Pulmonary Heart Disease

PULMONARY HEART DISEASE
edited by LEWIS J. RUBIN
MARTINUS NI]HOFF PUBLISHING

A MEMBER OF THE KLUWER ACADEMIC PUBLISHERS GROUP
Boston/The Hague /Dordrecht/Lancaster

Copyright 1984 @ by Martinus Nijhoff Publishing, Boston
Softcover reprint of the hardcover 1st Edition 1984
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,
or transmitted in any form or by any means, mechanical, photocopying, recording, or otherwise,
without written permission of the publisher, Martinus Nijhoff Publishing, 190 Old Derby Street,
Hingham, Massachusetts 02043
Distributors:
for North America for all other countries

Kluwer Academic Publishers Kluwer Academic Publishers Group
190 Old Derby Street Distribution Centre
Hingham, MA 02043 P.O. Box 322
3300 AH Dordrecht
The Netherlands
Library of Congress Cataloging in publication Data

Main entry under title:
Pulmonary heart disease.
Bibliography: p.
Includes index.
1. Cor pulmonale. I. Rubin, Lewis J. [DNLM:
1. Pulmonary heart disease. WG 420 P975]
RC685.C55P85 1984 616.1'2 84-1523
ISBN-13: 978-1-4612-9797-0 e-ISBN: 978-1-4613-2847-6
DOl: 10.1007/978-1-4613-2847-6
To my mother, Erna Rubin, and in memory of my father, Theodore Rubin.
CONTENTS
Preface xiii
1. Introduction 1
LEWIS J. RUBIN, M.D.
2. Pathophysiology of the pulmonary circulation 11

BRYDON J.B. GRANT, M.D.
DAVID R. DANTZKER, M.D.
3. Pathology of pulmonary heart disease 65

WAYNE E. TAYLOR, M.D.
4. Clinical evaluation 107

5. The electrocardiogram in chronic lung disease 117

FRANCIS X. MCGOWAN, M.D.
GALEN S. WAGNER, M.D.
6. Roentgenographic evaluation of pulmonary heart disease 135

CARL E. RAVIN, M.D.
7. Pulmonary function and exercise testing 177

TIMOTHY R. CHAPPELL, M.D.
8. Radionuclide angiocardiographic assessment of right and left ventricular

performance 223
vii
viii Contents
RICHARD A. MATTHAY, M.D.

HARVEY J. BERGER, M.D.
9. The left ventricle in chronic lung disease 247

EDGAR J. CALDWELL, M.D.
10. Arrhythmias in chronic lung disease 273

ERIC N. PRYSTOWSXY, M.D.
GEORGE KLEIN, M.D.
11. Acute cor pulmonale 285

WARREN R. SUMMER, M.D.
12. Therapy of pulmonary heart disease 325

ROBERT H. PETER, M.D.
13. Prognosis of pulmonary heart disease 355

RUSSELL DODGE, M.D.
BENJAMIN BURROWS,. M.D.
DOUGLASS MORRISON, M.D.
CONTRIBUTING AUTHORS
Harvey J. Berger, M.D.

Assistant Professor of Diagnostic Radiology and Medicine
Director of Cardiovascular Imaging
Yale University School of Medicine
New Haven, CT
Benjamin Burrows, M.D.

Professor of Medicine
Chief, Pulmonary Division
University of Arizona Health Sciences Center
Tucson, AZ
Edgar J. Caldwell, M.D.

Associate Professor of Medicine
University of Vermont School of Medicine
Pulmonary Division, Maine Medical Center
Portland, ME
Timothy R. Chappell, M.D.

Assistant Professor of Internal Medicine
University of Texas Health Science Center at Dallas
Assistant Chief, Pulmonary Section
ix
x Contributing Authors
Veterans Administration Medical Center

Dallas, TX
David R. Dantzker, M.D.

Chief, Pulmonary Division
University of Texas Health Science Center at Houston
Houston, TX
Russell Dodge, M.D.

Assistant Professor of Medicine
Pulmonary Division
Tucson, AZ
Brydon J.B. Grant, M.D.

Pulmonary Division
State University of New York at Buffalo School of Medicine
Buffalo, NY
George J. Klein, M.D.

Director, Clinical Electrophysiology Laboratory
Cardiac Investigation Unit
University Hospital
London, Ontario
Canada
Richard A. Matthay, M.D.

Associate Director, Pulmonary Division
Yale University School of Medicine
New Haven, CT
Francis X. McGowan, M.D.

Research Fellow in Cardiology
Department of Medicine
Duke University School of Medicine
Durham, NC
Douglass Morrison, M.D.

Assistant Professor of Medicine and Radiology
xi
Staff Cardiologist, Veterans Administration Medical Center

Tucson, AZ
Robert H. Peter, M.D.

Associate Director, Cardiovascular Laboratory
Durham, NC
Eric N. Prystowsky, M.D.

Director, Clinical Electrophysiology Laboratory
Krannert Institute of Cardiology
Indiana University School of Medicine
Indianapolis, IN
Carl E. Ravin, M.D.

Professor of Radiology
Director, Imaging Division
Durham, NC
Lewis J. Rubin, M.D.

Assistant Professor of Internal Medicine
Chief, Pulmonary Section, Veterans Administration Medical Center
Dallas, TX
Warren R. Summer, M.D.

Chief, Pulmonary-Critical Care Division
Louisiana State University Medical Center
New Orleans, LA
Wayne E. Taylor, M.D.

Assistant Professor of Pathology
Dallas, TX
Galen S. Wagner, M.D.

Cardiology Division
Durham, NC
PREFACE
Responsibility for the diagnosis and management of disorders of the pulmonary

circulation has become the shared domain of the pulmonologist, cardiologist,
surgeon, radiologist, pathologist, and, perhaps most important of all, the internist.
It is the general internist who is most likely to care for the majority of patients
with lung diseases that secondarily give rise to pulmonary heart disease, and it is
the internist who will first evaluate the patient with primary pulmonary hyperten-
sion or recurrent pulmonary thromboembolism who presents with nonspecific
complaints and may manifest subtle and nondiagnostic findings on preliminary
evaluation.
The burgeoning medical literature concerning aspects of the pulmonary circula-
tion, both clinical and investigative, is a reflection of the reawakening of great
interest in this field and has led to many new developments, both in our understand-
ing of cardiopulmonary pathophysiology and in the diagnosis and treatment of
pulmonary vascular diseases. This book is an attempt to provide the clinician with
a comprehensive overview of pulmonary heart disease from the perspective of
experts representing a variety of disciplines. It is intended to be thorough yet
clinically relevant. Individuals familiar with some facets of pulmonary heart disease
may gain insight into other aspects of this condition, whereas those unfamiliar with
this disorder may find this work useful as a general reference or as a resource to
address a specific question. Since there are few incontrovertible facts concerning
the approach to patients with pulmonary heart disease, emphasis has been placed
on general principles and guidelines rather than dogmatic recommendations.
xiii
xiv Preface
I am deeply indebted to my colleagues who have contributed their time, effort,

and knowledge to participate in this work. I have learned much from their writings
in the past, and I continue to do so. I am also deeply indebted to my secretary,
Becky Rendon, who despite all the other work and responsibility I seem to place
on her, quietly, efficiently, and reliably typed and organized draft after draft of
this work. Without her dedicated assistance this project could not have been
completed. Finally, I am grateful to Jeffrey Smith of Martinus Nijhoff Publishing
for his support and guidance through the stages of development and completion
of this work.
Lewis J. Rubin, M.D.

Dallas, Texas
1. INTRODUCTION
INTRODUCTION
The pulmonary circulation plays a pivotal role in the gas exchange function of the
lungs and, therefore, in the ultimate delivery of oxygen to the tissues. The central
location of the pulmonary circulation-interposed between the two sides of the
heart-coupled with its intimate relationship with the airspaces, renders this vascu-
lar network vulnerable not only to disorders that may affect it primarily, but to
conditions that alter the structure or function of the heart and lungs as well.
Disorders of the pulmonary circulation can be broadly classified based on the
segment of the circulation where the predominant pathophysiologic insult occurs
(table 1-1). Conditions that raise pulmonary venous pressure, such as left ventricu-
lar failure or mitral stenosis, may increase secondarily pulmonary arterial pressure.
The adult respiratory distress syndrome (ARDS) and other forms of noncardio-
genic pulmonary edema primarily attack the pulmonary capillary network, al-
though precapillary pulmonary hypertension may also occur in this setting. A
variety of diseases affect predominantly the pulmonary arterial tree, from the main
pulmonary arteries down to the small, muscular arterioles, ultimately producing
pulmonary arterial hypertension. This book will focus primarily on acute or
chronic respiratory conditions that affect the pulmonary circulation at a site proxi-
mal to the capillaries.
L.J. Ruhin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright @ 1984 Martinus NijhoffPuhlishing. Boston/The Hague/Dor-
drecht/Lancaster.
1
2 1. Introduction
Table 1-1. Classification of pulmonary

vascular diseases based on site of primary injury
Precapillary Postcapillary
Parenchymal lung diseases Left ventricular failure

Restrictive chest wall disease Mitral valve stenosis
Thromboembolic disease Left atrial myxoma or thrombus
Primary pulmonary hypertension Pulmonary veno-occlusive disease
Persistent fetal circulation
Congenital heart disease
Pulmonary vasculitis
High-altitude disease
Peripheral pulmonic stenosis
Pulmonary arteriovenous fistula
Background
Laennec (1) described the clinical and pathological features of emphysema in 1826
and noted that severe lung disease may produce heart failure:
All diseases which give rise to severe and long continued dyspnoea produce, almost
necessarily, hypertrophy or dilatation of the heart, through the constant efforts the organ
is called on to perform, in order to propel the blood into the lungs against the resistance
opposed to it by the cause of the dyspnoea.... When, however, diseases of the heart are
found to coexist with chronic pleurisy, phthisis, emphysema, and, in general, with chronic
disease of the lungs, it will usually be found, on close examination, that the latter are the
primary diseases. It follows from these, and other facts noticed under the head of emphy-
sema and pulmonary catarrh, that a neglected cold is frequently the original cause of the
most severe diseases of the heart. (1)
In 1901 Ayerza described a case of sclerosis of the pulmonary arteries in associa-

tion with profound cyanosis and chronic dyspnea (2). In 1931 Dr. Paul Dudley
White coined the term cor pulmonale to describe abnormalities of the heart resulting
from pulmonary parenchymal or vascular disease (3), thereby bringing this disorder
to wider clinical attention.
Definition
In 1960 an expert committee of the World Health Organization (WHO) convened
to address several issues concerning cor pulmonale. The deftnition of cor pulmonale
agreed upon by that committee has been widely accepted and is the most precise:
"Hypertrophy of the right ventricle resulting from diseases affecting the function
and / or structure of the lung, except when these pulmonary alterations are the result
of diseases that primarily affect the left side of the heart or of congenital heart
disease." (4)
The terms cor pulmonale and pulmonary heart disease can be used synonymously.
It is my preference to use pulmonary heart disease because cor pulmonale is fre-
quently equated with right-sided cardiac failure. It is worth emphasizing that right
heart failure is a late manifestation of pulmonary heart disease and need not be
3
present to entertain that diagnosis (5,6). The earliest and most consistent feature
of pulmonary heart disease is an elevation in pulmonary arterial pressure or
pulmonary vascular resistance, measured either at rest or during exercise. Increased
right ventricular afterload is what ultimately leads to right ventricular hypertrophy
and / or dilatation.
Classification
The classification of chronic pulmonary heart disease proposed by the WHO
committee is based on causative diseases and is shown, with minor modifications,
in table 1-2. These conditions include primary disorders of the pulmonary vascula-
ture as well as diseases which affect the lung parenchyma or the thoracic cage and
which interfere with efficient intrapulmonary gas exchange. While this classifica-
tion is useful because it emphasizes the different pathophysiologic mechanisms
responsible for the development of pulmonary heart disease, one should keep in
mind that considerable overlap in mechanisms may exist in some conditions. For
example, while sarcoidosis generally produces cor pulmonale as a result of pulmo-
nary interstitial disease, pulmonary vascular granulomata may cause or contribute
to pulmonary heart disease in this condition (7,8). Primary vascular involvement
with polymyositis, without evidence of parenchymal lung disease, has been re-
ported (9). Similarly, intravenous drug abusers may have evidence of interstitial
lung disease as well as primary vascular abnormalities (10,11). This point has great
relevance to the therapy of pulmonary heart disease, which will be discussed in
chapter 12.
INCIDENCE
While a variety of conditions may be associated with pulmonary heart disease,
patients with chronic obstructive lung diseases comprise the largest group in whom
cor pulmonale occurs, accounting for over 80% of cases in the United States and
other industrial countries (108-110).
An accurate estimate of the incidence of pulmonary heart disease is difficult to
obtain from the published literature. Since the populations studied are frequently
not randomly selected, there is a predilection to selecting out and reporting patients
who are at the greatest risk for developing cor pulmonale. For example, an autopsy
study from the Cleveland Cystic Fibrosis Center found that 15% of cystic fibrosis
patients dying at 25 years of age or older had overt right heart failure (23); while
these statistics convey the bleak prognosis of pulmonary heart disease, they provide
only limited information concerning the overall magnitude of the problem.
A major obstacle to determining the incidence of pulmonary heart disease has
been the lack of uniform agreement on the clinical criteria necessary for diagnosis.
Although the WHO definition of cor pulmonale may be the most accurate, its
clinical usefulness is limited. Right ventricular hypertrophy per se has no clinical
correlate, and overt clinical signs tend to appear late in the course of the disease.
The incidence varies from 5 to 40% of patients with chronic obstructive lung
disease in studies that defme cor pulmonale as the presence of clinical signs of right
Table 1-2. Classification of pulmonary heart disease according to causative diseases
"'"
1. Diseases affecting air passages of the lung and alveoli 2. Diseases affecting thoracic cage movement
a. Chronic obstructive pulmonary diseases (ref. 5, 12-19) a. Kyphoscoliosis (ref. 57, 58)
b. Cystic fibrosis (ref. 20-23) b. Thoracoplasty (ref. 52)
c. Congenital developmental defects (ref. 24) c. Pleural fibrosis (ref. 4)
d. Infiltrative or granulomatous diseases d. Neuromuscular weakness (ref. 59, 60)
(1) Idiopathic pulmonary fibrosis (ref. 25-28) e. Sleep apnea syndromes (ref. 61-64)
(2) Sarcoidosis (ref. 7,8, 29-31) f. Idiopathic hypoventilation (ref. 59, 65-67)
(3) Pneumoconiosis (ref. 12, 32-34)
(4) Scleroderma (ref. 35, 36)
(5) Mixed connective tissue disease (ref. 37-39)
(6) Systemic lupus erythematosus (ref. 40, 41)
(7) Rheumatoid arthritis (ref. 42-45)
(8) Polymyositis (ref. 9)
(9) Eosinophilic granuloma (ref. 46, 47)
(10) Malignant infiltration (ref. 48)
(11) Radiation (ref. 49)
e. Upper airways obstruction (ref. 50)
f. Pulmonary resection (ref. 51-53)
g. High-altitude disease (ref. 54-56)
3. Diseases affecting the pulmonary vasculature 4. Pressure on pulmonary arteries by mediastinal tumors, aneurysms,
a. Primary diseases of the arterial wall granulomata, or fibrosis (ref. 105-107)
(1) Primary pulmonary hypertension (ref. 68)
(2) Granulomatous pulmonary arteritis (ref. 29, 43, 44, 69, 70)
(3) Toxin-induced pulmonary hypertension
a) Aminorex fumarate (ref. 71-73)
b) Intravenous drug abuse (ref. 10, 11, 74-77)
(4) Chronic liver disease (ref. 78-85)
(5) Peripheral pulmonic stenosis (ref. 86)
b. Thrombotic disorders
(1) Sickle cell disease (ref. 87, 88)
(2) Pulmonary microthrombi (ref. 89)
c. Embolic disorders
(1) Thromboembolism (ref. 90-96)
(2) Tumor embolism (ref. 97-101)
(3) Other embolism (amniotic fluid, air) (ref. 102)
(4) Schistosomiasis and other parasites (ref. 103, 104)
5
heart failure (4,18,109,111). Additionally, the presence of edema is not specific for
cor pulmonale, and this sign is likely to be an insensitive marker when used as the
sole criterion for the presence of pulmonary heart disease.
The accurate diagnosis of pulmonary heart disease requires the measurement of
pulmonary arterial pressures and pulmonary blood flow, but these invasive proce-
dures are not indicated in every patient in whom that diagnosis is suspected.
Furthermore, the hemodynamic criteria for the diagnosis are not uniform: Siassi
et al. found that 50% of 34 children with cystic fibrosis had hemodynamic evidence
of cor pulmonale, defmed in that study as a mean pulmonary artery pressure
exceeding 20 mmHg (20). The WHO committee defmed pulmonary hypertension
as a mean pulmonary artery pressure of 25 mmHg or greater (4). In our laboratory
pulmonary hyptertension is considered to be present if the mean pulmonary arterial
pressure is greater than 25 mmHg with a pulmonary capillary wedge pressure less
than 15 mmHg, or a pulmonary vascular resistance of 3 units (approximately 240
dynes·sec·cm- S) or greater, measured either at rest or during exercise (112). We
defme right ventricular failure as the presence of hepatomegaly, ascites, or edema
in association with a right ventricular end-diastolic pressure of 10 mmHg or greater
(113). While these criteria may be somewhat rigid, they serve to differentiate
between patients with cor pulmonale and patients with combined biventricular
failure or those with mild hemodynamic abnormalities which may not significantly
contribute to mortality or warrant specific therapy. Additionally, these criteria
recognize the profound hemodynamic abnormalities that may develop during
exercise in some patients.
An additional difficulty encountered in clarifying the true incidence of pulmo-
nary heart disease is the tendency for both clinical and autopsy studies to reflect
the prevalence of certain lung diseases in specific geographic areas. Studies from
the industrial Midlands of Great Britain showing that 25 to 40% of patients
admitted to hospitals because of congestive heart failur/e had cor pulmonale (108,-
109) may reflect the high incidence of lung disease mthat region, estimated to be
67% of males and 26% of females. Similarly, a range in incidence from 0.9% of
cardiac autopsies in Massachusetts to 54% in Arizona may reflect the incidence of
lung disease in the Southwest, a popular region for people suffering from chronic
respiratory diseases (4). A reasonable estimate may be that pulmonary heart disease
accounts for approximately 10% of heart disease in the United States (114).
Although the exact incidence of pulmonary heart disease may not be known, this
entity clearly seems to be a clinically important complication of lung disease which
influences both morbidity and mortality. These important issues will be discussed
in greater detail in subsequent chapters.
CONCLUSION
By elaborating on some of the disparate data concerning the incidence of pulmo-
nary heart disease, I have deliberately set the stage for discussions in later chapters
of this book. While the goal of this work is to provide the reader with a
6 1. Introduction
comprehensive overview of various aspects of pulmonary heart disease, the unfor-

tunate fact remains that we still know very little about this disorder. Many issues
are unsettled and controversial, including the pathogenesis and therapy of cor
pulmonale. Weare, nevertheless, slowly but consistently enhancing our under-
standing of the pulmonary circulation. Advances are being achieved in the clarifica-
tion of pathogenesis and in the development of modalities for earlier diagnosis and
more effective management of pulmonary heart disease. The ultimate implementa-
tion of these developments in clinical practice will eventually render obsolete the
discussion of the incidence of cor pulmonale.
REFERENCES
1. Laennec R TH. A Treatise on the Diseases of the Chest and on Mediate Auscultation. Translated
by J. Forbes, London, 1834.
2. Garrison FH. Contributions to the History of Medicine. New York:Hafner Publications, 1966.
3. White PD. Heart Diseases. New York: MacMillan, 1931.
4. Chronic cor pulmonale: Report of an expert committee. Circulation 27: 594-615, 1963.
5. Bishop JM. Cardiovascular complications of chronic bronchitis and emphysema. Med Clin
North Am 57: 771-780, 1973.
6. Neal R W, Nair KG, Hecht HH. A pathophysiological classification of cor pulmonale: With
general remarks on therapy. Mod Concepts Cardiovasc Dis 37: 107-112, 1968.
7. Cornet A, Barbier JP, Debesse B, Even P, Carnot F, Neveu D. Sarcoidose avec hypertension
portale et hypertension arterielle pulmonaire. Ann Med Interne 131: 431-434, 1980.
8. Battesti JP, Georges R, Basset F, Saumon G. Chronic cor pulmonale in pulmonary sarcoidosis.
Thorax 33: 76-84, 1978.
9. Bunch TW, Tancredi RG, Lie]T. Pulmonary hypertension in polymyositis. Chest 79: 105-107,
1981.
10. Farber HW, Falls R, Glauser FL. Transient pulmonary hypertension from the intravenous
injection of crushed, suspended pentazocine tablets. Chest 80: 178-182, 1981.
11. Robertson CH, Reynolds RC, WilsonJE. Pulmonary hypertension and foreign body granulo-
mas in intravenous drug abusers: Documentation by cardiac catheterization and lung biopsy. Am
J Med 61: 657-664, 1976.
12. Harvey RM, Ferrer MI, Richards DW, Cournand A. Influence of chronic pulmonary disease
on the heart and circulation. Am J Med 10: 719-738, 1951.
13. Borden CW, Wilson RH, Ebert R V, Wells HS. Pulmonary hypertension in chronic pulmonary
emphysema. Am J Med 8: 701-709, 1950.
14. Heath D, Brewer D, Hicken P. Cor pulmonale in emphysema. In Mechanisms and Pathology.
Springfield, 11: Charles C Thomas, 1968.
15. Ferrer MI. Cor pulmonale (pulmonary heart disease): Present day status. Am Heart J 89:
657-664, 1975.
16. Bishop JM. Hypoxia and pulmonary hypertension in chronic bronchitis. Prog Resp Res 9:
10-16, 1975.
17. Segel N, Bishop JM. The circulation in patients with chronic bronchitis and emphysema at rest
and during exercise, with special reference to the influence of changes in blood viscosity and
blood volume on the pulmonary circulation. J Clin Invest 45: 1555-1568, 1966.
18. Fishman AP. Chronic cor pulmonale. Am Rev Respir Dis 114: 775-794, 1976.
19. Thomas AJ. Chronic pulmonary heart disease. Br Heart J 34: 653-657, 1972.
20. Siassi B, Moss AJ, Dooley RR. Clinical recognition of cor pulmonale in cystic fibrosis. J Pediatr
78: 794-805, 1971.
21. Symchych PS. Pulmonary hypertension in cystic fibrosis. Arch Path 92: 409-414, 1971.
22. Moss AJ, Harper WH, Dooley RR, Murray JF, MackJF. Cor pulmonale in cystic fibrosis of
the pancreas. J Pediatr 67: 797-807, 1965.
23. "GAP" Conference Report: Cor Pulmonale. Cystic Fibrosis Foundation, Rockville, Md, 1980.
24. Llamas R, Schwartz A, Gupta SK, Baum GL. Unilateral hyperlucent lung with polycythemia
and cor pulmonale. Chest 59: 690-692, 1971.
7
25. Crystal RG, Fulmer ]D, Roberts WC, Moss ML, Line BR, Reynolds HY. Idiopathic pulmo-
nary fibrosis. Ann Intern Med 85: 769-788, 1976.
26. Turner-Warwick M, Burrows B,]ohnson A. Cryptogenic fibrosing alveolitis: Clinical features
and their influence on survival. Thorax 35: 171-180, 1980.
27. Steinberg 1. Cor pulmonale due to pulmonary fibrosis simulating mediastinal tumor.]AMA 199:
158-160, 1967.
28. Heath D, Gillund TO, Kay]M, Hawkins CF. Pulmonary vascular disease in honeycomb lung.
] Path Bact 94: 42~30, 1968.
29. Smith L], Lawrence ]B, Katzenstein ALA. Vascular sarcoidosis: A rare cause of pulmonary
hypertension. Am] Med Sci 285: 38-44, 1983.
30. Levine BW, Saldana M, Hutter AM. Pulmonary hypertension in sarcoidosis. Am Rev Respir
Dis 103: 41~17, 1971.
31. Moffat RE, Sobonya RE, Chang CH]. Childhood sarcoidosis with fatal cor pulmonale. Pediatr
Radiol 7: 180-182, 1978.
32. Lapp NL, Seaton A, Kaplan KC, Hunsaker MR, Morgan WKC. Pulmonary hemodynamics in
symptomatic coal miners. Am Rev Respir Dis 104: 418-426, 1971.
33. Morgan WKC, Seaton A. Occupational Lung Disease. Philadelphia: W.B. Saunders, 1975.
34. Wells AL. Cor pulmonale in coal workers' pneumoconiosis. Br Heart] 16: 74-78, 1954.
35. Young RH, Mark G]. Pulmonary vascular changes in scleroderma. Am] Med 64: 998-1004,
1978.
36. Salemi R, Rodman GP, Leon DF, Shaver ]A. Pulmonary hypertension in the CREST syndrome
variant of progressive systemic sclerosis (scleroderma). Ann Intern Med 86: 394-399, 1977.
37. Kobayashi H, Sano T, Ii K, Hizawa K, Yarnanoi A, Otsuka T. Mixed connective tissue disease
with fatal pulmonary hypertension. Acta Pathol ]pn 32: 1121-1129, 1982.
38. Eulderink F, Cats A. Fatal primary pulmonary hypertension in mixed connective tissue disease.
Z Rheumatol4O: 25-29, 1981.
39. Jones MB, Osterholm RK, Wilson RB, Martin FR, Commers JR, Bachmayer ]D. Fatal
pulmonary hypertension and resolving immune-complex glomerulonephritis in mixed connec-
tive tissue disease. Am] Med 65: 855-863, 1978.
40. Perez RD, Kramer N. Pulmonary hypertension in systemic lupus erythematosus: Report of four
cases and review of the literature. Semin Arth Rheum 11: 177-181, 1981.
41. Turner-Stokes L, Turner-Warwick M. Intrathoracic manifestations of SLE. Clin Rheum Dis
8: 229-242, 1982.
42. Kay]M, Banik S. Unexplained pulmonary hypertension with pulmonary arteritis in rheumatoid
disease. Br] Dis Chest 71: 53-59, 1977.
43. Gardner DL, Duthie ]JR, MacLeod]. Pulmonary hypertension in rheumatoid arthritis: Report
of a case with intimal sclerosis of the pulmonary and digital arteries. Scott Med] 2: 183-188,
1957.
44. ]ordan]D, Snyder CH. Rheumatoid disease of the lung and cor pulmonale. Am] Dis Child
108: 174-180, 1964.
45. Onodera S, Hill ]R. Pulmonary hypertension. Report of a case in association with rheumatoid
arthritis. Ohio State Med] 61: 141-144, 1965.
46. Case Records of the Massachusetts General Hospital. N Engl] Med 298: 327-332, 1978.
47. Basset F, Corrin B, Spencer H, Lacronique], Roth C, Soler P, BattestiJP, Georges R, Chretien
]. Pulmonary histiocytosis X. Am Rev Respir Dis 118: 811-820, 1978.
48. Adelemus LR, Lee RE. Carcinomatosis of the lung with pulmonary hypertension. Pathoradi-
ologic spectrum. Arch Intern Med 119: 32-38, 1967.
49. Sweaney SK, Moss WT, Haddy F]. The effects of chest irradiation on pulmonary function.]
Clin Invest 38: 587-593, 1959.
50. Levy AM, Tabakin BS, Hanson ]S, Narkewicz RM. Hypertrophied adenoids causing pulmonary
hypertension and severe congestive heart failure. N Engl] Med 277: 506-511, 1967.
51. Jezek V, Ourednik A, Lichtenberg], Mostecky H. Cardiopulmonary function in lung resection
performed for bronchogenic cancer in patients above 65 years of age. Respiration 27: 42-50,
1970.
52. DiMaria G. The effect of pneumonectomy on the central hemodynamics. Prog Resp Res 5:
253-255, 1970.
53. Jezek V. Pulmonary hemodynamics in bronchogenic cancer before and after lung resection. Prog
Resp Res 5: 237-243, 1970.
8 1. Introduction
54. Penaloza D, Sime F. Chronic cor pulmonale due to loss of altitude acclimatization (chronic
mountain sickness). Am J Med 50: 723-743, 1971.
55. Sime F, Banchero N, Penaloza D, Gamboa R, Cruz J, Marticorena E. Pulmonary hypertension
in children born and living at high altitudes. Am J Cardioll1: 143-149, 1963.
56. 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 33: 249-262, 1966.
57. Bergofsky EH, Turino GM, Fishman AP. Cardio-respiratory failure in kyphoscoliosis. Medicine
38: 263-317, 1959.
58. Wanderman KL, Goldstein MS, Faber J. Cor pulmonale secondary to severe kyphoscoliosis in
Marfan's syndrome. Chest 67: 250-251, 1975.
59. Bergofsky EH. Cor pulmonale in the syndrome of alveolar hypoventilation. Prog Cardiovasc
Dis 9: 414-437, 1967.
60. Cherniack RM, Ewart WB, Hildes JA. Polycythemia secondary to respiratory disturbances in
poliomyelitis. Ann Intern Med 46: 720-727, 1957.
61. Coccagna G, Mantovani M, Brignami F, Parchi C, Lugaresi E. Continuous recording of the
pulmonary and systemic arterial pressure during sleep in syndromes of hypersomnia with
periodic breathing. Bull Eur Physiopathol Respir 8: 1159-1172, 1972.
62. Tilkian AG, Guilleminault C, Schroeder JS, Lehrman KL, Simmons FB, Dement We. Hemody-
namics in sleep-induced apnea. Ann Intern Med 85: 714-719, 1976.
63. Guilleminault C, Tilkian A, Dement We. The sleep apnea syndromes. Annu Rev Med 27:
465-484, 1976.
64. Guilleminault C, Eldridge FL, Simmons FB. Sleep apnea syndrome: Can it induce hemodynamic
changes? West J Med 123: 7-16, 1975.
65. Mellins RB, Balfour HH, Turino GM, Winters R W. Failure of automatic control of ventilation
(Ondine's Curse). Medicine 49: 487-504, 1970.
66. Lawrence LT. Idiopathic hypoventilation, polycythemia and cor pulmonale. Am Rev Respir
Dis 80: 575-581, 1959.
67. Naeye RL. Alveolar hypoventilation and cor pulmonale secondary to damage to the respiratory
center. Am J Cardiol 8: 416-419, 1961.
68. Voelkel NF, Reeves)T. Primary pulmonary hypertension. In Pulmonary Vascular Diseases,
Moser KM (ed). New York: Marcel Dekker, 1979, pp. 573-628.
69. Lupi HE, Sanchez TG, Horwitz S, Gutierrez FE. Pulmonary artery involvement in Takayasu's
arteritis. Chest 67: 69-74, 1967.
70. Celoria GC, Friedel1 GH, Sommers Se. Raynaud's disease and primary pulmonary hypertension.
Circulation 22: 1055-1059, 1960.
71. Smekal P, Standfuss K, Rau G. Incidence of primary pulmonary hypertension after intake
arninorexfumarat (Menocil). Prog Resp Res 5: 22-29, 1969.
72. Kay JM, Smith P, Heath D. Aminorex and the pulmonary circulation. Thorax 26: 262-270,
1971.
73. Heath D, Smith P. Pulmonary vascular disease. Med Clin N Amer 61: 1279-1307,
1977.
74. Houck RJ, Bailey GL, Daroca PJ, Brazda F,Johnson FB, Klein RC. Pentazocine abuse-Report
of a case with pulmonary arterial cel1ulose granulomas and pulmonary hypertension. Chest 77:
227-230, 1980.
75. Groth DH, MacKay GR, Crable ]V, Cochran TH. Intravenous injection of talc in a narcotics
addict. Arch Path 94: 171-178, 1972.
76. Bainborough AR, Jericho KWF. Cor pulmonale secondary to talc granulomata in the lungs
of a drug addict. Can Med Assoc J 103: 1297-1298, 1970.
77. Wedndt VE, Puro HE, Shapiro J. Angiothrombotic pulmonary hypertension in addicts: "Blue
velvet" addition. JAMA 188: 755-757, 1964.
78. McDonnel1 PJ, Toye PA, Hutchins GM: Primary pulmonary hypertension and cirrhosis: Are
they related? Am Rev Respir Dis 127: 437-441, 1983.
79. Lebrec D, Capron)p, Dhumeaux D, Benhamou)p. Pulmonary hypertension complicating
portal hypertension. Am Rev Respir Dis 120: 849-856, 1979.
80. Adam A, Patterson DLH. Pulmonary hypertension associated with hepatic cirrhosis and primary
acrocyanosis. J R Soc Med 74: 689-691, 1981.
9
81. Kibria G, Smith P, Heath D, Sagar S. Observations on the rare association between portal and
pulmonary hypertension. Thorax 35: 945-949, 1980.
82. Clinicopathologic Conference. Chronic active hepatitis and pulmonary hypertension. Am J Med
63: 6~13, 1977.
83. Rutmer JR, Bartschi JP, Niedermann R, Schnieder J. Plexogenic pulmonary arteriopathy and
liver cirrhosis. Thorax 35: 133-136, 1980.
84. Molden D, Abraham JL. Pulmonary hypertension: Its association with hepatic cirrhosis and iron
accumulation. Arch Pathol Lab Med 106: 328-331, 1982.
85. Senior RM, Britton RC, Turino GM, WoodJA, Langer GA, Fishman AP. Pulmonary hyper-
tension associated with cirrhosis of the liver and with portacaval shunts. Circulation 37: 88-96,
1968.
86. Eldredge WJ, TinglestaadJB, Robertson LW, Mauck HP, McCue CM. Observations on the
natural history of pulmonary arterial coarctations. Circulation 45: 404-409, 1972.
87. Collins FS, Orringer EP. Pulmonary hypertension and cor pulmonale in the sickle hemo-
globinopathies. Am J Med 73: 814-821, 1982.
88. Rowley PT, Enlander D. Hemoglobin S-C disease presenting as acute cor pulmonale. Am Rev
Respir Dis 98: 494-500, 1968.
89. Levin DL, Weinberg AG, Perkin RM. Pulmonary microthrombi syndrome in newborn infants
with unresponsive persistent pulmonary hypertension. J Pediatr 102: 299-303, 1983.
90. Paraskos JA, Adelstein SJ, Smith RE, Rickman FD, Grossman W, Dexter L, Dalen JE. Late
prognosis of acute pulmonary embolism. N Engl J Med 289: 55-58, 1973.
91. Wilhelmsen L, Selander S, Soderholm B, Paulin S, Varnauskas E, Werko L. Recurrent pulmo-
nary embolism. Medicine 42: 355-355, 1963.
92. Riedel M, Stanek V, Widimsky J, Prerovsky 1. Long-term follow-up of patients with pulmo-
nary thromboembolism. Chest 81: 151-158, 1982.
93. Fleischner FG. Recurrent pulmonary embolism and cor pulmonale. N Engl J Med 276: 1213-
1220, 1967.
94. Owen WR, Thomas W A, Castleman B. Unrecognized emboli to the lungs with subsequent
cor pulmonale. N Engl J Med 249: 919-926, 1953.
95. Dalen JE, Alpert JS. Natural history of pulmonary embolism. Prog Cardiovasc Dis 17: 259-270,
1975.
96. Falicox RE, Resnekov L, Petasnick J. Progressive pulmonary vascular obstruction and cor
pulmonale due to repeated embolism from axillary vein thrombosis. Ann Intern Med 73:
429-431, 1970.
97. Heath D, MacKinnon J. Pulmonary hypertension due to myxoma of the right atrium, with
special reference to the behavior of emboli of myxoma in the lung. Am Heart J 68: 227-235,
1964.
98. Marini JJ, Bilnoski W, Huseby JS. Acute cor pulmonale resulting from tumor microembolism.
West J Med 132: 77-80, 1980.
99. GrahamJP, Rotman HH, WegJG. Tumor emboli presenting as pulmonary hypertension. Chest
69: 229-230, 1976.
100. Vitale JCG, Bunuel Re. Pulmonary tumor emboli and cor pulmonale in primary carcinoma
of the lung. Cancer 38: 2105-2110, 1976.
101. Chakeres DW, Spiegel PK. Fatal pulmonary hypertension secondary to intravascular metastatic
tumor emboli. Amer J Roentgen 139: 997-1000, 1982.
102. Mendelson CL. Acute cor pulmonale and pregnancy. Clin Obstet Gynecolll: 992-1009,1968.
103. Obeyesekere I, Peiris D. Pulmonary hypertension and filariasis. Br Heart J 36: 67fr681, 1974.
104. Marchand E, Marcial-Rojas R, Rodriguez R, Polanco G, Diaz-Rivera RS. The pulmonary
obstruction syndrome in schistosoma mansoni pulmonary endarteritis. Report of flve cases. Arch
Intern Med 100: 965-980, 1957.
105. Cheris DN, Dadey JL. Fibrosing mediastinitis. An unusual cause for cor pulmonale. Amer J
Roentgen 100: 328-331, 1967.
106. Nelson WP, Lundberg GD, Dickerson RB. Pulmonary artery obstruction and cor pulmonale
due to chronic fibrous mediastinitis. Amer J Med 38: 279-284, 1965.
107. Arnett N, Bacos JM, Macher AM, Marsh HB, Savage DD, Fulmer JD, Roberts We. Fibrosing
mediastinitis causing arterial hypertension with pulmonary venous hypertension. Clinical and
necropsy observations. Amer J Med 63: 634-643, 1977.
10 1. Introduction
108. Fulton RM. The heart in chronic pulmonary disease. Q J Med 22: 43-58, 1953.
109. Flint FJ. Cor pulmonale. Lancet 2: 51-58, 1954.
110. Robin ED, Gaudio R. Cor Pulmonale. Disease-A-Month. Chicago: Year Book, 1970.
111. Behnke RH, Bristow JD, Carrieri V, Pierce JA, Sashara A, Soffer A. Resources for the optimal
care of acute respiratory failure. Circulation 43: A 185-A195, 1971.
112. Rubin IJ, Peter RH. Hemodynamics at rest and during exercise after oral hydralazine in patients
with cor pulmonale. Amer J Cardiol 47: 116-122, 1981.
113. Rubin IJ, Handel F, Peter RH. The effects of oral hydralazine on right ventricular end-diastolic
pressure in patients with right ventricular failure. Circulation 65: 1369-1373, 1982.
114. Ingram RH, Grossman GD. Chronic cor pulmonale. In The Heart, HurstJW (ed). New York:
McGraw-Hill, 1974, p. 1278.
2. PATHOPHYSIOLOGY OF THE PULMONARY CIRCULATION
BRYDON J.B. GRANT, M.D.

and
DAVID R. DANTZKER, M.D.
The pulmonary vasculature is the crossroad between the right and left ventricles.
Therefore, the circwnstances under which the pulmonary circulation operates are
the key to understanding the pathophysiology of pulmonary heart disease.
We shall begin by describing the functional characteristics of the normal pulmo-
nary circulation. We shall then proceed to delineate the pathological mechanisms
that lead to abnormalities in the pulmonary vasculature and their physiological
effects.
THE NORMAL PULMONARY CIRCULATION

The primary function of the pulmonary circulation is the exchange of oxygen and
carbon dioxide between blood and alveolar gas at rates that depend on the meta-
bolic demands of the systemic tissues. The performance of this task at minimal cost
in energy expenditure and with maximal efficiency is clearly advantageous.
To minimize the energy costs, the pulmonary circulation operates with input
pressures less than 20% of the systemic circulation. The low pressures render the
pulmonary vasculature susceptible to mechanical influences that modify the relation
between pressure and blood flow in the lung. These passive effects have practical
implications for the assessment of pulmonary vascular function.
To maximize gas exchange efficiency, the pulmonary vasculature has the ability
L.J. Ruhin (ed.), Pulmonary Heart Diseose. All rights rese"",d. Copyright @ 1984 Martinus NijhoffPuhlishing. Boston/The Hague/Dor-
drecht/Lancaster.
11
12 2. Pathophysiology of the Pulmonary Circulation
to control the distribution of blood flow within the lung by hypoxic pulmonary
vasoconstriction in order to improve the matching of perfusion to ventilation in
lung units. Active mechanisms that affect the pulmonary vasculature are of interest
because of the potential for pharmacological interventions in pulmonary heart
disease.
In this section we shall outline the functional aspects of the normal pulmonary
vasculature anatomy. Since the key variables for describing circulatory function are
pressure and flow, we shall introduce some concepts that are used to examine the
relation between pressure and flow in the pulmonary vasculature and then describe
the passive and active factors that affect this relation.
Functional consequences of pulmonary vascular anatomy

From an anatomic perspective, the pulmonary vasculature can be divided into two
parts: the pulmonary vascular tree and the pulmonary microcirculation. From a
physiological perspective, this division is not unreasonable. The nature of the
pulmonary vascular tree has implications for cardiac function, whereas the pulmo-
nary microcirculation is concerned with gas exchange.
Pulmonary vascular tree

LOCATION OF THE VASCULAR TREE WITHIN THE LUNG. The pulmonary arterial tree
lies in close proximity to the bronchial tree, as they both branch out from the hila
through the lung parenchyma to the alveolar-capillary interface. Both lie in the
same fascial sheath, together with lymphatic vessels. The pulmonary venous tree
branches in a similar manner, but it is separated from the pulmonary arterial and
bronchial trees (1). Both the pulmonary arterial and venous trees are surrounded
by a perivascular space (2), into which the pull of the parenchyma is transmitted.
The perivascular pressure of these intrapulmonary vessels is likely to be less than
pleural pressure to the extent that the vessel walls resist radial traction by the lung
parenchyma. Nevertheless, the general principle applies that there is a mechanical
interdependence between lung volumes and the volume of these vessels (3).
The close proximity of the pulmonary arterial tree to the bronchial tree has
functional implications that may be of importance in disease. Vasoactive substances
generated in the airway may diffuse across their walls and affect adjacent pulmonary
arteries. Some have suggested that the chemical mediators of asthma may act in this
manner (4). Similarly, substances generated in the pulmonary arteries may affect
airway smooth muscle. Besides diffusion, another pathway through which humoral
agents may pass between the bronchi and pulmonary vasculature is the anastomoses
between the bronchial and pulmonary circulation. These channels have been de-
scribed at the capillary level in both anatomic and physiologic studies (5,6). Since
bronchial blood flow is normally 1% of pulmonary blood flow (7), these anas-
tomoses are unlikely to have an important impact on pulmonary gas exchange, but
they may assist in maintaining the viability of lung tissue after pulmonary embo-
lism.
13
MORPHOMETRY. Despite the close relation between the bronchial and pulmonary
arterial trees, quantitative differences are found in their morphology as measured
from casts of postmortem material. The total cross-sectional area of the bronchial
tree ranges from 2 cm2 in the trachea to 280 cm2 in the terminal respiratory
bronchioles (8). The total cross-sectional area of the pulmonary arterial tree ranges
from 7 cm2 in the main pulmonary artery to an estimated 400 cm2 in the precapil-
lary arteries (9). These geometric differences have been used to calculate the
pressure drop along the bronchial and pulmonary arterial trees under Poiseuille
flow conditions (see later in this chapter). In contrast to the bronchial tree, most
of the pressure drop along the pulmonary arterial tree is believed to occur in the
small peripheral arteries (10). This fact suggests that the design of the pulmonary
vasculature is better suited to control blood flow distribution than the bronchial
tree is to control ventilation distribution, in order to match the ratio of ventilation
to perfusion within small lung units.
The greater total cross-sectional area of the precapillary arteries is due to exces-
sive branching of the pulmonary arterial tree compared with the bronchial tree
(10). This excessive branching gives rise to supernumerary vessels and is particularly
prominent at the lobular level where the number of branches that do not accom-
pany airways exceeds the number of branches that do by fourfold (11). This
arrangement results in a close proximity of the walls of these small vessels to
alveolar gas in neighboring alveoli. Therefore, a potential pathway exists whereby
alterations of alveolar gas tensions due to changes in local alveolar ventilation can
modify the local pulmonary blood flow by hypoxic pulmonary vasoconstriction
and control of the local ventilation-perfusion ratio (VA/Q).
FUNCTIONAL ANATOMY OF THE VASCULAR TREE. The idea of active control of the
pulmonary vessels is supported by the structure of their walls. It is sometimes stated
that the walls of pulmonary arteries are so thin (the walls of pulmonary veins are
even thinner [11]) relative to systemic arteries that active control of the pulmonary
vasculature is of little consequence. Indeed, bronchial arteries, which are part of
the systemic circulation, have a wall thickness four times greater than pulmonary
arteries of similar diameters (12). This difference probably reflects the greater
pressure in the systemic circulation rather than a lack of active control of the
pulmonary vasculature.
The structure of the media varies along the length of the pulmonary arterial tree.
In large arteries (greater than 2 mm external diameter) the media consists of
muscular-elastic tissue with seven elastic lamina and is about 2% of the internal
diameter (11). The elastic lamina between the innermost and outermost layers
become less distinct toward the periphery, and the media consists almost entirely
of smooth muscle in vessels with a diameter less than 2 mm. In vessels less than
0.2 mm, the thickness of the muscular arteries increases from 2 to 10%. Vessels
between 0.15 mm and 0.03 mm may be completely, partially, or nonmuscularized
(11). Arteries that accompany the respiratory bronchioles are invariably muscula-
rized. Therefore, the machinery exists for controlling the caliber of small pulmo-
nary arteries far out into the periphery of the lung.
.--. -,
,, "
,
",
\
I \
I ,
,
, I
• I
~==::2/=-_--===
Figure 2-1. Windkessel model of the pulmonary arterial tree and microcirculation. During
systole flow passes through the one-way valve (pulmonary valve). Some flow passes directly
through the narrow resistance (R), which represents the pulmonary microcirculation, and the
rest distends the elastic wall of the pulmonary arterial tree to a position shown by the dashed
line. During diastole inflow ceases and the valve closes. Blood leaves the elastic chamber as it
recoils and drives flow through the microcirculation. (Reproduced with permission from
Hemodynamics by W.R. Milnor, published by Williams & Wilkins, Baltimore.)
The adventitia of both pulmonary arteries and veins contains nerve fibers that
extend down to vessels of 0.03 mm in the human pulmonary tree (14). In most
mammalian species, both adrenergic and cholinergic endings have been located in
the outer parts of the media (15). Thus, histochemical evidence exists that the
pulmonary vasculature is innervated by the autonomic nervous system.
CONSEQUENCES FOR CARDIAC FUNCTION. The anatomy of the pulmonary vascular
tree indicates that it is not a rigid structure. This design is of particular importance
as far as cardiac function is concerned. Blood flows through the lung as the result
of rhythmic contraction of the heart. Flow would be impossible if the entire
pulmonary vasculature, including the pulmonary microcirculation, was composed
of rigid elements. During systole flow from the pulmonary circulation to the left
ventricle is interrupted due to closure of the mitral valve. During diastole, flow
into the left ventricle can proceed, but flow into the pulmonary artery ceases due
to closure of the pulmonary valve. Without a compliant element in the system to
store stroke volume during systole, pulmonary blood flow would cease no matter
how large the total cross-sectional area of the pulmonary vasculature.
The existence of a compliant element in the vasculature was recognized in the
eighteenth century (16) and came to be described as the Windkessel (17), (figure
2-1). A Windkessel is a compliant element in a fire hose that is designed to convert
the pulsatile flow of water into a steady stream. The Windkessel effect of the
pulmonary vasculature behaves similarly in the human lung: only 44% of the
amplitude of flow pulsations in the pulmonary artery is transmitted to the pulmo-
nary microcirculation (18), and 15% of the pulmonary arterial pressure pulsations
is transmitted through to the pulmonary veins (19). Therefore, the major portion
of the Windkessel appears to reside in the pulmonary arterial tree. Although the
15
pulmonary arterial tree does not anatomically resemble the Windkessel shown in
figure 2-1, it does appear to function in a somewhat similar manner (see the section
discussing consequences for cardiac function).
The primary function of the pulmonary vascular tree is to conduct blood to and
from the pulmonary microcirculation. Although evidence suggests that gas ex-
change may occur in the small pulmonary arteries (20-22), the pulmonary mi-
crocirculation is almost certainly the main location for this function.
The pulmonary microcirculation

LOCATION OF THE PULMONARY CAPILLARIES WITillN THE LUNG. The major portion
of the pulmonary capillaries are located in the alveolar septa. Therefore, the
perivascular pressure to which they are subjected has two components. In an axis
perpendicular to the free alveolar wall, capillaries are subjected to alveolar pressure
that tends to compress them. In the longitudinal axis capillaries are adjacent to the
interstitium and tend to be stretched when lung volume is increased, since pleural
pressure is transmitted through to the interstitium. The overall effect is that
increases in transpulmonary pressure (alveolar minus pleural pressure) caused by
raising alveolar pressure or by lowering pleural pressure tend to flatten these vessels
in the longitudinal axis of the alveolar septa. This effect does not occur in capillaries
that are not located in the alveolar septa.
Capillaries located at the comer of alveoli, where adjacent alveolar septa meet,
are less exposed to alveolar pressure because they are predominantly surrounded
by interstitial tissue (23). The perivascular pressure of these corner vessels may
include not only the interstitial pressure but also a component due to surface tension
of the alveoli. Alveoli are more likely to be hexagonal than spherical structures
(1). Therefore, from the law of Laplace, one can anticipate that surface forces will
be greater at the comers of alveoli where the radii of curvature are greater.
MORPHOMETRY. The small pulmonary arteries empty into the pulmonary capil-
lary bed, which provides a vast surface area for gas exchange. The alveolar surface
is approximately 80 m 2 when the lung is fully distended (1). The surface area of
human pulmonary capillaries occupies approximately 85% of this surface. This
value is midway between similar estimates of 91 % in the cat (24) and 78% in the
dog (25). These differences may well reflect methodological rather than species
differences. Because the major portion of the capillary bed is located in the alveolar
septa, these vessels are exposed to alveolar gas on both sides and a thin film of blood
is exposed to the alveolar gas. The distance between the alveolar gas and capillary
lumen is less than 0.001 mm (1); therefore, the distance for gas to traverse is
extremely short.
In the human lung each precapillary artery supplies a mesh of 1,000 capillary
segments with an approximate length of 0.012 mm and a mean diameter of 0.008
mm each. In the human a red cell must traverse several hundred microns before
reaching a postcapillary vessel (1). The distance between pre- and postcapillary
vessels in the dog and cat has been measured to be between 0.6 and 0.8 mm (26).
Capillary dimensions are considerably variable, as figure 2-2 demonstrates. Capil-
Figure 2-2. Tracing of part of the human capillary bed. (10 p.m = 0.01 mm.) (Reproduced
with permission from original publication in Z. Zellforsch 57: 64S--666, 1962.)
lary lengths vary between 0.0015 and 0.0325 nun, and diameters vary between
0.0005 and 0.0615 nun (1).
FUNCTIONAL ANATOMY OF THE PULMONARY MICROCIRCULATION. Figure 2-2 pro-
vides a surface view of the capillary bed. The manner in which blood traverses the
capillary bed can be viewed from two different perspectives: blood can be consid-
ered to flow through a network of interconnecting tubes (tube flow), in which each
element has variable dimensions and may have correspondingly variable mechanical
characteristics (27), particularly as blood has to negotiate numerous branch points.
This idea is particularly apparent from the upper portion of figure 2-2. Alterna-
tively, blood can be considered as flow through a sheet sandwiched between two
layers of capillary endothelium held together by posts of cellular and interstitial
tissue covered with endothelium, similar in structure to an underground parking
garage. This idea is particularly apparent in the lower portion of figure 2-2, where
the stippled area represents the film of blood and the white areas represent the posts
as viewed from the alveolar surface. Although we have illustrated the concepts of
tube flow and sheet flow in terms of structure of the pulmonary microcirculation,
the important differences between them are functional (28).
17
Tube flow is a more complex concept because it stresses inhomogeneity of the

microvasculature. At a constant lung volume, the cross-sectional area of the capil-
lary bed can be increased by one of two ways. Capillary segments that were
previously closed may be opened when a critical opening pressure is exceeded
(recruitment), or the caliber of open capillaries may be increased (distention). Sheet
flow is a simple concept because it stresses homogeneity of the pulmonary vascular
bed. At a constant lung volume, the cross-sectional area of the sheet can be increased
only by increasing the thickness of the sheet as measured in an axis perpendicular
to the alveolar septa. Therefore, sheet flow is a more tractable concept mathemati-
cally, but it considers only distention. Recruitment was not considered since it was
possible to explain experimental pressure-flow relations without needing to invoke
the concept of recruitment (29).
According to the principle of parsimony, sheet flow has a distinct advantage over
tube flow. The question of whether sheet flow is capable of describing all experi-
mental data related to pulmonary blood flow is still open. In our opinion, tube
flow and sheet flow should not be considered as two opposing theories but rather
as two different approaches to understanding the pulmonary microcirculation that
will eventually lead to the development of a unifying theory. As presently formu-
lated the lack of recruitment in the sheet flow concept is what concerns protagonists
of capillary network flow. Recruitment of capillaries has been observed histologi-
cally (25,30) and also under direct vision (31). On the other hand, these observa-
tions are limited to capillaries close to the pleural surface, which may not be
representative of the entire capillary bed.
Recruitment was thought to be due to progressive opening of small pulmo-
nary arteries, which have a wide range of critical opening pressures (32,33),
similar to the systemic circulation (34). Therefore, the degree of capillary fuling
would be dependent on pulmonary arterial pressures. More recently, this idea has
been challenged. Histological studies have revealed that the variation is as great
in capillary fuling within the area supplied by a single precapillary vessel (arteri-
ole) as it is between arteriolar domains (25). This fmding suggests that recruit-
ment occurs within the capillaries themselves rather than the arterioles. The pat-
tern of capillary filling has been examined theoretically (27). This analysis
indicates that recruitment can occur over a wide range of arterial pressures, even
though the critical opening pressure of a capillary segment is in the order of
only 0.02 em H 20. The key feature of this mathematical model is the variation
in the mechanical properties of the capillary segments. Frictional resistance due
to the cellular components of blood, particularly at branch points in the net-
work, is thought to be the basis for the critical opening pressure of capillar-
ies.
Distention of pulmonary capillaries was originally thought unlikely since the
caliber of capillaries in systemic tissues does not change, even when capillary
pressure is raised by 100 mmHg (35). Nevertheless, distention of pulmonary
capillaries has been observed histologically (24,25,30) with increasing pulmonary
arterial pressure up to 50 em H 20. This difference in the behavior of systemic and
pulmonary capillaries is probably related to the firm perivascular support of the
systemic capillaries compared to the relative lack of support of pulmonary capillar-

ies within the lung. Histological studies reveal that pulmonary capillaries within
the alveolar septa can increase in both the longitudinal (25) and perpendicular axes
(30).
In the systemic tissues recruitment is determined by the arterioles, and distention
of capillaries does not occur. In the lung recruitment is determined by the properties
of the capillary network, and distention does occur. We shall now explore some
of the consequences of this special nature of the pulmonary microcirculation on
gas exchange.
CONSEQUENCES FOR GAS EXCHANGE. The transit time of blood through the capil-
lary bed (0.3 to 12 seconds) varies inversely with pulmonary arterial pressure and
flow (31). These transit times were obtained by direct observation in canine
pulmonary capillaries rather than by indirect estimates from measures of pulmonary
capillary blood volume and flow. Even the most rapid transit time is estimated to
be sufficient to avoid compromising the gas exchange function of the lung by
incomplete equilibration of capillary blood with alveolar gas (31).
Earlier, we mentioned the variation of capillary ftlling that can occur both
within and between arteriolar domains (25). In addition, a variation of capillary
flow exists within a single lobule (the portion of lung subtended by a terminal
bronchiole) (36). Therefore, the question arises as to whether the variation of
capillary flow within the lobule impairs pulmonary gas exchange by causing
ventilation-perfusion mismatch. An uneven distribution of capillary flow will not
cause ventilation-perfusion inequality if gaseous diffusion within a lung unit is
sufficiently rapid to maintain homogeneity of alveolar gas tension in all parts of
the lung unit. Unfortunately, there is no clear answer to our question because the
size of this lung unit is not known. Beads of various diameters have been injected
into the pulmonary artery to determine if the level at which ventilation-perfusion
inequality occurs can be detected in the canine lung by the multiple inert gas
technique (37). This study suggests that the homogenous lung unit is subtended by
pulmonary arteries of 0.15 mm; these vessels were usually associated with respira-
tory bronchioles.
In addition to gas exchange, the pulmonary microcirculation has implications
for cardiac function since it is the major site of resistance to blood flow through
the lungs (38), as shown in ftgure 2-3. This subject will be discussed in the
following section.
Analysis of pressure-flow relations

Concept of pulmonary vascular resistance
The idea of characterizing the pulmonary vasculature in terms of resistance is
derived from the analogy to electrical circuitry. Resistance is voltage drop across
a DC circuit divided by current. In the lung pulmonary vascular resistance is the
mean pressure drop across the pulmonary vasculature (mean pulmonary artery
pressure minus mean left atrial pressure) divided by cardiac output. Under condi-
19
Arteries Capillaries Veins

II
•
II
0
...•
U
C
CIII
...
20
•... -
ta
u"
40
.~
CIII 0
>1/l 60
... .....
;..
CIII
c 80
0
j
::::I
Q. 100
I
Pulmonary 30-50 20 10 1020 30-50 Pulmonary
arterial venous
Diameter (um)
Figure 2-3. Profile of the distribution of vascular resistance along the pulmonary circulation in
the isolated perfused dog lung with pulmonary artery pressure of 16.7 em H 20, venous pressure
of 11.1 em H 20, and alveolar pressure of 5 em H 20. 46% of the total resistance resides in the
capillaries. (Reproduced with permission from reference 38, copyright 1980 by American
Association for the Advancement of Science.)
tions of Poiseuille flow, pulmonary vascular resistance can be calculated from the
following relation:
R = (8 u l)/(1T'r 4),
where u is viscosity of blood, 1 is the effective length, and r is the effective radius
of the pulmonary vasculature. Because resistance is inversely proportional to the
fourth power of the radius, it is particularly sensitive to changes of vessel caliber.
Apart from the fact that flow through the pulmonary circulation is pulsatile,
characterization of the circuit in terms of pulmonary vascular resistance has several
limitations.
First, Poiseuille flow conditions exist only in long narrow tubes. Although
turbulent flow is unlikely at rest because oflow Reynold's numbers in the pulmo-
nary vasculature, true laminar Poiseuille flow is unlikely to occur because of the
branching nature of the vascular tree (39). Resistance, therefore, is not necessarily
constant. The pressure drop is not linearly related to flow, but perhaps to some
power of flow rate.
Second, blood is not a Newtonian fluid because it contains cellular elements.
Therefore, viscosity is not a constant but will vary with flow and depends on the
geometry of the pulmonary vasculature (39). The apparent viscosity of blood
relative to plasma increases with hematocrit (40). Pulmonary vascular resistance of

the isolated lung can increase by 85% when hematocrit is increased from zero to
80%. Most of this increase occurs when the hematocrit exceeds 40% (41). The
influence of blood viscosity on pulmonary vascular resistance has important thera-
peutic implications; it is discussed in more detail in chapter 12.
Third, geometric representation of the pulmonary vasculature in terms of effec-
tive diameter and length ignores the fact that the pulmonary vessels are not rigid
tubes, but rather are subject to variation. This factor is probably the m:yor reason
for the nonlinearity in pressure-flow relations in the lung. Prior to discussing the
factors that affect this relation by altering the vascular geometry, we need to
consider the raison d'etre for variable vascular dimensions, that is, the pulsatility
of pulmonary blood flow-a property that pulmonary vascular resistance ignores.
Concept of pulmonary vascular impedance

The pressure in the main pulmonary artery (input pressure) is required not only
to overcome the resistive element, but also to distend the compliant component
of the vasculature (the Windkessel) and to overcome the inertia of blood in
order to produce the acceleration of blood that occurs during pulsatile flow.
Input impedance takes into account all these factors. Whereas pulmonary vascu-
lar resistance is defined as the ratio of the mean pressure drop across the pulmo-
nary vasculature to mean flow, input impedance is defmed as the ratio of the
pressure oscillations to flow oscillations at the input to any region of the circula-
tion (39). Input impedance has two components: the modulus, which is the ratio
of amplitudes of the pressure to flow oscillations, and the phase angle, which is
the extent to which the pressure cycle leads the flow cycle. At zero frequency no
pulsation occurs and input impedance becomes analogous to pulmonary vascular
resistance, with the exception that left atrial pressure is neglected. Input imped-
ance, therefore, can be considered an extension of the concept of pulmonary
vascular resistance.
Although the pulmonary pressure and flow waves are not sinusoidal in form,
they are periodic and can be represented by the sum of sinusoidal waves of the fIrst
fIve to ten harmonics of the fundamental frequency (heart rate). By measuring
simultaneously the instantaneous pressure and flow in the pulmonary artery, a plot
of input impedance against frequency can be obtained by the use of Fourier analysis.
A typical plot of pulmonary input impedance in a human subject without known
pulmonary vascular disease is shown in fIgure 2-4 (42). To understand the changes
of input impedance with frequency, introducing the concept of characteristic
impedance is useful.
Characteristic impedance is the same as input impedance when measured in
circumstances in which no waves are reflected back to the input (39). An infinitely
long elastic tube with uniform wall properties behaves in this manner. This
circumstance is similar to the movement of a long rope lying on the ground. A
vertical jerking movement at one end will cause a wave to travel toward the other
end, but this wave will diminish in amplitude as it travels along the rope. As a
21
Pulmonary arterial input impedance
100
,;;- 80
'5
..
'"
~ 60
'""
:;
~ 40
20
(Pressure leads)
-I (Flaw leads)
o 4 8
Frequency Hz
Figure 2-4. Typical pulmonary input impedance in man shown in the solid line. The dashed
line indicates the theoretical input impedance in a tube of similar characteristics as the
pulmonary vasculature but without wave reflections, so that input impedance at higher
frequencies is the characteristic impedance. Note that pulmonary input impedance at high
frequencies (solid line) fluctuates around the value of characteristic impedance. (Reproduced
with permission from Cardiovascular Fluid Dynamics, copyright American Press, Inc. [London]
Ltd., 1972.)
result, the wave will die out before it reaches the far end of the rope. In a long
elastic tube the pressure and flow waves will be in phase (no phase shift), but input
impedance changes with frequency. At zero frequency its value is the terminal
impedance of the system (pulmonary vascular resistance). As frequency increases
the input impedance tends toward the characteristic impedance of the system (39).
Transmission line theory indicates that the characteristic impedance is independent
of frequency and depends on the ratio of inertial to compliant properties of the
system. Characteristic impedance can be approximated by the following equation
(43):
Zc = V {(p/'1Tr 2)/(dS/dP)}.
where p is the density of blood, r is the radius of the tube, P is pressure, and S
is cross-sectional area; thus, p/7Tr 2 expresses the inertance per unit length of the
tube, and dS/ dP expresses the compliance of the tube per unit length. This equation
contains no resistance term because at very high frequencies the changes of pressure
and flow are so rapid that the inertial and compliant properties prevent flow
through the system. As any hi-fi enthusiast will attest, wave reflections will occur
when impedances are not matched. The extent of wave reflection can be expressed
as the reflection coefficient (r):
r = 1 - Zc/Z
1 + Zc/Z'
where Zc is the characteristic impedance and Z is the input impedance (43). At
very low frequencies the input impedance is close to the terminal impedance. In
the pulmonary circulation, characteristic impedance is less than the terminal imped-
ance; therefore, wave reflections can be anticipated at low frequencies. The pulmo-
nary vasculature is quite different from a uniform tube of infinite length. The
pulmonary arterial tree has a fmite length with numerous bifurcations and termi-
nates in the pulmonary microcirculation, where the major portion of pulmonary
vascular resistance resides (38). Therefore, opportunity is ample for wave reflections
to occur.
Although the forward pressure and flow waves are in phase at the input, the
measured pressure and flow waves become out of phase as a result of wave
reflections. Phase shifts occur because the measured pressure is the sum of the
forward and reflected waves. The extent of these phase shifts will depend on the
frequency, the average distance between the input and the sites of wave reflection,
and the speed of wave propagation, which, in turn, depends on the mechanical
properties of the vessel walls and the viscosity of the blood. As a result the modulus
and phase angle of input impedance fluctuates around characteristic impedance to
an extent that depends on the reflection coefficient and frequency (figure 2-4). But
as frequency increases, the input impedance tends toward the characteristic imped-
ance, and the reflection coefficient decreases toward zero at very high frequencies.
Therefore, wave reflection and the deviations of input impedance from characteris-
tic impedance diminish as frequency increases.
Characteristic impedance can be estimated from the input impedance, but it is
not a particularly useful measurement of vascular function because as the degree
of vascular distention increases, vascular compliance decreases and tends to diminish
the changes in characteristic impedance (44). Total arterial compliance can be
estimated by other techniques (45,46). Nevertheless, the changes of input imped-
ance with frequency have important implications for cardiac function. The after-
load for the right ventricle is not the terminal impedance of the system (pulmonary
vascular resistance), but rather the input impedance.
Consequences for cardiac function

The branching nature, asymmetry, and nonuniformity of the pulmonary arterial
tree are factors that tend to randomize the distances between the input and sites
of reflection, and thereby reduce the fluctuations of input impedance that occur
with frequency. As a result, the fluctuations of input impedance with frequency
are less than would be expected in a uniform tube with wave reflections, and the
23
..-
""
II)
as
~ 100
.-
.....E
oscillatory terms
a: 0
w
~
0 mean terms
a.. 100
u
..J
:)
<IIC 200
a:
c IN OUT DISSIPATED
>
z
Figure 2-5. Diagram that indicates the hydraulic power delivered by the right ventricle to
blood in the main pulmonary artery (in), the power present in the pulmonary veins (out), and
the difference, which represents the power used (dissipated) to produce pulsatile pulmonary
blood flow. The total power is separated into a component due to the oscillatory (pulsatile)
nature of flow (above), and a component due solely to mean flow (below). The shaded area
represents power due to pressure energy, and the unshaded area represents power due to kinetic
energy. (Based on the data supplied in reference 42.)
input impedance spectrum is somewhat similar to that of a Windkessel model,

which has no wave propagation. This result is useful because it avoids the right
ventricle having to operate at fIxed frequencies to minimize afterload.
From input impedance data the power delivered into the pulmonary artery has
been estimated for both dogs (47) and humans (41), and the results are fairly similar.
Approximately one-third of the total power delivered to the pulmonary artery is
contained in the pulsatile nature of human pulmonary blood flow (fIgure 2-5).
Therefore, the afterload on the right ventricle depends not only on the pulmonary
vascular resistance but also on the viscoelastic properties of the pulmonary arterial
tree and the inertia of blood.
Nearly all oscillatory power is dissipated because of the reduced pulsatility in
the pulmonary veins. Most of this power is not lost by simply accelerating blood
or simple distention of the pulmonary vasculature during systole. The inertial and
compliant components of the pulmonary circulation act as energy stores. This
energy is used to drive blood through the pulmonary vascular resistance as blood
decelerates and as the Windkessel discharges. On the other hand, some of this power
is dissipated to overcome the viscous resistive elements of the pulmonary vascular
wall. In terms of energy dissipated, the effective resistance of the pulmonary
vasculature has been estimated to be 20% greater than the conventionally measured
pulmonary vascular resistance (48). Possibly the resistive properties of the proximal
pulmonary arterial tree may increase and thereby place an additional load on the
right ventricle without a change in pulmonary vascular resistance as measured by
conventional techniques. Whether or not this ever occurs is unknown. Interest-

ingly, in experimental canine preparations, sympathetic nerve stimulation has been
shown to alter the characteristic impedance (49,50) and the compliance of the main
pulmonary artery (51) without a change of the pulmonary vascular resistance.
Although pulmonary vascular resistance neglects the pulsatile nature of pulmo-
nary blood flow, its measurement is technically easier than the measurement of
pulmonary vascular impedance. While the information gained from measurement
of pulmonary vascular impedance is of particular physiologic interest, its clini-
cal use in pulmonary vascular disease has yet to be clarified. Since pulmonary vas-
cular resistance appears to be the major impediment to pulmonary blood flow,
we shall proceed to describe the factors that can affect pulmonary vascular resis-
tance and thereby alter mean pressure-flow relations in the pulmonary vascular
bed.
Passive factors affecting pressure-flow relations

Lung volume
Pulmonary vascular resistance is dependent primarily on the geometry of the
vasculature, particularly the caliber of the vessels. The location of intrapulmonary
vessels renders them susceptible to mechanical influences that act passively to alter
their dimensions. The experiments of Macklin and of Howell and his colleagues
(52) demonstrated the contrasting changes of vascular dimensions with lung infla-
tion that depend on their size. The volumes of small pulmonary vessels decrease
with lung inflation whereas the volumes of large vessels increase. These experiments
led to classification of pulmonary vessels into two groups: the alveolar and the
extra-alveolar (53). This defmition is based on function rather than anatomy.
Alveolar vessels include all vessels in the alveolar septa that are exposed to alveolar
pressure. Extra-alveolar vessels probably include the rest of the intrapulmonary
vasculature and the corner vessels and are exposed to perivascular interstitial
pressure of the lung, which is dependent on the transmitted pleural pressure by
virtue of tissue attachments (54). The precise anatomic boundaries between these
two types of vessels are unknown.
Lung volume is dependent on transpulmonary pressure. Similar changes of lung
volume occur whether transpulmonary pressure is altered by increasing alveolar
pressure (positive pressure inflation) or by decreasing pleural surface pressure
(negative pressure inflation). Therefore, changes of pulmonary vascular resistance
with lung volume should be independent of the method of inflation because the
measurement of pulmonary vascular resistance uses the intraluminal pressure drop
across the pulmonary vasculature. On the other hand, the arterial and the venous
(or left atrial) pressures must be measured relative to another pressure, usually
atmospheric pressure. For negative pressure inflation, pulmonary vascular pressure
is related to alveolar pressure since it is identical to atmospheric pressure under static
conditions with the airway open. Under these circumstances the pressure on the
external circumference of the extra-alveolar vessels, and on the external circumfer-
25
ence of the alveolar vessel in the longitudinal axis of alveolar septa, decreases while
the pressure on the external circumference of the alveolar vessel in an axis perpen-
dicular to the alveolar septa remains constant. For positive pressure inflation,
pulmonary vascular pressure is related to pleural pressure. Under these conditions
similar changes on the external circumference of the alveolar and extra-alveolar
vessels occur with lung inflation, except that the pressure on alveolar vessels
increases in an axis perpendicular to the alveolar septa. As a result negative and
positive inflation cause differing effects on pulmonary vascular resistance by virtue
of differing effects on the caliber of the alveolar and extra-alveolar vessels rather
than lung volume per se. For example, negative pressure inflation from zero
transpulmonary pressure causes pulmonary vascular resistance to decrease at low
lung volumes as the caliber of extra-alveolar vessels increases by lung expansion,
but to increase at high lung volumes as the caliber of alveolar vessels is reduced
by stretching in an axis longitudinal to the alveolar septa. Pulmonary vascular
resistance is minimal at mid-lung volumes and tends to be related to the changes
oflung volume rather than transpulmonary pressure (55), which differs on inflation
and deflation due to the hysteresis that occurs as a result of alveolar surface tension.
On the other hand, positive pressure inflation causes pulmonary vascular resistance
to increase from low to high lung volume (56) because of the dominant effect of
alveolar pressure to compress the alveolar vessels in an axis perpendicular to the
alveolar septa. Not surprisingly, therefore, these changes of pulmonary vascular
resistance tend to be related to changes of transpulmonary pressure rather than to
lung volume.
Surfoce tension
The hysteresis of lung volume is thought to be related to changes in surface tension
on the alveolar surface since it occurs in air-filled, but not saline-filled, lungs (57).
Lloyd and Wright (58) showed that surface tension also affects the pulmonary
vessels. When alveolar pressure exceeds pulmonary arterial pressure, the alveolar
vessels are compressed. By staining the perfusate of an isolated lung with Evans
blue dye, the investigators showed that flow occurs even when alveolar pressure
exceeded pulmonary arterial pressure by 5 em ~O in the air-filled lung, but not
in the saline-fuled lung. They suggested that in the air-ft1led lung, surface tension
forces were responsible for maintaining the vessels' patency by counteracting the
effects of alveolar pressure. This idea has been confmned by others (59, 60) who
made use of the hysteresis effect of lung volume to change surface tension. In
addition, it has been shown that flow occurs through the corner vessels (61), which
are somewhat protected from alveolar pressure but are likely to be influenced by
alveolar surface tension due to their location.
These changes of pulmonary vascular resistance with lung volume demonstrate
the differing mechanical influence of lung mechanics on the alveolar and extra-
alveolar vessels. They also point out that the effects of alveolar pressure on the
alveolar vessels also depend on its relation to vascular pressures. We shall now
examine this relation in greater detail.
Alveolar pressure
Under physiological conditions three relations are possible between alveolar pres-
sure (P alv)' pulmonary arterial pressure (ParJ' and pulmonary venous pressure
(PveJ:
Palv> Part> Pven

Part> Palv > Pyen
Part> Pyen > Palv
All three conditions can exist within the adult human lung in the upright posture
because of the low pressures in the normal pulmonary circulation, and they have
been designated zones 1, 2, and 3, respectively (62). The vascular pressures at any
level in the lung will vary with lung height because of a hydrostatic gradient of
pressures due to the earth's gravitational field. The height of the lung is approxi-
mately 30 em, with the hilum midway between the apex and the base. A pulmonary
artery pressure of 14 em H 20 relative to the hilum would be insufficient to perfuse
the apex 15 em H 20 above the hilum. Palv (zero em H 20) would compress the
alveolar vessels, and Part at this level would be insufficient to overcome Palv; thus
no Bow would occur (zone 1 conditions). If Pyen (or left atrial pressure) is 5 em
H 20, Pyen at 5 em H 20 above the hilum would be zero. Above this level, where
P art exceeds P alv' Bow would depend on this pressure difference and would be
independent of Pyen (zone 2 conditions). Below this level Pyen exceeds Palv' and
now flow would depend on the pressure difference between Part and Pyen (zone 3
conditions).
The mechanisms that govern blood Bow under each of these three conditions
are illustrated in figure 2-6, based on data obtained with the radioactive xenon
gas method in the isolated perfused dog lung. This method estimates the blood
Bow through alveolar vessels per unit lung volume at each level. This value is
calculated using the ratio of the radioactivity of xenon gas that is evolved into
the alveolar space to blood Bow after vascular injection of the gas in solution and
the radioactivity after the alveolar space is equilibrated with radioactive xenon
gas by rebreathing. Under zone 1 conditions no Bow occurs because P alv com-
presses the alveolar vessels. Under zone 2 conditions, with Part greater than
Palv' P art is sufficient to start overcoming critical opening pressures. As Part in-
creases down zone 2 while alveolar pressure remains constant throughout the
lung, Bow increases. Histological measurements indicate that capillary recruit-
ment is the dominant factor that accounts for the increase in Bow under these
conditions (25,30).
The independence of Bow on the difference between Part and Pyen in zone 2
gave rise to the analogy to sluice Bow (63) or waterfall conditions (64) where
Bow is independent of the height of the fall. This independence of vascular
pressure drop means the concept of pulmonary vascular resistance may be some-
what misleading under these conditions (65). The measurement of pulmonary
27
2 WATERFAll
~
-<:::>
3 DlSTEILSIOIL
~
"'::::9
4 INTERSTITIAL
PRESSURE
~
.....4JllLo....
~
Blood F\ow---+
Figure 2-6. On the left is a diagram to show the factors that operate to modify blood flow in
zones 1, 2, 3, and 4 of a lung in the verticle position. On the right is a diagram to show
regional blood flow per unit lung volume at corresponding levels in the lung. (Reproduced
with permission from reference 72.)
vascular resistance with the balloon occlusion technique to obtain an estimate of

left atrial pressure may be erroneous when the catheter tip subtends the lung
under zone 1 or 2 conditions where there is a discontinuity between arterial and
venous beds (66).
Under zone 3 conditions flow becomes dependent on the difference between
Part and P yen since both pressures are greater than P alv and therefore negate the
compression effect of P alv on the alveolar vessels. Even though the difference
between Part and Pyen is constant at each level within zone 3, since they are both
exposed to the same hydrostatic gradient, flow increases down this zone. Histologi-
cal measurements of capillaries indicate that this increased flow is due primarily to
distention of capillaries down zone 3 and continues to occur up to pulmonary
arterial pressures of 50 em H 20 (25,30). This account explains most but not
necessarily all of the features of regional blood flow in zones 1, 2, and 3. For
example, no statistical difference was found in the increase of blood flow with lung
height between zone 2 and zone 3 (67), as would be expected if different mech-
anisms were operating. Sheet-flow theory can be used to account for this observa-
tion and provides an alternative explanation for the regional blood flow differences
in zones 1, 2, and 3 (68). Some investigators have suggested that flow in zone 3
is almost uniform (69,70), but their measurements were made before the existence
of zone 4 was fully appreciated.
He
150
."
.
-i"' ::::::-- • • • ..A;.AI
:::0 lC,
....I
C> x,"
~
....I
CI
100
'x,
~
C>
RV lC""
....I
~
C>
C>
C>
....I
ao 50
.....
.<:>
Bottom eX
o· I
20
I
15
I
10
I 1
0
5
LUNG DISTANCE (ems be(ow nb 21
Figure 2-7. Relative blood flow at different levels in the lung (ordinate) plotted against
vertical distance of the lung from the second rib anteriorly (abscissa) at total lung capacity
(TIC). functional residual capacity (FRC). and residual volume (RV). Each point represents
average data obtained in eight normal subjects in the upright posture. (Reproduced with
permission from reference 72.)
Interstitial pressure
Below zone 3, flow decreases down the lung despite increasing levels of Part and
Pyen in zone 4. This fmding indicates that flow can no longer be explained in terms
of the relation between alveolar and vascular pressures on alveolar vessels. It has
been suggested that the chief site of resistance shifts from the alveolar to the
extra-alveolar vessels (68). The location of the extra-alveolar vessels protects them
from alveolar pressure but exposes them to lung interstitial pressure. Two factors
that can increase interstitial pressure and reduce the caliber of extra-alveolar vessels
are a decrease in lung volume and the development of perivascular edema (71).
Changes of regional blood flow that occur with lung volume (72) and edema (73)
in upright human subjects support the idea that zone 4 is due to changes in the
extra-alveolar vessels. Zone 4 occurs at the dependent portion of the lung (72),
where the lung is less distended compared with the apex (74) due to the vertical
gradient of transpulmonary pressure (75). The height of zone 4 increases as lung
volume is reduced (73). At residual volume, where the extra-alveolar vessels are
likely to be the chief site of resistance (see earlier discussion on lung volume), zone
4 occupies the entire lung (figure 2-7): regional blood flow per unit lung volume
progressively increases from base to apex. A similar distribution of pulmonary
29
blood flow at all lung volumes has been observed in patients who have raised
venous pressure and are likely to have perivascular edema (73).
On the other hand, the cause of zone 4 in the isolated perfused lung is less readily
explained. The decreased blood flow in zone 4 in this preparation (71) cannot be
related to a reduction of regional lung volume because alveolar size is uniform
throughout the isolated lung (76). The decreasing flow toward the lung base could
be due to increasing interstitial pressure since the interstitium of the lung is a
continuum and subject to a hydrostatic gradient of pressures. This hypothesis lacks
morphometric support (77), and why the hydrostatic gradient of interstitial pres-
sure is not balanced by the hydrostatic gradient of pulmonary vascular pressures
is unclear.
Measurement of pressure:flow relations

Studies of regional blood flow within the vertical human lung are of considerable
interest because they demonstrate the spectrum of mechanical factors that influence
the normal pulmonary vasculature. In general, the mechanisms underlying these
changes can be explained although some caveats do exist. Regardless of the mech-
anisms involved, the results demonstrate how pulmonary vascular resistance is not
a constant, but can vary even when the pressure drop across the pulmonary
vasculature (~P) is constant.
At each level within the vertical lung, the difference between Part and Pyen is
constant due to the hydrostatic gradient of pressures within the pulmonary vascula-
ture. Therefore, as the levels of Part and Pyen increase proceeding down the lung
through zones 1 to 3, pulmonary vascular resistance decreases since Part minus
Pyen is constant and regional blood flow is increasing. As a result, pulmonary
vascular resistance is dependent on the levels of Part. This observation is of im-
portance clinically when pulmonary vascular resistance is used to determine
whether a therapeutic intervention has had a direct effect on pulmonary vascular
tone. If both P art and blood flow increase and pulmonary vascular resistance
decreases, the change of pulmonary vascular resistance cannot be attributed neces-
sarily to change of pulmonary vascular tone. The decrease of pulmonary vascular
resistance may be solely due to the mechanical effect of vascular pressures increas-
ing recruitment and/or distention of the pulmonary capillary bed. If both Part
and blood flow decrease and pulmonary vascular resistance increases, similar ar-
guments apply.
To determine whether changes of pulmonary vascular resistance are due to
passive mechanical factors or to active effects on pulmonary vascular tone, knowl-
edge of the passive pressure-flow relation of the pulmonary vasculature is required.
Several investigators have obtained this relation in human subjects (78-81). Rather
than producing a primary change of the level of vascular pressures and measuring
the secondary increase in flow, they produced a primary increase of blood flow
through one or both lungs by unilateral pulmonary artery occlusion and/or exer-
cise and measured the secondary changes of the pressure drop (~P). Usually, the
pressure-flow plot consists of only three points: two data points and the origin. This
Type II
o Q
.
Figure 2-8. Diagram to show the two types of pressure-flow relations that have been obtained
in supine human subjects. The pressure drop across the lungs (ordinate, .1P, which is the
difference between mean pulmonary arterial pr.essure and mean left atrial pressure) as plotted
against total pulmonary blood flow (abscissa, Q). Pulmonary vascular resistance is represented ~y
the slope of the interrupted lines. For type I curve pulmonary vascular r!=Sistance decreases as Q
increases. For type II curve pulmonary vascular resistance is constant as Q increases. (Modified
from graph given in reference 80.)
approach assumes that the interventions designed to increase blood flow do not
modify the pressure-flow relation by active mechanisms. These experiments were
conducted in the supine position so that the major portion of the lung was
presumably under zone 3 conditions.
Two types of relation have been observed (figure 2-8). In normal subjects with
increased alveolar pressure (78), in high-altitude dwellers with pulmonary hyper-
tension (81), and in patients with chronic bronchitis (78), the relation is initially
curvilinear and then becomes linear as blood flow increases (type I). The initial
curvilinear portion is presumed to be the result of recruitment and distention of
the pulmonary vasculature, but becomes linear as blood flow increases when these
processes are complete. In one study additional data points to obtain the shape of
this relation were obtained from spontaneous fluctuations of blood flow (80). On
this plot pulmonary vascular resistance is represented by the slope of a line between
a point on this relation and the origin. For the type I relation, pulmonary vascular
resistance decreases as blood flow increases. This pattern of a decreasing pulmonary
vascular resistance with increased flow is similar to that which occurs in different
lung regions. It should be noted that pulmonary vascular resistance, as convention-
ally calculated, continues to decrease even on the linear portion of this relation.
31
~P
400 cmH 2 0
Pyen= 0 Pyen=O
20
P A1y =4 P A1y = 4
300 Pyen =8
...
I Pyen =12
15
c
~ 200
E 10
a
100 5
o 10 20 30 o 100 200 300

Q ml.min- 1
Figure 2-9. Pressure-flovy relations in the isolated canine lung. The left-hand panel plots
pulmonary blood flow (Q) on the ordinate against pulmonary arterial pressure (P...J on the
abscissa at three levels of pulmonary venous pressure (Pv.,,). while alveolar pressure (Palv) is
constant for all three relations. The pressures indicate that the relation furthest to the left was
obtained under zone 2 conditions and that the other two relations were obtained under zone 3
conditions. The right-hand panel shows the same data plotted as the pressure difference between
pulmonary arterial and venous pressures (~P) on the ordinate against flow on the abscissa. This
plot is similar to that used in figure 2-8. (Drawn from data supplied in reference 63.)
The component of pulmonary artery pressure that is required to recruit and distend
the pulmonary vasculature is incorporated into the calculation of pulmonary
vascular resistance. In normal subjects (78,80) and in patients with mitral stenosis
(78), a rectilinear relation was obtained with an intercept close to the origin (type
II). It was suggested that in this circumstance the pulmonary vasculature was
completely recruited and distended. Consequently, pulmonary vascular resistance
was constant despite increases of blood flow (80).
The conclusion that the pulmonary vasculature in normal supine subjects is
completely recruited and distended is difficult to reconcile with measurements of
regional blood flow in the human lung and histological data from the canine lung.
We believe that this conclusion is unwarranted for the following reasons. First, we
shall examine the pressure-flow data obtained by Banister and Torrance (63) in the
isolated canine lung (figure 2-9, left-hand panel). These data have been replotted
(figure 2-9, right hand panel) as dP against blood flow (0) similar to the plot
shown in figure 2-8. The pressure-flow relation obtained under zone 2 conditions
is similar to the type I relation. The two pressure-flow relations obtained under
zone 3 conditions are similar to the type II relation. It is important to remember
that the data shown in figure 2-9 were obtained under experimental conditions that
were ideal for accurate data collection. Human pressure-flow data, on the other
hand, are obtained under conditions that subject the measurements to greater errors,
and usually only three points are available to describe the relation. Under these
circumstances, it seems unlikely that the subtle nonlinearities of pressure-flow
relations under zone 3 conditions can be detected. The pressure-flow relations in
human subjects are indistinguishable from a rectilinear relation (type II). This
statement does not exclude the possibility that the pressure-flow relation is curvilin-
ear (type I). Measurement error and the small number of data points available from
experiments of this kind render it difficult to detect the small departures from a
rectilinear relation that are likely to occur under these conditions. Therefore,
alterations of pulmonary vascular resistance with changes of pressure and flow
under zone 3 conditions cannot be ruled out.
We believe that the degree of nonlinearity of pressure-flow relations under zone
3 conditions has been overemphasized. Likewise, the resolution of human pressure-
flow data in normal subjects is inadequate to state that recruitment and distention
are virtually complete in zone 3.
The best way to ensure that a change of pulmonary vascular resistance reflects
an active rather than passive effect on the pulmonary vasculature is to measure
changes of flow when all mechanical factors that may affect the pulmonary vascula-
ture are held constant: pulmonary vascular pressures, alveolar pressure, lung vol-
ume, and interstitial pressure. While these criteria can be achieved only in the
experimental animal, they need not preclude clinical investigation. The studies of
human pressure-flow relations have demonstrated the possibility of obtaining useful
information about the behavior of the pulmonary circulation when experimental
design is appropriate. Changes of pulmonary vascular resistance in normal supine
subjects can probably be attributed to active rather than passive effects on the
pulmonary vasculature, but not in the presence of lung disease. Because of the
possibility that the pressure-flow relation in the diseased lung may be curvilinear,
it is not possible to attribute changes of pulmonary vascular resistance to active
effects on the pulmonary vasculature when both pressure and flows are changing
in the same direction and the pulmonary vascular resistance changes in the opposite
direction.
Consequences /or pulmonary gas exchange

Although the low pressures in the pulmonary circulation render the pulmonary
vasculature susceptible to mechanical influences and give rise to marked topograph-
ical variation of pulmonary blood flow distribution, they have little effect on
pulmonary gas exchange. In human subjects it has been estimated that regional
differences of ventilation and blood flow account for the small alveolar-arterial
oxygen gradient due to ventilation-perfusion inequality in the erect posture, but
these differences impair oxygen and carbon dioxide transfer by only 1.3 and 2.1 %,
respectively (82). The majority oflung disease is diffuse and has patchy distribution.
As a result, the impairment of pulmonary gas exchange that is primarily the result
of ventilation-perfusion inequality cannot be explained on a regional basis. The gas
exchange defect appears to be due to inhomogeneity within lung regions at the
level of small lung units. Although the mechanical factors that result in topograph-
33
ical differences of pulmonary blood flow distribution have little effect on pulmo-
nary gas exchange, they may be of importance in modifying the ventilation-
perfusion inequality that occurs at a local level within the lung. As we shall see,
to distinguish the cause for alterations of pulmonary blood flow distribution is
often difficult because both active and passive factors may be operating simultane-
ously. We shall illustrate this problem in two disease states: bronchial asthma and
atelectasis.
The topographical distribution of pulmonary blood flow in patients with bron-
chial asthma shows a patchy distribution (83-85). Areas of hypoperfusion may be
related to local increases of alveolar pressure or mechanical distortion of lung
architecture. Alternatively, local hypoperfusion may be related to pulmonary
vasoconstriction due to local alveolar hypoxia secondary to reduced local alveolar
ventilation, or perhaps due to vasospastic effects of chemical mediators (4,86). In
this circumstance distinguishing the relative roles of active and passive effects is
difficult. In atelectasis a similar problem exists, but the position has been clarified,
at least in animal models.
Passive mechanisms can be invoked to explain the decreased pulmonary blood
flow to an atelectatic lobe: a reduced lung volume increases pulmonary vascular
resistance. On the other hand, lobar collapse causes local hypoxia, which leads
to pulmonary vasoconstriction. At least two studies (87,88) indicate that the ac-
tive pulmonary vasoconstriction causes the decreased blood flow to the atelectatic
lobe. In this circumstance active mechanisms predominate over passive mechan-
Isms.
Active factors affecting pressure-flow relations

Because of the difficulties in distinguishing the active and passive factors that affect
the pulmonary circulation, much of the work in this section is the result of animal
experimentation, for which conditions are more easily controlled. We shall de-
scribe the effects of chemical, neural, and myogenic stimuli on the pulmonary
vasculature. The effects of pharmacological agents will not be discussed in this
chapter.
Humoral control
The ability of alveolar hypoxia to cause pulmonary vasoconstriction is generally
accepted to be the most important active effect on the pulmonary vasculature (89).
This response can occur at a local level within the lung and has been observed to
a varying extent in many mammalian species, including unanesthetized human
subjects with and without lung disease (90--92). The major portion of the hypoxic
pulmonary vascular response occurs at alveolar oxygen tension of25 to 150 mmHg
(93-95). In isolated perfused lungs pulmonary vasodilation has been observed in
many species below an alveolar oxygen tension of 25 mmHg (95,96), but this
degree of hypoxemia is unlikely to occur in human lung disease. The response
(figure 2-10) has been demonstrated to occur even within a small group oflobules
LOCAL Q
.
( ml.miri.')
5
,
•
•
o 50 100 150
LOCAL PA02 (mmHg)
Figure 2-10. The hypoxic pulmonary vascular response in a small group of lobules in the
South American raccoon (94). Local blood flow (Q) in the ordinate is plotted against the
alveolar oxygen tension (PA02) on the abscissa. The relation is drawn by eye. (Reproduced
with permission, from reference 131.)
(less than 1%of the lung) in the South American raccoon (93). Since the majority
of generalized lung disease causes hypoxemia due to ventilation-perfusion inequal-
ity between lung units at a subsegmental level, this result indicates that hypoxic
pulmonary vasoconstriction may be important for regulating local VAl Q ratios (see
the discussion on consequences for gas exchange that follows).
The debate remains unresolved (97) as to whether hypoxia has a direct effect
on pulmonary vascular smooth muscle or an indirect effect that occurs due to
release of a vasoconstrictor mediator or to suppression of a vasodilator substance
during hypoxia (98). Experiments on isolated pulmonary vascular smooth muscle
by the same investigator can be used to support either school of thought (99,100).
Histamine and prostaglandins now appear to be unlikely candidates as a unique
mediator of hypoxic pulmonary vasoconstriction, although they may playa modu-
lating role (101,102). Possibly, both direct and indirect mechanisms may coexist.
The hypoxic pulmonary vascular response is variable not only between species
but within one species, even over the course of a single experiment (103,104). The
state of the initial tone is one reason proposed to explain the variable nature of
the hypoxic response (105). If initial tone is high, then alveolar hypoxia may be
a less effective vasoconstrictor. Perhaps histamine and prostaglandins affect hypoxic
pulmonary vasoconstriction by altering vascular tone.
35
Some have debated the site of hypoxic pulmonary vasoconstriction, that is,
whether the chief site of vasoconstriction resides in the pulmonary arteries, in
capillaries, or in veins. The most compelling evidence indicates that it is in the small
pulmonary arteries associated with terminal and respiratory bronchioles (106,107).
Reports vary as to whether pulmonary arterial (mixed venous) blood oxygen
tension has any direct effect on the pulmonary vasculature. The difficulty with
experiments of this kind is determining whether a decrease of mixed venous oxygen
tension has a direct effect on the pulmonary vasculature or an indirect effect due
to secondary alterations of alveolar oxygen tension. Pulmonary arterial hypoxemia
has been shown to increase pulmonary vascular tone during ventilation with pure
oxygen (108). Under these circumstances alveolar oxygen tension will remain too
high to affect the vasoconstrictor response. In our opinion the claim that the direct
effect of precapillary hypoxemia is as equally effective as alveolar hypoxia in
causing pulmonary vasoconstriction is less certain.
The effects of CO 2 on the pulmonary vasculature have not been evaluated as
extensively as hypoxia but have been studied in human subjects (109). Difficulty
in identifying a pulmonary vascular response to CO 2 may be related in part to its
complex effect. It has been suggested that there may be two components to the
response to increased levels of CO 2: a direct vasoconstrictor effect due to molar
CO 2 and an indirect vasodilator effect due to an associated increase of hydrogen
ion concentration (110-112). The net effect in most manunals is hypercapnic
vasoconstriction. This response is of a lesser magnitude compared with the hypoxic
response (92) and is less likely to be as effective at controlling the distribution of
pulmonary blood flow within the lung because of the narrow range of alveolar
CO 2 tension within the lung compared with alveolar 02 tension. Nevertheless,
CO 2 can have marked effects on the time course (113) of the hypoxic pulmonary
vascular response and may enhance the hypoxic response (111).
The lung is the only organ that receives the entire cardiac output. Therefore,
one is not surprised that many substances are generated, stored, or metabolized by
lung tissue. Although the potential exists, the precise role of these compounds for
control of the pulmonary vasculature in the normal or diseased lung is unknown.
Histamine and prostaglandins are of particular interest because they are known to
be released from lung tissue in circumstances of clinical interest such as anaphylaxis.
Perivascular mast cells contain histamine, which may modulate hypoxic pulmo-
nary vasoconstriction. Exogenous histamine, by acting through HI receptors, causes
vasoconstriction under hyperoxic conditions, but can cause vasodilation through
~ receptors under hypoxic conditions (114). Again, this differential response may
be explained by differences in the initial vascular tone. Prostaglandins are synthe-
sized in the lung (115), and some of these compounds are vasoconstrictors (PGF2J
whereas others are vasodilators (PGE I and PGI:J. Prostaglandins may also
modify flow through the pulmonary circulation by their effects on platelets; for
example, thromboxane promotes while PGI2 inhibits platelet aggregation. Platelets
themselves contain vasoactive substances such as adenosine (116), and antiplatelet
serum has been shown to suppress the hypoxic pulmonary vascular response (117).
The presence of catecholamines and cholinesterase in the pulmonary vasculature

(15) is of particular interest in relation to its innervation by the autonomic nervous
system.
Neural control
Electrical stimulation of the autonomic nervous system has profound effects on
pulmonary blood flow in fetal lambs (118). Sympathetic stimulation reduces blood
flow to one lung to near zero whereas vagal stimulation can double blood flow
to that lung. Electrical stimulation of the sympathetic nerves decreases pulmonary
vascular compliance (51), but the effects on flow are less pronounced in the adult
compared with the fetus (119, 120). Recent data indicate that both alpha- and
beta-adrenergic receptors in pulmonary vascular smooth muscle are innervated and
can cause vasoconstriction and vasodilation, respectively (121). Electrical stimula-
tion of parasympathetic fibers in the vagus nerve has been shown to cause vasodila-
tion (122).
Although there are reflex pathways that act through efferent autonomic path-
ways in response to stimulation ofbaroreceptors and chemoreceptors by pharmaco-
logical stimuli, their pathophysiologic importance is uncertain (123). However,
two studies suggest that physiological stimuli can have an effect through neuronal
efferent pathways. Hypoxemia that is isolated to the systemic vascular bed increases
pulmonary artery pressure (124), and bilateral sympathectomy reduces the local
decrease of pulmonary blood flow in response to unilateral alveolar hypoxia in dogs
(125). At this time no evidence exists that autonomic innervation affects pressure-
flow relations in human subjects. Nevertheless, this possibility cannot be discounted.
Myogenic control
In the systemic circulation vascular smooth muscle tone increases when it is
stretched (126). At a local level this phenomenon may playa role in the autoregula-
tion of blood flow. In the human pulmonary circulation, myogenic control has
been implicated in the pathogenesis of pulmonary hypertension, where it may have
the deleterious effect of magnifying an increase of pulmonary vascular pressure
(127), although direct evidence is lacking. Of interest is the demonstration of a
myogenic reflex in the dog. Acute nonocclusive distention of the main pulmonary
artery causes reflex pulmonary vasoconstriction (128). Unlike the peripheral circu-
lation, this reflex appears to be mediated locally through sympathetic pathways
(129), but its behavior resembles myogenic control. On the other hand, pulmonary
microembolism, which raised the feline pulmonary artery pressure by 21 to 75%,
caused reflex vasodilation. This reflex is mediated through vagal afferents and by
beta-adrenergic efferent pathways (130).
Consequences for gas exchange

Since hypoxic pulmonary vasoconstriction seems to be the dominant active factor
affecting the pulmonary vasculature, we shall consider only the effects of this
37
mechanism on gas exchange. Earlier we indicated that hypoxic pulmonary vaso-

constriction may act to reduce blood flow to the atelectatic lobe; however, the fact
that this mechanism can operate within small lung units suggests that it may be
of importance in the degree of ventilation-perfusion inequality resulting from
more generalized lung disease. Although the precise mechanism of the hypoxic
pulmonary vascular response remains to be elucidated, with the aid of computer
models, some insight has been gained as to its consequences for pulmonary gas
exchange.
Hypoxic pulmonary vasoconstriction can be considered as part of a negative
feedback system that controls VA/Q ratios within a lung (131). For example, a
decrease of local alveolar ventilation reduces the local VAl Q ratio, which, in turn,
decreases local alveolar oxygen tension. The resulting decrease of local perfusion
due to hypoxic pulmonary vasoconstriction will tend to increase the local VAl Q
ratio back toward its former value. The efficiency of this system can be estimated
with techniques derived from control system theory and is expressed in terms of
gain due to feedback. A gain of zero indicates no benefit is derived from hypoxic
pulmonary vasoconstriction; that is, the change ofVA/Q ratio that occurs follow-
ing a change of local VA is as great with hypoxic pulmonary vasoconstriction as
without it. An infmitely high gain indicates that homeostasis is perfect, in that the
change of VAl Q ratio has been completely negated by hypoxic pulmonary vaso-
constriction. Figure 2-11 shows that the maximal gain for hypoxic pulmonary
vasoconstriction, calculated from the hypoxic pulmonary vascular responses in the
South American raccoon, falls between these two extremes. It is moderately effec-
tive at controlling local VAl Q ratios at a VAl Q ratio of close to 004, but the gain
diminishes at extremes of the VAl Q ratio. Only small changes of local alveolar
ventilation were used to calculate the gain of each VAl Q ratio because of non-
linearities in the system. Further analysis revealed that the maximal gain occurs at
moderately low VA/Q ratios, not because of the nonlinearity of the oxygen
dissociation or the hypoxic pulmonary vascular response curves, but because of the
slope of the oxygen dissociation curve. This dependence of gain on the VAl Q ratio
is important for the ability of hypoxic pulmonary vasoconstriction to modify
pulmonary gas exchange by causing a redistribution of pulmonary blood flow in
the presence of ventilation-perfusion inequality.
Von Euler and Liljestrand (132) considered that local hypoxic pulmonary
vasoconstriction acts to improve pulmonary gas exchange by diverting blood flow
to better ventilated lung units. Recent analysis has shown that this redistribution
of blood flow within the lung usually, but not always, has a beneficial effect on
pulmonary gas exchange. In certain circumstances the maximal effect of hypoxic
pulmonary vasoconstriction at moderately low VA/Q ratios may cause worsening
of arterial hypoxemia by diverting blood flow to lung units with even lower
VAl Q ratios if these lung units receive the preponderance of pulmonary blood flow
{131}.
Another potentially adverse effect of hypoxic pulmonary vasoconstriction on
pulmonary gas exchange is its tendency to reduce total pulmonary blood flow
Gfb
1.0
NEGATIVE
0.8
FEEDBACK
0.6
0.4
0.2
0.01 0.1
.VA/a.1.0
RATIO
10.0 100.0
Figure 2-11. Negative feedback effect of hypoxic pulmonary vasoconstriction on gas exchange
in a single lung unit, based on the relations shown in figure 2-10. The homeostatic effect on the
local ventilation-~rfusion ratio to small changes of local alveolar ventilation, expressed as gain.
due. to feedback (Gfb' on the ordinate), is plotted against the local ventilation-perfusion ratio (V
A/Q, on the abscissa on a log scale). Inspired gas was air and mixed venous gas tensions for
oxygen and CO 2 were 40 and 46 mmHg, respectively. (Reproduced with permission from
reference 131.)
(cardiac output). In a lung model in which overall ventilation, perfusion, oxygen

uptake, and carbon dioxide output are held constant, ventilation-perfusion inequal-
ity will result in an increase of arterial and mixed venous blood CO 2 gas tensions
and a decrease of arterial and mixed venous blood oxygen tension. If hypoxic
pulmonary vasoconstriction is incorporated into the model and total pulmC?nary
blood flow is not held constant, total flow will decrease with increasing VA/Q
inequality due to local reductions in blood flow associated with the development
of local alveolar hypoxia (131). Whereas hypoxic pulmonary vasoconstriction
under these circumstances improves the pulmonary gas exchange and reduces the
deterioration of arterial blood gas tensions, the fall in cardiac output has an adverse
effect on mixed venous oxygen tension. As a result the lung is less able to withstand
more severe degrees of ventilation-perfusion inequality (figure 2-12). In reality this
sequence of events is unlikely to occur because cardiac output increases when
hypoxemia and hypercarbia occur. As a result, the beneficial effects of hypoxic
pulmonary vasoconstriction on arterial blood gas tensions are retained, but its
39
Arterial
Blood Gas
Tensions
0.5 .5
100 ,
I
so
Mi xed Venous 60
Blood Gas
Tensions
40--=-"'-
20
0.5 1.0 1.5 2.5
Figure 2-12. Effect of local hypoxic pulmonary vasoconstriction on arterial (above, on the
ordinate) and mixed venous gas tensions (below, on the ordinate) plotted against the degree of
ventilation-perfusion inequality Qog.,SD) on the abscissae. The interrupted lines show the effects
of ventilation-perfusion inequality for arterial (Pa02 and PaC02) and mixed venous (PV02 and
P V C0 2) blood gas tensions without hypoxic pulmonary vasoconstriction. Overall alveolar
ventilation, pulmonary blood flow, oxygen uptake, and carbon dioxide output are held constant
at 5.1, 6.0,0.3, and 0.24 liters/min- t , respectively. The solid line shows the blood gas tensions
under the same conditions when local hypoxic pulmonary vasoconstriction is incorporated into
each of the 100 lung compartments and allowed to alter overall pulmonary blood flow.
(Redrawn from data provided in reference 131.)
adverse effects on mixed venous oxygen tension are avoided (figure 2-13). This
analysis demonstrates that hypoxic pulmonary vasoconstriction has the potential to
reduce the deleterious effects of lung disease on pulmonary gas exchange, particu-
larly when it operates in association with other regulatory mechanisms.
The price to be paid for hypoxic pulmonary vasoconstriction is the increased
energy demands placed on the right ventricle to pump blood through the pulmo-
\ /
80 \ ,, ,
,'''----P a CO2
~ ,, ,,
~
Arterial 60 ,
Blood Gas
Tensions
I
"
_0-'~<.
,,
40 '0,
,
20
'",, '0,
'''"':-P a 02
01
0.5 1.5 2.5 3.5
100
,
,,
,
/'----.py- CO2
80
,
!'
,-.-
/
Mixed Venous
BT'ood Gas
Tensions
60
40
- ..- .......... ,Y
--o...---o...... ~o...
--" ,
20 "-, ,
'O~Pv02
,
a
a 0.5 1.0 '.5 2.0 2.5 3.0 3.5
VA/a inequality (loge SO)
Figure 2-13. Effect of local hypoxic pulmonary vasoconstriction and overall pulmonary blood
flow (cardiac output) control by arterial blood gas tensions on the blood gas tensions in the
presence of increasing ventilation-perfusion inequality. The plots are similar to those shown in
figure 2-12. The interrupted line shows the blood gas tensions with no control mechanisms
operating under the same conditions described in the legends of figure 2-12. The continuous
lines show the blood gas tensions under the same conditions when both local (hypoxic
pulmonary vasoconstriction) and overall pulmonary blood flow control are incorporated into the
model. (Redrawn from data provided in reference 131.)
nary circulation. As larger portions of the lung become hypoxic, the ability of
hypoxic pulmonary vasoconstriction to redistribute flow diminishes and pulmo-
nary hypertension occurs (133,134). An extreme example is the global alveolar
hypoxia that occurs on ascent to high altitude, which leads to pulmonary hyperten-
sion and edema in susceptible individuals (135).
THE ABNORMAL PULMONARY CIRCULATION

The first part of this chapter carefully reviewed the physiology of the normal
pulmonary circulation, evaluating both passive and active processes. In this section,
we shall consider how these processes interact under abnormal conditions to influ-
ence function in the patient.
41
Anatomical influences on pathophysiology

Anatomical alterations of the pulmonary vascular bed are a common accompani-
ment to many diseases of the lung, both primary and secondary. As already
discussed, those changes will affect the pressure-flow relations within the lung and
may eventually lead to elevations in pulmonary artery pressure and cor pulmonale.
This section will discuss the physiological consequences of the two most common
mechanisms that lead to a reduction in the cross-sectional area of the pulmonary
vasculature.
Vascular remodeling
Any prolonged elevation of pulmonary artery pressure, regardless of the cause, will
lead to anatomic remodeling of the pulmonary vessels {136}. Since a complete
discussion of these morphological changes will be found in chapter 3, we will
confme ourselves here to discussing the consequences of chronic hypoxia as they
influence pathophysiology in clinical situations.
The acute and reversible effects of alveolar hypoxia on pulmonary vascular tone
have already been discussed. The beneficial effects of hypoxic vasoconstriction on
ventilation-blood flow matching {94} and the detrimental effects during the pro-
duction of high-altitude pulmonary edema {137} suggest the degree of its acute
influence on cardiopulmonary hemodynamics. In disorders characterized by
chronic pulmonary abnormalities where the presence of hypoxia is likely to be
prolonged, the vasoconstriction will eventually lead to significant anatomic
changes. These, in turn, result in progression of pulmonary hypertension and its
continued presence, even when the initial stimulus has been removed.
Much of our knowledge of these changes comes from descriptions of chronic
hypoxia induced in experimental animals. In rats exposed to hypobaric hypoxia,
a gradual but progressive increase in pulmonary artery pressures above the initial
acute response is found {138}. The initial response results from hypoxic vasocon-
striction and is rapidly reversible on return to normoxia. The progressive increase
correlates with changes in the anatomy of the small pulmonary arterioles and
arteries. With continued hypoxia the first change seen is an alteration in the
distribution of smooth muscle in the small pulmonary arterioles. In the normal lung
the muscle coat of these arteries progressively decreases along with the diameter.
Chronic hypoxic exposure results in an extension of the smooth muscle into the
partially muscularized and unmuscularized vessels so that a progressively longer
segment of the arterial tree is enveloped in a fully developed muscular coat {139}.
The new smooth muscle is thought to arise from the conversion of pericytes and
intermediate cells, which are normally found in nonmuscular arteries, into smooth
muscle cells {140}. At the same time an increase in the medial muscle coat of the
small pulmonary arteries and a decrease in their total number are seen by angio-
graphic techniques {141}. Other anatomical changes that have been described in
pulmonary hypertension, including medial hypertrophy of the larger arteries and
angiodestructive lesions, are not seen as a result of hypoxia. The absense of these
changes are most likely related to the relatively milder elevations of pulmonary
artery pressures that develop (142).
Whereas most of the morphological alterations resulting from hypoxic exposure
are reversible, the presence of these anatomical changes is likely to explain the
failure of patients with hypoxic lung disease to respond acutely to increased oxygen
levels with a reduction in pulmonary vascular resistance. Despite the general
impression that hypoxia is the major cause of pulmonary hypertension in patients
with hypoxemic lung disease, such as chronic obstructive pulmonary disease
(COPD), and the observation that in these patients the level of pulmonary artery
pressure is inversely correlated with the arterial P0 2 (143,144), an acute response
to oxygen is rarely seen. In patients with stable severe COPD, increasing the
inspired oxygen concentration (FI0z) not uncommonly leads to a fall in pulmo-
nary artery pressure, but this fall is usually accompanied by a fall in cardiac output
and no decrease in the calculated pulmonary vascular resistance (145,146). Even
during exacerbations of chronic disease with worsening hypoxemia, the acute
effects of supplemental 02 on pulmonary vascular resistance have been unimpres-
sive (147). On the other hand, these patients do appear to retain their ability to
exhibit hypoxic vasoconstriction; an increase in pulmonary arterial pressure was
observed when stable patients with COPD were exposed to 14 to 15% oxygen
(147). Additionally, it has been demonstrated that the administration of oxygen
could lessen the rise in pulmonary artery pressure seen during exercise in patients
with COPD (144).
If we accept that anatomical alterations prevent the acute reversal of hypoxia-
induced pulmonary hypertension, we must next look at the time course of reversal
of the anatomical changes with prolonged relief of the hypoxia. This pattern of
regression can, perhaps, be best described in its purest form in subjects born and
raised at high altitudes who return to sea level for prolonged periods (137).
Highlanders have been clearly shown to have pulmonary hypertension. The pulmo-
nary vascular resistance is highest in children and gradually falls to adult levels after
five years of age. This difference relates to the greater muscularity state of the fetal
pulmonary vascular bed. The fall of pulmonary vascular resistance to adult levels
following birth is duplicated in lowlanders, but over a markedly shorter period
of time. Healthy highlanders have a pulmonary vascular resistance 2 to 2.5 times
that of age-matched lowlanders, and this difference is exaggerated during exercise.
When highlanders return to sea level, an immediate decrease in pulmonary vascular
resistance of 15 to 20% occurs, which is ascribed to the release of hypoxic
vasoconstriction. The remainder of the pulmonary hypertension reverses slowly,
and the reversal is thought to involve both a decrease in the viscosity of the blood
as hypoxemia-induced polycythemia resolves and a regression of the musculariza-
tion of the terminal portions of the pulmonary arteries. After two years normal
sea-level pressures are invariably found. In experimental animals the recovery of
normal pressure after relief of chronic hypoxia is more rapid, although evidence
of continued anatomical changes is present for months (148).
In light of this behavior, one might expect prolonged oxygen therapy to result
43
in a similar gradual, but progressive, improvement in the pulmonary hypertension

of hypoxic lung disease. Whereas early, small, uncontrolled trials of prolonged
oxygen therapy (some for as short as four to eight weeks) suggested a similar
response (149,150), recent large, controlled studies have failed to demonstrate a
major salutory effect on pulmonary hemodynamics even after one year. Both the
British Medical Research Council trial (MRC) , which compared no oxygen with
15 hours of oxygen per day (151), and the Nocturnal Oxygen Therapy Trial
(NOTT), which compared 12 hours of oxygen with continuous therapy (152),
failed to demonstrate significant decreases in pulmonary vascular resistance. The
oxygen did, however, appear to reduce a further progression of the pulmonary
hypertension, although some data suggest that progression is a very slow process
even in untreated patients (153).
Before we conclude from these fmdings that hypoxia is therefore not the major
stimulus to pulmonary hypertension in patients with lung disease, we must look
more closely at the physiological mechanisms that are operating. We have already
mentioned that in studies of rats exposed to chronic hypoxia, a return to normoxia
resulted in a return to normal pressures, just as in the highlanders moving to sea
level; but for this to occur, the normoxia must be continuous. When the animals
were allowed only intermittent normoxia, even up to 16 hours per day, no
regression of pulmonary hypertension was found (154). Even the continuous
oxygen group in the NOTT trial averaged only 17.7 hours of actual oxygen use
per day. Thus, some of the failure to see a regression in the pulmonary hypertension
may be due to a continued intermittent hypoxic insult.
Equally interesting is the fmding that exposure of normal animals to as little as
eight hours of hypoxia daily leads to the development of cor pulmonale as
measured by the ratio of the weight of the right ventricle to the weight of the
left ventricle plus septum (155). Such short periods of hypoxia, along with increases
in pulmonary artery pressures, have been documented in patients with COPD
during sleep due to both apneic and hypopneic episodes (156). This could explain
the development of pulmonary hypertension in those patients despite the presence
of normal arterial blood gases during the day.
Another mechanism that may prevent the regression of cor pulmonale despite
even continued use of supplemental oxygen requires an understanding of the
relation between the ventilation-perfusion ratio eVA/OJ and the major stimulus for
hypoxic vasoconstriction, the alveolar P02 (P A Oz). The PA 02 of any lung unit
is highly dependent on its V.Alq (157). Figure 2-14 graphs the relat~on ~etween
the changes of P A02 with VAl Q ratio in lung units with different VAl Q ratios.
On room air! major changes ofPA02 with VA/Q occur over a rather narrow range
-between VA/Q ratios of 0.15 to 4.6.
Normal young subjects have a rather narrow distribution of VA/Q ratios cen-
tered on a mean of 0.78 and spanning no more than one decade along the VA/Q
axis graphed on a long scale. Older normals experience some increase in the span,
but little or no blood flow is found below a VA/Q of 0.1 (158). In patients with
COPD the distribution of VA/Q is markedly widened (159). Figure 2-15 re-
50
01 40
1:
~
'0
.... 30
~
.>
<0 20
8
:
s 10
a
0.01 0.1 1.0 10.0 100.0 1000.0
VA/Q RATIO
Figure 2-14. Values of the partial derivative (8PA02/8VA/Q) on the ordinate at different
ventilation-perfusion ratios on the abscissa. The partial derivative is a measure of t~e c~ge of
alveol:rr o:crgen tension (PA02) for a small change of ventilation-perfusion ratio (VA/ Q). At
each VA/Q ratio the inspired gas is air, and mixed venous blood gas tensions are 40 mmHg for
oxygen and .46 ipffiHg for carbon dioxide. Note the maximum value of the partial derivative
occurs at a VA/Q ratio of 0.4.
c:
'E
.... 1.0
..J •
Shunt
to.8'"
.75
~
o
U.
'C
o .50
o
CD
-
~
oCS
.25
--
c:
.2
.~
cQ):
o .001 .01 0.1 1.0 10 100
>
Ventilation - Perfusion Ratio
Figure 2-15. The ventilation-perfusion distribution of a patient with chronic obstructive
pulmonary disease during an episode of acute bronchitis.
45
produces the distribution of VAl Q obtained utilizing the technique of multiple

inert gas elimination (160) in a patient with severe COPD and cor pulmonale. In
this patient are !ung units that vary all the way from ~otal.ly unventilated (shunt)
and very low VA/Q to units with abnormally high VA/Q and unperfused units
(dead space). This pattern of VA/Q inequality is typical of patients with both
chronic bronchitis and emphysema (159). The presence of a shunt of this size is
unusual in stable patients with COPD but can develop in patients during even mild
exacerbations. The P A 02 in a large proportion of these lung units will be well
within the range in which significant hypoxic vasoconstriction should occur when
the patient is breathing room air.
The ability of an increase in the ~r02. to increase the Pa0 2 in patients with lung
disease depends on the underlying VA/Q distribution (157). In normal subjects or
in those with only mild lung disease, the Pa02 will rise in a linear fashion with
the Fr0 2. With progressively higher desree~ of VA/Q inequality, the increases in
Pa02 with Fr02 diminish. With severe VA/Q inequality little improvement in the
Pa02 may be seen until the FP2 reaches 0.4 or higher. This differing pattern of
response reflects the effect of an increasing Fr02 on the various VA/Q ratios of the
gas exchanging units within the lung. Lung units with normal or high VAl Q will
experience a rapid rise in P A02 as Fr02 is increased, while those with very low
VA/Q will be quite resistant to any change.
Perhaps this pattern can be most graphically demonstrated with a theoretical
example. Figure 2-16 plots the effect of increasing the FP2 on the Pa02 in a
theoretical lung with severe VA/Q inequality, similar in degree to the patient in
figure 2-15. For the purpose of this exercise, we have constrained minute ventila-
tion, cardiac output, and metabolic rate and have calculated the effect of a clinically
useful range of FP2 on the Pa02. The line labeled "lung" represents the change
in Pa02' and it is evident that a "clinically acceptable" level is reached with only
small increments in the Fr0 2. However, in the large number of lung units with
low VAl Q ratios, the P A02 remains essentially unchanged. Thus, the achievement
of an acceptable Pa0 2 in patients with chronic hypoxemic lung disease is likely
to leave the P A02 of a significantly large portion of the lung at levels at which
the stimulus for hypoxic vasoconstriction will remain intense. Because levels of
inspired 02 necessary to raise the P A02 of these units to levels high enough to tum
off hypoxic vasoconstriction are impractical, attempts at pharmacological manipu-
lation of vascular tone are now being attempted. They will be discussed in chapter
12.
Vascular bed obliteration

