(Infectious Agents and Pathogenesis) Ramon G. Canto, George R. Robinson II (Auth.), Herman Chmel, Mauro Bendinelli, Herman Friedman (Eds.) - Pulmonary Infections and Immunity-Springer US (1994)

Pulmonary Infections
and Immunity
INFECTIOUS AGENTS AND PATHOGENESIS
Series Editors: Mauro Bendinelli, University 01 Pisa
Herman Friedman, University 01 South Florida
COXSACKIEVIRUSES
A General Update
Edited by Mauro Bendinelli and Herman Friedman
FUNGAL INFECTIONS AND IMMUNE RESPONSES

Edited by Juneann W. Murphy, Herman Friedman, and
Mauro Bendinelli
MYCOBACTERIUM TUBERCULOSIS
Interactions with the Immune System
Edited by Mauro Bendinelli and Herman Friedman
NEUROPATHOGENIC VIRUSES AND IMMUNITY

Edited by Steven Specter, Mauro Bendinelli, and
Herman Friedman
PSEUDOMONAS AERUGINOSA AS AN OPPORTUNISTIC PATHOGEN

Edited by Mario Campa, Mauro Bendinelli, and
Herman Friedman
PULMONARY INFECTIONS AND IMMUNITY

Edited by Herman Chmel, Mauro Bendinelli, and
Herman Friedman
VIRUS-INDUCED IMMUNOSUPPRESSION
Edited by Steven Specter, Mauro Bendinelli, and
Herman Friedman
A Continuation Order Plan is available for this series. A continuation order will bring delivery of each
new volume immediately upon publication. Volumes are billed only upon actual shipment. For
further information please contact the publisher.
Pulmonary Infections
and Immunity
Edited by
Herman Chmel
St. Francis Medical Center
Trenton, New Jersey
Mauro Bendinelli
University of Pisa
Pisa, Italy
and
Herman Friedman
University oj South Florida
Tampa, Florida
Springer Science+Business Media, LLC

Library of Congress Cataloglng-In-Publlcatlon Data
Pulmonary Infections and Immunity I edited by Herman Chmel, Mauro

Bendinelll, and Herman Friedman.
p. cm. -- (Infectious agents and pathogenesis)
Includes bibliographical references and Index.
1. Respiratory Infectlons--Immunologlcal aspects. I. Chmel,

Herman. II. Bendinelll, Mauro. III. Friedman, Herman, 1931-
IV. Series.
[ONLM: 1. Respiratory Tract Infectlons--Immunology. 2. Lung
Olseases--Immunology. WF 140 P9828 19941
RC740.P85 1994
616.2'00479--dc20
ONLM/DLC
for Library of Congress 94-15377
CIP
ISBN 978-1-4899-1065-3 ISBN 978-1-4899-1063-9 (eBook)

DOI 10.1007/978-1-4899-1063-9
© 1994 Springer Science+Business Media New York

Originally published by Plenum Press, New York in 1994.
Softcover reprint of the hardcover 1st edition 1994
All rights reserved
No part of this book may be reproduced, stored in a retrieval system, or transmitted in any
form or by any means, electronic, mechanical, photocopying, microfilming, recording, or
otherwise, without written permission from the Publisher
Contributors
JOHN G. BARTLETT • Department of Medicine, Johns Hopkins University

School of Medicine, Baltimore, Maryland 21205
ROBERT W BRADSHER • Department of Medicine, University of Arkansas
for Medical Sciences, Little Rock, Arkansas 72205-7199
MIRIAM L. CAMERON • Division of Infectious Diseases, Duke U niversity
Medical Center, Durham, North Carolina 27710
RAMON G. CANTO • Pulmonary/Critical Care Division, The Pennsylvania
State University, M. S. Hershey Medical Center, Hershey, Pennsylvania 17033
STANLEY W CHAPMAN • Division of Infectious Diseases, University of
Mississippi Medical Center, Jackson, Mississippi 39216-4505
RICHARD D. CLOVER • Department of Family Medicine, The U niversity of
Texas Medical Branch at Galveston, School of Medicine, Galveston, Texas
77555-0853
S. J. CRYZ, JR. • Swiss Serum and Vaccine Institute, CH 3001 Berne, Switzer-
land
BURKE A. CUNHA • Infectious Disease Division, Winthrop-University Hos-
pital, Mineola, New York 11501; and SUNY School ofMedicine, Stony Brook,
New York 11790
STANLEY C. DERESINSKI • Department of Medicine, Stanford University
School of Medicine, Stanford, California, 94305; Department of Medicine,
Santa Clara Valley Medical Center, San Jose, California 95128; and AIDS
Community Research Consortium, Redwood City, California 94062
JERROLD J. ELLNER • Department of Medicine, Case Western Reserve Uni-
versity and University Hospitals, Cleveland, Ohio 44106-4984
LOUIS F. FRIES • Univax Biologics, Inc., Rockville, Maryland 20852
v
vi CONTRIBUTORS
ANN W. FUNKHOUSER • Laboratory of Infectious Disease, National Insti-

tute of Allergy and Infectious Disease, Bethesda, Maryland 20892
ELLIE J. C. GOLDSTEIN • Department of Medicine, Pulmonary Division
and Infectious Disease Division, St. john's Hospital and Health Center, Santa
Monica, California 90404; and UCLA School of Medicine, Los Angeles,
California, 90024
RICHARD H. GOLDSTEIN • Department of Medicine, Pulmonary Division
and Infectious Disease Division, St. john's Hospital and Health Center, Santa
Monica, California 90404; and UCLA School of Medicine, Los Angeles,
California, 90024
HAROLD M. HENDERSON • Division of Infectious Diseases, University of
Mississippi Medical Center, jackson, Mississippi 39216-4505
CAROL A. KEMPER • Department of Medicine, Stanford University School
of Medicine, Stanford, California 94305; Department of Medicine, Santa
Clara Valley Medical Center, Sanjose, California 95128; and AIDS Commu-
nity Research Consortium, Redwood City, California 94062
GERALD LANCZ • Department of Medical Microbiology and Immunology,
University of South Florida College of Medicine, Tampa, Florida 33612
RICHARD W. MCDONNELL • Department of Medicine, University of Ar-
kansas for Medical Sciences, Little Rock, Arkansas 72205-7199
JOHN R. PERFECT • Division of Infectious Diseases, Duke University Medi-
cal Center, Durham, North Carolina 27710
HERBERT Y. REYNOLDS • Department of Medicine, The Pennsylvania
State University, M. S. Hershey Medical Center, Hershey, Pennsylvania 17033
GEORGE R. ROBINSON 11 • Pulmonary/Critical Care Division, The Pennsyl-
vania State University, Hershey, Pennsylvania 17033
LORRY G. RUBIN • Department ofPediatrics, Schneider Children's Hospital
of Long Island jewish Medical Center, Long !sland Campus for the Albert
Einstein College of Medicine, New Hyde Park, New York, 11042
DAVID H. SHEPP • Department of Medicine, North Shore University Hospi-
tal, Cornell University Medical College, Manhasset, New York 11030
STEVEN SPECTER • Department of Medical Microbiology and Immunol-
ogy, University of South Florida College of Medicine, Tampa, Florida 33612
RICHARD V. SPERA,jR. • Department ofMedicine, North Shore University
Hospital, Cornell University Medical College, Manhasset, New York 11030.
Present address: Department of Internal Medicine, The Brooklyn Hospital
Center, Brooklyn, New York 11201
CONTRIBUTORS Vll
JANET E. STOUT • Special Pathogens Laboratory, Pittsburgh, Pennsylvania

15261
FREDERICK R. VOGEL • Vaccine Research and Development Branch, Divi-
sion of AIDS, Department of Health and Human Services, National Insti-
tutes of Health, Bethesda, Maryland 20892
ROBERT S. WALLIS • Department of Medicine, Case Western Reserve Uni-
versity and University Hospitals, Cleveland, Ohio 44106-4984
VICTOR L. YU • University of Pittsburgh, Infectious Disease Section, VA
Medical Center, Pittsburgh, Pennsylvania 15240
Preface to the Series
The mechanisms of disease production by infectious agents are presently the

focus of an unprecedented ßowering of studies. The field has undoubtedly
received impetus from the considerable advances recently made in the under-
standing of the structure, biochemistry, and biology of viruses, bacteria, fungi,
and other parasites. Another contributing factor is our improved knowledge of
immune responses and other adaptive or constitutive mechanisms by which hosts
react to infection. Furthermore, recombinant DNA technology, monoclonal anti-
bodies, and other newer methodologies have provided the technical tools for
examining questions previously considered too complex to be successfully tackled.
The most important incentive of all is probably the regenerated idea that
infection might be the initiating event in many clinical entities presently classified
as idiopathic or of uncertain origin.
Infectious pathogenesis research holds great promise. As more information
is uncovered, it is becoming increasingly apparent that our present knowledge of
the pathogenic potential of infectious agents is often limited to the most notice-
able effects, which sometimes represent only the tip of the iceberg. For example,
it is now weIl appreciated that pathological processes caused by infectious agents
may emerge clinically after an incubation of decades and may result from genetic,
immunologic, and other indirect routes more than from the infecting agent itself.
Thus, there is a general expectation that continued investigation will lead to the
isolation of newagents of infection, the identification of hitherto unsuspected
etiologic correlations, and eventually, more effective approaches to prevention
and therapy.
Studies on the mechanisms of disease caused by infectious agents demand a
breadth of understanding across many specialized areas, as weIl as much cooper-
ation between clinicians and experimentalists. The se ries Infectious Agents
ix
x PREFACE TO THE SERIES
and Patlwgenesis is intended not only to document the state of the art in this
fascinating and challenging field but also to help lay bridges among diverse areas
and people.
M. Bendinelli
H. Friedman
Preface
It is widely accepted that most microorganisms do not cause infection of the

respiratory systemunless other factors interfere with host defenses-factors such
as exposure to toxic chemicals, immunodeficiencies (acquired or induced), and
prolonged immobilizations. This contrasts with the fact that the warm, moist,
nutrient-rich milieu of the respiratory tract would seem to provide an ideal place
for growth of microorganisms, both pathogenic and opportunistic, widely pres-
ent in the external environment with which the respiratory system so extensively
communicates. Nonspecific defense mechanisms of the respiratory tract depend
on both anatomical architecture and numerous cellular and humoral effectors
important in preventing environmental agents, including infectious organisms,
from causing damage to the system. Nonspecific host mechanisms which defend
the lung facilitate clearance of noxious or pathogenic agents so they cannot
penetrate through the upper respiratory tract, and also destroy or neutralize any
contaminants that may penetrate. In addition, specific host immune defense
mechanisms are also extremely important in defending the respiratory system
from infection.
This book deals with basic science concepts and clinical implications of
infections of the respiratory apparatus, with special reference to the role of the
immune system. As respiratory infections are a major cause of morbidity and
mortality throughout the world, a primary purpose of this volume is to stimulate
renewed investigative interest in this area of biomedical research among clini-
cians, microbiologists, immunologists, and other medical scientists in general.
Leading investigators active in the area of respiratory infections have authored
one or more chapters in this book. Appreciation of the organisms involved in
respiratory infections is essential, we believe, in gaining a fuller understanding of
the clinical importance of these microbes and the types of relationships they
establish with diseased tissues. In addition, an understanding of the interactions
of such organisms with the host in general and the immune system in particular is
also necessary for evaluation of public health procedures and development of
xi
xii PREFACE
appropriate control measures for respiratory infections. In this regard, exciting

new developments are taking place in understanding the intricacies of host!
parasite relationships, including interaction of microorganisms with the complex
host respiratory system. It is therefore appropriate that the book opens with a
description of the general defense mechanisms of the respiratory tract and the
breaches in such defenses which may be exploited by opportunistic infections.
Subsequent chapters focus on specific infectious agents that may affect the
respiratory system, including gram positive and gram negative bacteria, both
aerobic and anaerobic. One chapter deals with Mycobacterium tuberculosis and
other mycobacteria, which are classically known to be important disease-causing
organisms in the lung. Tuberculosis, unfortunately, is making a "comeback" in
developed Western countries, because of increased numbers of immunocompro-
mised individuals, especially patients with AIDS. Chapters dealing with fungal
infections which cause respiratory diseases, including histoplasma and coccidiosis
infections, are also included. Chapters covering important viruses often associ-
ated with lung infections, including influenza viruses, follow.
The editors of this volume, as well as the authors of individual chapters, are
excited by recent developments concerning immune involvement in respiratory
infections caused by bacteria, fungi, and viruses, including opportunistic infec-
tions important in causing disease in immunosuppressed individuals. We believe
these findings will be of value not only for clinicians in the field, but also for active
investigators and biomedical scientists as well as students and health professionals
in general.
The editors wish to express thanks and gratitude to Ms. Ilona Friedman, who
continued to serve in an outstanding manner as Editorial Assistant for this
volume and for all the books of this series.
Herman Chmel
Mauro Bendinelli
Herman Friedman
Contents
1. Defense Meehanisms of the Respiratory Traet

RAMON G. CANTO, GEORGE R. ROBINSON 11,
and HERBERT Y. REYNOLDS
1. Introduetion.................................................. 1
2. Airway Defenses .............................................. 3
2.1. Aerodynamie Defenses and the Mucoeiliary Esealator ....... 3
2.2. Baeterial Mueosal Adherenee .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3. Immunologie Defense of the Airways ...................... 6
3. Lymphoeytes ............................... ',' . . . . . . . . . . . . . . . . . 7
3.1. Immunoglobulins........................................ 7
3.2. T Lymphoeytes .......................................... 9
3.3. Lymphoid Tissue and Cell-Mediated Defense ............... 10
3.4. Antigen-Presenting Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4. Defenses in the Alveolar Milieu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.1. Noneellular Components of the Alveolar Lining Fluid ....... 12
4.2. Nonimmune Opsonins .................................... 13
4.3. Immune Opsonins ....................................... 14
4.4. Complement System Components . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.5. Cells in the Alveolar Spaee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
5. Integrated Lung Defense Meehanisms .......................... 20
6. Conclusion.................................................... 23
Referenees .................................................... 24
xiii
xiv CONTENTS
2. The Immunology of Pneumococcal Pneumonia

CAROL A. KEMPER and STANLEY C. DERESINSKI
l. Introduction .................................................. 29
2. The Organism ................................................ 30
3. Pneumococcal Cell Components and the Pathogenesis of Infection 31
3.l. Cell Wall................................................ 31
3.2. Plasma Membrane ....................................... 32
3.3. Capsule................................................. 33
3.4. Cytoplasmic Factors ...................................... 33
4. Immune Defense .............................................. 34
4.l. Mucocal Barriers and Antibody Responses ................. 34
4.2. Opsonization and the Complement Cascades ................ 35
5. Disease States Predisposing to Severe Pneumococcal Infection ..... 37
5.1. Complement Deficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.2. Hypogammaglobulinemia ................................ 38
5.3. Alcoholism and Liver Disease ............................. 38
5.4. Viral Infections .......................................... 39
5.5. Postsplenectomy ......................................... 39
5.6. Sickle Cell Disease ....................................... 40
5.7. Acquired Immunodeficiency Syndrome .................... 40
5.8. Malignancies ............................................ 41
5.9. Other Conditions Which Predispose to Pneumococcal
Infection ................................................ 41
6. Conclusion.................................................... 42
References .................................................... 42
3. Pulmonary Infections Caused by Lancefield Group A Beta-

Hemolytic Streptococcus
RICHARD H. GOLDSTEIN and ELLIE J.C. GOLDSTEIN
l. Introduction .................................................. 51
2. Microbiology/Immunology ..................................... 51
3. Patterns of Hemolytic Streptococcal Pneumonia .................. 54
4. Epidemiology................................................. 57
5. Pathology..................................................... 58
6. Clinical Manifestations ......................................... 59
7. Complications................................................. 60
8. Treatment .................................................... 60
References .................................................... 61
CONTENTS xv
4. Pulmonary Infections Caused by Haemophilus inftuenzae

LORRY G. RUBIN
1. Introduction ................................................. 63
2. The Organism: Microbiology, Typing, and Virulence Determinants 64
3. Epidemiology of Colonization of H. inftuenzae ................... 66
4. Pathogenesis of Colonization and Infection with H. inftuenzae ..... 67
5. Immunity ................................................... 70
6. Clinical Syndromes ........................................... 71
6.1. Pneumonia............................................. 71
6.2. Chronic Bronchitis Exacerbation ......................... 74
6.3. Tracheitis and Tracheobronchitis ......................... 74
6.4. Epiglottitis ............................................. 75
7. Diagnosis .................................................... 75
8. Antibiotic Therapy ........................................... 76
9. Prevention.................................................. . 77
10. Summary ................................................... 78
11. References................................................... 79
5. Klebsiella Pneumonia
S. J. CRYZ, JR.
1. Etiological Agent .............................................. 85

2. Epidemiology and Clinical Significance .......................... 85
2.1. Antibiotic Resistance ..................................... 87
2.2. Pathogenesis of Klebsiella Pneumonia . . . . . . . . . . . . . . . . . . . . . . . 87
3. Klebsiella Virulence Factors ..................................... 88
3.1. Immunity to Klebsiella .................................... 89
4. CPS Vaccine .................................................. 90
5. Conclusion............................................... . . . . . 92
References .................................................... 93
6. Legionella pneumophila
JANET E. STOUT and VICTOR L. YU
1. The Disease .................................................. 97

2. Epidemiology and Transmission ................................ 97
3. Intracellular Niche ............................................ 100
4. Immune Defense Mechanisms .........................•........ 101
5. Virulence Factors ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
XVI CONTENTS
6. Vaccine Deve10pment .......................................... 105

References .................................................... 106
7. Anaerobic Bacterial Infections of the Lung

