early reduction in both filling pressure (62) and stroke volume with sustained altitude exposure is a reduction in plasma volume. Numerous studies have documented this adaptation, which ranges from 20 to 30% (97,98) (see also Chapter 15). Plasma volume is reduced within the first few hours

Figure 4 Left ventricular pressure-volume relationships. Point 1 on curve represents the normal supine position on the pressure-volume (left) and Starling (right) curves. Acute dehydration would cause a shift to point 2, to a less stiff region (smaller slope) of the P-V curve and a large fall in stroke volume (SV) during orthostasis. In contrast, hyperhydration would shift to point 3, where changes in SV would be modest despite large changes in LVEDP. For a given left ventricular volume, changes in the underlying P-V relationship could result in an increase in cardiac compliance (curve c) or decrease in compliance (curve b) due to intrinsic or extrinsic factors.

252

Wolfel and Levine

to days at altitude and is responsible for acute hemoconcentration (99). When this reduction in plasma volume is prevented with CO2 breathing (100), the reduction in stroke volume is also prevented. Over time, a gradual increase in red cell mass occurs, such that blood volume gradually increases. For example, in most of the studies from Pikes Peak there was approximately a 15–20% decrease in plasma volume throughout the period of residence and a smaller (2–5%) decrease in blood volume after 3 weeks at 4300 m due to the progressive increase in red cell mass (74,101). Figure 5 shows the relationship between the changes in blood and plasma volume and the reductions in stroke volume and cardiac output with sustained hypoxia at 4300 m. In these studies, the changes in blood volume accounted for some of the decrease in stroke volume after acclimatization, particularly during submaximal exercise. However, other mechanisms are probably also involved since the reduction in stroke volume at altitude is seen early in the acclimatization period and does not appear to be totally ameliorated by the gradual increase in blood volume over time. Alexander et al. infused 500 mL of dextran into two subjects acclimatized to 3100 m (80). In these two subjects, pulmonary capillary wedge pressure increased modestly, suggesting increased cardiac filling, and stroke volume was increased but not to sea level values. Unfortunately, the small numbers of subjects studied and the rather small volume increases and attendant pressure changes preclude definitive conclusions from these data. Moreover, as depicted in Figure 4, a rise in filling pressure does not necessarily equate with a rise in LVEDV, particularly if ventricular interdependence and pericardial constraints are operative (102). However, a recent study performed in young men in a hypobaric chamber (Operation COMEX) demonstrated that intravenous volume infusion improved maximal oxygen consumption at simulated altitudes of 7000 m, especially in those subjects with the lowest plasma volume from prolonged hypoxia exposure (103). Although no determinations of stroke volume were obtained with these infusions, the increase in exercise capacity was assumed to be related to increases in cardiac output from an increased intravascular volume, suggesting that both convection and diffusion limitations may be operative in the reduced physical performance at extreme altitudes. A second possible explanation for the reduction in stroke volume may be that it occurs as a consequence of the increase in heart rate. Chronic increases in heart rate, for example, in models using chronic pacing, are always associated with decreases in stroke volume, even before LV function deteriorates (104). This result may be due to redistribution of blood volume to the venous capacitance associated

Figure 5 Simple linear regression of changes in blood volume and plasma volume with changes in cardiac output and stroke volume during submaximal exercise between sea level and 18–20 days at 4300 m. Data taken from combined studies in 12 men on Pikes Peak in 1988 and 1991 (74,101). Changes in both cardiac output and stroke volume between sea level and 4300 m are correlated with changes in both plasma volume and total blood volume.

Cardiovascular System at High Altitude

253

254

Wolfel and Levine

with transiently increased cardiac output or as a function of increased peripheral vascular resistance. Supportive evidence for this concept comes from the observation that when subjects are beta-blocked prior to ascent to altitude (63), the increase in heart rate is prevented and the decrease in stroke volume is substantially reduced. However, this explanation is not sufficient, as stroke volume is reduced with sustained altitude exposure even at maximal exercise, when maximal heart rate is lower than at sea level. Finally, the reduced left ventricular filling at altitude may be due to impaired right ventricular function. Because the right heart is not well suited to geometric modeling, volume data similar to that of the LV are not available for the right ventricle at altitude. The role of the right ventricle may be particularly important because hypoxia causes pulmonary vasoconstriction and sustained pulmonary hypertension at altitude (105,106) (see also Chapter 11), resulting in numerous electrocardiographic reports of right ventricular strain or overload (60,67,107). Because the right ventricle does not have the contractile reserve available to the left ventricle, it may not be able to perform optimally against acute or subacute increases in afterload (108). Anecdotal reports of acute right ventricular dilatation have been made in climbers with high-altitude pulmonary edema, and RV dilatation has been documented by echocardiography even in well subjects at extreme altitude in OE II (85) and more recently in a group of Japanese climbers acclimatized in the field to altitudes above 6000 m (93). More dramatic echo evidence of RV failure has been obtained in some ultra-endurance athletes competing in a high-altitude endurance race (109). In fact, right ventricular function is probably particularly important at maximal exercise, when ventricular filling time is shortest. In a recent study (110), patients born without a right ventricle who had undergone the Fontan operation (passive conduit directly from the systemic veins into the pulmonary artery) had completely normal LV stroke volume and cardiac output response at 3100 m, both at rest and submaximal exercise, but had a depression in stroke volume at maximal exercise, indicating that right ventricular dysfunction could contribute to the reductions in stroke volume during exercise with sustained hypoxia. The potential significance of the right ventricle for supporting LV filling has recently been highlighted in some patients with congestive heart failure (111). In this study, patients with heart failure and depressed LV stroke volume were subjected to lower body negative pressure with pooling of blood in the lower part of the body. Contrary to what occurs in normal individuals, these patients had an increase rather than a decrease in cardiac output, associated with an increase in LVEDV and SV. Thus, in some patients RV dilation may impair LV filling through ventricular interdependence and pericardial constraint with shift of the interventricular septum from right to left. Although such an effect is theoretically possible at altitude, it is unlikely to play a major role in most individuals because LV filling pressure when measured has been uniformly low, providing evidence against pericardial constraint. However, much more work remains to be done to examine right ventricular function at altitude.

Cardiovascular System at High Altitude

255

Diastolic Function

Although LV systolic function is normal with sustained hypoxia, abnormalities of LV diastolic function could also contribute to the impaired ventricular filling. The filling of the heart during diastole is a dynamic, complex process involving several distinct hemodynamic phases and interactions among a number of variables. These include left ventricular chamber stiffness, dependent on both intrinsic myocardial and extrinsic factors such as pericardial and pulmonary mechanical constraint, but also active myocardial relaxation, atrioventricular coupling, the dynamic establishment of atrioventricular and intraventricular pressure gradients, and extracardiac factors. Abnormalities in intracellular calcium transport, secondary to hypoxia, could result in abnormal left ventricular relaxation or compliance. With brief exposure to varying degrees of acute hypoxia induced by breathing hypoxic gas mixtures, 10 young men were found to have abnormalities in left and right ventricular diastolic function on cardiac echo-Doppler studies (112). However, traditional Doppler measures of diastolic function are extremely load dependent (113), and it is difficult to sort out frank changes in relaxation and/or compliance from the well-described hemodynamics of acute altitude exposure. Probably the strongest evidence against abnormalities of diastolic function at altitude come from the low pulmonary capillary wedge pressures measured in OEII (62). However, it is possible that abnormalities of ventricular compliance or distensibility may be masked by the reduced ventricular volumes observed in subjects during sustained hypoxia (compare point 2 with point 5 in Fig. 4). At least two plausible mechanisms exist that could lead to reduced distensibility after sustained altitude exposure. The first is a reduction in physical activity with associated cardiac atrophy (96). Two weeks of bedrest deconditioning in normal healthy subjects results in both atrophy of the heart and reduced LV distensibility, associated with a reduction in LVEDV and SV at any given filling pressure. Although complete pressure/volume curves have not been obtained after sustained altitude exposure, examination of the reported PCW pressure and LVEDV at rest from the data in OEII (62,85) suggests a shift upward and to the left, i.e., an increase in pressure for any given volume, consistent with decreased ventricular distensibility. Certainly the confinement associated with chamber or mountain hut studies, as well as the hypoxia-induced decrease in aerobic power, could conceivably lead to cardiovascular deconditioning, particularly if the subjects were very fit and active prior to the studies. A second mechanism that could explain this decreased distensibility is increased pulmonary mechanical constraint associated with expansion of lung volumes. Data in dogs after pneumonectomy show dramatic increases in cardiac distensibility, and recent data from space flight suggest that removal of pulmonary mechanical constraint can also lead to a dramatic increase in distensibility in humans (114,115). Although there are no data available on the role of enhanced lung volume on diminishing cardiac distensibility, this mechanism could play some role because

256

Wolfel and Levine

of the typically large ventilatory adaptations seen with acclimatization to high altitude. B. Systemic Circulatory Changes—Sustained Hypoxia General Systemic Vascular Responses

The systemic circulation involves the entire circulatory system, including the heart and its interaction with regional vascular beds, which determines overall circulatory function. Although the local peripheral circulation appears to be precisely regulated by arterial oxygen content, the systemic circulation seems to behave differently. One of the key questions regarding the cardiovascular response to sustained hypoxia has been whether sympathetic activation persists or diminishes over the course of prolonged exposure to high altitude. Indirect measurements of blood and urinary catecholamines over 3 weeks on Pikes Peak suggest that circulating norepinephrine actually increases rather than decreases over time at altitude (15,16,116). Direct measurements of muscle sympathetic nerve activity (MSNA) have confirmed that with short-term (2–4 days) exposure to altitudes up to 4500 m, MSNA is slightly greater than with acute hypoxic exposure in healthy subjects (34). In contrast, MSNA is markedly augmented in climbers who are susceptible to high-altitude pulmonary edema, both acutely and even more so with sustained altitude exposure, forming a potential mechanistic link between the magnitude of sympathetic activation and the illnesses of high altitude (34). Most recently, MSNA was recorded from a relatively large number of well-acclimatized subjects who had been at altitudes above 5200 m for more than one month as part of the 1998 Danish Medical Research Expedition to the Andes. In contrast to the opinion that sympathetic activity would decrease over time, MSNA in these well-acclimatized subjects was dramatically elevated (117). These data suggest that the well-described increase in peripheral chemosensitivity, which has been documented for the ventilatory response to hypoxia, may lead not only to a progressive increase in ventilation during acclimatization (see Chapter 7) but also to a similar progressive increase in sympathetic nerve activity despite improvements in arterial oxygen content. This hypothesis needs to be tested by further experiments in subjects during sustained hypoxia. Regional Vascular Responses

Since oxygen delivery (blood flow ⫻ Cao 2 ) seems to be precisely regulated at a local level with acute hypoxia, it is reasonable to assume that the changes in peripheral blood flow with sustained hypoxia would also mirror the changes in Cao 2 , and in fact that is the case. The largest body of work examining the effect of sustained hypoxia on peripheral blood flow has been performed at 4300 m on the summit of Pikes Peak. Leg blood flow was measured using the thermodilution technique under both acute and more than 3 weeks at rest and during exercise (63,74,118). In general, these studies used a common paradigm: baseline measurements at sea level, followed by acute

Cardiovascular System at High Altitude

257

exposure to 4300 m, and then repeat measurements after 2–3 weeks at 4300 m. Based on these experiments, the following conclusions can be drawn: (1) leg blood flow was consistently elevated with acute hypoxia at the same metabolic demand (same oxygen uptake) compared to sea level, thereby maintaining oxygen delivery; (2) with acclimatization, the increase in arterial oxygen content was associated with a decrease in leg blood flow and a concomitant increase in oxygen extraction at the same metabolic demand; and (3) the addition of supplemental oxygen to acclimatized subjects resulted in a further reduction in leg blood flow, confirming the tight coupling of leg blood flow to oxygen content for any given oxygen requirement. Similar findings have been reported for the nonexercising forearm of men at the same high altitude (119), i.e., a progressive increase in forearm vascular resistance over 7 days of acclimatization to 4300 m (Fig. 6). Moreover, regardless of whether the upper or lower extremity was studied, and independent of metabolic activity, the observed peripheral vasoconstriction during acclimatization was related to elevated urinary and arterial norepinephrine levels, indicating that enhanced sympathetic stimulation was likely playing an important role (16,119). Thus, functional sympatholysis diminished with sustained hypoxia as acclimatization restored arterial O2 content and sympathetic activity appeared to increase. Intriguingly, hypocapnia might be playing an important role in mediating changes in limb blood flow and vascular resistance, as forearm flow and resistance were unaltered at altitude under conditions of normocapnic hypoxia (120). These data provide support for a potentially significant interaction between ventilatory and cardiovascular chemoreflexes. Finally, upon return to sea level after a sojourn at altitude, climbers who had been exposed to altitudes ranging from 3500 to 8400 m were found to have a 26– 34% reduction in muscle blood flow, determined by 133Xe clearance, during submaximal exercise (121), associated with persistent elevations in hemoglobin concentration and arterial oxygen content. Thus, even with restoration of normal Pao 2 , the surfeit of oxygen availability continues to regulate oxygen delivery via alterations in limb blood flow. Thus, the peripheral circulation develops adaptive responses with sustained hypoxia that maintain tissue oxygen uptake and substrate delivery. Both at rest and during exercise, in a small (forearm) or a large (legs) muscle bed, vasoconstriction appears to be a common mechanism during acclimatization as arterial oxygen content increases and a persistent sympathetic activation is unmasked by withdrawal of functional sympatholysis. Coronary Circulation

The coronary circulation has a number of features that distinguish it from the peripheral circulation and may influence the adaptation to hypoxia. First of all, coronary blood flow is regulated by myocardial oxygen demand, which in turn is determined by three main factors: heart rate, contractility, and wall stress. Wall stress itself is determined both immediately prior to the onset of contraction (preload), and after the onset of contraction (afterload). Whether myocardial oxygen demand is increased,

Figure 6 Sustained hypoxia at 4300 m results in a reduction in forearm compliance (upper left) and blood flow (upper right) along with increases in mean arterial pressure (MAP) (lower left) and forearm vascular resistance (lower right). Subjects were residents of 1600 m (Denver, CO). (Figures redrawn from original data in Ref. 119.)

258 Wolfel and Levine

Cardiovascular System at High Altitude

259

decreased, or unchanged during sustained hypoxia is not clear and depends on a complex interplay among these factors, some of which are increased (heart rate, contractility, blood pressure, afterload) while others are maintained at a lower level (preload). In addition, oxygen extraction is extremely high in the myocardium even at rest, and thus most of the adaptive range of myocardial metabolic activity must be met by alterations in coronary blood flow. For normal young individuals without coronary artery disease, this vasodilator reserve is adequate to meet myocardial oxygen demands despite extreme hypoxia, even under conditions of maximal exercise. Thus in both Operation Everest II (67) and the American Medical Research Expedition to Mount Everest (AMREE) (60), there were no symptoms or ECG evidence of ischemia in subjects exercising maximally (Vo 2 ⫽ 1.17 L/min) at 8848 m despite extraordinary metabolic derangements including an Sao 2 as low as 49% and pH of 7.52 (67). Similar to the peripheral circulation, with acute hypoxia coronary blood flow increases in proportion to the reduction in arterial oxygen content (122). However, with sustained hypoxia the situation is quite different. Studies performed in both high-altitude residents and recently acclimatized lowlanders suggest that coronary blood flow is decreased compared to sea level. For example, high-altitude Andean natives demonstrated a progressive decline in coronary blood flow at rest with increasing altitude, compared to sea level natives, reaching 30% at the highest altitudes (123). These findings were related to a lower heart rate, which contributed to a decreased myocardial O2 demand as well as to an increased red cell mass, which allowed a relatively greater coronary arteriovenous O2 difference. However, similar reductions in coronary blood flow have also been observed in newly acclimatized young men after only 10 days at 3100 m (Leadville, CO) (124). Like high-altitude natives, these recently acclimatized lowlanders also had a 30% decrease in coronary blood flow compared to sea level values, though without a concomitant reduction in heart rate and myocardial O2 consumption. Although coronary venous Po 2 was appropriately low (though not less at 3100 m that at sea level), coronary O2 extraction was increased as reflected by a decrease in coronary sinus O2 content and saturation. Thus, it appeared that coronary blood flow was regulated to maintain a constant myocardial O2 tension at high altitude, similar to the findings in the peripheral circulation during submaximal exercise when skeletal muscle metabolic rate is similarly high. Systemic Blood Pressure

There has been some controversy as to how the myriad of cardiovascular responses to sustained hypoxia affects systemic blood pressure. Most of the controversy, however, probably relates to the timing of blood pressure measurements. With acute exposure to altitude, as described above, blood pressure usually falls due to a decrease in systemic vascular resistance. However, with acclimatization stroke volume and cardiac output fall, even in the face of persistent tachycardia, and peripheral vascular resistance gradually rises as ventilatory and hematological acclimatization

260

Wolfel and Levine

restore arterial oxygen content towards normal. This acclimatization response may occur relatively rapidly, particularly at moderate altitudes of 3000–4000 m. Thus, persistent sympathetic activation due to hypoxia is no longer offset by functional sympatholysis and locally mediated vasodilation, and blood pressure rises over time (42). Increases in systemic arterial pressure and vascular resistance have been noted in numerous studies of high-altitude acclimatization. For example, native lowlanders who have spent 2–4 weeks at altitudes varying from 3000–5000 m have been shown consistently to have elevations in arterial pressure at rest and during submaximal and maximal exercise when compared to sea level values (63,65,74,78,118,125– 128). Conversely, high-altitude residents at 3100 m (Leadville, CO) have been shown to have lower systolic and diastolic blood pressures at rest and during submaximal exercise after 10 days at sea level (82). Because cardiac output is decreased rather than increased with sustained altitude exposure, studies using β-blockers prior to ascent have not observed a reduction in the systemic blood pressure elevation associated with chronic hypoxia (57,63), thus raising the importance of α-adrenergic vasoconstriction in this process. In support of this concept, healthy young women treated with the α-adrenergic blocker prazosin were found to have a decline in both diastolic and mean arterial pressure during exercise after 48 hours of hypoxia in a hypobaric chamber simulating an altitude of 4300 m when compared to the unblocked state (129). α-Adrenergic blockade also blunted the systemic pressor response to the inhalation of hypoxic gas (130). However, administration of prazosin to women prior to a 12-day sojourn at 4300 m blunted but did not abolish the rise in systemic blood pressure observed with ambulatory monitoring (131). Thus, sympathetic blockade with either α- or β-blockers was not sufficient to prevent the systemic pressor response at moderate high altitude. Moreover, few data exist regarding the relative importance of other neurohormones such as angiotensin, aldosterone, or vasopressin, all of which may be involved in some degree with raising blood pressure during sustained high-altitude exposure. These mediators may be particularly important in sustaining the pressor response to sustained hypoxia, as acute administration of 35% O2 after acclimatization to 4300 m did not return blood pressure to sea level values (118). Although several studies at altitudes between 3000 and 5000 m have reported increases in systemic arterial pressure, no increases in blood pressure or systemic vascular resistance have been observed at more extreme altitudes (106,132). At these extreme altitudes, the degree of hypoxia is profound and the acclimatization process cannot restore arterial oxygen content to normal as it can at lower altitudes due to the profound stimulus to peripheral vasodilation with reduction in peripheral resistance. Few data exist regarding blood pressure on recovery from altitude. A study of 47 men who were taken from sea level to 3658 m for a 10-day sojourn reported a progressive rise in resting diastolic blood pressure that was associated with an increase in total urinary catecholamines (125). On return to sea level, diastolic blood pressure remained elevated for several days while urinary catecholamines returned

Cardiovascular System at High Altitude

261

to normoxic levels within the first 24 hours. Insufficient data were available to provide a satisfying explanation and more research is needed in this area. Finally, elevations in systemic arterial pressure do not seem to persist in lowlanders who remain at altitude for several months or longer (133). Sympathetic activity may diminish with continued residence at the same altitude. For example, urinary catecholamine levels have been reported to return to sea level values in subjects who remained at altitudes greater than 3000 m for more than 90 days (134). However, no direct data regarding sympathetic nerve activity are available in sea level natives over such a prolonged period of time. In addition, highlanders who were native to this altitude had similar urinary catecholamines levels as lowlanders at sea level. Moreover, studies in high-altitude populations have shown that systemic hypertension is uncommon and blood pressures in highlanders tend to be lower than in sea level natives (135). Thus, elevations in systemic arterial pressure appear to be part of the early (i.e., days to weeks) acclimatization process to high altitude associated with heightened sympathetic stimulation but may gradually abate over time as the sea level native becomes more like a high-altitude native. In summary, although the mechanisms are not entirely clear, there is general agreement that over a period of days to weeks, the cardiovascular adaptations to sustained hypoxia include persistent and probably augmented sympathetic activation, resulting in tachycardia with a reduced stroke volume that together is not sufficient to maintain cardiac output at sea level values. Coronary and peripheral oxygen transport is maintained via increases in Cao 2, and regional blood flow is precisely regulated to support delivery of oxygen and substrate to the tissues. With acclimatization, a gradual reduction in coronary and peripheral blood flow is associated with sympathetic activation that is now unopposed by regional vasodilation, leading to increases in total peripheral resistance and systemic hypertension. Key Unanswered Questions—Sustained Hypoxia

Despite all the studies cited, the mechanism and feedback loops for the reduction in cardiac output between acute and sustained hypoxia have not been elucidated. Is it all a consequence of changes in preload, or are other factors involved, such as the central nervous system? Are there cellular mechanisms that govern blood flow and what variable is sensed? What is the physiological purpose for the apparent matching of tissue oxygen supply to demand? Which comes first—is heart rate increased from chemoreceptor mechanisms and stroke volume passively reduced, or is stroke volume reduced via neurohumoral mechanisms and heart rate increased reflexively? Is the systemic vasoconstriction with sustained hypoxia an adaptive or maladaptive response? Is the elevation in systemic blood pressure with sustained hypoxia of short duration completely governed by enhanced sympathetic activity, or are there local vascular factors that play a major role?

262

Wolfel and Levine IV. Cardiovascular Function During Exercise

As compared to sea level responses, submaximal exercise during acute hypoxia is associated with an increased heart rate and no decrease in stroke volume and therefore an increased cardiac output for any given absolute value of oxygen uptake. The same response has been shown with leg blood flow during acute hypoxia. However, with sustained hypoxia, submaximal exercise is associated with an increased heart rate but a decrease in stroke volume for any given absolute level of oxygen uptake. The usual sea level relationship between cardiac output and oxygen uptake is preserved in some but not all studies at more moderate altitudes of 2500–4500 m, while other studies report a reduced cardiac output for a given oxygen uptake. With more extreme hypoxia, altitudes greater than 5000 m, the cardiac output and oxygen uptake relationship is similar to sea level. Maximal exercise capacity is reduced to a similar degree with both acute or sustained hypoxia, although the mechanisms for the reduction differ depending on the duration of hypoxia. With acute hypoxia of short duration, heart rate, stroke volume, and cardiac output are generally maintained at sea level values with reductions in arterial oxygen content being responsible for the decrease in exercise capacity. With sustained hypoxia, reduction in maximal exercise capacity is associated with a decreased maximal heart rate and cardiac output that is as yet not clearly explained. Whether the reduction in cardiac output at maximal exercise with acclimatization is responsible for the decrease in maximal oxygen uptake or is a consequence of it is one of the most intriguing questions in regard to human adaptation to high altitude. One of the key features of the cardiovascular system that ultimately determines work capacity of the organism is its remarkable range of responsiveness: oxygen uptake can increase 10- to 20-fold from rest to maximal exercise depending on the state of physical conditioning. Cardiac output must increase by nearly an order of magnitude in order to supply skeletal muscle with substrate to support this metabolic demand. Adaptations that are adequate to ensure substrate delivery at rest may be inadequate during high levels of exercise. Many investigators have studied the acute and sustained effects of hypoxia on submaximal and maximal exercise performance, and a detailed review of exercise physiology in hypoxia is presented in Chapter 20. However, a number of important points related to cardiovascular function during exercise deserve emphasis and comment. V.

Acute Hypoxia—Submaximal Exercise

When compared to normoxic exercise, with acute hypoxia, submaximal exercise is associated with an increase in muscle blood flow, heart rate, and cardiac output for any given absolute work rate, thus enabling the maintenance of oxygen uptake at sea level values in the face of reduced arterial oxygen content (60,62,63,74,118) (Figs. 7,8). Exposure to hypoxia of short duration (hours) is not associated with any

Figure 7 Cardiac hemodynamic responses to acute hypoxia at simulated altitudes of 3048 m (523 torr) and 4572 m (429 torr) (9). The Vo 2workload relationship is maintained as at sea level, while both heart rate and cardiac output increases in a progressive fashion with greater hypoxia at a given workload. Stroke volume remains at or above sea level values. Although maximal Vo 2 is lower than at sea level; both maximal heart rate and cardiac output are unchanged compared to sea level with less than one hour of hypoxia exposure.

Cardiovascular System at High Altitude 263

Figure 8 Leg blood flow is greater for a given workload with acute hypobaric hypoxia compared to normoxia. During acute hypoxia, the relationship between leg Vo 2 and workload is similar to that at sea level. (Figures drawn from data from Ref. 118.)

264 Wolfel and Levine

Cardiovascular System at High Altitude

265

reduction in stroke volume. The magnitude of the increase in heart rate is proportional to the severity of hypoxia, and the changes in heart rate during submaximal exercise are similar to those of resting heart rate. One simple explanation for this increase in heart rate during submaximal exercise is that the same absolute work rate in hypoxia represents a greater relative work rate (greater percentage of maximal oxygen uptake) compared with normoxia, and relative oxygen uptake appears to govern the cardiovascular responses to exercise (136). Yet when workload is adjusted for % maximal oxygen uptake, heart rate is still somewhat greater in hypoxia than in normoxia (8). Increases in sympathetic stimulation, especially the elevation in arterial epinephrine, may be partly responsible for this increase in heart rate (15). However, studies in both humans and animals with β-adrenergic blockade during acute hypoxia continue to show some rise in cardiac output and heart rate with no effect on the vasodilator effect of hypoxia (12). Thus, peripheral factors involved in ‘‘functional sympatholysis’’ may also play a major role in the augmentation of cardiac output during acute hypoxia by reducing peripheral resistance and afterload. Despite the increases in cardiac output, conductive oxygen delivery is still somewhat reduced compared to sea level (8,9), notwithstanding the existence of substantial systemic and regional blood flow reserve during submaximal exercise. Therefore, oxygen uptake is preserved additionally by a decrease in venous Po 2, thereby increasing peripheral oxygen extraction (76). Intracardiac filling pressures during submaximal exercise in the setting of acute hypoxia are no different than with normoxia except for the elevation in pulmonary arterial pressures (9), thus excluding systolic or diastolic dysfunction as a cause of this relative reduction in oxygen delivery. Thus, for reasons that are not entirely clear, activation of the cardiovascular system during submaximal exercise is restrained during acute hypoxia, and oxygen utilization is preserved by a combination of both increased blood flow and increased peripheral extraction.

VI. Acute Hypoxia—Maximal Exercise All studies performed within an hour after exposure to hypoxia have shown that maximal heart rate and cardiac output are unchanged from sea level, although this concept has been challenged recently, at least with respect to maximal heart rate (137). Maximal oxygen uptake is decreased because of an obligatory reduction in maximal convective oxygen transport (8,9). Similar to the situation during submaximal exercise, oxygen delivery is decreased compared to sea level. However, at maximal exercise tissue oxygen extraction is unable to compensate for the reduced convective oxygen transport and maximal oxygen uptake falls (8,9). Intracardiac filling pressures remain at or below sea level values with a hypoxia-related rise in pulmonary arterial pressures (9).

266

Wolfel and Levine VII. Sustained Hypoxia—Submaximal Exercise

The key difference between acute and sustained hypoxia during exercise is the apparent ‘‘cardiac sparing’’ (reduced heart rate, stroke volume, and cardiac output) that occurs due to adaptations in other organ systems such as increased ventilation, enhanced tissue oxygen extraction, and increases in hemoglobin concentration. Unlike with acute hypoxia, there is less cardiac output and limb blood flow at a given absolute work rate, and conductive oxygen delivery is supported to a greater extent by the increase in oxygen content in arterial blood. However, the degree of hypobaric hypoxia (PB), the duration of exposure, and the type of exercise, i.e., submaximal or maximal, influence the specific physiological changes in cardiovascular function during exercise over time at high altitude. As opposed to acute hypoxia, where cardiac output is elevated for a given oxygen uptake, with sustained hypoxia the cardiac output decreases so that over time at altitudes ⬎5000 m, the relationship between cardiac output and oxygen uptake is similar to that at sea level (61,62,66,106,132,138). In contrast, at more moderate altitudes (3100–4500 m), where there is both less direct hypoxic vasodilation and less sympathetic activation, the decrease in cardiac output with acclimatization may result in an output that is actually less than that at sea level for the same oxygen uptake (63,74,78,80,126) (Fig. 9), although the magnitude of reported decreases varies widely (127,128,139,140) (see Table 4). However, studies on climbers who have spent an extended period of time at high altitude have demonstrated consistent reductions in both stroke volume and cardiac output at several submaximal workloads even upon return to sea level, indicating that there is an alteration in the cardiac output–oxygen uptake relationship from prolonged exposure to hypoxia (141,142). Although heart rate remains elevated at the same absolute or relative submaximal work rate during sustained hypoxia exposure (63,75,143), stroke volume is universally lower than at sea level (Fig. 10) and is primarily responsible for the reduction in cardiac output during acclimatization. Similar to observations made at rest, pulmonary capillary wedge pressure is also decreased during submaximal exercise, indicating that reductions in cardiac filling or preload and not decreased contractility are responsible for the decrease in stroke volume. Thus, even with activation of the muscle pump during exercise, central blood volume remains reduced during sustained hypoxia resulting in a decreased stroke volume. One of the intriguing regulatory questions regarding the cardiovascular response to exercise after acclimatization to hypoxia is how the balance between systemic and local blood flow and oxygen extraction is determined. Although at least some of the reduction in cardiac output may be offset (if not induced) by an increased oxygen-carrying capacity of the blood, total oxygen delivery virtually always is reduced compared to sea level, and there is a greater dependence on peripheral extraction (66,132,138) (Fig. 11). This response occurs in spite of substantial heart rate and blood flow reserve, which could, if more substantially engaged, restore oxygen delivery to normal levels (63,74).

Cardiovascular System at High Altitude

267

Figure 9 Figures drawn from combined data of 12 male subjects in studies in 1988 and 1991 at 4300 m (63,74). Submaximal exercise was at the same absolute workload (100 W) which was 50% Vo 2max at sea level and 65% Vo 2max at both acute and chronic altitude. (a) Cardiac output and oxygen delivery decreases with prolonged hypoxia. (b) Heart rate increases while stroke volume falls. (c) Arterial oxygen content falls dramatically with acute hypoxia but recovers with acclimatization, while mixed venous oxygen content decreases with acute hypoxia and remains at the same level with sustained hypoxia. (d) Mean arterial pressure and vascular resistance decrease with acute hypoxia but both rise dramatically with prolonged hypoxia.

268

Wolfel and Levine

Cardiovascular System at High Altitude

269

In fact, with prolonged acclimatization at the same altitude, there appears to be a reduction over time in heart rate at the same level of submaximal exercise and submaximal Vo 2, which contributes to the reduction in cardiac output (63,74, 128,143). Administration of supplemental O2 during submaximal exercise in acclimatized subjects residing at high altitude has resulted in a further reduction in heart rate, suggesting that oxygen content of the blood is an important determinant of the heart rate response (62,66). Augmented vagal tone does not appear to be responsible for this reduction in exercise heart rate during acclimatization as heart rate increases with atropine to the same degree, both at sea level and after acclimatization to 4300 m (99). However, acute β blockade while at altitude does reduce submaximal heart rate to near sea level values (63,65), emphasizing the importance of hypoxia-induced sympathetic activation in this adaptation. The reductions in heart rate with acclimatization are therefore probably related to a decreased responsiveness to sympathetic stimulation as a result of downregulation of cardiac β-adrenergic receptors (69,70). The mechanism by which arterial oxygen content regulates both central and peripheral blood flow during exercise remains unknown. Studies in animals suggest that it is the active oxygen-carrying capacity of red blood cells and not the hematocrit that influences changes in blood flow and vascular resistance (144–146). More investigation is required to determine the sensing mechanism and factors responsible for this regulation of blood flow during submaximal exercise with prolonged hypoxia. VIII. Sustained Hypoxia—Maximal Exercise Virtually all studies of maximal exercise after acclimatization to sustained hypoxia show a reduction in maximal cardiac output (see Table 5). In contrast to the hemodynamic changes during submaximal exercise, reductions in both heart rate and stroke volume are responsible for the reduced cardiac output at maximal exercise to varying degrees depending on the altitude in question. For example, at moderate altitudes of 3000–4000 m, reductions in stroke volume are predominant (126,127,128,143) and are similar in magnitude to those observed during submaximal exercise. However, at altitudes of ⱖ4000 m, reductions in maximal heart rate appear to play an increasingly important role (62,66,147,148). This reduction in maximal cardiac output, compounded by decreased arterial oxygen content, results in a decrease in peak

Figure 10 Cardiac hemodynamic responses during exercise with progressive hypoxia exposure in the Operation Everest II study (62). The Vo 2-workload (upper left) and the cardiac output–workload relationships (upper right) are maintained during submaximal exercise at all altitudes. With progressive hypoxia, heart rate increases (lower left) and stroke volume decreases (lower right). The response of stroke volume to progressive exercise differs between sea level and extreme hypoxia. Contrast these findings with the hemodynamic responses with acute hypoxia in Figure 7.

Figure 11 With progressive hypoxia exposure, oxygen uptake is maintained by an increase in oxygen extraction (a-v o 2 /CaO 2) as oxygen delivery is decreased compared to sea level. (From Ref. 138.)

270 Wolfel and Levine

Cardiovascular System at High Altitude

271

O2 delivery. Thus, despite appropriate acclimatization, maximal oxygen uptake remains reduced compared to sea level, and similar to the reduction observed with acute hypoxia in all studies at ⬎3000 m, though the reductions in conductive oxygen delivery are offset to some extent by increased tissue O2 extraction. Still unclear is whether the reduction in cardiac output at maximal exercise is a primary cause of or secondary to the reduction in maximal oxygen uptake at high altitude (149). In contrast to the elevated heart rate at rest and during submaximal exercise, there is a consistent reduction in maximal heart rate at virtually all altitudes studied (Fig. 12). The degree of reduction in heart rate is dependent on the duration and severity of sustained hypoxia exposure with the greatest reductions at the most extreme altitudes. Many studies have reported reductions in maximal heart rate after 2–3 weeks residence at 4300 m (63,74,126–128,150). At altitudes of ⬎5800 m there is a 25% reduction in maximal heart rate associated with a ⬎60% reduction in maximal oxygen uptake (62,66,147,148). The mechanism for this reduction in maximal heart rate could be related to a direct depressant effect of hypoxia itself on either the sinus node or central cortical irradiation (central command), a secondary effect of reduced work capacity, and therefore reduced afferent feedback from exercising skeletal muscle, or specific alterations in autonomic nervous system activity associated with sustained hypoxia. In practice, these different mechanisms have been very difficult to differentiate. Studies with inhalation of supplemental oxygen to simulate normoxic conditions after prolonged exposure to hypoxia have yielded mostly expected results. For example, at an altitude of 4300 m, abrupt restoration of normoxic conditions in

Figure 12 Maximal heart rate falls with sustained hypoxia exposure. The average duration of high-altitude exposure in each study was at least 2 weeks. Maximal heart rate began to decrease above altitudes of 4000 m. Numbers identify each specific study in the reference list.