A second common mechanism for anatomic alteration of the pulmonary vascular
bed is the occlusion or obliteration of pulmonary vessels. While the cross-sectional
area of the pulmonary vascular bed may be reduced by many conditions-includ-
ing vasculitis, congenital stenosis, parasitic infestation, and self-injected particles by
drug addicts-the two most common conditions in which this mechanism is
thought to play a significant role are emphysema and thromboembolic disease.
120
100
80
40
20
o .21 .24 .28 .35
Figure 2-16. The effect of increasing the inspired oxygen fraction (F10 2) on the arterial blood
and end-capillary blood in individual lung units in a patient with COPD. While the arterial
P0 2 is increased to clinically accep~ablt; levels by an F10 2 as low as .28, the end-capillary blood
in individual lung units with low VA/Q ratios are relatively unchanged.
Whatever the morphometric technique used to measure the alveolar surface,

emphysema is found to result in a marked loss of alveolar surface area (161). The
amount of loss depends, obviously, on the degree of disease, as well as on the
pattern, with greater reductions found in panlobular as contrasted to centrilobular
emphysema. Reductions in the surface area of as much as 50% have been found,
with decreases from 70 meters 2 to 32 meters2, and decreases in the number of alveoli
from 273 X 106 to as low as 16 X 106• Because the walls of the alveoli are made
up almost entirely by the mesh of pulmonary capillaries, this decrease must result
in a decrease in the capillary cross-sectional area of a similar magnitude. Because
a large proportion of the resistance to flow is attributed to the capillary bed, this
obliteration has been assumed to playa role in the development of cor pulmonale
in COPD.
Many studies have been designed to test this hypothesis. All are based on
postmortem observations relating morphometric measurements of the number of
alveoli or the alveolar surface area to the presence of cor pulmonale as defined by
the presence of right ventricular hypertrophy. Right ventricular hypertrophy is
47
often assessed by measuring the thickness of the right ventricular free wall. This
measurement has been shown to be less sensitive than a measurement of the weight
of the right ventricle or, preferably, the ratio of the weight of the right ventricle
to the weight of the left ventricle plus the septum after excluding hearts with
significant left ventricular hypertrophy (162).
Some studies have indeed found a positive correlation between the degree of
emphysema and right ventricular weight (163,164). However, these studies invari-
ably have included individuals with no history of COPo. Examination of these
data suggests that the correlation is due in great part to the inclusion of a large
number of subjects with no emphysema and normal right ventricular weight.
Studies in which only patients with known lung disease were included failed to
demonstrate any relationship between right ventricular hypertrophy and the degree
of emphysema. In fact, evidence of cor pulmonale was more likely to be found
in the presence of centrilobular emphysema than in panlobular, despite its less
destructive nature (161,165).
A strong correlation has been found between right ventricular hypertrophy and
muscularization of the media of the small pulmonary arterioles in patients with
COPO and cystic fibrosis (161,166). As mentioned earlier, increased musculariza-
tion of the media is typically seen in chronic hypoxic states and suggests that
vasoconstriction rather than anatomical obliteration is a more important etiology
of the pulmonary hypertension in patients with copo.
At first glance, pulmonary embolic disease would appear to present a clear
situation in which the alterations in pulmonary vascular resistance are the direct
effect of vascular obliteration. As might be predicted from the pressure-flow
relations in the normal supine lung, a direct relationship between the loss of
vascular cross-sectional area (as estimated by angiography) and the increase in
pulmonary artery pressure has been found following pulmonary emboli in patients
who are free of underlying cardiopulmonary disease (167). No apparent obstruc-
tion threshold for the elevation of pressure in this study was found, but there
appeared to be a limit. No patients were found to have a mean pulmonary artery
pressure greater than 40 mmHg. The authors suggested that 40 mmHg represented
the pressure limit that could be achieved by the normal right ventricle. This would
appear to be a reasonable approximation. In a similar study of acute cardiopulmo-
nary hemodynamics following acute pulmonary emboli, four of eight patients with
a mean pulmonary artery pressure of 40 mmHg or greater died, three within hours
of the acute event (168).
The hemodynamic consequences of acute embolic disease in patients with preex-
isting cardiovascular disease are accentuated (169). The degree of pulmonary hyper-
tension is much greater for a lesser degree of occlusion, often exceeding a mean
pressure of 40 mmHg. In addition, there is no apparent correlation of the pressure
rise with the degree of occlusion. This fmding is not surprising since the pressure-
flow characteristics in these patients may differ considerably from normals, with
a steeper slope and nonlinear relation. Some have suggested that the fmding of high
pulmonary pressures out of proportion to the degree of embolic occlusion infers

the presence of chronic embolic disease or other preceding cardiopulmonary abnor-
malities (170).
A confounding factor in the interpretation of the relation between vascular
occlusion and pulmonary hypertension in these patients is the presence of hypox-
emia. A widened alveolar-arterial gradient for oxygen is almost always seen in
patients with significant embolic disease and can be attributed to either ventilation-
perfusion mismatch or the presence of shunt. In studies in which it has been looked
for, a good correlation has been found between the elevation of pulmonary artery
pressure and the fall in Pa02 (167). In addition, a recent study compared the
elevation of pulmonary artery pressures in a group of patients with less than 25%
vascular obstruction due to embolic disease with a matched group of normals and
a group of patients with other acute illnesses initially thought to be embolic but
angiographically excluded (171). The elevation of pulmonary artery pressure above
that of the control group was similar in both patient groups and correlated well
with the decrease in Pa02. The additional role of hypoxic vasoconstriction as a
cause of acute pulmonary embolic hypertension probably accounts for the higher
pressures seen following thromboembolization as compared with an equal degree
of obstruction achieved by balloon occlusion studies (172).
The course of the elevated pulmonary artery pressure following embolization
demonstrates a gradual but progressive reduction over the first few weeks. This
improvement is coincident with a decrease in the degree of obstruction due to
breaking and dissolution of the clots, as well as to an improvement in overall gas
exchange (168). The long-term prognosis in most patients is favorable. In a study
of 76 patients with proven embolic disease, 74% had an initial mean pulmonary
artery pressure of 30 mmHg or less (173). A reduction in the pressures was found
in all on n~peat catheterization. However, in those patients with high pressures at
initial catheterization, a persistence or worsening of the hypertension was common,
and 16 of 20 died as a result. As discussed previously, this group of patients likely
represents those with preexisting disease or recurrent embolization. The develop-
ment of vascular remodeling in response to the high pressure is likely to represent
an additional contribution to the maintenance of increased pressure. Finally, the
ability to reduce pulmonary vascular resistance in selected patients with vasodilators
suggests an added role of increased vascular tone (174).
Functional influence on pathophysiology
Mechanical factors
Both the heart and lungs are enclosed within the thorax and are thus subject to
the alterations in pressures occurring during respiration. The heart and large intra-
thoracic vessels are exposed to pleural pressure, while the remainder of the extra-
alveolar vessels are surrounded by pulmonary interstitial pressure. When transpul-
monary pressure is low (at low lung volume), interstitial pressure approaches
pleural pressure but becomes exceedingly more negative as transpulmonary pressure
49
(and thus lung volume) increases (175). The intra-alveolar vessels, in contrast, are
exposed to alveolar pressure which remains close to atmospheric pressure during
normal spontaneous ventilation.
While we reference our clinical measurements of intravascular pressures to
atmospheric pressure, it is important to realize that from the view of the heart, the
true distending pressure is the transmural pressure, or the difference in pressure
across the wall of the cardiac chamber or vessel. These transmural pressures deter-
mine both the preload and afterload on the heart and influence the transvascular
movement of fluid in the lungs (175).
During normal inspiration ventricular stroke volume falls (176). This fall is
thought to be the result of two mechanisms. First, the decrease in pleural pressure
leads to an increase in the afterload on both the right and left ventricles. This results
from an increase in the transmural pressure in the intrathoracic vessels while the
pressure around the intra-alveolar and extrathoracic systemic vessels remains rela-
tively unchanged. Even et al. (177) likened this situation to a water pump in a
valley filling a reservoir on the top of the hill: "If a pump (the heart) in a valley
(the thorax) has to fill a reservoir on the hill (the alveolar or peripheral vessels),
the pressure that the pump has to develop must be measured by reference to the
level of the pump and not to the level of the hill. So the lower the valley, the
higher the energy needed to fill the reservoir, and the lower (i.e., more negative)
the intrathoracic pressure, the higher the cardiac afterload."
In addition to this increase in afterload, the increased venous return that occurs
during inspiration leads to a shift in the intraventricular septum toward the left
ventricle. This shift decreases left ventricular compliance and further contributes
to the fall in stroke volume (178). Experimental evidence suggests that during
normal quiet breathing this effect of ventricular interdependence is what plays the
major role in the inspiratory fall in stroke volume (179).
In the presence of obstructive lung disease, major alterations in transpulmonary
pressures may occur with a concomitant exaggeration of the influence of respiration
on cardiovascular function. For example, it has long been appreciated that pulmo-
nary artery pressure increased during the development of acute asthma. The increase
was initially assumed to result from the high positive pressures generated as the
patient attempted to expire through obstructed airways. However, measurements
of pleural pressure in these patients have now shown that the major alteration is
a marked increase in the negative pressure during inspiration (180,181) (table 2-1).
This results from an increase in resting lung volume and is thought to occur as an
attempt to maintain airway patency in the face of severe bronchoconstriction (180).
Table 2-1. Pleural pressures in obstructive lung disease (em HP)
Tidal inspiratory pressure Tidal expiratory pressure
Acute asthma -22 to -34 -2 to +7

COPD (rest) -5 to -20 -5 to +6
COPD (exercise) -15 to -32 +7 to +48
The markedly negative pleural pressure increases the hydrostatic gradient which
must be overcome, and thus the heart must develop increased pressure to maintain
flow through the alveolar and extrathoracic systemic vessels. Similar changes in
transpulmonary pressure have been seen in patients with COPD, espeically during
exercise (table 2-1). In contrast to normal respiration, this increase in ventricular
afterload is thought to play the dominant role in the decreased stroke volume seen
during obstructed breathing (179). A further impediment to cardiac output result-
ing from increased lung volume may occur due to direct impairment of ventricular
filling due to compression of the heart by the hyperinfiated lungs. This may also
limit the unstressed volume of the pericardium, exaggerating the effect of the
inspiratory increase in right ventricular volume on left ventricular compliance
(182).
To some extent, these detrimental effects on cardiac function can be compensated
for by an increase in venous return. However, the increasing negative pleural
pressure has only a limited ability to increase the gradient to venous inflow because
of the tendency of the large intrathoracic portion of the veins to collapse as the
negative pressure increases: "the waterfall effect" (183). In addition, studies in
individuals with grossly overinflated lungs have shown an inspiratory decrease in
venous return from the inferior vena cava which is thought to result from an
anatomic constriction of the vena cava by the flattened diaphragm (184). Finally,
the positive expiratory pleural pressures that develop, especially during exercise,
may further impair venous return.
Biochemical
While gas exchange is the most widely appreciated function of the lung, it has
become increasingly clear that the lung also participates in the metabolism and
production of many hormones and mediators (185). Clearly evident is the existence
of an active mechanism for the uptake and degredation of 5-hydroxytryptamine
(5-HT) and norepinephrine. The presence of angiotensin-converting enzyme
(ACE) is important in the hydrolysis of angiotensin I to the active octapeptide,
angiotensin II. In addition, this enzyme is able to metabolize the vasoactive media-
tor, bradykinin, to inactive peptides and dipeptides. The mast cells in the lung are
actively involved in the production of histamine, as well as slow-reacting substance
of anaphylaxis (SRS-A), eosinophilic leukocyte chemotactic factor of anaphylaxis
(ECF-A), and other proteolytic enzymes. They are related in a number of situa-
tions, including hypoxia and asthma.
The lung is probably also prominent in the synthesis and degradation of eicosan-
oids from both the cyclo-oxygenase and lipoxygenase pathways. Recently, addi-
tional vasoactive polypeptides have been discovered in lung extracts. These include
a vasodilator agent identical to vasoactive intestinal peptide (VIP) and vasocon-
strictive substances yet to be characterized (186).
With the exception of the mast cell, the site of most of this metabolic activity
appears to be the pulmonary endothelial cells. This metabolic activity seems to be
a characteristic of endothelial cells in general, and the importance of the lung,
51
therefore, may lie in its anatomy. The large cross-sectional area of the pulmonary
vascular bed necessary for gas exchange also results in a large concentration of
endothelial cells available for metabolism. In addition, since the entire blood flow
passes through the lungs, its products are capable of generalized effects, as opposed
to other vascular beds which are more likely to exert local control.
The exact role of this activity in overall body homeostasis is unclear. For
example, even though the ACE in the lung is responsible for both the production
of a vasoconstrictor (angiotensin) and the destruction of a potent vasodilator
(bradykinin), whether this plays a significant role in the overall mediation of blood
pressure is unclear. Additionally, while many of these hormones and mediators are
capable of dilatation and constriction of both the pulmonary vessels and airways,
the role of these substances in the modulation of gas exchange is unclear, despite
the fmdings that mediators such as histamine are released during lung hyperinflation
(187,188). In addition, their role in lung injury is unclear. Many of them, especially
certain arachidonic acid metabolites, are released in response to endotoxin, pulmo-
nary embolism, aspiration, and other acute insults (115). While many of these
substances have the capability of increasing vascular permeability, their role in the
production of the adult respiratory distress syndrome (ARDS) is unclear. Some
studies have suggested that prostaglandins are released during mechanical ventila-
tion with positive end-expiratory pressure (PEEP) and may be responsible for
myocardial depression and the observed fall in cardiac output (189).
Additionally, the development of lung disease has been shown to interfere with
the metabolic function of the lungs. The total endothelial surface area and the time
of exposure of the blood to these cells appear to be important in modulating both
the synthetic and degradative abilities of the lungs. Inhibition of oxidative metabo-
lism in the lung has been shown to impair the metabolism of 5-HT and the synthesis
oflung lipids (190). In rats given monocrotaline to induce pulmonary hyperten-
sion, a reduction in ACE levels in both serum and lung has been found (191), as
well as a decreased removal of 5-HT and noradrenaline (192). However, clinical
consequences of this reduced metabolic activity on either the lung or any other
body function have yet to be conclusively demonstrated.
Physiological consequences of an abnormal pulmonary circulation

Any abnormality of the pulmonary vascular bed is likely to be translated into
alterations in both cardiac and pulmonary function. The effects on both right and
left ventricular function will be covered extensively in later chapters; therefore, we
will limit our discussion here to the lung.
Consequences for pulmonary gas exchange

The abnormal pulmonary circulation can result in alterations in gas exchange
through two mechanisms. Partial or complete obliteration of pulmonary vessels or
changes in pulmonary perfusion pressure can directly influence the blood flow
distribution and thus the matching of ventilation and blood flow within gas
exchanging units of the lung. In addition, by reducing cardiac output at rest or

interfering with the normal response to increased metabolic demands, arterial
hypoxemia may be exaggerated due to a widening of the arterial-venous oxygen
difference.
While hypoxemia is characteristic of lung disease in which abnormal pulmonary
hemodynamics are an important by-product, to separate the contribution of the
pulmonary vascular disease from the underlying parenchymal lung disease is diffi-
cult. In patients with chronic bronchitis, emphysema (159), or cystic fibrosis (193),
no difference in the pattern ofVA/Q inequality was found between those patients
with and those without pulmonary hypertension.
Patients in whom pulmonary vascular disease exists in the absence of significant
parenchymal disease should provide a clearer representation of the isolated effects
of the vascular disease on gas exchange. In acute thromboembolic disease VAl Q
inequality has been proposed as the most likely mechanism of the abnormal
alveolar-arterial gradient for 02 (A-aDOz) (194,195). The mismatch of ventilation
and blood flow is best explained by the redistribution of blood flow from the
embolized to nonembolized lung units. Partial or total occlusion of blood flow
would result·in the development of lung units with high VA/Q ratios or increased
dead space (ventilated, but unperfused lung units) while hypoxemia would occur
only if redistribution of blood flow to nonembolized units was sufficient. This can
be seen by referring to the theoretical lung model in figure 2-17. The almost total
occlusion of blood flow from a relatively small number oflung units would result
in physiological dead space but not significant hypoxemia. However, partial occlu-
sion of the large portion of the pulmonary vascular bed would lead to the
development of primarily low VAl Q units with resultant hypoxemia. Examples
of both extremes have been seen in experimentally induced thromboembolism in
dogs (195) (figure 2-18). In most cases a combination of these patterns is seen, so
that increases in both high and low. VJQ units are found (figure 2-19). An
additional influence on the pattern ofVA/Q inequality found following pulmonary
embolization is the size of the embolus (37). In a study of resin bead embolization
in dogs, emboli smaller than 0.15 mm never caused high VA/Q units whereas larger
emboli always did. Smaller emboli were concluded to obstruct vessels within a
functional gas exchange unit with collateral perfusion, preventing the development
of high VA/Q. While similar measurements of VA/Q distribution have not been
made in patients with acute thromboemboli, most likely comparable mechanisms
are at work.
The VA/Q distributions have been measured in two patients with massive
pulmonary embolization and hemodynamic instability with pulmonary hyperten-
sion (196). Hypoxia resulted from the presence of a large shunt. Since both patients
had a low pulmonary artery occlusion pressure and a normal chest x-ray, the
etiology of the shunt is unknown. It may result from diffuse microatelectasis
secondary to alveolar duct construction (197) or from alterations in surfactant,
which have been described experimentally (198). Alternatively, it could result from
pulmonary edema similar to that seen at high altitude (137) or following glass bead
53
SMALL LARGE
COMPARTMENT COMpjlRTMENT
"db
V, 1.0 5.0
HOMOGENEOUS LUNG
0 10 5.0
PRE· EMBOLIZATiON
V,/O 1.0
Po, 104 ~ 104

ARTERIAL PO, 104
ARTERIAL Peoz 39
~
Peoz 39 39
"db
v, 10 5.0
SMALL COMPARTMENT
0 5.9
EMBOUZED
V,/O 100 0.9
ARTERIAL Paz 98
Paz 142 98
ARTERIAL Pea, 40
~
Peoz 16 40
"db
V, 5.0
LARGE COMPARTMENT
0 3.5 2.5
EMBOUZED
V,/O 0.3 2.0
ARTERIAL Po, 66
r-"'.
Po, 58 123
ARTERIAL Peoz 40
~ 33
Peo, 44
Figure 2-17. Two-compartment lung model with ventilation and blood flow matched in the
top panel. In the middle panel blood flow to the small compartment has been reduced ~y ~O%
and diverted to the large compartme.nt. This will result in the development of a high VA/Q
unit, but a minimal decrease in the VA/Q of the larger compartment (1 to 0.9). Thus, no
significant hypoxemia would ensue. In the bottom panel, flow to the larger compartment has
been decreased by 50% and diverted t<? tht; small compartment. This leads to 58% of the blood
flow distribution to a lung unit with VA/Q of 0.3 and significant hypoxemia. In each case the
P0 2 and PC02 of the mixed venous blood are assumed to remain constant at 40 and 45 mmHg,
respectively. (Reproduced with permission from reference 62.)
embolization in experimental animals (199). In both situations increased capillary

permeability resulting from endothelial damage due to the uneven application of
high pulmonary artery pressures has been the suggested mechanism of pulmonary
edema.
Patients with long-standing vascular obliteration due to recurrent thromboem-
bolism or primary pulmonary hypertension have minimal abnormalities of their
VAl Q distribution despite the marked reduction in pulmonary vascular cross-
sectional area (200) (figure 2-20). In the majority of patients studied, only 10 to
15% of the cardiac output was found perfusing shunt or low VA/Q units. How-
ever, when vasodilators were given a reduction in pulmonary vascular resistance
was inevitably associated with an increase in VA/Q inequality (201) (figure 2-21).
Therefore, a part of the reason for the minimal VA/Q inequality may be the
PRE -EMBOLIZATION POST-EMBOLIZATION
0.4 DEAD DEAD

SPACE SPACE
""I 29.
0.3 Vlnlllflli(JII
81f1f1d {Ifl'
0.2
c:
~
..J
.. 0.1
...9
Q
o
o o QI LO 10.0 o 0.01 0.1
!
().()I
iii
~ 0.4 ·OCAD DEAD
SPACE SPACE
Jrl 41.
~
~
..J 0.3
Vlnlilflli(JII
~
....>z
0.2
0.1
,,-17" Q8"
• SMI
o J....,t .rSlNtnI
......_ _r/L-.-----'-~'/>- J:..,t......-..,IL-.----.-_,I-
o 0.01 0.1 1.0 100 100.0 0 001 0.1 1.0 10.0 1000
VENTILATION-PERFUSION RATIO
Figure 2-18. The VA/Q distributions pre- and postemboliza~on !n two dogs. In the top case
embolization led predominantly to the development of high VA/Q units and a siglJifi~t fall in
the arterial paz. In the bottom case the major alteration was the creation of low VA/Q units
and the development of hypoxemia. (Modified from reference 62.)
presence of reflex vasoconstriction functioning to maximize gas exchange. Since

these patients did not show similar increases in VA/Q inequality in response to
oxygen, the stimulus is unlikely to ~e hypoxic vasoconstriction. .
Despite the mild abnormalities ofVA/Q inequality, most patients with oblitera-
tive pulmonary vascular disease have a widened A-aD0 2, and many are signifi-
cantly hypoxic. A major cause of this hypoxemia is the widened arterial-venous
oxygen difference that results from abnormal cardiac performance.
The arterial P02 represents a composite of events occurring in each gas exchang-
ing unit. In each of these units, the end capillary P02 is determined by the Fr02'
the VA/Q ratio, and the mixed venous P02 (PV0:z). With all other factors
,:onstant, the end capillary P02 will vary with the PV02, depending on i~ VA!
Q ratio. The resultant effect on the Pa02 will depend on the overall VAl Q
55
BEFORE EMBOliZATION
0.3 61.1.
·ec 0.2
'"
..J
• 0.1
o
...I
&&..
IN SHllIIT
Q
o O~~-- . .--~~,-~~--r--
o
...I o 0.001 0.01 0.1 1.0 10.0 100.0
II)
Q
Z AFTER EMBOLIZATION
C
:!: 0.2
z
o
~
C
..J
~
ffi 0.1
>
( SNIlIIT
0.1.
o 0001 0.01 0.1 1.0 10.0 100.0
VENTILA TlON- PERFUSION RAT 10

Figure 2-19. The distribution of ventilation and blood flow in a dog before and after . .
embolization with autologous clot. Following embolization, there is a marked increase in VA/Q
inequality but no increase in shunt of dead space. (Reproduced with permission from reference
62.)
distribution, with the PV02 having increasing influence as the degree of VAl Q
inequality or the amount of shunt increases. As an example, the Pa02 of patient
V J. in figure 2-20 was 58 mmHg with a PV0 2 of 24 mmHg. With the same
minimally abnormal VA/Q but a PV0 2 of 40 mmHg, her Pa02 would have been
88 mmHg. A similar influence of the Pv02 on arterial oxygen has been seen in
patients with COPD (202) and cystic fibrosis (193) with mixed venous hypoxemia
as a result of cor pulmonale.
The influence of the Pv02 becomes even more important during exercise. In
normal subjects exercise results in a widening of the arterial-venous 02 difference
since increased cardiac output by itself is unable to provide the necessary increase
in oxygen delivery (203). In patients with cor pulmonale, the cardiac output is less
able to increase with increasing metabolic demands, and the fall in Pv0 2 for any
--•.
~
1.0
V.J.
!
Dead
CIT = 1.84 I/min Space
0
'liE = 8.41 I/min 41.7%
lJ... 0.8 Baseline
"C
o
m·e 0.6
O~
cO';/;
-'-
o ~ 0.4
'-III' : :
0.2
Shunt
6.1%
0.0 0.01 0.1 1.0 10.0 100.0
Venti lotion - Perfusion Ratio

Figure 2-20. The ventilation-perfusion (VA/OJ distribution in a patient with chronic. em~olic
pulmonary hypertension. Despite marked vascular obstruction, there is only minimal VA/Q
inequality.
level of exercise is even greater. Measurements of the VA/Q distribution in patients

with a variety of chronic pulmonary disorders suggest that during exercise the
degree of VA/Q inequality remains the same or improves (159,193,204). Despite
this, the arterial P02 often falls. This fall can be explained to a great degree by
the lower Pv02 and its ability to amplify the effects of the remaining low VAl
Q units and shunt on arterial oxygenation.
In summary, this chapter has attempted to review the physiology of the normal
and abnormal pulmonary vascular bed in order to emphasize its influence on
cardiopulmonary function. Clearly, the complexity of factors influencing this
vascular network, coupled with its relative inaccessibility, has left a great number
of questions unanswered. Hopefully, continued experimentation and improvements
in technology will begin to fill in these voids in our knowledge; but the readers
should keep these present gaps in our understanding of pulmonary vascular func-
tion clearly in mind when interpretating the results of clinical investigations.
REFERENCES
1. Weibel ER. Morphometry of the Human Lung. New York: Academic Press, 1963.
l'
2. Von Hayek H. The Human Lung. New York: Hafner Press, 1960.
3. Mead Takishima T, Leith D. Stress distribution in lungs: A model of pulmonary elasticity.
J App Physiol 28: 596-608, 1970.
4. Grant BJB, Fortune J, West JB. Effect oflocal antigen inhalation and hypoxia on lobar blood
flow in allergic dogs. Am Rev Respir Dis 122: 39-46, 1980.
5. Miller WS. The Lung, 2nd edition. Springfield, Ill: Thomas, 1947.
57
1.2
ES.
QT:&2.251/mln
1.0
V£ '17.94 IImin
Dtod
BASELINE S_
0.8 37.6%
POO2a90
0.6
0.4
c:::
~ 0.2
~ Shunt
~ 0.3%
.. 0.0 0.01 0.1 1.0 10.0 100.0
•
~ 1.2
.., E.S
o Or' 2.58
ffi 1.0 v. -582
ttl NITROPRUSSIDE
Dead
~ 0.8 Space
35.7%
c:
.2
~ 06
c:
~
0.4
0.2
Shunt
~% -
0.0 0.01 0.1 1.0 10.0 100.0
Ventilation- Perfusion Ratio
Figure 2-21. The distribution of VA/Q ratios in a patient with idiopathic pulmonary
hypertension before and du~ing. intravenous nitroprusside infusion. During nitroprusside. inf~ion
there is a widening of the VA/Q distribution with an increase in the perfusion of low VA/Q
units. (Reproduced from Dantzker DR, Bower JS, Prog Resp Res 1981, 16: 264-266, with
permission.)
6. Modell HI, Beck K, Butler J. Functional aspects of canine bronchial-pulmonary vascular

communications. J App. Physiol 50: 1045-1051, 1981.
7. Cudkowicz 1. Bronchial arterial blood flow in man. Med Thorac 19: 582-597, 1962.
8. Horsfield K, Cumming G. Morphology of the bronchial tree in man. J Appl Physiol 24:
373-383, 1968.
9. Singhal S, Henderson R, Horsfield K, Harding K, Cumming G. Morphology of the human
pulmonary arterial tree. Circ Res 33: 190-197, 1973.
10. Horsfield K. Morphology of branching trees related to entrophy. Respir Physiol 29: 179-184,
1977.
11. Reid 1. Structural and functional reappraisal of the pulmonary artery system. In The Scientific
Basis of Medicine-Annual Reviews. London: Atholone Press, 1968.
12. Wagenvoort CA. Morphologic changes in intrapulmonary veins. Hum Pathol 1: 205-213,
1970.
13. Weibel ER. Morphometry of the pulmonary circulation. Prog Resp Res 5: 2-12, 1969.
14. Spencer H, Leof D. The innervation of the human lung. J Anat (London) 98: 599-608, 1964.
15. Hebb C. Motor innervation of the pulmonary blood vessels in mammals. In The Pulmonary
Circulation and Interstitial Space, Fishman AP, Hecht HH (eds). Chicago: University of
Chicago Press, 1969.
16. Hales S. Statistical essays: Containing haemastaticks. Reprinted (1964) No. 22 History of
Medicine Series, Library of'New York Academy of Medicine, Hafner NY, 1733.
17. Frank O. Die Grundform des arteriellen Pulses Erste Abhandlung. Mathematische Analyse. Z
BioI 37: 483-526, 1899.
18. Lee G de J, Dubois AB. Pulmonary capillary blood flow in man. J Clin Invest 34: 1380--1390,
1955.
19. Caro CG, Harrison GK, Mognoni P. Pressure wave transmission in the human pulmonary
circulation. Cardiovasc Res 1: 91-HlO, 1967.
20. Staub NC. Gas exchange vessels in the cat lung. Fed Proc 29: 107, 1961.
21. Sobol BJ, Bottex G, Emirgil C, Gissen H. Gaseous diffusion from alveoli to pulmonary vessels
of considerable size. Circ Res 3: 71-79, 1963.
22. Jameson AG. Gaseous diffusion from alveoli into pulmonary arteries. J Appl Physiol 19:
44~56, 1964.
23. Gil J. Morphological aspects of alveolar microcirculation. Fed Proc 37: 2462-2465, 1978.
24. Sobin SS, Tremer HM, Fung Y-C. The morphometric basis of the sheet-flow concept of the
pulmonary alveolar microcirculation in the cat. Circ Res 26: 397-414, 1970.
25. Warrell DA, Evans JW, Clarke RO, Kingsby GP, West JB. Pattern of filling in the pulmonary
capillary bed. J Appl Physiol 32: 346--356. 1972.
26. Staub NC, Schultz EL. Pulmonary capillary length in dog, cat and rabbit. Respir Physiol 5:
371-378, 1968.
27. West JB, Schneider AM, Mitchell MM. Recruitment in networks of pulmonary capillaries. J
Appl Physiol 39: 976--984, 1975.
28. Fung Y-C, Sobin SS. Theory of sheet flow in lung alveoli. J Appl Physiol 26: 472-488, 1969.
29. Fung Y-C, Sobin SS. Pulmonary alveolar sheet elasticity. Circ Res 30: 951-969, 1972.
30. Glazier JB, Hughes JMB, Maloney JE, West JB. Measurement of capillary dimensions and blood
volume in rapidly frozen lungs. J Appl Physiol 26: 65-76, 1969.
31. Wagner WW Jr, Latham LP, Gillespie MN, Gunther JP, Capen RL. Direct measurement of
pulmonary capillary transitions. Science 218: 379-381, 1982.
32. Fowler KT. The vertical gradient of perfusion in the erect human lung. J Appl Physiol 20:
1163-1172, 1965.
33. Permutt S, Caldini P, Maseri A, Palmer WH, Sasamori T, Zierler K. Recruitment versus
distention in the pulmonary vascular bed. In The Pulmonary Circulation and Interstitial Space,
AP Fishman, HH Hecht (eds). Chicago: University of Chicago Press, 1969, pp. 375-387.
34. Burton AC. On the physical equilibrium of small blood vessels. Am J PhysioI164: 319-329, 1951.
35. Zweifach BW, Intaglietta M. Mechanics of fluid movement across single capillaries in the rabbit.
Microvasc Res 1: 83-101, 1968.
36. Wagner P, Read J, McRae J. Stratified distribution of blood flow in secondary lobules of the
rat lung. J Appl Physiol 22: 1115-1123, 1967.
37. Young I, Mazzone RW, Wagner PD. Identification of functional lung unit in the dog by
gradual vascular embolization. J Appl Physiol 49: 132-141, 1980.
38. Bhattacharya J, Staub NC. Direct measurement of microvascular pressures in the isolated
perfused dog lung. Science 210: 327-328, 1980.
39. McDonald DA. Blood Flow in Arteries. Baltimore: Williams & Wilkins, 1974.
40. Whitmore RL. Rheology of the Circulation. Oxford: Pergamon Press, 1968.
41. Murray JF, Karp RB, Nadel JA. Viscosity effects on pressure-flow relations and vascular
resistance in dogs' lungs. J Appl Physiol 27: 336--341, 1969.
42. Milnor WR, Conti CR, Lewis KB, O'Rourke MF. Pulmonary arterial pulse wave velocity
and impedance in man. Circ Res 25: 637-649, 1969.
43~ WesterhofN, Sipkema P, Wlzinga G, MurgoJP, GiolmaJP. Arterial impedance. In Quantita-
tive Cardiovascular Studies, MHC Henry, DR Gross, DJ Patel (eds). Baltimore: University Park
Press, 1979, pp. 111-150.
44. VanDenBos GC, westerhof M, Randall OS. Pulse wave reflection: Can it explain the difference
between systemic and pulmonary pressure and flow wave? A study in dogs. Circ Res 51:
479-485, 1982.
45. Engelberg J, Dubois AB. Mechanics of pulmonary circulation in isolated rabbit lungs. Am J
Physiol 196: 401-414, 1959.
46. Reuben SR, Gersh BJ, SwadlingJP, Lee G de J. Measurement of pulmonary arterial distensibility
in the dog. Cardiovasc Res 4: 473-481, 1970.
59
47. Milnor WR, Bergel DN, Barginer JD. Hydraulic power associated with pulmonary blood flow
and its relations to heart rate. Circ Res 19: 467-480, 1966.
48. Skalak R, Weiner F, Morkin E, Fishman AP. The energy distribution in the pulmonary
circulation. II. experiments. Phys Med BioI 11: 437-449, 1966.
49. Pace JB. Sympathetic control of pulmonary vascular impedance in anesthetized dogs. Circ Res
29: 555-568, 1971.
50. Pace JB, Cox RH, Alvarez-Vara F, Kaareman G. Influence of sympathetic stimulation on
pulmonary hydraulic input power. Am J Physiol 222: 196--201, 1972.
51. Ingram RH, Szidon JP, Skalak R, Fishman AP. Effect of sympathetic nerve stimulation on the
pulmonary arterial tree of the isolated lobe perfused in situ. Circ Res 22: 801-815, 1968.
52. HowellJBL, Permutt S, Proctor DF, Riley RL. Effect of inflation of the lung on different parts
of pulmonary vascular bed. J Appl Physiol 16: 71-76, 1961.
53. Mead J, Whittenberger JL. Lung inflation and hemodynamics. In Handbook of Physiology,
Respiration Volume I. Washington DC: American Physiological Society, 477-486, 1964.
54. Riley RL. Effect of lung inflation upon the pulmonary vascular bed. In Pulmonary Structure
and Function, De Reuch AVS, O'Conner M (eds). London: Churchill, 1962, pp. 261-272.
55. Thomas LJ Jr, Griffo ZJ, Roos A. Effect of negative-pressure inflation of the lung on pulmonary
vascular resistance. J Appl Physiol 16: 451-456, 1961.
56. Whittenberger JL, McGregor M, Berglung E, Borst HG. Influence of state of inflation of the
lung on pulmonary vascular resistance. J Appl Physiol 15: 878-882, 1960.
57. Mead J, Whittenberger JL, Radford EP Jr. Surface tension as a factor in pulmonary volume-
pressure hysteresis. J Appl Physiol 10: 191-196, 1957.
58. Lloyd TC, Wright GW. Pulmonary vascular resistance and vascular transmural gradient. J Appl
Physiol 15: 241-245, 1960.
59. BrudermanJ, Sommers K, Hamilton WK, Tooley WH, Butler J. Effect of surface tension on
circulation in the excised lungs of dogs. J Appl Physiol 19: 707-712, 1964.
60. Pain MCF, West JB. Effect of the volume history of the isolated lung on distribution of blood
flow. J Appl Physiol 21: 1545-1550, 1966.
61. Rosenzweig DY, Hughes JMB, Glazier JB. Effect of transpulmonary and vascular pressures on
pulmonary blood volume in isolated lung. J Appl Physiol 28: 553--560, 1970.
62. West JB, Dollery CT, Naimark A. Distribution of blood flow in isolated lung; relation to
vascular and alveolar pressures. J Appl Physiol 19: 713--724, 1964.
63. Banister J, Torrance RW. The effects of tracheal pressure upon flow: Pressure relations in the
vascular bed of isolated lungs. Quart J Exp Physiol 45: 352-357, 1960.
64. Permutt S, Bromberger-Bamea B, Bane NH. Alveolar pressure, pulmonary venous pressure and
the vascular waterfall. Med Thorac 19: 239-260, 1962.
65. DeBono EF, Caro CG. Effect oflung-inflating pressure on pulmonary blood pressure and flow.
AmJ Physiol205: 1178-1186, 1963.
66. Tooker J, Huseby J, Butler J. The effect of Swan-Ganz catheter height on the wedge pressure-left
atrial pressure relationship in edema during positive pressure relationship in edema during
positive pressure ventilation. Am Rev Respir Dis 117: 721-725, 1978.
67. Hughes JMB, Glazier JB, Maloney JE, West JB. Effect of extra-alveolar vessels on distribution
of blood flow in dog lung. J Appl Physiol 25: 701-712, 1968.
68. Fung Y-C, Sobin SS. Pulmonary alveolar blood flow. Circ Res 30: 470-490, 1972.
69. Anthonisen NR, Milic-Emili J. Distribution of pulmonary perfusion in erect man. J Appl
Physiol 21: 7@-766, 1966.
70. Hoppin FG Jr, York E, Kuhl DE, Hyde RW. Distribution of pulmonary blood flow as affected
by transverse (+Gx) acceleration. J Appl Physiol 22: 465-474, 1967.
71. Permutt S. Effect of interstitial pressure of the lung on pulmonary circulation. Med Thorac 22:
118-131, 1965.
72. Hughes JMB, Glazier JB, Maloney JE, West JB. Effect of lung volume on the distribution of
pulmonary blood flow in man. Respir Physiol 4: 58-72, 1968.
73. Hughes JMB, Glazier JB, Rosenzweig DY, West JB. Factors determining the distribution of
pulmonary blood flow in patients with raised venous pressure. Clin Sci 37: 847-858, 1969.
74. Milic-Emili J, Henderson JAM, Dolovich MB, Trop D, Kaneko K. Regional distribution of
inspired gas in the lung. J Appl Physiol 21: 749-759, 1966.
75. Milic-Emili J, Mead J, Turner JM. Topography of esophageal pressure as a function of posture
in man. J Appl Physiol 19: 212-216, 1964.
76. Glazier JB, Hughes JMB, Maloney JE, West JB. Vertical gradient of alveolar size in lungs of
dogs frozen intact. J Appl Physiol 23: 694-705, 1967.
77. Ritchie BC, Schauberger G, Staub NC. Inadequacy of perivascular edema hypothesis to account
for distribution of pulmonary blood flow in lung edema. Circ Res 24: 807-814, 1969.
78. Harris P, Segel N, Bishop JM. The relation between pressure and flow in the pulmonary
circulation in normal subjects and in patients with chronic bronchitis and mitral stenosis.
Cardiovasc Res 2: 73-83, 1968.
79. Widimsky J. Pressure, flow and volume changes of the lesser circulation during pulmonary
artery occlusion in healthy subjects and patients with pulmonary hypertension. Prog Resp Res
5: 224-236, 1969.
80. Even P, Divoux P, RuffF, Caubarrere I, deVemejoul P, Brouet G. The pressure-flow relation-
ship of the pulmonary circulation in normal man and in chronic obstructive pulmonary disease.
Effects of muscular exercise. Scand J Respir Dis 77 (Suppl): 72-76, 1971.
81. Lockhart A, ZeIter M, Mensch-Dechene J, Antezana G, Paz-Zamora, Vargas E, Coudert J.
Pressure-flow-volume relationships in pulmonary circulation of normal highlanders. J Appl
Physiol 41: 449-456, 1976.
82. West JB. Regional differences in gas exchange in the lung of erect man. J Appl Physiol 17:
893-898, 1962.
83. Woolcock AJ, McRaeJ, MorrisJG, Read]. Abnormal pulmonary blood flow distribution in
bronchial asthma. Austral Ann Med 15: 196-203, 1966.
84. Henderson LL, Tauxe WN, Hyatt RE. Lung scanning in asthmatic patients with 131I_MAA.
S Med J 6: 795-804, 1967.
85. Mishkin F, Wagner HN, Jr. Regional abnormalities in pulmonary arterial blood flow during
acture asthmatic attacks. Radiology 88: 142-144, 1976.
86. Grant BJB, Levinson ML, Liss HP, Reid CM. Effect of hypoxic vascular tone on lobar
pulmonary blood flow response to local antigen challenge (abstract). Am Rev Respir Dis 125:
271, 1982.
87. Barer GR, Howard P, McCurrie JR, Shaw JW. Changes in the pulmonary circulation after
bronchial occlusion in anesthetized dogs and cats. Circ Res 25: 747-764, 1969.
88. Benumof JL. Mechanism of decreased blood flow to atelectatic lung. J Appl Physiol 46:
1047-1048, 1979.
89. Fishman AP. Vasomotor regulation of the pulmonary circulation. Ann Rev Physiol 42: 211-
220, 1980.
90. Blackmore WS, Carlens E, Bjorkman S. Effect of unilateral rebreathing of low oxygen gas
mixtures upon pulmonary blood flow in man. Surg Forum Proc 5: 691-696, 1955.
91. Ulmer S, Wenke A. Bronchospirometrische Untersuchungen zur Frage der gasspannungsabhan-
gigen Durchblutun~regulation der Alveolarkapillaren. Arch Kreislaufforsh 26: 256-270, 1957.
92. Arborelius M Jr. Kr 5 in the study of pulmonary circulation and ventilation during unilateral
hypoxia. Scand J Resp Dis 62 (Suppl): 105-108, 1966.
93. Barer GR, Howard P, Shaw JW. Stimulus-response curves for the pulmonary vascular bed to
hypoxia and hypercapnia. J Physiol (London) 211: 139-155, 1970.
94. Grant BJB, Davies EE, Jones H, Hughes JMB. Local regulation of pulmonary blood flow and
ventilation-perfusion ratios in the coati mundi. J Appl Physiol 40: 216-228, 1976.
95. Sylvester]T, Harabin AL, Peake MD, Frank RS. Vasodilator and vasoconstrictor responses to
hypoxia in isolated pig lungs. J Appl Physiol 49: 820-825, 1980.
96. Peake MD, Harabin AL, Brennan Nj, Sylvester]T. Steady state vascular responses to graded
hypoxia in isolated lungs of five species. J Appl Physiol 51: 1214-1219, 1981.
97. Fishman AP. Hypoxia and the pulmonary circulation. Circ Res 38: 221-231, 1976.
98. Wier EK. Does normoxic pulmonary vasodilation rather than hypoxic pulmonary vasoconstric-
tion account for the pulmonary pressor response to hypoxia? Lancet 1: 476-477, 1978.
99. Lloyd TC Jr. Hypoxic pulmonary vasoconstriction: Role of perivascular tissue. J Appl Physiol
25: 560--565, 1968.
100. Lloyd TC Jr. Responses to hypoxia of pulmonary arterial strips in nonaqueous baths. J Appl
Physiol 28: 566-569, 1970.
101. Tucker A, Weir EK, Reeves]T, Grover RF. Failure of histamine antagonists to prevent hypoxic
pulmonary vasoconstriction. J Appl Physiol 40: 496-500, 1976.
102. Weir EK, McMurtry IF, Tucker A, Reeves]T, Grover RF. Prostaglandin synthetase inhibitors
do not decrease hypoxic pulmonary vasoconstriction. J Appl Physiol 41: 714-718, 1976.
61
103. Unger M, Atkins M, Briscoe W A, King TKC. Potentiation of pulmonary vasoconstriction with
intermittent repeated hypoxia. J Appl Physiol 43: 662--667, 1977.
104. Miller MA, Hales CA. Stability of alveolar hypoxic vasoconstriction with intermittent hypoxia.
J Appl Physiol 49: 846-850, 1980.
105. Daly I de B, Hebb C. Pulmonary and Bronchial Vascular Systems. London: Arnold, 1966, p.
201.
106. Kato M, Staub NC. Response of small pulmonary arteries to unilobar hypoxia and hypercapnia.
Circ Res 19: 426-440, 1966.
107. Glazier JB, Murray JF. Sites of pulmonary vasomotor reactivity in the dog during alveolar
hypoxia and serotonin and histamine infusion. J Clin Invest 50: 2550-2558, 1971.
108. Hyman AL, Higashida RT, Spannhake EW, Kadowitz PJ. Pulmonary vasoconstrictor responses
to graded decreases in pre-capillary blood P02 in intact-chest cats.] Appl Physiol51: 1009-1016,
1981.
109. Durand J, Ladurie ML, Ranson-Bitker B. Effects of hypoxia and hypercapnia on the repartition
of pulmonary blood flow in supine subjects. Prog Resp Res 5: 156-165, 1969.
110. Viles PH, Shepherd]T. Evidence for a dilator action of carbon dioxide on pulmonary vessels
of the cat. Circ Res 2: 325-332, 1968.
111. Viles PH, Shepherd]T. Relationship between pH, p02 and pC02 on the pulmonary vascular
bed of the cat. Am J Physiol 215: 1170-1176, 1968.
112. Barer GR, Shaw JW. Pulmonary vasodilator and vasoconstrictor actions of carbon dioxide. J
Physiol (London) 213: 633--645, 1971.
113. Benumof JL, Mathers JM, wahrenbrock EA. Cyclic hypoxic pulmonary vasoconstriction
induced by concomitant carbon dioxide changes. J Appl Physiol 41: 466-469, 1976.
114. Tucker A, Weir EK, ReevesJT, Grover RF. Histamine H 1- and H 2-receptors in pulmonary
and systemic vasculature of the dog. J Appl Physiol 229: 1008-1013, 1975.
115. Hyman AL, Spannhake EW, Kadowitz PJ. Prostaglandins and the lung. Am Rev Respir Dis
117: 111-136, 1978.
116. Mentzner RM, Rubio R, Berne RM. Release of adenosine by hypoxic canine lung tissue and
its possible role in the pulmonary circulation. Am J Physiol 229: 1625-1631, 1975.
117. Weir EK, McLzoch J, Seavy ], Cohen JJ, Grover RF. Platelet antiserum inhibits hypoxic
pulmonary vasoconstriction in the dog. J Appl Physiol 41: 211-215, 1976.
118. Colebatch HJH, Dawes GS, Goodwin JW, Nadeau RA. The nervous control of the circulation
in the fetal and newly expanded lungs of the lamb. J Physiol 178: 544-562, 1965.
119. Ingram RH, Szidon JP, Skalak R, Fishman AP. Effects of sympathetic nerve stimulation on the
pulmonary arterial tree of the isolated lung perfused in situ. Circ Res 22: 801--815, 1968.
120. Daly I de B, Duke H, Hebb CO, WeatheralJ. Pulmonary vasomotor fibres in the sympathetic
chain and its associated ganglia in the dog. Q] Exp Physiol 34: 285-313, 1948.
121. Hyman AL, Nandiwada P, Knight DS, Kadowitz PJ. Pulmonary vasodilator response to
catecholarnines and sympathetic nerve stimulation in the cat. Circ Res 48: 407-415, 1981.
122. Daly I de B, Hebb C. Pulmonary vasomotor fibres in the cervical vagosympathetic nerve of
the dog. Q J Exp Physiol 37: 19-43, 1952.
123. Downing SE, Lee JC. Nervous control of the pulmonary circulation. Am Rev Physiol 42:
199-210, 1980.
124. Wilcox BR, Austin WG, Bender HW. Effect of hypoxia on pulmonary artery pressure of dogs.
Am J Physiol 207: 1314-1318, 1964.
125. Kazerni H, Bruecke PE, Parsons E. Role of autonomic nervous system in the hypoxic response
of the pulmonary vascular bed. Respir Physiol 15: 245-254, 1972.
126. Bayliss WM. On the local reaction of the arterial wall to changes of internal pressure.] Physiol
(London) 28: 220-231, 1902.
127. Harris P. Patent ductus arteriosus with pulmonary hypertension. Brit Heart J 17: 85-92,
1955.
128. Hyman AL. Pulmonary vasoconstriction due to nonocclusive distention of large pulmonary
arteries in the dog. Circ Res 23: 401-413, 1968.
129. Juratsch CE, ]engo ]A, Laks MM. Role of autonomic nervous system and l'ulmonary artery
receptors in production of experimental pulmonary hypertension. Chest 71 (Suppl): 265-269,
1977.
130. Kealey GP, Brody MJ. Studies on the mechanism of pulmonary vascular responses to miliary
pulmonary embolism. Circ Res 41: 807--814, 1971.
131. Grant BJB. Effect of local pulmonary blood flow control on gas exchange: Theory. J Appl
Physiol 53: 1100-1109, 1982.
132. Euler von US, Liljestrand G. Observations on the pulmonary arterial blood pressure in the cat.
Acta Physiol Scand 12: 304-320, 1946.
133. Marshall BE, Marshall C. Continuiry of response to hypoxic pulmonary vasoconstriction. J Appl
Physiol49: 189-196, 1980.
134. Marshall BE, Marshall C, BenumofJ, Saidman LJ. Hypoxic pulmonary vasoconstriction in dogs:
Effect of lung segment size and oxygen tension. J Appl Physiol 51: 1543-1551, 1981.
135. Grover RF, Hyers TM, McMurtry IF, Reeves]T. High altitude pulmonary edema. In Pulmo-
nary Edema, Fishman AP, Renkin EM (eds). Bethesda: American Physiological Sociery, 1979.
136. Reid LM. The pulmonary circulation: Remodeling in growth and disease. Am Rev Respir Dis
119: 531-546, 1979.
137. Heath DA, Williams DD. Man at High Altitudes. New York: Churchill Livingstone, 1977.
138. Hislop A, Reid 1. New fmdings in pulmonary arteries of rats with hypoxia-induced pulmonary
hypertension. Br J Exp Pathol 57: 542-554, 1976.
139. Rabinovitch M, Gamble W, Nadas AS, Mettinen OS, Reid 1. Rat pulmonary circulation after
chronic hypoxia: Hemodynamic and structural features. AmJ Physiol236: H818-H827, 1979.
140. Meyrick B, Reid 1. The effect of continued hypoxia on rat pulmonary artery circulation. Lab
Invest 38: 188-200, 1978.
141. Heath D. Ultrastructure in pulmonary hypertension. In Pulmonary Circulation in Health and
Disease, Cumming G, Bonsignore G (eds). New York: Plenum Press, 1980, pp. 233-248.
142. Williams JF, Behnke RH. The effect of pulmonary emphysema upon cardiopulmonary hemody-
namics at rest and during exercise. Ann Intern Med 60: 824-842, 1964.
143. Burrows B, Kettel LJ, Niden AH, Rabinowitz M, Diener CF. Patterns of cardiovascular
dysfunction in chronic obstructive lung disease. N Engl J Med 286: 912-918, 1972.
144. Wilson RH, Hoseth W, Dempsey M. The effects of breathing 99.6% oxygen on pulmonary
vascular resistance and cardiac output in patients with pulmonary emphysema and chronic
hypoxia. Ann Intern Med 42: 629-637, 1955.
145. Aubier M, Murciano D, Milic-Emili J. Effects of the administration of 02 on ventilation and
blood gases in patients with chronic obstructive pulmonary disease during acute respiratory
failure. Am Rev Respir Dis 122: 747-754, 1980.
146. Whitaker W. Pulmonary hypertension in congestive heart failure complicating chronic lung
disease. Quart J Med 23: 57-72, 1954.
147. Abraham A, Cole R, Green I, Hedwotth-Whitty R, Clarke S, Bishop J. Factors contributing
to the reversible pulmonary hypertension of patients with acute respiratory failure studied by
serial observations during recovery. Circ Res 24: 51-60, 1969.
148. Abraham AS, Kay JM, Cole RB, Pincock AC. A haemodynamic and pathological study of the
effect of chronic hypoxia and subsequent recov~ry on the heart and pulmonary vasculature of
the rat. Cardiovasc Res 5: 95-102, 1971.
149. Levine BE, Bigelow B, Hamstra R, Beckwitt H, Mitchell R, Nett L, Stephen T, Petry TI.
The role of long-term continuous oxygen administration in patients with chronic airway
obstruction with hypoxemia. Ann Intern Med 66: 639-650, 1967.
150. Stark RD, Finnegan P, Bishop JM. Long-term domiciliary oxygen in chronic bronchitis with
pulmonary hypertension. Brit Med J 1: 467-470, 1973.
151. Medical Research Council: Long-term domiciliary oxygen therapy in chronic hypoxic cor
pulmonale complicating chronic bronchitis and emphysema. Lancet 1: 681-685, 1981.
152. Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen therapy in hypox-
emic chronic obstructive lung disease. Ann Intern Med 93: 391-398, 1980.
153. Weitzenblum E, Loiseau A, Hitth C, Mirhom R, Rasaholinjanahary J. Course of pulmonary
hemodynamics in patients with chronic obstructive pulmonary disease. Chest 75: 656-662, 1979.
154. Kay JM. Effect of intermittent normoxia on chronic hypoxic pulmonary hypertension, right
ventricular hyperttophy and polycythemia in rats. Am Rev Respir Dis 121: 993-1001, 1980.
155. Nattie EE, Bartlett D, Johnson K. Pulmonary hypertension and right ventricular hyperttophy
caused by intermittent hypoxia and hypercapnia in the rat. Am Rev Respir Dis 118: 653-658,
1978.
156. Boysen PG, Block AJ, Wynne ]W, Hunt LA, Flick MR. Nocturnal pulmonary hypertension
in patients with chronic obstructive pulmonary disease. Chest 76: 536-542, 1976.
157. West JB. Ventilation-perfusion relationships. Am Rev Respir Dis 116: 919-943, 1977.
63
158. Wagner PD, Laravuso RB, Uhl RR, West JB. Continuous distributions of ventilation-perfu-
sion ratios in normal subjects breathing air and 100% 02. J Clin Invest 54: 54-68, 1974.
159. Wagner P, Dantzker D, Dueck D, Calusen J, West J. Ventilation-perfusion inequality in chronic
obstructive pulmonary disease. J Clin Invest 59: 203-216, 1977.
160. Wagner P, Saltzman H, West J. Measurement of continuous distributions of ventilation-
perfusion ratios: Theory. J Appl Physiol 36: 588--599, 1974.
161. Harris P, Heath D. The Human Pulmonary Circulation. New York: Churchill Livingstone,
1977, pp. 504-521.
162. Bove KE, Scott RC. The anatomy of chronic cor pulmonale secondary to intrinsic lung disease.
Prog in Cardiovasc Dis 9: 227-238, 1966.
163. Mitchell RS, Stanford RE, Silvers GW, Dart G. The right ventricle in chronic airway obstruc-
tion: A clinicopathologic study. Am Rev Respir Dis 114: 147-154, 1976.
164. Scott KWM. A pathological study of the lungs and heart in fatal and nonfatal chronic airways
obstruction. Thorax 31: 70-79, 1976.
165. Cullen JH, Kaemmerlen]T, Daoud A, Katz HL. A prospective clinical-pathological study of
the lungs and heart in chronic obstructive lung disease. Am Rev Respir Dis 102: 190-204, 1970.
166. Symchych PS. Pulmonary hypertension in cystic fibrosis. Arch Path 92: 409-414, 1971.
167. McIntyre K, Sashara A. The hemodynamic response to pulmonary embolism in patients without
prior cardiopulmonary disease. Am J Cardiol 28: 288--294, 1971.
168. Stanek V, Riedel M, Widimsky J. Hemodynamic monitoring in acute pulmonary embolism.
Bull Europ Physiopath Resp 14: 561-572, 1978.
169. McIntyre KM, Sasahara AA. Determinants of the cardiovascular responses to pulmonary embo-
lism. In Pulmonary Thromboembolism, Moser K, Stein M (eds). Chicago: Year Book, 1973,
pp. 144-159.
170. Mcintyre KM, Sasahara AA. The ratio of pulmonary arterial pressure to pulmonary vascular
obstruction. Chest 71: 692-697, 1977.
171. Alpert JS, Godtfredsen J, Ockene IS, Anas J, Dalen JE. Pulmonary hypertension secondary to
minor pulmonary embolism. Chest 73: 795-797, 1978.
172. Laros CD, Swierenga J. Temporary unilateral pulmonary artery occlusion in the pre-operative
evaluation of patients with bronchial carcinoma. Med Thorac 24: 269-283, 1967.
173. Riedel M, Stanek V, Widimsky J, Prerovsky 1. Long-term follow-up of patients with pulmo-
nary thromboembolism. Chest 81: 151-158, 1982.
174. Dantzker DR, Bower JS. Partial reversibility of chronic pulmonary hypertension caused by
pulmonary thromboembolic disease. Am Rev Respir Dis 124: 129-131, 1981.
175. Permutt S. Mechanical influences of water accumulation in the lungs. In Pulmonary Edema,
Fishman AP, Renkin EM (eds). Bethesda: American Physiological Society, 1979, pp. 175-193.
176. Gabe II, Gault JH, Ross J, Mason DT, Mills q, Schillingford]p, Braunwald E. Measurement
of instantaneous blood flow velocity and pressure in conscious man with a catheter-tip velocity
probe. Circulation 40: 603-614, 1969.
177. Even P, Sors H, Safran D, Reynaud P. Interaction between ventilation and circulation in
bronchial asthma and pulmonary emphysema. In Pulmonary Circulation in Health and Disease,
Cumming, G, Bonsignore G (eds). New York: Plenum Press, 1980, pp. 279-293.
178. Summer WR, Permutt S, Sagana K, Shoukas A, Bromberger-Bamea B. Effects of spontaneous
respiration on canine left ventricular function. Circ Res 45: 719, 1979.
179. Scharf SM, Brown R, Saunders N, Green LH. Effects of normal and loaded spontaneous
inspiration on cardiovascular function. J Appl Physiol 47: 582-590, 1979.
180. Permutt S. Some physiological aspects of asthma: Bronchomuscular contraction and airway
calibre. In Identification of Asthma (Ciba Symposium). London: Churchill, 1971, pp. 63-85.
181. Stalcup SA, Mellins RB. Mechanical forces producing pulmonary edema in acute asthma. N
Engl J Med 297: 592-596, 1977.
182. Robotham]1, Mitzner W. A model of the effects of respiration on left ventricular performance.
J Appl Physiol 46: 411-418, 1979.
183. Guyton AC, Adkins LJ. Quantitative aspects of the collapse factor in relation to venous return.
Am J Physiol 177: 523-527, 1954.
184. MakJUauan FK, Palmer WH, McGregor M. Influence of respiration on venous return in
pulmonary emphysema. Circulation 33: 8-16, 1966.
185. Junod AF. Metabolism, production and release of hormones and mediators in the lung. Am Rev
Respir Dis 112: 93-108, 1975.
186. Said SI, Mutt V. Relationship of spasmogenic and smooth muscle relaxant peptides from normal
lung to other vasoactive compounds. Nature 265: 84, 1977.
187. Hauge A. Role of histamine in hypoxic pulmonary hypertension in the rat. Circ Res 22:
371-383, 1968.
188. Said SI. Release of biologically active materials from the lung: Release induced by physical and
chemical stimuli. In Metabolic Function of the Lung, Bakhle YS, Vane JR (eds). New York:
Marcel Dekker, 1977, pp. 297-320.
189. Dunham BM, Grindlinger GA, Utsunomiya T, Krausz MM, Hechtman HB, Shepro D. Role
of prostaglandins in positive end-expiratory pressure-induced negative inotropism. Am J Physiol
241: H783-H788, 1981.
190. Fisher AB, Steinberg H, Bassett D. Energy utilization of the lung. AmJ Med 57: 437-446,1974.
191. Kay JM, Keane PM, Suyama KL, Gauthier D. Angiotension-converting enzyme activity and
evolution of pulmonary vascular disease in rats with monocrotaline pulmonary hypertension.
Thorax 37: 88-96, 1982.
192. Gillis CN, Huxtable R, Roth R. Effects of monocrotaline pretreatment of rats on removal of
5-hydroxytryptamine and noradrenaline by perfused lung. Brit J Pharmacol 64: 435-443, 1978.
193. Dantzker DR, Patten GA, Bower JS. Gas exchange at rest and during exercise in adults with
cystic fibrosis. Am Rev Respir Dis 125: 400-405, 1982.
194. Kafer ER. Respiratory function in pulmonary thromboembolic disease. AmJ Med 47: 904-915,
1969.
195. Dantzker DR, Wagner PD, Tornabene VW, Alazraki NP, West JB. Gas exchange after
pulmonary thromboembolization in dogs. Circ Res 42: 92-103, 1978.
196. D' Alonzo GE, Bower JS, Dantzker DR. Gas exchange alterations in acute massive pulmonary
embolism in humans (abstract). Chest 82: 223, 1982.
197. Nadel JA, Colebatch HJH, Olsen CR. Location and mechanism of airway constriction after
barium sulfate microembolism. J Appl Physiol19: 387-394, 1964.
198. Chernik V, Hodson WH, Greenfield LJ. Effects of chronic pulmonary artery ligation on
pulmonary mechanics and surfactant. J Appl Physiol 21: 1315-1320, 1966.
199. Ohkuda F, Nakahora K, Weidner WJ, Binder A, Staub NC. Lung fluid exchange after uneven
pulmonary artery obstruction in sheep. Circ Res 43: 152-161, 1978.
200. Dantzker DR, Bower J. Mechanisms of gas exchange abnormality in patients with chronic
obliterative pulmonary vascular disease. J Clin Invest 64: 1050-1055, 1979.
201. Dantzker DR, Bower JS. Pulmonary vascular tone improves V A/Q matching in obliterative
pulmonary hypertension. J Appl Physiol 51: 607-613, 1981.
202. Mithoefer JC, Ramirez C, Cook W. The effect of mixed venous oxygenation on arterial blood
in chronic obstructive pulmonary disease. Am Rev Respir Dis 117: 259-264, 1978.
203. Jones NL, Campbell EJM, Edwards RHT, Robertson DG. Clinical Exercise Testing. Philadel-
phia: W.B. Saunders, 1975.
204. Wagner PD, Dantzker DR, Dueck R, Depolo JI., Wasserman K, West JB. Distribution of
ventilation-perfusion ratios in patients with interstitial lung disease. Chest 69: 256--257, 1976.
3. PATHOLOGY OF PULMONARY HEART DISEASE
WAYNE E. TAYLOR, M.D.
A discussion of the pathology of pulmonary heart disease must be centered on the

pulmonary vasculature since, in the disease processes relevant to this chapter, the
heart is an innocent bystander to events taking place in the lungs. Frequently, the
cardiac changes are important and in many cases may constitute the dominant
clinical fmdings. But, the lungs, and the pulmonary vasculature in particular, are
where the most prominent pathologic changes usually reside. These changes may
be intrinsic to the lung, such as in primary pulmonary hypertension, or secondary
to abnormalities in other parts of the body, as in peripheral venous thrombosis with
pulmonary thromboembolic disease. Exceptions to this generalization might be
congenital heart disease or ischemic heart disease with chronic left ventricular
failure, in which the main problem is intrinsic to the heart. These conditions may
undoubtedly be associated with pulmonary vascular overload, right ventricular
hypertrophy, and right ventricular failure; however, these intrinsic cardiac prob-
lems are not ordinarily included in defmitions of pulmonary heart disease, or, as
it is more commonly known, cor pulmonale (1).
The view is generally accepted that congenital heart disease and acquired diseases
of the left ventricle must be excluded before the terms cor pulmonale or pulmonary
The critical review and helpful suggestions ofL. Max Buja, M.D., are gratefully acknowledged. The photographic
expertise of Dan Rodriguez and Linda Bolding and the typing efforts of Jean Redding are also recognized.
L.J. Ruhin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright © 1984 Martinus NijhofJPuhlishing. Boston/The Hague/Dor-
drecht/Lancaster.
65
66 3. Pathology of Pulmonary Heart Disease
heart disease can be applied in a proper sense (2). This approach is both customary
and convenient, but it neglects the fact that changes in the pulmonary vasculature
associated with congenital cardiac diseases may be remarkably similar, if not
identical, to changes thought to represent intrinsic pulmonary vascular disease (3).
Furthermore, for some congenital heart lesions, such as atrial septal defects, to
present as occult causes of pulmonary vascular disease is not unusual (4). For these
reasons, and for the sake of a more complete understanding of the response of the
pulmonary vasculature to increased load, a brief description of the changes in the
pulmonary vasculature that result from m:yor congenital heart lesions will be
included to the extent deemed appropriate. Otherwise, the bulk of this chapter will
concentrate on the pulmonary vascular changes of pulmonary heart disease.
One must recognize that pulmonary vascular overload and the resultant pulmo-
nary heart disease represent dynamic processes that may become clinically apparent
late in their natural course. One might divide this course into three stages. The
earliest stage is represented by a purely functional increase in pulmonary vascular
load, usually evidenced by pulmonary arterial hypertension, although in some cases
the load might be an increased volume of flow. Since the elevations of arterial
pressure are either low or intermittent, little compensatory effort is required by the
right ventricle to accommodate. Anatomically, any cardiac change is trivial, and
pulmonary vascular alteration is mild. Clinically, this stage of pulmonary heart
disease is silent. This stage could hardly be called cor pulmonale, yet it represents
the initial phase.
In the second stage of pulmonary heart disease, the increased pulmonary vascular
load provokes right ventricular hypertrophy, but this compensatory mechanism
allows the load to be handled without a reduction in output. Anatomically, the
right ventricular wall is thickened and the total right ventricular weight is in-
creased. The pulmonary vasculature reflects significant changes of hypertension.
Few clinical signs or symptoms are present, certainly none that reflects overt failure
of the right heart. Cor pulmonale is an appropriate term for this point in the
evolution of cardiac disease (5).
The third stage is characterized by the presence of right ventricular hypertrophy,
to which may be added the additional element of failure. The pulmonary load has
increased beyond the point at which hypertrophy alone can maintain output, and
cardiac dilation is now required to increase ejection volume according to the
Frank-Starling mechanism (6). If this proves inadequate, then the overt signs and
symptoms of right heart failure become clinically apparent. Anatomically, the right
ventricle is now hypertrophied and dilated, and the pulmonary vasculature contains
pronounced changes (5).
These three stages correlate quite well with the anatomic changes and reasonably
well with the clinical fmdings. Clearly, these stages represent a continuum of
change rather than discrete leaps from one to another. Additionally, reversing the
direction of change may be possible in some instances, thus providing further
evidence of the dynamic factors at work.
Also important is the examination of cor pulmonale from the standpoint of real
67
time, to borrow a term from the computer world, in which it might be divided
into acute, subacute, or chronic forms. Acute cor pulmonale occurs in a limited
number of clinical settings; for instance, sudden massive pulmonary thromboem-
bolism. In this condition right ventricular overload develops so rapidly that com-
pensatory hypertrophy cannot take place. Under the acute strain the ventricle
dilates and fails (7,8). Chronic cor pulmonale, on the other hand, is what is usually
implied when the unqualified term cor pulmonale is used. This condition arises over
long periods of time, usually months and often years. Progressive increases in
pulmonary resistance result in the compensatory changes described previously (1,2).
Interposed between these two extremes is subacute cor pulmonale, a term
initially reserved for right ventricular hypertrophy as a result of tumor emboli
(9-13), but more recently used for other conditions that may give rise to rapidly
progressive increases in pulmonary resistance for which time is sufficient for the
heart to undergo hypertrophy (14). This condition usually develops over a few
weeks or months.
In a practical sense these categorizations are somewhat arbitrary, but they may
have clinical application in sorting out the etiology of some forms of pulmonary
heart disease, and they serve to emphasize the dynamic interplay between the heart
and pulmonary vasculature.
With this introduction the remainder of the chapter will describe the salient
pathologic features of pulmonary heart disease. This discussion will include a brief
review of the normal pulmonary vasculature, alterations due to aging and other
influences, a general discussion of the changes caused by pulmonary arterial hyper-
tension, and how these changes may be graded and categorized by etiology. A brief
description of other organs affected by cor pulmonale and a discussion of the use
of the open lung biopsy in evaluating pulmonary heart disease will complete this
chapter.
ANATOMY OF THE PULMONARY VASCULATURE
Pulmonary arteries
In general the course of the pulmonary arteries parallels the dichotomous branching
pattern of the bronchial tree, so that each bronchial, or bronchiolar, airway is
accompanied by at least one and sometimes two or more arterial branches. This
pattern is maintained from the level of the main right and left pulmonary arteries
and the mainstem brochi, to the point at which the respiratory bronchioles and
pulmonary arterioles terminate into their respective functional units, alveolar ducts
or sacs, and alveolar capillaries (15,16). However, not all branches of the pulmonary
artery remain adjacent to the wall of a bronchus, there being many so-called
supernumerary branches that diverge at right angles from the bronchovascular area
to enter the pulmonary parenchyma directly, that is, without an accompanying
bronchial airway. While these supernumerary branches may arise at any point along
the course of the pulmonary artery from the proximal part of the pulmonary
segment outward, the bulk arise within the last several centimeters of the bron-
chovascular tree. The supernumerary branches are numerous and, in fact, may
outnumber the conventional branches of the pulmonary artery (17,18). This
branching pattern may explain why the diameter of the conventional artery nar-
rows more rapidly than does the diameter of its companion bronchus as they reach
the peripheral parts of the lung.
Besides being classified as conventional or supernumerary, the pulmonary arter-
ies may be classified by their size, or histologic structure, as proposed by Brenner
in 1935 (15). If classified by the structure of their middle layer or media, arteries
may be either elastic or muscular. The elastic arteries are large and usually have
an external diameter of 1 mm or greater if the poorly demarcated adventitia is
neglected, and the measurement is made at its widest point from one external elastic
lamina to the opposite. Muscular arteries range from 0.1 mm to 1.0 mm in external
diameter, and those with a diameter of less than 0.3 are considered small muscular
arteries. The arteries smaller than 0.1 mm in diameter are usually termed arterioles
although they differ considerably in structure from the arterioles of the systemic
circulation, as will be described later. Histologically, the large or elastic arteries
consist of three layers: an outer adventitia of fibrous tissue with occasional small
blood vessels and nerves, a distinct media composed of layers of elastic tissue
arranged in laminae, and a thin intima formed by a single layer of endothelial cells
applied directly to an internal elastic lamina. In vessels of 5 mm to 6 mm in
diameter, the media may contain 16 to 20 elastic laminae, whereas vessels of 1 mm
in diameter rarely have more than 3 or 4 laminae. Smooth muscle is evident in
the media, but compared to the elastic tissue it is not prominent.
Muscular arteries are composed of the same layers as the elastic arteries and differ
mainly by the prominence of smooth muscle in the media. Elastic tissue is present,
but it is more conspicuous at the inner and outer zones of the media, where it forms
discrete internal and external elastic laminae. The intima consists of a single layer
of endothelium applied to the internal elastic lamina, and the adventitia is a layer
of fibrous tissue, often fairly thick but still poorly segregated from other fibrous
tissue present in the bronchovascular structure.
A third category of artery, the arteriole, is much more difficult to defme
histologically. These small vessels consist of tubes of endothelium with a spirally
wound elastic fibril that is continuous with the external elastic lamina of its parent
muscular artery, and occasional eccentrically placed smooth muscle cells that are
rarely apparent in vessels less than 0.08 mm to 0.06 mm in diameter. No well-
defmed circular band of smooth muscle is characteristic of the systemic arterioles.
The recognition of these vessels as arterioles depends in large part on measuring
their diameter or demonstrating that they arise from larger muscularized arteries.
While Brenner's classification of the pulmonary arteries has formed the basis of
many studies of the pulmonary vascular disorders, evidence suggests that it is too
simple. Some have pointed out that the vessels between 1.0 mm and 0.5 mm are
generally "transitional" arteries and have a medial structure intermediate between
that of the typical elastic and muscular arteries (3,17). Likewise, the transition from
muscular arteries to arterioles is not clear-cut. While unusual, tiny vessels with a
69
diameter of 0.025 mm may on occasion have a muscularized wall (19). The age
of the patient and whether the artery is a conventional or a supernumerary vessel
may also be important in determining whether a muscular media will be present
(17). One suggestion to clarify the nomenclature of these vessels is to call all of
these precapillary vessels arteries, with the additional comment that they are either
muscularized, incompletely muscularized, or nonmuscularized (17). Nevertheless,
the divisions originally proposed by Brenner have been used by many authorities
in the field (3,20-23), are strongly entrenched in the literature, and can still be
useful subcategorizations of the pulmonary arterial tree.
Pulmonary capillaries
The pulmonary alveolar capillaries constitute the business end of the pulmonary
vascular tree, but are relatively simple compared with the other pulmonary vascular
structures. They consist of a single layer of endothelium supported by a basement
membrane. Their lumens are somewhat larger than the diameter of a red blood cell
(seven microns), but they can be distended to several times that size (15).
Pulmonary veins
The pulmonary venules form the anatomic counterpart of the pulmonary arteriole,
and up to a diameter of 0.15 mm to 0.25 mm they possess the same histologic
structure. Unless they are located within a lobular septum, or serial sections reveal
they are connected to larger more clearly defined veins, they cannot be reliably
distinguished from pulmonary arterioles. In a general way the vessels in the
periphery of a pulmonary acinus are more likely to be venules, whereas similar
vessels of the same size in the central part of the acinus are more likely to be
arterioles. Veins larger than 0.25 mm gradually accumulate smooth muscle in the
media, although never to the extent that an artery of the same caliber does nor as
organized to the extent seen in arteries. More connective tissue is seen in the wall
of the vein than in its counterpart artery, and the elastic fibers are more irregularly
placed. Veins from 1 mm to 5 mm in diameter possess an internal elastic lamina,
a relatively thin media with scattered bundles of smooth muscle, and a thick
adventitia of fibrous tissue. Based on histology alone, telling these larger veins from
their arterial counterparts is usually no problem (15).
Bronchial vessels
The bronchial arteries arise from the systemic circulation and course as small vessels
that run within the outer bronchial wall. Histologically, they may be distinguished
from pulmonary arteries by their smaller size, relatively thick muscular media, and
possession of a single, well developed internal elastic lamina. These vessels extend
to the level of the bronchioles, where they are no longer apparent in histologic
sections. Capillary networks fed by the bronchial arteries supply the entire wall
of the bronchus and its supporting tissues, and even form collateral connections
with capillaries of adjacent alveoli (24). The capillary network drains into a plexus
of bronchial veins which anastomose freely with the pulmonary veins, thereby
effectively delivering the blood from the bronchial circulation back to the left
heart. Anastomoses between the bronchial and pulmonary arterial systems exist, but
are difficult to demonstrate except by use of specialized techniques since they
usually constitute a small component of the vascular tree (15,25).
Pulmonary lymphatics
Lymphatic vessels consist of thin-walled structures lined by endothelium, but
without a basement membrane. They parallel the blood vascular system, except that
they extend only as far as the alveolar ducts where they apparently end directly
in the interstitium. They are ordinarily inconspicuous, unless distended by fluid.
They form rich plexuses in the bronchi, interlobular septa, and pleura (15).
CHANGES IN THE PULMONARY VASCULATURE UNRELATED TO

ABNORMAL PULMONARY VASCULAR LOAD
Two major changes are seen in the pulmonary arterial tree with aging. The first
of these changes occurs following birth and closure of the ductus arteriosus and
foramen ovale. As the pressure in the pulmonary circulation falls, the thick muscu-
lar media characteristic of the fetal arteries becomes less apparent, and by six to
eighteen months resembles that of the adult. An appreciation of these normal
physiologic variations is critical in interpretation of biopsy material from infants
and young children; otherwise, these changes might be mistaken for those of
pulmonary hypertension. For a more detailed discussion of these changes, the reader
is referred to other references (17,23,26-28).
A second change with age is seen at the opposite end of life. Brenner (15)
described increased sclerosis occurring primarily in the elastic arteries after age 40.
Others, concerned mainly with the muscular arteries, have found that the overall
diameter of these arteries increases in thickness and exhibits greater fluctuation in
individual diameters when aged lungs are compared to young lungs. This change
is especially prominent after age 65 (29). The increased thickness is accounted for
by intimal fibrosis since the media is only minimally thickened (30). The media
is clearly affected by aging, but the main alteration is an increase in collagen and
hyalin, which replaces the smooth muscle and elastin rather than adds to it, and
any thickening is usually much less than that seen with medial hypertrophy due
to pulmonary hypertension (29).
Other factors that seem to affect pulmonary arteries include lung weight and
cigarette smoking. One study has shown that increased wall thickness occurs in
some lungs with increased weights. Because the heavier lungs were also from older
individuals, this finding may represent an aging phenomenon. The lungs were
heavier probably because of edema, with the swelling accounting for some of the
thickening (31). Experimentally, edema produced by increased venous pressure is
known to cause an increase in arterial wall thickness (32). A number of studies have
found thickening and fibrosis of the muscular arteries, with increased numbers of
muscularized arterioles, increased amounts of medial smooth muscle, and increased
71
intimal sclerosis in the lungs of smokers when compared to control lungs from
nonsmokers (33,34). In some cases the degree of vascular change parallels the degree
of small airways disease and obstructive emphysema, suggesting that the major
determinant of the vascular change may be parenchymal disease rather than some
direct effect of the smoke (35). Clarification of this point will have to await further
investigation.
Changes are commonly found in the pulmonary arteries of individuals with
bronchial carcinoma. Both medial hypertrophy and severe intimal fibrosis may be
found in the lung with the tumor. These fmdings may be even more striking in
the lobe with the tumor, but they are not seen in the opposite lung. The cause is
not known with certainty, but it appears to be a local disturbance and may result
from bronchial artery Ipulmonary artery shunts or from the effects of thrombi,
which are increased in cases of lung cancer (36).
Alterations in the pulmonary venous system also occur as a result of aging. The
main feature is increased intimal fibrosis, which becomes evident as early as 20 years
of age (15,37).
GRADING THE RESPONSE OF THE PULMONARY VASCULATURE TO

INCREASED PRESSURE OR FLOW
While differences may exist in the response of the pulmonary vasculature to
increased pressure or flow depending on the underlying etiology, the similarities
of response are more impressive than the differences. In general, there is a graded
response of the arterial bed which correlates quite well with the severity and
duration of increased pulmonary load. Although many of these changes had been
recognized for a long time, they were organized by Heath and Edwards (38) in
1957 into a grading system of pulmonary hypertension based solely on a study of
the muscular arteries and the arterioles. This system was devised initially for
evaluation of cases of congenital heart disease, but Heath and Edwards noted that
it appeared to be applicable to cases of primary pulmonary hypertension as well.
Indeed, this system has become the most widespread method for grading the
changes of pulmonary hypertension. A brief summary of this system will serve two
purposes: to familiarize the reader with the grading and to introduce the rather
stereotyped responses of the muscular pulmonary arteries and arterioles to increased
pulmonary vascular load.
Grade one
Two main fmdings are characteristic of this grade: the extension of smooth muscle
cells into arterioles that normally do not have a visible muscle coat and increased
thickness of the medial smooth muscle in the small muscular arteries. As a response
to pulmonary hypertension, even arterioles with a diameter of 0.03 mm may possess
a distinct and completely muscularized media (figure 3-1). Because these vessels
now look different from the pulmonary veins, and because they are more promi-
nent, at first glance the lung appears to have increased numbers of arterioles (21,39).
Those arteries that are normally muscularized exhibit hypertrophy of the medial
Figure 3-1. Mild pulmonary hypertension. This pulmonary arteriole with a diameter of
approximately 0.06 rom possesses a distinct medial layer of circumferential smooth muscle.
Extension of smooth muscle into these normally nonmuscularized vessels represents an early
response to increased pulmonary artery pressure. This is a Heath and Edwards Grade 1 change.
(Elastic stain; bar length indicates 0.04 rom.)
smooth muscle (figure 3-2) although considerable variation in wall thickness may
exist from one artery to another. There are no intimal changes in this stage, but
to see some degree of increased adventitial fibrous tissue in both the arterioles and
muscular arteries is not uncommon.
Grade two
In addition to the changes of medial hypertrophy found in the previous grade, there
are now intimal changes seen in the small muscular arteries (less than 0.3 mm in
diameter) and in the arterioles. Proliferations of plump cells resembling endothelial
cells are prominent in these vessels and appear to occlude the lumens in some vessels
(figures 3-3 and 3-4). As an average, the degree of muscular hypertrophy is greater
than that seen in grade 1, although there is substantial variation from one vessel
to another.
Grade three
At this level of injury, three main changes separate this phase from grade 2. First,
the character of the intimal proliferation changes from a predominantly cellular
composition to one of less cellular fibrous tissue, and fmally to one of acellular
73
Figure 3-2. Mild pulmonary hypertension. The small muscular pulmonary artery has a diameter
of !-'pproximately 0.26 mm. The muscular media is substantially thickened, and represents over
30% of the external diameter of the artery. This percentage is calculated by multiplying the
average medial thickness (as determined by the distance between the inner and outer elastic
laminae) by two, dividing the product by the average external diameter of the artery
(determined by the distance from one external elastic lamina to the opposite external lamina).
and expressing the result as a percentage. In an uninjected lung greater than 10% medial
thickness is abnormal (58). No intimal change is present. This is a Heath and Edwards grade 1
change. (Elastic stain; bar length indicates 0.1 mm.)
fibrous tissue. Also, some increase in elastic material is often evident in the in-
tima, and the internal elastic lamina may be split in regions. A second change is
the extension of the proliferation of fibrous intimal tissue into the larger muscu-
lar arteries (0.3 mm to 0.5 mm in diameter). These masses of tissue may be either
eccentric or concentric, and in some areas may be florid enough to occlude
vessels or to leave only a pinpoint lumen. Some of the concentric proliferations
of intimal fibrous tissue resemble an "onionskin" (figures 3-5 and 3-6). At first
the proliferations are focal, but as the disease progresses they become widespread.
The third and fmal change represents a continuation of medial hypertrophy, so
that in this grade the media attains its greatest relative thickness, in some cases
being up to 30% of the external diameter of the vessel. Further increases in
pressure may produce alterations in the vessel wall (as described in grades 4
through 6), but will not produce any further degree of medial hypertrophy.
Variation in extent of medial and intimal changes is still found from vessel to
vessel, with some vessels exhibiting grade 1 lesions, and some grade 2, as well as
the more severe grade 3 lesions. As the disease advances some vessels may begin
Figure 3-3. Mild pulmonary hypertension. This pulmonary arteriole (diameter approximately
0.06 rom) has not only a distinct muscular media, but there is also a cellular intimal
proliferation that compromises the lumen, leaving a faintly apparent opening. This is a Heath
and Edwards grade 2 change. (Elastic stain; bar length indicates 0.04 rom.)
Figure 3-4. Mild pulmonary hypertension. The small muscular artery in this illustration has an
external diameter of approximately 0.28 rom and a medial thickness of 30%. Also present is a
subsrantial proliferation of cellular intimal tissue. Note the perpendicular arrangement of intimal
nuclei relative to the elastic lamina. This is a Heath and Edwards grade 2 change. (Elastic stain;
bar length indicates 0.1 rom.)
Figure 3-5. Concentric laminar intimal fibrosis. The pulmonary artery (0.14 mm diameter)
depicted here has a distinct muscular media, but the most prominent change is a concentric
proliferation of cellular intimal tissue, forming an "onionskin" pattern. This is a relatively early
lesion since there are still a large number of nuclei visible in the intima. This change is
characteristic of Heath and Edwards grade 3. (Elastic stain; bar length indicates 0.04 mm.)
Figure 3-6. Concentric laminar intimal fibrosis. The small muscular artery in this
photomicrograph demonstrates essentially the same lesion as depicted in figure 3-5, except that
the intimal proliferation is much less cellular and much more fibrotic. Sometimes, substantial
amounts of elastic tissue are present in the older lesions. This is still a grade 3 change in the
Heath and Edwards system. (Elastic stain; bar length indicates 0.1 mm.)
to show generalized dilation, which should not be confused with the more spe-
cific and focal "dilatation lesions" to be found in more advanced grades (that is,
grades 4 to 6).
Grade four
Beginning with this grade proliferation is no longer the major reaction to in-
creased pulmonary load. The main new features of this grade and the higher
grades are specific "dilatation lesions." The best known of these is the plexiform
lesion, which is the hallmark of grade 4 and represents a focal change within a
dilated thin-walled arteriole or muscular artery. It consists of a saclike dilated
wall of the parent vessel and a knot of tortuous small channels and intervening
stroma resembling a glomerular tuft that fill the sac (figures 3-7 and 3-8). Proxi-
mal to the plexiform lesion, the artery has the other changes of pulmonary
hypertension, such as medial hypertrophy, fibrous intimal proliferation, and gen-
eralized dilation. At or distal to the lesion is frequently a thrombus, which may
eventually organize and be incorporated into the plexiform lesion. Elsewhere,
the arteries exhibit changes described before, including proliferative intimal le-
sions and muscular medial hypertrophy.
Grade five
In addition to the plexiform lesion, three other specific dilatation lesions were
described by Heath and Edwards as characteristic of grade S. These include the
angiomatoid lesions, cavernous lesions, and the transformation of branches of
hypertrophied muscular arteries into dilated vessels resembling veins. All of these
lesions and the plexiform lesion give rise to thin-walled vessels resembling sinusoids
that deliver blood to the alveolar capillaries. The presence of these congested
structures makes the lung appear extremely vascular. More detailed descriptions of
these lesions are presented in previous publications (3,38,40-42). Other changes
representative of this grade are the appearance of medial fibrosis of the muscular
arteries and hemosiderin deposits within alveolar macrophages. The hemosiderosis
is rarely as prominent as that seen in chronic passive pulmonary congestion due
to left heart failure.
Grade six
Besides the cumulative lesions of the previous grades, the hallmark of grade 6 is
the presence of focal fibrinoid necrosis of muscular pulmonary arteries. If the
necrosis is recent, the smudgy or glassy red appearance of fibrinoid in hematoxylin
and eosin-stained sections can be seen replacing part of all of the medial muscle.
An acute inflammatory infiltrate of neutrophils and sometimes eosinophils gives
this lesion a striking resemblance to the arteritis of polyarteritis nodosa (figure 3-9).
Later lesions show less inflammatory infiltrate and often have superimposed
thrombi. Organization of these thrombi with capillaries and fibrous tissue may
produce structures resembling the plexiform lesion. Fibrinoid necrosis may occur
Figure 3-7. Plexiform lesions. A cross section through this pulmonary artery (0.18 mm
diameter) shows the complex pattern of vascular channels within a dilated vessel that is typical
of the plexiform lesion of Heath and Edwards grade 4. (Elastic stain; bar length indicates 0.04
mm.)
Figure ~. Plexiform lesion. This is a lower magnification of the same type lesion depicted in
figure 3-7. except that the parent artery is more clearly visible. (Elastic stain; bar length
indicates 0.1 mm.)
Figure 3-9. Fibrinoid necrosis. The entire wall of this muscularized pulmonary artery, cut at a
branch point, is involved by fibrinoid necrosis. An inflammatory infiltrate containing neutrophils
indicates that this lesion is probably recent. Older lesions often have less inflammatory infiltrate
and demonstrate some evidence of organization by granulation tissue or fibrosis. This is a grade
6 lesion by the Heath and Edwards system. (Hematoxylin and eosin stain; bar length indicates
0.1 mm.)
in the presence of variable mixtures of other lesions indicative of any of the

previously described grades.
CRITIQUE OF THE HEATH AND EDWARDS GRADING SYSTEM

Although the Heath and Edwards grading system represented a major step forward
in the understanding of pulmonary vascular diseases, it has not been accepted
without criticism.
One objection raised has been to the implications of the grading system with
regard to severity and prognosis. Generally, a reasonable correlation exists between
grades 1, 2, and 3 and the clinical severity and duration of pulmonary hypertension.
Furthermore, evidence suggests that the lesions found in these grades may be partly
or completely reversible, providing that the underlying problem can be alleviated
(4~5). Grades 4 through 6, on the other hand, seem to represent "fixed" disease,
and once established will persist even if the underlying pulmonary hypertension
is controlled (46). However, there is little correlation of individual grades 4
through 6 with progression of disease or with prognosis since the presence of any
of the three advanced grades is usually associated with severe disease and a poor
79
prognosis. A suggestion was made that grades 4 through 6 be combined into a single
grade 4 to simplify the grading system and remove the implication that grade 6
is worse than grade 4 (47). This approach has been used in some pathological
evaluations and correlates well with the way primary physicians use the system to
assess the feasibility of surgery in congenital heart diseases (48).
An additional problem with grade 6 is that a considerable number of studies now
indicate that fibrinoid necrosis may precede the formation of the plexiform lesion.
In essence, it is thought that extreme elevations in pulmonary artery pressure,
perhaps even transiently, may result in necrosis, which then provides a nidus for
thrombus formation. Organization of this area results in a plexiform mass of
capillary-sized vessels within a dilated artery. That the plexiform lesion should be
confused with organizing thrombi is then easily understood since both may have
a similar origin (3,49-51).
Another shortcoming of the Heath and Edwards grading system has emerged
because of the recognition of slightly different patterns of pulmonary vascular
lesions that are dependent on the underlying cause. This has proven to be true with
regard to congenital heart diseases (52-54) and with some other forms of pulmo-
nary heart disease (55). Heath and Edwards recognized certain differences between
patients with atrial septal defects and other congenital shunts (38). Even more
critical may be the recognition that a grading system applicable to congenital heart
disease, and perhaps primary pulmonary hypertension, may not be applicable to
other types of pulmonary heart disease. For instance, the Heath and Edwards
grading system takes into account changes that occur only in the arterial circulation.
Complete evaluation of the lung lesions of pulmonary heart disease requires
examination of veins, lymphatics, and parenchyma, not just arteries. This point
should become even clearer during the subsequent section on classification of
pulmonary vascular disease.
Finally, the Heath and Edwards system is qualitative and depends on a relative
assessment of medial hypertrophy as well as on the presence or absence of certain
features (such as plexiform lesions) to determine grade, rather than some more
quantitative method. The hazards of attempting to quantify medial hypertrophy
in relative terms should be obvious. Assuming one minimizes any artifacts due to
the preparation of the tissue, one must still make a subjective judgment about the
amount of medial smooth muscle relative to what is considered "normal." Like-
wise, the presence or absence of a given lesion may be a relatively crude measure,
and in some cases may be misleading. For example, mild intimal fibrosis in a few
pulmonary arteries does not carry the same prognostic significance as does more
severe and widespread intimal fibrosis, yet it may be accorded the same grade (53).
One might also question whether the presence of a single plexiform lesion in a lung
biopsy is sufficient to render a grade 4 categorization and to deny the possible
benefit of surgery to some cases of congenital heart disease.
A number of attempts to grade pulmonary vascular disease by more objective
methods have been tried. Brenner (15) measured the medial thickness of arteries
relative to their external diameter, excluding adventitia, but his measurements may
be falsely high since he used uninjected and probably underinflated lungs from
autopsy. More consistent data had been obtained by either injecting the lungs
through the pulmonary artery at constant pressure (31,56-59) or by simply assuring
that the lungs are adequately inflated before they are fixed (table 3-1) (52,58,60,61).
Not only can studies of this type provide information about the amount of smooth
muscle, they can also provide some idea of the overall number of arterioles in a
unit volume of lung.
Rather unexpectedly, a significant decrease in the number of arterioles has been
found in some types of pulmonary vascular disease, especially that associated with
certain congenital heart diseases (56,57,61). Needless to say, there is no part of the
Heath and Edwards grading system that can take arterial density into account, and
a modified grading system has been proposed. The modification is essentially a
refmement and expansion of grade 1, with an additional category for number of
arterioles. In this system grade A is characterized by abnormal extension of muscle
into peripheral arterioles, grade B by increased thickness of the media, and grade
C by reduction in number of arterioles per unit volume oflung (58,61). The main
use of this grading system has been the preoperative evaluation of congenital
cardiac shunts for possible surgical correction; it has not found great use in other
pulmonary vascular diseases. In fact, some recent studies question the role of
decreased numbers of arterioles in producing pulmonary hypertension with certain
congenital heart diseases (62,63). Nevertheless, the increased amount of objective
information obtained by these kinds of methods might prove of value in evaluating
other pulmonary vascular diseases and should be encouraged.
The problems discussed above have led at least one authority to recommend
abandoning the Heath and Edwards system altogether (47). To do so may be
somewhat unrealistic since the system is so strongly entrenched both in the litera-
ture and in common practice. The recommendation likewise fails to consider that
Table 3-1. Medial thickness of normal pulmonary arteries
% medial Method of
Study Size vessels thickness' preparation
Heath & Best (21) 0.1-0.3 mm 2.8-3.1% Inflated with

formalin
Wagenvoort (59) Less than 0.1 mm 4.64 ± 1.08% Barium injection
(15-20 mm Hg
pressure)
Warnock and 0.1-2.0 mm 2.4 ± 0.6% Barium injection (100
Kunzmann (31) em water pressure)
Rabinovitch, et aI. 0.05-0.1 mm 8'Yc (Newborn) Barium injection (100
(58) 4% (4 mo.-adult) em water pressure)
Rabinovitch, et al. 0.05-0.1 mm 20% (Newborn) Biopsy specimen
(58) 10% (4 mo.-adult) (uninjected)
1Percent medial thickness is calculated by multiplying the average medial thickness (distance between inner and
outer elastic laminae) by two, dividing the product by the arterial diameter (distance from one outer elastic lamina
to the opposite one), and multiplying by 100 to convert the answer to a percentage fraction. If the artery is
sectioned tangentially, then the minimum thickness and shortest diameter should be used in these calculations.
81
the grading system has usefulness in at least several areas, including primary
pulmonary hypertension. The objections do serve to emphasize the fact that any
grading system should be only one aspect of an overall appraisal of the severity,
prognosis, and etiology of any given example of pulmonary vascular disease.
THE CLASSIFICATION OF PULMONARY VASCULAR DISEASE

Pulmonary vascular diseases have been classified by a number of different methods
over the past several decades, a pattern reflecting the overlap of pathologic features
from one category to another, the difficulty of using clinical fmdings to separate
categories, and our incomplete understanding of the pathogenesis of pulmonary
vascular diseases. Some have sought to separate pulmonary vascular diseases based
mainly on pathologic changes (42,52,55), and others have used mainly clinical
associations (64). The majority of classifications have used some combination of
pathologic and clinical findings (20,22,41,42,46). No single classification has seemed
to satisfy the needs of all concerned, possibly because some are of more use to the
primary physician and others to the pathologist. For example, the pathologist can
use the information obtained from a lung biopsy to determine whether the patient
has plexogenic pulmonary arteriopathy, chronic pulmonary thromboembolic hy-
pertension, or pulmonary veno-occlusive disease, three conditions that may be
impossible to separate purely by clinical means. On the other hand, ventricular
septal defects and primary pulmonary hypertension are clinically separable diseases,
yet both are causes of plexogenic pulmonary arteriopathy and appear histologically
indistinguishable to the pathologist limited to examination of a lung biopsy. One
can reasonably expect both primary physician and pathologist to be familiar with
more than just one side of the problem in order to effectively and accurately
evaluate any given case. In this chapt~r, however, emphasis will be given to
identification of categories on the basis of pathologic changes identifiable in lung
biopsy specimens or tissue obtained by autopsy.
Plexogenic pulm.onary arteripathy

This category of pulmonary vascular disease represents a spectrum of pathologic
changes, which are fundamentally those described by Heath and Edwards and
others, in association with congenital heart diseases and "primary" pulmonary
hypertension (4,38,53,65-68). All the changes previously described under grading
of pulmonary vascular disease may occur in this class, including medial hypertro-
phy, intimal fibrosis, and dilatational lesions, but the two distinctive features that
separate this group from others are the plexiform lesion and concentric laminar
intimal fibrosis. This latter change was described by Heath and Edwards in their
grade 3 of pulmonary hypertension, but it was not considered a discrete lesion until
later (68). It consists of a focal proliferation of intimal cells, collagen, and elastic
fibers that form an onionskin pattern in muscular pulmonary arteries.
This class is sometimes referred to as vasoconstrictive pulmonary hypertension
since in the early stages vasoconstriction may account for many of the lesions. If the
cause of vasoconstriction can be removed, the medial hypertrophy and cellular

intimal proliferation may regress (43-45). However, the appearance of plexiform
lesions usually denotes irreversibility with severe and progressive disease. By defIni-
tion, this category of pulmonary vascular disease can be recognized with certainty
only after the appearance of rather specific markers: plexiform and onionskin lesions.
An exception is young children with congenital heart defects and left-to-right
shunts, in which plexiform lesions and concentric laminar intimal fibrosis are seen
only rarely, with the most dominant change being severe medial hypertrophy
(46,61).
Other clinical associations with plexogenic pulmonary arteriopathy include the
pulmonary hypertension that occurs with some cases of portal hypertension, due
to cirrhosis or portal vein stenosis (69-74), and the pulmonary hypertension that
follows ingestion of certain drugs (46,75).
Chronic embolic pulmonary hypertension

In the United States the most common cause of embolic pulmonary hypertension
is probably chronic thromboembolism involving small arteries and arterioles. This
condition is, however, an infrequent and occult subset of pulmonary thromboem-
bolic disease that is not to be confused either clinically or pathologically with the
disease produced by embolization of medium- or large-sized thrombi (7,76,77).
Chronic thromboembolism involving small arteries or arterioles is often clinically
silent and frequently escapes detection until the signs and symptoms of cor pul-
monale or right heart failure become evident (3). The characteristic histologic
changes are the residue of organized and partly recanalized thrombi, which are
eccentric plaques of intimal fibrosis, and fibrous bands that form septa subdividing
the arterial lumens (figures 3-10 to 3-12) (78). Additional features include the
nonspecific findings due to vascular overload, such as medial hypertrophy and
proliferative intimal changes. In general, the medial changes are much less promi-
nent than the changes seen in plexogenic pulmonary arteriopathy, pulmonary
venous hypertension (see the following section), or pulmonary hypertension due
to chronic hypoxia (see the section discussing hypoxic pulmonary hypertension).
Because the hallmark intimal lesions may be focal, it is easy to underestimate the
degree of pulmonary vascular impairment that may be present. When faced with
evidence of increased pulmonary resistance, and lacking the histologic changes of
other forms of pulmonary hypertension, the presence of even a few clear-cut
eccentric plaques and intravascular fibrous septa should suggest thromboembolic
pulmonary hypertension (3,65,66,78,79).
One should also include as a cause of embolic pulmonary hypertension any other
material-foreign, intrinsic, or parasitic-that may be embolized chronically and
result in the occlusion of small arteries or arterioles. In the Far East, schistosomiasis
accounts for many cases of pulmonary heart disease and may be the most significant
cause of embolic lung disease worldwide (22,42,80). Other causes include emboli
of various tissues, such as trophoblastic cells (81), and malignant tumors (9-13),
which may on occasion produce subacute cor pulmonale (figures 3-13 and 3-14).
Detection of emboli of marrow or fat in autopsy or lung biopsy specimens is not
Figure 3-10. Organizing pulmonary thromboembolus. A thromboembolus within a small
muscular pulmonary artery is undergoing organization. Elongated nuclei within the thrombotic
mass represent macrophages. fibroblasts. and endothelial cells. Notice that the thrombus occupies
an eccentric position in the vessel. the result of partial lysis and organization. Compare the
location of this lesion to the eccentric fibrous plaque in figure 3-11. (Hematoxylin and eosin
stain; bar length indicates 0.1 mm.)
Figure 3-11. Eccentric fibrous plaque. A larger muscular pulmonary artery contains an eccentric
fibrous plaque. the residual of a completely organized pulmonary thromboembolus. In other
areas there were more recent thromboemboli in varying stages of organization. Many
thromboemboli undergo complete lysis. leaving behind no evidence of their presence. (Elastic
Figure 3-12. Intraluminal fibrous septa. The small muscular pulmonary artery contains fibrous
bands that bridge the lumen and are characteristic of organized thromboemboli. In some
examples only a single fibrous strand is apparent; in others the proliferation may be so complex
that it may be mistaken for the plexiform lesion. (Hematoxylin and eosin stain; bar length
indicates 0.1 mm.)
Figure 3-13. Syncytial trophoblast. Within a small pulmonary vessel is a multinucleated

syncytial trophoblastic cell. Numerous vessels were filled with these recently embolized cells.,
This specimen came from a young woman who died of "respiratory failure" following delivery
of a term infant. (Hemotoxylin and eosin stain; bar length indicates 0.04 mm.)
85
Figure 3-14. Trophoblastic embolus, older lesion. This photomicrograph is from the same case
illustrated in figure 3-13. The lesion in the center is a dilated and inapparent arteriole filled
with trophoblasts and thrombotic material. This lesion has been present for some time, probably
several days or a week, proving that embolization occurred prior to delivery. The right
ventricle was dilated, but unfortunately, the weight was not determined. This case is an example
of "acute" cor pulmonale. If right ventricular hypertrophy had been present, it would be an
example of "subacute" cor pulmonale. (Hematoxylin and eosin stain; bar length indicates 0.1
mm.)
uncommon, but these emboli rarely account for chronic pulmonary hypertension
and usually represent incidental findings. If massive, marrow or fat emboli are often
fatal (82-84). Of more significance may be the emboli of foreign material, such
as talc and fiber, that result from intravenous drug abuse. These emboli eventually
result in destruction of enough pulmonary vessels to produce pulmonary hyperten-
sion (85-87), but whether this is the only cause of pulmonary hypertension in drug
abusers is not clear. At least some may have pulmonary vascular changes from the
drugs per se, as with users of propylhexedrine, for whom evidence suggests a
vasoconstrictive component (88-90).
Pulmonary venous hypertension

Any condition that chronically impedes venous return to the left ventricle, or
otherwise results in transmission of abnormally high volumes of blood to the
pulmonary veins, may result in pulmonary arterial hypertension. Some common
causes include mitral stenosis, left ventricular failure (due to ischemic heart disease
or hypertensive heart disease), left atrial myxoma, and stenosis of the pulmonary
veins (4,20,37,53,91-93).
In this state the pulmonary vessels may contain a variety of changes, but the key
histologic hallmark is the presence of severe and diffuse abnormalities of both

pulmonary arteries and veins. Increased pulmonary venous pressure results in
medial muscular hypertrophy and intimal fibrosis to such a degree that the veins
resemble arteries, a condition frequently referred to as "arterialization" of the veins.
While narrowed, the veins are still patent in most cases.
The pulmonary arteries exhibit medial hypertrophy, often to nearly the same
degree as seen in primary pulmonary hypertension and congenital heart disease with
left-to-right shunts. Muscle is extended into arterioles, and intimal fibrosis may be
either eccentric or concentric. Usually absent are ~y plexiform lesions, necrotizing
arteritis, and areas of concentric laminar intimal fibrosis.
The capillaries are subjected to increased hydrostatic pressure, which results in
increased lymph formation, with distended lymphatic vessels, and hemorrhage by
diapedesis. Hemosiderin-laden alveolar macrophages are present and sometimes
numerous enough to give the lung a brown cast when examined grossly. Interstitial
fibrosis occurs in long-standing cases and may be severe enough to be confused with
idiopathic pulmonary fibrosis if the other findings are not appreciated (figures 3-15
to 3-17).
Another term frequently used by pathologists in referring to this class of pulmo-
nary vascular disease is chronic passive pulmonary congestion. It can sometimes be
mistaken for idiopathic diffuse interstitial pulmonary fibrosis (37,60).
Pulmonary veno-occlusive disease

The cause of this condition is not known although evidence suggests that it is not
a single entity but the result of several different diseases, including; inflammatory
diseases and, possibly, viral infections (94,95). The dominant histologic fmding is
obliteration of pulmonary veins and venules by intimal fibrosis, and sometimes by
thrombi that can be found in varying states of organization. In contrast to pulmo-
nary venous hypertension, the changes tend to be focal and affect the intima more
than the media of the veins. Not all veins are affected, although an elastic stain often
reveals that the extent of involvement is greater than suspected from routine
sections. If the arteries are affected, the change is usually mild and often limited
to a small degree of medial hypertrophy. As in pulmonary venous hypertension,
there is hemosiderosis, lymphatic dilatation, and interstitial fibrosis, but these
changes, like the involvement of the veins, tend to be patchy within a lobe or lung
and vary considerably from one patient to another (figures 3-18 and 3-19). In a
limited biopsy this condition might be mistaken for either chronic passive conges-
tion or idiopathic pulmonary hemosiderosis (3,96,97).
Hypoxic pulmonary hypertension

This category is probably the most complex from the standpoint of the many
different clinical syndromes that may be associated with it. The common denomina-
tor of these syndromes is alveolar hypoxia, although other stimuli for vasoconstric-
tion may be present as well. Among the conditions known to result in alveolar
hypoxia and hypertension are chronic airways obstruction, abnormalities of the
87
Figure 3-15. Chroniclassive congestion of the lungs. A section of lung from a patient with
rheumatic valvulitis an untreated mitral stenosis demonstrates thickening of the alveolar walls
due to a combination of factors, including congestion, interstitial edema, and interstitial fibrosis.
The dark, heavily pigmented alveolar macrophages contain hemosiderin, the result of
intra-alveolar hemorrhage by diapedesis. Such cells, often referred to as "heart-failure cells," are
prominent in cases of chronic passive congestion of the lungs. (Hematoxylin and eosin stain; bar
length indicates 0.1 mm.)
chest bellows, neuromuscular abnormalities, and residence at high altitude, to

mention some of the better known (98-105). A more complete list of etiologies
can be found on table 1-2. The characteristic abnormality in tissue sections is the
presence of longitudinal bundles of muscle in the intima of small muscular arteries
and arterioles (figure 3-20). These bundles are thought to be an almost pathogno-
monic marker of hypoxic pulmonary hypertension (106,107). Longitudinal muscle
bundles within the media may be seen in other types of pulmonary vascular disease
and do not carry the same diagnostic significance as intimal bundles do (3). Medial
hypertrophy in the small muscular arteries and extension of smooth muscle into
arterioles are present, but the arteriolar change is exaggerated when this condition
is compared to other categories, such as plexogenic pulmonary arteriopathy. The
medial hypertrophy of muscular arteries is mild at best. Plexogenic lesions are not
present, and intimal fibrosis is not conspicuous (41,46,106,107).
Pulmonary vasculopathy associated with parenchymal diseases of the lung

Some parenchymal diseases result in scarring and destruction of pulmonary vessels.
If the scarring is widespread and severe, pulmonary hypertension could result.
Diseases sometimes listed under this title include diffuse interstitial pulmonary
Figure 3-16. Venous sclerosis in chronic passive congestion of the lungs. This is from the same
case illustrated in figure 3-15. A small venule with intimal sclerosis but patent lumen is
illustrated. (Elastic stain; bar length indicates 0.1 mm.)
Figure 3-17. Venous sclerosis in chronic passive congestion of the lungs. A larger vein with
medial thickening and intimal sclerosis is shown in the photomicrograph. The elastic lamina is
more apparent than normal, giving this vein a strong resemblance to pulmonary arteries of the
same caliber. In other areas of the section, arterial changes were evident, indicating both arteries
and veins were affected in this case (same as shown in figures 3-15 and 3-16). (Elastic stain; bar
length indicates 0.2 mm.)
89
Figure 3-18. Pulmonary veno-occlusive disease. In the central portion of the photo!11icrograph
is an arc-shaped interlobular septum containing a markedly dilated lymphatic vessel. Also
evident are focal collections of hemosiderin-laden alveolar macrophages, patchy interstitial
fibrosis, and interstitial inflammation. Such findings in pulmonary veno-occlusive disease are
extremely variable from one area of lung to another. (Hematoxylin and eosin stain; bar length
indicates 0.2 mm.)
fibrosis, obstructive emphysema, and sarcoidosis (108). It is now apparent that

pulmonary hypertension in these types of pulmonary diseases is rarely the result
of destruction of blood vessels alone, but results from a combination of tissue
destruction and alveolar hypoxia (19,109-111). Histologic examination of lung
tissue in these diseases reveals the changes of hypoxic pulmonary hypertension
(extension of smooth muscle into arterioles, minimal medial hypertrophy of mus-
cular arteries, and development of longitudinal bundles of smooth muscle in the
intima of arterioles and small muscular arteries) as well as the changes of the
underlying parenchymal disease (104,106-108). Clinically, the pulmonary hyper-
tension of these patients may respond to correction of hypoxia (2,20). There are,
however, a small number of cases in which severe parenchymal destruction may
directly produce pulmonary hypertension. These include severe and long-standing
pulmonary tuberculosis, advanced silicosis or anthrasilicosis, and certain other
granulomatous diseases (3) .
Pulmonary vasculopathy in conditions with decreased pulmonary arterial flow