JOHN G. BARTLETT
1. Introduction .................................................. 113

2. Incidence..................................................... 113
3. Pathophysiology............................................... 116
4. Clinical Features .............................................. 117
5. Laboratory Diagnosis .......................................... 118
6. Bacteriology .................................................. 119
7. Treatment .•.................................................. 121
8. Prognosis ..................................................... 124
References .................................................... 125
8. Immunology of M. tuberculosis and Other Mycobacteria

ROBERT S. WALLIS and JERROLD J. ELLNER
1. Introduction................................................. 129
2. Mononuclear Phagocyte Immunoregulatory Properties ........... 129
3. Suppressor Lymphocytes in Tuberculosis ....................... 133
4. Mononuclear Phagocyte-Mycobacterial Interactions .............. 133
5. Cell-Mediated Cytotoxicity .................................... 134
6. Vitamin D and Mycobacterial Killing . . . . . . . . . . . . . . . . . . . . . . . . . .. 134
7. 'VI) T Cells ................................................... 135
8. Mycobacterial Antigens ....................................... 135
9. Human Monoclonal Antibodies .. ............ .. .. .............. 137
10. Mononuclear Phagocyte-Activating Mycobacterial Proteins . . . . . . .. 138
11. Mycobacterium avium .......................................... 141
References ................................................... 142
9. Bordetella pertussis
FREDERICK R. VOGEL
1. Introduction.................................................. 149
2. Virulence Factors and Protective Antigens ....................... 150
2.1. Pertussis Toxin .......................................... 150
CONTENTS xvii
2.2. Filamentous Hemagglutinin ............................... 150

2.3. Adenylate Cyclase Toxin.. ... ......... .. .. ................ 151
2.4. Heat-Labile Toxin........................................ 151
2.5. Tracheal Cytotoxin ....................................... 151
2.6. Pertactin ................................................ 152
2.7. Agglutinogens........................................... 152
2.8. Pertussis Endotoxin ...................................... 152
3. Clinical Manifestations of Pertussis .............................. 152
3.1. Transmission ............................................ 152
3.2. Complications ........................................... 153
3.3. Antibiotic Therapy ....................................... 153
4. Pertussis Vaccines ............................................. 153
4.1. Whole Cell Vaccines ...................................... 153
4.2. Acellular Pertussis Vaccines ............................... 154
5. Pertussis and AIDS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
References .................................................... 155
10. Coxiella burnetii and Q Fever

ANN W FUNKHOUSER and LOUIS F. FRIES
1. Introduction................................................. 159
2. E pidemiology and Transmission ............................... 161
2.1. Animal Infection ....................................... 162
2.2. Human Infection ....................................... 162
4. Acute Q Fever ............................................... 164
4.1. Pneumonia............................................. 164
4.2. Hepatitis............................................... 165
4.3. Neurologie Manifestations ............................... 166
4.4. Other Uncommon Clinical Correlates ..................... 166
4.5. Pathology .............................................. 167
5. Chronic Q Fever ............................................. 167
5.1. Endocarditis ........................................... 167
5.2. Other Chronic Q Fever Syndromes ....................... 169
6. Diagnosis .................................................... 169
7. Therapy..................................................... 171
8. Immunology................................................. 172
9. Vaccines..................................................... 174
10. Summary ................................................... 175
References ................................................... 177
xviii CONTENTS
11. The Chlamydial Pneumonias

BURKE A. CUNHA
1. Introduction.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
2. Microbiology.... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
3. Pathophysiology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
4.1. Infant Chlamydiai Pneumonia ........................... 187
4.2. Psittacosis.............................................. 189
4.3. Chlamydia pneumoniae Pneumonia ......................... 191
5. Summary ................................................... 193
References ................................................... 195
12. Histoplasma capsulatum

STANLEY W. CHAPMAN and HAROLD M. HENDERSON
1. Introduction ................................................. 197

2. History ............ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
3. Mycology.................................................... 198
4. Host Defense ................................................ 198
5. Epidemiology................................................ 199
6.1. Acute Pulmonary Histoplasmosis ......................... 201
6.2. Chronic Pulmonary Histoplasmosis ....................... 201
6.3. Mediastinal Granuloma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 205
6.4. Mediastinal Fibrosis ..................................... 205
6.5. Pericarditis............................................. 205
6.6. Disseminated Histoplasmosis ................... . . . . . . . . .. 207
7. Diagnosis .................................................... 207
7.1. Skin Tests .............................................. 207
7.2. Tests for Antibodies ....................... . . . . . . . . . . . . .. 208
7.3. Tests for Antigen ....................................... 209
7.4. Culture and Histopathology ............................. 209
8. Treatment................................................... 211
8.1. Disseminated Histoplasmosis ............................. 211
8.2. Nondisseminated Forms of Histoplasmosis ................ 212
References ................................................... 212
CONTENTS xix
13. Blastomyces dermatitidis and Paracoccidioides brasiliensis

ROBERT W. BRADSHER and RICHARD W. MCDONNELL
1. Introduction ................................................. 217

2. Mycology.................................................... 217
3. Pathophysiology.............................................. 218
4. Epidemiology................................................ 218
5. Manifestations ............................................... 219
6. Clinical Markers ............................................. 219
6.1. Identification of the Organism ........................... 219
6.2. Serology ............................................... 220
6.3. Skin Testing ............................................ 221
7. Immunology..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
7.1. In Vivo Markers ......................................... 222
7.2. In Vitro Lymphocyte Studies ............................. 223
7.3. In Vitro Polymorphonuclear Studies ....................... 225
7.4. In Vitro Monocyte and Macrophage Studies ................ 227
8. Conclusions.................................................. 231
References ................................................... 232
14. The Immunology of Coccidioidomycosis

STANLEY C. DERESINSKI
1. Introduction ................................................. 239

2. 240
Coccidioides immitis ............................................
3. Immune Response ........................................... 241
4. Determinants of Outcome of Coccidioidal Infection-Hypothesis 243
References ..•................................................ 244
15. Pulmonary Cryptococcosis: Pathophysiological and Clinical

Characteristics
MIRIAM L. CAMERON and JOHN R. PERFECT
1. Introduction ................................................. 249

2. Organism ................................................... 250
3. Host Defenses ............................................... 253
4. Clinical Presentations ......................................... 258
4.1. Immunocompetent Host .. . .. .. .. .. .. .. .. . . . .. .. .. .. .. ... 258
4.2. Immunocompromised Host without HIV Infection ......... 264
4.3. Immunocompromised Host as a Result of HIV Infection ... 265
xx CONTENTS
5. Conclusion .................................................. 267

References ................................................... 267
16. Influenza Viruses

RICHARD V. SPERA, JR., and DAVID H. SHEPP
1. Introduction................................................. 281
2. Epidemiology................................................ 282
3. Clinical Overview ............................................ 282
3.1. Uncomplicated Influenza ................................ 282
3.2. Complications of Influenza .............................. 283
4. Viral Structure and Function .................................. 284
5. Pathogenic Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 285
5.1. Transmission ........................................... 285
5.2. Histopathology ......................................... 286
5.3. Antigenie Variation ..................................... 286
5.4. Effect of Infection on Immunologie Defenses .............. 287
5.5. Effect of Bacterial Infection on Influenza Virus .. . . . . . . . . .. 289
5.6. Effect of Infection on Respiratory Physiology .............. 289
6. Host Immune Response to Influenza Virus ..................... 289
6.1. Protective Immunity to Influenza . . . . . . . . . . . . . . . . . . . . . . . .. 289
6.2. MucosaIImmunity...................................... 290
6.3. Humoral Immunity ..................................... 291
6.4. Durability of Antibody Responses ........................ 291
6.5. Cell-Mediated Immunity ................................ 292
6.6. Immunopathogenesis.................................... 293
7. Prevention and Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 293
7.1. Vaccination............................................. 293
7.2. Antiviral Agents ........................................ 296
References ................................................... 298
17. Parainfluenza Viruses

RICHARD D. CLOVER
1. Introduction................................................. 309
2. Virology..................................................... 309
3. Epidemiology................................................ 310
4. Clinical Manifestations ........................................ 311
5. Diagnosis .................................................... 311
6. Immunity ................................................... 312
7. Treatment and Prevention . .. .. .. .. .... ..... .. .. .. .. ....... .. .. 312
References ................................................... 314
CONTENTS xxi
18. Varicella-Zoster Virus

GERALD LANCZ and STEVEN SPECTER
1. Introduction ................................................. 319

2. Properties of the Virus ....................................... 320
3. Virus Replication .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 322
4. Epidemiology of Varicella Zoster Virus Infections ............... 323
5. Pathogenesis................................................. 323
5.1. Varicella ............................................... 323
5.2. Herpes Zoster .......................................... 324
5.3. Clinical Course ......................................... 325
6. Diagnosis .................................................... 329
7. Therapy..................................................... 330
8. Prevention................................................... 331
References ................................................... 333
Index............................................................ 339
1
Defense Mechanisms of
the Respiratory Tract
RAMON G. CANTO, GEORGE R. ROBINSON 11,
and HERBERT Y. REYNOLDS
1. INTRODUCfION
The human respiratory tract, with its primary function of moving ambient air
into intimate contact with blood for gas exchange, is constantly confronted with a
multitude of noxious agents and elements that abound in the environment,
including a variety of microbial pathogens. Moreover, the nasooropharynx is
heavily colonized with a diverse group of microorganisms that can be aspirated
into the lower airways. It is therefore remarkable that respiratory infections are
not more common and are usually not serious for most humans. The host defense
apparatus of the lung is responsible for this protection. This system consists of
structural, mechanical, secretory, and cellular mechanisms that are designed to
eliminate or contain the majority of these pathogenic agents (Table I). The
human host is rendered more susceptible to pulmonary infections if any of the
system's components malfunctions or if the system is overwhelmed by a new
microbe, a particularly virulent microbial strain, or a large inoculum dose.
Three distinct portions make up the human respiratory tree: the nasooro-
pharynx, the conducting airways and the alveolar units. The nasooropharynx and
the larynx comprise the upper airway, which begins at the point of air intake at
the alae nasi and the lips of the mouth. Here, each breath is warmed and
RAMON G. CANTO and GEORGE R. ROBINSON II • Pulmonary/Critical Care Division; and

HERBERT Y. REYNOLDS • Department of Medicine, The Pennsylvania State University, M.S.
Hershey Medical Center, Hershey, Pennsylvania 17033.
Pulmonary Infections and Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
2 RAMON G. CANTO et al.
TABLE I
Defense Mechanisms of the Respiratory Tract
Defense mechanism Defect Potential infection
Airway Defenses
Aerodynamic barriers Endotracheal intubation, Aspiration, direct entry of
tracheostomy microbes into airways
Mucociliary clearance Ciliolytic microbes, ciliary Mycoplasma, stagnant
defects, cystic fibrosis secretions, bronchiectasis,
recurrent infections
Airway reflexes Depression of cough reflex Poor removal of secretions
Preferential bacterial Poor nutrition, fever, local pH Colonization and eventual
adherence infection
Local immunoglobulin IgA deficiency Viral infections
proteetion (IgA)
Alveolar Defenses
Alveolar macrophages Immunosuppression Pneumocystis carinii,
Legionella sp mycobacteria
Polymorphonuclear Granulocytopenia, defective Poor inflammatory response,
neutrophils PMNs gram-negative bacilli and
fungal infection
Surfactant Deficiency due to decreased Probably a negative effect on
synthesis or increased usage alveolar macrophage
functions, atelectasis
Immune opsonins (lgG) IgG deficiency Pneumonia with encapsulated
bacteria
Complement H ypocomplementemia Decreased dearance of
pneumococci and
Pseudomonas aeruginosa,
increased susceptibility to
infection
Lymphocytes AIDS Pneumocystis carinii
pneumonia, fungal infection
Augmenting Mechanism
Generation of an Granulocytopenia or defective Same as for PMNs
inflammatory PMNs, complement
response deficiency, defective
Iymphocytes
Adapted from Reynolds.l
humidified then rapidly moved through the conducting airways. The trachea
marks the beginning of the conducting airways, which include 16 generations of
bronchi and respiratory bronchioles. These contain approximately 200 ml of air.
Distal to the respiratory bronchioles, the airstream reaches the alveolar ducts and
diffuses into the alveoli where gas exchange occurs. These alveolar units have an
enormous surface area of more than 100 m 2 and can inflate to accommodate 3 to
4 liters of gas volume. Air is then expired, augmented with carbon dioxide; water
RESPIRATORY DEFENSE MECHANISMS 3
vapor is extracted as the air courses out of the respiratory tract. Blood supply to
the conducting airways is provided by the bronchial arteries whereas the pulmon-
ary arterial system supplies the alveoli.
2. AIRWAY DEFENSES
The respiratory tract contains a complex series of mechanisms that defend

the lung by minimizing contamination of the lower tract from above, by facilitat-
ing clearance of those noxious or pathogenic agents that do penetrate through the
upper airway defenses and by destroying or neutralizing any contaminants that
may remain. In the nasooropharynx the defensive scheme consists of the removal
of particulate matter from inspired air, the resistance to microbial growth on the
mucous membranes, and the mechanisms that prevent aspiration.
2.1. Aerodynamic Defenses and the Mucociliary Escalator

The earliest set of defenses met by potentially pathogenic microbial agents
are the anatomicalbarriers aerodynamically arranged to filter out and remove
particulate matter from the airstream. Particles suspended in air tend to setde
back onto surfaces if they are larger than 20 nm in diameter size. Smaller particles
of 10 nm diameter or less account for most of the material inhaled in the air. The
filtration process begins in the nose, where large particles of more than 10 nm in
size are either trapped and removed by the nasal vibrissae or impacted on the
surface of the nasal septum and turbinates. This process is facilitated by the
design of the nasal passageways, which prornotes turbulent flow. As the airstream
changes direction in the nasopharynx, most of the remaining particles, 5-10 nm
in size, impact on the posterior pharyngeal wall, where the tonsillar lymphoid
ring helps in their final elimination. The airstream again encounters turbulent
flow as it enters the upper trachea and the narrowed supraglottic portion, causing
further particle deposition on the mucociliary blanket. Below the vocal cords,
airflow slows and becomes laminar. From this point, sedimentation becomes more
important than impaction and most particles of 0.2 to 5 nm are removed as the
airstream proceeds distally. Both the number and the angulations of the bron-
chial branches are important determinants of the sedimentation process. In the
alveolar ducts and alveoli, the smallest particles randomly distribute by Brownian
motion, unaffected by inertia and gravitational forces. Most are exhaled rapidly.
An important adjunct to the aerodynamic defenses are the airway reflexes.
Cough, which is mediated by the vagus nerve, is elicited by stimulation of
chemical and mechanical receptors located in the subepithelium of the hypo-
pharynx, larynx, and tracheobronchial tree. Under normal circumstances, the
laryngeal apparatus allows very few oropharyngeal secretions to pass into the
trachea. Moreover, the mucociliary clearance mechanism is highly efficient.
Cough assists this mechanism when secretions become excessive or when foreign
material enters the trachea. In a similar fashion, the sneezing reflex helps clear
4 RAMON G. CANTO cl al.
the nasal passageways of secretions and trapped matter. Reflex bronchoconstric-