272

Wolfel and Levine

recently acclimatized subjects led to an increase in heart rate and cardiac output with a definite increase in maximal workrate and oxygen uptake (128). However, cardiac output was still less than at sea level. Similar findings were seen at an altitude of 5800 m with improvement in heart rate and oxygen uptake at maximal exercise with oxygen inhalation (66). In contrast, with more prolonged, progressive exposure to more extreme altitude, there was little benefit in heart rate and work capacity with supplemental oxygen (62). It is likely, however, that prominent cardiovascular deconditioning played a substantial role in reducing work capacity in these studies. The degree of restoration of maximal heart rate with acute oxygen administration in acclimatized subjects may depend on the degree of attenuation of sympathetic activity (65)—if sympathetic activity is already near maximal, then supplemental oxygen results in relatively minimal additional tachycardia. Thus, autonomic nervous system activity likely plays either a primary or a facilitative role in reducing maximal heart rate with prolonged hypoxic exposure. In addition to the known blunting of the heart rate response to enhanced sympathetic activity with prolonged hypoxia, early studies suggested an additional enhancement of parasympathetic activity with chronic hypoxia, as maximal heart rate was restored to sea level values with atropine in recently acclimatized subjects to 4600 m (151). Moreover, in lifelong high-altitude residents, atropine produces a greater increase in exercise heart rate compared to recently acclimatized subjects (152), suggesting an enhancement of vagal influences. However, studies with both β-adrenergic blockade and parasympathetic blockade in climbers residing at altitudes of ⬎5000 m suggest that the degree of parasympathetic modulation of the heart is intimately influenced by the extent of cardiac sympathetic activity (65). In an elegant study by Savard et al., climbers who had the greatest reduction in adrenergic responsiveness also had the greatest increase in maximal heart rate with atropine. Conversely, if sympathetic responsiveness was minimally reduced, then the response to atropine was less prominent. Thus, the interplay between sympathetic and parasympathetic regulation of heart rate during maximal exercise makes a simple determination of a single, responsible neural pathway somewhat tenuous. Interestingly, in all the studies where maximal heart rate was increased with atropine, there was no improvement in maximal oxygen uptake, suggesting that the decreased maximal heart rate with acclimatization is unlikely to be a primary cause of the reduction in peak work capacity. Finally, a key unanswered question is the degree to which the reduction in maximal cardiac output at high altitude contributes to the reduction in maximal oxygen uptake. At present, most evidence points to the converse: that the reduction in oxygen availability/utilization is the primary determinant of systemic and regional blood flow (149). Lines of evidence that support this hypothesis include (1) the relationship between cardiac output and oxygen uptake remains similar to that at sea level, even at maximal exercise at extreme altitude; (2) addition of supplemental oxygen results in an immediate increase in work rate, oxygen uptake, heart rate, and cardiac output; (3) increases (65,151) or decreases (143) in heart rate by pharmacological intervention fail to alter maximal oxygen uptake; and (4) even autologous

Cardiovascular System at High Altitude

273

blood transfusion at altitude, which increases blood volume and red cell mass, fails to increase maximal oxygen uptake at altitude (153), arguing for a primary role of reduced diffusion gradients for oxygen as the cause of reduced oxygen utilization. Issues related to distribution of cardiac output and degree of regional autonomic activation during maximal exercise with prolonged hypoxia remain to be elucidated before this hypothesis can be universally accepted (154). In summary, during exercise at high altitude, the cardiovascular system functions to provide the delivery of oxygen in relation to the degree of oxygen utilization. During acute hypoxia, blood flow is increased to levels greater than expected for a given oxygen uptake at sea level. With prolonged hypoxia, there appears to be a ‘‘cardiac-sparing’’ effect on cardiac output and limb blood flow in relation to the degree of improvement in arterial oxygenation by the acclimatization process. There is a complex interaction between oxygen delivery, diffusion, and extraction to maintain oxygen uptake during submaximal exercise. At maximal exercise, oxygen uptake is reduced but the relationship between cardiac output and oxygen uptake remains similar to sea level and the factors responsible for the reduction in cardiac output remain to be determined. There clearly is an interplay between the autonomic nervous system and the degree of hypoxia which influences the cardiovascular responses to exercise with prolonged hypoxia. A. Key Unanswered Questions—Exercise During Acute and Sustained Hypoxia

What is the mechanism for the increased blood flow responses during acute hypoxia? Is it all sympathetic stimulation? What factors regulate the degree of peripheral oxygen extraction during both acute and sustained hypoxia? Peripheral venous oxygen tension remains the same with acute and sustained hypoxia. Why? What are the relative roles of diminished cardiac response to sympathetic stimulation and enhanced parasympathetic stimulation with sustained hypoxia? Do the changes in cardiac output and peripheral blood flow play a primary or secondary role in the reduced exercise capacity during sustained hypoxia? IX. High-Altitude Residents and Populations Exposure to high altitude over many years leads to alterations in hemodynamic, autonomic, metabolic, and coronary circulatory changes that may enhance tolerance to chronic hypoxia. High-altitude natives taken to sea level have a persistent reduction in stroke volume and cardiac output despite the introduction of normoxic conditions. Thus, vascular remodeling and other structural changes may occur with longterm residence at altitude that are not readily reversible with normoxia. An enhanced utilization of glucose by the heart in high-altitude natives results in greater oxygen efficiency (more high-energy phosphate production per molecule of oxygen) and may be an important factor in the preservation of cardiac function with prolonged hyp-

274

Wolfel and Levine

oxia. Autonomic balance favors enhanced parasympathetic and reduced sympathetic activity in high-altitude natives, though definitive studies have not yet been done. The significance of these findings for cardiovascular function has not been determined. There may be favorable changes in the coronary circulation of high-altitude natives that enhance myocardial blood flow and aid in myocardial protection. The combination of reduced coronary blood flow and enhanced vascularity may be a favorable adaptation to chronic hypoxia. The mechanisms for these coronary responses remain to be determined. Finally, the distinction between lifelong acclimatization to hypoxia versus population-based, genetic adaptations in the phenotypic cardiovascular changes of the high-altitude native have yet to be determined. The study of high-altitude natives provides an opportunity to examine more prolonged exposure to chronic hypoxia and to differentiate the cardiovascular adaptations both quantitatively and qualitatively from the short-term (days to weeks) acclimatization responses discussed above. In particular, the responses to normoxia (sea level) in such individuals can provide information on the reversibility of the cardiovascular adaptations to chronic hypoxia. These populations are usually limited to elevations of 3000–4500 m. Interpretation of information from high-altitude residents is influenced by many factors including the altitude of residence, the physical activity patterns of the subjects, and the possible role of genetic changes that may have resulted in advantageous physiological adaptations to a low oxygen environment. Men of European ancestry whose parents moved from low altitude to Leadville, Colorado, differ genetically from natives of the Peruvian Andes and inhabitants of the Tibetan plateau whose ancestors have spent several generations at high altitude, though it is not clear whether these differences are a direct result of highaltitude residence. Early reports on the cardiovascular function of the Sherpas of Nepal suggested that this high altitude population might have unique physiological characteristics that explained their superior exercise capacity under conditions of extreme hypoxia (6,132). These responses include a greater heart rate and cardiac output at maximal exercise along with lower minute ventilation, a higher PCO2, and a normal blood pH compared to recently acclimatized lowlanders. Intriguingly, many of these adaptations of Asian highlanders are directionally opposite to those obtained in lowlanders who have acclimated well to prolonged hypoxia. Whether inhabitants of high regions elsewhere in the world behave similarly is a question for future study. In general, hemodynamic changes with prolonged hypoxia in acclimatized lowlanders are similar in high altitude residents. One of the key differences between high altitude natives and recently acclimatized lowlanders is the substantially greater red cell mass in the former (see Chapter 3). This difference appears to result in a more prominent reduction in cardiac volumes. For example, the higher the hematocrit with progressive altitude residence, the lower the cardiac output secondary to decreases in stroke volume (155). These changes appear to persist with lifelong exposure to high altitude as Tibetans at 3,658 m have lower stroke volumes at rest and during exercise compared to sea level natives examined in the Operation Everest II study, even when corrected for differences in body size (62,156) (Fig. 13). Cardiac

Figure 13 Native Tibetans have higher heart rates and lower stroke volume indices at rest and during exercise compared to North American men. Cardiac index, however, is similar between the groups. Vo 2 values were similar at identical workloads in both subject groups. Figures drawn from data on sea level subjects from Operation Everest II (62) and from lifelong residents of Tibet at 3658 m (156).

Cardiovascular System at High Altitude 275

276

Wolfel and Levine

index, however, was similar for a given oxygen uptake in the high-altitude residents compared to acclimatized sea level natives, as a greater heart rate offset the smaller stroke volume. This observation confirms the similarity in regulatory mechanisms between the two populations and provides additional evidence supporting the reliability of the finding of smaller stroke volume. The differences in stroke volume between these two diverse populations might also be explained by differences in fitness between the North American subjects and the Tibetans. For example, the Operation Everest II subjects were generally very fit and included a number of endurance-trained athletes, resulting in a mean peak oxygen uptake of 51.2 ⫾ 9.0 (SD) mL/kg/min at sea level. In contrast, the Tibetan subjects were drawn from the general population in Lhasa and had a peak oxygen uptake more typical of untrained subjects of 43.4 ⫾ 4.0 mL/kg/min. Comparing the stroke volumes between these highlanders and lowlanders also provides indirect evidence against the hypothesis that impaired right ventricular function could be partially responsible for the reduction in left ventricular stroke volume. For example, when compared to recently acclimatized subjects or highaltitude residents in both North and South America, the pulmonary artery pressures of Tibetan natives were significantly lower at similar O2 tensions (156). Thus, if pulmonary hypertension was responsible for impairing left ventricular filling, the Tibetans would be expected to have higher, rather than lower stroke volumes. When high-altitude residents are taken to sea level, their cardiac outputs and stroke volumes are lower than those of sea level natives (80,82,157). With shortterm residence at sea level, cardiac output and stroke volume increase but they remain lower than in sea level natives, raising the possibility of a structural remodeling. However, most of the change in stroke volume could be attributed to changes in plasma volume (9), and differences in fitness, diet, blood pressure, and other factors between the two populations make definitive conclusions from short-term deacclimatization studies uncertain. Interestingly, the breathing of a high-oxygen mixture to simulate sea level conditions while the subjects remained at altitude did not result in any significant hemodynamic changes (82). However, when cardiac hemodynamics were studied in high-altitude natives after 2 years of residence at sea level, the findings were similar to sea level natives, indicating that the majority of the cardiovascular changes with chronic hypoxia are reversible with prolonged exposure to normoxia. This observation argues against any permanent structural changes or fundamental genetic differences (158). One of the key questions regarding long-term exposure to chronic hypoxia is whether the dramatic sympathetic activation recently reported with short-term acclimatization persists or gradually abates (see discussion above). This response may be very different in individuals born in a hypoxic environment, compared with those who were born at sea level but have migrated to high altitudes. Studies with β-adrenergic blockade and atropine in Tibetans compared to newly acclimatized Han Chinese suggest greater parasympathetic activity in the high-altitude natives (152), manifested by greater increases in heart rate with atropine. Heart rate variability studies have also shown a greater respiratory sinus arrhythmia (20), though such

Cardiovascular System at High Altitude

277

differences simply may be a function of differences in respiratory rate and tidal volume. This is an important area for future research. One of the most intriguing findings in high-altitude natives relates to alterations in myocardial metabolism. Early studies in native high-altitude Peruvians suggested a greater reliance on carbohydrate metabolism for energy sources by the resting myocardium (159). This difference was suggested by a greater coronary sinus respiratory quotient (0.81 in lowlanders vs. 0.91 in high-altitude residents) as well as greater myocardial extraction of lactate and glucose compared to free fatty acids. More recent studies on Sherpa men using 31P-magnetic resonance spectroscopy also support this theory: lower ratios of phosphocreatine to ATP found in these natives are consistent with a threefold increase in free ADP concentration in the myocardium (160). These metabolic conditions would accommodate the higher enzyme kinetic constants (K M ) of the enzymes phosphoglycerate kinase and pyruvate kinase, indicating enhanced capacity for carbohydrate metabolism. Similar findings were also reported in both Quechua and Sherpa subjects in whom an enhanced glucose uptake compared to sea level natives was observed with positron emission tomography (161). It remains to be determined whether this alteration in myocardial substrate utilization is a phenotypic expression of an altered genetic myocardial adaptation in high-altitude natives or a common acclimatization response seen with more shortterm exposure to high altitude. The persistence of myocardial preference for glucose after 3–4 weeks of residence at sea level in these high-altitude natives would support a more permanent adaptation (160,161). On the other hand, with short-term (3 weeks) exposure to a moderate altitude of 4300 m there is a definite increase in glucose and lactate utilization by skeletal muscle (162–164). Whether there is also an early shift in myocardial metabolism with short-term hypoxia exposure is unknown. The functional significance of these observations is unclear, however, since myocardial function has been demonstrated to be entirely normal in sea level natives performing maximal exercise under extreme hypoxia. Limited studies available in high-altitude natives also suggest that there are changes in coronary anatomy and physiology that may be protective in the setting of chronic hypoxia. Resting coronary blood flow has been shown to decline progressively with increasing altitude (123). However, myocardial oxygen consumption was also lower at rest in these high-altitude natives, a finding that differs from studies of recently acclimatized newcomers to altitude (124). Differences in methodology and the possible inclusion of patients with chronic mountain sickness make these data difficult to interpret. Pathological studies on high-altitude natives also demonstrate that lifelong exposure to hypoxia is associated with a more abundant coronary vascular bed with a greater density of peripheral branching of smaller coronary vessels (165). The significance of these findings remains unclear and is hard to separate from population differences in diet, physical activity, and genetic susceptibility to coronary artery disease. However, it is possible that coronary adaptations to chronic hypoxia, similar to the increased collateralization seen with chronic myocardial ischemia (166), could result in more effective myocardial vascularization with greater surface area of oxygen diffusion. Such adaptations could be responsible

278

Wolfel and Levine

for a possible protective effect of long-term residence at high altitude against coronary artery disease (167). In summary, there are various hemodynamic, autonomic, metabolic, and possibly coronary structural adaptations that occur in long-term residents at high altitude that may contribute to preservation of cardiovascular function. The time course and the factors responsible for many of these changes have not been determined. Despite the known increase in pulmonary artery pressure with sustained hypoxia, certain high-altitude populations appear to have a blunted hypoxia-pressor response in the pulmonary vascular bed. The heterogeneous responses to sustained high-altitude exposure may relate to different genetic susceptibility as well as to alterations in controlling mechanisms of cardiovascular function. A. Key Unanswered Questions—High-Altitude Residents and Populations

What is the time course for the various cardiovascular adaptations observed in high-altitude natives? Are there actual genotypic changes that result from generations of chronic hypoxia exposure, or are there changes reversible with prolonged normoxic exposure? How do alterations in autonomic activity influence the cardiovascular adaptations to chronic hypoxia in high-altitude residents? X.

Clinical Correlation

Although the cardiovascular system appears to function normally even during exercise at extreme altitude, the effect of acute and sustained hypoxia on cardiac function in patients with underlying cardiovascular disease has not been well defined. The role of hypoxia in the induction of myocardial ischemia, cardiac arrhythmia, and heart failure in these patients is based on limited available clinical data; however, the application of theoretical concepts will provide a clearer understanding of potential pathophysiological responses (see Chapter 25). Although the cardiovascular system appears to function well at high altitude in healthy individuals, even during high-intensity exercise at extreme altitudes, exposure to acute and prolonged hypoxia in patients with cardiovascular disease could result in the new onset or accentuation of symptoms when compared to living at sea level. Clinical issues of particular concern include the provocation of myocardial ischemia, the occurrence of unstable coronary syndromes including acute myocardial infarction, susceptibility to cardiac arrhythmias, and the occurrence or worsening of heart failure. Exposure to high altitude presents several potentially deleterious stresses that may exacerbate the clinical manifestations of underlying cardiac disease. These include the degree of hypoxia itself, alkalosis, heightened sympathetic activity, elevated systemic blood pressure and heart rate, and increased blood viscosity. Exercise at altitude also may be an important factor, as any submaximal work-

Cardiovascular System at High Altitude

279

load at altitude is at a higher percentage of maximal oxygen uptake than at sea level and thereby requires increased cardiac work. However, the ‘‘cardiac-sparing’’ effects of acclimatization (reductions in heart rate and cardiac output) as well as the known decreases in heart size with prolonged hypoxia exposure may diminish any potential cardiac risks. A. Myocardial Ischemia and Infarction

In patients with known coronary disease, exposure to the hypoxia of high altitude may result in myocardial ischemia, especially during exercise. There is limited information available to evaluate the risk of ischemia in patients with coronary disease. Extreme hypoxia alone does not appear to cause any clinical or ECG evidence of ischemia in the absence of atherosclerotic disease (60,67). However, the vasomotor response of coronary arteries, especially via endothelial-dependent mechanisms, has been shown to be abnormal in patients with atherosclerotic coronary disease. For example, intracoronary infusion of acetylcholine, an endothelial-dependent vasodilator in normal vessels, causes vasoconstriction in diseased coronary vessels (168). A similar observation has been noted during exercise at sea level in coronary patients where vasoconstriction, rather than vasodilation, can occur in diseased coronary segments (169). A potential mechanism for this abnormal coronary vasoconstriction is enhanced sympathetic activity (170), such as occurs during exercise. There clearly is enhanced sympathetic activity on arrival at high altitude and further activation with prolonged exposure. Whether this increased sympathetic activity influences coronary vasomotor tone at high altitude is not known; however, the combination of exercise and hypoxia could theoretically result in enhanced coronary vasoconstriction in patients with CAD. The role of other vasoconstrictors such as endothelin and angiotensin II in coronary vasomotor control during exercise with hypoxia has not been investigated, but they do appear to be important in other hyperadrenergic states such as heart failure. Despite this condition of heightened sympathetic activity during hypoxia, counterregulatory mechanisms may serve to blunt or prevent adverse coronary vascular tone. Acute exposure of endothelial cells to hypoxia leads to production of nitric oxide (NO), an endothelial-dependent vasodilator (171,172). NO production may serve to protect the coronary tree from vasoconstriction early during hypoxia, but there are no long-term data on NO production by endothelial cells during chronic hypoxia. The pathogenesis of acute coronary syndromes including myocardial infarction are related to atherosclerotic plaque disruption in a coronary vessel. Local factors that may lead to these events include a lipid-rich plaque, vasoconstriction, residual thrombus in the area of the plaque, and the degree of stenosis (173). Systemic factors are also important and include increased catecholamine levels, an abnormal coagulation and fibrinolysis profile, abnormal metabolic states (diabetes, homocysteinemia), and shear forces in the vessels as can occur with hypertension. Although there have not been any reported coagulation abnormalities with hypoxia (174), other factors such as hypertension, heightened sympathetic activity, and possi-

280

Wolfel and Levine

bly coronary vasoconstriction could lead to plaque disruption in the susceptible patient with lipid-laden plaques while at altitude. Exercise may be a critical factor as it has been found to be an important cause of serious plaque rupture at sea level (175). Despite these theoretical reasons for increased cardiac risk at high altitude, the limited available data suggest that altitude exposure is generally well tolerated by patients with coronary disease. Coronary patients who were residents at 1600 m developed objective evidence of myocardial ischemia at a lower exercise workload upon initial arrival at 3100 m (176). However, after 5 days of acclimatization to the higher altitude, their ischemic pattern returned to their baseline level (177). In these partially altitude-acclimatized individuals, there was no change in the ischemic threshold (heart rate–systolic blood pressure product at the onset of ischemia) with acute or prolonged exposure to hypoxia. In contrast, sea level natives with coronary artery disease studied under conditions of acute hypoxia in a hypobaric chamber at a simulated altitude of 2500 m did develop myocardial ischemia at a lower hemodynamic threshold compared to normoxic conditions, raising the possibility of coronary vasoconstriction during hypoxia (11). However, these patients when studied after 5 days of residence at 2500 m only developed ischemia at their sea level threshold. Thus, it appeared that the initial adverse response to acute hypoxia was reversible with more prolonged exposure. Finally, in 90 male subjects over the age of 40 years, who underwent ECG-telemetry monitoring during alpine skiing at 2500 m, the frequency of silent ST-segment depression was 5.6% (178). There were no adverse clinical events in these men who were exercising at ⬎80% of their maximal sea level heart rate. Unfortunately, no sea level exercise ECG information was available in these men, preventing a direct comparison. B. Cardiac Arrhythmias

There is limited information on the occurrence of cardiac arrhythmias in heartdiseased patients at high altitude. Other than occasional premature ventricular and atrial beats (67), there does not appear to be an increased risk of arrhythmias with altitude exposure in normal subjects despite significant hypoxia and alkalosis. In elderly patients with either an increased risk for or known coronary disease, there were no significant arrhythmias noted by either short-term ECG monitoring at rest or exercise electrocardiography with acute or more prolonged exposure to 2500 m (11). In these same subjects, there were no hypoxia-related abnormalities in resting signal–averaged ECG recordings, a sensitive marker for the presence of a myocardial electrophysiological substrate conducive for arrhythmia production. No occurrence of significant cardiac arrhythmias has been reported in any study of heart disease patients at high altitude; however, most of these studies did not include patients with significant left ventricular systolic dysfunction, a group more prone to develop arrhythmias. In one small study of patients with a reduced mean left ventricular ejection fraction (LVEF) of 39%, no arrhythmias were reported during exercise at 2500 m (179). Further work at similar and higher altitudes is required

Cardiovascular System at High Altitude

281

in a larger number of patients with a greater clinical likelihood of cardiac arrhythmias before any definitive statement can be made about the risks of arrhymogenesis with altitude exposure. C. Heart Failure

Although the myocardium functions normally under conditions of acute and sustained hypoxia, it is unclear how a heart with reduced ventricular systolic function would tolerate the increased demands of acute and prolonged hypoxia. The heightened sympathetic activity, demand for increased cardiac output with acute hypoxia, and the systemic vasoconstriction that occurs with more prolonged exposure may result in further deterioration of ventricular function and may produce symptoms of decompensated heart failure. Although the usual cardiac response to prolonged hypoxia is a reduction in end-diastolic volume with a maintained or increased ejection fraction (85), echocardiographic studies in coronary subjects living in Denver, CO (1600 m), who spent 5 days at 3100 m demonstrated an increase in end-diastolic and end-systolic dimensions with a fall in LVEF from 51% at 1600 m to 37% at 3100 m (180). These results have not been confirmed in another study at 2500 m in which sea level residents with known coronary disease studied after 5 days at altitude had no echocardiographic abnormalities with sustained moderate hypoxia (11). Patients with known left ventricular dysfunction who were exercised at 2500 m were found to have the same decrement in exercise capacity as normal age-matched subjects (179). No adverse cardiovascular events were noted after 2 days at moderate altitude in these patients. Recently, sea level residents with chronic heart failure were studied while exercising at different degrees of acute normobaric hypoxia equivalent to altitudes varying from 1000 to 3000 m (181). These patients, with reduced exercise capacities when compared to normal subjects, had a similar response to progressive hypoxia in regards to arterial oxygen saturation, exercise ventilation, heart rate, and blood pressure. In normal subjects there was a 3% decrement in maximal exercise capacity for each 1000 m in elevation. Heart failure patients with a sea level maximal uptake greater than 15 mL/kg/min had a similar decrement of 5% per each 1000 m. However, patients with maximal oxygen uptake less than 15 mL/kg/min at sea level had a more marked reduction in exercise capacity, with an 11% reduction for each 1000 m of elevation. This reduction in exercise capacity probably reflected the inability to increase cardiac output in response to acute reductions in arterial oxygenation. No data are available in more prolonged hypoxia exposure in these heart failure patients. In addition, all these patients were on active heart failure medication that may have modified the cardiovascular responses to hypoxia. D. Clinical Guidelines

Based on the above information, patients with cardiac disease should be evaluated at sea level prior to ascent to altitude to insure a stable disease state (182) (see also Chapter 25). In addition, patients with known coronary disease should limit their

282

Wolfel and Levine

physical activity during the first few days at altitude to allow the favorable effects of acclimatization to occur. Guidelines for patients with left ventricular dysfunction are more difficult to determine, but patients with mild to moderate ventricular dysfunction should be able to tolerate moderate altitude without serious adverse consequences. In questionable cases, nocturnal oxygen supplementation should be considered and medications used at sea level should be continued at altitude. Patients with more significant cardiac impairment at sea level, especially with significant exercise limitation, should not travel to high altitude in the absence of supplemental oxygen to simulate normoxic conditions. In summary, there is limited information on altitude tolerance in patients with cardiovascular disease. Hypoxia-induced coronary vasoconstriction is a particularly intriguing concept that needs to be explored. The limited data would suggest that moderate altitude exposure (2500–3100 m) is well tolerated by stable coronary patients with normal or moderately depressed ventricular function. Although difficult to obtain, more information is needed in this important area of high-altitude medicine. E.

Key Unanswered Questions—Clinical Correlations

Can hypoxia alone lead to instability of coronary vascular lesions and vasomotor tone with the precipitation of unstable coronary syndromes and possibly acute myocardial infarction? Does the cardiac response to exercise at sea level in coronary disease patients predict the response during either acute or sustained hypoxia? What are the implications of altitude exposure on patients with reduced ventricular function at sea level? Are they at increased risk for arrhythmias, cardiogenic pulmonary edema, pulmonary hypertension, and right heart failure? Does exposure to acute or sustained hypoxia increase the likelihood of arrhythmias in patients with cardiovascular disease? What is the electrophysiological mechanisms of cardiac conduction during hypoxia? What guidelines are reliable to assess cardiac risk at high altitude? Acknowledgments This chapter is dedicated to the memory of Herbert Hultgren, M.D., who was a pioneer in the investigation of cardiovascular responses to high altitude in a variety of settings. His work in lowlanders exposed to high altitude, in high-altitude natives, and in patients with high-altitude pulmonary edema has provided important information that has contributed to our understanding of the functioning of the cardiovascular system and systemic circulation at altitude. His mentoring and constant guidance in the pursuit of high-altitude research as well as his genuine concern for his colleagues will be greatly missed. We also wish to acknowledge the investigators and subjects of the various Pikes Peak studies, Operation Everest II, and the Tenth Moun-

Cardiovascular System at High Altitude

283

tain Division Study. Their efforts have greatly contributed to the advancement of knowledge in this important area of high-altitude physiology and medicine.

References 1. Bert P. La Pression Barometrique. Masson, Paris, 1878. English translation by Hitchcock MA and Hitchcock FA. Columbus, Ohio: College Book Co., 1943:343–346. 2. Douglas CG, Haldane JA, Henderson Y, Schneider EC. Physiological observations made on Pikes Peak, Colorado, with special reference to adaptation to low barometric pressures. Phil Trans R Soc London Ser B 1913; 203:185–381. 3. Barcroft J, Binger CA, Bock AV, Doggart JH, Forbes HS, Harrop G, Meakins JC, Redfield AC. Observations upon the effect of high altitude on the physiological processes of the human body, carried out in the Peruvian Andes, chiefly at Cerro de Pasco. Phil Trans R Soc London Ser B 1923; 211:351–480. 4. Grollman A. Physiological variations of the cardiac output of man. VII. The effect of high altitude on the cardiac output and its related functions: an account of experiments conducted on the summit of Pikes Peak, Colorado. Am J Physiol 1930; 93:19–40. 5. Doyle JT, Wilson JS, Warren JV. The pulmonary vascular responses to short-term hypoxia in human subjects. Circulation 1952; 5:263–270. 6. Wescott RN, Fowler NO, Scott RC, Hauenstein VD, McGuire J. Anoxia and human pulmonary vascular resistance. J Clin Invest 1951; 19:957–970. 7. Vogel JA, Harris CW. Cardiopulmonary responses of resting man during early exposure to high altitude. J Appl Physiol 1967; 22:1124–1128. 8. Stenberg J, Ekblom B, Messin R. Hemodynamic response to work at simulated altitude, 4,000 m. J Appl Physiol 1966; 21:1589–1594. 9. Wagner PD, Gale GE, Moon RE, Torre-Bueno JR, Stolp BW, Saltzman HA. Pulmonary gas exchange in humans exercising at sea level and simulated altitude. J Appl Physiol 1986; 61:260–270. 10. Wagner JA, Miles DS, Horvath SM. Physiological adjustments of women to prolonged work during acute hypoxia. J Appl Physiol 1980; 49:367–373. 11. Levine BD, Zuckerman JH, deFilippi CR. Effect of high-altitude exposure in the elderly: The Tenth Mountain Division Study. Circulation 1997; 96:1224–1232. 12. Richardson DW, Kontos HA, Raper AJ, Patterson JL Jr. Modification by betaadrenergic blockade of the circulatory responses to acute hypoxia in man. J Clin Invest 1967; 46:77–85. 13. Koller EA, Drechsel S, Hess T, Macherel P, Boutellier U. Effects of atropine and propranolol on the respiratory, circulatory, and ECG responses to high altitude in man. Eur J Appl Physiol 1988; 57:163–172. 14. Cunningham WL, Becker EJ, Kreuzer F. Catecholamines in plasma and urine at high altitude. J Appl Physiol 1965; 20:607–610. 15. Mazzeo RS, Bender PR, Brooks GA, Butterfield GE, Groves BM, Sutton JR, EE Wolfel, Reeves JT. Arterial catecholamine responses during exercise with acute and chronic high-altitude exposure. Am J Physiol 1991; 261:E419–E424. 16. Mazzeo RS, Wolfel EE, Butterfield GE, Reeves JT. Sympathetic response during 21 days at high altitude (4,300 m) as determined by urinary and plasma catecholamines. Metabolism 1994; 43:1226–1232.

284

Wolfel and Levine

17. Leuenberger U, Gleeson K, Wroblewski K, Prophet S, Zelis R, Zwillich C, Sinoway L. Norepinephrine clearance is increased during acute hypoxemia in humans. Am J Physiol 1991; 261:H1659–H1664. 18. Hughson RL, Yamamoto Y, McCullough RE, Sutton JR, Reeves JT. Sympathetic and parasympathetic indicators of heart rate control at altitude studied by spectral analysis. J Appl Physiol 1994; 77:2537–2542. 19. Yamamoto Y, Hoshikawa Y, Miyashita M. Effects of acute exposure to simulated altitude on heart rate variability during exercise. J Appl Physiol 1996; 81:1223–1229. 20. Passino C, Bernardi L, Spadacini G, Calciati A, Robergs R, Anand I, Greene R, Martignoni E, Appenzeller O. Autonomic regulation of heart rate and peripheral circulation: comparison of high altitude and sea level residents. Clin Sci 1996; 91(suppl):81– 83. 21. Ponchia A, Noventa D, Bertaglia M, Carreta R, Zaccaria M, Miraglia G, Pascotto P, Buja G. Cardiovascular neural regulation during and after prolonged high altitude exposure. Eur Heart J 1994; 15:1463–1469. 22. Farnelli CC, Kayser B, Binzoni T, Cerretelli P, Girardier L. Autonomic nervous control of heart rate at altitude (5,050 m). Eur J Appl Physiol 1994; 69:502–507. 23. Perini R, Milesi S, Biancardi L Veicsterinas A. Effects of high altitude acclimatization on heart rate variability in resting humans. Eur J Appl Physiol 1996; 73:521–528. 24. Yamamoto Y, Hughson RL, Sutton JR, Houston CS, Cymerman A, Fallen EL, Kamath MV. Operation Everest II: an indication of deterministic chaos in human heart rate variability at simulated high altitude. Biol Cybern 1993; 69:205–212. 25. Lipsitz LA, Hsahimoto F, Lubowsky LP, Mietus J, Noody GB, Appenzeller O, Goldberger AL. Heart rate and respiratory rhythm dynamics on ascent to high altitude. Br Heart J 1995; 74:390–396. 26. White DB. Autonomic cardiac regulation during acute and chronic hypoxia assessed by spectral analysis of R-R interval variability. In: Houston CS, Coates G, eds. Hypoxia: Women at Altitude. Burlington, VT: Queen City Printers Inc, 1997:124–133. 27. Brown TE, Beightol LA, Koh J, Eckberg DL. Important influence of respiration on human R-R interval power spectra is largely ignored. J Appl Physiol 1993; 75:2310– 2317. 28. Levine BD, Zhang R, Zuckerman JH, Morrow M. The effect of cardiac mechanics on heart rate variability. Circulation 1996; 94(8):I–368. 29. Piepoli M, Sleight P, Leuzzi S, Valle F, Spadacini G, Passino C, Johnston J, Bernardi L. Origin of respiratory sinus arrhythmia in conscious humans: an important role for arterial carotid baroreceptors. Circulation 1997; 95:1813–1821. 30. Grassman E, Blomqvist CG. Absence of respiratory sinus arrhythmia: a manifestation of the sick sinus syndrome. Clin Cardiol 1983; 6:151–154. 31. Eckberg DL. Sympathovagal balance: a critical appraisal. Circulation 1997; 96:3224– 3232. 32. Somers VK, Mark AL, Abboud FM. Interaction of baroreceptor and chemoreceptor reflex control of sympathetic nerve activity in normal humans. J Clin Invest 1991; 87: 1953–1957. 33. Saito M, Mano T, Iwase S, Koga K, Abe H, Yamazaki Y. Responses in muscle sympathetic activity to acute hypoxia in humans. J Appl Physiol 1988; 65:1548–1552. 34. Duplain H, Vollenweider L, Delabays A, Nicod P, Bartsch P, Scherrer U. Augmented sympathetic activation during short-term hypoxia and high altitude exposure in subjects susceptible to high altitude pulmonary edema. Circulation 1999; 99:1713–1718.

Cardiovascular System at High Altitude

285

35. Roach R, Vissing SF. Unpublished results, personal communication, 1998. 36. Eckberg DL, Bastiw H III, Scruby AE. Modulation of human sinus node function by systemic hypoxia. J Appl Physiol 1982; 52:570–577. 37. Sommers VK, Abboud FM. Chemoreflex control of cardiac vagal activity. In: Levy MN, Schwartz PJ, eds. Vagal Control of the Heart. Armonk, NY: Futura Publishing Co, 1994:403–416. 38. Spyer KM. Neural organisation and control of the baroreceptor reflex. Rev Physiol Biochem Pharmacol 1981; 88:23–123. 39. Kahler RL, Goldblatt A, Braunwald E. The effects of acute hypoxia on the systemic venous and arterial systems and on myocardial contractile force. J Clin Invest 1962; 41:1553–1563. 40. Mifflin SW, Felder RB. Synaptic mechanisms regulating cardiovascular afferent inputs to solitary tract nucleus. Am J Physiol 1990; 259:H653–H661. 41. Asano A, Mazzeo RS, McCullough RE, Wolfel EE, Reeves JT. Relation of sympathetic activation to ventilation in man at 4,300 m altitude. Aviat Space Environ Med 1997; 68:104–110. 42. Hultgren H. The systemic circulation. In: High Altitude Medicine. Stanford, CA: Hultgren Publications, 1997:33–63. 43. Pool PE, Covell JW, Chidsey CA, Braunwald E. Myocardial high energy phosphate stores in acutely induced hypoxic heart failure. Circ Res 1966; 19:221–229. 44. Heistad DD, Abboud FM. Circulatory adjustments to hypoxia. Circulation 1980; 61: 463–470. 45. Rowell LB, Saltin B, Kiens B, Christensen NJ. Is peak quadriceps blood flow in humans even higher during exercise during hypoxia? Am J Physiol 1986; 251:H1038– H1044. 46. Knight DR, Schaffartzik W, Poole DC, Hogan MC, Bebout DE, Wagner PD. Effects of hyperoxia on maximal leg O2 supply and utilization in men. J Appl Physiol 1993; 75:2586–2594. 47. Hartley LH, Vogel JA, Landowne M. Central, femoral, and brachial circulation during exercise in hypoxia. J Appl Physiol 1973; 34:87–90. 48. Mancia G, Mark AL. Arterial baroreflexes in humans. In: Shepherd JT, Aboud FM, eds. Handbook of Physiology, Section 2: The Cardiovascular System, Vol. 3. Betheseda, MD: American Physiological Society, 1983:795–813. 49. Nicholas R, O’Meara P, Calonge N. Is syncope related to moderate altitude exposure? JAMA 1992; 268:904–906. 50. Remensnyder JP, Mitchell JH, Sarnoff SJ. Functional sympatholysis during muscular activity. Circ Res 1962; 11:370–380. 51. Hansen J, Thomas GD, Harris SA, Parsons WJ, Victor RG. Differential sympathetic neural control of oxygenation in resting and exercising human skeletal muscle. J Clin Invest 1996; 98:584–596. 52. Thomas GD, Hansen J, Victor RG. ATP-sensitive potassium channels mediate contraction-induced attenuation of sympathetic vasoconstriction in rat skeletal muscle. J Clin Invest 1997; 99:2602–2609. 53. Thomas GD, Victor RG. Nitric oxide mediates contraction-induced attenuation of sympathetic vasoconstriction in rat skeletal muscle. J Physiol 1997; 506:817–826. 54. Hackett JG, FM Abboud, AL Mark, Schmid PG, Heistad DD. Coronary vascular responses to stimulation of chemoreceptors and baroreceptors: evidence for reflex activation of vagal cholinergic innervation. Circ Res 1972; 31:8–17.

286

Wolfel and Levine

55. Ellsworth ML, Forrester T, Ellis CG, Dietrich HH. The erythrocyte as a regulator of vascular tone. Am J Physiol 1995; 269:H2155–H2161. 56. Hultgren H. Circulatory responses to acute hypoxia in normal subjects at sea level. In: High Altitude Medicine. Stanford, CA: Hultgren Publications, 1997:34. 57. Wolfel EE, Selland MA, Mazzeo RS, Reeves JT. Systemic hypertension at 4,300 m is related to sympathoadrenal activity. J Appl Physiol 1994; 76:1643–1650. 58. Nielsen AM, Bisgard GE, Vidruk EH. Carotid chemoreceptor activity during acute and sustained hypoxia in goats. J Appl Physiol 1988; 65:1796–1802. 59. Hoon RS, Balasubramanian V, Mathew OP, Tiwari SC, Sharma SC, Chadha KS. Effect of high-altitude exposure for 10 days on stroke volume and cardiac output. J Appl Physiol 1977; 42:722–727. 60. Karliner JS, Sarnquist FF, Graber DJ, Peters RM Jr, West JB. The electrocardiogram at extreme altitude: experience on Mount Everest. Am Heart J 1985; 109:505– 513. 61. Houston CS, Riley RL. Respiratory and circulatory changes during acclimatization to high altitude. Am J Physiol 1947; 140:565–588. 62. Reeves JT, Groves BM, Sutton JR, Wagner PD, Cymerman A, Malconian MK, Rock PB, Young PM, Houston CS. Operation Everest II: preservation of cardiac function at extreme altitude. J Appl Physiol 1987; 63:531–539. 63. Wolfel EE, Selland MA, Cymerman A, Brooks GA, Butterfield GE, Mazzeo RS, Grover RF, Reeves JT. O2 extraction maintains O2 uptake during submaximal exercise with β-adrenergic blockade at 4,300 m. J Appl Physiol 1998; 85:1092–1102. 64. Moore LG, Cymerman A, Huang SY, McCullough RE, McCullough RG, Rock PB, Young A, Young P, Weil JV, Reeves JT. Propranolol blocks metabolic rate increase but not ventilatory acclimatization to 4,300 m. Respir Physiol 1987; 70:195–204. 65. Savard GK, Areskog NH, Saltin B. Cardiovascular response to exercise in humans following acclimatization to extreme altitude. Acta Physiol Scand 1995; 154:499– 509. 66. Pugh LGCE, Gill MB, Lahiri S, Milledge JS, Ward MP, West JB. Muscular exercise at great altitudes. J Appl Physiol 1964; 19:431–440. 67. Malconian M, Rock P, Hultgren H, Donner H, Cymerman A, Groves BM, Reeves JT, Alexander J, Sutton JR, Nitta M, Houston CS. The electrocardiogram at rest and exercise during a simulated ascent of Mount Everest (Operation Everest II). Am J Cardiol 1990; 65:1475–1480. 68. Bristow MR, Ginsburg R, Minobe W, Cubiccioti RS, Sageman WS, Lurie K, Billingham ME, Harrison DC, Stinson EB. Decreased catecholamine sensitivity and βadrenergic receptor density in failing human hearts. N Engl J Med 1982; 307:205– 211. 69. Voelkel NF, Hegstrand L, Reeves JT, McMurtry IF, Molinoff PB. Effects of hypoxia on density of β-adrenergic receptors. J Appl Physiol 1981; 50:363–366. 70. Kacimi R, Richalet JP, Corsin A, Abousahl I, Crozatier B. Hypoxia-induced downregulation of β-adrenergic receptors in rat heart. J Appl Physiol 1992; 73:1377–1382. 71. Richalet JP, Larmignat P, Rathat C, Keromes A, Baud P, Lhoste F. Decreased cardiac response to isoproterenol infusion in acute and chronic hypoxia. J Appl Physiol 1988; 65:1957–1961. 72. Antezana AM, Kacimi R, Le Trong JL, Marchal M, Abousahl I, Dubray C, Richalet JP. Adrenergic status of humans during prolonged exposure to the altitude of 6,542 m. J Appl Physiol 1994; 76:1055–1059.