A number of congenital cardiac lesions may have decreased pulmonary arterial
flow, including tetralogy of Fallot, pulmonic stenosis, tricuspid atresia, and com-
Figure 3-19. Pulmonary veno-occlusive disease. A pulmonary venule has been obliterated by a
proliferation of fibrous connective tissue within the lumen. Such vessels are easily overlooked
unless specifically sought. Elastic stains are especially helpful in this search and should always be
performed. In this case the venules were more severely affected than the larger veins. A febrile
episode preceded the onset of respiratory symptoms in this patient, suggesting the possibility of a
viral or inflammatory etiology. No venous thrombi were present in this case, but they may be
prominent in others. Same case as illustrated in figure 3-18. (Elastic stain; bar length indicates
0.1 mm.)
plete transposition of the great vessels with intact septa (112,113). In each case the
pulmonary arterial flow is decreased and substantial collateral flow develops
through the bronchial circulation. The pulmonary arteries are widely dilated and
exhibit medial atrophy that in some vessels is reflected by complete absence of
smooth muscle (52,55,114). As a secondary change, there are often thrombi in
varying stages of organization, and, indeed, patients with tetralogy of Fallot seem
especially prone to thrombosis of both arteries and veins (115).
PROBLEMS IN THE CLASSIFICATION OF PULMONARY VASCULAR DISEASE

A number of pitfalls await the pathologist who tries to interpret these categories
too precisely or to separate them too definitively according to mechanism. Several
examples will serve to illustrate this point.
For many years destruction of the capillary bed in parenchymal diseases was
believed to be the main reason for increased pulmonary artery pressure, as men-
tioned before. Hypoxia is now recognized as an important factor in determining
91
Figure 3-20. Hypoxic pulmonary hypertension. The pulmonary arteriole shown in this
photomicrograph contains an eccentric bundle of longitudinally oriented smooth muscle in the
intima, a fmding that is virtually pathognomonic of hypertension induced by alveolar hypoxia.
This section came from a patient with asbestosis and cor pulmonale. In addition to the diffuse
interstitial pulmonary fibrosis, extensive pleural fibrosis was present (see figure 3-21). (Elastic
pulmonary artery pressures in these cases. This explains the difficulty in correlating
the severity of parenchymal disease to the severity of pulmonary hypertension (42).
This also accounts for the classification of interstitial pulmonary fibrosis by some
authors under diseases with destruction of vasculature, and by other authors as an
hypoxic form of pulmonary vascular disease (115). One may further compound
interpretation if there are additional causes of alveolar hypoxia, such as severe
pleural fibrosis, which could interfere with the normal bellows action of the chest
wall and result in hypoventilation (figure 3-21) .
Cor pulmonale in many cases, if not most cases, is the result of more than one
factor (116). In mitral stenosis, there is frequently evidence that pulmonary venous
hypertension and hypoxic vasoconstriction are both important in producing vascu-
lar changes (43-45). Again, deciding which of these two factors is most important
in producing physiologic or histologic abnormalities may be very difficult. As a
fmal example, to examine a tissue section and separate primary changes from
secondary changes is often difficult. Patients with pulmonary venous hypertension
may develop thromboembolic disease to complicate the pathologic picture, just as
patients with primary pulmonary hypertension develop this same complication. In
Figure 3-21. Asbestosis. Diffuse interstitial pulmonary fibrosis with early honeycombing is
illustrated in this photograph, as well as extensive pleural fibrosis. Pulmonary vascular changes
of hypoxic hypertension were evident microscopically (see figure 3-20), and the right ventricle
was severely hypertrophied and dilated. (Bar length indicates 5 mm.)
some instances of severe, long-standing primary pulmonary hypertension, the

extent of secondary thromboembolization may render impossible a clear-cut diag-
nosis of plexogenic pulmonary arteriopathy (60,65,66,79).
CARDIAC CHANGES DUE TO PULMONARY VASCULAR DISEASE

The right ventricle responds to increased loads in much the same manner as the
left ventricle does to systemic hypertension (6). An initial stage of compensatory
myocardial hypertrophy occurs during which the individual muscle cells enlarge
and the ventricular wall thickens, thus maintaining a normal output to the pul-
monary circulation. During this time, the external appearance of the heart
changes little if any, except, perhaps, to assume a slightly more globular shape.
Cross sections through the ventricle reveal a normal-sized chamber with a thick-
ened wall measuring at least 4 mm and often greater than 6 mm, exclusive of
epicardial fat (15) . As the load increases, and maximum compensatory hypertro-
phy is reached, the right ventricle dilates and pulmonary output is maintained by
the Frank-Starling mechanism (6) . Externally, the heart is now distinctly globu-
lar (figure 3-22). If opened by cuts made in the direction of blood flow, the
surface area of the chamber appears markedly increased, a reflection of the dila-
tion that has occurred (figure 3-23). The wall mayor may not appear thickened
93
Figure 3-22. External view of the heart in cor pulmonale. If the right ventricle is only
modestly hypertrophied and not dilated, the heart probably will be normal externally. In this
advanced case of cor pulmonale, the presence of severe right ventricular hypertrophy and, more
importantly, dilation due to failure, gives the heart a distinctly globular appearance. In some
cases this change is detected by chest roentgenograms.
at this point, depending on the degree of dilatation, which, if severe, can mask
even marked hypertrophy. If the heart is opened by cross sections, the effects of
dilation are revealed by the assumption of a rounded configuration by the cavity,
rather than by the sharply angled crescent shape of the normal right ventricle
(figure 3-24).
Several factors make it easy to overlook right ventricular hypertrophy. Mild
changes are rarely detected because they are rather subtle, and by comparison to
the large bulk of the left ventricle the right heart seems diminutive. If the right
heart is hypertrophied but also dilated, then the wall is thinned and measurements
of thickness may be false estimates of the total right ventriclar mass. Conversely,
in some cases of acute cardiac dilation without hypertrophy (that is, acute massive
thromboembolism), the presence of a large dilated right ventricle may mislead the
A,14- 082
Figure 3-23. Cor pulmonale, heart opened anteriorly. The left ventricle is normal. Notice the
markedly increased surface area of the right ventricle (indicated by arrow) in comparison to the
left. The right ventricle in this example is both hypertrophied and dilated.
Figure 3-24. Cor pulmonale, heart cut in cross section. Notice the rounded contour of the
right ventricular cavity (indicated by arrow), which is typical of dilation. The normal right
ventricle is a crescent-shaped thin-walled structure. Both hypertrophy and dilatation of the right
ventricle are present in this case.
95
unwary into overestimating the bulk of this chamber. Finally, in left ventricular
hypertrophy there is always some mild increase in right ventricular thickness and
muscle mass that is unrelated to any evidence of pulmonary vascular disease, left
heart failure, or increased right ventricular load. This increase probably results from
the anatomic extension of some left ventricular myofibers into the right ventricle,
a fmding present in all normal hearts. If not aware of this phenomenon, one might
mistake this physiologic response for evidence of pulmonary heart disease (117,-
118).
Clearly, then, a more defmitive method of detecting hypertrophy must be used
if reliable data are to be obtained. The method recommended by the World Health
Organization (68) is that used by Fulton (119) and others. Simply, the heart is
trimmed of excess epicardial fat, the atria and valves are removed, and the right
ventricle is dissected free from the left ventricle and interventricular septum, which
is considered as a single unit. Weights are determined on a sensitive balance, and
the ratio between the left and right is calculated. By this method the upper limit
of normal weight for the right ventricle is 65 gms, and the ratio of left ventricle
to right ventricle is 2.3 to 1 up to 3.3 to 1 (119). To eliminate borderline cases
80 gms has been recommended as the minimum weight before a diagnosis of
hypertrophy is made (68). Mild right ventricular hypertrophy is probably a reason-
able diagnosis if the weight is less than 80 gms but the LV /RV ratio is less than
2 to 1. Moderate hypertrophy is present if the weight is 80 to 100 gms, and severe
hypertrophy if the weight is greater than 100 gms (111). In both situations the
LV / R V ratio is less than 2 to 1 as long as hypertrophy is confmed to the right
ventricle. If concomitant left ventricular hypertrophy is present, the LV / R V ratio
may be in the normal range of 2.3-3.3 to 1 (119).
In order to diagnose cor pulmonale pathologically, as defined by the World
Health Organization and as used in this chapter, careful dissection of both ventricles
with measurement of weights and calculation of LV /RV provides the best docu-
mentation of isolated right ventricular hypertrophy. Unless these simple measures
are taken, many cases of cor pulmonale will most likely be missed.
Left ventricular hypertrophy is a common fmding with many different causes
and might occur in many cases of pulmonary heart disease by coincidence alone
(1,2). This subject is addressed in detail in chapter 9.
VISCERAL CHANGES DUE TO CHRONIC RIGHT VENTRICULAR FAILURE

Once cor pulmonale is established and right ventricular failure becomes chronic,
a number of other organ systems may be affected by chronic venous congestion.
Chronic passive congestion of the liver is characterized by atrophy of the hepatic
cords and expansion of the sinusoids near the centrilobular areas, which are the last
to be perfused and therefore received poorly oxygenated blood (figure 3-25).
Grossly, the liver may be enlarged due to vascular engorgement, but once the
capsule is cut, blood runs freely from the surface and the weight is usually less than
normal. The cut surface has the appearance of a sliced nutmeg, and the liver in this
state is sometimes referred to as "nutmeg liver" (figure 3-26). Clinically, there may
Figure 3-25. Chronic passive congestion of the liver. This low-power photomicrograph
demonstrates marked atrophy of the hepatic cords and cells in the centrilobular areas (indicated
by arrow). The centrilobular sinusoids are dilated and congested. corresponding to the sunken
dark areas seen grossly (see figure 3-26). Hepatocytes in the periportal areas remain nonnal.
(Hematoxylin and eosin stain; bar length indicates 0.2 rnm.)
Figure 3-26. Chronic passive congestion of the liver. The total weight of this liver was well
below normal, reflecting both loss and atrophy of the centrilobular hepatocytes. Alternating
areas of dark (congested centrilobular sinusoids) and light (nonnal hepatocytes and sinusoids in
the periportal areas) give the liver its characteristic "nutmeg" appearance.
97
be evidence of mild hepatic dysfunction, and during acute exacerbations of right

heart failure, necrosis of centrilobular hepatocytes occurs. In some long-standing
cases of cor pulmonale, repeated episodes of hepatic injury results in cirrhosis
(120,121).
The pancreas may exhibit atrophy of the exocrine cells as the result of decreased
supply of oxygen to these areas (figure 3-27) (122). Additionally, the spleen may
be affected and exhibit fibrosis and hemosiderosis, a condition termedfibrocongestive
splenomegaly (121). Visceral dysfunction in this setting is not often associated with
significant clinical problems, since cardiopulmonary malfunction is a more immedi-
ate threat to life and correction of these abnormalities leads to improvement of the
other visceral dysfunctions.
USE OF THE OPEN-LUNG BIOPSY IN EVALUATING PULMONARY VASCULAR

DISEASE
Unquestionably, one of the main roles of the lung biopsy has been in evaluation
of the pulmonary vasculature to determine operability in certain cases of congenital
heart disease. This area is rather specialized and fairly controversial as to the timing
of the biopsy and often as to the determination of whether surgery should be
performed. An adequate discussion of this area is beyond the scope of this chapter,
although a number of references dealing with it have been cited previously (56-
58,60,61,112,113).
If congenital heart lesions are excluded, the most useful role of the open-lung
biopsy is probably the separation of three clinically similar entities-plexogenic
pulmonary arteriopathy, chronic pulmonary thromboembolism, and pulmonary
veno-occlusive disease. All three may have pulmonary arterial hypertension and cor
pulmonale, but have either normal or minimally elevated pulmonary capillary
wedge pressures. The symptoms and clinical signs are often inconclusive for a
diagnosis (41,65,66,79). This distinction may be important since evidence shows
that anticoagulant therapy may slow or prevent progression of chronic thrombo-
embolic disease (123), and current studies indicate that treatment of some cases of
plexogenic pulmonary arteriopathy with vasodilating drugs may be possible (124).
If drugs eventually become a mainstay in the therapy of this latter disorder, the
open-lung biopsy may be an important tool in predicting response. Advanced
changes such as plexiform lesions, and perhaps concentric laminar intimal fibrosis
affecting a large number of arteries, may be associated with relatively little re-
sponse, just as surgery is often not effective in advanced cases of congenital heart
diseases (3,60).
In individual cases the open-lung biopsy may give valuable and often unex-
pected information. For example, figure 3-28 is a photograph of a lung biopsy
from a 35-year-old woman with moderate pulmonary hypertension and normal
left heart pressures. It was taken during the surgical repair of a 4 X 2 em atrial
septal defect. No significant medial changes were evident in any size pulmonary
artery, and fibrotic intimal changes tended to be patchy, with some vessels being
completely occluded, others having eccentric plaques, and rare vessels containing
Figure 3-27. Chronic passive congestion of the pancreas. Blood enters the pancreatic lobules in
the center and drains to the periphery. As in the liver the last cells to be perfused undergo
atrophy. Notice the cells in the outer part of the lobule are much smaller than the cells in the
center. (Hematoxylin and eosin stain; bar length indicates 0.1 mm.)
Figure 3-28. Foreign body reaction, intravenous drug abuser. This pulmonary arteriole contains
multinucleated foreign body giant cells in addition to a plug of fibrous connective tissue. By
polarized light microscopy, birefringent crystals were also apparent in the giant cells and
elsewhere within vascular lumens. Such changes are common in persons who inject crushed
tablets intravenously and may cause pulmonary hypertension. (Elastic stain; bar length indicates
0.1 mm.)
99
intraluminal septa. More importantly, each histologic section contained between

five and ten intravascular deposits of birefringent material, often with a foreign
body granulomatous reaction, and associated fibrin thrombus. By energy-dispersive
x-ray analysis, the platelike crystalline material was composed of magnesium and
silicon in the same proportions as found in the mineral talc. The intravascular
location of the crystals, and their large size (30 microns), is consistent with a
hematogenous origin (125). Talc is a common filler for various oral drugs, which
are often crushed, "dissolved" in water, and injected intravenously by illicit drug
users (85). The atrial septal defect, although large, played little or no role in causing
the patient's pulmonary hypertension, but served to confuse the clinical picture
until an incidental biopsy during surgery revealed the true nature of her disease.
No clear-cut criteria exist to determine when a biopsy is indicated. Certainly,
in many cases of clinical primary pulmonary hypertension, the biopsy is helpful
in making both diagnostic and therapeutic decisions. In other cases of unknown
or possibly mixed etiologies, the biopsy may help to clarify the clinical evaluation.
Ultimately, the decision to biopsy must be made following a careful overall
assessment of each individual case, with the possible benefits in terms of therapy
or prognosis weighed against the risks of the procedure.
Once the decision to biopsy has been made, several important steps ensure that
the pathologist receives an optimally selected and prepared sample. In most cases
the determination of which lobe to biopsy is not critical. Several studies indicate
that upper and lower lobes provide similar information and that the peripheral 1
to 2 em of lung contains an adequate number of small vessels to make a valid
sampling (56,60). An exception to this rule may be in pulmonary venous hyperten-
sion'in which the vascular changes appear exaggerated in the dependent lobes (3).
Avoiding the lingula and perhaps the right middle lobe may also be wise. The
experience of many pathologists has been that these portions of the pulmonary tree
are frequently the site of pneumonias and other inflammatory changes that leave
nonspecific scars, and thereby complicate the interpretation of the biopsy (21,126).
Once the site is selected, the lung should be fully inflated before it is clamped and
excised. Expansion of the alveoli helps to prevent collapse of the vessels and
provides a more uniform specimen than the crushed, atelectatic "blob" sometimes
given to the pathologist. Measurements of medial thickness are reproducible if this
method is followed. While still clamped, the tissue should be immersed in buffered
fixative (56). Multiple sections should be taken, but a small amount should be saved
for electron microscopy. In certain conditions another portion of fresh lung might
be snap frozen and saved for immunofluorescent studies or for other procedures
that might be indicated in a specific situation. In addition to the routine hematox-
ylin and eosin-stained sections, sections stained for elastic tissue are mandatory.
Ideally, all sections should be screened by polarized light microscopy for birefrin-
gent material. The final pathology report should reflect not only the alterations of
the pulmonary vessels, but also of the parenchyma and bronchi. Some assessment
of the most likely cause or causes of the pulmonary vascular changes should be
given, as well as an assessment of the extent, severity, and activity of the changes.
In some cases a formal grading system should be used (such as the Heath and
Edwards system or some modification), as well as more quantitative measurements
of medial muscle thickness (table 3-1). In a biopsy specimen medial thickness
greater than 10% of the external diameter of the artery in an adult is indicative
of pulmonary hypertension (58). The density of arteries per volume of lung may
be important in some cases. In other circumstances, the calculation of medial
cross-sectional area may give more reliable information about medial muscle mass
than measurements of percentage thickness (21,59). For diseases such as embolic
pulmonary hypertension, a more subjective assessment of the severity may be
sufficient since the most important information in this case is not the grade of
disease but the classification.
SUMMARY
Our understanding of the pulmonary vascular changes that underlie pulmonary
heart disease allows the recognition of seven different classes of disease based on
histologic examination of lung tissue: plexogenic pulmonary arteriopathy, chronic
embolic pulmonary hypertension, pulmonary venous hypertension, pulmonary
veno-occlusive disease, hypoxic pulmonary hypertension, pulmonary vasculopathy
associated with parenchymal diseases, and pulmonary vasculopathy associated with
decreased pulmonary flow. In some cases congenital heart defects may be the
underlying cause of pulmonary vascular disease, and right ventricular hypertrophy
in these cases is not properly called cor pulmonale. In other cases right ventricular
enlargement results from either pulmonary vasoconstriction or mechanical block-
age of the pulmonary vessels and is termed cor pulmonale, or pulmonary heart
disease. The concept that cor pulmonale frequently results from destruction of
pulmonary parenchyma has proven inadequate, and most parenchymal diseases are
now known to produce hypertension because of alveolar hypoxia. In some cases
more than one mechanism may be important in producing pulmonary hyperten-
sion. Open-lung biopsy may be helpful in certain situations, providing information
regarding the cause of pulmonary heart disease and helping to determine therapy.
Recommendations are made regarding the proper handling of open-lung biopsy
specimens and the assessment of right ventricular hypertrophy at the time of
necropsy. Grading of pulmonary vascular disease is appropriate, in certain situa-
tions, and may be improved by the use of rather simple quantitative measurements.
REFERENCES
1. World Health Organization. Chronic cor pulmonale. Report of an expert committee of the
World Health Organization. Circulation 27: 594--615, 1963.
3. Wagenvoort CA, Wagenvoort N. Pathology of Pulmonary Hypertension. New York: John
Wiley, 1977.
4. Edwards JE. Functional pathology of the pulmonary vascular tree in congenital cardiac disease.
Circulation 15: 164-196, 1957.
5. Fineberg MH, Wiggers CJ. Compensation and failure of the right ventricle. Am Heart J 11:
225-263, 1936.
101
6. Braunwald E, Ross J, Sonnenblick EH. Mechanisms of Contraction of the Normal and Failing
Heart. Boston: Little, Brown, 1976.
7. Moser KM. Pulmonary embolism. Am Rev Respir Dis 115: 829-852, 1977.
8. Mack I, Harris R, Katz LN. Acute cor pulmonale in the absence of pulmonary embolism. Am
Heart J 39: 664-677, 1950.
9. Altemus LR, Lee RE. Carcinomatosis of the lung with pulmonary hypertension. Arch Int Med
119: 32-38, 1967.
10. Brill IC, Robertson TD. Subacute cor pulmonale. Arch Int Med 60: 1043-1057, 1937.
11. Kane RD, Hawkins HK, Miller JA, Noce PS. Microscopic pulmonary tumor emboli associated
with dyspnea. Cancer 36: 1473-1482, 1975.
12. Morgan AD. The pathology of subacute cor pulmonale in diffuse carcinomatosis of the lungs.
J Path Bact 61: 75-84, 1949.
13. Winterbauer RH, Elfenbein lB, Ball WC. Incidence and clinical significance of tumor emboli-
zation to the lungs. Am J Med 45: 271-290, 1968.
14. Bank L, Benatar SR. Subacute cor pulmonale in hypersensitivity pneumonitis. A case report.
S Afr Med J 62: 768-769, 1982.
15. Brenner 0. Pathology of the vessels of the pulmonary circulation. Arch Int Med 56: 211-237,
457-497, 724-752, 976--1014, 1189-1241, 1935.
16. Weibel ER, Gomez DM. Architecture of the human lung. Science 137: 577-585, 1962.
17. Reid LM. The pulmonary circulation: Remodeling in growth and disease. Am Rev Respir Dis
119: 531-546, 1979.
18. Elliott FM, Reid L. Some new facts about the pulmonary artery and its branching pattern. Clin
Radiol 16: 193-198, 1965.
19. Semmens M, Reid L. Pulmonary arterial muscularity and right ventricular hypertrophy in
chronic bronchitis and emphysema. Br J Dis Chest 68: 253-263, 1974.
20. Harris P, Heath D. The Human Pulmonary Circulation. Edinburgh: Churchill Livingstone,
1977.
21. Heath D, Best PV. The tunica medica of the arteries of the lung in pulmonary hypertension.
J Path Bact 76: 165--174, 1958.
22. Katzenstein AL, Askin FB. Surgical Pathology of Non-neoplastic Lung Disease, Major Problems
in Pathology Series (Volume 13). Philadelphia: W.B. Saunders, 1982, pp. 281-313.
23. Civin WH, Edwards JE. The postnatal sttuctural changes in the intrapulmonary arteries and
arterioles. Arch Path 51: 192-200, 1951.
24. Reid A, Heard BE. Preliminary studies of human pulmonary capillaries by India ink injection.
Med Thorac 19: 215-219, 1962.
25. Cudkowicz L. Bronchial arterial circulation in man. Normal anatomy and responses to disease.
In Pulmonary Vascular Diseases, Moser KM (ed). New York: Marcel Dekker, 1979, pp.
111-232.
26. Wagenvoort CA, Neufield HN, Edwards JE. The sttucture of the pulmonary arterial tree in
fetal and early postnatal life. Lab Invest 10: 751-762, 1961.
27. Naeye RL. Arterial changes during the perinatal period. Arch Path 71: 121-128, 1961.
28. Rudolph AM. The changes in the circulation after birth. Their importance in congenital heart
disease. Circulation 41: 343-354, 1970.
29. SiIr.ons P, Reid L. Muscularity of pulmonary artery branches in the upper and lower lobes of
the normal young and aged lung. Br J Dis Chest 63: 38-44, 1969.
30. Wagenvoort CA, Wagenvoort N. Age changes in muscular pulmonary arteries. Arch Path 79:
524-528, 1965.
31. Warnock ML, Kunzmann A. Changes with age in muscular pulmonary arteries. Arch Path Lab
Med 101: 175-179, 1977.
32. West JB, Dollery CT, Heard BE. Increased pulmonary vascular resistance in the dependent zone
of the isolated dog lung caused by perivascular edema. Circ Res 17: 191-206, 1965.
33. Auerbach 0, Stout AP, Hammond EC, Garfinkel L. Smoking habits and age in relation to
pulmonary changes. Rupture of alveolar septums, fibrosis and thickening of walls of small
arteries and arterioles. N Engl J Med 269: 1045-1054, 1963.
34. Naeye RL, Dellinger WS. Pulmonary arterial changes with age and smoking. Arch Path 92:
284-288, 1971.
35. Hale KA, Niewoehner DE, Cosio MG. Morphologic changes in the muscular pulmonary
arteries. Relationship to cigarette smoking, airway disease, and emphysema. Am Rev Respir Dis
122: 273-278, 1980.
36. Wagenvoort CA, Wagenvoort N. Pulmonary arteries in bronchial carcinoma. Arch Path 79:
529-533, 1965.
37. Wagenvoort CA. Morphologic changes in intrapulmonary veins. Human Path 1: 205-213, 1970.
38. Heath D, Edwards JE. The pathology of hypertensive pulmonary vascular disease. A description
of six grades of structural changes in the pulmonary arteries with special reference to congenital
cardiac septal defects. Circulation 18: 533-547, 1958.
39. Heath D, Whitaker W. The pulmonary vessels in patent ductus arteriosus. J Path Bact 70:
285--290, 1955.
40. Edwards JE. Pathology of chronic pulmonary hypertension. In Pulmonary Pathology Decen-
nial, Sommers SC (ed). New York: Appleton-Century-Crofts, 1975, pp. 489-513.
41. Edwards WD, Edwards JE. Recent advances in the pathology of the rulmonary vasculature.
In The Lung. Structure, Function and Diseases, Thurlbeck WM, Abel MR (eds). Baltimore;
Williams & Wilkins, 1978, pp. 235--261.
42. Naeye RL. Pulmonary hypertension and vascular disease. In The Heart, Edwards JE, Lev M,
Abell MR (eds). Baltimore: Williams & Wilkins, 1974, pp. 297-326.
43. Ramirez A, Grimes ET, Abelmann WHo Regression of pulmonary vascular changes following
mitral valvuloplasty. An anatomic and physiologic case study. Am J Med 45: 975--982, 1968.
44. Dalen JE, Matloff JM, Evans GL, Hoppin FG, Bhardwaj P, Harken DE, Dexter L. Early
reduction of pulmonary vascular resistance after mitral valve replacement. N Engl J Med 277:
387-'394, 1967.
45. Goodale F, Sanchez G, Friedlich AC, Scannell JG, Myers GS. Correlation of pulmonary
arteriolar resistance with pulmonary vascular changes in patients with mitral stenosis before and
after valvulotomy. N Engl J Med 252: 979-983, 1955.
46. Heath D, Smith P. Pulmonary vascular disease. Med Clin N Am 61: 1279-1307, 1977.
47. Wagenvoort CA. Grading of pulmonary vascular lesions--A reappraisal. Histopathology 5:
595--598, 1981.
48. Lamb D. Cor pulmonale and pulmonary hypertension. In The Pathology of the Heart, Pomer-
ance A, Davies MJ (eds). Oxford; Blackwell Scientific Publications, 1975, pp. 167-198.
49. Fishman AP. Plexiform lesions. Path Microbiol 43: 242-245, 1975.
50. Ruttner JR, Bartschi P, Niederman R, Schneider J. Plexogenic pulmonary arteriopathy and liver
cirrhosis. Thorax 35: 133-136, 1980.
51. Saldana ME, Harley RA, Liebow AA, Carrington CB. Experimental extreme pulmonary
hypertension and vascular disease in relation to polycythemia. Am J Path 52: 935-981, 1968.
52. Wagenvoort CA. Classification of pulmonary vascular lesions in congenital and acquired heart
disease. Adv in Cardiol 11: 48-55, 1974.
53. Wagenvoort CA. Hypertensive pulmonary vascular disease complicating congenital heart dis-
ease. A review. Cardiovasc Clinics 5: 43-60, 1973.
54. Yamaki S, Wagenvoort CA. Plexogenic pulmonary arteriopathy. Significance of medial thick-
ness with respect to advanced pulmonary vascular lesions. Am J Path 105: 70-75, 1981.
55. Wagenvoort CA. Classifying pulmonary vascular diseases. Chest 64: 503-504, 1973.
56. Haworth SG, Reid L. A morphometric study of regional variation in lung structure in infants
with pulmonary hypertension and congenital cardiac defects. Brit Heart J 40: 825-831, 1978.
57. Haworth SG, Sauer U, Buhlmeyer K, Reid L. Development of the pulmonary circulation in
ventricular septal defect. A quantitative structural study. Am J Cardiol 40: 781-788, 1977.
58. Rabinovitch M, Haworth SG, Vance Z, Vawter G, Castaneda AR, Nadas AS, Reid LM. Early
pulmonary vascular changes in congenital heart disease studied in biopsy tissue. Human Path
11: 499-509, 1980.
59. Wagenvoort CA. Vasoconstriction and medial hypertrophy in pulmonary hypertension. Circu~
lation 22: 535-546, 1960.
60. Wagenvoort CA. Lung biopsy specimens in the evaluation of pulmonary vascular disease. Chest
77: 615--625, 1980.
61. Rabinovitch M, Haworth SG, Castaneda AR, Nadas A, Reid L. Lung biopsy in congenital heart
disease. A morphological approach to pulmonary vascular disease. Circulation 58: 1107-1122,
1978.
62. Takahashi T, Wagenvoort CA. Density of muscularized arteries in the lung. Its role in congeni-
tal heart disease and its clinical significance. Arch Path Lab Med 107: 23-28, 1983.
103
63. Takahashi T, Wagenvoort N, Wagenvoort CA. The density of muscularized pulmonary arteries
in normal lungs. Arch Path Lab Med 107: 19-22, 1983.
64. Fishman AP. Pulmonary Diseases and Disorders. New York: McGraw-Hill, 1982, pp. 863-882.
65. Anderson EG, Simons G, Reid L. Primary and thromboembolic pulmonary hypertension. A
quantitative pathological study. J Path 110: 273-293, 1973.
66. Wagenvoort CA, Wagenvoort N. Primary pulmonary hypertension. A pathologic study of the
lung vessels in 156 clinically diagnosed cases. Circulation 42: 1163-1183, 1970.
67. Glenn WWL, Liebow AA, Lindskog GE. Pulmonary vascular changes in congenital heart
diseases. In Thoracic and Cardiovascular Surgery with Relared Pathology. New York: Apple-
ton-Century-Crofts, 1975, pp. 918-937.
68. World Health Organization. Primary Pulmonary Hypertension. Report of a WHO Meeting,
October 15-17, 1973. Hatano S, Strasser T (eds), 1975, pp. 7-45.
69. Levine OR, Harris RC, Blanc W A, MelIins RB. Progressive pulmonary hypertension in
children with portal hypertension. J Ped 83: 964-972, 1973.
70. Naeye RL. Primary pulmonary hypertension with coexisting portal hypertension. A retrospec-
tive study of six cases. Circulation 22: 376-384, 1960.
71. Segel N, Kay JM, Bayley lJ, Paton A. Pulmonary hypertension with hepatic cirrhosis. Br Heart
J 30: 575-578, 1968.
72. Senior RM, Britton RC, Turino GM, Wood JA, Langer GA, Fishman AP. Pulmonary hyper-
tension associated with cirrhosis of the liver and with portacaval shunts. Circulation 37: 88-96,
1968.
73. Morrison EB, Gaffney FA, Eigenbrodt EH, Reynolds RC, Buja LM. Severe pulmonary
hypertension associared with macronodular (postuecrotic) cirrhosis and autoimmune
phenomena. Am J Med 69: 513-519, 1980.
74. Cohen MD, Rubin Lj, Taylor WE, Cuthbert JA. Primary pulmonary hypertension. A unique
case associated with extrahepatic portal hypertension. Hepatology 3: 588-592, 1983.
75. Fishman AL. Dietary pulmonary hypertension. Circ Res 35: 657-660, 1964.
76. Gorham LW. A study of pulmonary embolism. Arch Int Med 108: 8-22, 189-207,418-426,
1961.
77. Hampton AO, Castleman B. Correlation of postmortem chest teleoroentgenograms with au-
topsy findings. Am J Roentgenol 43: 305-325, 1940.
78. Kom D, Gore I, Blenke A, Collins DP. Pulmonary arterial bands and webs. An unrecognized
manifestation of organized pulmonary emboli. Am J Path 40: 129-151, 1962.
79. Edwards WD, Edwards JE. Clinical primary pulmonary hypertension. Three pathologic types.
Circ 56: 884-888, 1977.
80. Sadigursky M, Andrade ZA. Pulmonary changes in schistosomal cor pulmonale. Am J Trop Med
Hyg 31: 779-784, 1982.
81. Roffman BY, Simons M. Syncytial trophoblastic embolism associated with placenta increta and
pre-eclampsia. Am J Ob Gyn 104: 1218-1220, 1969.
82. Warren S. Fat embolism. Am J Path 22: 69437, 1946.
83. Schinella RA. Bone marrow emboli. Arch Path 95: 386-391, 1973.
84. Rappaport H, Raum M, HorrellJB. Bone marrow embolism. AmJ Path 27: 407-433,1951.
85. Tomashefski JF, Hirsch CS. The pulmonary vascular lesions of intravenous drug abuse. Human
Path 11: 133-145, 1980.
86. Arnett EN, Battle WE, Russo]V, Roberts WC. Intravenous injection of talc-containing drugs
intended for oral use. Am J Med 60: 711-718, 1976.
87. Lewman LV. Fatal pulmonary hypertension from intravenous injection of methylphenidate
(Ritalin) tablets. Human Path 3: 67-70, 1972.
88. Robertson CH, Reynolds RC. Wilson JE. Pulmonary hypertension and foreign body granulo-
mas in intravenous drug abusers. Documentation by cardiac catheterization and lung biopsy. Am
J Med 61: 657-664, 1976.
89. Sturner WQ, Spruill FG, Garriott JC. Two propylhexedrine-associated fatalities. Benzedrine
revisited. J Forensic Sci 19: 572-574, 1974.
90. Taylor WE, Cohle S. Personal observations.
91. Parker F, Weiss S. Nature and significance of structural changes in the lungs in mitral stenosis.
Am J Path 12: 573-598, 1936.
92. Andrews EC. Five cases of an undescribed form of pulmonary interstitial fibrosis caused by
obstruction of the pulmonary veins. Bull Johns Hopkins Hosp 100: 28-42, 1957.
93. Goodwin RA, Nickell ]A, Des Prez RM. Mediastinal fibrosis complicating healed pulmonary
histoplasmosis and tuberculosis. Medicine 51: 227-246, 1972.
94. Wagenvoort CA. Pulmonary veno-occlusive disease. Entity or syndrome? Chest 69: 82-86,
1976.
95. McDonnell P], Summer WR, Hutchins GM. Pulmonary veno-occlusive disease. Morphological
changes suggesting a viral cause. ]AMA 246: 667-671, 1981.
96. Carrington CB, Liebow AA. Pulmonary veno-occlusive disease. Human Path 1: 322-324,1970.
97. CrissmanJD, Koss M, Carson RP. Pulmonary veno-occlusive disease secondary to granuloma-
tous venulitis. Am] Surg Path 4: 93--99, 1980.
98. Arias-Stella], Saldana M. The terminal portion of the pulmonary arterial tree in people native
to high altitude. Circulation 28: 915--925, 1963.
99. Amad KH, Brennan ]C, Alexander JK. The cardiac pathology of chronic exogenous obesity.
Circulation 32: 740-745, 1965.
100. Bergofsky EH, Turino GM, Fishman AP. Cardiorespiratory failure in kyphoscoliosis. Medicine
38: 263--317, 1959.
101. Turino GM, Goldring RM, Fishman AP. Cor pulmonale in musculoskeletal abnormalities of
the thorax. Bull NY Acad Med 41: 959-980, 1965.
102. Levine OR, Simpser M. Alveolar hypoventilation and cor pulmonale associated with chronic
airway obstruction in infants with Down's syndrome. Clin Pediatr 21: 25-28, 1982.
103. Naeye RL. Alveolar hypoventilation and cor pulmonale secondary to damage to the respiratory
center. Am] Cardiol 8: 416-419, 1961.
104. Hicken P, Health D, Brewer DB, Whitaker W. The small pulmonary arteries in emphysema.
] Path Bact 90: 107-114, 1965.
105. Moss A]. The cardiovascular system in cystic fibrosis. Pediatrics 70: 728-741, 1982.
106. Heath D. Longitudinal muscle in pulmonary arteries.] Path Bact 85: 407-412, 1963.
107. Hasleton PS, Health D, Brewer DB. Hypertensive pulmonary vascular disease in states of chronic
hypoxia.] Path Bact 95: 431-440, 1968.
108. Heath D, Smith P. Pulmonary vascular disease secondary to lung disease. In Pulmonary Vascular
Diseases, Moser KM (ed). New York: Marcel Dekker, 1979, pp. 387-426.
109. Hicken P, Heath D, Brewer DB. The relation between the weight of the right ventricle and
the percentage of abnormal air space in the lung in emphysema.] Path Bact 92: 519-528, 1966.
110. Hicken P, Brewer DB, Heath D. The relation between the weight of the right ventricle of the
heart and the internal surface area and number of alveoli in the human lung in emphysema. ]
Path Bact 92: 529-546, 1966.
111. Millard], Reid L. Right ventricular hypertrophy and its relationship to chronic bronchitis and
emphysema. Brit] Dis Chest 68: 103--110, 1974.
112. Wagenvoort CA, Nauta], van der Schaar P], Weeda HWH, Wagenvoort N. Vascular changes
in pulmonic stenosis and tetralogy of Fallot in lung biopsies. Circulation 36: 924-932, 1967.
113. Wagenvoort CA, Nauta], van der Schaar P], Weeda HWH, Wagenvoort N. The pulmonary
vasculature in complete transposition of the great vessels judged from lung biopsies. Circulation
38: 746--754, 1968.
114. Wagenvoort CA, Wagenvoort N. Pulmonary vascular bed. Normal anatomy and responses to
disease. In Lung Biology in Health and Disease, Moser K (ed). New York: Marcel Dekker, 1979,
pp.l-l09.
115. Rich AR. A hitherto unrecognized tendency to the development of widespread pulmonary
vascular obstruction in patients with congenital pulmonary stenosis (tetralogy of Fallot). Bull
Johns Hopkins Hosp 82: 389-401, 1948.
116. Hecht HW. Heart failure and lung disease. Circulation 14: 265--290, 1956.
117. Reiner L, Mazzoleni A, Rodgiruez FL, Freudenthal RR. The weight of the human heart I.
Normal cases. Arch Path 68: 58-73, 1959.
118. Reiner L, Mazzoleni A, Rodriguez FL, Freudenthal RR. The weight of the human heart. II.
Hypertensive cases. Arch Path 71: 180--201, 1961.
119. Fulton RM, Hutchinson EC, Jones AM. Ventricular weight in cardiac hypertrophy. Br Heart
] 14: 413-420, 1952.
120. Lambert RA, Allison BR. Types oflesion in chronic passive congestion of the liver. Bull Johns
Hopkins Hosp 27: 350--356, 1916.
121. Moschowitz E. The morphology and pathogenesis of cardiac fibrosis of the liver. Ann rnt Med
36: 933--955, 1952.
105
122. Von Glahn WC, Chobot R. The histological alterations of the pancreas in chronic passive
congestion. Am J Path 1: 373-381, 1925.
123. ShuckJW, WalderJS, Kam TH, Thomas HM. Chronic persistent pulmonary embolism. Report
of three cases. Am J Med 69: 790-794, 1980.
124. Rubin LJ, Peter RH. Oral hydralazine therapy for primary pulmonary hypertension. N Engl
J Med 302: 69-73, 1980.
125. Abraham jL, Brambilla C. Particle size for differentiation between inhalation and injection
pulmonary talcosis. Environ Res 21: 94-96, 1980.
126. Gaensler EA, Carrington CB. Open biopsy for chronic diffuse infIltrative lung disease. Clinical,
radiologic-physiologic correlations in 502 cases. Ann Thor Surg 30: 411-426, 1980.
4. CLINICAL EVALUATION
LEWIS J. RUBIN, MD.
The development of sophisticated imaging techniques and other noninvasive labo-

ratory markers for disease has greatly enhanced our diagnostic capabilities, but a
meticulously obtained medical history and a carefully performed physical examina-
tion remain the first and most important steps in a patient evaluation. While no
single piece of information can be elicited from a history or physical examinauon
that is diagnostic of pulmonary heart disease, these bedside techniques, coupled with
other tests discussed in later chapters, may suggest the presence of cor pulmonale
and provide clues to its potential cause.
HISTORY
No symptoms are related specifically to the presence of right ventricular hypertro-
phy. Symptoms do not develop until hemodynamic compromise occurs, and this
may occur months or years after the initiation of the stimulus for right ventricular
pressure overload. Furthermore, the symptoms that develop with pulmonary hy-
pertension or, ultimately, with right ventricular failure are frequently nonspecific.
The keys to the successful clinical diagnosis of pulmonary heart disease are a high
index of suspicion in patients at risk for its development, coupled with the presence
of those signs and symptoms sufficiently compatible with the diagnosis to warrant
further noninvasive or invasive investigation.
L.J. Rubin (ed.j, Pulmonary Heart Disease. All riglrts reserved. Copyright @ 1984 Martinus NijhoffPubiishing. Boston/The Hague/Dor-
dre,ht/La.,aster.
107
108 4. Clinical Evaluation
Dyspnea
Dyspnea is the most common symptom of patients with pulmonary heart disease
(1-5). Patients with normal lung function and mild to moderate pulmonary
hypertension may have dyspnea only with exertion. Dyspnea at rest occurs in
patients with chronic respiratory diseases in the presence or absence of pulmonary
heart disease, or in patients with severe pulmonary hypertension resulting from
primary vascular disease. An increase in effort-related dyspnea in a patient with
chronic pulmonary disease may be suggestive of the development of coexistent
heart disease, particularly if no demonstrable change has occurred in airway dynam-
ics. Cor pulmonale or left ventricular dysfunction may be contributing factors in
this setting.
Dyspnea is a common presenting complaint of patients with primary pulmonary
hypertension and pulmonary heart disease secondary to chronic thromboembolism.
The mechanism responsible for dyspnea and tachypnea in these conditions is
unclear; it is unrelieved by the administration of 100% oxygen (6), but improves
with interventions that reduce pulmonary vascular resistance (7,8). Some have
suggested that dyspnea is caused by the stimulation of intrapulmonary receptors
(6,9), somehow affected by the presence of pulmonary vascular disease. Alterna-
tively, dyspnea may be related to the impairment in tissue oxygen delivery, which
is common in these patients, since improving systemic oxygen delivery by increas-
ing cardiac output frequently results in a reduction in breathlessness.
Syncope
Some patients with pulmonary heart disease seek medical attention because of
dizziness or syncope. These symptoms may be effort related or, particularly in
patients with chronic obstructive lung disease, may be induced by paroxysms of
coughing (10). The mechanism responsible appears to be a decrease in cerebral
blood flow resulting from the inability of the right heart to increase its output with
exertion, coupled with the peripheral vasodilation that accompanies exertion (11,-
12). In cough-related syncope the increased intrathoracic pressure reduces venous
return to the heart, resulting in a decreased cardiac output. Arrhythmias or conduc-
tion defects may also cause or contribute to syncope.
Chest pain
Several studies have reported the association of substernal or parasternal chest pain,
frequently exertional, with pulmonary hypertension from a variety of causes
(10,13-15). This pain is often indistinguishable from left ventricular ischemic pain
and has been referred to as "right-heart angina." Right ventricular ischemia is most
likely responsible and is most likely caused by the increased oxygen demands of
a hypertrophied right ventricle exceeding its oxygen delivery. Dilatation or disten-
tion of the main pulmonary trunk has also been suggested as a possible mechanism
for pain (14), but there is little evidence to support this. In the author's experience,
precordial pain is more common in those conditions that produce the highest
109
pulmonary arterial pressures, such as primary pulmonary hypertension or recurrent

thromboembolism. A recent study by Zimmerman et al. (16) suggested that the
incidence of chest pain was no greater in patients with pulmonary hypertension
than in the general population, but this was a retrospective evaluation, and the
accuracy and reliability of the histories is questionable.
Hemoptysis
Hemoptysis occurs frequently in patients with chronic bronchitis and does not
necessarily imply the presence of an abnormality in the pulmonary circulation. The
constellation of hemoptysis, dyspnea, and pleuritic chest pain suggests the diagnosis
of an acute pulmonary embolism to most clinicians. But the absence of these
symptoms does not exclude recurrent thromboembolism as the cause of pulmonary
heart disease since many patients with pulmonary heart disease secondary to recur-
rent thromboembolism experience no symptoms until they present with the non-
specific signs and symptoms of right heart failure (17-20). Hemoptysis can occur
in the advanced stages of primary pulmonary hypertension, probably due to
rupture of dilated capillaries or arterioles that are under increased pressure (5).
other symptoms
Other symptoms included palpitations, either from arrhythmias or from the sensa-
tion of an enlarged heart "pounding in the chest," and hoarseness, resulting from
compression of the recurrent laryngeal nerve by enlarged proximal pulmonary
arteries (21). Early satiety, epigastric or right upper quadrant pain, nausea, and
vomiting occur occasionally, due to hepatic congestion and distention of Glisson's
capsule. RaynilUd's phenomenon is said to occur in 30% of patients with primary
pulmonary hypertension (22), but in my experience with 20 patients with primary
pulmonary hypertension, only one had Raynaud's phenomenon. Headache was
observed in 55% of patients with chronic obstructive lung disease in Flint's series
(10) and was attributed to the vascular dilatation produced by hypoxia and hyper-
carbia.
PHYSICAL EXAMINATION
General
The general appearance of a patient may provide important clues regarding the
pathogenesis of pulmonary heart disease. Chest cage deformities with limited
respiratory excursion suggest hypoventilation secondary to chest wall disease.
Obesity, somnolence, and plethora are suggestive of obesity-hypoventilation syn-
drome or sleep apnea. The appearances of the "pink puffer" and "blue bloater" with
obstructive lung disease are familiar to most clinicians.
Cyanosis
Cyanosis is often one of the first signs observed by the physician and may be
prominent in patients with pulmonary heart disease. Several potential causes for
cyanosis exist: The first cause is arterial hemoglobin desaturation secondary to

impaired intrapulmonary gas exchange. The most common cause of central cyano-
sis in this category is chronic bronchitis, where arterial oxyhemoglobin saturations
as low as 70 to 75% while breathing room air are not uncommon. Hypoxemia
to this degree is unusual in other forms of pulmonary heart disease. The second
is right-to-Ieft shunting through a patent foramen ovale. As right-sided heart
pressures increase, shunting across the foramen ovale can occur in all forms of
pulmonary heart disease and contribute to hypoxemia and cyanosis. The third cause
is peripheral cyanosis secondary to a low cardiac output. In this situation the arterial
hemoglobin saturation may be relatively normal, but the chronic low-flow state
requires an increased extraction of oxygen at the tissue level and results in a high
concentration of desaturated hemoglobin in the capillaries, thus producing periph-
eral cyanosis. This condition is more common in primary pulmonary vascular
diseases.
Vital signs
Tachypnea is common in primary pulmonary hypertension and thromboembolic
pulmonary hypertension (23), whereas most patients with obstructive lung disease
have hypoventilation. Tachycardia is common, particularly late in the course of
the disease when right heart failure may be present. Systemic hypertension is said
to be uncommon in patients with severe parenchymal lung disease (24,25). This
may be due to the inability of the diseased lung to metabolize or synthesize
vasocative substances, such as angiotensin. An exception to this may be patients
with sleep apnea syndrome and cor pulmonale, who may also have systemic
hypertension (26-28). Most patients with severe pulmonary hypertension and a
reduced cardiac output have low systemic blood pressures.
GenerM exanrlnanon
Patients with severe derangements in gas exchange may be somnolent, confused,
and alarmingly sedate. A flapping tremor, similar to asterixis in patients with
hepatic failure, may be present (2). A puffy facies may be present, and exophthal-
mos has been noted, both due to increased venous pressure (29). On fundoscopic
examination these patients may have dilated and darkly congested retinal veins (10)
and papilledema from the effects of hypoxia and hypercarbia (30).
Cardiovascular exanrlnanon
The cardiovascular examination may be quite impressive. A visible right ventricu-
lar impulse may be appreciated, either in the parasternal region or, commonly in
patients with obstructive airways disease, in the epigastrium. The neck veins may
be distended, but this condition does not necessarily imply right ventricular failure:
patients with obstructive airways disease and increased intrathoracic pressures may
have jugular venous distention in the absence of cor pulmonale. A prominent A
111
wave in the jugular venous pulse is a useful sign, implying the presence of a
noncompliant right ventricle. Prominent CV waves are generally felt to reflect the
presence of tricuspid insufficiency, but Sherman et al. (31) suggested an alternate
explanation. They analyzed 70 right atrial curves obtained by cardiac catheteriza-
tion in patients with cor pulmonale, of whom 46% had peripheral venous conges-
tion, and found only one case with true tricuspid regurgitation. They suggested
that the clinical signs frequently attributed to tricuspid insufficiency may be ex-
plained by a sudden reversal of flow from the right atrium to the veins as a
consequence of pressure variations within a dilated right atrium.
Palpation of the precordium may reveal a parasternal right ventricular lift, and,
on occasion, a palpable pulmonic component of the second heart sound (P~. On
auscultation P 2 may be loud and delayed, particularly if a right bundle branch block
is present. A right-sided fourth heart sound and pulmonic ejection click are fre-
quently audible. A right ventricular gallop (S3) implies right ventricular failure
(32).
It should be emphasized that these signs are more prominent in patients with
primary vascular disease. In patients with severe obstructive airways disease, auscul-
tation of either normal or abnormal sounds is often impossible, owing both to the
hyperinflation of the chest and noisy airway sounds.
Murmurs of tricuspid or pulmonic insufficiency may be heard. Tricuspid insuffi-
ciency is a holosystolic murmur, generally best heard at the sternal border and
increasing in intensity with inspiration (Carvallo sign) (33). The murmur may also
be augmented by manual hepatic pressure. The Graham Steell murmur of pulmonic
insufficiency is a short, early diastolic murmur heard at the parasternal region or
second left intercostal space.
An enlarged and tender liver is suggestive of venous congestion. Patients with
obstructive lung disease and hyperinflation have low diaphragms and may have a
palpable liver without congestion. Hepatic pulsation suggests tricuspid insuffi-
ciency.
Edema
The presence of edema is frequently and, often incorrectly, ascribed to right heart
failure. Although this condition is no doubt frequently the case, other causes for
edema should be considered. We have performed right heart catheterization on six
patients referred to us with severe chronic obstructive lung disease, edema, and "cor
pulmonale," only to find that hemodynamic measurements, both at rest and with
exercise, were normal. Venous insufficiency and hypoalbuminemia, particularly in
patients with cystic fibrosis (34), are other potential causes for edema. Campbell
et al. (35) suggested that in patients with obstructive lung disease, edema was more
closely related to hypercarbia than the presence of right heart failure. They sug-
gested that an increased PC02 would promote the exchange of H+ for Na+ in
the renal tubules, expanding the intravascular volume and potentially predisposing
to edema formation.
....
....
N
Table 4-1. Clinical features of most common etiologies for pulmonary hypertension I
Diagnosis Cor pulmonale Recurrent Primary pulmonary Left ventricular Congenital heart
secondary to pulmonary hypertension failure disease with
disturbances in thromboembolism Eisenmenger's
ventilation complex
History Dyspnea (rest and Dyspnea, usually Dyspnea, usually Dyspnea (on Dyspnea, especially
exertion), cough exertional; fatigue; exertional; fatigue; exertion or rest), exertional;
(dry or chest pain; chest pain; paroxysmal syncope; chest
productive), syncope or syncope; nocturnal dyspnea, pain; hemoptysis
fatigue, weight dizziness Raynaud's fatigue, weight
gain. phenomenon gain
Physical examination Findings consistent Tachypnea; R V Tachypnea; R V Tachypnea; left-sided Tachypnea; central
with underlying heave; loud P2; heave; loud P2; S4' S3; rales; cyanosis; R V lift;
disease, i.e., right-sided S4' S3; right-sided S4' S3; edema loud P2;
wheezes, ronchi, TR or PR TR or PR right-sided S4' S3;
dry rales; central murmurs; edema; murmurs; edema; TR,PR
cyanosis; Edema; peripheral cyanosis peripheral cyanosis murmurs;
clubbing clubbing
Pulmonary function Abnormal, usually Normal or mildly Normal or mildly Minimally abnormal Normal; DLCO
tests severe defect; sleep abnormal in abnormal in absence of may be low
study in suspected absence of coexistent lung
sleep apnea coexistent airways disease
disease; DLCO
may be low
Arterial blood gases Hypoxemia, usually Moderate hypoxemia Mild hypoxemia Mild hypoxemia, Moderate to severe
severe in COPD; with respiratory with respiratory often with mild hypoxemia;
hypercapnia in alkalosis alkalosis respiratory minimal change
COPD; alkalosis with 100% O 2
polycythemia respiratory
alkalosis;
polycythemia
EKG R VH or nonspecific RVH RVH Nonspecific changes; RVH
changes LV ischemic
changes, i.e.,
previous infarct
Chest roentgenogram Evidence of Normal lung fields; Normal lung fields; Pulmonary vascular Cardiomegaly;
parenchymal enlarged R V; enlarged R V; congestion; dilated pulmonary
disease dilated proximal dilated proximal enlarged LV arteries; normal
pulmonary arteries pulmonary arteries lung fields
Radionuclide Abnormal R VEF; Abnormal R VEF; Abnormal R VEF; Abnormal LVEF; Abnormal R VEF;
angiocardiogram normal LVEF normal LVEF normal LVEF normal or Normal LVEF;
abnormal R VEF shunt may be
detected
Echocardiogram Dilated RA and Dilated RA and Dilated RA and Dilated LV; normal Dilated R V; shunt
RV; normal LV; RV; normal LV; RV; normal LV; or abnormal R V may be
normal valves; normal valves normal valves demonstrable by
(right heart not contrast study
always visualized)
Cardiac Pulmonary Pulmonary Pulmonary Mild pulmonary Pulmonary

catheterization hypertension; hypertension; hypertension; hypertension; hypertension;
normal capillary emboli by normal capillary elevated shunt
wedge angiography (or wedge; no pulmonary demonstrable by
lung scan) significant shunt; capillary wedge serial dye curves
no emboli by or contrast
angiography (or injection
lung scan)
'The studies listed may not be necessary for all patients. Other tests, i.e., open-lung biopsy, may be necessary for diagnosis in some patients.
...
...
...,
Clubbing
Digital clubbing is common in many forms of chronic pulmonary disease and does
not imply the presence of pulmonary heart disease in this setting. Clubbing is
unusual in patients with either primary pulmonary hypertension or thromboem-
bolic disease, and its presence in the face of significant pulmonary hypertension
justifies a search for other causes, such as congenital heart disease, parenchymal lung
disease, or cirrhosis with pulmonary hypertension.
CONCLUSION
The diagnosis of pulmonary heart disease at the bedside is frequently difficult and
occasionally impossible. The clinician must extract information from the history
and physical exam, piece it together into a unifying hypothesis, and couple this
with an index of suspicion. The clinician may then be justified in performing
additional noninvasive or invasive tests, which may both facilitate making the
diagnosis and clarify the etiology. The features of the history, physical examination,
and laboratory tests of several of the etiologies for pulmonary hypertension which
are frequently considered are listed in table 4-1. These tests, and their roles in
diagnosis and management of patients with suspected pulmonary heart disease, are
discussed in subsequent chapters.
REFERENCES
1. W orId Health Organization: Chronic cor pulmonale. Report of an expert committee. Circula-
tion 27: 594-615, 1963.
2. Oakley CM, GoodwinJF. Current clinical aspects of cor pulmonale. AmJ Cardiol20: 842-852,
1967.
3. Neal R W, Nair KG, Hecht HH. A pathophysiological classification of cor pulmonale: With
general remarks on therapy. Mod Concepts Cardiovasc Dis 37: 107-112, 1968.
4. Yu PN. Primary pulmonary hypertension: Report of 6 cases and review of the literature. Ann
Intern Med 49: 1138-1161, 1958.
5. Voelkel N, Reeves Jr. Primary pulmonary hypertension. In Pulmonary Vascular Diseases, KM
Moser, ed. New York; Marcel Dekker, 1979, pp. 609-611.
6. Obrecht HG, Scherrer M, Gurtner HP. Der Gasaustausch in der Lunge bei der prirnar vascularen
Form des chronischen Cor pulmonale. Schweiz Med Wochenschr 98: 1999-2007, 1968.
J Med 302: 69-73, 1980.
8. Klinke WP, Gilbert JAL. Diazoxide in primary pulmonary hypertension. N Engl J Med 302:
91-92, 1980.
9. Guz A. Regulation of respiration in man. Ann Rev Physiol 37: 303--323, 1975.
10. Flint FJ. Cor pulmonale. Lancet 2: 51-58, 1954.
11. Howarth S, Lowe JB. Mechanism of effort syncope in primary pulmonary hypertension and
cyanotic congenital heart disease. Br Heart J 15: 47-54, 1953.
12. Dresdale DT, Schultz M, Michtom RJ. Primary pulmonary hypertension. I. Clinical and hemo-
dynamic study. Am J Med 11: 686--705, 1951.
13. Lasser RP, Genkins G. Chest pain in patients with isolated pulmonic stenosis. Circulation 15:
258-266, 1957.
14. Viar WN, Harrison TR. Chest pain in association with pulmonary hypertension. Circulation 5:
1-11, 1952.
15. Ross RS. Right ventricular hypertension as a cause of precordial pain. Am Heart J 61: 134-135,
1961.
115
16. Zimmennan D, Parker BM. The pain of pulmonary hypertension. Fact or fancy? J Amer Med
Assoc 246: 2345-2346, 1981.
17. Owen WR, Thomas WA, Castleman B. Unrecognized emboli to the lungs with subsequent cor
pulmonale. N Engl J Med 249: 919-926, 1953.
18. Freilich JK, Szatkowski J. Pelvic vein thrombosis with pulmonary vascular lesions simulating
primary pulmonary hypertension. Am J Cardiol 7: 297-301, 1961.
19. Davison PH, Armitage GH, McIlveen DJS. Chronic cor pulmonale due to silent pulmonary
embolism. Lancet 2: 224-226, 1956.
20. Riedel M, Stanek V, Widimsky J, Prerovsky I. Long-term follow-up of patients with pulmonary
thromboembolism. Chest 81: 151-158, 1982.
21. Wilmshurst PT, Webb-Peploe MM, Corker RJ. Left recurrent laryngeal nerve palsy associated
with primary pulmonary hypertension and recurrent pulmonary embolism. Br Heart J 49:
141-143, 1983.
22. Walcott G. Burchell HB, Brown AL. Primary pulmonary hypertension. AmJ Med 49: 70-79,
1970.
23. Fanta CH, Wright TC, McFadden ER. Differentiation of recurrent pulmonary emboli from
chronic obstructive lung disease as a cause of cor pulmonale. Chest 79: 92-95, 1981.
24. Anderson AE, Bedrossian CWM, Foraker AG. Systemic blood pressure in subjects with and
without emphysema. Am Rev Respir Dis 103: 576-578, 1971.
25. Schrijen F, Uffholtz H, PoluJM, Poincelot F. Pulmonary and systemic hemodynamic evolution
in chronic bronchitis. Am Rev Respir Dis 117: 25-31, 1978.
26. Tilkian AG, Guilleminault C, Schroeder JS, Lehrman KL, Simmons FB, Dement WC. Hemody-
namics in sleep-induced apnea. Ann Intern Med 85: 714-719, 1976.
27. Guilleminault C, Tilkian A, Dement WC. The sleep apnea syndromes. Ann Rev Med 27:
465-484, 1976.
28. Guilleminault C, Eldridge FL, Tilkian A, Simmons FB, Dement WC. Sleep apnea syndrome due
to upper airway obstruction. Arch Intern Med 137: 296-300, 1977.
29. Spain DM, Handler BJ. Chronic cor pulmonale. Arch Intern Med 77: 37-65, 1946.
30. Simpson T. Papilledema in emphysema. Br Med J 2: 639-641, 1948.
31. Sherman WT, Ferrer MI, Harvey RM. Competence of the rricuspid valve in pulmonary heart
disease (cor pulmonale). Circulation 31: 517-522, 1965.
32. Gupta BD, Moodie DS, Hodgman JR. Primary pulmonary hypertension in adults. Clinical
features, catheterization fmdings and long-term follow-up. Cleve Clin Quart 47: 275-284,1980.
33. Cha SD, Gooch AS. Diagnosis of rricuspid regurgitation. Arch Intern Med 143: 1763-1768, 1983.
34. Fleisher DS, DiGeorge AM, Bamess LA, Cornfeld D. Hypoproteinemia and edema in infants with
cystic fibrosis of the pancreas. J Pediat 64: 341-348, 1964.
35. Campbell EJM, Short DS. The cause of edema in "cor pulmonale." Lancet 1: 1184-1186, 1960.
5. THE ELECTROCARDIOGRAM IN CHRONIC LUNG DISEASE
FRANCIS X. MCGOWAN, M.D.

and
GALEN S. WAGNER, M.D.
A discussion of the electrocardiogram (ECG) must consider that the parameters

such as axis and amplitude of the waves on the electrocardiogram are the net result
of both the "cardiac generator" and the electrical properties of the surrounding
tissues, as recorded at fixed, arbitrary points on the body surface. This interplay
of heart and thorax becomes especially prominent in the presence of pulmonary
disease, which may produce either (a) heart disease (cor pulmonale) by causing
elevation of pulmonary arterial pressure that overloads the right ventricle, and Ior
(b) thoracic alterations (due to the hyperinflation of emphysema) that modify the
electrical signal transmitted to the body surface from the heart. In theory one
should be able to distinguish between the changes produced on the ECG by these
markedly different factors. The confusion apparent in the literature indicates,
however, that such a distinction is quite difficult. This confusion originates from
discrepancies in the criteria used for the clinical diagnosis of entities such as
emphysema, chronic bronchitis, and cor pulmonale. Even the definitions of these
terms vary. Patients commonly do not have a "pure" form of lung disease;
therefore, the ECG criteria designed to separate these entities are often derived from
patients with mixed disease and are considered to indicate "chronic obstructive
The authors wish to express appreciation to Nancy B. Hindman for editorial review and to Gail McKinnis for
preparation of the manuscript.
L.J. Rubin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright © 1984 Martinus NijhofJPublishing. Boston/The Hague/Dor-
drecht/Lancaster.
117
118 5. The Electrocardiogram in Chronic Lung Disease
pulmonary disease" (COPD). Either emphysema or bronchitis may be the major

or sole process initially, but most patients eventually enter the common pathway
of airways obstruction and parenchymal destruction, with the resultant air-trap-
ping, hypoxia, acidosis, hypercapnia, and pulmonary hypertension that cause both
hyperinflation and cor pulmonale.
Factors other than pulmonary disease also confound the diagnostic capability of
the ECG. For example, cardiac abnormalities are present in many patients. Addi-
tionally, there are unpredictable and often canceling effects of such factors as age,
body habitus, concurrent disease status, and drug and electrolyte disturbances.
Indeed, the value of the ECG in diagnosis of pulmonary disease has been considered
limited to those conditions that were relatively stable over time and did not alter
the conductivity of the thorax or change the electrical position of the heart, such
as primary pulmonary hypertension, interstitial lung disease, and alveolar hypoven-
tilation with normal lungs (1). These pulmonary and extrapulmonary considera-
tions have prevented the development and the evaluation of the sensitivity and
specificity of ECG criteria for either cor pulmonale or emphysema. Never~eless,
this chapter will attempt to demonstrate that the ECG is a useful tool in the
assessment of these abnormalities and that ECG variations induced by fluctuations
in these disease states can sometimes be employed in evaluating disease severity and
response to therapy. The limitations of clinical, hemodynamic, and radiographic
criteria for the diagnosis of both COPD-produced cor pulmonale and emphysema
serve to maximize the importance of accuracy in ECG interpretation. This chapter
discusses the most useful criteria presently available for the diagnosis by ECG of
both emphysema and cor pulmonale secondary to COPD. The use of the ECG in
acute asthma and cor pulmonale from other causes will also be examined. Changes
on the standard ECG will be considered primarily, with reference to the vectorcar-
diogram only when specific and sensitive criteria have been developed.
THEORETICAL ASPECTS
Pulmonary emphysema and the ECG

Although hyperinflation is widely stated to induce substantial rotation of the heart
upon its longitudinal and transverse axes (2), there is little supporting radiographic
or autopsy evidence. Rotation sufficient to cause the observed magnitude of ECG
changes is probably anatomically impossible (3). However, hyperinflation does
cause the heart to assume a lower and more vertical position in the chest (2,3). This
is the result of several factors, including flattening of the diaphragm and pulmonary
herniation into the anterior and superior mediastinum caused by the increased
intrathoracic volume (3). The appearance of "verticality" is enhanced by asymmet-
ric movements of the thorax relative to the heart, with the largest cardiac-to-
thoracic distances occurring laterally and posteriorly. The combined result of these
factors is a more anterior and superior electrode position in relation to the cardiac
dipole. This alteration in recording contributes (a) to a "negative" cardiac location
119
due to the descended heart with resultant negative deflections (rS, QS, qrS) in leads
I-III, aVF, and V4-6; (b) to right axis deviation (RAD); (c) to posterior displace-
ment of the transitional complex (and therefore the mean QRS vector) in the
horizontal plane from its normal location near V3, often past V6 (termed "clock-
wise rotation" of the heart); and (d) to diminished limb lead voltages due to more
posterior net force orientation placing them more perpendicular to the frontal
plane.
Other investigators have considered the electrical conduction properties of the
tissues (for example, muscles, lungs) surrounding the heart responsible for transmis-
sion of cardiac potentials to the body surface. Evidence suggests that air trapping
serves to electrically isolate the heart. Hyperinflation approximately triples the
electrical resistance of normal lung tissue (1), while increased anteroposterior
thoracic diameter both increases the necessary conduction distance and alters con-
duction pathways. These factors are primarily responsible for the decreased ECG
voltages in those diseases with increased lung volumes.
The lung, a nonhomogeneous conductor under normal conditions, may also
contribute to changes in vector direction when it is randomly fIlled with em-
physematous air sacs of low conductivity. Grant (4) postulated that the insulation
effect resulting from voluminous, poorly conducting lung tissue surrounding the
heart would concentrate the electric fIeld along the vertical axis of the medias-
tinum. This would minimize anterior, posterior, and laterally directed vector
components, yielding largely superior or inferior forces. The net result would be
determined by extent and location of hyperinflation, anatomic cardiac position,
chamber enlargement, and so forth. Rudy et al. (5) have confIrmed that diminished
voltage, both overall and especially in the anterior and posterior vectors, can
accompany hyperinflation. Similar fmdings have been reported to explain surface
potential abnormalities occurring with cystic fIbrosis (without apparent R VH) (6)
and the variations induced by normal respiration (7). In a canine preparation
without either R VH or cardiac rotation, hyperinflation produced decreased volt-
ages in both the frontal plane and the left precordial R waves and superior and
posterior migration of the QRS vectors (8).
Cor pulmonale and the ECG

Pulmonary hypertension and right ventricular hypertrophy (RVH) are the normal
situation in the fetus and newborn, serving to maintain the fetal circulatory pattern
and thus shunt blood (via the foramen ovale and ductus arteriosus) away from the
uninflated lungs to the low resistance systemic circulation and placenta. Birth results
in (a) loss of the low resistance placental "sink," (b) lung inflation, and (c) increased
arterial oxygen saturation. These events convert the pulmonary circuit to one of
low pressure/low resistance, elevate the systemic pressure and resistance, and
thereby terminate the fetal circulation. Subsequently, left ventricular predominance
results from the increased pressure work of the left ventricle (LV) in comparison
to that of the right ventricle (RV). The high compliance of the pulmonary circuit
is maintained by the vast number of recruitable capillary beds with their minimal
autonomic and muscular tone.
As may be inferred from neonatal events, oxygen is a potent pulmonary
vasodilator. Conversely, hypoxemia, produced by the ventilation-perfusion (V/O)
mismatch of COPD, is the major stimulus for pulmonary vasoconstriction and
pulmonary hypertension. This relationship underlies most of the effects of pul-
monary disease on heart disease. At first, the hypoxemia and pulmonary hyper-
tension may be episodic, occurring only nocturnally (9) or with exercise (10),
and thus the R V disturbance may also be intermittent. Diminished ventilation,
due to airways obstruction and to increased dead space from destroyed alveoli,
induces hypercapnia and acidosis, both of which have a vasoconstrictive effect
and thereby potentiate the pulmonary hypertensive effect of hypoxemia (11,12).
Pulmonary arterial pressure can be approximately predicted by an equation that
requires only the results of arterial blood gas analysis (13). Pulmonary hyperten-
sion increases RV pressure work (systolic overload), which, if it is sustained
chronically, produces the predominant response of muscular hypertrophy of the
RV free wall (14).
Other consequences of COPD contribute to a R V diastolic (volume) over-
load, via hypoxemia-induced compensatory erythrocytosis and intravascular vol-
ume expansion. These processes increase cardiac output and pulmonary blood
flow via the Frank-Starling mechanism. The R V reaction to volume overload is
dilatation, especially of the inflow tract (14). Right bundle branch block
(RBBB)---either complete or incomplete-is characteristic of R V volume over-
load (15).
Irreversible anatomic constriction and Ior destruction of the pulmonary vascular
bed had been considered to be the primary cause of pulmonary heart disease in
COPD. These factors are now known to playa minimal role until later in the
disease course. The relatively new concept of the reversibility of many of the
cardiac sequelae of cor pulmonale with early diagnosis and therapy has been a
significant cause of decreased patient morbidity and mortality (16). This under-
standing is a major stimulus to fmd methods for early diagnosis of cor pulmonale
using the ECG and other techniques.
Unfortunately, the ECG changes specific for these events have not yet been well
defined. The inability to develop reliable ECG criteria for the detection of cor
pulmonale is a result of three causes. First, the masking effects of the larger LV
in the adult necessitates a greater degree of R V abnormality for it to become
apparent; concurrent presence of systemic hypertension will enhance this masking
effect. Second, coexistent diseases such as coronary artery disease with myocardial
infarction also complicate ECG interpretation. Finally, the variable severity of
coexistent emphysema-induced hyperinflation influences the electroconductive
effects that alter axes and voltages. These problems have confounded the execution
of well-designed studies leading to ECG criteria that are both sensitive and specific
for the diagnosis of cor pulmonale.
121
DIAGNOSIS BY EeG
ECG diagnosis of emphysema

The above discussion makes apparent that attempts to detect emphysema on the
ECG would meet with difficulty for two reasons: the identification of patients with
occult cor pulmonale but no emphysema (low specificity) and the failure to detect
those with true emphysema (low sensitivity). One or both deficiencies limit the
usefulness of the majority of ECG parameters for emphysema previously described.
Failure to carefully obtain historical, radiographic, blood-gas, pulmonary function
test (PFT), or hemodynamic correlations with postmortem findings has been the
most common cause of limitation to these efforts. Attempts to define criteria
specific for emphysema (3,17-20) have been confounded by one or more of the
following: (a) patients with a significant amount of mixed disease-emphysema
plus bronchitis, systemic hypertension, and so on; (b) the unknown degree and
type(s) ofPFT abnormality, acidosis, hypoxia, pulmonary hypertension, R VH, and
so on; and (c) the ECG criteria used to rule out a significant contribution ofRVH
to the abnormalities observed are of proven low sensitivity and! or specificity in
patients with pulmonary disease (see next section of this chapter). While many of
the ECG criteria thus derived may be present in patients with emphysema, most
are of relatively low sensitivity; additionally, these criteria also tend to "detect"
a significant number of normals as well as patients with chronic bronchitis, congeni-
tal heart disease, and myocardial infarction.
The criteria of Selvester and Rubin (21) (table 5-1 and figure 5-1) were
established with consideration of many of the above difficulties. These criteria
achieve approximately 65% sensitivity for diagnosis of emphysema and 95%
specificity for exclusion of emphysema in normal controls and in patients with
congenital heart disease or myocardial infarction (21,22). This good perform-
ance relative to other systems is most likely the result of combining criteria of
both P and QRS axes with QRS amplitude criteria. The study population of
Selvester and Rubin was carefully assessed by clinical and autopsy means to
assure the inclusion of patients with either "pure" emphysema or "pure"
RVH.
The usefulness of these ECG criteria for emphysema is indirectly supported by
many other studies. The normal P wave axis lies between + 5° and + 50° in >
99% of adults and is unaffected by age, sex, or body habitus (23). Rightward shift
of the P wave (generally considered to be > + 60°) correlates well with the
presence and degree of abnormal PFTs (24-28) but much less so with congenital
heart disease, mitral stenosis, or right ventricular enlargement (17,21,26,27). As
discussed above, both posterior orientation of the mean Q RS vector in the horizon-
tal plane (sometimes called "clockwise rotation") and diminished QRS voltages
can occur in pure hyperinflation due to the diminished and distorted electrical
conductivity of the hyperinflated lungs. Approximately 98% of normal subjects
will have an equally biphasic QRS complex by V4 (23); thus, no more than 2%
Table 5-1. Selvester and Rubin's proposed criteria for pulmonary emphysema
Definite Possible
P-axis > +60· in limb leads' P-axis > + 60· in limb leads
AND AND
EITHER EITHER
1) Rand S amp. ~0.7mv in limb R and S amp. ~ 0.7mv in limb
A
1 leads
AND
All) leads
OR
2) R amp. ~ 0.7mv in lead V6 2) R amp. ~ 0.7mv in lead V6
OR
B) SV4 ~ RV4
'leads J, II, III, aVR, a VL, aVF
"
Jm
1
I AVR ~ H-fI--t-t-t+-H
n AVl
•
L.... ,-
"'''.'"
m AVF Vol ' .

, ::.: ~-
Figure 5-1. Criteria of Selvester and Rubin (table 5-1) and of Spodick (table 5-2) are
satisfied, but none of the criteria for cor pulmonale are satisfied. Frontal plane axis is + 90· and
the S amplitude is > the R amplitude in lead V4 (Selvester and Rubin). P pulmonale is
present, the frontal P axis is +90·, and the S waves are at least 2 mm in duration in all of the
precordial leads (Spodick).
would meet the SV4 ~ R V4 criterion. This index of posterior QRS orientation
also appears to be independent of age, sex, and body habitus. This criterion is
frequently observed as a reversible abnormality in acute status asthmaticus in the
presence of a normal heart and normal right heart pressures (29,30).
Potential limitations of the Selvester and Rubin criteria have also been cited.
Frontal plane P wave axis and horizontal plane transitional complex shifts may be
transient and thus occur as a result of acute airways obstruction as well as with
emphysematous destruction (23,24). Posterior QRS orientation has also been
shown to correlate with autopsy or other EeG evidence of RVH(2,28,31). Low
123
voltage has been considered an important but not consistent feature of the ECG
in emphysema. However, voltage normally decreases with age (23), myocardial
infarction, and in patients with a vertical (but otherwise normal) heart and no
pulmonary disease (2,25,32). However, the P wave is usually normal in these
conditions (33). Thus, the strong methodology of the Selvester and Rubin system
permits identification of patients with "pure" emphysema and implementation of
multiple criteria for the diagnostic scoring of an individual ECG. These features
are absent in the methods of Littman (3), Wasserburger et al. (19), and others.
Although few patients have "pure" emphysema, many present with emphysema as
the dominant process. Further, the need to establish and assess diagnostic criteria
based upon a "pure" population is obvious.
Spodick's descriptors of "diffuse lung disease" (17,18,34) are presented in table
5-2 and illustrated in figure 5-2. Differences from those of Selvester and Rubin's
system include (a) substitution of "P pulmonale" (P wave > 2.5 mm in leads II,
III, and aVF) or a "Gothic" P wave (single-peaked P of normal size in II, III, and
aVF; (b) inclusion of rightward QRS axis shift or "axis illusion"; (c) absence of
a low voltage criterion; and (d) addition of an abnormally deep S wave persisting
across the left precordium. The frontal plane P wave amplitude is < 2 mm in 99%
of normal adults. Like rightward P axis shift, P pulmonale has been correlated with
abnormal PFT's indicative of obstruction. P pulmonale, but not rightward P axis
shift, has been described in many forms of both pulmonary (35) and nonpulmo-
nary-induced (36) cor pulmonale and is therefore a less specific criterion for
emphysema. Gothic P waves are difficult to defme and can occur in normals. The
specificities of the QRS criteria (b,c,d above) are also uncertain. Rightward shifts
of the QRS axis, like P pulmonale, may occur with either emphysema or cor
pulmonale (3,27,29,32,37-39). "Axis illusion" is defmed as a frontal plane axis
between _910 and 1800 and is not specific for emphysema since it could be due
either to extreme right or to left axis deviation. When due to emphysema, it is
thought to result from either (a) the insulating effect of the lungs concentrating
electrical forces along the vertical axis of the mediastinum (4) (see earlier section)
and/or (b) the emphysema-produced posterior orientation of the mean QRS axis,
so that small superior or inferior alterations of the angle of this vector in the sagittal
plane can project the frontal plane superiorly (- 800 to - 90j or inferiorly (+
800 to +90j (3,17,40). Precordial S waves may also arise from both pulmonary
and cardiac causes.
Spodick reported 93% sensitivity and 96% specificity for these criteria; how-
ever, the population was mixed (emphysema with associated bronchitis, bronchiec-
tasis, or pulmonary fibrosis). Thus, the title "Criteria for Diffuse Lung Disease"
is accurate, and these criteria should not be considered as specific for emphysema
as should the Selvester and Rubin criteria. In general, the performance of the
Spodick criteria is inconsistent, as some studies have shown that low specificity is
achieved in normals and in patients with congenital heart disease, R VH, and/or
RBBB (21,22), while other studies reaffirm the high sensitivity and specificity
originally reported (40,41).
Table 5-2. Spodick criteria for diffuse lung disease
Any of:
1. P pulmonale or "gothic" P waves in II, III and aVF

OR
2. a) Frontal P wave axis ~ + 700
OR
b) ~2 mm S waves in V t -V 6
OR
3. a) Frontal P wave axis +60 to +690
AND
b) ~2 mm S waves in V t -V 6
AND
c) QRS axis vertical or "axis illusion"
.•.-.-.-.. ". -. =
.:: :
•.
I - : : .i AVR
II f-+-1F-F1+-t---tl AVL
AVF ~+++-II-+-H
Figure 5-2. Only the Spodick criteria for "diffuse lung disease" are satisfied. P pulmonale is
present, and S wave amplitude is at least 2 mm in all of the precordial leads (table 5-2).
The P-T wave (3,27,42) (the deflection corresponding to atrial repolarization)

is considered abnormal if there is ;;:: 1.0 mm negative deflection in leads II, III, or
aVF during sinus rhythm either alone or as part of "Wasserburger's Pentalogy"
(11,19) . This finding has been considered pathognomonic for pulmonary emphy-
sema. However, it also occurs in normals with tachycardia and infrequently in
patients with cor pulmonale (2,18,26).
Currently available are methods designed to separate emphysema from cor
pulmonale and from other nonpulmonary problems. The P-wave change of right-
ward axis shift distinguishes emphysema from normal and also somewhat from cor
pulmonale and other causes of R VH. In R VH a rightward P axis will almost
125
always be associated with QRS changes characteristic of R VH (see below). How-

ever, this P-wave abnormality can be affected by exacerbation and remission of the
hyperinflated state. Since R VH is usually associated with anterior and rightward
QRS forces, the presence of posterior orientation of the QRS complex is of value
in diagnosis of emphysema. The combination on ECG of P-wave changes with
diminished QRS voltages will separate emphysema from normal and also from
both cor pulmonale and nonpulmonary heart disease in the absence of emphysema.
Similarly, "axis illusion" occurring with either the P-wave changes or low voltage
may be considered indicative of emphysema.
Diagnosis of cor pulmonale

The diagnosis of cor pulmonale by ECG has been very difficult. The difficulties
have been due to the same methodological problems encountered with emphysema,
and also to the effects of both the normal and abnormal left ventricle. The
preceding discussion of ECG parameters for emphysema indicated that many of
the criteria considered indicative of right-sided enlargement (such as P pulmonale,
deep left precordial S waves, and inferior and rightward QRS axis) may be
mimicked by emphysema alone. The most common criteria for R VH were devel-
oped primarily from subjects with either congenital heart disease or mitral stenosis.
Not surprisingly, therefore, these criteria have performed poorly in patients with
cor pulmonale: they have achieved low sensitivity (often < 50%) when tested in
populations with cor pulmonale proven clinically or by autopsy and attained low
specificity in normals (2,16,21,19,43). Attempts to relate abnormalities in pulmo-
nary arterial pressure, hypoxia, acidosis, R V wall thickness, and so on, with ECG
evidence of cor pulmonale have usually demonstrated a weakly positive correlation
(44-46). Three of the most commonly cited sets of criteria (39,47,48) are presented
in table 5-3, and examples of electrocardiograms meeting these various criteria are
shown in figure 5-3. A brief discussion of the validity of these individual ECG
abnormalities follows.
Right axis deviation (RAD) of the QRS

Although early investigators were hesitant to use RAD as indicative of cor pul-
monale (44,47,49), subsequent postmortem analysis (35,37-39,43) has shown that
RAD is sensitive for R VH due to COPD (25) (approximately 60 to 80%).
Additionally, RAD has a significantly positive correlation to right ventricular
thickness (37,50). The normal adult frontal plane QRS axis has been reported to
lie between -300 and + 1100 (35,38,39) with nearly 99% in one large population
leftward of + 800 (23). Thus, the specificity of "RAD > + 110"" is quite high,
and use of this extreme criterion is responsible for needlessly diminished sensitivity.
Right axis deviation, and especially the magnitude of rightward shift, may fluctuate
with exacerbations in pulmonary arterial pressure (32,46) and! or hyperinflation
(3,27,32).
...
~
Table 5-3. Sampler of Reported BCG Criteria for R VH
Criteria type Sokolow and Lyon (47) I Milnor (39)2 Phillips and Burch (48)3
QRS axis RAD (+110· to -80j RAD (110· to -90j RAD in a radiologically enlarged heart
QRS duration <0.12 Incomplete RBBB with rSR'
Voltage lead I
or II
Abnormal III
deflection aVR R >5.0 mm
aVL R :;: 5.0 mm and !ST and T
aVF R :;: 5.0 mm and !ST and T
V, R ; 7.0 mm or R > 5.0 mm with !ST and T R/S or R' /S > 1.0 R> QorS
S <2.0 mm and
R/S > 1 (age > 5) R or R' > 0.5 m V
intrinsicoid .04 to .07
V2 intrinsicoid.04 to .07
R > 5.0 mm and !ST and T
V3 R ~5.0 mm and !ST and T
V4
Vs R in Vs or V6 :S;4.0 mm
Sin Vs or V6 ~7.0 mm
V6 R/S in Vs or V6 < 1.0
RV, + SVs or SV6 >10.5mm (age >5)
R/S in Vs + R/S in V, :s;0.04
Notes: 1Method of applying system not stipulated.

2QRS duration < 0.12 sec; either axis or voltage criterion.
l"Fairly conclusive of RVH."
127
AVLI I
III AVR .
AVl.
• 1+ .1-:
r ]
.. H:
Figure 5-3. Neither the top set nor the bottom set meets either the Selvester and Rubin or the
Spodick criteria. However, the top set meets both the Sokolow and Lyon and the Milnor
criteria (table 5-3) for RVH, and the bottom set meets the Phillips and Burch criteria for
RVH. The top set shows marked right axis deviation, S wave amplitude < 2 mm, and RIS
ratio > 1 in lead VI: delayed intrinsicoid deflection in lead V2; and R amplitude :0:; 4 mm and
RIS ratio <1 in both Vs and V6 (Sokolow and Lyon) . The QRS duration <0.12 sec and the
marked right axis deviation satisfy the criteria of Milnor. In the bottom set neither the Sokolow
and Lyon nor the Milnor criteria are satisfied; however, the marginal right axis deviation and
incomplete right bundle branch block with prominent R' in VI satisfy the Phillips and Burch
criteria for "fairly conclusive of R VH."
Axis illusion
The potential for frontal plane axis between - 91 0 and 1800 due to a vertical cardiac
position and hyperinflation was discussed previously. Since this phenomenon has
been noted in approximately 10% of patients with cor pulmonale from COPD
(35,39), it is impossible to determine whether it is due to cor pulmonale, concurrent

emphysema, a left ventricular abnormality, or some combination thereo£
RBBB
Incomplete RBBB (characterized by RSR' in lead V t of 0.10-0.11 sec) has been
considered characteristic of cor pulmonale due to the specific problem of volume
overload and resultant crista supraventricularis hypertrophy (15,48). Since the crista
is depolarized relatively late in the cardiac cycle, it is less likely to be obscured by
the larger LV; therefore, incomplete RBBB is a sensitive indicator of cor pul-
monale (2,16,31,35,45). RBBB occurs independent of the degree of airway ob-
struction (18,27) and is present in approximately 15% of patients with cor pul-
monale (35,43). Some studies (16,32,35) have stressed the responsive nature of this
parameter to exacerbations and subsequent remissions of the disease early in its
course. The overall incidence of complete RBBB (~0.121 sec) in cor pulmonale
is much lower (::::;; 5%) and is often associated with coexisting left ventricular
disease (35,39,43).
Voltage measurements
The range of normal limits of QRS amplitudes has been presented by many
investigators (23,47,51,52). The amplitude limits of the right precordial R waves
or the left precordial S waves are rarely exceeded in cor pulmonale due to COPO
until late in the course when significant R V free wall hypertrophy is present
(39,43,48). This is probably due to the effects of coexisting emphysema, as illus-
trated in figure 5-4. When amplitudes of this magnitude (table 5-3) occur, they
are indicative of significant cor pulmonale (37). The R/S criteria are somewhat
more useful (25,32,39,43) (sensitivity and specificity approximately 50%), most
likely because the ratio method circumvents the problem of diminishing absolute
voltage due to emphysema. As indicated by Milnor (39), the efficacy of QRS
amplitude considerations can be enhanced by combination of absolute amplitude
and amplitude ratio criteria (for example, neither R/S in V t > 1.0 nor V t voltage
> 0.5 mV is individually abnormal in a significant number of patients, but their
simultaneous occurrence is almost always abnormal).
Abnormal QRS deflections

Prominent S waves in leads I-III ("SCS2-S3 pattern"), when due to cor pulmonale,
are, like incomplete RBBB, thought to result from hypertrophy of the crista
supraventricularis. The SCS2-S3 pattern may be an artifact of the position of the
recording electrode relative to the descended cardiac position in the hyperinflated
state. While the presence of this fmding correlates with the duration and severity
of cor pulmonale (2,25,43), it has also been observed both in normals and in R VH
due to congenital heart disease (53).
Cor pulmonale may also simulate myocardial infarction by Q-wave deflections
( ~ 0.03 sec), especially in the right precordial leads and in II, III, and aVF
129
lilillll' 'Ii Hi' iii Iii ',ii jl!m~ 1=1: Ii!! ,' .. '!:
, II\! 'II: lilllH ;m ji!,!W I ii i,i! I'; 1i\ I: :,'
I V, H!; ~~ I ~:. Ill; :i!i H,I li:! l!ilm :l:j ,::1 !~i .~:.
IllillliilUlIHlli1 ;~ ': :~i.: ;~. '1:,: hi" :t~f :' • ll~: '::
I
111;1111 :.['WI. . iii III';:' ", llI!:'£!1,

1m in! Hi! Ii! Iii Iii;' I Ii Ii'" I ' :i> 'Iii
I II I'i' i!.' 1:1 ' d\
n Ii!I IIfl
,I'
111 '11':" , I,
;1!,jUl "
.I
-'-'-+ I
' Il ij 'r ~
If !.'
aVL~[~~
':: ~~~,~
\! , ~
Figure 5-4. Criteria for emphysema, diffuse lung disease, and for R VH are all satisfied in this
example.
(43,44,48), as illustrated in figure 5-5. These Q waves have been noted to occur
as frequently as 18% in cor pulmonale (35) and were found to be the most valuable
predictor of RVH in cor pulmonale in a study with autopsy correlation (43) . No
method is available to distinguish such QRS changes due to cor pulmonale from
those of myocardial infarction; therefore, history, comparison with old ECGs, and
the like (54) are important.
S-T segment and T-wave changes

T-wave inversion is a frequent finding in most patients with COPD. Flattened or
inverted T waves in the right precordial leads and in leads II and III may occur
in 10 to 60% of patients with cor pulmonale (31,32,35) . Since the normal T wave
is upright in all of these leads (51 ,52) (with V t normally upright by age 16), this
finding is rarely present in normal subjects. S-T segment depression in these same
leads also occurs with or without concomitant T -wave changes-although much
less often (32,43) . The hallmark of both of these changes is their lability. Fre-
quently, they have been observed during experimental hyperventilation, infusion
n ll-H-H-++1l AVl l**lil$m
AVF
Figure 5-5. Criteria for pulmonary emphysema, common diffuse lung disease, and R VH are all
satisfied by this example. In addition Q waves are present in Vt-V J with minimal R waves in
VC V6 mimicking anterior and lateral myocardial infarction.
of a beta-adrenergic agonist (55,56), and acute status asthmaticus (29). Apparently,

S-T and T abnormalities are the result of an acute decompensation of pulmonary
and/or cardiac status (32) and are mediated via neurohumoral and chemical
(hypoxemia, acidosis) means. These S-T segment and T-wave changes have been
considered to signify RV systolic overload with "strain" (31,57): a correlation with
either autopsy-proven R VH and/or elevated pulmonary artery pressure has been
demonstrated by some studies (32,43), but not others (35,46).
In assessing the merits of these various methods (table 5-3), Milnor's (39) criteria
appears to offer the best combination of practical simplicity and theoretical validity.
However, although his methodology was careful, his population had R VH due
to both pulmonary and nonpulmonary causes. When tested in a large number of
subjects with cor pulmonale by Padmavati et al. (35), the sensitivity of the Milnor
criteria was quite low (15%). In contrast, the combined "fairly conclusive" and
"strongly suggestive" groups of the Phillips and Burch criteria (48) derived from
a population with COPD achieved approximately 60% sensitivity in that same
study (35).
An interesting approach is that of Van Lingen and Palmhert (43), who identified
and ranked 33 criteria for R VH in patients with autopsy-proven cor pulmonale.
They reported high sensitivity and specificity using one or more of these criteria
(predominantly voltage); however, such a system would have applicability in
computerized ECG interpretation only after validation in larger test populations.
Another interesting perspective is provided by Kilcoyne et al. (32), who attempted
to use the severity-related fluctuations in several of the above parameters for early
131
Table 5--4. Reversible ECG criteria for the diagnosis of cor pulmonale (32) 1
With arterial oxygen saturation < 85% and mean pulmonary artery pressure (PAM) ~ 25
mmHg2
1. RAD or rightward shift of > 30· as compared to a prior tracing
2. Flattened or inverted T waves in right precordial leads
3. S-T segment depression in leads II, III, and aVF
4. Transitory RBBB
1. One or more criteria seen; will usually revert with improvement in O 2 saturation and pulmonary hypertension.
2. Calculated from ABG (12) as PAM = 40.363 - O.713H - 3.0085 + +
1028HS, where H = [H J in mEq/1,
S = % arterial hemoglobin desaturation.
diagnosis of cor pulmonale (table 5-4). These changes were largely reversible if
hypoxemia and/or pulmonary hypertension improved.
Asthma
It is interesting to examine the changes that occur during an acute asthmatic attack
as a model for airways obstruction with (relatively) normal heart and lungs.
P-wave changes have been the most frequent abnormality observed, with rightward
P-axis shift and P pulmonale occurring at incidences of 50 to 75% and 20 to 30%,
respectively. These abnormalities correlated well with the degree ofPFT abnormal-
ity (29,30,58,59). The following criteria also occurred in less than 25% of these
acute asthmatic episodes: S-T segment and T-wave changes, posterior QRS orienta-
tion, RAD of the P wave or QRS complex, RBBB, and one or more voltage
criteria for R VH (29,30,58,59). All of these abnormalities were transient in the
majority of patients, resolving in several hours to two weeks. Also, in two studies
of chronic, institutionalized asthmatic patients examined in the nonacute state
(60,61), low P- and R-wave amplitudes were consistent fmdings without other
evidence of either hyperinflation or cor pulmonale.
VCG analysis has been reported to indicate abnormalities in a large number of
asthmatic patients with normal ECGs (59,60,62). In addition to diminished ampli-
tudes in the chronic state, acute abnormalities include posterior orientation of the
QRS loop (30 to 90%), patterns of right atrial and right ventricular enlargement,
and T-loop changes (20%).
ECG IN OTHER STATES PRODUCING PULMONARY HYPERTENSION

Elevated pulmonary arterial pressure in the absence of COPD can be due to
idiopathic vasoconstriction or to phenomena, both resulting in either an intravascu-
lar or muscular obstruction of the pulmonary circulation (for example, sarcoid,
recurrent thromboembolism, chronic pulmonary venous hypertension, etc.) (63-
65). These factors produce severe anatomic restriction and decreased compliance of
the pulmonary vascular bed; pulmonary heart disease is thereby produced primarily
by this elevated pulmonary vascular resistance, and secondarily by the resultant
hypoxemia. The R V is confronted with a systolic overload, compensates initially
with hypertrophy, and proceeds toward failure via dilatation. The ECG reflects
this "classic" cor pulmonale with (a) RAD of the QRS (> 110, in 50 to 100%;
(b) R VH (either Sokolow and Lyon [47] or Milnor [39] criteria) in > 90%; and
(c) T-wave inversion (75%) (63-66). The RVH criteria might be expected to
perform much better in this type of cor pulmonale than in that induced by COPD
due both to the severity of the pressure overload and to the absence of the vagaries
of emphysema. Abnormalities of P-wave axis or amplitude are very uncommon
until late in the disease process, when P pulmonale (but not P-wave RAD) may
occur in about 50% (64,66) of patients. This P pulmonale probably represents
severe and prolonged R V overload, but similar fmdings have been observed in
normals at high altitude (67).
REFERENCES
1. Holford FD. The electrocardiogram in lung disease. In Pulmonary Diseases and Disorders,
Fishman AP (ed). New York: McGraw-Hill, 1980, pp. 139.
2. Burch GE, DePasquale NP. The electrocardiographic diagnosis of pulmonary heart disease. Am
] Cardiol 11: 622-638, 1963.
3. Littman D. The electrocardiographic fmdings in pulmonary emphysema. Am ] Cardiol 5:
339-348, 1960.
4. Grant RP. Left axis deviation. An electrocardiographic-pathologic correlation study. Circulation
14: 233--249, 1956.
5. Rudy Y, Wood R, Plonsey R, Liebman J. The effect of high lung conductivity on electrocardio-
graphic potentials. Results from human subjects undergoing bronchopulmonary lavage. Circula-
tion 65: 440-445, 1982.
6. Flaherty]T, Blumenschein SD, Spock A, Canent R V, Gallie TM, Boineau]p, Spach MS. Cardiac
potentials in pulmonary disease. Overdistention of the lung versus cor pulmonale (right ventricu-
lar hypertrophy). Am] Cardiol 20: 29-38, 1967.
7. Flaherty]T, Blumenschein SD, Alexander AW, Gentzler RD, Gallie TM, Boineau]p, Spach
MS. Influence of respiration on recording cardiac potentials. Isopotential surface-mapping and
vectorcardiographic studies. Am] Cardiol 20: 21-28, 1967.
8. Toyama], Okada A, Nagata Y, Okajima M, Yamada K. Electrocardiographic changes in
pulmonary emphysema. Effects of experimentally induced overinflation of the lungs on QRS
compelxes. Am Heart] 87: 606-613, 1974.
9. Boysen PG, Block A], Wynne ]W, Hunt LA, Flick MR. Nocturnal pulmonary hypertension
10. Matthay RA, Berger H], Davies RA, Loke], Zaret BL, Gottschalk A. Effect of steady-state
exercise on right and left ventricular performance in chronic obstructive pulmonary disease. Chest
77: 303, 1980.
11. Fishman AP, Fritts HW]r, Cournand A. Effects of breathing carbon dioxide upon the pulmonary
circulation. Circulation 22: 220-225, 1960.
12. Enson Y, Giutini C, Lewis ML, Morris TQ, Ferrer MI, Harvey RM. The influence of hydrogen
ion concentration and hypoxia on the pulmonary circulation.] Clin Invest 43: 1146-1162, 1964.
13. Harvey RM, Enson Y, Betts R, Lewis ML, Rochester DF, Ferrer MI. Further observations on
the effects of hydrogen ion on the pulmonary circulation. Circulation 35: 1019-1027, 1967.
14. Cabrera E, Monroy]R. Systolic and diastolic loading of the heart. I. Physiologic and clinical
data. Am Heart] 43: 661-668, 1952.
15. Cabrera E, Monroy]R. Systolic and diastolic loading of the heart. II. Electrocardiographic data.
Am Heart] 43: 669-686, 1952.
16. Ferrer MI. Cor pulmonale (pulmonary heart disease): Present-day status. Am Heart] 89: 657-664,
1975.
17. Spodick DH. Electrocardiographic studies in pulmonary disease. I. Electrocardiographic abnor-
malities in diffuse lung disease. Circulation 29: 1067-1072, 1959.
18. Spodick DH, Hauger-Klevene ]H, Tyler ], Muench H, Dorr CA. The electrocardiogram in
pulmonary emphysema. Relationship of characteristic electrocardiographic fmdings to severity
of disease as measured by degrees of airway obstruction. Am Rev Respir Dis 88: 14, 1963.
133
19. Wasserburger RH, Kelly JR, Rasmussen HK, Juhl JH. The electrocardiographic pentalogy of
pulmonary emphysema. A correlation of roentgenographic findings and pulmonary function
studies. Circulation 20: 831-841, 1959.
20. Phillips R W. The electrocardiogram in cor pulmonale secondary to pulmonary emphysema. A
study of 18 cases proved at autopsy. Am Heart J 56: 352-371, 1958.
21. Selvester RH, Rubin HB. New criteria for the electrocardiographic diagnosis of emphysema and
cor pulmonale. Am Heart J 69: 437-447, 1965.
22. Cowdery TS, Wagner GS, Pratt PC, Pressley JC, Ideker RE, Selvester RH. The evaluation of
quantitative electrocardiographic criteria for pulmonary emphysema. Manuscript in preparation.
23. Lamb LE. Electrocardiography and Vectorcardiography. Philadelphia: W.B. Saunders, 1965, pp.
95-166.
24. Chapman TT. Lung function tests and a vertical P wave axis in the electrocardiogram. Thorax
29: 106-109, 1974.
25. Kamper D, Chou T-C, Fowler NO, Witt RL, Bloomfield S. The reliability of electrocardio-
graphic criteria of chronic obstructive lung disease. Am Heart J 80: 445-452, 1970.
26. Calatayud JB, Abad JM, Khoi NB, Stanbro WW, Siler HM. P wave changes in chronic
obstructive pulmonary disease. Am Heart J 79: 444-453, 1970.
27. Caird Fl, Wilcken DEL, Williams RS. The electrocardiogram in diffuse interstitial disease of
the lung and its relation to pulmonary function. Am J Cardiol 10: 14--19, 1962.
28. Ray CT, Horan LG, Flowers NC. An early sign of right ventricular enlargement. J Electrocardiol
3: 57-64, 1970.
29. Rebuck AS, Read J. Assessment of management of severe asthma. AmJ Med 51: 788-798, 1971.
30. Gunstone RF. Right heart pressures in bronchial asthma. Thorax 26: 39-45, 1971.
31. Coelho E, Amram SS, E Sa AB, Mendes JCF, Tavares V. Electrocardiographic and vectorcardio-
graphic alterations in chronic cor pulmonale. Am J Cardiol 10: 20-29, 1962.
32. Kilcoyne MM, Davis AL, Ferrer MI. A dynamic electrocardiographic concept useful in the
diagnosis of cor pulmonale. Result of a survey of 200 patients with chronic obstructive pulmo-
nary disease. Circulation 42: 903-924, 1970.
33. Wool P. Electrocardiographic appearance in acute and chronic pulmonary heart disease. Br Heart
J 10: 87, 1948.
34. Spodick DH. Electrocardiographic studies in pulmonary disease. II. Establishment of criteria for
the influence of diffuse lung disease. Circulation 20: 1073-1078, 1959.
35. Padmavati S, Raizada V. Electrocardiogram in chronic cor pulmonale. Br Heart J 34: 658-667,
1972.
36. Chou T-C, Helm RA. The pseudo P pulmonale. Circulation 32: 96-105, 1965.
37. Millard FJC. The electrocardiogram in chronic lung disease. Br Heart J 29: 43-50, 1967.
38. Soffer A. Range of frontal plane QRS axis. Electrocardiograms of subjects in multiphasic
screening program. Chest 72: 477-479, 1977.
39. Milnor R W. Electrocardiogram and vectorcardiogram in right ventricular hypertrophy and right
bundle branch block. Circulation 16: 348-367, 1957.
40. Shmock CL, Pomerantz B, Mitchell RS, Pryor R, Maisel JC. The electrocardiogram in emphy-
sema with and without chronic airways obstruction. Chest 60: 328-334, 1971.
41. Shmock CL, Mitchell RS, Pomerantz B, Pryor R, Maisel JC. The electrocardiogram in chronic
airways obstruction. The role of bronchitis and emphysema. Chest 60: 335-340, 1971.
42. Rasmussen HK, Juhl JH. The T-A wave of the adult electrocardiogram. An expression of
pulmonary emphysema. Am Heart J 54: 875-886, 1957.
43. Van Lingen B, Palrnhert R. The electrocardiogram in isolated right ventricular hypertrophy due
to chronic respiratory disease. S Afr Med J 48: 1480, 1974.
44. Meyers GB, Klein HA, Stofer BE. The electrocardiographic diagnosis of right ventricular
hypertrophy. Am Heart J 35: 1, 1948.
45. Johnson JB, Ferrer MI, West JR, Cournand A. The relation between electrocardiographic
evidence of right ventricular hypertrophy and pulmonary arterial pressure in patients with
chronic pulmonary disease. Circulation 1: 536-550, 1949.
46. Tobis JM, Danahy DT, Aronow WS, Chetty K, Glauser F. Comparison of noninvasive with
hemodynamic data in patients with pulmonary hypertension due to chronic obstructive pulmo-
nary disease. Am J Med Sci 278: 19-25, 1979.
47. Sokolow M, Lyon TP. The ventricular complex in right ventricular hypertrophy as obtained by
unipolar precordial and limb leads. Am Heart J 38: 273-294, 1949.
48. Phillips JH, Burch GE. Problems in the diagnosis of cor pulmonale. Am Heart J 66: 818-832,
1963.
49. Roman GT Jr, Walsh 1], Massie E. Right ventricular hypertrophy. Correlation of electrocardio-
graphic and anatomic fmdings. Am J Cardiol 7: 481-487, 1961.
50. Mazzoleni A, Wolff R, Wolff L, Reiner L. Correlation between component cardiac weights and
electrocardiographic patterns in 185 cases. Circulation 30: 808-829, 1964.
51. Simonson E. Differentiation Between Normal and Abnormal in Electrocardiography. St. Louis:
C.V. Mosby, 1961.
52. Van Lingen B, Palmhert R. New standards for electrocardiograms of adult males without
evidence of cardiac hypertrophy. S Afr Med] 48: 1413, 1974.
53. Lipman BS, Massie E. Clinical Scalar Electrocardiography. Chicago: Year Book, 1959, p. 205.
54. Phillips ]H, Burch GE. Selected problems in the management of emphysema complicated by heart
disease in addition to cor pulmonale. Am Heart J 64: 828-841, 1962.
55. Lary D, Goldschlager N. Electrocardiographic changes during hyperventilation resembling myo-
cardial ischemia in patients with normal coronary arteriograms. Am Heart] 87: 383-390, 1974.
56. Biberman L, Sarma RN, Surawicz B. T wave abnormalities during hyperventilation and iso-
proterenol infusion. Am Heart] 81: 166-174, 1971.
57. Armen RN, Kantor M, Weiser NJ. Pulmonary heart disease with emphasis on ECG diagnosis.
Circulation 17: 164-175, 1958.
58. Siegler D. Reversible electrocardiographic changes in severe acute asthma. Thorax 32: 328-332,
1977.
59. Rao M, Steiner P, Paydar M, Padre R, Torres RR. Electrophysiological changes in the heart
of children with bronchial asthma during the acute attack and later during remission.] Asthma
Res 11: 113, 1974.
60. Quivers WW, Linde LM, Sapin SO, Heimlich EM. The electrocardiogram and vectorcardiogram
in children with severe bronchial asthma. Am] Cardiol 14: 616-621, 1964.
61. Miyamoto T, Kass I, Hoffman MS. Electrocardiogram in asthmatic children. Circulation 22:
90--95, 1960.
62. Criep LH, Silverblatt M. The effect of bronchial asthma on the heart, with special reference to
the spatial vectorcardiogram.] Allergy 31: 191-210, 1960.
64. Sleeper ]C, Orgain ES, McIntosh HD. Primary pulmonary hypertension. Circulation 26: 1358-
1369, 1962.
65. Walcott G, Burchell HB, Brown AL. Primary pulmonary hypertension. Am] Med 49: 70--79,
1970.
66. Thadani U, Burrow C, Whitaker W, Heath D. Pulmonary veno-occlusive disease. Quart] Med
44: 133-159, 1975.
67. Penaloza D, Gamboa R, Marticorena E, Echevarria M, Dyer ], Gutierrez E. The influence of
high altitudes on the electrical activity of the heart. Electrocardiographic and vectorcardiographic
observation in adolescence and adulthood. Am Heart J 61: 101-115, 1961.
6. ROENTGENOGRAPHIC EVALUATION OF PULMONARY HEART
DISEASE
CARL E. RAVIN, M.D.
Despite the virtual explosion of newer imaging modalities and the continued
importance of history, physical examination, and electrocardiography, the plain
chest radiograph remains one of the most important methods for evaluating the
patient with pulmonary and associated cardiac disease. The chest radiograph pro-
vides one of the most reproducible methods for evaluating the heart, lungs, and
pulmonary vasculature. Morphologic alterations in these structures provide signifi-
cant clues not only to underlying pathologic changes but also to physiologic
changes that might otherwise be difficult, if not impossible, to detect. Accurate
interpretation of the chest fum requires an understanding of normal anatomy and
physiology as well as of potential pathologic alterations. When combined with the
remainder of the patient's clinical and laboratory evaluation, the chest radiograph
frequently enables the physician to make an accurate diagnostic evaluation of a
variety of pathophysiologic alterations.
THE NORMAL CHEST RADIOGRAPH