tion minimizes the entry of inhaled matter into the more distal parts of the
tracheobronchial tree. Also, by reducing the cross-sectional area of the bronchi,
cough or forced expiration becomes more effective in pushing endobronchial
debris mouthward.
After aerosolized particles are trapped in the airway lining surface, mucocili-
ary dearance and other mechanisms will begin working on their removal or
inactivation. The complex lining surface consists of a mucosallayer, the lamina
propria below the basement membrane, and a rich network ofvascular and nerve
supply. Except in the nares and the larynx where the surface epithelium is
squamous, a pseudostratified, columnar, ciliated, mucus-secreting epithelium
covers the lining of the airways. induding the nasal turbinates. As the airways
become smaller, the epithelium becomes thinner and less stratified and the cells
ass urne a more cuboidal shape with shorter cilia. Eventually, it flattens into a
single layer of type 1 pneumocytes that line the alveolar ducts and alveoli.
The ciliated respiratory epithelium acts in concert with the secretory cells to
provide the defense mechanism known as the mucociliary transport system. Each
ciliated epithelial cell contains a tuft of about 200 cilia that beat rapidly at an
average rate of 600 times aminute. 2 The coordinated beating movement continu-
ously sweeps and propels the overlying mucous blanket and other admixed debris
toward the oropharynx at a rate of 0.5 to 1 mm/min in the small airways and 5
to 20 mm/min in the larger airways. 3 In normal man, 10 to 100 ml of mucous
secretions pass up the trachea into the hypo pharynx every day and are mostly
swallowed. 4 ,5
Ultrastructural studies of cilia reveal a wheel-like arrangement ofnine outer
microtubule doublets connected to each other by nexin links (Fig. 1). The smaller
of the doublet has two short dynein arms. Radial spokes extend from each outer
microtubule pair and converge on the sheath surrounding a central microtubule
pair which coordinates the sliding mechanism responsible for normal ciliary
movement. A variety of congenital structural abnormalities may render cilia
completely immobile or significantly dysfunctional, thus altering the rate of
mucociliary dearance. Defects have been identified in the dynein arms as seen in
Kartagener's syndrome, in the radial spokes and in the microtubular apparatus.
The abnormalities are collectively known as the ciliary dyskinetic syndromes. 6
Ciliary function can also be depressed or impaired by physical or chemical agents
such as cigarette smoke and other pollutants, drugs such as anticholinergic agents
and alcohol, and 100% oxygen. Ciliolytic microorganisms such as Mycoplasma and
several viruses as weIl as certain soluble bacterial products such as pyocyanin
derived from Pseudomonas aeruginosa mayaiso inhibit ciliary activity. 7 Abnor-
malities in the quality of mucus and the mucous blanket also lead to impaired
mucociliary dearance, as seen in cystic fibrosis and other causes ofbronchiectasis.
The mucous blanket is mainly produced by the goblet cells and the mucous
secretory cells and glands. The goblet cells are dispersed throughout the epithe-
lial surface in a ratio of approximately five ciliated cells for every goblet cello The
goblet cells release their mucus by rupture of the cell surface whereas mucous
-'I:-T-.---- peripheral
doublet
-f------;;#--='--++--- central
microtubules
' - - - r , f - - - - - inner dynein arm
nexin link
spoke
outer dynein arm
FIGURE 1. Diagram of the cross-section of a cilium or of the central portion of a sperm tail. The
assembly of the nine outer microtubular pairs and the two central microtubules is held together
by three kinds of connections: the dynein arms, the nexin links, and the spokes. (From Wilson et
al.,7 with permission. )
gland secretions make their way onto the surface epithelium through ciliated
ducts. Clara cells are nonciliated bronchiolar secretory cells found only in the
terminal bronchioles, an area where goblet cells are usually sparse. The serous
cell is another secretory cell that is present in human fetus but has not been
identified in the human adult. 8
Absorption of fluid along the airways is important to maintain an equilib-
rium, keeping the surface moist while minimizing changes in net volume. The
absorption process includes escape of moisture from the mucosal surface,
lymphatic drainage, and local absorption probably mediated by microvillous
brush cells. 9
Much of the barrier function that restricts the penetration or absorption of
particles or gases into the subepithelium of the bronchial mucosa is provided by
the tight apical junctions between the epithelial cells. 1O These junctions can be
breached by inhaled irritants, vasoactive substances, or an antigen-antibody
reaction, resulting in increased permeability. Mechanical deformation of the
junctions by the rapid discharge of mucus from swollen goblet cells can also
cause a transient increase in mucosal permeability.
The basement membrane lies beneath the pseudostratified epithelium and
provides structural support. It is rather permeable, allowing ready access to the
lamina propria below. The lamina propria is a complex area containing various
plasma cells and mast cells and the basal portions of the bronchial glands. These
bronchial mucous glands have a volume about 40 times that of goblet cells ll which
6 RAMON G. CANTO el al.
has led to the assumption that they may be quantitatively more important in
mucus secretory function.
The normal mucociliary transport mechanism can clear entrapped particu-
late matter rapidly, removing half of a given number of particles within minutes.
In the nonciliated alveoli, clearance is much slower. Nonetheless, almost all
material that is deposited on the normal mucous blanket is removed within 24 hr.
The exact physiological and neurological mechanisms that control ciliary func-
tion and mucus production remain incompletely understood.
2.2. Bacterial Mueosal Adherenee

A variety of aerobic and anaerobic bacteria inhabit the mucosal surface of the
nasooropharynx. 12 Common isolates include Neisseria sp., a variety of Streptococcus
sp., Corynebacteria, Staphylococcus sp., and Moraxella catarrhalis. In people with
chronic airways disease, Streptococcus pneumoniae and Haemophilus sp. can likewise
be recovered. Fungi and protozoa have also been found. Quantitatively, anaerobic
bacteria abound in the gums and around the teeth, reaching 108 mieroorganisms
per ml of oral secretions. Gram-negative bacilli and viruses are not common. This
complex residential bacterial flora appears to symbiotically exist in the normal
human host for reasons that remain unclear.I 3
A healthy mucosa is able to regulate the bacterial flora and is not overrun by
microorganisms. The ability of the nasopharyngeal epithelium to preferentially
allow adherence of relatively nonpathogenic microbes is an important host
defense. Bacteria have various mechanisms for attaching to mucosal cells. These
include surface adhesions that fit into various sugar or glycosamine receptors or
special microbial adaptations such as pili that promote contact. Some bacteria
produee specific proteases or exotoxins whieh help clear away cell surface secre-
tions such as IgA, thus facilitating contact with the special reeeptors.I 3 Several
other factors such as pH, redox potential, temperature, composition of the oral
seeretions, inhibitory substanees such as lysozyme, and the different intrinsie
abilities of bacteria are thought to be contributory in the adherenee process.I 4
Undoubtedly, the nutritional status of the host, which may determine cellular rate
of mitotic renewal and regeneration, is an important factor too.
As long as the host defense apparatus is intact, the microbial population in
the nasooropharynx will not cause infection. Furthermore, when normal hosts are
exposed to high numbers of gram-negative baeteria, the organisms are effectively
cleared within a few hours. 15 Patients who are eritically ill, postoperative patients
and patients with acute infections tend to permit increased numbers of gram-
negative bacteria to adhere to the epithelial surfaee. 16 This leads to colonization
which in turn, predisposes the host to infection.
2.3. Immunologie Defense of the Airways

In the framework of the airway mucosal surface lie various immunologie
structures. Certainly, the respiratory tract is an immunologie organ because all the
ingredients needed to produce an immune response can be identified.
3. LYMPHOCYTES
In the human body, lymphoeytes are found in the respiratory tree from the
nose to the alveoli. They may be found as individuallymphoeytes, as small groups
or as paratraeheal and hilar lymph nodes. Bronehoalveolar lavage studies have
shown T eells to be the predominant lymphoeyte in the alveoli outnumbering B
eells by more than 10:1 and natural killer eells by more than 100:1.
3.1. Immunoglobulins
The B eells are the effeetors ofhumoral immune response. They produee the
five families of immunoglobulins. (IgG, IgA, IgD, IgE, IgM). These eells arise
from the same stern eell as T eells but then travel to the liver and spleen for
differentiation in utero and in the bone marrow in adult life. The immunoglobu-
lins on the B-eell surfaee are responsible for binding antigens and almost any
other immunogenie moleeule. Unlike the T lymphoeytes, immunoglobulins do
not need the m<tior histoeompatibility eomplex moleeule to reeognize antigen.
Immunoglobulins are glyeoproteins with two disulfate-bonded heavy (H)
ehain subunits joined by interehanged disulfate bonds to two light (L) ehain
subunits to form a tetramer (Fig. 2). The variable regions are at the distal ends
of the tetramer and this is where antigens are bound. The Fe region binds to
surfaee Fe reeeptors on T eells, maerophages, neutrophils, and eosinophils.
Light ehains have one variable (VL) and one eonstant (CL) region. Heavy
ehains have one variable (VH) and three to four eonstant (CH) regions depending
upon the immunoglobulin type. The eonstant regions are the same sequenee in
strueture as all other ehains in that isotype and subclass. Complement is bound to
portions of the constant regions of IgGs and IgMs. The variable regions have
seetions of extreme variability where antigens are bound.
Immunoglobulins are divided into five groups. IgM is a large moleeule with
five immunoglobulin monomers grouped as a pentamer and joined by a] ehain.
IgM is the first immunoglobulin to appear in an immune response (primary
response) and is the first immunoglobulin to form in neonates. IgG is a monomer
that funetions as a secondary antibody response whieh oecurs after reehallenge
with an antigen. There are four subtypes of IgG ealled IgGI' IgG2 , IgG 3 , and
IgG4 • With respeet to the proportion of IgG, the greatest is IgG] and the least is
IgG4 • The IgG group eomprises about 75% of serum immunoglobulins. IgA is
found as a polymer (two IgA monomers linked by aJ ehain). IgA is found in body
seeretions (tears, nose, GI traet including saliva, milk) and has two subtypes. IgA]
is found eireulating in the serum while IgA 2 is most prevalent in the seeretions
listed above. IgD is found as a monomer but only in minute quantities. Its fune-
tion is not yet known. IgE is also a monomer found in minute quantities. IgE has a
great affinity for mast eells when stimulated by allergens. This eauses the eharae-
teristie wheal and Rare response seen in allergie individuals.
In the respiratory traet, immunoglobulins A, G, and E are the important
classes. They are seereted by plasma eells dispersed throughout the lamina
propria and make their way up through the basement membrane and epithelium
x,y ~
..?"V x-<-r; 00
~
x,y x-<-r;
..?"V
Hinge Region
(Cleavage
Si/es)
Fab I "
Complement Binding Region

CH 2
I I
(') (')
Fe ~ ~
'E. 'E.
:::J :::J
~
CH 2
~
o
52
z
COO· COO· d
~
FIGURE 2. Schematie model of an IgG molecule. VL and VH indicate the variable regions. whereas CL and CH indicate the
j:.
eonstant regions. The hinge region is more flexible and more exposed to enzymes and ehemieals; it is the area where c1eavage oceurs
to produee Fab and Fe fragments.
before reaching the airway surface where they are mixed with mucus. In non-
smokers, washings ofthe nasal mucosa have shown that IgA accounts for approx-
imately 10% and IgG approximately 3.5% of protein content.I7 Lacrimal secre-
tions, earwax, and stimulated parotid fluid also have a high content of IgA.I8
Salivary gland secretions in the oral pharynx contain mostly IgA and little IgG.
The local humoral response in the upper airways where relatively large parti-
des are deposited is predominantly IgA antibody. Most of the IgA in the mucosal
surface is in the dimeric form and binds to secretory component produced by
serous cells to form secretory IgA. The functions of IgA in protecting the airways
are not entirely dear. It is not an efficient opsonin and does not potentiate the
function of complement. It can agglutinate microbial partides, which may
enhance mucociliary dearance. By itself, it can neutralize several kinds of respira-
tory viruses induding rhinovirus, influenza, and respiratory syncytial virus. It is
also involved in the mechanisms of preferential bacterial adherence.
Immunoglobulin G, which is considerably smaller than secretory IgA, does
not require a secretory vehide as a transporting mechanism. It acts primarily in
the mucosa itself to limit invasion by microorganisms that reach the epithelium. It
is a good opsonin and usually fixes complement. Its concentration in airway
secretions may rise a hundredfold by change in the vascular permeability. The
presence of normal amounts of IgG must compensate adequately for the function
of secretory IgA, because most IgA-deficient hosts are asymptomatic and do not
have increased susceptibility to infection. 19 In contrast, patients who are defi-
cient in IgG or certain IgG subdasses usually require medical attention because
of recurrent respiratory infections.
The relative proportions of IgA and IgG change throughout the bronchial
tree. In the trachea and bronchial divisions, IgA accounts for approximately 10%
of the protein content in secretions while IgG content is about 1%. In the alveoli,
IgG content increases to 10-15% while IgA decreases to 5% of the protein content
in secretions. Perhaps this design emphasizes a greater relative importance of the
protective functions of IgG.
3.2. T Lymphocytes
The T lymphocytes' major function in host defense is to differentiate invad-
ing foreign antigens from self. Once identified as foreign, the T cell marks the
antigen upon which further components of the immune response mechanism act.
When there is a breakdown in the recognition of self versus nonself, autoimmune
diseases can result. Other functions of T cells indude both positive and negative
regulation of other branches ofhost responses. This is done through the differen-
tiation of T cells into T4 helper (CD4) cells and T4 suppressor (CD8) cells as
described below. T cells are also involved in Type IV delayed hypersensitivity
reactions.
T cells require the presence of antigens on the cell surface of an antigen-
presenting cell to be recognized. Since these antigen-presenting cells must be
encoded within the major histocompatibility complex (MHC), differentiation of
self from nonself is enhanced. The T cells have specific receptors (TCR) which
actually find the antigens. The ability to recognize the large diversity of antigens
the T cell can be exposed to results from rearrangement of the TCR genes. This
process is similar to that performed by antibodies. Unlike the immunoglobulins,
the TCR needs to have antigens presented in combination with the autologous
major MHC proteins for recognition. It appears that the TCR recognizes the two
parts (MHC and antigen) as a unit and not as two separate entities. The inabil-
ity of the TCR to function properly is thought to result in T-cellieukemias and in
autoimmune diseases such as rheumatoid arthritis. 20
T cells are divided into two subgroups, CD4 and CD8. CD4 cells are also
known as the helper cells as they are found to increase immunoglobulin response
and they are involved with the release ofvarious lymphokines. 21 The CD4 gene is
found on chromosome 12 which codes for production of the molecule. 22 The CD8
suppressor T cell, in distinction to the CD4 cell, suppresses the immune response.
A lack of CD4 cells is thought to be one of the major mechanisms of immune
deficiency in the Acquired Immunodeficiency Syndrome (AIDS). The CD8 gene
is coded on chromosome 2. 23
Recent studies have identified a type of cell combining features ofboth T-cell
subgroups. These cells are called natural killer (NK) cells. They can effect
spontaneous cytotoxicity toward certain tumors and virus-infected cells with-
out the help of MHC. NK cells are found in small numbers in the lungs but their
function has yet to be identified. 24
3.3. Lymphoid Tissue and Cell-Mediated Defense
Generally, the respiratory tract has not been considered an important organ
of the lymphoreticuloendothelial system. In truth, with its ample lymphoidal
components, it is a significant part of this system. Lymphoid tissue in the
respiratory tree appears in various forms and inc1ude tonsils and adenoids in the
nasooropharynx, lymph nodes around the trachea, carina, and mainstem bronchi
(the hilar lymph-node complex), submucosal aggregates spaced along the airways
(bronchus-associated lymphoid tissue), and detachable or free lymphoid cells on
the alveolar surface. The follicular organization and capsule present in the lymph
nodes that drain the lungs are similar to other systemic lymph node tissues.
Bronchus-associated lymphoid tissue (BALT) is not nearly as prominent in
humans as in rodents, rabbits, and fowl. They are relatively similar in structure to
Peyer's patches in the small intestine. 25 Although BALTwas first recognized more
than 100 years ago, its structure and function are not completely known. 25-27 The
strategie location ofBALT at branching points in the conducting airways suggests
that these structures playa role in local immunity, at least in the afferent limb. At
these locations, BALT may intercept particulate matter deflected from the air-
stream. Their specialized sticky covering, which is devoid of ciliated cells and
goblet cells, appears suitable for trapping and retaining substances. 28
BALT has been evaluated most extensively in rodents and rabbits. It consists
of isolated aggregates of one or two follicles immediately below the epithelium.
The overlying epithelium is histologically different from nearby tissue as it tends

to be more flattened and irregularly shaped. The epithelium mayaiso lack cilia
and glandular tissue and goblet cells are absent over the central point of the
follicle. 29 In the BALT follicles, cells are mainly small and medium-size lympho-
cytes and the follicles lack the capsule and germinal centers characteristic of true
lymph nodes. There are also occasional macrophages and scattered reticulum
cells,3° The macrophages appear to initially trap the antigens and then process
and present them to the appropriate lymphocytes. T and B lymphocytes have
been identified in cell suspensions derived from rabbit BALT. B cells tend to
predominate in these suspensions, with a B-cell to T-cell ratio similar to that
found in Peyer's patches-approximately 20% of the lymphocytes are T cells and
40-78% are B cells. 27 This is in distinction to lymph nodes or peripheral blood, in
which 70% of lymphocytes are T cells and 15-20% are B cells. Rabbit BALT
preparations have also been examined for the presence of various classes of
immunoglobulins. IgA and IgM have predominated with IgG being less nu-
merous. 27 .29 Immunoglobulin-bearing cells could migrate directly from BALT
to the lamina propria along the respiratory tract and produce local antibody for
the mucosal surface.
Summary findings suggest that BALT function in the immune response may
be flexible or versatile. BALT may serve as a reservoir for various immunocompe-
tent cells, including cells with the potential to secrete IgG, IgA and IgE. Reserva-
tion about the universal importance of BALT is appropriate since there is
considerable species variation in the physical prominence of BALT in the major
airways. Chickens, mice, rats, and especially rabbits appear to be ideal models,
with abundant lymphoid tissue in the trachea and the bronchi. However, BALT is
poorly developed in neonates and germ-free animals. Study in the higher animal
species is continuing.
3.4. Antigen-Presenting Cells

Many antigenic substances are inhaled and must be removed and inactivated
by clearance mechanisms in the airways or they may persist and possibly initiate
an immune response. For this immune response to develop, the antigen must be
captured, processed, and then presented to a compatible T lymphocyte by an
antigen-presenting cell (APC). A number of cells are capable of presenting
antigens to the T lymphocytes, including macrophages, Langerhans cells, den-
dritic cells, fibroblasts, endothelial cells, and B lymphocytes. 31 Antigens can be
inhaled and deposited at any level of the respiratory tree so that any of several cell
type& are equipped to deal with them. The overall importance of surface phago-
cytosis in the respiratory tract is unclear and may be dependent on species
variation. BALT, phagocytic macrophages, and dendritic cells are most likely
involved in this process. BALT has been previously considered. A few macro-
phages may reside on the mucosal surface since they are retrieved by bronchial
lavage,32 They represent either senescent or impotent alveolar macrophages on
their way to expulsion from the lungs or airway macrophages. In recent studies,
viable airway macrophages that are morphologically distinct from alveolar macro-
phages have been found to reside at the air/surface interface of major human
airways.33 Their number is small and little is known of their function. Whether
they are better antigen-presenting cells than their alveolar counterparts (which
are not considered to be potent APCs) remains to be seen.
In humans, the most effective antigen-presenting cells appear to be the
dendritic cells. These are elongated cells with several slender cytoplasmic pro-
ces ses extending from the ceIl body. They represent approximately 1% of the
epithelial cells in the mucosa of large airways and are interspersed among the
columnar cells. 31 They are also found in vascular walls, in the visceral pleura, and
distaIly as far as the alveolar septa. Cigarette smoking may increase the number
and activity of these cells. 34 Mouse studies have demonstrated that dendritic cells
have good accessory function in stimulating antigen-specific T cells. Because the
dendritic cellular processes do not reach the mucosal surface, antigens would
need to adhere and actually penetrate the mucosa to make contact with these
processes. This would suggest that the antigens must remain in place or actually
attach to the mucosal surface in order for the antigen-presenting system to begin
operation. This requirement for sticking and penetration of the mucosa is, of
course, opposed by the airway mucosal barrier effect, mucociliary dearance, and
other airway defense mechanisms that remove many inhaled antigens prior to
their actual penetration.
4. DEFENSES IN THE ALVEOLAR MILIEU
The conducting airways terminate at the level of the respiratory bronchioles,

an area of transition before the airstream reaches the alveoli. Airflow velocity at
this point slows markedly to allow air molecules to diffuse into the alveolar spaces
so gas exchange can occur. The reduction in airflow velocity is caused by the
enormous increase in the cross-sectional area of the airspaces (to about 100 m 2
and severalliters of volume). Proximal airway defense mechanisms no Ion ger
operate. There is a gradual thinning of the lining epithelium from a pseudo-
stratified ciliated type to a single-Iayer surface with disappearance of goblet ceIls
and mucus-producing glands. The blood supply also changes over to the pulmon-
ary circulation. With the absence of glandular and cellular secretory components
and the proximity of a rich vascular and lymphatic network, defensive strategy
changes. Local and intravascular humoral and phagocytic components induding
macrophages, inflammatory cells, surfactant, immunoglobulins, and comple-
ment factors take over the defensive tasks.
4.1. Noncellular Components of the Alveolar Lining Fluid