Cardiovascular System at High Altitude

287

73. Maher JT, Manchanda C, Cymerman A, Wolfe DL, Hartley LH. Cardiovascular responsiveness to β-adrenergic stimulation and blockade in chronic hypoxia. Am J Physiol 1975; 228:477–481. 74. Wolfel EE, Groves BM, Brooks GA, Butterfield GE, Mazzeo RS, Moore LG, Sutton JR, Bender PR, Dahms TE, McCullough RE, McCullough RG, Huang SY, Sun SF, Grover RF, Hultgren HN, Reeves JT. Oxygen transport during steady-state submaximal exercise in chronic hypoxia. J Appl Physiol 1991; 70:1129–1136. 75. Houston CS, Sutton JR, Cymerman A, Reeves JT. Operation Everest II: man at extreme altitude. J Appl Physiol 1987; 63:877–882. 76. Wagner PD. A theoretical analysis of factors determining VO2 max at sea level and altitude. Respir Physiol 1996; 196:329–343. 77. Fulco CS, Cymerman A, Rock PB, Farese G. Hemodynamic responses to upright tilt at sea level and high altitude. Aviat Space Environ Med 1985; 56:1172–1176. 78. Klausen K. Cardiac output in man at rest and work during and after acclimatization to 3,800 m. J Appl Physiol 1966; 21:609–616. 79. Alexander JK, Grover RK. Mechanism of reduced cardiac stroke volume at high altitude. Clin Cardiol 1983; 6:301–303. 80. Alexander JK, Hartley LH, Modelski M, Grover RF. Reduction in stroke volume during exercise in man following ascent to 3,100 m altitude. J Appl Physiol 1967; 23: 849–858. 81. Balasubramanian V, Behl A, Das GS, Waldhwa AK, Mathew OP, Hoon RS. Effect of digoxin and diuretics on high altitude left ventricular dysfunction. Circulation 1978; 57:1180–1185. 82. Hartley LH, Alexander JK, Modelski M, Grover RF. Subnormal cardiac output at rest and during exercise in residents at 3,100 m altitude. J Appl Physiol 1967; 23:839– 848. 83. Fowles RE, Hultgren HN. Left ventricular function at high altitude examined by systolic time intervals and M-mode echocardiography. Am J Cardiol 1983; 52:862–866. 84. Buch J, Egeblad H, Hansen P, Kjaergard H, Waldorf S, Steiness E. Correlation between changes in systolic time intervals and left ventricular end-diastolic diameter after preload reduction. Br Heart J 1980; 44:668–671. 85. Suarez J, Alexander JK, Houston CS. Enhanced left ventricular systolic performance at high altitude during Operation Everest II. Am J Cardiol 1987; 60:137–142. 86. Webster KA, Bishopric NH. Molecular regulation of cardiac myocyte adaptation to chronic hypoxia. J Mol Cell Cardiol 1992; 24:741–752. 87. Mazer CD, Stanley WC, Hicker RF, Neese RA, Cason BA, Demas KA, Wisneski JA, Gertz EW. Myocardial metabolism during hypoxia: Maintained lactate oxidation during increasing glycolysis. Metabolism 1990; 39:913–918. 88. Wiesner RJ. Adaptation of mitochondrial gene expression to changing cellular energy demands. News Physiol Sci 1997; 12:178–184. 89. Bunn HF, Poyton RO. Oxygen sensing and molecular adaptation to hypoxia. Physiol Rev 1996; 76:839–887. 90. Kacimi R, Long CS, Karliner JS. Chronic hypoxia modulates the interleukin-1β stimulated inducible nitric oxide synthase pathway in cardiac myocytes. Circulation 1997; 96:1937–1943. 91. Lau S, Patnaik N, Sayen R, Mestril R. Simultaneous over-expression of two stress proteins in rat cardiomyocytes and myogenic cells confers protection against ischemiainduced injury. Circulation 1997; 96:2287–2294.

288

Wolfel and Levine

92. Zimmerman LH, Levine RA, Farber HW. Hypoxia induces a set of stress proteins in cultured endothelial cells. J Clin Invest 1991; 87:908–914. 93. Hirata K, Ban T, Jinnouchi Y, Kubo S. Echocardiographic assessment of left ventricular function and wall motion at high altitude in normal subjects. Am J Cardiol 1991; 68:1692–1697. 94. Sagawa K, Maughan L, Suga H, Sunagawa K. Physiologic determinants of the ventricular pressure-volume relationship. In: Sagawa K, Maughan L, Suga H, Sunagawa K, eds. Cardiac Contraction and the Pressure-Volume Relationship. New York: Oxford University Press, 1988:110–170. 95. Mirsky I. Assessment of diastolic function: suggested methods and future considerations. Circulation 1984; 69:836–841. 96. Levine BD, Zuckerman JH, Pawelczyk JA. Cardiac atrophy after bed-rest deconditioning: A non-neural mechanism for orthostatic intolerance. Circulation 1997; 96:517– 525. 97. Jung RC, Dill DB, Horton R, Horvath SM. Effect of age on plasma aldosterone levels and hemoconcentration at altitude. J Appl Physiol 1971; 31:593–597. 98. Surks MI, Chinn KSK, Matoush LO. Alterations in body composition in man after acute exposure to high altitude. J Appl Physiol 1966; 21:1741–1746. 99. Grover RF, Weil JV, Reeves JT. Cardiovascular adaptations to exercise at high altitude. Exer Sports Sci Rev 1986; 14:269–302. 100. Grover RF, Reeves JT, Maher JT, McCullough RE, Cruz JC, Denniston JC, Cymerman A. Maintained stroke volume but impaired arterial oxygenation in man at high altitude with supplemental CO2. Circ Res 1976; 38:391–396. 101. Grover RF, Selland MA, McCullough RG, Dahms TE, Wolfel EE, Butterfield GE, Reeves JT, Greenleaf JE. β-Adrenergic blockade does not prevent polycythemia or decrease in plasma volume in men at 4,300 m altitude. Eur J Appl Physiol 1998; 77: 264–270. 102. Tyberg JV. Coupled or uncoupled pericardial constraint: Effects on cardiac chamber interactions. J Appl Physiol 1996; 81:1797–1798. 103. Richalet JP. Operation COMEX. In: Hypoxia: Into the Next Millennium, Proceedings of the 11th International Hypoxia Symposium, Jasper, Alberta, Canada, 1999. 104. Ross J Jr, Linhart JW, Braunwald E. Effect of changing heart rate in man by electrical stimulation of the right atrium: studies at rest, during exercise, and with isoproterenol. Circulation 1965; 32:549–556. 105. Hultgren H, Kelly J, Miller H. Pulmonary circulation in acclimatized man at high altitude. J Appl Physiol 1965; 20:233–238. 106. Groves BM, Reeves JT, Sutton JR, Wagner PD, Cymerman A, Malconian MK, Rock PB, Young PM, Houston CS. Operation Everest II: elevated high-altitude pulmonary resistance unresponsive to oxygen. J Appl Physiol 1987; 63:521–530. 107. Das BK, Tewari SC, Parashar SK, Akhtar M, Grover DN, Ohri VC, Dutta SK, Chatterjee JC. Electrocardiographic changes at high altitude. Indian Heart J 1983; 35:30– 37. 108. Abel FL, Waldhausen JA. Effects of alterations in pulmonary vascular resistance on right ventricular function. J Thorac Cardiovasc Surg 1967; 54:886–894. 109. Davila-Roman VG, Guest TM, Tuteur PG, Rowe WJ, Ladenson JH, Jaffe AS. Transient right but not left ventricular dysfunction after strenuous exercise at high altitude. J Am Coll Cardiol 1997; 30:468–473. 110. Garcia JA, McMinn SB, Zuckerman JH, Fixler DE, Levine BD. The role of the right

Cardiovascular System at High Altitude

111.

112. 113. 114.

115. 116.

117. 118.

119. 120.

121.

122.

123.

124. 125.

126. 127. 128. 129.

289

ventricle during hypobaric hypoxic exercise: insights from patients after the Fontan operation. Med Sci Sports Exerc 1999; 31:269–276. Atherton JJ, Moore TD, Lele SS, Thomson HL, Galbraith AJ, Belenkie I, Tyberg JV, Frenneaux MP. Diastolic ventricular interaction in chronic heart failure. Lancet 1997; 349:1720–1724. Cargill RI, Kiely DG, Lipworth BJ. Adverse effects of hypoxaemia on diastolic filling in humans. Clin Sci 1995; 89:165–169. Garcia MJ, JD Thomas, Klein AL. New Doppler echocardiographic applications for the study of diastolic function. J Am Coll Cardiol 1998; 32:865–875. Buckey JC, Gaffney FA, Lane LD, Levine BD, Watenpaugh DE, Wright SJ, Yancy CW Jr, Blomqvist CG. Central venous pressure in space. J Appl Physiol 1996; 81: 19–25. White RJ, Blomqvist CG. Central venous pressure and cardiac function during spaceflight. J Appl Physiol 1998; 85:738–746. Mazzeo RS, Brooks GA, Butterfield GE, Podolin DA, Wolfel EE, Reeves JT. Acclimatization to high altitude increases muscle sympathetic activity both at rest and during exercise. Am J Physiol 1995; 269:R201–207. Sander M, Hansen J, Saltin B. Unpublished observations. Bender PR, Groves BM, McCullough RE, McCullough RG, Huang SY, Hamilton AJ, Wagner PD, Cymerman A, Reeves JT. Oxygen transport to exercising leg in chronic hypoxia. J Appl Physiol 1988; 65:2592–2597. Weil JV, Byrne-Quinn E, Battock DJ, Grover RF, Chidsey CA. Forearm circulation in man at high altitude. Clin Sci 1971; 40:235–246. Cruz JC, Grover RF, Reeves JT, Maher JT, Cymerman A, Denniston JC. Sustained venoconstriction in man supplemented with CO2 at high altitude. J Appl Physiol 1976; 40:96–100. Cerretelli P, Marconi C, Deriaz O, Giezendanner D. After effects of chronic hypoxia on cardiac output and muscle blood flow at rest and exercise. Eur J Appl Physiol 1984; 53:92–96. Hellems HK, Ord JW, Talmers FN, Christensen RC. Effects of hypoxia on coronary blood flow and myocardial metabolism in normal human subjects. Circulation 1963; 16:893–898. Moret PR. Coronary blood flow and myocardial metabolism in man at high altitude. In: Porter R, Knight J, eds. High Altitude Physiology: Cardiac and Respiratory Aspects. Edinburgh: Churchill Livingstone, 1971:131–148. Grover RF, Lufschanowski R, Alexander JK. Alterations in the coronary circulation of man following ascent to 3,100 m altitude. J Appl Physiol 1976; 41:832–838. Sharma SC, Balasubramanian V, Mathew OP, Hoon RS. Serial studies of heart rate, blood pressure, and urinary catecholamine excretion on acute induction to high altitude (3,658 m). Indian J Dis Chest 1977; 19:16–20. Vogel JA, Hartley LH, Cruz JC, Hogan RP. Cardiac output during exercise in sealevel residents at sea level and high altitude. J Appl Physiol 1974; 36:169–172. Vogel JA, Hansen JE, Harris CW. Cardiovascular responses in man during exhaustive work at sea level and high altitude. J Appl Physiol 1967; 23:531–539. Saltin B, Grover RF, Blomqvist CG, Hartley LH, Johnson RL Jr. Maximal oxygen uptake and cardiac output after 2 weeks at 4,300 m. J Appl Physiol 1968; 25:400–409. Wolfel EE, Reeves JT, Lindenfeld J, Rock PB, Fulco C, Muza S, Butterfield GE, Braun B, Zamudio S, Moore LG. Women at altitude: Reduction in peak exercise systemic

290

130.

131.

132. 133. 134. 135.

136.

137. 138.

139.

140. 141.

142.

143.

144. 145. 146.

Wolfel and Levine blood pressure with alpha-adrenergic blockade upon exposure to simulated altitude. FASEB J 1998; 12:A56. Muza SR, Rock PB, Butterfield GE, Wolfel EE, Moore LG. Women at Altitude: Effect of alpha-blockade on the ventilatory and systemic arterial pressure responses to hypoxia and hypercapnia. 11th International Hypoxia Symposium Proceedings, Jasper, Alberta, Canada, 1999. Wolfel EE, Tokeshi J, Meertens M, McCullough RG, Rock PB, Butterfield GE, Moore LG. Alpha-adrenergic blockade limits the increase in ambulatory blood pressure observed in young women after short-term residence at 4,300 m altitude. FASEB J 1999; 13:A784. Pugh LGCE. Cardiac output in muscular exercise at 5,800 m (19,000 ft). J Appl Physiol 1964; 19:441–447. Schneider EC, Hedblom CA. Blood pressure with special reference to high altitudes. Am J Physiol 1908; 23:90–104. Sharma SC, Hoon RS, Balasubramanian V, Chadha KS. Urinary catecholamine excretion in temporary residents of high altitude. J Appl Physiol 1978; 44:725–727. Marticorena E, Ruiz L, Severino J, Galvez J, Penaloza D. Systemic blood pressure in white men born at sea level: changes after long residence at high altitudes. Am J Cardiol 1969; 23:364–368. Lewis SF, Taylor WF, Graham RM, Pettinger WA, Schutte JE, Blomqvist CG. Cardiovascular responses to exercise as functions of absolute and relative work load. J Appl Physiol 1983; 54:1314–1323. Roach R. Exercise at high altitude. Symposium on Mountain Medicine. Med Sci Sports Exer 1999; 31(suppl):S–39. Sutton JR, Reeves JT, Wagner PD, Groves BM, Cymerman A, Malconian MK, Rock PB, Young PM, Walter SD, Houston CS. Operation Everest II: oxygen transport at extreme simulated altitude. J Appl Physiol 1988; 64:1309–1321. Thomson JM, Dempsey JA, Chosy LW, Shahidi NT, Reddan WG. Oxygen transport and oxyhemoglobin dissociation during prolonged muscular work. J Appl Physiol 1974; 37:658–664. Dempsey JA, Thomson JM, Forster HV, Cemy FC, Chosy LW. HBO2 dissociation in man during prolonged work in chronic hypoxia. J Appl Physiol 1975; 38:1022–1029. Ferretti G, Boutellier U, Pendergast DR, Moia C, Minetti AE, Howald H, di Prampero PE. Oxygen transport system before and after exposure to chronic hypoxia. Int J Sports Med 1990; 11(suppl 1):S15–S20. Steinacker JM, Liu Y, Boning D, Halder A, Maassen N, Thomas A, Stauch M. Lung diffusion capacity, oxygen uptake, cardiac output and oxygen transport during exercise before and after an Himalayan expedition. Eur J Appl Physiol 1996; 74:187–193. Moore LG, Cymerman A, Shao-Yung H, McCullough RE, McCullough RG, Rock PB, Young A, Young PM, Bloedow D, Weil JV, Reeves JT. Propranolol does not impair exercise oxygen uptake in normal men at high altitude. J Appl Physiol 1986; 61:1935–1941. Fan F, Chen RYZ, Schuessler GB, Chien S. Effects of hematocrit variations on regional hemodynamics and oxygen transport in the dog. Am J Physiol 1980; 238:H545–H552. Lindenfeld J, Weil JV, Travis VL, Horwitz LD. Hemodynamic responses to normovolemic polycythemia at rest and during exercise in dogs. Circ Res 1985; 56:793–800. Lindenfeld J, Weil JV, Horwitz LD. Oxygen content, not viscosity regulates cardiac output during induced polycythemia in dogs. Clin Res 1983; 31:201A.

Cardiovascular System at High Altitude

291

147. Cerretelli P. Limiting factors to oxygen transport on Mount Everest. J Appl Physiol 1976; 40:658–667. 148. West JB, Boyer SJ, Graber DJ, Hackett PH, Maret KH, Milledge JS, Peters RM Jr, Pizzo CJ, Samaja M, Sarnquist FH, Schoene RB, Winslow RM. Maximal exercise at extreme altitudes on Mount Everest. J Appl Physiol 1983; 55:688–698. 149. Wagner PD. Why does the maximal exercise cardiac output fall during altitude residence and is it important? Proceedings of the 11th International Hypoxia Symposium, Jasper, Alberta, Canada, 1999. 150. Reeves JT, Grover RF, Cohn JE. Regulation of ventilation during exercise at 10,200 feet in athletes born at low altitude. J Appl Physiol 1967; 22:546–554. 151. Hartley LH, Vogel JA, Cruz JC. Reduction of maximal exercise heart rate at altitude and its reversal with atropine. J Appl Physiol 1974; 36:362–365. 152. Zhuang J, Droma T, Sutton JR, McCullough RE, McCullough RG, Groves BM, Rapmund G, Janes C, Sun SF, Moore LG. Autonomic regulation of heart rate response to exercise in Tibetan and Han residents of Lhasa (3,658 m). J Appl Physiol 1993; 75:1968–1973. 153. Young AJ, Sawka MN, Muza SR, Boushel R, Lyons T, Rock PB, Freund BJ, Waters R, Cymerman A, Pandolf KB, Valeri CR. Effects of erythrocyte infusion on VO2 max at high altitude. J Appl Physiol 1996; 81:252–259. 154. Rowell LB. Why does the maximal exercise cardiac output fall during altitude residence and is it important? Proceedings of the 11th International Hypoxia Symposium, Jasper, Alberta, Canada, 1999. 155. Moret P, Covarrubias E, Coudert J, Duchosal F. Cardiocirculatory adaptations to chronic hypoxia. Acta Cardiol 1972; 27:285–305. 156. Groves BM, Droma T, Sutton JR, McCullough RG, McCullough RE, Zhuang J, Rapmund G, Sun S, Janes C, Moore LG. Minimal hypoxic pulmonary hypertension in normal Tibetans at 3,658 m. J Appl Physiol 1993; 74:312–318. 157. Vogel JA, Hartley LH, Cruz JC. Cardiac output during exercise in altitude natives at sea level and high altitude. J Appl Physiol 1974; 36:173–176. 158. Sime F, Penaloza D, Ruiz L. Bradycardia, increased cardiac output, and reversal of pulmonary hypertension in altitude natives living at sea level. Br Heart J 1971; 33: 647–657. 159. Moret P. Myocardial metabolism: acute and chronic adaptation to hypoxia. Med Sport Sci 1985; 19:48–63. 160. Hochachka PW, Clark CM, Holden JE, Stanley C, Ugurbil K, Mennon RS. 31P magnetic resonance spectroscopy of the Sherpa heart: A phosphocreatine/adenosine triphosphate signature of metabolic defense against hypoxia. Proc Natl Acad Sci USA 1996; 93:1215–1220. 161. Holden JE, Stone CK, Clark CM, Brown WD, Nickles RJ, Stanley C, Hochachka PW. Enhanced cardiac metabolism of glucose in high-altitude natives: adaptation against chronic hypoxia. J Appl Physiol 1995; 79:222–228. 162. Brooks GA, Butterfield GE, Wolfe RR, Groves BM, Mazzeo RS, Sutton JR, Wolfel EE, Reeves JT. Increased dependence on blood glucose after acclimatization to 4,300 m. J Appl Physiol 1991; 70:919–927. 163. Brooks GA, Butterfield GE, Wolfe RR, Groves BM, Mazzeo RS, Sutton JR, Wolfel EE, Reeves JT. Decreased reliance on lactate during exercise after acclimatization to 4,300 m. J Appl Physiol 1991; 71:333–341. 164. Brooks GA, Wolfel EE, Groves BM, Bender PR, Butterfield GE, Cymerman A, Maz-

292

165.

166. 167. 168.

169.

170.

171. 172. 173. 174. 175.

176.

177. 178. 179.

180.

181.

182.

Wolfel and Levine zeo RS, Sutton JR, Wolfe RR, Reeves JT. Muscle accounts for glucose disposal but not lactate appearance during exercise after acclimatization to 4,300 m. J Appl Physiol 1992; 72:2435–2445. Arias-Stella J, Topilsky M. Anatomy of the coronary circulation at high altitude. In: Porter R, Knight J, eds. High Altitude Physiology: Cardiac and Respiratory Aspects. Edinburgh: Churchill Livingstone, 1971:149–157. Fuster V, Frye RL, Kennedy MA, Connolly DC, Makin HT. The role of collateral circulation in the various coronary syndromes. Circulation 1979; 59:1137–1144. Mortimer EA Jr, Monson RR, MacMahon B. Reduction in mortality from coronary heart disease in men residing at high altitude. N Engl J Med 1977; 296:581–585. Ludmer PL, Selwyn AP, Shook TL, Wayne RR, Mudge GH, Alexander RW, Ganz P. Paradoxical vasoconstriction induced by acetylcholine in atherosclerotic coronary arteries. N Engl J Med 1986; 315:1046–1051. Gage JE, Hess OM, Murakami T, Ritter M, Grimm J, Krayenbuehl HP. Vasoconstriction of stenotic coronary arteries during dynamic exercise in patients with classic angina pectoris: reversibility by nitroglycerin. Circulation 1986; 73:865–876. Vita JA, Treasure CB, Yeung AC, Vekshtein VI, Fantasia GM, Fish RD, Ganz P, Selwyn AP. Patients with evidence of coronary endothelial dysfunction as assessed by acetylcholine infusion demonstrate marked increase in sensitivity to constrictor effects of catecholamines. Circulation 1992; 85:1390–1397. Pohl U, Busse R. Hypoxia stimulates the release of endothelium-derived relaxant factor (EDRF). Am J Physiol 1989; 256:H1595–H1600. Busse R, Pohl U, Kellner, Klemm U. Endothelial cells are involved in the vasodilator response to hypoxia. Pflugers Arch 1983; 397:78–80. Fuster V. Mechanisms leading to myocardial infarction: Insights from studies of vascular biology. Circulation 1994; 90:2126–2146. Bartsch P, Waber U, Haeberli A, et al. Enhanced fibrin formation in high altitude pulmonary edema. J Appl Physiol 1987; 63:752–757. Burke AP, Farb A, Malcom GT, Liang Y, Smialek JE, Virmani R. Plaque rupture and sudden death related to exertion in men with coronary artery disease. JAMA 1999; 281:921–926. Morgan BJ, Alexander JK, Nicoli SA, Brammell HL. The patient with coronary disease at altitude: observations during acute exposure to 3,100 m. J Wilderness Med 1990; 1:147–153. Alexander JK. Personal communication. Unpublished results. Grover RF, Tucker CE, Robinson-McGroarty S, Travis RR. The coronary stress of skiing at high altitude. Arch Intern Med 1990; 150:1205–1208. Erdmann J, Sun KT, Masar P, Niederhauser H. Effects of exposure to altitude on men with coronary artery disease and impaired left ventricular function. Am J Cardiol 1998; 81:266–270. Alexander JK, Abinader EG, Sharif DS, Morgan BJ, Brammell HL. Left ventricular function in coronary heart disease at high altitude. Circulation 1988; 78(suppl II):II6. Agostoni P, Cattadori G, Guazzi M, Bussotti M, Conca C, Lomanto M, Marenzi G, Guazzi MD. Exercise capacity in heart failure patients during an hypoxia-altitude simulation test. J Am Coll Cardiol 1999; 33(suppl A):195A. Hultgren HN. Coronary heart disease and trekking. J Wilderness Med 1990; 1:154– 161.

10 The Pulmonary Circulation at High Altitude

JOHN T. REEVES and KURT R. STENMARK University of Colorado Health Sciences Center Denver, Colorado

I.

Introduction

The pulmonary circulation changes profoundly at high altitude as pulmonary hypertension appears. Modest pulmonary hypertension at high altitude may be beneficial in some circumstances, but severe pulmonary hypertension is nearly always detrimental. On arrival at altitude, acute hypoxia raises pulmonary arterial pressure, and with the chronic hypoxia of continued residence, pressure rises even more. But the relationship between the acute and the chronic hypoxic responses is far from clear. We hypothesize that the key to the magnitude of the pulmonary hypertension at altitude is how the pulmonary arteriolar walls thicken in the transition from arrival to chronic residence. Understanding this transition is difficult because both acute hypoxic vasoconstriction and chronic thickening of the arteriolar walls are extremely variable within and between species, and the mechanisms for either are only dimly seen. Our approach in the following review is to use available data to describe the lung circulation before, during, and after the transition from arrival at altitude to chronic residence and to examine possible mechanisms involved. We will use human data where possible as well as animal data where necessary. The susceptibility of 293

294

Reeves and Stenmark

the lung circulation to pulmonary hypertension at altitude seems to be greatest in the newborn because the mechanisms which have maintained high lung pressures and resistance in fetal life are powerfully retained in the neonate (69). We will therefore examine in detail the altitude effects in the very young. Most of the relevant human data are from young adults, from whom vascular reactivity and the responses to exercise provide clues to altered lung circulatory control at altitude, and these also will be examined in some detail. The aged may be susceptible to the pulmonary hypertensive effects of altitude when they suffer the ravages of heart and lung diseases as presented elsewhere (95). We refer the reader to a recent review by Grover (27) and other excellent reviews (23,30,69,92,93,108,115,116,123,124). Perhaps to appreciate where we are and how far we have come, it is important to look specifically at the history of the lung circulation at high altitude. In doing so, we are struck by two facts: first, the history is quite brief, and second, the important foundations of our knowledge were laid by South American investigators working in the Andes.

II. History The general history of humans at high altitude going back for several centuries is reported elsewhere in this volume. The body of pulmonary circulatory research conducted in humans between the years 1928 and 1962, particularly by investigators based in Lima, Peru, is remarkable and has been amply confirmed by subsequent workers. In order to follow the development of ideas, we have attempted to highlight the salient features of these studies and put them into context with other concepts simultaneously developed at low and high altitudes around the world. Observations related to the lung circulation in humans at high altitude began about 1928 (64–67) when Carlos Monge (Lima) described heart failure and an increase in the second heart tone in a subset of men with chronic polycythemia (often called Monge’s disease), thus raising the possibility of increased pulmonary arterial pressure. Hurtado, a colleague of Monge, noted in a 1942 review (48) that no direct measurement had been made of the pulmonary arterial pressure at high altitudes, but its possible elevation may be inferred from the permanent congestive condition (46,47) and the greater volume of blood contained in the lungs (47) and from the high frequency of right ventricular preponderance in the electrocardiograms taken in apparently healthy natives (12). High altitude residence was considered the culprit, because all signs and symptoms resolved when the polycythemic patient went to live at low altitude (46). Hurtado had referred to the large Argentinean expedition of 1936 to the Bolivian Altiplano led by Capdehourat (12) (Fig. 1). Electrocardiograms had been obtained at approximately 4000 m in young men who were asymptomatic and who had passed the preemployment physical examination to work in the mine at Catavi. The electrocardiograms suggested right ventricular hypertrophy because they showed a mean QRS electrical axis in the frontal plane (standard leads I, II, and

Pulmonary Circulation

295

Figure 1 Photograph of E. L. Capdehourat, right, showing the President of Argentina some of the equipment for the expedition to the Bolivian Altiplano (12).

III) exceeding ⫹90° in more than 20% of the 78 subjects examined, four times the percentage expected in sea level dwellers (7). Histograms of the electrical axis (Fig. 2) show the skewing to the right in the Bolivian high-altitude natives. Subsequent electrocardiographic studies in children and adults of Leadville, Colorado (3100 m), showed findings similar to those obtained at 4000 m in Peru (84). To our knowledge, the electrocardiograms from the Altiplano were the first quantitative data ever reported that suggested right ventricular hypertrophy (and hence pulmonary hypertension) in normal human residents at high altitude. In 1938 Rotta reported that chest radiographs showed larger hearts in normal residents living at high altitude in Peru than in residents at sea level (Lima), but he was unable to specify which chambers were enlarged. However, at the 1946 Congress of Cardiology in Mexico City he presented a comprehensive study (100) of young men native to 4540 m in Morococha, Peru, compared with those living in Lima at 200 m. The distribution of transverse cardiac diameters indicated radiographic cardiac enlargement at altitude (Fig. 3A). Further, using cardiac fluoroscopy and electrocardiography, he specified that the right ventricle was enlarged. Also, he obtained phonocardiograms, which suggested the presence of pulmonary hypertension, and he noted an elevated venous pressure, which indicated increased resistance to filling of the right ventricle. The two decades of studies from the high altitudes of Bolivia and Peru were prescient because they predicted (48) elevation of pulmonary

296

Reeves and Stenmark

Figure 2 Histogram of the electrocardiographic frontal mean QRS axis as reported from normal adult residents of the Bolivian Altiplano (12) (crossed circles), and sea level (New Orleans, 7) (filled circles). An axis of ⱖ90° is considered right axis deviation and is consistent with a large right ventricular muscle mass. Seen is that the high-altitude populations are skewed toward greater right axis deviations. QRS axis from 3100 m (not shown) (84) is more to right than at sea level but less rightward than from the Bolivian Altiplano. (Adapted from Refs. 7,12.)

arterial pressure before hypoxia was known to cause acute pulmonary vasoconstriction and before cardiac catheterization was available to measure pulmonary arterial pressure. However, there were rapid developments elsewhere. In 1946 von Euler and Liljestrand demonstrated that acute hypoxia raised the pulmonary arterial pressures of cats (119). About the same time Cournand reported measuring pulmonary arterial pressure in humans during cardiac catheterization (15), and in 1947 his laboratory found acute hypoxia raised human pulmonary arterial pressure (for reviews, see Refs. 23, 72). Maybe acute hypoxia on arrival at high altitude increased pulmonary arterial pressure and continued residence increased it further. Rotta and colleagues lost little time in setting about to make the measurements, and in 1949 (99) and 1952 (98) they reported the first pulmonary arterial pressures ever obtained in any species at high altitude. The full reports in English (80,97) were a tour de force in that pulmonary arterial pressures and blood flows were measured (1) in native resi-

Pulmonary Circulation

297

dents of Lima, (2) in native high-altitude residents of Morococha at 4540 m, (3) in newcomers living for 1 year at Morococha, (4) in residents of Morococha with chronic mountain sickness (Monge’s disease), and (5) in some of the high-altitude residents during the breathing of 35% oxygen to simulate sea level Po 2. Their measurements established the co-existence of arterial oxygen desaturation and pulmonary hypertension (Fig. 3B). Thus residents at high altitude with low oxygen saturations also had elevated pulmonary arterial pressure, supporting the concept that chronic hypoxia caused chronic pulmonary hypertension. In addition, the data showed that newcomers to Morococha had less hypoxemia and less pulmonary hypertension than did healthy people born and living there (Fig. 3B). Further, two subjects who had lived in Morococha for more than 25 years and who had excessive polycythemia (chronic mountain sickness) had the most severe hypoxemia and the highest pulmonary arterial pressures. Taken together (Fig. 3B), the results indicated that the elevated pressure was related to the severity of the hypoxia and that contributing factors were likely to be the duration of residence at altitude and the presence of polycythemia. Rotta et al. must have expected these findings based on prior clinical, laboratory, and literature findings. However, they could not have expected the poor pulmonary vasodilator responses when their subjects breathed a high oxygen mixture (97). While 35% oxygen tended to lower pulmonary arterial pressure in the various groups, the cardiac output fell more than the pressure in each instance. Thus, supplemental oxygen tended to raise, not lower, total pulmonary resistance! How were they to explain, then, that acute reversal of the chronic hypoxia failed to reduce the elevated vascular resistance? The suggestion by Rotta et al. (97) showed considerable insight: ‘‘anoxia cannot be considered as a possible isolated factor in the pulmonary hypertension.’’ Clearly these authors realized that the pulmonary hypertension of chronic hypoxia was something more than the maintenance of acute hypoxic vasoconstriction. One possibility they considered was that 15–25 minutes of high oxygen might not be sufficient to reverse the elevated resistance, and, indeed, in a subsequent study (80), Sime et al. showed normalization of pressures and resistances in normal or polycythemic men taken for 2 years to sea level (104). As discussed below, we continue to debate the sequence of events and mechanisms by which chronic pulmonary hypertension at high altitude replaces acute hypoxic vasoconstriction. So far, the measurements and observations in the Andes had been largely confined to adult men, but events unfolding simultaneously in England pointed to the importance of studying children at high altitude, especially in the immediate newborn period. In Cambridge, Barcroft, who had begun to study the fetal circulation in the late 1920s, and knowing of the low Po 2 in fetal blood, referred to ‘‘Mt. Everest in utero’’ (9). Geoffrey Dawes, who continued the Barcroft tradition, began his classic studies in the lamb in 1948 at the Nuffield Institute in Oxford. Research in his laboratory showed in the near-term fetus that inflating the lung with air or oxygen at birth caused pressure to fall in the pulmonary artery, but not in the aorta (17). Within a few days after a normal birth of lambs, pulmonary pressure had fallen

298

Reeves and Stenmark

Figure 3 Early measurements from Peru relating to the lung circulation. (A) Histograms of the transverse diameter of the heart from frontal chest roentgenograms in 200 native residents of Lima at 200 m elevation (filled circles) and in 200 native residents of Morococha at 4540 m elevation (unfilled circles). (Adapted from Ref. 100.) Subjects at high altitude, but not at sea level, had hearts that were larger than predicted normal. (B) Mean pulmonary arterial pressures as related to systemic arterial oxygen saturations from individual adult native residents of Lima (filled circles), from Lima residents who had lived at Morococha for one year (partially filled circles), from 24-year-old healthy residents of Morococha (unfilled circles), and from two Morococha residents older than 25 years with chronic mountain sickness (CMS, ⫹). Shown is the curvilinear regression (unbroken line) through all data. Seen is that

Pulmonary Circulation

299

to the stable nadir that would be maintained throughout life (17). Similar findings were obtained in human newborns (1). One potentially important mechanism for survival of mammalian species might be hypoxic vasoconstriction in the fetal lung to limit blood flow while the organ was relatively dormant and fluid filled. However after birth, high lung blood flow becomes essential for oxygenation of the newborn, and reversal of hypoxic pulmonary vasoconstriction is one of several mechanisms promoting the increased lung blood flow. But what if the neonate breathed an hypoxic gas after birth? In 1957 Rowe showed (101) that even brief inhalation of lowoxygen mixtures caused large pulmonary hypoxic vasopressor responses. Thus, children born at high altitude might not enjoy a rapid relief of their fetal pulmonary hypertension because their alveoli would be filled with air at a low Po 2. The newborns at altitude might be denied the full vasodilating effect of the high Po 2 available to newborns at sea level. It was important, then, that the Peruvian investigators look for clues relative to the persistence of pulmonary hypertension following birth at high altitude. Their strategy was first to use noninvasive measurements to compare, for varying postnatal ages, children born at sea level with those born at altitude. Naturally they turned to the children in Lima and Morococha, respectively (77,78). Using the electrocardiogram to provide an index of right ventricular muscle mass, they found that at birth there was comparable right axis deviation at birth in both Lima and Morococha. These findings were consistent with high pulmonary artery pressures and thick right ventricles (Fig. 3C), known to exist in the fetus. However, findings from the two populations quickly diverged. Of 350 children examined in Lima, the right axis deviation rapidly regressed after birth to normal sea level values (Fig. 3C), but the axes of the 190 children in Morococha did not. Rather, in Morococha the right axis deviation persisted near the newborn levels. The data implied a rapid atrophy of the right ventricle at sea level, but not at 4540 m. Even in 14-year-olds in Morococha, the electrocardiographic axis was not different from that of the newborn. The inspired Po 2 in Morococha is of the order of 80 mmHg, and the alveolar Po 2 would therefore be expected to be between approximately 40–50 mmHg, which apparently is low enough to delay or even prevent the normal postnatal regression of the right ventricle. Because the electrocardiogram is an insensitive indicator of ventricular hypertrophy, Arias-Stella and Recavarren (6) sought direct evidence by weighing the right and left ventricles from children of Lima and Morococha who died from accidents or acute illness. At birth, the ratios of right ventricular to total ventricular weights

pressures appeared to increase with hypoxemia, were higher in natives than in newcomers, and were highest in CMS patients. (Adapted from Ref. 97.) (C) Mean frontal electrocardiographic axis of the QRS complex at various ages in healthy residents of Lima (filled circles) and Morococha (unfilled circles). Seen is that the right axis deviation persisted after birth at high altitude but not at sea level. (Adapted from Ref. 77.)