Accurate interpretation of the chest radiograph in patients with pulmonary heart
disease is based on an understanding of the normal radiograph (1). We strongly
believe that the standard radiographic examination of the chest should include both
a frontal and lateral projection, as these two views, taken at right angles to one
L.J. Rubi. (ed.), Pulmonary Heart Disease. All rights reserved. Copyright @ 1984 Marlinus NijiroffPubiishing. Boston/The Hague/Dor-
drecht/Lancaster.
135
136 6. Roentgenographic Evaluation of Pulmonary Heart Disease
another, potentially provide for a complete look at the underlying thorax and
intrathoracic contents. Obtaining only a single frontal view may result in the
missing of significant lesions, particularly in the anterior mediastinum and the
retrocardiac and subdiaphragmatic portions of the lung.
Frontal Projection
By convention, the routine chest radiograph is obtained at a six-foot focal-film
distance (distance between the x-ray tube and the film). In order to keep magnifica-
tion of the heart to a minimum, the heart is placed as close as possible to the film
so that the x-ray beam passes through the patient from back to front (posterio-
anterior or PA film). Turning the patient so that the back is close to the film and
the beam passes from front to back (antero-posterior or AP film) will result in some
magnification of cardiac structures because they are now further from the film. In
addition, it is generally true that when an AP film is obtained, the focal-film
distance is frequently less than six feet, again contributing to increased magnifica-
tion of the cardiac image. Inasmuch as cardiac size constitutes a major radiographic
clue to the pathophysiologic status of the heart, recognition of magnification effect
and spurious increase in heart size is of considerable importance. One way in which
this can be judged is to recognize the difference between PA and AP film. Because
of factors related to patient positioning and divergence of the x-ray beam from
its source at the x-ray tube, determination of whether a film has been obtained PA
or AP can often be made by analysis of the lower cervical spine (2) (figure 6-1).
In a PA film the posterior arches of the lower cervical spine are generally seen,
whereas with an AP film the vertebral bodies of the lower cervical spine will
usually be imaged. If the film is obtained in a PA projection, it will generally be
done at a six-foot focal-fum distance. On the other hand, if the film is obtained
in an AP projection, some spurious magnification of the cardiac shadow is likely
to occur.
One also needs to evaluate the film for technical adequacy. To begin with, the
appropriateness of the overall radiographic exposure should be analyzed. Films that
have been "overexposed" will tend to have black lung fields, and, as a result, it
will be difficult to correctly analyze the pulmonary vasculature or to visualize
changes in the pulmonary_ parenchyma. Films that have been "underexposed" will
show the mediastinum to be too white, and lesions behind the heart or in the
subdiaphragmatic areas of the lung will be impossible to visualize. A film is
generally appropriately exposed if one can just see the vertebral bodies and inter-
vertebral disc spaces through the heart. In overexposed films the vertebral bodies
and intervertebral disc spaces stand out very clearly, and in underexposed films they
cannot be seen at all. If one is just able to visualize them, however, the film will
generally have been appropriately exposed and the risks of missing lesions in the
retrocardiac and subdiaphragmatic areas reduced.
In addition to checking exposure, one should evaluate the film for rotation and
ensure that the patient has made an adequate inspiratory effort. Rotation is most
easily evaluated by noting the relationship between structures known to be anterior
A
B
Figure 6-1. Recognizing a PA versus AP film.
A. On the P A film the posterior arches of the lower cervical spines are imaged.
B. On the AP film, however, the vertebral bodies of the lower cervical spine are better
visualized.
and those known to be posterior. In the chest the medial ends of the clavicles are
readily recognizable anterior structures that parallel the midline, and the dorsal
spines of the vertebral bodies are easily identifiable midline posterior structures. In
a nonrotated film the dorsal spines will project midway between the medial ends
of the clavicles.
Recognition of rotated films is extremely important, as rotation will distort the
normal mediastinal and cardiac contours; if unrecognized, rotated films can inap-
propriately suggest enlargement of various cardiac chambers or mediastinal struc-
tures (figure 6-2). In addition, rotation frequently tends to make the lung on the
side toward which the patient is rotated appear hyperlucent (figure 6-3). Such
hyperlucency, if real, would potentially be of great significance and certainly
would require further investigation. Therefore, one should recognize that such
hyperlucency can be due to technical factors (rotation) and is not necessarily related
to underlying pathologic alteration.
With regard to satisfactory inspiratory result, accurate assessment can generally
be made by noting the number of visualized ribs. In general, an adequate inspiratory
result has been achieved if nine to ten posterior ribs or six or seven anterior ribs
are visualized above the level of the hemidiaphragm. When one sees eleven or
twelve ribs posteriorly, the possibility of hyperinflation should be considered
strongly. However, there is a potential difficulty with equating hyperinflation on
the chest film with obstructive lung disease, since the patient is instructed to take
as deep a breath as possible and to hold it for the fUm. Athletic individuals or those
engaging in regular exercise programs frequently can take a very deep breath, deep
enough to drive the diaphragm to a low level and, indeed, to even flatten its
mimicking the appearance often seen with chronic obstructive pulmonary disease.
If, however, the patient is not an athlete or regularly engaging in a physical fitness
program, observation of a significant inspiratory result on the chest film should
suggest the possibility of obstructive lung disease.
Several chambers of the heart routinely can be assessed on the frontal projection
(figure 6-4A). The right cardiac border is formed by the right atrium and generally
projects slightly to the right of the vertebral column. It is in anatomic contact with
the adjacent right middle lobe. The left-sided contours from above downward are
formed by the aortic arch, the pulmonary artery, the left atrial appendage, and the
left ventricle. The left ventricle is in anatomic contact with the lingula. The right
ventricle cannot be directly assessed on the frontal projection as it is not border
forming. The left atrium, which is also not border forming in general, can also
not be directly assessed. Occasional clues to left atrial size are available, however,
by noting enlargement of the left atrial appendage or alterations in the bronchial
angle. In general, the atrium must increase in size by two to three times before it
can be recognized radiographically.
Lateral projection
The lateral view provides a right-angle projection to the routinely obtained frontal
view and, as such, is a most important view of the chest. In truth, when reviewing
A
Figure 6-2. Effects of rotation on the appearance of mediastinal and cardiac contours.
A. The patient is rotated slightly to the left, accentuating the left cardiac silhouette.
B. The same patient is rotated to the right, markedly changing the appearance of the mediastinal
and cardiac silhouettes. The right-sided cardiac structures now appear far more prominent than
they did in A.
...
t5
?'
i
1
i"tTl
<
e-
..,c::
g.
:;
g.,
c::
§'"
~
..::!
:r:
g
tl
~.
A
Figure 6-3. Effect of rotation. This three-part figure illustrates the effect B. With the phantom turned toward the right. the right lung appears
of rotation on the appearance of the final chest radiograph. The studies more lucent than the left.
were perfonned on a phantom so that changes in the configuration of C. With the phantom rotated to the left. the left lung appears more
soft tissues cannot be responsible for the changes seen. lucent than the right. This series of films indicates that rotation can result
A. Equal lucency of both lungs is demonstrated with the phantom in the spurious appearance of hyperIucency in the lung toward the side
position without rotation. which the patient is rotated.
141
chest radiographs, most physicians spend the vast majority of their time and effort
analyzing the frontal projection and rarely spend much time with the lateral view.
This approach seems attributable to the fact that they do not understand the
anatomy as projected on the lateral view and, therefore, are uncomfortable with
its analysis. The primary confusion on the lateral view occurs in the area of the
hilum. The normal anatomic structures in the hilum are illustrated in figure 6-4B.
The lateral view is a very useful projection for evaluating the size of the pulmonary
artery, which, as will be seen later, can provide significant clues as to alterations
in pulmonary blood flow and/or pressure. The lateral view is extremely valuable
in evaluation of the "prominent hilum," in that it allows distinction between an
enlarged hilum due to arterial prominence and one due to adenopathy and/or mass
(figure 6--5). In the former situation, the pulmonary arteries are enlarged but the
normal, relatively radiolucent area beneath the left upper lobe bronchus remains
clear. The overall appearance is that of a "horseshoe" (figure 6--5B). In the setting
of adenopathy and/or mass, the area projected beneath the left upper lobe bronchus
is frequently radio-opaque and the overall appearance resembles a "doughnut"
(figure 6--5D).
The lateral view permits direct analysis of the two cardiac chambers, which were
not well visualized on the frontal projection because they were not border forming
(figure 6-4B). That is, the right ventricle, positioned anteriorly projects directly
behind the sternum on the lateral view. The left atrium, which sits behind the rest
of the cardiac mass and is not normally border forming on the frontal view, can
also be readily visualized on the lateral view. Visualization is enhanced by filling
the esophagus, which is in anatomic contact with the left atrium, with barium. In
addition, the left ventricle can be seen forming the lower cardiac border in the
lateral view.
Evaluating the size of the left ventricle on the lateral view is frequently possible
by applying a sign described by Hoffman and Rigler in 1965, which has come to
be known as "Rigler's sign." In their initial article (3), Hoffman and Rigler
described two measurements that could provide clues as to the left ventricular
enlargement. The measurement most often referred to (their measurement A) is
defined as the distance which the left ventricle extends posteriorly to the posterior
border of the inferior vena cava measured at a point 2 em cephalad to the crossing
of the vena cava and the left ventricle. The other sign they described, which has
not received as much attention (measurement B), is defmed as the distance of the
crossing of the inferior vena cava and the left ventricle caudad to the left leaf of
the diaphragm. In their study they noted that if measurement A was 1.8 em or
greater, there was a strong probability that the left ventricle was enlarged. If a good
inspiratory film had been obtained and measurement B was noted to be 0.75 em
or less, the left ventricle also was probably enlarged. Although the measurements,
particularly measurement A, are frequently referred to, a number of factors poten-
tially limit their application. To begin with the lateral film must meet several
requirements in order to utilize the sign. The film must be a true lateral projection
(without any rotation), obtained in deep inspiration and penetrated sufficiently to
142
c
Figure 6-4. Cardiac anatomy as demonstrated on routine radiographic pulmonary artery (LPA) is seen arching over the left mainstem bronchus.
projections.
C. The left posterior oblique view is demonstrated. Here the chest is
A. A frontal projection demonstrates that the border-forming structures rotated so that the interventricular septum is turned on end. The right
are the right atrium (RA) on the right and, from the top down on the ventricle (RV) is border forming on the right side of the cardiac
left, the aortic arch (AO), pulmonary artery (PA), and left ventricle silhouette and the left ventricle (LV) border forming to the left.
(LV).
D. On the right posterior oblique view, the septum has been rotated so it
B. The lateral projection demonstrates the right ventricle (RV) border is oriented en face and the two ventricles project over one another. The
forming in the retrosternal space. Posteriorly, one can identify the inferior pulmonary artery (PA) and outflow tract form the upper portion of the
vena cava (IVC), the left ventricle (LV), and the left atrium (LA). In the cardiac border and are well seen in this projection. The left atrium (LA)
hilar region the right pulmonary artery (RPA) is seen in front of the is turned so that it may be visualized as well.
lucent circle formed by the left mainstem bronchus (arrow), and the left
...e
144
ID
c o
Figure 6-5. Radiographic analysis of the "prominent" hilum. Figures C and 0 illustrate the prominent hilum secondary to adenopathy
and mass. On the frontal view (C) the etiology of the prominent hilum
Figures A and B demonstrate a prominent hilum secondary to enlarged
(that is, enlarged arteries vs. adenopathy) is again unclear. However, on
pulmonary arteries. While the etiology of the large hilum on the frontal
the lateral view (0), the normal "horseshoe" configuration is now seen
projection is difficult to diagnose, the lateral projection demonstrates the
to be obliterated by a large mass (arrows) giving an overall configuration
classic "horseshoe" configuration formed by the right (RPA) and left
of a "doughnut," with the normal lucent left mainstem bronchus
(LPA) pulmonary arteries with relative radiolucency beneath the level of
occupying the position of the "hole" in the doughnut.
the left mainstem bronchus (arrow). ....
""
'"
see the retrocardiac space clearly. Rotation and suboptimal inspiration can both
create false positive results. In addition, enlargement of the right ventricle often
displaces the left ventricle posteriorly, giving the false impression ofleft ventricular
enlargement. Even when none of the above factors is present, the original study
by Hoffman and Rigler had an 11 % false positive rate and a 19% false negative
rate for measurement A. Measurement B had a false positive rate of 41 % and a
false negative rate of 32%. Therefore, our conclusion has been that although the
Rigler's sign may occasionally be helpful in analysis of left ventricular size on the
chest radiograph, a number of false positives and false negatives are associated with
its use even if the films are technically appropriate for its application. In addition,
one must be particularly cautious applying this sign in situations in which the film
is rotated or when there has been a suboptimal inspiratory result or when the right
ventricle is enlarged. This last consideration (right ventricular enlargement) is
obviously of particular concern in patients with obstructive pulmonary disease and
cor pulmonale.
Oblique views
While frontal and lateral views are the standard projections obtained for analysis
of both pulmonary and cardiac structures, additional information can often be
obtained from oblique views (figures 6-4C and 6-4D). This is particularly true
in analysis of the heart and great vessels. The interventricular septum runs at an
angle of approximately 45° to the plane of the x-ray fum; therefore, rotating the
patient toward the 45° angle will enable one to look "down the septum" and cause
the right ventricle to form one cardiac border and the left ventricle to form the
other (figure 6-4C). This view (known as a left posterior-antero oblique) is
obtained by placing the patient's left side against the fum cassette and rotating the
right side out at an angle of 45° with the x-ray beam passing from back to front.
The 45° obliquity may be increased to 50 or 55° on occasion to avoid projecting
the heart over the spine. Ordinarily, the left ventricle will project away from the
spine, and the right ventricle will be seen to form a smooth convexity with the
ascending portion of the aortic arch. This view provides maximum clarity of the
bifurcation of the trachea, the arch of the aorta, and the posterior basilar portion
of the left ventricle. Although both ventricles can be evaluated in this view, the
standard lateral view often provides similar information, in that the right ventricle
is border forming in the retrosternal space and the left ventricle forms the lower
portion of the cardiac shadow.
The right posterior-antero oblique view of the chest (figure 6-4D)---obtained
by placing the patient's right side against the film and angling the left side back
45°, with the beam passing from back to front-places the interventricular septum
parallel to the film plane. In this situation both ventricles are projected over one
another. This view demonstrates to good advantage the proximal portions of the
main pulmonary artery as well as the left atrium.
Both of these views may be of help in assessing cardiac chamber enlargement,
but were of more importance historically, prior to the advent of echocardiography.
147
With the latter modality, to directly assess chamber size is now possible; therefore,
use of oblique views that provide only indirect evidence of specific chamber
enlargement is not required as commonly as in the past. The views remain useful
in situations in which echocardiographic or other direct assessment of cardiac
chambers is not available. The major views, however, with which all physicians
should be familiar, are the frontal and lateral projections discussed above.
THE ABNORMAL CHEST RADIOGRAPH IN PULMONARY HEART DISEASE
Radiographic features of chronic obstructive pulmonary disease

Identification of patients suffering from chronic obstructive lung disease (COPD)
utilizing the radiograph alone is frequently difficult and at times may be impossible.
Classically, three radiographic signs have been associated with COPD (4-6): (a)
hyperinflation (excess air in the lungs), (b) alterations in pulmonary vascularity
such that there are localized areas of oligemia, and (c) bullae. These three criteria,
although still widely used to suggest a diagnosis of COPD, clearly are neither
particularly sensitive nor specific in establishing the diagnosis (7). To begin with,
"hyperinflation" or excess air in the lungs is difficult to assess. The problem is
inherently difficult because one is attempting to assess obstruction to air flow from
a static film. In essence, assessment assumes that if the lungs are hyperinflated on
the radiograph, the patient cannot effectively empty them, but this sort of reasoning
is clearly circuitous and potentially filled with inaccuracies. The situation is compli-
cated further by the fact that chest films are exposed at total lung capacity; that
is, the patient is asked to take as deep a breath as possible and to hold it while the
film is made. The patient with good respiratory mechanics may be able to take quite
a large breath and appear to be hyperinflated on the resulting radiograph. There-
fore, one must evaluate apparent hyperinflation with particular caution in athletic
patients as well-trained athletes can frequently mimic all of the radiographic
features of hyperinflation---even those with only moderate athletic training may
appear to be hyperinflated. As noted previously, a good inspiration is generally
thought to have resulted when nine or ten posterior ribs or six to seven anterior
ribs are exposed above the level of the hemidiaphragm. If greater than this number
of ribs are exposed, one may suspect hyperinflation (figure 6--6). Some investigators
(8) have suggested that hyperinflation and emphysema are most certainly present
when the diaphragms are inverted (figure 6-7). This is generally the case, but if
one requires diaphragmatic inversion before diagnosing hyperinflation, one will
miss many patients with underlying emphysema; thus the criteria seem unusually
strict, although relatively specific.
In addition to diaphragmatic depression or inversion, which is commonly taken
as a sign of hyperinflation, the lateral view frequently will show an increase in the
"retrostemal clear space" in the setting of hyperinflation. Unfortunately, this sign
is very dependent on configuration of the bony thorax and may be mimicked in
normal patients who have a "dorsum rotundum" or kyphotic deformity of the
thoracic spine. Simon and co-workers (9) have noted that the width of the
Figure 6-6. Classic radiographic fmdings of chronic obstructive pulmonary disease. There is
hyperinflation of the lungs. areas of focal oligemia. with a tendency for the vessels to be
markedly diminished in the periphery of the lungs. The heart is small and narrow as it has been
elongated by depression of the hemidiaphragms secondary to hyperinflation.
retrosternal space must be 4.5 em or greater before it can be reliably considered

an indicator of obstructive airways disease. When the retrosternal space is this large,
flattening of the hemidiaphragm is generally also noted so that the sign is rarely
of use by itself.
With regard to focal oligemia, there clearly are other causes for destruction of
localized areas of the vascular bed in addition to emphysema. If areas of focal
oligemia are identified in patients with apparent hyperinflation, however, the
criteria become much more specific (figure 6-8). Unfortunately, evaluating such
oligemia is frequently difficult unless it is relatively widespread and severe, in which
case the diagnosis in general is no longer a mystery either clinically or radiograph-
ically.
Finally, bullae, although frequently seen in association with emphysema and
other chronic obstructive pulmonary diseases, are, in and of themselves, not
sufficient criteria to establish a diagnosis of COPD. Thurlbeck and others (4,10)
have pointed out that bullae may exist in otherwise normal lungs and that their
identification, therefore, does not establish the diagnosis of COPD. However, the
A B
Figure ~7. Classic radiographic features of emphysema. There is marked The heart is relatively narrow due to the depressed position of the
hyperaeration of the lungs with flattening and inversion of the diaphragm. The vascular distribution is no longer normal and there are
hemidiaphragms. On the frontal view (figure A), the costal insertions of both focal areas of oligemia as well as some change in normal vascular
the slips of diaphragmatic muscle can be visualized (arrow), while on the distribution pattern. ....
lateral view (figure B), there is almost inversion of the hemidiaphragms. ~
Figure 6-8. Classic fmdings of chronic obstructive pulmonary disease with hyperaeration of the
lungs, flattening of the hemidiaphragrns, and focal areas of oligemia. The oligemia reflects areas
of pulmonary parenchymal destruction secondary to emphysema.
presence of bullae in conjunction with apparent hyperinflation and focal oligemia

lends further credence to the diagnosis of COPO (figure 6-9) .
In summary, although the classic signs of hyperinflation, focal oligemia, and
bullae when seen together in the appropriate clinical circumstance can certainly
establish the diagnosis of chronic obstructive pulmonary disease, one should recog-
nize that these signs are neither particularly sensitive nor individually specific for
this disease process.
One additional radiographic fmding occasionally seen in association with
chronic obstructive pulmonary disease is the so-called "saber-sheath" trachea. Oe-
scribed by Greene in 1978, this static defonnity, consisting of marked coronal
narrowing associated with sagittal widening, was found to have a strong association
with chronic obstructive pulmonary disease (11) . Of 60 patients with the saber-
sheath trachea configuration, 57 (95%) had clinical evidence of COPO. Of these
57 patients 26 (45%) lacked the other conventional radiographic criteria generally
associated with COPO. Thus, the saber-sheath configuration of the trachea may
be very useful in the radiographic identification of patients with COPO. The
deformity consists of marked narrowing of the trachea in the frontal projection
beginning abruptly at the level of the anterior thoracic inlet (figures 6-10A and
A
B
Figure 6-9. Severe bullous lung disease. The large bullae compress the remaining lung
parenchyma (figure A). This compression is dramatically illustrated by the pulmonary
arteriogram (figure B) showing how the remaining parenchyma of the lung is displaced against
the heart and mediastinum by the large bullae.
III
c
Figure 6-10. Saber-sheath trachea. PA (figure A) and lateral (figure B) front to back plane. This is better illustrated on the close up views in
view of the chest demonstrate marked narrowing of the trachea (arrows) figures C and D. ....
from side to side at the level of the thoracic inlet and widening in the
'"'"
6-10C). Associated with the narrowing on the frontal projection is apparent

widening of the trachea in the sagittal plane, as seen on the lateral view (figures
6-10B and 6-10D). The changes begin at the anterior thoracic inlet, and the
configuration suggests that increased thoracic pressure may be responsible for the
deformity. However, postmortem examinations show that the trachea is rigid and
maintains its abnormal shape even when intrathoracic pressure is not increased.
Radiographic features of pulmonary arterial hypertension

"Redistribution" of pulmonary Mood .flow
The earliest changes of pulmonary arterial hypertension are best recognized radio-
graphically in the pulmonary vascular bed. Normally, because the pulmonary
vascular bed is a low-pressure, low-resistance circuit, the effects of gravity are
sufficient to divert blood flow toward the lung bases when the patient is standing
erect (12,13). As a result, on the chest radiograph exposed with the patient standing
erect, there are larger and more numerous blood vessels in the lung bases as
compared to those seen in the upper lungs (figure 6-11). In general, the size of
vessels at the lung base exceeds those at the lung apex by two to three times. If
the patient is placed supine, removing the gravitational effects that normally exist
from apex to base with the patient standing erect, vessel size and number will now
appear more nearly equal between upper and lower lung zones (figure 6-12). This
change reflects recruitment of previously underutilized vessels in the upper lung
zone and reflects the fact that, in persons standing erect, generally a considerable
amount of pulmonary vascular reserve is available in the upper lung zones for
recruitment should the need arise. For example, as pressure in the pulmonary circuit
increases, the critical opening pressure of previously unused pulmonary vessels is
exceeded and flow begins to be diverted toward the upper lung (since this is where
the largest amount of vascular reserve exists in a normal individual standing erect).
Therefore, the earliest change evident radiographically in pulmonary arterial hy-
pertension is apparent equalization of flow between upper and lower lung zones,
which is evidenced as an increase, relative to normal, in the size and number of
vessels seen in the upper lungs (14). Similar changes can also be seen in a variety
of situations in addition to pulmonary arterial hypertension. The most common
of these include pulmonary venous hypertension, states in which there is increased
pulmonary blood flow through the lungs such as a left-to-right intracardiac shunt
(figure 6-13), and pulmonary parenchymal destruction.
Similar changes are also seen, however, in situations in which the lungs and
pulmonary vascular bed are affected diffusely and in which specifically there is no
predilection for the underlying disease process to affect the lung bases. Obviously,
if the vascular bed at the lung bases is selectively destroyed by the disease process,
an obligatory redistribution of pulmonary blood flow to the upper lungs will result
(figure 6-14). This is true for diseases such as pulmonary emboli (which, because
most of the blood flow generally goes to the bases, are more common in the lower
lung zones), Eisenmenger's physiology in reaction to left-to-right shunts (again
A
Figure 6-11. Normal distribution of pulmonary vascularity. The effect of gravity superimposed
on the low-resistance, low-pressure pulmonary vascular bed is illustrated by the normal chest
radiograph shown in figure A. Note that the vessels at the lung bases are two to three times the
size of those in the upper lung zones. Figure B demonstrates how pulmonary vascular
distribution may be analyzed by evaluating vessels along a line (arrowheads) equidistant from
the hilum. Again, the size and number of vessels at the lung bases is normally significantly
greater than those seen in the upper lung zones. Alterations in this normal relationship should
suggest underlying pathophysiologic abnormality.
156
c
Figure 6-12. Figure A demonstrates the normal pattern of pulmonary with the patient standing erect (E) can be contrasted against the larger
vascular distribution with the patient standing erect (E). Note that the and more numerous vessels noted when the patient is placed supine (S)
vessels are larger and more numerous at the lung base than in the lung and flow redistributes to the upper lung zone.
apex as flow is diverted by gravitational effects superimposed on the Figure D demonstrates the changes associated with redistribution in the
low-resistance, low-pressure pulmonary circuit. lower lobes as there are fewer vessels seen at the lung base with the
Figure B shows that with the patient placed supine (S) there is a patient supine (S) as contrasted with the larger, more numerous vessels
tendency toward equalization of pulmonary vascular distribution between seen when the patient is standing erect (E) and there is more flow to the
upper and lower lung zones. lung base. As illustrated here, the changes of "redistribution" are much
easier to recognize in the upper lung zones than at the lung bases.
Figure C illustrates to better advantage the changes noted with close-up
views of the left upper lung zone. The relative sparsity of vessels seen
...~
.....
00
""
Figure 6-13. Redistribution of pulmonary blood flow secondary to

left-to-right intracardiac shunt.
A. Note that the frontal view (shows redistribution of pulmonary blood B. Note that the apparent prominence of the hila can be completely
flow as evidenced by the upper lobe vessels which are larger and more attributed to enlarged pulmonary arteries by looking at the lateral view,
numerous than normal. The lower lobe vessels are also enlarged and there which demonstrates the area beneath the left mainstem bronchus (arrow)
is prominence of the main and central pulmonary arteries, reflecting to be relatively radiolucent while the right (RPA) and left (LPA)
increased flow and early development of pulmonary arterial hypertension. pulmonary arteries are enlarged compared to normal.
159
Figure 6-14. Obligatory redistribution of pulmonary blood flow secondary to destruction of

pulmonary parenchyma at the lung base. The patient suffers from Agammaglobulinemia and has
suffered repeated pulmonary infections. These infections have occurred primarily at both lung
bases and resulted in destruction of the pulmonary parenchyma in these areas. As a result blood
is diverted to areas that remain normally perfused, which in this case are primarily in the upper
lung zones, and hence the apparent redistribution.
because most of the flow is directed toward the lung base), chronic pulmonary
venous hypertension (in which the earliest changes are generally noted at the lung
bases), and certain lung disease, such as alpha-I-antitrypsin deficiency, which, while
diffuse, seem to affect the lung bases more severely early in its course (figure 6-15).
Two groups of patients with pulmonary arterial hypertension have been iden-
tified in which a generalized reduction of the pulmonary vascular bed occurs
without a predilection for either upper or lower lung zones. These two groups
include patients with primary pulmonary arterial hypertension and patients with
central alveolar hypoventilation.
Patients with primary pulmonary arterial hypertension suffer from a generalized
reduction in cross-sectional area of the pulmonary arterial bed (15,16). The exact
etiology of the disease has not been established, although it is known that there
is a significant reduction in the luminal size of small pulmonary arteries and
arterioles with associated medial muscular hypertrophy and intima proliferation.
The resulting decrease in cross-sectional area of the pulmonary arterial bed in turn
results in the development of pulmonary arterial hypertension. The process occurs
Figure ~15. Patient with alpha-1-antitrypsin deficiency. There is redistribution of pulmonary

blood flow to the upper lung zones. The lungs are hyperinflated and there are focal areas of
oligemia, particularly at both lung bases. This radiographic picrure is very suggestive of
alpha-1-antitrypsin deficiency, which radiographically affects the lung bases preferentially. As a
result of the destruction of lung parenchyma at the lung base, there is obligatory redistribution
of flow to the upper lung zones which does not reflect either congestive heart failure or
pulmonary arterial hypertension.
diffusely throughout the lungs, and there is no obvious predilection for the lung
bases. Radiographs in these patients demonstrate equalization of vessel size and
number between upper and lower lung zones as well as dilatation of the main and
central pulmonary arteries with apparent tapering or disproportionate narrowing
of the distal branches, all of which reflect the pulmonary arterial hypertension
(figures 6-16 to 6-18) .
Patients with central alveolar hypoventilation (Ondine's curse) (17-20) suffer
from a dysfunction of automatic control of respiration. This causes a generalized
alveolar hypoxia, which is more pronounced when the patient sleeps. The low
alveolar oxygen tension, which occurs secondary to alveolar hypoventilation, in
turn causes vasoconstriction of the small pulmonary arterioles adjacent to the
alveolar spaces. Such vasoconstriction, occurring diffusely throughout the lungs,
significantly reduces the cross-sectional area of the pulmonary arterial bed and
produces pulmonary arterial hypertension. Chest radiographs of these patients
demonstrate the radiographic changes classically associated with pulmonary arterial
hypertension, including enlargement of the main and central pulmonary arteries
161
Figure 6-16. Primary pulmonary hypertension. Note the prominent main and centtal
pulmonary arteries and the equalization of flow between upper and lower lung zones. The
vessels supplying the periphery of the lungs appear disporportionately small, a feature often
described as "pruning."
with disproportionate narrowing of the more distal branches (14) (figure 6-19) .
As in other causes of pulmonary arterial hypertension, the vessels supplying the
upper lung zones are larger than normal and are nearly equal in size to those in
the lower lung fields. This change in vessel size is a reflection of redistribution of
pulmonary blood flow which is now more equally distributed between upper and
lower lung zones.
Enlargement of central pulmonary arteries

In association with the development of pulmonary arterial hypertension and cor
pulmonale, the main and central pulmonary arteries dilate. Dilation is presumably
possible because of their elastic nature and is readily apparent on routine frontal
and lateral chest radiographs. In general, the main pulmonary artery as well as the
first- and second-order branches are most noticeably affected. The lateral view is
often very helpful in confirming that the increased opacities noted in the hilar
regions are due to enlarged arteries as opposed to adenopathy or other mass (figure
6-20). On the lateral view the normal horseshoe configuration is maintained, but
the arteries are enlarged compared to normal. The area beneath the left upper lobe
....
~
A B
Figure 6-17. Severe pulmonary arterial hypertension. B. The close-up demonstrates calcification secondary to atherosclerotic
change within the main pulmonary artery. Such calcification is seen only
A. Note the markedly enlarged central pulmonary arteries on the frontal
in the setting of long-standing and severe pulmonary arterial hypertension.
vIew.
A
Figure 6-18. Patient with primary pulmonary hypertension. The cardiomegaly is solely
attributable to right ventricular enlargement.
A. The main and central pulmonary arteries are enlarged on the frontal projection.
B. Although both hila appear prominent, the lateral view shows that the normal "horseshoe"
configuration about the left mainstem bronchus is maintained although the arteries are larger
than normal.
."
~
. .
Figure 6-19. A patient with central alveolar hypoventilation (Ondine's curse). Note the
cardiomegaly, which is entirely attributable to right ventricular enlargement. The main and
central pulmonary arteries are enlarged and there is equalization of flow between upper and
lower lung zones. The vessels supplying the periphery of the lungs are disproportionately small.
The patient has been treated with phrenic nerve pacer (arrow) .
bronchus on the lateral view remains relatively radiolucent since arteries do not
normally occupy this area.
Somewhat similar changes can be seen with left-to-right shunts and reflect
dilatation of the central pulmonary vascular bed due to increased flow. These
changes are most marked, however, as pulmonary arterial hypertension supervenes
secondary to vasoconstriction of more distal small peripheral arteries and arterioles
(Eisenmenger's physiology). The largest main and central pulmonary arteries are
generally identified in the setting of left-to-right shunts with Eisenmenger's physi-
ology, but may occasionally be seen with pure pulmonary arterial hypertension
alone.
The combination of enlarged main and central pulmonary arteries, as well as
equalization of pulmonary blood flow between upper and lower lung zones, with
disproportionate narrowing of the distal arterial branches suggests strongly the
diagnosis of pulmonary arterial hypertension.
Numerous attempts have been made to predict the degree of pulmonary arterial
hypertension or the actual pulmonary artery pressure by developing various in-
A
Figure 6-20. Frontal (A) and lateral (B) views of a patient with primary pulmonary
hypertension demonstrating massively enlarged central pulmonary arteries. Although appearing
masslike on the frontal projection, on the lateral view the enlarged pulmonary arteries (arrows)
maintain their normal horseshoe configuration with an area of relative lucency seen beneath the
level of the left mainstem bronchus (arrow) , suggesting that the changes seen on the frontal
view are due solely to enlargement of the pulmonary arteries.
dexes to compare the size of the pulmonary arteries at the hilar and/or right
intralobar pulmonary artery level with the cardiothoracic ratio. The results of most
of these investigations indicate that enlargement of the arteries is a reasonably good
predictor of elevated pulmonary artery pressure, but the various indexes cannot
predict accurately specific pulmonary artery pressure (21,22).
Cardiac changes
On the standard chest roentgenogram, the distinction between right atrial and right
ventricular enlargement is frequently difficult. For most purposes the right-sided
cardiac chambers are dealt with as a single unit. Pulmonary arterial hypertension
primarily affects the right ventricular contour and results in hypertrophy and,
eventually, dilatation of this chamber. This process is evidenced by an apparent
"filling in" of the retrostemal space on the lateral view by the enlarging right
ventricle. On the frontal view the heart may also appear enlarged, particularly
when associated tricuspid insufficiency and right atrial enlargement occur. In the
setting of right-sided cardiac enlargement, the heart tends to have an upward tilt
of the left apex whereas with left ventricular enlargement the heart appears to
extend more inferiorly. One should also be aware that enlargement of the right
ventricle will appear to displace the left ventricle posteriorly on the lateral view,
often giving the false impression that the left ventricle, rather than the right
ventricle, is enlarged. Therefore, in the setting of right ventricular enlargement,
one should be careful in assessing the size of the left ventricle, particularly on the
lateral projection. In fact, Murphy and co-workers (23) found that even for trained
observers, none of the roentgenographic criteria usually suggested for the specific
diagnosis of right or left ventricular hypertrophy was reliable in the setting of
chronic bronchitis and emphysema. Obtaining oblique views may be of help in
making the distinction between right and left ventricular enlargement, but this has
not yet been tested.
GENERALIZED ALVEOLAR HYPOVENTILATION SYNDROMES

Low alveolar oxygen tension is a potent stimulus to vasoconstriction of adjacent
pulmonary arterioles. The body uses this mechanism in an attempt to divert blood
flow away from poorly oxygenated alveoli toward those with better ventilation.
Vasoconstriction of small pulmonary arterioles, and the resultant increased vascular
resistance, is sufficient to rapidly divert blood flow in the low-pressure, low-
resistance pulmonary circuit. This system functions very well in situations in which
localized areas are hypoventilating. However, with generalized alveolar hypoventi-
lation, a diffuse vasoconstriction results, which markedly reduces the overall cross-
sectional area of the pulmonary vascular bed and potentially can produce pulmo-
nary arterial hypertension.
Generalized alveolar hypoventilation can occur from a variety of causes. In-
cluded among these are low inspired oxygen, as may be seen in individuals living
at high altitudes, failure of adequate ventilation secondary to the central alveolar
167
hypoventilation syndrome discussed above, and a variety of mechanical restrictive

diseases which compromise respiratory mechanics to such a degree that generalized
alveolar hypoventilation results.
Mechanical restrictive diseases

Kyphoscoliosis
Severe kyphoscoliosis can markedly reduce the ability of the lungs to be adequately
ventilated. As a result a generalized alveolar hypoventilation occurs, and a reactive
generalized vasoconstriction results in response. As illustrated in figure 6-21 this
combination of changes can eventually result in pulmonary arterial hypertension
with the classic radiographic fmdings of dilatation between upper and lower lung
zones and a relative diminution of pulmonary vascularity in the periphery of the
lungs.
pickwickian syndrome
The massive obesity associated with Pickwickian syndrome restricts the ability of
the lungs to ventilate normally (24,25). The sheer bulk of the patient's body
increases the work of breathing significantly, and hypoventilation tends to occur.
As a result, alveolar oxygen tension throughout the lungs falls, and the reactive
vasoconstriction in the small, adjacent pulmonary arterioles results in a significant
reduction in the cross-sectional area of the pulmonary vascular bed. Pulmonary
arterial hypertension supervenes, and again seen are the classic radiographic features
of dilatation of the main and central pulmonary arteries, equalization of pulmonary
blood flow between upper and lower lung zones, and a relative diminution of
pulmonary vasculature supplying the lung periphery (figure 6-22).
DESTRUCTION OF THE PULMONARY VASCULAR BED

As noted during previous discussions, any disease that destroys the pulmonary
vascular bed, if severe enough, can reduce the size of the pulmonary vascular bed
so significantly that pulmonary arterial hypertension results and eventually cor
pulmonale supervenes. The two best examples of such diseases are emphysema and
chronic pulmonary emboli. Emphysema may directly destroy portions of the
vascular bed, and, due to destruction of functional lung units, causes areas of low
oxygen tension, which in turn cause vasoconstriction of adjacent small pulmonary
arterioles (figure 6-23). Thus, the reduction in overall size of the pulmonary
vascular bed with emphysema occurs from both the direct effects of pulmonary
vascular destruction and the indirect effect of pulmonary arteriolar vasoconstric-
tion. Indeed, the latter effect may be the major cause of pulmonary arterial
hypertension in these patients (26).
With chronic pulmonary emboli the occlusion of multiple small pulmonary
arteries and arterioles progressively reduces the size of the pulmonary vascular bed
and ultimately can cause arterial hypertension. This persistent hypertension in turn
affects the right-sided cardiac chambers and ultimately results in cor pulmonale. On
A
B
Figure 6-21. Patient with gibbus deformity of the spine and secondary pulmonaty atterial
hypertension. Frontal (A) and lateral (B) views of a patient with healed tuberculosis of the
spine and a resulting severe gibbus deformity, whose respiratory mechanics ate severely
compromised by the abnormalities of the bony thorax. As a result there is chronic alveolat
hypoventilation, and the resulting atteriolat vasoconstriction has caused pulmonary atterial
hypertension. This is evidenced radiographically by the very large central pulmonary atteries.
The pulmonary atteries ate so latge that they resemble masses on the frontal view (A), but can
be distinguished from masses on the lateral view (B) by noting that the atea beneath the left
mainstem bronchus is relatively radiolucent.
169
Figure ~22. Pickwickian syndrome. This massively obese patient clinically suffers from chronic
alveolar hypoventilation of the type associated with the Pickwickian syndrome. Note the large
prominent main and central pulmonary aneries, radiographically suggesting the diagnosis of
pulmonary arterial hypertension.
occasion, radiography enables recognition of the irregular occlusion of pulmonary

arteries secondary to emboli and suggests the appropriate diagnosis (figure 6-24).
CONGESTIVE HEART FAILURE
Left-sided failure
Failure of the left ventricle to perform adequately is manifest initially by evidence
of increased pressure in the pulmonary venous system and subsequently by accumu-
lation of fluid in interstitial and alveolar spaces. In addition the heart is often seen
to enlarge in conjunction with left ventricular failure.
As the left ventricle begins to fail, two phenomena occur. Left ventricular
end-diastolic pressure increases. Pressure subsequently rises in the left atrium in
order to maintain the gradient required to ensure filling of the left ventricle during
diastole. As there are no valves between the left atrium and the pulmonary veins,
pressure increases in the atrium are reflected in the pulmonary venous bed. In order
to maintain an appropriate pressure gradient across the vascular bed from the
....
C!
?'
,,"
i
f.
§
o
....,
"::l
c:
3
j
g
o
I'
A
Figure 6-23. Patient with emphysema and redistribution of pulmonary pulmonary arterial hypertension. At catheterization, the wedge pressures
blood flow. Thelatient has generalized emphysema as evidenced by were normal, excluding congestive heart failure and the pulmonary artery
hyperaeration an flattening of the hemidiaphragms. There is pressures were moderately elevated consistant with pulmonary arterial
redistribution of blood flow to the upper lungs, suggesting the possibility hypertension.
of either venous hypertension and associated left heart failure, or
A B
Figure 6-24. Patient with pulmonary arterial hypertension secondary to which suggests a mass on the frontal projection. The radiographic features
multiple recurrent pulmonary emboli. are those of pulmonary arterial hypertension, but the fact that there are
A. On the frontal view note the prominent main and central pulmonary large areas with no vascular supply and no obvious lung disease should
arteries. suggest the correct diagnosis of multiple pulmonary emboli.
B. on the lateral view note the large right pulmonary artery (RPA), ...
~
Figure 6-25. Congestive heart failure. Note the cardiomegaly and redistribution of flow to the
upper lung zones. The vessels in the upper lungs are equal in size to those at the lung bases, but
there is no prominence at the central pulmonary arteries to suggest pulmonary arterial
hypertension. This patient suffers from left ventricular failure with a wedge pressure of 25
mmHg.
arterial to venous side, pressures eventually rise on the arterial side as well. Radio-
graphically, these alterations in pulmonary vascular pressure, superimposed on a
low-resistance, low-pressure system, will be seen as "redistribution" of pulmonary
blood flow to the upper lung zones (figure 6-25). This phenomenon is manifest
radiographically as an increase in size and number of pulmonary vessels supplying
the upper lungs (27,28). In general this phenomenon can be recognized radiograph-
ically when left atrial pressure (as determined by pulmonary wedge pressure)
exceeds 15 mmHg. The radiographic changes become more pronounced as pressures
are elevated above this level, and, eventually, plasma oncotic pressure is exceeded
and fluid begins to leak into the extravascular interstitial space. In a normal patient
this process occurs at approximately 25 to 26 mmHg, but in patients with low
protein or with increased capillary permeability, such leakage may occur with
lower intravascular pressures. The fluid first accumulates in the interstitial spaces,
where it is cleared by lymphatic drainage. Radiographic features at this stage
include indistinct vessel margins and peribronchial "cuffing," reflecting thickening
of the peribronchial interstitial space by the edema fluid. If fluid leakage occurs at
173
Figure 6-26. Bat wing pulmonary edema. Demonstration of the so-called "bat wing" pattern
of pulmonary edema. The air space edema is bilaterally symmetrical and perihilar in distribution
with sparing of the lung periphery. This pattern, although seen to be classic, is unusual for
cardiogenic pulmonary edema and is more frequently seen in patients with uremia. In the case
illustrated, however, no evidence of renal failure was found, and the edema was on the basis of
left ventricular failure following myocardial infarction.
a rate greater than can be managed by lymphatic drainage, fluid begins to accumu-
late in the alveolar spaces. As this happens confluent air space opacities are noted
radiographically. These generally occur first at the lung bases and, on occasion, have
a somewhat central distribution which has been described as a "bat wing" pattern.
This latter appearance, although visually very striking, is the exception rather than
the rule and when it occurs is more commonly seen in patients with renal failure
than congestive heart failure (figure 6-26).
The radiographic pattern is altered by the presence of underlying pulmonary
disease. Edema will occur only in those areas that are being perfused by the
pulmonary arterial system. Therefore, if there is bullous disease or emphysema, and
particularly if destruction is localized, pulmonary edema will spare the destroyed
area. As a result of this sparing, the radiographic patterns of pulmonary edema may
be very unusual in patients with obstructive or destructive lung disease (figure
6-27), and the appropriate diagnosis may not be suggested unless the observer
recognized or is aware of the underlying lung disease.
....
~
Figure 6-27. Atypical pulmonary edema superimposed on chronic radiographic distribution suggests the possibility of pneumonia, but there
obstructive pulmonary disease. Frontal (A) and lateral (B) views of the were no clinical symptoms to suggest infection and all of the air space
chest demonstrate marked hyperaeration, focal areas of oligemia, and opacity was felt attributable to pulmonary edema secondary to congestive
bullous formation throughout the lungs. The patient is in congestive heart heart failure. The changes resolved completely with appropriate medical
failure, but the edema is accumulating primarily in the left upper lobe, management of the congestive heart failure.
which is the best perfused portion of the lung in this patient. The
175
Right-sided heart failure

When the right heart fails, changes are manifest on the systemic side of the
circulation. Such changes are generally easier to detect clinically than radiograph-
ically, although occasionally there are radiographic clues. In addition to enlarge-
ment of the right-sided cardiac chambers, one may note dilatation of the superior
vena cava, reflecting venous engorgement, as well as enlargement of the azygous
vein, again reflecting venous engorgement. In addition, particularly if there is
associated ascites or hepatic engorgement, one may observe a right-sided pleural
effusion. Generally, however, as noted above, detecting symptoms of right-sided
cardiac failure (peripheral edema, venous engorgement, hepatic enlargement, and
so on) by physical examination is easier than determination from the chest radio-
graph.
CONCLUSION
In sununary, the plain chest radiograph remains a significant cornerstone in the
evaluation of patients with pulmonary heart disease. It provides a consistent,
reproducible, low-risk procedure that enables assessment of a variety of cardiac and
pulmonary abnormalities. An understanding of the pathophysiologic alterations
operative in these patients is critical to accurate interpretation of the radiograph.
A wide variety of alterations related to various pulmonary diseases may be imaged
on the chest film, but very important among these are relatively subtle alterations
in the pulmonary vascular pattern which may provide significant clues to underly-
ing pathophysiologic alterations. Most important to recognize among these altera-
tions is "redistribution" or increased flow relative to normal in the upper lung
zones. This alteration reflects early pressure changes and can be seen in association
with both pulmonary venous and pulmonary arterial hypertension. Arterial hyper-
tension, however, is associated with dilatation of the main and first- and second-
order branches of the pulmonary artery and a relative diminution of the peripheral
arterial tree. These findings are most striking radiographically and strongly suggest
pulmonary arterial hypertension. When associated with dilatation of right-sided
cardiac chambers and obvious pulmonary disease, the possibility of cor pulmonale
may be legitimately suggested from the chest radiograph.
REFERENCES
1. Ravin CEo Disease signs in chest radiography. Optimization of Chest Radiography. U.S.
Department of Health and Human Services publication (FDA) 80-8124, pp. 3-7, 1980.
2. Kattan KR, WiotJF. How was this chest roentgenogram taken, AP or PA? AmJ Roentgenol
117: 843-845, 1973.
3. Hoffman RB, Rigler LG. Evaluation of left ventricular enlargement in the lateral projection of
the chest. Radiology 85: 93-100, 1965.
4. Heitzman ER, Markarian B, Solomon J. Chronic obstructive pulmonary disease: A review,
emphasizing roentgen pathologic correlations. Rad Clin of North Am 11: 49-75, 1973.
5. Gamsu G, Nadd JA. The roentgenologic manifestations of emphysema and chronic bronchitis.
Med Clin North Am 57: 719-733, 1973.
6. Thurlbeck WM, Simon G. Radiographic appearance of the chest in emphysema. Am J Roent-
genol 130: 429-440, 1978.
7. Herman PG. The chest roentgenogram: Its role in evaluating cardiomegaly and chronic obstruc-
tive disease. Chest 71: 689-690, 1977.
8. Nicklaus TM, Stowell DW, Christiansen WR, Renzetti AD Jr. The accuracy of the roent-
genologic diagnosis of chronic pulmonary emphysema. Am Rev Respir Dis 93: 889-899, 1966.
9. Simon G, Pride NB, Jones NL, Raimondi AC. Relation between abnormalities in the chest
radiograph and changes in pulmonary function in chronic bronchitis and emphysema. Thorax
28: 15-23, 1973.
10. Thurlbeck WM. Chronic airflow obstruction in lung disease. In Major Problems in Pathology.
Philadelphia: W.B. Saunders, 1976.
11. Greene R. "Saber-sheath" trachea: Relation to chronic obstructive pulmonary disease. Am J
Roentgenol 130: 441-444, 1978.
12. Anthonisen NR, Milic-Emil]. Distribution of pulmonary perfusion in erect man. J Appl Physiol
21: 760-766, 1966.
13. West JB, Dollery CT, Naimark A. Distribution of blood flow in isolated lung: Relation to
vascular and alveolar pressures. J Appl Physiol 19: 713-724, 1964.
14. Ravin CE, Greenspan RH, Mcloud TC, Lange RC, Langou RA, Putman CEo Redistribution
of pulmonary blood flow secondary to pulmonary arterial hypertension. Invest Rad 15: 29-33,
1980.
15. Kuida H, Dammin GJ, Haynes FW, Rapapart E, Dexter L. Primary pulmonary hypertension.
Am J Med 23: 166-182, 1957.
16. Walcott G, Burchell HB, Brown AL. Primary pulmonary hypertension. Am J Med 49: 70-79,
1970.
17. Mellins RB, Balfour HH, Turino GM, Winders RW. Failure of automatic control ventilation
(Ondine's curse). Medicine 49: 487-504, 1970.
18. Rhoads GG, Brody JS. Idiopathic alveolar hypoventilation: Clinical spectrum. Ann Intern Med
71:271-278, 1969.
19. Rodman T, Close HP. The primary hypoventilation syndrome. AmJ Med 26: 808-817, 1959.
20. Rodman T, Resnick ME, Berkowitz RD, Fennelly TF, Olivia]. Alveolar hypoventilation due
to involvement of the respiratory center by obscure disease of the central nervous system. Am
J Med 32: 208-217, 1962.
21. Lupi E, Dumont C, Tejada VM, Horwitz S, Galland F. A radiologic index of pulmonary arterial
hypertension. Chest 68: 28-31, 1975.
22. Chetty KG, Brown SE, Light R W. Identification of pulmonary hypertension in chronic obstruc-
tive pulmonary disease from routine chest radiographs. Am Rev Respir Dis 126: 338-341,1982.
23. Murphy ML, Boger J, Adamson JS, Jr., Rubin S. Evaluation of cardiac size in chronic bronchitis
and pulmonary emphysema. Chest 71: 712-717, 1977.
24. Burwell CS, Robin ED, Whaley RD, Bickelmann AG. Extreme obesity associated with alveolar
hypoventilation-A Pickwickian syndrome. Am J Med 21: 811-818, 1956.
25. Sieker HO, Estes EH, KeIser GA, Mcintosh HD. A cardiopulmonary syndrome associated with
extreme obesity. J Clin Invest 34: 916, 1955.
26. Szidon]p, Fishman AP. Cor pulmonale associated with COPD. Postgrad Med 54: 113-118, 1973.
27. Lavender ]P, Doppman J, Shawdon H, Steiner RE. Pulmonary veins in left ventricular failure
and mitral stenosis. Br J Radiol 35: 293-302, 1962.
28. Logue RB, Rogers]V, Gay BB. Subtle radiographic signs of left heart failure. Am Heart J 65:
464-473, 1963.
7. PULMONARY FUNCTION AND EXERCISE TESTING
TIMOTHY R. CHAPPELL, M.D.
The objective of this chapter is to examine the role of pulmonary function

and exercise testing in the clinical evaluation of patients with pulmonary heart
disease. Although much has been written about both static and dynamic pul-
monary function testing, the significance of this literature relative to patients
with pulmonary heart disease is limited, because frequently no differentiation
was made between patients with and without pulmonary heart disease or be-
cause the diagnosis of pulmonary heart disease was based on clinical fmdings
rather than hemodynamic measurements. Whenever possible, I will distinguish
patients with chronic lung disease alone from those with pulmonary hyperten-
sion in order to keep the focus of this review on pulmonary function and ex-
ercise testing, as these tests specifically relate to patients with pulmonary heart
disease.
Methodologic and technical aspects of these tests will not be considered here in
any detail. Rather, I will emphasize their clinical significance and role in the
diagnostic evaluation of patients with pulmonary heart disease, with particular
attention given to both the relationship between measured abnormalities and the
pathophysiology of pulmonary hypertension and the potential role of these tests
in the clinical identification of patients with cor pulmonale.
LJ. Ruhin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright @ 1984 Martinus NijhoffPuhlishing. Boston/The Hague/Dor-
drecht/Lancaster.
177
178 7. Pulmonary Function and Exercise Testing
PULMONARY FUNCTION TESTING

Following a detailed history and careful physical examination, the diagnostic
workup of a patient with known or suspected pulmonary heart disease should
include complete pulmonary function testing. At a minimum the workup should
consist of routine spirometry, arterial blood gas analysis, and measurement of
diffusing capacity. Each test is easily obtained at very little, if any, risk to the
patient. Based on the analysis of these results, the clinician may decide that more
complex and time-consuming procedures, such as measurement of lung volumes,
a sleep study, exercise testing or even right heart catheterization, are also indicated.
Much of our understanding of the physiologic mechanisms responsible for the
development of pulmonary hypertension is based on the relationship between
pulmonary function tests and hemodynamic measurements. The potential clinical
significance of these relationships had been obscured in the past by the dismal
prognosis and lack of effective treatment of cor pulmonale. Recent reports suggest-
ing that systemic vasodilators may improve the hemodynamic abnormalities of
some patients with pulmonary hypertension (1-5) have reawakened interest in cor
pulmonale and resulted in renewed efforts to develop better methods to identify
potentially treatable patients and to follow noninvasively the course of their disease
(6-9). The possibility that pulmonary function tests may be useful in this capacity
deserves additional review.
Relationship between pulmonary function tests and hemodynamic

abnonnalities
The demonstration of a relationship between pulmonary function tests and hemo-
dynamic measurements would be useful clinically since it would imply that pulmo-
nary function tests could serve as a noninvasive screening test in the evaluation of
patients suspected of having pulmonary heart disease and justify performing more
invasive or elaborate diagnostic procedures. Potential limitations to such an ap-
proach seem obvious.
In general, the overlap of pulmonary function test abnormalities in patients with
chronic lung disease with and without cardiac involvement makes it unlikely that
those with cor pulmonale can be specifically identified on the basis of pulmonary
function tests alone. Similarly, the wide range of abnormalities in patients with cor
pulmonale could make it difficult to predict accurately the severity of the underly-
ing pulmonary hypertension in an individual patient. Furthermore, pulmonary
function test abnormalities may not accurately reflect the complex interactions
between the many physiologic disturbances that are felt to produce pulmonary
hypertension. For example, patients frequently have mild abnormalities of several
pulmonary function tests that are clinically unimpressive when considered alone,
but that result in severe pulmonary hypertension when their physiologic effects are
combined. By focusing on one parameter, hemodynamic abnormalities would be
underestimated and potentially even missed in many patients.
The demonstration of a correlation between parameters such as forced vital
capacity (FVC) and pulmonary vascular resistance (PVR) in patients with severe
179
pulmonary function abnormalities would not necessarily imply a causal relation-

ship because other abnormalities, which could be equally important in the develop-
ment of pulmonary hypertension, such as arterial hypoxemia, may also be present.
As a result, most such relationships merely reflect the severity of the underlying
lung disease rather than independently identify the presence of pulmonary hyper-
tension.
A relationship between pulmonary function tests and hemodynamic measure-
ments would also not necessarily imply that a common mechanism producing
pulmonary hypertension was operative in all patients with comparable abnormali-
ties. For example, the physiologic consequences of a reduction in FEV1.0 are
considerably different in patients with chronic obstructive pulmonary disease
(COPD) compared to those with restrictive lung disease. The development of
pulmonary hypertension in patients with COPD is generally attributed to the
severe hypoxemia and hypercarbia resulting from ventilation perfusion mismatch-
ing, combined with the effects of increased intrathoracic pressure and variable
degrees of vascular destruction. In contrast, a comparable reduction in FEV1.0 in
a patient with restrictive lung disease such as pulmonary fibrosis may be associated
with only moderate arterial hypoxemia and a relatively normal carbon dioxide
tension, while fibrosis and obliteration of the pulmonary capillary bed playa more
important role in the development of pulmonary hypertension (10,11). Apprecia-
tion of such differences has a considerable bearing not only on the likelihood that
pulmonary hypertension may be present, but on the choice of and expected
response to conventional therapy, such as bronchodilators, chest physiotherapy, and
supplemental oxygen (11-17).
Many studies have demonstrated significant relationships between spirometry,
lung volumes, arterial blood gases and diffusing capacity, and a variety ofhemody-
namic parameters. However, medications and therapy such as aminophylline (15),
oxygen (15), or vasodilators (1-4,16), which might be expected to affect either
ventilation or the pulmonary circulation and thus alter the correlation between
pulmonary function tests and hemodynamic measurements, are rarely described in
the methods of most reports. As a result correlations in treated patients may be
clinically misleading if applied to patients who are not on similar therapy.
Interpretation of the literature is further complicated by frequent inclusion of
clinically unstable patients in studies consisting of otherwise chronically stable
patients. In addition some patients considered to have cor pulmonale may also have
hemodynamic evidence of left ventricular failure. The amount of data and methods
used for measurements, as well as the reporting of either absolute or percent
predicted values, frequently make it difficult to compare otherwise similar groups
of patients. As a result most correlations have been demonstrated using only a small
number of patients.
Finally, criteria used to exclude from study patients who have comparable
pulmonary function test abnormalities but normal hemodynamics are often not
clarified. Without this information the potential value of reported correlations as
a screening procedure for pulmonary hypertension cannot be truly determined.
W
0:
:::> 60 •
•• •
•
If)
If)
·••••.,. •
W
0: •
0.. 40
.-.•........ •••••• ••• •

>-
0: I. ••
W ~
~
<{ 20 • ... •
>-
0:
<{
•
Z
o o+-~--~--?-~--~
o 1-0 2·0
2
-.J
:::> FEV 1 (UTRES)
0..
Figure 7-1. An inverse relationship exists between the FEV1.0 (liters) and mean pulmonary
artery pressure (nunHg) in patients with chronic obstructive lung disease. (Reproduced with
This vital point must always be kept in mind when interpreting the clinical
significance of the many relationships that have been described in the literature.
Routine spirometry and lung volumes

FORCED EXPIRATORY VOLUME IN ONE SECOND (FEVl.O). An inverse relationship
has been noted consistently between the FEV 1.0 and hemodynamic measurements
in patients with COPD (18-22). The relationship between FEVl.O and the mean
pulmonary artery pressure (PAPm) in 41 patients with chronic bronchitis is shown
in figure 7-1 (19). In general pulmonary artery pressure was elevated in these
patients when the FEVl.O was less than 1 liter. Others have reported that the
pulmonary vascular resistance (PVR) is elevated when resting FEV 1.0 is less than
70% of predicted (22). Thus, the FEV1.0 may be helpful in screening patients with
COPD for the presence of pulmonary hypertension, but the scatter of points in
this figure illustrates the problems of relying only on such relationships to deter-
mine accurately the severity of vascular disease in a given patient.
The marked elevation in PAPm in some patients with a low FEV1.0 is presuma-
bly related to additional factors, including arterial blood gas abnormalities and
different levels of blood flow through a variably destroyed or obstructed pulmo-
nary capillary bed. Harris et al. (19) also demonstrated a significant relationship
between airway resistance (expressed as 1!FEVl.O) and PVR, suggesting that
compression of blood vessels due to increasing alveolar pressure also played a role,
181
but its physiologic significance is controversial, especially compared to other factors

such as hypoxemia (23,24).
A similar relationship generally does not exist in patients with restrictive lung
disease, even when FEV1.0 is reduced to levels comparable to that seen in patients
with COPD (10,22). This difference may be due, at least in part, to the relative
absence of severe arterial blood gas abnormalities until late in the course of many
restrictive diseases, particularly those associated with pulmonary fibrosis. However,
since the development of pulmonary hypertension in patients with restriction
secondary to neuromuscular or chest wall abnormalities is in large part due to
chronic hypoxemia and respiratory acidosis, correlations reflecting the severity of
alveolar hypoventilation seem more likely to be useful in this setting although this
has not been systematically examined.
LUNG VOLUMES. The ratio of residual volume (RV) to total lung capacity (TIC),
or the R V /TIC, has also been shown to correlate with both mean PA pressure
and total pulmonary resistance in patients with COPD (25), but not in patients
with restriction (10,26). This fmding is inconsistent, however, since other studies
have found no relationship between lung volume and hemodynamic parameters
(23,27-31). Correlations with pulmonary hemodynamics have not been reported
in COPD patients, using either the residual volume or total lung capacity alone
(25,28), presumably because large variations in R V and TIC are common in these
patients. In contrast, a relationship between measurements of lung volume such as
vital capacity and hemodynamic parameters does exist in patients with restriction,
both with and without parenchymal disease.
VITAL CAPACITY (VC). An inverse relationship between the vital capacity (VC)
and both pulmonary artery pressure and resistance has been reported in patients
with restriction. In 32 patients with a variety of conditions associated with inter-
stitial lung disease, Enson (11) demonstrated a relationship between the vital
capacity and the pulmonary artery diastolic-wedge gradient (figure 7-2). When
vital capacity was greater than 80% of predicted, no hemodynamic abnormalities
were noted; between 50 and 80%, pulmonary vascular resistance was generally
increased, even if resting mean pulmonary artery pressure was still normal. Pul-
monary hypertension was invariably present, even at rest, when the vital capacity
was below 50% of predicted. There was no evidence that hypoxic vasoconstric-
tion played a major role in the development of pulmonary hypertension in these
patients (11,32).
The relationship implied by figure 7-2 is potentially misleading because the
development of pulmonary hypertension in these patients is generally considered
to be due to more than a loss of lung volume. Significant abnormalities in diffusing
capacity were probably also present (26,33-35), although DLCO measurements
were not reported (11). Furthermore, while hypoxic vasoconstriction was not
important in these patients, it is uncertain whether this observation applies as well
to patients with other types of restrictive lung disease, such as kyphoscoliosis (36),
where hypoxemia is more common. For these reasons VC should be considered
in the initial evaluation of patients with restrictive lung disease, but the presence
30 • PROGRESSIVE SYSTEMIC
SCLEROSI S
•
• SARCOID
• MI SCELLANEOU S I NTER-
STITI-AL LUNG 01 SEASE
r=0.S7 p<.OOl
20
PULMONARY
DIASTOLIC
GRADIENT
(mm Hgi
•
10
00 20 40 60 80
VITAL CAPACITY (OBSERVED I PREDICTED x1001
Figure 7-2. An inverse relationship exists between the vital capacity (expressed as a percentage
of predicted) and pulmonary vascular resistance (expressed as the pulmonary artery
diastolic-wedge pressure gradient) in patients with a variety of diffuse interstitial lung diseases
resulting in restriction. (Reproduced with permission from reference 11.)
and severity of significant hemodynamic abnormalities cannot be excluded based

on this measurement alone.
The relationship between VC and pulmonary circulatory abnormalities is far less
certain in patients with COPD (22,27). In most reports the correlation between
the vital capacity and mean PA pressure is poor, even when the VC is reduced to
a level comparable to that seen with restriction (21,37). Once again, unreported
differences in arterial blood gases or diffusing capacity may account for such
discrepancies. Along these lines, in one study in which a correlation between VC
and PAPm was noted, the only other factors independently related to PAPm were
arterial oxygen tension (PaOz) and pH (21).
Arterial blood gases

There is greater agreement as to the nature and significance of relationships between
arterial blood gas and hemodynamic abnormalities in patients with either obstruc-
tion or restriction. Unfortunately, most reports do not include results of spirome-
try, information that could be used to confirm and extend the other correlations
and relationships described.
ARTERIAL OXYGENATION. An inverse relationship is consistently noted between
the level of arterial oxygenation, either oxygen tension (PaOz) or saturation
183
lEST
r = -0.81 ,:·0."
SEE ,5.2 SEE '1.0
':<0.001 '''000$
r: +0.70
SEE '1.3
':<0.005
°
/./.
0/
all
065 75 85 95 lOS
OXYGEN SAT. (% I pH
Figure 7-3. Relationship between arterial blood gases and pH and mean pulmonary artery
pressure (PA mean), total pulmonary resistance (TPR), and pulmonary vascular resistance (PVR)
in patients with COPD. Oxygen saturation (%) showed the best correlation with hemodynamic
measurements. However, the clinical significance of these correlations, particularly pH, is limited
by the wide scatter of points. (Reproduced with permission from reference 22.)
(SaO:J, and pulmonary artery pressure and resistance (10,18,20--23,25-27,29,30,-

36,38-51). In general, the best correlations are noted using oxygen saturation, and
the relationship is potentially strong enough to be clinically useful (figure 7-3) in
patients with COPD (22). Its clinical value in patients with restriction is less certain
because of the small number of patients in most reports. Furthermore, the possible
role that reductions in diffusing capacity may also play in the development of
pulmonary hypertension in these patients, particularly those with pulmonary fibro-
sis, is rarely addressed.
Enthusiasm about the clinical significance of the relationship between arterial
oxygenation and hemodynamic abnormalities in patients with chronic lung disease
should be tempered by a concern that many of the studies in which the strongest
correlations were noted included patients who were either clinically unstable or
recovering from an episode of right ventricular failure (39,44). Although the
relationship appears to exist whether or not right heart failure is present (30,39),
increases in PAP and reductions in oxygen saturation tend to be greater both during
right heart failure and in the setting of an acute exacerbation of the underlying
lung disease (39). Resultant correlations may be misleading when applied to
chronically stable patients (52). The importance of abnormalities that develop as

patients deteriorate, such as hypercarbia and acidosis (13,20,24), may explain why
oxygen administration to patients with cor pulmonale usually reduces less PA
pressure during an exacerbation than it does with recovery (28,39,53), even at
comparable levels of oxygenation.
ACID BASE STATUS. A relationship between both arterial pH (21,24,32,49,51,54)
and carbon dioxide tension (paCO~ (20-22,25,39,40,55-60) and either PAPm and
PVR has been noted in patients with COPD by some authors, but not by others
(figure 7-3) (10,31,44,61). Reports in which such relationships are absent mainly
include clinically stable patients with minimal blood gas abnormalities or a com-
pensated chronic respiratory acidosis. The effects of pH and PaC02 in chronic lung
diseases other than COPD are not as readily identified because the variable degrees
of hypoxemia or destruction of the capillary bed (46,47) make it difficult to sort
out the isolated effects of either one or both.
The combined effects of acid base disturbances and hypoxia on the pulmonary
circulation have been more convincingly demonstrated (23,28,29,39,43,62). Enson
et al. reported a strong correlation (r = 0.830) between both hydrogen ion
concentration (H +) and oxygen saturation and mean pulmonary artery pressure
in 43 patients with chronic obstructive pulmonary disease. In comparison, similar
correlations using either H+ or 02 saturation alone were (r = 0.609) and (r =
-0.798), respectively (32). This finding suggests that hypoxemia and acidosis are
potentiating stimuli to vasoconstriction and that the pulmonary arterial bed is
relatively insensitive to reductions in arterial saturation at low concentrations of
hydrogen ion, but extremely sensitive to reductions in arterial saturation when
hydrogen ion concentrations are increased (24,32,42).
The interaction between hypoxia and acidosis is strong enough that estimation
of pulmonary artery pressure may be possible on the basis of these two parameters
without the need for right heart catheterization (42) (figure 7-4). However, the
predictive value of these relationships is especially poor if patients are evaluated
during an acute illness (63). Furthermore, it has not been shown that pressures can
be similarly predicted in patients with chronic lung diseases other than COPD. The
interrelationship between arterial hypoxemia and acidosis has also been used to
calculate pulmonary artery diastolic pressure using measurements of arterial satura-
tion, hydrogen ion concentration, and the pulmonary capillary wedge pressure
(PCW) (49), but the inclusion ofPCW in this equation renders moot the possibil-
ity of deriving pulmonary artery pressure from arterial blood gas results without
performing right heart catheterization.
Thus, while the presence of pulmonary hypertension can be suggested by the
relationship between these variables, estimation of its severity can be misleading
because additional factors other than hypoxia and acidosis may act to increase
pulmonary artery pressure in some patients. In general, the diagnosis of pulmonary
hypertension should be considered in patients with an Sa02 of less than 85% if
any degree of acidosis is also present.
185
60
PULMONARY
ARTERY MEAN
40
PRESSURE
mmHg
20
o
[H]+ mIiEQ/'':-l,jI-L20----..L30------'4-0-------J5L.0----6.L0-
pH 7.70 7 .53 7.40 7.30 7.22
Figure 7-4. In patients with COPD a relationship exists between mean pulmonary artery
pressure and both hydrogen ion concentration and artetial oxygenation. A normal or high pH in
the face of diminished arterial saturation is associated with a relatively lower mean pulmonary
artery pressure than would be seen if acidosis were present. The sevetity of pulmonary
hypertension in such patients can be estimated to within 5 to 10 mmHg using this graph.
(Reproduced with permission from reference 42.)
Diffusing capacity (DLCO)

Potentially useful correlations between the diffusing capacity (OLCO) and hemo-
dynamic measurements have been noted in several studies consisting only of a small
number of patients (10,22). The amount of data in the literature is very limited
because OLCO is frequently not reported and not measured in every patient in each
series or because the measurements are made by methods that are sufficiently
different to render quantitative comparisons between otherwise similar patients
difficult. Exercise OLCO measurements may be particularly valuable in the evalua-
tion of these patients and will be discussed in greater detail later in this chapter.
The relative lack of data is also, at least in part, due to difficulty in obtaining
accurate measurements in patients with chronic lung disease. Many technical and
physiologic factors are known to directly affect OLCO measurement in these
patients. Clinically, the most important of these include nonuniformity of the lung
(ventilation perfusion (V/Q) mismatch) (64,65), any degree of alveolar filling
(edema, blood), thickening of the alveolar capillary membrane (fibrosis or edema),
reduction in available hemoglobin (anemia or significant levels of carboxyhemo-
globin), and changes in pulmonary capillary blood volume (reduced cardiac output
or pulmonary embolism). To eliminate more easily correctable technical factors,
hemoglobin and carboxyhemoglobin levels should also be measured in patients
with an abnormal OLCO. The physiologic significance of the other abnormalities
can be only roughly estimated from other parameters such as the chest x-ray,
arterial blood gases or, if available, lung biopsy.
Single-breath measurements (OLCOSB) are the most widely reported (66--68).
Results are easily obtained, but they are subject to error due to ventilation perfusion
mismatching in patients with COPO. Measurements of OLCO sB may not be
possible during exercise or in patients unable to inspire the necessary volume of
test gas or who are so dyspneic that a 10-second breath-holding maneuver cannot
be completed. Steady state measurement (OLCO ss) (69) is a reasonable alternative
in these patients. Results are easily obtained, and steady state measurements can be
performed during exercise, but like single-breath measurements they are influenced
by ventilation perfusion mismatching (63,64).
Rebreathing measurements (OLCO RB) are very accurate, reproducible, and less
affected by ventilation perfusion mismatching due to the mixing process of the
rebreathing maneuver itself, and can be performed during exercise. Rapid accumu-
lation of carbon monoxide in the patient's blood may occur, however, when
repeated measurements are made. An additional obstacle to widespread use of
rebreathing has been the need for a computer to perform the complex calculations
involved (70,71).
The potential clinical significance of the OLCO in the evaluation of patients
with cor pulmonale is suggested by the demonstration of correlations between the
DLCO (percent predicted) and a variety of hemodynamic abnormalities measured
both at rest and during exercise (10,22). For example, in a small group of patients
with sarcoidosis (10) (figure 7-5), only one patient-who had a OLCO of 61 %
-had normal hemodynamics both at rest and during exercise, suggesting that a
OLCO ofless than 60% of predicted was the threshold below which hemodynamic
abnormalities were present. When the OLCO is less than 40% of predicted, the
likelihood increases that one or more hemodynamic measurements may be abnor-
mal. Hemodynamic abnormalities are invariably present when the OLCO is less
than 20 to 33% of predicted (10,35).
Similar fmdings were noted in 18 patients with COPO, but the threshold below
which rest or exercise PVR was elevated was a OLCO 70% of predicted (22).
Presumably, hemodynamic abnormalities were noted in these patients at this rela-
tively higher OLCO due to associated arterial blood gas abnormalities (52). In
support of this, and in contrast to the patients with sarcoidosis, arterial oxygen
saturation also correlated with hemodynamic measurements (r=0.88), and this
correlation improved (r = 0.97) using a multiple regression analysis using OLCO
and oxygen saturation versus PVR (22).
The subdivision of the OLCO into two components-Om, or the membrane
component of carbon monoxide transfer from alveolar gas into the blood, and Vc,
the capillary blood volume-may enable a more meaningful analysis of abnormali-
187
ao
•• ..3.
~
\
" at.
" 001
SEt '51
30
'">
A-
20
MI ,
10203040 so no 70 l010JO'OSOo07(
DCO PP
Figure 7-5. The relationship between diffusing capacity expressed as a percentage of predicted
nonnal (DCOpp) and a variety of hemodynamic measurements in a small number of patients
with sarcoidOSIS. Parameters include mean pulmonary artery pressure at rest (P AMR) and during
exercise (PAME), pulmonary vascular resistance during exercise (PVRE), log of resting
pulmonary vascular resistance (PVRRlog), and total pulmonary resistance during exercise (TPRE
lo~. All but one patient whose DLCO was 40% of predicted had an elevated pulmonary artety
pressure or resistance at rest or during exercise. The mean DLCO for the group was 33%
(range 16-91%). (Reproduced with permission from reference 10.)
ties of the pulmonary capillary bed (72,73). Theoretically, pulmonary capillary

destruction or a reduction in the size of the capillary bed would proportionately
decrease Dm and Vc, whereas diseases involving the alveolar capillary membrane
would decrease Dm to a greater degree than the Vc (64). Such clear distinctions
are frequently blurred because many diseases demonstrate variable degrees of both
abnormalities.
Available reports are suggestive that these two measurements may be clinically
useful in the evaluation of patients with pulmonary heart disease. For example, V c
is normal or increased in patients with mitral stenosis whereas a decrease in Vc has
been associated with increased PVR (74,75). These changes, however, as well as
changes in Dm related to the amount of associated capillary thickening and fibrosis,
are variable (76-79). There is a disproportionately greater decrease in Dm com-
pared with Vc in patients with pulmonary fibrosis (64,76), but whether this
relationship persists or whether V c is comparably decreased when pulmonary
hypertension is also present is unknown. Extremely variable changes in both
parameters have been reported in patients with COPD, although Dm may be
decreased more than Vc in emphysema (64,80).
Both Dm and Vc have been reported to be reduced in patients with either
primary pulmonary hypertension and recurrent pulmonary embolism, presumably
due to intense precapillary vasoconstriction, partial destruction of the capillary bed,
or both, but the number of patients is small and other reports have demonstrated
more variable changes (76,74,81,82). Changes in Vc have been used to monitor the
results of thrombolytic therapy in patients with pulmonary embolism (83), but a
relationship between decreased Vc and pulmonary vascular resistance in these
patients has not been reported. Whether V c would also be useful in monitoring
vasodilator therapy of patients with either primary or secondary pulmonary hyper-
tension is an unknown but intriguing possibility.
Correlations derived from compiled data

To examine the clinical significance of the many relationships between pulmonary
function tests and hemodynamic measurements that have been reported and to
further defme their potential use as part of the evaluation of patients with cor
pulmonale, I have included in table 7-1 the individual patient data from each series
that has been reviewed as well as data from several other reports in which individ-
ual patient pulmonary function and hemodynamic measurements are given, but
possible correlations were not attempted (54,84-87).
Only patients with significant pulmonary hypertension (resting PAPm > 20
mmHg) who could be classified as having either obstruction or restriction were
included. Normals, clinically unstable patients, and patients with an elevated
pulmonary capillary wedge pressure (> 16 mmHg) were excluded. While the
total number of patients with restriction is considerably less than those with
obstruction, the combined total in both groups is larger than any other series and
thus should be useful in evaluating the correlations described in smaller series.
Table 7-1 demonstrates the relationships that exist between pulmonary function
tests and mean pulmonary artery pressure. Similar correlations using either total
pulmonary resistance or pulmonary vascular resistance were, in general, less signifi-
cant, and because these measurements were provided in fewer number of patients,
the results are not shown.
In patients with obstruction the strongest correlation with the results of spirome-
try was found using the FEVl.O (r = -0.48), but in patients with restriction it
189
Table 7-1. Coefficients of correlation between various pulmonary function tests and mean
pulmonary artery pressure (PAPm)
Obstruction (n=507) Restriction (n=73)

(PAPm = 38 ± 11 mmHg) (PAPm = 36 ± 9 mmHg)
VC -0.24 (n=150) -0.46 (n=18)

VC% -0.42 (n=97) -0.13 (n=34)
FEV 1.0 -0.48 (n=56) -0.21 (n=27)
FEVl.oo;. -0.09 (n = 84) -0.14 (n=l1)
RV 0 0.14 (n = 90)
TIC 0.21 (n=103) -0.14 (n=17)
RV/TIC 0.24 (n=150)
DLCO ss -0.25 (n = 45) -0.13 (n=20)
DLCOSB -0.43 (n=15) -0.34 (n = 27)
pH 0.05 (n=68) 0.03 (n = 27)
P0 2 -0.50 (n=l04) -0.45 (n=43)
pC0 2 0.30 (n=210) -0.09 (n=50)
02 sat % -0.69 (n=150) -0.55 (n = 33)
VC=vitai capacity, VC(%) = percent predicted VC, FEVl.0=forced expiratory volume in one second, FEV
1.0(%) = percent predicted FEV1.0' R V = residual volume, TLC = total lung capacity, R V /TLC = ratio of R V
to TLC, DLCOSB = single breath diffusing capacity, DLCO SS = steady state diffusing capacity, pH = arterial pH,
P02=arteriai oxygen tension, pC02=arterial carbon dioxide tension, 02 sat %=percent oxygen saturation.
was with the vital capacity (= - 0.46). However, in both groups there was such
a wide range of PAPm in patients with the most severe abnormalities that PAPm
could not be accurately determined in an individual patient on the basis of spirome-
try alone.
No correlation was found using either the DLCO ss or DLCO SB in either group,
but the lack of a relationship does not diminish the value of this measurement in
the evaluation of these patients. This apparent discrepancy with the literature may
be due to the small number of patients in other reports.
Arterial oxygenation, measured as either the P0 2 or as oxygen saturation,
provided the best correlations in both groups. Correlation using oxygen saturation
may be stronger because the interaction between pH and PaC0 2 on the pulmonary
vasculature is also reflected in their effects on oxygen saturation through the oxygen
hemoglobin dissociation curve. Figure 7-6 demonstrates that while the strongest
correlation, oxygen saturation and PAPm, is highly significant (p < 0.00(1) and
the presence of pulmonary hypertension is strongly suggested on the basis of this
relationship, the severity of pulmonary hypertension in any given patient is more
difficult to determine. Furthermore, because patients with arterial hypoxemia to a
degree comparable to that seen in many of these patients do not invariably have
cor pulmonale, it is important to reemphasize that the presence of pulmonary
hypertension cannot always be ascertained on the basis of this relationship alone.
Pulmonary function results in patients with pulmonary hypertension

Results of pulmonary function tests from 41 patients with pulmonary hypertension
who have recently been evaluated in our laboratory are shown in table 7-2. Patients
were classified according to established spirometric criteria (88,89) into four clinical
70 + y= -0.845 (Xl + 102.88

65 + r= -0.74
+
60
• •
55 • • +
5 • •
• •
+
•
0
• •
PAPm
45
• • •• + +
(mmHg) 40 +
•• 0
35 TABLE I:
• OBSTRUCTION
30
• RESTRICTION
TABLE D:
•
o OBSTRUCTION
•
o RESTRICTION
25 • MIXED
+ NORMAL
20
L., ,
50
I
60
,
70 80
I
90
, ,
100
O2 SAT (%)
Figure 7-6. Relationship between oxygen saturation (02 sat %) and mean pulmonary artery
pressure (PAPm) from patients in table I shown in figure (closed symbols) with obstruction (n
=150) and restriction (n=33). The correlation coefficient (r= -0.74) and equation of the
regression line derived from these data are shown. Data from table II shown in figure (open
symbols) are superimposed and demonstrate that patients from our laboratory with obstruction
(n=18), restriction (n=9), and mixed obstruction and restriction (n=7) do not appear to be
significantly different from the group as a whole. However, there is little or no relationship
between 02 sat % and PAPm in patients with normal spirometry (n=7), emphasizing that
factors other than arterial hypoxemia contribute to hemodynamic abnormalities in some patients.
categories: (a) obstruction, (b) restruction, (c) a mixture of both obstruction and
restriction, and (d) normal. Measurements oflung volume confirmed each patient's
classification and further defined underlying abnormalities, particularly in those
with mixed obstruction and restriction. Single-breath diffusing capacity (DLCO SB)'
arterial blood gases, and mean pulmonary artery pressure (P APm) are also shown.
The relatively large standard deviation for each parameter is indicative of a wide
range of disease severity in these patients.
Eighteen patients classified as having obstructive lung disease included ten with
clinical features of both chronic bronchitis and emphysema, six with cystic fibrosis,
one with emphysema due to alpha-l-antiprotease deficiency, and one with chronic
asthma. Patients classified as having restrictive lung disease included five with
idiopathic pulmonary fibrosis documented by lung biopsy, three with primary
pulmonary hypertension, and one with sleep apnea. Seven patients were classified
as having a mixture of obstructive and restrictive lung disease, including one each
with scleroderma and chronic bronchitis, extensive fibrotic changes secondary to
tuberculosis in addition to chronic bronchitis, sarcoidosis, angiographically
Table 7-2. Pulmonary function tests in 41 patients with pulmonary hypertension l
Arterial Blood Gases

DLCO SB
Age FVC FEV1.0 TLO RV' (ml/min/ pH pCO, PO, 0, sat PAPm 3
(years) (liters) (liters) (liters) (liters) mmHg) (units) (torr) (torr) (%) (mmHg)
Obstructive lung disease (n= 18)

55±14 2.5+0.9 1.2±0.6 6.8+1.5 4.0±1.7 20.0±3.9 7.39±0.04 44±11 51±13 85±5 39±8
(59±19%) (40±16%) (110±21%) (175±42%) (75±16%)
Restrictive lung disease (n=9)

48±9 204+0.5 2.2+0.7 3.9+0.6 1.9+0.3 9.1±204 704O±0.05 37±8 60±9 90±6 33±6
(60±15%) (67±18%) (62±12%) (94±18%) (31±7%)
Mixed obstruction and restriction (n=7)

54±9 204+0.9 1.6+0.7 4.5+1.0 2.8+0.7 14.9+5.5 7042±0.05 36±9 56±8 89±4 35±7
(61±26%) (54±21%) (7±23%) (112±26%) (57±19%)
Normal spirometry (n=7)

45±17 3.7+004 3.1+0.4 5.8±1.0 2.0+0.7 16.0+3.2 7044±0.02 33±5 73±14 93±3 54±11
(89±5%) (91±4%) (97±11%) (104±19%) (62±25%)
lData presented as mean±standard deviation.

'Measured by close circuit helium dilution.
3Resting supine measurements.
FVC = forced vital capacity, FEV1.0=forced expiratory volume in one second, TLC = total lung capacity, RV=residuai volume, DLCO SB = single breath diffusing capacity,
pH=arterial pH, pCO,=arterial carbon dioxide tension, PO,=arteriai oxygen tension, 0, sat=arterial oxygen saturation, PAPm=mean pulmonary artery pressure.
....
~
....
conftrmed recurrent pulmonary emboli, and chronic intravenous drug abuse. Two
patients had mixed defects of an undetermined etiology. Seven patients with
normal spirometry included one with angiographically documented recurrent pul-
monary embolism and six with primary pulmonary hypertension.
Ohstructive lung disease

Our patients with obstruction demonstrated varying degrees of hyperinflation and
air trapping which, in some patients, resulted in a secondary reduction in forced
vital capacity (FVC). These patients also had the most severe arterial hypoxemia
and hypercarbia, but were, for the most part, well compensated as suggested by
the arterial pH and a history of at least a six-week period of clinical stability prior
to catheterization.
The OLCO was decreased, but the relatively large standard deviation makes the
clinical signiftcance of this abnormality uncertain, particularly compared with the
other three groups. The wide range of abnormalities is presumably due to the type
of diseases causing obstruction. Severe reductions in OLCO have been reported in
patients with emphysema (52), but only one of our patients was a classic "pink
puffer" (90,91). The remaining patients had predominant clinical features of
chronic bronchitis, or bronchiectasis in the case of patients with cystic ftbrosis
(90,91). The OLCO has been reported to be normal or moderately decreased in
cystic ftbrosis or bronchiectasis (64,90,92).
Restrictive lung disease

Reductions in FVC in patients with restriction were similar in magnitude to that
seen in patients with obstruction, but the FEV1.0 and lung volumes conftrm the
absence of signiftcant airways obstruction and air trapping. In the ftve patients with
pulmonary ftbrosis, progressive parenchymal destruction resulted in both a loss of
lung volume and progressive obliteration of pulmonary capillaries, demonstrated
by the marked reductions in OLCO. Moderate arterial hypoxemia was generally
present. The relative severity and late stage of the underlying disease are suggested
by evidence of CO 2 retention in some patients.
One of our patients had severe restriction secondary to morbid obesity, which
was further complicated by obstructive sleep apnea. Obesity, particularly in associa-
tion with hypoventilation syndromes, is a recognized cause of pulmonary heart
disease (93). Because some patients with sleep apnea are not obese, we consider a
sleep study to be part of the evaluation of patients with clinical evidence of
pulmonary hypertension without an apparent etiology. A sleep study may also be
useful in documenting severe nocturnal desaturation, both with and without apneic
episodes, in patients with chronic obstructive pulmonary disease (94,95). Based on
the assumption that this episodic hypoxemia can lead eventually to cor pulmonale
or cause further clinical deterioration in patients with pulmonary hypertension, we
routinely prescribe nocturnal oxygen supplementation under these circumstances
(15,96,97) .
193
Mixed obstructive and restrictive lung disease

Patients with mixed obstruction and restriction had reductions in FVC similar to
that seen in the other two groups, but less evidence of airways obstruction and air
trapping than patients with obstruction. Abnormalities ofDLCO and arterial blood
gases both generally fell between those of patients with either obstruction or
restriction. All of these fmdings are consistent with the various combinations of
underlying diseases clinically documented in these patients.
Normal spirometry
The group with normal spirometry included patients with severe pulmonary
vascular abnormalities, predominantly primary pulmonary hypertension. Moderate
arterial hypoxemia was generally present, but PaC0 2 tended to be normal or low.
The most notable abnormality in this group as a whole appeared to be a reduction
in DLCO, but this was highly variable. The lowest DLCO (21%) we observed
in any of these 41 patients was in the patient with chronic recurrent pulmonary
embolism. Patients with pulmonary hypertension secondary to intracardiac defects
or valvular heart disease may also have normal spirometry (98), but these condi-
tions were excluded prior to catheterization and are not represented in this series.
Primary pulmonary hypertension

Patients with primary pulmonary hypertension or chronic recurrent pulmonary
embolism would be expected to have normal pulmonary function tests, but several
recent reports suggest that a small percentage of these patients may demonstrate
restriction. Three of our patients with primary pulmonary hypertension had reduc-
tions in FVC ranging from 55% to 72% of predicted (81,99-101). Although the
mechanism of restriction is poorly defmed, it has been attributed to the loss of
vascular distensibility, which, by mechanical coupling, also decreases distensibility
of the airspaces (101), encroachment of distended vessels into the airspaces (102),
or associated chest wall weakness (103). No obvious clinical or physiologic differ-
ences were found between the patients with restriction and those with normal
spirometry. Not unexpectedly, no correlation was found between the results of any
pulmonary function test and pulmonary arterial pressure or resistance (98). Variable
changes in the DLCO have also been reported, but reductions below 60% of
predicted (35,98-100,104-106) appear to be uncommon, even in patients with
severe pulmonary hypertension.
Clinical application of pulmonary function testing

The spectrum of lung diseases represented by these 41 patients reflects our process
of patient selection and referral. Because the physical examination, chest radio-
graph, and electrocardiogram were used as the initial screening evaluation, it is
probable that patients with pulmonary hypertension who had little or no abnor-
malities in any of these three parameters were not selected for further study, even
if pulmonary function test abnormalities were comparable to those of the patients
who eventually underwent right heart catheterization. The significant effect of our
referral pattern is demonstrated by the nine patients with primary pulmonary
hypertension. However, our cohort is also more typical in that, as in most other
series, the most common underlying lung disease was COPD.
The only significant relationship noted between PFT results and hemodynamic
measurements was with the level of oxygenation in patients with obstruction
(r = - 0.58, P < 0.05). This fmding emphasizes that while no single test oflung
function is adequate to evaluate the complex physiologic abnormalities of patients
with pulmonary heart disease, arterial oxygenation appears to be the most signifi-
cantly related to the presence of cor pulmonale of all the tests available.
Spirometry does not reflect the physiologic impact of abnormalities suggested
by measurements of the DLCO and arterial blood gases. While the pattern of
abnormality on spirometry may suggest an underlying anatomic and physiologic
basis for the development of pulmonary heart disease, it is often not specific in
evaluating patients in whom pulmonary hypertension may have developed as a
result of primary vascular process, abnormalities of respiratory drive, or neuromus-
cular disease. If the severity of spirometric abnormalities were the sole criterion to
be considered, a number of our patients with relatively mild defects might have
gone undetected. Pulmonary hypertension was documented in patients with a wide
range of obstruction or restriction. On the other hand, severe spirometric abnor-
malities do not conclusively prove the presence of underlying pulmonary hyperten-
sion. Several of our patients with severe airways obstruction had only borderline
elevations of pulmonary artery pressure. We have also seen seven patients with
pulmonary function abnormalities comparable to these pulmonary hypertensive
patients who were found to have normal pulmonary artery pressures at right heart
catheterization.
Abnormalities of the DLCO or arterial blood gases were the major clue to the
presence of underlying pulmonary vascular disease in many of our patients. Some
of the many limitations of the DLCO have been described, but its potential value
should also be emphasized. More work is needed to defme its clinical significance,
particularly in terms of using measurements of Dm and Vc to further interpret
reductions in DLCO in patients with cor pulmonale.
Arterial blood gas abnormalities are particularly important because they quan-
titate, at least in part, the roles of hypoxia and acidosis in the development and
severity of pulmonary hypertension. One or more arterial blood gas measure-
ments were abnormal to some degree in most of our patients. However, the
clinical significance of arterial blood gas abnormalities in terms of the presence
and severity of pulmonary hypertension in any given patient must be viewed
with caution, despite the fact that significant correlations are frequently noted.
To demonstrate, individual data from our 41 patients have been plotted on figure
7-6. While most of our patients (table 7-2) do not appear to be significantly
different from the group (table 7-1) as a whole, the wide scatter of points
emphasizes the complex interaction of other factors that relate to the develop-
ment of pulmonary hypertension in some of these patients, particularly the pa-
tients with normal spirometry.
195
Summary
Pulmonary function tests are an important part of the evaluation of patients with
pulmonary heart disease. Results of simple tests, including spirometry, lung
volumes, DLCO, and arterial blood gases, are useful both to determine the type
and to quantitate the severity of the underlying lung disease. Relationships between
the various test results and the pulmonary circulation suggest that pulmonary
function abnormalities, particularly severe abnormalities, may correlate to some
degree with the presence and the severity of underlying pulmonary hypertension.
Isolated abnormalities of pulmonary function such as spirometry are very diffi-
cult to interpret in the context of pulmonary heart disease. For this reason the
clinical evaluation of these patients should include, at a minimum, the several tests
described. The results of all these tests must be interpreted in light of results from
other diagnostic studies such as the electrocardiogram and chest x-ray. A severe
abnormality in one or more pulmonary function tests, combined with evidence of
right ventricular hypertrophy or large pulmonary arteries on a chest x-ray, in a
patient with suspected pulmonary hypertension would make a strong case for the
presence of cor pulmonale.
On the other hand, demonstration of mild abnormalities or even normal pulmo-
nary function tests does not exclude the possibility that pulmonary hypertension
is present, particularly if the ECG and chest x-ray are abnormal. More importantly,
this opens up the differential diagnosis to a variety of diseases not associated with
parenchymal destruction, such as sleep apnea syndromes, congenital heart disease,
recurrent pulmonary embolism, and, rarely, primary pulmonary hypertension. In
our experience these patients provide the greatest diagnostic challenge. All too
often this type of patient simply carries the clinical diagnosis of cor pulmonale of
unknown etiology despite the unfortunate fact that, with aggressive diagnostic
evaluation to determine the underlying disease process, a specific treatment may
be available.
The possibility that a patient may have pulmonary hypertension should always
be considered in patients with severe pulmonary function test abnormalities even
if the chest roentgenogram and electrocardiogram are normal. The specificity of
both criteria is such that an absence of abnormalities should not prevent a complete
evaluation if clinically indicated. Results of exercise testing in this setting can
provide valuable clues both to the functional severity of the underlying lung disease
and in some cases to the presence of otherwise undetected pulmonary vascular
abnormalities.
The possibility that a patient may have pulmonary hypertension should be
considered if there is severe obstruction (FEV 1.0 < 1 liter or < 70% predicted)
or restriction (FVC < 40% of predicted) and significant reductions in DLCO
( < 40--60% and < 70% of predicted in patients with restriction or obstruction,
respectively). The addition of severe arterial hypoxemia (Sa02 < 85%), particularly
when it is present in combination with acidosis and hypercarbia, considerably
increases the likelihood that hemodynamic abnormalities are present. However,
beyond these generalizations, pulmonary function tests do not appear to be useful
in clearly distinguishing patients with severe lung disease alone from those with
secondary cardiac involvement, and may only provide a rough estimate of the
severity of pulmonary hypertension if it is present. As a result, it seems unlikely
that any of these relationships can completely eliminate the need for more sensitive
or specific evaluation of these patients.
EXERCISE TESTING
Exercise testing provides an objective assessment of the complex interaction of
ventilation and gas exchange, the central and peripheral circulation, blood gas
transport, and tissue metabolism under controlled circumstances of increasing or
maximal physiologic stress. An abnormality in anyone or in a combination of these
processes results in quantitative alterations in a variety of measurements that are
an important part of the evaluation of patients with chronic lung disease.
Cardiovascular and ventilatory changes that occur during exercise in patients with
chronic lung disease have been extensively studied, but the exercise response of
patients with pulmonary heart disease, in particular, is very poorly characterized
because the diagnosis is frequently based on clinical findings. Hemodynamic meas-
urements are either not done during exercise or are made at different levels of exercise
or workload, making it difficult to compare and interpret results from different
series. Because the exercise response of patients with cor pulmonale is frequently
assumed to be similar to patients with severe chronic lung disease, much of our
current understanding is extrapolated from patients who may not be even compara-
ble.
All of these points, coupled with a general reluctance to exercise patients with
clinical evidence of cor pulmonale, account for an amazing lack of information
relating to exercise testing in patients with pulmonary heart disease. As a result
specific recommendations are not possible, and suggestions tend to be based more
on theory than on fact.
General clinical objectives of exercise testing in patients with chronic lung
disease include the following needs:
a. To establish the pathophysiologic mechanism(s} causing abnormal gas exchange.

b. To establish functional capacity or, as a corollary, the degree of disability.
c. To monitor disease progression and/or response to treatment.
d. To establish the presence of associated ischemic heart disease or abnormal airway
reactivity.
In the context of patients with pulmonary heart disease, the clinical objectives and
thus the value of exercise testing are more limited.
The degree of disability is clinically obvious in most patients with pulmonary
hypertension secondary to chronic parenchymal lung disease, even to the point that
functional limitations are so severe that an exercise test may not be possible. Given
the age of many of these patients, the possible presence of associated left ventricular
197
Table 7-3. Sununary of measurements made in stage 1 to stage 4 exercise tests

Stage 1: Heart rate (HR), blood pressure (BP), electrocardiogram (ECG), minute ventilation (VE),
tidal volume (VT), symptoms (oxygen saturation by ear oximetry and measurements of
DLCO, if available).
Stage 2: HR, BP, EKG, VE, VT, ~ymptoms and mixed expired 02 apd CO 2 concentrations for
oxygen consumption (V0 2), carbon dioxide production (VC0 2), respira~ory quotient
(RQ), oxygen pulse (02 pulse) and ventilatory equivalent for 02 (VE/V0 2).
Stage 3: Same as stage 2 with arterial blood gases (ABG), lactate levels, calculation of VD/VT, and
P (A-a) 02'
Stage 4: Same as stage 3 with right heart catheterization for all hemodynamic measurements, mixed
venous blood sampling, and Fick cardiac output (thermodilution if available).
dysfunction is a frequent clinical consideration but may be difficult to detect on

the basis of an exercise test alone, particularly in patients with overt right ventricu-
lar failure. Exercise testing has been used to document a response to treatment such
as oxygen, but the physiologic significa,nce of the results is unclear, and whether
exercise testing can be used to monitor the results of other treatments, such as
systemic vasodilators, is as yet unproven.
Exercise testing is perhaps most useful in establishing the pathophysiologic
mechanisms responsible for abnormal gas exchange and functional limitation.
Exercise testing may also prove to be useful in identifying and evaluating patients
with cor pulmonale, but it is still too early to recommend its routine use for this
purpose.
Practical aspects of exercise testing in patients with pulmonary heart disease

Basic principles and technical aspects of exercise testing are beyond the scope of
this chapter, and the reader is referred to several excellent reviews that provide
greater detail (107-112). However, several practical considerations concerning
exercising patients with pulmonary heart disease deserve additional comment.
Patient safety is the first and foremost concern. As with all clinical tests, the
information obtained must justify the potential risks, discomfort, or difficulty of
the procedures involved. This is particularly true in patients with cor pulmonale,
who may be susceptible to clinical deterioration concomitant with or resulting
from an exercise test. Patients must be exercised using a carefully established
routine, since oversights in technique or procedure result in poorly collected data
and may cause complications or delay the recognition and proper treatment of
complications. Patients must be closely monitored, and both resuscitation equip-
ment and personnel skilled in its use should be immediately available at all times
(112).
A reasonable clinical approach to exercise testing in these patients is a standard-
ized series of tests termed stages 1 to 4 (table 7-3), originally described by Jones
(110,112). Easily obtained measurements such as minute ventilation (VE) and heart
rate are common to all stages, whereas invasive procedures such as arterial puncture
and right heart catheterization are done only in later stages.
Stage 1 is brief and therefore usually tolerated even by patients with the most
severe lung disease. Simple, noninvasive measurements are made during a progrer
sive incremental power test to the patient's symptom-limited maximum. Stage 2
is also noninvasive, but involves more complex measurements such as oxygen
consumption (VOz) during steady state exercise at a level determined during stage
1. Stage 3 essentially repeats stage 2 measurements, but adds arterial blood sampling.
Stage 4 is the most complex and invasive, requiring right heart catheterization for
central venous blood sampling and hemodynamic measurements. In our lab patients
undergo a stage 4 evaluation as part of research protocols. The risk of either right
heart catheterization or exercise testing in an experienced research setting is very
low, even when performed together (113).
Two basic exercise protocols can be used to evaluate these patients: steady state
and progressive incremental, or multistage (108,112,114). In steady state testing the
patient exercises at a constant workload for at least 3 minutes before measurements
are recorded. After resting 30 minutes or until baseline returns, exercise is repeated
at a higher workload, and this process continues until either the patient cannot
complete at least 3 minutes of exercise at a higher load or predetermined end points
such as maximum heart rate are reached. In order to ensure that excessive fatigue
does not determine the maximum load in a patient with limited cardiopulmonary
reserve, repeat testing may be necessary on the following day beginning at 50%
of the previous maximum load, increasing to the previous maximum, and if this
is easily obtained, proceeding on to progressively higher loads as tolerated. Both
the cycle ergometer and treadmill are suitable for this protocol.
In multistage testing workload is progressively increased at intervals as short as
one minute until the patient is either unable to continue or predetermined end
points are reached (108,115). The cycle ergometer is best suited for this protocol.
This technique offers the advantage of rapidly assessing the maximal limitations of
a patient who is likely to be exhausted by repeated periods of exercise. This
consideration is important in the evaluation of patients with pulmonary heart
disease, who might be expected to have a very limited cardiopulmonary reserve.
By utilizing equipment that ,rapidly measures expired oxygen and carbon dioxide
in combination with computer analysis of data, the multistage method can be used
to determine VE , V0 2, and carbon dioxide production (VCOz) on a breath-by-
breath basis. With this technique maximum oxygen consumption (V0 2 max) and
anaerobic threshold (AT) can be visually identified (109,116).
Many patients with chronic lung disease have such severe limitations that
obtaining the steady state necessary for measurements in stages 2 through 4 may
be difficult. The possible effect this may have on the interpretation of data can be
minimized if steady state measurements are repeated at several different power
outputs. Most measurements made using breath-by-breath analysis during incre-
mental tests have been shown to be comparable to those obtained after 4 minutes
of steady state exercise (117). In severely limited subjects cycle ergometry using
a multistage protocol consisting of very small increments of power (50 kpm) may
be the only way to obtain meaningful information (112,115).
Cycle ergometers and the treadmill are the most commonly used modes of
199
exercise in patients with lung disease. We use the cycle because the patient's arm,
wrist, and torso are relatively stable, thereby reducing the effect of body motion
on hemodynamic tracings. In addition, breathing through a mouthpiece and re-
peated blood sampling are more easily accomplished. The most significant differ-
ence between the two is that a higher V02 max is measured using the treadmill.
This is apparently due to greater work efficiency, familiarity with walking, and
less thigh muscle fatigue, a limiting factor on the cycle (112). Because the difference
is small and patients with severe chronic lung disease can rarely exercise at these
work levels, this consideration is generally unimportant (112,118).
We also routinely insert an arterial line in order to facilitate repeated blood
sampling, minimize patient discomfort, and reduce the amount of time lost due
either to attempting to locate an artery or to the difficulty in sampling arterial
blood in an exercising patient. To avoid complications adequate circulation to the
hand should always be confirmed using the Allen's test (119), arterial catheters
should never be flushed by syringe (120), and the catheter should be immediately
removed if excessive patient discomfort, redness, or blanching develops in the
region of the catheter tip at any time.
The standard objective end point of the exercise test is V0 2 max (108,122).
Predictions ofV0 2 max based on age or extrapolations from a heart rate measured
during submaximal exercise are unreliable because maximum heart rate is highly
variable in patients with chronic lung disease (20,123,125). While determining
V0 2 max in each patient is desirable, a reasonable clinical alternative is to determine
the symptom-limited maximum oxygen consumption. In fact, pushing patients
with cor pulmonale beyond the point at which clinical symptoms develop may be
particularly unwise. We had one patient terminate exercise due to symptoms that
developed only after mixed venous oxygen saturation fell to less than 10%!
Unfortunately, measurements based on the presence or absence of subjective
symptoms may be subject to considerable error due, for example, to psychological
or nutritional factors and poor effort. However, intra- and inter-individual varia-
bility can be minimized, and objective, reproducible end points can be developed
by relating all measurements to the work rate or V0 2 at which they were made.
Our experience has been that a careful explanation of the procedures and the
patient's awareness of the seriousness of the underlying disease result in good
cooperation and consistent results.
Closely related to concerns about patient effort are physiologic alterations due
to the general deconditioning of most patients with chronic lung disease. Abnor-
malities that result from deconditioning are particularly troublesome because of
their similarity to the abnormalities seen with intrinsic cardiac disease, thus render-
ing distinction between the two etiologies difficult (124,125). Since there is no
evidence to suggest that patients with pulmonary heart disease are more severely
deconditioned than patients with chronic lung disease alone, exercise duration does
not, by itself, signify the presence of pulmonary hypertension. Furthermore, the
exercise capacity of most patients is so limited that their capability to exercise long
enough at work levels high enough to cause a training effect is unlikely, even with
repeated exercise testing (123). However, this should not be taken as a rationale
for not encouraging patients with cor pulmonale to be active (127), particularly
if an exercise test can be used to suggest a safe level of activity for each individual.
The level of oxygen consumption at which minute ventilation and expired
e02 increase in response to increasing lactate concentrations has been described as
the anaerobic threshold (AT) (125,128). However, because anaerobic metabolism
may be important at the start of exercise before ventilatory and circulatory adapta-
tions occur, it is unlikely that, as suggested by this term, there is ever a point or
threshold below which tissue lactate is not present (129). Peak blood lactate levels
are seen three to five minutes after maximal exercise is reached because diffusion
of lactate from muscle is slow and results in a delayed appearance in the blood
during exercise.
Significant increases in blood lactate normally do not occur before reaching at
least 50% ofV0 2 max. In contrast, lactate concentrations do not increase until very
high V0 2 in trained individuals, but in poorly conditioned subjects, lactate levels
begin to rise at very low V0 2• On this basis the AThas been suggested as an index
of physical conditioning (109,125,130). However, chronic lung disease limits the
clinical value of this concept. Ventilatory limitations are frequently so severe that
patients either fatigue and stop exercising before AT is reached or their VE increases
so rapidly in response to increased ve0 2 that exercise is terminated due to dyspnea
which develops as maximum obtainable levels of ventilation are approached, even
at low workloads (112,128).
Many patients with lung disease also develop severe hypoxemia during exercise
(123,131). In this circumstance supplemental oxygen administration generallyal-
lows patients to do comparable work at a lower VE with less dyspnea, resulting
in increased exercise duration or higher work rates. Outpatient ambulatory oxygen
is generally prescribed if both improved exercise tolerance and oxygen saturation
are demonstrated on repeat exercise testing (123,132,133). More specific recommen-
dations for using supplemental oxygen during exercise cannot be made until more
data are available, particularly from long-term studies. In more practical terms,
supplemental oxygen makes the determination of V0 2 more difficult because even
small errors in measuring either inspired or exhaled oxygen concentrations result
in large errors in the calculation of V0 2•
Cardiopulmonary interactions during exercise

The complex interactions that occur during exercise are schematically shown in
figure 7-7 (111). The rate at which the interconnecting gears should turn is
determined by the metabolic demands of the tissues (107,111,116). If demand
exceeds the ability to transport oxygen to the tissues, work can continue, at least
temporarily, utilizing anaerobic mechanisms. While the major abnormalities of
patients with cor pulmonale involve the heart and lungs, other abnormalities may
also be important in some patients.
Metabolic or endocrine abnormalities of skeletal muscle can limit exercise, but
limitation at this level is more commonly related to diminished blood flow due
201
MUSCLE 02 a C02 VENTILATION

ACTIVITY DELIVERY (VA + YO" YE)
PERIPH.
CIRC.
Figure 7-7. Schematic illustration of the complex physiologic interactions that are
physiologically linked during exercise in a system represented by the gears. (Reproduced with
to peripheral vascular disease or medications that interfere with sympathetic regula-

tion of regional blood flow. Inability to increase local perfusion to exercising
muscle results in clinical symptoms of claudication or early fatigue, while objective
measurements demonstrate a low AT. No evidence suggests that patients with
either chronic lung disease or pulmonary heart disease have specific abnormalities
at this level.
Inadequate cardiac output may severely limit the exercise tolerance of patients
with left or right ventricular failure or both. Left ventricular dysfunction due to
cardiomyopathies, hemodynamically significant valvular lesions, and ischemia re-
sults in an inability to appropriately increase cardiac output at any given level of
exercise. Increases that do occur are generally the result of greater increases in heart
rate than stroke volume. In patients with cor pulmonale, increases in cardiac output
are limited by increases in pulmonary vascular resistance and right ventricular
afterload.
Pulmonary circulatory abnormalities such as chronic recurrent emboli or pri-
mary pulmonary hypertension can also limit exercise tolerance. The inability either
to recruit additional unused vessels or to further dilate existing vessels due to
destruction, obstruction, or vasoconstriction results in increasing pulmonary artery
pressure and resistance as blood flow is increased. Exercise testing appears to be
particularly well suited to detecting pulmonary vascular disease, even in its earliest
stages, when resting hemodynamics are normal.
Ventilatory limits are generally reflected by resting pulmonary function test
abnormalities. Severe abnormalities must be present before exercise tolerance is
considered to be significantly impaired on the basis of lung disease alone (108,-
114,134). As a result, exercise testing is better at detecting subclinical cardiac disease
than at detecting early pulmonary parenchymal disease.
Complex relationships derived from resting pulmonary function tests such as the
resting FEVl.O' FVC, DLCO, VD/VT, and predicted V0 2 max have been used
to estimate the degree of exercise limitation due to the lungs (114,124,134-136).
However, such approaches are limited because assumptions must be made about
cardiac output and the homogeneity of the distribution of ventilation, perfusion,
and diffusing capacity throughout the lung that may not be clinically valid in
patients with cor pulmonale. Additionally, many patients increase their FEV1.0 and
FVC above resting values during exercise (112,134), resulting in considerable
underestimation of maximum ventilatory capacity.
Malfunction of one of the interacting gears in figure 7-7 results in adaptations
in the others that may make continued exercise possible. Abnormalities can be
measured and interpreted in a way that helps localize the primary defect-be it
the peripheral tissues and circulation, heart, pulmonary circulation, or the lungs
(108). Figure 7-7 does not demonstrate which tests (stages 1-4) may be useful and
how they relate to underlying abnormalities or whether they can be used to
distinguish patients with chronic lung disease from those with cor pulmonale (112).
Exercise-induced abnormalities in pulmonary heart disease
Peripheral tissues and circulation

Patients with chronic lung disease appear to have a delayed increase compared to
normals in oxygen consumption and the appearance of arterial hypoxemia at the
start of exercise (124,137). However, this is due to deconditioning rather than a
specific abnormality of cellular metabolism, organ function, or the presence of cor
pulmonale.
Increased resting oxygen consumption is noted in many patients and is generally
related to an increased work of breathing (108). Increases in oxygen consumption
during exercise, particularly in patients with severe airflow obstruction, may be an
additional limiting factor. The percentage of total body oxygen consumption
attributable to work of breathing may be as high as 20% (range of 10 to 55%)
in some patients; this high percentage may have important physiologic conse-
quences in terms of the availability of oxygen to other tissues, particularly in
patients with low cardiac outputs (138). While high oxygen consumption has been
suggested to occur in patients with left ventricular dysfunction, it probably is also
related to the presence or type of symptoms and the severity of exercise limitation
seen in patients with cor pulmonale (139). .
=
Calculation of systemic oxygen transport (SOT cardiac output [Q] X arterial
oxygen content [Ca02D is helpful in understanding the effects of lung disease on
exercise tolerance, but this formula does not account for such factors as regional
shifts of blood flow (shunting), increased tissue oxygen extraction, and local effects
of pH at the tissue level that may also be important in some patients (140). SOT
may be inadequate in patients with chronic ~g disease due to worsening arterial
hypoxemia, while patients with pulmonary heart disease may, in addition, have an
inadequate cardiac output (123,140-146). A low resting mixed venous oxygen
tension (Pv0 2), which has been suggested to be due to an inadequate cardiac output
203
response to chronic hypoxemia, is an important prognostic factor in survival of

patients with COPD (140,141). Because Pv0 2 and SOT may also be low in
patients with cor pulmonale, they could also be important prognostic factors, but
such a relationship has not been f1I'1l1ly established.
Right ventricular Junction

Evaluation of the right ventricle has, until recently, required catheterization and
hemodynamic measurements. In addition, virtually all that is known about abnor-
mal function has been generated from patients with COPD (147). The develop-
ment of radionuclide angiography has added considerably to our understanding of
right ventricular function, but it is also limited because methods for determining
volume are technically difficult in patients with abnormal, dilated right ventricles
(148-150).
The right ventricle is a flow generator pump (151) and its Starling function
curve is likely to be considerably different from that of the left ventricle (152),
both because its geometry results in large changes in volume with little muscle fiber
shortening and its relatively thin wall is not capable of generating pressures equiva-
lent to those of the left heart.
Increased afterload is probably the earliest abnormality in the development of
cor pulmonale and results in right ventricular hypertrophy, impaired contractility
(153), cardiac dilation, and increasing preload (154,155). However, because large
increases in diastolic volume result in small changes in end-diastolic pressure (156)
and right ventricular output increases as end-diastolic volume is increased (151),
cardiac output can often be maintained until severe increases in afterload have
developed (155). As a result, hemodynamic evidence of ri~ht ventricular dysfunc-
tion must be assessed in terms of both cardiac output (Q) and right ventricular
end-diastolic pressure (RVEDP) (147).
Right ventricular function appears to be preserved in most patients with COPD,
and both Q and R VEDP are normal at rest and during exercise (147,157). The
change in cardiac output during exercise is appropriate for the change in V0 2
(30,147) and for the cardiac output increase as the result of increasing stroke volume
(SV). Patients with normal Q and R VEDP at rest but elevated R VEDP during
exercise are considered to have latent right ventricular failure. Overt R V failure
is characterized at rest by a low Q and elevated RVEDP (29,39,44,158) and during
exercise by a blunted increase in Q and SV, further increases in RVEDP (157,159),
increased oxygen consumption relative to cardiac output (30), widening davOz
implying that any increases in Q that are noted are inadequate (160), and marked
increases in PVR (150,161).
While R VEDP is generally accepted as hemodynamic evidence of right ven-
tricular dysfunction, resting elevations can be misleading in some patients. For
example, fluid retention or increased blood volume can increase R VEDP in the
absence of failure. On the other hand, R VEDP may be normal in patients with
frank right heart failure if ventricular compliance is increased due to cardiac
dilation (147). Increases in RVEDP during exercise are considered to be a more
sensitive indicator of right ventricular dysfunction and are more common in

patients with electrocardiographic evidence of right ventricular hypertrophy, even
in the absence of other clinical findings (147). The vectorcardiogram may be even
more sensitive of early hemodynamic abnormalities, even in the absence of R VH
(162).
Stroke volume is normal in patients with COPD (108) and is maintained in
the setting of increased afterload during exercise by an increase in end-diastolic
volume. Stroke volume is abnormal in patients with severe R V failure, but there
is some confusion about SV responses during exercise due to the effects of decon-
ditioning, body position (supine versus upright), and the work rates at which
measurements were made (108,111,163). Oxygen pulse, which is related to stroke
volume (V0 2!HR = SV X !laVO~, may provide a noninvasive means to
assess this parameter. 02 pulse is generally normal in patients with COPD (108),
but whether this also holds true in patients with cor pulmonale is less certain. A
reduction in 02 pulse may be useful in separating patients with COPD from those
with cor pulmonale, but it probably will not distinguish right from left ventricular
dysfunction because a diminished cardiac reserve is common to both conditions
(112).
Like elevations in RVEDP, a depressed right ventricular ejection fraction
(RVEF) also is seen more commonly during exercise (142,164,165). In one study
(150) 15 of 30 patients with normal resting R VEF had abnormal responses during
exercise. A relationship was found between the severity of the reductions in
FEV1.0 and Pa0 2 {but not PaC0 2 or exercise SaO~ and RVEF, but even some
patients with mild obstruction had severe reductions in R VEF during exercise.
Patients with an abnormal R VEF reached comparable levels of oxygen consump-
tion and heart rate at a lower submaximal workload than patients with normal R V
function. Measurements during maximal exercise are generally not attempted
because the development of lactic acidosis itself has a deleterious effect on myocar-
dial contractility and PVR (132).
Pulmonary circulation
The most widely recognized abnormalities associated with pulmonary heart disease
are increases in pulmonary arterial pressure and vascular resistance during exercise
(20,30,46,87,147,157,160,161,166,167), which are found in up to 50% of patients
with COPD in whom hemodynamic measurements are made at rest and during
exercise (162).
If resting pulmonary artery pressures are elevated, increases during exercise tend
to be large, suggesting that in later stages of disease even small changes in cardiac
output cause excessive increases in pressure (figure 7-8). The steeper slope of the
regression line for decompensated patients may be due to associated hypoxemia and
acidosis that develop as patients deteriorate (147), resulting both in further increases
in PAP and PVR and decreases in Q. The absolute change in pulmonary artery
pressure does not distinguish between patients with and without right ventricular
failure, although resting PAP tends to be higher and larger changes are more likely
in the latter.
205
Ppa, mmHg
7 •
60
50
"0 •
o , , ,
" 8
Q 1 min-1 m-2
10
Figure 7-8. The relationship between cardiac index (Q) and mean pulmonary artery pressure
(lip.) at rest (open symbols) and during exercise (closed symbols). In group I 25 patients with
clironic bronchitis (circles). In group II 14 patients with chronic bronchitis and a history of
right heart failure (squares). In group III 10 patients with chronic bronchitis and clinical right
heart failur~ (triangles). Hatched area .represents 95% confidence interval of normal values. The
increase in Ppa is larger at any given Q as patients deteriorate and develop cor pulmonale.
(Reproduced with permission from reference 47.)
Patients with primary pulmonary hypertension characteristically have an ex-

tremely high pulmonary vascular resistance and very low cardiac output even
compared to patients with pulmonary hypertension secondary to chronic lung
disease (30,141,143,144). The resistance to flow is so extreme that increases in
cardiac output during exercise are blunted, often so severely that systemic blood
pressure and tissue perfusion decrease, producing clinical symptoms (170). Some
degree of the intense vasoconstriction in these patients may be partially reversible
with vasodilators. Exercise testing has been used to demonstrate a higher cardiac
output and lower PVR at comparable V0 2 after treatment (4).
Similarly, after being treated with vasodilators, some patients with secondary
pulmonary hypertension are capable of increasing oxygen transport at a lower right
ventricular afterload both at rest and during exercise (1). Significant reductions in
PVR are consistently noted, but changes in pulmonary artery pressure are more
variable (1,2). R VEDP is also frequently decreased, and changes correlate with
reduction in pulmonary resistance even when pulmonary artery pressure is un-
changed (172). Because changes in R VEF have also been shown to correlate with
reductions in pulmonary vascular resistance, radionuclide techniques may prove to
be the most valuable method of noninvasively measuring afterload in these patients
(173,174), both at rest and during exercise.
Patients with normal resting hemodynamics who develop pulmonary hyperten-

sion during exercise may represent a special group of patients for evaluation (175).
Long-term studies of patients with cor pulmonale suggest that the disease progres-
sion in these patients is usually slow (20,176). If vasodilators ultimately prove to
be useful in the management of cor pulmonale, will this type of patient require
treatment and is therapy likely to be even more beneficial at this apparently earlier
stage? Conversely, does waiting until resting pressures are abnormal diminish the
potential benefits of therapy? This possibility emphasizes the potential future
significance of exercise testing in the evaluation of these patients, particularly if
noninvasive methods of evaluating pulmonary artery pressure, flow, or right
ventricular function are developed.
Ventilation
MECHANICAL FACTORS. Mechanical abnormalities such as muscular weakness,
decreased lung compliance, or increased airways resistance limit exercise by reduc-
ing ventilatory capacity, although reductions in ventilatory capacity are not specific
for patients with cor pulmonale. The measurement of FEV1.0 allows estimation of
ventilatory capacity through its relationship to the maximum voluntary ventilation
(MVV) (MVV = FEVl.O X 35) (114,124), but MVV is often underestimated by
this formula in patients with .an FEV1.0 less than 1 liter (112,124). MVV is clinically
useful if its relationship to VE and the subjective sensation of dyspnea are consid-
ered. Breathlesmess is usually first noted when the "dyspnea index" (VE/MVV X
1(0) exceeds 50% (177,180). Exercise-induced bronchospasm should always be
considered in patients who develop dyspnea at aVE well below levels predicted
using the resting FEVl.O (179,180). Beyond this point, breathlessness is not a specific
fmding of cor pulmonale although it is frequently severe and out of proportion
to other clinical findings and may be an important diagnostic clue, particularly in
patients with primary pulmonary hypertension (170).
Excessive positive intrathoracic pressure may develop during expiration in
patients with severe obstruction. Direct transmission of this pressure to the heart
and blood vessels results in an overestimation of intravascular pressures, including
the pulmonary capillary wedge, even if electronic integration of the systolic and
diastolic pressure is used (181). For this reason, PCW measured during exercise
is generally considered to be unreliable in the setting of COPD and frequently
leads to clinical confusion about the presence of associated left ventricular dys-
function (18,19,31,43,46,150,157,167,182). The magnitude of the contribution of
increased intrathoracic pressure to the development of cor pulmonale by directly
increasing pulmonary vascular resistance is still controversial (25,123,158,159,-
182-184).
RATIO OF DEAD SPACE TO TIDAL VOLUME (VD/VT). The ratio of dead space to tidal
volume (VD /VT) is normally 25% to 30% at rest and falls during exercise because
tidal volume (VT) increases while dead space (VD), which is predominantly fixed
(or anatomic), changes minimally (106,107). In contrast, VD/VT may increase
during exercise in some patients with lung disease (185) (figure 7-9), due to large
207
~(%)
VT
50 ,
-- ---'
......... I
...... I
~.>--- ,
40
,................ . . . . . . ------------- ,,-'
_.---r
30
20
10
L,{---,--I- - ' - - - - - ' - - - - - " - - - - - - - - - - . . . J . . . . - . - . . . . . L - - - . . . . J