By broncho-alveolar lavage, a lipoprotein fluid film can be recovered from
the alveolar spaces.3 5 The lavage fluid, which is a diluted sam pie of the alveolar
epitheliallining fluid, contains various noncellular substances that are important
for normal alveolar defense function. These include surfactant, other phospho-
lipids, neutral lipids, enzymes, and proteins including albumin, transferrin,
alpha-l antitrypsin, immunoglobulins, and complement. Proteins in respiratory
secretions come from two major sources: local synthesis by B lymphocytes and
plasma cells in the airway lumen and transudation of plasma proteins through the
capillary endothelial-alveolar epithelial interface. The degree to which transuda-
tion occurs depends largely on molecular size and configuration and on the
integrity of the blood-air barrier which is normally selectively permeable but may
undergo changes in association with inflammation, irritation, or injury. Immuno-
10gicaIly, three different proteinaceous groups are important nonimmune opso-
nins, immune opsonins, and components of the complement system.
4.2. Nonimmune Opsonins

At least three substances can coat or opsonize particles in a nonspecific way
that could increase phagocytic uptake. These nonimmune opsonins include
surfactant, fibronectin fragments,36 and perhaps, C-reactive protein,37 although
this is not present in normal bronchoalveolar lavage (BAL) fluid. 38 The most
important of these is surfactant.
Surfactant is synthesized solely in the Type 11 pneumocytes. It is composed of
structurally heterogenous, phospholipid-rich lipoproteins and forms an insoluble
film at the surface of the alveolar lining fluid, modifying surface tension to
promote lung expansion and prevent collapse. Recent surfactant research has
recognized the interaction of the surfactant system with the lung host defense and
immune system,39 It appears that the role of surfactant in lung defense is
multi dimensional.
Studies have demonstrated that surfactant isolated from normallungs has
different effects on alveolar macrophage function. 40-43 The clinical significance
of this surfactant-alveolar macrophage interaction is incompletely understood
and studies investigating its mechanisms continue. This ultimately relates to the
potential impact of exogenous surfactant therapy on the host defense system.
Recent data indicate that the surfactant protein SP-A plays a major role in the
interaction with alveolar macrophages. 44-46
Investigations of the role of surfactant in the immune system and inflamma-
tory responses of the host suggest a suppressant function. For example, when
endotoxin-stimulated human alveolar macrophages were exposed to synthetic
surfactant, the release of inflammatory cytokines from these cells was signifi-
candy inhibited. 47
In addition to its potential role in immunoregulation and macrophage
function, surfactant may have a more basic function in lung defense. Because
surface tension is markedly reduced at end expiration, the surface film naturally
moves from the alveoli to the bronchioles. Particulate matter trapped in this film
as weIl as damaged cells may then be removed from the alveoli by mucociliary
clearance. 39 Also, various surfactant components may specifically bind to micro-
organisms and influence their removal.
14 RAMON G. CANTO el al.
4.3. Immune Opsonins

The more specific (immune) opsonins in the alveolar lining fluid are identi-
fied as antibodies. Quantitatively, immune opsonins in the lower respiratory tract
and alveoli are principally IgG which accounts for up to 15% of the total protein
content of the alveolar lining fluid. As in serum, IgG1 is most abundant (66%),
with IgG 2 next (28%), and IgG3 and IgG4 accounting for 2 and 5%, respectively.48
Immunoglobulin G is an excellent opsonin and also fixes complement. It facili-
tates phagocytosis and promotes lytic destruction of microbes and other antigens.
In addition to locally produced antibody, plasma IgG also contributes to lung
defense as an important component of the inflammatory transudate.
The content of IgA in the epithelial lining fluid is much lower and IgA
receptors on alveolar macrophages are few, suggesting a less important role of
IgA in alveolar space defense. 49 .50 Immunoglobulin A is not a good opsonin and
does not fix complement. Its main role in lung defense is in the upper and
conducting airways. Immunoglobulin E is also consistently identified in broncho-
alveolar lavage sampling but its role in the defensive scheme of the respiratory
tract is not yet known.
4.4. Complement System Components

Because titers of complement factors are low in normal BAL fluid, it is
uncertain how essential this system is in creating lysis that could disrupt a
microbe, enhancing opsonic phagocytosis, or creating immune complexes. In the
normal host, complement components probably enter the alveolar milieu by
diffusion through the normal blood-air barrier. In vitro, cultured alveolar macro-
phages can also synthesize certain complement components. 51 U se of the alterna-
tive pathway for complement activation in the lung is supported by analysis of
available precursor molecules in BAL fluid 52 that can produce several active
fragments important for opsonin-dependent receptor-mediated phagocytosis
(C3b) and for inflammation by generating chemotactic activity (C5a). Factor B,
also from the alternative pathway, is found consistently in BAL fluid. It is a
homologue of classical pathway factor C2 and serves as a growth factor for
antigen-stimulated B lymphocytes 53 and as a cofactor for human macrophage
spreading .54
It is clear that complement plays an important role in the defense against
pulmonary infections. In mice rendered hypocomplementemic by cobra venom
factor, impaired pulmonary clearance of Streptococcus pneumoniae and Pseudo-
monas aeruginosa has been demonstrated.5 5•56 Pulmonary clearance of Staphylo-
coccus aureus and Klebsiella pneumonia is apparently not affected by complement
depletion. 57
4.5. Cells in the Alveolar Space

In a typicallavage of the lungs of anormal nonsmoker, approximately 10 to
20 million respiratory cells are recovered.58 Macrophages represent 85% of the
differential count, appearing in a variety of sizes and forms. The rest are
lymphocytes and a small number of neutrophils. Eosinophils and basophils are
rarely detected.
4.5.1. Alveolar Macrophages
The alveolar macrophage is one of many mononuclear phagocytes that

comprise the reticuloendothelial system. Besides the lung macrophages, other
resident members of the macrophage cell line include the Kupffer cells in the
liver, microglial cells in the central nervous system, Langerhans cells in the skin,
and osteoclasts in bone. Macrophages themselves are derived from circulating
monocytes that are originally derived from the bone marrow. Alveolar prolifera-
tion of macrophages has also been demonstrated although this probably consti-
tutes only a small number of total macrophages. This intrapulmonary production
may, however, be of consequence during severe lung inftammation.
The macrophage is the resident phagocyte in the lungs. Lung macrophages
are found in the alveoli, the airways, the pulmonary intravascular component, the
connective tissue, and the pleura. Airway macrophages have been discussed
previously. Pulmonary intravascular macrophages are found in the lung capil-
laries. They have been identified in several animal species and recently have been
studied in humans. 59 These macrophages are noted to adhere firmly to the
vascular endothelial cell and are morphologically different from blood mono-
cytes. Although their true role in lung defense still needs to be determined,
animal studies have shown that they are able to clear circulating bacteria and
other particles from the pulmonary bloodstream. 6o They mayaiso play an
important part in the pathogenesis of adult respiratory distress syndrome. In the
connective tissue, the macrophages are usually situated in the subepithelial
surface or in the cross-epithelial barriers. Because of their location, they are not
directly exposed to airborne particles. They have intimate contact with elastin
and collagen, which may explain their role in the pathogenesis of emphysema.
These macrophages have been found to possess an increased ability to synthesize
DNA as compared to alveolar macrophages in vitro. Their role in lung defense is
otherwise ill defined. Very little is known about the pleural macrophages, espe-
cially about their role in the pathogenesis of pleural disease.
The pulmonary alveolar macrophages are the primary phagocytic cells in the
alveolar space. They occupy less than 1% of the alveolar surface but because of
their intrinsic roving capacity and their ability to permeate through intra-alveolar
pores, they can provide omnipresent coverage for alveolar defense. They are
reusable and enjoy a life span of several months to years. They play a dual role in
lung defense: They are the first line ofphagocytic defense and efficient immune
effector cells. Alveolar macrophages are not a homogeneous cell type but consist
of subpopulations that also have different functional characteristics. How these
subpopulations vary at certain ages of the host or in different disease states is
largely unknown. Some of these cells seem to be more involved in phagocytosis
while others are primarily involved with media tor release. Dysfunction depends
on different receptor expressions from the cell membranes of the different
sub populations. The number of these different subpopulations will vary accord-
ing to the stress placed on the lung. In acute inftammation, an increased number
of small monocyte-like macrophages is noted, whereas in chronic lung diseases,
the macrophages tend to be larger and more mature. 61
Macrophages can interact with other cells and molecules through a large
number of secretory products and the expression of several surface receptors
(Table 11). The number of receptors is likely to expand as research continues.
These receptors are known to couple with various secondary messengers and
activate a cascade of intercellular processes. The secretory products generated by
alveolar macrophages number more than 100. These include hormones, enzymes
such as proteases, and anti-proteases, peptides, major histocompatibility complex
proteins, bioactive lipids, lectin-binding molecules, and oxygen radicals. These
oxygen metabolites include hydrogen peroxide, hydroxyl, and superoxide and
are produced as a result of the respiratory burst associated with the marked
increase in oxygen uptake by the macrophage during phagocytosis.
Lysozyme is the major secretory enzyme of the alveolar macrophage, ac-
counting for about 25% ofthe protein material released. 63 Lysozyme is a glycos-
aminidase that is thought to be bactericidal for many gram-positive organisms.
Other secretory enzymes include elastases and collagenases. Angiotensin-
converting enzyme is also known to be released by alveolar macrophages and is
increased in patients with pulmonary sarcoidosis and in cigarette smokers. 64
Alveolar macrophages also produce antiproteases to maintain balance within the
lung. The major antiprotease is the alpha 2M antiprotease which is able to bind
the majority of proteases and some elastases.
In the inftammatory state, the alveolar macrophages release several polypep-
tide hormones including cytokines, interleukin-l, and tumor necrosis factor.
These factors are involved in the activation of lymphocytes, polymorphonuclear
neutrophils (PMNs), fibroblasts, and tumor cells. When activated, alveolar macro-
phages also produce cyclooxygenase- and lipooxygenase-derived proteins which
convert arachidonic acid to metabolites called eicosanoids. These are phlogistic
factors that are important mediators in the immune and inftammatory response.
Through phagocytosis and production of oxygen radicals and proteases, the
alveolar macrophages are able to remove most of the small particles in the distal
airways and alveoli. Activation of alveolar macrophages may oecur after direct
exposure to certain materials such as endotoxin or in response to cytokines
released by sensitized T lymphocytes. The activated macrophage is a larger cell
with a larger array of surface membrane receptors and hydrolytic enzymes,
making it a more efficient phagocytic cell with enhanced bactericidal or bacterio-
static activity. In the activated state, alveolar macrophages show increased activity
against such pathogens as Mycobacterium tuberculosis, LegioneUa pneumophila, Pneu-
mocystis carinii, Toxoplasma gondii, Listeria monocytogenes, Cryptococcus neoformans,
and cytomegalovirus. They also become cytotoxic for virus-infected cells.
Alveolar macrophages also synthesize and release chemotactie faetors for
PMNs, which they effectively recruit to reinforce defense. The release of these
neutrophil chemotactic factors is stimulated by phagocytosis ofbacterial particles
TABLE 11
Major Products Released
by Alveolar Macrophages
Cytokines
Interleukin-l alpha and beta
Interleukin-6
Tumor necrosis factor
Alpha and gamma interferon
Colony-stimulating growth factors
Transforming growth factor
Fibroblast growth factor
Neutrophil-activating factor
Enzyme-releasing peptide
Neutrophil chemotactic factor
Platelet-derived growth factor
Enzymes
Lysozyme, beta-glucoronidase
Acid hydrolases
Angiotensin converting enzyme
Elastase: serine and metalloenzyme
Collagenase: fibroblastlike and gelatinase
Plasminogen activator
Cathepsin L
Biologically active lipids
Cyclooxygenase metabolites
Thromboxane A2, prostaglandins
Lipooxygenase metabolites
5-hydroxyeicosatetraenoic acid
Leukotrine B4, C4, D4
Platelet-activating factor
Oxygen metabolites
Superoxide, hydroxyl, H 2 0 2
Proteins
Antiproteases
Alphai proteinase inhibitor
Alpha2 macroglobulin
Plasminogen activator inhibitor
Collagenase inhibitor
Glycoprotein: fibronectin
Complement components: C2, C4
Binding proteins: transferring, ferritin, apolipoprotein A
Other inhibitors
Free fatty acids
Antioxidants-glutathione
Coagulation factors-factors V and VII
1,25 Dihydroxyvitamin D3
Adapted from Sibille and Reynolds. 62
18 RAMON G. CANTO ct al.
such as Staphylococcus aureus and by the formation of immune complexes. This

release of chemotactic factors is augmented if the partides are first opsonized by
IgG or complement. In addition to this chemotactic function, alveolar macro-
phages also have cytotoxic effects against human tumor celllines. This activity is
thought to be related to release of oxygen radicals, especially hydrogen peroxide.
4.5.2. Lymphocytes
The cellular profile of the alveolar lining fluid is about 10% lymphocytes.
The heterogeneity of these lymphocytes recovered in BAL fluid has been studied
more thoroughly as phenotypic identification became readily possible with mono-
donal antibody reagents. Their functional characteristics have also been better
defined from analysis of cytokine production. The composition ofT lymphocytes
in the alveoli approximates that found in normal peripheral blood. Seventy
percent are T cells, of which the majority are of the T-helper/inducer subset. The
resulting T-helper to T-suppressor cell ratio is about 1:5. In lung diseases that
significantly affect lymphocytes, substantial differences in this ratio can occur.
Granulomatous diseases such as sarcoidosis65 and hypersensitivity pneumonitis66
or certain immunosuppressive states such as the acquired immunodeficiency
syndrome67 •68 are examples of such conditions. A mixture of plasma cells and B
lymphocytes constitutes the rest of the lymphocytic population of the alveolar
lining fluid. Some of these seem to contain surface immunoglobulins that can be
released as needed.69 The remainder of lung lymphocytes are untypeable and
may represent "null" cells.
4.5.3. Neutrophils
Polymorphonudear neutrophils are the human body's major host defense
against invading microorganisms. Their appearance outside of the circulation
suggests that an abnormal situation exists. In the normal lung, the largest
component ofPMNs are indeed in the vascular compartment. A small number of
PMNs are found in the air spaces at all times, however. In bronchoalveolar lavage
fluid, neutrophils account for up to 2% of lavageable cells. 58 These neutrophils
have migrated to the alveolar milieu in response to continuous low-grade genera-
tion of chemotactic factors, most likely by the alveolar macrophages. In the event
of an inflammatory situation such as an infection, the magnitude of PMN
migration is increased under the influence of several other chemotactic factors
released by complement components in addition to those generated by the
macrophages. In the alveolar space, these cells might be considered as a second-
ary line of phagocytic defense recruitable into the alveoli to augment the phago-
cytic function of the alveolar macrophages. The majority of neutrophils in the
peripheral blood have cell-surface receptors for IgG Fe. There do not appear to
be receptors for the other dasses of immunoglobulin. The neutrophils are drawn
to these receptors and then are able to start the inflammatory process. Both the
classic and alternative complement pathways can also generate complement

fragments which attract and activate neutrophils.
To provide ready access for the defense of the host, neutrophils are able to
diapedese through the walls of small blood vessels and migrate toward the
invasive organisms to be ingested. In addition, they are able to adhere to the
vascular endothelium and thereby have the ability to demarginate during an
invasion. Once the neutrophils recognize an invader, they attach to it and initiate
phagocytosis. Once the invader is phagocytized, neutrophils will discharge cyto-
plasmic granules which contain a vast array of digestive enzymes which, when
released, can act either intracellularly or extracellularly to kill and degrade
microorganisms.
Through their secretory products, neutrophils neutralize microorganisms
via two mechanisms. One mechanism is oxygen dependent and the other is
oxygen independent. The oxygen-dependent system relies on the neutrophils'
ability to secrete toxic oxygen metabolites in what is known as the respiratory
burst. The process involves release ofhigh-energy electrons which reduce oxygen
to superoxide. Superoxide is then broken down in the presence of water to
produce hydrogen peroxide which is the toxic component to invading organisms.
In the presence of myeloperoxidase from the primary granules and chloride ion,
hydrogen peroxide can be broken down to form hypochlorous acid. This is also
toxic to invading organisms. The oxygen-independent system includes the acid
hydrolases such as elastases and proteases, which will degrade proteins in various
microorganisms, and lysozyme which breaks down the cell walls in certain other
bacteria.
Because neutrophils have very little capacity for controlling their own fune-
tion, they are major mediators of host tissue injury when inappropriately regu-
lated. The same enzymes and reaetive oxygen species that are produced and
released by these cells in defending the host are capable of randomly destroying
host cells.
4.5.4. Eosinophils
Eosinophils share similar morphology, lysosomal constituents, phagocytic
capacity, oxidative metabolism, and most chemotactic responses with the neutro-
phils. There are, however, m~or differenees between these two eell types. Little is
known about the natural function of eosinophils. Eosinophils appear to be
associated with allergie reactions and parasitie infestations. They have a Ion ger
life span than neutrophils and unlike neutrophils they ean recireulate. Eosino-
phils normally comprise about 1-3% of eirculating leukocytes; however, in
allergie patients, lO-20% eosinophilia can oceur.
Less than 1% of the cells found in the alveolar lining fluid are eosinophils. 58
This number inereases in conditions associated with eosinophilia. Like neutro-
phils, these eells migrate to the lung under ehemotaetie influenee. Eosinophilie
granules contain, among other constituents, eosinophil peroxidase whieh eata-
lyzes the oxygenation of many substanees by hydrogen peroxide. Major basic