300

Reeves and Stenmark

showed no difference between high- and low-altitude children (Fig. 4A), supporting the electrocardiographic findings. Also supporting the electrocardiogram were the decreasing ratios of right to total ventricular weight as children grew older at sea level and the much smaller decreases in the weight ratios with age in Morococha (Fig. 4A). The results were consistent with the persistence of right ventricular hypertrophy up to 10 years, the age of the oldest children from whom hearts were obtained. Thus, the electrocardiogram and the postmortem examination of the right ventricle pointed to a persistence of pulmonary hypertension following birth, but there had been no histological examinations of the pulmonary vascular tree at different ages after birth at high altitude, and the time was ripe for such an examination. In 1951 in Minnesota, Civin and Edwards (13) had shown for humans that fetal pulmonary arterioles had thick media but that the media became thin after birth. Wagenvoort (120), then in Edwards’s laboratory, published quantitative measurements showing the time course of the postnatal thinning to adult levels by 6–12 weeks after birth. Because chronic hypoxia after birth at high altitude might sustain lung arteriolar thickening, Arias-Stella and Saldan˜a (5) examined the lungs from persons dying at low and high altitudes. Their findings were that arteriolar medial thickness decreased, as expected, after birth at sea level, but at high altitude thickened media persisted for many months and, to some extent, throughout life (Fig. 4B). Taken together, all of these data suggested the persistence of fetal pulmonary hypertension after birth at high altitude, but there had been no pressure measurements in healthy infants following birth at high altitude. The measurements available from sea level showed a prompt fall in pressure within minutes to hours after birth (Fig. 5A) (1,102,103). To obtain high-altitude data, cardiac catheterization measurements were made in healthy children in Morococha by Sime et al. (105) and reported initially by Penalosa et al. in 1962 (80). They showed that the pulmonary arterial pressures were sustained near the fetal levels for weeks and months after birth (Fig. 5A). These early studies in humans suggested that infants might be particularly susceptible to chronic hypoxic pulmonary hypertension. Cardiac catheterization studies conducted nearly simultaneously at 3100 m in Leadville, Colorado (117), showed that elevated pulmonary arterial pressures persisted for years in adolescents and young adults (Fig. 5A). Thus by June of 1962, South American investigators, together with those from Colorado, had set in place principles relating to the lung circulation at high altitude: 1. 2. 3. 4.

Residence at high altitude raised resting pulmonary arterial pressure in newcomers and in the native born. Certain life-long residents who had hypoxemia and excessive polycythemia (Monge’s disease) also had excessive pulmonary hypertension. The pulmonary hypertension in normal or polycythemic individuals could be reversed by a period of residence near sea level. In children born at high altitude, there were failures of the normal regression in right ventricular hypertrophy, the regression of pulmonary arterio-

Pulmonary Circulation

301

Figure 4 Early measurements from Peru relating to the right ventricle and hypertrophy of lung arterioles. (A) Ratios of the weight of the right ventricle to total ventricular weight from persons of various ages dying acutely and thought to be previously healthy as obtained in Lima (sea level, filled circles) and Morococha (4540 m, unfilled circles). Curvilinear regression lines are drawn through each data set. Seen are that high right ventricular weight ratios persist at high altitude but not at sea level. (Data recalculated from Ref. 6.) (B) Pulmonary arteriolar medial smooth muscle in residents of various ages dying acutely in Lima (filled circles) or Morococha (unfilled circles). Seen is that the media thickness is increased at high altitude. (Data recalculated from Ref. 5.)

302

Reeves and Stenmark

Figure 5 Pulmonary circulatory pressures in residents of various altitudes. (A) Resting pulmonary arterial pressures at various ages following birth in normal residents of sea level (unbroken line schematized from Ref. 75), 3100 m (replotted from Refs. 23,100), and 4540 m (replotted from Ref. 88). Compared to sea level, pulmonary hypertension persists after birth at both high altitudes shown. (B) Resting pulmonary arterial pressures (open circles) and pulmonary arterial minus wedge (filled circles) in native residents of various altitudes. (Data replotted from Refs. 8,36,44,97,117.)

Pulmonary Circulation

303

lar vascular smooth muscle, and the fall in pulmonary arterial pressure and resistance. 5. As discussed below, exercise, even in healthy residents, was accompanied by greater pulmonary hypertension than at sea level and a blunting of an exercise-related fall in vascular resistance. Subsequent investigators who have measured pulmonary hemodynamics have shown the altitude-related pulmonary hypertension in terms of pulmonary arterial pressure as well as the pressure gradient from pulmonary artery to wedge (Fig. 5B). However, the early investigations from the Andes set the challenge for the search for the mechanisms by which high altitude altered the pulmonary circulation. The mechanisms must take account of differences between acute hypoxic pulmonary vasoconstriction and chronic altitude-induced hypertrophy/hyperplasia of the pulmonary arterioles, both of which seem particularly to affect the newborn. The current chapter focuses on these mechanisms.

III. Mechanisms A. Acute Hypoxic Pulmonary Vasoconstriction Direct Effects on Smooth Muscle

The question of the mechanism of hypoxic pulmonary vasoconstriction has been the subject of intense investigation around the world since von Euler’s report (119). Hypoxic vasoconstriction is intrinsic, and probably unique, to the lung circulation. Early investigations by Duke and Killick (18) and subsequently by others, established that lungs removed from the body and artificially perfused, with or without blood (58), retained their capacity for hypoxic pulmonary vasoconstriction. In the isolated lung, the pressor response develops progressively over 1–5 minutes during exposure to alveolar hypoxia and tends to regress rather more rapidly on return to normoxia (Fig. 6A). Further, isolated arterioles (⬍300 mm diameter) from lungs, but not from systemic organs, constricted in response to hypoxia (54). At issue was whether the constriction was mediated by a direct effect of hypoxia on the vascular smooth muscle cells or indirectly by vasoconstrictor substances from other cells. In evaluating this issue, Archer suggested criteria (3,4) that need to be satisfied. 1. The response is intrinsic to the lungs and lung arterioles. 2. The threshold for the response is a Po 2 of approximately 55 mmHg in isolated lungs (an alveolar Po 2 of 65 mmHg in intact man). 3. There is a rapid onset and offset of the constriction. 4. Calcium moves from outside to inside the smooth muscle cell. 5. There is depolarization of the smooth muscle membrane (Fig. 6B). While naturally occurring vasoactive substances, including catecholamines, histamine, angiotensin II, 5-hydroxytryptamine, prostaglandins, and leukotrienes, all

304

Reeves and Stenmark

Figure 6 (A) Schematized representation of hypoxic pulmonary pressor responses in blood perfused isolated rat lungs before (control) and after the administration of the calcium channel blocker, verapamil, and a compound facilitating calcium entry in to cells, Bay K8644. (Adapted from Refs. 60,124.) (B) Schematized decrease in the transmembrane potential of isolated pulmonary vascular smooth muscle cells during exposure to hypoxia. (Adapted from Ref. 124.)

have been found to influence the magnitude of the vasoconstrictor response, none of them have been shown to satisfy the above criteria (3,4,116,124). The possibility that the mechanism was not mediated through vasoactive substances, but more directly via the vascular smooth muscle was suggested by the requirement for extracellular calcium in the vasoconstrictor response to hypoxia (59,60). Blockade of calcium channels by verapamil inhibited the hypoxic pulmonary vasoconstriction in isolated lungs, while facilitation of calcium entry by a calcium channel agonist, such as the compound Bay K8466, augmented the response (Fig. 6A). Hypoxia appeared to depolarize the pulmonary vascular smooth muscle cell (Fig. 6B), allowing the entry of calcium, which then initiated the contraction (33,54,55). For the carotid body type I cells (i.e., the O 2-sensing cells), the possibility that the calcium entered the cell via a voltage-dependent K ⫹ channel was indicated by the experiments of Lopez-Barneo (53). If, as suggested by Torrance (111), there

Pulmonary Circulation

305

is a mechanism for O 2 sensing that is common to several different organs, then the pulmonary arterioles might be controlled in a way analogous to the carotid body. In smooth muscle from small (⬍600 mm) pulmonary arteries (82,83,124), and even from isolated pulmonary arterial smooth muscle cells (55,82), evidence has been accumulated supporting the role for K ⫹ channels in the hypoxic vasoconstrictor response. It seems likely that hypoxia inhibits an outward potassium current, resulting in depolarization of the pulmonary vascular smooth muscle cell membrane, which then allows calcium entry through voltage-dependent calcium channels followed by contraction (for review, see Ref. 124). By contrast, in systemic arteries hypoxia increases potassium outflow through ATP-dependent potassium channels resulting in membrane hyperpolarization and smooth muscle relaxation. These studies seem to be pointing to differences in K ⫹ channel populations and functions in large versus small lung arteries and in systemic versus pulmonary arterioles (124). We are moving toward a greater understanding of the mechanisms of hypoxic pulmonary vasoconstriction, where the emphasis is now on the membrane of the smooth muscle cell itself, and a modulating role for vasoactive substances (116). The questions now are more focused, relating to the type and structure of the K ⫹ channels involved, their mechanism of activation, how they transduce the Ca ⫹ entry, how hypoxia is specifically involved in the processes, and the influences of various modulating mechanisms. Modulation of Hypoxic Pulmonary Vasoconstriction

Hypoxic pulmonary vasoconstriction is widely variable among healthy persons (90). It also varies markedly with age of an individual and among the different mammalian species. The variability is probably related to the multiple factors that influence the magnitude of the response. The K ⫹ and Ca 2⫹ ion channels themselves are under multiple influences such as vascular tone, cytochrome P-450 enzymes, and a variety of eicosanoids (32,33). In addition, neighboring cells influence smooth muscle. The capacity of the endothelium to produce nitric oxide (NO) or prostacyclin with increasing shear stress, the release of constrictor or dilator substances from mast cells, or the presence of inflammatory mediators will all affect the tone of the lung arterioles and their response to acute hypoxia. The amount of the vascular smooth muscle and even matrix proteins are determinants of the response. Review of these and other factors can be found elsewhere (3,4,115,123). B. Transition from Acute Hypoxic Vasoconstriction to Chronic High-Altitude Pulmonary Hypertension

Acute hypoxic vasoconstriction and its modulation by chemical substances, as discussed above, are mechanisms likely set in motion when individuals first arrive at high altitude. But if chronic high-altitude pulmonary hypertension is not simply the persistence of acute vasoconstriction, then we must look to other mechanisms. As the focus of this review is related to the transition from acute to chronic hypertension, we must consider how the transition is controlled and whether it is reversible. Fi-

306

Reeves and Stenmark

nally, there is the philosophical question of whether high-altitude pulmonary hypertension serves a useful purpose. As is often the case with questions, they are easier to pose than to answer. However, the starting point must be a description of the time course of the pulmonary arterial pressure changes at high altitude in living individuals. Pulmonary Arterial Pressure Changes over Time at Altitude

Changing pulmonary arterial pressure over time at high altitude would be compatible with the concept that chronic pulmonary hypertension is not just the maintenance of acute hypoxic vasoconstriction. If, compared to arrival at altitude, for example, pressure subsequently increased more than could be accounted for by an increase in flow, then either functional or structural changes in the lung vascular bed would be expected. Selected and representative data from three species are examined below. Humans

In young men having cardiac catheters in place at 3800 m, pulmonary arterial pressure doubled from the sea level value in the first 24 hours and then remained elevated at this level for 3 days (Fig. 7) (50). Cardiac output rose on arrival and subsequently fell, resulting in a progressive increase over 3 days at altitude in the calculated total pulmonary resistance. Figure 7 also compares the measurements in these recent arrivals with life-long altitude residents of Leadville, Colorado (117). Even though the Leadville residents were at only 3100 m, they had higher pulmonary arterial pressures and higher calculated total pulmonary resistances than observed in subjects residing over 3 days at the higher altitude of 3800 m (50). Taken together, the data suggest that the pulmonary vascular resistance in humans increases over the first days at altitude, but with life-long residence, pressure and resistance increase further. Cattle

Because of their large size, cattle can have repeated catheterizations with relative ease, and as in humans, measurements in cattle have been made in the actual highaltitude environment. Thus, for example, hemodynamic data are available in 10 yearling steers brought from near sea level to reside for up to 9 weeks at 3900 m on Mt. Evans, Colorado (24). Pulmonary arterial pressure progressively rose (Fig. 7). In three animals, one of which developed right heart failure, pressures reached mean values of approximately 100 mmHg. Cardiac output was unchanged or decreased at altitude, and therefore total pulmonary resistance progressively rose (Fig. 7). The average increase in pressure was 47 mmHg at 3900 m, i.e., greater than the 9 mmHg increase caused by acute hypoxia at low altitude. A prior study (128) in cattle taken from sea level to 3050 m also showed large increases in pulmonary pressure and resistance. Rats

When placed in altitude or hypoxic chambers, rats’ pulmonary arterial pressure can be measured daily, but the catheterization procedure is usually done when the rats

Pulmonary Circulation

307

Figure 7 Left panels: Pulmonary arterial pressure (PPA, top), blood flow (Q, middle), and total pulmonary resistance (TPR, bottom) in humans during 3 days at 3800 m. (Adapted from Ref. 41.) Also shown for comparison, in filled circles, are pressure, flow, and resistance group means from young adults living an average of 16 years in Leadville, Colorado, at 3100 m elevation. (Adapted from Ref. 117.) Right panels: Pulmonary arterial pressures (PPA, top), blood flow (Q, middle), and total pulmonary resistance (TPR, bottom) in cattle during 8week residence at 3900 m. (Adapted from Ref. 24.)

308

Reeves and Stenmark

have been removed to room air. In reported studies, whether or not the rats breathed hypoxic gas at the time of measurement (to simulate the hypoxia at which they had been kept), pulmonary arterial pressure progressively rose (Fig. 8). For example, in one study (87), after approximately one month at half atmosphere, mean pressures approached 60 mmHg, a value higher than seen during acute hypoxia in sea level rats. Thus, measurements in humans and other species point to the same general conclusion, namely, that pulmonary hypertension is progressive over time at altitude. Further, during chronic high-altitude residence the pressures are higher than can be accounted for by acute hypoxic vasoconstriction at low altitude. Clearly, factors not present during acute hypoxia operate during the development of chronic highaltitude pulmonary hypertension.

Figure 8 Measurements in rats during development and regression of hypoxic pulmonary hypertension. (A & B) Percent right ventricular of total ventricular weight during the development (unfilled circles, panel A, left) of hypoxic pulmonary hypertension and its regression after return to normoxia (filled circles, panel B, right) as collected from several studies (39,42,68,70,87,88,106,114). (C & D) Mean pulmonary arterial pressure during the development (unfilled or cross-filled circles, panel C, left) of hypoxic pulmonary hypertension and its regression after return to normoxia (filled circles, panel D, right) in rats as collected from several studies (68,70,87,88,106,114). (Cross-filled circles are data from Ref. 88, where hypoxia was maintained during measurement.)

Pulmonary Circulation

309

Factors (Excluding Wall Thickening) Contributing to Chronic Hypoxic Pulmonary Hypertension Polycythemia

Increasing polycythemia increases the viscosity of blood and could increase pulmonary arterial pressure. One wonders whether polycythemia that occurs over time in humans and most other mammals has a major role in the development of altituderelated pulmonary hypertension. In rabbits, a species with a brisk erythropoietic response to chronic hypoxia, there was a parallel increase in right ventricular systolic pressure and hematocrit (changes of 65 and 62%, respectively) during several weeks at 4300 m, suggesting that blood viscosity played an important role in the pulmonary hypertensive response (91). Polycythemia and pulmonary hypertension co-existed as central features in chronic mountain sickness (Fig. 3B). However, cattle with pulmonary arterial pressures increasing by more than 300% during high-altitude residence had no increase in hematocrit (24,128). Hultgren reported no relationship of pulmonary arterial pressure or resistance to hematocrit in 30 healthy residents at 3750 m in Peru (45). Thus, an increase in viscosity may contribute to, but is not necessary for, pulmonary hypertension at high altitude. Augmented Hypoxic Pulmonary Vasoconstriction

Another possibility to account for the high pulmonary pressures at altitude is that hypoxic pulmonary vasoconstriction increases with continued residence in an hypoxic environment. If so, either the hypoxic stimulus or the response to it increases. For normal subjects, we can dismiss an increase in the severity of the hypoxic stimulus, because from the time of arrival at altitude to a residence of many years duration, hypoxemia does not increase. The question of an increased response to a given hypoxic stimulus (stronger vasoconstriction) is more difficult to answer for humans, because we do not have hypoxic pressor responses at sea level and at altitude in the same subjects. However, resistance changes are available from Leadville, Colorado, at 3100 m (100) in 15 subjects, who breathed 13% O 2 (Pi O2 ⫽ 59 mmHg) for a few minutes. As a group, total pulmonary resistance increased by 28% (Fig. 9A). Eight of the subjects did not increase pressure with hypoxia, and for them an acute hypoxic pulmonary pressor response was absent (Fig. 9B). The other 7 subjects had increases of less than 100%. The little human data available in normal altitude residents suggest the acute hypoxic pulmonary vasoconstrictor response is not stronger than in sea level residents. However, there may be subsets of the population with strong responses, as in Leadville children, who develop high-altitude pulmonary edema when they get upper respiratory infections (21). For chronically hypoxic rats, the answer is clearer than in humans. When rats that have been at high altitude for 4–6 weeks have their lungs removed and artificially perfused with blood, the acute hypoxic pressor responses are blunted (59,114) (Fig. 10). Thus, for hypoxic Po 2 values in the physiological range, pressor reactivity is reduced, not increased. However, pressor responses are increased to other vasoconstrictor agonists such as angiotensin II (Fig. 10), norepinephrine, or prostaglandin

310

Reeves and Stenmark

Figure 9 Effect of acute hypoxia or normoxia in persons exposed chronically to high altitude. (A) Measurements of total pulmonary resistance at various altitudes in persons breathing air (filled circles), 13% oxygen (cross), or high oxygen mixtures (unfilled circles). (From Refs. 29,44,97,117.) (B) Individual pulmonary arterial pressures and flows in 15 young adults breathing air (filled circles) and then 13% oxygen (unfilled circles) in Leadville, Colorado, at 3100 m. (From Ref. 100.)

F 2a . These results are paradoxical. The increased arteriolar muscularity at altitude (96) allows greater responses to vasoactive substances, yet the response to hypoxia is reduced. Further, the mystery is enhanced by the remarkable finding that when rats that had been at altitude for weeks were returned to low altitude for several hours, pressor responses to hypoxia increased well above those of the low-altitude controls (Fig. 10). The reasons for the blunting by hypoxia and the posthypoxic enhancement are not clear. However, the results for many normal humans and for

Pulmonary Circulation

311

Figure 10 Pulmonary pressor responses to hypoxia or angiotensin II in isolated lungs from low-altitude rats (filled symbols), rats living for 4 weeks at high altitude (unfilled symbols), or rats that have been removed for 3–5 days from the hypoxic environment (crosses). (Adapted from Ref. 59.)

rats are compatible with the concept that during high-altitude residence the pressor response to a given hypoxic stimulus is not increased and, further, that the greater pulmonary hypertension with chronic than acute hypoxia is not explained by a stronger acute hypoxic vasoconstriction. Blunted Hyperoxic Vasodilation

The question of hypoxic reactivity can also be addressed by examining the ‘‘other side of the coin,’’ namely the ability of high oxygen to reverse acutely chronic hypoxic pulmonary hypertension (97). Several studies suggest that while human residents of high altitude eliminate hypoxia by breathing O 2-enriched mixtures for 5–25 minutes, they fail to attain normal sea level pulmonary hemodynamics (29,44). Furthermore, the results from the collected studies suggest that the relief of hypoxia usually lowers the pulmonary blood flow along with the pressure so that resistance is not changed (Fig. 9A). Similar findings are reported for chronically hypoxic neonatal calves with severe pulmonary hypertension, namely, 5–10 minutes of 100% oxygen fail to lower pulmonary vascular resistance (107). These data from both human and animal studies with both increasing and decreasing hypoxic stimuli lead to the concept that chronic pulmonary hypertension cannot be explained fully by increasing vasoreactivity. That is, an increasing contractile response to a given hypoxic stimulus may not account for altitude-related pulmonary hypertension. In fact, the results often point in the opposite direction, namely that chronic high-altitude exposure blunts specifically and reversibly the

312

Reeves and Stenmark

ability of acute hypoxia to raise pulmonary arterial pressure. The change from an hypoxic-responsive to an hypo-responsive lung circulation may develop in a few days. In the newborn calf the relatively fixed component of hypoxic pulmonary hypertension increases within 2 weeks (Fig. 11). Although acute vasomotor reactivity to hypoxia persists at high altitude to a variable extent among subjects and among species, altitude-related pulmonary hypertension appears to be more than just sustained acute hypoxic pulmonary vasoconstriction. Is Acute Hypoxic Vasoconstriction Required for the Development of Pulmonary Hypertension at High Altitude?

In humans, there is a very large variation in the pressor response to acute hypoxia (Fig. 12A) and also in the magnitude of chronic high altitude-related pulmonary hypertension (Fig. 12B), but the measurements are in different populations, and the relation of the acute to the chronic response cannot be established. However, a clue that the acute response may be dissociated from the chronic response is suggested from data in high-altitude residents. No correlation existed for the relation of pulmonary pressor response to 13% or 14% O 2 to the resting pulmonary arterial pressure in 21 healthy residents of Leadville, Colorado (117) (Fig. 12C) or in 5 residents of Lhasa, Tibet, respectively (28). These data neither established nor denied that the acute hypoxic response determines the chronic response to altitude. However, they have indicated that the acute response was variable whether or not high altitude– related pulmonary hypertension was present, and suggested that in the high altitude

Figure 11 Development during 2 weeks of altitude exposure of an increased fixed component (unresponsive to 1 hour of normoxia and/or normoxia plus acetylcholine infusion) in newborn calves with severe pulmonary hypertension. (Data redrawn from Ref. 20.)

Pulmonary Circulation

313

Figure 12 Human variability in pulmonary hypertension with hypoxia. (A) Measurements (n ⫽ 186) of pulmonary vascular resistance (PVR) in 50 persons near sea level and without lung disease, breathing air and a series of hypoxic gases. Shown for 5 mmHg classes of pulmonary arterial pressure are mean and one standard deviation. (Adapted from Ref. 90.) (B) Distribution of pulmonary arterial pressure in 37 residents of Leadville, Colorado, illustrating the variation in chronic hypoxic pulmonary hypertension in this population. (Adapted from Ref. 117.) (C) Percent change in pulmonary arterial pressure in 25 of the persons shown in B, above, during breathing of acute hypoxia, 13% O 2. The pressor response was variable, but most of the cohort increased pressures less than 50%.

314

Reeves and Stenmark

resident factors other than the strength of the acute response might determine the chronic response. In cattle, by contrast, a strong relationship was seen between the acute hypoxic response at low altitude to the severity of chronic pulmonary hypertension observed when the individuals were taken to altitude (127). Thus, steers with small acute hypoxic pressor responses developed less pulmonary hypertension at altitude than those with large acute responses. However, the significance of the observation is not clear because the cattle came from two distinct strains bred specifically to be either susceptible or resistant to hypoxic pulmonary hypertension. Further, the susceptible breed had from an early age thicker pulmonary arterioles than the resistant breed (113). Thus, factors other than the strength of the acute hypoxic response could have determined the responses both to acute and to chronic hypoxia in the two herds. There are interesting findings in the rat, where banding the left main pulmonary artery to reduce its lumen markedly reduces the flow to the left lung and the arterial pressure distal to the band. When such rats were exposed to chronic hypoxia, the arteriolar walls in the unbanded right lung were greatly thickened, as expected (86). However, distal to the band, arteriolar walls in the left lung were not thicker than those in the left lung in comparably banded rats at sea level. Presumably, reducing pressure and flow to one lung ‘‘protected’’ the arterioles from the chronic hypoxia–related wall thickening. If acute hypoxic vasoconstriction occurred and was sustained in both lungs, but only one lung developed remodeled artrioles, then acute vasoconstriction would not account for all of the chronic remodeling. Presumably high pressure (or flow) is required. A possible scenario is that acute hypoxic vasoconstriction raises pressure, which initiates a series of events, which then lead to wall thickening. However, many factors contribute to acute hypoxic vasoconstriction and to the chronic pulmonary hypertension at altitude, and in each individual and species the factors may differ. Therefore, we must be cautious in assigning a role for the acute response in determining altitude-related pulmonary hypertension. Thickening of the Pulmonary Arteriolar Wall in Adolescents and Adults

It is generally accepted that chronic hypoxia increases the thickness of the pulmonary arterioles walls. Thickened walls encroach upon and narrow the lumen, with the result that pulmonary arterial pressure and resistance increase. Evidence for such a scenario includes the following: 1.

2.

The thickness of the medial layer in the arteries, but much less in veins, increases over time at high altitude in humans and animals, and the increase is relatively greater in the smaller than in the larger arterioles (34,39,62,121). Thus, the greatest smooth muscle response is at the level of the arteriolar resistance vessels, where lumenal narrowing would be most effective in raising resistance. Perhaps even more important, precapillary arterioles that are poorly muscularized or even devoid of smooth muscle at low altitude become muscu-

Pulmonary Circulation

3.

4.

5.

6.

7. 8.

9.

315

larized at high altitude and could be particularly effective in obstructing the lung circulation (34,42,62,63,106,121). The development of a thick smooth muscle layer lags some hours or days behind the increase in pressure, consistent with an initial vasoconstriction, followed by a muscularization process (61,87,106). When animals with altitude-induced pulmonary hypertension return to low altitude, both pressure and medial thickness decrease, compatible with a causative role for vascular smooth muscle in the hypertension (42,88). With relief of hypoxia the decrease in pressure and medial thickness occur over many weeks, a time course consistent with smooth muscle atrophy, and not just relief of vasoconstriction (87,88,106). Administration of agents such as heparin and collagen synthesis inhibitors, which are not vasodilators but which inhibit the increase of smooth muscle, also blunt the rise in pressure at altitude (30,31,37,49). Because the perivascular space is limited, increasing wall thickness of some magnitude must encroach upon the lumen (108). All layers, including the adventitia (85,108,109), participate in the increasing wall thickness at high altitude, and where measurements have been made lumenal narrowing has been observed (108,109). Electron microscopy has demonstrated endothelial enlargement (38), muscular evaginations (68), and longitudinal muscle in the intima (121), all of which would act to narrow the lumena of small arteries.

With so many pieces of evidence from so many laboratories, it is perhaps surprising that the case is not airtight for structural remodeling of the vascular wall as the principal mechanism for chronic hypoxic pulmonary hypertension. However, some uncertainty remains. Whether there is (42,62) or is not (68) permanent loss of small arteries with chronic hypertension is hotly debated. Also to be resolved is the question of chronic hypoxic vasoconstriction, where there could be an hypoxiarelated increased tonus, which requires more than a few minutes to be relieved by high oxygen. Finally, we are not aware of studies that describe, for various sized vessels, the extent of hypoxic wall thickening that will result in narrowed lumena, nor are there data establishing the relationship (for various sized arterioles) of lumenal narrowing to increased resistance in chronic hypoxia. Thus, increased wall thickening occurs in chronic hypoxia, it may be of great magnitude, and it is strategically located to obstruct the lung circulation, but relevant studies still need to be done. Regression of Altitude-Related Chronic Pulmonary Hypertension and Vascular Wall Thickening Measurements in Humans

At issue is the difficult question of whether the pulmonary circulation becomes entirely normal when an individual comes to live at low altitude following a period

316

Reeves and Stenmark

of residence at high altitude. For 11 healthy male natives of 4540 m going for 2 years to sea level (80), the answer was that the group mean pulmonary arterial pressure fell from 24 to 12 mmHg and calculated pulmonary vascular resistance became normal. Rats (39) and calves (109) with chronic hypoxic pulmonary hypertension brought to low altitude have shown normal resting hemodynamics and normal vasculature by light microscopy. However, there are clues that following relief of chronic hypoxia the lung circulation may retain abnormal features, depending on the severity of the pulmonary hypertension in hypoxia, the duration of the residence in normoxia, the kind of examination performed, and the subsequent stresses to which the lung circulation is subjected. For example, when a resident at 4540 m with chronic mountain sickness went for 2 months at sea level, he lowered his pulmonary arterial pressure from 62 to 24 mmHg, but not to normal levels. More remarkable is the case of an apparently healthy 15-year-old native of Leadville, Colorado, who by chance was discovered to have mean resting and exercising pulmonary arterial pressures of 44 and 109 mmHg at 3100 m (25) (Fig. 13A). Eleven months after she had moved to sea level, her pressure had fallen to 17 mmHg. Although this pressure at sea level was in the upper normal range, there were two clues that her lung circulation was not entirely normal. First, during exercise her pulmonary arterial pressure rose to 38 mmHg, about double that predicted for her estimated workload. Second, 6 months following her return to Leadville, her resting and exercising pressures at 3100 m were 49 and 72 mmHg. The pressures on return to Leadville in relation to O 2 uptakes (Fig. 13A) indicated that her lung circulation had returned to that originally observed at 3100 m. The findings suggested that 11 months at sea level were not sufficient to restore a completely normal lung circulation as judged by her response to exercise and the challenge of returning to high altitude. Measurements in Experimental Animals

Particularly in young rats exposed to chronic hypoxia, pulmonary arterial pressures may not become normal even after several months of normoxia (88). Similar but less impressive results have been seen in older rats (88). Light microscopic studies also suggest that chronic high-altitude rats retain a larger than normal number of muscularized arterioles after return to sea level (106). Further, when young rats exposed for only a few days to altitude are returned to sea level for many months and then reexposed to altitude, they develop pulmonary hypertension of greater magnitude and with greater rapidity than do rats that have never had altitude exposure (40,41). These data in rats are reminiscent of a report of Grover et al. (25). Even though the species, the relative ages, and the exposure times are different, the results taken together raise the possibility that once the lung circulation has been exposed to chronic hypoxia, some individuals or species retain the capacity for abnormal responses to subsequent hypoxic stress. One mechanism for the development of smooth muscle in precapillary vessels was reported by Sobin et al. (106) in a remarkable electron micrographic study in the rat. They found that by 24 hours of either normobaric or hypobaric hypoxia,

Pulmonary Circulation

317

Figure 13 Regression of hypoxic pulmonary vascular changes on removal to low altitude. (A) Pulmonary arterial pressures and blood flows at rest and during exercise in a 15-yearold girl from Leadville, at discovery in Leadville (open circles), following 11 months residence near sea level (filled circles), and 6 months after return to Leadville (partially filled circles). (Adapted from Ref. 25.) (B) Measurements on a logarithmic time scale showing (filled symbols) the development of medial thickness of normally nonmuscular pulmonary arterioles over 9 months at high altitude, and the partial regression of the thickness after 9 months recovery at low altitude. Open circles show the pattern of increase of interstitial fibroblasts neighboring these arterioles for the first part of the high-altitude exposure. (Adapted from Ref. 106.)

318

Reeves and Stenmark

precapillary vessels of the diameter (⬃25 m), which normally had no smooth muscle, began to show smooth muscle within their walls, a consequence of migration and transformation of adventitial fibroblasts. As the smooth muscle continued to increase over 9 months of altitude exposure, so did wall thickness (Fig. 13B). But when some of the rats were returned to sea level for 9 months, smooth muscle was still found in normally nonmuscular arteries (Fig. 13B). The findings were in general agreement with those of Rabinovitch et al. (87), who reported by light microscopy that by day 2 of simulated altitude, smooth muscle began to appear in some arterioles that were normally nonmuscular. They also found that the proportion of these muscularized arterioles increased along with the increase in pulmonary arterial pressure. A subsequent report (88) indicated that following recovery in normoxia, smooth muscle was found in normally nonmuscular arterioles. Taken together, these findings raise the possibility that once very distal, normally nonmuscular arterioles become muscularized at altitude, they may retain some smooth muscle for a very long time after return to sea level. Origin of the Arteriolar Smooth Muscle

How does the smooth muscle get into arterioles that usually contain none? Two routes, which are not mutually exclusive, have been proposed. The first is that there is direct peripheral extension from normally muscular arteries to progressively smaller and smaller arterioles (40). The other proposed route is that cells that resemble fibroblasts (pericytes, transitional cells, intermediate cells) exist in the in the wall or interstitium surrounding the nonmuscular arteries. These cells increase in number with hypoxia, invade the arteriolar wall, and become transformed into smooth muscle cells (61–63,106). Evidence favoring the scenario are electronmicrographs in rats (106) showing a threefold increase in fibroblast-like cells in the interstitium within 24 hours of hypoxic exposure (Fig. 13B). The decrease in fibroblast number after 24 hours (Fig. 13B) is accompanied by an increase in muscularity within the arterioles, as though there had been migration of fibroblasts from the interstitium to the cell wall, where they were transformed into smooth muscle. Migration and transformation are dynamic processes that cannot be seen in fixed tissues and therefore must be inferred. However, fibroblasts, including those in the pulmonary arterial adventitia, are known to proliferate and migrate in response to several vasoconstrictors and growth factors (76). Fibroblasts are also potentially pleomorphic and are considered sometimes to be precursors to smooth muscle cells (109). If further research confirms that hypoxia causes fibroblast migration and transformation to medial smooth muscle cells, then one would wonder if the process is poorly reversible to account for an occasional slow and incomplete resolution of pressure and muscularization on return to normoxia. Hypertrophy/Hyperplasia of the Newly Formed Muscle

No matter whether the first smooth muscle cells arrive at the previously nonmuscular arterioles by direct extension along the vascular tubes or by migration of other cells,

Pulmonary Circulation

319

it seems likely that smooth muscle cells will increase in number and/or size. At issue are the factors that promote the increasing muscularization. Stretch of walls of conduit arteries from rats will rapidly induce DNA synthesis of the pulmonary arterial smooth muscle in 1–3 days (125). In young calves with hypoxic pulmonary hypertension, the ratio of diploid to tetraploid cells in 1-mm-diameter arteries was increased, suggesting that hypertrophy (75) and not only hyperplasia had occurred. Stretch of pulmonary arteries also induces increased production of adventitial matrix within hours (112). Although neither flow cytometry studies to assess ploidy nor in vitro experiments of connective tissue synthesis are reported for the smallest arterioles, available studies raise the possibility that physical factors can influence wall thickness in pulmonary arteries. Chemical factors are certainly important, as recent work has shown for angiotensin II (AII) (70,71). Earlier literature was clear that inhibiting angiotensin II formation by administering angiotensin-converting enzyme (ACE) inhibitors reduced hypoxic pulmonary hypertension (130). However, equally clear was the fact that rat lung angiotensin content and ACE activity decreased with chronic hypoxia (74). It was paradoxical that angiotensin was important in producing hypoxic pulmonary hypertension, and yet its concentration decreased in hypoxic lung. The resolution of the paradox was that loss of angiotensin II from the parenchyma decreased its content in whole lung, yet there was an impressive increase in angiotensin II activity and expression in the arterioles (70) in parallel with the developing chronic pulmonary hypertension. Of particular interest was that the finding was limited to the smallest arterioles as determined by immunohistochemistry and in situ hybridization. Further, the development of new muscle was substantially blocked by ACE or by AII receptor inhibition. AII is only one participant in a cascade whereby paracrine and autocrine growth factors (22) enhance lung arteriolar muscularity. Thickening of the Pulmonary Arterial Wall in the Newborn A Unique Circulation in Structure and Function

Abundant evidence exists that the newborn lung circulation differs from that in the adult in having, e.g., (1) cuboidal rather than flattened endothelial cells (38), (2) increased medial arteriolar muscularity, (3) increased arteriolar adventitial thickness, (4) reduced cross-sectional area of the microvascular bed, (5) an augmented vasoconstrictor response to hypoxia and other vasoactive agents, and (6) an augmented capacity to develop chronic pulmonary hypertension in response to a variety of stimuli (88,107,114). In addition, the newborn at sea level demonstrates its restricted lung vascular bed when it stands and begins to walk only a few minutes after birth. Pulmonary arterial pressures increase from 47 mmHg at rest to 66 mmHg during walking (89). Mild exercise in older calves (89), and as discussed below in adult humans, causes little increase in pressure. Also, newborn cattle develop more quickly than do adults a blunted response to pulmonary vasodilators at high-altitude exposure (19). The newborn lung circulation clearly differs from that of the adult.

320

Reeves and Stenmark

A Unique Capacity for Smooth Muscle Replication

Vascular smooth muscle replication rates are extremely high (up to 80% in 24 hours) in the fetus (rat aorta) and decrease toward birth (14). Further, early in gestation the smooth muscle replicates in an autocrine fashion, independent of stimulation by known growth factors. As the fetus matures and replication slows, the cells become dependent on growth factors from other cells. At birth, for both the rat and the calf, smooth muscle replication is still higher than in the adult (14,108). When the newborn calf is placed in an hypoxic environment, smooth muscle replication increases. Of particular interest, at least for the large pulmonary arteries, is that not all medial cells appear to increase their replication rate, but mitosis occurs in rather selective populations within the media (126). These observations raise the possibility that particular smooth muscle phenotypes, which might persist from fetal life, are retained in the newborn and are stimulated by the hypoxic environment to an active replicative state. Preliminary data from cell cultures indicate a remarkable developmentally related variation in capacity for pulmonary arteriolar smooth muscle replication where the replication rates are high in the term fetus, which exceed the normoxic newborn, which exceed the adult (129). Perhaps most remarkable of all is that the hypoxic newborn has even higher rates than those of the term fetus. The preliminary data further suggest that a protein kinase C (PKC) pathway, which has been stimulated by chronic hypoxia in the newborn calf pulmonary arteries, is uniquely related to the capacity for smooth muscle cell replication (129). Replication of Fibroblasts and Matrix Production in the Neonate

As noted above, fibroblasts lie in the adventitia of the pulmonary arterioles in close relation to the smooth muscle cells. Fibroblasts too are stimulated in chronic hypoxia (16,85). Further as in the case of the smooth muscle, specific subpopulations of fibroblasts proliferate, where the rates of proliferation are greatly enhanced in the hypoxic newborn as compared to the adult (16). Both fibroblasts and the smooth muscle cells show increased production of matrix in the chronically hypoxic calf, and there may well be interaction of the fibroblast and matrix with the smooth muscle cell. In a sense, then, the pulmonary circulation at birth is a ‘‘loaded gun’’ with a large amount of arteriolar smooth muscle poised to constrict and to replicate. For the neonate, chronic hypoxia may, in fact, pull the trigger.