100 200 300 400 500 600 700 800 900
REST EXERCISE WORK LOAD (kg.-M'/min.)
Figure 7-9. Increasing ratio of dead space to tidal volume (VD/VT) durin~ exercise has been
suggested as a means to diagnose early pulmonary vascular disease patients (hatched line) and
normals (solid line). See text for discussion. (Reproduced with permission from reference 185.)
increases in physiologic dead space resulting from the development of high-

ventilation-to-perfusion (V/Q) ratios in areas of the lung with increases in ventila-
tion greater than the increases in blood flow (105,186,187). An increase in VD/VT
during exercise appears to be related to the presence of pulmonary vascular abnor-
malities, even if resting VD /VT is normal. For this reason changes in VD /VT
during exercise have been suggested as an early screening test for pulmonary
vascular disease, even in patients with normal resting hemodynamics (185) and little
or no evidence of pulmonary heart disease on physical exam, chest x-ray, or
electrocardiogram.
The sensitivity ofVD/VT in this setting has recently been challenged (188), but
the discrepancy between these studies may be related only to the functional integ-
rity of the pulmonary capillary bed, as suggested by differences in measurements
of the DLCO (185). For example, assuming that the pulmonary capillary bed is
relatively intact in patients with little or no reduction in DLCO, increasing flow
during exercise will further distend or recruit capillaries in underperfused areas. As
a result V/Q will be either unchanged or improved and VD/VT will fall. In
contrast, in patients with reductions in DLCO due to severe capillary destruction
or obstruction, resting perfusion may have already maximally recruited or dis-
tended all available remaining capillaries, particularly if resting pressures are ele-
vated. In this setting increasing flow will cause further increases in pressure, but
not improve V/ Q, so that VD /VT will neither change nor increase. In one study
in which VD /VT decreased during exercise in patients with pulmonary vascular
disease, the mean DLCO was 65% of predicted (188) compared to mean values
of 50% in a study of comparable patients in whom VD /VT increased during
exercise (185).
While resting DLCO measurements are also important clues to the presence of
underlying vascular disease, a low DLCO does not necessarily mean that VD /VT
will increase during exercise. Additionally, an elevated VD /VT at rest that in-
creases further during exercise does not by itself diagnose the presence or determine
the severity of cor pulmonale. Finally, the magnitude of changes in VD /VT in
patients with chronic parenchymal lung disease varies widely, even when associated
vascular involvement is present.
Along similar lines, changes in the DLCO and Vc in response to cold or changes
in body position have also been used to suggest early pulmonary involvement of
systemic diseases such as scleroderma. For example, patients with Raynaud's phe-
nomenon increase Vc and DLCO after exposure of an extremity to cold, presuma-
bly due to central redistribution of blood during peripheral vasoconstriction (189).
This response is absent in patients with both Raynaud's phenomenon and sclero-
derma or scleroderma alone (189,190), although recent studies suggest that this
actually may also be a normal response (191). Patients with scleroderma also do
not increase DLCO when changing from a sitting to supine position.
While this is generally interpreted to be evidence of a lack of distensibility of
pulmonary vasculature secondary to structural vascular changes, hemodynamic
measurements have not been done to rule out changes in Q, reflex pulmonary
vasodilation, or abnormal peripheral vascular response as possible mechanisms for
these observations. While these changes may be evidence of early pulmonary
vascular involvement, they have not been shown to be also specific for the presence
of cor pulmonale, even though approximately 50% of patients with scleroderma
have pulmonary hypertension in the absence of overt clinical signs (33). Another
unknown factor is whether these patients would increase VD /VT during exercise,
although this would seem likely (190).
An increase in physiologic dead space is also an important cause of dyspnea both
at rest and during exercise because VE is increased by the need to maintain an
adequate alveolar ventilation (VA = VE[1- VD/VT]) and normal PaC0 2, partic-
ularly during exercise (111,114,185,192). As a result, patients with severe COPD
have an excessive VE at any level of V0 2 or an abnormal ventilatory equivalent
for oxygen (VE/VO~, which, in some patients, is also associated with a higher
dyspnea index.
DIFFUSING CAPACITY (DLCO). DLCO increases in a nearly linear relationship
with V0 2 during exercise in normal patients (66,193,194) due predominantly to
increases in Vc, but evidence that Dm is also increased suggests that a combination
of both capillary recruitment and distention is occurring (193). In patients with
severe lung disease, the absolute increase in DLCO during exercise is often blunted,
particularly in patients with emphysema (122). As previously discussed the signifi-
cance of changes in DLCO must always be interpreted in light of the difficulty
209
in obtaining accurate measurements in some patients, particularly since resting

V/Q abnormalities can increase during exercise.
If the DLCO reflects the size of the functioning pulmonary capillary bed, then
changes in DLCO during exercise should vary inversely with pulmonary vascular
resistance at higher pulmonary blood flows, particularly when the capacity of the
capillaries is exceeded (126). In other words, as the number of available capillaries
that are being perfused approaches its upper limit, continued increases in blood flow
would result in rising pulmonary vascular resistance and pulmonary artery pressure.
Thus, patients with the lowest maximum DLCO during exercise would also be
expected to have the highest PVR.
Pulmonary blood volume (PBV) does not increase during exercise in patients
with a variety of chronic lung diseases who have both resting pulmonary hyperten-
sion and a low DLCO (59 + 13%), despite large increases in PAPm. In contrast,
in similar patients with relatively normal resting pressures and DLCO (89 + 7%),
PBV increases by 20% during exercise without changing PAPm. This fmding is
at least consistent with the hypothesis that lung vessel distensibility is reduced or
that PBV is near its maximum capacity at rest in patients with a low DLCO and
high PAPm, and that further increases in blood flow result only in increasing PAP
and PVR (195).
A significant relationship between exercise DLCO and PVR has been demon-
strated in patients with COPO (22) and sarcoidosis (10) (figure 7-5). However,
due to the small number of patients, the clinical significance of these reports is
unclear, and broad generalizations to patients with other types of chronic lung
disease cannot be made at this time. Furthermore, interactions between changes in
arterial oxygen saturation, pH and PaC0 2, and pulmonary vascular resistance that
are also likely to be present during exercise may make such a relationship difficult
to prove. Finally, the question remains of whether symptom-limited exercise in
these patients also represents the level of cardiac output at which this effect may
occur.
If a significant relationship between DLCO and PVR can be documented,
exercise OLCO may provide important clues to the nature of increased pulmonary
vascular resistance in some patients and may be useful in determining how a patient
may respond to vasodilator therapy. An increase in exercise OLCO might imply
a reversible (potentially treatable) process and that further capillary recruitment and
distention is still possible. In contrast, a fixed OLCO might indicate anatomically
irreversible changes, suggesting that the response to increases in cardiac output as
the result of therapy would be limited.
The OLCO is also a reflection of an important cause of hypoxemia in some
patients (196,197). Alveolar capillary block is the term frequently used to describe
the concept of thickening of the alveolar capillary membrane resulting in an
impaired diffusion of oxygen (196). However, most studies suggest that, despite
the descriptive nature of this term, severe V/Q abnormalities that are also seen with
many of the diseases associated with alveolar capillary block are probably the most
important cause of hypoxemia, at least at rest (114,199).
The relationship between oxygen saturation and the capillary transit time of
SEA LEVEL 3100 METERS

lOOr----:::=::!:==t
If 80
z
Q
~ 60
a:::
;:)
~
en 40
Capillary Capillary
Transit Time-: Transit Tim~
I I
°O~--~--~L-~-L~
0.2 0.4 0.6 0 0.2 0.4 0.6
TIME IN SECONDS SPENT IN LUNG CAPILLARIES
Figure 7-10. The relationship between oxygen saturation (%) of pulmonary capillary blood
and the time (seconds) spent in transit through the lung capillaries. Calculations were done
assuming a normal maximum cardiac output and diffusing capacity. The effect of decreasing the
gradient between alveolar and capillary oxygen tension is demonstrated using two different PA
02 based on an altitude of 3,100 meters and sea level of 110 mmHg and 70 mmHg,
respectively. See text for discussion of the clinical implications of decreasing PA0 2 and DLCO
and changing cardiac output based on this relationship. (Reproduced with permission from
reference 114.)
blood through the lungs during heavy exercise in normals at both sea level (PA0 2
= 110 mmHg) and 3,100 meters (PA0 2 = 70 mmHg) is shown in figure 7-10
(114). Blood remains in the capillaries nearly twice as long as required for complete
oxygen saturation, even at a very high cardiac output. As a result cardiac output,
theoretically, could double (or capillary transit time could be halved) before
end-capillary oxygen saturation would decrease. In contrast, the driving pressure
between the alveolus and capillary blood at 3,100 m is low enough that the time
spent in the lung capillaries at high cardiac outputs may be insufficient for equilibra-
tion and arterial hypoxemia could develop, particularly if further increases in
cardiac output were possible (114) or alveolar oxygen tension fell even lower.
The clinical implications of figure 7-10 are that the DLCO could theoretically
fall to less than 50% of predicted at sea level before maximum exercise tolerance
would be impaired by arterial hypoxemia due to diffusion abnormalities alone
(114). Additional reductions in P A 0zdue to abnormal distribution of ventilation
that also occur in many of the chronic lung diseases associated with a reduction
in DLCO could precipitate arterial hypoxemia and limit exercise tolerance at lower
211
cardiac outputs in these patients. Patients with an abnormal DLCO who desaturate
during exercise generally have both severe diffusion and V/o. abnormalities rather
than alveolar capillary block alone (198).
A patient with a low DLCO who also desaturates during exercise might be at
risk to develop hypoxemia during vasodilator therapy since cardiac output may
increase, V/o. may deteriorate, or both may occur (1,16). The interaction between
cardiac output and arterial blood gas abnormalities in patients with cor pulmonale
is so complex that this concept may be too simplistic to be clinically useful. The
effect of a low cardiac output in the setting of right ventricular failure would cause
capillary transit time to lengthen and mixed venous oxygen to fall. The combined
/0.
action of these two effects, coupled with V abnormalities and all changes related
to vasodilators, would determine end-capillary oxygen saturation (114,196,197).
ARTERIAL BLOOD GASES. The strength of the correlations between resting arterial
oxygen saturation and PAPm (figure 7-6) suggests that arterial blood gas abnor-
malities are a major determinant of pulmonary artery pressure, particularly at low
oxygen saturations. However, arterial blood gas changes during exercise do not
appear to be specific for cor pulmonale except that further deterioration could
worsen pulmonary hypertension at any level of exercise (131,137,146,167). A fall
in exercise PaOz has also been associated with decreased survival in patients with
COPD (143).
The alveolar arterial oxygen gradient (P[A-a]0z) is the most useful means to
clinically assess arterial hypoxemia during exercise in patients with chronic lung
disease because it is unaffected by hyperventilation. Normally, P (A-a)02 decreases
slightly with mild exercise due to improved V/o., but during heavy work it begins
to increase due to a reduced mixed venous oxygenation (Pv0z). A widened
P(A-a)02 at rest does not necessarily imply that an irreversible abnormality is
present unless it either does not change or increases significantly during exercise
(112).
The mechanism(s) responsible for hypoxemia and thus an increased P(A-a)02
include the following:
a. Anatomic right-to-Ieft shunt.

b. Abnormal diffusion.
c. Abnormal V/o..
d. Low Pv02•
The presence of a shunt can be excluded in patients who are not at risk for CO 2
narcosis by measuring PaOz while breathing 100% oxygen. The possible role of
diffusion abnormalities has been briefly discussed. Abnormal V/o. is the most
common and usually the most clinically obvious cause of hypoxemia in patients
with chronic lung disease and is discussed in greater detail in chapter 2.
The distribution of V/o., measured using the multiple inert gas elimination
method, is not significantly changed during exercise in patients with COPD. As
a result mixed venous P0 2 contributes significantly to the level of arterial oxygena-
tion (58). In contrast, similar studies in patients with V/Q abnormalities secondary
to cystic fibrosis demonstrated substantial alterations in the pattern ofV/Q distribu-
tion during exercise in some patients which resulted in improved gas exchange and
minimal changes in arterial P0 2. The resting V/Q pattern of distribution in these
patients appeared to be unrelated to clinical status, degree of airways obstruction,
or pulmonary artery p'ressure (45). In the patients with COPD, type A or "pink
puffers" had high V/Q areas, and type B or "blue bloaters" more variably had
distinct low and/or high V/Q areas (58). No diffusion impairment for oxygen was
noted in either group (45,58).
The potential significance of a reduction in Pv02 is much more difficult to detect
clinically. Pv02 is most commonly reduced as a result of a decreased arterial
Pa02' decreased cardiac output, increased oxygen consumption, or increased blood
flow to tissues with high oxygen extraction (123). A reduction in Pv02 secondary
to inadequate cardiac output in patients with pulmonary heart disease may have
a greater role in the development of arterial hypoxemia than in other patients with
chronic lung disease. However, the range of cardiac output and oxygen consump-
tion in these latter patients is such that a low Pv02 does not appear specific for
cor pulmonale (201).
The level of oxygenation of mixed venous blood may contribute to the modula-
tion of pulmonary vascular tone either by acting directly on vascular smooth
muscle or indirectly by decreasing arterial oxygenation (Pa02) (202), alveolar
oxygen tension (PAO~ (203), or both. Thus, a low Pv02 may also playa major
role in the development of the increased pulmonary vascular resistance in patients
with pulmonary heart disease.
Summary
Despite a paucity of data, it is clear that some measurements made during exercise
testing (stages 1-4) can be helpful in the diagnostic evaluation of some patients with
suspected or known pulmonary hemodynamic abnormalities. For example, an
increase in VD /VT during exercise strongly suggests the presence of underlying
pulmonary vascular disease, particularly if the DLCO is also abnormal. In some
patients exercise DLCO appears to correlate with pulmonary resistance, but more
work is needed to firmly establish this relationship before it can be clinically useful.
Abnormalities of resting pulmonary function tests in combination with either of
these abnormalities would be a strong clinical indication for further investigation,
possibly including right heart catheterization.
At the present time hemodynamic measurements made during stage 4 are still
necessary to document the severity of pulmonary hypertension and the presence
of right ventricular failure and to confirm the presence or absence of significant
left ventricular dysfunction. Evidence strongly suggests that radionuclide angiogra-
phy may also be useful in this regard, but more work is needed to better define
its role, particularly as part of an exercise test.
Possibly, there may be a physiologic profile of exercise abnormalities that is
typical of a patient with pulmonary heart disease (108,111). For example, patients
213
with cardiac limitations typically have a normal Pa02 and a normal or decreased
PaC02, normal MVV, while V0 2 max, AT, and oxygen pulse are decreased.
VD /VT may be normal or increased at rest but usually decreases during exercise.
In contrast, patients with ventilatory limitation~ ty'pically have an abnormal
Pa02 and PaC02 and decreased MVV, a high VE /V0 2 but a normal AT and
oxygen pulse. VD /VT is more likely to be increased at rest, but subsequent changes
during exercise are variable (125).
In contrast to patients solely limited by either heart or lung disease, the physi-
ologic proftle of patients with pulmonary heart disease may prove to be indicative
of both a ventilatory and cardiac limitation. That is, ventilatory parameters will
be abnormal in addition to an increased heart rate, abnormal oxygen pulse, and a
low AT. Whether these measurements are sufficiently distinct or speciftc in a
clinical setting to separate patients with lung disease from those with cor pulmonale
is unknown, but deserves additional investigation.
At the present time the exercise response of patients with pulmonary heart disease
does not appear to be sufficiently speciftc to distinguish patients with chronic lung
disease from those with cor pulmonale. Results do quantify pathophysiologic
abnormalities that may explain symptoms, but this does not necessarily imply that
these abnormalities also account for the development of pulmonary hypertension.
As a result the current role of exercise testing in the evaluation of patients with
cor pulmonale overlaps with the clinical objectives for exercise testing in patients
with severe chronic lung disease. With continued research and a better understand-
ing of many of the complex physiologic interactions discussed, the clinical indica-
tions and diagnostic importance of exercise testing in patients with cor pulmonale
seem likely to be signiftcantly expanded in the near future.
REFERENCES
1. Rubin Lj, Peter RH. Hemodynamics at rest and during exercise after oral hydralazine in patients
with cor pulmonale. Am j Cardiol 47: 116-122, 1981.
2. Sturani C, Bassein L, Schiavina M, Gunella G. Oral nifedipine in chronic cor pulmonale
secondary to severe chronic obstructive pulmonary disease. Chest 84: 135-142, 1983.
3. Pearl RG, Rosenthal MH, Schroeder JJ, Ashton JP A. Acute hemodynamic effects of nitroglyc-
erin in pulmonary hypertension. Ann Intern Med 99: 9-13, 1983.
4. Rubin Lj, Peter RH. Oral hydralazine therapy for primary pulmonary hypertension. N Engl
j Med 302: 69-73, 1980.
5. Kadowitz Pj, Hyman AL. Hydralazine and the treatment of primary pulmonary hypertension.
N Engl j Med 307: 1357-1358, 1982.
6. Morrison D, Goldman S, Wright AL, Henry R, Sorenson S, Caldwell j, Ritchie J. The effect
of pulmonary hypertension on systolic function of the right ventricle. Chest 84: 250-257, 1983.
7. Berger Hj, Matthay RA, Loke j, Marshall RC, Gottschalk A, Zaret B. Assessment of cardiac
performance with quantitative radionuclide angiocardiography: Right and left ventricular ejec-
tion fraction with reference to fmdings in chronic obstructive pulmonary disease. Am j Cardiol
41: 897-905, 1978.
8. Nanda NC, Gramiak R, Robinson n, Shah PM. Echocardiographic evaluation of pulmonary
hypertension. Circulation 50: 575-581, 1974.
9. Hirschfeld S, Meyer R, Schwartz DC, Korfhagen j, Kaplan S. The echocardiographic assessment
of pulmonary artery pressure and resistance. Circulation 52: 642-650, 1975.
10. Emirgil C, Sobol Bj, Herbert WH, Trout K. The lesser circulation in pulmonary fibrosis
secondary to sarcoidosis and its relationship to respiratory function. Chest 60: 371-378, 1971.
11. Enson Y, Thomas HM, III, Bosken CH, Wood JA, leRoy EC, Blanc W A, Wigger HJ, Harvey
RM. Pulmonary hypertension in interstitial lung disease: Relation of vascular resistance to
abnormal lung structure. Trans Assoc Am Phys 88: 248-255, 1975.
12. Ferrer MI, Harvey RM, Cathcart RT, Webster CA, Richards DW, Cournand A. Some effects
of digoxin upon the heart and circulation in man. Digoxin in chronic cor pulmonale. Circulation
1: 161-186, 1950.
13. Harvey RM, Ferrer MI. A clinical consideration of cor pulmonale. Circulation 21: 236-255,
1960.
14. Ferrer MI. Cor pulmonale (pulmonary heart disease): Present day starus. Am Heart J 89:
657-664, 1975.
15. Continuous nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease in a
clinical trial. Ann Intern Med 93: 391-398, 1980.
16. Packer M, Greenberg B, Massie B, Dash H. Deleterious effects of hydralazine in patients with
pulmonary hypertension. N Engl J Med 306: 1326-1331, 1982.
17. Parker JO, Kelkan K, West po. Hemodynamic effects of aminophylline in cor pulmonale.
Circulation 33: 17-25, 1966.
18. Evans TO, VanDerReis L, Selzer A. Circulatory effects of chronic pulmonary emphysema. Am
Heart J 66: 741-747, 1963.
19. Harris P, Segel N, Green K, Housley E. The influence of the airways resistance and alveolar
pressure on the pulmonary vascular resistance in chronic bronchitis. Cardiovasc Dis 2: 84-92,
1968.
21. Timsit G, Neukirch F, Castillon D, Perron M, Verdier F, Drutel P, Legrand M, Botto MJ,
Lesobre R. Statistical correlation between PAP, spirographic data and arterial and mixed venous
blood gases in 22 subjects with chronic obstructive lung disease. Prog Resp Res 9: 105-111,
1975.
22. Emirgil C, Sobol BJ, Herbert WH, Trout KW. Routine pulmonary function studies as a key
to the starus of the lesser circulation in chronic obstructive pulmonary disease. Am J Med 50:
191-199, 1971.
23. Borden CW, Wilson RH, Ebert RU, Wells HS. Pulmonary hypertension in chronic pulmonary
emphysema. Am J med 8: 701-709, 1950.
24. Harvey RM, Enson Y, Betti R, Lewis ML, Rochester DF, Ferrer MI. Further observations
on the effects of hydrogen ion on the pulmonary circulation. Circulation 35: 1019-1027,
1967.
25. Yu PNG, Lovejoy FW,Joos HA, Nye RE, McCann WS. Studies of pulmonary hypertension.
I. Pulmonary circulatory dynamics in patients with pulmonary emphysema at rest. J Clin Invest
32: 130-137, 1953.
26. Emirgil C, Sobol BJ, Williams MH Jr. Long-term study of pulmonary sarcoidosis-The ef-
fect of steroid therapy as evaluated by pulmonary function studies. J Chronic Dis 22: 69-86,
1969.
27. Stem RC, Borkat G, Hirschfeld SS, Boat TF, Matthews LW, LiebmanJ, Doershuk CF. Heart
failure in cystic fibrosis. Am J Dis Child 134: 267-272, 1980.
28. Mounsey JPD, Ritzman LW, Selverstone NJ, Briscoe WA, McLemore GA. Circulatory changes
in severe pulmonary emphysema. Br Heart J 14: 153-172, 1952.
29. Harvey RM, Ferrer MI, Richards DW Jr, Cournand A. Influence of chronic obstructive
pulmonary heart disease on the heart and circulation. Am J Med 10: 719-738, 1951.
30. Williams JF, Behnke RH. The effect of pulmonary emphysema upon cardiopulmonary hemody-
namics at rest and during exercise. Ann Int Med 60: 824-842, 1964.
31. Lim TPK, Brownlee WE. Pulmonary hemodynamics in obstructive lung disease. Dis Chest 53:
113-125, 1968.
32. Enson Y, Giuntini C, Lewis ML, Morris TQ, Ferrer MI, Harvey RM. The influence of
hydrogen ion concenttation and hypoxia on the pulmonary circulation. J Clin Invest 43:
1146-1162, 1964.
33. Sackner MA, Akgun N, Kimbel P, Lewis DH. The pathophysiology of scleroderma involving
the heart and respiratory system. Ann Intern Med 60: 611-630, 1964.
34. Herbert FA, Nalunias BB, Gaensler EA, MacMahon HE. Pathophysiology of interstitial pulmo-
nary fibrosis. Arch Int Med 110: 628-648, 1962.
215
35. Williams MH, Adler .D, Colp C. Pulmonary function studies as an aid in the differential
diagnosis of pulmonary hypertension. Am J Med 47: 378-383, 1969.
36. Turino GM, Goldring RM, Fishman AP. Cor pulmonale in musculoskeletal abnormalities of
the thorax. Bull NY Acad Med 41: 959-980, 1965.
37. Harris P. Relation of airways resistance and pulmonary vascular resistance. Bull Physio-Path
Resp 4: 65-74, 1968.
38. Harris P, Heath D. The Human Pulmonary Circulation, 2nd edition. Edinburgh; Livingstone,
1977, pp. 522-546.
39. Whitaker W. Pulmonary hypertension in congestive heart failure complicating chronic lung
disease. Quart J Med 23: 57-72, 1954.
40. Aber GM, Baylery 1], Bishop JM. Inter-relationships between renal and cardiac function and
respiratory gas exchange in obstructive airways disease. Clin Sci 25: 159-170, 1963.
41. Coumand A. Some aspects of the pulmonary circulation in normal man and in chronic cardi-
opulmonary diseases. Circulation 2: 641-657, 1950.
42. Ferrer Ml. Disturbances in the circulation in patients with cor pulmonale. Bull NY Acad Med
41: 942-958, 1965.
43. Lewis CS, Samuels AJ, Daines MC, Hecht H. Chronic lung disease, polycythemia and congestive
heart failure. Circulation 6: 874-887, 1952.
44. Herles F,Jezek V, Daum S. Site of pulmonary resistance in cor pulmonale in chronic bronchitis.
Br Heart J 30: 654-660, 1968.
45. Dantzker DR, Patten GA, Bower JS. Gas exchange at rest and during exercise in adults with
cystic fibrosis. Am Rev Respir Dis 125: 400-405, 1982.
46. Segel N, Bishop JM. The circulation in the patients with chronic bronchitis and emphysema
at rest and during exercise, with special reference to the influence of changes in blood viscosity
and blood volume on the pulmonary circulation. J Clin Invest 45: 1555-1568, 1960.
47. Fowler NO, Wescott RN, Scott RC, Hess E. The cardiac output in chronic cor pulmonale.
Circulation 6: 888-893, 1952.
48. Enson Y. Pulmonary heart disease: Relation of pulmonary hypertension to abnormal lung
structure and function. Bull NY Acad Med 53: 551-566, 1977.
49. Harvey RM, Enson Y, Ferrer MI. A reconsideration of the origins of pulmonary hypertension.
Chest 59: 82-94, 1971.
51. Lockhardt A, Sestier F, Sentissi M, Gauthier .D, Schrijen F. Effects of exercise and postural
changes on pulmonary hemodynamics in patients with chronic lung disease. Scand J Resp Dis
77 (suppl): 77-81, 1971.
52. Williams MH Jr, Zohman LR. Cardiopulmonary function in chronic obstructive emphysema.
Am Rev Respir Dis 80: 689-699, 1959.
53. Westcott RN, Fowler NO, Scott RC, Harrenstein VD, McGuireJ. Anoxia and human pulmo-
nary vascular resistance. J Clin Invest 30: 957-970, 1951.
54. Bergofsky EH, Lehr DE, Fishman AP. The effect of changes in hydrogen ion concentration on
the pulmonary circulation. J Clin Invest 41: 1492-1502, 1962.
55. Fishman AP, Fritts WHJr, Cournand A. Effects of breathing carbon dioxide upon the pulmo-
nary circulation. Circulation 22: 220--225, 1960.
56. Abraham AS, Cole RB, Green ID, Hedworth-Whitty RB, Clarke SW, Bishop JM. Factors
contributing to the reversible pulmonary hypertension of patients with acute respiratory failure
studied by serial observations during recovery. Circ Res 24: 51-60, 1969.
57. Fishman AP, McClement J, Himmelstein A, Cournand A. Effects of acute anoxia on the
circulation and respiration in patients with chronic pulmonary disease studied during the steady
state. J Clin Invest 31: 770--781, 1952.
58. Wagner PD, Dantzker DR, Dueck R, ClausenJL, WestJB. Ventilation-perfusion inequality
in chronic obstructive pulmonary disease. J Clin Invest 59: 203-216, 1977.
59. Harvey RM. The influence of hydrogen ion in the control of pulmonary artery pressures in
patients with obstructive lung disease. Med Thorac 22: 108-117, 1965.
60. Housley E, Clarke SW, Hedworth-Witty RB, Bishop JM. Effect of acute and chronic acidemia
and associated hypoxia on the pulmonary circulation of patients with chronic bronchitis. Bull
Physio Path Resp 5: 347-361, 1969.
61. Paul G, Varnauskas SA, Forsberg R, Sannerstedt R, Widimsky J. Effect of carbon dioxide
breathing upon the pulmonary circulation in patients with mitral valve disease. Clin Sci 26:
111-120, 1964.
62. Lloyd TC Jr. Influence of blood pH on hypoxic pulmonary vasoconstriction. J Appl Physiol
21: 358-364, 1966.
63. Sibald W), Paterson NAM, Holliday RL, Anderson RA, Lobb TR, Duff ]H. Pulmonary
hypertension in sepsis. Chest 73: 583-591, 1978.
64. McNeill RS, Rankin J, Forster RE. The diffusing capacity of the pulmonary membrane and
the pulmonary capillary blood volume in cardiopulmonary disease. Clin Sci 17: 465-482, 1958.
65. Bates DV. The uptake of carbon monoxide in health and in emphysema. Clin Sci 11: 21-32,
1952.
66. Marshall R. A comparison of methods of measuring the diffusing capacity of the lungs for
carbon monoxide. Investigation by fractional analysis of the alveolar air. J Clin Invest 37:
397-407, 1958.
67. Ogilvie CM, Forster RE, Blakemore WS, Morton JW: A standardized breath holding technique
for the clinical measurement of the diffusing capacity of the lung for carbon monoxide. J Clin
Invest 36: 1-17, 1957.
68. Bates DV, Varvis q, Donevan RE, Christie RV. Variations in the pulmonary capillary blood
volume and membrane diffusion component in health and disease.J Clin Invest 39: 1401-1412,
1960.
69. Filley FG, MacIntosh DJ, Wright GW. Carbon monoxide uptake and pulmonary diffusing
capacity in normal subjects at rest and during exercise. J Clin Invest 33: 530-539, 1954.
70. Sackner MA, Greeneltch D, Heiman M, Epstein S, Atkins N. Diffusing capacity, membrane
diffusing capacity, capillary blood volume, pulmonary tissue volume and cardiac output mea-
sured by a rebreathing technique. Am Rev Respir Dis 111: 157-165, 1975.
71. Lewis BM, Lin TH, Noe FE, Hayford-Welsing EJ. The measurement of pulmonary diffusing
capacity for carbon monoxide by a rebreathing method.] Clin Invest 38: 2073-2086, 1959.
72. Roughton FJW, Forster RE. Relative importance of diffusion and chemical reaction rates in
determining rate of exchange of gases in the human lung, with special reference to true diffusing
capacity of pulmonary membrane and volume of blood in the lung capillaries. J Appl Physiol
11: 290-302, 1957.
73. Lewis BM, Lin TH, Noe FE, Komisaruk R. The measurement of pulmonary capillary blood
volume and pulmonary membrane diffusing capacity in normal subjects; the effects of exercise
and position. J Clin Invest 37: 1061-1070, 1958.
74. McCredie RM. The pulmonary capillary bed in various forms of pulmonary hypertension.
Circulation 33: 854-861, 1966.
75. McCredie RM. The diffusing characteristics and pressure-volume relationships of the pulmonary
capillary bed in mitral valve disease. J Clin Invest 43: 2279-2289, 1964.
76. Burgess JH. Pulmonary diffusing capacity in disorders of the pulmonary circulation. Circulation
39: 541-550, 1974.
77. MacIntosh DJ, Sinnott JC, Milne IG, Reid EAS. Some aspects of disordered pulmonary function
in mitral stenosis. Ann Int Med 49: 1294-1304, 1958.
78. Riley RL, Johns q, Cohen G, Cohn JE, Carroll DG, Shepard RH. The diffusing capacity of
the lungs in patients with mitral stenosis studied postoperatively. J Clin Invest 35: 1008-1014,
1956.
79. Parker F Jr, Weiss S. The nature and significance of structural changes in the lungs in mitral
stenosis. Am J Path 12: 573-598, 1936.
80. Jain BP, Pande IN, Guleria JS. Membrane diffusing capacity and pulmonary capillary blood
volume in chronic obstructive lung disease. Am Rev Respir Dis 105: 900-907, 1972.
81. Herzog H, Daum S. Lung function in primary obstructive pulmonary hypertension. Prog Resp
Res 5: 400-405, 1970.
82. Sundstrom G, Zauner CW, Arborelius M Jr. Decrease in pulmonary diffusing capacity during
lipid infusion in healthy men. J Appl Physiol 34: 816--820, 1973.
83. Sharma GVRK, Burleson VA, Sasahara AA. Effect of thrombolytic therapy on pulmonary
capillary bed volume in patients with pulmonary embolism. N Engl J Med 303: 842-845, 1980.
84. Parker JO, Ashekian PB, DiGiorgi S, West RO. Hemodynamic effects of aminophylline in
chronic obstructive pulmonary disease. Circulation 35: 365-372, 1967.
85. Stockley RA, Finnegan P, Bishop JM. Effect of intravenous terbutaline on arterial blood gas
217
tensions, ventilation, and pulmonary circulation in patients with chronic bronchitis and cor
pulmonale. Thorax 32: 601-605, 1977.
86. Goldring RM, Fishman AP, Turino GM, Cohen HI, Denning CR, Andersen DH. Pulmo-
nary hypertension and cor pulmonale in cystic fibrosis of the pancreas. J Ped 65: 501-524,
1964.
87. Hickman JI, Cargill WHo Effect of exercise on cardiac output and pulmonary arterial pressure
in normal persons and in patients with cardiovascular disease and pulmonary emphysema. J Clin
Invest 27: 10-23, 1948.
88. Kanner RE, Morris AH. Clinical pulmonary function testing. Salt Lake City Intermountain
Thoracic Society. VI: 37-93, 1975.
89. Crapo RO, Morris AH, Gardner RM. Reference spirometric values using techniques and
equipment that meet ATS recommendations. Am Rev Respir Dis 123: 659-664, 1981.
90. Thurlbeck WM, Henderson JA, Fraser RG, Bates DV. Chronic obstructive lung disease: A
comparison between clinical, roentgenologic, functional and morphologic criteria in chronic
bronchitis, emphysema, asthma and bronchiectasis. Medicine 49: 81-145, 1970.
91. Filley GF, Beckwith HJ, Reeves]T, Mitchell RS. Chronic obstructive bronchopulmonary
disease. II. Oxygen transport in two clinical types. Am J Med 44: 26-38, 1968.
92. Bates DV. Impairment of respiratory function in bronchial asthma. Clin Sci 11: 203-207,
1952.
93. Guilleminault C, Eldridge FL, Simmon FB, Dement WC. Sleep apnea syndrome: Can it induce
hemodynamic changes? West J Med 123: 7-16, 1975.
94. Koo KW, Sax DS, Snider GL. Arterial blood gases and pH during sleep in chronic obstructive
pulmonary disease. Am J Med 58: 663-670, 1975.
95. Flick MR, Block AJ. Continuous in vivo monitoring of arterial oxygenation in chronic
obstructive lung disease. Ann Int Med 86: 725--730, 1977.
96. Boysen PG, Block AJ, Wynne JW, Hunt LA, Flick MR. Nocturnal pulmonary hypertension
97. Kearley R, Wynne JW, Block AJ, Boysen PG, Lindsey S, Martin C. The effect of low-flow
oxygen on sleep-disordered breathing and oxygen desaturation. Chest 78: 682-685, 1980.
98. McIlroy MB, Apthorp GH. Pulmonary function in pulmonary hypertension. Br Heart J 20:
397-402, 1958.
99. Sleeper JC, Orgain ES, McIntosh HD. Primary pulmonary hypertension. Review of clinical
features and pathologic physiology with a report of pulmonary hemodynamics derived from
repeated catheterization. Circulation 26: 1358-1369, 1962.
100. Fowler NO, Black-Schaffer B, Scott RC, Gueron M. Idiopathic and thromboembolic pulmo-
nary hypertension. Am J Med 40: 331-345, 1966.
101. Scharf SM, Feldman NT, Graboys TB, Wellman.u. Restrictive ventilatory defect in a patient
with primary pulmonary hypertension. Am Rev Respir Dis 118: 409-413, 1978.
102. Benjamin.u, Murtagh PS, Proctor DF, Menkes JA, Permutt S. Pulmonary vascular interdepend-
ence in excised dog lobes. J Appl Physiol 37: 887--894, 1974.
103. Yernault JC, Detroyer A. Mechanics of breathing in patients with primary pulmonary hyperten-
sion. Am Rev Respir Dis 119: 320-321, 1979.
104. Llamas R, Swenson EW. Diagnostic clues in pulmonary thrombembolism evaluated by angio-
graphic and ventilation-blood flow studies. Thorax 20: 327-336, 1965.
105. Jones NL, GoodwinJF. Respiratory function in pulmonary thromboembolic disorders. Br Med
J 1: 1089-1093, 1965.
106. NadelJA, Gold WM,Jennings DB, Wright RR, Fudenberg HH. Unusual disease of pulmonary
arteries with dyspnea. Am J Med 41: 440-447, 1966.
107. Wasserman K, VanKessel AL, Burton GG. Interaction of physiological mechanisms during
exercise. J Appl Physiol 22: 71--85, 1967.
108. Wasserman K, Whipp BJ. Exercise physiology in health and disease. Am Rev Respir Dis 112:
219-249, 1975.
109. Wasserman K, Whipp BJ, Koyal SN, Beaver WL. Anaerobic threshold and respiratory gas
exchange during exercise. J Appl Physiol 35: 236-243, 1973.
110. Jones NL. Exercise testing in pulmonary evaluation: Rationale, methods and normal respiratory
response to exercise. N Engl J Med 293: 541-544, 647-650, 1975.
111. Brown HV, Wasserman K. Exercise performance in chronic obstructive pulmonary diseases.
Med Clin North Am 65: 525--547, 1981.
112. Jones NL, Campbell EJM. Clinical Exercise Testing. Philadelphia: W. B. Saunders, 1982.
113. Harvey M, Levine RJ. The risk of research procedures: Methodologic problems and proposed
standards. Clin Res 31: 126-139, 1983.
114. Johnson RL Jr. Exercise testing in lung disease. In Diagnostic Techniques in Pulmonary Disease,
Sackner MA (ed). New York: Marcel Dekker, 1980, pp. 473-501.
115. Sue DY, HansenJE, Blais M, Wassennan K. Measurement and analysis of gas exchange during
exercise using a programmable calculator. J Appl Physiol 49: 456-461, 1980.
116. Wassennan K. Breathing during exercise. N Engl J Med 298: 780-785, 1978.
117. Furuike AN, Sue DY, HansenJE, Wassennan K. Comparison of physiologic deadspace/tidal
volume ratio and alveolar-arterial P02 difference during incremented and constant work exer-
cise. Am Rev Respir Dis 125: 579-583, 1982.
118. Shuey CB Jr, Pierce AK, Johnson RL Jr. An evaluation of exercise tests in chronic obstructive
lung disease. J Appl Physiol 27: 256-261, 1969.
119. Downs JB, Rackstein AD, Klein EF Jr, Hawkins IF Jr. Hazards of radial artery catheterization.
Anesthesiology 38(3): 283-286, 1973.
120. Lowenstein E, Little JW, III, Hing HL. Prevention of cerebral embolization from flushing radial
artery cannulas. N Engl J Med 285: 1414-1415, 1971.
121. Bedford RF, Wollman H. Complications of percutaneous radial artery cannulation: An objec-
tive prospective study in man. Anesthesiology 38: 228-236, 1973.
122. Marcus JR, Mclean RL, Duffell GM, Ingram RH. Exercise perfonnance in relation to the
pathophysiologic type of chronic obstructive pulmonary disease. Am J Med 14-22, 1970.
123. Hughes RL, Davison R. Limitations of exercise reconditioning in COLD. Chest 83: 241-249,
1983.
124. Spiro SG, Hahn HL, Edwards RHT, Pride NB. Cardiorespiratory adaptations at the start of
exercise in normal subjects and in patients with chronic obstructive bronchitis. Clin Sci Mol
Med 47: 165-172, 1974.
125. Wassennan K. Physiology of gas exchange and exertional dyspnea. Clin Sci 61: 7-13, 1981.
126. Rebuck AS, Vandenberg RA. The relationship between pulmonary artery pressure and physi-
ologic deadspace in patients with obstructive lung disease. Am Rev Respir Dis 107: 423-428,
1973.
127. Pierce AK, Taylor HF, Archer RK, Miller WF. Responses to exercise training in patients with
emphysema. Arch Int Med 113: 28-36, 1964.
128. Kanarek D, Kaplan D, Kazemi H. The anaerobic threshold in severe chronic obstructive lung
disease. Bull Europ Physiopath Resp 15: 163-169, 1979.
129. DiPrampero PE, Davies CTM, Cerretelli P, Margaria R. An analysis of 02 debt contracted in
submaximal exercise. J Appl Physiol 29: 547-551, 1970.
130. Whipp BJ, Wassennan K. Oxygen uptake kinetics for various intensities of constant workload.
J Appl Physiol 33: 351-356, 1972.
131. Minh YD, Lee HM, Dolan GF, Light RW, Bell J, Vasquez P. Hypoxemia during exercise
in patients with chronic obstructive pulmonary disease. Am Rev Respir Dis 120: 787-794,
1979.
132. Belman MJ, Wasserman K. Exercise training and testing in patients with chronic obstructive
pulmonary disease. Basics of RD 10: 38-43, 1982.
133. Bradley BL, Garner AE, Billie D, Mestas JM, Fonnan M. Oxygen-assisted exercise in chronic
obstructive lung disease. Am Rev Respir Dis 118: 239-243, 1978.
134. Jones NL, Jone G, Edwards HT. Exercise tolerance in chronic airway obstruction. Am Rev
Respir Dis 103: 477-491, 1974.
135. Annstrong BW, Workman]N, Hurt HH Jr, Roemich WR. Clinico-physiologic evaluation
of physical working capacity in persons with pulmonary disease. Am Rev Respir Dis 93: 90-99,
223-233, 1966.
136. Wehr KL, Johnson RL Jr. Maximal oxygen consumption in patients with lung disease. J Clin
Invest 58: 880-890, 1976.
137. Young IH, Woolcock AJ. Arterial blood gas tension changes at the start of exercise in chronic
obstructive pulmonary disease. Am Rev Respir Dis 119: 213-221, 1979.
138. Field S, Kelly SM, Macklem PT. The oxygen cost of breathing in patients with cardiorespiratory
disease. Am Rev Respir Dis 126: 9-13, 1982.
139. McIlroy MB. Dyspnea and the work of breathing in diseases of the heart and lungs. Prog
Cardiovasc Dis 1: 284-297, 1959.
219
140. Kawakami Y, Kishi F, Yamamoto H, Miyamato K. Relation of oxygen delivery, mixed venous
oxygenation, and pulmonary hemodynamics to prognosis in chronic obstructive pulmonary
disease. N Engl J Med 208: 1045-1049, 1983.
141. Bergofsky EH. Tissue oxygen delivery and cor pulmonale in chronic obstructive pulmonary
142. Tenney SM, Mithoefer JC. The relationship of mixed venous oxygenation to oxygen transport:
With special reference to adaptations to high altitude and pulmonary disease. Am Rev Respir
Dis 125: 474-479, 1982.
143. Kawakami Y, Terai T, Yamamoto H, Murao M. Exercise and oxygen inhalation in relation
to prognosis of chronic obstructive pulmonary disease. Chest 81: 182-188, 1982.
144. Mitchell RS, Webb NC, Filley GF. Chronic obstructive bronchopulmonary disease. III. Factors
influencing prognosis. Am Rev Respir Dis 89: 878-896, 1964.
145. Renzetti AD Jr, McClementJH, Li HBD. The Veterans Administration Cooperative Study of
pulmonary function. III. Mortality in relation to respiratory function in chronic obstructive
146. Rafferstin B, Escourrou P, Legrand A, Duroux P, Lockhart A. Circulatory transport of oxygen
in patients with chronic airflow obstruction exercising maximally. Am Rev Respir Dis 125:
426-431, 1982.
147. Jezek V, Schrijen F, Sadoul P. Right ventricular function and pulmonary hemodynamics during
exercise in patients with chronic obstructive bronchopulmonary disease. Cardiology 58: 20-31,
1973.
148. Berger HJ, Matthay RA, Pytlik LM, Gotrschalk A, Zaret BL. First pass radionuclide assessment
of right and left ventricular performance in patients with cardiac and pulmonary disease. Semin
Nucl Med 9: 275-295, 1982.
149. Matthay RA, Berger HJ, Loke J, Gottschalk A, Zaret BL. Effects of aminophylline upon right
and left ventricular performance in chronic obstructive pulmonary disease: Noninvasive assess-
ment by radionuclide angiocardiography. Am J Med 65: 903-910, 1978.
150. Matthay RA, Berger HJ, Davies RA, Loke J, Mahler DA, Gottschalk A, Zaret BL. Right and
left ventricular exercise performance in chronic obstructive pulmonary disease: Radionuclide
assessment. Ann Intern Med 93: 234-239, 1980.
151. Piene H, Sund T. Flow and power output of right ventricle facing load with variable input
impedance. Am J Physiol 237: H125-H130, 1979.
152. Sarnoff SJ, Berglund E. Ventricular function. I. Starling's law of the heart studied by means
of simultaneous right and left ventricular function curves in the dog. Circulation 9: 706-718,
1954.
153. Smith DE, Bissett jK, Phillips JR, Doherty JE, Murphy ML. hnproved right ventricular systolic
time intervals after digitalis in patients with cor pulmonale and chronic obstructive pulmonary
disease. Am J Cardiol 41: 1299-1304, 1978.
154. Jezek V. Rate of right ventricular isometric pressure increase in patients with chronic cor
pulmonale studied by ventricular derivative curves. Cardiologia 48: 461-470, 1966.
155. Rubin LJ. Cardiovascular effects of vasodilator therapy of pulmonary artery hypertension.
Clinics in Chest Medicine 4: 309-314, 1983.
156. Goldberg HS, Rabson J. Control of cardiac output by systemic vessels. Am J Med 47: 696-702,
1981.
157. Horsfield K, Segel N, Bishop JM. The pulmonary circulation in chronic bronchitis at rest and
during exercise breathing air and 80% oxygen. Clin Sci 43: 473-483, 1968.
158. Lockhart A, Trareva M, Nader F, Leblanc P, Schrijen F, Sadoul P. Elevated pulmonary artery
wedge pressure at rest and during exercise in chronic bronchitis: Fact or fancy. Clin Sci 37:
503-517, 1969.
159. Rao BS, Cohn KE, Eldridge FL, Hancock EW. Left ventricular failure secondary to chronic
160. Shaw DB, Grover RF, Reeves JT, Blount G Jr. Pulmonary circulation in chronic bronchitis
and emphysema. Br Heart J 27: 674-683, 1965.
161. Widirnsky J, Riedel M, Stanek V. Central hemodynamics during exercise in patients with
restrictive pulmonary disease. Bull Europ Physiopath Resp 13: 369-379, 1977.
162. Wilson JR, Mason UG, Bahler RC, Chester EH, Picken lJ, Baum GL. Vectorcardiographic
detection of early hemodynamic abnormalities in chronic obstructive pulmonary disease. Chest
76: 160-165, 1979.
163. Chapman CB, Fisher]N, Sproule BJ. Behavior of stroke volume at rest and during exercise
in human beings. J Clin Invest 39: 1208-1213, 1960.
164. Matthay RA, Berger Hj, Davies RA, Loke J, Zaret BL, Gottschalk A. Effect of steady state
exercise on right and left ventricular performance in chronic obstructive pulmonary disease.
Chest 77 (suppl): 303-304, 1980.
165. Brent BN, Berger Hj, Matthay RA, Mahler D, Pytlik L, Zaret BL. Physiologic correlates of
right ventricular ejection fraction in chronic obstructive pulmonary disease: A combined ra-
dionuclide and hemodynamic study. Am J Cardiol 50: 255-262, 1982.
166. Riley RL, Himmelstein A, Motley HL. Studies of the pulmonary circulation at rest and during
exercise in normal individuals and in patients with chronic pulmonary disease. Am J Physiol
152: 372-382, 1948.
167. Kitchin AH, Lowther CP, Matthews MB. The effects of exercise of breathing oxygen enriched
air on the pulmonary circulation in emphysema. Clin Sci 21: 93--106, 1961.
168. Burkart F. The exercise test in patients with obstructive pulmonary hypertension. Prog Resp
Res 5: 395-399, 1970.
169. Hermiller JB, Bambach D, Thompson MJ, Huss P, Fontana ME, Magorien RD, Unverferth
DV, Leier CV. Vasodilators or prostaglandin inhibitors in primary pulmonary hypertension.
Ann Int Med 97: 480-489, 1982.
170. Gazetopoulos N, Salonikides N, Davies H. Cardiopulmonary function in patients with pulmo-
nary hypertension. Br Heart J 36: 19-28, 1974.
171. Cotes JE, Pisa Z, Thomas AJ. Effect of breathing oxygen upon cardiac output, heart rate,
ventilation, systemic and pulmonary blood pressure in patients with chronic lung disease. Clin
Sci 25: 305-321, 1963.
172. Rubin LJ, Handel F, Peter RH. The effects of oral hydralazine on right ventricular end-diastolic
173. Rubin LJ, Nicod P, Hillis LD, Firth BG. The acute and chronic treatment of primary pulmonary
hypertension with nifedipine: A hemodynamic and scintigraphic evaluation. Ann Int Med 99:
433--438, 1983.
174. Morrison D, Sorenson S, Caldwell J, Wright N, Ritchie J, Kennedy JW, Hamilton G. The
normal right ventricular response to supine exercise. Chest 82: 686-691, 1982.
175. Martorana PA, Schaper J, Van Even P. The effect of physical exercise on elastase-induced
emphysema in hamsters. Am Rev Respir Dis 120: 1209-1213, 1979.
176. Weitzenblum E, Loiseau A, Hirth C, Mirhom R, Rasaholinjanahary J. Course of pulmonary
hemodynamics in patients with chronic obstructive pulmonary disease. Chest 75: 656-662, 1979.
177. Fishman AP, Ledlie]F. Dyspnea. Bull Europ Physiopath Resp 15: 789-804, 1979.
178. Gandevia B. Ventilatory response to exercise and the results of standardized exercise test in
chronic obstructive lung disease. Am Rev Respir Dis 88: 66, 406-408, 1963.
179. McFadden ERJr, Ingram RHJr. Exercise-induced asthma. N EnglJ Med 301: 763--769, 1979.
180. McFadden ER Jr. Exercise-induced asthma. Am J Med 68: 471-472, 1980.
181. Rice DL, Awe Rj, Gaasch WH, Alexander JK, Jenkins DE. Wedge pressure measurement in
obstructive pulmonary disease. Chest 66: 628-632, 1974.
182. Buda AJ, Pinskey MR, Ingels NB Jr, Daughters GT II, Stinson EB, Alderman EL. Effect of
intrathoracic pressure on left ventricular performance. N Engl J Med 301: 453--459, 1979.
183. Permutt S. Relation between pulmonary arterial pressure and pleural pressure during the acute
asthmatic attack. Chest 63 (suppl): 25-28, 1973.
n,
184. Culver BH, Marini Butler J. Lung volume and pleural pressure effects on ventricular function.
J Appl Physiol 50: 630--635, 1981.
185. Nadel JA, Gold WM, Burgess]H. Early diagnosis of chronic pulmonary vascular obstruction.
Am J Med 44: 16-25, 1968.
186. Heilman RS, Tabakin BS, Hanson JS, Naeye RL. Alterations of circulatory and ventilatory
dynamics in pulmonary vascular obstruction secondary to recurrent pulmonary emboli. Am J
Med 32: 298-303, 1962.
187. Dantzker DR, Bower JS. Mechanisms of gas exchange abnormality in patients with chronic
obliterative pulmonary vascular disease. J Clin Invest 64: 1050--1055, 1979.
188. Mohsenifar Z, Tashkin DP, Levy SE, Bjerke RD, Clements PJ, Furst D. Lack of sensitivity of
measurements ofVD /VT at rest and during exercise in detection of hemodynamically significant
pulmonary vascular abnormalities in collagen vascular disease. Am Rev Respir Dis 123: 508-
512, 1981.
221
189. Wise RA, Wigley F, Newball HH, Stevens MB. The effect of cold exposure on diffusing
capacity in patients with Raynaud's phenomenon. Chest 81: 695-698, 1982.
190. Ettinger WH, Wise RA, Stevens MB, Wigley FM. Absence of positional changes in pulmonary
diffusing capacity in systemic sclerosis. Am J Med 75: 305-312, 1983.
191. Miller MJ. Effect of the cold pressor test on diffusing capacity; comparison of normal subjects
and those with Raynaud's disease and progressive systemic sclerosis. Chest 85: 264-266, 1983.
192. Jones NL. Pulmonary gas exchange during exercise in patients with chronic airway obstruction.
Clin Sci 31: 39-50, 1966.
193. Johnson RL Jr, Spicer WS, Bishop JM, Forster RE. Pulmonary capillary blood volume, flow
and diffusing capacity during exercise. J Appl Physiol 15: 893-902, 1970.
194. Bjure J. Pulmonary diffusing capacity for carbon monoxide in relation to cardiac output in man.
Scand J Clin Lab Invest 17 (suppl 81): 1-113, 1965.
195. Schrijen F, Urtiaga B. Pulmonary blood volume in chronic lung disease. Chest 81: 545-549,
1982.
196. Austrian R, McClementJH, Renzetti AD Jr, Donald KW, Riley RL, Coumand A. Clinical
and physiologic features of some types of pulmonary diseases with impairment of alveolar-
capillary diffusion. Am J Med 667-685, 1951.
197. Shepard RH, Cohn JE, Cohn G, Armstrong BW, Carroll DG, Denoso H, Riley RL. The
maximal diffusing capacity of the lung in chronic obstructive disease of the airways. Amer Rev
Tuberc 71: 249-259, 1955.
198. Finley TN, Swenson EW, Comroe JH Jr. The cause of arterial hypoxemia at rest in patients
with "alveolar capillary block syndrome." J Clin Invest 41: 618-622, 1962.
199. Johnson RL Jr. Pulmonary diffusion as a limiting factor in exercise stress. Circ Res 21 (suppl
11): 154-160, 1967.
200. DeGraff AC, Taylor HF, Ord]W, Chuang TH, Johnson RL Jr. Exercise limitation following
extensive pulmonary resection. J Clin Invest 44: 1514-1522, 1965.
201. Minh VD, Chun D, Dolan GF, Lee HM, Vasquez P. Mixed venous oxygenation, exercise, body
posture and V / Q ratio in chronic obstructive pulmonary disease. Am Rev Respir Dis 124:
226-231, 1981.
202. Hyman AL, Higashida RT, Spannhake EW, Kadowitz PJ. Pulmonary vasoconstrictor responses
to graded decreases in precapillary blood P0 2 in intact chest cat.J Appl Physiol51: 1009-1016,
1981.
203. Benumof]1, Pirlo AF, Johanson I, Trousdale FR. Interaction of Pv02 with Pa02 on hypoxic
pulmonary vasoconstriction. J Appl Physiol 51: 871-874, 1981.
8. RADIONUCLIDE ANGIOCARDIOGRAPHIC ASSESSMENT OF RIGHT
AND LEFT VENTRICULAR PERFORMANCE
RICHARD A. MATIHAY, M.D.

and
HARVEY J. BERGER, M.D.
A lack of noninvasive methodology suitable for evaluating ventricular structure

and function has hindered efforts to clarify the natural history and pathogenesis of
cardiovascular performance in cor pulmonale patients. Pulmonary arterial hyper-
tension and right ventricular dysfunction generally are the most significant abnor-
malities in these individuals although, in some cases, left ventricular function is
impaired as well (1-3). This chapter, a compilation of two previous articles (4,5),
reviews radionuclide techniques for assessing right and left ventricular performance
and the application of these techniques in patients with chronic obstructive pulmo-
nary disease and cor pulmonale.
ANATOMIC CONSIDERATIONS
The triangular, crescent-shaped right ventricular chamber is a structure bounded
by the convex interventricular septum and the concave free wall. Its surface-area-
to-volume ratio is quite large. Contraction of this chamber involves three inde-
pendent phenomena. With the first, the trabeculae and papillary muscles push the
tricuspid valve plane down toward the apex. This action shortens the longitudi-
nal axis of the chamber, but has little effect on actual ejection. Second, the free
wall of the chamber moves toward the convex surface of the septum and is
responsible for the major portion of right ventricular contraction. Finally, the
L.J. Rubin (ea.), Pulmonary Heart Disease. All rights reserved. Copyright @ 1984 Marlinus NijhoffPuhlishing. Boslon/The Hague/Dor-
drechl/Lan,asler.
223
224 8. Radionuclide Angiocardiographic Assessment of Ventricular Perfonnance
circular fibers of the thicker left ventricle contract, increasing the curvature of
the interventricular septum. While this activity enhances the bellows action of
the free wall, its relative significance in overall right ventricular performance
remains unknown (6).
Because of its configuration, the right ventricle is ideal for ejecting relatively
large amounts of blood with minimal myocardial shortening (4). However, this
chamber is not well suited to the high intracavitary pressures that develop with
increases in pulmonary arterial pressure and pulmonary vascular resistance. The
tension of the right ventricular myocardium would have to be greater than that
of the left ventricle under similar conditions. Normally, the pulmonary vasculature
provides relatively moderate resistance to flow from the right ventricle. However,
a sustained increase in pulmonary arterial pressure culminating in pulmonary
arterial hypertension inevitably leads to cor pulmonale and right heart failure (4).
The surface of the left ventricle is small relative to its intracavitary volume. A
reduction in the diameter of the main axis, generally resulting from contraction
involving the circumferential muscle bundles, is responsible for approximately
85% of the volunie changes. The left ventricle is an excellent pressure pump,
capable of ejecting against the high-resistance systemic circuit (6).
CONTRAST ANGIOGRAPHIC TECHNIQUES

No single technique is ideal for contrast angiographic assessment of right ven-
tricular volumes. Angiographically, the right ventricle has been defmed as a
prism, a pyramid, parallelopiped, and an ellipse (7-9). Volume assessment based
on area length measurements and Simpson's rule have correlated relatively well
with information drawn from postmortem casts; however, a wide variation exists
in the reported normal ranges. Most validation studies have been based on hearts
free of cardiac disease, an issue of particular importance when the heart is dilated.
In certain pathological states the trabeculae are of relatively less importance, and
the crescent-shaped right ventricle expands to a more ovoid cross section while
the ventricular silhouette undergoes little change (10). Contrast angiographic
studies present other technical problems as well. Right ventricular injections tend
to induce premature ventricular contractions. The extensive trabeculations of the
chamber create problems in reproducibly defming its silhouette; additionally, the
shape of the right ventricle is extremely irregular and variable among patients
(figure 8-1) (11).
Radionuclide techniques
Radionuclide techniques for evaluation of cardiovascular function fall into two
broad categories: (a) analysis of the first transit through the central circulation
(first-pass technique) or (b) analysis of the total equilibrium blood pool subsequent
to intravascular labeling (equilibrium gated technique). Technical details for these
methods may be found elsewhere (12-21).
225
ANTERIOR
Figure 8-1. Traced outlines of contrast angiographic studies in six normal patients without
evidence of cardiovascular disease. The shaded areas represent the right ventricle (RV). The left
ventricle (LV) is also indicated. These diagrams were obtained directly from contrast
angiographic studies in the anterior projection. Note the wide variation in the shape and
configuration of the normal right ventricle. (Modified, with permission, from Dotter and
Steinberg [11].)
First-pass technique
Quantitative analysis in first-pass radionuclide angiocardiography is based on the
principles of the indicator-dilution technique (12), which assume a homogeneous
mixing of the radioactive tracer with blood. To ensure adequate mixing a compact
bolus injection is essential. When the mixture is uniform alterations in externally
detected radioactive counts are proportional to changes in chamber volume. Be-
cause radioactivity is segregated both temporally and anatomically within each
chamber, both right and left ventricular performance can be evaluated from the
same study. Left ventricular ejection fraction, regional wall motion and end-
diastolic volume, right ventricular ejection fraction, and other indexes of systolic
and diastolic ventricular function can be determined by assessment of the high-
frequency components of the time-activity curve (12-17). A separate injection of
a technetium-99-m labeled radio tracer is necessary for each first-pass study. The
major limitation of this technique is the relatively low count rates of the raw data
since analysis is based on several cardiac cycles. This factor is particularly significant
when conventional single-crystal scintillation cameras are used. However, the
226 8. Radionuclide Angiocardiographic Assessment of Ventricular Performance
computerized multicrystal camera, composed of 294 individual sodium iodide

crystals, permits accumulation of count rates up to approximately 450,000 counts/-
sec and is ideal for dynamic first-pass studies. Unfortunately, this system is not
portable, thus precluding bedside studies. The first-pass technique has been modified
to allow data acquisition in conjunction with the patient's electrocardiogram. This
adaptation, the gated first-pass technique, is based partially on the principles of
gated equilibrium blood pool imaging and is more suitable to the single-crystal
camera (18). For first-pass studies the patient may be upright or supine and in any
position relative to the detector. This variability is possible because of the temporal
sequence of radioactivity within individual cardiac chambers and is a potential
advantage in analysis of regional wall motion. For single-position studies we use
the anterior position, which places the heart close to the detector to maximize
count-rate detection. Further, this study is readily performed during upright bicy-
cle exercise, permits evaluation of right and left ventricular function from a single
injection, and allows assessment of regional wall motion. To determine right
ventricular ejection fraction, the 30° right anterior oblique position is used often
since it provides the best separation between the right atrium and the right ventri-
cle. Once a suitable region of interest has been identified in the left or right
ventricle, radioactivity is assessed only when it is in that chamber. This allows
exclusion of activity from overlapping structures. After correcting for background
noncardiac activity, the computer generates a time-activity curve, similar to a
ventricular volume curve (figure 8-2).
A major recent advance has been the development of short half-life tracers
suitable for first-pass studies. Although several have been described (21), the initial
results with gold-195m have been the most encouraging. This radionuclide has a
half-life of 30.5 sec and can be eluted easily from a self-shielded desk-top generator,
thus allowing multiple sequential first-pass studies with extremely favorable radia-
tion dosimetry.
Equilibrium gated technique

In gated cardiac blood pool imaging, the total equilibrium blood pool is imaged
several times during cardiac contraction. The process is done by synchronizing the
collection of scintillation data with a marker of cardiac contraction (18). Elec-
trocardiographic events generally have an established relationship to the mechanical
activity of the heart. Thus, repetitive sampling of certain phases in the cardiac cycle
from each of many beats can be performed until the image has sufficient count
density. The entire intravascular blood pool is labeled, usually with technetium-
99m in vivo labeled autologous red blood cells. Development of time-activity
curves requires assessment of the 45° left anterior oblique position study since this
is the only view that separates activity within the two ventricles. In this position
the left atrium contributes relatively little to left ventricular counts due to its
distance from the detector and its location superior and posterior to the left
ventricle. This technique is used frequently for evaluation of global and regional
left ventricular function, including end-diastolic volume. The left anterior oblique
227
TIME - ACTIVITY CURVES

~ 2800r--'---v--,,--~~r-~---r--'---r-~
Cf)
W
c:::
W
~ RV
Z
oIL. 0 OL.-----"_ _ _ _ __
Z
o
~ 2000 r--,----,,------,.-----.,,------..-,r-__.---r----.------,...-----,
c:::
Z
Cf)
LV
~
Z
~
o
u
o
t-----i
TIME (SECONDS) 0.5
Figure 8-2. First-pass radionuclide angiocardiography. Illustrated are high count rate
time-activity curves generated from the right and left ventricular regions of interest after
peripheral venous injection of technetium-99-m radiotracers. Peaks in the curves represent
end-diastolic counts, and valleys represent end-systolic counts. The maximal count rate in each
curve corresponds to 100%. Right and left ventricular ejection fractions are calculated from
counts at end-diastole and end-systole. (RV, right ventricle; LV, left ventricle.) (Reproduced
with permission from Berger et al. [12].)
equilibrium blood pool has been employed recently to determine right ventricular
ejection fraction since the right atrium contributes substantially to right ventricular
counts in this position (19,20) (figure 8-3). Overlap of the right atrium and right
ventricle occurs throughout the cardiac cycle, even with a variable region of
interest (11,18) (figure 8--4). These factors should be considered when assessing
results obtained with this method.
Thallium-201 imaging
Another nuclear approach useful in chronic obstructive pulmonary disease
(COPD) is thallium-201 myocardial imaging. Regional myocardial blood flow
and myocardial mass determine distribution of this tracer. Due to the greater mass
of the left ventricle, it is visualized clearly at rest while normally the right ventricle
is not seen. The increased right ventricular blood flow during exercise permits
demonstration of this chamber. Visualization of the resting right ventricle was first
described in patients with pulmonary arterial hypertension by Cohen et al. (22).
Additional fmdings included thickening of the right ventricular free wall in the
most severe cases of pulmonary arterial hypertension and right ventricular hyper-
trophy (figure 8-5).
228 8. Radionuclide Angiocardiographic Assessment of Ventricular Perfonnance
ANT
LAO
ED ES
Figure 8-3. Gated equilibrium cardiac blood pool imaging in a patient with cor pulmonale.
Images are shown in the anterior (ANT) and the left anterior oblique (LAO) positions at
end-diastole (ED) and end-systole (ES). Note the marked right ventricular dilatation and diffuse
dysfunction. The left ventricle is nonnal.
More recent studies have assessed thallium-201 imaging at rest in individuals

with right ventricular volume and pressure overload. Qualitative analysis demon-
strated a correlation between marked right ventricular visualization and the highest
right ventricular systolic pressure and pulmonary vascular resistance (23). These
findings indicate a relationship between the degree of right ventricular visualization
and the magnitude of right ventricular hypertrophy. By demonstrating that myo-
cardial thallium-201 uptake directly reflects myocardial mass in chronically
hypoxic rats with right ventricular hypertrophy, Rabinovitch et al. (24) showed
the usefulness of this approach for quantifying and grading the degree of right
ventricular hypertrophy.
229
c
Figure S-4. Traced outlines of contrast angiocardiogram obtained in the left anterior oblique
projection. The outlines were traced directly from frame-by-frame contrast angiograms. The
right ventricle is shown as the shaded area.
A. Frame obtained 1.5 seconds after contrast injection, with minimal activity is noted in the
right ventricle. The right atrium is predominantly opacified.
B. Frame obtained 2 seconds after injection, with the right ventricle, pulmonaty conus, and
pulmonary arteries opacified. Note the substantial overlap between right ventricle and right
atrium in the left anterior oblique projection.
C. Frame obtained 2.5 seconds after injection, with the outflow tract of the right ventricle
superimposed on the right atrial appendage. However, there still is superimposition of the right
atrium and right ventricle in this projection. This first-pass contrast angiographic study
highlights the potential problems encountered in equilibrium cardia! pool imaging. (Modified,
with permission, from Dotter and Steinberg [11] .)
APPLICATION OF RADIONUCLIDE ANGIOCARDIOGRAPHY IN CHRONIC

OBSTRUCTIVE PULMONARY DISEASE
Right ventricular ejection fraction at rest

Right ventricular dysfunction is a principal hemodynamic abnormality in cor
pulmonale (2,3). First-pass radionuclide angiocardiography has been used to assess
right ventricular ejection fraction in COPD patients (15,16,25-27). In 36 patients
with chronic obstructive pulmonary disease studied in our laboratory (16), right
ventricular ejection fraction ranged from 19% to 71% (figure 8-6); 19 of these
individuals had a low (less than 45%) right ventricular ejection fraction. In 10
patients who had clinical evidence of cor pulmonale, right ventricular ejection
fraction averaged 35 + 2% (mean+standard error of the mean); this was signifi-
cantly lower than that of the other patients. All cor pulmonale patients demon-
strated a depressed right ventricular ejection fraction. However, 9 of 26 patients
with severe ventilatory impairment but no clinical evidence of cor pulmonale also
had a depressed right ventricular ejection fraction. Acute respiratory decompensa-
tion and cor pulmonale subsequently developed in four of these nine individuals.
In contrast, cor pulmonale did not develop within two years of the original study
in any of the patients with a normal right ventricular ejection fraction.
ANT
LAO
LLAT
Figure 8-5. Resting thallium-201 myocardial perfusion images in a patient with chronic
obstructive pulmonary disease. Images are shown in the anterior (ANT), 45° left anterior oblique
(LAO), and left lateral (LLAT) positions. Note the marked right ventricular thallium-201
uptake seen in all three positions. Using quantitative computerized analysis, the right ventricular
background ratio for thallium-201 was 1.6 to 1. The patient had an abnormal right ventricular
ejection fraction (37%). (Reproduced with permission from Berger and Matthay [4].)
Right ventricular dysfunction was related to the severity of ventilatory impair-

ment and arterial hypoxemia (16). Of19 patients with a forced expiratory volume
of less than 1 liter in one second (FEV 1), 14 demonstrated a low right ventricular
ejection fraction. In all 19 individuals the ejection fraction averaged 42 3%, +
significantly less than that in the 17 patients with FEV1 of 1 liter or more (52
+ 2%). On the other hand, FEV 1 was 37 +
of the expected value in patients
231
I I
•
..... 70 r- • -
~
••• ••
z
Q • ••
u 60
~ l- •• -
ct •
~
""1
0:
~ Mean±SO ••
z •• •
0 50 r- •• • -
~ •• ••
u •
..,
w •
w • ••
• ••
0: 40 r- •• -
ct •
...J ••
;:) ••
u ••
0: ••
~
z 30 r- -
w
>
~ N = 50 N = 36
:I:
<!)
0: 20 r- • -
I 1
NORMAL COPO
Figure 8-6. Right ventricular ejection fraction in 50 normal control subjects without evidence
of cardiopulmonary disease and in 36 patients with chronic obstructive pulmonary disease
(COPD). The mean±standard deviation (SD) is shown for the normal group. Using 2±
standard deviations, the lower limit of normal is 45%. Note the wide variation in ejection
fraction values for patients with chronic obstructive pulmonary disease. Nineteen of these
patients demonstrated abnormal right ventricular performance at rest. (Reproduced from Berger
et al. [16])
with an abnormal right ventricular ejection fraction, compared with 54 +

4% of
the predicted value in the other subjects. Similar significant differences were appar-
ent for forced vital capacity and arterial oxygen tension. Slutsky et al. (15) recently
confirmed the association between right ventricular ejection fraction at rest and the
severity of ventilatory impairment. Neither study detected a meaningful correla-
tion between right and left ventricular ejection fraction (15,16).
In a study of 20 patients, Brent et al. (27) evaluated the pathophysiologic
correlates of right ventricular ejection fraction. They found a strong inverse
linear correlation between right ventricular ejection fraction and right ventricular
afterload as demonstrated by peak or mean pulmonary artery pressure (r = -
0.81) and pulmonary vascular resistance index (r = -0.73) (figure 8-7). In
addition, there were inverse correlations between right ventricular ejection fraction
and parameters reflecting preload, either right ventricular end-diastolic volume
index (r = -0.56), or mean right atrial pressure (r = -0.51) (figure 8-8).
Cardiac index, the ratio of peak pulmonary arterial pressure to right ventricular
end-systolic volume index, arterial oxygen tension, and left ventricular ejection
fraction had no relationship to right ventricular ejection fraction. These results
indicate that aside from intrinsic contractile influences, right ventricular ejection
fraction in COPD patients is largely dependent upon afterload, but less so upon
preload.
Right ventricular performance during exercise

Frequently, abnormalities in right ventricular performance become apparent only
during exercise. Cardiac catheterization techniques have demonstrated elevated
pulmonary arterial, pulmonary capillary wedge, and right ventricular end-diastolic
pressures during exercise stress (28,29). Therefore, first-pass radionuclide angiocar-
diography was used to study 30 patients with chronic bronchitis and emphysema
at rest and during upright bicycle exercise (30). The exercise protocol differed from
that normally employed in evaluating patients with coronary or valvular heart
disease, since different physiologic factors in each disorder limit exercise (that is,
ventilatory impairment in chronic obstructive pulmonary disease and myocardial
ischemia in coronary artery disease). To assure relatively steady state aerobic
conditions, a single-stage submaximal exercise test with on-line monitoring of
minute ventilation and respiratory exchange ratio was used. These precautions were
taken to avoid the potentially harmful and independent effects on myocardial
contractility of uncompensated lactic acidosis caused by anaerobic metabolism.
Also, most COPD patients generally limit themselves to workloads below the
anaerobic threshold.
Based on data from 25 controls, normal exercise ventricular reserve was deter-
mined to be an absolute increase in ejection fraction of 5% or more, regardless of
the exercise protocol or value of ejection fraction at rest (30,31) (figure 8-9). Of
30 COPD patients 23 demonstrated an abnormal right ventricular response to
submaximal exercise (figures 8-10 and 8-11). Although airways obstruction and
arterial hypoxemia were substantially greater in patients with abnormal right
ventricular exercise reserve than in those with normal reserve, abnormal right
ventricular responses were apparent even in some individuals with mild airways
obstruction. Olvey et al. (26) have reported confirmatory data.
These fmdings, coupled with the known hemodynamic responses, indicate that
altered right ventricular afterload is the major factor governing the right ven-
tricular response to exercise. Mahler et al. (32) employed pulmonary artery ca-
theterization in subjects with mild to moderate chronic obstructive pulmonary
disease and showed an abnormal rise in pulmonary artery pressure during exer-
cise. This increase in right ventricular afterload was linked to an abnormal right
ventricular ejection fraction response (abnormal right ventricular reserve) during
exercise. However, it should be noted that abnormal right ventricular exercise
233
A. B.
r = -O.BI
n= 20 55
r=-0.73
55
• • n= 20
• • •
~
.. •• • • •
45
45
• \.
lL.
UJ
, • • •• •
>
a:
35 •
• •• •
35 • ••••
• • ••
25 25
~:::
0 20 40 60 80 0 200 400 600 800 1000

PPASP (mmHg) PVRI (dynes· sec ·cm 5 . m 2)
Figure 8-7. A. Relation of right ventricular ejection fraction (RVEF) and indices of right
ventricular afterIoad as measured by peak pulmonary arterial systolic pressure (PPAP).
B. Pulmonary vascular resistance index (PVRI). Note the good inverse relation between right
ventricular ejection fraction and right ventricular afterload. (N = number of patients; r =
correlation.) (Reproduced with permission from Brent et al. [27].)
A. B.
r = -0.56 r=-0.51
n=20 n= 20
•
•
• • •
• • • • • •
• •• • •
• • ••
40 60 80 100 120 140 160 o 4 8 12 16

RVEDVI (ml/m2) RAP (mm HQ)
Figure 8-8. The relation of right ventricular preload as measured by ventricular end-diastolic
volume index (RVEDVI) (panel A) and mean right atrial pressure (RAP) (panel B). Note the
good inverse relation between right ventricular ejection fraction and right ventricular
end-diastolic volume index and right atrial pressure. (Reproduced with permission from Brent et
al. [27].)
NORMAL CONTROL SUBJECTS

RV EJECTION FRACTION LV EJECTION FRACTION
• MAXIMAL (N=14:
% %
o SUBMAXIMAL (N=lil
90 90
80 80
70 70
60 60
50 50
p<O.OOI p< 0.001
REST EXERCISE REST EXERCISE
Figure 8-9. Right ventricular (RV) and left ventricular (LV) ejection fraction at rest and
exercise in normal control subjects. Fourteen patients exercised using a single-stage submaximal
protocol. Both right and left ventricular ejection fractions increased by 5% in each patient.
irrespective of the exercise protocol. On the basis of these data. normal exercise ventricular
reserve is defined as an absolute increment in ejection fraction of at least 5%. (p=probability.)
(Reproduced with permission from Matthay et aI. [30].)
reserve may be a normal physiologic response to augmented afterload rather than

evidence of intrinsic right ventricular myocardial dysfunction.
Thallium-201 imaging
Recently, 71 COPD patients were evaluated to determine the significance of right
ventricular visualization in thallium-201 images (33). Computerized thallium-201
images were compared with resting right ventricular ejection fraction data deter-
mined by first-pass radionuclide angiocardiography, electrocardiography, and pul-
monary status. The right ventricle was demonstrated in 52 of these patients. Mild
uptake was defined as a right-ventricular-to-background ratio of less than 1.5. Of
11 patients with electrocardiographic evidence of right ventricular hypertrophy,
9 had significant right ventricular uptake and 2 had mild uptake. The former group
had significantly greater right ventricular dysfunction, arterial hypoxemia, and
airways obstruction compared with patients with normal thallium-201 images. Of
23 patients with substantial right ventricular uptake, 20 had abnormal right ven-
tricular ejection fraction while only 9 demonstrated electrocardiographic evidence
235

%
%
N= 30 • •
80 80
70 70
60 60
50 50
40 40
30 <>30
p = NS p<O.OI
REST EXERCISE REST EXERCISE
Figure 8-10. Right ventricular (RV) and left ventricular (LV) ejection fractions at rest and
submaximal exercise in 30 patients with chronic obstructive pulmonary disease. Data in
individual patients are shown as closed circles connected hy solid lines. The mean values are shown
at the sides of each panel. For the overall group right ventricular ejection fraction was
unchanged with exercise, whereas left ventricular ejection fraction increased normally.
(Reproduced with permission from Matthay et al. [30].)
of right ventricular hypertrophy. In contrast, of 19 patients with normal thallium-

201 images, only 3 had abnormal right ventricular ejection fraction; none had
electrocardiographic evidence of right ventricular hypertrophy.
These findings indicate that marked right ventricular uptake in thallium-201
images occurs frequently in chronic obstructive pulmonary disease and suggest that
right ventricular dysfunction and hypertrophy may be present more often than
suggested by electrocardiogram. Normal thallium-201 images in COPD patients
generally are associated with preserved right ventricular function. Further studies
are underway to determine (a) whether acute interventions that change right
ventricular afterload will affect right ventricular thallium-201 visualization and (b)
the relationship between pulmonary arterial pressure and the magnitude of right
ventricular uptake (figure 8-5).
Left ventricular performance

Leflllentricular ejection fraction at rest
The effects of chronic obstructive pulmonary disease on left ventricular function
are more complicated and controversial (see chapter 9). Left ventricular hypertro-
phy at postmortem examination has been noted in several reports (34,35). How-
236 8. Radionuclide Angiocardiographic Assessment of Ventriculat Performance
0.. 100
~
0 N=30
a:
(.!)
:L: 80
u
«
w
z
60
....
C/)
Z
W
....
« 40
0..
LL.
0
....
z 20
w
u
a:
®
w
0.. 0
RV LV RV LV
ABNORMAL RESTING ABNORMAL EXERCISE
EJECTION FRACTION RESERVE
Figure 8-11. Incidence of ventriculat performance abnormalities at rest or during exercise in 30

patients with chronic obstructive pulmonary disease. The shaded bars indicate right ventriculat
(RV) performance and the open bars left ventriculat (LV) performance. Note that the
predominant hemodynamic abnormality involves right ventriculat function, especially during
exercise. Abnormal left ventriculat performance was relatively uncommon. (Reproduced with
permission from Berger and Matthay [4].)
ever, the majority of studies have shown that left ventricular performance is
unaltered in patients with chronic obstructive pulmonary disease as long as no other
associated cardiac abnormalities are present (16,36--42). First-pass radionuclide
angiocardiography studies in our laboratory (16,30) and others (15,38) have estab-
lished that resting left ventricular ejection fraction is usually normal (figures 8-10
and 8-11). In each study, however, specific patients had abnormal left ventricular
function despite the absence of concomitant systemic hypertension or coronary
artery disease. Furthermore, abnormalities in early systolic left ventricular function
have been described in some individuals with severe chronic obstructive pulmonary
disease without underlying left-sided heart disease (15). The potential mechanisms
responsible for this observation are discussed in chapter 9.
Left ventricular ejection fraction during exercise

Left ventricular exercise reserve during single-stage submaximal exercise was evalu-
ated using techniques described earlier for right ventricular performance. Findings
were abnormal in only 6 of 30 patients studied, 4 of whom demonstrated normal
237
function at rest (30). For the group as a whole, left ventricular ejection fraction
increased normally (figures 8-10 and 8-11). No correlation was found between
abnormalities in left ventricular exercise reserve and pulmonary function or arterial
blood gases.
Effect oj increased preload and after/oad

In COPD patients investigators have found normal left ventricular responses to the
stress of augmented preload with infusion of dextran (39) and augmented afterload
with methoxamine-induced vasoconstriction (40). In these cases, left ventricular
ejection fraction did not diminish even in patients with cor pulmonale.
Biventricu1ar performance in acute respiratory failure
Left ventricular function

Relatively few radionuclide studies have been reported on ventricular function
during acute respiratory failure (38,41,42). Steele et al. (38) measured left ventricu-
lar ejection fraction using first-pass radionuclide angiography in 92 patients with
acutely decompensated chronic obstructive pulmonary disease (figure 8-12). Fewer
than 50% of those studied demonstrated abnormal resting left ventricular ejection
fraction. Thirteen of the 26 had concomitant systemic hypertension or coronary
artery disease. Thus, only 13 of 92 patients exhibited abnormal left ventricular
performance without concomitant left-sided heart disease. Additional studies in the
few patients with left ventricular dysfunction should clarify the prognostic signifi-
cance of these results.
Biventricular performance
In describing the qualitative characteristics of a radiographic examination, Ellis and
Steele (43,44) determined left ventricular ejection fraction, pulmonary blood vol-
ume, pulmonary transit time, and cardiac index in 125 patients with acute respira-
tory failure. A significant drop in pulmonary blood volume associated with a
shorter pulmonary transit time generally occurred in patients with major pulmo-
nary embolism; left ventricular ejection fraction usually remained normal. A total
or partial loss of the distinct right and left ventricular peaks associated with a
normal pulmonary blood volume was characteristic of right ventricular dysfunc-
tion. Mean pulmonary artery pressure was substantially higher, and marked right
ventricular dilatation with minimal hypertrophy was apparent at autopsy. Left
ventricular ejection fraction was normal. Abnormalities in left ventricular perform-
ance were significant in 25% of these patients. Acute ventilatory failure was linked
with concomitant myocardial infarction in most instances. These fmdings illustrate
the potential value of using radionuclide studies of biventricular function as a
complement to conventional hemodynamic monitoring in patients with acute
respiratory failure.
100
90 00
0
0
0
0
80 00000o
00
~
c:0 0
!00
:;::; 70 0 000
(,) 0
~ 0 0
0
u. 0
000
c:: 0 0
8
0
:;::; 60 0
0
00000o
000
ooooof.-
U 00
III 0
00
W •
...
.!!! SO
oooe
oe
8
0000
0
-
::l 00
U
0':::
oe
0 00
c::
- or•
00
~ ..0 oe
III
...J as 0
30
0
•
20 ODDO . . . . . . .
Stable Acute
Figure 8-12. Left ventricular ejection fraction at rest in 28 patients with stable chronic
obstructive pulmonary disease and in 92 patients with acutely decompensated cor pulmonale.
Data in patients with definite concomitant left-sided heart disease are shown in closed circles,
data in patients with no evidence of left-sided heart disease are shown in open circles. Note that
most patients with an abnormal left ventricular ejection fraction have concomitant docwnented
left-sided heart disease (in most cases hypertensive cardiovascular disease or myocardial
infarction). (Reproduced with permission from Steele et al. [38].)
EVALUATION OF THERAPEUTIC INTERVENTIONS

The underlying cause of respiratory insufficiency should be the target of major
therapy in cases of acute and chronic respiratory failure. Reversing hypoxia and
acidosis, thereby relieving alveolar hypoventilation, may diminish pulmonary
arterial hypertension. Combating the blood gas abnormalities associated with respi-
ratory insufficiency frequently produces improvement in right-sided heart failure.
The prognosis for individuals with long-standing pulmonary arterial hypertension
is poor; therefore, efforts to improve the hemodynamic status of such high-risk
patients are recommended.
239
Response to digitalis
Administration of digitalis and its derivatives to cor pulmonale patients is a
controversial issue (see chapter 12). Green and Smith (45) reviewed the literature
and found no defInite evidence that cardiac glycosides are useful except in cases
of concomitant left ventricular failure. Radionuclide techniques may provide the
solution to the problem, and preliminary data are available from two studies. Ellis
et al. (25) reported no improvement in radionuclide-assessed right ventricular
ejection fraction after the intravenous administration of 0.01 mg/kg of ouabain in
11 patients with severe chronic obstructive pulmonary disease and cor pulmonale.
However, their study evaluated only the acute effects of this digitalis glycoside.
Recently, Mathur et al. (46) used the equilibrium blood pool technique to evaluate
left and right ventricular ejection fractions in 15 cor pulmonale patients. Both
values improved signifIcantly after eight weeks of digitalization. However, four
individuals who responded with substantially augmented left ventricular ejection
fraction (from 43% to 56%) and right ventricular ejection fraction (from 30%
to 37%) appeared to account for the overall improvement observed. The other 11
patients exhibited no change in either left or right ventricular ejection fraction.
Baseline left ventricular ejection fraction was considerably lower in patients who
responded than in those who did not. These initial fIndings seem to support the
conclusion that digitalis improves right ventricular function in obstructive airways
disease only in the presence of simultaneous left ventricular dysfunction (45).
Additional studies are needed to confIrm these results.
Response to theophylline
First-pass radionuclide studies have been performed to evaluate the effects on
biventricular function of intravenously administered aminophylline, a potent bron-
chodilator (47). Data on arterial blood gases, ventilatory function, and theophylline
blood levels, as well as left and right ventricular ejection fractions, were obtained.
Studies were performed in the control state and after a 30-minute intravenous
infusion of 9 mg/kg aminophylline in 15 patients with chronic bronchitis and
emphysema, including 4 individuals with cor pulmonale. In 8 of the 15 patients,
right ventricular ejection fraction at rest was normal. Aminophylline markedly
improved the right ventricular ejection fraction from 45 ± 4% to 52 + 4%
(fIgure 8-13). Aminophylline normalized right ventricular ejection fraction in six
of the eight individuals with basal right ventricular dysfunction. Concomitant
increases in left ventricular performance were noted in 11 of 15 patients. These
alterations in biventricular function were noted despite only minimal improvement
in FEV 1 and forced vital capacity. Additionally, while arterial carbon dioxide
tension dropped, presumably due to a central mechanism involving hyperventila-
tion, arterial oxygen tension underwent little change.
The effect of ventilatory status alterations on improvement in cardiac function
was evaluated further in a comparable study performed in fIve normal patients who
had no evidence of cardiopulmonary disease (47). Aminophylline infusion in these
80 80
70
60
./
e
er.---+-
70
60
e
50 e
e---;-'--"o..c;..., 50
40 40
30 30
p< 0.001
p< 0.001
20 L - - - _ L - -_ _ _-'-----------' 20 '-----'--------'-----'
CONTROL AMINO- CONTROL AMINO-
PHYLLINE PHYLLINE
Figure 8-13. Effect of aminophylline upon right and left ventricular performance in 15 patients
with chronic obstructive pulmonary disease. Data obtained during the control state and at the
conclusion of the aminophylline infusion are depicted by closed circles connected by solid lines.
The means are represented by open circles at the sides of each panel. Note that right ventricular
and left ventricular ejection fractions increased significantly with aminophylline. (Reproduced
with permission from Matthay et al. [47].)
control subjects produced similar increases in right and left ventricular ejection
fractions, while ventilatory performance and arterial oxygen tension remained
unaltered. A direct inotropic effect or a secondary improvement in ventricular
function as a result of reduced pulmonary and systemic vascular resistances is the
most likely explanation for improvement in left and right ventricular ejection
fractions. Moderate enhancement of biventricular performance was maintained
during long-term therapy up to four months in a study using oral sustained-release
theophylline (48).
Response to vasodilators and bet~ agonists

Several theoretical advantages of vasodilators and beta-adrenergic stimulants should
result in improved cardiovascular performance in patients with COPD and pulmo-
241
nary hypertension. These agents may decrease right ventricular afterload through
(a) direct dilation of the pulmonary vasculature, (b) recruitment of underperfused
pulmonary vessels, or (c) reduction of alveolar pressure.
First-pass radionuclide angiocardiography and pulmonary artery catheteriza-
tion with a thermodilution catheter have been useful in assessing the acute
effects of these agents. The vasodilators nitroglycerin, nitroprusside, and hy-
dralazine (49,50) and the betaz-adrenergic agent terbutaline (51) were adminis-
tered to patients with chronic bronchitis and emphysema and right ventricular
dysfunction. The venodilators, nitroglycerin, and nitroprusside reduced cardiac
output and right ventricular end-diastolic volume indices, but did not alter
right ventricular afterload (49,50). Nitroglycerin also resulted in diminished ar-
terial oxygen tension and systemic oxygen transport values (50). The arteriolar
dilator hydralazine produced a reduction in pulmonary vascular resistance and
increases in right ventricular ejection fraction, cardiac index, and systemic oxy-
gen transport (49).
In one study subcutaneous terbutaline, considered a selective betaz agonist,
increased biventricular ejection fractions, lowered pulmonary and systemic vascular
resistances, and enhanced systemic oxygen transport (51) (figures 8-14 and 8-15).
In another study Teule and Majid (52) administered terbutaline intravenously in
patients with severe COPo. They found a significant reduction in mean pulmonary
artery pressure and pulmonary vascular resistance accompanied by enhanced cardiac
output and resting peak expiratory flow rate. These two trials (51,52) suggest that
a decreased biventricular afterload, rather than a positive inotropic effect on the
myocardium, is primarily responsible for the beneficial effects of terbutaline on
systolic ventricular function. Thus, the principal value of terbutaline is its ability
to lower right and left ventricular afterload, thereby improving cardiac perform-
ance and oxygen delivery to the tissues, in addition to its role in augmenting
airways flow in COPO patients. Further studies should determine the influence of
oral and inhaled terbutaline and other beta-adrenergic agents on the cardiovascular
system of patients with obstructive airways disorders.
Response to low-flow oxygen

Continuous low-flow oxygen therapy has been useful in treating chronic obstruc-
tive pulmonary disease (53-56). Petty and Finigan (56) reported overall clinical
improvement with this therapy. Ellis et al. (25) found a substantial reduction in
pulmonary arterial pressure and pulmonary vascular resistance in six patients with
cor pulmonale during six months of continuous low-flow oxygen therapy. Further,
they reported a simultaneous increase in right ventricular ejection fraction assessed
by first-pass radionuclide angiocardiography (39% to 56%). By augmenting ox-
ygenation and lowering pulmonary vascular resistance, right ventricular ejection
fraction increased due to reduced right ventricular afterload.
Olvey et al. (26) presented preliminary data on the effects of acute oxygen
administration in 10 patients with chronic hypoxic lung disease. They found that
right ventricular ejection fraction increased with exercise while they were breath-
ing oxygen. Should further studies establish that these fmdings are valid for
MEAN PULMONARY RIGHT VE!'ITRICULAR PULMONARY VASCULAR
ARTERY PRESSURE END DIASTOLIC VOLUME RESISTANCE INDEX
(mm Hg) INDEX ( ml/m2) (dynes· sec 'cm -5 ·m 2 )
175
~
40
150 750
-<>-
:::=:=--=----
125
30
00- 00- 00-
100 -<>- 500
20
~ 75
00-
------- ~
50 250
10
p= NS 25 poNS
C T C T C T
Figure 8-14. Acute effects of terbutaline on right ventricular preload and afterload. Mean
pulmonary arterial pressure and right ventricular end-diastolic volume index are unaffected by
terbutaline. In contrast, pulmonary vascular resistance index is decreased significantly. (C =
control state; NS = not significant; p = probability; T = 30 minutes after terbutaline
administration.) (Reproduced with permission from Brent et al. [51].)

(%) (%)
50
75- ~
50~ -o-~
40
30
25 I-
20
p<.01 p<.005
I I
C T C T
Figure 8-15. Acute effects of subcutaneous terbutaline on right ventricular (RV) and left
ventricular (LV) ejection fraction. Both right and left ventricular ejection fraction increased
significantly with terbutaline. (C=control state; NS=not significant; p=probability; T=30
minutes after terbutaline administration.) (Reproduced with permission from Brent et al. [51].)
243
long-term low-flow oxygen therapy as well, the therapeutic value of this technique
in chronic obstructive pulmonary disease would seem apparent.
CONCLUSION
Radionudide angiocardiography provides the opportunity to noninvasively evalu-
ate cardiac function in the setting of pulmonary heart disease. These techniques may
prove useful in the darification of patterns of cardiovascular dysfunction in chronic
lung disease, in the early diagnosis of pulmonary heart disease, and in the evaluation
of therapy.
REFERENCES
1. Thomas AJ. Chronic pulmonary heart disease. Br Heart J 34: 653-657, 1972.
2. Renzetti AD Jr, McClement]H, Litt BD. The Veterans Administration cooperative study of
pulmonary function. III. Mortality in relation to respiratory function in chronic obstructive lung
disease. Am J Med 41: 115-129, 1966.
3. Weitzenblum E, Loiseau A, Hirth C, et aI. Course of pulmonary hemodynamics in patients with
chronic obstructive pulmonary disease. Chest 76: 656--662, 1979.
4. Berger Hj, Matthay RA. Noninvasive radiographic assessment of cardiovascular function in acute
and chronic respiratory failure. Am J Cardiol 47: 950-962, 1979.
5. Matthay RA, Berger HJ. Cardiovascular function in cor pulmonale. Clin Chest Med 4: 269-295,
1983.
6. Rushrner R. Functional anatomy and control of the heart. In Cardiovascular Dynamics (4th
edition), Rushrner R (ed). Philadelphia: W.B. Saunders, 1976, pp. 89-98.
7. FedinzJ, Godin R, Cohn P, et aI. Right ventricular performance in patients with coronary artery
disease. Circulation 52: 608-615, 1975.
8. Arcilla RA, Tsai P, Thilenius 0, et aI. Angiographic method for volume estimation of right and
left ventricles. Chest 60: 446-454, 1971.
9. Gentzler R, Briselli M, Gault J. Angiographic estimation of right ventricular volume in man.
Circulation 50: 324-330, 1974.
10. Jaffee CC, Ellis K. Angiographic quantitation of ventricular volume, shape, and mass. Curr Prob
Radiol 4: 1-55, 1974.
11. Dotter CT, Steinberg I. The normaI angiocardiograrn. In Angiocardiography, Dotter CT, Stein-
berg I (eds). New York: Harper & Brothers, 1953, p. 50.
12. Berger HJ, Matthay RA, Pytlik L, et aI. First-pass radionuclide assessment of right and left
ventricular performance in patients with cardiac and pulmonary disease. Semin Nucl Med 9:
175-195, 1979.
13. Tobninick E, Schelbert HR, Henning H, et aI. Right ventricular ejection fraction in patients with
acute anterior and inferior myocardiaI infarction assessed by radionuclide angiography. Circula-
tion 57: 1078-1084, 1978.
14. Steele P, Kirch D, LeFree M, Battock D. Measurement of right and left ventricular ejection
fractions by radionuclide angiocardiography in coronary artery disease. Chest 70: 51-56, 1976.
15. Slutsky RA, Ackerman W, Karliner JS, et aI. Right and left ventricular dysfunction in patients
with chronic obstructive lung disease. Assessment by first-pass radionuclide angiography. Am J
Med 68: 197-205, 1980.
16. Berger HJ, Matthay RA, Locke J, et aI. Assessment of cardiac performance with quantitative
radionuclide angiocardiography: Right ventricular ejection fraction with reference to findings
in chronic obstructive pulmonary disease. Am J Cardiol 41: 897-905, 1978•.
17. Van Dyke D, Anger HO, Sullivan RW, et aI. Cardiac evaIuation from radioisotope dynamics.
J Nucl 13: 585-592, 1972.
18. Strauss HW, McKusick KA, Boucher CA, et aI. Of linens and laces. The eighth anniversary of
the gated blood pool scan. Semin Nucl Med 9: 296--309, 1979.
19. Slutsky R, Hooper W, Gerger K, et aI. Assessment of right ventricular function at rest and during
exercise in patients with coronary heart disease: A new approach using equilibrium radionuclide
angiography. Am J Cardiol 45: 63-71, 1980.
20. Maddahi J, Berman DS, Matsuoka DT, et al. A new technique for assessing right ventricular
ejection fraction using rapid multiple gated equilibrium cardiac blood pool scintigraphy. Descrip-
tion, validation and findings in chronic coronary artery disease. Circulation 60: 581-589, 1979.
21. Wackers FJ, Giles RW, Hoffer PB, et al. Gold-195m, a new generator produced short-lived
radionuclide for sequential assessment of ventricular performance in first-pass radionuclide angi-
ocardiography. Am J Cardiol 45: 63, 1980.
22. Cohen HA, Baird MG, Rouleau JR, et al. Thallium-201 myocardial imaging in patients with
pulmonary hypertension. Circulation 54: 790-795, 1976.
23. Kondo M, Kubo A, Yamazaki H, et al. Thallium-201 myocardial imaging for evaluation of right
ventricular overloading. J Nucl Med 19: 1197-1203, 1978.
24. Ravinovitch M, Fisher K, Gamble W, et al. Thallium-201 quantitation of right ventricular
hypertrophy in chronically hypoxic rats. Radiology 130: 223-225, 1979.
25. Ellis JH, Jr, Kirch D, Steele PP. Right ventricular ejection fraction in severe chronic airway
obstruction. Chest 71: 281S-282S, 1977.
26. Olvey AK, Reduto LA, Stevens PM, et al. First pass radionuclide assessment of right and left
ventricular ejection fraction in chronic pulmonary disease. Effect of oxygen upon exercise
response. Chest 78: 4-9, 1980.
27. Brent BN, Berger HJ, Matthay RA, et al. Physiologic correlates of right ventricular ejection
fraction in chronic obstructive pulmonary disease: A combined radionuclide and hemodynamic
study. Am J Cardiol 50: 255-262, 1982.
28. Jezek V, Schrijen F, Sadoul P. Right ventricular function and pulmonary hemodynamics during
exercise in patients with chronic obstructive pulmonary disease. Cardiology 58: 20-31, 1973.
29. Jezek V, Schrijen F. Left ventricular function in chronic obstructive pulmonary disease with and
without cardiac failure. Clin Sci Mol Med 45: 267-279, 1973.
30. Matthay RA, Berger HJ, Davies RA, et al. Right and left ventricular exercise performance in
chronic obstructive pulmonary disease: Radionuclide assessment. Ann Intern Med 93: 234-239,
1980.
31. Berger HJ, Reduto LA, Johnstone DE, et al. Global and regional left ventricular response to
bicycle exercise in coronary artery disease. Assessment by quantitative radionuclide angiocardiog-
raphy. Am J Med 66: 13-21, 1979.
32. Mahler DA, Matthay RA, Berger HJ, et al. Right ventricular exercise performance: A combined
hemodynamic and radionuclide assessment. Am Rev Respir Dis 125: 80, 1982.
33. Berger H, Wackers F, Mahler D, et al. Right ventricular visualization on thallium-201 myocar-
dial images in chronic obstructive pulmonary disease: Relationship to right ventricular function
and hypertropl>y. Circulation 62: 103, 1980.
34. Rao BS, Cohn KE, Eldridge FL, et al. Left ventricular function secondary to chronic pulmonary
disease. JAMA 45: 229-241, 1968.
35. Murphy ML, Adamson J, Hutcheson F. Left ventricular hypertrophy in patients with chronic
bronchitis and emphysema. Ann Intern Med 481: 307-313, 1974.
36. Christianson LC, Shah A, Fisher VJ. Quantitative left ventricular cineangiography in patients
with chronic obstructive pulmonary disease. Am J Med 66: 399-404, 1979.
37. Burrows B, Kettel LJ, Niden AH, et al. Patterns of cardiovascular dysfunction in chronic
obstructive lung disease. N Engl J Med 286: 912-917, 1972.
38. Steele P, Ellis JH, Van Dyke D, et al. Left ventricular ejection fraction in severe chronic
obstructive airways disease. Am J Med 59: 21-28, 1975.
39. Matthay RA, Ellis JH, Steele PP. Effect of dextran loading on left ventricular performance in
chronic obstructive pulmonary disease. Am Heart J 92: 730-736, 1976.
40. Matthay RA, Ellis JH, Steele PP. Methoxamine-induced increase in afterload. Effect on left
ventricular performance in chronic obstructive pulmonary disease. Am Rev Respir Dis 117:
871-877, 1978.
41. Unger K, Shaw D, Karliner JS. Evaluation of left ventricular performance in acutely ill patients
with chronic obstructive lung disease. Chest 68: 135-142, 1975.
42. Kline LE, Crawford MH, MacDonald WJ Jr. Noninvasive assessment of left ventricular perform-
ance in patients with chronic obstructive pulmonary disease. Chest 72: 558-564, 1977.
43. Ellis JH Jr, Steele PP. Value of combined hemodynamic and radiographic studies in acute
respiratory failure. In Nuclear Cardiology: Principles and Methods, Serafini AH, Gibson AJ,
Smoak WM (eds). New York: plenum Medical Books, 1977, p. 187.
245
44. Ellis JH Jr, Steele PP. Quantitative radiocardiography in major pulmonary throboembolism.
Chest 69: 575-581, 1976.
45. Green LH, Smith TW. The use of digitalis in patients with pulmonary disease. Ann Intern Med
87: 459-465, 1977.
46. Mathur PN, Pugsley SO, Powles ACP, et al. Effect of digitalis on right and left ventricular
fnnction in chronic cor pulmonale. Ann Intern Med 95: 282-288, 1981.
47. Matthay RA, Berger RJ, Loke J, et al. Effects of aminophylline upon right and left ventricular
performance in chronic obstructive pulmonary disease: Noninvasive assessment by radionuclide
angiocardiography. Am J Med 65: 903-910, 1978.
48. Matthay RA, Berger HJ, Davies R, et al. Oral theophylline improves biventricular performance
in chronic obstructive pulmonary disease. Am Heart J 104: 1022-1026, 1982.
49. Brent B, Berger RJ, Matthay RA, Pytlik L, Zaret B. Contrasting acute effects of vasodilators
(nitroglycerin, nitroprusside, and hydralazine) on right ventricular performance in patients with
chronic obstructive pulmonary disease and pulmonary hypertension: A combined radionuclide
hemodynamic study. Am J Cardiol 51: 1682-1689, 1983.
50. Matthay RA, Brent BN, Berger HJ, et al. Hemodynamic effects of nitroglycerin in patients with
chronic obstructive lung disease, pulmonary artery hypertension, and right ventricular dysfunc-
tion. Am Rev Respir Dis 123: 78, 1981.
51. Brent BN, Mahler DA, Berger RJ, et al. Augmentation of right ventricular performance in
chronic obstructive pulmonary disease by terbutaline: A combined radionuclide and hemody-
namic study. Am J Cardiol 50: 313-319, 1982.
52. Teule G], Majid PA. Hemodynamic effects of terbutaline in chronic obstructive airways disease.
Thorax 35: 536-542, 1980.
53. Anthonisen NR. Hypoxemia and 02 therapy. Am Rev Respir Dis 126: 729-733, 1982.
54. Flenley DC, Douglas NJ, Lamb D. Nocturnal hypoxemia and long-term domiciliary oxygen in
blue and bloated bronchitics. Chest 77: 305-307, 1980.
55. Nocturnal Oxygen Therapy Trial Group: Continuous or nocturnal oxygen therapy in hypoxemic
chronic obstructive lung disease. Ann Intern Med 93: 391-398, 1980.
56. Petty TL, Finigan MM. The clinical evaluation of prolonged ambulatory oxygen therapy in
patients with chronic airway obstruction. Am J Med 45: 242-252, 1968.
9. THE LEFf VENTRICLE IN CHRONIC LUNG DISEASE
EDGAR J. CALDWELL, MD.
Many symptoms associated with left ventricular dysfunction, such as exertional

dyspnea, orthopnea and cough, may similarly be attributed to lung disease (1,2).
For example, any increase in dyspnea perceived as a significant change by a patient
with chronic lung disease can represent an important and sometimes difficult
challenge to the physician. For such patients, few reliable clinical guides are
available for separating breathlessness due to progression of lung disease from
superimposed left ventricular dysfunction. Physical fmdings, such as crackles and
tachycardia, are common to both entities. The problem is compounded in older
patients, in whom the distinction between cardiac and pulmonary disease becomes
blurred by changes accompanying aging. This issue is underscored by Rapaport:
"Perhaps the most difficult problem in the differential diagnosis of dyspnea arises
when the patient with chronic lung disease has cardiac disease, independently, as
well" (2). The physician must ask the question, Does a change in dyspnea represent
progression of pulmonary disease or is this a new assault on the left ventricle?
In the late eighteenth century, a number of observers noted that patients dying
from emphysema and asthma had cardiac enlargement. Laennec noted that hyper-
trophy and dilatation of the heart and dilatation of the pulmonary artery occurred
This chapter was prepared during a sabbatical leave as a visiting associate professor of medicine, Pulmonary
Division, the University of Texas Health Science Center at Dallas.
L.J. Rubin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright ® 1984 Martinus NijhoffPublishing. Boston/The Hague/Dor-
drecht/Lanlaster.
247
248 9. The Left Ventricle in Chronic Lung Disease
as a sequel to emphysema. Of the four cases he described, one had a normal heart
and the other three had biventricular enlargement (3).
By the early 1940s clinicians and pathologists agreed that the heart was apt to
be enlarged in pulmonary emphysema (4-6). Unfortunately, many autopsy series
focused on ventricular wall thickness as the criteria for hypertrophy (7-12).
Ventricular weight is a more accurate reflection of hypertrophy when compared
to wall thickness because of the variability in wall thickness and the nonuniformity
of measurement site selection by investigators (13-17). These considerations were
ignored in most studies evaluating the incidence of left ventricular hypertrophy
in chronic obstructive pulmonary disease (COPD) throughout the 1960s and
resulted in studies reporting that left ventricular hypertrophy occurred in 24% to
97% of patients with COPD (10-12,18). The cause ofleft ventricular hypertrophy
was not readily apparent, but most authors speculated that hypoxemia, ventricular
interdependence, or a combination of both factors might be responsible.
In this chapter we will examine the effect of chronic lung disease on the left
ventricle, the relationships between the right and left ventricles in chomic lung
disease and potential mechanisms that may be involved in the development of left
ventricular dysfunction in the setting of chronic respiratory disease.
ANATOMIC EVIDENCE FOR LEFT VENTRICULAR CHANGES IN CHRONIC

LUNG DISEASE
Ventricular anatomy
A brief overview of gross ventricular anatomy is in order to appreciate the
interaction between the right and left ventricles in normal as well as disease states.
To state that the left ventricle is a complex muscular organ is a gross underesti-
mation. Interest in its anatomy has a long and noble history, with many opinions
concerning the appropriate geometry. Studies of left ventricular architecture ex-
tend back over 300 years (19,20).
The ventricles are the major source of energy to propel blood through the
pulmonary and systemic vasculature, albeit at different pressure loads. The right
and left ventricles act as two pumps side by side, each possessing a unique structure.
Flow of blood through the right ventricle is unidirectional; blood enters at one
orifice and leaves by another, with inflow and outflow tracts forming an angle of
90° with each other. In contrast, left ventricular flow is essentially bidirectional:
blood enters and leaves from the same orifice separated by a thin membrane, the
anterior leaflet of the mitral valve situated between the mitral and aortic orifices
of the left ventricle (21). Ventricular flow differences are shown in figure 9-1. The
ventricles are composed of inlet, trabecular, and outlet components. An internal
view indicates the ventricular inlet extending from the atrioventricular annulus to
the origins of the papillary muscles. Beyond this point the apical parts of the
ventricles represent trabecular components, coarse trabeculations in the right ventri-
cle, and a fmely trabeculated left ventricle. The outlet component supports the
arterial valve. In the left ventricle the outlet component is contiguous with the inlet
249
RIGHT VENTRICLE LEFT VENTRICLE
Figure 9-1. The directions of the inflow and outflow tracts of the right and left ventricles
from the frontal view. (From reference 21, with permission.)
component-aortic-mitral continuity-but in the right ventricle the outlet com-

ponent is a complete muscular structure, the infundibulum. A longitudinal section
through the ventricular walls shows nonuniform thickness. The thickest region of
the left ventricle is at the base, and the muscle of the ventricular wall becomes
exceedingly thin toward the apex. At the apex there is a potential continuity
between the endo- and epicardium. In the right ventricle the lower portion of the
anterior wall in the region of the anterior sulcus is extremely thin (19).
The superficial muscle fibers have a complex orientation. Fibers on the diaphrag-
matic aspect of the left ventricle pass longitudinally until they approach the apex,
where they become circumferential and form the vortex of the left ventricle. A
similar architecture of the apical fibers is also seen on the right side. Anterior fibers
of both ventricles run in an oblique fashion, interdigitating in some areas, particu-
larly in the infundibulum (20,22,23).
In contrast to the complexity of the superficial fibers, the arrangement of deeper
fibers is more simple. Circumferentially arranged middle fibers are confmed to the
left ventricle and septum. The greatest thickness of circumferential fibers is located
at the left ventricular base, where they encircle inlet and outlet of blood flow.
Moving toward the apex, the middle layer gradually becomes thinner, so that apical
to the insertion of the papillary muscles, the superficial and subendocardial layers
form the full thickness of the wall. The free right ventricular wall has only
superficial and subendocardial layers.
The subendocardial muscle layer of the left ventricle is continuous with the
superficial fibers of the apical vortex. Muscle fibers forming the trabeculae run
longitudinally. Subendocardial fibers run more obliquely to the long axis deep to
and between the trabeculae. Figure 9-2 depicts these ventricular muscle layers and
the direction of fiber orientation in a highly schematized manner. The papillary
MIDDLE LAYER
SUPERFICIAL LAYER
LV
RV POST
LT+RT
ANT
SUBENDOCARDIAL LAYER
Figure 9-2. Muscle fiber direction of the left ventricle. Upper left panel: Fiber direction of the
superficial myocardial muscle layer. Upper right panel: The solid arrows depict fiber direction of
the middle myocardial muscle layer. The dashed arrows show relation of superflcial layer
orientation. Lower panel: Subendocardial muscle fiber direction is shown.
muscles of the right ventricle arise from the subendocardial fibers, but are less well
formed than in the left ventricle.
The septum is formed from the subendocardial fibers of right and left ventricle,
together with a middle layer composed of circumferential fibers continuous with
those from the corresponding layer of the free wall of the left ventrIcle. The septum
therefore "belongs" to the left ventricle by virtue of its thick middle layer (23).
The cardiac fibrous skeleton supports the leaflets of the atrioventricular valves
and attaches them to the ventricular mass. The muscle fibers of the ventricles arise
by branching from one another rather than taking origin from the fibrous skeleton.
This feature is in concert with the origin of the heart as an expanded blood vessel.
Fiber orientation is maintained during systole and diastole despite the change in
wall thickness. This type of muscle architecture allows for only minimal tension
development between successive myocardial layers during contraction.
Ventricular pathology
Credible anatomic studies of the left ventricle in patients with COPD began to
appear in the late 1960s (24-28) when the method of Fulton for left ventricular
anatomic assessment (13) became widely accepted. This technique separates the atria
251
from the ventricles, removes extraneous tissue from the outer cardiac surface
(vessels and fat), and separates the right free ventricular wall froIll the septum along
with excision of the septum from the left ventricle. These three structures are then
weighed. Recently, the septum and left ventricle have been combined by omitting
amputation of the septum. The incidence of left ventricular hypertrophy in studies
using this method ranged from 11 to 37% in populations with COPO. Table 9-1
was constructed by including right and left ventricular weights from patient data
presented by the respective authors. In each series patients were excluded for clinical
evidence of systemic hypertension, coronary artery disease, or left ventricular
valvular disease. A ventricle was considered to be either normal or hypertrophied
based on the ventricular weight using the criteria of Fulton et al. (13). When data
from comparable groups are pooled and reexamined, the following pattern
emerges: the overall incidence of left ventricular hypertrophy in patients with
COPO is approximately 26%, with virtually all left ventricular changes occurring
in association with right ventricular hypertrophy.
Murphy et al. (26) studied postmortem prevalence of left ventricular hypertro-
phy in 72 patients with an unequivocal clinical diagnosis of COPo. Thirty-three
patients (46%) had right ventricular hypertrophy, and 20 patients (28%) had left
ventricular hypertrophy, including 12 patients with hypertensive, arteriosclerotic
heart, or aortic valve disease. Thus, in this study, left ventricular hypertrophy in
patients with COPO usually resulted from associated disease states. In 8 patients
(11 %), however, the etiology of left ventricular hypertrophy was unknown.
Cardiac growth in humans involves an increase in the number of myocardial
cells (hyperplasia) throughout fetal life and following birth until the age of three
months. Thereafter, cardiac growth is associated with progressive myocardial cell
enlargement (hypertrophy). Hypertrophy appears to be the primary method by
which myocardial mass increases during normal growth from infancy to adulthood
(29). Subsequent increases in ventricular mass result from the addition of new
contractile proteins in series and parallel within a virtually constant population of
cardiac muscle cells.
The nature of left ventricular hypertrophy in disease states was further examined
by Ishikawa and coworkers (27), who measured fiber diameter as a function of
ventricular size. In their study of adult normal hearts from patients with known
systemic hypertensive cardiovascular disease and from patients with cor pulmonale
due to COPO, they found that as the right ventricle gained mass the diameter of
the myocardial fibers of the right ventricle increased. The diameter of left ventricu-
lar fibers increased along with right ventricle fiber diameter changes, but not to
the same degree. The changes in left ventricle fiber diameter in patients with COPO
did not mirror the fiber diameter changes seen in "pure left ventricular hypertro-
phy" resulting from hypertensive cardiovascular disease, but left ventricle fiber
diameter changes in those hearts of COPO patients with combined increased right
and left ventricular mass were increased compared to the controls (27). Septal fiber
diameter changes followed the left ventricle rather than the right ventricle, a
phenomenon that has been noted previously with ventricular mass changes.
tv
""tv
Table 9-1. The association of gross left ventricular changes with right ventricular changes in chronic lung disease with cor pulmonale
RVH/LV RV/LV RVH/LVH RV/LVH

Author Total # # of % of # of % of # of % of # of % of
(ref) patients patients total patients total patients total patients total
Bignon 40 16 (40) 6 (15) 16 (40) 2 (5)

(24)
Ishikawa 55 26 (47) 18 (33) 10 (18) (2)
(27)
Millard 46 23 (50) 16 (35) 7 (15) o (0)
(28)
R V = right ventricular weight is < 75 gm.