protein is an arginine-rieh polypeptide that is found mainly in the secondary
granules of the eosinophil. This agent is thought to mediate eosinophilie adher-
enee and toxicity in its defense meehanisms against helminths.
In the lung, there are essentially six oecasions in whieh involvement by
eosinophils is noted. These are allergie granulomatosis (Churg-Strauss syn-
drome), chronie eosinophilie pneumonia, allergie bronehopulmonary aspergil-
losis, drug-indueed pulmonary eosinophilia, tropical eosinophilia seeondary to
parasites, and acute pulmonary eosinophilia eaused by Ascaris lumbricoides. No
specifie eause is known for the onset of ehronie eosinophilie pneumonia or allergie
granulomatosis. The other four are eaused either by a known organism or drug.
5. INTEGRATED LUNG DEFENSE MECHANISMS
The following is an overview of sequential events that may typieally occur as

the respiratory tract defends against an invading antigenic mierobial pathogen.
The microbe is either inhaled or aspirated through the nasopharynx or
conducting airways. There it meets aerodynamic airway defenses and may impact
or stick at some point along the respiratory route because of air turbulence or
inertial and gravitation al forces. If the mueociliary transport system is unable to
remove the particle and it remains in contaet with the mueosa long enough to
irritate the epithelial surface and open up a tightjunction, it may gain aeeess into
the submucosa where it can make contact with the cellular processes of the
dendritic cells. The microbe mayaiso land fortuitously on BALT lymphoepithe-
lium. Some mieroorganisms have other meehanisms by which they gain entranee
into the respiratory tract. For example, Mycoplasma species and Bordetella pertussis
bind to specific reeeptors on cilia and eventually impair eiliary function. Strep-
tococcus pneumoniae does not attach to cilia but produces IgA protease that may
paralyze cilia. IgA protease also selectively degrades IgA l thereby facilitating
mucosal adherenee. Hemophilus injluenzae, Neisseria meningitides and some gram-
negative bacilli like Pseudomonas aeruginosa also elaborate IgA proteases. A variety
of virus es are ciliotoxic or may direetly penetrate the epithelial cells. Once
captured by the dendritie eell, the antigen is proeessed and presented to the
appropriate T lymphocyte for further action.
If the microorganism manages to elude the maze of airway defenses and
reaches the alveolar surface, a sequence of interaeting events among loeal im-
mune components will oecur that could esealate the immune response in a
eumulative fashion, if needed (Fig. 3). First, the microbe comes in contaet with the
alveolar wall where alveolar expansion and contraction is likely to roll it along in
the fluid film that overlays the epithelial surface. The chance exists for the
microbe to become enmeshed and coated with opsonins like surfactant. Eneapsu-
lated bacteria including Streptococcus pneumoniae, Hemophilus injluenzae, and Kleb-
siella pneumoniae and other microbes such as Staphylococcus aureus, Pseudomonas
aeruginosa and eertain fungi are all ingested more readily by phagocytes when
URT
Mechanical
Factors
Muco-ciliary
Transport
AERODYNAMIC
FILTRATION ~
ALVEOLUS
Lymphokines
FIGURE 3. An enlarged representation of the alveolar unit. Factors responsible for the
clearance of bacteria (B) in the upper respiratory tract (URT) are listed. Alveolar defense
mechanisms are quite different. A bacterium that escapes removal by upper airway defenses and
is deposited in the alveolus may be opsonized by surfactant secreted by Type II pneumocytes (II)
and/or immunoglobulins and complement components. It is then phagocytized by the resident
alveolar macrophage (AM). If the AM cannot contain or kill the phagocytized microbe, cytokines
released by immune Iymphocytes (T-LYM) may "activate" the AM and stimulate its bactericidal
capacity. In addition, the AM can liberate chemotactic factors which attract nearby polymorpho-
nuclear neutrophils (PMNs), thus initiating an inftammatory response. (Reprinted with permis-
sion from Reynolds. 70 )
Nl
Nl
BLOOD
MONOCYTEL
PRECURSER
Soluble IL- 2 receptors
~
~
o
~
z
(IL-4,5,6) d
~
~
they are coated with antibody, especially IgG.71 The major phagocyte is the
alveolar macrophage. If the macrophage cannot destroy or contain the microbe
after ingesting it, its bacteriostatic or bactericidal activity can be enhanced by its
interaction with cytokines produced by T lymphocytes in the lung (Fig. 4). A
predictable group of obligatory intracellular microorganisms is not contained
efficiendy by tissue macrophages of the reticuloendothelial system. In the lung,
these microbes include Mycobacterium tuberculosis, Legionella pneumophila, Pneumo-
cystis carinii, human immunodeficiency virus (HIV), and others previously men-
tioned. If the microbe is antigenie, the macrophage can process and present it to
an appropriate lymphocyte that will begin an immune response, the process
ending perhaps in specific antibody. Diseases that disable lymphocytic func-
tion such as the acquired immunodeficiency syndrome disrupts the normal
macrophage-Iymphocyte interaction hence the propensity for opportunistic lung
infections such as Pneumocystis carinii pneumonia.
Finally, the alveolar macrophage can release chemotactic factors that attract
other phagocytic cells especially the polymorphonuclear neutrophils into the
alveolar space. This is predicated on the availability of normal PMNs which are
normally provided by the marginated intravascular pool. Obviously, in the granu-
locytopenic patient or in the host with defective PMNs, the inflammatory reaction
is suboptimal, which predisposes the host to infection with gram-negative bacte-
ria and fungi.
6. CONCLUSION
The occurrence of respiratory infections is largely determined by the integ-

rity of the lung's defense mechanisms. The interaction between pathogenic
microorganisms and the respiratory tract is adynamie one. Various microbes
utilize different offensive strategies to gain a foothold in the host. These are
countered by an impressive array of defensive maneuvers that include mechanical
barriers, immunologie defenses, and, when required, an appropriate inflamma-
F1GURE 4. Summary of the interactions between the alveolar macrophage (AM) and lympho-
cyte in the alveolar space. An AM differentiates from a monocyte precursor with the help of 1,25
Vit.D3 and other maturation factors. Not only is the AM the major phagocyte in the alveolar
space; it is also an immune effector cell capable of inftuencing the function of other cells
especially when it is stressed or 'activated' by the T-helper lymphocyte (TH) through the action of
monokines such as interferon and migration inhibitory factor (MIF). In similar fashion, T-sup-
pressor lymphocyte (TS) can exert an opposite effect. The AM is involved in three important
immune interactions. First, it can process and present antigen to an appropriate major histocom-
patible (MHC) T-cell for further immune action. Second, the AM can be activated. And third,
the activated AM can stimulate other T-cell functions such as attracting other T-cells to the alve-
olar site, proliferation under interleukin-6 and growth factors (CF), activation of dormant killer
lymphocytes, and induction of B-cell differentiation yielding antibody. The various circuits as
discussed are complex and likely to be incomplete. Reprinted with permission from Reynolds. 7o
tory response. Considering the frequency of microbial challenge, this defense

system is remarkably efficient.
Past and ongoing research have provided asolid understanding of host-
microbial interactions in the lung. As further studies unfold, more insight will be
gained into how specific defects in the lungs' defense mechanisms lead to
pulmonary disease. Research on therapeutic interventions and preventive strate-
gies also need to keep up to decrease the substantial impact of respiratory
infections on patient morbidity and mortality as weIl as on socioeconomic aspects
of health care.
REFERENCES
1. Reynolds, H. Y., 1983, Altered lung host defenses and infection, in: Respiratory Inftctions:
Diagnosis and Management (J. E. Pennington, ed.), Raven, New York, pp. 1-23.
2. Rutland, w., Griffin, M., and Cole, P.]., 1982, Human ciliary beat frequency in epithelium
from intrathoracic and extrathoracic airways, Am. Rev. Respir. Dis. 125:100-105.
3. Wanner, A., 1977, Clinical aspects of mucociliary transport (state of the art), Am. Rev. Respir.
Dis. 116:73-125.
4. Reid, L., 1965, Natural history of mucus in the bronchial tree. Arch. Environ. HeaUh. 10:
265-272.
5. Toremalm, N. G., 1960, The daily amount of tracheobronchial secretions in man. Acta
Otolaryngol (Suppl). 158:43-46.
6. Eliasson, R., Mossberg, B., Camner, P., and Afzelius, B. A., 1977, The immotile cilia
syndrome: A congenital ciliary abnormality as an etiologic factor in chronic airway infections
and male sterility. N. Engl. J Med. 297:1-6.
7. Wilson, R., Pitt, T., Taylor, G., Watson, D., MacDermott,]., Sykes, D., Roberts, D., and Cole, P.,
1987, Pyocyanin and I-hydroxyphenazine produced by Pseudomonas aeruginosa inhibit the
beating of human respiratory cilia in vitro. J Clin. Invest. 79:221-229.
8. Gail, D. B., and Lenfant, C.]. M., 1983, Cells of the lung: Biology and clinical implications
(state of the art), Am. Rev. Respir. Dis. 127:366-387.
9. Reid, L., and ]ones, R., 1979, Bronchial mucosal cells, Fed. Proc. 38:191-196.
10. Hogg,]. C., Pare, P. D., and Boucher, R. C., 1979, Bronchial mucosal permeability, Fed. Proc.
38:197-201.
11. Reid, L., 1960, Measurement of the bronchial mucous gland layer: A diagnostic yardstick in
chronic bronchitis, Thorax 15:132-141.
12. Mackowiak, P. A., 1982, The normal microbial flora, N. Engl. J Med. 307:83-93.
13. Reynolds, H. Y., 1987, Bacterial adherence to respiratory tract mucosa-a dynamic inter-
action leading to colonization, Semin. Respir. Med. 2:8-19.
14. Gibbons, R.]., and van Houte,]., 1975, Bacterial adherence in oral microbial ecology, Ann.
Rev. Microbiol. 29:19-44.
15. LaForce, F. M., Hopkins,]., Trow, R., and Wang, W L. L., 1976, Human oral defenses against
gram-negative rods, Am. Rev. Respir. Dis. 124:929-935.
16. ]ohanson, W G., Higuchi, ]. H., Chaudhuri, T. R., and Woods, D. E., 1980, Bacterial
adherence to epithelial cells in bacillary colonization of the respiratory tract, Am. Rev. Respir.
Dis. 121:55-63.
17. Merrill, W. W, Goodenberg, D., Strober, w., Matthay, R. A., Naegel, G. P., and Reynolds,
H. Y., 1980, Free secretory component and other proteins in human lung lavage, Am. Rev.
Respir. Dis. 122:156-161.
18. Tomasi, T. B., and Plaut, A. G., 1985, IgA in mucosal immunity, in: Advanees in Host Defense
Meehanisms, 0.], Gallin and A. S. Fauci, eds.), Raven, New York, pp. 31-61.
19. Kaliner, M. A., 1992, Human nasal host defense and sinusitis. J Allergy. Clin. Immunol.
90:424-430.
20. Stamenkovic, 1., Stegagno, M., Wright, K. A., Krane, S. M., Amento, E. P., Colvin, R. B.,
Duquesnoy, R. ]., and Kurnick,]. T., 1988, Clonal dominance among T-Iymphocyte infil-
trates in arthritis, Proe. Natl. Acad. Sei. USA 85:1179-1183.
21. Reinherz, E. L., Kung, P. C., Goldstein, G., and Schlossman,]. F.,1979, Further characteriza-
tion of the human inducer T-cell subset defined by monoclonal antibody, J Immunol. 123:
2894-2896.
22. Kozbor, D., Finan, ]., Nowell, P. C., and Croce, C. M., 1986, The gene encoding the T4
antigen maps to human chromosome 12.J Immunol. 136:1141-1143.
23. Parnes,]. R., 1989, Molecular biology and function of CD4 and CD8, Adv. Immunol. 44:
265-311.
24. Takeuchi, M., Nagai, S., and Izumi, T., 1988, Effect of smoking on natural killer cell activity
in the lung, Chest 94:688-693.
25. Bienenstock,]., and Befus, D., 1984, Gut and bronchus-associated lymphoid tissue, Am. J
Anat. 170:437-445.
26. Bienenstock,]., 1984, Bronchus-associated lymphoid tissue, in: Immunology of the Lung and
Upper Respiratory Tract O. Bienenstock, ed), McGraw-Hill, New York, pp. 96-118.
27. Bienenstock,]., Clancy, R. L., and Perey, D. Y. E.,1976, Bronchus-associated lymphoid tissue
(BALT): Its relationship to mucosal immunity, in: Immunologie and Infeetious Reactions in
the Lung, (C. H. Kirkpatrick and H. Y. Reynolds, eds.) Marcel Dekker, New York, pp.
29-58.
28. Racz, P., Tenner-Racz, K., Myrvik, Q. N., and Fainter, L. K.,1977, Functional architecture of
bronchial-associated lymphoid tissue and Iymphoepithelium in pulmonary cell-mediated
reactions in the rabbit, J Retieuloendothel. Soe. 22:59-83.
29. Bienenstock,]., 1985, Bronchus-associated lymphoid tissue, Int. Areh. Allerg. Appl. Immunol.
76(Suppl 1): 62-69.
30. Racz, P., Tenner-Racz, K., Myrvik, Q. N., Shannon, B. T., and Love, S. H., 1978, Sinus
reactions in the hilar lymph node complex of rabbits undergoing a pulmonary cell-mediated
immune response: Sinus macrophage dumping reaction, sinus Iymphocytosis and imma-
ture sinus histiocytosis, J Retieuloendothel. Soe. 24:499-525.
31. Sertl, K., Takemura, T., Tschachler, E., Ferrans, v.]., Kaliner, M. A., and Shevach, E. M.,
1986, Dendritic cells with antigen-presenting capability reside in airway epithelium, lung
parenchyma and visceral pleura,J Exp. Med. 163:436-451.
32. Fick, R. B., Richerson, H. B., Zavala, D. C., and Hunninghake, G. W, 1987, Broncho-alveolar
lavage in allergie asthmatics, Am. Rev. Respir. Dis. 135:1204-1209.
33. Rankin, ]. A., Marcy, T., Rochester, C. L., Sussman, ]., Smith, S., Buckley, P., and Lee, D.,
1992, Human airway macrophages, Am. Rev. Respir. Dis. 145:928-933.
34. Soler, P., Moreau, A., Basset, F., and Hance, A.]., 1989, Cigarette smoking-induced changes
in the number and differentiated state of pulmonary dendritic cells/Langerhans cells, Am.
Rev. Respir. Dis. 139:1112-1117.
35. Reynolds, H. Y., and Chretien,]., 1984, Respiratory tract fluids: Analysis of content and
contemporary use in understanding Jung diseases. Disease-A-Month 30:1-103.
36. Rennard, S. 1., Hunninghake, G. W, Bitterman, P. B., and Crystal, R. G., 1981, Production of
fibronectin by the human alveolar macrophage: A mechanism for the recruitment of
fibroblasts to sites of tissue injury in the interstitial lung diseases, Proe. Natl. Aead. Sei. USA
78:7147-7151.
37. Mold, C., Rogers, C. P., Kaplan, R. L., and Gewurz, H., 1982, Binding ofhuman C-reactive
protein to bacteria, Infeet. Immun. 38:392-395.
26 RA MON G. CANTO et al.
38. Bell, 0. Y., Haseman, J. A., Spock, A., McLennan, G., and Hook, G. E. R., 1981, Plasma
proteins of the broncho-alveolar surface of the lungs of smokers and non-smokers, Am. Rev.
Respir. Dis. 124:72-79.
39. Lewis,J. F., and Jobe, A. H., 1993, Surfactant and the adult respiratory syndrome (state of the
art), Am. Rev. Respir. Dis. 147:218-233.
40. Coonrod, J. 0., Lester, R. L., and Hsu, L. C., 1984, Characterization of the extracellular
bactericidal factors of rat alveolar lining material,] Glin. Invest. 74:1269-1279.
41. Jonsson, S., Musher, 0. M., Goree, A., and Lawrence, E. C., 1986, Human alveolar lining
material and antibacterial defenses, Am. Rev. Respir. Dis. 133:136-140.
42. Sherman, M. P., D~mbola, J. B., Aeberhard, E. E., and Barrett, C. T., 1988, Surfactant
therapy of newborn rabbits impairs macrophage bactericidal activity,] Appl. Physiol. 56:
137-145.
43. Speer, C. P., Gotze, B., Curstedt, T., and Robertson, B., 1991, Phagocytic functions and
tumor necrosis factor secretion of human monocytes exposed to natural porcine surfactant
(Curosurf), Pediatr Res. 30:69-74.
44. Weber, H., Heilmann, P., Meyer, B., and Maier, K. L., 1990, Effect of canine surfactant
protein (SP-A) on the respiratory burst of phagocytic cells, FEBS Lett. 270:90-94.
45. van Iwaarden,J. F., Welmers, B., Verhoef,J., Haagsman, H.P., and van Golde, L. M. G., 1990,
Pulmonary surfactant protein A enhances the host-defense mechanism of rat alveolar
macrophages, Am.] Respir. Gell Mol. Biol. 2:91-98.
46. Hoffman, R. M., Claypool, W. 0., Katyal, S. L" Singh, G., Rogers, R. M., and Dauber,J. H.,
1987, Augmentation of rat alveolar macrophage migration by surfactant protein, Am. Rev.
Respir. Dis. 135:1358-1362.
47. Thomassen, M. J., Lee, F., Meeker, 0., Antal, J., Connors, M., and Wiedemann, H., 1992,
Synthetic surfactant (Exosurf) inhibits endotoxin-stimulated cytokine secretion by human
alveolar macrophages (abstract), Am. Rev. Respir. Dis. 145:A876.
48. Merrill, W. w., Naegel, G. P., Olchowski,J.J., and Reynolds, H. Y.,1985, Immunoglobulin G
subclass proteins in serum and lavage fluid of normal subjects: Quantitation and compari-
son with immunoglobulins A and E, Am. Rev. Respir. Dis. 131:584-591.
49. Reynolds, H. Y., Atkinson, J. P., Newball, H. H., and Frank, M. M., 1975, Receptors for
immunoglobulin and complement on human alveolar macrophages,] Immunol. 114:1813-
1819.
50. Sibille, Y., Chatelain, B., Staquet, P., Merrill, W. w., Delacroix, 0. L., and Vaerman,J. P.,1989,
Surface IgA and Fc-alpha receptors on human alveolar macrophages from normals and
patients with sarcoidosis, Am. Rev. Respir. Dis. 139:740~747.
51. Cole, F. S., Matthews, W. J., Rossing, T. H., Gash, D. J., Lichtenberg, N .A., and Pennington, J.
E., 1983, Complement biosynthesis by human bronchoalveolar macrophages, Glin. Immunol.
Immunopathol. 27:153-159.
52. Robertson, J., Caldwell, J. R., Castle, J. R., and Waldman, R. H., 1976, Evidence for the
presence of components of the alternate (properdin) pathway of complement activation in
respiratory secretions,] Immunol. 117:900-903.
53. Peters, M. G., Ambrus, J. L., Fauci, A. S., and Brown, E. J., 1988, The Bb fragment of
complement factor B acts as aBceIl growth factor,] Exp. Med. 168:1225-1235.
54. Sundsmo, J. S., and Gotze, 0., 1981, Human monocyte spreading induced by factor Bb of the
alternative pathway of complement activation,] Exp. Med. 154:763-777.
55. Gross, G. N., Rehn, S. R., and Pierce, A. K., 1978, The effect of complement depletion on
lung c1earance of bacteria,] Glin.lnvest. 62:373-378.
56. Toews, G. B., and Vial, W. C., 1984, The role of C5 in polymorphonuclear leukocyte
recruitment in response to Streptococcus pneumoniae, Am. Rev. Respir. Dis. 129:82-86.
57. Toews, G. B., and Pierce, A. K., 1984, The fifth component of complement is not required
for c1earance of Staphylococcus aureus, Am. Rev. Respir. Dis. 129:597-601.
58. Reynolds, H. Y., and Newball, H. H., 1974, Analysis of proteins and respiratory cells
obtained from human lungs by bronchiallavage,j Lab. Clin. Med. 84:559-573.
59. Dehring, D. j., and Wismar, B. L., 1987, Intravascular macrophages in pulmonary capillaries
in humans, Am. Rev. Respir. Dis. 139:1027-1029.
60. Warner, A. E., Molina, R. M., and Brain, j. D., 1987, Uptake of blood-borne bacteria by
pulmonary intravascular macrophages and consequent inflammatory responses in sheep,
Am. Rev. Respir. Dis. 136:683-690.
61. Brannen, A. L., and Chandler, D. B., 1988, Alveolar macrophage subpopulations responsive-
ness to chemotactic stimuli, Am. j Patlwl. 132:161-166.
62. Sibille, Y., and Reynolds, H. Y., 1990, Macrophages and polymorphonuclear neutrophils in
lung disease and injury (state of the art), Am. Rev. Respir. Dis. 141:471-501.
63. Gordon, S., 1980, Lysozyme and plasminogen activator: Constitutive and induced secretory
products of mononuclear phagocytes, in: Mononuclear Phagocytes Part Il (R. van Furth, ed.)
Nijhoff, The Hague, The Netherlands, pp. 1273-1294.
64. Hinman, L. M., Stevens, C. A., Matthay, R. A., and Gee, j. B. L., 1979, Angiotensin-
convertase activities in human alveolar macrophages: Effects of cigarette smoking and
sarcoidosis, Science 205:202-203.
65. Hunninghake, G. W, and Crystal, R. G., 1981, Pulmonary sarcoidosis: A disorder mediated
by excess helper T-Iymphocyte activity at sites of disease activity, N. Eng. j Med. 305:
429-434.
66. Reynolds, H. Y., Fulmer,j. D., Kazmierowski,j. A., Roberts, W C., Frank, M. M., and Crystal,
R. G., 1977, Analysis of cellular and protein components of bronchoalveolar lavage fluid
from patients with idiopathic pulmonary fibrosis and hypersensitivity pneumonitis, j Clin.
Invest. 59:165-175.
67. Wallace, j. M., Barbers, R. G., Oishi, j., and Prince, H., 1984, Cellular and T Iymphocyte
subpopulation profiles in bronchoalveolar lavage fluid from patients with acquired immuno-
deficiency syndrome and pneumonitis, Am. Rev. Respir. Dis. 130:786-790.
68. Young, K. R., Jr., Rankin, j. A., Naegel, G. 1'., Paul, E. S., and Reynolds, H. Y., 1985,
Bronchoalveolar lavage cells and proteins in patients with acquired immunodeficiency
syndrome-an immunologic analysis, Ann. Intern. Med. 103:522-533.
69. Rankin, j. A., Naegel, G. 1'., Schrader, C. E., Matthay, R. A., and Reynolds, H. Y., 1983,
Airspace immunoglobulin production and levels in bronchoalveolar lavage fluid of normals
and patients with sarcoidosis, Am. Rev. Respir. Dis. 127:442-448.
70. Reynolds, H. Y., 1985, Respiratory infections may reflect deficiencies in host defense
mechanisms, Disease-A-Month 31:1-98.
71. Reynolds, H. Y., 1988, Immunoglobulin G and its function in the human respiratory tract,
Mayo Clin. Proc. 63:161-174.
72. Reynolds, H. Y., 1986, Lung immunology and its contribution to the immunopathogenesis of
certain lung diseases,j Allergy Clin. Immunol. 78:833-847.
2
The Immunology of
Pneumococcal Pneumonia
CAROL A. KEMPER and STANLEY C. DERESINSKI
1. INTRODucnON
Streptococcus pneumoniae is a leading cause of community acquired pneumonia,