IV. Hemodynamics in Exercising Adult Humans The foregoing discussion has been largely limited to rest, but life requires physical exertion. Based on the preceding sections, the following discussion assumes that high altitude residence increases resistance to blood flow through the lungs, whether the increase might be from hypoxic tone, loss of vasodilating capacity, blood viscosity, and/or vascular remodeling. Exercise is a stress that not only increases pulmonary blood flow, but also introduces stimuli (decreased mixed venous Po 2, increased neural activity, augmented release of vasoactive substances), which could affect the

Pulmonary Circulation

321

lung circulation (93). The question is, therefore, whether altitude-related changes in the lung circulation alter its function during exercise, and if so, what are the consequences? To approach this question we must first look at the lung circulation in normal persons exercising at sea level. A. Exercise in Normal Subjects at Sea Level

A recent extensive review (93) has suggested that at sea level the lung circulation is markedly affected by rising downstream pressures in the left heart. Review of published data for men and women performing supine and sitting cycle exercise of varying intensities indicated: 1. The pulmonary wedge (left atrial) pressure rose with increasing oxygen uptake and pulmonary blood flow. 2. Pulmonary arterial pressure correlated so strongly with wedge pressure that nearly 90% of the variation in the pulmonary arterial pressure could be accounted for by the variation in wedge pressure. 3. The pulmonary gas-exchanging surface, the arterioles, capillaries, and venules, are all thin walled and distensible in humans and other mammals. 4. The increasing microvascular pressures with increasing exercise distend and recruit the microcirculation, which results in increasing diffusion surface area and improved pulmonary oxygen transport. Inquiry into the mechanism of the increased wedge pressure (in supine and sitting subjects) suggested that it arose from the left heart, possibly largely from the inherent limitations to ventricular filling by the normal pericardium (93). The evidence that implicated pericardial limitation was: 1. The wedge pressures correlated strongly with the left ventricular end diastolic pressure at rest and during exercise, suggesting the wedge pressure was a reliable indicator of left ventricular filling pressure. 2. With progressively heavier exercises, the left ventricular end-diastolic pressure rose progressively, but the end-diastolic volume did not. This suggested that left ventricular filling was on the steep portion of the pressure-volume curve. 3. Stroke volume related to end-diastolic volume and not to end-diastolic pressure, confirming that during moderate and heavier exercises filling of the left ventricle was limited. 4. Pericardial pressure-volume curves from animal experiments were compatible with a close application of a poorly compliant pericardium to the ventricles. 5. In animals, removal of the pericardium led to increased ventricular volumes and, during exercise, increased stroke volumes and oxygen uptakes. Taken together, the data suggested that the left heart and particularly the pericardium limited exercise-related ventricular filling, thereby raising filling pressures, which

322

Reeves and Stenmark

increased pulmonary wedge pressure and pulmonary arterial pressure, dilated the thin walled pulmonary microcirculation and contributed importantly to the increased pulmonary O 2 diffusing capacity (93). B. The Pulmonary Circulation During Exercise on Acute Exposure to High Altitude

For persons arriving at high altitude, the question is whether acute hypoxia alters the pulmonary circulatory response to exercise. On the one hand, hypoxic pulmonary vasoconstriction would be expected to raise resting resistance. On the other hand, if wedge and pulmonary arterial pressures increased during exercise, the increasing intralumenal pressures could oppose vasoconstriction of delicate thin-walled arterioles (10,57). In dog lobes, as the left atrial pressure was progressively increased, the acute hypoxic pulmonary pressor response diminished, and became extinguished at about 25 mmHg (10), compatible with the concept that an increased microvascular pressure had opposed the hypoxic pulmonary vasoconstriction. Human data (73) (Fig. 14) indicate that for a given pulmonary flow, during exercise in subjects acutely breathing hypoxic gas mixtures, the pulmonary arterial pressures (P PA), wedge pressures (P W ), and driving pressures (P PA ⫺ P W ) were not different from those during exercise in normoxia. The administration of almitrine, a pharmaceutical agent that augments the ventilatory response to hypoxia and may augment acute pulmonary vasoconstriction, had no effect on the exercise pressures (Fig. 14). Also, data reported by Wagner et al. (122) indicated that during an acute exposure to high altitude in a chamber, exercise pulmonary arterial pressures and wedge pressures, as related to flow, were similar to those observed in the same subjects at sea level (Fig. 14). The concept arises that acute hypoxia, by altering neither the rise in wedge pressure during exercise nor the relation of wedge to pulmonary arterial pressure, has little influence on the control of the exercise pulmonary circulation in normal humans. If so, one might speculate that a microcirculatory pressure rise during exercise opposes and offsets hypoxic vasoconstriction in the normal, thin-walled, lung arterioles. C. Exercise and the Pulmonary Circulation During Altitude Acclimatization

As discussed above, the transition from acute hypoxic pulmonary vasoconstriction to structurally altered lung arterioles begins during the first few days at high altitude and results in higher resting pulmonary arterial pressure and resistance. Thus, one might expect that remodeling of the arterioles could affect the pressure-flow relationships in the lung during exercise. Human pulmonary arterial pressures during exercise through the first 3 days at altitudes of 3900 and 4300 m are available from Kronenberg et al. (50) and Vogel (118), respectively. These measurements show that for a given flow, the altitude exposure is associated with a higher pulmonary arterial pressure during exercise (Fig. 15). There was a parallel upward shift in pressure-flow diagram, which, by analogy with the Starling resistor (for review, see

Pulmonary Circulation

323

Figure 14 Measurements against blood flow in humans at rest and during exercise in six humans breathing air (normoxia, filled circles), during acute altitude exposure (unfilled circles), and during altitude exposure following the administration of an agent (almitrine, hypox ⫹ al, partially filled circles), which putatively potentiates hypoxic pulmonary vasoconstriction. (A) Mean pulmonary arterial pressure. (B) Pulmonary wedge pressure. (C) Pressure gradient from artery to wedge. (Data from Refs. 73,122.)

324

Reeves and Stenmark

Figure 15 Pulmonary arterial pressures and flows in healthy men at rest and during exercise at low altitude (filled circles) and after 1–3 days at high altitude. Correlation coefficients are shown. (Data from Refs. 50,118.)

Ref. 81) may be interpreted as an increase in capillary resistance. Unfortunately, wedge pressures were not reported in these studies, but one expects the site of the increased resistance to be in the microcirculation. D. Lung Circulation in Newcomers to Extreme Altitude, Operation Everest II

Important insight into how chronic hypoxia increased pulmonary vascular resistance and affected the lung circulation was obtained by making sequential measurements in healthy individuals over time as they were taken to extreme, simulated high altitude in Operation Everest II (OE II) (29,43,94). Over nearly 7 weeks, healthy men were progressively decompressed in an altitude chamber to the equivalent elevation of Mt. Everest. Pulmonary hemodynamics were measured at sea level, again at a barometric pressure of 347 mmHg (⬃6100 m), at 282 mmHg (⬃7620 m), and at 253 mmHg (8848 m, not shown). At rest and during exercise, pulmonary arterial pressures increased with increasing altitude (Fig. 16A). Mean resting and exercise pulmonary arterial pressures for the subjects having measurements made at 282 mmHg reached 34 and 54 mmHg, respectively. At comparable pulmonary blood flows, resting and exercise pulmonary arterial pressures at sea level during normoxia and acute altitude exposure were only approximately 15 and 20 mmHg, respectively. Thus, normal subjects exposed chronically to extreme altitude developed moderately severe pulmonary arterial hypertension. Wedge pressures measured during exercise at high altitude were less than at sea level (Fig. 16B). With increasing exertion, the blunted rise in wedge pressure

Pulmonary Circulation

325

Figure 16 Pressure measurements against flow from Operation Everest II in normal residents of sea level (filled circles), after 3–4 weeks at 6100 m (partially filled circles), and after 6–7 weeks at approximately 7620 m (open circles). (A) Pulmonary arterial pressure. (B) Wedge pressure. (C) Pressure gradient from pulmonary artery to wedge. (Data from Refs. 29,92,94.)

326

Reeves and Stenmark

Figure 17 Variability in pulmonary arterial pressure and flow at rest and during exercise in young adult residents of Leadville (partially filled circles), compared to group mean values of normal persons at sea level (filled circles). (Data from Refs. 92,117.)

contributed strongly to the increase in the pulmonary artery to wedge pressure gradient (P PA ⫺ P W ) at altitude. The low wedge pressures during exercise at altitude differed from the increasing wedge pressure with exercise in normoxic subjects at sea level as well as in those exposed to acute hypoxia or acute altitude (compare Fig. 14 and 16B). These observations raised the possibility that time at altitude allowed the development of some factor that inhibited the rise in wedge pressure with altitude. At sea level the pressure gradient driving flow through the lung (P PA ⫺ P W ) increased slightly from rest to exercise as the blood flow tripled (Fig. 16C). Therefore, pulmonary vascular resistance fell by more than 50% (29). However, at the altitudes shown the higher exercise-induced pulmonary arterial pressures and the lower wedge pressures resulted in large increases in driving pressure. The P PA ⫺ P W pressure gradient during exercise increased nearly fivefold from sea level to the 7620 m altitude (Fig. 16C). The slope of the pressure-flow line increased, compatible with an increasing precapillary resistance to blood flow (81). Oxygen inhalation for 5 minutes had little effect on pulmonary vascular resistance. Taken together, the findings are compatible with structural changes in the walls of the lung arterioles at these very high altitudes (29). E.

Pulmonary Circulation During Exercise in High-Altitude Natives

Reports from Peru in 1962 and 1963 (79,80) indicated that more severe pulmonary hypertension was induced by relatively mild exercise in high altitude (4540 m) than

Pulmonary Circulation

327

Figure 18 Group mean resting and exercise pressures and flows from normal persons native to various altitudes. (A) Pulmonary arterial pressure. (B) Pulmonary wedge pressure. (C) Pressure gradient from artery to wedge. (Adapted from Refs. 8,44,80,92,117.)

328

Reeves and Stenmark

in sea level residents. A more detailed study from the same group was published in 1966 (8). Lockhart et al. (52) reported data from La Paz, Bolivia (3750 m). Exercise pressures in North American residents of 3100 m were reported by Vogel (117) and Hartley (36). Data from Tibetan residents of Lhasa (3600 m) were reported by Groves et al. (28). The exercise pulmonary arterial pressures from all altitudes were clearly elevated above those at sea level (Fig. 17A). With increasing altitude of residence, the exercise pulmonary arterial pressures tended to increase, with the highest pressures being seen at the greatest altitude of 4540 m. In addition, for the highest pressures, the pressure-flow lines (Fig. 18A) were not parallel to that at sea level, but the slopes were greater, suggesting increased precapillary resistance (81). The collected data also show great variability in exercise pulmonary arterial pressures. For example, in Leadville at 3100 m, the high school students studied by Vogel et al. (117) had much higher pressures than the 29-year-old men studied by Hartley et al. (36). Even when age was minimized as a factor by just considering the high school students (117), interindividual variability was very great (117), where for example one student had an exercise pressure of 23 mmHg, while for a comparable exercise intensity in another student, the pressure reached 68 mmHg (Fig. 17). In this cohort of students, much of the variation in the exercise pressure could be attributed to the variation in the resting pressure and resistance, where the higher the resting pressures or resistances, the greater the exercise pressures and resistances (respectively, r ⫽ 0.6 and 0.7, p ⬍ 0.01). Another factor related to the observed variability in exercise pressures was the population being studied. For example, young Tibetan men at 2658 m had relatively low exercise pulmonary arterial pressures (Fig. 18A) (28). The authors proposed that populations long resident at high altitude developed relatively low resistance lung circulations as an adaptation to hypoxia (see Sec. VI). An important observation from the reviewed data was that the exercise wedge pressures at sea level tended to be higher than those at altitude (Fig. 18B). (Factors relating to the changes in wedge pressure during exercise at sea level versus those at altitude are considered in Sec. VI. F.) The lower exercise wedge pressures during exercise at altitude than at sea level contributed to the wider pressure gradient from pulmonary artery to wedge (P PA ⫺ P W ) observed at altitude (Fig. 18C). Thus, the higher exercise-related pressure gradient at high altitude than at sea level resulted both from higher pulmonary arterial pressures and lower wedge pressures. At 4540 m the calculated vascular resistance did not fall with exercise and tended to increase (8). These data suggest that the regulation of the lung circulation is likely different for residents at altitude than for those at sea level (see Sec. IV.F). If so, the altered regulation of the lung circulation during exercise in altitude residents may be due to their increased pulmonary vascular smooth muscle, as suggested by several pieces of evidence. For example, breathing high oxygen acutely lowered the exercise resistance in Leadville natives (Fig. 19), even though it had little effect at rest. Apparently during exercise there developed an increased vascular tone, which could be reversed by high oxygen. Further insight into the mechanism

Pulmonary Circulation

329

Figure 19 Decrease during breathing 44% O 2 in exercise pulmonary arterial pressure in normal young adult residents of Leadville. (Data from Ref. 117.)

of altered control was provided by Lockhart et al. (51,52) in a remarkable study in natives of La Paz (3750 m). Exercise increased both pressure and resistance, and assuming blood flow was equally divided between the two lungs, they could estimate resistance to flow through one lung. They then approached the question of whether increased pressure per se could dilate the lung circulation in high altitude natives. To increase pressure they forced the entire cardiac output through one lung using a balloon catheter (Fig. 20). During exercise with all flow diverted to one lung, higher pressures were achieved and the lung vascular resistance fell, even though there was more severe hypoxemia with occlusion. The results were compatible with a pressure-induced dilation of the arteriolar vessels. In a second group of subjects, high oxygen breathing also reduced pulmonary vascular resistance during exercise, compatible with relief of hypoxic vasoconstriction (Fig. 20). From consideration of all the data available to them, the authors concluded that both alveolar and mixedvenous oxygen tensions contributed to the pulmonary vascular tone during exercise. They also concluded that raising the pressure could dilate the lung circulation of the high-altitude native. Of interest, the fall in exercise resistance with balloon occlusion of one main artery was comparable to that observed during oxygen breathing without balloon occlusion (Fig. 20). Apparently either oxygen breathing or high pressure during air breathing could lower exercise resistance, though in neither case to the sea level normal values. The results are compatible with the concept that increased arteriolar smooth muscle could account for some of the differences between sea level and altitude residents in lung vascular function.

330

Reeves and Stenmark

Figure 20 Pressure gradient during exercise from pulmonary artery to wedge against the flow through one lung in normal residents of 3750 m in La Paz. Shown passing through the origin are isopleths of resistance from 1 through 4 units, as labeled. Estimated exercise flow through one lung did not change when subjects changed from air (control) to high O 2 (O 2), but pressure and resistance fell. Exercise with unilateral occlusion of one main pulmonary artery (occlusion) raised flow and pressure to the contralateral lung, but resistance fell. (Data from Refs. 51,52.)

F. Factors Related to the Lung Circulation During Exercise: Normoxia, Acute Hypoxia, and Chronic Hypoxia

For a given pulmonary blood flow during exercise, pulmonary hemodynamics in normal sea level residents are markedly different from from the hemodynamics in subjects exposed to chronic hypoxia (altitude residents and subjects in OE II), summarized in Table 1. As indicated in the previous sections, hypoxia alone does not account for the difference, because subjects exposed to acute hypoxia, i.e., breathing hypoxic gases, or exposed briefly to hypobaria in a chamber have exercise hemodynamics no different from sea level residents (Table 1). At issue are the factors that cause pulmonary hemodynamics to become altered over time at altitude. A key issue is the rise in wedge pressure with exercise at sea level but not at altitude (see Secs. IV.D,E). A review of sea level exercise wedge pressure measurements (93) found them to relate well to left ventricular end diastolic pressures, indicating that the elevations with exercise were not artifacts, but represented increased ventricular filling pressures. Other factors proposed (93) to raise wedge and left atrial diastolic pressure during exercise at sea level included (1) increased left ventricular filling volume and increased afterload pressure, both of which would in-

Pulmonary Circulation

331

Table 1 Changes with Exercise in Hemodynamic Variables That Affect the Lung Circulation a

Wedge pressure (PW) Pulmonary arterial pressure (PPA) Pressure gradient (PPA-PW) Stroke volume

Normoxia

Hypoxia, Acute

Hypoxia, Chronic

⫹⫹⫹ ⫹⫹⫹ ⫾ ⫹⫹⫹

⫹⫹⫹ ⫹⫹⫹ ⫾ ⫹⫹⫹

⫾ ⫹⫹⫹⫹⫹ ⫹⫹⫹⫹⫹ ⫹

a Summary for normal subjects living at sea level (normoxia), for subjects breathing hypoxic mixtures or having brief exposure to hypobaria in an altitude chamber (hypoxia, acute), and for residents at high altitude or subjects at extreme altitude in Operation Everest II (hypoxia, chronic). Shown are approximate changes with exercise relative to normal resting sea level values. Residents at high altitude do not include subjects from the Tibetan population (see Sec. VI).

crease ventricular filling pressure via the Starling mechanism, (2) pericardial limitation of left ventricular stretch during the filling phase, and (3) the presence of a pressure gradient across the mitral valve with very high exercise blood flows. On going from rest to very mild exercise, both stroke volume and left ventricular filling pressure rose, consistent with operation of the Starling mechanism. On going from mild to moderate and heavy exercise, wedge and left ventricular end-diastolic pressures rose, but stroke volume did not increase further. For example, in Operation Everest II at sea level, stroke volumes reached a plateau at about 150 mL, considered to represent pericardial limitation of ventricular filling. The literature review suggested that in many (but not all) subjects at sea level, pericardial limitation to left ventricular filling may have accounted for much of the rise in left atrial pressure during exercise at moderate intensities. If so, an increase in stroke volume with exercise would be necessary for the full increase in wedge pressure to occur. Key issues for the present chapter are the factors that cause the wedge pressures to be less with chronic hypoxia than with normoxia or acute hypoxia. Considering the measurements from OE II, we probably can rule out left ventricular dysfunction at altitude as causing the blunted increase in wedge pressure with exercise, because cardiac function was preserved in these subjects (94,110) and, further, left heart failure should have raised, not lowered, wedge pressure. End-systolic, enddiastolic, and stroke volumes were smaller in these subjects during chronic hypoxia than at sea level. Heart rates for a given cardiac output were higher than at sea level, and stroke volumes were smaller. At 7620 m altitude, stroke volumes of less than 150 mL were not associated with elevations of wedge pressure. Possibly these smaller stroke volumes [and smaller end-diastolic volumes (110)] represented ventricular volumes not at the limits of pericardial compliance. Based on data from Operation Everest II, a wedge pressure versus stroke volume curve may be drawn to approximate for these subjects the compliance characteristics of their pericardia (Fig. 21). An hypothesis to be tested, therefore, is that the wedge pressure will not

332

Reeves and Stenmark

Figure 21 Wedge pressure against stroke volume from Operation Everest II. A regression line through all the data and the regression equation are shown. (Data from Refs. 29,94.)

rise in altitude-exposed subjects until left ventricular filling becomes restricted by the pericardium. One wonders why the stroke volumes during exercise are less in altitude than in sea level residents. Exercise stroke volumes may become reduced within a few days of exposure to moderate altitude (26). The reduction in stroke volume at altitude accompanies a reduction in plasma volume and has been prevented by preventing plasma volume reduction (26). Whether or not the increasing pulmonary vascular resistance contributes to the reduction in exercise stroke volume during chronic altitude exposure is not known. However, one might speculate that contributing to the reduced stroke volume over time at altitude are several factors, including tachycardia from increased sympathetic tone, reduced plasma volume, and possibly increased pulmonary vascular resistance.

V.

Schema and Unanswered Questions

A. Schema

As indicated from the above considerations, the behavior of the lung circulation at altitude, either at rest or during exercise, depends on the vascular remodeling which occurs over time. The remodeling results in both structural and functional alteration of the lung circulation. That alteration occurs primarily in lung arterioles and may be said to involve at least four general steps (131):

Pulmonary Circulation

333

1. Detection by the vessel wall of altered physical and hemodynamic forces 2. The relay of the signal to the endothelial, smooth muscle, and fibroblast cells involved in the process of remodeling 3. Synthesis of substances that may initiate and promote cell division and hypertrophy and other substances that oppose these changes 4. Alteration and thickening of the composition of vessel wall cells and matrix These general steps may be incorporated into a tentative and rather simplified schema (Fig. 22). The schema, though incomplete, indicates that all cell types within the vascular wall participate in the altitude-related changes, that there are factors such as hypertrophy, proliferation, matrix synthesis, and chemotaxis, which contribute to thickening of the cell wall, and other factors that oppose thickening (11). Further, there are general influences that contribute to the interindividual and inerspecies variability of the vascular remodeling. The most important of these influences seem to be the postnatal age of the individual, the duration of the hypoxic exposure, and genetic susceptibility. Examples of chemical mediators are indicated.

Figure 22 Hypothetical schema for steps by which chronic hypoxia at altitude leads to remodeling of the pulmonary arterioles. (Adapted from Ref. 11.)

334

Reeves and Stenmark B. Unanswered Questions

The schematic overview in Figure 22 simply illustrates likely pathways for vascular remodeling but does not portray the mechanisms involved in these pathways. It is fair to say that for each of the arrows shown, the mechanisms are not understood or are, at best, poorly understood. The mechanisms that control alterations in the lung circulation from chronic hypoxia remain important physiological and clinical questions. Questions also remain as to how the remodeled lung circulation behaves in humans at altitude. Does the increased pulmonary hypertension in altitude residents distend the thickened arterioles, or is the pressure effect offset by hypoxic vasoconstriction? At sea level each mmHg intralumenal pressure rise increases the diameter of the microcirculatory vessels by approximately 2% (for review, see Ref. 93). The result is dilation and recruitment, which contribute to an increased diffusing capacity. We need to know to what extent increasing intralumenal pressure increases diameter in lung vessels at altitude. If the microcirculation is not well distended during exercise in altitude residents, is diffusing capacity impaired thereby? Further, a role specifically during exercise has been proposed (51,52) for pulmonary vasoconstriction from hypoxic alveoli and hypoxemic mixed venous blood. One wonders if exercise alters the vasoconstriction/vasodilation ratio toward constriction. Clarification is needed for roles of structural changes in the arteriolar walls, the capacity for passive stretch, and induced vasomotor tone. Finally, does the increased peripheral lung resistance constitute a bottleneck for cardiac output in the high-altitude dweller, and is this a factor in the failure of wedge pressure to rise with exercise? These and other questions continue to emphasize the need for further clinical and basic research. VI. Teleology [I]f you take a just little piece of lung and make it hypoxic, it will constrict (its blood vessels) very strongly—maybe 80% of the blood flow will be diverted away from that region of lung. But it makes hardly any difference to pressure because it’s such a small piece. If you climb a mountain, however, and have the whole lung hypoxic . . . you can’t divert any blood flow from one region of the lung to another. All you can do is increase pressure (56).

One wonders what useful purpose is served by the increased pulmonary arterial pressure that, with few exceptions, is found in people and animals living at high altitude. Of course, we are not sure of the answer to this question. When there is a relatively large vertical distance from the bottom to the top of the lung, as in resting upright adult humans or in cattle, the high arterial pressure may make the lung blood flow more even by increasing flow to the top of the lung, possibly improving gas exchange (27). But one can hardly imagine that such an advantage would accrue to the rat, mouse, or guinea pig, which have very small lungs. With the resting

Pulmonary Circulation

335

pulmonary arterial pressure already elevated at altitude, one wonders what further advantage would result from a large increase during exercise. Perhaps large increases in pulmonary arterial pressure are not advantageous and may even be hurtful. Most populations of mammals, including humans, which have been at high altitude for thousands of years have relatively little pulmonary hypertension. For example, Tibetans, a population with perhaps the longest residence at altitude, have relatively little exercise pulmonary hypertension (28). The yak, a member of the bovine species in the Himalayas, has low resting pulmonary artery pressure and thin-walled pulmonary arterioles (20,35). A related species, the llama, which is native to the Andes, has a small but significant rise in pressure on going from low to high altitude (27). That harmful pulmonary hypertension occurs at high altitude is most clearly seen in infants and young adults, particularly from poorly adapted populations. For example, fatal pulmonary hypertension on the Tibetan plateau was observed in infants born to Chinese migrants of Han ancestry and but rarely in infants of Tibetan parentage (40). Chronic pulmonary hypertension leading to right heart failure was seen in young Indian soldiers stationed high in the Himalayas (2). These findings in humans and those discussed above in young rats and newborn calves reinforce the susceptibility of the young to high-altitude pulmonary hypertension as originally reported from the Andes. The concept is that in the perinatal period there is potential for rapid proliferation of vascular cells and increased production of matrix (109). There are in the newborn muscular arterioles well-developed mechanisms for hypoxic pulmonary vasoconstriction—mechanisms likely needed for a healthy fetus which are still present at birth. Thus, the pulmonary circulation of the newborn is unfortunately positioned to be susceptible to the hypoxia of high altitude. One can imagine that high-altitude species may need enough chronic hypoxic pulmonary hypertension to get them through the fetal and newborn periods, but not enough to cause right heart failure in infancy and adolescence.

References 1. Adams FH, Lind J. Physiologic studies on the cardiovascular status of normal newborn infants (with special reference to the ductus arteriosus). Pediatrics 1957; 19:431–437. 2. Anand IS, Chandrashekhar Y. Subacute mountain sickness syndromes: role of pulmonary hypertension. In: Sutton JR, Coates G, Houston CS, eds. Hypoxia and Mountain Medicine, eds. Burlington, VT: Queen City Printers, 1992:241–251. 3. Archer SL, Weir EK. Mechanisms in hypoxic pulmonary hypertension. In: Wagenvoort CA, Denolin H, eds. Pulmonary Circulation: Advances and Controversies. Amsterdam: Elsevier, 1989:87–113. 4. Archer SL, McMurtry IF, Weir EK. Mechanisms of acute hypoxic and hyperoxic changes in pulmonary vascular reactivity. In: Weir EK, Reeves JT, eds. Pulmonary Vascular Physiology and Pathophysiology, New York: Marcel Dekker, 1989:241– 290.

336

Reeves and Stenmark

5. Arias-Stella J, Saldana M. The muscular pulmonary arteries in people native to high altitude. Conference on the Normal and Abnormal Pulmonary Circulation June 13– 16, 1962. Med Thorac 1962; 19:292–301. 6. Arias-Stella J, Recavarren S. Right ventricular hypertrophy in native children living at high altitude. Am J Pathol 1962; 41:55–64. 7. Ashman R, Byer E. The normal human ventricular gradient I. Factors which affect its direction and its relation to the mean QRS axis. Am Heart J 1943; 25:16–35. 8. Banchero N, Sime F, Pen˜alosa D, Cruz J, Gamboa R, Marticorena E. Pulmonary pressure, cardiac output, and arterial saturation during exercise at high altitude and at sea level. Circulation 1966; 33:249–262. 9. Barcroft J. Researches on Prenatal Life. Oxford: Blackwell Scientific Publications, 1946. 10. Benumof JL, Wahrenbrock EA. Blunted hypoxic pulmonary vasoconstriction by increased lung vascular pressures. J Appl Physiol 1975; 38:846–850. 11. Bishop JE, Reeves JT, Laurent GJ. Pulmonary Vascular Remodelling. London: Portland Press, 1995. 12. Capedehourat EL, Morera V, Nanclares A, Cotino LE, Coppola A, Martinez A. Estudio electrocardiografico. In: Estudios sobre la ‘‘Biologia del Hombre de Altitud.’’ Buenos Aires: Ministerio de Justica e Instruccio´n Pu´blica, 1937:235–263. 13. Civin WH, Edwards JE. Postnatal structural changes in intrapulmonary arteries and arterioles. Arch 1951; 51:192. 14. Cook CL, Weiser MCM, Schwartz PE, Jones CL, Majack RA. Developmentally timed expression of an embryonic growth phenotype in vascular smooth muscle cells. Circ Res 1994; 74:189–196. 15. Cournand A. Measurement of the cardiac output in man using right heart catheterization, description of the technique, discussion of the validity and of place in the study of the circulation. Fed Proc 1945; 4:207. 16. Das M, Dempsey EC, Stenmark KR. Unique growth properties of pulmonary arterial adventitial fibroblasts from neonatal calves with pulmonary hypertension. Am J Res Crit Care Med 1995; 151:A733. 17. Dawes GS, Mott JC, Widdicombe JG, Wyatt DG. Changes in the lungs of the newborn lamb. J Physiol 1953; 121:141–162. 18. Duke H, Killick EM. Pulmonary vascular responses of isolated perfused cat lungs to anoxia. J Physiol 1952; 117:303–316. 19. Durmowicz AG, Orton EC, Stenmark KR. Progressive loss of vasodilator responsive component of pulmonary hypertension in neonatal calves exposed to 4570 m. Am J Physiol 1993; 265:H2175–2183. 20. Durmowicz AG, Hofmeister S, Kadyraliev TK, Aldashev AA, Stenmark KR. Functional and structural adaptation of the yak pulmonary circulation to residence at high altitude. J Appl Physiol 1993; 74:2276–2285. 21. Fasules JW, Wiggins JW, Wolfe RR. Increased lung vasoreactivity in children from Leadville, Colorado after recovery from high altitude pulmonary edema. Circulation 1985; 72:957–962. 22. Fishel RS, Thourani V, Eisenberg SJ, Shai SY, Corson MA, Nabel EG, Bernstein KE, Berk BC. Fibroblast growth factor stimulates angiotensin converting enzyme expression in vascular smooth muscle cells. J Clin Invest 1995; 95:377–387. 23. Fishman AP. A physician-scientist’s tale. In: Wagner WW Jr, Weir EK, eds. The Pul-

Pulmonary Circulation

24. 25. 26.

27.

28.

29.

30. 31. 32.

33.

34. 35.

36. 37. 38. 39.

40.

41.

337

monary Circulation and Gas Exchange. Armonk, NY: Futura Publishing Co., 1994: 381–387. Grover RF, Reeves JT, Will DH, Blount SG Jr. Pulmonary vasoconstriction in steers at high altitude. J Appl Physiol 1963; 18:567–574. Grover RF, Vogel JHK, Voigt GC, Blount SG Jr. Reversal of high altitude pulmonary hypertension. Am J Cardiol 1966; 18:928–932. Grover RF, Reeves JT, Maher JT, McCullough RE, Denniston JC, Cymerman A. Maintained stroke volume but impaired arterial oxygenation in man at high altitude with supplemental CO 2. Circ Res 1976; 38:391–396. Grover RF. Pulmonary hypertension: The price of high living. In: Wagner WW Jr, Weir EK, eds. The Pulmonary Circulation and Gas Exchange. Armonk, NY: Futura Publishing Co, 1994:317–342. Groves BM, Droma T, Sutton JR, McCullough RG, McCullough RE, Zhuang J, Rapmund G, Sun S-F, Janes C, Moore LG. Minimal hypoxic pulmonary hypertension in normal Tibetans at 3658 m. J Appl Physiol 1993; 74:312–318. Groves BM, Reeves JT, Sutton JR, Wagner PD, Cymerman A, Malconian MK, Rock PB, Young PM, Houston CS. Operation Everest II: elevated high-altitude pulmonary resistance unresponsive to oxygen. J Appl Physiol 1987; 63:521–530. Hales CA. Pulmonary hypertension: vasoconstriction and vascular remodeling. Semin Resp Med 1985; 7:136–140. Hales CA, Kradin RL, Brandsetter RD, Zhu Y. Impairment of hypoxic pulmonary artery remodeling by heparin in mice. Am Rev Respir Dis 1983; 128:747–751. Harder DR, Gebremedhin D, Roman RJ. Influence of physical and environmental factors on ion channels in arterial smooth muscle. In: Weir EK, Hume JR, Reeves JT, eds. Ion Flux in Pulmonary Vascular Control. New York: Plenum Press, 1993:105– 110. Harder DR, Madden JA, Dawson CA. Hypoxic induction of Ca 2⫹-dependent action potentials in small pulmonary arteries of the cat. J Appl Physiol 1985; 59:1389– 1393. Harris P, Heath D. The Human Pulmonary Circulation. 2d ed. Edinburgh: Churchill Livingstone, 1977. Harris P. The pulmonary circulation of some domestic animals at high altitude: the real story. In: Wagner WW Jr, Weir EK, eds. The Pulmonary Circulation and Gas Exchange. Armonk, NY: Futura Publishing Co, 1994:283–315. Hartley LH, Alexander JK, Modelski M, Grover RF. Subnormal cardiac output at rest and during exercise in residents at 3100 m altitude. J Appl Physiol 1967; 23:839–848. Hassoun PM, Thompson BT, Hales CA. Partial reversal of hypoxic pulmonary hypertension by heparin. Am Rev Respir Dis 1992; 145:193–196. Haworth SG. Pulmonary vascular remodeling in neonatal pulmonary hypertension. Chest 1988; 93:1335–1385. Heath D, Edwards C, Winston M, Smith P. Effects on the right ventricle, pulmonary vasculature, and carotid bodies of the rat exposed to, and on recovery from, simulated high altitude. Thorax 1973; 28:24–28. Heath D. Pulmonary vascular disease in Sheffield, the Andes, Tibet, and Tanzania. In: Wagner WW Jr, Weir EK, eds. The Pulmonary Circulation and Gas Exchange. Armonk, NY: Futura Publishing Co, 1994:265–282. Herget J, Hample V. Perinatal hypoxia increases hypoxic pulmonary vasoconstriction

338

42. 43. 44. 45. 46. 47. 48. 49.

50.

51. 52.

53.

54. 55.

56. 57. 58.

59.

60. 61. 62.

Reeves and Stenmark in adult rats recovering from chronic exposure to hypoxia. Am Rev Respir Dis 1990; 142:619–624. Hislop A, Reid L. Changes in the pulmonary arteries of the rat during recovery from hypoxia-induced pulmonary hypertension. Br J Exp Path 1977; 58:653–662. Houston CS, Sutton JR, Cymerman A, Reeves JT. Operation Everest II: man at extreme altitude. J Appl Physiol 1987; 63:877–883. Hultgren HN, Kelly J, Miller H. Effect of oxygen upon pulmonary circulation in acclimatized man at high altitude. J Appl Physiol 1965; 20:239–243. Hultgren HN, Kelly J, Miller H. Pulmonary circulation in acclimatized man at high altitude. J Appl Physiol 1965; 20:233–238. Hurtado A. Studies at high altitude. Am J Physiol 1932; 100:487–505. Hurtado A. Aspectos Fisiologicos e Pathologicos de las Vida en la Altura. Editorial Rimac, Lima, 1937. Hurtado A. Chronic mountain sickness. J Am Med Assoc 1942; 120:1278–1282. Kerr JS, Ruppert CL, Tozzi CA. Reduction of chronic hypoxic pulmonary hypertension in the rat by an inhibitor of collagen production. Am Rev Respir Dis 1987; 135:300– 306. Kronenberg RS, Safar P, Lee J, Wright F, Noble W, Wahrenbrock E, Hickley R, Nemoto E, Severinghaus JW. Pulmonary artery pressure and alveolar gas exchange in man during acclimatization to 12,470 ft. J Clin Invest 1971; 50:828–837. Lockhart A, Saiag B. Altitude and human pulmonary circulation. Clin Sci London 1981; 60:599–605. Lockhart A, Zelter M, Mensch-Dechene J, Antezana G, Paz-Zamora M, Vargas E, Coudert J. Pressure-flow relationships in pulmonary circulation of normal highlanders. J Appl Physiol 1976; 41:449–456. Lopez-Barneo J, Lopez-Lopez JR, Urena J, Gonzales C. Chemotransduction in the carotid body: K ⫹ current modulated by PO 2 in type I chemoreceptor cells. Science 1988; 242:580–582. Madden JA, Dawson CA, Harder DR. Hypoxia-induced activation in small isolated pulmonary arteries from the cat. J Appl Physiol 1985; 59:113–118. Madden JA, Vadula MS, Kurup VP. Effects of hypoxia and other vasoactive agents on pulmonary and cerebral artery smooth muscle cells. Am J Physiol 1992; 7:L384– L393. Davis D. At the head of the table. Penn Med 1995; (fall):12–21. Marshall C, Marshall B. Site and sensitivity of hypoxic pulmonary vasoconstriction. J Appl Physiol 1983; 55:711–716. McMurtry IF, Davidson AB, Reeves JT, Grover RF. Inhibition of hypoxic pulmonary vasoconstriction by calcium antagonists in isolated rat lungs. Circ Res 1976; 38:99– 104. McMurtry IF, Petrun MD, Reeves JT. Lungs from chronically hypoxic rats have decreased pressor responses to acute hypoxia. Am J Physiol 1978; 235:H104– H109. McMurtry IF. BAY K 8644 potentiates and A23187 inhibits hypoxic vasoconstriction in rat lungs. Am J Physiol 1985; 18:H741–H746. Meyrick B, Reid L. Hypoxia and incorporation of 3H-thymidine by cells of the rat pulmonary arteries and alveolar wall. Am J Pathol 1979; 96:51–70. Meyrick B, Reid L. The effect of continued hypoxia on rat pulmonary arterial circulation. Lab Invest 1978; 38:188–200.