R VH = right ventricular hypertrophy where weight is 2:: 75 gm.
LV = left ventricular weight is < 190 gm.
LVH = left ventricular hypertrophy where weight is > 190 gm.
253
Experimental animal studies using the hypoxia model to produce right ventricu-
lar hypertrophy have shown left ventricular weight changes similar to that seen
in patients with COPD (30). In addition to heart weight changes, focal myocardial
ischemic lesions have been noted by both Samad and Noehren (11) and Moret et
al. (31). The etiology of these focal degenerative changes is unknown but they
appear to be independent of coronary artery disease. Of particular relevance is a
recent animal study by Larson and coworkers (32) demonstrating significant left
ventricular mass increases with two experimental models of right ventricular
pressure overload. Right ventricular pressure overload (RVPO) was produced by
repeated intravenous injections of silica particles or by intraperitoneal injections of
monocrotaline, which produces R VH in rodents via a toxin-induced pulmonary
artery fibrosis. Volume overload to the right ventricle was accomplished by surgi-
cally implanting a shunt graft between the pulmonary artery outflow tract and
right atrium. These models developed not only significant R VH when compared
to controls but also significant increases in left ventricular mass. Furthermore,
ventricular mass changes were correlated with plasma epinephrine concentration,
irrespective of any alterations in hemodynamic parameters. Norepinephrine levels
did not correlate with either right or left hypertrophy in any of the overload
models. Propranolol, a beta-adrenergic antagonist, attenuated myocardial growth
in the silica pressure-overload model either by (a) selective beta-receptor blockade
of the myocardium and/or (b) decreased release of epinephrine from the adrenal
medulla. Similarly, epinephrine levels have been shown to increase and parallel the
development of LVH in response in aortic constriction in the dog, and propranolol
blocked LVH in the acute phase (first 96 hours) (33).
Since an increased myocardial mass is predominantly the result of enlargement
rather than the multiplication of the myocardial cells, some authors have suggested
that this response may be detrimental since diffusion distances for nutrients must
increase. The issue of capillary density or myocardial fiber to capillary ratio is
controversial because human studies are rare. The capillary density does appear to
maintain a 1 to 1 fiber-capillary relationship during hypertrophy in the adult
human heart (34).
Does ventricular hypertrophy mean dysfunction?

Does ventricular hypertrophy, of necessity, equal ventricular dysfunction? When
the hyperfunction of most organs take place in a healthy individual in response to
physiological stress, it is ordinarily transient and disappears after cessation of the
stress. In contrast, the compensatory cardiac hyperfunction caused by valvular
destruction or by persistently high arterial (pulmonary or systemic) blood pressure
is protracted because the lesions that have developed in the body are irreversible,
and the persistence of hyperfunction becomes necessary to life (35,36).
Folborth, cited by Meerson, noted that in organs subjected to high functional
strain, repetition of functional stresses may cause two opposite states to develop.
When the organ has had time to attain a state of complete restoration after a
preceding exhaustion, it becomes adapted and its working capacity is gradually
enhanced. When it is subjected to sustained activity before having attained a

condition of stable restoration, it then becomes chronically exhausted (35). This
state of affairs does not apply to compensatory hyperfunction of the left ventricle.
Despite unremitting strenuous activity, the phenomenon of exhaustion is observed
only very late. For example, many patients with valvular heart disease or hyperten-
sion live for decades with their condition and are capable of performing considera-
ble physical work. In contrast, mortality may rise to 70% within three years of
the development of cor pulmonale.
Studies of compensatory cardiac hyperfunction in the animal model have dealt
primarily with the left ventricle, although the sequences of cardiac exhaustion from
persistent physical stress such as a sustained increase in ventricular afterload are
remarkably comparable to that described in some clinical studies of cor pulmonale,
particularly the histologic finding of focal myocardial fibrosis (37).
In summary, pathologic evidence of left ventricular hypertrophy occurs in
varying incidence in association with chronic cor pulmonale. Recent experimental
data suggest that this type of left ventricular hypertrophy may develop in response
to elevated blood levels of epinephrine. Experimental data also suggest that hyper-
trophy per se does not always produce organ dysfunction although dysfunction
may occur as organ exhaustion develops from persistent physical stress.
PHYSIOLOGIC AND CLINICAL EVIDENCE OF LEFT VENTRICULAR

DYSFUNCTION
Experimental data and review of animal modeling

To gain insight into how the left ventricle might be affected in cor pulmonale,
animal modeling must incorporate the primary stimulus of pressure overload to
the right ventricle. Other determinants of ventricular function in cor pulmonale
include the levels of hydrogen ion concentration and blood carbon dioxide and
oxygen tensions.
The effect of hypoxia on ventricular myocardium is complex. Mild to moderate
hypoxia, or oxygen saturation levels to 50%, produces an increase in myocardial
contractility, evidenced by a raised left ventricular systolic pressure (LVSP) in the
denervated heart preparation (38). A reduction in LVSP occurs with severe
hypoxia, or oxygen levels below 50% saturation (39). The ventricular myocardium
is tolerant of increased hydrogen ion concentration without concomitant hypoxia,
but it is particularly susceptible to depressed function when both oxygen and
hydrogen ion levels are deranged. Although changes in myocardial metabolism
engendered by hypoxemia are not well understood, acute acidosis appears to affect
ventricular function both by exerting a direct negative inotropic effect on the left
ventricle and by reducing ventricular responsiveness to exogenous catecholamines
(38-41).
Increasing the blood carbon dioxide tension produces a consistent fall in maxi-
mum left ventricular dP / dt and an increase in left ventricular end-diastolic pressure
(figure 9-3) (42). Measurements following beta blockade support an earlier view
255
Aortic
flow.
Iml. sec-I) ':Ll\-
Stroke
volume
Imll
Maximum
left ventricular
I_I
2000
LI1
dP/dt
ImmHg . sec-II 0
']I\ !
Left
ventricular
pressure
ImmHg) J
L-.J L----.J
I~ sec 10 sec
Figure 9-3. Chart recording of changes in hemodynamic variables following the start of
administration of 5% CO 2 at the time indicated by the arrow. (From reference 42, with
permission.)
of Noble et al. (43) that the initial transient depression of the heart produced by
hypercarbia is a direct effect and the subsequent stimulation is an adrenergic
mediated response.
The cardiopulmonary unit consists of three an1tomic subunits, two muscular
pumps in-series, which are mechanically coupled by the third subunit, the lungs.
Although the right and left ventricular pumps normally perform at different
pressure levels, the anatomy of the heart allows for direct interaction between
ventricles. As we have seen the myocardium is a syncytium of muscle fibers tethered
within a supportive network of collagen. The contraction, or shortening, of these
muscle fibers results in the generation of pressure within the ventricular chambers,
thickening of the wall, and a reduction in ventricular volume. The load on the
muscular wall and its fibers is a function of both the chamber pressure and the
configuration of the chamber. For example, the myocardium surrounding an
enlarged ventricle has to develop more force to generate a given level of pressure
than does a normal-sized heart. Wall stress, force per unit of cross-sectional area
of myocardium, is homogeneously distributed throughout the ventricle favoring
uniform wall motion. In a similar fashion stress is uniformly distributed across the
interventricular septum. However, septal stress will be a function of the right
ventricular forces acting on the septal surfaces. A marked gradient in transseptal

surface forces would promote a shift of the septum to the chamber with less force
and potentially obliterate this ventricle. Septal displacement, which occurs as a
result of a disturbance in the transseptal gradient in force, will also be accompanied
by changes in shape, size, and effective compliance of each ventricle.
The magnitude and time course of wall force and stress vary throughout the
cardiac cycle. At end-diastole the diastolic configuration and pressure of either
ventricle determine the distending force, or preload, on its wall. Preload determines
muscle fiber length before contraction. The force the myocardium must support
during ejection-the afterload-is a function of the corresponding change in
chamber configuration and pressure. Normally, afterload peaks shortly after the
onset of ejection. Because chamber size decreases more than pressure increases,
afterload decreases during the remainder of ejection. Hence, the heart is normally
able to partially unload itself during its contraction (44).
During systole the pressure in one ventricle will also influence the systolic
pressure developed in the other. The systolic interaction between ventricles particu-
larly applies to the influence that the left ventricle exerts on right ventricular
pressure development. The influence of right ventricular systolic pressure is appar-
ent only when the pressure generated by the left ventricle is reduced to hypotensive
levels and is comparable with right ventricular systolic pressure. This interaction
is minimal under normal physiologic conditions.
Diastolic interaction between the ventricles is more pronounced with an in-
creased right ventricular volume. When the transseptal force gradient favors a
rightward shift of the septum, as with acute left ventricular distention, diastolic
pressure in the right ventricle either increases or remains invariant despite a reduc-
tion in chamber volume. The reverse is also true for the left ventricle. The septal
surface forces also serve to buttress the septum during systole and retard its motion
so that the pressure developed by either ventricle becomes greater.
The viscoelastic properties of the pericardium impose a physical constraint on
acute cardiac distention. In the experimental setting ventricular interdependence
markedly decreases when the pericardium is removed. Other changes noted with
pericardial removal include smaller changes in left ventricular diastolic pressure for
a given increase in afterload and decreased interaction between right ventricular
distention and left ventricular afterload. The opposite extreme of the removal of
the pericardium is typified by pericardial tamponade, when even small changes in
right ventricular volume and left ventricular afterload are manifested by large
increases in left ventricular diastolic pressure and a marked decrease in left ventricu-
lar stroke volume (44,45).
With each inspiratory decrease in pleural pressure during spontaneous respira-
tion, the right ventricle distends because of an increased venous return and an
increased right heart afterload. This causes the left ventricle to be effectively stiffer,
increasing left ventricular diastolic pressure, decreasing pulmonary venous return,
and, therefore, decreasing the left ventricular stroke volume. This phenomenon
interacts with the effective increase in left ventricular afterload created by the
257
decrease in pleural pressure. The relationship between left ventricular and pleural
pressures was further expanded by Permutt and Buda et al., who demonstrated that
during situations of dynamic alterations in intrapleural pressure, transmural left
ventricular pressure (PT~-that is, the left ventricular pressure (PLV) relative to
intrapleural pressure (PpJ, or PLY minus PpL-more accurately reflects left ven-
tricular afterload than does systolic aortic pressure. Thus, large sustained changes
in intrathoracic pressure may affect left ventricular function not only by altering
ventricular filling but also by changing afterload. When left ventricular transmural
pressure falls, afterload decreases; when left ventricular pressure increases, afterload
increases. In the presence of large changes in intrathoracic pressure in man, one must
consider their influence on left ventricular transmural pressure and the resulting
effects on cardiac performance (46,47).
Brisket disease in calves, a relatively "pure" form of pulmonary hypertensive
heart disease occurring as a result of chronic exposure to moderately high altitudes,
provides an animal model to examine the left ventricle in pulmonary heart disease.
On occasion these animals develop an elevated pulmonary capillary wedge, left
atrial, and left ventricular end-diastolic pressure. The mechanism of left ventricular
involvement in brisket disease is not known. Several possibilities were suggested
by Hecht et al., including conforming restrictions of the pericardium or excessive
pericardial fluid; however, hemodynamic measurements and autopsy studies did not
support such a mechanism. Although hypoxia is crucial to t;!!e development of
brisket disease, the degree of hypoxemia was not striking and the correlation
between the level of arterial oxygen saturation and LVEDP was poor. An increase
in collateral pulmonary blood flow as a cause of left ventricular dysfunction has
been suggested in cystic lung disease and bronchiectasis, but no lesions of this type
were found on autopsy. Unfortunately, no clear answer was obtained (48).
In both acute and chronic right ventricular pressure overload models, derange-
ments in left ventricular performance have been observed. This is manifested in the
chronic preparation by a diminution of left ventricular peak systolic pressure, peak
dP Idt, and calculated peak systolic wall stress developed during isovolumic con-
traction over a wide range of left ventricular end-diastolic pressures (49). Bemis
and coworkers found that left ventricular end-diastolic pressure and geometry were
significantly altered by acute changes in right ventricular filling pressures. Aug-
mented right ventricular filling resulted in an apparent decrease in left ventricular
compliance, producing a direct shift of the interventricular septum toward the left
ventricle. The magnitude of change in shape, albeit small, produced a substantial
change in left ventricular filling pressure (46,50). Left ventricular volume changes
are illustrated in figure 9-4. Stool and coworkers (51), using an animal model of
acute pulmonary hypertension, found that left ventricular stroke volume decreased
when the mean pulmonary arterial pressure was increased to 60 mmHg. Right
ventricular end-diastolic and end-systolic volumes increased approximately 50%
above control values at this level of mean pulmonary artery pressure, while right
ventricular ejection fraction remained essentially unchanged.
During acute pulmonary hypertension a characteristic distortion of left ventricu-
[wr -
80
EF 047 047 047 046 041 037
70
_ 60
-',
1
u
u
sv ,
...... ...
;:;
~
~o
I
r ---
::I
..J 40
0
>
~ 30
ESV
--- I --- - ~
20
10
10 20 30 40 ~O 60
MEAN PULMONARY ARTERY PRESSURE (mmHo)
Figure 9-4. Effect of progressive pulmonary hypertension on left ventricular volumes. Volumes
represent mean data from eight dogs and brackets represent the standard error of the mean.
(EDV = end-diastolic volume; EF = ejection fraction; ESV = end systolic volume; LV = left
ventricle; SV = stroke volume.) (From reference 51, with permission.)
lar geometry accompanies the increase in right ventricular volumes. Specifically,

a disproportionate shortening of the septal-lateral axis occurs at both end-diastole
and end-systole at all levels of pulmonary artery pressure. This selective geometric
change was attributed to a bulging of the interventricular septum toward the cavity
of the left ventricle and may contribute to the apparent decrease in left ventricular
compliance seen in volume-loaded animal preparations (51).
Clinical data review

Despite intense interest in left ventricular function in the setting of cor pulmonale,
the evidence for left ventricular dysfunction developing as a sequela of chronic
pulmonary hypertension has been felt to be extremely weak among reviewers and
editorial writers over the past decade (52-58).
Since pulmonary heart disease commonly occurs in smokers in their fifth decade
or beyond, several investigators have suggested that coexistent coronary artery
disease could be the cause ofleft ventricular dysfunction in these patients (59-63).
No differences have been found in the incidence of coronary artery disease in
patients with and without COPD. Although two autopsy studies have shown that
fewer patients with COPD die from acute myocardial infarction compared to
patients without lung disease (59,61), there is little support for the contention that
the chronic hypoxia and/or hypercapnia associated with advanced COPD may
result in the development of intercoronary anastomoses which improve coronary
collateral circulation (61).
In addition to autopsy studies, numerous workers have utilized the cardiac
259
Table 9-2. Catheterization data of patients at rest with COLD and cor pulmonale!
Total number Number of patients with PCWP

Author (ref) of patients or LVEDP > 16 mmHg at rest2
Jezek (64) 19 4
Rubin (65) 38 5
Christianson (66) 19
Kline (67) 20 2
Louridas (88) 13 1
Rice (89) 7 1
Krayenbuehl (70) 10 2
Dexter (71) 15
Herles (72) 92 12
Rao (73) 4 3
Segel (74) 15 3
Lockhart (75) 86 12
Matthay (76) 9
Baum (77) 15 4
Parker (78) 9
Parker (79) 10
Davies (80) 12
Jezek (81) 32 1
Evans (82) 47 3
Williams (83) 16
Khaja (84) 20
Moret (31) 22 1
Horsfield (85) 17 3
Weitzenblum (86) 32
Stem (87) 29 5
Total 608 62
'Percent abnormal PCWP and/or LVEDP 62/608 = 10%.

2LVEDP, left ventricular end-diastolic pressure
PCWP, pulmonary capillary wedge pressure
catheter to defme left ventricular function. Table 9-2 shows the incidence of left
ventricular dysfunction at rest in several studies in which cardiac catheterization
was performed (33,64-89). A pulmonary capillary wedge pressure (PCWP) of 16
mmHg or greater is considered indicative of left ventricular dysfunction, based on
data generated from the Duke University catheterization laboratory: in 619 patients
undergoing routine right and left heart catheterization for workup of clinical chest
complaints and who had normal hemodynamics, normal coronary angiograms, and
normal left ventricular function as determined by biplane angiography, the mean
PCWP was 7.7 + 2.8 mmHg (SO) and ranged from 1 to 16 mmHg (90). It could
be argued, however, that a PCWP of > 16 mmHg is a liberal criterion for LV
dysfunction. The high incidence of LV dysfunction in some reports may be
explained on the basis of using a lower PCWP as evidence for LV dysfunction and
acceptance of exercise data (91,92). Accurate measurements ofPCWP are difficult
to obtain during exercise in patients with COPD because of the large swings in
intrathoracic pressure.
The left ventricle in right-sided cardiac disease

Right ventricular pressure overload is seen in congenital pulmonic valve stenosis,
and isolated cases of a depressed left ventricular ejection fraction (LVEF) have been
reported: in one study 3 of21 patients (14%) had LVEF below 56%, but the mean
LVEF was 63%, which is comparable to normal mean LVEF data. On the other
hand, in children with pulmonic stenosis and right-to-Ieft shunt, the LVEF appears
to be more significantly depressed. In a retrospective analysis of young adults with
mild pulmonic stenosis, 6 out of 7 patients had an elevation of the left ventricular
end-diastolic pressure (LVEDP) both at rest and during exercise (93,94).
Experimental right ventricular pressure overload produces anatomic changes in
the left ventricle at the cellular level. Pulmonary artery banding caused increases
in cell length, decreases in sarcomere length in both left and right ventricles, and
significant increases in the intercalated disc width. Left ventricular weight in-
creased, but to a lesser extent and at a slower rate than in the right ventricle. At
the same filling pressure, both left and right ventricular volumes were significantly
greater in the pulmonary artery banded hearts compared to normal hearts.
Left ventricular biochemical abnormalities have also been reported in the experi-
mental pressure overload-induced right ventricular hypertrophy state. Myocardial
collagen, as measured by hydroxyproline levels, was increased in both the right and
the unstressed left ventricle. Left ventricular myofibrillar ATPase activity was
significantly depressed in the presence of right ventricular failure, and left ventricu-
lar myofibrillar ATPase activity was lowest in the hearts with the lowest right
ventricular myofibrillar ATPase activity. In a similar animal preparation, signifi-
cantly lower levels of total creatine and creatine phosphate were found in the left
ventricles of cats with right ventricular failure. Numerous workers have found
reduced stores of norepinephrine in both left and right ventricles of dogs with heart
failure produced by pulmonary artery banding (95). We previously saw that
experimentally induced right ventricular pressure overload decreased cardiac out-
put, LVEF, stroke volume, end-diastolic volume, and end-systolic volume. The
reductions in left ventricular volume were associated with left ventricular axis
changes measured by angiocardiography (51). Thus, left ventricular end-diastolic
geometry is distorted by right ventricular pressure overload, and the left ventricle
does not return to its normal shape during systole. This might be the cause, in part,
for the dysfunction or reduced mechanical efficiency of the left ventricle in the
setting of pulmonary hypertension.
The cause for left ventricular dysfunction in total anomalous pulmonary venous
return and large atrial septal defects is somewhat more complicated. Here, left
ventricular end-diastolic volume (LVEDV) and LVEF are significantly decreased,
but whether these changes are due to LV underloading with resultant decrease in
LV size and systemic flow or reflect a basic alteration in myocardial contractile state
is difficult to ascertain (96).
The effect of an atrial septal defect on left ventricular function in adults and
children seems to be more straightforward. Left ventricular dysfunction-mani-
261
fested by decreases in systemic blood flow, LVEF, and left ventricular stroke
volume-and abnormal interventricular septal motion by angiocardiography and
echocardiography have been observed. The cause of left ventricular failure in atrial
septal defect, first reported clinically by Dexter (97), was postulated to be a reverse
Bernheim's syndrome: the encroachment on the left ventricular cavity by the
interventricular septum causes an interference with filling. Patients undergoing
surgical closure of an atrial septal defect showed inappropriate cardiac output
responses to exercise postoperatively compared to age-matched control subjects.
These data suggest that left ventricular dysfunction may be present after operation,
even in patients who are asymptomatic (94).
In a group of 25 adults with atrial septal defect who were studied at rest and
exercise preoperatively, only eight patients had normal biventricular function (98).
Left ventricular dysfunction was defmed as an LVEDP greater than 16 mmHg, an
increase in systemic cardiac output of less than 550 ml/min for each 100 ml/min
increase in oxygen consumption during exercise, or a decrease in left ventricular
stroke volume from rest to exercise associated with an increase in LVEDP. Right
ventricular function was considered abnormal if the right ventricular stroke vol-
ume decreased from rest to exercise in association with an increase in right ventricu-
lar end-diastolic pressure. The authors favored the concept that the functional and
structural impairment that occurs in response to overload of one ventricle produces
a generalized myocardial response with involvement of both ventricles. Other
studies evaluating left ventricular function in atrial septal defect have speculated
that a decrease in LV compliance resulting from decreased preload contributed to
left ventricular dysfunction (97,99,100).
The left ventricle in COPD

Williams et al. (83) performed right and left heart catheterization in 16 patients
with COPD who had no evidence of systemic hypertension, valvular heart disease,
previous myocardial infarction, or a history of angina pectoris. The patients were
subdivided into three groups: group 1 consisted of five patients with a mean
pulmonary artery pressure of 20 mmHg or less; group 2 was composed of 5 patients
with a mean pulmonary artery pressure of greater than 20 mmHg, but without a
history of right ventricular failure; and the remaining group consisted of six
patients with pulmonary hypertension (mean pulmonary artery pressure equal to
or greater than 30 mmHg) and a history of right ventricular failure. When
ventricular function curves were generated by plotting stroke volume against left
ventricular end-diastolic pressure (LVEDP) and stroke work against LVEOP be-
fore and during an infusion of methoxamine, the responses were similar in patients
with chronic cor pulmonale and in normal patients. The authors concluded that
the left ventricle performs normally in COPO, regardless of right ventricular
hemodynamics.
In a similar study by Matthay et al. (76), methoxamine infusion in patients with
COPO produced no change in stroke volume and left ventricular end-diastolic
volume. Left ventricular ejection fraction determined by first-pass radionuclide
techniques was unchanged during methoxamine infusion. The authors concluded

that latent ventricular dysfunction was not present. Vallette et al. (92) studied
incipient left ventricular dysfunction in 25 patients with COPD by producing
acute plasma volume expansion and used atrial pacing to unmask coronary artery
disease. Increasing heart rate to 80% of the theoretical maximal heart rate did not
produce chest pain or changes in ventricular repolarization. Plasma expansion,
achieved by intravenous infusion of macromolecular gelatin at a rate of 30 ml / min
up to 1 liter, increased cardiac output by an average of 32% without significantly
changing heart rate. Both right atrial and pulmonary capillary wedge pressures
increased significantly. One patient complained of chest tightness and wheezing
associated with a pulmonary wedge pressure of 20 mmHg, a condition that neces-
sitated discontinuation of the infusion. One other patient had abnormal ventricular
performance, although specifics were not given. The authors concluded that left
ventricular failure is uncommon in patients with COPD. The incidence of left
ventricular dysfunction in this study (2/25, 8%) is comparable to the level of
dysfunction reported in table 9-2. Matthay et al. (101) studied the left ventricular
response to volume expansion with dextran in patients with COPD, including four
with documented coronary artery disease. Left ventricular function following
dextran infusion did not deteriorate when compared to control values, but the
responses of this group were not compared to a control group. The study by Davies
and Overy (80) also supports the concept of normal left ventricular function in
chronic cor pulmonale. They measured left heart pressures and subtracted eso-
phageal pressure to account for alterations in intrathoracic pressure due to altered
lung mechanics. They found left ventricular end-diastolic pressure and cardiac
index were normal in patients with chronic lung disease. Based on these findings
Davies and Overy concluded that resting left ventricular performance was normal
in COPD.
Baum et al. (77) examined left ventricular performance during an acute in-
crease in afterload produced by a graded infusion of angiotensin. Left ventricular
performance was evaluated using single-plane cineventriculography, and left ven-
tricular function curves were constructed by plotting changes in stroke work
index (SWI) against LVEDP at different afterloads. Of fifteen patients only one
manifested a normal left ventricular response. Only one of ten patients who
underwent coronary arteriography and ventriculography had both normal coro-
nary angiograms and ventriculography, five had an increased left ventricular wall
thickness, five had increased left ventricular end-diastolic volume, but only two
patients had defmite evidence of coronary artery narrowing. These authors con-
cluded that left ventricular dysfunction occurred in the setting of lung disease
and that it was not influenced by the presence of right-sided heart failure. This
study is clearly in disagreement with the studies of Williams et al. (83) and
Matthay et al. (76). The impaired left ventricular function found in Baum's
study may have been the result of the ventricular depressant effect of angiotensin,
although this effect is generally considered to be minor and should not have
influenced the left ventricle to the degree noted in this study. In addition, the
263
authors imply that their patients had more severe airways disease compared to
patients in the other two studies.
Rao et al. (73) reported eight patients with severe COPO and clinical left
ventricular failure. None of these patients had systemic hypertension or coronary
or valvular heart disease. Of four patients who underwent right heart catheteriza-
tion, three had an elevated pulmonary capillary wedge pressure. Postmortem
examination in five patients showed hypertrophy and dilatation of both ventricles.
These authors could fmd no clear correlation with hypoxemia, hypercapnia, or
infection, but concluded that left ventricular failure may develop in the presence
of right ventricular failure secondary to chronic lung disease. Unfortunately,
separate ventricular weights were not obtained in the autopsy study, and ventricular
wall thickness was used as the criterion for hypertrophy. Lamb (14) has demon-
strated that there is no correlation between weight and thickness for individual
ventricles.
Left ventricular performance curves were also created by Khaja and Parker (84)
in 20 patients with COPO, nine with cor pulmonale. In plotting stroke work
against LVEOP, they found that left ventricular performance was slightly reduced
in the chronic cor pulmonale group compared to normals. Those patients without
cor pulmonale had ventricular function curves similar to normals. In plotting
stroke volume against LVEOP, left ventricular performance was markedly de-
pressed in those patients who had chronic lung disease and cardiac involvement.
In each case the patients without cor pulmonale had depressed slopes, while patients
with cor pulmonale showed even lower slopes than the other two groups. These
data showed depressed left ventricular function in some patients with COPo. The
patients with cor pulmonale had markedly elevated values for total pulmonary
resistance, and these authors suggested that decreased left ventricular performance
was a manifestation of reduced right ventricular work output which is the result
of the increased pressure work.
The study by Krayenbuehl et al. (70) offers the opportunity to examine left
ventricular function in association with primary pulmonary hypertension or pul-
monary hypertension resulting from thromboembolic disease. In this study values
for mean pulmonary arterial pressure and pulmonary vascular resistance were
higher than in most reports of patients with COPo. Left ventricular systolic
function in patients with chronic pulmonary hypertension was judged to be normal
at rest because the maximal velocity of contractile element shortening, determined
from left ventricular total pressure-velocity curves and volumetric measurements
derived from cineangiography, did not differ from normals. Similarly, left ven-
tricular ejection determined from cineangiograms was preserved at rest. In contrast
to left ventricular systolic function at rest, diastolic function appeared to be clearly
abnormal. Plotting LVEOP against volume showed decreased chamber compliance.
Measurements of left ventricular contractile function during isometric exercise,
Vmax, and VPM (peak velocity of shortening) increased only slightly and insignifi-
cantly in the patients with pulmonary hypertension, in marked contrast to control
subjects. The authors could only speculate as to the cause of impaired left ventricu-
lar contractile reserve, since hypoxemia was not a consistent fmding and coronary
artery disease, although not excluded, was considered unlikely. Septal bulging and
hypertrophy were noted on cineangiography and echocardiography and were
thought to play an important role in altering left ventricular geometry.
Angiocardiography
Christianson and co-workers (66) carried out a quantitative cineangiographic study
of the left ventricle in 19 patients with severe COPD. Approximately one-half of
the group (n=9) had a documented episode of right-sided heart failure previously.
Their methods enabled them to evaluate wall motion and thickness for assessment
of myocardial contractility. In addition, pulmonary capillary wedge and/or left
ventricular end-diastolic pressures were monitored, allowing a complete evaluation
of left ventricular contractility. Left ventricular end-diastolic pressure was elevated
and ejection fraction was significantly depressed in 3 of 19 patients either at rest
or exercise. These numbers are admittedly small, and quantitative indices of left
ventricular contractility did not correlate well with hemodynamic abnormalities.
Left ventricular wall motion was abnormal in the patients with depressed ejection
indices. Coronary arteriograms were not helpful in elucidating a reason for ven-
tricular dyskinesis. These authors suggest that chronic pressure overload to the right
ventricle does not alter left ventricular contractility at rest. However, the exercise-
induced left ventricular dyskinesis remains to be explained. The angiographic data
obtained by Krayenbuehl et al. (70) is in agreement with the measurements of
LVEF and mean velocity of circumferential fiber shortening at the minor axis of
the left ventricle by Christianson et al. (66).
Studies involving noninvasive techniques: Nuclear angiocardiography

With the addition of noninvasive tools to evaluate left ventricular function,
potential sources of error begin to influence data in an adverse fashion and further
confound the issue. First-pass radionuclide assessment of left ventricular ejection
fraction has been investigated by many workers (67,102-110). The focus on
patients with COPD without known coronary artery disease or cardiomyopathy
suggests an incidence of depressed left ventricular ejection fraction at rest or during
an episode of superimposed respiratory failure of 31 %, and 41 % during exercise
(table 9-3). In the study by Steele et al. (104), LVEF was measured in 28 patients
with stable COPD and in 92 patients with COPD and superimposed acute respira-
tory failure. In those patients with a severely depressed LVEF «40%),64% had
clinical or autopsy evidence of coronary artery disease. Twenty-three patients had
a moderately depressed LVEF (41% to 54%). None of the patients with a modest
depression of LVEF who died had coronary artery disease at autopsy. The authors
found no relationship between LVEF and arterial oxygen or carbon dioxide
tensions or pH and were unable to. ascribe an etiology for a depressed LVEF in
the absence of concomitant coronary artery disease.
Olvey et al. (103) found an abnormal LVEF response to exercise in one-half
265
Table 9-3. Left ventricular ejection fraction (LVEF)
Abnormal/Normal!
Author (ref) Rest Exercise Comments
Olvey (103) 0/18 9/18 Upright exercise bicycle ergometer

Kline (67) 3/20 2 patients were noted to have
tricuspid regurgitation
Steele (104) 51/120 See text for discussion
Matthay (105) 7/24 Upright exercise bicycle ergometer
Berger (106) 9/35
Slutsky (108) 3/20
Slutsky (109) 6/12 Supine exercise bicycle ergometer
Mathur (110) 4/15
Totals 70/228 22/54
% abnormal LVEF 31 41
labnormal LVEF ~ 50%
of their patients, which was attributed in part to possible latent coronary artery
disease.
In a group of 15 patients with stable COPD and evidence of right ventricular
dysfunction studied by Mathur et al. (110),4 had significantly depressed LVEF.
The authors did not attribute a cause for the depressed LVEF.
Slutsky et al. (109) studied three groups of patients during exercise with gated
equilibrium radionuclide angiography and right heart catheterization. The patients
included a group with angiographically demonstrated proven coronary artery
disease (CAD), a group with stable severe COPD, and a group with both severe
COPD and CAD. During exercise the patients with CAD had significant increases
in PCWP while both LV end-diastolic and end-systolic volumes increased and
ejection fraction was unchanged. The LVEF response to exercise was abnormal in
50% of COPD patients while the PCWP did not significantly change. Left
ventricular end-diastolic and end-systolic volumes both decreased. In patients with
both COPD and CAD exercise LVEF responses were abnormal, but, in contrast
to COPD patients without CAD, both LV end-systolic and end-diastolic volumes
increased. The authors concluded that the depression of left ventricular ejection
fraction in the COPD patients with and without CAD was due to different
mechanisms. Exercise-induced LV dysfunction associated with LV dilation is prob-
ably due to concomitant CAD, whereas a reduction in LV volume during exercise
implies the absence of abnormalities in contractility. Exercise produces a severe
exacerbation of pulmonary hypertension and increases the pressure overload on the
right ventricle in COPD, producing RV dilation, a reduced stroke volume, and
R V failure. The reduction in left ventricular preload likely explains the abnormal
left ventricular volume response to exercise in COPD alone.
Chipps et al. (111) examined left ventricular function in 21 patients with cystic
fibrosis. Because of their age, the presence of atherosclerotic or hypertensive disease
was unlikely. Radionuclide determination of left ventricular ejection fraction was
abnormal in four patients at rest and in three during exercise. The authors specu-
lated that myocardial fibrosis, which has been reported in some patients with cystic
fibrosis, may diminish the compliance of the ventricle and cause LV dysfunction.
Differences in techniques, including single-crystal or multi crystal scintillation
cameras, studies in the supine or upright positions, and bedside studies of critically
ill patients make comparing results from these studies difficult.
Studies involving noninvasive techniques: Echocardiography

The echocardiogram has not been used widely in the evaluation of the influence
of obstructive lung disease on the left ventricle. Brinker et al. (112) showed that
acute right ventricular afterload augmentation using the Mueller maneuver dis-
places the interventricular septum leftward and alters the left ventricle size and
shape in normals, a fmding that supports the concept of ventricular interdependence
in normal man. Clinical and experimental studies of left ventricular function in
chronic pulmonary hypertension (113,114) have added little to the clarification of
this problem.
Studies involving noninvasive techniques: Systolic time indices

The systolic time index (STI), described by Weissler and co-workers (115), is an
ideal noninvasive bedside procedure to evaluate left ventricular function. As a
parameter of left ventricular contractility, however, it falls far short of the mark.
These studies are difficult to perform on patients with severe COPD because of
technical problems related to lung hyperinflation (116). Numerous studies have
been carried out on patients with varying degrees of COPD and have shown that
the systolic time indices are indeed abnormal (117-120). Unger et al. (117) studied
a group of patients with severe COPO who had a recent increase in dyspnea.
Systolic time indices were measured as part of an investigation into the cause of
the change in dyspnea and showed a prolonged pre-ejection period index (PEPI)
in 65%, while the left ventricular ejection time index (LVETI) was reduced in
88%. The ratio of the pre-ejection time to the left ventricular ejection time
(PEP /LVET) was within the normal range in only 12% of the patients. Among
the patients with a normal PCWP both at rest and exercise, 32% had a normal
PEPI, 5% had a normal LVETI, and 5% had a normal PEP/LVET ratio. Of a
smaller group of patients who were noted to have an elevated PCWP at rest or
exercise, 43% had a normal PEPI while 29% had a normal LVETI and PEP !LVET
ratio. No significant differences were found in the systolic time indices between
the two groups. The study by Unger et al. is in agreement with the fmdings of
Louridas et al. (88), who found that the STI was abnormal in the majority of
patients with COPO, despite apparently normal left ventricular performance in
most. They concluded that STI cannot be used to identify patients who have an
elevated PCWP. These abnormalities, however, reflect changes in left ventricular
preload without describing the contractile state of the ventricle. Conditions of left
ventricular preload are again related to pathophysiologic changes that have oc-
curred in the lung and right heart as a consequence of COPO (121-122).
267
CONCLUSIONS
The anatomic evidence supports the contention that the left ventricle does increase
in size when a right ventricular pressure overload stress occurs in the presence of
chronic lung disease. The degree of left ventricular hypertrophy is less than would
occur if this chamber were the direct recipient of a pressure overload stimulus. The
macroscopic, microscopic, and clinical evidence for hypertrophy of the left ventri-
cle secondary to right ventricular hypertrophy has been presented. The develop-
ment of the left ventricular lesion associated with experimental right ventricular
hypertrophy has been demonstrated, and epinephrine has been implicated as a
possible mediator for cardiac myocyte hypertrophy.
The issue of left ventricular dysfunction in the clinical setting, however, is not
settled. Left ventricular filling pressures appear to be abnormally elevated in only
10% of patients with right ventricular failure and/or hypertrophy due to chronic
lung disease. The evidence for LV dysfunction from indirect hemodynamic tech-
niques is at variance with the catheterization information; this may be due to more
stringent catheterization criteria for LV dysfunction.
What has unfolded is a spectrum of left ventricular function in the presence of
cor pulmonale. To be sure, enlargement occurs in about one-third of cases, but
hypertrophy per se does not necessarily lead to ventricular dysfunction. It is
appropriate to determine left ventricular function using strict cardiac catheteriza-
tion criteria in the presence of right ventricular failure when all of the issues of
cor pulmonale have been clinically addressed and dyspnea remains unchanged.
REFERENCES
1. Rapaport E. Dyspnea: Pathophysiology and differential diagnosis. Prog Cardiovasc Dis 8:
532-545, 1971.
2. Fishman AP, Ledlie]F. Dyspnea. Bull Europ Physiopath Resp 15: 789-804, 1979.
3. Parkinson], Hoyle C. The heart in emphysema. Quart] Med 6: 59-86, 1937.
4. Kountz WB, Alexander HL, Prinzmetal M. The heart in emphysema. Am Heart] 1: 163-172,
1936.
5. Parker RL. Pulmonary emphysema: A study of its relation to the heart and the pulmonary
arterial system. Ann Intern Med 14: 795-809, 1940.
6. Scott R W, Garvin CF. Cor pulmonale: Observations in fifty autopsy cases. Am Heart] 22:
56-63, 1941.
7. Spain DM, Handler B]. Chronic cor pulmonale: Sixty cases studied at necropsy. Arch Intern
Med 77: 37--65, 1946.
8. Spatt SD, Grayzel DM. Cor pulmonale; observations on forty-two autopsied patients. Am] Med
5: 252-255, 1948.
9. Zimmerman HA, Ryan]M. Cor pulmonale: A report of an additional 52 cases compared with
a previous study of 50 cases. Dis Chest 20: 286-289, 1951.
10. Michelson N. Bilateral ventricular hypertrophy due to chronic pulmonary disease. Dis Chest
38: 435--446, 1960.
11. Samad lA, Noehren TH. Focal myocardial necrosis in the cor pulmonale of emphysema. Dis
Chest 48: 376-379, 1965.
12. Fluck DC, Chandrasekar RG, Gardner FV. Left ventricular hypertrophy in chronic bronchitis.
Br Heart] 29: 92-97, 1966.
13. Fulton RM, Hutchinson EC, Jones AM. Ventricular weight in cardiac hypertrophy. Br Heart
] 14: 413-420, 1952.
14. Lamb D. Heart weight and assessment of ventricular hypertrophy. In Recent Advances in
Clinical Pathology (series 6), Dyke SC (ed). London: Churchill Livingstone, 1973, pp. 133-148.
15. Bove KE, Scott RC. The anatomy of chronic cor pulmonale secondary to intrinsic lung disease.
Prog Cardiovasc Dis 9: 227-238, 1966.
16. Dadgar SK, Tyagi SP. Importance of heart weight, weights of cardiac ventricles and left
ventricle plus septum/right ventricle ratio in assessing cardiac hypertrophy. Jap Heart J 20:
63-73, 1979.
17. Reiner L, Mazzoleni A, Rodriguez FL, Frudenthal RR. The weight of the human heart. II.
Hypertensive cases. Arch Path 71: 180-201, 1961.
18. Mitchell RS, Stanford RE, Silvers GW, Dart G. The right ventricle in chronic airway obstruc-
tion: A clinicopathologic study. Am Rev Respir Dis 114: 147-154, 1976.
19. Greenbaum RA, Ho SY, Gibson DG, Becker AE, Anderson RH. Left ventricular fibre architec-
ture in man. Br Heart J 45: 248-263, 1981.
20. Rushmer RF. Functional anatomy and control of the heart. In Cardiovascular Dynamics, 4th
edition. Philadelphia: Saunders, 1971, pp. 76--131.
21. Grant RP. Architectonics of the heart. Am Heart J 46: 405-431, 1953.
22. Navaratham V. The structure of cardiac muscle. In Developments in Cardiovascular Medicine,
Dickinson CJ, Marks J (eds). Baltimore: University Park Press, 1978, pp. 119-120.
23. Streeter DD Jr. Gross morphology and fiber geometry of the heart. In Handbook of Physiology,
Berne RM, Sperelakis N (eds). Baltimore: Williams & Wilkins, 1979, pp. 61-112.
24. Bignon J, Khoury F, Even P, Andre J, Brouet G. Morphometric study in chronic obstructive
bronchopulmonary disease. Am Rev Respir Dis 99: 669-695, 1969.
25. Murphy ML, Boger J, Adamson JS Jr, Rubin S. Evaluation of cardiac size in chronic bronchitis
and pulmonary emphysema. Chest 71: 712-717, 1977.
bronchitis and emphysema. Ann Intern Med 81: 307-313, 1974.
27. Ishikawa S, Fattal GA, Popiewicz J, Wyatt]p. Functional morphometry of myocardial fibers
in cor pulmonale. Am Rev Respir Dis 105: 358-367, 1972.
28. Millard FJC. The electrocardiogram in chronic lung disease. Br Heart J 29: 43-50, 1967.
29. Grossman W. Cardiac hypertrophy: Useful adaptation or pathologic process? Am J Med 69:
576--584, 1980.
30. Widimsky J, Urbanova D, Ressi J, Ostadel B, Pelouch V, Prochazka J. Effect of intermittent
altitude hypoxia on the myocardium and lesser circulation in the rat. Cardiovasc Res 7: 709-909,
1973.
31. Moret DR, Bopp P, Panos J. Coronary circulation and myocardial metabolism in chronic cor
pulmonale. In Fifth World Congress of Cardiology Symposia, pp. 158-171.
32. Larson DF, Womble JR, Copeland JG, Russel DH. Concurrent left and right ventricular
hypertrophy in dog models of of right ventricular overload. J Thoracic Cardiovasc Surg 84:
543-547, 1982.
33. Womble JR, Haddox MK, Russell DH. Epinephrine elevation in plasma parallels canine cardiac
hypertrophy. Life Sci 23: 1951-1958, 1978.
34. Pearlman ES, Weber KT, Janicki JS. Quantitative histology of the hypertrophied human heart.
Fed Proc 400: 2042-2047, 1981.
35. Meerson FZ. Compensatory hyperfunction of the heart and cardiac insufficiency. Circ Res 10:
250-258, 1962.
36. Grimm AF, Kubota R, Whitehorn WV. Properties of myocardium in cardiomegaly. Circ Res
12: 118-124, 1963.
37. Beznak M. Cardiac output in rats during development of cardiac hypertrophy. Circ Res 6:
207-212, 1958.
38. Ng ML, Levy MN, De Geesi H, Zieske H. Effects of myocardial hypoxia on left ventricular
performance. Am J Physiol 211: 45-50, 1966.
39. Downing SE, Tilner NS, Gardner H. Influence of hypoxemia and acidemia on left ventricular
function. Amer J Physiol 210: 1327-1335, 1966.
40. 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 41: 1553-1563, 1962.
41. Wildenthal K, Mierzwiak DS, Myers R W, Mitchell JH. Effects of acute lactic acid acidosis on
left ventricular performance. Amer J Physiol 214: 1352-1359, 1968.
42. Van den Bos C, Drake AJ, Noble NIM. The effect of carbon dioxide upon myocardial
contractile performance, blood flow and oxygen consumption. J Physiol (Lond) 287: 149-161,
1979.
269
43. Noble NIM, Trenchard D, Guz A. Effect of changes in PaC0 2 and Pa02 on cardiac perform-
ance in conscious dogs. J Appl Physiol 22: 147-152, 1966.
44. Weber KT,JanickiJS, ShroffS, Fishman AP. Contractile mechanics and interaction of the right
and left ventricles. Am J Cardiol 47: 686--695, 1981.
45. Robotham JL. Cardiovascular disturbances in chronic respiratory insufficiency. Am J Cardiol
47: 941-949, 1981.
46. Permutt S. Relation between pulmonary arterial pressure and pleural pressure during the acute
asthmatic attack. Chest 63: 255-285, 1973.
47. Buda AJ, Pinsky MR, Ingels NB Jr, Daughters GT, II, Stinson EB, Alderman EL. Effect of
intrathoracic pressure on left ventricular performance. N Engl J Med 301: 453-459, 1979.
48. Hecht HH, Kuida H, Tsagaris lJ. Brisket disease. IV. Impairment of left ventricular function
in a form of cor pulmonale. Trans Assoc Am Physicians 65: 63-276, 1962.
49. Kelly DT, Spotnitz M, Beiser GD, Pierce JE, Epstein SE. Effects of chronic right ventricular
volume and pressure loading on left ventricular performance. Circulation 44: 403-412,
1971.
50. Bemis CE, Serur JR, Borkenhagen D, Sonnenblick EH, Urschel CWo Influence of right
ventricular filling pressure on left ventricular performance and dimension. Circ Res 34: 498-504,
1974.
51. Stool EW, Mullins CB, Leshin SJ, Mitchell JH. Dimensional changes of the left ventricle during
acute pulmonary arterial hypertension in dogs. Am J Cardiol 33: 866-875, 1974.
52. Weisse AB. Contralateral effects of cardiac disease affecting primarily either the left or right
chambers of the heart. Am Heart J 87: 654-660, 1974.
53. Ferrer MI. Cor pulmonale (pulmonary heart disease): Present day status. Am Heart J 89:
657-664, 1975.
54. Fishman AP. The left ventricle in chronic bronchitis and emphysema (editorial). N EnglJ Med
285: 402-404, 1971.
55. Thomas AJ. Chronic pulmonary heart disease (editorial). Br Heart J 34: 653-657, 1972.
56. Kachel RG. Left ventricular function in chronic obstructive pulmonary disease. Chest 74:
286-290, 1978.
57. Alpert JS. Pulmonary hypertension and cardiac function in chronic obstructive pulmonary
disease (editorial). Chest 75: 651-652, 1979.
59. Mitchell RS, Walker SH, Maisel JC. The causes of death in chronic airway obstruction. II.
Myocardial infarction. Am Rev Respir Dis 98: 611-612, 1968.
60. Kassis E. Systetnic hypertension, left ventricular hypertrophy and myocardial infarction in
patients with chronic obstructive lung disease. Scand J Resp Dis 58: 324-329, 1977.
61. Samad lA, Noehren TH. Myocardial infarction in pulmonary emphysema. Dis Chest 47: 26-29,
1965.
62. Bhargava RK, Woolf CR. Relationship between ischemic heart disease and chronic diffuse
pulmonary disease. Chest 59: 254-261, 1971.
63. Karpick Rj, Pratt PC, Asmundsson T, Kilburn KH. Pathological fmdings in respiratory failure.
Goblet cell metaplasia, alveolar damage and myocardial infarction. Ann Intern Med 72: 189-197,
1970.
64. Jezek V. Left ventricular volumetry and function in chronic cor pulmonale. Cor Vasa 23:
94-103, 1981.
65. Rubin LJ, Chappell TR. Personal communication.
66. Christianson LC, Shah A, Fisher VJ. Quantitative left ventricular cineangiography in patients
with chronic obstructive pulmonary disease. Am J Med 66: 399-404, 1979.
67. Kline LE, Crawford MH, MacDonald WJ Jr, Schelbert H, O'Rourke RA, Moser KM. Nonin-
vasive assessment ofleft ventricular performance in patients with chronic obstructive pulmonary
disease. Chest 72: 558-564, 1977.
68. Boushy SF, North LB. Hemodynamic changes in chronic obstructive pulmonary disease. Chest
72: 565-570, 1977.
69. Jezek V, Schrijen F. Hemodynamic effect of deslanoside at rest and during exercise in patients
with chronic bronchitis. Br Heart J 35: 2-8, 1973.
70. Krayenbuehl HP, Turina J, Hess 0. Left ventricular function in chronic pulmonary hyperten-
sion. Am J Cardiol 41: 1150-1158, 1978.
71. Dexter L, Whittenberger JL, Gorlin R, Lewis BM, Haynes FW, Spiegel RJ. The effect of
chronic pulmonary disease (cor pulmonale and hypoxia) on the dynamics of the circulation in
man. Trans Assoc Am Physicians 54: 226--236, 1951.
72. Herles F, Jezek V, Daum S. Site of pulmonary resistance in cor pulmonale in chronic bronchitis.
Brit Heart J 30: 654-660, 1968.
73. Rao BS, Cohn KE, Eldridge FL, Hancock EW. Left ventricular failure secondary to chronic
and during exercise, with special reference to the changes in blood viscosity and blood volume
on the pulmonary circulation. J Clin Invest 45: 1555-1568, 1966.
75. Lockhart A, Tzareva M, Nader F, LeBlanc P, Schrijen F, Sadoul P. Elevated pulmonary artery
wedge pressure at rest and during exercise in chronic bronchitis: Fact or fancy. Clin Sci 37:
503-517, 1969.
76. Matthay RA, Ellis JH Jr, Steele PP. Methoxamine-induced increase in afterload. Effect on left
ventricular performance in chronic obstructive pulmonary disease. Am Rev Respir Dis 117:
871-877, 1978.
77. Baum GL, Schwartz A, Llamas R, Castillo C. Left ventricular function in chronic obstructive
lung disease. N Engl J Med 285: 361-365, 1971.
78. Parker JO, Kelkar K, West RO. Hemodynamic effects of aminophylline in cor pulmonale.
Circulation 33: 17-25, 1966.
79. Parker JO, Ashekian PB, DiGiorgi S, West RO. Hemodynamic effects of aminophylline in
chronic obstructive pulmonary disease. Circulation 35: 365-372, 1967.
80. Davies H, Overy HR. Left ventricular function in cor pulmonale. Chest 58: 8-14, 1970.
81. Jezek V, Schrijen F. Left ventricular function in chronic obstructive pulmonary disease with and
without cardiac failure. Clin Sci Mol Med 45: 267-279, 1973.
82. Evans TO, Van der Reis L, Selzer A. Circulatory effects of chronic pulmonary emphysema.
Am Heart J 66: 565-570, 1977.
83. Williams JF, Childress RH, Boyd DL, Higgs LM, Behnke RH. Left ventricular function in
patients with chronic obstructive pulmonary disease. J Clin Invest 47: 1143--1153, 1968.
84. Khaja F, Parker JO. Right and left ventricular performance in chronic obstructive lung disease.
Am Heart J 82: 319-327, 1971.
85. Horsfield K, Segel N, Bishop JM. The pulmonary circulation in chronic bronchitis at rest and
during exercise breathing air and 80% oxygen. Clin Sci 34: 473-483, 1968.
86. Weitzenblum E, Loiseau A, Hirth C, Mirhom R, Rasaholinjanahary J. Course of pulmonary
hemodynamics in patients with chronic obstructive pulmonary disease. Chest 75: 656--662, 1979.
87. Stem RC, Borkat G, Hirschfeld SS, Boat TF, Matthews LW, LiebmanJ, Doershuk CF. Heart
88. Louridas G, Patakas D, Stavropoulos C. Left ventricular function in patients with chronic
obstructive pulmonary disease. Cardiology 67: 73--80, 1981.
89. Rice Dl, Kearney DJ, Awe RJ,Jenkins DE. Left ventricular function during respiratory failure.
Crit Care Med 5: 193--198, 1977.
90. Peter RH. Unpublished data, Duke University Catheterization Laboratory-personal commu-
nication.
92. Vallette H, Raffestin B, Lockhart A. Cardiac function in chronic bronchitis: Effects of pacing
and plasma expansion. Clin Sci 60: 371-375, 1981.
93. Nakazawa M, Marks RA, Isabel-Jones J, Jarmakani JM. Right and left ventricular volume
characteristics in children with pulmonary stenosis and intact ventricular septum. Circulation
53: 884-890, 1976.
94. Herbert WH, Yellin E. Left ventricular diastolic pressure elevation consequent to pulmonary
stenosis. Circulation 40: 887-892, 1969.
95. Bove AA, Santamore WP. Ventricular interdependence. Prog Cardiovasc Dis 23: 365-388, 1981.
96. Graham TP,JarmakaniJM, Canent RV. Left heart volume characteristics with a right ventricu-
lar volume overload. Total anomalous pulmonary venous connection and large atrial septal
defect. Circulation 45: 389-396, 1972.
97. Dexter L. Atrial septal defect. Br Heart J 18: 209-225, 1956.
98. Levin AR, Liebson PR, Ehlera KH, Diamani B. Assessment of left ventricular function in
secundum atrial septal defect: Evaluation by determination of volume, pressure and external
systolic time indices. Pec!iat Res 9: 894-899, 1975.
271
99. Flamm MD, Cohn KE, Hancock EW. Ventricular function in atrial septal defect. Am] Med
48: 286-294, 1970.
100. Popio KA, Gorlin R, Teichholz LE, Cohn PF, Bechtel D, Herman MV. Abnormalities of left
ventricular function and geometry in adults with an atrial septal defect. Am ] Cardiol 36:
302-308, 1975.
101. Matthay RA, Ellis]H ]r, Steele P. Effect of dextran loading on left ventricular performance
in chronic obstructive pulmonary disease. Am Heart] 92: 730-736, 1976.
102. Marshall RC, Berger H], Costin]C, Freedman GS, Wolberg], Cohen LS, Gottschalk A, Zaret
B. Assessment of cardiac performance with quantitative radionuclide angiocardiography. Circu-
lation 56: 820-829, 1977.
103. Olvey SK, Reduto LA, Stevens PM, Deaton W], Miller RR. First-pass radionuclide assessment
of right and left ventricular ejection fraction in chronic pulmonary disease. Chest 78: 4-9,1980.
104. Steele P, Ellis]H ]r, Van Dyke D, Sutton F, Creagh E, Davies H. Left ventricular ejection
fraction in severe chronic obstructive airways disease. Am] Med 59: 21-28, 1975.
105. Matthay RA, Berger H], Davies RA, Loke], Zaret BL, Gottschalk A. Effect of steady state
exercise on right and left ventricular performance in chronic obstructive pulmonary disease.
Chest 77: 303, 1980.
106. Berger H], Matthay RA, Loke], Marshall RC, Gottschalk A, Zaret B. Assessment of cardiac
performance with quantitative radionuclide angiocardiography: Right ventricular ejection frac-
tion with reference to fmdings in chronic obstructive pulmonary disease. Am] Cardiol 41:
897-905, 1978.
107. Berger H], Matthay RA. Noninvasive radiographic assessment of cardiovascular function in
acute and chronic respiratory failure. Am] Cardiol 47: 950-962, 1981.
108. Slutsky RA, Ackerman W, Karliner ]S, Ashburn WL, Moser KM. Right and left ventricular
dysfunction in patients with chronic obstructive lung disease. Assessment by first-pass radionu-
clide angiography. Am] Med 68: 197-205, 1980.
109. Slutsky R, Hooper W, Ackerman, Ashburn W, Gerber K, Moser K, Karliner J. Evaluation of
left ventricular function in chronic pulmonary disease by exercise gated equilibrium radionuclide
angiography. Am Heart] 101: 414-420, 1981.
110. Mathur PN, Powles P, Pugsley SO, McEwan MP, Campbell E]M. Effect of digoxin on right
ventricular function in severe chronic airflow obstruction. Ann Intern Med 95: 283-288, 1981.
111. Chipps BE, Alderson PO, Roland]A, Yang S, van Aswegen A, Martinez CR, Rosenstein B].
Noninvasive evaluation of ventricular function in cystic fibrosis.] Pediatr 95: 379-384, 1979.
112. Brinker ]A, Weiss JL, Lappe DL, Rabson ]L, Summer WR, Permutt S, Weisfeldt ML. Leftward
septal displacement during right ventricular loading in man. Circulation 61: 626-633, 1980.
113. Helak ]W, Reichek N. Quantitation of human left ventricular mass and volume by two-
dimensional echocardiography: In vitro anatomic validation. Circulation 63: 1398-1407, 1981.
114. Liboff M], St John Sutton M, Weber KT, ]aniciki ]S, Reichek N. The effect of chronic
pulmonary hypertension on left ventricular size and dynamics (abstract). Circulation 66: II-237,
1982.
115. Weissler AM, Harris WS, Schoenfeld CD. Bedside technics for the evaluation of ventricular
function in man. Am] Cardiol 23: 577-583, 1969.
116. Pigott V, Spodick DH. Effects of normal breathing and expiratory apnea on duration of the
phases of cardiac systole. Am Heart] 82: 786-793, 1971.
117. Unger K, Shaw D, Karliner ]S, Crawford M, O'Rourke RA, Moser KM. Evaluation of left
ventricular performance in acutely ill patients with chronic obstructive lung disease. Chest 68:
135-142, 1975.
118. Ahmed SS, Levinson GE, Schwartz CO, Ettinger PO. Systolic time intervals as measures of
the contractile state of the left ventricle in man. Circulation 46: 559-571, 1972.
119. Alpert ]S, Rickman FD, Howe JP, Dexter L, Dalen ]E. Alteration of systolic time intervals in
right ventricular failure. Circulation 50: 317-323. 1974.
120. Hooper GH, Whitcomb ME. Systolic time intervals in chronic obstructive pulmonary disease.
Circulation 50: 1205-1209, 1974.
121. Alpert], Crawford M, Karliner], O'Rourke RA. STI in COPD (letter to editor). Circulation
52: 354, 1975.
122. Bahler RC. Assessment of left ventricular function in chronic obstructive pulmonary disease.
Chest 68: 132-133, 1975.
10. ARRHYTHMIAS IN CHRONIC LUNG DISEASE
ERIC N. PRYSTOWSKY, M.D.

and
GEORGE J. KLEIN, M.D.
Patients who have pulmonary disorders, especially those with chronic obstructive
pulmonary disease, often have atrial and ventricular arrhythmias. However, the
pathogenesis of these arrhythmias is complex and may involve many factors. For
example, these individuals often have anatomic cardiac abnormalities, such as right
ventricular hypertrophy, with or without associated congestive heart failure. Ciga-
rette smoking is protean in patients with chronic obstructive pulmonary disease,
and, consequently, atherosclerotic heart disease may be present with its attendant
complications. Hypoxia and hypercapnia occur in these patients and, under appro-
priate circumstances, can be arrhythmogenic. Patients may receive drugs that can
cause arrhythmias. For example, diuretics may lead to hypokalemia, and amino-
phylline and adrenergic bronchodilator drugs may create the environment neces-
sary for arrhythmias to emerge. Finally, patients with pulmonary disease are prone
to any arrhythmias that occur in other people, for example, atrioventricular nodal
reentry and Wolff-Parkinson-White reentry.
We have divided this chapter into three parts. In the first part, we review
The authors greatly appreciate the secretarial assistance of Nancy S. Lineback. This work was supported in part
by the Hennan C. Krannert Fund, Indianapolis, Indiana; by Grants HL--06308 and HL--07812 from the National
Heart, Lung and Blood Institute of the National Institutes of Health, Bethesda, Maryland; by the Attorney General
of Indiana Public Health Trust and by the Roudebush Veterans Administration Medical Center, Indianapolis,
Indiana; and by a Grant-in-Aid from the American Heart Association, Indiana Affiliate, Inc., Indianapolis, Indiana.
L.J. Rubin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright © 1984 Martinus Nijh4JPubiishing. Boston/The Hague/Dor-
drecht/Lancaster.
273
274 10. Arrhythmias in Chronic Lung Disease
pathophysiology of arrhythmias in patients with pulmonary disease. The second

part of the chapter discusses the specific arrhythmias noted in these patients, and
the third section discusses therapy.
PATHOPHYSIOLOGY
Review of the literature demonstrates a very complex interaction between periph-
eral and central reflex mechanisms, as well as direct effects caused by hypoxia and
hypercapnia. It is very difficult to determine which factor or combination of factors
is responsible for the initiation of a particular arrhythmia in man. Thus, although
the following abnormalities are discussed under separate headings, one must re-
member that interactions between these factors may be highly significant for the
emergence of arrhythmias.
Reflex mechanisms
Activation of peripheral chemoreceptors results in responses that are highly inte-
grated with central nervous system activity (1). Other factors also influence the
responses to carotid body stimulation. For example, carotid body reflex cardioinhi-
bition is modulated by changes in breathing. Thus, the bradycardia due to activa-
tion of the peripheral chemoreceptors can be enhanced in the presence of apnea.
Therefore, any situation that results in apnea, combined with activation of the
cardioinhibitory reflexes, can potentially lead to serious clinical consequences such
as prolonged cardiac asystole (1). Such clinical conditions may include the Pick-
wickian syndrome, Ondine's curse, and the sleep apnea syndrome (1).
The sleep apnea syndrome is an interesting condition that often results in various
forms of arrhythmias, especially bradyarrhythmias (2). Zwillich and colleagues (3)
recently investigated the mechanisms which caused bradycardia during sleep apnea
in six patients. In general, periods of apnea were associated with sinus bradycardia,
although shorter episodes of apnea were accompanied by no or only minimal
bradycardia. Further, there appeared to be a direct relationship between the length
of the apneic spells and the degree of bradycardia. A corresponding relationship
between degrees of bradycardia and oxygen desaturation was also found. Supple-
mental oxygen appeared to abolish most of the bradycardia during apneic spells.
Slowed heart rate, also occurred in some normal patients who were evaluated
during periods of breath holding that resulted in oxygen desaturation (3). This
bradycardia was blocked by administration of atropine prior to breath hold.
Bradycardia with breath hold also was abolished when supplemental oxygen was
given. The authors concluded that the bradycardia of sleep apnea results from the
combined effect of cessation of breathing in the presence of hypoxemia.
Hypoxia
Research on responses of the intact animal to hypoxia has demonstrated various
results, depending on the species studied as well as the integrity of the nervous
system (4,5). For example, hypoxia produces an increase in heart rate and cardiac
275
output and may cause a decrease in blood pressure in man, but blood pressure often
is better maintained in the dog and cat (5). In the rabbit P0 2 greater than 35 mmHg
results in tachycardia, but more severe hypoxemia results in bradycardia (5). In
patients with chronic obstructive pulmonary disease, hypoxia most often is as-
sociated with other abnormalities, and arrhythmogenesis is probably multifactorial.
The following is a review of some effects of hypoxia on various cardiac tissues.
Samuelsson (6) evaluated the outcome of systemic hypoxia on action potential
duration and heart rate in dogs. Intracardiac suction electrodes were used to record
monophasic action potentials from the right atrium and right ventricle. Systemic
hypoxia was induced by exchanging the air in the ventilation system for 100%
nitrogen. When the nervous system was intact, hypoxia induced a pronounced
decrease in heart rate which was associated with a marked shortening of the atrial
monophasic action potential duration, but the ventricular action potential duration
increased with the slowing of heart rate as expected. These results were similar to
those obtained with vagal nerve stimulation. When cardiac pacing was employed
to keep heart rate constant during hypoxia, the atrial monophasic action potential
duration was decreased, but the duration of the ventricular monophasic action
potential was not changed. Parasympathetic denervation and hypoxia resulted in
an increased heart rate. The author concluded that severe systemic hypoxia resulted
in a substantial increase in parasympathetic as well as sympathetic activity in the
heart, but that parasympathetic activity predominated on the sinus node and atrium.
Studies in man have shown that the parasympathetic nervous system is prepotent
on the sinus node, but the effects of resting parasympathetic tone on ventricular
refractoriness is less pronounced (7). Additionally, enhanced parasympathetic tone
shortens atrial refractoriness in man (8).
The central sites responsible for the apparent vagal-induced bradycardia caused
by hypoxia were investigated in cats (9). The area mediating this effect appears to
be in the lateral mesencephalic reticular formation, for bilateral lesions in this area
either decreased the extent of bradycardia or caused reversal to tachycardia. Bilat-
erallesions in the medial mesencephalic reticular formation failed to alter brady-
cardia in cats.
The effects of hypoxia also have been evaluated in isolated rabbit sinus nodal
preparations using intracellular microelectrodes to record action potentials from
sinus nodal cells (10). These preparations are superfused with nutrient solutions
containing various amounts of oxygen. Induced hypoxia decreased the rate of
spontaneous sinus nodal impulse formation primarily by decreasing the slope of
diastolic depolarization. Further, the amplitude of the sinus nodal action potentials
was diminished and the conduction time from the atrium to the sinus node was
prolonged. The authors noted that these changes occurred fairly promptly.
Atrioventricular nodal conduction is also affected by hypoxia. In dogs subjected
to severe hypoxia during atrial pacing, second-degree AV nodal block occasionally
occurred (6). Investigations of isolated AV nodal preparations during hypoxia
show conflicting results (10-12). Senges et al. (10) exposed rabbit AV nodal cells
to a superfusate with P0 2 of 40 mmHg and noted a marked reduction in action
potential amplitude as well as a substantial decrease in the slope of diastolic

depolarization. AV nodal conduction time was prolonged at all pacing cycle
lengths as compared with control values, and AV nodal conduction block occurred
at slower pacing rates. Further, the effective refractory period of the AV node
significantly increased from 140 + 34 to 340 + 180 msec. Of note, hypoxia
affected AV nodal conduction relatively quickly, and changes were seen by 10
minutes.
The rapidity of hypoxia-induced changes in AV nodal conduction demonstrated
by Senges et al. (10) contrasted with results obtained from Kohlhardt and Haap
(11). These authors performed microelectrode studies on isolated rabbit AV nodal
preparations in the presence of hypoxic solution. They found marked effects on
AV nodal transmembrane action potential characteristics in the presence of hypoxia
-for example, a decrease in Vrnax • The effects occurred over a long period of time,
up to 90 minutes, which contrasted with a marked reduction in automaticity,
occurring within 20 minutes. Of interest, isoproterenol superfusion or the presence
of a high external glucose concentration in the superfusate countered the effects
of hypoxia.
Belardinelli and associates (12) investigated AV nodal conduction during
hypoxia in the guinea pig heart. The P0 2 of the superfusate was 19 to 25 mmHg.
After 5 minutes of hypoxia, the AH interval (approximate AV nodal conduction
time) prolonged, and second-degree AV block eventually developed. Aminophyl-
line added to the solution decreased the prolongation of conduction induced by
hypoxia.
Conflicting results were offered on the effects of hypoxia on the ventricle.
Rogers et al. (13) studied ventricular fibrillation threshold determinations in canine
ventricles during hypoxia. This methodology has many potential problems. It
consists of introducing a train of stimuli in the right ventricle using different
current levels and then timing the pulses to occur during the vulnerable period of
the cardiac cycle. These trains of impulses, initiated at progressively higher current
levels until ventricular fibrillation occurs, are done before and after interventions.
In their study (13) hypoxia was produced by ventilating the animal with low
oxygen mixtures. Their results demonstrated that hypoxia did not change the
amount of current necessary to produce ventricular fibrillation; that is, the ven-
tricular fibrillation threshold was the same before and after hypoxia. In contrast,
Murnaghan (14) studied the isolated rabbit heart and noted that hypoxia signifi-
cantly lowered the ventricular fibrillation threshold; that is, less current was needed
to induce ventricular fibrillation in the presence of hypoxia. The effect of anoxia
on the ventricular fibrillation threshold was not prevented with propranolol or
pindolol, or by previous treatment with reserpine (14). Differences in the results
between these two studies may be explained in part by experimental methodology.
Mauck and associates (15,16) examined the central nervous system's role in
arrhythmogenesis during hypoxia. When the mesencephalon of the cat was stimu-
lated during hypoxia (P0 2 40 mmHg), a significant increase in heart rate and
arterial pressure resulted (15). During stimulation and hypoxia, 11 of 20 cats
277
developed abnormal heart rhythms, which consisted of sinus pauses, multiformed

ventricular premature complexes, and salvos of either ventricular or junctional
tachycardia. When stimulation occurred in the presence of normal oxygenation,
heart rate and blood pressure increased but cardiac rhythm disturbances were not
noted. In the presence of propranolol and hypoxia, stimulation of the mesencepha-
lon elicited no changes in heart rate (15).
A subsequent study from Mauck et al. (16) investigated the interactions of the
autonomic nervous system and hypoxia on mesencephalic stimulation in cats.
During ventilation with room air mesencephalic stimulation produced a significant
increase in heart rate and arterial pressure, but cardiac arrhythmias were not
observed. During combined hypoxia and hypercapnia, stimulation of the midbrain
evoked multiple cardiac arrhythmias, such as sinus depression, supraventricular
tachycardia, and ventricular tachycardia. Administration of propranolol abolished
the cardiac arrhythmias induced with mesencephalic stimulation. These authors
(15,16) concluded that acute hypoxia was the major determinant of arrhythmo-
genesis, but the initiation of arrhythmias was markedly associated with enhanced
sympathetic tone.
Acidosis and alkalosis

Hypoxia without any disturbance of acid-base balance is unusual in pulmonary
disorders. Thus, quite possibly many arrhythmias are a combination of oxygen lack
as well as acid-base abnormalities. Rogers et al. (13) investigated the effects of
metabolic and respiratory acidosis and alkalosis on ventricular fibrillation threshold
in dogs. Metabolic acidosis decreased the ventricular fibrillation threshold, but
respiratory acidosis caused no change in the ventricular fibrillation threshold.
Interestingly, respiratory acidosis in the presence of hypoxia increased the ventricu-
lar fibrillation threshold above control; that is, ventricular fibrillation was harder
to initiate under these experimental conditions. Metabolic alkalosis increased the
ventricular fibrillation threshold. These data suggest that respiratory acidosis in
association with hypoxia may protect the canine ventricle from ventricular fibrilla-
tion-rather surprising results. Obviously, these data do not suggest that one
should abandon the basic tenets of treating patients who have pulmonary disease
and arrhythmias, that is, correct the underlying metabolic and respiratory abnor-
malities. Rather, these results demonstrate that much more research is needed to
define the precise cause or causes of arrhythmias in the complex pulmonary patient.
ARRHYTHMIAS
Chronic obstructive pulmonary disease

Patients with chronic obstructive pulmonary disease (COPD) constitute a complex,
generally older population with a high prevalence of associated medical and,
specifically, heart disease (17-19). Not surprisingly, they manifest the whole spec-
trum of cardiac arrhythmias, including tachycardias and bradycardias (19-21). No
arrhythmias can be considered specific for COPD, and the diversity of arrhythmias
~
,
I
,
I ,
I I
,
III
I ! ---:
I
,
,
I ,
I
10 " ~'~-- ,
.Jj~ '._ b-.
..j.A.L d ~AiJ;K ,'7!\;-,.-' .~ I
~. ~ ~m~.'·
I
.. ,
I
~!+'t+~1"~i ;YT~ I
++11+
I
, + h J'F.~J ~-H
fl. -"'~k :d"'}:. ~VF'
lrf
i !
V
I
!
I
,
'tIL~: I
v,·. ,L~iLttu
i f-
i I -LL I
Figure 10-1. Atrial tachycardia. This arrhythmia occurred sporadically in a 50-year-old woman
with chronic lung disease. The atrial rate is approximately 160 beats per minute and the
ventricular response is variable ("PAT with block"). Note that atrial tachycardia persists in the
presence of AV block, proving that the ventricle is not required in the perpetuation of the
arrhythmia.
observed probably reflects their multifactorial pathogenesis (20). The prevalence

of detected arrhythmias approaches 100% with prolonged ECG monitoring (19-
21).
Atrial arrhythmias have been found in 52 to 65% of patients using ECG
recordings of 10 to 72 hours (19,21). The majority of patients have atrial extrasys-
toles while fewer have atrial fibrillation, atrial flutter, or other forms of supraven-
tricular tachycardia. The published reports have not attempted to clarify the
mechanism of supraventricular tachycardia. We and others (22) have observed that
supraventricular tachycardia in patients with lung disease is generally ectopic atrial
tachycardia as opposed to atrioventricular or AV nodal reentry. That is, atrial
tachycardia persists in the presence of AV nodal block, showing that the AV node
most likely is not needed in the tachycardia circuit (figure 10-1). It is not known
whether such tachycardias result from atrial reentry, triggered activity, or abnormal
automaticity.
Ventricular arrhythmias have been documented in approximately 70% of pa-
tients (19,21). Ventricular ectopic activity most frequently is observed, while
ventricular tachycardia (three or more consecutive PVCs) occurs in approximately
10% of patients. Kleiger et al. (21) noted that ventricular arrhythmias in stable
COPD were similar in frequency and severity to those observed in a control group
279
2:50 HI< 10'+
Figure 10-2. Multifocal atrial tachycardia. This arrhythmia was documented on ambulatory
monitoring in a 45-year-old female with no apparent medical disease. The ventricular response is
irregular, and at least three P wave morphologies are observed.
of patients recovering from acute myocardial infarction, and they suggested that
these arrhythmias might portend a poor prognosis. Diener and Burrows (23) noted
that 8% of 200 patients with COPD died unexpectedly over long-term follow-up,
but could not determine whether these deaths were due to bradycardia, tachycardia,
or neither. In truth, the prognostic significance of ventricular ectopy in stable
COPD is not known, and possible modification of mortality with antiarrhythmic
drug therapy remains speculative.
Bradycardia has been observed less frequently in this patient group. Five of
thirty-five patients studied by Holford and Mithoefer (19) had "significant" sinus
arrest, junctional bradycardia, or Mobitz II AV block, whereas other investigators
(20,21) examining ambulatory records of patients with COPD did not emphasize
the occurrence of these arrhythmias.
Acute respiratory failure in COPD

Atrial or ventricular arrhythmias were observed during ambulatory ECG monitor-
ing in 94% of a population of patients with COPD and acute respiratory failure
(19). The frequency of atrial and ventricular arrhythmias in this group is under-
scored by the observation of arrhythmias recorded on a single 12-lead ECG in
approximately 50% of COPD patients presenting with acute respiratory failure
(24,25). The spectrum of arrhythmias in this setting is probably similar to that of
stable COPD. However, multifocal atrial tachycardia (MAT), a relatively uncom-
mon arrhythmia, is observed most frequently in the setting of acute respiratory
failure (26-28). Diagnostic criteria for this arrhythmia (figure 10-2) include the
following:
a. Heart rate greater than l00/min.

b. Three or more discrete P wave morphologies with an isoelectric baseline.
c. Irregular P-P intervals.
MAT also can be observed in congestive heart failure, in the critically ill patient
from a variety of causes, and even in patients without apparent associated disease.
MAT can be short-lived (seconds) or sustained for months and can terminate
abruptly or by gradual slowing. The arrhythmia is usually associated with a poor
prognosis in respiratory failure, but is probably a "marker" for severe cardiopulmo-
nary disease rather than a significant factor in the patient's demise.
The presence of ventricular arrhythmias on a routine ECG is associated with
high in-hospital mortality (70%) and long-term mortality (100% over 2Y2 years)
in COPD patients with acute respiratory failure (25). However, the clinical signifi-
cance of these arrhythmias other than possibly as a clinical marker is not known.
For example, only one of sixteen patients suffering a witnessed cardiac arrest in an
intensive care setting (five with documented ventricular fibrillation) could be
resuscitated (25). The latter suggests to us that ventricular fibrillation is the culmi-
nation of end-stage cardiorespiratory disease and not a primary arrhythmia that
could be prevented or reversed with antiarrhythmic medication.
Arrhythmias in other respiratory disease
Bronchial asthma
Arrhythmias other than sinus tachycardia in acute bronchial asthma are rare (29-31)
and generally confmed to simple atrial or ventricular ectopy. Moderate use of
epinephrine and/or aminophylline does not have a significant impact on arr-
hythmias (31,32), although an "epidemic" of sudden deaths was observed previ-
ously in asthmatic patients using potent beta-adrenergic aerosols, which are no
longer available (33). Mortality in acute asthma is related to the severity of the
acute obstruction and not to arrhythmias (29-31).
Pulmonary embolism
Unexplained atrial arrhythmias, especially atrial flutter, have been reported to be
a frequent accompaniment of recurrent pulmonary embolism (34,35), whereas
ventricular arrhythmias are infrequently seen (35). We have recently observed a
patient presenting with unexplained syncope that was shown to be related to
ventricular tachycardia (figure 10-3). This patient subsequently was documented
to have recurrent pulmonary embolism without organic heart disease. Ventricular
tachycardia did not recur after appropriate therapy for pulmonary embolism.
Arrhythmias probably have poor specificity and sensitivity for the diagnosis of
pulmonary embolism, but the latter diagnosis should be considered for any unex-
plained arrhythmias, especially atrial flutter.
Primary pulmonary hypertension

Little data are available on the contribution of arrhythmias to mortality in this
syndrome. Kanemoto (36) reviewed the electrocardiograms of 101 patients with
primary pulmonary hypertension and found no arrhythmias. Sinus tachycardia
occurred in 13 patients and was an ominous sign.
281
LL
mn-floor'1I ~ .n
..H I
. "',...
10 11 14 "IU lDI - •• ) III .:1"
r 11
I
~l ~ II ! 1 1 II lilil J.
III . 1m!
I
! ~
Ij
I I
: i
.
._. - --
M
Figure 10-3. Ventricular tachycardia. This arrhythmia was recorded during a presyncopal spell
in a 60-year-old man with recurrent pulmonary embolism. The tachycardia rate exceeds 300
beats per minute. It is remotely possible that this arrhythmia represents atrial Hutter with 1 to 1
conduction and aberration.
Mechanical ventilation
Osborn et al. (37) reviewed arrhythmic complications occurring in 150 monitored
patients with a variety of illnesses undergoing ventilation postthoracotomy. Eight
patients had malignant ventricular arrhythmias, that is, ventricular tachycardia or
fibrillation, and significant hypercarbia was present in five of these patients. Wasser-
man (38) also emphasized the importance of alkalosis and acidosis in the genesis
of arrhythmias in patients receiving assisted ventilation therapy.
THERAPY
The most important therapeutic goal in patients with arrhythmias and pulmonary
disease is to correct any underlying disturbances in oxygenation or electrolyte and
acid-base balance. Often, these measures alone will alter the arrhythmogenic milieu
and resotre normal sinus rhythm. If arrhythmias persist despite optimal treatment
of the underlying pulmonary disorder, specific antiarrhythmic therapy may be
necessary, and, with few exceptions, the decision to treat, as well as the choice of
antiarrhythmic drug, is guided by the same principles used to treat patients without
pulmonary disease.
Supraventricular tachycardia can be classified into two broad categories-ar-
rhythmias in which AV nodal conduction is necessary for maintenance of the
tachycardia, notably AV nodal reentry and WPW reentry, and arrhythmias that
do not require AV nodal conduction for maintenance of tachycardia but utilize
the AV node to conduct impulses to the ventricle, for example atrial flutter, atrial
fibrillation, or multifocal atrial tachycardia. Drugs such as verapamil, digitalis, and
beta blockers impair AV nodal conduction and may terminate or prevent the
occurrence of AV nodal and WPW reentry. These same drugs usually will not
terminate the latter group of arrhythmias, but are useful to block AV nodal
conduction of some of the atrial impulses and thereby decrease the ventricular rate.
Verapamil is a very effective agent for both groups of supraventricular tachycardia
(39,40) and is probably the drug of choice in patients with COPD, assuming no
contraindications for its use are present. Digitalis should be used with caution
because patients with severe pulmonary disease may have an increased sensitivity
to digitalis (41), and beta blockers should be avoided whenever possible. Prevention
of either group of supraventricular arrhythmias often is difficult and may require
treatment with quinidine, disopyramide, or procainamide.
Treatment of patients who have ventricular arrhythmias is controversial. Al-
though most cardiologists would agree that patients with a history of sustained
ventricular tachycardia or cardiac arrest require long-term antiarrhythmic therapy,
little data are available regarding the necessity to treat patients with nonsustained
ventricular tachycardia or only premature ventricular complexes. If these latter
arrhythmias cause substantial symptoms, attempts to suppress them seem justified.
However, antiarrhythmic drugs, especially quinidine, can be arrhythmogenic and
result in life-threatening ventricular arrhythmias. Further, prospective intervention
trials to demonstrate that therapy prevents sudden death in these patients are not
available. Thus, at present we do not recommend antiarrhythmic drug treatment
for asymptomatic patients who have only premature ventricular complexes. Ther-
apy may be useful for patients with nonsustained ventricular tachycardia and should
be considered if the runs of tachycardia are long and/or rapid or if other factors
such as substantial myocardial dysfunction are present.
REFERENCES
1. Daly M, Angell-James JE, Elsner R. Role of carotid-body chemoreceptors and their reflex
interactions in bradycardia and cardiac arrest. Lancet 1: 764-767, 1979.
2. Guillerninault C, Connally SJ, Winkle RA. Cardiac arrhythmia and conduction disturbances
during sleep in 400 patients with sleep apnea syndrome. Am J Cardiol 52: 490-494, 1983.
3. Zwillich C, Devlin T, White D, Douglas N, Weill, Martin R. Bradycardia during sleep apnea
-characteristics and mechanism. J Clin Invest 69: 1286-1292, 1982.
4. Coleridge JCG, Coleridge HM. Chemoreflex regulation of the heart. In Handbook of Physiol-
ogy, Section 2: The Cardiovascular System, Berne RM, Sperelakis N, Geiger SR (eds). Bethesda,
Md: American Physiological Society, 1979, pp. 653-676.
5. Komer PI. Central nervous control of autonomic cardiovascular function. In Handbook of
Physiology, Section 2: The Cardiovascular System, Berne RM, Sperelakis N, Geiger SR (eds).
Bethesda, Md: American Physiological Society, 1979, pp. 691-739.
6. Samuelsson R. Effects of severe systemic hypoxia on myocardial excitation. Acta Physiol Scand
88: 267-280, 1973.
7. Prystowsky EN, Jackman WM, Rinkenberger RL, Heger], Zipes DP. Effect of autonomic
blockade on ventricular refractoriness and atrioventricular nodal conduction in man. Evidence
supporting a direct cholinergic action on ventricular muscle refractoriness. Circ Res 49: 511-518,
1981.
8. Prystowsky EN, Naccarelli GV,Jackman WM, Rinkenberger RL, Heger], Zipes DP. Enhanced
parasympathetic tone shortens atrial refractoriness in man. Am J Cardiol 51: 96-100, 1983.
9. Hall RE, Rubinstein EH, Sonnenschein RR. Abolition of hypoxic vagal bradycardia by lateral
mesencephalic lesions in spinal cats. Am J Physiol 237: R15-R19, 1979.
10. SengesJ, Mizutani T, Pelzer D, BrachmannJ, SonnhofU, Kubler W. Effect of hypoxia on the
sinoatrial node, atrium, and atrioventricular node in the rabbit heart. Circ Res 44: 856-863, 1979.
11. Kohlhardt M, Haap K. The influence of hypoxia and metabolic inhibitors on the excitation
process in atrioventricular node. Basic Res Cardiol 75: 712-727, 1980.
12. Belardinelli L, Belloni FL, Rubio R, Berne RM. Atrioventricular conduction disturbances
during hypoxia-Possible role of adenosine in rabbit and guinea pig heart. Circ Res 47: 684-691,
1980.
13. Rogers RM, Spear]F, Moore EN, Horowitz LH, Sonne JE. Vulnerability of canine ventricle
to fibrillation during hypoxia and respiratory acidosis. Chest 63: 986-994, 1973.
14. Murnaghan MF. The effect of anoxia on the ventricular fibrillation threshold in the rabit isolated
heart. Br J Pharmacol 54: 41~20, 1975.
283
15. Mauck HP, Szumski AJ, Fu TC, Forbes JE, Clendenin MA. The effect of acute hypoxia on
cardiac dysrhythmias induced by mesencephalic stimulation. Am Heart J 87: 197-202, 1974.
16. Mauck HP Jr, Szumski AJ, Forbes JE, Clendenin MA, Newton RA. Effects of hypercapnea and
of hypercapnea in combination with hypoxia on midbrain-induced cardiac dysrhythmias. Am
Heart J 92: 315-323, 1976.
17. Steele P, EllisJH, Van Dyke D, Sutton F, Creagh E, Davies H. Left ventricular ejection fraction
in severe chronic obstructive airways disease. Am J Med 59: 21-28, 1975.
18. Kachel RG. Coronary arterial disease and arrhythmias in chronic obstructive lung disease. Chest
76: 242, 1979.
19. Holford FD, Mithoefer JC. Cardiac arrhythmias in hospitalized patients with chronic obstructive
pulmonary disease. Am Rev Respir Dis 108: 879--885, 1973. .
20. Biggs FD, Lefrak SS, Kleiger RE, Senior RM, Oliver GC. Disturbances of rhythm in chronic
lung disease. Heart Lung 6: 256--261, 1977.
21. Kleiger RE, Senior RM. Long-tenn electrocardiographic monitoring of ambulatory patients
with chronic airway obstruction. Chest 65: 483-487, 1974.
22. Goldberg LM, Bristow]D, Parker BM, et al. Paroxysmal atrial tachycardia with atrioventricular
block. Its frequent association with chronic pulmonary disease. Circulation 21: 499, 1960.
23. Diener CF, Burrows B. Further observations on the course and prognosis of chronic obstructive
lung disease. Am Rev Respir Dis III: 719-724, 1975.
24. Khokhar N. Cardiac arrhythmias associated with acute respiratory failure in chronic obstructive
pulmonary disease. Milit Med 146: 856--858, 1981.
25. Hudson LD, Kurt TL, Petty TL, Genton E. Arrhythmias associated with acute respiratory failure
in patients with chronic airway disease. Chest 63: 661-665, 1973.
26. Shine KI, Kastor JA, Yurchak PM. Multifocal atrial tachycardia. Clinical and electrocardio-
graphic features in 32 patients. N Engl J Med 279: 344-349, 1968.
27. Phillips J, Spano J, Burch G. Chaotic atrial mechanism. Am Heart J 78: 171-179, 1969.
28. Lipson MJ, Naimi S: Multifocal atrial tachycardia (chaotic atrial tachycardia). Circulation 42:
397-407, 1970.
29. Crompton GK, Grant IWB. Observations on the management of acute bronchial asthma. Br J
Dis Chest 73: 157-163, 1979.
30. Grossman J. The occurrence of arrhythmias in hospitalized asthmatic patients. J Allergy Clin
Immunol 57: 310--317, 1976.
31. Josephson GW, Kennedy HL, MacKenzie EJ, Gibson G. Cardiac arrhythmias during the treat-
ment of acute asthma. A comparison of two treatment regimens by a double blind protocol. Chest
78: 429-435, 1980.
32. Assael R, Martr JM, Okeson GC. Frequency of isoproterenol hydrochloride-induced cardiac
arrhythmia in 19 patients with chronic obstructive pulmonary disease: A prospective study. Am
Rev Respir Dis 111: 743-747, 1975.
33. Stolley PD. Asthma mortality. Why the United States was spared an epidemic of deaths due to
asthma. Am Rev Respir Dis 105: 883-885, 1972.
34. Johnson JC, Flowers NC, Horan LG. Unexplained atrial Hutter: A frequent herald of pulmonary
embolism. Chest 60: 29-34, 1971.
35. Brown AK, Newton P, Hamilton EAG, Anderson V. Recurrent pulmonary thromboembolism
presenting with cardiac arrhythmias. Thorax 34: 380--383, 1979.
36. Kanemoto N, Sasamoto H. Arrhythmias in primary pulmonary hypertension. Jpn Heart J 20:
765-775, 1979.
37. Osborn], RaisonJCA, BeaumontJO, HilIJ?' Kerth WJ, Popper RW, Gerbode F. Respiratory
causes of "sudden unexplained arrhythmia' in postthoracotomy patients. Surgery 69: 24-28,
1971.
38. Wassennan K. Cardiovascular manifestations of respiratory insufficiency. Clin Notes Respir Dis
12: 3-10, 1973.
39. Rinkenberger RL, Prystowsky EN, Heger], Troup PJ, Jackman WM, Zipes DP. Effect of IV
and chronic oral verapamil administration in patients with a variety of supraventricular tachyarr-
hythmias. Circulation 62: 996--1010, 1980.
40. Klein GJ, Gulamhusein S, Prystowsky EN, Carruthers SG, Donner AP, Ko PT. Comparison of
the electrophysiological effects of intravenous and oral verapamil in patients with paroxysmal
supraventricular tachycardia. Am J Cardiol 49: 117-124, 1982.
41. Green LJ, Smith TW. The use of digitalis in patients with pulmonary disease. Ann Intern Med
87: 459-465, 1977.
11. ACUTE COR PULMONALE
WARREN R. SUMMER, M.D.
"Cor pulmonale" refers to heart disease caused by diseases that involve primarily
the lung or its vascular supply ("cor" meaning heart and "pulmonale," lungs). The
most widely accepted definition of cor pulmonale is the one suggested by the
WHO Technical Series report in 1961 (1). This definition, however, raises an
important question: Is cor pulmonale a clinical or morphological diagnosis? For
example, when evidence of pulmonary hypertension is present but actual documen-
tation of ventricular hypertrophy is impossible, can one consider the patient to be
suffering from cor pulmonale? Autopsies often show a predominance of right
ventricular dilation with little, if any, hypertrophy. Do these patients have cor
pulmonale? For all practical purposes, the answer to this question is yes. Since, from
a clinical viewpoint, right ventricular hypertrophy or dilatation is difficult to
document, any proven sustained elevation in pulmonary artery pressure may be
considered evidence for chronic cor pulmonale. For this chapter we will consider
any acute elevation in pulmonary artery pressure or pulmonary vascular resistance,
with or without documented right ventricular hypertrophy or an increased ven-
tricular volume, to be indicative of acute cor pulmonale.
PATHOGENESIS
The pathogenesis of acute cor pulmonale usually begins when a chronic pulmonary
disease produces a reduced capacity of the pulmonary vascular bed, leading to an
L.J. Ruhin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright © 1984 Martinus NijhofJpuhlishing. Boston/The Hague/Dor-
drecht/Lancaster.
285
286 11. Acute Cor Pulmonale
PATHOGENESIS OF COR PULMONALE
~ CHRONIC L!G DISEASE
~ ~
~ ~ :~I:~RS~::N~~
"'UC"ON ON
HY\" PULMONARY lVASCULAR BED
"LVCYTH'M.. ANO /
HYPERVISOSITY _ _ _ _ .. / '
~ PULMONARY HYPERTENSION
~
HYPERTROPHY AND DILATATION
OF THE RIGHT VENTRICLE
~
RIGHT VENTRICULAR FAILURE
Figure 11-1. Pathogenesis of cor pulmonale.
increased resistance to blood flow and pulmonary hypertension. This process can
be best documented by shifts in pressure flow relationships observed with at least
two levels of cardiac output (2). The increased pressure load on the right ventricle
eventually leads to myocardial hypertrophy and right ventricular failure (figure
11-1). The most common disease associated with these changes is chronic obstruc-
tive pulmonary disease (COPD). The mechanism responsible for this effect in
COPD has been shown to be due predominantly to alveolar hypoxia (3). Exactly
how hypoxia produces the pulmonary vasoconstriction is not known, but is
thought to be sensed directly, through locally reduced alveolar oxygen tension
(4,5).
Hypoxic pulmonary vasoconstriction may be a direct effect on smooth muscle
contraction (4-6) or may be mediated through the stimulation of chemoreceptors
(7,8) and/or local release of vasoconstricting mediators such as catecholamines,
histamine, and arachidonic acid metabolites. The release of undefmed substances
associated with mast cell degranulation has also been implicated in hypoxic vaso-
constriction (4,5,9-11). An alternative hypothesis to explain the relationship be-
tween hypoxia and pulmonary vasoconstriction suggests that normoxia or pulsatile
flow produces pulmonary vascular relaxation through the enzymatic or mechanical
release of vasodilators (12-15), and removal of these influences during hypoxia
results in vasoconstriction. Although no unified hypothesis exists to clarify ade-
quately the mechanisms responsible for hypoxic pulmonary hypertension, the
continued effort to identify agents or circumstances that attenuate or augment
hypoxic vasoconstriction is important in order to fmd a means for treating patients
287
with acute or chronic cor pulmonale. Regardless of the intermediary steps leading
to vascular muscular contraction, recent studies suggest that calcium ions play an
integral role in acute hypoxic vasoconstriction (6,16).
Prolonged pulmonary vasoconstriction results in muscularization of the small
arterioles (17). A significant degree of muscularization and, therefore, sustained
pulmonary hypertension takes a significant period of time either to develop or to
recede. Apparently, however, tremendous individual variability is found in pulmo-
nary vasoreactivity. A survey of people living at high altitude shows that while
some may have a mean pulmonary arterial pressure less than 30 mmHg, others have
pressures greater than 60 mmHg (normal PA pressure at sea level is 20 mmHg).
Most individuals with an elevated PA pressure are clinically "healthy" and are able
to take part in athletic events. Clinical studies of patients with cor pulmonale have
demonstrated an inverse correlation between the arterial P02 and PA pressure
(18,19). Similarly, when one plots P a0 2 against right ventricular diameter or fiber
thickness in animals with experimentally induced cor pulmonale, the curve is
identical to the plot of Pa0 2 versus PA pressure (20). This correlation, however,
is never very tight because of the wide individual variations. Another factor
influencing the less than tight inverse correlation between P a0 2 and PA pressure
is the problem of sampling error. A random arterial P a0 2 does not reflect the total
hypoxic experience of patients. Flick et al. (21) have shown that the PP2 varies
with the level of activity and may fall to low levels during sleep.
Pulmonary artery pressure seems to have a bimodal distribution among hypoxic
subjects. Lindsay and Read (22) found that approximately half the population falls
into a group that responds to hypoxia with acute vasoconstriction. If these subjects
have COPD, their prognosis is worse than that of patients who have similar degrees
of airflow limitation and are poor responders to hypoxia. The implied assumption
is that those responders who develop COPD are more likely to develop cor
pulmonale and have a worse outcome. Thus, cor pulmonale complicates COPD
only in patients who are "significant responders" to alveolar hypoxia, through its
tendency to produce vasoconstriction, muscularization of pulmonary arterioles, and
pulmonary hypertension. Once muscularization occurs in the pulmonary arteries
with increased pulmonary arteriolar resistance (a narrowed vascular bed), further
acute reductions in oxygen tension associated with such events as acute respiratory
failure result in severe pulmonary hypertension (23), right ventricular strain, and
possibly failure.
Recent studies have suggested that acute hypoxemia may produce an increased
pulmonary artery pressure by a Starling resistorlike effect at the level of small
vessels (24). This result has been demonstrated in an isolated lung model, in which
acute hypoxia produced an increase in critical opening pressure and a parallel shift
in fluid filtration. A parallel association between hypoxic vasoconstriction and
increased fluid filtration (lung leak) at any level of blood flow suggests that acute
hypoxic constriction occurs downstream from fluid exchanging vessels. Although
this downstream acute hypoxic vasoconstrictive effect appears well documented in
an isolated lung preparation, clinical evidence of acute pulmonary edema following
sudden hypoxia has been limited to studies evaluating individuals at high altitude.
An increased fluid filtration in most cases of acute pulmonary hypertension may
be prevented by the prior muscularization of pulmonary arterioles and increased
resistance in the upstream vessels. High-altitude pulmonary edema may occur after
return to altitude because of recent upstream vasodilation resulting from a period
of normoxia.
The pathogenesis of other forms of pulmonary hypertension has been less well
studied. Vascular obstruction is the major cause of the reduced pulmonary bed
following thromboembolism, but some element of vasoconstriction may contribute
to both acute and chronic occlusive states (25,26). The effect of new emboli on
pulmonary vascular resistance is different in patients with normal lungs compared
to those with previously injured lungs (27).
Secondary vasoactive factors could have a profound effect when the pulmonary
vascular bed has already been narrowed by preexisting derangements. In the adult
respiratory distress syndrome (ARDS), an increased pulmonary artery pressure may
result from perivascular edema, release of mediators, thromboembolism, or hypoxia
(28-32). Pulmonary artery pressure can be elevated relative to atmospheric pressure
because of respiratory therapy interventions that increase pleural pressure. In addi-
tion, transmission of an increased alveolar pressure to the pulmonary capillaries
with the use of positive end-expiratory ventilation (PEEP) may result in elevations
in pulmonary arterial transmural pressure.
Vascular obliteration frequently occurs in sarcoidosis, scleroderma, idiopathic
fibrosis, or pneumoconiosis although recent data suggest that vasoconstriction may
also playa role in some patients (32,33). In scleroderma the performance of
thermodilution cardiac output determinations with iced injections has been as-
sociated with acute pulmonary vasoconstriction (34), presumably by a pulmonary
vascular Raynaud's phenomenonlike effect. Primary pulmonary hypertension is
one of the most interesting diseases producing a progressive reduction of the
pulmonary arterial bed. Various hypotheses for the increase in pulmonary vascular
resistance have been suggested although no single factor has been isolated (35).
Acute exacerbations of pulmonary hypertension, possibly due to vasoconstriction,
have been documented in these patients after spontaneous withdrawal of vasodila-
tor drugs.
CONDITIONS ASSOCIATED WITH ACUTE COR PULMONALE

Pulmonary embolism is the most common event that leads to acute cor pulmonale
in an otherwise normal pulmonary vascular bed. Another cause of acute cor
pulmonale in a previously normal individual is ARDS. This syndrome may be
associated with either a normal or elevated pulmonary artery pressure (36,37). In
most other conditions associated with acute cor pulmonale, a gradual loss of
pulmonary vasculature and the development of arteriolar muscular hypertrophy
have occurred. Following a new additional insult, which would be minor in an
otherwise normal individual, a marked increase in pulmonary vascular resistance
289
Table 11-1. Etiological conditions associated with acute cor pulmonale
A. Major disease categories associated with acute hypoxic pulmonary vasoconstriction

1. Asthma
2. Bronchiectasis
3. Chronic obstructive pulmonary disease
4. Mountain sickness
5. Cystic fibrosis
6. Kyphoscoliosis
7. Neuromuscular disease
8. Obesity hypoventilation syndrome
9. Pleuropulmonary fibrosis
10. Fixed upper airways obstruction
B. Major disease categories producing anatomic obstruction and/or reduction of pulmonary
vascular bed
1. Pulmonary embolism
Thrombotic: acute massive, usually multiple, symptomatic recurrent, rare multiple,
small, recurrent
Other: tumor, fat, bone marrow, ova, Gaucher's cells, foreign body
2. Adult respiratory distress syndrome (ARDS) (various causes)
3. Pulmonary fibrosis (many diseases)
4. Pulmonary Iymphangitic spread of carcinoma
5. Primary pulmonary hypertension
6. Pulmonary hypertension due to predominant vascular disease in which conditions usually
produce interstitial fibrosis
-Progressive systemic sclerosis (CREST)
-Sarcoidosis
-Intravenous injection of medication intended for oral use
7. Vasculitis
8. Disease categories producing loss of pulmonary veins
-Mediastinal fibrosis
-Multicentric fibrosclerosis
-Idiopathic veno-occlusive disease
-Tumor compressing veins
as well as right ventricular decompensation may occur. The additional insult is

frequently an exacerbation of the same process that produced the primary rise in
vascular resistance. The most common conditions resulting in clinically significant
pulmonary hypertension and, therefore, acute cor pulmonale are listed in table
11-1. Conditions associated with hypoxic pulmonary vasoconstriction are the most
frequently recognized. Episodes of acute cor pulmonale occurring in patients with
preexisting vascular obstruction or vascular obliteration are often due to infection,
worsening hypoxemia, fluid overload, or small degrees of progression of the
underlying disease. Recognition of the effects of all contributing factors is impor-
tant to treat the patient appropriately and to set expectations in regard to potential
patient improvement.
A number of diseases may result in loss of pulmonary veins (table 11-1). These
conditions usually present clinically as unexplained pulmonary hypertension with
progressive shortness of breath (38,39). An acute episode of cor pulmonale, how-
ever, is frequently the event that results in a defmitive workup and ultimate
diagnosis. Unfortunately, the same acute event that leads to the clinical recognition
of the diagnosis of pulmonary veno-occlusive hypertension often results in the
death of the patient.
CLINICAL FEATURES
Presenting symptoms
Right-sided heart failure is the third most common cause of heart disease and
accounts for approximately 10 to 30% of hospital admissions for heart failure (40).
Acute cor pulmonale, with or without overt right-sided failure, usually occurs in
the setting of chronic cor pulmonale and is often precipitated by acute respiratory
failure (ARF). It is therefore important to remember that ARF may develop from
a host of physiologic derangements that result from numerous pathogenic events
(41). Thus, the clinical presentation of acute cor pulmonale is markedly influenced
by the underlying disease that results in ARF, and these respiratory events usually
overshadow the cardiovascular consequences.
A background of chronic heart, lung, musculoskeletal, or neuromuscular disease
is usually present. Patients with COPD usually have a history of cough, sputum
production, heavy cigarette consumption, shortness of breath, wheezing, and occa-
sionally ankle edema. Nonspecific chest pain is common. Acute pleuritic chest pain
suggests a pneumothorax, pneumonia, or pulmonary embolism as the cause of the
acute cardiopulmonary decompensation. Status asthmaticus may result in acute cor
pulmonale, but rapidly progressive respiratory failure is usually the life-threatening
process in this setting. A history of previous episodes of respiratory failure and the
prior use of therapeutic interventions, such as assisted ventilation or bronchodila-
tors, often indicate the etiology of ARF and help guide future therapeutic deci-
sions. The overall level of consciousness and ease of arousal may help differentiate
between primary respiratory failure or primary CNS events and secondary respira-
tory insufficiency. Sedative drug overdose, a common cause of ARF, is rarely
associated with significant acute cor pulmonale. Most neuromuscular diseases that
result in pulmonary heart disease have a protracted period of weakness prior to the
development of respiratory failure. The rapidity of onset of symptoms is very
helpful in considering the presence of pulmonary embolism, but sudden dyspnea,
in itself, is far from diagnostic of thromboemboli. In addition, most patients with
cor pulmonale secondary to thromboemboli probably have several weeks or
months of recurrent embolic events prior to the event that produced clinically
obvious heart failure. The majority of patients with massive emboli die prior to
hospitalization or within the first hour of embolism. Those lasting long enough
to reach medical attention are usually severely dyspneic and tachypneic, markedly
agitated, and in a shock state.
Patients with ARDS and significant cor pulmonale have symptoms very similar
to patients without substantially elevated pulmonary artery pressures. Most patients
with ARDS and a high initial pulmonary vascular resistance are extremely hypoxic
and require aggressive respiratory support. Patients with progressive or persistent
291
pulmonary hypertension are often those with numerous secondary complications.

The overall mortality is high among these patients, but symptoms directly attribut-
able to cor pulmonale are difficult to identify (42,43).
Physical examination
Isolated, acute cor pulmonale produces numerous physical fmdings that make the
diagnosis relatively obvious. The best prototype for such an event is massive
pulmonary embolism (44,45), with marked hyperpnea with a respiratory rate of
up to 40 per minute and a tidal volume as high as 1 liter. Such an extreme situation,
due to marked loss of lung perfusion, results in huge increases in physiologic dead
space. Heart rate is also usually very rapid, and the pulse pressure is narrowed.
Pulsus paradoxus may be present in any patient with hyperpnea secondary to
respiratory or metabolic etiologies (46), but it is most profound with the acute cor
pulmonale associated with asthma (47).
The jugular venous pulse is usually elevated with an accentuated A wave.
Although internal jugular venous valves have been well described (48), they usually
do not impair visualization of the height and pattern of the pulse. In fact, patients
with long-standing right-sided failure may develop incompetent valves. Cannon
waves or absence of A waves is rare although an irregular rhythm, such as multifo-
cal atrial tachycardia, may make assessment of the venous pulse difficult. A promi-
nent A wave is one of the most helpful indicators of reduced ventricular compliance
(or overdistention) and is more commonly seen than an elevated venous pressure
in acute cor pulmonale. Prominent A waves may help distinguish multifocal atrial
tachycardia, which is often difficult to ascertain by electrocardiography, from atrial
fibrillation. In patients with right ventricular failure, the degree of right atrial
pressure can usually be estimated within 4 em of water pressure by evaluation of
the jugular venous pressure (49). Unfortunately, the use of the accessory muscles
of respiration in acute cor pulmonale often makes evaluation of both the contour
and the level of the venous pulse difficult. Increased accessory muscle activity is
most pronounced in asthma and upper airways obstruction although prominent
contractions of sternocleidomastoid and scaleneus muscles occur with all forms of
respiratory distress. Tricuspid regurgitation is usually not seen in the venous pulse
during acute exacerbations of cor pulmonale, except in patients with primary
pulmonary hypertension.
Cardiac examination of isolated acute cor pulmonale often reveals a gentle and
sustained parasternal lift. This fmding should not be confused with the brief tap
that can be appreciated in thin individuals or patients with the straight back
syndrome. The parasternal lift can often be appreciated best by placing one hand
over the other, in cardiac resuscitation fashion, at the left parasternal region and
applying firm pressure to the area while watching the rise and fall of one's fingers.
The pulmonary outflow tract and the pulmonary component of the second heart
sound (P ~ may be palpated occasionally in patients with acute cor pulmonale.
Cardiac auscultation in isolated acute cor pulmonale usually reveals a narrowed
inspiratory splitting of the second sound. However, because of the increased
amplitude of P2, splitting may still be appreciated during expiration. In most

patients P2 will be greater in intensity than A2 in the pulmonic area, but a reduced
intensity of the pulmonic component has also been reported (50). The pulmonic
component of the second heart sound is frequently heard in the mitral area as well
as the pulmonic area. In the presence of significant right-sided heart failure, right
ventricular ejection is prolonged and pulmonary valve closure may be delayed (51).
A third or fourth heart sound prominent during inspiration is usually heard along
the left sternal border and subxyphoid area. A quadruple rhythm from both S3 and
S4 gallops is occasionally appreciated and is usually best heard during inspiration.
The murmur of tricuspid regurgitation, a high-pitched midsystolic murmur along
the lower right border and subxyphoid region, is common in acute cor pulmonale
and is best heard during inspiration. Unfortunately, most cases of acute cor pul-
monale are associated with underlying conditions and chest configurations that
distort or attenuate observations made by physical examination.
Overall inspection will establish the patient's sensorium and breathing pattern.
Paradoxic diaphragmatic motion occurs frequently in ARF secondary to COPD
but also occurs with neuromuscular disease, obesity hypoventilation syndrome, and
during agonal respiration (52). Unresponsive patients with cor pulmonale are
usually suffering from CO 2 narcosis, although I have seen one patient with the
obesity hypoventilation syndrome in such deep sleep that he was admitted with
the diagnosis of unexplained coma. Most patients with the obesity hypoventilation
syndrome have a characteristic breathing pattern due to intermittent obstruction
(53) or have a labored, noisy respiratory breathing pattern due to partial airways
obstruction. A general chest examination is also helpful in describing findings
suggestive of COPD, severe asthma, upper airways obstruction (fixed or variable),
pneumothorax, or pleural effusion. Neurologic examination can usually determine
the presence of a chronic underlying process that has predisposed to acute respira-
tory failure and detectable cor pulmonale. This situation is most often observed
in patients who have progressive amyotrophic lateral sclerosis or poliomyelitis and
develop and acute respiratory infection. Acute myasthenia gravis, the Guillain-
Barre syndrome, or drug overdose rarely show clinically significant acute cor
pulmonale since acute hypoxia and hypercapnia results in modest elevations in
pulmonary vascular resistance, which are generally well tolerated by the normal
right heart. Patients with the Guillain-Barre syndrome have an increased risk of
embolism, which should be suspected if acute cor pulmonale develops.
Peripheral edema is variable in acute cor pulmonale, and its presence usually
depends on the extent of the underlying diseases. The mechanism of fluid retention
is controversial and may not be solely due to right heart failure (54). Edema is most
profound in patients with classical Pickwickian syndrome.
Laboratory assessment
Laboratory abnormalities in acute cor pulmonale demonstrate a spectrum, depend-
ing on the precipitating cause or underlying conditions. Many of the specialized
laboratory techniques are discussed more extensively in other chapters of this book.
293
The hemoglobin level is often increased in patients with chronic hypoxia from
COPD or obesity hypoventilation syndrome, but it is usually low with connective
tissue diseases or multiple tumor emboli. Upper gastrointestinal bleeding and
nonspecific bone marrow depression are common with severe underlying disease
and often overwhelm the erythropoietic-stimulating effects of chronic hypoxemia.
Fluid retention may expand plasma volume and result in an underestimation of the
true red cell mass. In ARDS the hemoglobin is usually low, due to a combination
of bleeding, hemolysis, and underproduction. No characteristic hematologic pat-
terns are seen with most causes of acute cor pulmonale, although disseminated
intravascular coagulation has been reported to occur frequently in both ARDS and
massive pulmonary emboli (32,55). Erythrocyte morphology may help in the
diagnosis of pulmonary thrombosis in patients with sickle anemia and acute pulmo-
nary hypertension by showing an increased number of fragmented red blood cells
such as helmet cells and triangular cells (56).
No specific electrolyte changes occur with acute cor pulmonale. Hypokalemia
and metabolic alkalosis commonly result from the use of diuretics in patients with
COPD. Superimposed ARF in these patients usually produces a mixed respiratory
acidosis, in contrast to the pure respiratory acidosis seen in untreated patients with
hypoxic-hypercapnic ARF. Acute respiratory alkalosis occurs in most cases of acute
cor pulmonale that are not associated with severe airways obstruction and may be
accompanied by an increased anion gap due to increased production of both
pyruvic acid and lactic acid (57).
The blood urea nitrogen may be elevated secondary to prerenal azotemia if the
cardiac output is significantly diminished. Acute hepatic congestion can produce
elevations of serum transaminases, bilirubin, or alkaline phosphatase. Very high
arterial ammonia levels are seen in hypoxic patients with acute cor pulmonale when
cardiac output is reduced to a degree sufficient to result in congestion of the liver.
Liver congestion persistent over days will result in a prolongation of the prothrom-
bin time.
Arterial blood gases reflect the underlying respiratory process. Patients with
acute hypoxic pulmonary hypertension usually have a severe worsening of their
chronic hypoxemia although on rare occasion the Pa02 may be only slightly
depressed from baseline levels. These patients usually have an acute respiratory
acidosis, which probably contributes to the degree of pulmonary hypertension (11).
If pulmonary vascular obliteration produces acute cor pulmonale either from
embolism, vasculitis, edema, or pneumonitis, both the PaC02 and P a0 2 are usually
low. The reason for a reduced Pa0 2 following vascular occlusion is not clear, but
may be due to the development of anatomic pulmonary arterial-venous fistulae in
the parenchyma or pleural surface, opening of a patent foramen ovale, or the
development of macro- and microatelectasis in areas of remaining blood flow
producing new low VA/Q regions. Usually both VA/Q mismatch and shunts are
present since the hypoxemia is frequently nearly completely reversed by the ad-
ministration of 100% oxygen (58,59). A decreased P aC02, despite the increased
physiologic dead space, is seen with acute vascular occlusion and is caused by the
294 11. Acute Cor Puhnonale
hyperventilation resulting from the stimulation of irritant or deflating receptors.