sinusitis, otitis media, and meningitis. The organism was first independently
isolated by Pasteur and by Sternberg in 1880, and its association with lobar
pneumonia was uncovered just a few years later.! There are an estimated
150,000-570,000 cases of pneumococcal pneumonia in the United States each
year and an estimated annual attack rate of 68-260 cases per 100,000 popula-
tion. 2 Closed communities may experience higher attack rates. The male to
female ratio is 3:2. The highest age-specific incidence of infection is seen in
infants younger than 2 years of age, but, in adults, the risk of infection increases
with advancing age. 3 Despite the availability of effective therapy, pneumonia
caused by this organism remains potentially life-threatening with an overall case
fatality rate of 5-7%. Both underlying disease, and youth or old age contribute
significantly to higher mortality rates, and in some patient groups (e.g., alco-
holics), the case fatality rate may be as high as 25%.
Oropharyngeal colonization by S. pneumoniae can be demonstrated, with the
use of careful bacteriologic techniques, in as many as 70% of individuals. Its
prevalence varies with a number of factors including age (adults are more
CAROL A. KEMPER and STANLEY C. DERESINSKI • Department of Medicine, Stanford

University School of Medicine, Stanford, California, 94305; Department of Medicine, Santa
Clara Valley Medical Center, San Jose, California, 95128; and AIDS Community Research
Consortium, Redwood City, California 94062.
Pulmonary Infictions and Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
29
30 CAROL A. KEMPER and STANLEY C. DERESINSKI
frequently colonized while children are more likely to become persistent carriers),
season (peak rates occur in the winter and the early spring in temperate climates),
and the presence of upper respiratory tract infection. An individual may become
simultaneously colonized with organisms of more than one capsular type. 4 Most
infections occur after recent colonization by the organism rather than after
prolonged carriage. 5 This is especially apparent in infants, 15% of whom develop
acute infection following acquisition of a new strain, usually within one month
of the initial colonization. 6 The factors that affect the progression from coloniza-
tion to invasive disease are the subject of this chapter.
The mechanisms by which the normal commensal existence of Streptococcus
pneumoniae in the upper respiratory tract of healthy individuals is disrupted and
the organism produces disease remains generally uncertain. 7.8 The m<tior compo-
nents of the antibacterial defense mechanisms of the respiratory tract include:
aerodynamic filtration, the epiglottic closure and cough reflexes, the mucociliary
transport system, professional phagocytes (polymorphonuclear leukocytes and
monocytes/macrophages), humoral and cellular immunity, and a variety of anti-
bacterial respiratory tract secretions. Pneumonia occurs as a consequence of
aspiration of organisms from the oropharynx which manage .to evade these
multiple overlapping lines of defense to reach the alveoli and multiply.
2. THE ORGANISM
Pneumococci, which are aerobic and facultatively anaerobic, have complex

metabolic requirements, primarily generating their energy requirements by lactic
acid fermentation. 9 Replication of the organism occurs within a narrow tempera-
ture range of 25° to 42°C, but optimally occurs at a temperature of 37°C and a
pH of 7.6. They lack cytochromes but utilize oxygen via a flavoenzyme system to
make H 2ü 2 • Since the organism does not produce catalase, an exogenous source
(e.g., erythrocytes) is required for the survival of the organism in culture media.
This latter fact may account for the absence of this strict parasite in the environ-
ment. lO
The cell envelope consists of a cell membrane and a cell wall, encompassed
by a polysaccharide capsule of varying mass. The capsule, present on all clinical
isolates of the organism, accounts for type specificity. Twenty-one of the 83
identified capsular serotypes (1, 3,4, 6A, 6B, 7F, 8, 9N, 9V, lOA, BA, 12F, 14, 15B,
17F, 18C, 19A, 19F, 20, 22F,and 23F) accountformore than 80% ofpneumococcal
bacteremias in adults in the United States.!l The degree of encapsulation depends
on the strain and the growth conditions. Immunologic cross-reactivity of some
of the pneumococcal capsular polysaccharides occurs with the capsular material
of other alpha hemolytic streptococci, the capsular polysaccharides of Klebsiella
and Salmonella spp., and with certain blood group substances. l2.13
Associated with the cell envelope is an amidase that functions as the organ-
ism's major autolytic enzyme. This enzyme, N-acetylmuramyl-L-alanine amidase,
is normally quiescent, but it is activated during ceIl-waIl morphogenesis and is
IMMUNOLOGY OF PNEUMOCOCCAL PNEUMONIA 31
responsible for the removal of the tetrapeptide from muramic acid in the cell wall
during organism replication and division. The choline moiety of the cell-wall
teichoic acid is essential for this reaction. Substitution of ethanolamine for choline
in the cell wall causes resistance to autolysis and impairs normal cellular division. 9
Activation of autolysis by surface-active agents forms the basis for the bile
solubility test used to distinguish S. pneumoniae from other alpha hemolytic
streptococci.
3. PNEUMOCOCCAL CELL COMPONENTS AND

THE PATHOGENESIS OF INFECTION
The predominant histologic feature of pneumococcal pneumonia is an

intense inflammatory response characterized by a polymorphonudear leukocyte
infiltration of alveolar spaces. Components of the organism's cell envelope are of
prime importance to the induction of this inflammatory response and to the
virulence of the organism.
3.1. Cell Wall

The cell wall, composed of interlinked peptidoglycan and covalendy bound
teichoic acid, is responsible for eliciting the intense inflammatory reaction to the
organism. A m'Üor component of the cell wall is C-polysaccharide, which is cova-
lently linked to the peptidoglycan and which can activate the alternate comple-
ment pathway.l4 An important component of C-polysaccharide is the phosphoryl-
choline residue of teichoic acid which is a potent inducer of antibody response 15
and is also critical to the autolytic capability of the organism. 9 ,l6 The antibodies
induced by this component, which are present in most healthy adults, do not
appear to protect against pneumococcal invasion in humans or in micep,18
The cell-wall components elicit production of IL-l, but not tumor necrosis
factor (TNF), by human monocytes in vitro. 19 The C-polysaccharide binds the
acute phase reactant, C reactive protein (hence its name).20 The role of pneumo-
coccal cell-wall components in eliciting an inflammatory response has been most
extensively studied in animal models of meningitis. In these models, they are
responsible for the rapid migration of polymorphonudear leukocytes into the
cerebrospinal fluid, as weIl as the production of IL-l, TNF, prostaglandins, and
the activation of the complement cascade. 21 ,22
Intrapulmonary replication is not necessary for the induction of the inflam-
matory response. Encapsulated strains rapidly multiply to high concentrations
and induce an intense inflammatory response (as determined from protein
concentration and leukocyte number in bronchoalveolar lavage fluid). Unencap-
sulated strains are deared rapidly from rabbit lungs, but they induce an inflam-
matory response similar to that of encapsulated strains. A similar degree of
inflammation is induced by heat-killed pneumococci. 23
Purified whole pneumococcal cell wall is a potent inducer of the inflamma-
tory response in rabbit lung. In one study, the peak influx ofleukocytes occurred
4 hr after intratracheal inoculation and the peak protein concentration was
reached at 24 hr. Various individual ceIl-waIl components are also capable of
eliciting an inflammatory response. The leukocyte and pro tein exudative re-
sponse to a mixture of ceIl-wall components, such as teichoie acid attached to the
glycan backbone of the pneumococcal cell wall and a variety of small-stem
peptides (products of penicillin-induced ceIl-wall hydrolysis), was similar to the
response to the intact cell wall. Soluble ceIl-wall components (e.g., a mixture of
purified soluble peptidoglycan and teichoic acid) induced leukocytosis but did
not cause a protein influx, indicating that the two effects are not necessarily
related. In addition, further degradation of insoluble products did not elicit an
inflammatory response. Capsular polysaccharide of type 2 and 3 strains elicited
an equivalent, albeit delayed, inflammatory response relative to that induced by
the cell wall. A minimal inflammatory response (10-15% greater than control
animals) resulted from the intratracheal instillation of either cell-pneumococcal
membrane or cytoplasmic contents. 23
In addition to the ability of purified pneumococcal cell wall to activate
complement, induce the secretion of IL-l from macrophages, and bring about
other proinflammatory effects, it also induces production of platelet activating
factor (PAF) in a rabbit model of pneumonia. 23a PAF in turn is known to increase
vascular permeability and to cause the local accumulation of platelets and leuko-
cytes. Note that PAF, like pneumococcallipoteichoic acid, contains phosphoryl-
choline and that like the latter, PAF binds to C-reactive protein. In contrast to the
neutralizing effect of lipoteichoic acid on C-reactive protein, however, the com-
plex with PAF enhances the platelet-aggregating activity of the latter. 23b
Purified pneumococcal cell wall is a potent inducer of procoagulant activity
of cultured human umbilical vein endothelial ceIls.23c This may, in part, account
for local small blood vessel occlusion that may be seen in the lung, as weIl as
instances of dis semina ted intravascular coagulation in some patients.
Cell wall-associated autolysin (N-acetylmuramyl-L-alanine amidase) may
significantly contribute to pneumococcal virulence, most likely by mediating the
release of cytoplasmic proteins, pneumolysin, and neuraminidase, during cell
lysis. 7,8,24,25 Immunization of mice with autolysin provided a modest but signifi-
cant protection against intraperitoneal challenge with S. pneumoniae, but failed to
provide protection in mice subsequently challenged with a genetically modified
mutant strain of pneumococcus unable to express active pneumolysin. 25
3.2. Plasma Membrane