Pulmonary Circulation

339

63. Meyrick BO, Perkett EA. The sequence of cellular and hemodynamic changes of chronic pulmonary hypertension induced by hypoxia and other stimuli. Am Rev Respir Dis 1989; 140:1486–1489. 64. Monge C, Heraud C, Encinas E, Hurtado A. Estudios fisiologicos sobre el hombre de los Andes. Facultad de Medicina, 1928. 65. Monge C. La enfermedad de los Andes (sindromes eritremicos). Med Per Lima 1929; 2:7. 66. Monge C. La enfermidad de los Andes (sindromes eritremicos). Facultad de Medicina, Lima, 1928. 67. Monge C. Les erythremies de l’altitude. Paris Cie, 1929. 68. Mooi W. Muscular evaginations and disappearing arteries in hypoxia. In: Heath D, ed. Aspects of Hypoxia. Liverpool: Liverpool University Press, 1986:1–18. 69. Morin FC, Stenmark KR. Persistent pulmonary hypertension of the newborn: stateof-the-art. Am J Respir Crit Care Med 1995; 151:2010–2032. 70. Morrell NW, Atochina EN, Morris KG, Danilov SM, Stenmark KR. Angiotensin converting enzyme expression is increased in small pulmonary arteries of rats with hypoxia-induced pulmonary hypertension. J Clin Invest 1995; 96:1823–1833. 71. Morrell NW, Morris KG, Stenmark KR. Role of angiotensin converting enzyme and angiotensin II in development of hypoxic pulmonary hypertension. Am J Physiol 1995; 269:H1186–H1194. 72. Motley HL, Cournand A, Werko L, Himelstein A, Dresdale D. The influence of short periods of induced acute anoxia upon pulmonary artery pressures in man. Am J Physiol 1947; 150:315–320. 73. Naeije R, Me´lot C, Georges Niset, Delecroix M, Wagner PD. Mechanisms of improved arterial oxygenation after peripheral chemoreceptor stimulation during hypoxic exercise. J Appl Physiol 1993; 74:1666–1671. 74. Oparil S, Narkates AJ, Jackson RM, Ann HS. Altered angiotensin converting enzyme in lung and extrapulmonary tissues of hypoxia adapted rats. J Appl Physiol 1988; 65: 218–227. 75. Orton EC, LaRue SM, Ensley B, Stenmark K. Bromodeoxyuridine labeling and DNA content of pulmonary arterial medial cells from hypoxia-exposed and nonexposed healthy calves. Am J Vet Res 1992; 53:1925–1930. 76. Peacock AJ, Dawes KE, Shock A, Gray AJ, Reeves JT, Laurent GJ. Endothelin-1 and endothelin-3 induce chemotaxis and replication of pulmonary artery fibroblasts. Am J Respir Cell Mol Biol 1992; 7:492–499. 77. Penalosa D, Gamboa R, Dyer J, Echevarria M, Marticorena E. The influence of high altitudes on the electrical activity of the heart I. Electrocardiographic and vectorcardiographic observations in the newborn, infants, and children. Am Heart J 1960; 59:111– 128. 78. Penalosa D, Gamboa R, Dyer J, Echevarria M, Marticorena E. The influence of high altitudes on the electrical activity of the heart II. Electrocardiographic and vectorcardiographic observations in adolescence and adulthood. Am Heart J 1961; 61:101–128. 79. Penaloza D, Sime F, Banchero N, Gamboa R, Cruz J, Marticorena E. Pulmonary hypertension in healthy man born and living at high altitudes. Am J Cardiol 1963; 11:150. 80. Penaloza D, Sime F, Banchero N, Gamboa R. Pulmonary hypertension in healthy man born and living at high altitudes, Fifth Annual Conference on Research in Emphysema, Aspen, CO, June 13–16, 1962. Med Thorac 1962; 19:449–460. 81. Permutt S. Pulmonary mechanics and the pulmonary blood vessels. In: Wagner W,

340

82. 83. 84. 85. 86.

87.

88.

89. 90.

91. 92.

93.

94.

95. 96.

97. 98.

99.

Reeves and Stenmark Weir EK, eds. The Pulmonary Circulation and Gas Exchange. Armonk, NY: Futura Publishing Co., 1994:147–165. Post JM, Gelband CH, Hume JR. [Ca 2⫹] i inhibition of K ⫹ channels in canine pulmonary artery. Circ Res 1995; 77:131–139. Post JM, Hume JR, Archer SL, Weir EK. Direct role for potassium channel inhibition in hypoxic pulmonary vasoconstriction. Am J Physiol 1992; 262:C882–C890. Pryor R, Weaver WF, Blount SG Jr. Electrocardiographic observation of 493 residents living at high altitude (10150 feet). Am J Cardiol 1965; 16:494–499. Rabinovitch M. Elastase, remodeling of the extracellular matrix, and pulmonary hypertension. Semin Respir Crit Care Med 1994; 15:199–206. Rabinovitch M, Konstam MA, Gamble WJ, Papanicolau N, Aronovitz MJ, Treves S, Reid L. Changes in blood flow affect vascular response to chronic hypoxia in rats. Circ Res 1983; 52:432–441. Rabinovitch M, Gamble WJ, Nadas AS, Miettinen OS, Reid L. The rat pulmonary circulation after chronic hypoxia. Hemodynamic and structural features. Am J Physiol 1979; 236:H818–H827. Rabinovitch M, Gamble WJ, Miettinen OS, Reid L. Age and sex influence on pulmonary hypertension of chronic hypoxia and on recovery. Am J Physiol 1981; 240:H62– H72. Reeves JT, Leathers JE. Pulmonary hypertension with muscular exercise in the newborn calf. J Appl Physiol 1965; 20:249–252. Reeves JT, Groves BM. Approach to the patient with pulmonary hypertension. In: Weir EK, Reeves JT, eds. Pulmonary Hypertension. Mt. Kisko, NY: Futura, 1984:1– 44. Reeves JT, Grover EB, Grover RF. Circulatory responses to high altitude in the cat and rabbit. J Appl Physiol 1963; 18:575–579. Reeves JT, Dempsey JA, Grover RF. Pulmonary circulation during exercise. In: Weir EK, Reeves JT, eds. Pulmonary Vascular Physiology and Pathophysiology. New York: Marcel Dekker, 1989:107–133. Reeves JT, Taylor AE. Pulmonary hemodynamics and fluid exchange in the lungs during exercise. In: Rowell LB, Shepherd JT, eds. Handbook of Physiology, Section 12, Exercise: Regulation and Integration of Multiple Systems. New York: American Physiological Society, Oxford University Press, 1996:585–613. Reeves JT, Groves BM, Sutton JR, Wagner PD, Cymerman A, Malconian MK, Rock PB, Young PM, Houston CS. Operation Everest II: preservation of cardiac function at extreme altitude. J Appl Physiol 1987; 63:531–539. Regensteiner JG, Moore LG. Migration of the elderly from high altitude in Colorado. J Am Med Assoc 1985; 253:3124–3128. Reid LM. Structural remodeling of the pulmonary vasculature by environmental change and disease. In: Wagner WW Jr, Weir EK, eds. The Pulmonary Circulation and Gas Exchange. Armonk, NY: Futura Publishing Co, 1994:77–110. Rotta A, Canepa A, Hurtado A, Velasquez T, Chavez R. Pulmonary circulation at sea level and at high altitudes. J Appl Physiol 1956; 9:328–336. Rotta A, Canepa A, Velasquez T, Hurtado A, Aste Salazar H, Chavez R. Presio´n de la arteria pulmonar en el hombre que vive a 4500 metros de altitud. Rev Argentina Cardiol 1952; 19:374–375. Rotta A, Miranda A, Chavez R. Heart catheterization at high altitudes. Symposium International de Biologı´a de Altitud, 1949.

Pulmonary Circulation

341

100. Rotta A. Physiologic condition of the heart in the natives of high altitudes. Am Heart J 1947; 33:669–676. 101. Rowe RD, James LS. The pattern of response of pulmonary and systemic arterial pressures in newborn and older children to short periods of hypoxia. AMA J Dis Child 1957; 93:13. 102. Rudolph AM, Auld PAM, Gulinko RJ, Paul MH. Pulmonary vascular adjustments in the neonatal period. Pediatrics 1961; 28:28–34. 103. Saling E. Neue Untersuchungsergebnisse u¨ber den Kreislauf des Kindes unmittelbar nach der Geburt. Arch Gyna¨k 1961; 194:287–306. 104. Sime F, Penalosa D, Ruiz L. Bradycardia, increased cardiac output, and reversal of pulmonary hypertension in altitude natives living at sea level. Br Heart J 1971; 33: 647–657. 105. Sime F, Banchero N, Penaloza D, Gamboa R, Cruz J, Marticorena E. Pulmonary hypertension in children born and living at high altitudes. Am J Cardiol 1963; 11:143– 149. 106. Sobin S, Tremer HM, Hardy JD, Chiodi HP. Changes in arteriole in acute and chronic hypoxic pulmonary hypertension and recovery in rat. J Appl Physiol 1983; 55:1445– 1455. 107. Stenmark KR, Fasules J, Hyde DM, Voelkel NF, Henson J, Tucker A, Reeves JT. Severe pulmonary hypertension and arterial adventitial changes in newborn calves at 4300 m. J Appl Physiol 1987; 62:821–830. 108. Stenmark KR, Dempsey EC, Badesch DB, Frid M, Mecham RP, Parks WC. Regulation of pulmonary vascular wall cell growth: developmental and site specific heterogenicity. Eur Respir Rev 1993; 16:629–637. 109. Stenmark KR, Mecham RP, Durmowicz AG, Parks WC. Persistence of the fetal pattern of tropoelastin gene expression in severe neonatal pulmonary hypertension. J Clin Invest 1994; 93:1234–1242. 110. Suarez A, Alexander JK, Houston CS. Enhanced ventricular systolic performance at high altitude during Operation Everest II. Am J Cardiol 1987; 60:137–142. 111. Torrance RW. The idea of a chemoreceptor. In: Fishman AP, Hecht HH, eds. The Pulmonary Circulation and Interstitial Space. Chicago: University of Chicago Press, 1969:223–239. 112. Tozzi CA, Poiani GJ, Harangozo AM, Boyd CD, Riley DJ. Pressure-induced connective tissue synthesis in pulmonary artery segments is dependent on intact endothelium. J Clin Invest 1989; 84:1005–1012. 113. Tucker A, McMurtry IF, Reeves JT, Alexander AF, Will DH, Grover R. Lung vascular smooth muscle as a determinant of pulmonary hypertension at high altitude. Am J Physiol 1975; 228:762–767. 114. Tucker A, Migally N, Wright ML, Greenlees KJ. Pulmonary vascular changes in young and aging rats exposed to 5486 m altitude. Respiration 1984; 46:246–257. 115. Voelkel NF. Mechanisms of hypoxic pulmonary hypertension. Am Rev Respir Dis 1986; 133:1186–1195. 116. Voelkel NF. Pulmonary hypertension: From phenomenology to a molecular understanding of disease mechanisms. In: Weir EK, Hume JR, Reeves JT, eds. Ion Flux in Pulmonary Vascular Control. New York: Plenum Press, 1993:1–14. 117. Vogel JHK, Weaver WF, Rose RL, Blount SG, Grover RF. Pulmonary hypertension on exertion in normal man living at 10,150 feet (Leadville, Colorado). Med Thorac 1962; 19:461–477.

342

Reeves and Stenmark

118. Vogel HK, Goss JE, Mori M, Brammel HL. Pulmonary circulation in normal man with acute exposure to high altitude (14,260 feet). Circulation 1966; 34:III-233. 119. von Euler US, Liljestrand G. Observations on the pulmonary arterial blood pressure in the cat. Acta Physiol Scand 1946; 12:301–320. 120. Wagenvoort CA, Neufeld HN, Edwards JE. The structure of the pulmonary arterial tree in fetal and early postnatal life. Lab Invest 1961; 10:751. 121. Wagenvoort CA, Wagenvoort N. Diseases of the pulmonary circulation. In: Silver MD, ed. Cardiovascular Pathology. New York: Churchill Livingstone, 1983. 122. Wagner PD, Gale GE, Moon RE, Torre-Bueno J, Stolp BW, Saltzman H. Pulmonary gas exchange in humans exercising at sea level and simulated altitude. J Appl Physiol 1986; 61:260–270. 123. Wagner W. Pulmonary circulatory control through hypoxic vasoconstriction. Semin Respir Med 1985; 7:124–135. 124. Weir EK. The mechanism of acute hypoxic pulmonary vasoconstriction: the tale of two channels. FASEB J 1995; 9:183–189. 125. Weiser MCM, Majack RA, Tucker A, Orton EC. Static tension is associated with increased smooth muscle cell DNA synthesis in rat pulmonary arteries. Am J Physiol 1995; H1133–H1138. 126. Whorley JD, Frid MG, Orton EC, Belknap JK, Stenmark KR. Hypoxia selectively induces proliferation in a specific subpopulation of smooth muscle cells in bovine neonatal pulmonary arterial media. J Clin Invest 1995; 96:273–281. 127. Will DH, Hicks JL, Card CS, Reeves JT, Alexander AF. Correlation of acute with chronic hypoxic pulmonary hypertension in cattle. J Appl Physiol 1975; 38:495–498. 128. Will DH, Alexander AF, Reeves JT, Grover RF. High altitude induced pulmonary hypertension in normal cattle. Circ Res 1962; 10:172–177. 129. Xu Y, Stenmark KR, Walchak SJ, Ruff LJ, Dempsey EC. Pulmonary artery smooth muscle cells from hypoxia-exposed neonatal calves retain and/or re-express fetal-like growth properties. Submitted. 130. Zakheim RM, Mattioli L, Molteni A, Mullis KB, Bartley J. Prevention of pulmonary vascular changes of chronic alveolar hypoxia by inhibition of angiotensin I-converting enzyme in the rat. Lab Invest 1975; 33:57–61. 131. Zhao L, Winter RJD. Pulmonary vascular remodeling. In: Pulmonary Circulation. London: Chapman & Hall Medical, 1996.

11 Cerebral Circulation at High Altitude

JOHN W. SEVERINGHAUS University of California San Francisco, California

I.

Introduction

The effect of high altitude upon CBF, appearing simple at first, has proved complex. With acute ascent to about 4000 m altitude, cerebral blood flow (CBF) rises by 30– 60%. Over the next few days to weeks it falls, and after months to years it is not different from sea level normal values, despite chronic severe hypoxemia. This fall is not due to insensitivity to hypoxia because in natives of altitude, CBF falls in response to acute hyperoxia. There is no evidence of decreased cerebral metabolic rate for oxygen (CMRo 2), which remains normal even in hypoxia severe enough to obtund consciousness and increase brain lactic acid. In this chapter we examine the physiological responses of cerebral circulation to short- and long-term hypoxia. Cerebral vessels receive conflicting signals in acute hypoxia: While hypoxia is a vasodilator, the hypoxic hyperventilation lowers Pco 2 , causing cerebral vasoconstriction. The magnitude of change of CBF depends upon the relative strengths of four reflex mechanisms: (1) hypoxic ventilatory response (HVR), (2) hypercapnic ventilatory response (HCVR), (3) hypoxic cerebral vasodilation, and (4) hypocapnic cerebral vasoconstriction. In addition, the ventilatory responses to O 2 and CO 2 interact. For example, hypercapnia greatly increases the hypoxic ventilatory response, 343

344

Severinghaus

and hypocapnia diminishes it. The variations in these four sensitivities between individuals may determine individual susceptibility to altitude sickness and explain why some subjects have failed to show increased CBF in acute hypoxia. The task of this chapter is to examine the evidence about the several mechanisms and their interactions, consider the possible roles played by them in the pathological cerebral signs and symptoms one encounters in climbing and living on mountains, and identify questions about these phenomena begging further investigation. How does hypoxia cause vasodilation? To what extent can hypocapneic vasoconstriction limit hypoxic vasodilation, and can this cause injury? Does hypoxia directly dilate arterioles or require ECF (extracellular fluid) messengers such as H ⫹, K ⫹, adenosine, or NO? Does high CBF cause or contribute to acute mountain sickness (AMS), high-altitude cerebral edema (HACE), or brain injury? Do some of the signs or symptoms of AMS or HACE depend on the CBF response to altitude? Does acetazolamide help in acclimatization by increasing CBF at altitude? Are there other unrecognized mechanisms downregulating CBF in chronic hypoxia? Does brain decrease its inherent metabolic rate in response to chronic hypoxia?

II. Brain Oxidative Biochemistry and Its Effect on Flow CMRo 2 is maintained constant despite hypoxia by the Pasteur effect by which increased ADP stimulates glycolysis. Isolated mitochondria are able to maintain a constant O 2 consumption until solution Po 2 falls below 0.5 torr (1). Average cortex tissue Po 2 is about 9 mmHg as determined with recessed, calibrated gold-plated microelectrodes (2). Neuronal mitochondrial cytochrome is normally not fully saturated with O 2 , and thus the redox state is not fully oxygenated, as shown by the rise in NAD with vasodilation (3). Thus, even in normoxia the brain exhibits some ‘‘anaerobic metabolism,’’ detected by measuring the difference between arterial and cerebral venous lactate. Lactic acid crosses the capillary endothelial blood-brain barrier (BBB) using a stereospecific, facilitated, saturable, passive transporter (4).

III. Normal Cerebral Blood Flow Regulation CBF regulation involves responses in both large arteries and precapillary sphincters. Smooth muscles of arteries larger than about 50 µm constrict or dilate in response to variation of intra-arterial pressure, exhibiting the curious stretch response, which actually reduces lumen diameter when arterial pressure rises. This response is independent of the sympathetic innervation (5,6). These arteries regulate pressure supplied to the microcirculation but do not control flow. Rather, terminal arterioles and precapillary sphincters control flow, moment to moment, in response to local chemical and neural mediators (7). At rest, red cells travel through the capillary network in 100–300 ms along 150–500 µm paths (8). Blood flow in individual capillaries in brain is cyclic, resulting in oscillations in tissue Po 2 (4–10 min ⫺1) (9–11). Vasodi-

Cerebral Circulation at Altitude

345

lators increase the velocity of the red cells (8), the number of capillaries perfused at any moment, and the length of each open period (12), suggesting this open-shut system to be a major component of cerebral autoregulation by which flow is independent of pressure between approximately 60 and 150 mmHg (13), but rapidly responsive to metabolic demands partly mediated via locally released NO (14). During and following hypoxia, autoregulation may be disrupted (15,16). CBF increases in response to both decreased O 2 supply and increased O 2 demand. With neuronal activation, O 2 consumption rises within milliseconds, and local CBF rises within 1–2 s (14). The hyperemic response to step hypoxia is almost equally rapid (17). ECF adenosine concentration also rises within a minute, but the distance between neurons and arterioles is too great for diffusion of neuronally released adenosine or lactic acid H ⫹ to arterioles to account for the rapid flow increase. K ⫹ is also a possible mediator (see Sec. III.C). A. Hypoxia

During acute hypoxia, red blood cell (RBC) velocity increases in all brain capillaries (using intravital microscopy) (8). CBF rises similarly in proportion to the severity of isocapnic hypoxia in all mammals. However, when Pco 2 is not controlled during hypoxia, there is considerable variability between individuals and species because of the wide variability of the HVR, which determines how far Pco 2 falls. A strong HVR and major fall of Pco 2 in experimental subjects led to an early impression that hypoxic vasodilation was a threshold phenomenon, scarcely beginning until Pao 2 had fallen to about 30 mmHg (18). In their original study in young male volunteers using their N 2 O method (19), Kety and Schmidt found a 35% increase in CBF while breathing 10% O 2 (resulting in Sao 2 ⫽ 65%), which caused Paco 2 to fall from 40 to 36 mmHg. Reported responses in humans at altitude have varied from no immediate increase to rises of the order of 30–60% during the initial hours or days. In awake sheep after 30 minutes of hypoxia to Pao 2 ⫽ 40 mmHg, CBF increased by as much as 250% in spite of a 6 mmHg fall of Paco2 (20). CBF rose more than 300% in rats at Pao 2 ⫽ 24 mmHg (21). Some effects on CBF of acute hypoxia may persist after reoxygenation. Trojan and Kapitola (22) found residual high blood flow through the medulla, cerebellum, subcortical regions, and cerebral cortex in adult rats 20 hours after completing an 8-hour exposure to 7000 m altitude (in a chamber). The direct vasodilating effect of hypoxia on CBF may be separated from the negative feedback change of Paco 2 that usually accompanies hypoxia by adding CO 2 to the inspired air to keep end-tidal Pco 2 constant, a procedure known as isocapnic hypoxia. Using isocapnia in nine healthy male volunteers, Cohen et al. (23) showed that a step reduction of Pao 2 to 34.6 ⫾ 1.6 mmHg (SE) (⬇66% Sao 2) increased CBF about 70% (from 0.45 to 0.77 mL g ⫺1 min ⫺1) accompanied by a 27% rise in glucose consumption (CMR glu) and a fourfold rise in cerebral lactate production (CMR lac), with no significant fall of CMRo 2. This isocapnic rise of CBF was twice that reported at the same Sao 2 by Kety and Schmidt (19) when Paco 2 was

346

Severinghaus

allowed to fall. In a direct comparison of the effect of Pco 2 on hypoxic vasodilation in sheep, Yang et al. (24) reported that step reduction of Pao 2 to 40 mmHg increased CBF 156% in isocapnia but only 68% when Pco 2 was allowed to fall freely (from 36 to 28 mmHg). These several studies provide a simple guideline that the effect of isocapnia is to double the hypoxic increase of CBF compared to poikilocapnia. The relationship of CBF to Pao 2 is hyperbolic, rather like the hypoxic ventilatory response. Since the ventilatory response proved to be a linear function of arterial Sao 2 , several studies have sought similar linearity for the plot of CBF vs. Sao 2. Ashwal et al. (25) found that, in fetal lambs made hypoxic by acute maternal isocapnic hypoxemia, CBF was an approximately linear function of fetal Sao 2 down to nearly zero, at which point flow was increased to about 250% of control. However, Jensen et al. (26) found the relationship between flow and Sao 2 to be hyperbolic in humans. In six normal males at sea level and during 5 days at 3810 m altitude using 5-minute steps to four levels of isocapnic hypoxia, they measured the relationship of CBFv (v for velocity, measured by transcranial doppler, TCD) to Sao 2 (Fig. 1). Acute hypoxia to 70% Sao 2 increased CBFv by 35% at sea level and by 46% after 5 days at altitude, indicating a 34% increase in the sensitivity of cerebral vessels to acute hypoxia induced by this 5-day acclimatization period. At sea level the acute

Figure 1 CBFv% (% of control CBF velocity by TCD) in response to 5 minute steps to four levels of isocapnic step hypoxia in normal subjects after 3–4 days at 3810 m altitude. The hyperbolic empirical relationship is: CBFv% 100(1 ⫹ X[(60/(S aO 2-40)) ⫺ 1]). Factor X was found to average 0.35 ⫾ 0.11 at sea level and 0.46 ⫾ 0.08 after 5 days at altitude. X may be interpreted as the fractional increase in CBFv induced by acute isocapnic hypoxia at 70% Sao 2. (From Ref. 26.)

Cerebral Circulation at Altitude

347

increase of CBF was about two-thirds of the sensitivity reported by Cohen et al. 30 years earlier. Hypoxia had surprisingly little effect on the high energy state of cortex, cerebellum, or brain stem in lightly anesthetized rats at 35, 29, and 23 mmHg, as determined by a rapid (⬍5 s) in situ freezing method (27). At the lowest Po 2 (23 mmHg), ADP showed small increases, but no change in ATP or AMP was detected. Lactate rose from ⬍2 to 13 mM kg ⫺1, pyruvate levels tripled, and regional CBF rose to five to seven times control in all three regions. Autoradiographic methods were used to show that the vasodilatory response to both hypercapnia and hypoxia was greatest among brain stem gray matter structures, intermediate among cortical and diencephalic gray matter structures, and least in white matter (28). In subcortical white matter, hypoxia increased glucose consumption fivefold but flow less than twofold, suggesting localized anaerobic glycolysis and acidosis. In highly active regions of gray matter, in comparison, glucose consumption rose less than 50% while flow rose 200–500%. This observation may fit with other evidence that white matter is especially vulnerable to hypoxic damage (29,30). B. Anemia

Isovolemic hemodilution increases RBC velocity three- to fourfold and increases RBC flux to a moderate degree, with a relatively small decrease in capillary hematocrit, under normal and compromised arterial blood supply (8). Hemodilution increases cerebral blood flow in both normal and polycythemic patients (31–33). In newborn lambs this effect was extremely variable. Reduction of hematocrit from 55% to 10% by means of exchange transfusions increased CBF sevenfold in a few animals, but scarcely doubled it in others (34). Both the lowered viscosity and lowered O 2 content are presumed to contribute, but because of the strong effect of O 2 content, it has been difficult to demonstrate a separate effect of viscosity. The role of viscosity was reported to be insignificant in a study of the influence of severe acute normovolemic anemia on CBF and CMRo 2 in normocapnic rats under nitrous oxide anesthesia (35,36). The arterial hemoglobin content was reduced to values of about 12, 9, 6, and 3 g dL ⫺1 by replacing blood with equal volumes of plasma. The CBF increase was proportional to the reduction in hemoglobin content, rising to five to six times normal in extreme anemia, but CMRo 2 remained unchanged. Cerebral venous Po 2 and Sao 2 did not decrease below normal values, indicating that tissue hypoxia did not develop. At 3 g dL ⫺1 tissue, lactate content increased moderately, and associated changes in other carbohydrate metabolites and in amino acids suggested that tissue hypoxia was present. However, since the concentrations of phosphocreatine and adenine nucleotides remained constant, this hypoxia must have been slight. Since the increase in CBF at hemoglobin concentrations of below 9 g dL ⫺1 was far in excess of that expected from the decrease in viscosity, hyperemia was thought to be due primarily to the reduction in arterial O 2 content. More recently, MetHb has been used to separate the effects of viscosity from O 2 content. Red cells containing metHb cannot deliver oxygen and thus lower O 2

348

Severinghaus

content without affecting viscosity (37). Two groups of awake anemic lambs were transfused, one with normal blood, the other with metHb-inactivated red cells. In both groups, Hct was raised from about 20 to 40%. CBF fell 50% using normal O 2carrying erythrocytes but also fell 28% with MetHb cells, suggesting that about half of the effect must be due only to viscosity. The effect of viscosity was quite variable, being greatest in those lambs with the highest CBF. The role of viscosity in reducing CBF during polycythemia has not been reported but would be of interest for those with high-altitude polycythemia in view of the report that CBF in altitude natives has the same inverse linear relationship of Hct as in sea level natives (38). A compilation of data from various studies (39) of anemic, normal, and polycythemic subjects revealed a hyperbolic fourfold variation of (A-V)∆ O2 with hematocrit (Hct) (Fig. 2). From these data, CBF values may be estimated to vary with Hct from 1.0 to 0.29 mL g ⫺1 min ⫺1 by the following process: normal global CBF in humans at sea level is 0.45 mL g ⫺1 min ⫺1 at normal Hct ⫽ 45%. CMRo 2 is independent of oxygen supply down to a critical threshold. Because (A-V)o 2 difference (vol%) is a hyperbolic function of Hct, CBF is inversely proportional to Hct (38).

Figure 2 The (A-V)∆O 2 content during acute normoxia in various groups including natives of the Bolivian altiplano as a function of Hct. The normal relationship of CBF to Hct derived from these data is given in the text. 䊊: anemia; 䉭: normal mean; 䊉: chronic polycythemia (altitude); 䊐: chronic polycythemia (other). (From Ref. 39.)

Cerebral Circulation at Altitude

349

By assuming (A-V)∆O 2 ⫽ CMRo 2 /CBF, where CMRo 2 is invariant, the relationship of normoxic global CBF to Hct from these data may be approximated by: CBF ⫽ [0.675 ⫺ 0.005 Hct] (mL ⋅ g ⫺1 ⋅ min ⫺1) In order to use this relationship to factor out Hct in studies at varying times and in varying groups with other normal control CBF values (Q 45), the relationship may be written: Q 45 ⫽ 90CBF/[135-Hct] C. Carbon Monoxide and Its Effect on the O 2 Dissociation Curve

CO hypoxia is a more potent vasodilator than either hypoxia or anemia, actually increasing O 2 delivery (CaO 2 ⫻ CBF) above normal. Koehler et al. (40) compared CBF responses to hypoxia induced by low alveolar Po 2 or by carbon monoxide in awake adult sheep. The arterial O 2 content (CaO 2) was equally reduced stepwise to 50–60% of the control value. CBF and O 2 delivery increased more with CO hypoxia than with alveolar hypoxia. This CBF difference was proportional to the magnitude of the leftward shift of the oxygen dissociation curve (ODC) in those given CO. Despite the higher O 2 delivery with CO hypoxia, CMRo 2 fell by 16% in adults but was maintained in newborns, while CMRo 2 was unchanged during alveolar hypoxia in both age groups. CO lowers capillary Po 2 by inducing a leftward shift of ODC (oxygen dissociation curve), especially at low saturation, because it eliminates the S shape at the bottom. For example, 20% COHb reduces the Po 2 of blood with 24% O 2Hb from 17 (normal) to 12 mmHg. In normal blood, Po 2 ⫽ 12 mmHg at 13% O 2Hb. The critical Po 2 is approximately 12 mmHg, below which CMRo 2 begins to fall. Thus, COHb severely limits unloading of O 2 from Hb. This is a considerably greater effect than other causes of left-shifted ODC, such a low 2,3-DPG. The importance of the position of the ODC as a determinant of cerebral blood flow supports the presence of a highly sensitive, tissue Po 2 –dependent mechanism regulating the cerebral circulation. To test whether the CO effect was simply due to the shift of the ODC or to additional cellular effects, Koehler et al. (40,41) performed isovolemic exchange transfusions on unanesthetized newborn lambs, replacing their high–O 2-affinity hemoglobin with low-affinity adult sheep donor blood. Exchange transfusion resulted in an average increase in P 50 of 10 mmHg and in a 14% decrease of regional cerebral blood flow and cerebral O 2 delivery. Induction of CO hypoxia (20–40% COHb) after the exchange transfusion returned P 50 to the control level, and restored both cerebral O 2 delivery and fractional O 2 extraction to the pretransfusion values. Thus it is the fall in P 50 , rather than a direct tissue effect of CO, which is responsible for the relative cerebral overperfusion during CO hypoxia. They concluded that capillary Po 2 , not capillary O 2 content, determined CBF, an interpretation similar to that of Woodson using hypocapnic left shifts of the ODC (see Sec. III.D) (42).

350

Severinghaus D. Mediators of Acute Hypoxic Cerebral Vasodilation

How does reduced capillary Po 2 dilate arterioles? Attention is now focused on those glial cells whose feet are wrapped around arterioles and whose heads are in tight contact with local neurons (13). Flow is increased by local neuronal activity (43– 45), increasing ECF [K ⫹] (46), adenosine (45,47), and hypercapnia (48). Other vasoactive messengers might include intravascular NO generated in endothelia (44,49) and a variety of autocoids and cytokines (13). While hypoxia probably operates through several of these mechanisms, the evidence to be presented suggests that it may have other still unknown direct vasodilating mechanisms. Potassium

When a neuron depolarizes, K ⫹ ions diffuse out from the neurons through the ECF space into adjacent glial cells whose cellular membranes are freely permeable to K ⫹. Experimental elevation of ECF K ⫹ up to about 15 mmol L ⫺1 causes cerebral vasodilation, while higher levels cause vasoconstriction (46,50). This is one putative mechanism by which neuronal activity signals the glial cell to cause vasodilation. The effector at ‘‘glial feet’’ may be either K ⫹ or adenosine (46,51). In cerebral blood vessels, four different potassium channels have been described, one responding to ATP, one to calcium, and two types of rectifier channels, the ATP channels being most clearly linked to the hypoxic response (51,52). However, I have found no evidence that hypoxia causes a rise in ECF K ⫹ in brain. Lactic Acid

As noted in Sec. III.A, normally CBF limits brain tissue Po 2 such that ECF and CSF lactate average 1.5 mM, twice the arterial level. Increasing CBF by hyperoxic hypercapnia lowers CSF lactate to arterial levels. One might expect altitude hypoxia to increase brain lactate production. Early studies in dogs led to the conclusion that lactic acidosis was the main hypoxic vasodilator (15), but studies at high altitude have not supported this possibility. Cain and Dunn (53) showed that acute exposure of dogs to 21,000 ft altitude (chamber) produced only transient increases in brain and cerebral venous blood lactate, the small rises having returned to control levels within 8 hours. Silver recorded rat brain ECF Po 2 , pH, [K ⫹], [Cl ⫺], [Ca 2⫹], lactate, and blood flow using ion-specific microelectrodes (17). Increased blood flow in response to close arterial injection of hypoxic blood occurred within 1–2 seconds of the fall in tissue Po 2 before changes in either pH or potassium. The immediate effect of step hypoxia was a rise of local CBF and ECF pH, excluding lactic acid as the mediator of vasodilation. The same conclusion was reached by Siesjo et al. using different methods (rapid freezing and brain chemical analysis) (43,50). Adenosine

Hypoxia to Pao 2 ⫽ 45 mmHg does not significantly reduce ATP, CP, or the cerebral metabolic rate for O 2 (CMRo 2) (54). It is therefore somewhat surprising to find that

Cerebral Circulation at Altitude

351

the final breakdown product of ATP, adenosine, is a potent cerebral endogenous vasodilator (47). In both brief (30 s) and longer (5 min) hypoxia, brain adenosine increases rapidly from 0.34 ⫾ 0.08 (SE) to 1.65 ⫾ 0.33 mM kg ⫺1, paralleling temporally the changes in cerebral blood flow (47). By using extremely rapid freezing (within 1 s) this group found that adenosine concentrations remained stable between Pao 2 200 and 100 mmHg, doubled at Pao 2 ⫽ 50 mmHg, and increased sevenfold when Pao 2 reached 30 mmHg. In prior studies using a slower in situ freezing technique, no increases in adenosine or its metabolites had been detected during comparable hypoxia. Three other groups have now documented an adenosine-hypoxic vasodilator link (55–57). Furthermore, pial arterial dilation by hypoxia measured through an implanted cranial window in rats was enhanced by inhibiting adenosine uptake and blocked by adenosine blockade with theophylline (57). Thus, adenosine plays an important role in hypoxic cerebral vasodilation, probably being expressed by glial cells at their capillary feet. Nitric Oxide

Nitric oxide, formerly called endothelial-derived relaxing factor (EDRF ), is generated by endothelial cells and acts locally on the arteriolar smooth muscle. While NO is intimately involved in the CBF response to hypercapnia and hypocapnia (58), it seems to play a less easily defined role in hypoxic vasodilation (8,33,49). NO inhibition by N-nitro-l-arginine methyl ester (L-NAME) did not alter the increases in CBF elicited by ventilation with 8% oxygen for 25 s (59), but the selective nNOS inhibitor 7-nitroindazole completely blocked hypoxic hyperemia in other studies (33,60). Other Speculative Mechanisms

It now seems probable that cerebral vasodilation in hypoxia is signaled through glial cells which ‘‘connect’’ neurons to the nearest arteriolar smooth muscle cells. Adenosine, H ⫹, K ⫹, and NO may serve as vasodilator messengers from glia to smooth muscle, but of these only adenosine and NO are confirmed transmitters in hypoxia. It remains unknown whether oxygen supply to some critical metabolic process must be impaired to cause vasodilation, or whether some non–oxygenconsuming way of sensing Po 2 is at work. There are preliminary reports that oxygen directly affects the opening of potassium channels, but that mechanism is still unexplained since there is no known chromophore or metal reaction site on K ⫹ channels. E. Effects of Hypocapnia on CBF The Pco 2 –pH ECF —CBF Relationship

In normoxia, CBF falls about 3–4% per mmHg decrease of Paco 2 from normal (48). ECF H ⫹ is assumed to mediate the response to Pco 2 (61–63). Local vasodilation occurs with topically applied weak fixed acids (13). A fall of Pco 2 to 20 mmHg reduces CBF by half, but further hypocapnia has little effect, vasoconstriction being

352

Severinghaus

limited by tissue hypoxia. Hypercapnia increases CBF to a maximum two to three times normal. Brain Pco 2 is about 6 mmHg higher than Paco 2. At normal Paco 2 of 40 mmHg, a fall of 1 mmHg causes a rise of 0.01 pH in CSF and ECF. The bloodbrain barrier consists of endothelial cells with tight junctions, which effectively block arterial blood nonrespiratory acid-base changes from both the arteriolar smooth muscle cells and from most of the brain ECF (62,64,65). Even in fetal lambs, where the blood-brain barrier is generally thought to be incompletely formed, and thus more permeable, variation of pHa from 6.56 to 7.59 had no detectable influence on CBF at constant Paco 2 (25). The relationship between brain ECF pH and CBF at altitude was difficult to unravel, in part because investigators have studied lumbar CSF as a proxy for brain ECF, and partly because of the potential for error in measuring pH in poorly buffered CSF and its variation with Pco 2 before and during sampling. The evidence now suggests that as long as the carotid chemoreceptor drive remains elevated by hypoxia, the balance of respiratory drives is tilted, the increased ventilation causing medullary respiratory center and cerebral ECF pH to remain above normal. This alkalosis presumably continues to exert a vasoconstricting effect, in opposition to hypoxic vasodilation (66,67). Acclimatization gradually increases Pao 2 as ventilation rises. In addition, capillary Po 2 may rise if blood pH falls, reducing the alkaline Bohr shift. P 50 (measured at pH ⫽ 7.4) was found to be elevated 4 mmHg during the early days of acclimatization according to Lenfant et al. (68) but 2 mmHg or less in a similar test by Weiskopf et al. (69). Site of Action of CO 2: Neural Cells or Endothelium?