Pulmonary function testing has been reported to aid in the differential diagnosis
of pulmonary hypertension (60-62) (chapter 7). These tests, however, are usually
of limited help in acute cor pulmonale, except to determine the presence and degree
of airways obstruction. Restrictive ventilatory defects can occur with neuromuscu-
lar or chest wall diseases, while restrictive lung disease with reduced diffusing
capacity is seen in pulmonary fibrosis, multiple pulmonary emboli, and primary
pulmonary hypertension. Lung volumes can be quantitatively helpful, but are not
diagnostically specific.
Electrocardiographic evidence of right ventricular hypertrophy is often unim-
pressive among patients with COPD, especially in the presence of associated
coronary artery disease. Criteria for pulmonary hypertension in adults without
other underlying diseases have been reported (63) and are discussed in chapter 5.
Changes from previous electrocardiograms consistent with acute cor pulmonale are
qualitatively but not quantitatively helpful in diagnosing a new increase in right
ventricular afterload (figure 11-2).
Transient arrhythmias are common in these patients, but persistent abnormalities
other than sinus tachycardia are unusual (3). Approximately half of patients with
acute respiratory failure have major supraventricular and ventricular arrhythmias
(64). Supraventricular arrhythmias, primarily atrial tachycardia and multifocal
atrial tachycardia, are slightly more common than ventricular arrhythmias. Patients
with ventricular arrhythmias have a poorer prognosis and should be closely moni-
tored. The management of arrhythmias in the setting of pulmonary heart disease
is discussed in chapter 10.
Numerous radiographic indices for pulmonary hypertension and right ventricu-
lar hypertrophy have been developed (51,65). However, indices that correlate the
size of proximal pulmonary arteries with the presence or absence of pulmonary
hypertension are generally unreliable and quantitatively nonlinear. Patients with
huge pulmonary arteries usually have pulmonary hypertension (figure 11-3), but
false positives have been recorded with most published criteria (51,65). Frequently,
patients with a radiographically unimpressive vascular pattern may have extremely
elevated pulmonary arterial pressures (figure 11-4). Unfortunately, only an anteri-
or-posterior film taken with portable equipment may be possible in many patients
with acute cor pulmonale, and serial chest films are not very helpful in assessing
the acute progression or regression of pulmonary hypertension or in evaluating the
effects of acute therapeutic interventions in this setting. In a number of cases of
documented cor pulmonale, the chest film was interpreted as representing conges-
tive heart failure (figure 11-5).
Recently, Toombs and Miller (66) have pointed out that changes in the size and
shape of the supradiaphragmatic portion of the inferior vena cava (IVe) reflect
pressure or volume overload of the right heart. In our experience the presence of
a convex backward IVC on the lateral chest film is evidence for increased right
atrial transmural pressure and is a reliable indicator of cor pulmonale under
appropriate circumstances (figure 11-6).
295
11
-- -----
. .
.-'--1
Ii II
I I I~ }I' •. :
2 ill I
j" ':
-=
11 1;IIIHf.
'.I'I!II!
dJj~. :' i
.----
fA •
, /' 1:01IIII '
, !
-I
I
' IIIRI:
,
--- - ,
I i IrIllr, 'i
l i:II:U!1
,
•• 1
'
- - I
Figure 11-2. Patient is a 56-year-old white female with severe COPD who presented with
ARF and cor pulmonale. Routine EeG one month prior to admission showed an axis of 75°,
poor R wave progression in precordial leads, and an S in V6 of 4 mm. The P waves were 2
mm in height with rounded tops. On discharge her axis was 90°, the S in V6 was 5 mm, and
the R/S ratio in VI was 1. Her P waves were 2.5 and slightly peaked.
A variety of specialized noninvasive techniques can be used in an attempt to

evaluate the presence and degree of pulmonary hypertension and right ventricular
performance. These techniques include systolic and diastolic time intervals,
echocardiography, first-pass and gated nuclear cardiography, and thallium perfu-
sion scanning. Most of these studies can detect moderate pulmonary hypertension
in chronic cor pulmonale, but little is known about the positive and negative
predictive values of these techniques in acute cor pulmonale. Systolic time intervals
by doppler or echocardiographic techniques are more satisfactory in children, for
/1-"3 - 7~
57?- ;2..,;;. -;1. ~o
Figure 11-3. Posterior-anterior chest film of a 62-year-old male with severe COPD and cor
pulmonale. Postmortem examination showed marked muscular hypertrophy of pulmonary
arterioles and multiple segmental and subsegmental emboli.
whom the pulmonary valve echocardiogram is easily obtained (67-70). Echocardi-

ography has also been used effectively to assess right ventricular diastolic wall
thickening (71). Radionuclide techniques give a higher percentage of satisfactory
studies than echocardiograms in adults with cor pulmonale secondary to COPO
(72,73). First-pass radionuclide angiocardiography produces a more reliable estima-
tion of the right ventricular ejection fraction than gated cardiac blood pool
scanning. However, this latter study gives a better estimate of right ventricular
end-diastolic volume by comparing left and right images in the LAO projection.
Radionuclide studies, however, are quite expensive and, with the exception of very
specialized units, require transporting patients to the imaging devices. In our unit
where portable equipment is available, first-pass and gated angiocardiography have
helped assess left and right ventricular performance and apparent responses to
therapy during acute respiratory failure. Thallium perfusion, along with gated
blood pool scans, has been very helpful in the diagnosis of cor pulmonale and in
297
Figure 11-4. Posterior-anterior chest film in a 52-year-old woman with unexplained pulmonary
hypertension. The mean pulmonary arterial pressure was 58 mmHg.
differentiating left versus right heart disease in complex cardiopulmonary derange-

ments (74-76) .
Recently, we have examined the use of krypton 81m (Kr-81m) for continuous
monitoring of right ventricular function. Kr-81m, with a 13-second half-life,
correlates well with Tc-99m first-pass studies and can be used repeatedly during
physiologic and pharmacologic interventions without background activity from
previous injections (77). First-pass studies combined with direct hemodynamic
measurements may answer questions concerning factors affecting the contractility
or afterload of the right ventricle (73).
DIAGNOSIS
In 1936 Prinzmetal stated the diagnosis of cor pulmonale by clinical examination
is difficult. Definitive diagnosis of acute cor pulmonale, with recent elevation in
pulmonary artery pressure, can be made only by direct hemodynamic measure-
Figure 11-5. Antero-posterior chest film in a 64-year-old male diagnosed as having congestive
heart failure. Treatment included 240 mg lasix and digoxin. Cardiac catheterization revealed a
pulmonary artery pressure of 90/50 mmHg with a pulmonary capillary wedge pressure of 15
mmHg. FEV\ performed after catheterization was 0.5 L, and PaC0 2 was 54 mmHg.
ments. Confirmation of the presence of acute pulmonary hypertension often re-

quires both pressure and flow measurements.
Hemodynamic features
The hemodynamic abnormalities in patients with acute cor pulmonale require a
brief discussion. The respiratory group from Nancy, France, has performed
hemodynamic studies repeatedly in a series of patients with significant COPD
(78). This study took place over a five-year period and the patients were reca-
theterized at three-year intervals. Approximately one-half of these patients ini-
tially had normal pulmonary arterial pressures. The hemodynamics in this group
were stable, with no new development of pulmonary hypertension. Longer peri-
ods of observation in this group would probably show similar stability since all
m:uor studies of patients with significant chronic lung disease have revealed that
approximately 50% do not have cor pulmonale at postmortem examination. This
study is probably of a tight little group who represent the natural history of
patients with COPD and who are inherently "poor responders" to a hypoxic
stimulus.
299
Figure 11-6. Upper panel: Left lateral chest film in a 61-year-old patient with severe COPD, a
mean PAP of 35 mmHg, and a right atrial pressure of 6 mmHg. The inferior vena cava is
concave toward the spine, although somewhat straightened compared to most normals. Lower
panel: Same patient four months later during an episode of acute cor pulmonale, mean PAP 60
mmHg, and right atrial pressure 12 mmHg. The inferior vena cava is now bowed backward
(convex toward spine).
Approximately 50% of patients with COPO will have cor pulmonale at autopsy
(79). Hemodynamic data from this group are usually divided into data obtained
from patients with and without heart failure. Catheterization is usually performed
when the patient's condition is stable (80). Patients without any evidence of heart
failure usually have a mean PA pressure of appro~imately 27 mmHg, and the
pressure increases with exercise. The right atrial pressure and the right ventricular
stroke work index are normal at rest and increase with exercise (81). These patients'
cardiac status appears to approximate a normal cardiac function curve although
they seem to extend higher on that normal curve, especially with exercise. A
number of patients with this type of hemodynamic profile have also undergone
repeated right heart catheterization (78). Interestingly, these patients with COPD
and compensated pulmonary hypertension have also been found to be extremely
stable over time.
Hemodynamic measurements in patients with significant right heart failure have
also been performed when their conditions are relatively stable. In contrast, mean
PA pressures are usually > 40 mmHg and increase significantly with minimal
exercise. The right atrial pressures are elevated (average 9 mmHg) and increase
further with exercise. The stroke work index may be either slightly depressed or
normal at rest, but it does not rise normally with exercise. These patients usually
have a short survival and all have cor pulmonale at autopsy.
Hemodynamic measurements in patients with acute cor pulmonale secondary to
COPD have not been frequently reported. The pulmonary artery pressure, resist-
ance, and cardiac output vary, depending on the study and are obviously deter-
mined by group selection. When hemodynamics are measured during an acute
exacerbation of COPD and repeated during recovery, acute pulmonary vasocon-
striction can be shown to be occurring at the height of the illness. Mean pulmonary
artery pressure is usually greater than 50 mmHg, but levels over 60 mmHg are not
uncommon and levels above 90 mmHg have been recorded. The pulmonary
vascular resistance is usually three to four times normal. The cardiac output remains
normal in most cases, but high and slightly low values also occur.
Experimental and clinical studies in cor pulmonale with elevated right atrial
pressure have revealed another interesting hemodynamic fmding: a direct correla-
tion between the wedge and right atrial pressures (78,80): the higher the right atrial
pressure, the more abnormal the left atrial pressure. This relationship may be
explained by ventricular interdependence (82,83) although left ventricular dysfunc-
tion secondary to right heart disease has been a recurring issue (84,85) (see chapter
9). Volume overload associated with cor pulmonale does not have as significant
an effect on gas exchange or alveolar ventilation as it does in left-sided failure (86).
Pulmonary artery pressure and resistance can also be elevated with acute pulmo-
nary emboli or ARDS. The hemodynamic measurements in pulmonary embolism
are markedly influenced by underlying cardiopulmonary disease (87,88). This is
probably less important for ARDS since patients with significant underlying
pulmonary disease usually do not survive the initial insult leading to increased
capillary permeability. Left heart disease with decreased oxygen delivery contrib-
utes adversely to survival in both pulmonary emboli and ARDS (89). Patients
without significant previous left or right heart disease usually have an elevated
cardiac output. With severe pulmonary hypertension and right ventricular over-
load, venous return and cardiac output will fall and shock may ensue. A reduced
intravascular volume status and elevations in pleural pressure can play key roles
in producing significant reductions in cardiac output (36).
During spontaneous inspiration the right heart distends because of increased
venous return, decreasing left ventricular compliance through ventricular inter-
dependence. Although this effect may change left-sided pressure-volume relation-
ships and reduce left ventricular stroke volume, its clinical significance in acute cor
301
pulmonale has not been documented. Spontaneous respiration with a falling pleural
pressure has been shown to directly influence left ventricular performance by
increasing the impedance to left ventricular ejection (90). This effect is uncommon
with ARF and COPD, in which pleural pressure falls minimally; in contrast, it
is common with ARDS and status asthmaticus, in which pleural pressure swings
are large (46,47,90).
During intermittent positive pressure breathing (IPPB), cardiac output decreases
primarily as the result of an increased pleural pressure and a decreased venous
return. This condition will be most prominent in patients with acute cor pulmonale
and high pulmonary compliance and low circulatory volume. Although IPPB may
transiently increase left ventricular stroke volume during inspiration, the steady
state effect is clearly a reduction in venous return. The reduction in venous return
can be corrected by reestablishment of the appropriate gradient for venous return
with fluid administration (91,92). Thus, patients with COPD on assisted ventilation
who show transient decreases in systemic blood pressure with inspiration are
frequently volume depleted.
High tidal volumes during artificial ventilation contribute to a reduction in the
effective pulmonary vascular bed by increasing pulmonary artery pressure (93,94).
Since the effect of alveolar pressure on pulmonary artery pressure is dependent on
the mean transpulmonary pressure, this phenomenon can be minimized by reducing
tidal volume and prolonging expiratory time (47,91,92).
Hemodynamic measurements during artificial ventilation in ARDS have been
extensively studied (42,95,96). Positive end-expiratory pressure reduces cardiac
output in proportion to the reduction in venous return and does not appreciably
depress ventricular function (37,91,97,98). In this setting cardiac output can be
increased through fluid administration (99,100). Levels of transpulmonary pressure
above left atrial transmural pressure produce a back pressure similar to elevating
wedge pressure. Prevention of excessive alveolar pressure and distention will
minimize the effects of airway pressure on pulmonary artery pressure.
Newer modes of assisted ventilation such as high frequency ventilation have been
introduced in order to improve gas exchange while potentially producing less
barotrauma and circulatory alterations (101-103). Although peak airway pressure
may be lower with high-frequency ventilation compared to conventional ventila-
tion, hemodynamics are similar if the mean airway pressures are identical (103).
More experience is necessary to evaluate the circulatory effects of high-frequency
ventilation, but to date improvement in hemodynamics does not appear to be one
of its advantages.
TREATMENT
Therapy for acute cor pulmonale is obviously influenced by the underlying eti-
ology and pathogenesis of the pulmonary hypertension. Treatment directed specifi-
cally at the elevated pulmonary artery pressure and its effect on right ventricular
performance should be instituted only when it has a reasonable potential to im-
prove patient outcome without excessive risk or cost. Much has been written about
Table 11-2. Therapeutic goals in acute cor pulmonale
1. Reverse hypoxemia
2. Optimize oxygen delivery
3. Diagnose and improve underlying physiologic dysfunction
-Correct airflow limitation
-Correct ventilation perfusion imbalance
-Correct pH
-Prevent new and dissolve old thromboemboli
-Increase alveolar ventilation
-Reduce physiologic dead space
4. Identify precipitating events
5. Adjust fluid volume
6. Prevent complications
7. Direct reduction of pulmonary vascular resistance
risk-cost-benefit ratio in the treatment of critically ill patients (104), but little is
known about outcome for most therapies, including those directed toward treat-
ment of cor pulmonale. Thus, although the tendency is to do more, as more
therapeutic modalities arise, one should temper enthusiasm unless prior experience
suggests the reasonable possibility of treatment success. Occasionally, when the
situation appears desperate, new and theoretically helpful modalities may be tried.
An outline of therapeutic goals in the treatment of acute cor pulmonale is listed
in table 11-2.
Correction of hypoxemia
The major cause of acute cor pulmonale and threat to survival in ARF is severe
hypoxemia, and primary attention should be paid to its reversal. Treatment of ARF
has been extensively reviewed in numerous textbooks and monographs (41,105).
Conservative treatment with low flow 02 is now well documented as successful
in alleviating hypoxemia and improving patient survival (106,107). Although the
PaCO2 will rise in over two-thirds of these patients, continued conservative therapy
is often still successful, and failure of this approach identifies the group of patients
with an extremely high mortality regardless of the mode of treatment, including
assisted ventilation (108). The arterial PC02 often rises following 02 therapy in
patients who need 02 the most-those with the lowest Pa02 during an episode
of ARF (109,110). Concern over a rising PaC02 should not prevent relief of
hypoxia since oxygen is essential for tissue viability and carbon dioxide produces
little direct physiologic injury.
The conventional explanation for this rise in PaC02 has been a suppression of
the chemoreceptor anoxic respiratory drive (109). Some improvement in the pH
of cerebrospinal fluid with relief of 02-dependent lactic acidosis may also contrib-
ute to the reduced respiratory drive. Recently, Murciano et al. (111) have suggested
that the increase in PaCO~ is ?ue to the oxygen-induced decrease in pulmonary
vascular resistance in low VA/Q regions, whose increased flow then "steals" blood
from the high VA/Q areas and results in an increase in physiologic dead space.
Either mechanism could explain why patients with the lowest Pa0 2 are most likely
303
to increase P aC02, but only the "steal" hypothesis would correlate a rise in PaC02
with a fall in pulmonary vascular resistance. Regardless of the mechanism for
increased PaC02' the correction of hypoxemia has the highest priority. If resistance
falls markedly as it may in some patients, cardiac output and 02 delivery are usually
further augmented, allowing for improved peripheral oxygenation at any level of
Pa0 2. Since both Pa0 2 and oxygen delivery have physiologic consequences, inter-
ventions should be made that take into account Pa0 2, 02 delivery, pulmonary
vascular resistance, PaC02 and pH, and their effects in a particular patient. Al-
though any rise in CO2 may contribute to an increase in pulmonary vascular
resistance, this effect may be overshadowed by an increased alveolar oxygen ten-
sion. However, if the PaC0 2 increases sufficiently to reduce the alveolar 02 tension
to baseline levels, the therapeutic effects of supplemental 02 may be negated and
the pulmonary vascular resistance may be higher than when therapy was first
instituted.
Relief of hypoxemia improves pulmonary vasoconstriction and reduces pulmo-
nary vascular input impedance (112). The initial hemodynamic effects of removing
alveolar hypoxia are rapid (113), and in stable COPD the pulmonary vasoconstric-
tive effect induced by acute hypoxia is reversed within one hour. However, the
ultimate effect of supplemental oxygen on pulmonary vascular resistance may take
months (114). In acute cor pulmonale the administration of supplemental 02 will
produce more acute pulmonary vasodilatation than oxygen administered at a
similar F10 2 on subsequent days (115) although the most marked fall in PA pressure
and resistance occurs within the first five to seven days of relieved hypoxia.
Long-term continuous 02 is required to reduce muscularization of the pulmonary
arteries and to further reduce pulmonary vascular resistance (114).
Hypoxic vasoconstriction is not present in normals at high altitude until PA02
falls below 80 mmHg, and vasoconstriction continues until the PA02 reaches 50
mmHg. In animal models and newcomers to a higher altitude, lower 02 tensions
produce increasing vasoconstriction. Little is known about the relationship between
the extent of hypoxic vasoconstriction and the levels of acute hypoxia in patients
with known muscular hypertrophy of the pulmonary arteries. From acute animal
experiments a P a02 of 25 mmHg produces maximum vasoconstriction, and pro-
gressive vasodilation occurs when Pa02 is increased from 50 mmHg up to approxi-
mately 150 mmHg (116-118). Since many patients tolerate high PP2 (> 80
mmHg), especially after 24 hours of treatment, levels of 02 considered "excessive"
for tissue oxygenation may be helpful in relieving the elevated pulmonary resist-
ance in severe acute cor pulmonale. Certainly, when such a patient is receiving
assisted ventilation to maintain a desired PaC0 2, a higher Fr02 than would be
desired during weaning may be beneficial in relieving pulmonary vasoconstriction.
The criteria used to determine the need for subsequent respiratory support are
described elsewhere (105,117).
Although hypoxia is usually present and often severe following acute massive
pulmonary embolization, correction of the Pa02 with supplemental 02 does little
to reduce pulmonary arterial pressure. The lack of effect of 02 has also been seen
in experimental pulmonary embolism (118). PEEP may correct the hypoxemia

following emboli in animal experiments (118), but the effect of increased transpul-
monary pressure on pulmonary vascular dynamics has not been closely examined
in clinical studies. Careful monitoring of cardiac output is required under these
circumstances since a rise in pulmonary artery or pleural pressures following PEEP
can be devastating.
The pulmonary vascular resistance is often, but not uniformly, elevated in
patients with ARDS (37,38). The pathogenesis is complex and seems to be second-
ary to a variety of insults resulting in serious microvascular damage (119). Correc-
tion of hypoxemia with supplemental 02 rarely affects pulmonary vascular resist-
ance in this setting, but I have seen a dramatic reduction in some patients. An
increase in lung volume following PEEP may shift the intercept of the pulmonary
arterial pressure-flow curve to higher or lower pressures (94,120). This change in
pressure at any particular cardiac output may be used as a guide in choosing
"optimum" PEEP (100).
Acute hypoxia has other physiologic effects that may complicate cor pulmonale,
and its reversal may improve cardiopulmonary status and overall patient outcome
(table 11-3). Correction of hypoxia is dictated by knowledge of baseline oxygen
tension, initial Pa0 2 on admission and the patient's underlying physiologic state
(age, hemoglobin, history of peripheral and central vascular disease, and so on). The
events precipitating ARF and cor pulmonale should be defmed and reversed (table
11-4). Many of the cases with COPD and no obvious precipitating cause are
probably due to both progression of the underlying disease and intermittent exacer-
bations which eventually lead to acute cor pulmonale and ARF through a combi-
nation of the physiologic derangements listed in table 11-3.
Reversal of precipitating causes

Acute cor pulmonale usually develops in the background of COPD following
some event that worsens ventilation-perfusion imbalance, decreases alveolar venti-
lation, increases metabolic demands, or increases vascular resistance. Most of these
events lead to a decreased Pa0 2, increased P aC0 2 with a decreased pH, increased
lung volume, or increased work of breathing with secondary increase in cardiac
output. Reversal of precipitating factors is an indirect but effective means of
correcting the physiologic parameters and will reduce pulmonary artery pressure
and improve right ventricular failure.
Infection may alter airways resistance through bronchial inflammation, increased
airway reactivity (almost always viral), and increased airway secretions. Secretions
usually play a minor role in altering gas exchange or airways resistance, but in
extreme cases may lead to anoxia and severe respiratory failure. Infection may also
increase oxygen consumption and carbon dioxide production secondary to fever
and increased metabolic rate, placing an increased demand on a limited respiratory
reserve. Reduction of bacterial inflammation with antibiotics will help reduce
secretions and metabolic requirements. Postural drainage, frequent turning, and
physical therapy with vibration may help some patients who demonstrate increased
305
Table 11-3. Physiologic effects of acute hypoxia
Pulmonary circulation
Vasoconstriction
Systemic circulation
Decreased 02 content
Decreased 02 delivery
Increased peripheral resistance with intact chemoreceptors
Arrhythmias---depends on underlying cardiac status
Decreased myocardial contractility p.0 2 < 30 mmHg and underlying cardiac status
Lactic acidosis p.02 < 30 mmHg
Kidney
Decreased free water clearance (SIADH)
Decreased GFR
Lung
Increased airways resistance (blocked by atropine)
eNS
Decreased cognitive functioning; obtundation
Endocrine system
Increased catecholamine secretion
Blood and bone marrow
Demargination of WBC
Increased marrow release of WBC
Increased RBC production
Table 11-4. Precipitating factors

of acute cor pulmonale
(Most factors precipitate ARF first)

1. Infection
2. Bronchospasm
3. Sedation
4. Uncontrolled 02
5. Surgery
6. Fluid overload and polycythemia
7. Electrolyte and acid-base disturbances
8. Pneumothorax
9. Pulmonary embolus
10. Arrhythmias
11. Unknown
expectoration and removal of secretions (121-123). These modalities should be

limited to individual cases where improvement can be documented. Physical ther-
apy may fatigue some patients and worsen arterial blood gases and should not be
applied indiscriminately (124-126).
Bronchospasm is clearly the major problem in asthma, but may also be significant
in some patients with COPD. New P2-selective adrenergic agents administered by
nebulization will reverse most components of airway reactivity. These drugs have
little cardiotoxic effects, and their airway effects last four to six hours. No convinc-
ing evidence exists that oral beta-adrenergic agents have additive effects when
combined with nebulized therapy, and they probably produce more cardiac irrita-
bility (127-129). The role of sympathomimetics as inotropic agents or pulmonary
vasodilators is discussed in chapter 12. However, the amount of drug and degree
of monitoring for bronchodilation must be kept distinct from that used for vasodi-
lation. Aminophylline may be additive to beta agonists in some cases of severe
asthma (130) and, at high therapeutic levels, appears to significantly improve
airflow in patients with ARF (131). Aminophylline is also a respiratory stimulant
and diuretic that can aid in the management of respiratory failure. Its effects as a
pulmonary vasodilator and attenuator of hypoxic pulmonary hypertension are
generally accepted, although recent experiments have questioned its effectiveness
(132). Aminophylline may be very difficult to manage in older patients with cor
pulmonale and liver congestion (133,134), however, and its volume of distribution
may change during stages of acute respiratory failure and recovery (135). Toxicity
is not uncommon in the critically ill, and seizures and cardiac arrhythmias have
significant morbidity. Some of the effects of aminophylline may be related to
increased catecholamine secretion (136). Recently, it has been suggested that
aminophylline may contribute to reduced pulmonary antibacterial defenses in
respiratory patients (137).
Acidosis can augment pulmonary vasoconstriction, result in fluid shifts from the
peripheral to the central circulation, and may also diminish the responsiveness to
beta agonists in asthma. In isolated lung preparations a rise in pulmonary vascular
resistance can be related to hydrogen ion concentrations (H+) (5), but in anesthe-
tized sheep the increased pulmonary artery pressures associated with increased
(H+) are related to catecholamine secretion as well (11). Severe acidosis may also
impair cardiac contractility. Correction of respiratory acidosis is often not an
immediate priority unless the pH is < 7.2 and the PaC02 continues to rise.
Bicarbonate administration is usually not indicated in these patients since this
therapy may merely depress respiration and actually further reduce cardiac output
by producing greater intracellular myocardial acidosis (138). Similarly, correction
of lactic acidosis with sodium bicarbonate may depress cardiac function without
improving survival (139,140).
Alkalosis depresses respiration, but usually has little clinical effect on pulmonary
artery pressure. Isolated alkalosis tends to attenuate hypoxic vasoconstriction. Re-
cent studies in our laboratory, using an oxidant injury model of ARDS, suggest
that alkalosis may potentiate the release of cyclooxygenase mediators, resulting in
an augmentation of pulmonary artery pressure.
Electrolyte abnormalities may aggravate cardiac arrhythmias and acid-base ab-
normalities, but they have little direct effect on pulmonary vascular resistance.
Calcium transmembrane potentials may regulate hypoxic vasoconstriction (6), but
acute pulmonary hypertension does not appear to be related to serum calcium
levels.
Polycythemia is a frequent manifestation of chronic hypoxemia, but the in-
creased red cell mass has little effect on pulmonary resistance until the hematocrit
exceeds 56%. The rheologic effect of polycythemia on vascular resistance is more
significant at high blood flows. Treatment by venesection to restore the hematocrit
toward normal produces a small decrease in pulmonary artery pressure (141), but
307
the total and central blood volumes are more important in the adjustment of
pulmonary artery pressure and flow than blood viscosity. No consistent changes
occur in PaOZ or PaCO Z after phlebotomy (142), and a slight reduction in oxygen
transport may occur (143). Phlebotomy may improve cerebral blood flow in these
patients (144).
Fluid overload
Patients with acute cor pulmonale frequently retain large amounts of fluid which
appears out of proportion to the extent of right ventricular dysfunction (54).
Although the mechanism for fluid retention has not been fully elucidated, recent
studies have documented elevated levels of antidiuretic hormone in both congestive
heart failure (145) and acute respiratory failure (146). Careful reversal of the fluid
overload may improve pulmonary compliance, alveolar ventilation, and ventila-
tion-perfusion ratios (86,147), as well as reduce pulmonary arterial pressure, right
ventricular filling pressure, and wall stress.
Acute cor pulmonale is seldom a life-threatening disorder, and diuretics should
be used with caution (148) since oxygen delivery will rarely be improved by
diuresis and secondary electrolyte abnormalities will often be aggravated. Absolute
or relative dehydration is probably more worrisome a problem in patients with
ARF and should be avoided. Volume depletion itself does not precipitate cor
pulmonale but may lead to a sequence of events that worsens tissue oxygenation
and airway function and reduces perfusion to high VAl Q units with physiologic
deterioration leading to higher pulmonary vascular resistance.
A number of other factors may be involved in the development of acute
respiratory failure and cor pulmonale. These include muscle fatigue (149), malnu-
trition, arrhythmias often induced by digitalis toxicity, sedation, negative inotropic
or bronchodilator drugs, and electrolyte disturbances such as hypophosphatemia
(150,151). Obviously, reversal of an acute pneumothorax in a patient with respira-
tory failure and cor pulmonale has the highest priority since this complication may
worsen arterial blood gases and decrease cardiac output, leading to a rapid demise.
Tension pneumothorax per se does not usually produce pulmonary hypertension
or reduce venous return (152).
Special conditions
A number of conditions may result in acute cor pulmonale or contribute to an
already elevated pulmonary vascular resistance. Sleep apnea or hypopnea may
produce severe cor pulmonale in patients with the obesity hypoventilation syn-
drome (153) or aggravate hypoxic vasoconstriction in patients with COPD (154-
156). Treatment of the obesity hypoventilation syndrome is still unsatisfactory, but
clinical improvement may be achieved with progesterone (157); tricyclic an-
tidepressants (158); a nasal pharyngeal airway, nasal CPAP (159); weight loss,
uvulopaletectomy (160,161); and oxygen (162). In cases with severe cor pulmonale
and life-threatening arrhythmias, tracheostomy is often required (163). No therapy
has been evaluated satisfactorily for the patient with COPD and sleep apnea or
hypopnea (164) exacerbating pulmonary heart disease. Optimization of waking
cardiopulmonary status is the only therapeutic approach currently agreed upon and
alcohol, beta blockers, and sedatives should be avoided in these patients.
Thromboembolism is a common autopsy froding in patients with ARF second-
ary to COPD or ARDS {165}, but its exact role in patient outcome is not clear.
Prescott et al. (f~6) have recently reported that COPD patients admitted with ARF
who survive do not have an increased incidence of clinically detectable venous
obstruction and, by inference, should not develop pulmonary emboli. In 250
consecutive autopsies on patients dying in our MICU, we found a 33% incidence
of major {subsegmental or larger} thromboemboli which were rarely diagnosed
premortem. Over two-thirds of these patients had respiratory failure. A similar
incidence of undetected premortem clots has been reported by others (167). The
significance of thromboemboli found at autopsy is not clear, and mortality may
or may not be influenced by embolic events. Studies without concurrent controls
reporting prevention of pulmonary emboli with low-dose heparin in intensive care
units are difficult to interpret.
Patients with clinical acute cor pulmonale due primarily or secondarily to
thromboembolism should be treated with fibrinolytic therapy (168,169), unless
contraindicated {170}. If these patients deteriorate during subsequent heparin ther-
apy or have significant bleeding, they are candidates for interruption of the inferior
vena cava {171}. The major therapeutic strategy for patients with acute thrombo-
embolic cor pulmonale other than dissolution of clots and prevention of recurrence
is maintenance of adequate venous return, which may require massive fluid ad-
ministration. Oxygenation should also be optimized and can usually be accom-
plished with a high F10 2, but occasionally the use of PEEP is required. PEEP can
effectively reduce the right-to-Ieft shunt in patients with pulmonary emboli by
opening perfused atelectatic units, but it may also complicate hemodynamic man-
agement; its use requires careful monitoring.
Patients with acute respiratory failure secondary to neurologic disorders are
uncommon and should be treated with assisted ventilation and oxygen in a manner
simialr to patients with COPD since elevation of the pulmonary vascular resistance
is usually the result of similar physiologic derangements.
Digitalis
The treatment of cor pulmonale with digitalis is discussed in chapter 12. Many
experimental models and most patients with acute cor pulmonale have a normal
cardiac output in spite of elevated filling pressures {172,173}. Demonstrating a
substantial clinical effect of digitalis in these patients has been difficult. However,
in patients with poor ventricular performance, digitalis may reduce ventricular
filling pressure and improve cardiac output {148,174}. Such improvement in myo-
cardial performance may be associated, however, with further elevations in pulmo-
nary artery pressure and resistance due to direct vasoconstriction {175}. Addition-
ally, arrhythmias secondary to digitalis occur frequently, even in the face of
309
"therapeutic" serum levels (176). A variety of factors, such as hypoxia, sympathetic

discharge, and electrolyte disturbances, may contribute to the rhythm disturbances
(176). Thus, digitalis should be considered only in patients with overt left ventricu-
lar failure. The benefit will not be dramatic and digitalis should not be relied upon
as a sole therapeutic measure.
Therapy with vasoactive drugs

Currently, the only established means of reducing pulmonary hypertension is by
treating the underlying disease and improving alveolar and arterial oxygen tension
(113,177,178). If pulmonary hypertension is of recent onset or recently accelerated
so that it is poorly tolerated by the right ventricle, vasodilator therapy may tide
the patient over until the acute aggravating problem is resolved.
The physiologic basis of vasodilator therapy for heart failure has been recently
reviewed (179). Drugs that relax arterial smooth muscle should result in an increase
in ejection fraction, an increased stroke volume, and a reduced end-systolic ven-
tricular volume. Concomitant venous relaxation will shift blood into the periphery
and reduce ventricular end-diastolic volume. A reduction in ventricular volume
in both diastole and systole should relieve edema by decreasing upstream capillary
pressure, decrease myocardial wall stress and oxygen consumption, improve dias-
tolic perfusion of the myocardium, and, perhaps, reduce the stimulus for myocar-
dial hypertrophy.
Simonneu et al. (116) have recently examined the hemodynamic effects of
nifedipine in patients with acute hypoxic respiratory failure. As might be an-
ticipated from animal experiments with acute hypoxia (16), they found that
nifedipine produced a decrease in PA 'pressure and resistance and an increase in
cardiac output. The administration of supplemental oxygen produced a comparable
reduction in pulmonary artery pressure. The quantitative and qualitative hemody-
namic similarities in these studies between 02 and nifedipine, both inhibitors of
hypoxic vasoconstriction, could be due to the fact that a high Fr02 (mean Pa 0 2
240 mmHg) and relatively small doses of nifedipine (20 mg) were administered.
Whether higher concentrations of nifedipine can produce vasodilation beyond that
produced by oxygen was not determined. However, high concentrations of 02 may
be impractical in patients with acute respiratory failure, thus limiting the full
usefulness of the vasoactive properties of oxygen in clinical practice. The relative
value of vasodilator therapy alone or in some combination with 02 remains
speculative. We have found that the reversal of acute hypoxic pulmonary hyper-
tension with nifedipine in an animal model exceeds the effects of other vasodilators
(16), making calcium channel blockers excellent agents for future investigation in
patients with pulmonary hypertension. The dosage required to completely abate
acute hypoxic pulmonary vasoconstriction was well within the recommended
therapeutic dosage. In clinical practice we have administered 30 to 40 mg of
nifedipine to several patients with ARF resulting from COPD or the obesity
hypoventilation syndrome with excellent hemodynamic results. After an adequate
volume status was achieved, sequential doses of 10 mg orally every 30 minutes,
to a total of 30 to 40 mg, were given while systemic arterial pressure and right
atrial pressure were monitored. These patients uniformly manifested a decrease in
pulmonary vascular resistance and an increase in cardiac output. Over half demon-
strated a fall in PA pressure, as reported by Simonneu. The Pa0 2 fell only slightly,
and in four patients not receiving respiratory assistance, the PaCO2 also fell. The
effects on PaC02 may be due to the increased respiratory drive resulting from a
decreased Pa02 or from a change in airway function resulting in a reduced
physiologic dead space. All patients also received supplemental 02 to maintain a
P a02 between 50 and 80 mmHg, as well as theophylline and nebulized bron-
chodilators. Although none of these patients developed systemic hypotension, this
drug's long duration of action makes it difficult to regulate in ARF. Although
nifedipine is not a bronchodilator, it may attenuate acute airways reactivity (180).
Intravenous nitroglycerin (TNG) has recently been introduced into clinical
practice. It is a superb venous and systemic vasodilator with a very short half-life
(181,182). We examined the effects of intravenous nitroglycerin in seven patients
with ARF within 12 hours of admission to the MICU. All patients received
low-flow 02 to maintain an average Pa02 of 80 mmHg, intravenous aminophylline
(5.6 mg/kg load followed by 0.5 mg/kg/hr infusion) and nebulized betaz-agonists
every 4 hours (183). For one hour prior to and throughout the study, no nebulized
therapy was administered. Increasing concentrations of TNG were administered
until mean arterial pressure fell 10 mmHg. These patients experienced a 16%
decrease in mean pulmonary artery pressure, a 40% decrease in PVR, and a 20%
increase in cardiac output. The mean Pa02 fell approximately 20 mmHg, but
because supplemental 02 was given, the P a02 remained at satisfactory levels (>
60 mmHg) in all patients. Spirometry improved in three patients, but was un-
changed in the group as a whole. This unpredictable airway response is consistent
with previous studies in acute asthma (184). The difference between the effects of
TNG in patients with COPD (72) and the above results can probably be explained
by the tendency of TNG to decrease cardiac output in stable patients due to
vasodilation when intravascular volume is low. In contrast, all of our patients had
elevated right ventricular filling pressures and most were edematous. Other
vasodilators have been shown to be effective in some patients with hypoxic
pulmonary hypertension (72,73,185,186), but the advantage of intravenous nitro-
glycerin over these other agents may be the ability to titrate effective doses and
its short half-life of two to three minutes (181). The infusion rate is generally begun
at 5 p.g/min and increased every 5 minutes until a 5-10 mmHg reduction in mean
systemic arterial pressure is obtained. The average dose in patients with acute cor
pulmonale has been 50 p.g/min. The effects on the pulmonary circulation are then
determined by hemodynamic measurements. Side effects other than a reduction in
Pa0 2 have not been observed.
Clinical interest is also increasing in the beneficial pulmonary and cardiac
hemodynamic effects of established bronchodilation. Aminophylline and beta ago-
nists improve right and left ventricular performance (72,73,185,186) in addition to
reducing acute hypoxic pulmonary vasoconstriction (187). These drugs might be
311
of primary value in patients with a very low cardiac output and systemic hypoten-
sion. However, most patients with ARF are already receiving these medications,
a possible reason why patients with acute cor pulmonale do not generally have a
reduced cardiac output. The current usage of these drugs makes it difficult to
consider increasing doses to improve cardiac performance. Intravenous nitroglyc-
erin may have additional hemodynamic effects in patients with ARF who are
receiving usual doses of nebulized beta agonists and intravenous aminophylline,
although prospective studies are required before any vasoactive drug should be
routinely recommended for patients with acute cor pulmonale secondary to
COPD.
Isoproterenol reduces pulmonary vascular resistance following experrimental
acute massive pulmonary embolism (118). Isoproterenol may be an ideal vasodila-
tor for this condition because of its additional inotropic effects on the right
ventricle. Because of poor left ventricular filling (obstruction to flow through the
lung), the associated chronotropic effect of isoproterenol is also probably advanta-
geous as long as heart rate does not exceed 150 beats per minute. Other vasodilators
have produced a decrease in pulmonary vascular resistance in patients with more
stable pulmonary vascular occlusive disease (26,94,188). However, any venous
dilation resulting in a reduction in venous return in patients with acute massive
emboli and systemic hypotension could be disastrous. Drugs with negative in-
atropic effects should probably be avoided in the acute setting, especially if systemic
hypotension is present.
Patients with ARDS and severe pulmonary hypertension may benefit from
reducing fluid flux by a reduction in left atrial pressure (96,189,190). Lowering
mean pulmonary artery pressure and thus mean microvascular pressure can also
reduce lung leakage (191,192). Recent experimental evidence suggests that hypoxia
itself increases lung fluid filtration (24,193). On rare clinical occasions an elevation
of alveolar pressure with PEEP may also reduce mean pulmonary artery pressure,
but more often increasing lung volume results in an elevation of pulmonary arterial
pressure (94) with either an unchanged (92,194) or an increased fluid flux (120).
Although a variety of vasoconstricting substances have been shown to be released
by the lungs of ARDS patients (195,196), blockade of mediator release has not been
effective clinically. A number of nonspecific vasodilators have been shown to
reduce pulmonary vascular resistance in models of ARDS-induced pulmonary
hypertension (197,198). If patients with pulmonary hypertension require cardiac
augmentation either because of the use of PEEP or coexistant myocardial dysfunc-
tion, selection of a vasoactive agent with combined inotropic and vasodilating
properties is probably wise (73,187). Some of these substances may actually reduce
fluid flux as well as the driving pressures for fluid filtration (192,199,200).
Assisted ventilation
Patients with acute cor pulmonale may require assisted ventilation for their primary
pulmonary condition. Assistance is most frequently instituted to increase ventila-
tion and augment CO 2 elimination in patients with COPD or asthma, but may
be implemented in the form of positive end-expiratory pressure to open alveolae

and reduce right-to-Ieft shunting in patients with ARDS and pulmonary embolism.
PEEP has also been suggested for the treatment of asthma (201). Recently, high-
frequency ventilation has been introduced in an effort to improve gas exchange in
patients with ARDS, bronchopleural fistulae, and COPD (5,102). In addition to
the usual indications and complications associated with assisted ventilation, patients
with acute cor pulmonale must be managed with consideration given to optimizing
their circulatory status. The m:yor physiologic abnormality in acute cor pulmonale
is an increased PVR and pulmonary arterial pressure, producing increased right
ventricular intracavitary tension. The right ventricle dilates using its volume re-
serve in an attempt to meet the afterload, which further increases right ventricular
wall stress (73,83,96,202). In acute cor pulmonale the increases in right ventricular
volume and pressure require a sudden increase in mean systemic pressure to main-
tain an appropriate venous return (91,97). Depending on the circulatory volume
and compliance of the venous bed, the gradient for venous return is very susceptible
to increases in pleural pressure. Thus, minor changes in pleural pressure during
artificial ventilation may produce a severe decrease in right ventricular stroke
volume and output. A large right ventricle may also change the compliance of the
left ventricle and impede its filling (83,202). Left ventricular venous return may
also be impaired by a high PVR and result in a reduced left ventricular stroke
volume. Any elevation in alveolar pressure in a zone 2 state (frequently observed
because of air trapping in COPD [203], asthma [47], and with pulmonary emboli
during PEEP) will further increase the back pressure to right ventricular emptying,
placing a greater load on this ventricle and, consequently, further decrease venous
return to the left heart (92,97).
Inspiration is associated with numerous cardiovascular effects, depending on
whether pleural pressure increases or decreases, whether the lung is in a predomi-
nantly zone 2 or zone 3 condition, and whether the mean systemic pressure and
gradient for venous return is high or low (46,83,91). The effects of inspiration will
also be determined by the function of the right and left ventricles and their
respective afterloads (46,81,203) and by the presence of cardiac and pulmonary
interdependence (83,202). The resting lung volume and the magnitude of the
inspiratory change dictated by the end-expiratory, mean, and peak transpulmonary
pressures will affect all of the above factors in a quantitative fashion. Cournand
correlated the decrease in cardiac output during IPPB ventilation with the peak
airways pressure and the ratio of inspiration to expiration, since these factors are
the m:yor modifiers of pleural pressure and venous return is the main determinant
of cardiac output in the normal ventricle (3,4). Minimizing peak and mean airways
pressure by reducing tidal volume will prevent a significant reduction in venous
return so that the right heart can meet the increased impedance associated with acute
cor pulmonale. Minimizing changes in alveolar pressure relative to pleural pressure
(PALT~ will also minimize the load on the right ventricle and maintain left
ventricular venous return in most patients with zone 2 pulmonary conditions. In
COPD with ARF mean pleural and mean transpulmonary pressures can be reduced
313
by increasing the inspiratory flow rate, a process that will produce a shorter period
oflung inflation and leave sufficient time for complete airway emptying (204,205).
Numerous studies have now demonstrated that left ventricular performance is not
significantly altered by assisted ventilation and end-expiratory pressure (96,98,206).
The effectiveness of high-frequency ventilation in reducing the risk of baro-
trauma in the usual patient requiring assisted ventilation is still in the investigative
stage, and this mode has not been proven superior to conventional ventilation.
Although peak airways pressure may be high during high-frequency ventilation,
this pressure is extremely transient and probably not transmitted to the alveolae.
Compared to high-frequency ventilation, conventional ventilation produces higher
peak alveolar pressures, but for the same mean transpulmonary pressure the circula-
tory effects of high-frequency ventilation are probably not significantly different
than conventional ventilation, although research in this area is in its infancy (102).
At very high frequencies and driving pressures, air trapping may occur with jet
ventilation. We have seen two patients with severe ARDS who could be ade-
quately ventilated with high-frequency jet ventilation but could not maintain an
adequate blood pressure, presumably because of high alveolar and/or pleural
pressures. The chest films at this time demonstrated marked hyperinflation and the
appearance of bronchopulmonary dysplasia. In contrast, both of these patients
maintained a better blood pressure and cardiac output with conventional ventila-
tion although adequate gas exchange could not be sustained. Both patients expired
and were found to have severe pulmonary congestion, hyaline membranes, and
pulmonary fibrosis consistent with end-stage ARDS.
Assessment of response to therapy

Hemodynamic data gathered through direct measurements with balloon flotation
catheterization of the pulmonary artery are now used routinely in the care of
critically ill patients (207). It is generally agreed that the physiologic data obtained
by direct measurements are essential to management of many patients (208).
Recently, Connors et al. (207) have demonstrtated that the hemodynamic status
of leu patients based on indirect clinical assessments frequently conflicts with
direct measurements, a fmding that may dictate changes in therapy.
Knowledge of hemodynamic measurements allows the clinician to maximize
cardiac output and oxygen delivery by optimizing the preload and afterload (179).
One can also reduce pulmonary vascular resistance by optimizing alveolar oxygen
tension using vasodilators and adjusting PEEP. Hydrostatic pressure within the
venous and capillary bed can also be monitored to minimize pulmonary edema
(96). In addition, knowledge of systemic arterial pressure can prevent hypoperfu-
sion of vital organs, such as the kidneys, brain, and heart. Repeated measurements
as therapy proceeds are usually more helpful than isolated measurements.
The therapeutic goals vary with different causes of cor pulmonale. In the
majority of patients cor pulmonale is secondary to COPD, and symptomatic
improvement usually occurs following improved oxygenation and reduced work
of breathing (107). The resolution of either parenchymal inflammation following
infection or the resolution of acute vascular obstruction following pulmonary

emboli also decreases the degree of dyspnea. Improved right ventricular emptying
as hypoxic vasoconstriction subsides does not seem to influence symptoms, although
acute, severe overload may be associated with right ventricular subendocardial
ischemia (209). Any increase in systemic blood flow and 02 delivery should have
salutory effects on organ function. However, except for possible increased urine
output and better cognitive function, the effects of improving organ perfusion have
not been well documented.
Assessment of response to therapy for cor pulmonale depends on objective
measurements and an understanding of the physiologic concepts operative in heart
failure (179,208). The right ventricle appears to maintain its output if adequate
filling is allowed during most imposed afterloads. Only massive pulmonary embo-
lism overwhelms the capacity of the right heart to deliver blood to the periphery, yet
even this reduction in blood flow can often be augmented with sufficient fluid
administration. Since the right heart usually provides adequate blood flow, the
question arises, What is the cost of such flow to the rest of the body and to the heart
itself? Fortunately, as right ventricular distention occurs to allow the heart to meet a
finite load, only peripheral edema or liver congestion results, rather than the
pulmonary edema associated with left ventricular distention. Therefore, an adequate
response to therapy is indicated by symptomatic improvement with reestablishment
of vital organ function. A cardiac output sufficient to prevent generalized lactic
acidosis or tissue ischemia is usually easily maintained. Reducing pulmonary artery
pressure or resistance is a secondary priority relative to increasing pulmonary blood
flow. A reduced pulmonary artery pressure is probably helpful over the long run, but
usually unnecessary acutely. Monitoring the patient and observing the gradual fall in
pulmonary vascular resitance with 02 or fibrinolytic therapy is usually satisfactory.
Therapeutic manipulation of right ventricular function should be reserved for
patients with a severe, acute decrease in right ventricular ejection fraction and poor
tissue perfusion. In these patients vasodilation may be tried to decrease right
ventricular load and increase organ perfusion. Arterial hypoxia may worsen (96,-
116,189,198) with potential adverse effects, and blood gases should be monitored
carefully during vasodilator therapy. Vasoactive drugs with inotropic and pulmo-
nary dilating effects may be preferred, if arrthymias can be averted.
Worsening blood gases or pulmonary edema may occur with inadvertant fluid
administration and should be corrected by gentle diuresis. Excessive fluid loss must
be carefully avoided since small changes in venous return can produce marked
reductions in cardiac output .and lethal metabolic or electrolyte abnormalities.
Although changes in venous return may be relatively unimportant in severe left
ventricular failure, the right ventricle rarely appears to be operating on a markedly
flattened portion of the Frank-Starling curve. Thus, small changes in right ventricu-
lar pressure, often aggravated by elevated pleural pressure secondary to artificial
ventilation, can produce severe reductions in cardiac output.
No studies have been conducted to evaluate the effects of treatment outcome
with and without invasive monitoring. However, the inadequacy of clinical evalu-
315
Table 11-5. Complications associated

with treatment of ARF and cor pulmonale
Pulmonary
Aspiration of gastric contents
Atelectasis (especially LLL)
Barotrauma, pneumothorax
Tracheal hemorrhage or injury
Thromboemboli
Circulatory
Arrhythmias
Hypotension
Hypoxemia (vasodilation, i0 2 extraction)
Liver dysfunction (i NH 3, t drug clearance)
Volume overload
Metabolic
Alkalosis (hyperventilation, N-G suction, diuretics)
Hypokalemia, hypophosphatemia
Hematologic
Bleeding from anticoagulation
Clotting from stasis, intravascular lines
Gastrointestinal
Gastric distension (aerophagia)
Bleeding
Fecal impaction
Infection
Nosocomial, mostly respiratory
Urinary and systemic also frequent
Intubation
Aspiration
Airway injury (upper or lower)
Monitoring
Vascular injury (lacteration or hemothorax)
Lung perforation (pneumothorax)
Infection (local, bacteremic, embolic)
Hyperalimentation
Volume overload
Acute fatty liver (pain, nausea)
Elevated transaminase
Hypophosphatemia
Toxic
Many drugs, especially aminophylline, antibiotics, digitalis
Cimetidine (affects metabolism of other drugs)
ation in predicting underlying hemodynamics (207) suggests that appropriate

monitoring should be obtained in order to optimize treatment. It is imperative that
data obtained by right heart catheterization be accurate and that complications are
minimized (210-213). As soon as this form of monitoring is no longer necessary,
it should be withdrawn.
Complications
A number of complications are associated with therapy of ARF and cor pulmonale.
Most of these complications have been extensively reviewed elsewhere (41), and
the major complications encountered during treatment are listed in table 11-5.
Whether any particular untoward event is due to ill-advised therapy per se or is
a result of poor response to state-of-the-art treatment is often hard to determine.
Most complications occur in association with assisted ventilation, but this is obvi-
ously a selected group of the most severely ill patients. Many of the complications
are so interrelated that their organization as to site, etiology, or ultimate conse-
quence is impossible. For example, volume overload is frequently seen during
attempts to correct hypokalemia or during hyperalimentation and results in marked
circulatory congestion, pulmonary hypertension, and hepatic congestion, which can
also cause a decreased drug clearance. Volume overload also results in increased
hypoxemia and either hyperventilation or hypoventilation, which can both pro-
duce arrhythmias and seizures. The insertion of lines to monitor the effects of
hypervolemia may also result in additional complications.
In general, it must be reemphasized that acute cor pulmonale itself is usually not
the limiting factor in patient outcome. However, a circulatory demise is often the
result of overzealous therapy for all aspects of ARF. Attention must be paid to
careful monitoring of the responses to any therapeutic intervention, and considera-
tion should be given to its discontinuation if adequate improvement is not forth-
coming. Drugs such as aminophylline (214) are often utilized because of hopeful
salutary effects on isolated organ systems, but the use of these drugs may be further
complicated by the addition of other medications such as cimetidine (215), which
may actually increase sensitivity to bronchospasm and potentiate hypoxic vasocon-
striction (5). Acute and chronic cor pulmonale has created many controversies
because specific physiologic effects on the host are hard to isolate and attempts to
reverse hemodynamic dysfunction with treatment may not influence outcome. The
best a physician can do is to treat the whole patient and not reverse anyone
parameter too quickly or too extensively.
REFERENCES
1. World Health Organization. Chronic cor pulmonale. A report of the expert committee.
Circulation 27: 594-615, 1963.
3. Fisherman AP. Chronic cor pulmonale. Am Rev Respir Dis 114: 775-794, 1976.
4. Fishman AP. Hypoxia on the pulmonary circulation. How and where it acts. Circ Res 38:
221-231, 1976.
5. Bergofsky EH. Active control of the normal pulmonary circulation. In Pulmonary Vascular
Diseases, Moser KM (ed). New York: Marcel Dekker, 1979, pp. 233-278.
6. McMurtry IF, Stanbrook HS, Rounds S. The mechanism of hypoxic pulmonary vasoconstric-
tion: A working hypothesis. In Oxygen Transport to Human Tissues, Leoppky JA, Riedesel ML
(eds). Amsterdam: Elsevier North-Holland, 1982, pp. 77-90.
7. Levitzky MG. Chemoreceptor stimulation and hypoxic pulmonary vasoconstriction in conscious
dogs. Resp Physiol 37: 151-160, 1979.
8. Levitzky MG, Newell q, Krasney JA, Dutton RE. Chemoreceptor influence on pulmonary
blood flow during unilateral hypoxia in dogs. Resp Physiol 31: 345-356, 1977.
9. Kivity S, Souhrada JF. Plasma and platelets potentiate a hypoxic vascular response of the isolated
lung. J Appl Physiol 51: 875-880, 1981.
317
10. Ahmed T, Oliver W]r, Frank BL, Robinson M], Wanner A. Hypoxic pulmonary vasoconstric-
tion in conscious sheep: Role of mast cell degranulation. Am Rev Respir Dis 126: 291-297,
1982.
11. O'Brodovich HM, Stalcup SA, Pang LM, Mellins RB. Hemodynamic and vasoactive mediator
response to experimental respiratory failure.] Appl Physiol 52: 1230-1236, 1982.
12. Weir EK. Does normoxic pulmonary vasodilatation rather than hypoxic vasoconstriction
account for the pulmonary pressor response to hypoxia. Lancet 1: 476-477, 1978.
13. Gregory T], Newell ]C, Hakim TS, Levitzky MG, Sedransk N. Attenuation of hypoxic
pulmonary vasoconstriction by pulsatile flow in dog lungs.] Appl Physiol53: 1583-1585, 1982.
14. Harabin AL, Peake MD, Sylvester ]T. Effect of severe hypoxia on the pulmonary vascular
response to vasoconstrictor agents.] Appl Physiol 50: 561-565, 1981.
15. Knoblauch A, Sybert A, Brennan N], Sylvester ]T, Gurtner GH. Effect of hypoxia and CO
on a cytochrome P-450-mediated reaction in rabbit lungs.] Appl Physiol 51: 1635-1642,
1981.
16. Kennedy T, Summer W. Inhibition of hypoxic pulmonary vasoconstriction by nifedipine. Am
] Cardiol 50: 864-868, 1982.
17. Wagenvoort CA, Wagenvoort N. Hypoxic Pulmonary Hypertension. Pathology of Pulmonary
Hypertension. New York: John Wiley, 1977, pp. 232-259.
18. Enson Y, Giuntini C, Lewis ML, Morris TQ, Ferrer MI, Harvey RM. The influence of
hydrogen ion concentration and hypoxia on the pulmonary circulation. ] Clin Invest 43:
1146--1162, 1964.
19. Yu PNG, Lovejoy FW ]r,]oos HA, Nye RE]r, McCann WS. Studies of pulmonary hyperten-
sion. I. Pulmonary circulatory dynamics in patients with pulmonary emphysema at rest.] Clin
Invest 32: 130-137, 1953.
20. Ishikawa ST, Fattal GA, Popiewicz], Wyatt ]P. Functional morphometry of myocardial fibers
in cor pulmonale. Am Rev Respir Dis 105: 358-367, 1972.
21. Flick MR, Block A]. Continuous in-vivo monitoring of arterial oxygenation in chronic
obstructive lung disease. Ann Intern Med 86: 725-730, 1977.
22. Lindsay DA, Read]. Pulmonary vascular responsiveness in the prognosis of chronic obstructive
lung disease. Am Rev Respir Dis 105: 242-251, 1972.
23. Abraham AS, Cole RB, Green 10, Hedworth-Whitty RB, Clarke SW, Bishop ]M. Factors
contributing to the reversible pulmonary hypertension of patients with acute respiratory failure
studied by serial observations during recovery. Circ Res 24: 51-60, 1969.
24. Mitzner W, Sylvester ]T. Hypoxic vasoconstriction and fluid filtration in pig lungs.] Appl
Physiol 51: 1065-1071, 1981.
25. Moser KM. Pulmonary vascular obstruction due to embolism and thrombosis. In Pulmonary
Vascular Diseases, Moser KM (ed). New York: Marcel Dekker, 1979, pp. 341-386.
26. Dantzker DR, Bower ]S. Partial reversibility of chronic pulmonary hypertension caused by
27. McIntyre KM, Sasahara AA. Hemodynamic and ventricular responses to pulmonary embolism.
Prog Cardiovasc Dis 17: 175-190, 1974.
28. Gurtner GH, Knoblauch A, Smith PL, Sies H, Adkinson NF. Oxidant and lipid-induced
pulmonary vasoconstriction mediated by arachidonic acid metabolites. Submitted, ] Appl
Physiol.
29. Brigham KL, Newman ]H, Snapper ]R, Ogletree ML. Metabolites of arachidonic in the
pathophysiology of the pulmonary circulation. In Advances in Prostaglandin Tromboxane and
Leukotriene Research, Oates]H (ed). New York: Raven Press, 1982, pp. 357-366.
30. Huttemeier PC, Watkins WD, Peterson MB, Zapol WM. Acute pulmonary hypertension and
lung thromboxane release after endotoxin infusion in normal and leukopenic sheep. Circ Res
50: 688-694, 1982.
31. Rinaldo ]E, Rogers RM. Adult respiratory distress syndrome: Changing concepts oflung i~ury
and repair. N Engl] Med 306: 900-909, 1982.
32. Robin ED. Some basic and clinical challenges in the pulmonary circulation. Chest 81: 357-363,
1982.
33. Risk C, Peterson M, Woods B, Rie M, Kong 0, Watkins WD. Thromboxane and prostacyclin
(epoprostenol) during exercise in diffuse pulmonary fibrosis. Lancet 2: 1183-1185, 1982.
34. Naslund M], Pearson TA, Ritter ]M. A documented episode of pulmonary vasoconstriction in
systemic sclerosis. Johns Hopkins Med] 148: 78-80, 1981.
35. Voelkel N, Reeves JT. Primary pulmonary hypertension. In Pulmonary Vascular Diseases,
Moser KM (ed). New York: Marcel Dekker, 1979, pp. 573-628.
36. Zimmerman GA, Morris AH, Cengiz M. Cardiovascular alterations in the adult respiratory
distress syndrome. Am J Med 73: 25-34, 1982.
37. Utsonomiya T, Krausz MM, Valeri CR, Shepro D, Hechtman HB. Treatment of aspiration
pneumonia with ibuprofen and prostacyclin. Surgery 90: 170-176,1981.
38. Carrington CB, Liebow AA. Pulmonary veno-occlusive disease. Hum Pathol 1: 322-324,
1970.
39. McDonnell PJ, Summer WR, Hutchins GM. Pulmonary veno-occlusive disease: Morphological
changes suggesting a viral cause. JAMA 246: 667-671, 1981.
40. Inter-Society Commission for Heart Disease Resources. Primary prevention of pulmonary heart
disease. Circulation 4: A-17, 1970.
41. Moser K. Acute respiratory failure with hypercapnia. In Respiratory Emergencies, Moser KM,
Spragg RG (eds). St. Louis: CV Mosby, 1982, pp. 82-93.
42. Zapol WM, Snider MT. Pulmonary hypertension in severe acute respiratory failure. N Engl
J Med 296: 476-480, 1977.
43. Lamy M, Fallat Rj, Koeniger E, Dietrich HP, Ratliff jL, Eberhart RC, Tucker HJ, Hill]D.
Pathologic features and mechanisms of hypoxemia in adult respiratory distress syndrome. Am
Rev Respir Dis 114: 267-284, 1976.
44. Bell WR, Simon TL, DeMets DL. The clinical features of submassive and massive pulmonary
embolism. Am J Med 62: 355-360, 1977.
45. Stein PD, Willis PW, III, DeMets DL. History and physical examination in acute pulmonary
embolism in patients without preexisting cardiac or pulmonary disease. Am J Cardiol 47:
218--223, 1981.
46. Wise RA, Robotham jL, Summer WR. Effects of spontaneous ventilation on the circulation.
Lung 159: 175-186, 1981.
47. Summer WR. Physiological changes in the acute asthmatic attack. In Status Asthmaticus, Weiss
EB (ed). Baltimore: University Park Press, 1978, pp. 81-99.
48. Fisher J, Vaghaiwalla F, Tsitlik J, Levin H, Brinker J, Wesifeldt M, Yin F. Determinants and
clinical significance of jugular venous valve competence. Circulation 65: 188--196, 1982.
49. Davison R, Cannon R. Estimation of central venous pressure by examination of jugular veins.
Am Heart J 87: 279-282, 1974.
50. Sutton G, Harris A, Leatham A. Second heart sound in pulmonary hypertension. Br Heart J
30: 743-756, 1968.
51. Gregoratos G, Karliner JS, Moser KM. Mechanisms of disease and methods of assessment. In
Pulmonary Vascular Diseases, Moser KM (ed). New York: Marcel Dekker, 1979, pp. 279-339.
52. Roussos C, Macklem PT. The respiratory muscles. N Engl J Med 307: 786-797, 1982.
53. Remmers JE, Anch AM, deGroot WJ. Respiratory disturbances during sleep. Clin Chest Med
1: 57-72, 1980.
54. Campbell EJM, Short DS. The cause of oedema in cor pulmonale. Lancet 1: 1184-1186,1960.
55. Wilson JE III, Frenkel EP, Pierce AK, Johnson RL Jr, Winga ER, Curry GC, Mierzwiak DS.
Spontaneous fibrinolysis in pulmonary embolism. J Clin Invest 50: 474-480, 1971.
56. Barreras L, Diggs LW, Bell A. Erythorcyte morphology in patients with sickle cell anemia and
pulmonary emboli. JAMA 203: 569-573, 1968.
57. Kreisberg RA. Lactate hemeostasis and lactic acidosis. Ann Intern Med 92: 227-237, 1980.
58. Wilson JE III, Pierce AK, Johnson RL, Winga ER, Harrell WR, Curry GC, Mullins CB.
Hypoxemia in pulmonary embolism: A clinical study. J Clin Invest 50: 481--491, 1971.
59. Jardin F, Bardet J, Sanchez A, Blanchet F, Broudarias]p, Margairaz A. Massive pulmonary
embolism without arterial hypoxemia: Pathophysiology in two cases. Intens Care Med 3: 77-80,
1977.
60. Williams MH, Adler J], Colp C. Pulmonary function studies as an aid in the differential
diagnosis of pulmonary hypertension. Am J Med 47: 378-383, 1969.
61. Nadel JA, Gold WM, Burgess JH. Early diagnosis of chronic pulmonary vascular obstruction.
Am J Med 44: 16-25, 1968.
62. Burgess]H. Pulmonary diffusing capacity in disorders of the pulmonary circulation. Circulation
49: 541-550, 1974.
63. Rees HA, Thomas Aj, Rossiter C. The recognition of coronary heart disease in the presence
of pulmonary disease. Br Hearr J 26: 233-240, 1964.
319
64. Hudson LD, Kurt TL, Petty TL, Genton E. Arrhythmias associated with acute respiratory failure
in patients with chronic airway obstruction. Chest 63: 661-665, 1973.
65. Chetty KG, Brown SE, Light RW. Identification of pulmonary hypertension in chronic
obstructive pulmonary disease from routine chest radiographs. Am Rev Respir Dis 126: 338-
341, 1982.
66. Toombs BD, Miller SW. Clinical implications of the convex supradiaphragmatic inferior vena
cava. Radiology 132: 577-581, 1979.
67. Hirschfeld S, Meyer R, Schwartz DC, Korfhagen J, Kaplan S. Measurement of right and left
ventricular systolic time intervals by echocardiography. Circulation 51: 304-309, 1975.
68. Tsuda S. Studies on right ventricular performance by noninvasive measurement of systolic and
diastolic time intervals in patients with chronic right ventricular overloading. Jap Cir J 42:
1319-1320, 1978.
69. Curtiss EI, Reddy PS, O'Toole]D, Shaver JA. Alterations of right ventricular systolic time
intervals by chronic pressure and volume overloading. Circulation 53: 997-1103, 1976.
70. Acquatella H, Schiller NB, Sharpe DN, Chatterjee K. Lack of correlation between echocardio-
graphic pulmonary valve morphology and simultaneous pulmonary arterial pressure. Am J
Cardiol 43: 946-950, 1979.
71. Prakash R: Determination of right ventricular wall thickness in systole and diastole: Echocardio-
graphic and necropsy correlations in 32 patients. Br Heart J 40: 1257-1261, 1978.
72. Berger HJ, Matthay RA. Noninvasive radiographic assessment of cardiovascular function in
acute and chronic respiratory failure. Am J Cardiol 47: 950-962, 1981.
73. Brent BN, Berger HJ, Matthay RA, Mahler D, Pytlik L, Zaret BL. Physiologic correlates of
right ventricular ejection fraction in chronic obstructive pulmonary disease: A combined ra-
dionuclide and hemodynamic study. Am J Cardiol 50: 255-262, 1982.
74. Cohen HA, Barid MG, Rouleau JR, Fuhrmann CF, Bailey IK, Summer WR, Strauss HW, Pitt
B. Thallium-201 myocardial imaging in patients with pulmonary hypertension. Circulation 54:
790-795, 1976.
75. Ohsuzu F, Handa S, Kondo M, Yamazaki H, Tsugu T, Kubo A, Takagi Y, Nakamura Y.
Thallium-201 myocardial imaging to evaluate right ventricular overloading. Circulation 61:
620-625, 1980.
76. Rigo P, Murray M, Taylor DR, Weisfeldt ML, Kelly DT, Strauss HW, Pitt B. Right
ventricular dysfunction detected by gated scintiphotography in patients with acute inferior
myocardial infarction. Circulation 52: 268-274, 1975.
77. Wong DF, Natarajan TK, Koller D, Summer W, Tibbits P, Kasecamp W, Lamb J, OlynykJ,
Philp MS, Wagner HN. Intravenous krypton-81 m for continuous monitoring of right ventricu-
lar function. J Am Coli Cardioll(2): 578, 1983.
78. Schrijen F, uffholtz H, Polu JM, Poincelot F. Pulmonary and systemic hemodynamic evolution
in chronic bronchitis. Am Rev Respir Dis 117: 25-31, 1978.
bronchitis and emphysema. Ann Inter Med 81: 307-313, 1974.
80. Herles F,Jezek V, Daum S. Site of pulmonary resistances in cor pulmonale in chronic bronchitis.
Br Heart J 30: 654-660, 1968.
81. Khaja F, Parker JO. Right and left ventricular performance in chronic obstructive lung disease.
Am Heart J 82: 319-327, 1971.
82. Taylor RR, Covell JW, Sonnenblick EH, Ross J Jr. Dependence of ventricular distensibility
on filling of the opposite ventricle. Amer J Physiol 213: 711-718, 1967.
83. Robotham JL. Cardiovascular disturbances in chronic respiratory insufficiency. Am J Cardiol
47: 941-949, 1981.
84. Unger K, Shaw D, Karliner JS, Crawford M, O'Rourke RA, Moser KM. Evaluation of left
ventricular performance in acutely ill patients with chronic obstructive pulmonary disease. Chest
68: 135-142, 1975.
85. Kline LE, Crawford MH, MacDonald WJ, Schelbert H, O'Rourke RA, Moser KM. Noninva-
sive assessment of left ventricular performance in patients with chronic obstructive pulmonary
disease. Chest 72: 558-564, 1977.
86. Turino GM, Goldring RM, Heinemann HO. Water, electrolytes and acid-base relationships in
chronic cor pulmonale. Prog Cardiovasc Dis 12: 467-483, 1970.
87. McIntyre KM, Sasahara AA. The ratio of pulmonary arterial pressure to pulmonary vascular
obstruction: Index of preembolic cardiopulmonary status. Chest 71: 692-697, 1977.
88. McIntyre KM, Sasahara AA. Determinants of right ventricular function and hemodynamics after
pulmonary embolism. Chest 65: 534-543, 1974.
89. Steele P, Ellis JH, Van Dyke D, Sutton F, Creagh E, Davies HL. Left ventricular ejection
fraction in severe chronic obstructive airways disease. Am J Med 59: 21-28, 1975.
90. SUl.iner WR, Permutt S, Sagawa K, Shoukas AA, Bromberger-Barnea B. Effects of spontane-
ous respiration on canine left ventricular function. Circ Res 45: 719-728, 1979.
91. Goldberg HS, RabsonJ. Control of cardiac output by systemic vessels: Circulatory adjustments
to acute and chronic respiratory failure and the effect of therapeutic interventions. Am J Cardiol
47: 696-702, 1981.
92. Cournand A, Motley ML, Werko L, Richards DW Jr. Physiologic studies of the effect of
intermittent positive pressure breathing on cardiac output in man. AmJ Physiol152: 162-174,
1948.
93. Whittengerber JL, McGregor M, Berglund E, Borst HG. Influence of state of inflation of the
lung on pulmonary vascular resistance. J Appl Physiol 15: 878-882, 1960.
94. Graham R, Skoog C, Oppenheimer L, Rabson J, Goldberg HS. Critical closure in the canine
pulmonary vasculature. Circ Res 50: 566-572, 1982.
95. Kumar A, Falke KJ, Green B, Aldredge CF, Laver MB, Lowenstein E, Pontoppidan H.
Continuous positive-pressure ventilation in acute respiratory failure: Effects on hemodynamics
and lung function. N Engl J Med 283: 1430-1436, 1970.
96. Wood LDH, Prewitt RM. Cardiovascular management in acute hypoxemic respiratory failure.
Am J Cardiol 47: 963-972, 1981.
97. Scharf SM, Caldini P, Ingram RH Jr. Cardiovascular effects of increasing airway pressure in
the dog. Am J Physiol 232: H35-43, 1977.
98. Jardin F, Farcot JC, Boisante L, Curten N, Margairaz A, Bourdarias JP. Influence of positive
end-expiratory pressure on left ventricular performance. N Engl J Med 304: 387-392, 1981.
99. Qvist J, Pontoppidan H, Wilson RS, Lowenstein E, Laver MB. Hemodynamic responses to
mechanical ventilation with PEEP. Anesth 42: 45-55, 1975.
100. Suter PM, Fairley HB, Isenberg MD. Optimum end-expiratory airway pressure in patients with
acute pulmonary failure. N Engl J Med 292: 284-289, 1975.
101. Sjostrand U. High-frequency positive-pressure ventilation (HFPPV): A review. Crit Care Med
8: 345-364, 1980.
102. Carlon GC, Kahn RC, Howland WS, Ray C Jr, Turnbull AD. Clinical experience with
high-frequency jet ventilation. Crit Care Med 9: 1-6, 1981.
103. Hoff BH, Smith RB, Bunegin L, Cherry D. High-frequency ventilation in dogs with open
chests. Crit Care Med 10: 517-521, 1982.
104. Knaus WA, Draper EA, Wagner DP, Zimmerman JE, Birnbaum ML, Cullen DJ, Kohles MK,
Shin B, Snyder JV. Evaluating outcome from intensive care: A preliminary multihospital
comparison. Crit Care Med 10: 491-496, 1982.
105. Sykes MK, McNicol MW, Campbell EJM (eds). Respiratory Failure. London: Blackwell
Scientific Publications, 1976.
106. Sykes MK, McNicol MW, Campbell EJM (eds). Oxygen therapy. In Respiratory Failure.
London: Blackwell Scientific Publications, 1976, pp. 127-150.
107. Campbell EJM. The management of acute respiratory failure in chronic bronchitis and emphy-
sema. Am Rev Respir Dis 96: 626-639, 1967.
108. Sluiter HJ, Blokzijl EJ, van Dijl W, van HaeringenJR, Hilvering C, Steenhuis EJ. Conservative
and respirator treatment of acute respiratory insufficiency in patients with chronic obstructive
109. Rudolph M, Banks RA, Semple SJG. Hypercapnia during oxygen therapy in acute exacerbations
of chronic respiratory failure. Lancet 2: 483-486, 1977.
110. Anthonisen NR. Pulmonary perspective: Hypoxemia and O 2 therapy. Am Rev Respir Dis 126:
729-733, 1982.
111. Aubier M, Murciano D, Milic-EmiliT, Touaty E, Dagfous J, Pariente R, Derenre JP. Effects
of the administration of O~ on ventilation and blood gases in patients with chronic obstruc-
tive pulmonary disease dunng acute respiratory failure. Am Rev Respir Dis 122: 747-754,
1980.
112. Wilson RH, Hoseth W, Dempsey ME. Effects of breathing 99.6% oxygen on pulmonary
vascular resistance and cardiac output in patients with pulmonary emphysema and chronic
hypoxia. Ann Intern Med 42: 629-637, 1955.
321
113. Haneda T, Nakajima T, Shirato KJ, Onodera S, Takishima T. Effects of oxygen breathing on
pulmonary vascular input impedance in patients with pulmonary hypertension. Chest 83:
520-527, 1983.
114. Abraham AS, Cole RB, Bishop JM. Reversal of pulmonary hypertension by prolonged oxygen
administration to patients with chronic bronchitis. Circ Res 23: 147-157, 1983.
115. Abraham AS, Kay JM, Cole RB, Pincock AC. Hemodynamic and pathological study of the
effect of chronic hypoxia and subsequent recovery of the heart and pulmonary vasculature of
the rat. Cardiovas Res 5: 95-102, 1971.
116. Simonneau G, Escourrous P, Duroux P, Lockhart A. Inhibition of hypoxic pulmonary vasocon-
striction by nifedipine. N Engl J Med 304: 1582-1585, 1981.
117. Summer WR. Acute respiratory failure (ARF). In The Principles and Practice of Medicine,
Harvey AM, Johns RJ, McKusick Va, Owens AH, Ross RS (eds). New York: Appleton-
Century-Crofts, 1980, pp. 398-407.
118. Caldini P. Pulmonary hemodynamics and arterial oxygen saturation in pulmonary embolism.
J Appl Physiol 20: 184-190, 1965.
119. Snow RL, Davies P, Pontoppidan H, Zapol WM, Reid L. Pulmonary vascular remodeling in
adult respiratory distress syndrome. Am Rev Respir Dis 126: 887-892, 1982.
120. Toung 1JK, Saharia P, Mitzner W A, Permutt S, Cameron JL. The beneficial and harmful effects
of positive end-expiratory pressure. Surgery 147: 518-524, 1978.
121. Holody B, Goldberg HS. The effects of mechanical vibration physiotherapy on arterial oxygena-
tion in acutely ill patients with atelectasis or pneumonia. Am Rev Respir Dis 123: 372-375,
1981.
122. Remolina C, Khan AU, Santiago TV, Edelman NH. Positional hypoxemia in unilateral lung
123. Martin q, Ripley H, Reynolds J, Best F. Chest physiotherapy and the distribution of ventila-
tion. Chest 69: 174-178, 1976.
124. Graham WGB, Bradley DA. Efficacy of chest physiotherapy and intermittent positive pressure
breathing in the resolution of pneumonia. N Engl J Med 299: 624-627, 1978.
125. Connors AF Jr, Hammon WE, Martin RJ, Rogers RM. Chest physical therapy. The immediate
effect on oxygenation in acutely ill patients. Chest 78: 559-564, 1980.
126. Chulay M, Brown J, Summer W. Effect of postoperative immobilization after coronary arter
bypass surgery. Crit Care Med 10: 176-179, 1982.
127. Shim C, Williams MH. Bronchial response to oral versus aerosol metaproterenol in asthma. Ann
Intern Med 93: 428-431, 1980.
128. VanArsdel PP, Paul GH. Drug therapy in the management of asthma. Ann Intern Med 87: 68-74,
1977.
129. Schild HO, Hawkins DF, Mongar JL, Heryheimer H. Reactions of isolated human asthmatic
lung and bronchial tissue to specific antigen. Lancet 2: 376-382, 1951.
130. Rossing TH, Fanta CH, McFadden ER Jr. A controlled trial of the use of single versus combined
drug therapy in the treatment of acute episodes of asthma. Am Rev Respir Dis 123: 190-194,
1981.
131. Vozeh S, Kewitz G, Perruchoud A, Tschan M, Kopp C, Heitz M, Follath F. Theophylline serum
concentration and therapeutic effect in severe acute bronchial obstruction: The optimal use of
intravenously administered aminophylline. Am Rev Respir Dis 125: 181-184, 1982.
132. Benumof JL, Trousdale FR. Aminophylline does not inhibit canine hypoxic pulmonary vaso-
constriction. Am Rev Respir Dis 126: 1017-1019, 1982.
133. Hendeles L, Weinberger M. Avoidance of adverse effects during chronic therapy with theophyl-
line. Brit J Resp Dis 61: 103-119, 1980.
134. Ouslander J. Theophylline dosage in the elderly. N Engl J Med 301: 435, 1979.
135. Vallner ], Speir WA, Jr., Kolbeck RC, Harrison NH, Bransome ED Jr. Effect of pH on the
binding of theophylline to serum proteins. Am Rev Respir Dis 120: 83-86, 1979.
136. Higbee MD, Uumar M, Galant SP. Stimulation of endogenous catecholamine release by theop-
hylline: A proposed additional mechanism of action for theophylline effects. J Allerg Clin
Immunol 70: 377-382, 1982.
137. Nelson S, Summer WR,Jakab GJ. Pulmonary antibacterial defenses are acutely suppressed by
aminophylline. Crit Care Med 11: 221, 1983.
138. Bishop RL, Weisfeldt ML. Sodium bicarbonate administration during cardiac arrest: Effect on
arterial pH, PC02 and osmolaliry. JAMA 235: 506-509, 1976.
139. Arieff AL, Leach W, Park R, Lazarowitz VC. Systemic effects of NaHC03 in experimental
lactic acidosis in dogs. Amer J Physiol 242: F586-F591, 1982.
140. Park R, Arieff AI. Treatment of lactic acidosis with dichloroacetate in dogs. J Clin Invest 70:
853-862, 1982.
141. Murray JF, Karp RB, NadelJ. Viscosity effects on pressure-flow relations and vascular resistance
in dogs' lungs. J Appl Physiol 27: 336-341, 1969.
143. Chetty KG, Brown KE, Light RW. Improved exercise tolerance of the polycythemic lung
patient following phlebotomy. Am J Med 74: 415-420, 1983.
144. York EL,Jones RL, Menon D, Sproule BJ. Effects of secondary polycythemia on cerebral blood
flow in chronic obstructive pulmonary disease. Am Rev Respir Dis 121: 813-818, 1980.
145. Szatalowicz VL, Arnold PE, Chaimovitz C, Bichet D, Berl T, Schrier RW. Radioimmunoassay
of plasma arginine vasopressin in hyponatremic patients with congestive heart failure. N Engl
J Med 305: 263-266, 1981.
146. Szatalowicz VL, GoldbergJP, Anderson RJ. Plasma antidiuretic hormone in acute respiratory
failure. Am J Med 72: 583-587, 1982.
147. Sladen A, Laver MB, Pontoppidan H. Pulmonary complications and water retention in pro-
longed mechanical ventilation. N Engl J Med 279: 448-453, 1968.
148. Wynne JW. The treatment of cor pulmonale. JAMA 239: 2283-2285, 1978.
149. Cohen CA, Zagelbauam G, Gross D, Roussos CH, Macklem PT. Clinical manifestations of
inspiratory muscle fatigue. Am J Med 73: 308-316, 1982.
150. Knochel JP. Hypophosphatemia in the alcoholic. Arch Intern Med 140: 613-614, 1980.
151. NewmanJH, NeffTA, Ziporin P. Acute respiratory failure associated with hypophosphatemia.
N Engl J Med 296: 1101-1103, 1977.
152. Gustman P, Yerger L, Wanner A. Immediate cardiovascular effects of tension pneumothorax.
Am Rev Respir Dis 127: 171-174, 1983.
153. Sharp]T, Barocas M, Chokroverty S. The cardiorespiratory effects of obesity. Clin Chest Med
1: 103-118, 1980.
154. Guilleminault C, Cummiskey J, MottaJ. Chronic obstructive airflow disease and sleep studies.
Am Rev Respir Dis 122: 397-406, 1980.
155. DeMarco FJJr, WynneJW, BlockJ, Boysen PG, Taasan VC. Oxygen desaturation during sleep
as a determinant of the "blue and bloated" syndrome. Chest 79: 621-625, 1981.
156. Guilleminault C, Kurland G, Winkle R, Miles LE. Severe kyphoscoliosis, breathing and sleep.
The "Quasimodo" syndrome during sleep. Chest 79: 626-630, 1981.
157. Strohl KP, Hensley MJ, Sanders NA, Scharf SM, Brown R, Ingram RH Jr. Progesterone
administration and progressive sleep apnea. JAMA 245: 1230-1232, 1981.
158. Smith PL, Haponik EF, Allen RP, Bleecker ER. The effect of protriptyline in sleep-disordered
breathing. Am Rev Respir Dis 127: 8-13, 1983.
159. Sullivan CE, Berthon-Issa M, Jones FG, Eves L. Reversal of obstructive sleep apnoea by
continuous positive airway pressure applied through the nares. Lancet 1: 862-865, 1981.
160. Orr WC, Martin RJ. Obstrutive sleep apnea associated with tonsillar hypertrophy in adults.
Arch Intern Med 141: 990-992, 1981.
161. Haponik EF, Smith PH, Bohlman ME, Allen RP, Goldman SM, Bleeker ER. Computerized
tomography in obstructive sleep apnea. Am Rev Respir Dis 127: 221-226, 1983.
162. Martin RJ, Sanders MH, Gray BA, Pennock BE. Acute and long-term ventilatory effects of
hyperoxia in the adult sleep apnea syndrome. Am Rev Respir Dis 125: 175-180, 1982.
163. MottaJ, Guilleminault C, SchroederJS, Dement WC. Tracheostomy and hemodynamic changes
in sleep induced apnea. Ann Intern Med 89: 454-458, 1978.
164. Fleetham J, West P, Mezon B, Conway W, Rorh T, Kryger M. Sleep, arousals and oxygen
desaturation in chronic obstructive pulmonary disease. Am Rev Respir Dis 126: 429-433, 1982.
165. Mitchell RS, Silvers GW, Dart GA, Petty TL, Vincent TN, Ryan SF, Filley GF. Clinical and
morphological correlations in chronic airway obstruction. Am Rev Respir Dis 97: 54-62, 1968.
166. Prescott SM, Richards KL, TikoffG, ArmstrongJD Jr, ShigoekaJW. Venous thromboembolism
in decompensated chronic obstructive pulmonary disease. Am Rev Respir Dis 123: 32-36, 1981.
167. Pingleton SK, Bone RC, Pingleton WW, Ruth WE. Prevention of pulmonary emboli in a
respiratory intensive care unit: Efficacy of low-dose heparin. Chest 79: 647-650, 1981.
323
168. Greenfield LJ, Jay SJ. Immediate management of massive pulmonary embolism. Chest 82:
775-778, 1982.
169. Sharma GVRK, Cella G, Parisi AF, Sasahara AA. Thrombolytic therapy. N Engl J Med 306:
1268-1276, 1982.
170. Sasahara AA. The case for fibrinolytic therapy. J Cardiovasc Med 5: 794, 1980.
171. Bomalaski JS, Martin GJ, Hughes RL, Yao JST. Inferior vena cava interruption in the manage-
ment of pulmonary embolism. Chest 82: 767-774, 1982.
172. Ferrer MI, Harvey RM, Cathcart RT, Webster CA, Richards DW Jr, Cournand A. Some effects
of digoxin upon the heart and circulation in man: Digoxin in chronic cor pulmonale. Circulation
1: 161-186, 1950.
173. Hecht HH, Kuida H, Tsagaris 1J. Brisket disease. IV: Impairment of left ventricular function
in a form of cor pulmonale. Trans Assoc Amer Physicians 75: 263-276, 1962.
174. Howarth S, McMichael J, Sharpey-Schafer EP. Effects of oxygen, venesection and digitalis in
chronic heart failure from disease of the lungs. Clin Sci 6: 187-196, 1948.
175. Haas F, Foster WM, Bergofsky EH. Direct effects of ouabain on the pulmonary vasculature
and its enhancement of the vasoconstrictor response to hypoxia. Prog Resp Res 9: 273-284,
1975.
176. Williams JF Jr, Boyd DL, Border JF. Effect of acute hypoxia and hypercapnic acidosis on
the development of acetylstrophanthidin-induced arrhythmias. J Clin Invest 47: 1885-1894,
1968.
177. Krongrad E, Helmholz HF Jr, Ritter DG. Effect of breathing oxygen in patients with severe
pulmonary vascular obstructive disease. Circulation 47: 94-100, 1973.
178. Morrison G, Macartney F. Effects of oxygen administration, bicarbonate infusions, and brief
hyperventilation on patients with pulmonary vascular obstructive disease. Br Heart J 41: 584-
593, 1979.
179. Cohn]N. Physiologic basis of vasodilator therapy for heart failure. Am J Med 71: 135-146,
1981.
180. Fanta CH, Venugopalan CS, Lacouture PG, Drazen JM. Inhibition of bronchoconstriction in
the guinea pig by a calcium channel blocker, nifedipine. Am Rev Respir Dis 125: 61-66, 1982.
181. Flaherty ]T. Intravenous nitroglycerine. Johns Hopkins Medical J 151: 36-40, 1982.
182. Flaherty ]T, Magee PA, Gardner TL, Potter A, MacAllister NP. Comparison of intravenous
nitroglycerin and sodium nitroprusside for treatment of acute hypertension developing after
coronary artery bypass surgery. Circulation 62: 1072-1077, 1982.
183. Summer W, Lappe D, Change D, Keunedy T, Frank R, Terry P, Sylvester J. Reduction of
pulmonary vascular resistance with intravenous nitroglycerin. Am Rev Respir Dis 119: 389,
1979.
184. Kennedy T, Summer WR, Sylvester]T, Robertson D. Airway response to sublingual nitroglyc-
erin in acute asthma. JAMA 246: 145-147, 1981.
185. Rutherford]D, Vatner SF, Braunwald E. Effects and mechanisms of action on aminophylline
on cardiac function and regional blood flow distribution in conscious dogs. Circulation 63:
378-387, 1981.
186. Rubin Lj, Peter RH. Hemodynamics at rest and during exercise after oral hydralazine in patients
with cor pulmonale. Am J Cardiol 47: 116-122, 1981.
187. Furman WR, Summer WR, Kennedy TP, Sylvester]T. Comparison of the effects of dobuta-
mine, dopamine and isoproterenol on hypoxic pulmonary vasoconstriction in the pig. Crit Care
Med 10: 371-374, 1982.
188. Vaage J, Piper PJ. The release of prostaglandin-like substances during platelet aggregation and
pulmonary microembolism. Acta Physiol Scand 94: 8-13, 1975.
189. Prewitt RM, McCarty J, Wood LDH. Treatment of acute lower pressure pulmonary edema
in dogs. J Clin Invest 67: 409-418, 1981.
190. Miller WC, Rice DL, Unger KM, Bradley BL. Effect of PEEP on lung water content in
experimental noncardiogenic pulmonary edema. Crit Care Med 9: 7-9, 1981.
191. Tate RM, Vanbenthysen KM, Shasby DM, McMurtry IF, Repine JE. Oxygen-radical-mediated
permeability edema and vasoconstriction in isolated perfused rabbit lungs. Am Rev Respir Dis
126: 820-826, 1982.
192. Demling RH. The pathogenesis of respiratory failure after trauma and sepsis. Surg Clin N Amer
60: 1373-1390, 1980.
193. Furrukh IS, Adkinson NF, Michael JR, Summer WR, Murray PA, Gurtner GH. Effect of
vasodilators on pulmonary vasoconstriction induced by arachidonic acid metabolites. Abstracted,

presented at Crit Care Meetings, May 1983.
194. Bresack MA, Bland RD. Alveolar hypoxia increases lung fluid filtration in unanesthetized
newborn lambs. Circ Res 46: 111-116, 1980.
195. Utsunomiya T, Krausz MM, Dunham B, Valeri CR, Levine L, Shepro D, Hechtman HB.
Modification of inflammatory response to aspiration with ibuprofen. Am J Physiol 243: H903-
910, 1982.
196. Utsunomiya T, Krausz MM, Valeri CR, et al. Treatment of aspiration pneumonia with ibu-
profen and prostacyclin. Surgery, 170-176, 1981.
197. Colley PS, Cheney FZW Jr, Hlastala MP. Pulmonary gas exchange effects of nitroglycerin in
canine edematous lungs. Anesthes 55: 114-119, 1981.
198. Prewitt RM, Wood LDH. Effect of sodium nitroprusside on cardiovascular function and
pulmonary shunt in canine oleic acid pulmonary edema. Anesthes 55: 537-541, 1981.
199. Svensjo E, Persson CGA, Rutili G. Inhibition of bradykinin-induced macromolecular leakage
from postcapillary venules by a beta2 adrenoreceptor stimulator terbutaline. Acta Physiol Scand
101: 504-506, 1977.
200. Raymond RM, Jandhyala BS, Grega GJ. The interrelationship among bradykinin, various
vasoactive substances, and macromolecular permeability in the canine forelimb. Microvas Res
19: 329-337, 1980.
201. Martin JG, Shore S, Engle LA. Effects of continuous positive airway pressure on respiratory
mechanics and pattern of breathing in induced asthma. Am Rev Respir Dis 126: 812-817, 1982.
202. Weber KT, Janicki JS, Shroff S, Fishman A. Contractile mechanics and interaction of the right
and left ventricles. Am J Cardiol 47: 686-695, 1981.
203. Mathru M, Rao TLK, El-Eter AA, Pifarre R. Hemodynamic response to changes in ventilatory
patterns in patients with normal and poor left ventricular reserve. Crit Care Med 10: 423-426,
1982.
204. Pepe PE, Marini.D. Occult positive end-expiratory pressure in mechanically ventilated patients
with airflow obstruction. Am Rev Respir Dis 126: 166-170, 1982.
205. Connors AF, McCaffree DR, Gray BA. Effect of inspiratory flow rate on gas exchange during
mechanical ventilation. Am Rev Respir Dis 124: 537-543, 1981.
206. Rankin JS, Olsen CO, Arentzen CE, Tyson GS, Maier G, Smith PK, Hammon JW, Davis JW,
McHale PA, Anderson RW, Sabiston DC Jr. The effects of airway pressure on cardiac function
in intact dogs and man. Circulation 66: 108-120, 1982.
207. Connors AF, McCaffree DR, Gray BA. Evaluation of right heart catheterization in the critically
ill patient without acute myocardial infarction. N Engl J Med 308: 263-267, 1983.
208. Shaver JA. Hemodynamic monitoring in the critically ill patient. N Engl J Med 308: 277-279,
1983.
209. Gold FL, Bache RJ. Transmural right ventricular blood flow during acute pulmonary artery
hypertension in the sedate dog. Circ Res 51: 196-204, 1982.
210. McGoon MD, Benedetto PW, Greem BM. Complications of percutaneous central venous
catheterization: A report of two cases and review of the literature. Johns Hopkins Med J 145:
1-6, 1979.
211. Mitchell SE, Clark RA. Complications of central venous catheterization. Am J Radiol 133:
467-476, 1979.
212. Swan HJC, Ganz W. Complications with flow-directed balloon-tipped catheters. Ann Intern
Med 91: 494, 1979.
213. Forrester JS, Diamond G, Chatterjee K, Swan HJC. Medical therapy of acute myocardial
infarction by application of hemodynamic subsets. N Engl J Med 295: 1356-1362, 1976.
214. Sigrist D, Thomas D, Howell S, Roussos C. The effect of aminophylline on inspiratory muscle
contractility. Am Rev Respir Dis 126: 46-50, 1982.
215. Reitberg DP, Bernhard DH, Schentag.D. Alteration of theophylline clearance and half-life by
cimetidine in normal volunteers. Ann Intern Med 95: 582-585, 1981.
12. THERAPY OF PULMONARY HEART DISEASE

and
ROBERT H. PETER, M.D.
Since impaired intrapulmonary gas exchange is the principle cause of pulmonary

heart disease secondary to parenchymal lung disease, measures directed at improv-
ing gas exchange are the mainstay of therapy for these patients. Patients with
thromboembolic disease or primary pulmonary hypertension pose more difficult
therapeutic challenges because (a) gas exchange is frequently not seriously impaired,
(b) the underlying process may no longer be reversible, or (c) the pathogenesis is
unknown, making specific therapy impossible.
A carefully planned, aggressive medical regimen can result in gratifying im-
provement in many patients with pulmonary heart disease. These patients are,
however, quite susceptible to interventions that may disrupt the delicate balance
between cardiovascular and respiratory functions. For these reasons a detailed
review of therapy is appropriate.
BRONCHODILATORS
Theophylline
Theophylline and other xanthine derivatives can improve intrapulmonary gas
exchange in patients with obstructive lung disease through several mechanisms,
including bronchodilatation with improved airflow (1), enhanced mucociliary
clearance (2), and improved diaphragmatic function (3). Aminophylline is widely
L. J. Rubin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright @ 1984 Martinus NijhoffPubiishing. Boston/The Hague/Dor-
drecht/Lancaster.
325
326 12. Therapy of Pulmonary Heart Disease
used in patients with reversible airways disease in this country; moreover, it may
also be useful in patients with nonreversible airways disease because of its other
potentially beneficial effects on gas exchange (2,3).
In addition to its effects on respiratory function, theophylline also has cardiovas-
cular effects that are noteworthy. Aminophylline produces systemic vasodilation
and increases cardiac output in patients with left ventricular failure (4). Parker et
al. (5) infused aminophylline in eight patients with left ventricular failure and
demonstrated an increase in cardiac output and a reduction in pulmonary arterial
and both right and left ventricular end-diastolic pressures. The fall in pulmonary
artery pressure in these patients, however, may have been the result of improved
left ventricular emptying rather than a direct effect on pulmonary vascular tone.
Parker et al. (6) administered aminophylline to nine patients with chronic
obstructive lung disease and cor pulmonale and found that both pulmonary arterial
and right ventricular pressures fell significantly. Heart rate increased while cardiac
index was unchanged, and mean systemic arterial pressure fell by approximately
10 mmHg. As a result pulmonary vascular resistance and right ventricular stroke
work decreased significantly. Interestingly, arterial oxygen saturation was un-
changed despite an increase in alveolar ventilation and a fall in arterial peo2• The
authors concluded from their observations that aminophylline exerted a direct
pulmonary vasodilator effect in their patients. Harris (7,8) has suggested an alterna-
tive hypothesis: the increased intrathoracic pressure present in patients with severe
airways obstruction could contribute to the hydrostatic pressures in the cardiac
chambers, and bronchodilators such as aminophylline could lower transmural
pulmonary vascular pressures by improving airflow and decreasing intrathoracic
pressure. Additionally, an indirect beneficial hemodynamic effect resulting from
improvement in gas exchange cannot be excluded.
If aminophylline is used, its administration should be monitored by serum drug
levels, and the dose should be adjusted to avoid toxicity. Patients with severe
hypoxemia and hepatic or renal dysfunction secondary to right heart failure are
particularly susceptible to adverse effects, especially arrhythmias or cardiac ectopy
(9-11).
Beta-adrenergic agonists
Isoproterenol
Isoproterenol is a potent beta-adrenergic agonist that has been used for bronchodila-
tation in patients with reversible obstructive airways disease. Isoproterenol also
exerts a pulmonary vasodilator effect in isolated perfused lungs and in intact
animals with pulmonary vasoconstriction induced by alveolar hypoxia (12,13).
Williams et al. (14) evaluated the responses to intravenous isoproterenol 5-10
mcg/min in 18 subjects with emphysema and cor pulmonale. Isoproterenol de-
creased pulmonary arterial pressure and pulmonary vascular resistance and increased
cardiac output at rest and with exercise. Mean systemic blood pressure was un-
changed, and, not unexpectedly, heart rate increased. Most interestingly, arterial
327
oxygenation improved. The magnitude of pulmonary hypertension was modest in

this series, however: the baseline mean pulmonary arterial pressure was 25 mmHg,
and the pulmonary vascular resistance was 350 dyne sec em- 5• Additionally, as
with other drugs having airways as well as vascular effects, a beneficial effect on
gas exchange as the mechanism responsible for the hemodynamic improvement
cannot be excluded. This study, nevertheless, demonstrated the potential reversibil-
ity of pulmonary heart disease and has provided the rationale for the use of
isoproterenol and other vasodilators in other forms of pulmonary heart disease.
Herrera and his colleagues (15) evaluated the effects of a single 3 mg parenteral
dose of isoproterenol in 15 patients with ventricular septal defects with pulmonary
hypertension. Four patients manifested reductions in pulmonary artery pressure and
vascular resistance concomitant with an increase in pulmonary blood flow. These
patients subsequently underwent repair of their septal defects and had an average
reduction in mean pulmonary artery pressure of 31 mmHg. The authors concluded
that isoproterenol can be used to distinguish patients with ventricular septal defects
whose pulmonary vascular dynamics may improve as a result of surgery from those
with fixed pulmonary vascular lesions.
Shettigar et al. (16) reported a patient with primary pulmonary hypertension
who manifested both symptomatic and hemodynamic improvement over a three-
year period with sublingual isoproterenol. Subsequently, Daoud and associates (17)
found that isoproterenol 0.5 to 2 mcg/min intravenously reduced the pulmonary
vascular resistance in five of six patients with primary pulmonary hypertension, but
only one patient responded to chronic therapy using sublingual isoproterenol.
Because of the brief duration of action of sublingual isoproterenol, this patient took
up to 20 tablets daily in order to maintain symptomatic improvement. Herrera et
al. (18) reported two patients who demonstrated dramatic and sustained responses
to sublingual isoproterenol although side effects limited the use of the drug to doses
no greater than 15 mg every four hours (figures 12-1 and 12-2). Person and
Proctor (19) reported a patient who responded to chronic sublingual isoproterenol
in doses of 20 mg every 2 hours. The usefulness of sublingual isoproterenol has
been limited by its short duration of action coupled with the frequent occurrence
of tremor, palpitations, and headache. Additionally, isoproterenol can paradoxi-
cally produce an increase in pulmonary arterial pressure when the inotropic and
chronotropic cardiac effects predominate over the pulmonary vasodilator effects
(17). Nevertheless, in patients who respond favorably to isoproterenol, it may be
useful in preventing or improving exertional-related symptoms.
Terbutaline
Terbutaline is a beta-adrenergic agonist with /3 2 selectivity widely used in the
management of obstructive airways disease. Stockley et al. (20) found that intrave-
nous terbutaline produced significant reductions in pulmonary vascular resistance
without affecting pulmonary artery pressures in a group of patients with chronic
bronchitis and cor pulmonale. Cardiac index and the oxygen content of mixed
venous blood increased while arterial oxygen saturation remained unchanged.
+-+
95 D-a P"<O.05
85
~
75
65 &
CJt
%
e 55
e
I:
25
f5
Figure 12-1. Relationship between mean pulmonary artery pressure (PAP) and cardiac index
(Cl) before (B) and after (A) isoproterenol. (From reference 18, with permission.)
Studies by Brent et al. (21) and by Tuele and Majid (22) suggested that terbutaline
reduces both right and left ventricular afterload in patients with obstructive airways
disease. Both of these studies reported significant reductions in pulmonary vascular
resistance and increases in cardiac output and systemic oxygen delivery in patients
with obstructive lung disease who were given subcutaneous terbutaline.
Terbutaline produced hemodynamic benefit in a patient with primary pulmo-
nary hypertension (19), but this observation has not been confirmed. We reserve
the use of terbutaline for patients with demonstrable airflow obstruction that is
reversible since side effects, particularly tremor and palpitations, are both common
and frequently bothersome to patients.
CARDIAC GLYCOSIDES
The role of cardiac glycosides in the treatment of pulmonary heart disease has been
controversial for many years (23). Ferrer et al. (24) demonstrated that acute
intravenous digoxin reduced right ventricular end-diastolic pressures in a series of
patients with cor pulmonale, and both right ventricular and pulmonary artery
329
3000
P"<0.05
2750
250
+""+
2250
\~.
10 2000
I
E
u 1750
..Do 1500
~
1250
IX
ct 1000
11.
750
200
3
C I.L.min.M 2
B......-A
--IV ~
- - - - Sublino Normol
Figure 12-2. Relationship between cardiac index (el) and pulmonary arteriolar resistance
(PAR) before (B) and after (A) infusion or sublingual administration of isoproterenol. (From
reference 18, with permission.)
pressures were lower at restudy after several weeks of combined therapy with
digitalis and intensive pulmonary toilet. Patients with cor pulmonale and normal
right ventricular end-diastolic pressure, however, have no response to digitalis
(25,26). Smith et al. (27) demonstrated that ouabain reduced right ventricular
systolic time intervals, thus suggesting that right ventricular function was im-
proved, but the significance of this finding is unclear since the right ventricular
end-diastolic pressures were normal and unchanged after ouabain in their patients.
Additionally, ouabain produced no other significant hemodynamic changes in their
patients.
Aside from studies showing no benefit from digitalis, there is reason to believe
that glycosides may be harmful in the setting of pulmonary heart disease, particu-
larly when it is the result of chronic lung disease. Digitalis produces a small but
consistent increase in pulmonary vascular resistance (26,28,29) which appears to be
a direct effect. Coates and his colleagues (30) found that chronic oral digoxin
administration in ten patients with cystic fibrosis produced no change in maximum
work capacity during bicycle exercise. They did note, however, that cardiac output
and stroke volume decreased during steady state exercise after digoxin and that
these hemodynamic alterations were accompanied by a deterioration in arterial

oxygen tension. Their observations are consistent with a digoxin-induced increase
in pulmonary vascular resistance during exercise that produced an increased right
ventricular work and diminished exercise capacity.
A number of studies have suggested that the risk of digitalis toxicity is enhanced
by the presence of cor pulmonale. Rodensky and Wasserman (31) found that 22
of 92 patients with digitalis toxicity had chronic hypoxemic lung disease, and half
of that group had clinical evidence of cor pulmonale. Beller et al. (32) found that
53% of patients with cardiac rhythm disturbances consistent with digitalis toxicity
had pulmonary disease. Holford and Mithoefer (33) reported that nearly one-half
of a series of patients with chronic obstructive pulmonary disease who underwent
72-hour electrocardiographic monitoring had arrhythmias, and 63% were taking
cardiac glycosides. Klein et al. (34) suggested that digitalis toxicity may be present
at lower serum digoxin levels in patients with pulmonary heart disease. A combina-
tion of hypoxemia, exogenous and endogenous catecholamine excess, and hypoka-
lemia from concomitant diuretic use appear to be the major predisposing factors
to digitalis toxicity in the setting of chronic pulmonary disease (23).
Is there a role for cardiac glycosides in the management of patients with cor
pulmonale? A recent study by Mathur and his associates (35) puts this problem in
perspective. Using equilibrium radionuclide angiocardiography, they showed that
digoxin had no effect on right ventricular ejection fraction when isolated right
ventricular dysfunction was present. However, digoxin therapy resulted in im-
proved right ventricular function when combined right and left ventricular failure
was present (figure 12-3). Thus, cardiac glycosides are likely to be useful only in
the setting of biventricular failure, but they should be used cautiously in patients
with chronic respiratory diseases because the risk of toxicity is increased (36). The
role of digitalis as an antiarrhythmic agent in pulmonary heart disease was discussed
in chapter 10.
DIURETICS
Diuretics have been advocated in the management of pulmonary heart disease
because they can reduce the degree of salt and water retention and may also exert
independent effects on alveolar ventilation and pulmonary vasomotor tone (37-
42). Noble et al. (37) reported improvement in both symptoms and arterial blood
gases following the administration of furosemide to four patients with acute
respiratory failure and edema, but the presence of concomitant left ventricular
dysfunction was not excluded. Whitman and his associates (38) found that diuretics
reduced central venous pressure in patients with cystic fibrosis, but they found no
consistent changes in either arterial blood gases or pulmonary hemodynamics as a
result of diuresis.
As emphasized previously (see chapter 4), the presence of peripheral edema in
the setting of chronic respiratory disease does not imply failure of the right
ventricle as a pump. Indeed, while six of the seven patients reported by Whitman
et al. (38) had elevated right atrial pressure and all seven had edema, the cardiac
331
18
l'
14
13
12
~
I ••
11
10
I ••
7
4
3
2
1
RIGHT VENTRICULAR LEFT VENTRICULAR