The plasma membrane envelops the cytoplasm and is composed of li po-
teichoic acid, lipid, and various proteins. It also includes the Forssman antigen, a
lipoteichoic acid with the lipid portion embedded in the cell membrane. Forss-
man antigen is a potent inhibitor of pneumococcal autolysin and induces an
inflammatory response in rabbit lung. This substance, however, is effective only in
vitro and it is not believed to be a major chemotactic agent during infection. 23
3.3. Capsule
Capsular polysaccharide is of great importance both with regard to the

virulence of the organism and the induction of antibody against it. As indicated
above, unencapsulated strains are cleared rapidly from rabbit lungs while encap-
sulated strains multiply to high concentrations. 23 Transposon Tn916 mutagenesis
of type 3 S. pneumoniae results in astrain phenotypically identical to the parent
strain except that it is incapable of producing capsular polysaccharide. This
mutation virtually abrogates virulence in mice (change in LD 50 from 1 CFU to
> 5 X 107 CFU).26
In contrast to the ability of the pneumococcal cell wall to elicit an inßamma-
tory response, the capsular polysaccharide is relatively inert in this regard. It
does, however, inhibit opsonization and pulmonary clearance of pneumococci. 27
Capsular polysaccharide from types 2 and 3 S. pneumoniae, each at a concentra-
tion of 1.0 mcg/mL, absorb specific immunoglobulin in vitro and thus impair
opsonization by polymorphonuclear leukocytes. Preincubation of polymorpho-
nuclear leukocytes with type 3 capsular polysaccharide at a concentration of 10.0
mcg/ml im pairs the bactericidal activity against the homologous organism. 28
The amount of capsular material is important in determining the ability of
the organism to invade, but it is not the sole determinant. Despite similar amounts
of capsular material, type 37 (the capsule of which is composed predominantly
of a glucose homopolymer) is relatively avirulent while type 3, with a capsule
composed of a polymer of glucose and glucuronic acid (repeated units of
D-glucuronic acid beta 1~4 linked to n-glucose joined by beta 1~3 glucosidic
bonds) is highly invasive. 29 Thus the invasiveness of organism appears to be more
dependent on the relative composition of the capsule than the amount of capsular
material,3o Other factors, including immunogenicity,31 alternative complement
pathway activation,32 and resistance to phagocytosis33 depend to varied degrees
on the capsular type. Most pneumococcal polysaccharides possess acidic groups
and are negatively charged. 29 Their resultant polar hydrophilic character may
interfere with cell-cell interaction and, consequently, with phagocytosis.
3.4. Cytoplasmic Factors
Pneumolysin, genetically related to its ancestral cousin, streptolysin, appears

to be an important pneumococcal virulence factor. 7,8,34,35 This thiol-activated
cytolytic toxin of approximately 53 kDa, which is released during bacteriallysis,
binds to cholesterol in the eukaryotic membrane and forms trans-membrane
channels, leading to host celllysis. 36 The channels induced by pneumolysin in
planar lipid bilayers have been functionally characterized. 35a Low concentrations
of pneumolysin inhibit the function of professional phagocytes (polymorpho-
nuclear leukocytes and monocyte-derived macrophages) including chemotaxis,
respiratory burst, and bactericidal activity37,38 and inhibit antibody synthesis by B
cells. 34 Pneumolysin-specific antibodies increase during infection, indicating in
vivo production of this protein.
Pneumolysin activates the classic complement cascade in the absence of

toxin-specific antibodies.34 Pneumolysin also causes slowing of human ciliary
beating in vitro at concentrations as low as 5 ng/mI.39 Organ cultures of human
turbinate tissue indicate that sufficient pneumolysin is produced du ring bacterial
colonization to slow ciliary beating. 40 At higher concentrations in vitro, it is
cytotoxic to nasopharyngeal epithelium and pulmonary endothelial cells. 4o,4l
Recombinant pneumolysin injected directly into rat lung induced an inftamma-
tory histologie picture identical to that associated with pneumococcal pneumonia,
whereas heat-inactivated pneumolysin did not. 42 Pneumolysin is cytotoxic to
bovine pulmonary endothelial cells in vitro and may thus contribute to the alveolar
hemorrhage commonly seen in pneumococcal pneumonia. 43a
Point mutations that resulted in modified toxin mitigated both its ability to
activate complement and the degree of histologie inftammation. Furthermore,
pneumolysin-deficient mutants of S. pneumoniae are significantly less virulent
in mice after intra nasal or intraperitoneal challenge as determined by clearance
of organism from blood and by survival time. Immunization of mice with native
and recombinant pneumolysin conferred limited protection against intranasal
challenge with virulent pneumococci in a serotype-independent fashion. 25 ,34
The role of pneumococcal neuraminidase, which removes terminal sialic
acid residues, as a virulence factor is uncertain. Immunization of mice with
formaldehyde-pretreated neuraminidase yielded a small but significant increase
in survival time (compared to untreated controls), but the median survival of mice
who had received both neuraminidase and pneumolysin was not significantly
different from those that had received pneumolysin alone. 43
4. IMMUNE DEFENSE
Antibody response, activation of the complement cascade, opsonization and

phagocytosis, and the inftammatory response as a consequence of infection with
S. pneumoniae are presented in this section. The role of mechanical defense
barriers, such as epiglottic closure, cough defenses, and ciliary action are not
discussed here.
4.1. Mucosal Barriers and Antibody Responses

Respiratory secretions contain lysozyme capable of killing S. pneumoniae by
causing breakdown of the cell wall. While triggering of pneumococcal autolysin
may contribute to this effect, it is not essential. 44
S. pneumoniae appears to adhere to human nasopharyngeal epithelial cells
via receptors, although the extent and specificity ofthis adherence is uncertain. 45
The adherence of the organism to nasopharyngeal epithelial cells is impaired by
preincubation of the bacteria with nasopharyngeal secretions. Inhibition of
adherence is associated with the presence of secretory IgA, but not with IgG.lo,46
S. pneumoniae produces an IgA protease specific for human IgA l that cleaves a
prolyl-threonyl peptide bond at position 227-228. To the extent that secretory

IgA plays a role in preventing invasion by the organism, this protease may be an
important virulence factor. 47- 50
Organ cultures of human ciliated nasal turbinate tissue indicate, however,
that the generation of a thick gelatinous mucosal layer, in the presence of
compromised ciliary function (largely caused by pneumolysin), may be more
important for bacterial colonization than is specific cell adherence. 40
Pneumococcal capsular polysaccharides are T-cell independent antigens that
elicit IgG 2 as the predominant subclass of protective antibody.29 The LD 50 of
S. pneumoniae in T lymphocyte-deficient, pneumococcus-naive nude mice is
identical to that of control (+/+) mice. 51 Capsular type-specific IgG antibodies
will not eliminate organism carriage when acquisition has predated the develop-
ment of the serological response, but these antibodies tend to prevent further
acquisition of homologous strains. 4 This trait may explain the greater risk of
invasive disease with a new serotype of S. pneumoniae compared to that associated
with prolonged carriage. Antibodies to C polysaccharide do not, however, appear
to confer protective immunity against intraperitoneal inoculation of pneu-
mococci in mice.I 8
Antibodies to pneumococcal surface protein A, a 43-kDa surface protein,
afford a degree of protection in mice, whereas those directed against the Forss-
man antigen are not protective in a murine model. The epitopes of the pneumo-
coccal surface protein A reside in the N-terminal half of the molecule. 52- 54
Suppressor T cells mayaiso playa role in the host response to this organism. 55
Increased IL-2 receptor expression has been demonstrated on both CD8+ lym-
phocytes and on other peripheral blood lymphocytes after the administration of
pneumococcal polysaccharide vaccine in humans; this indirectly suggests that
suppressor T cells are activated in response to vaccine administration. 56
4.2. Opsonization and the Complement Cascades

Pneumococci are able to activate both the classical and alternative comple-
ment pathways; either pathway results in the binding of C3b fragments to the
organism. 8 ,57 Efficient killing of S. pneumoniae by professional phagocytes re-
quires opsonization both by capsular type-specific antibody (IgG and IgM) and
by complement.5 8 This is true even though the pneumococcus, like other gram-
positive bacteria, is resistant to the bactericidal and lytic activities of complement.
Complement is, however, critical to the opsonization of the organism, making it
palatable to professional phagocytes. Although activation of the classical comple-
ment pathway by antibody represents the major host defense against the organ-
ism, the classic and alternative pathways playa cooperative role in this process and
opsonization is likely to take place more rapidly and extensively when both
pathways are activated.
Activation of the classical pathway primarily requires the presence of pneu-
mococci encoated by anticapsular antibodies of either the IgG or IgM subclass.
IgM appears to be more potent than IgG. Despite the demonstrated ability of
rabbit type-specific antibody to activate the classic complement pathway, human

convalescent sera has been reported to be ineffective in this regard. 59 This
observation may be related to the relative insensitivity of the assay conditions
utilized in the studies utilizing human sera. Evidence suggests, however, that the
classic pathway may be activated by other mechanisms. The acute-phase reactant,
C-reactive protein, provides protection in nonimmune mice challenged with S.
pneumoniae. 60 C-reactive protein forms complexes with pneumococcal C polysac-
charide, apart of the Forssman antigen, in the presence of Ca2+. This complex
can activate the classic complement pathway in the absence of antibody in a
manner similar to that of immune complex activation. 61 ,62 The synthesis of
acute-phase reactants, such as C-reactive protein, by the liver is induced primarily
by interleukin-6, serum concentrations of which have been demonstrated to rise
markedly in pigs injected intravenously with type 6b S. pneumoniae. 63
Pneumolysin (and streptolysin) can also activate the classic complement
pathway in the absence of toxin-specific antibodies. 34 This occurs as a result of
antibody binding (via Fc) to a region of either of these toxins homologous to
C-reactive protein. Indeed, two noncontiguous segments ofthe pneumolysin pro-
tein share limited sequence homology with the C-terminal region of C-reactive
protein. 64 By activating the complement cascade in the fluid layer, the organism
may effectively be diverting complement away from itself, mitigating complement-
mediated phagocytosis. This finding may explain the observation noted above
that immunization of mice with C-reactive protein provides partial protection
against infection caused by S. pneumoniae60 ; immunization with pneumolysin is
similarly protective. 25 ,43
In contrast, the alternative complement pathway may be activated by the cell
wall of the organism in the absence of type-specific antibody.65 Cell wall constitu-
ents, in particular teichoic acid, are potent activators of the alternative path-
way,66,67 but capsular polysaccharide is generally unable to activate the alternative
pathway and the cell membrane does so only at very high concentrations. 68
Bronchopulmonary lavage fluid contains IgG, IgA, IgA secretory component,
and elements of the alternative complement pathway.68-71 Thus, in the non-
immune host, clearance of organisms from the lung may be media ted by direct
activation of the alternative complement pathway70 or by triggering of the
alternative complement pathway as a consequence of nonimmune absorption of
IgA onto the surface of the organism. 72
Either the classic or alternative cascades result in generation of complement
and the binding of C3 fragments to the organism.5 7 Complement, probably
generated via the alternative pathway, plays a key role in nonimmune clearance.5 8
After activation of C3 via the alternative pathway, C3b is fixed to the cell wall,
beneath the capsule of the organism. In this position, neither the fixed IgG Fc
fragment nor C3b are accessible to receptors on professional phagocytes, because
of the intervening capsule. 62 ,68,73 Fixation at this site may make it relatively
inaccessible to antibody and phagocytes.
Bacteria reaching the alveoli are initially confronted by tissue macrophages
and, subsequently, by polymorphonuclear leukocytes. Studies in mice indicate
that C5 plays an important role in protection against mortality, ar least after

intraperitoneal injection. The mechanism of protection may be related to chemo-
tactic generation and migration of polymorphonudear leukocytes to infected
alveoli. 74 For example, C5a, which is generated by activation of the complement
pathway, is a potent chemotactic factor. In nonimmune animal models, C5a attracts
polymorphonudear leukocytes to the alveoli where phagocytosis of bacteria occurs
in 12-36 hours, presumably in the absence of organism-specific antibody.75
Decomplementation of mice with cobra venom factor prior to inhalation al
challenge with type 3 S. pneumoniae led to more severe infection than in normo-
complementemic controls, and no dearance of either IgG- or IgM-coated bacte-
ria was observed. 76 In rabbits receiving intratracheal instillation of pneumococcal
cell wall components, pretreatment with complement-depleting cobra venom
factor or with methylprednisolone only minimally decreased the inflammatory
response at 24 hr (60 and 73% ofthe control response, respectively). These results
in complement-depleted animals suggest that chemotaxins other than comple-
ment may be relatively unimportant in the early phase of recruitment of polymor-
phonudear leukocytes to the alveoli in response to bacterial challenge.
On the other hand, limited evidence suggests that leukotrienes mayaiso play
an important chemotactic role in pneumococcal pneumonia. Pretreatment with
CGP 6258, a specific inhibitor of cydooxygenase (which results in the production
of prostaglandins), increased the leukocyte concentration in lavage fluid 427%
relative to control at 24 hr, whereas didofenac sodium had minimal effect,
reducing the count to 77% of control. Indomethacin, by contrast, reduced the
leukocyte count to 29% of control. 23 Thus, substances that interfere with the
lipoxygenase pathway and the generation of leukotrienes may negatively impact
leukocyte chemotaxis, and substances that preferentially enhance leukotriene
formation may result in increased chemotaxis.
Failure to kill the organism within the confines of the respiratory tract and
pulmonary parenchyma may allow it to reach the bloodstream. Once bacteremia
occurs, dearance depends on the same components of the immune system,
although the role of polymorphonudear leukocytes may be diminished. Neutro-
penia does not appear to be an important risk factor for pneumococcal disease. 77
5. DISEASE STATES PREDISPOSING TO SEVERE

PNEUMOCOCCAL INFECI'ION
Debilitating causes of all sorts render individuals more susceptible. Alcoholism is

perhaps the most potent predisposing factor. Robust, healthy men are, however,
often attacked. And, Probably for each one of us it is a batde between the degree of
resistance and the virulence of the organism which we harbor. A catarrh of the
upper passages, exposure, alcoholism, etc., weaken the defenses, and give the ever-
present enemy a chance, either for a frontal attack in the lungs, in an acute
pneumonia, or to make a Banking assault on some unprotected region, causing a
peritonitis, otitis, sinusitis, etc.
William Osler78
Prior to the AIDs epidemic, conditions such as those cited by Osler were
present in 33-80% of patients hospitalized with pneumococcal infections as
weIl as in a majority of the fatalities. ll A number of disease and immune-deficient
states that predispose individuals to pneumococcal infection are discussed in this
section, including complement deficiency, hypogammaglobulinemia, neutro-
penia, alcoholism and chronic liver disease, asplenia, sickle cell disease and the
acquired immune deficiency syndrome (AIDS).
5.1. Complement Deficiency

Patients with C3 deficiency, either as a consequence of congenital absence or
increased consumption of C3 because of C3b/C4b inactivator deficiency, are at
increased risk of pneumococcal infection. 57 Congenital C2 deficiency, the most
common of the inherited complement deficiencies, also predisposes to pneumo-
coccal bacteremia. 79 One patient who had a combined C2 deficiency and low
serum concentration of IgG 2 had recurrent infections with encapsulated organ-
isms including S. pneumoniae. Partial deficiency of factor B may playa role in some
patients. 80
5.2. H ypogammaglobulinemia
S. pneumoniae and other encapsulated bacteria are the most frequent bacterial
pathogens in patients who have hypogammaglobulinemia,sl Consistent with evi-
dence of its importance in the response to pneumococcal infection, IgG 2 defi-
ciency appears to increase the risk of pneumococcal infection. In most reported
cases of IgG2 deficiency, however, other antibody deficiencies also were present.
Interestingly, preexisting IgG2 deficiency does not preclude an IgG2 response to
administration of pneumococcal polysaccharide vaccine. 82 Patients with multiple
myeloma and chronic lym phocytic leukemia are also at increased risk of pneumo-
coccal infection, primarily as the result of hypogammaglobulinemia. 83
5.3. Alcoholism and Liver Disease

Alcohol im pairs the function ofalveolar macrophages,84 inhibits the migra-
tion of polymorphonuclear leukocytes into the lung,85 and im pairs ciliary func-
tion,s6 Total IgG and IgG 1 concentrations in bronchoalveolar lavage fluid were
decreased (relative to controls) in 70% of patients with alcoholic liver disease,s7
Leukopenia is common in alcoholic patients with pneumococcal sepsis and is
associated with high mortality.88 Ethanol can cause bone-marrow suppression and
there is evidence that it inhibits T-Iymphocyte production of colony-stimulating
factor. 89,90
Feeding Sprague-Dawley rats ethanol for 7 days prior to and 10 days after
intratracheal challenge with type 3 S. pneumoniae led to an increased severity of
infection relative to that of nonintoxicated controls. 91 The ethanol-fed animals
showed impaired pulmonary clearance of the organism (lOO-fold less than con-
trols) assoeiated with a greater frequency ofbacteremia, capsular polysaccharide
antigenemia, impaired clearance of organism from the bloodstream, and a
higher mortality rate. This occurred despite histologic evidence of an intense
polymorphonuclear leukocyte infiltration in the lungs of intoxicated rats. Further
studies using this model demonstrate that neutrophil chemotaxis was severely
impaired in vitro, but recruitment to infected rat lungs was not affected. 92 Serum
complement levels were not affected by ethanol administration, a finding which is
also true in ethanol-intoxicated human alcoholics without liver disease. 93 .94
Patients with eirrhosis appear predisposed to severe pneumococcal disease.
Among the general immune defieits in patients with eirrhosis are impaired
function of polymorphonuclear leukocytes and monocytes (depressed chemo-
taxis, phagocytosis, and intracellular killing) and reduced serum opsonic activity
and complement levels. 95 ,96 Induction of eirrhosis in rats by intragastric adminis-
tration of CCL4 led to a much-reduced LD 50 in response to intratracheal
challenge with type 3 S. pneumoniae, particularly those rats that had developed
ascites. While serum total hemolytic complement remained normal in eirrhotic
rats without ascites, those with ascites had a value that was 37% of controls. While
the defects were worse in the hypocomplementemic animals, the defieits in the
normocomplementemic cirrhotic rats without ascites suggest a different defect in
the response to S. pneumoniae, such as impaired polymorphonuclear leukocyte
function. 97
Human eirrhotics are known to have impaired function of their reticulo-
endothelial system, a system that is important in the clearance of S. pneumoniae
from the bloodstream. 98 The fixed tissue macrophages ofthe liver are important
in the reticuloendothelial system with regard to removal of S. pneumoniae from the
bloodstream. 99 Hepatic clearance of pneumococei from blood in the guinea pig is
dependent on adequate opsonization of the organism with C3b. 74
5.4. Viral Infections