In order to answer this controversial question. Severinghaus and Lassen (70) determined the time constant of the CBF response to step hypocapnia. Their rationale was that the washout of CO 2 from the endothelium would be very rapid compared to its predicted 1–2 minutes 63% washout from bulk brain tissue. In gray matter, CBF ⬃1 mL g ⫺1 min ⫺1, which predicts a CO 2 washout time constant of 1 minute. White matter CBF is about one third that of gray. CBF was estimated from the jugular vein to arterial Sao 2 difference, assuming constant CMRo 2. Following sudden hyperventilation to a constant end-tidal Pco 2 of 20 mmHg, the CBF time constant was 20–25 seconds (70). This suggested that the flow regulation responds directly to arteriolar smooth muscle ECF pH, not to gray matter Pco 2. Effect of pH on Blood O 2 Affinity

Blood O 2 affinity is quantified as P 50 , the Po 2 resulting in 50% O 2 saturation of Hb. Under standard conditions, at pH ⫽ 7.4, T ⫽ 37°C, P 50 in adults at sea level averages 26.6 mmHg. Affinity changes induced by pH, temperature, and 2,3-DPG affect Po 2 proportionally at all levels of O 2 saturation, so a 10% rise of P 50 means multiplying by 1.10 all Po 2 values of the ODC. Acute respiratory alkalosis not only decreases cerebral blood flow, but also increases the affinity of hemoglobin for oxygen (Bohr effect), which thereby lowers capillary and brain tissue Po 2 and increases brain and

Cerebral Circulation at Altitude

353

blood lactate concentrations (68,71). For example, P 50 falls 20% to 21.3 mmHg at pH ⫽ 7.57 and BE ⫽ ⫺10 mM L ⫺1 (the values obtained at Pb ⫽ 250 mmHg in volunteers acclimatized in a chamber simulating ascent of Everest) (72). An acute left shift in the ODC when combined with hemorrhagic shock diminished shock tolerance with evidence of increased sympathetic outflow and increased mortality (42). In 10 newborn lambs Rosenberg (73) showed that as PaCO 2 was reduced from 46 to 12 mmHg, CMRo 2 was unchanged, while brain lactate production increased. Exchange transfusion of normal rats with low P 50 blood resulted in a major increase in cerebral flow. At high altitude the standard P 50 (at pH ⫽ 7.4) rises slightly, but because of hyperventilation alkalosis, the net effect is a left shift in most subjects. At altitude, a left shift increases O 2 loading in the lung about enough to compensate for its impairment of unloading in tissues. Interaction of Hypocapnic Vasoconstriction and Bohr Effect on O 2 Delivery

The effect on O 2 delivery of the conflicting responses of hypoxic vasodilation, hypocapnic vasoconstriction, and alkaline Bohr shift depends on the relative strengths of these three regulatory mechanisms. Each is subject to interindividual variability, and this may be important in functioning at high altitude. There may be pharmacological ways of limiting the vasoconstriction or Bohr shift that would ease life at extreme altitude. A comparative physiological approach has been used to identify patterns of the balance of these three effects in species that tolerate altitude well. Some birds, especially bar-headed geese, fly over Everest at altitudes as high as 9000 m. Grubb et al. (74) reported that artificially ventilated Pekin ducks have weaker hypocapnic cerebral vasoconstriction than mammals but a mammal-like hypoxic vasodilation (75). Faraci et al. (76) compared the effects of hypoxia in awake bar-headed geese (Anser indicus) and Pekin ducks (Anas platyrhynchos). While Paco 2 decreased to about 7 mmHg at 28 mmHg Pao 2 in both, arterial O 2 content (CaO 2) in geese (10.4 mL dL ⫺1) was significantly higher than in ducks (4.1 mL dL ⫺1) due to a left-shifted ODC in the geese. O 2 delivery to the brain of geese was the same as or higher than that of ducks, even though cerebral blood flow increased more than fivefold in ducks but less than threefold in geese. Faraci and Fedde then confirmed Grubb’s report, finding that in artificially ventilated, normoxic geese, hypocapnia to as low as Paco 2 ⫽ 7 mmHg did not significantly reduce CBF below normal (77), although hypercapnia (Paco 2 ⫽ 60 mmHg) tripled CBF. These geese also were found to have a greater cerebral capillary density and higher diffusing capacity, permitting them to consume O 2 at the lower capillary Po 2 resulting from their strongly left-shifted ODC. Bickler and Julian (11) studied the effect of extreme hypocapnia on cortical blood flow and tissue Po 2 (using O 2 and H 2 microelectrodes) in anesthetized normoxic geese (Anser domesticus). CBF decreased as Paco 2 fell to 20 mmHg but was not affected by further Pco 2 reduction, similar to the findings of Faraci and Fedde. This threshold for maximum vasoconstriction has been reported in mammals, includ-

354

Severinghaus

ing humans, and has been attributed to reduced O 2 delivery since CBF will fall farther with 100% inspired O 2. Despite the plateau in CBF during severe hypocapnia, tissue Po 2 continued to decline as Paco 2 fell but may have reached a plateau below 15 mmHg (Fig. 3). They noted that, at very low Pco 2 , reduction of Pao 2 to 30–40 mmHg had no significant effect on brain tissue Po 2 , which may be interpreted as suggesting that CBF increased (not measured), or that more lactic acid entered the capillary blood, maintaining Po 2 by the Bohr effect, as it does in exercising muscle. Thus, in high-flying geese, CBF is less reduced by hypocapnia than in mammals. During severe hypocapnic hypoxia, the left-shifted ODC delivers a higher Sao 2 to brain at very low Pao 2. The brain apparently tolerates the low capillary Po 2 well, flow being less than in ducks at similar Paco 2 and Pao 2. Apparently, geese utilize higher O 2 content despite a lower Po 2 because diffusing capacity and capillary density are greater. Unfortunately, as far as I could determine, none of the studies of awake bar-headed geese have measured CBF during the combination of severe hypocapnia and hypoxia expected at 9000 m. Role of Nitric Oxide in Hypercapnia

NO is involved in the CBF increase caused by hypercapnia (13,58,78). NO affects both major cerebral arteries and arteriolar sphincters (79). In rats an inhibitor of NO

Figure 3 Effect of normoxic hypocapnia on duck brain tissue Po 2 (implanted polarographic electrodes). Although CBF reduction reached a plateau at about 20 mmHg, as in humans, brain tissue Po 2 continued to fall in some birds with Pco 2 reduction, due largely to the leftshift of the ODC curve (Bohr effect of high pH). Symbols represent different birds. (From Ref. 11.)

Cerebral Circulation at Altitude

355

synthase decreased normocapnic CBF by 40% despite an increase of arterial blood pressure and attenuated CO 2 reactivity (49,80). However, the vasoconstriction of hypocapnia is unimpaired by blockade of NO synthesis. Acetazolamide and CBF

Acetazolamide (Diamox  ) is widely used to reduce AMS during acclimatization and improve sleep at altitude. Its mechanism of action on several aspects of blood gas transport and acid-base balance are complex and have led to controversy about how it helps the sojourner at altitude. Acetazolamide blocks carbonic anhydrase (81), primarily in red blood cells and renal tubules, but also in brain cells. Unloading of CO 2 from HCO 3 ⫺ in red cells is delayed in the lung, so blood CO 2 content is increased but alveolar Pco 2 is reduced. PaCO 2 rises slowly after blood leaves the lung so its value in tissues depends on the time from lung to tissue. Respiration is stimulated (82,83) by a rise of Pco 2 in the medullary respiratory chemoreceptors, as in all tissues, because metabolic CO 2 cannot rapidly be transported and converted in red cells into HCO 3 ⫺ (84). The ventilatory rise improves oxygenation during acute hypoxia (85) without an increase in lactate (86,87). Acetazolamide increases CBF in dogs (88). In 10 adult patients given 1 g acetazolamide, CBF rose by 66% within 30 minute (89). Oral administration of 1 g of acetazolamide to eight normoxic subjects caused a 38% increase in CBF (90, 91). However, at sea level continued administration does not sustain high CBF in normoxic volunteers (90,92). Acetazolamide increases brain bicarbonate concentration (93) and raises tissue Pco 2 about 6 mmHg, while Paco 2 is significantly lower when blood reaches capillaries having left the lung at the low alveolar Pco 2. Acetazolamide increases cortex surface Po 2 16–20 mmHg, eliminating a proposed explanation of the vasodilation as due to Bohr shift hypoxia (94). Acetazolamide causes brain carbonic acidosis (95), reducing brain surface pH by 0.1 pH within 3 minutes (96) (Fig. 4), while intracellular pH (MRS using 31P) does not fall (limit of detection ⫾0.06) (97). Several authors have suggested that little of the benefit of acetazolamide at altitude is due to increased CBF, but that the beneficial effects are more due to increased ventilation raising Pao 2 , affording a significant increase of the arterial oxygen content (98). However, the definitive studies at altitude remain to be conducted, since one must consider the effect of the drug on ventilation, Pco 2 , Po 2 , Sao 2 , acid-base balance, and CBF together. My impression is that CBF is higher with acetazolamide at altitude at the lower Pco 2 it causes than it would have been at that Pco 2 without the drug. IV. CBF Changes During Acclimatization A. Time Course of CBF Change at Altitude

The effect of altitude on CBF depends in part on the changes of Po 2 and Pco 2. With rapid ascent to high altitude, ventilation immediately increases slightly reducing

356

Severinghaus

Figure 4 Acetazolamide (25 mg/kg IV in a dog) causes a rapid acidification of brain ECF (surface flat pH electrode) even when tissue Pco 2 (flat surface electrode) is held constant (by increased ventilation). This effect cannot be explained by the effect on red cell carbonic anhydrase and indicates a rise of tissue carbonic acid due to delay of dehydration of metabolically produced H ⫹ and HCO 3 ⫺ to CO 2 within the brain. (From Ref. 96.)

Paco 2. During the next hours, days, and weeks the process of acclimatization results in further increase in ventilation, rise of Po 2 and fall of Pco 2 , affecting CBF. While the literature implies considerable variability of this CBF response, in general, in humans at moderate altitudes, CBF usually rises initially and then falls slowly over the first week at altitude. Do the concurrent blood gas changes account for the observed CBF time course? The first measurements of CBF during acclimatization to high altitude were obtained by Severinghaus et al. in seven normal men using the N 2O method of Kety and Schmidt (19,99). After 6–12 hours at 3810 m, CBF was increased 24% from the sea level value with Paco 2 ⫽ 35.0, Pao 2 ⫽ 43.5 mmHg (Fig. 5). Acute normoxia caused CBF to return to sea level values, although Paco 2 remained low (35.1 mmHg). After 3–5 days at this altitude, CBF had decreased but was still 13% above sea level control values (Pao 2 ⫽ 51.2, Paco 2 ⫽ 29.7 mmHg). Acute normoxia again caused CBF to fall to sea level control, while Paco 2 rose only to 30.9 mmHg. The continued hyperventilation with acute normoxia demonstrates a resetting of the medullary ventilatory CO 2 chemoreceptor ‘‘set point.’’ When 4% CO 2 was added to the normoxic gas, Paco 2 rose from 30.9 to 35.2 mmHg. CBF increased 34% above sea level control. We may estimate that at sea level, to increase CBF 34% would require Paco 2 to be raised about 7 mmHg. Thus the set point of the relationship of CBF to Paco 2 had been reset downward by about 10 mmHg, a resetting similar to that of ventilatory control.

Cerebral Circulation at Altitude

357

Figure 5 CBF in five newcomers after 6–12 hours and 3–5 days at 3810 m altitude. The acute responses to normoxia and normoxic hypercapnia show the resetting of hypercapnic response to a lower Pco 2. (Replotted data from Ref. 99.)

This time course of prompt rise and gradual decline of CBF with hypoxia has been observed by others. Jensen et al. reported that in 12 lowlanders ascending to 3475 m, CBF was increased 24% at 24 hours and was still 4% above sea level values after 6 days (100). In nine others acclimatized to 3200 m, ascent to 4785– 5430 m increased CBF 53% above estimated sea level values. Similarly, in 13 soldiers transported to 3700 m altitude, Roy et al. (101) found that CBF was 40% above control at 12–36 hours of hypoxia, and diminished to 4% above control after 4 days. Gradual fall of CBF with time in chronic hypoxia has also been reported by others (102–105). The ventilatory and blood gas changes in acute hypoxia suggest that the rise of CBF should occur within minutes and then decrease over days. However, this dominance of initial vasodilation has not always been evident. Huang et al. (102) used TCD to measure the CBFv in the internal carotid and vertebral arteries of six healthy men (Fig. 6). After 2–3 days a CBF increase of 20% was observed. Subsequently (days 4–12) CBFv declined to values close to those observed originally at sea level. However, 2–4 hours after arrival at the summit of Pikes Peak (4300 m), they reported (but excluded from the plot of Fig. 6) that CBFv in both arteries was only slightly increased above sea level values, even though the lowest arterial O 2 saturations (Sao 2) were observed on arrival. This lack of prompt increase of CBF

358

Severinghaus

Figure 6 CBF measured by TCD in men during acclimatization on Pikes Peak was only slightly elevated on day 1, but increased on days 2 and 3, followed by a reduction on days 4–12. The failure to promptly rise at altitude was unexpected and unexplained, but was presumed to indicate a vasoconstrictive effect of the initial hypocapnia. (From Ref. 102.)

was thought to be probably due to an initial vigorous hypocapnic vasoconstriction. Attempts to assess the effect of exercise on CBF have been performed on lowlanders on Pikes Peak and comparing Han to Tibetans in Lhasa (106,107). These studies are insufficient to conclude that increases of CBF exceed those that might be expected from the increased regional brain neural activity of motor cortex involved in work. Serial studies using TCD face problems of probe replacement, changing vessel diameter, and altered peak-to-mean Doppler pulse flow signal ratio when heart rate and arterial pressure rise (108). In nine healthy subjects, Koppenhagen (109) investigated the effect on CBF of ascent in a chamber over 25–30 minutes. They found no change in CBF after 2 hours at 4000 m and a 10% decrease after 2 hours at 7000 m. Why CBF did not increase acutely was not explained, but again might be because in these subjects hypocapnic cerebral vasoconstriction dominated acutely. Pco 2 was not reported. Furthermore, in animal studies not all investigators have observed the tendency toward normalization of CBF with time, especially with more extreme hypoxia. LaManna et al. (110) noted that in rats, despite an increased Hct and thus increased oxygen-carrying capacity, regional cerebral blood flow remained elevated after 4 weeks in a chamber at 0.5 atm, was still elevated after 4 hours of normoxia despite increased Hct, and only returned to normal after 24 hours of normoxia. Aritake et al. reported that, in cats, CBF was elevated even after 4.5 months of hypoxia,

Cerebral Circulation at Altitude

359

with a final inspired O 2 concentration of 8% and Hct ⫽ 56% (111), and again flow was still elevated when these animals were made acutely normoxic. Finally, as we describe below, CBF is not elevated in either natives or longterm residents of high altitude despite low Sao 2 and Pao 2. Can we account for the slow reduction of CBF in terms of the known physiological factors regulating cerebral arteriolar tone? B. Mechanisms Normalizing CBF with Time

The initial rise of CBF in hypoxia and its variability are consistent with what is known about acute responses to combinations of low Po 2 and low Pco 2. Nevertheless, Pao 2 remains normal below sea level, and acute hyperoxia does slow CBF in altitude natives (39). Why then does CBF not remain elevated? A major question is whether there are other factors than changing blood gases that cause CBF to decline toward sea level values during long-term hypoxia. I cite eight variables that may be involved: Rising PaO 2 and Falling PaCO 2

At a constant altitude, the acclimatization process results in a rise of ventilation, which increases Pao 2 and reduces Paco 2. In some subjects this may be aided by resolution of acute pulmonary dysfunction or HAPE. Qualitatively at least both changes would be expected to decrease CBF. Of the two, Krasney et al. suggest that Paco 2 may be the more important (112). In sheep during 4 days of constant arterial hypoxia, there was no fall of CBF if Paco 2 was not permitted to fall. When Pco 2 was allowed to fall while arterial Po 2 was held constant, CBF did fall with time (Fig. 7). The Resetting of CSF Acid-Base Balance

The effect of Paco 2 on both CBF and ventilatory control have been shown to be mediated by ECF and CSF pH (99). Peripheral carotid body hypoxic stimulation at altitude causes an immediate fall of Paco 2 , which elevates ECF pH. This alkalinity decreases central ventilatory drive and constricts cerebral arterioles, counteracting the hypoxic vasodilation. The subsequent acclimatization process involves a resetting of the relationship of pH to Paco 2 by a change of ECF and CSF HCO 3 ⫺. CSF HCO 3 ⫺ falls, typically from 24 to 18 mM/L at 4000 m in about 5 days, mostly during the first day. Paco 2 falls in proportion to the fall of CSF HCO 3 ⫺, while CSF pH remains alkaline. This may be regarded as a metabolic acidosis compensating for the respiratory alkalosis of brain ECF, but it is not initially accompanied by an equivalent blood metabolic acidosis. With continued hypoxia, neural output from carotid body hypoxic chemosensitivity is upregulated. Over about 2 weeks this may double the hypoxic ventilatory response (113). This further reduces Paco 2 , which increases CSF alkalinity and in

360

Severinghaus

Cerebral Circulation at Altitude

361

turn presumably results in additional fall of CSF HCO 3 ⫺. The rise of Pao 2 with time slightly offsets the net effect of this increased gain in sensitivity. The balance of these two processes determines whether CSF pH falls, stays constant, or rises. The consensus is that CSF pH remains alkaline or even rises during the first few weeks, because the second process, carotid body upregulation, although slower, usually is stronger than the first, the compensatory metabolic acidosis. The resulting alkalosis is probably responsible for the falling CBF. When a subject acclimatized to altitude is made acutely normoxic, the removal of the hypoxic drive to peripheral chemoreceptors raises Paco 2 enough to cause CBF and CSF pH to fall to normal (114). Because CSF HCO 3 ⫺ is low and can only change slowly, this residual metabolic acidosis of CSF and ECF cause ventilation to remain somewhat elevated. Calculations suggest that the pH decrement at ventral medullary chemoreceptors needed to account for the continued hyperventilation and hypocapnia is about 0.005 units (114), too small to affect CBF measurably. Rising P 50

An increase of P 50 facilitates unloading of O 2 to tissue, but the overall benefit depends on both pulmonary loading and tissue unloading, and this is dependent upon the altitude involved. At moderate elevations (3500–5000 m), a right shift of the ODC should raise tissue Po 2 because uptake in the lung is less affected, Pa O2 being on a still relatively flat part of the ODC. At higher altitude and lower alveolar Po 2 , Sao 2 should be adversely affected by a right-shifted ODC, offsetting the advantage to unloading of oxygen at the tissues. At about 4000 m, a P 50 increase of 2–4 torr (69,115) might contribute slightly to tissue oxygenation, thereby enabling decreasing CBF with time. Rising Hct

CBF is inversely proportional to Hct at altitude as at sea level (see Sec. III.B), but the increase of Hct is too slow and modest at the moderate altitudes of these studies to play a significant role in this short-term adaptation of CBF. Autonomic Nervous System

Based on studies of sheep at simulated high altitude, Curran-Everett et al. (116) reported that increased sympathetic tone might increase cerebrovascular resistance, causing decreased CBF over time. However, no evidence exists that sympathetic tone increases with time at altitude, and in fact Ueno et al. found decreased adrener-

Figure 7 The CBF response to a step to steady hypoxia depends upon whether Paco 2 is held constant or not. With isocapnia, CBF continues to rise while with poikilocapnia, it slowly falls. (From Ref. 104.)

362

Severinghaus

gic receptor density in major cerebral arteries in pregnant sheep and their fetuses after 110 days at 3810 m (117). Capillary Morphology

A hypothetical remodeling of the microcirculation resulting in shorter critical diffusion paths between capillaries and mitochondrial cytochrome might enhance oxygen flux comparable to that shown in muscle with training (118). This possibility is supported by evidence for angiogenesis with prolonged hypoxia (119–121). After 49 days exposed to 7% O 2 , rat cerebral cortex, striatum, and hippocampus showed a significant increase of vessel density, but the changes were small despite the prolonged and severe hypoxia. One would not predict much improvement in diffusing capacity in humans acclimatizing to altitude, though a change in vascular density may influence CBF in natives of altitude. Biochemical Changes

In sheep, 4 days of constant hypoxia at constant normal Pco 2 resulted in sustained elevation of CBF (112). I have found no evidence of changes in intermediary metabolic pathways coupling hypoxia and CBF. If any adaption to hypoxia had occurred, one would anticipate reduction of cerebral vasodilation in response to an acute hypoxic challenge. But the contrary was found—CBF sensitivity actually increased in response to acute hypoxic (Fig. 1) after 5 days at 3810 m (26,122). Brain Metabolic Rate

If CMRo 2 were to decrease with prolonged hypoxia, CBF could also fall. In Quechua natives of the Peruvian altiplano, Hochachka et al. (123) found regional glucose metabolic rates reduced by as much as 20% by positron emission tomography (PET). The greatest differences from sea level dwellers were found in areas associated with higher cortical functions. The authors speculate that low cerebral metabolic rate could protect against hypoxia as is presumed to be the case in cold-blooded amphibians and reptiles. Acute hypoxia does not decrease CMRo 2 , nor does CMRo 2 appear to be decreased with acclimatization of lowlanders. No direct measurements of CMRo 2 in highlanders have been reported. To summarize, the reasons for gradual fall of CBF toward sea level values with time at altitude are likely multifactorial. Major contributors are the gradually falling Pco 2 with associated rises of Pao 2 and possibly CSF pH. Increases of P 50 and Hct may also play a lesser role. However, those factors that have been measured fail to explain why CBF is normal despite severe hypoxia in natives of high altitude. Whether individuals with low HVR who are more hypoxic and less hypocapnic at altitude have higher CBF than those with more vigorous HVR is not known. The cerebral vasoconstrictive effect of a strong HVR on CBF in climbers at extreme altitudes has been proposed as possibly contributing to subsequent residual neuropsychological deficits reported in some mountaineers (124) (see Sec. VII).

Cerebral Circulation at Altitude V.

363

Cerebral Circulation with Prolonged Hypoxia

Since CBF falls to normal with time at altitude in newcomers, it is not surprising that it has usually been found normal or below normal in natives of high altitude. The challenge is to determine whether the low flow is adequately explained by known factors. CBF is inversely related to Hct both at sea level (31) and in natives at high altitude (Fig. 2). At normal sea level Hct values, CBF was the same in altitude natives as the accepted normal in sea level natives. Marc-Vergnes et al. (125) reported a 20% lower CBF in 16 natives of the Bolivian altiplano compared with sea level natives. They made no correction for increased Hct. In five of their group, cisternal CSF pH averaged 7.357, no different from sea level values. They noted that the effects on CBF of increased and decreased alveolar O 2 were greater than in sea level natives, a contrast to the diminished change of ventilation with change of Po 2 in these lifelong residents of high altitude. CBF was reduced despite decreased ventilation and higher Pco 2 in those with a blunted HVR. Brain oxygen delivery is affected by hemoglobin oxygen affinity. Acute hypoxia elevates P 50 2–4 mmHg in newcomers, which tends to raise tissue Po 2. Winslow et al. reported P 50 values averaging 31.2 ⫾ 1.9 mmHg in natives of the Andean altiplano at pH ⫽ 7.4, 2.0 mmHg higher that group’s sea level average (115). This could help explain a chronically low CBF. In eight normal adult natives of Cerro de Pasco, Peru (4300 m), with a mean Hct ⫽ 58%, Milledge and Sørensen (39) found that breathing 100% O 2 increased the arterial-internal jugular O 2 content difference from 7.89 ⫾ 1.01 to 9.58 ⫾ 1.17 mL dL ⫺1. This was not caused by hyperventilation induced by hyperoxia, since these subjects, on average, showed no change in Paco 2 with 100% O 2. Their mean (a-v) O 2 content difference while breathing air at 4300 m was greater than that of normals at sea level, indicating a subnormal CBF. Assuming no change in CMRo 2 , they showed an 18% decrease of CBF with hyperoxia. Thus, vasodilation by hypoxia persists after a lifetime of exposure to high altitude. Because newcomers develop increasing carotid chemoreceptor drive, and therefore a lower Paco 2 than natives, their CSF pH remains somewhat alkaline for weeks or months. In some natives and long-term residents at altitude, a gradual loss of peripheral hypoxic chemosensitivity reduces the ventilatory drive and slowly normalizes arterial and CSF pH. Natives of the Peruvian altiplano were found to have normal arterial and CSF pH (126) but elevated pHa (7.44) by Winslow (115). The subjects of Sørensen’s study (38) showed no abnormalities in brain glucose or lactate metabolism compared with sea level values. The lactate/glucose index of 3.7 was unchanged by either one hour of breathing 100% O 2 or in some subjects by short periods of reduced inspired O 2 at altitude (LaPaz, 3800 m). In calves, CBF is normal in chronic hypoxia and is insensitive to acute Po 2 changes, while in cats and rats CBF may remain elevated in chronic hypoxia and fail to fall with acute normoxia. Human studies suggest that after long residence, or birth, at high altitude, CBF in chronic hypoxia may be normal or subnormal. This

364

Severinghaus

low flow is not explained by Pco 2 , CSF pH, or Hct. But in such persons CBF remains sensitive to acute changes of Po 2 in either direction. We are left to attribute the unexpectedly low flow to the slightly elevated P 50 , biochemical changes, increased diffusing capacity with increased capillarity, and/or reduced CMRo 2. Determining the roles of these factors in high altitude natives is a reasonable research target. VI. CBF in AMS and HACE: Cause or Effect? The possible relationships of CBF to altitude illnesses have been much discussed and studied to answer such questions as: Does the high CBF at altitude cause or exacerbate AMS and HACE (122,127,128)? Do subjects with AMS have a poor ventilatory response to hypoxia, a lower Sao 2 and a higher Pco 2 , which should result in higher CBF (129)? A. Does CBF Differ in Those with AMS or HACE Versus Those Without?

Reeves et al. (130) found no correlation of CBF with headache in newcomers at altitude. Jensen et al. reported that, in nine normal subjects, ascending from 3200 to 4785–5430 m, CBF increased 53% above estimated sea level values. Increases in CBF were similar in subjects with or without AMS (100). Baumgartner et al. (131) assessed CBFv of both middle cerebral arteries by TCD in 23 subjects at 490 m and over 3 days after a rapid ascent to the Capana Margarita laboratory on Monta Rosa (4559 m). After ascent, CBFv increased 48% in the subjects who developed AMS and 27% in subjects who did not. The rise of CBFv correlated inversely with arterial Po 2 but was unrelated to blood pressure, Paco 2 , or [Hb]. They concluded that subjects with AMS had a higher CBFv than healthy subjects due to a lower arterial Po 2 , presumably representing a poorer HVR. It is unlikely that studies of this kind can answer the question of etiology of AMS and HACE because those conditions may impede CBF by the increased intracranial pressure. B. Attempts to Treat AMS with CO 2

For many years (since Mosso proposed acapnia to be the cause of AMS in 1898) (132), there has been a lingering suspicion that hypocapnia contributed to the illness. Attempts to test this have been made by adding CO 2 to inspired air during hypoxic exposure. To determine the role of CO 2 , Yang et al. (24) exposed chronically instrumented ewes to 96 hours of hypoxia (Pao 2 ⫽ 40 mmHg) in an environmental chamber. One group was permitted to become hypocapnic (Paco 2 ⫽ 27 mmHg), while the other was kept eucapnic (Paco 2 ⫽ 37 mmHg). AMS, estimated from food and water intakes and behavior, occurred in 9 of 12 with hypocapnia and 9 of 9 with

Cerebral Circulation at Altitude

365

normocapnia ( p ⬍ 0.05). Intracranial pressure and the ICP-sagittal sinus pressure gradient increased in AMS sheep in both groups. CBF was greater in the normocapnic animals. Cerebral O 2 and glucose uptakes were normal in both groups. Brain edema, reflected by elevated wet-to-dry tissue weight ratios, occurred only in AMS sheep. They concluded that AMS is associated with cerebral edema despite normal brain aerobic metabolism and that CO 2 supplementation at constant Pao 2 did not reduce AMS, intracranial pressure, or brain tissue edema. Maher et al. (133) tested whether prevention of hypocapnia and alkalosis would ameliorate the symptoms of acute mountain sickness (AMS). Five subjects were exposed to simulated high altitude for 4 days but kept isocapnic by adding 3.8% CO 2 to the chamber. Four other subjects were allowed to become hypocapnic. Alveolar oxygen tensions (55–60 mmHg) were held equal (by adjusting barometric pressure). They reported that severity of symptoms was ‘‘clearly greater’’ in normocapnic than in hypocapnic subjects (no symptom scoring was done). Harvey et al. (134) added 3% CO 2 to inhaled air in six subjects with AMS during a medical expedition to 5400 m. However, they made no effort to keep alveolar or arterial Po 2 constant, so the resulting higher alveolar Pco 2 and rise of ventilation increased alveolar oxygen tension by 24–40%. Symptoms of AMS were rapidly relieved. In three subjects cerebral blood flow increased by 17–39%, so that oxygen delivery to the brain would have been considerably improved. They pronounced their findings a ‘‘rediscovery’’ despite the preceding paper by Maher et al. using constant alveolar Po 2 showing an opposite result. The claim of beneficial effect of addition of 3% CO 2 to air was soon challenged by Ba¨rtsch et al. (135), who randomly allocated 20 mountaineers with AMS at 4559 m to three treatment groups: (1) with 33% O 2 , (2) with 3% CO 2 in air, and (3) an air control. Thirty-three percent O 2 significantly relieved symptoms of AMS and reduced CBF, but CO 2 addition did not significantly ameliorate AMS, despite the rise of Pco 2 , ventilation, and alveolar Po 2. The differing results in these studies appear to be explained by differences in Po 2 and Sao 2 caused by experimental design. When CO 2 is added to inspired air, it only slightly reduces Pi O2 but raises Pa O2 by inducing hyperventilation. Such studies find an improvement in or less AMS. When Pao 2 is constant at differing Paco 2 levels, AMS is found to be more severe with the higher Paco 2. This is explained in part by the effect in the lung of the higher Paco 2 , which reduces Sao 2 (due to the Bohr effect) at constant Pao 2. While this pulmonary loss is partly offset by higher capillary Po 2 , facilitating unloading, the net effect at the altitude used in these studies is to reduce capillary Po 2. At constant Pao 2 and Sao 2 , there is no evidence of a beneficial effect of isocapnia, despite their (assumed) higher CBF. It now seems clear that low Pco 2 is beneficial rather than contributing to AMS. C. The Proposed Role of High CBF in Causing AMS

The relationships of hypoxia and CO 2 to intracranial dynamics and CBF was investigated in sheep and goats by Yang et al. (78). In six unanesthetized goats subjected

366

Severinghaus

to hypobaric hypoxia at a simulated 4000 m altitude for 2 hours CBF increased from 1.46 ⫾ 0.11 to 1.85 ⫾ 0.08 mL g ⫺1 min ⫺1. ICP increased from 15.4 ⫾ 1.8 to 27.4 ⫾ 8.8 cm H 2O. With both 2 and 72 hours of hypoxia exposure, cerebral water content increased and intracranial compliance fell (136). In a review of this work, Krasney (128) postulated that HACE is in some way due to hypoxic cerebral vasodilation and elevated cerebral capillary hydrostatic pressure, which results in reduced brain compliance with compression of intracranial structures in the absence of altered global brain metabolism. He postulated that these primary intracranial events elevate peripheral sympathetic activity, which acts neurologically in the lung, possibly in concert with pulmonary capillary stress failure, to cause HAPE, and in the kidney to promote salt and water retention. He proposed that the adrenergic responses are likely modulated by striking increases of aldosterone, vasopressin, and atrial natriuretic peptide, compounded by increase in sympathetic neural activity. D. Evidence That High CBF Without Hypoxia Fails to Cause AMS

To test whether high CBF alone could cause cerebral edema, Yang and Krasney (137) kept sheep for 4 days in elevated CO 2 (52–55 mmHg). CBF remained about twice normal, and CMRo 2 was increased both during exposure and in the posthypercapnic period. They observed no symptoms like those of AMS or HACE, although brain water rose slightly from 79.8 ⫾ 0.24 to 80.3 ⫾ 0.2% ( p ⬍ 0.05). High CBF is unlikely to be the cause of HACE, and other factors must be sought. E.

Possible Role of VEGF and Angiogenesis in HACE

Retinal petechial hemorrhages found in many climbers at extreme altitude suggest a pathological process involving cerebral circulation, which may mirror changes throughout the brain. These hemorrhages may be a result of the first step in a process called angiogenesis (138). Tissue hypoxia is the initiating stimulus of angiogenesis, initiating growth of new capillaries into hypoxic tissue. Hypoxia causes expression in rat brain of vascular endothelial growth factor (VEGF), formerly termed vascular permeability factor (VPF ) (139). VEGF attacks and dissolves capillary basement membranes, permitting plasma and red cell leakage. Harik et al. found increased microvascular protein at one week and increased DNA in adult rats after 3 weeks at 0.5 A (119), and Mironov et al. reported increased capillary segment length (121). Dexamethasone inhibits angiogenesis. Twelve reports show that it is effective in preventing and treating AMS and HACE (138). VII. Brain Injuries at Extreme Altitude A major unanswered question is whether hypocapnic cerebral vasoconstriction plus the Bohr shift of the ODC contribute to AMS, neuropsychological functional limitation, or brain cellular injury. There is evidence that the brain is injured in some

Cerebral Circulation at Altitude

367

climbers at extreme altitudes, but it is not clear whether low blood flow may be involved (124). Song et al. (140) reported cerebral thrombi in several climbers who had gone higher than 5000 m for longer than 3 weeks. They speculated that the cause was hemoconcentration resulting from secondary polycythemia and dehydration at altitude. In the fetus and infant, hypoxemia, whether from high altitude or other causes, is associated with increased cerebrovascular morbidity. Longo et al. (141) compared cerebral arteries obtained from normoxic and chronically hypoxic sheep adults and fetuses. Long-term hypoxemia was associated with generalized increase in basesoluble protein (5–50%), a depression of the maximum potassium-induced tensions (16–49%), and a depression of the relaxation responses to S-nitroso-N-acetylpenicillamine (1–11%), which releases nitric oxide into solution upon hydration. They concluded that chronic hypoxemia depresses cerebral vascular smooth muscle and endothelial hypoxic response to a greater extent in the fetus than in adults. VIII. Personal Comment Regarding Hypoxic Distress During the past decade my colleagues and I have tested pulse oximeter accuracy by multiple brief (2–3 min) steps to Sao 2 values as low as 50% in several hundred subjects in order to calibrate instruments. I have noted that acute hypoxia uniformly evokes greater anxiety and hyperventilation on the first trial. Step hypoxia, even with voluntary hyperventilation and low Paco 2 , may be less well tolerated at 75% on the first run than at 55% on later runs. Since we allow only 1–4 minutes at high Po 2 between runs, we presume this subsequent relief of distress might be related to the phenomenon called hypoxic ventilatory decline in (HVD) in which ventilatory drive fades over 5–20 minutes to half or less at constant Po 2 and Pco 2. As a subject of hypoxic ventilatory or CBF experimentation, I have noted extreme anxiety on some hypoxic tests at levels of saturation that I found easily tolerable at other times. Hypoxia stimulates the sympathetic nervous system and catechol secretion. I note this observation because HVD may play a role in the variability of the rise in CBF in acute hypoxia, even at altitude, primarily through its effect on ventilation and Pco 2 since most evidence suggests that catecholamines and sympathetic nerves have no direct effect on CBF. IX. Summary Cerebral vasodilation by arterial hypoxemia occurs within 1 second and is mediated primarily by adenosine. Its site of expression is unknown but must be very close to the arteriolar smooth muscle in view of the rapid response, which tends to implicate the glial feet that invest all arterioles. H ⫹ from lactic acidosis may also contribute later. K⫹ and NO, known cerebral vasodilators, are not proved to be involved in hypoxic response. K⫹ dilates with neural activation, and NO with hypercapnia. During the first hours at altitude, CBF rises 30–60% and then over days declines

368

Severinghaus

toward normal sea level values. The magnitude of the initial rise depends on the altitude and four factors: the individual cerebral vascular and ventilatory sensitivities to both O 2 and CO 2. The reason for the subsequent reduction of flow remains poorly understood, but it is at least partly explained by the concurrent rise of Pao 2 and fall of Paco 2 , though not by polycythemia or reduced sensitivity of brain or cerebral vessels to hypoxia. Later effects of hypoxia with effects on CBF include reduced CSF HCO 3 ⫺ , increasing and then decreasing peripheral chemoreceptor sensitivity, and rising Hct. None of the known factors modulating CBF (including polycythemia) can fully account for the evidence that in several studies, natives of high altitude and acclimatized lowlanders have been found to have normal or subnormal CBF (compared with sea level normals). It seems probable that some remodeling of the capillaries increases local O 2-diffusing capacity. Reduced brain CMRo 2 has been reported in humans native to high altitude, which could help explain the low CBF. AMS and HACE may be accompanied by high CBF, because Pao 2 is lower, but high CBF (if caused by hypercapnia) fails to cause AMS or HACE. Hypocapnic vasoconstriction may contribute to the subtle brain injury seen in mountaineers after return from extremely high altitudes.

References 1. Li Y, Naqui A, Frey TG, Chance B. A new procedure for the purification of monodisperse highly active cytochrome c oxidase from bovine heart. Biochem J 1987; 242: 417–423. 2. Fennema M, Wessel JN, Faithful NS, Erdmann W. Tissue oxygen tension in the cerebral cortex of the rabbit. Adv Exp Med Biol 1989; 248:451–460. 3. Bickler PE, Litt L, Severinghaus JW. Effects of acetazolamide on cerebrocortical NADH and blood volume. J Appl Physiol 1988; 65:428–433. 4. Nemoto EM, Severinghaus JW. Stereospecific permeability of rat blood-brain barrier to lactic acid. Stroke 1974; 5:81–84. 5. Harder DR. Pressure-dependent membrane depolarization in cat middle cerebral artery. Circ Res 1984; 55:197–202. 6. Faraci FM, Heistad DD. Regulation of large cerebral arteries and cerebral microvascular pressure. Circ Res 1990; 66:8–17. 7. Ngai AC, Winn HR. Modulation of cerebral arteriolar diameter by intraluminal flow and pressure. Circ Res 1995; 77:832–840. 8. Hudetz AG. Blood flow in the cerebral capillary network: a review emphasizing observations with intravital microscopy. Microcirculation 1997; 4:233–252. 9. Clark LC, Misrahy G, Fox RP. Chronically implanted polarographic electrodes. J Appl Physiol 1958; 13:85–91. 10. Ozanne GM, Vilnis V, Severinghaus JW. Implications of O 2 wave shapes and synchrony for regulation of cerebral microcirculation. In: Harper M, Jennett WB, Miller JD, Rowan JO, eds. Blood Flow and Metabolism in the Brain. Edinburgh: ChurchillLivingston, 1975:9.3–9.7. 11. Bickler PE, Julian D. Regional cerebral blood flow and tissue oxygenation during hypocarbia in geese. Am J Physiol 1992; 263:R221–225.