EJEcnON FRACTION EJECTION FRACTION
Figure 12-3. Digoxin increased right ventticular ejection fraction in the four patients (open
bar) whose left ventticular ejection fraction improved, but not in the 11 patients (closed bar)
whose left ventricular ejection fraction did not change. (From reference 35, with permission.)
index was normal or high in all but two. Additionally, decreasing intravascular
volume as a result of aggressive diuresis can reduce right ventricular preload and
decrease cardiac output (39). The striking sensitivity of the right ventricle to
alterations in preload is demonstrated by the effects of changing position: patients
with pulmonary heart disease experience marked decreases in cardiac output and
stroke volume upon assuming a sitting or upright position, due, to a large extent,
to the reduction in venous return (4~5).
Potent diuretics such as furosemide can produce hypokalemia and, by enhancing
the reabsorption of bicarbonate in the kidney, produce a metabolic alkalosis that
may be poorly tolerated by patients with chronic hypercapnia and hypoxemia (46).
Diuretics should be reserved for patients with pulmonary congestion, which may
contribute to respiratory failure, or for patients with massive volume overload and
symptomatic edema which is refractory to measures directed at improving ox-
ygenation and gas exchange. The doses used initially should be conservative, and
potassium supplements should be given as needed.
SUPPLEMENTAL OXYGEN
The goal of supplemental oxygen therapy for pulmonary heart disease is twofold:
to ameliorate symptoms related to tissue oxygen deprivation and to alleviate
1.00
c 0.90
"..
0 0.80
.
c-
o
Q. 0.70
-;;
;> 0.60
..
;>
:I 0.50
'J)
~
;> 0.40
~
:I 0.30
e:I
0.20
u
0.10
3 6 9 12 15 18 21 24 27 30 33 36 39 42 45
Time from Randomization (months)
Figure 12-4. Overall mortality in continuous Q2 therapy group (open circles) and nocturnal
02 therapy group (open squares). Ordinate is fraction of patients surviving; abscissa is time from
randomization or duration of treatment. (From reference 50, with permission.)
hypoxic pulmonary vasoconstriction. Several early studies showed hemodynamic

and clinical improvement in small groups of patients with chronic obstructive lung
disease treated with supplemental oxygen, but the hemodynamic effects were
variable and often took several weeks of oxygen use before they were achieved
(47,48).
The British Medical Research Council Study compared survival rates in patients
given oxygen for 15 hours per day with patients given no oxygen (49). The
five-year mortality was 45% in the oxygen-treated group compared to 67% in the
untreated group. Pulmonary vascular resistance tended to increase in the untreated
group while it remained unchanged or slightly decreased in the treated group.
The Nocturnal Oxygen Therapy Trial (NOTT) was a multicenter study com-
paring the effects of continuous with nocturnal oxygen therapy in chronically
hypoxemic patients (50). After a mean follow-up period of 19 months, nearly twice
as many patients receiving nocturnal oxygen died compared to the group receiving
continuous oxygen (41% vs. 22%) (figure 12-4). Measurements of motor coordi-
nation, memory, and other intellectual functions were also significantly improved
in the continuous oxygen group. Curiously, pulmonary vascular resistance de-
creased only modestly (mean decrease 11 %) after six months of continuous oxygen.
Furthermore, patients with the least disturbed pulmonary hemodynamics appeared
to derive the greatest overall benefit from oxygen.
333
+10
+5
-5
0 ..-..
..--
00
OOOCXXX>
-10 (XX)
0
Torr -15
0
-20 0
-25
0
-30 0
-35 oR
-NR
-40
l1 PAP
Figure 12-5. Changes in mean pulmo~ artery pressure (~PAP) after breathing oxygen for
24 hours in individual subjects. The PAP falls more than 5 nunHg in the responders (R) (open
circles) and remains unchanged or even rises in the nonresponders (NR) (closed circles). (From
reference 51, with permission.)
These studies demonstrate clearly that patients with hypoxemia and cor pul-
monale may benefit from the use of oxygen, at least in terms of improved survival
and intellectual function. The inclusion criteria used by the NOTT study are useful
in selecting patients who may benefit from such therapy: (a) hypoxemia, that is,
Pa0 2 less than 55 mmHg while breathing room air, (b) presence of edema, (c)
hematocrit greater than 55%, and (d) P pulmonale on electrocardiogram. Ashutosh
et al. (51) have recently suggested that a hemodynamic response to short-term
oxygen administration (a change in pulmonary artery pressure of 5 mmHg or
greater or a maximal oxygen consumption during exercise of 6.5 ml/min/kg after
24 hours of 28% oxygen) was useful in identifying patients who were likely to
improve with long-term oxygen administration and may be an additional criterion
for selecting candidates for oxygen therapy (figure 12-5).
Oxygen should be used for at least 16 to 18 hours daily, and most of its use
should be during sleep since nocturnal hypoxemia is common in chronic obstruc-
tive lung disease (52-54) and may produce arrhythmias, sudden death, or contrib-
ute to the development or maintenance of a pulmonary hypertensive state. Fre-

quently, oxygen has little hemodynamic effect, and the vascular responsiveness to
oxygen may wane with time (55). Nevertheless, continued supplemental oxygen
therapy may be appropriate in such patients for its other beneficial symptomatic
effects (50,56).
A variety of oxygen delivery systems are commercially available. For long-term
home use, oxygen concentrators have the advantages of overall cost of use, effi-
ciency, and mobility compared to cylinders, but neither system is suitable for
ambulatory use. Liquid oxygen is available in portable systems, but it is also
expensive. In our medical center the use of portable systems is reserved for active
and ambulatory patients who wish to spend a significant proportion of the day
away from home.
Little evidence has been shown that supplemental oxygen is beneficial in non-
hypoxemic patients with pulmonary heart disease. Although there have been
isolated reports of hemodynamic responses to oxygen in patients with primary
pulmonary hypertension, our experience with oxygen in this population has been
to the contrary. A recent study on the effects of oxygen on pulmonary vascular
impedance by Haneda et al. (57) included two patients with primary pulmonary
hypertension who had modest reductions in pulmonary artery pressure with oxy-
gen despite the absence of hypoxemia. Thus, determining the response to oxygen
may be useful as part of the therapeutic evaluation of patients with nonhypoxic
pulmonary heart disease.
PHLEBOTOMY
Systemic oxygen transport is often maintained in chronic hypoxic states by the
development of polycythemia. Hyperviscosity resulting from polycythemia, how-
ever, can lead to further increases in pulmonary vascular resistance. Segal and
Bishop (58) found that exercise cardiac output was improved after venesection in
some patients, and Harrison et al. (59) demonstrated a significant decrease in
pulmonary vascular resistance during exercise associated with an increased exercise
tolerance after phlebotomy. Weisse and his associates (60) have shown that isovo-
lumic reduction in hematocrit from a mean of 61 % to 50% resulted in significant
reductions in pulmonary artery pressure and pulmonary vascular resistance in 12
patients with cor pulmonale. These investigators also found that no additional
improvement was achieved by further reducing the hematocrit to the normal range
(mean value, 44%). A recent study by Chetty and his associates (61) demonstrated
improved exercise tolerance after phlebotomy in 13 polycythemic patients with
severe chronic obstructive lung disease. Although central hemodynamics were not
measured, the authors reasoned that the observed increases in exercise duration,
workload tolerated, and maximal oxygen consumption after phlebotomy were
likely to be the result of decreasing pulmonary artery pressure and increasing right
ventricular ejection fraction.
Phlebotomy is not indicated in all patients with pulmonary heart disease, even
if polycythemia is present. When polycythemia is the sequela of chronic hypox-
335
emia, as it usually is in cor pulmonale, continuous oxygen therapy will often

result in a reduction in hematocrit although significant reductions may occur
only after 6 to 12 months of oxygen use (50). When phlebotomy is entertained
for persistent or severe polycythemia that is felt to be contributing to cor pul-
monale, the hematocrit should be reduced isovolumically to levels between 50 to
55%, the level at which the improvement in blood viscosity and right ventricu-
lar function appears to outweigh the decrease in the oxygen-carrying capacity of
arterial blood (60,61).
ANTICOAGULANTS
Pulmonary thromboembolism is not only a cause of pulmonary heart disease, but
it also may contribute to worsening pulmonary hypertension and death in patients
with other etiologies for cor pulmonale (62-64). Patients with pulmonary hyper-
tension are susceptible to venous thrombosis because stasis and increased blood
volume are frequently present. The sluggish pulmonary blood flow may also
predispose to pulmonary vascular thrombosis in situ. Whatever the responsible
mechanism, even a small additional vascular obstruction produced by thrombosis
is poorly tolerated by patients with preexistant compromise of the pulmonary
circulation.
Most authors recommend chronic anticoagulation in patients with pulmonary
heart disease secondary to chronic recurrent thromboembolism. Unless contrain-
dicated, oral anticoagulation with warfarin is the treatment of choice, but bleeding
complications occur frequently (65,66). Experience to date with chronic subcutane-
ous heparin is limited, but recent studies suggest that adjusted-dose heparin (ad-
justed to obtain a partial thromboplastin time one and one-half times control) may
be as effective as warfarin in preventing recurrence and is associated with a lower
incidence of adverse effects (67).
The clinical differentiation between primary pulmonary hypertension and mi-
cro-theomboembolic disease is difficult at best and often impossible. Additionally,
patients with primary pulmonary hypertension tend to die suddenly (68,69), sug-
gesting the possibility of acute thrombosis in the pulmonary vasculature. Although
anticoagulation does not produce any hemodynamic effects, a recent retrospective
study from the Mayo Clinic suggests that survival may be prolonged with an-
ticoagulation in primary pulmonary hypertension (70). These data are preliminary,
and additional studies are needed to resolve this issue. We do not anticoagulate
patients with parenchymal lung disease and cor pulmonale unless thromboembol-
ism is documented.
One of the most difficult diagnostic problems facing the clinician caring for
acutely ill patients with chronic pulmonary heart disease is the question of an acute
thromboembolism versus an exacerbation of the underlying disease. Symptoms,
such as dyspnea or even hemoptysis, are not diagnostic, and worsening hypoxemia
or signs of increasing right-sided heart failure may be due to progression of the
primary process. Electrocardiography, chest radiography, and ventilation-perfusion
lung scanning are not generally helpful in this setting. Although both right heart
catheterization and pulmonary arteriography are hazardous in patients with pulmo-

nary hypertension (68,69), selective arteriography performed with hand injections
can frequently be done safely. A trial of anticoagulation may be considered when
the risks of performing the procedure appear to outweigh the potential benefit and
the index of suspicion for thromboembolism is high. The use of thrombolytic
agents is particularly appealing when dealing with patients with acute thromboem-
bolism superimposed on chronic pulmonary heart disease, but they should be
reserved for patients with documented thromboembolism because of the risk of
major hemorrhage.
PULMONARY REHABILITATION
Physical conditioning has been demonstrated to improve work capacity in patients
with chronic obstructive pulmonary disease, but the results of the hemodynamic
effects of rehabilitation in patients with cor pulmonale have been disappointing.
Although the total work performed is increased at a lower maximal heart rate,
several studies have failed to demonstrate any hemodynamic benefit from an
aggressive program of physical conditioning using either a stationary bicycle or
walking treadmill exercise (71,72). Nevertheless, patients may feel better, and they
should be encouraged to perform submaximal exercise to the level their symptoms
will allow, preferably as part of a monitored rehabilitation program.
VASODILATORS
The earliest clinical descriptions of the hemodynamic abnormalities in patients with
pulmonary heart disease included descriptions of attempts to lower pulmonary
arterial pressure using systemic vasodilators. The development of potent systemic
vasodilators led to their use in treating left ventricular failure, and this approach
has been applied with increasing frequency to conditions affecting exclusively the
right side of the heart and the pulmonary circulation. The rationale for vasodilator
therapy for pulmonary heart disease is based on the demonstration that right
ventricular dysfunction results from an increased afterload rather than an impair-
ment in contractility (73,74), and that pulmonary vasoconstriction may constitute
a variable but often significant component of the elevation in pulmonary vascular
resistance (75). A drug that decreases pulmonary vasomotor tone would reduce
right ventricular afterload and increase pulmonary blood flow. Systemic vascular
resistance is likely to be reduced by vasodilators as well, but a minimal change may
occur in systemic blood pressure if cardiac output increases. Although no specific
or even preferential pulmonary vasodilator has been identified, several vasodilators
have been used to treat patients with pulmonary heart disease resulting from a
variety of causes, including conditions in which active vasoconstriction would not
be expected to playa major role. Since the mechanisms responsible for the develop-
ment of pulmonary vascular disease are incompletely understood and are likely to
be multiple, therapy with vasodilators remains empiric and is frequently unsuccess-
ful.
337
Acetylcholine
Acetylcholine was one of the first agents shown to produce pulmonary vasodilata-
tion. Harris (76) found that intrapulmonary infusion of acetylcholine produced
transient reductions in pulmonary artery pressure in patients with pulmonary
hypertension resulting from a variety of causes. Systemic arterial pressure was not
affected, presumably due to the rapid inactivation of acetylcholine by circulating
cholinesterase. Fritts and his co-workers (77) demonstrated that a continuous
intrapulmonary infusion of acetylcholine in normals breathing 12% oxygen re-
duced pulmonary artery pressure and Behnke et al. (78) found that acetylcholine
infused into the pulmonary artery at a rate of 2 to 8 mg/min produced reductions
in rest and exercise pulmonary artery pressures in patients with emphysema and
pulmonary hypertension. Marshall (79) found that intrapulmonary acetylcholine
infused at a rate of 2 mg/min produced a 60% reduction in total pulmonary
resistance in a patient with primary pulmonary hypertension. Samet et al. (80) and
Wood (81) reported similar results in two and five of six patients, respectively,
with primary pulmonary hypertension. These studies suggested that vasoconstric-
tion is present and at least transiently reversible in some patients with this disease.
Samet and Bernstein (82) subsequently reported a patient who was unresponsive
to acetylcholine three years after an initial catheterization demonstrated vasoreac-
tivity with acetylcholine.
The obvious limitations to the clinical utility of acetylcholine lie in its rapid
inactivation in the blood and the lack of a suitable orally active analog. Neverthe-
less, the demonstration of pulmonary vascular responsiveness to acetylcholine
confirmed that some forms of pulmonary heart disease are characterized by varying
degrees of potentially reversible vasoconstriction. These observations prompted and
justified the search for more effective and practical pulmonary vasodilating agents.
Alpha-adrenergic antagonists
Tolazoline
Tolazoline (priscoline) is an alpha-adrenergic antagonist and vasodilator that has
been used to reduce pulmonary vascular resistance in a variety of disease states. In
1951 Dresdale and his associates (83) reported that two 50 mg doses of tolazoline
produced a marked reduction in pulmonary arterial pressure and vascular resistance
in a patient with primary pulmonary hypertension. Three years later the same
group reported similar hemodynamic results with tolazoline administered to a
patient with emphysema and cor pulmonale and to a patient with an atrial septal
defect and Eisenmenger's syndrome (84).
Widimsky and his colleagues (85) infused tolazoline 10 to 30 mg into the
pulmonary artery in 15 patients with chronic lung disease, with mean pulmonary
arterial pressures ranging from 16 to 68 mmHg, and found that pulmonary arterial
pressure fell and cardiac output increased in almost every patient. Kelminson et al.
(86) reported that parenteral tolazoline produced marked reductions in pulmonary
arterial pressure in six patients with cystic fibrosis and pulmonary heart disease.
Additionally, tolazoline combined with breathing 100% oxygen produced better

results than when either modality was used alone.
The usefulness of tolazoline has three limitations. First, the published experience
is limited with tolazoline administered by any route other than intrapulmonary or
intravenous. We have seen one patient who responded to intravenous tolazoline
but had no hemodynamic or symptomatic response to tolazoline 50 mg orally four
times daily for five months. Second, as with every vasodilator reported to be useful
in pulmonary heart disease, profound and refractory systemic hypotension can
develop after parenteral tolazoline. Finally, arterial oxygenation can deteriorate
after the administration of tolazoline (85). The mechanisms responsible for
vasodilator-induced hypoxemia and its implications are discussed subsequently in
this chapter.
Phentolamine
Phentolamine is an alpha-receptor antagonist that is biologically active after oral
administration. Gould and his associates (87) found that an infusion of phentola-
mine in patients with emphysema and cor pulmonale produced significant reduc-
tions in pulmonary vascular pressures and resistance, but the hemodynamic abnor-
malities in his patients were modest (baseline mean pulmonary artery pressure 22.8
mmHg and pulmonary vascular resistance 240 dyne sec cm- S). Cardiac index and
arterial blood gases did not change, and mean systemic blood pressure was main-
tained, but pulmonary capillary wedge pressure fell. Van Mieghem et al. (88)
infused phentolamine in 13 patients with chronic obstructive lung disease and cor
pulmonale and reported that a three-day infusion produced significant reductions
in pulmonary artery pressure and vascular resistance without changing arterial
blood gases or spirometry. Hemodynamic parameters returned to baseline after the
infusion was discontinued. Side effects, including dizziness, nausea, diarrhea, and
"nervousness," developed in five patients.
Ruskin and Hutter (89) demonstrated hemodynamic improvement in response
to phentolamine 5 mg intravenously over 10 minutes in a 29-year-old man with
primary pulmonary hypertension who was evaluated both at rest and during
exercise. This patient was subsequently treated with phentolamine 50 mg by mouth
every three hours, to a total dose of250 mg per day, and reported improved activity
tolerance. The hemodynamic response was sustained at repeat catheterization after
seven months of oral therapy and symptomatic improvement persisted at 20-month
follow-up. Although this report was encouraging, subsequent experience with
phentolamine was disappointing, showing a paradoxical increase in pulmonary
artery pressure with a clinical decline, intolerable side effects, fatal hypotension, or
no objective benefit (90,91).
Nitrates
Knapp and Gmeiner (92) first reported that nitroprusside infused in nine patients
with pulmonary hypertension caused a dose-related decline in pulmonary artery
339
pressure. Left atrial or pulmonary capillary wedge pressures were not measured,
however. Additionally, cardiac output did not change while heart rate increased.
The fall in stroke volume observed is, therefore, consistent with a decreased preload
resulting from the venodilating properties of nitroprusside.
Dantzker and Bower (93) administered nitroprusside to five patients with
chronic pulmonary heart disease secondary to recurrent pulmonary thromboembol-
ism and reported significant reductions in pulmonary artery pressure and pulmo-
nary vascular resistance. Three of these patients were treated chronically with
long-acting nitrate preparations, either nitropaste or isosorbide dinitrate, with
improved exercise tolerance and a reduction in symptoms. Fourrier et al. (94)
treated 13 patients with cor pulmonale and acute respiratory failure with intrave-
nous nitroglycerin 1.5 mg for 1 to 4 minutes and noted that pulmonary arterial
pressure fell. Cardiac index decreased consistent with a fall in preload, and the net
effect appears to have been a reduction in systemic oxygen transport. Similarly,
Konietzko et al. (95) reported that isosorbide 5 mg by mouth administered to 10
patients with pulmonary heart disease secondary to pulmonary fibrosis resulted in
a decline in pulmonary artery pressure at rest and during exercise, but the alveolar-
arterial oxygen difference (A-aDO~ increased and cardiac output decreased
slightly.
Although nitrates may be useful in selected individuals, cardiac output appears
to fall frequently with these agents due to their effects on venous capacitance.
Additionally, hypoxemia occurs commonly with their use (96-101).
Diazoxide
Diazoxide is a potent vasodilator that has been used to treat acute hypertensive
emergencies (102). Wang et al. (103) infused up to 515 mg of diazoxide into the
pulmonary artery in three patients with primary pulmonary hypertension and
demonstrated that cardiac output increased and pulmonary artery pressure fell. One
patient who had thromboembolic disease had no response to diazoxide. Two
patients were treated with oral diazoxide in divided doses of up to 600 mg daily
and were symptomatically improved. Klinke and Gilbert (104) reported similar
results in a patient treated with oral diazoxide.
Honey et al. (105) evaluated the responses to diazoxide in nine patients with
primary pulmonary hypertension. They found that pulmonary vascular resistance
fell when up to 600 mg were given parenterally. Of seven patients treated with
oral diazoxide in divided daily doses ranging from 150 to 600 mg, five had side
effects that precluded continuation of therapy. Only two patients had symptomatic
and hemodynamic improvement without limiting side effects. Hall and Petch (106)
reported a patient in whom diazoxide returned all hemodynamic parameters to
normal and prior symptoms abated. Upon discontinuation of the drug after 15
months of treatment, symptoms promptly recurred and hemodynamic abnormali-
ties were present once again. These investigators suggested that treatment may be
successful in the early stages of the disease, before structural abnormalities predomi-
nate over vasoconstriction.
The hazards of diazoxide administration, detailed in several studies, include

profound hypotension and death after intravenous or intrapulmonary injection
(91,107,108).
Hydralazine
Hydralazine is a vasodilator that has been used as both an acute and chronic
antihypertensive agent. In 1956 Judson et al. (109) showed that hydralazine 14 to
50 mg given intravenously increased both cardiac output and renal blood flow in
seven patients with congestive heart failure and in four patients with chronic
pulmonary disease. Total pulmonary resistance fell after hydralazine in three of the
four patients with lung disease although pulmonary arterial pressure actually
increased from 30 to 37 mmHg.
In 1980 we reported four patients with primary pulmonary hypertension who
responded to oral hydralazine 50 mg four times daily (110). Pulmonary vascular
resistance fell as cardiac output increased while pulmonary arterial pressure was
essentially unchanged. The hemodynamic effects persisted at repeat cardiac catheter-
ization after three to six months of oral therapy. Clinical improvement was noted
as well, including a reduction in breathlessness and the frequency of exertional chest
pain or syncope.
Herrera and his associates (111) evaluated the acute effects of hydralazine 0.33
mg/kg injected into the pulmonary artery in twelve patients with primary pulmo-
nary hypertension and found that six experienced a significant reduction in pulmo-
nary vascular resistance. These patients were treated with oral hydralazine, and all
showed clinical improvement at follow-up of up to eight months. Five of these
patients underwent repeat cardiac catheterization after eight months of chronic
treatment, which showed persistence of the hemodynamic effects.
We reported 12 patients with cor pulmonale secondary to chronic lung disease
in whom oral hydralazine improved pulmonary hemodynamics both at rest and
during exercise (45). A recent study by Brent and his associates (112) demonstrated
that hydralazine decreased pulmonary vascular resistance and increased cardiac
output and right ve~tricular ejection fraction (measured by radionuclide angiocar-
diography) in a group of stable patients with chronic obstructive pulmonary disease
and cor pulmonale. Similar observations have been made in isolated cases of
pulmonary heart disease resulting from talc granulomatosis (113), systemic lupus
erythematosis (114), scleroderma (115), massive acute pulmonary thromboembol-
ism (116), and the eRST syndrome (117). Hydralazine has also been shown to
reduce right ventricular end-diastolic pressure in patients with pulmonary heart
disease (118) (figure 12-6), possibly by either promoting a diuresis secondary to
increased renal blood flow or decreasing the regurgitant volume of tricuspid or
pulmonic valve insufficiency.
As with all other vasodilators, the results with hydralazine are not uniformly
favorable, and the administration of this drug to patients with pulmonary heart
disease is not without risk (119,120). Packer et al. (121) reported 13 patients with
precapillary pulmonary hypertension who were given up to 600 mg of hydralazine.
341
~ MEANtSD
N·14
25 p<O.0001
20
C'
:::I:
E 15
E
a..
o
w 10
>
a::::
CONTROL HYDRALAZINE
Figure 12-6. Right ventricular end-diastolic pressure (RVEDP) before (control) and after
hydralazine. 50 mg orally every 6 hours for 48 hours. (From reference 118. with permission.)
Although pulmonary vascular resistance fell by 21 %in the group, several patients
experienced severe side effects, and fatal systemic hypotension occurred in one
patient. Only one patient who was able to tolerate oral hydralazine experienced
clinical improvement.
Calcium channel blockers

The calcium channel blockers inhibit the contractile activity of vascular smooth
muscle and dilate systemic and coronary arteries (122). The three calcium channel
blockers currently available (nifedipine, verapamil, and diltiazem) have somewhat
different effects on vascular smooth muscle, cardiac contractility, and the cardiac
conduction pathways (123). Both nifedipine and verapamil have been shown to
inhibit hypoxia-induced pulmonary vasoconstriction in isolated lung preparations
(124,125).
Landmark et al. (126) found no consistent change in pulmonary vascular re-
sistance when 2 to 12 mg of verapamil were given to thirteen patients with
pulmonary hypertension (nine with primary pulmonary hypertension, two with
Eisenmenger's syndrome, and one with pulmonary fibrosis). The fall in mean
pulmonary artery pressure from 57 to 50 mmHg was largely due to a decrease

in cardiac output.
Diltiazem lowered pulmonary arterial pressure and increased cardiac output in
a young woman with primary pulmonary hypertension (127). The hemodynamic
response to 10 mg intravenously was maintained with chronic treatment using 30
mg orally three times daily. After 11 months of treatment, the patient was asymp-
tomatic, and radiographic and electrocardiographic abnormalities were no longer
present. Crevey et al. (128) evaluated the hemodynamic and gas exchange effects
of parenteral diltiazem in five patients with severe pulmonary heart disease (four
with primary pulmonary hypertension and one with recurrent pulmonary throm-
boembolism) and found minimal effects at rest or during exercise. Gas exchange
was not affected by the administration of diltiazem.
Camerini et al. (129) found that nifedipine 20 mg sublingually produced a
marked decrease in pulmonary vascular resistance and a modest fall in pulmonary
artery pressure in a young woman with primary pulmonary hypertension. A total
oral dose of 100 mg per day produced symptomatic improvement that persisted
during a follow-up period of three months. DeFeyter et al. (130), Wise (131), and
McLeod et al. (132) have reported similar results, but Berkenboom and his associ-
ates (133) have also reported a failure using nifedipine.
Our experience with nifedipine in the treatment of primary pulmonary hyper-
tension has been mixed: of nine patients evaluated with sublingual nifedipine, six
had responses, two had no response, and one paradoxically raised pulmonary
arterial pressure as cardiac output increased (134). The six acute responders have
been treated with oral nifedipine 40 to 120 mg per day; five had persistent
hemodynamic and symptomatic improvement while one patient did not.
The potential usefulness of calcium channel blockers in treating patients with
hypoxic pulmonary hypertension was demonstrated by Simonneau and his col-
leagues (135) who gave 20 mg sublingually to 13 patients with chronic bronchitis
and cor pulmonale. Pulmonary artery pressure fell and cardiac output increased in
most patients. No additional effects were found when supplemental oxygen was
added, suggesting that nifedipine dilated pulmonary vessels constricted by hypoxia.
Young et al. (136) have compared the effects of the three calcium channel
blockers on enhanced pulmonary vascular tone in dogs, produced either by hypoxia
or the infusion of prostaglandin F2u • They concluded that nifedipine was a more
potent pulmonary vasodilator than either verapamil or diltiazem. This also appears
to be the case in published clinical studies thus far.
Captopril
Captopril, an inhibitor of angiotensin-converting enzyme, is now frequently used
to treat systemic hypertension and left ventricular failure. Ikram et al. (137)
recently reported that captopril reduced pulmonary arterial pressure in five patients
with primary pulmonary hypertension, although cardiac output decreased in three.
Right ventricular ejection fraction, estimated using radionuclide scintigraphy,
increased from 39.5% to 59.8% after captopril, and exercise tolerance was im-
343
proved in one patient. Kokubu et al. (138) reported similar effects in one patient,
and Niarchos et al. (139) reported a 26% fall in pulmonary resistance after a single
dose of captopril in six patients with pulmonary hypertension "secondary to
collagen vascular disease."
Rich et al. (140) and Leier and associates (141) found that captopril produced
no pulmonary hemodynamic effects in their patients with primary pulmonary
hypertension. They concluded that since the major effects of captopril were on the
systemic vascular bed, patients with pulmonary hypertension were unlikely to
respond to this drug.
Prostaglandins
The prostaglandins are 20 carbon fatty acids produced in a variety of tissues,
including the lungs (142, 143). Both prostaglandins 12 (PGI 2, prostacyclin) and E1
(PGE 1) produce pulmonary vasodilatation in experimental pulmonary hyperten-
sion (144,145), and prostaglandins have been suggested to playa role in the
physiologic modulation of pulmonary vasomotor tone (146).
Watkins et al. (147) demonstrated that the infusion of prostacyclin into the right
atrium of an eight-year-old girl with pulmonary hypertension produced a marked
reduction in pulmonary vascular resistance. Rubin and his colleagues (148) infused
prostacyclin in doses ranging from 2 to 12 ng/kg/min in seven patients with
primary pulmonary hypertension and found that prostacyclin produced a dose-
related pulmonary vasodilation (figure 12-7). Prostacyclin had no specificity for
the pulmonary vascular bed, however, and systemic vascular resistance decreased
as well. The hemodynamic effects of prostacyclin were maintained during a 24-
to 48-hour infusion in three patients, but hemodynamic parameters returned to
baseline within minutes of discontinuation of the infusion (figure 12-8).
Naeije and his associates (149) administered PGE 1 intravenously to 26 patients
with decompensated chronic obstructive lung disease and cor pulmonale and
demonstrated a reduction in the degree of pulmonary hypertension without a
deterioration in gas exchange. Interestingly, the effects of PGE 1 were blunted in
patients on mechanical ventilation, and PGE 1 infusion in normal volunteers breath-
ing 12% oxygen did not block hypoxic pulmonary vasoconstriction.
Prostaglandin E1 has also been used to reduce elevated pulmonary vascular
resistance in patients with primary pulmonary hypertension (150), scleroderma
(151), and mitral valve disease (152).
As with acetylcholine the clinical utility of prostaglandins is limited since no
orally active analogue is currently available and the effects of an intravenous
infusion are dissipated within minutes of discontinuation. These agents may be
useful, however, in the management of acute pulmonary hypertensive states or in
evaluating pulmonary vasoreactivity in patients prior to administering longer
acting, nontitratable vasodilators (148).
What is the role of vasodilator therapy in the management of patients with
pulmonary heart disease? Despite a wealth of literature on the subject, it remains
to be defmed. Although some patients clearly demonstrate both hemodynamic and
TOTAL SYSTEMIC TOTAL PULMONARY
RESISTANCE RESISTANCE
+30 +30
+20 +20 0
! 1LI
X
1LI +10 (!)
+10
(!)
z 0 i z 0
I0
C( C(
::J: :I: t A
0 -10 t 0 -10
....Z 0
....Z 0
•
•
x x
1LI
0
-20
I
0
1LI
0 -30
-20
i
A
•
t
0: 6 0: 0
UJ 1LI i A
Q.. Q. -40 0
0 A
0
X a -!SO & 0
0
A A
-60
0 A
0 2 4 8 8 10 12 0 2 4 8 8 10 12
PG1 2 (no/ko/min) PGI Z (no/ko/min)
Figure 12-7. Effects of intravenous prostacyclin on total systemic and total pulmonary
resistance. Values are percentage changes compared with control. The 0 ng/kg/min dose
represents the effects of glycine buffer (vehicle). Symbols represent individual patients studied.
{From reference 148, with permission.}
3°1
20
,t
I
18 I
I ~ 8 no/ko/min.
16 I 0---0 8 no/ko/min.
I
I 6----6 10 no/ko/min.
-
e;; 14 I
c: 12
;:,
r
~
a::: 10
....D.. 8 I
I
P
I I
I I
6 /
I
I
4
0 8 16 24 32 40 48
HOURS OF INFUSION
Figure 12-8. Effects of a continuous infusion of prostacyclin on total pulmonary resistance

(TPR) in three patients. Interrupted lines represent discontinuation of the infusion and
subsequent control measurements. (From reference 148, with permission.)
345
symptomatic improvement in response to vasodilators, other patients clearly do

not. Most likely, vasoconstriction predominates in some patients whereas others
have irreversible vascular deformation. It is virtually impossible to determine
which lesion predominates in an individual patient prior to evaluating responses
to vasodilators. Although Herrera et al. (111) postulated that those patients with
the most severe derangements in pulmonary hemodynamics were the least likely
to respond to vasodilators, suggesting the presence of "ftxed" pulmonary vascular
obstruction, this may not necessarily be the case; several studies have shown
hemodynamic responses to vasodilators in patients with degrees of pulmonary
hypertension comparable to Herrera's nonresponders (103-105,129,148).
Open-lung biopsy has been advocated both to clarify the etiology in suspected
cases of primary pulmonary hypertension and to determine the likelihood of
vasoreactivity based on the vascular lesions present (153). No evidence has been
offered to date, however, that the pathologic ftndings correlate with hemodynamic
data or clinical outcome in patients without intracardiac shunts. Additionally,
surgical procedures performed under general anesthesia are poorly tolerated by
patients with pulmonary heart disease (69,154). These points must be weighed
against the potential beneftt of making a pathologic diagnosis in an individual
patient.
Several other questions concerning vasodilator use remain unanswered: Which
agent is best for long-term therapy? What are the effects of chronic therapy on
activity tolerance, quality of life, and, most important of all, survival? Which
hemodynamic response to vasodilators can be considered to be truly "beneftcial"?
The extensive experience with vasodilators in recent years, both good and bad,
suggests guidelines for the use of these agents:
a. Patients with the diagnosis of primary pulmonary hypertension are considered

candidates by us for vasodilator therapy once the diagnosis is made since no other
modality of treatment has been demonstrated, even inconsistently, to be useful. In
contrast, patients with chronic pulmonary disease and cor pulmonale are considered
to be candidates only after there is both clinical and hemodynamic evidence of
sustained pulmonary hypertension despite an aggressive regimen aimed at treating
the underlying disease.
b. Right heart catheterization is performed in order to document the presence
and severity of pulmonary heart disease and to evaluate the hemodynamic responses
to drug administration.
c. The risk of acute adverse effects is greatest when long-acting, nontitratable
vasodilators are given parenterally. Our preference is to administer these drugs by
mouth, starting with small doses and gradually increasing the dose to reasonable
levels based on published experience, as the patient's clinical course allows. Alterna-
tively, the acute use of titratable, short-acting potent intravenous agents such as
prostacyclin may be useful, if available, in safely determining a patient's vasoreac-
tivity.
d. It seems clear that patients may respond to one orally active agent but not
to another. Although evaluation of the responses to several potentially useful drugs

is useful in an individual patient, one must exhibit care in avoiding the overlap
of drugs with additive vascular effects. We prefer to perform repeat studies after
the effects of a previous drug are clearly no longer present.
e. We consider a beneficial hemodynamic response to a vasodilator to be present
if pulmonary vascular resistance falls by 20% or more and cardiac output increases
or remains unchanged and if mean pulmonary arterial pressure decreases or remains
unchanged. Clearly, the ideal response is an increase in cardiac output accompanied
by a decrease in pulmonary artery pressure, but this is frequently not the case. We
have found that right ventricular end-diastolic pressure fell after hydralazine even
when pulmonary arterial pressure was unchanged, and the reduction in end-
diastolic pressure correlated with the reduction in pulmonary vascular resistance
(118). Similar observations have recently been made by Brent et al. (112) using
radionuclide angiocardiographic techniques to evaluate right ventricular function.
The potential deleterious effects of vasodilators on gas exchange is a vexing

problem. The mechanism responsible appears to be a worsening in ventilation-
perfusion ratios (VA/Q) and/or shunt as a result of reducing pulmonary vascular
resistance and increasing blood flow to poorly ventilated lung units (iSS). On the
other hand, the detrimental effect of increasing blood flow to areas of poor
ventilation can be offset by an increase in mixed venous oxygen content resulting
from an increase in total cardiac output. Several studies have shown that arterial
oxygenation can be preserved or even improved with some vasodilators (150,156)
(figure 12-9) while others, particularly the nitrates, consistently worsen arterial
oxygenation, albeit to a variable degree (96,97,99-101). Clearly, a drug that
worsens arterial oxygenation or decreases systemic oxygen transport has limited
usefulness in a patient with chronic hypoxic pulmonary heart disease, even if it
reduces pulmonary vascular resistance or pressure.
SURGICAL TREATMENT
Pulmonary endarterectomy for chronic, recurrent thromboembolism (157-158)
and heart-lung transplantation (159-160) have been performed successfully in a
small number of patients. This approach is not uniformly successful and can be
performed at present at only a few institutions. Nevertheless, these techniques offer
promise and should be seriously considered for patients with end-state pulmonary
heart disease which is refractory to all other therapeutic approaches.
CONCLUSION
The management of pulmonary heart disease requires the physician to appreciate
the pathophysiology of the cardiopulmonary insult and to be familiar with the
pharmacology of the agents used. The primary approach to therapy remains di-
rected at treating the underlying disorder. Other, more controversial interventions,
such as vasodilator therapy and surgical approaches, should be reserved for patients
347
1.0 LIM IN
•...
0
IL
N.R.'} 81 BEFORE NIFEDlPINE
..... mee.ured, 42
V
V~: 32 ...
Pe02 mmHI
Q 'predlcted,41
0
0
iii .
05 SHUNT
1
~
0 VENTILATION -
Z .20 ...
0
~
...C
~
..,Z BLOOD FLOW,
.
>
0
i i i i i i
0 0.01 0.1 1.0 10.0 100.0
VENTILATION· PERFUSION RATIO
1.0 L/MIN
•...
0
IL
N. R. '} 81 AFTER NIFEDIPINE
..... mee.ured, 41
V
VD :,1'I.
T
Pe°2 mmHI
0 ..... predIcted, 48
0
•... G.,
0
VENTILATION -
~
0
z SHUNT
0
~ 1
...
c
• 10 ...
~
z BLOOD
~
0
0 0.01 0.1 1.0 10.0 100.0
VENTILATION· Pl!RFUSION RATIO
Figure 12-9. VA/Q distributions before and after nifedipine, 20 mg sublingually, in a patient
with primary pulmonary hypertension. (From reference 150, with permission.)
who fail to respond to these measures and should be performed by individuals who
are familiar and experienced with these techniques.
REFERENCES
1. Paterson JW, Woolcock AJ, Shenfield GM. Bronchodilator drugs. Am Rev Respir Dis 120:
1149-1188, 1979.
2. Serafini SM, Wanner A, Michaelson ED. Mucociliary transport in central and intermediate size
airways: Effect of aminophyllin. Bull Eur Physiopathol Respir 12: 415-422, 1976.
3. Aubier M, Troyer AD, Sampson M, Macklem PT, Roussos C. Aminophylline improves
diaphragmatic contractility. N Engl J Med 305: 249-252, 1981.
4. Darby TD, Sprouse JH, Walton RP. Evaluation of sympathetic reflex effects on the inotropic
action of nitroglycerin, quinidine, papaverine, aminophylline and isoproterenol. J Pharmacol
Exp Ther 122: 386--395, 1958.
5. Parker JO, Kelly G, West RO. Hemodynamic effects of aminophylline in heart failure. Am
J Cardiol 17: 232-239, 1966.
6. Parker JO, Kekar K, West RO. Hemodynamic effects of aminophylline in cor pulmonale.
Circulation 33: 17-25, 1966.
7. Harris P. Relation of airways resistance and pulmonary vascular resistance. Bull Physiopath
Respir 4: 65-73, 1968.
8. Harris P. Principles of management of cor pulmonale. Chest 58 (Suppl 2): 437-440, 1970.
9. PowellJR, Vozeh S, Hopewell P, Costello J, Sheiner LB, Riegelman S. Theophylline disposi-
tion in acutely ill hospitalized patients. The effect of smoking, heart failure, severe airway
obstruction and pneumonia. Am Rev Respir Dis 118: 229-238, 1978.
10. Horowitz LN, Spear JF, Moore EN, Rogers R. Effects of aminophylline on the threshold for
initiating ventricular fibrillation during respiratory failure. Am J Cardiol 35: 376-379, 1975.
11. Banner AS, Sunderrajan EV, Agarwal MK, Addington WW: Arrhythmogenic effects of orally
administered bronchodilators. Arch Intern Med 139: 434-437, 1979.
12. Silove ED, Inoue T, Grover RF. Comparison of hypoxia, pH, and sympathomimetic drugs on
bovine pulmonary vasculature. J Appl Physiol 24: 355-365, 1968.
13. Rubin LJ, Lazar ]D. Nonadrenergic effects of isoproterenol in dogs with hypoxic pulmonary
vasoconstriction. J Clin Invest 71: 1366-1374, 1983.
14. Williams JF, White DH, Behnke RH. Changes in pulmonary hemodynamics produced by
isoproterenol infusion in emphysematous patients. Circulation 28: 396-403, 1963.
15. Herrera EL, Sandoval J, Seoane M, Bialostozky D, Attie F. The role of isoproterenol in the
preoperative evaluation of high-pressure, high-resistance ventricular septal defect. Chest 81:
42-46, 1982.
16. Shettigar UR, Hultgren HN, Specter M, Martin R, Davies DH. Primary pulmonary hyperten-
sion. Favorable effect of isoproterenol. N Engl J Med 295: 1414-1415, 1976.
17. Daoud FS, ReevesJT, Kelly DB. Isoproterenol as a potential pulmonary vasodilator in primary
pulmonary hypertension. Am J Cardiol 42: 817-822, 1978.
18. Herrera EL, Bialostozky D, Sobrino A. The role of isoproterenol in pulmonary artery hyperten-
sion of unknown etiology (primary). Chest 79: 292-296, 1981.
19. Person B, Proctor RJ. Primary pulmonary hypertension. Responses to indomethacin, terbutaline
and isoproterenol. Chest 76: 601-602, 1979.
20. Stockley RA, Finnegan P, Bishop JM: Effect of intravenous terbutaline on arterial blood gas
tensions, ventilation, and pulmonary circulation in patients with chronic bronchitis and cor
pulmonale. Thorax 32: 601-605, 1977.
21. Brent BN, Mahler DA, Berger HJ. Augmentation of right ventricular perfonnance in chronic
obstructive pulmonary disease by terbutaline: A combined radionuclide and hemodynamic study.
Am J Cardiol 50: 313-319, 1982.
22. Teule GJ], Majid PA. Hemodynamic effects of terbutaline in chronic obstructive airways disease.
Thorax 35: 536-542, 1980.
23. Green LH, Smith TW. The use of digitalis in patients with pulmonary disease. Ann Intern Med
87: 459-465, 1977.
24. Ferrer MI, Harvey RM, Cathcart RT, Webster CA, Richards DW, Cournand A. Some effects
of digoxin upon the heart and circulation in man: Digoxin in chronic cor pulmonale. Circulation
1: 161-186, 1950.
25. Howarth S, McMichael J, Sharpey-Shafer EP. Effects of oxygen, venesection and digitalis in
chronic heart failure from disease of the lungs. Clin Sci 6: 187-196, 1948.
26. Jezek V, Schrijen F. Hemodynamic effect of deslanoside at rest and during exercise in patients
with chronic bronchitis. Br Heart J 35: 2-8, 1973.
27. Smith DE, Bissett lK, Phillips JR, Doherty JE, Murphy ML. Improved right ventricular systolic
time intervals after digitalis in patients with cor pulmonale and chronic obstructive pulmonary
disease. Am J Cardiol 41: 1299-1304, 1978.
28. Haas F, Foster WH, Bergofsky EH. Direct effects of ouabain on the pulmonary vasculature and
its enhancement of the vasoconstrictor response to hypoxia. Prog Resp Res 9: 273-276 1975.
29. Linde LM, Goldberg SJ, Gaal P, Momma K, Takahashi M, Sarna G. Pulmonary and systemic
hemodynamic effects of cardiac glycosides. Am Heart J 76: 356-364, 1968.
30. Coates AL, Desmond K, Asher MI, Hortop J, Beaudry PH. The effect of digoxin on exercise
capacity and exercising cardiac function in cystic fibrosis. Chest 82: 543-547, 1982.
31. Rodensky PL, Wassennan F. Observations on digitalis intoxication. Arch Intern Med 108:
171-188, 1961.
349
32. Beller GA, Smith TW, Abelmann WH, Haber E, Hood WB. Digitalis intoxication. A prospec-
tive clinical study with serum level correlations. N Engl] Med 284: 989-997, 1971.
33. Holford FD, Mithoefer ]C. Cardiac arrhythmias in hospitalized patients with chronic obstruc-
tive pulmonary disease. Am Rev Respir Dis 108: 879-885, 1973.
34. Klein MD, Lown B, Barr I, Hagemeijer F, Garrison H, Axelrod P. Comparison of serum digoxin
level measurement with acetylstrophanthidin tolerance testing. Circulation 49: 1053-1062, 1974.
35. Mathur PN, Powles RCP, Pugsley SO, McEwan MP, Campbell E]M. Effect of digoxin on
right ventricular function in severe chronic airflow obstruction. Aun Intern Med 95: 283-288,
1981.
36. Hargreave FE. Digitalis and cor pulmonale. Br Med] 2: 943, 1965.
37. Noble 1M, Trenchard D, Guz A. The value of diuretics in respiratory failure. Lancet 2: 257-260,
1966.
38. Whitman V, Stern RC, Bellet P, Doershuk CF, Liebman], Boat TF, Borkat G, Matthews LW.
Studies on cor pulmonale in cystic fibrosis. I. Effects of diuresis. Pediatrics 55: 83-85, 1975.
39. Heinemaun HO. Right-sided heart failure and the use of diuretics. Am] Med 64: 367-370, 1978.
40. Bhatia ML, Singh I, Manchanda SC. Effect of furosemide on pulmonary blood volume. Br Med
] 2: 551-555, 1969.
41. Dikshit K, Vyden JK, Forrester ]S. Renal and extrarenal hemodynamic effects of furosemide
in congestive heart failure. N Engl] Med 288: 1087-1090, 1973.
42. Stern RC, Borkat G, Hirschfeld SS, Boat TF, Matthews LW, Liebman], Doershuk CF. Heart
failure in cystic fibrosis. Am] Dis Child 134: 267-272, 1980.
43. Riley RL, Himmelstein A, Motley HL, Weiner HM, Cournand A. Studies of the pulmonary
circulation at rest and during exercise in normal individuals and in patients with chronic
pulmonary disease. Am] Physiol 152: 372-382, 1948.
44. Lee HM, Minh VD. Effect of body position on resting hemodynamics in patients with chronic
obstructive lung disease. Fed Proc 38 (3): 1378, 1979.
45. Rubin LJ, Peter RH. Hemodynamics at rest and during exercise after oral hydralazine in patients
with cor pulmonale. Am] Cardiol47: 116-122, 1981.
46. Bear R, Goldstein M, Phillipson E, Ho M, Hammeke M, Feldman R, Handelsman S, Halperin
M. Effect of metabolic alkalosis on respiratory function in patients with chronic obstructive lung
disease. Canad Med] 117: 900--903, 1977.
47. Burrows B. Arterial oxygenation and pulmonary hemodynamics in patients with chronic
airways obstruction. Am Rev Respir Dis 110: 64-70, 1974.
48. Findley L], Whelan DM, Moser KM. Long-term oxygen therapy in COPD. Chest 83: 671-674,
1983.
49. British Medical Research Council Working Party. Long-term domiciliary oxygen therapy in
chronic hypoxemic cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1:
681-686, 1981.
50. Nocturnal Oxygen Therapy Trial Group. Continuous or nocturnal oxygen therapy in hypox-
emic chronic obstructive lung disease: A clinical trial. Aun Intern Med 93: 391-398, 1980.
51. Ashutosh K, Mead G, Dunsky M. Early effects of oxygen administration and prognosis in
chronic obstructive pulmonary disease and cor pulmonale. Am Rev Respir Dis 127: 399-404,
1983.
52. Boysen PG, Block A], Wynne ]W, Hunt LA, Flick MR. Nocturnal pulmonary hypertension
53. Demarco F], Wynne JW, Block A], Boysen PG, Taasan Vc. Oxygen desaturation during sleep
as a determinant of the "blue and bloated" syndrome. Chest 79: 621-625, 1981.
54. Wynne JW, Block A], Hemenway], Hunt LA, Flick MR. Disordered breathing and oxygen
desaturation during sleep in patients with chronic obstructive lung disease (COLD). Am] Med
66: 573-580, 1979.
55. Anderson PB, Campbell RHA, Brennan SR, Howard P. The problem of pulmonary hyperten-
sion in chronic cor pulmonale. Prog Resp Res 9: 37-40, 1975.
56. Bishop ]M. Hypoxia and pulmonary hypertension in chronic bronchitis. Prog Resp Res 9:
10-16, 1975.
57. Haneda T, Nakajima T, Shirato K, Onodera S, Takashima T. Effects of oxygen breathing on
pulmonary vascular input impedance in patients with pulmonary hypertension. Chest 83:
520-527, 1983.
58. Segal N, Bishop ]M. The circulation in patients with chronic bronchitis and emphysema at rest
59. Harrison BDW, Davis J, Madgwick RG, Evans M. The effects of therapeutic decrease in packed
cell volume on the responses to exercise of patients with polycythemia secondary to lung disease.
Clin Sci Mol Med 45: 833-847, 1973.
60. Weisse AB, Moschos CB, Frank MJ, Levinson GE, Cannilla JE, Regan 1]. Hemodynamic effects
of staged hematocrit reduction in patients with stable cor pulmonale and severely elevated
hematocrit levels. Am J Med 58: 92-98, 1975.
61. Chetty KG, Brown SE, Light RW. Improved exercise tolerance of the polycythemic lung
patient following phlebotomy. Am J Med 74: 415-420, 1983.
62. The Urokinase Pulmonary Embolism Trial. Circulation 47, Suppl II: 84, 1973.
63. Bell WR, Simon TL, DeMets DL. The clinical features of submassive and massive pulmonary
emboli. Am J Med 62: 355-360, 1977.
64. Walcott G, Burchell HB, Brown AL. Primary pulmonary hypertension. AmJ Med 49: 70-79,
1970.
65. Bynum L, Wilson JE. Low-dose heparin in the long-term management of venous thromboem-
bolism. Am J Med 67: 553-556, 1979.
66. Hull R, Delmore T, Genton E. Warfarin sodium versus low-dose heparin in the long-term
tteattnent of venous thromboembolism. N Engl J Med 301: 855-859, 1979.
67. Hull R, Delmore T, Carter C, HirshJ, Genton E, Gent M, Turpie G, McLaughlin D. Adjusted
subcutaneous heparin versus warfarin sodium in the long-term tteattnent of venous thrombosis.
N Engl J Med 306: 189-194, 1982.
68. Gupta BD, Moodie DS, Hodgman JR. Primary pulmonary hypertension in adults. Cleve Clin
Q 47: 275-284, 1980.
69. Fowler NO, Black-Schaffer B, Scott RC, Gueron M. Idiopathic and thromboembolic pulmo-
nary hypertension. Am J Med 40: 331-345, 1966.
70. Steele PM, Fuster V, Edwards WD. Idiopathic pulmonary hypertension: Correlation of patho-
logical type, anticoagulant therapy and outcome in 120 patients. J Am Coli Cardiol 1: 735,
1983.
71. Alpert JS, Bass H, Szucs MM, Banas JS, Dalen JE, Dexter L. Effects of physical ttaining on
hemodynamics and pulmonary function at rest and during exercise in patients with chronic
obstructive pulmonary disease. Chest 66: 647-651, 1974.
72. Chester EH, Belman MJ, Bahler RC, Baum GL, Schey G, Buch P. Multidisciplinary tteattnent
of chronic pulmonary insufficiency. Chest 72: 695-702, 1977.
73. Stein PD, Sabbah HN, Marzilli M, Anbe DT. Effect of chronic pressure overload on the
maximal rate of pressure fall of the right ventricle. Chest 78: 10-14, 1980.
74. Stein PD, Sabbah HN, Anbe DT, Marzilli M. Performance of the failing and nonfailing right
ventricle of patients with pulmonary hypertension. Am J Cardiol 44: 1054-1055, 1979.
76. Harris P. Influence of acetylcholine on the pulmonary arterial pressure. Br Heart J 19: 272-278,
1957.
77. Fritts HW, Harris P, Claus RH, Odell JE, Coumand A. The effect of acetylcholine on the
human pulmonary circulation under normal and hypoxic conditions. J Clin Invest 37: 99-110,
1958.
78. Behnke RH, Williams]F, White DH. The effects of acetylcholine infusion upon cardiac
dynamics in patients with pulmonary emphysema. Am Rev Respir Dis 87: 57-62, 1963.
79. Marshall RJ, Helmholz HF, Shepherd ]T. Effect of acetylcholine on pulmonary vascular
resistance in a patient with idiopathic pulmonary hypertension. Circulation 20: 391-395, 1959.
80. Samet P, Bernstein WH, WidrichJ. Inttacardiac infusion of acetylcholine in primary pulmonary
hypertension. Am Heart J 60: 433-439, 1960.
81. Wood P. Pulmonary hypertension, with special reference to the vasoconstrictive factor. Br Heart
J 20: 557-570, 1958.
82. Samet P, Bernstein WHo Loss of reactivity of the pulmonary bed in primary pulmonary
hypertension. Am Heart J 66: 197-199, 1963.
83. Dresdale DT, Schultz M, Michton RJ. Primary pulmonary hypertension. I. Clinical and hemo-
dynamic study. Am J Med 11: 686-705, 1951.
84. Dresdale DT, Michtom RJ, Schultz M. Recent studies in primary pulmonary hypertension
351
including pharmacodynamic observations on pulmonary vascular resistance. Bull NY Acad Med

30: 195-207, 1954.
85. Widimsky J, Kasalicky J, Valach A, Dejdar R, Vyslouzil Z, Lukes M. Effect of priscoline on
the pulmonary circulation in cor pulmonale. Br Heart J 22: 571-578, 1960.
86. Kelminson LL, Cotton EK, Vogel JHK. The reversibility of pulmonary hypertension in patients
with cystic fibrosis. Observations on the effects of tolazoline hydrochloride. Pediatrics 39: 24-35,
1967.
87. Gould L, DeMartino A, Gomprecht RF, Umali F, Michael A. Hemodynamic effects of phen-
tolamine in cor pulmonale. J Clin Pharm 12: 153-157, 1972.
88. VanMieghem W, DeBacker G, DeWispelaere B, Billiet L, Cosemans J. Phentolamine infusion
in cor pulmonale due to chronic obstructive pulmonary disease. Acta Cardiologica 33: 253-262,
1978.
89. RuskinJN, Hutter AM. Primary pulmonary hypertension tteated with oral phentolamine. Ann
Intern Med 90: 772-774, 1979.
90. Cohen ML, Kronzon I. Adverse hemodynamic effects of phentolamine in primary pulmonary
hypertension. Ann Intern Med 95: 591-592, 1981.
91. Gould L, Chokshi AB, Pate! S, Gomes GI. Hemodynamic evaluation of vasodilator drug therapy
in primary pulmonary hypertension. Am Heart J 102: 300, 1981.
92. Knapp E, Gmeiner R. Reduction of pulmonary hypertension by nittoprusside. Int J Clin
pharmacol 15: 75-80, 1977.
93. Dantzker DR, Bower JS. Partial reversibility of chronic pulmonary hypertension caused by
94. Fourrier F, Chopin C, Durocher A, Dubois D, Watte! F. Inttavenous nittoglycerin in acute
respiratory failure of patients with chronic obsttuctive lung disease, secondary pulmonary
hypertension and cor pulmonale. Intensive Care Med 8: 85-88, 1982.
95. Konietzko N, Schlehe H, Harich B, Matthys H. Effect of isosorbide dinitrate on hemodynamics
and respiration of patients with coronary artery disease and of patients with chronic cor
pulmonale. Respiration 32: 368-377, 1975.
96. Wegyandt GR, Kopman EA, Ludbrook P A. Mechanism of nitroglycerin-induced hypoxemia.
Cath and Cardiovasc Diag 6: 387-395, 1980.
97. Kochukoshy KN, Chick TW, Jenne JW. The effect of nitroglycerin in gas exchange on chronic
obstructive pulmonary disease. Am Rev Respir Dis 111: 177-183, 1975.
98. Naeije R, Me!ot C, Mols P, Hallemans R. Effects of vasodilators on hypoxic pulmonary
vasoconstriction in normal man. Chest 82: 404-410, 1982.
99. Hales CA, Westphal D. Hypoxemia following the administration of sublingual nitroglycerin.
AmJ Med 65: 911-918, 1978.
100. Chick TW, Kochukoshy KN, Matsumoto S, Leach JK. The effect of nitroglycerin on gas
exchange, hemodynamics, and oxygen transport in patients with chronic obsttuctive pulmonary
disease. Am J Med Sci 276: lOS-Ill, 1978.
101. Annest SJ, Gottlieb ME, Rhodes GR, Paloski WH, Barie P, NewellJC, Shah DM. Nitroprus-
side and nitroglycerin in patients with post-traumatic pulmonary failure. J Trauma 21: 1029-
1031, 1981.
102. Koch-Weser J. Drug therapy: Diazoxide. N Engl J Med 294: 1271-1274, 1976.
103. Wang SWS, PohiJEF, Rowlands DJ, Wade EG. Diazoxide in treatment of primary pulmonary
hypertension. Br Heart J 40: 572-574, 1978.
104. Klinke WP, Gilbert JAL. Diazoxide in primary pulmonary hypertension. N Engl J Med 302:
91-92, 1980.
105. Honey M, Cotter L, Davies N, Denison D. Clinical and hemodynamic effects of diazoxide in
primary pulmonary hypertension. Thorax 35: 269-276, 1980.
106. Hall DR, Petch MC. Remission of primary pulmonary hypertension during treatment with
diazoxide. Br Med J 282: 1118, 1981.
107. Buch J, Wennevold A. Hazards of diazoxide in pulmonary hypertension. Br Heart J 45:
401-403, 1981.
108. Rubino JM, Schroeder JS. Diazoxide in treatment of primary pulmonary hypertension. Br Heart
J 42: 362-363, 1979.
109. Judson WE, Hollander W, Wilkins R W. The effects of intravenous apresoline (hydralazine)
on cardiovascular and renal function in patients with and without congestive heart failure.
Circulation 13: 664-674, 1956.
110. Rubin L), Peter RH. Oral hydralazine therapy for primary pulmonary hypertension. N Engl
) Med 302: 69-73, 1980.
111. Herrera EL, Sandoval), Seoane M, Bialostzky D. The role of hydralazine therapy for pulmonary
arterial hypertension of unknown cause. Circulation 65: 645-650, 1982.
112. Brent BN, Berger H), Matthay RA, Mahler D, Pytlik L, Zaret BL. Contrasting acute effects
of vasodilators (nitroglycerin, nitroprusside, and hydralazine) on right ventricular performance
in patients with chronic obstructive pulmonary disease and pulmonary hypertension: A com-
bined radionuclide-hemodynamic study. Am) Cardiol 51: 1682-1689, 1983.
113. Farber H, Glauser FL. Oral hydralazine ameliorates the exercise-induced pulmonary hyperten-
sion in patients with talc granulomatosis. Chest 80: 387, 1981.
114. Santini D, Fox D, Kloner RA, Konstam M, Rude RE, Lordi BH. Pulmonary hypertension
in systemic lupus erythematosus: Hemodynamics and effects of vasodilator therapy. Clin Cardiol
3: 406--411, 1980.
115. Naslund M), Pearson TA, Ritter )M. A documented episode of pulmonary vasoconstriction in
systemic sclerosis. Johns Hopkins Med) 148: 78-80, 1981.
116. Bates ER, Crevey B), Sprague FR, Pitt B. Oral hydralazine therapy for acute pulmonary
embolism and low output state. Arch Intern Med 141: 1537-1538, 1981.
117. McWeeney), Finnegan P. Vasodilator therapy of pulmonary hypertension in the CRST
syndrome. Irish) Med Sci 151: 151-154, 1982.
118. Rubin L), Handel F, Peter RH. The effects of oral hydralazine on right ventricular end-diastolic
119. Kronzon I, Cohen M, Winer HE. Adverse effect of hydralazine in patients with primary
pulmonary hypertension. )AMA 247: 3112-3114, 1982.
120. Hermiller )B, Bambach D, Thompson M), Huss P, Fontanat ME, Magonen RD, Unverferth
DV, Leier CV. Vasodilators and prostaglandin inhibitors in primary pulmonary hypertension.
Ann Intern Med 97: 480-489, 1982.
121. Packer M, Greenberg B, Massie B, Dash H. Deleterious effects of hydralazine in patients with
pulmonary hypertension. N Engl) Med 306: 1326-1331, 1982.
122. Stone PH, Antman EM, Muller )E, Braunwald E. Calcium channel blocking agents in the
treatment of cardiovascular disorders. Part II. Hemodynamic effects and clinical applications.
Ann Intern Med 93: 886-904, 1980.
123. Henry PD. Comparative pharmacology of calcium antagonists: Nifedipine, verapamil and
diltiazem. Am) Cardiol46: 1047-1058, 1980.
124. Kennedy T, Summer W. Inhibition of hypoxic pulmonary vasoconstriction by nifedipine. Am
) Cardiol 50: 864-868, 1982.
125. McMurtry I, Davidson A, Reeves), Grover R. Inhibition of hypoxic pulmonary vasoconstric-
tion by calcium antagonists in isolated rat lungs. Circ Res 38: 99-104, 1976.
126. Landmark K, Refsum AM, Simonsen S, Storstein O. Verapamil and pulmonary hypertension.
Acta Med Scand 204: 299-302, 1978.
127. Kambara H, Fujimoto K, Wakabayashi A, Kawai C. Primary pulmonary hypertension: Benefi-
cial therapy with diltiazem. Am Heart) 101: 230-231, 1981.
128. Crevey B), Dantzker DR, Bower )S, Popat KD, Walker SD. Hemodynamic and gas exchange
effects of intravenous diltiazem in patients with pulmonary hypertension. Am) Cardiol 49:
578-583, 1982.
129. Camerini F, Alberti E, Klugmann S, Salvi A. Primary pulmonary hypertension: Effects of
nifedipine. Br Heart) 44: 352-356, 1980.
130. DeFeyter P), Kerkkamp H], delong )P. Sustained beneficial effect of nifedipine in primary
pulmonary hypertension. Am Heart) 105: 333-334, 1983.
131. Wise )R. Nifedipine in the treatment of primary pulmonary hypertension. Am Heart) 105:
693-694, 1983.
132. McLeod AA, Wise )R, Daly K, )ewitt D. Nifedipine in treatment of primary pulmonary
hypertension. Br Heart) 45: 619, 1981.
133. Berkenboom G, Sobolski), Stoupel E. Failure of nifedipine treatment in primary pulmonary
hypertension. Br Heart) 47: 511, 1982.
134. Rubin L), Nicod P, Hillis LD, Firth BG. Treatment of primary pulmonary hypertension with
nifedipine: A hemodynamic and scintigraphic evaluation. Ann Intern Med 99: 433--438, 1983.
135. Simonneau G, Escourrau P, Duroux P, Lockhart A. Inhibition of hypoxic pulmonary vasocon-
striction by nifedipine. N Engl) Med 304: 1582-1585, 1981.
353
136. Young TE, Lundquist LJ, Chesler EC, Weir EK. Comparative effects of nifedipine, verapamil
and diltiazem on experimental pulmonary hypertension. Am J Cardiol 51: 195-200, 1983.
137. Ikram H, Maslowski AH, Nicholls MG, Espiner EA, Hull FTL. Hemodynamic and hormonal
effects of captopril in primary pulmonary hypertension. Br Heart J 48: 541-545, 1982.
138. Kokubu T, Kazatani Y, Hamada M, Matsuzaik K, Ito T, Nishimura K, Ochi T, Daimon F,Joh
T. Is captopril effective in primary pulmonary hypertension? JPN Circ J 46: 1095-1097, 1982.
139. Niarchos AP, Whitman HH, Goldstein JE, Laragh]H. Hemodynamic effects of captopril in
pulmonary hypertension of collagen vascular disease. Am Heart J 104: 834-838, 1982.
140. Rich S, Martinez J, Lam W, Rosen KM. Captopril as treatment for patients with pulmonary
141. Leier CV, Bambach D, Nelson S, Hermiller JB, Huss P, Magorien RD, Unverferth DV.
Captopril in primary pulmonary hypertension. Circulation 67: 155-161, 1983.
142. Ferreira SH, Vane JR. Prostaglandins: Their disappearance from and release into the circulation.
Nature 216: 868-873, 1967.
143. Piper P, Vane JR. The release of prostaglandins from lung and other tissues. Ann NY Acad Sci
180: 363-385, 1971.
144. Lock JE, Olley PM, Coceani F. Direct pulmonary vascular responses to prostaglandins in the
conscious newborn lambs. Am J Physiol 238: H631-H638, 1980.
145. Hyman AL, Kadowitz PJ. Pulmonary vasodilator activity of prostacyclin (pGI2) in the cat. Circ
Res 45: 404-409, 1979.
146. Weir EK, Grover RF. The role of endogenous prostaglandins in the pulmonary circulation.
Anesthesiology 48: 201-212, 1978.
147. Watkins WD, Peterson MB, Crone RK, Shannon DC, Levine L. Prostacyclin and prostaglandin
El for severe idiopathic pulmonary artery hypertension. Lancet 1: 1083, 1980.
148. Rubin LJ, Groves BM, Reeves JT, Frosolono MF, Handel F, Cato AE. Prostacyclin-induced
acute pulmonary vasodilation in primary pulmonary hypertension. Circulation 66: 334-338,
1982.
149. Naeije R, Melot C, Mols P, Hallemans R. Reduction in pulmonary hypertension by prostaglan-
din El in decompensated chronic obstructive pulmonary disease. Am Rev Respir Dis 125: 1-5,
1982.
150. Melot C, Naeije R, Mols P, Vandenbossche JL, Denolin H. Effects of nifedipine on ventilation/-
perfusion matching in primary pulmonary hypertension. Chest 83: 203-207, 1983.
151. Baron M, Skrinskas G, Hyland P, Urowitz MB. Effects of prostaglandin El and other vasodila-
tor agents in pulmonary hypertension of scleroderma. Br Heart J 48: 304-305, 1982.
152. Szczeklik J, Dubiel JS, Mysik M, Pyzik Z, Krol R, Horzela T. Effects of prostaglandin El on
pulmonary circulation in patients with pulmonary hypertension. Br Heart J 40: 1397-1401,
1978.
153. Wagenvoort CA. Lung biopsy specimens in the evaluation of pulmonary vascular disease. Chest
77: 614-625, 1980.
154. Nihill MR. Pulmonary hypertension and pulmonary vascular disease. Chest 77: 581-582, 1980.
155. Dantzker DR, Bower JS. Pulmonary vascular tone improves VA/Q matching in obliterative
pulmonary hypertension. J Appl Physiol 51: 607-613, 1981.
156. Miller MJ, Chappel TR, Rubin LJ. The effects of vasodilator therapy with hydralazine on gas
exchange in cor pulmonale. Clin Res 31: 419A, 1983.
157. Utley JR, Spragg RG, Long WB, Moser KM. Pulmonary endarterectomy for chronic thrombo-
embolic obstruction: Recent surgical experience. Surgery 92: 1096-1102, 1982.
158. Moser KM, Spragg RG, Utley J, Dailey PO. Chronic thrombotic obstruction of major pulmo-
nary arteries. Ann Intern Med 99: 299-305, 1983.
159. Reitz BA, Wallwork JL, Hunt SA, Pennock JL, Billingham ME, Oyer PE, Stinson EB,
Shumway NE. Heart-lung transplantation. Successful therapy for patients with pulmonary
vascular disease. N Engl J Med 306: 557-564, 1982.
160. Jamieson SW, Reitz BA, Oyer PE, Billingham M, Modry D, Baldwin J, Stinson EB, Hunt
S, Theodore J, Bieber CP, Shumway NE. Combined heart and lung transplantation. Lancet 1:
1130-1131, 1983.
13. PROGNOSIS OF PULMONARY HEART DISEASE
RUSSELL DODGE, M.D.,

BENJAMIN BURROWS, M.D.,
and
DOUGLASS MORRISON, M.D.
Pulmonary heart disease complicates a variety of pulmonary disorders. While

investigators of the natural history of diseases such as emphysema (1) and idiopathic
pulmonary fibrosis (2) have carefully described the progressive worsening of
symptoms and pulmonary function of their patients, progression of cardiovascular
abnormalities has not been so closely studied. The authors have usually simply
noted the mortality differences in patients with and without "cor pulmonale." In
these clinical studies "cor pulmonale" was usually not clearly defmed, often de-
pending on the clinical judgment of the investigator. Here we will equate the term
cor pulmonale with overt pulmonary heart disease.
The cursory reporting on pulmonary heart disease that was included in studies
of the prognosis of pulmonary disorders stemmed in part from the authors' primary
interest in the pulmonary disease. Also, unlike pulmonary function tests, evalua-
tions of cardiovascular status require invasive techniques that are not readily
performed repeatedly in a longitudinal study. A few authors have measured vascu-
lar and cardiac parameters at the onset and completion of a study. These researchers
reported actual pulmonary artery and right heart pressure measurements, but
avoided classifying patients as to the presence or absence of "cor pulmonale." As
researchers increasingly use noninvasive techniques of cardiovascular evaluation,
such as radionuclide angiocardiography and echocardiography, more quantitative
L.J. Rubin (ed.), Pulmonary Heart Disease. All rights reserved. Copyright @ 1984 Martinus NijhoffPubiishing. Boston/The Hague/Dor-
drecht/Lancaster.
355
356 13. Prognosis of Pulmonary Heart Disease
data will appear in the literature. In the present chapter, in an attempt to describe
the prognosis of cor pulmonale, we review the natural history studies of pulmonary
diseases often resulting in cor pulmonale in conjunction with reports emphasizing
the cardiovascular status of patients.
Patients with pulmonary obstructive disorders predominate the literature of
pulmonary heart disease, and this review reflects that emphasis. We have described
these patients in some detail because, among all patients with pulmonary heart
disease, patients with pulmonary obstructive disease benefit most by diagnosis and
treatment of their pulmonary hypertension. Oxygen therapy prolongs survival in
these patients although the mechanism is unclear, and this result has not been
conclusively demonstrated in the treatment of heart disease complicating other
pulmonary diseases.
We have divided the patients with pulmonary heart disease secondary to the
nonobstructive diseases into two groups: patients with restrictive lung diseases and
purely hypoventilatory disorders and patients with primary pulmonary vascular
diseases. Among diseases in the first category, including diseases of the chest wall
and nervous system, prognosis correlates roughly with the severity of the primary
disorder. Among patients with pulmonary vascular diseases, prognosis varies, but
most patients die within a few years of developing sustained severe pulmonary
hypertension. While this division oversimplifies the relationship between the lungs
and the pulmonary vascular system, especially in conditions such as idiopathic
kyphoscoliosis and progressive systemic sclerosis (PSS) , it is the same division used
in approaches to therapy, which if successful, influence prognosis. The pulmonary
heart disease secondary to restrictive disorders is generally treated indirectly by
therapy directed at the primary problem. Steroid therapy for diffuse interstitial
fibrosis and tricyclic therapy for sleep apnea are examples of such treatment. The
treatment of isolated pulmonary vascular disease is directed at the elevated pulmo-
nary artery pressure.
PROGNOSIS OF PATIENTS WITH PULMONARY HEART DISEASE

SECONDARY TO CHRONIC OBSTRUCTIVE DISEASE
Investigators have studied the relation of emphysema and the other obstructive
diseases to heart disease since Laennec's observations in 1819 (3). While they have
disagreed on the frequency and etiology of heart failure in these diseases, the
authors have consistently reported that the complication adversely affects prognosis.
Scott and Garvin (4), for example, reported that 43 of the 50 patients (86%) they
studied at autopsy had died while in their first attack of heart failure. In this report
heart failure was diagnosed when venous distention, hepatic enlargement, and
edema appeared. The patients' attacks lasted from six to eight months before death.
In another autopsy series with a review of patients' histories, Fulton (5) reported
that 20 of 49 patients with emphysema or bronchiectasis died with severe conges-
tive heart failure. None of the patients lived longer than 18 months after diagnosis
of pulmonary heart disease. According to Fulton, other patients in his study died
with little or no heart disease. Acute infections and ventilatory failure caused death
357
in these patients, who in general had a shorter duration of chest disease than had
the patients with heart failure. Fulton postulated that the patients with shorter
illness duration had rampant lung disease, with death occurring before right ven-
tricular failure could develop.
Other authors reported results similar to Fulton's. A study of patients from New
Dehli (6) found that survival after onset of heart failure averaged 19 months. In
this 1959 report, however, a few patients lived several years with pulmonary heart
disease.
Evidence that survival was gradually increasing continued with a report by
Stevens and his co-workers (7) in 1963. They reviewed the records of all patients
admitted to Massachusetts General Hospital between 1950 and 1956 with a diagno-
sis of cor pulmonale. Of 61 patients 47 had bronchitis and emphysema; the others
had restrictive disorders or nervous system disease with hypoventilation. The
average length of survival was 3.8 years after peripheral edema appeared. Survival
did not differ when patients with obstructive diseases were compared to the rest
of the group. Many of the patients had concomitant carbon dioxide retention, but
the authors did not report the cause of death in their patients.
Stevens and his co-authors attributed the improved survival time in their patients
to improved therapy. Assisted mechanical ventilation, for example, was employed
during episodes of carbon dioxide retention, a therapy previous authors generally
did not use. Aerosol bronchodilators also had become standard therapy by the time
these patients were studied. Improvements in intermittent oxygen therapy may
have also contributed to the results.
Improved survival in patients with obstructive disease has not, however, consis-
tently occurred. In 1980 Stern and co-workers (8) reported on the survival of 170
patients with cystic fibrosis. Fifty-five of the patients, mostly over 15 years of age,
died with overt pulmonary heart disease. Heart failure correlated well with hypox-
emia, and mean survival from onset of cardiac dilatation was eight months.
Over the past 20 years, since the report of Stevens et al., oxygen therapy has
evolved from intermittent to continuous and has become a standard therapy for
patients with obstructive pulmonary diseases. Enthusiasm for this therapy stemmed
from research that showed that oxygen relieved hypoxemic vasospasm in these
patients (9). Figure 13-1 shows that even brief oxygen therapy markedly reduces
the pulmonary artery pressure (PAP) increase seen in patients when they exercise
although it does little to resting PAP. Several groups of investigators have found
that continuous oxygen improves survival in chronic airways disease (10-12). If
oxygen prolongs survival by relief from pulmonary hypertension and right heart
failure, patients with overt pulmonary heart disease should be most benefited.
Surprisingly, then, in a large recent study that grouped patients according to
various cardiopulmonary characteristics (11), patients with marked pulmonary
vascular abnormalities did not show particular benefit from oxygen.
The Nocturnal Oxygen Therapy Trial (NOTT) attempted to confirm in a
controlled study the benefits of continuous oxygen (11), and the report, published
in 1980, clearly showed the benefits of continuous oxygen versus only nocturnal
w 80 • Air breathing
a::
I
o Oxygen breathing \
\
70 tChange with exercise
f
J
60
>
a:: 50 \
J!!,.. \
t)
\
IrJD
~~ \
1)
\
>E 40 \
l1 [) t) ti
~g
30
~
i
20
.:~
~
w
10
~
0
Figure 13-1. Effects of exercise on mean pulmonary artery pressure (PAPm) in 11 patients
with COPD with and without 02 breathing. Effects of 02 on resting PAPm may be seen by
comparing baseline values for the subjects connected by dashed lines. Effects on exercise PAPm
may be seen by comparing the tips of the arrows for each subject (representing their exertional
PAPm), also connected by dashed lines. (From reference 9, with permission of the authors and
publisher.)
use. Of a total of 203 patients, nearly twice as many had died in the nocturnal
oxygen group after a mean 19.3 months of follow-up. All patients had at least
moderately severe hypoxemia (pa0 2 < 55 mmHg) or a Pa02 < 59 mmHg plus
edema or polycythemia or cor pulmonale before entry. Continuous oxygen ther-
apy, however, did not excessively benefit the subgroup of patients who initially
showed evidence of pulmonary heart disease. For example, among patients with
relatively high pulmonary vascular resistances, the mortality rate was 45.2% in
those treated with nocturnal oxygen and 38.6% in those treated with continuous
oxygen (p = .11). Most of the improved survival occurred among patients with
relatively low pulmonary vascular resistances. From these data we cannot conclude
that continuous oxygen treatment for pulmonary heart disease is responsible for
the benefits oxygen produces in patients with obstructive diseases. Both the NOTI'
study and a British study of somewhat similar design suggested that the effect of
oxygen on other organ systems, such as the central nervous system, may explain
the results (11,12). On the other hand, a more complete evaluation of the pulmo-
nary circulation, such as with exercise measurements, may have identified a group
of patients with hemodynamic abnormalities and a significantly improved survival
with oxygen.
The reports of the NOTI' study group and Medical Research Council Working
Party are among the studies that have assessed the independent contribution of
359
,. "' ...... ,c .... "' .. fIV. _ _ - . . . .
I.J.•
$
'23" 5 i-j i ,'a" ';-"Q';'-a

y- y- y-
Figure 13-2. Effect of cor pulmonale (CP) on survival in subjects less than 65 years of age
divided into three groups according to their percentage predicted postbronchodilator forced
expiratory volume in 1 sec ('Yo pb FEV!). (c=with, and s=without.) (From reference 13,
with permission of the authors and publisher.)
pulmonary vascular disease (and indirectly, pulmonary heart disease) to the mortal-
ity of the obstructive lung diseases. Most of the earlier investigations, such as those
of Stevens and co-workers (7), recognized that while pulmonary heart disease was
associated with a bad prognosis, the heart disease was incidental to the severe lung
disease. Stevens stated: "The prognosis of pulmonary heart failure depends largely
on the outcome of the underlying chronic pulmonary disease and the rapidity with
which it progresses."
The view that pulmonary heart disease only incidentally complicates the ob-
structive diseases and has no independent prognostic value is a controversial one.
Workers other than Stevens, with better control of dependent variables, reached
different conclusions. Traver, Cline, and Burrows (13) examined the usefulness of
a multitude of variables in predicting mortality in patients with emphysema or
chronic bronchitis. After controlling for age the investigators found that the
percent predicted forced expiratory volume in one second (FEV1) best indicated
prognosis. As figure 13-2 shows, however, among the subjects under age 65, the
presence or absence of cor pulmonale further improved predictions of mortality.
According to the figure, one-half of the patients with an FEV1 below 30.5%
(roughly 1.0 liter) and cor pulmonale died within 2.2 years, while the 50%
cumulative survival was over five years for those patients with similar pulmonary
function but without cor pulmonale.
As figure 13-2 and other cited studies demonstrate, the presence or absence of
pulmonary heart disease suffers as a predictor of mortality because it is a discontinu-
ous variable. Consequently, investigators have measured PAP and right heart
function, continuous variables, to predict more accurately the effects of pulmonary
vascular abnormalities on mortality.
Weitzenblum (14) studied the pulmonary hemodynamics of a group of patients

with chronic obstructive disease. He reported that survival was more frequent in
patients with an initial mean PAP below 20 mmHg. In this study mean FEV 1
predicted survival as well as mean PAP did, perhaps because the small number of
patients produced only two categories of patients grouped by PAP and no im-
provement on classification by the presence or absence of pulmonary heart disease.
Weitzenblum studied 64 of these patients with repeat catheterizations, with an
average follow-up of five years (14). These patients were healthier than his original
group, with a mean FEV 1 of about 1.2 liters. The average increase in resting mean
PAP was 3.1 mmHg after an average of 5 years. Among the patients with an initial
PAP of greater than 20 mmHg, the increase was similar. Weitzenblum concluded
that among his patients, who were studied only when at rest, mild pulmonary
hypertension progressed very little over several years of observation and that he
could not detect a subgroup of patients in which rapid change in PAP was a
determinant of prognosis.
While simple measurements of PAP in patients at rest have provided little
information about pulmonary heart disease, more sophisticated catheterization
studies have provided prognostic data. Burrows and his co-workers studied 50
patients with chronic airways obstruction with measurements of PAP and cardiac
output (15). Calculation of pulmonary vascular resistance correlated better with
subsequent mortality than any lung function measurement. For example, no person
with a pulmonary vascular resistance greater than 550 dyne sec cm- s survived
more than three years after catheterization. The patients with extremely high PAP
who maintained average or high cardiac outputs and vascular resistances were
patients who had a clinical diagnosis of cor pulmonale. These patients had severe,
chronic arterial hypoxemia in addition to an obstructive ventilatory impairment,
and the hypoxemia may have produced the higher cardiac outputs and vascular
resistances. The authors concluded that if patients' obstructive disease was compli-
cated by "cor pulmonale, hypoxemic type" a particularly bad prognosis was likely.
Kawakami and his associates (16) recently reported that patients with chronic
obstructive lung disease and mixed venous hypoxemia, an indicator of impaired
tissue oxygen delivery, had a poorer prognosis than patients with a normal mixed
venous oxygen tension.
Other authors have shown that patients' prognosis is related to more than
spirometric impairment. Lindsay and Read (17) reported on 246 patients with
chronic airways disease. They measured the pulmonary vascular response to
hypoxia and grouped their patients into responders (those whose PAP rose with
hypoxia) and nonresponders. When patients were grouped according to level of
FEV1, they found that responders had a worse prognosis. Among patients with an
FEV 1 ranging from 1.0 to 1.5 liters, all nonresponders but only 57% of responders
survived three years.
Research on right ventricular function in patients with obstructive airways
disease has also provided some data on the prognosis of pulmonary heart disease.
Radionuclide studies of right ventricular ejection fraction (RVEF) have shown
361
that patients with normal ejection fractions did not develop clinical pulmonary
heart disease in two years of follow-up (18) while most patients with overt disease
already present had a low R VEF when studied. Also, the same researchers have
reported that four of nine patients with a low R VEF and obstructive lung disease
progressed to overt pulmonary heart disease (19). These patients had markedly
abnormal pulmonary function and arterial blood gases at initial evaluation. These
results suggest that further studies, using techniques such as radionuclide injections,
will provide information on the earliest events that occur in patients destined to
develop pulmonary heart disease.
The studies of Burrows et al., Kawakami et al., and Lindsey and Read have
emphasized the importance of the heart and pulmonary circulation in the prognosis
of chronic airways disease. Based on these studies, we can postulate that patients
with severe hypoxemia develop pulmonary heart disease and die earlier than
patients with equally severe respiratory impairment (percent predicted FEV1) but
less severe absolute hypoxemia. The mechanisms of hypoxemia in these patients
might include hypoventilation (especially at night) (20), alveolar hypoventilation
secondary to shallow breathing (21,22), and severe ventilation-perfusion mismatch-
ing. In addition to the studies we have cited to support the central role of
hypoxemia in prognosis, other types of studies also are compatible. For example,
death rates from emphysema are much higher at higher altitude than among
comparable patients living at higher ambient oxygen concentration (23). To
confirm the importance of hypoxemia, observations are needed of patients earlier
in their course, before hypoxemia or altered pulmonary hemodynamics occur. Not
yet clear is whether patients who react acutely to hypoxemia with pulmonary
vasoconstriction are the same ones who, if they develop sustained hypoxemia, are
especially likely to develop persistently elevated pulmonary vascular resistance and
cor pulmonale.
PROGNOSIS OF PATIENTS WITH PULMONARY HEART DISEASE

SECONDARY TO RESTRICTIVE LUNG DISEASES, DISEASE OF THE
THORACIC CAGE, AND OTHER HYPOVENTILATION DISORDERS
As in most reports of the natural history of chronic airways disease, articles on the
prognosis of restrictive pulmonary disorders give scant attention to the complica-
tion of pulmonary heart disease. Stack and his co-workers (2), for example, in a
report on a group of 96 patients with cryptogenic fibrosing alveoli tis, did not
describe pulmonary heart disease as more than an associated fmding in their patients.
The authors concluded that response to therapy and indicators of the severity of
pulmonary disease best predicted mortality. Similarly, in another study, Carrington
and his co-workers (24) reported on the "Natural History and Treated Course of
Usual and Desquamative Interstitial Pneumonia." The authors commented exclu-
sively on the pulmonary diagnoses, histologies, and treatments of their patients.
In a paper emphasizing the clinical characteristics of patients with cryptogenic
fibrosing alveoli tis, Turner-Warwick and co-workers (25) did find that heart
disease correlated with poor prognosis. Among the 20 patients who presented with
"right heart failure," the median survival was 4 months. Among the 186 patients
without heart failure, the median survival was 53 months. In addition, the authors
reported that right axis deviation on the electrocardiogram predicted a mean
survival of 22 months while a normal axis meant a survival of 59 months. Among
the patients with heart disease, spirometric impairment did not relate to survival.
Based on the limited clinical literature to date, we cannot state that pulmonary
heart disease has an independent effect on the prognosis of the interstitial lung
diseases, but the data of Turner-Warwick et al. are suggestive. Other methods of
inquiry, such as pathologic studies, suggest that progression of the fibrosis obliter-
ates the pulmonary vascular bed as well as the lung architecture, culminating in
severe disease of both systems. As Fishman (26) and others have commented, heart
failure in the interstitial diseases is associated with severe lung disease and a short
survival.
As in the interstitial diseases, the development of cor pulmonale in patients with
thoracic cage disorders is a bad prognostic sign. Bergofsky has grouped patients
with thoracic cage disorders into mechanical and neuromuscular categories (27).
Diseases such as kyphoscoliosis and fibrothorax are chronic, slowly progressive
disorders in which pulmonary heart disease frequently occurs as a consequence of
chronic hypoxemia. Bergofsky et al. (28) have reported that among six patients
with severe kyphoscoliosis and untreated congestive heart failure, three died within
one year. While controlled trials of therapy with oxygen and various methods of
mechanical ventilation have not demonstrated that these therapies affect prognosis,
one would expect oxygen to have a beneficial effect. As with patients with chronic
airways disease, relief of hypoxemia and acidosis produces a fall in pulmonary
artery pressure in these patients, as Bergofsky and his co-workers (28) have shown
in individual cases. Purely neuromuscular diseases, such as amyotrophic lateral
sclerosis and myasthenia gravis, generally either progress too rapidly or are success-
fully treated before pulmonary heart disease appears.
Cor pulmonale has also been reported among patients with other forms of
hypoventilation. Most common among these are the various diseases collectively
termed sleep disorders (29). While the occurrence of sudden death in obese, somno-
lent patients with pulmonary heart disease has been described (30), the general
prognosis of sleep disorder patients is unknown. Study of these patients has inten-
sified over the past 20 years, but efforts have centered on cross-sectional studies.
While such reports have produced results that suggest that pulmonary heart disease
is a serious complication of the sleep disorders, prognostic observations can appear
only with longitudinal studies.
Catheterization studies in patients have demonstrated the striking changes in the
pulmonary circulation that can occur during sleep, especially in patients with
underlying pulmonary disease. Coccagna and Lugaresi (31) reported that during
some sleep stages patients with chronic airways disease developed mean pulmonary
artery pressures of over 80 mmHg. Such pressure changes were accompanied by
hypoxemia. In other studies sleep hypoxemia (32) occurred most frequently in the
"blue bloaters," patients who others might label "cor pulmonale, hypoxemic type."
363
Serious arrhythmias have also been observed in these patients (33). As mentioned
above these studies have prompted Block (20) to speculate that the patients who
progress from moderate chronic airways disease to cor pulmonale may do so
because of nocturnal hypoxemic episodes. While differences among patients in
breathing patterns during sleep may explain a disposition to pulmonary heart
disease, other differences, such as in lung mechanics, may also help explain the
development of heart disease in only some patients with chronic airways obstruc-
tion. A slow, deep breathing pattern has been shown to be common in patients with
an emphysematous type of disease who have very compliant lungs, and this pattern
is associated with better preserved alveolar ventilation and higher arterial oxygen
levels than that in patients with chronic obstructive bronchitis (21). The inflamma-
tory changes present in the lungs of subjects with a more "bronchitic" type of
disease may preclude a loss of recoil and dictate a more rapid, shallow breathing
pattern with a larger dead space ventilation, leading to a worsening of blood
gas abnormalities and the more likely development of pulmonary heart disease
(21,22).
PROGNOSIS OF PULMONARY HEART DISEASE IN PATIENTS WITH

PRIMARY PULMONARY VASCULAR DISEASE
Unlike the patients described up to this point, the prediction of prognosis for most
patients with primary pulmonary vascular disease is not confounded by parenchy-
mal lung disease or neuromuscular disorders. With less independent variables at
work, prognosis should better relate to the severity of pulmonary heart disease, but
such a relationship is not clear from the literature. For example, among patients
with primary pulmonary hypertension, prognosis is unrelated to the level of
pulmonary artery pressure at presentation, as figure 13-3 demonstrates. As seen in
this figure, patients with primary pulmonary hypertension do not usually seek
medical attention until the hemodynamics are markedly abnormal (34). Most
patients also have clinical signs of pulmonary heart disease at their initial presenta-
tion and survive an average of two to three years after the onset of symptoms (34)
although regression of pulmonary hypertension and prolonged survival have oc-
curred (35-37). Voelkel and Reeves (34) have summarized serial catheterization
data and reported that pulmonary hypertension in general persisted as the vascular
resistance rose in 12 of16 patients. Eventually, the failing right ventricle gradually
depressed cardiac output, setting the stage for sudden death or a chronic low-output
syndrome.
The therapy of primary pulmonary hypertension has been reported to alter
favorably the hemodynamic abnormalities of the disease in individual cases (38-
39), but, as yet, prognosis clearly has not been affected. In chapter 12 the topic was
fully discussed, and the reader should keep in mind the reports of spontaneous
remission and prolonged survival when evaluating the various therapies.
Pulmonary hypertension and pulmonary heart disease have been associated with
progressive systemic sclerosis (PSS), systemic lupus erythematosus, dermatomyosi-
tis, and rheumatoid arthritis (40-43). While these diseases can also produce pulmo-
14 • SYSTOLIC
X DIASTOLIC
12
PA PRESSURE
(mmHol 100
80
MEAN
60
SURVIVAL AFTER
I" CATHETERIZATION HOURS DAYS 6 18-20 4-5 7-11
MONTHS YEARS
Figure 13-3. The magnitude of pulmonary hypertension at the time of initial heart
catheterization is not correlated with the duration of subsequent survival. (From reference 34,
with permission of the authors and publisher.)
nary fibrosis, Salemi and her co-workers (40) and Sackner et al. (44) found a poor
correlation between the severity of pulmonary hypertension and the severity of the
interstitial lung disease. Salemi et al., for example, reported on 10 patients with
the CREST syndrome (Calcinosis, Raynaud's phenomenon, esophageal dysfunc-
tion, sclerodactyly, telangiectasia), a variant ofPSS. The patients had no pulmonary
fibrosis, but all had abnormal pulmonary hemodynamics. Five of the patients died
from 1 to 38 months after catheterization. The authors directly attributed four of
these deaths to pulmonary hypertension because the autopsies demonstrated severe
pulmonary heart disease. These patients, unlike most with primary pulmonary
hypertension, had a prolonged clinical course before pulmonary hypertension was
recognized.
Reports of the association of pulmonary heart disease with the other collagen
vascular diseases have not clearly linked early mortality to this complication. Such
an association, however, will probably appear with further study of affected
patients.
PROGNOSIS OF ACUTE PULMONARY HEART DISEASE SECONDARY TO

LUNG SURGERY, PULMONARY EMBOLISM, AND ARDS
Animal models, which produced the data shown in figure 13-4 (45), warned
thoracic surgeons performing the thorocotomies of the 1930s and 1940s that the
extent of vascular resection predicted the likelihood of development of acute
pulmonary heart disease and death in their patients. The pressure-percent resection
365
800
1&1 700 0 ____ 0 PUPPIES

It:
::> . - . IDEALIZED CURVE
Ul 600
Ul .J
1&1 0 '£-x ADULT 0065
~ It:
> ~ 1500
It: Z
1&1 0
~O
It: ~ 400
«Z
> 1&1
It: ~
« 1&1300
Z Il.
o
5
~
200
Il.
100~~~--~--------~------~--------~
o 25 50 75 100
PERCENT PULMONARY RESECTION
Figure 13-4. Pulmonary artery pressures in adult dogs and puppies following varying degrees
of pulmonary resection. (From reference 45, with permission of the authors and publisher.)
curve for animal lung shows that little change in pulmonary artery pressure occurs
until resection exceeds 50%, with further resection producing exponential increases
in pressure. According to the model resection of more than 75% oflung tissue (and
accompanying vasculature) produces a fourfold increase in pulmonary artery pres-
sure and, in many animals, death secondary to heart failure.
The animal model correlates well with reports from the medical literature on
the limits of pulmonary resection. Death secondary to cor pulmonale occurred
frequently in patients whose lung resection exceeded 60% to 75% of the paren-
chyma (46). As expected, preoperative pulmonary heart disease also predicted a
poor result. Pecora and Brook reported the mortality of patients with mean
pulmonary artery pressures exceeding 36 mmHg was ten times higher than the
death rate of other patients (47).
While the presence of marked pulmonary hypertension preoperatively and the
development of elevated pressures postoperatively both adversely affect the prog-
nosis of lung surgery patients, they are not synonymous with progressive pulmo-
nary heart disease. Burrows et al. (48) have demonstrated the ability of some
patients to maintain significant elevations of right ventricular pressures for years
after surgery. Thus, while guidelines exist to deter surgery on patients at high risk
of postoperative heart failure (49), they are only rough estimates of risk and
individual patients have defied these predictions (50).
In a manner similar to surgical resection, pulmonary embolism can abruptly
produce pulmonary heart disease and high risk of death. In animal models massive
embolus causes right heart failure, shock, and death (51). Haggart and Walker (52),
using progressive pulmonary artery obstruction with clamps, showed that animals
tolerated up to 65% vascular obstruction before an abrupt rise in pulmonary
pressure, right heart failure, and death occurred. While no absolute level of
pulmonary artery pressure determined mortality, diastolic pressures over 30 mmHg
occurred only in animals with lethal obstruction.
Studies of pulmonary embolism in humans also have shown that fatal pulmonary
heart disease develops in patients with massive obstruction or less massive obstruc-
tion and underlying cardiopulmonary disease. Patients without underlying disease
who survive to be studied rarely have mean PAP exceeding 40 mmHg or overt
right heart failure (53). Thus, by inference, the prognosis of acute pulmonary heart
disease with more severe obstruction and hypertension is bleak. Pulmonary artery
pressures over 40 mmHg and elevated right ventricular diastolic pressures have been
recorded in surviving patients with underlying cardiopulmonary disease and, most
likely, underlying pulmonary hypertension, but in general these patients have a
worse prognosis than patients without underlying disease (54). The relative change
in pulmonary artery pressure rather than the absolute level seems to determine
prognosis.
Acute pulmonary heart disease in patients with adult respiratory distress syn-
drome (ARDS) , like patients with lung resection or embolism, has prognostic
importance. Zapol and Snider (55), in a study of 30 patients with severe respiratory
failure, found abnormally high pulmonary artery pressure. Because of low lung
compliances and low capillary wedge pressures in these patients, the authors be-
lieved the pulmonary hypertension did not reflect transmitted airway or left atrial
pressure but rather a combination of vasoconstriction, increased interstitial pressure,
and microvascular obliteration by microemboli or fibrosis. Patients whose pulmo-
nary hypertension remained or worsened did not survive. Several patients died with
mean pulmonary artery pressures over 50 mmHg and indicators of pulmonary heart
disease, such as high central venous pressures and low cardiac indices. As in patients
with other forms of interstitial lung disease, no independent effect of pulmonary
heart disease upon survival in ARDS was demonstrated by Zapol and Snider (55).
The strong association between sustained pulmonary hypertension and death is
the subject of continuing investigation (56), however, and as the number of studied
patients increases, an independent prognostic effect may be found.
REFERENCES
1. Burrows B, Earle RH. Prediction of survival in patients with chronic airway disease. Am Rev
Respir Dis 99: 865-871, 1969.
2. Stack BHR, Choo-kang YFj, Heard BE. The prognosis of cryptogenic fibrosing alveolitis.
Thorax 27: 535-542, 1972.
3. Laennec RTH. Trait' de l'ausculation mediate. Tr 1830. New York.
4. Scott RW, Garvin CF. Cor Pulmonale: Observations in fifty autopsy cases. Am Heart j 22:
56-63, 1941.
5. Fulton RM. Heart in chronic pulmonary disease. Quart j Med 22: 43-58, 1953.
6. PadmaVati S, Pathak SN. Chronic cor pulmonale in New Delhi: Study of 127 cases. Circulation
29: 343-352, 1959.
367
7. Stevens PM, Terplan M, Knowles JH. Prognosis of cor pulmonale. N Engl J Med 269: 1289-
1291, 1963.
8. Stem RC, Burkat G, Hirschfeld SS, Boat TF, Matthews LW, LiebbmanJ, Doershuk CF. Heart
9. Burrows B. Arterial oxygenation and pulmonary hemodynamics in patients with chronic airways
obstruction. Am Rev Respir Dis 110: 64-70, 1974.
10. Petry TL, Finigan MM. The clinical evaluation of prolonged ambulatory oxygen therapy in
patients with chronic airway obstruction. Am J Med 45: 242-252, 1968.
11. Nocturnal Oxygen Therapy Trial Group. ContinuoUs or nocturnal oxygen therapy in hypoxemic
chronic obstructive lung disease. Ann Intern Med 93: 391-398, 1980.
12. Medical Research Council Working Parry. Long-term domiciliary oxygen therapy in chronic
hypoxic cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1: 682--686, 1980.
13. Traver GA, Cline MG, Burrows B. Predictors of mortaliry in chronic obstructive pulmonary
disease. Am Rev Respir Dis 119: 895-902, 1979.
14. Weitzenblum E, Hirth C, Ducolone A, Mirhom R, Rasaholinjanahary J, Ehrhart M. Prognostic
value of pulmonary artery pressure in chronic obstructive pulmonary disease. Thorax 36: 752-
758, 1981.
15. Burrows B, Kettel LJ, Niden AH, Rabinowitz M, Diener CF. Patterns of cardiovascular dysfunc-
tion in chronic obstructive lung disease. N Engl J Med 286: 912-917, 1972.
16. Kawakami Y, Kishi F, Yamamoto H, Miyamoto K. Relation of oxygen delivery, mixed venous
oxygenation, and pulmonary hemodynamics to prognosis in chronic obstructive pulmonary
17. Lindsay DA, Read J. Pulmonary vascular responsiveness in the prognosis of chronic obstructive
18. Berger HJ, Matthay RA, Loke J, Marshall RC, Gottschalk A, Zaret BL. Assessment of cardiac
performance with quantitative radionuclide angiocardiography: Right ventricular ejection frac-
tion with reference to findings in chronic obstructive pulmonary disease. Am J Cardiol 41:
897-905, 1978.
19. Berger HJ, Matthay RA. Noninvasive radiographic assessment of cardiovascular function in acute
and chronic respiratory failure. Am J Cardiol 47: 950-962, 1981.
20. Block AJ, Boysen PG, Wynne JW. The origins of cor pulmonale: A hypothesis. Chest 75:
109-110, 1979.
21. Burrows B, Saksona FB, Diener CF. Carbon dioxide tension and ventilatory mechanics in chronic
obstructive lung disease. Ann Int Med 65: 685-700, 1966.
22. Parot S, Miara B, Milic-Emili J, Gautier H. Hypoxemia, hypercapnia, and breathing patterns in
patients with chronic obstructive pulmonary disease. Am Rev Respir Dis 126: 882-886, 1982.
23. Moore LG, Rohr AL, MaisenbachJK, ReevesJT. Emphysema mortaliry is increased in Colorado
residents at high altitude. Am Rev Respir Dis 126: 225-228, 1982.
24. Carrington CB, Gaensler EA, Coutu RE, Fitzgerald MX, Gupta RG. Natural history and treated
course of usual and desquamative interstitial pneumonia. N Engl J Med 298: 801-809, 1978.
25. Tumer-Warwick M, Burrows B, Johnson A. Cryptogenic fibrosing alveolitis: Clinical features
and their influences on survival. Thorax 35: 171-180, 1980.
26. Fishman AP. Chronic cor pulmonale: According to antecedent disorders. In Pulmonary Diseases
and Disorders, Fishman A (ed). New York: McGraw-Hill, 1980, pp. 863-882.
27. Bergofsky EH. Respiratory failure in disorders of the thoracic cage. Am Rev Respir Dis 119:
643-669, 1979.
28. Bergofsky EH, Turino GM, Fishman AP. Cardiorespiratory failure in kyphoscoliosis. Medicine
38: 263-325, 1959.
29. Phillipson EA, Bowes G. Sleep disorders. In Update: Pulmonary Diseases and Disorders, Fishman
A (ed). New York: McGraw-Hill, 1982, pp. 257-273.
30. Miller A, Granda M. In-hospital mortaliry in the Pickwickian syndrome. AmJ Med 56: 144-150,
1974.
31. Coccagna G, Lugaresi E. Arterial blood gases and pulmonary and systernic arterial pressure during
sleep in chronic obstructive pulmonary disease. Sleep 1: 117-124, 1978.
32. Douglas NJ, Calverley PMA, Loggett RJE, Brash HM, Flenley DC, Brezinova V. Transient
hypoxemia during sleep in chronic bronchitis and emphysema. Lancet 1: 1--4, 1979.
33. Flick MR, Block AJ. Nocturnal vs diurnal cardiac arrhythmias in patients with chronic obstruc-
tive pulmonary disease. Chest 75: 8--11, 1979.
34. Voelkel NF, ReevesJT. Primary pulmonary hypertension. In Pulmonary Vascular Disease, Moser
KM (ed). New York: Marcel Dekker, 1979, pp. 573-628.
35. Walcott G, Burchell HG, Brown AL. Primary pulmonary hypertension. AmJ Med 49: 70-79,
1970.
36. Suarez LD, Sciondro EE, Llera JJ, Perosio AM. Long-term follow-up in primary pulmonary
37. Bourdillon PD, Oakley CM. Regression of primary pulmonary hypertension. Br Heart J 38:
264-270, 1976.
J Med 302: 69-73, 1980.
39. Shettigar VR, Hultgren HN, Spector M, Martin R, Davies RH. Primary pulmonary hyperten-
sion: Favorable effect of isoproterenol. N Engl J Med 295: 1414-1415, 1976.
40. Salemi R, Rodnon GP, Leon DF, Shaver JA. Pulmonary hypertension in the crest syndrome
variant of progressive systemic sclerosis (scleroderma). Ann Int Med 86: 394-399, 1977.
41. Aitchison JD, Williams AW. Pulmonary changes in disseminated lupus erythematosus. Ann
Rheum Dis 15: 26-32, 1956.
42. Caldwell IW, AitchisonJD. Pulmonary hypertension in dermatomyosis. Br HeartJ 18: 273-276,
1956.
43. Jordan D, Snyder CH. Rheumatoid disease of the lung and cor pulmonale. Am J Dis Child 108:
174-180, 1964.
44. Sackner MA, Akgun N, Kimbel P. The pathophysiology of scleroderma involving the heart and
respiratory system. Ann Int Med 60: 611-630, 1964.
45. SchillingJA. Pulmonary resection and sequelae of thoracic surgery. In Handbook of Physiology,
Section 3: Respiration, Volume II; Fenn WO, Rahn H, (eds). Baltimore: Williams & Williams,
1965, pp. 1531-1552.
46. Adams WE, Perkins JF, Flores A, Chao P, Castellanos M. Significance of pulmonary hypertension
as a cause of death following pulmonary resection. J Thoracic Surg 26: 407-418, 1953.
47. Pecora DV. Evaluation of cardiopulmonary reserve in candidates for chest surgery. J Thorac
Cardiovasc Surg 44: 60-66, 1962.
48. Burrows B, Harrison RW, Adams WE, Humphreys EM, Long ET, Reimann AF. The post-
pneumonectomy state. Clinical and pathological observations in thirty-six cases. Am J Med 28:
281-297, 1960.
49. Olsen GN, Block AJ, Swanson EW, Castle JR, Wynne WJ. Pulmonary function evaluation of
the lung resection candidate: A prospective study. Am Rev Respir Dis 111: 379-387, 1975.
50. Bjork VO. Partial resection of the only remaining lung with the aid of respirator treatment. J
Thorac Cardiovasc Surg 39: 179-188, 1954.
51. Just-Viera JO, Yeager GH. Massive pulmonary embolism. II. Predictable mortality and cardiopul-
monary changes in dogs breathing room air. Ann Surg 159: 636-644, 1964.
52. Haggart GE, Walker AM. The physiology of pulmonary embolism as disclosed by quantitative
occlusion of the pulmonary artery. Arch Surg 6: 764-783, 1923.
53. Sasahara AA, Cannilla JE, Morse RH, Sidd JJ, Tremblay GM. Clinical and physiologic studies
in pulmonary thromboembolism. Am J Cardiol 20: 10-20, 1967.
54. Paraskos JA, Adelstein SJ, Smith RE, Ruchman FD, Grossman W, Dexter L, Dalen JE. Late
prognosis of acute pulmonary embolism. N Engl J Med 289: 55-58, 1973.
55. Zapol WM, Snider MT. Pulmonary hypertension in severe acute respiratory failure. N Engl J
Med 296: 476-480, 1977.
56. Zimmerman GA, Morris AH, Cengiz M. Cardiovascular alterations in the adult respiratory
distress syndrome. Am J Med 73: 25-34, 1982.
INDEX
Acetylcholine, 337, 343 Angiotensin infusion, effect on left ventricular per-

Acute respiratory failure (ARF), 264, 290, 302 formance, 262
Adenopathy, hilar Anomalous venous return, 260
radiographic distinction from vascular enlargement, Antidiuretic hormone, 307
161-166 Arterial blood gases
Adrenergic innervation of the pulmonary circulation, in acute respiratory fail ute, 293
14,36 correlation with pulmonary hemodynamics, 182-
Adult Respiratory Distress Syndrome (ARDS) 184, 189-190, 194
pulmonary hypertension in, I, 288-290, 300, 366 during exercise, 211
role of vasoactive substances in, 51 Arteriovenous fistulae, pulmonary, 293
Aging, effects on pulmonary circulation, 70, 71, 248 Asthma
Airway resistance, 180-181, 304 acute cor pulmonale, 49, 289
Allen's test, 199 arrhythmias in, 280
Alpha-I-antitrypsin deficiency electrocardiogram in, 130-131
hemodynamics in, 190-191 pulmonary artery pressure in, 49
radiographic features of, 159 pulmonary blood flow in, 33
Alveolar-capillary block, 209, 211 Atelectasis, pulmonary blood flow in, 33, 293
Alveolar hypoventilation (Ondino's Curse), 160 Atrial septal defect, 66, 97, 260-261
Aminophylline. See Thepphylline Atrioventricular (AV) node, 275-276
Aminorex fumarate, 4 Atropine, 274
Ammonia, arterial, 293 Axis illusion, 123, 125, 127-128
Amyotrophic lateral sclerosis, 292, 362 Azygos vein, enlargement radiographically of, 175
Anaerobic threshold, 198, 200, 213
Anesthesia, 345 "Bat wing" pulmonary edema, 173
Angiomatoid lesion, 76 Beta blockers. see Propranolol
Angiotensin converting enzyme, 50-51, 342 Bradycardia, 274, 279
369
370 Index
Bradykinin, 50 Dextran, augmented preload using, 237

Brisket disease, 257 Diazoxide, 339-340
Bronchial circulation Diffusing capacity (DLCO)
blood Bow, 12 correlation with pulmonary hemodynamics, 185-
histology of, 69-70 189, 192-193, 195,207-212
wall thickness in, 13 exercise, 207, 212
Bronchiectasis, 289, 356 Raynaud's phenomenon, 208
Bronchogenic carcinoma, effects of, on pulmonary vas- scleroderma, 208
culature, 71 Digitalis, 239, 308-309, 328-330
Bullae, radiographic evaluation of, 148-150 antiarrhythmic effects, 281, 308
toxicity, 307, 330
Cannon A waves, 291 Diltiazem, 341-342
Capillary transit time, 209-210 Disopyramide, 282
Captopril, 342-343 Disseminated intravascular coagulation (DIe), 293
Carbon dioxide Diuretics, 307, 330-331
effects on the pulmonary circulation, 35 Dyspnea, 108, 109, 247, 266, 290, 335
effects on ventricular function, 254-255 "Dyspnea index", 206, 208
Carotid body, 274
Carvallo sign, 111 Echocardiogram, 266
Catecholamines, 36, 286 Edema
Cavernous lesion, 76 pedal, 4-5, 111, 292, 333, 357
Chest pain, 108, 340 perivascular, 28-29
Cholinergic innervation of the pulmonary vasculature, pulmonary. See Pulmonary edema
14 Eisenmenger's complex, 112-113
Cholinesterase, 36 radiographic features, 154, 159, 164
Cimetidine, 316 vasodilato~ effects, 337, 341
Cirrhosis Embolism, 4. See also Thromboembolism
plexogeni<: lesions in, 82 Eosinophilic granuloma, 4
clubbing in, 114 Eosinophilic leukocyte chemotactic factor of ana-
Clockwise rotation of the heart, 119, 121 phylaxis (ECF-A), 50
Clubbing, 114 Epinephrine, 253-254, 280
Congenital development defects, pulmonary heart dis- Equilibrium gated radionuclide angiocardiography,
ease with, 4 224, 226--227
Congenital heart disease, plexogenic lesions in, 80, Exercise
82 hemodynamic effects, 120, 300
Continuous positive airway pressure (CPAP), nasal, in hypoxemia during, 56
sleep apnea, 307 influence of mixed venous oxygen (Pv0 2), 55-56
Coronary artery disease, 251, 258, 262, 264-265 testing, 196--200
CRST syndrome, 340, 364 VA / Q relationship during, 56
Cyanosis, 109-11 0 Exolphthalmos, 110
Cystic Fibrosis
edema in, 111, 330 Fibrillation
effects of drug therapy, 329, 337 atrial, 281, 291
hemodynamics in, 190-191 ventricular, 276, 277, 280, 281
hypoxemia in, 52, 55 Fibrocongestive splenomegaly, 97
left ventricular function in, 265-266 First pass radionuclide angiocardiography, 224, 225-
mortality with right heart failure, 3, 357 226, 295
pulmonary heart disease in, 3, 4, 5, 289 Foramen ovale, patent, 110, 293
Forced expiratory volume (FEV,), 180-181, 188-189,
Dead space-tidal volume ratio (VD/VT) , 206--208, 195, 202, 204, 230, 359-360
212,291 Fourier analysis, 20
Dermatomyositis, 363 Frank-Starling mechanism, 66, 92, 120, 203, 314
371
Gold-195m, 226 Liver disease, 4. (See also Ci"hosis)

"Gothic" P waves, 123 Lung biopsy, in evaluation of pulmonary vascular dis-
Graham-Steell murmur, 111 ease, 81, 97-100
Granulomatous pulmonary arteritis, 4 Lung volumes, correlation with vascular disease, 24-
Guillan-Barre syndrome, 292 25, 181
Lymphangitic carcinomatosis, 289
Heath-Edwards classification, 71-78 Lymphatics, pulmonary, 12, 70, 86
Hematocrit, effects on pulmonary vascular resistance,
20,334-335 Malignant infiltration, 4. See also Lymphangitic car-
Hemoptysis, 109, 335 cinomatosis
Hemosiderosis, 76, 86 Malnutrition, 307
Heparin, 308 Maximal voluntary ventilation (MVV), 206, 213
High altitude disease, 2, 4, 30, 40-42, 289, 303 Mediastinal fibrosis, 4, 289
High frequency ventilation, 301, 312-313 Methoxamine, augmented afterload with, 237, 261
Histamine, 34, 35, SO, 286 Mitral stenosis, 1, 2, 31, 85, 91
Hoarseness, 109 Mixed connective tissue disease, 4
Hydralazine, 241, 340--341 Monocrotaline, pulmonary artery fibrosis produced by,
Hydroxyproline, 260 253
Hyperinflation, radiographic detection, 147 Mountain sickness. See High altitude disease
Hypertension, systemic, in sleep apnea syndrome, Mueller maneuver, 266
110 Myasthenia gravis, 292, 362
Hypokalemia Myogenic reflex, 36
arrhythmias, 273 Myxoma, left atrial, 2, 85
diuretic-induced, 293, 331
Hypophosphatemia, 307 Nifedipine, 309-310, 341-342
Hypoxemia Nitroglycerin, 241, 310, 339
exercise-associated, 200, 202 Nitroprusside, 241, 338-339
mixed venous, 35, 38-39, 55, 360 Norepinephrine, 50, 51, 253, 260
Hypoxic pulmonary vasoconstriction (HPVC), 12, 13,
33-35, 37-40, 286 Ob.;sity hypoventilation syndrome. See Sleep apnea
pathology of, 86-90 and pickwickian syndrome
Oligema, focal, 148
Impedance Oxygen pulse, 204, 213
characteristic, 20-24 Oxygen therapy
input, 20-24, 303 in acute respiratory failure, 302-303
Incidence, of pulmonary heart disease, 3-5 effects on pulmonary circulation, 43-45, 331-334,
Inferior vena cava (IVC) interruption, 308 356-357
Interdependence, ventricular, 49, 255-256, 300 effects on survival, 357-358
Intermittent positive pressure breathing (IPPB), 301, evaluation by radionuclide angiocardiography, 241-
312 243
Intravenous drug use, 3, 4, 85, 192, 289
Isoproterenol, 276, 311, 326--327 P pulmonale, 122, 132, 333
lsosorbide dinitrate, 339 Papilledema, 110
Pericardium, 50, 256--257
Krypton 81m, 297 Phentolamine, 338
Kyphoscoliosis, 4, 181, 289, 356, 362 Phlebotomy, 307, 334-335
radiographic features of, 167 Pickwickian syndrome, 167, 292. See also Sleep apnea
Pleural
Lactate effusion, in right heart failure, 175
blood levels during exercise, 200, 204, 232 fibrosis, 4, 91, 289
respiratory drive, 302 pressure, effects on pulmonary circulation, IS, 24,
Left atrial enlargement, radiographic detection of, 138 49-50, 256--257, 301, 312
372 Index
Plexifonn lesion, 76, 81 Retrosternal clear space, 147-148

Plexogenic pulmonary arteriopathy, 81-82, 97 Rheumatoid arthritis, 4, 363
Pneumoconiosis, 4, 288 Right axis deviation, 125, 362
Pneumothorax, 290, 305, 307 Right bundle branch block, 120, 128
Poliomyelitis, 292 Right ventricular ejection fraction
Polycythemia, 42, 306, 334 correlation with spirometry, 204
Polymyositis, 3, 4 during exercise, 204, 232-234
Portal vein stenosis, plexogenic lesions in, 82 effects of digitalis, 239, 330
Positive end expiratory pressure (PEEP) effects of vasodilators, 340, 342
increased pulmonary artery pressure by, 288, 304 nonnal levels, 229
improved oxygenation with, 304, 308 resting levels, 229-232
reduced cardiac output by, 301 techniques of measurements, 225-227
release of prostaglandins by, 51 Rigler's sign, 141, 146
Poiseuille flow, 13, 18-19
Premature ventricular contractions, 277, 278-279 S,-S,-S, pattern, 128
with right ventricular injections, 224 Sabre-sheath trachea, 150, 154
Primary pulmonary hypertension, 4, 291 Sarcoidosis, ~, 89, 186,209,288
anticoagulation, 335 Schistosomiasis, 4, 82
pathology, 81 Sclerodenna, 4, 288, 356, 363
pulmonary function tests in, 193 diffusing capacity in, 208
prognosis, 362 Sedatives, 290, 307, 308
response to oxygen, 334 Septum, interventricular, 249-251, 255-257, 261,
response to vasodilators, 327, 328, 337, 338, 339, 266
340, 341, 342, 343 Sheet flow, 16-17, 27
ventilation perfusion distribution, 53 Sickle cell anemia, 4, 293
ventricular function, 263-264 Silicosis, 89
Procainamide, 282 Simson's rule, 224
Progesterone, 307 Sinus node, 275
Propranolol, 253, 276, 277 Sleep apnea, 4, 43, 109, 192, 292, 307, 362
Propylhexedrine, 85 arrhythmias in, 274
Prothrombin time, 293 Slow reacting substance of anaphylaxis (SRS-A), 50
Prostaglandins, 34, 35, 50-51, 342, 343 Smoking, effects on the pulmonary vasculature, 70-
Pulmonary edema 71
with acute respiratory failure, 314 Starling resistor, 287
high altitude, 41, 287-288 Subacute cor pulmonale, 67
Pulmonary fibrosis, 4, 190-191, 288-289, 341, 355, Supernumerary vessels, 13, 67-69
361-362 Syncope, 108, 340
Pulmonary resection, 4 Systemic lupus erythematosus, 4, 340, 363
Pulmonic stenosis, 4, 89, 260 Systemic oxygen transport, 108, 202-203, 241, 303,
Pulsus paradoxus, 291 305, 307, 313-314, 328, 334, 339
Systemic time indices, 266, 295
Quinidine, 282
Tachycardia
Radiation, 4 multifocal atrial (MAT), 279-280, 281, 290, 294
Raynaud's phenomenon, 109 supraventricular, 278
pulmonary vasoconstriction, 288 ventricular, 280, 281, 282
Recruitment, 17, 29 Talc granulomatosis, 340
"Redistribution", radiographic evaluation, 154, 159, Terbutaline, 241, 327-328
172 Thallium-201, 227-228, 234-235, 295
Reserpine, 276 Theophylline, 239-240, 306, 310, 316, 325-
Restrictive lung disease, 192. (See also Pulmonary fibro- 326
sis or specific disease) arrhythmias from, 273, 276, 280
373
Thoracoplasty, 4 Vasculitis, 2, 45, 289, 293

Thromboembolism, 4, 288, 290, 308 Vasoactive intestinal peptide (VIP), 50
anticoagulation, 335-336 Vasodilators, 336--346
arrhytlunias with, 280 in acute cor pulmonale, 309-311
course of hemodynamic changes, 48 effects on distribution of ventilation and perfusion,
endarterectomy, 346 53
left ventricular function in, 263-264 effects on exercise, 205, 211
massive, acute, 291, 292, 303, 314, 340, 365- radionuclide evaluation of effects, 240-241
366 Vectorcardiogram, 131, 204
pathology of, 82-85 Veins, pulmonary, 12, 69, 289
radiographic features, 167, 169 hypertension, 85-86, 99, 154, 159
radionuclide studies in, 237 Venesection. See Phlebotomy
relationship with pulmonary hypertension, 47 Veno-occlusive disease, 2, 86, 97, 289
thrombolytic therapy, 188, 308, 336 Ventilatory equivalent, 208, 213
ventilation perfusion relationships in, 52-53 Ventricular septal defect, 327
Tolazoline, 337-338 Verapamil, 281, 341
Tracheostomy, 307 Vital capacity, 181-182, 189, 192-193, 195, 202
Transit time, pulmonary, 18, 209-210, 237 V0 2 max, 198-200, 213, 333, 334
Transplantation, heart-lung, 346
Tricuspid insufficiency, 111, 291, 340 Warfarin, 335
Tricyclic antidepressants, 307, 356 Wasserburger's pentalogy, 124
Trophoblastic emboli, 82 Waterfall effect, 26, 50
Tube flow, 16--17 WHO report, 2, 95, 285
Tumor emboli, 67, 293 Windkessel effect, 14, 20, 23
Wolff-Parkinson-White syndrome, 273, 281
Upper airway obstruction, 4, 289, 292
Uvulopaletectomy, 307 Xenon, radioactive gas method, 26

Pulmonary Heart Disease

Uploaded by

Copyright:

Available Formats

Pulmonary Heart Disease

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pulmonary Heart Disease

Uploaded by

Copyright:

Available Formats

PULMONARY HEART DISEASE

PULMONARY HEART DISEASE

edited by LEWIS J. RUBIN

MARTINUS NI]HOFF PUBLISHING

Boston/The Hague /Dordrecht/Lancaster

for North America for all other countries

Library of Congress Cataloging in publication Data

Pulmonary heart disease.

2. Pathophysiology of the pulmonary circulation 11

3. Pathology of pulmonary heart disease 65

4. Clinical evaluation 107

5. The electrocardiogram in chronic lung disease 117

6. Roentgenographic evaluation of pulmonary heart disease 135

7. Pulmonary function and exercise testing 177

8. Radionuclide angiocardiographic assessment of right and left ventricular

RICHARD A. MATTHAY, M.D.

9. The left ventricle in chronic lung disease 247

10. Arrhythmias in chronic lung disease 273

11. Acute cor pulmonale 285

12. Therapy of pulmonary heart disease 325

13. Prognosis of pulmonary heart disease 355

Harvey J. Berger, M.D.

Benjamin Burrows, M.D.

Edgar J. Caldwell, M.D.

Timothy R. Chappell, M.D.

Veterans Administration Medical Center

David R. Dantzker, M.D.

Russell Dodge, M.D.

Brydon J.B. Grant, M.D.

George J. Klein, M.D.

Richard A. Matthay, M.D.

Francis X. McGowan, M.D.

Douglass Morrison, M.D.

Staff Cardiologist, Veterans Administration Medical Center

Robert H. Peter, M.D.

Eric N. Prystowsky, M.D.

Carl E. Ravin, M.D.

Lewis J. Rubin, M.D.

Warren R. Summer, M.D.

Wayne E. Taylor, M.D.

Galen S. Wagner, M.D.

Responsibility for the diagnosis and management of disorders of the pulmonary

I am deeply indebted to my colleagues who have contributed their time, effort,

Lewis J. Rubin, M.D.

LEWIS J. RUBIN, M.D.

Table 1-1. Classification of pulmonary

Parenchymal lung diseases Left ventricular failure

In 1901 Ayerza described a case of sclerosis of the pulmonary arteries in associa-

comprehensive overview of various aspects of pulmonary heart disease, the unfor-

BRYDON J.B. GRANT, M.D.

THE NORMAL PULMONARY CIRCULATION

Functional consequences of pulmonary vascular anatomy

Pulmonary vascular tree

The pulmonary microcirculation

Tube flow is a more complex concept because it stresses inhomogeneity of the

systemic capillaries compared to the relative lack of support of pulmonary capillar-

Analysis of pressure-flow relations

Arteries Capillaries Veins

relative to plasma increases with hematocrit (40). Pulmonary vascular resistance of

Concept of pulmonary vascular impedance