Cerebral Circulation at Altitude

369

12. Frankel HM, Garcia E, Malik F, Weiss JK, Weiss HR. Effect of acetazolamide on cerebral blood flow and capillary patency. J Appl Physiol 1992; 73:1756–1761. 13. Wahl M, Schilling L. Regulation of cerebral blood flow: a brief review. Acta Neurochirurg 1993; 59 (suppl):3–10. 14. Cholet N, Seylaz J, Lacombe P, Bonvento G. Local uncoupling of the cerebrovascular and metabolic responses to somatosensory stimulation after neuronal nitric oxide synthase inhibition. J Cereb Blood Flow Metab 1997; 17:1191–1201. 15. Kogure K, Scheinberg P, Reinmuth OM, Fujishima M, Busto R. Mechanisms of cerebral vasodilatation in hypoxia. J Appl Physiol 1970; 29:223–229. 16. Freeman J, Ingvar DH. Elimination by hypoxia cerebral blood flow autoregulation and EEG relationship. Exp Brain Res 1968; 5:61–71. 17. Silver IA. Cellular microenvironment in relation to local blood flow. Ciba Found Symp 1978; 56:49–67. 18. Shimojyo S, Scheinberg P, Kogure K, Reinmuth OM. The effects of graded hypoxia upon transient cerebral blood flow and oxygen consumption. Neurology 1968; 18:127– 133. 19. Kety SS, Schmidt CF. The effects of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J Clin Invest 1948; 27:484–492. 20. Iwamoto J, Curran-Everett DC, Krasney E, Krasney JA. Cerebral metabolic and pressure-flow responses during sustained hypoxia in awake sheep. J Appl Physiol 1991; 71:1447–1453. 21. Johannsson H, Siesjo BK. Blood flow and oxygen consumption of the rat brain in profound hypoxia. Acta Physiol Scand 1974; 90:281–282. 22. Trojan S, Kapitola J. [Reactive hyperemia in the brain of rats after high altitude hypoxia]. Sb Lek 1990; 92:97–102. 23. Cohen PJ, Alexander SC, Smith TC, Reivich M, Wollman H. Effects of hypoxia and normocarbia on cerebral blood flow and metabolism in conscious man. J Appl Physiol 1967; 23:183–189. 24. Yang SP, Bergo GW, Krasney E, Krasney JA. Cerebral pressure-flow and metabolic responses to sustained hypoxia: effect of CO 2. J Appl Physiol 1994; 76:303–313. 25. Ashwal S, Dale PS, Longo LD. Regional cerebral blood flow: studies in the fetal lamb during hypoxia, hypercapnia, acidosis, and hypotension. Pediatr Res 1984; 18:1309– 1316. 26. Jensen JB, Sperling B, Severinghaus JW, Lassen NA. Augmented hypoxic cerebral vasodilation in man during five days at 3810 M altitude. J Appl Physiol 1996; 80: 1214–1218. 27. Macmillan V, Salford LG, Siesjo BK. Metabolic state and blood flow in rat cerebral cortex, cerebellum and brainstem in hypoxic hypoxia. Acta Physiol Scand 1974; 92: 103–113. 28. Cavazzuti M, Duffy TE. Regulation of local cerebral blood flow in normal and hypoxic newborn dogs. Ann Neurol 1982; 11:247–257. 29. Hamilton AJ, Cymmerman A, Black PM. High altitude cerebral edema. Neurosurgery 1986; 19:841–849. 30. Myers RE. Four patterns of perinatal brain damage and their conditions of occurrence in primates. Adv Neurol 1975; 10:223–234. 31. Harrison MJ. Influence of haematocrit in the cerebral circulation. Cerebrovasc Brain Metab Rev 1989; 1:55–67.

370

Severinghaus

32. Chapler CK, Cain SM. The physiologic reserve in oxygen carrying capacity: studies in experimental hemodilution. Can J Physiol Pharmacol 1986; 64:7–12. 33. Todd MM, Farrell S, Wu B. Cerebral blood flow during hypoxemia and hemodilution in rabbits: different roles for nitric oxide? J Cereb Blood Flow Metab 1997; 17:1319– 1325. 34. Holzman IR, Tabata B, Edelstone DI. Blood flow and oxygen delivery to the organs of the neonatal lamb as a function of hematocrit. Pediatr Res 1986; 20:1274– 1279. 35. Borgstrom L, Johannsson H, Siesjo BK. The influence of acute normovolemic anemia on cerebral blood flow and oxygen consumption of anesthetized rats. Acta Physiol Scand 1975; 93:505–514. 36. Johannson H, Siesjo BK. Brain energy metabolism in anesthetized rats in acute anemia. Acta Physiol Scand 1975; 93:515–524. 37. Hudak ML, Koehler RC, Rosenberg AA, Traystman RJ, Jones MDJ. Effect of hematocrit on cerebral blood flow. Am J Physiol 1986; 251:H63–H70. 38. Sørensen SC, Lassen NA, Severinghaus JW, Coudert J, Zamora MP. Cerebral glucose metabolism and cerebral blood flow in high-altitude residents. J Appl Physiol 1974; 87:305–310. 39. Milledge JS, Sørensen SC. Cerebral arteriovenous oxygen difference in man native to high altitude. J Appl Physiol 1972; 32:687–689. 40. Koehler RC, Traystman RJ, Zeger S, Rogers MC, Jones MDJ. Comparison of cerebrovascular response to hypoxic and carbon monoxide hypoxia in newborn and adult sheep. J Cereb Blood Flow Metab 1984; 4:115–122. 41. Koehler RC, Traystman RJ, Rosenberg AA, Hudak ML, Jones MDJ. Role of O 2-hemoglobin affinity on cerebrovascular response to carbon monoxide hypoxia. Am J Physiol 1983; 245:H1019–1023. 42. Woodson RD. Functional consequences of altered blood oxygen affinity. Acta Biol Med Ger 1981; 40:733–736. 43. Siesjo BK, Johannsson H, Norberg K, Salford I. Brain function metabolism and blood flow in moderate and severe arterial hypoxia. In: Ingvar DH, Lassen NA, eds. Brain Work. The Coupling of Function, Metabolism and Blood Flow in the Brain. Copenhagen: Munksgaard, 1975:101–125. 44. Faraci FM, Brian JEJ. Nitric oxide and the cerebral circulation. Stroke 1994; 25:692– 703. 45. Janigro D, Wender R, Ransom G, Tinklepaugh DL, Winn HR. Adenosine-induced release of nitric oxide from cortical astrocytes. Neuroreport 1996; 7:1640–1644. 46. Kuschinsky M, Wahl M, Bosse O, Thurau K. Perivascular potassium and pH as determinants of local pial arterial diameter in cats. A microapplication study. Circ Res 1972; 31:240–247. 47. Winn HR, Rubio R, Berne RM. Brain adenosine concentration during hypoxia in rats. Am J Physiol 1981; 241:H235–242. 48. Reivich M. Arterial Pco 2 and cerebral hemodynamics. Am J Physiol 1964; 206:25– 34. 49. Buchanan JE, Phillis JW. The role of nitric oxide in the regulation of cerebral blood flow. Brain Res 1993; 610:248–255. 50. Astrup J, Heuser D, Lassen NA, Nilsson BB, Norberg K, Siesjo BK. Evidence against H ⫹ and K ⫹ as main factors for the control of cerebral blood flow: a microelectrode study. Ciba Foundation Symp 1978; 56:313–337.

Cerebral Circulation at Altitude

371

51. Taguchi H, Heistad DD, Kitazono T, Faraci FM. ATP-sensitive K ⫹ channels mediate dilatation of cerebral arterioles during hypoxia. Cir Res 1994; 74:1005–1008. 52. Kitazono T, Faraci FM, Taguchi H, Heistad DD. Role of potassium channels in cerebral blood vessels. Stroke 1995; 26:1713–1723. 53. Cain SM, Dunn JE. Transient arterial lactic acid changes in unanesthetized dogs at 21,000 feet. Am J Physiol 1964; 206:1437–1440. 54. Hamer J, Wiedemann K, Berlet H, Weinhardt F, Hoyer S. Cerebral glucose and energy metabolism, cerebral oxygen consumption, and blood flow in arterial hypoxaemia. Acta Neurochir (Wien) 1978; 44:151–160. 55. Morii S, Ngai AC, Ko KR, Winn HR. Role of adenosine in regulation of cerebral blood flow: effects of theophylline during normoxia and hypoxia. Am J Physiol 1987; 253:H165–175. 56. Laudignon N, Farri E, Beharry K, Rex J, Aranda IV. Influence of adenosine on cerebral blood flow during hypoxic hypoxia in the newborn piglet. J Appl Physiol 1990; 68: 1534–1541. 57. Meno JR, Ngai AC, Winn HR. Changes in pial arteriolar diameter and CSF adenosine concentrations during hypoxia. J Cereb Blood Flow Metab 1993; 13:214–220. 58. Underwood MD, Iadecola C, Sved A, Reis DJ. Stimulation of Cl area neurons globally increases regional cerebral blood flow but not metabolism. J Cereb Blood Flow Metab 1992; 12:844–855. 59. McPherson RW, Koehler RC, Traystman RJ. Hypoxia, alpha-2-adrenergic, and nitric oxide dependent interactions on canine cerebral blood flow. Am J Physiol 1994; 266: H476–482. 60. Hudetz AG, Shen H, Kampine JP. Nitric oxide from neuronal NOS plays critical role in cerebral capillary flow response to hypoxia. Am J Physiol 1998; 274:H982–989. 61. Betz E, Kozak R, Heredia A. Die Bedeutung des pH der Gehrinrinde fu¨r die Regulation der corticalen Durchblutung. Pflugers Arch Ges Physiol 1966; 289:R3. 62. Harper AM, Bell RA. Effect of metabolic acidosis and alkalosis on the blood flow through the cerebral cortex. J Neurol Neurosurg Psychiatry 1963; 26:341–352. 63. Lambertsen C, Semple SJG, Smyth MG, Gelfand R. H ⫹ and Pco 2 as chemical factors in respiratory and cerebral circulatory control. J Appl Physiol 1961; 16:473–478. 64. Mitchell RA, Carman CT, Severinghaus JW, Richardson BW, Singer MM, Shnider S. Stability of cerebrospinal fluid pH in chronic acid base disturbances in blood. J Appl Physiol 1965; 20:27–30. 65. Severinghaus JW. Regulation of respiration and CSF acid base balance. In: Loeppke JA, Riedesel ML, eds. Oxygen Transport to Human Tissues. New York: Elsevier North Holland, Inc, 1982; 231–241. 66. Weiskopf RB, Gabel RA, Fencl V. Alkaline shift in lumbar and intracranial CSF in man after 5 days at high altitude. J Appl Physiol 1976; 41:93–97. 67. Forster HV, Dempsey JA, Chosy LW. Incomplete compensation of CSF H ⫹ in man during acclimatization to high altitude (4300 M). J Appl Physiol 1975; 38:1067– 1072. 68. Lenfant C, Torrence JD, Reynafarje C. The shift of the O 2 Hb dissociation curve at altitude: mechanism and effect. J Appl Physiol 1971; 30:625–631. 69. Weiskopf RB, Severinghaus JW. Lack of effect of high altitude on hemoglobin oxygen affinity. J Appl Physiol 1972; 33:276–277. 70. Severinghaus JW, Lassen NA. Step hypocapnia to separate arterial from tissue Pco 2 in the regulation of cerebral blood flow. Circ Res 1967; 20:272–278.

372

Severinghaus

71. Kennealy JA, McLennan JE, Loudon RG, McLaurin RL. Hyperventilation-induced cerebral hypoxia. Am Rev Respir Dis 1980; 22:407–412. 72. Houston CS, Sutton JR, Cymerman A, Reeves JT. Operation Everest II. Man at extreme altitude. J Appl Physiol 1987; 63:877–882. 73. Rosenberg AA. Response of the cerebral circulation to profound hypocarbia in neonatal lambs. Stroke 1988; 19:1365–1370. 74. Grubb B, Mills CD, Colacino JM, Schmidt-Nielsen K. Effect of arterial carbon dioxide on cerebral blood flow in ducks. Am J Physiol 1977; 232:H596–601. 75. Grubb B, Jones JH, Schmidt-Nielsen K. Avian cerebral blood flow: influence of the Bohr effect on oxygen supply. Am J Physiol 1979; 236:H744–749. 76. Faraci FM, Kilgore DLJ, Fedde MR. Oxygen delivery to the heart and brain during hypoxia: Pekin duck vs. bar-headed goose. Am J Physiol 1984; 247:R69–75. 77. Faraci FM, Fedde MR. Regional circulatory responses to hypocapnia and hypercapnia in bar-headed geese. Am J Physiol 1986; 250:R499–504. 78. Yang YB, Sun B, Yang Z, Wang J, Pong Y. Effects of acute hypoxia on intracranial dynamics in unanesthetized goats. J Appl Physiol 1993; 74:2067–2071. 79. Faraci FM. Role of endothelium-derived relaxing factor in cerebral circulation: large arteries vs. microcirculation. Am J Physiol 1991; 261:H1038–H1042. 80. Wang Q, Paulson OB, Lassen NA. Effect of nitric oxide blockade by NG-nitro-Larginine on cerebral blood flow response to changes in carbon dioxide tension. J Cereb Blood Flow Metab 1992; 12:947–953. 81. Maren TH, Mayer E, Wadsworth BC. Carbonic anhydrase inhibition. I. The pharmacology of Diamox, 2-acetylamino-1,3,4-thiadiazole-5-sulfonamide. Bull Johns Hopkins Hosp 1954; 95:199–255. 82. Tomashefski JF, Chinn HI, Clark RT. Effect of carbonic anhydrase inhibition on respiration. Am J Physiol 1954; 177:451–454. 83. Mithoefer JC. Inhibition of carbonic anhydrase: its effect on carbon dioxide elimination by the lungs. J Appl Physiol 1959; 14:109–115. 84. Brzezinski J, Kjallquist A, Siesjo BK. Mean carbon dioxide tension in the brain after carbonic anhydrase inhibition. J Physiol 1967; 188:13–24. 85. Carter ET, Clark RT. Effects of carbonic anhydrase inhibition during acute hypoxia. J Appl Physiol 1958; 13:47–52. 86. Cain SM, Dunn JE. Increase of arterial oxygen tension at altitude by carbonic anhydrase inhibition. J Appl Physiol 1965; 20:882–884. 87. Cain SM, Dunn JE. Low doses of acetazolamide to aid accommodation of men to altitude. J Appl Physiol 1966; 21:1195–1200. 88. Posner JB, Plum F. Toxic effects of carbon dioxide and acetazolamide in hepatic encephalopathy. J Clin Invest 1960; 39:1246–1258. 89. Ehrenreich DL, Gurns RA, Alman RW, Fazekas JF. Influence of acetazolamide on cerebral blood flow. Arch Neurol 1961; 5:125–134. 90. Friberg L, Kastrup J, Rizzi D, Jensen JB, Lassen NA. Cerebral blood flow and endtidal Pco 2 during prolonged acetazolamide treatment in humans. Am J Physiol 1990; 258:H954–959. 91. Lassen NA, Friberg L, Kastrup J, Rizzi D, Jensen JJ. Effects of acetazolamide on cerebral blood flow and brain tissue oxygenation. Postgrad Med J 1987; 63:185–187. 92. Huang SY, McCullough RE, McCullough RG, Micco AJ, Manco-Johnson M, Weil JV, Reeves JT. Usual clinical dose of acetazolamide does not alter cerebral blood flow velocity. Respir Physiol 1988; 72:315–326.

Cerebral Circulation at Altitude

373

93. Kjallquist A, Siesjo BK. Increase in the intracellular bicarbonate concentration in the brain after acetazolamide. Acta Physiol Scand 1966; 68:255–267. 94. Cotev S, Lee J, Severinghaus JW. The effects of acetazolamide on cerebral blood flow and cerebral tissue Po 2 , Anesthesiology 1968; 29:471–477. 95. Severinghaus JW, Hamilton RN, Cotev S. Carbonic acid production and the role of carbonic anhydrase in decarboxylation in brain. Biochem J 1969; 114:703–705. 96. Bickler PE, Litt L, Banville DL, Severinghaus W. Effects of acetazolamide on cerebral acid-base balance. J Appl Physiol 1988; 65:422–427. 97. Vorstrup S, Jensen KE, Thomsen C, Henricksen O, Lassen N, Paulson OB. Neuronal pH regulation: constant normal intracellular pH is maintained in brain during low extracellular pH induced by acetazolamide—31P NMR study. J Cerebral Blood Flow Metab 1989; 9:417–421. 98. LaManna JC, McCracken KA. Carbonic anhydrase inhibition and cerebral cortical oxygenation in the rat. Adv Exp Med Biol 1990; 277:335–343. 99. Severinghaus JW, Chiodi H, Eger Eld, Brandstater B, Hornbein TF. Cerebral blood flow in man at high altitude. Role of cerebrospinal fluid pH in normalization of flow in chronic hypocapnia. Circ Res 1966; 19:274–282. 100. Jensen JB, Wright AD, Lassen NA, Harvey TC, Winterborn MH, Raichle ME, Bradwell AR. Cerebral blood flow in acute mountain sickness. J Appl Physiol 1990; 69: 430–433. 101. Roy SB, Guleria JS, Khanna PK, Talwar JR, Manchanda SC, Pande JN, Kaushik VS, Subba PS, Wood JE. Immediate circulatory response to high altitude hypoxia in man. Nature 1968; 217:1177–1178. 102. Huang SY, Moore LG, McCullough RE, McCullough RG, Micco AJ, Fulco C, Cymerman A, Manco-Johnson M, Weil JV, Reeves JT. Internal carotid and vertebral arterial flow velocity in men at high altitude. J Appl Physiol 1987; 63:395–400. 103. Bocking AD, Gagnon R, White SE, Homan J, Milne KM, Richardson BS. Circulatory responses to prolonged hypoxemia in fetal sheep. Am J Obstet Gynecol 1988; 159: 1418–1424. 104. Krasney JA, Jensen JB, Lassen NA. Cerebral blood flow does not adapt to sustained hypoxia. J Cereb Blood Flow Metab 1990; 10:759–764. 105. Manohar M, Parks CM, Busch M, Bisgard GE. Bovine regional brain blood flow during sojourn at a simulated altitude of 3500 m. Respir Physiol 1984; 58:111–122. 106. Huang SY, Tawney KW, Bender PR, Groves BM, McCullough RE, McCullough RG, Micco AJ, Manco-Johnson M, Cymerman A, Greene ER, et al. Internal carotid flow velocity with exercise before and after acclimatization to 4,300 m. J Appl Physiol 1991; 71:1469–1476. 107. Huang SY, Sun S, Droma T, Zhuang J, Tao JX. McCullough RG, McCullough RE, Micco AJ, Reeves JT, Moore LG. Internal carotid arterial flow velocity during exercise in Tibetan and Han residents of Lhasa (3,658 m). J Appl Physiol 1992; 73:2638– 2642. 108. Poulin MJ, Syed RJ, Robbins PA. The use of transcranial doppler to assess cerebral blood flow during exercise in humans. FASEB J 1998; 12:A1116. 109. Koppenhagen K. Cerebral blood flow under hypobaric conditions: effects of pentoxifylline (‘Trental’ 400). Pharmatherapeutica 1984; 4:1–5. 110. LaManna JC, McCracken KA, Strohl KP. Changes in regional cerebral blood flow and sucrose space after 3–4 weeks of hypobaric hypoxia (0.5 ATM). Adv Exp Med Biol 1989; 248:471–477.

374

Severinghaus

111. Aritake K, Mayer HM, Fritschka E, Cervos-Navarro J, Takakura K. [Cerebral hemodynamics in chronic hypoxic hypoxia]. No To Shinkei 1986; 38:363–369. 112. Krasney JA, Hajduczok G, Miki K, Matalon S. Peripheral circulatory response to 96 hours of eucapnic hypoxia in conscious sheep. Respir Physiol 1985; 59:197– 211. 113. Sato M, Severinghaus JW, Powell FL, Xu FD, Spellman MJ. Augmented hypoxic ventilatory response in men at altitude. J Appl Physiol 1992; 73:101–107. 114. Crawford RD, Severinghaus JW. CSF pH and ventilatory acclimatization to altitude. J Appl Physiol 1978; 45:275–283. 115. Winslow RM, Monge CC, Statham NJ, Gibson CG, Charache S, Whittembury J, Moran O, Berger RL. Variability of oxygen affinity of blood: human subjects native to high altitude. J Appl Physiol 1981; 51:1411–1416. 116. Curran-Everett DC, Meredith MP, Krasney JA. Acclimatization to hypoxia alters cerebral convective and diffusive O 2 delivery. Respir Physiol 1992; 88:355–371. 117. Ueno N, Zhao Y, Zhang L, Longo LD. High altitude-induced changes in α1-adrenergic receptors and Ins(1,4,5)P3 responses in cerebral arteries. Am J Physiol 1997; 272: R669–R674. 118. Knight DR, Schaffartzik W, Poole DC, Hogan MC, Bebout DE, Wagner PW. Effects of hyperoxia on maximal leg O 2 supply and utilization in man. J Appl Physiol 1993; 75:2586–2594. 119. Harik SI, Hritz MA, LaManna JC. Hypoxia-induced brain angiogenesis in the adult rat. J Physiol 1995; 485:525–530. 120. Patt S, Sampaolo S, Theallier-Janko A, Tschairkin I, Cervos-Navarro J. Cerebral angiogenesis triggered by severe chronic hypoxia displays regional differences. J Cereb Blood Flow Metab 1997; 17:801–806. 121. Mironov V, Hritz MA, LaManna JC, Hudetz AG, Harik SI. Architectural alterations in rat cerebral microvessels after hypobaric hypoxia. Brain Res 1994; 660:73–80. 122. Lassen NA. Increase of cerebral blood flow at high altitude: its possible relation to AMS. Int J Sports Med 1992; 13 (suppl 1):S47–48. 123. Hochachka PW, Clark CM, Brown WD, Stanley C, Stone CK, Nickles RJ, Zhu GG, Allen PS, Holden JE. The brain at high altitude: hypometabolism as a defense against chronic hypoxia? J Cerebral Blood Flow Metab 1994; 14:671–679. 124. Hornbein TF, Townes BD, Schoene RB, Sutton JR, Houston CS. The cost to the central nervous system of climbing to extremely high altitude. N Engl J Med 1989; 321:1714– 1719. 125. Marc-Vergnes JP, Antezana G, Coudert J, Gourdin D, Durand J. [Cerebral blood flow and energy metabolism, and acid-base equilibrium in cerebrospinal fluid in high altitude residents]. J Physiol (Paris) 1974; 68:633–654. 126. Severinghaus JW, Carcelen AB. Cerebrospinal fluid in man native to high altitude. J Appl Physiol 1964; 19:319–321. 127. Foulke GE. Altitude-related illness. Am J Emerg Med 1985; 3:217–226. 128. Krasney JA. A neurogenic basis for acute altitude illness. Med Sci Sports Exerc 1994; 26:195–208. 129. Otis SM, Rossman ME, Schneider PA, Rush MP, Ringelstein EB. Relationship of cerebral blood flow regulation to acute mountain sickness. J Ultrasound Med 1989; 8:143–148. 130. Reeves JT, Moore LG, McCullough RE. Headache at high altitude is not related to internal carotid arterial blood velocity. J Appl Physiol 1985; 59:909–913.

Cerebral Circulation at Altitude

375

131. Baumgartner RW, Ba¨rtsch P, Maggiorini M, Waber U, Oelz O. Enhanced cerebral blood flow in acute mountain sickness. Aviat Space Environ Med 1994; 65:726–729. 132. Mosso A. Life of Man on the High Alps. London: T. Fisher Unwin, 1898. 133. Maher JT, Cymerman A, Reeves JT, Cruz JC, Denniston JC, Grover RF. Acute mountain sickness: increased severity in eucapnic hypoxia. Aviat Space Environ Med 1975; 46:826–829. 134. Harvey TC, Raichle ME, Winterborn MH, Jensen J, Lassen NA, Richardson NV, Bradwell AR. Effect of carbon dioxide in acute mountain sickness: a rediscovery. Lancet 1988; 2:639–641. 135. Ba¨rtsch P, Baumgartner RW, Waber U, Maggorini M, Oelz O. Comparison of carbondioxide-enriched, oxygen-enriched, and normal air in treatment of acute mountain sickness. Lancet 1990; 336:772–775. 136. Curran-Everett DC, Iwamoto J, Meredith MP, Krasney JA. Intracranial pressures and O 2 extraction in conscious sheep during 72 h of hypoxia. Am J Physiol 1991; 261: H103–H109. 137. Yang SP, Krasney JA. Cerebral blood flow and metabolic responses to sustained hypercapnia in awake sheep. J Cerebral Blood Flow Metabolism 1995; 15:115–123. 138. Severinghaus JW. Hypothetical roles of angiogenesis, osmotic swelling, and ischemia in high-altitude cerebral edema. J Appl Physiol 1995; 79:375–379. 139. Xu FP, Severinghaus JW. Rat brain VEGF expression in alveolar hypoxia: possible role in high-altitude cerebral edema. J Appl Physiol 1998; 85:53–57. 140. Song SY, Asaji T, Tanizaki Y, Fujimaki T, Matsutani M, Okeda R. Cerebral thrombosis at altitude: its pathogenesis and the problems of prevention and treatment. Aviat Space Environ Med 1986; 57:71–76. 141. Longo LD, Hull AD, Long DM, Pearce WJ. Cerebrovascular adaptations to high-altitude hypoxemia in fetal and adult sheep. Am J Physiol 1993; 264:R65–72.

12 The High-Altitude Brain

MARCUS E. RAICHLE

THOMAS F. HORNBEIN

Washington University Medical Center St. Louis, Missouri

University of Washington School of Medicine Seattle, Washington

I.

Introduction

While a sojourn at altitude can be an enjoyable and at times thrilling experience, brief visits to even modest altitudes quickly remind most of us of our limitations. Most obvious is the breathlessness we experience on climbing a short flight of stairs or navigating a modest incline in a trial. More sinister, though, because it is usually not apparent to us, is the decline in some of our most sophisticated cognitive skills due to the effect of altitude on the function of our brain. In many instances safety in these spectacular yet hazardous environments is dependent upon proper and timely decisions about what courses of action to take and what to avoid. Such decisions depend upon a functioning brain. Some of the tragedies visited upon humans at altitude, such as the events on Everest in 1996, may have resulted from a critical loss of judgment as the result of brain hypoxia (1). That the mortality during the descent from the summit of Everest is three times greater for those who did not use oxygen than in those who did lends credence to this possibility (2). This chapter focuses on the effect of high altitude on human brain function. What we address in this review are not the high metabolic demands of the resting brain but the additional metabolic and circulatory requirements associated with per377

378

Raichle and Hornbein

ception, thought, and action. Lack of sufficient oxygen is certainly a plausible explanation for the changes in perception, thought, and action observed, given the brain’s large appetite for oxygen and the reduced availability of oxygen at altitude. As will become apparent, the exact mechanism or mechanisms underlying the neurobehavioral abnormalities associated with high altitude remain largely unknown. A variety of factors undoubtedly contribute. We will enumerate these factors in the hope that future research can focus on promising aspects of the problem. We begin this review with a look at the behavioral effects of altitude on brain function as well as several sea level conditions that mimic some of the effects of altitude. We then consider the metabolic requirements for normal brain function and, finally, the manner in which the function of the brain might be disrupted during and after exposure to high altitude. II. Effect of Acute Hypoxia on Behavior Nothing caught the attention of the public regarding the potential dangers of an acute exposure to high altitude more than did the exploits of early balloonists. The flavor of these escapades is captured in the commentary by Glaisher on the experiences of himself and his colleague Coxwell in 1862 as they ascended to heights thought to be in excess of the summit of Mt. Everest (3): In consequence, however, of the rotatory motion of the balloon, which had continued without ceasing since leaving the earth, his valve-line had become entangled, and he (Coxwell) had to leave the car and mount into the ring to readjust it. I then looked at the barometer, and found its reading to be 9 and 3/4 inches, still decreasing fast, implying a height exceeding 29,000 feet. Shortly after I laid my arm upon the table, possessed of its full vigour, but on being desirous of using it I found it powerless—it must have lost its power momentarily; trying to move the other arm, I found it powerless also. Then I tried to shake myself, and succeeded, but I seemed to have no limbs. In looking at the barometer my head fell over my left shoulder; I struggled and shook my body again, but could not move my arms. Getting my head upright for an instant only, it fell on my right shoulder; then I fell backwards, my back resting against the side of the car and my head on its edge. In this position my eyes were directed to Mr. Coxwell in the ring. When I shook my body I seemed to have full power over the muscles of the back, and considerably so over those of the neck, but none over either my arms or my legs. As in the case of the arms, so all muscular power was lost in an instant from my back and neck. I dimly saw Mr. Coxwell, and endeavored to speak, but could not. In an instant intense darkness overcame me, so that the optic nerve lost power suddenly, but I was still conscious, with as active a brain as at the present moment whilst writing this. I thought I had been seized with asphyxia, and believed I should experience nothing more, as death would come unless we speedily descended: other thoughts were entering my mind when I suddenly became unconscious as on going to sleep.

Thirteen years after Glaisher and Coxwell’s flight, the Zenith took off from Paris carrying Tissandier, Croce-Spinelli, and Sivel and bags containing oxygen

The High-Altitude Brain

379

provided by Paul Bert to enable a high ascent. But hypoxia dulled the mind and the three forgot to use the oxygen. When the balloon settled back to earth only Tissandier was alive to describe their journey toward the stars, his two companions dead from a surfeit of too little oxygen. While the experience of Glaisher and Coxwell tends to suggest that the major impairments short of unconsciousness were, first, motoric and then visual, the experience of Tissandler, Croce-Spinelli, and Sivel (4) more truly capture much of the subsequent literature. This literature suggests that a critical impairment in higher cognitive functions is the earliest and most insidious consequence of exposure to high altitude. We can suspect that Glaisher experienced the same difficulties, one of the signs of which is failure to recognize one’s own limitations. Triggered by these early balloon ascents to explore the troposphere and compelled particularly by concerns about pilot performance in aircraft of World War II, the literature on real-time effects of acute hypoxia is venerable, largely descriptive, and limited in scope. Much early work is reviewed by Stickney and Van Liere (5) and Tune (6). Seminal among these early studies is the account published by McFarland (7) in 1937 as the first in a series of papers on the effect of hypoxia on mental performance at high altitude. Utilizing a variety of observations and tests of sensory, motor, and cognitive function, McFarland observed that individuals taken rapidly (hours) to 15,000–16,500 feet exhibited impairment in both simple and complex psychological performance. Motor functions, such as handwriting, were also impaired but to a lesser extent, and sensory modalities were affected little if at all. We will focus upon a sample of more recent studies that fundamentally reinforce these earlier observations. Because of the well-recognized sensitivity of dark adaptation to acute hypoxia and concern about the consequence to night vision of pilots, increasingly sophisticated inquiries into visual deficits of graded acute hypoxia have accumulated in the literature. Kobrick and coworkers have examined the effects on peripheral as well as central vision (8). The time taken to respond to peripheral visual stimuli increases with increasing altitude between sea level and 17,000 feet. The slowing is less when the individual is concentrating on a central visual target, perhaps because the central task increases alertness. The authors concluded that the hypoxic-induced delay is a consequence of a perceptual deficit rather than an effect on retinal receptors per se. This theme of perceptual deficit recurs, as we shall see. Other consequences of acute hypoxia have been reviewed by Ernsting (9) and others. Response or reaction times are a usual component of such studies. In general, reaction times to a variety of visual or auditory stimuli are increased with progressive hypoxia, with the effect being greater the more demanding the cognitive task involved in the response. An intriguing approach evaluating acute hypoxia to arterial saturations equivalent to up to 15,000 feet while keeping end-tidal carbon dioxide tension constant demonstrated a hypoxic dose–related decrement in finger-tapping but with no significant effect on digit symbol and a number of other tests noted to be abnormal during or after sustained hypoxic exposure (10). Unfortunately, no normally hypocapnic group was studied to determine the extent to which a higher

380

Raichle and Hornbein

cerebral blood flow with isocapnia might have mitigated the magnitude of the changes. Even as low as 5000 feet (simulated), response to a visual positioning test administered during light work was slowed and more variable, although with learning these differences disappeared (11). Other described decrements are in short-term memory storage and recall. Fowler et al. (12), in a study designed to assess storage and retrieval from short-term memory, concluded that acute hypoxia did not directly impair either of these processes but rather slowed ‘‘the central executive or working memory, thereby reducing its capacity to process information.’’ They also posed the intriguing question whether this impairment of ‘‘working memory’’ is a direct effect of hypoxia on what they referred to as the central executive or a consequence of depression of an earlier, preprocessing stage in the reception of information. Relating this study with Kobrick’s observation that retinal receptors seemed unaffected by acute hypoxia suggests that the decrements in function might be occurring at higher levels of processing of information. Yet a ‘‘preprocessing’’ failure fits with the observations of Crow and Kelman (13), who noted that at 12,000 feet the error rate of recall of a short-term memory test was the same after a 32-second delay as after a 2-second delay, suggesting that ‘‘the defect may be mainly a failure of registration (failure to record a single digit, perhaps failure to perceive?) rather than a uniform loss of information in the interval before recall.’’ This possible explanation may apply as well to observations of effects of more sustained hypoxia on shortterm memory (see below) and conceivably also to residual deficits noted after the hypoxia has been relieved. While the generic response to hypoxia, either acute or chronic, is one of slowed or impaired performance, not all behavioral tests are altered, even with rather extreme hypoxia. Indeed some tests actually show improvement. For example, response times on simple reaction time paradigms are shortened. Usually such improvement is ascribed to learning effects, but a number of observations, extending over many decades, indicate better performance with acute hypoxia even when learning effects are unlikely or are adequately controlled for. In a recent study, an acute, mild hypoxic challenge (3450 m equivalent altitude) resulted in shortened time to recognize briefly presented letters (14), replicating earlier observations (15). As Kelman and Crow pointed out subsequently, this absence of deterioration was observed only for simpler tests requiring little learning or complex cognition (16). At this stage one can only speculate about where and how acute hypoxia may be acting to excite or inhibit pathways to account for at least a transitory improvement in visual perception. With improvements in aircraft design, including cabin pressurization, rapid ascent to extremely high altitudes became increasingly possible after World War II. This capability caused researchers to worry about the possibility of sudden cabin depressurization and its effects on pilot performance. How much useful time does an individual have before loss of consciousness prevents the institution of life-saving measures? Luft and colleagues explored this issue by studying rapid decompression to 55,000 feet from 30,000 feet in healthy individuals breathing 100% oxygen (17). From their work and a review of the work of others they concluded that there was

The High-Altitude Brain

381

a minimum latency or ‘‘free interval’’ prior to the loss of consciousness, regardless of the altitude achieved. During the time to unconsciousness (TUC), no discernible impairments in behavior (they had their subjects performing a writing task) could be detected, and the subjects reported no symptoms. In their entire series of experiments on decompression to 55,000 feet (17), unconsciousness occurred regularly after a latent period of 15–17 seconds provided that the exposure time exceeded 6 seconds. Reasoning from estimates of the systemic circulation time and the desaturation time of hemoglobin, they were led to conclude that the brain continues to function unimpaired for several seconds after the oxygen supply in the surrounding capillaries has reached its lowest level. Obviously, this is such a short period of time that detailed analysis of behavior would be difficult. Nevertheless, such timing information is important in eventually understanding how sudden increases in metabolic demands might be met by a vasculature that tends to respond in seconds rather than milliseconds (18–21). Another important aspect of this work on the TUC was the observation that humans born and living at altitude have significantly longer TUC than do their sea level counterparts (22). Chronic adaptations to altitude will become important when we consider the neurobiology of high-altitude behavioral symptoms later in this chapter. III. Effect of More Sustained Hypoxia on Behavior A. Experiences at Altitude: Awake

The history of Mt. Everest climbs is replete with anecdotal reports of cognitive impairments (23,24). Greene, the physician for the 1933 expedition, quoted Hingston’s observation on Everest in 1924 (23): ‘‘Though the mind is clear, yet there was a disinclination for effort. It was far more pleasant to sit about than to do a job of work that required thought. . . . Though mental work is a burden at high altitudes, yet with an effort it can be done. . . . The main effect of altitude is a mental laziness that determination can overcome.’’ Hallucinations, particularly the sensed presence of an extra companion, surface periodically in this high-altitude literature, though to be sure, altitude is not essential to such wanderings of the mind (23): (Frank Smythe) was convinced when climbing alone, and of course without oxygen equipment, at about 28,000 feet (8530 metres), that he had with him a friendly companion, to whom at one point he offered a sandwich. He also had visual hallucinations, seeing above the ridge of the mountains some curious dark bodies which came to be known as ‘‘Frank’s pulsating teapots.’’ Yet his judgment was otherwise unimpaired. At 28,100 feet (8565 metres), with the summit less than 1000 feet (305 metres) above him, he was able soberly to consider the pros and cons of continuing, weighing his feeling of physical fitness and the early hour against the bad condition of the snow, and coming to the unwelcome conclusion that he must return. Smythe never suffered from the irritability that occasionally appeared in all the rest of us who went high. It seemed that chronic anoxia increased his placidity. At low altitudes a little inclined to take offense, at great ones he emanated an infectious happiness.

382

Raichle and Hornbein

Smythe, acclimatized, was climbing at an altitude almost identical to that which rendered Glaisher unconscious some 70 years earlier. And Glaisher’s companion, Coxwell, was still functioning even as Glaisher was fading away. These contrasting observations indicate that adaptations to high altitude can take place given an appropriate period of acclimatization, and, additionally, that individual susceptibilit