Pneumococcal pneumonia is a recognized complication of influenza virus
infection. Infection of mice with influenza A virus enhances adherence of pneu-
mococei to tracheal epithelial cells.l°o Simultaneous infection with respiratory
syncytial virus and S. pneumoniae may occur.l° 1,102 An epidemiological study
suggests that parainfluenzae type 3 viral infection markedly predisposed chim-
panzees to invasive pneumococcal infection. 103
5.5. Postsplenectomy
The risk of pneumococcal bacteremia in splenectomized adults is reported to
be 0.92-2.1 cases/lOOO population/year.l°4-106 Pneumococcal infection in the
asplenic individual is often overwhelming and fatal. Aerosol challenge with S.
pneumoniae in splenectomized CD-I mice was examined in one study.107 Relative
to sham-operated controls, asplenic mice demonstrated impaired pulmonary

clearance and increased translocation to tracheobronchiallymph nodes. Partial
splenectomy led to intermediate results with evidence that pulmonary clearance
and animal survival were directly related to the amount of preserved splenic
tissue. 107 These data are consistent with the spleen having a systemic effect on
clearance of encapsulated organisms.
5.6. Sickle Cell Disease

The risk of pneumococcal bacteremia in children with sickle ceIl disease is
0.86 cases/1000 population/year;I08 approximately 300-500 times that of chil-
dren without sickle cell disease. Patients with this hemoglobinopathy have func-
tional and, eventuaIly, virtual anatomie asplenia. 109 In addition to splenie dysfune-
tion, some patients with sickle cell anemia have depressed serum-opsonic activity
mediated by either the classic or the alternative pathway or, in some cases, both. HO
This combination of deficits would appear to put such patients at a severe
disadvantage since studies in a murine model demonstrate synergism between
the spleen and serum complement in the clearance of pneumoeocci from the
bloodstream after intravenous challenge. lll
5.7. Acquired Immunodeficiency Syndrome

A retrospective review of the microbiology laboratory records of ten San
Francisco hospitals found a rate of pneumococcal bacteremia among HIV in-
fected adults (20-55 years old) of 9.4/1000 patient years.I 12 Only 43% of these
patients had AIDS at the time they developed pneumocoeeal bacteremia. This
rate is approximately 100 times that reported in San Francisco prior to the AIDS
epidemie. There was no difference in serotype prevalence compared to that of
cases reported for the general population. Mortality in AIDS patients with
pneumococcal pneumonia is reportedly higher (57%) than in any other group of
individuals, largely because of the oecurrence of bacteremia and septic shoek
early in the course of disease.I 13
B-cell dysfunction oeeurs early in HIV infection and worsens with progres-
sion to AIDS.I 14 Polyclonal hypergammaglobulinemia is commonly present. In
vitro studies indicate redueed proliferative responses to phytohemagglutin,
pokeweed mitogen (a T-eell-dependent B-cell mitogen), and to Staphylococcus
aureus Cowan strain 1 (a B-cell mitogen). B-ceIl produetion of immunoglobulin in
response to pokeweed mitogen is diminished whereas spontaneous production of
antibody is higher than in controls.I 15,1l6 Despite the usual finding of hyper-
gammaglobulinemia, selective IgG 2 deficiency may oecur, albeit rarely. Patients
with such a deficieney are reported to be at further increased risk of pneumococ-
cal disease.I 17 HIV-infected individuals with AIDS have reduced antibody re-
sponses to pneumococcal vaccine relative to those with asymptomatie lentivirus
infeetion. Zidovudine, administered for at least 4 weeks, improves the response to
vaecination. 118
Patients with HIV infection have elevated complement fragment to parent

ratios (C4d/C4, Ba/factor B, and C3d/C3), indicating classic complement path-
way activation, with greater activation paralleling the progression of HIV infec-
tion. 1l9
5.8. Malignancies
Some patients with neoplastic disease may develop severe pneumococcal
infection. 120 Patients with multiple myeloma and chronic lymphocytic leukemia
are specifically predisposed to infections with encapsulated organisms. Prior to
1973 the pneumococcus was the single most frequent pathogen, accounting for
27.6% of all infections, in patients with multiple myeloma and chronic lympho-
cytic leukemia. The predominant immunologie defect accounting for this obser-
vation is the presence of hypogammaglobulinemia. Additional defects have,
however, been described. These include impaired granulocyte chemotaxis, mi-
gration and adherence, impaired opsonization, and defects in C3 activation and
deposition. 121 The epidemiology of infection in patients with multiple myeloma
has, however, changed. The prolongation of life, increased hospitalization, and
chemotherapy have led to an increase in the importance of gram-negative enteric
organisms as pathogens late in the course of myeloma.t 22
5.9. Other Conditions Which Predispose to Pneumococcal Infection

S. pneumoniae is a frequent cause of pneumonia in the elderly. The elderly
have impaired antibody response to foreign antigens, including pneumococcal
polysaccharide as well as evidence of impaired polymorphonuclear leukocyte
chemotaxis and phagocytosis. 123
Patients with chronic cardiac disease, particularly congestive heart failure,
are thought to be at increased risk of pneumococcal pneumonia. The presence of
increased intraalveolar fluid had been thought to playa role. Of interest, however,
is that digoxin, commonly used in the treatment of congestive heart failure,
impairs mobilization of polymorphonuclear leukocytes in CD-l mice. Treatment
with digoxin prior to intratracheal or intravenous challenge with type 3 S.
pneumoniae led to decreased numbers of granulocytes in peripheral blood rela-
tive to pneumococcal-infected controls not receiving digoxin. Chemotaxis was not
impaired. 124
Pneumococcal infection occurred in five of 129 heart transplant recipients
between March 1985 and December 1987 in the Utah Affiliated Hospitals
Cardiac Transplant Program.t 25
The increased riskofinfection, particularly peritonitis and sepsis, caused by
S. pneumoniae and Eseheriehia eoli in patients with idiopathic nephrotic syndrome
appears to be related to impaired opsonic activity as a result of low concentra-
tion of factor B. 57 ,126 The relative risk of infection roughly corresponds to periods
of heavy proteinuria, suggesting that factor B, in conjunction with albumin, is
excreted in the urine.
6. CONCLUSION
A combination of capsular, cell wall, and cytoplasmic virulence factors

appears to account for the ability of S. pneumoniae to cause invasive, potentially
lethai disease, even in hosts without apparent immunocompromise. The polysac-
charide capsule of S. pneumoniae is an impediment to phagocytosis of the organ-
ism and is relatively inert in its contribution to the elicitation of an effective
inflammatory response. The capsule does, however, elicit type-specific protective
antibody formation. Components of the complex cell wall, on the other hand, are
potent provocateurs of an intense inflammatory response, including the elicita-
tion of cytokines, triggering of the complement cascade, and chemotactic re-
sponses with resultant influx of polymorphonuclear leukocytes. This inflamma-
tory response, while critical to the control of the organism, may contribute
significantly to injury to host tissues resulting from this infection. Lysis of the
organism as a result of the immune response to the S. pneumoniae can cause the
release of additional cell wall components, leading to further inflammation.
Furthermore, lysis of the organism leads to release, from cytoplasm, of pneumo-
lysin, which im pairs organism clearance by virtue of its adverse effect on ciliary
function of respiratory tract epithelial cells and its inhibition of phagocytic
function. Pneumolysin is, in addition, capable of directly lysing host cells.
A successful host response to S. pneumoniae involves mechanical barriers such
as an intact cough reflex and effective ciliary clearance, local secretory factors
such as lysozyme and secretory IgA, effective opsonization by type specific
antibody and complement and, finally, phagocytosis and killing by professional
phagocytes. The eventual outcome of infection with S. pneumoniae depends upon
the complex interplay of its diversity of virulence mechanisms and the multi-
layered mosaic of the host response.
AcKNOWLEDGMENT. We are indebted to Barbara Houghton, M.L.S., Medical

Staff Librarian, Sequoia District Hospital, Redwood City, CA 94062.
REFERENCES
1. Austrian R., 1960, The Gram stain and the etiology of pneumonia: An historical note,
Bacteriol. Rev. 24:261-265.
2. 1984, Pneumococcal polysaccharide vaccine, Morb. Mortal. Wkly. Rep. 33:273-276, 281.
3. Burnman, L. A., Norrby, R, and Trollfors, B., 1985, Invasive pneumococcal infections:
Incidence, predisposing factors, and prognosis, Rev. Infect. Dis. 7(2):133-142.
4. Austrian, R., 1986, Some aspects ofthe pneumococcal carrier state,J Antimicrob. Chemother.
18(Suppl A):35-45.
5. Gray, B. M. and Dillon, H. C.,Jr., 1989, Natural history ofpneumococcal infections, Pediatr.
Infect. Dis. J 8(Suppll):S23-S25.
6. Gray, B. M., Converse, G. M., III, and Dillon, H. C., Jr., 1980, Epidemiologie studies of
Streptococcus pneumoniae in infants: Acquisition, carriage, and infection during the first 24
months of life, J Infect. Dis. 142:923-933.
7. Boulnois, G.]., 1992, Pneumococcal proteins and the pathogenesis of disease caused by
Streptococcus pneumoniae,J Gen. Microbiol. 138:249-259.
8. Bruyn, G. A. w., Zegers, B. ]. M., and van Furth, R., 1992, Mechanisms of host defense
against infection with Streptococcus pneumoniae, Glin. Infoct. Dis. 14:251-262.
9. Austrian, R, 1980, Pneumococci. in: Microbiology, Third Edition, (B. D. Davis, R. Dul-
becco, H. N. Eisen, and H. S. Ginsberg, eds.) Harper and Row, Philadelphia. pp. 596-606.
10. Johnston, R. B., Jr., 1991, Pathogenesis of pneumococcal pneumonia, Rev. Infect. Dis.
13(SuppI6):S509-S517.
11. Mufson, M. A., 1981, Pneumococcal infections,JAMA 246:1942-1948.
12. Ginsburg, 1., 1981, Streptococcus, in: Medical Microbiology and Infoctious Diseases, (A. L.
Braude, C. E. Davis, and]. Fierer, eds.) W. B. Saunders, Philadelphia, pp. 281-295.
13. Lee, C.]., Koizumi, K., Henrichsen,]., Perch, B., Lin, C. S., and Egan, w., 1984, Capsular
polysaccharides of nongroupable streptococci that cross-react with pneumococcal group
19, J Immunol. 133(5):2706-2711.
14. Fischer, H. and Tomasz, A., 1985, Peptidoglycan crosslinking and teichoic acid attachment
in Streptococcus pneumoniae,J Bacteriol. 163:46-54.
15. Leon, M. A. and Young, N. M., 1971, Specificity for phosphorylcholine of six murine
myeloma proteins reactive with pneumococcus C polysaccharide and beta-lipoprotein,
Biochemistry 10:1424-1429.
16. Tomasz, A., 1981, Surface components of Streptococcus pneumoniae, Rev. Infoct. Dis. 3:
190-211.
17. Musher, D. M., Watson, D. A., and Baughn, R E., 1990, Does naturally acquired IgG
antibody to cell wall polysaccharide protect human subjects against pneumococcal infec-
tion? J Infoct. Dis. 161:736-740.
18. Chen Szu, S., Schneerson, R, and Robbins, ]. B., 1986, Rabbit antibodies to the cell wall
polysaccharide of Streptococcus pneumoniae to fail to protect mice from lethaI challenge with
encapsulated pneumococci, Infect. Immun. 54(2):448-455.
19. Riesenfeld-Orn, 1., Wolpe, S., Garcia-Bustos,]. F., Hoffmann, M. K., and Tuomanen, E.,
1989, Production of interleukin-l but not tumor necrosis factor by human monocytes
stimulated with pneumococcal cell surface components, Infoct. Immun. 57:189-193.
20. Gotschlich, E. C. and Liu, T. Y., 1967, Structural and immunological studies on the
pneumococcal C polysaccharide, J Biol. Ghern. 242:463-470.
21. Riesenfeld-Orn, 1., Tuomanen, E., Tomasz, A., Hengstler, B., and Zal, 0., 1985, The
relative role of bacterial cell wall and capsule in the induction of inflammation in pneu mo-
coccal meningitis, J Infect. Dis. 151:535-540.
22. Tuomanen, E., Liu, H., Hengstler, B., Zak, 0., and Tomasz, A., 1985, The induction of
meningeal inflammation by components of the pneumococcal cell wall, J Infect. Dis.
151:859-868.
23. Tuomanen, E., Rich, R., and Zak, 0., 1987, Induction of pulmonary inflammation by
components of the pneumococcal cell surface, Am. Rev. Respir. Dis. 135:869-874.
23a.Cabellos, C., Madntyre, D. E., Forrest, M., Burroughs, M., Prasad, S., and Tuomanen, E.,
1992, Differing roles for platelet-activating factor du ring inflammation of the lung and
subarachnoid space. The special case of Streptococcus pneumoniae, J Glin. Invest. 90:
612-618.
23b.Kohayakawa, M. and Issoue, K., 1986, Augmentation of 1-0-alkyl-2-0-acetyl-sn-glycero-3-
phosphocholine (PAF)-induced human platelet activation by C-reactive protein, Thromb.
Res. 41:649-657.
23c.Geelen, S., Bhattacharya, C., and Tuomanen, E., 1992, Induction of procoagulant activity
on human endothelial cells by Streptococcus pneumoniae, Infect. Immun. 60:4179-4183.
24. Berry, A. M., Lock, R A., Hansman, D., and Paton,]. C., 1989, Contribution of autolysin to
virulence of Streptococcus pneumoniae, Infoct. Immun. 57:2037-2042.
25. Lock, RA., Hansman, D., and Paton,]. C., 1992, Comparative efficacy of autolysin and
pneumolysin as immunogens protecting mice against infection by Streptococcus pneumoniae,
Microbial Patlwgen. 12:137-143.
26. Watson, D. A. and Musher, D. M., Interruption of capsule production in S. pneumoniae
serotype 3 by insertion of transposon Tn916, Inftct. Immun. 58:3135-3138.
27. Ansfield, M.]., Woods, n E., and Johanson, W G., Jr., 1977, Lung bacterial clearance in
murine pneumococcal pneumonia, Inftct. Immun. 17:195-204.
28. Esposito, A. L. and Clark, C. A., 1990, The effects of capsular polysaccharide on the
capacity of serum to support the killing of type 3 Streptococcus pneumoniae by human
neutrophils, Apmis 98:812-822.
29. Lee, C. ]., Banks, S. n, and Li,]. P., 1991, Virulence, immunity, and vaccine related to
Streptococcus pneumoniae, Crit. Rev. Microbiol. 18:89-114.
30. Knecht, ]. C., Schiffman, G., and Austrian, R, 1970, Some biological properties of
pneumococcus type 37 and the chemistry of its capsular polysaccharide, J Exp. Med.
132:475-487.
31. Heffron, R, 1979, in: Pneumonia, witk special reftrence to pneumococcus lobar pneumonia,
Harvard University Press, Cambridge, Massachusetts.
32. Fine, D. P., 1975, Pneumococcal type-associated variability in alternate complement path-
way activation, Inftct. Immun. 12:772-778.
33. Giebink, G. S., Verhoef,]., Peterson, P. K., and Quie, P. G., 1977, Opsonic requirements for
phagocytosis of Streptococcus pneumoniae types VI, XVIII, XXIII, and XXv, Infect. Immun.
18:291-297.
34. Boulnois, G. ]., Paton,]. C., Mitchell, T.]., and Andrew, P. w., 1991, Structure and func-
tion of pneumolysin, the multifunctional, thiol-activated toxin of Streptococcus pneu-
moniae, Mol. Microbiol. 5(11):2611-2616.
35. Walker, ]. A., Allen, R. L., Flamagne, P., Johnson, M. K., and Boulnois, G. ]., 1987,
Molecular cloning, characterization, and complete nucleotide sequence of the gene for
pneumolysin, the sulfhydryl-activated toxin of Streptococcus pneumoniae, Infect. Immun.
55(5):1l84-1189.
35a.Korchev, Y. E., Bashford, C. L., and Pasternak, C. A., 1992, Differential sensitivity of
pneumolysin-induced channels to gating by divalent cations, J Membrane Biol. 127:
195-203.
36. Saunders, F. K., MicheIl, T. ]., Walker, ]. A., Andrew, P. W, and Boulnois, G. ]., 1989,
Pneumolysin, the thiol-activated toxin of Streptococcus pneumoniae, does not require a thiol
group for in vitro activity, Infoct. Immun. 57:2547-2552.
37. Nandoskar, M., Ferrante, A., Bates, E. ]., Hurst, N., and Paton, ]., 1986, Inhibition of
monocyte respiratory burst, degradation, phospholipid methylation and bactericidal activ-
ity by pneumolysin, Immunology 59:515-520.
38. Paton,]. C. and Ferrante, A., 1983, Inhibition of human polymorphonuclear leukocyte
respiratory burst, bactericidal activity and migration by pneumolysin, Inftct. Immun.
41:1212-1216.
39. Feldman, C., MitchelI, T.]., Andrew, P. W, Boulnois, G.]., Read, R C., Todd, H., Cole, P.].,
and Wilson, R., 1990, The effect of Streptococcus pneumoniae pneumolysin on human
respiratory epithelium in vitro, Microbial Patlwgen. 9:275-284.
40. Feldman, C., Read, R., Rutman, A., Jeffery, P. K., Brain, A., Lund, v., Mitchell, T. ].,
Andrew, P. W, Boulnois, G. ]., Todd, H. C., et al., 1992, The interaction of Streptococcus
pneumoniae with intact human respiratory mucosa in vitro, Eur. Respir. J 5(5):576-583.
41. Rubins, F. B., Duane, P. G., Charboneau, n, and Janoff, E. N., 1992, Toxicity of pneumo-
lysin to pulmonary endothelial cells in vitro, Infect. Immun. 60(5):1740-1746.
42. Feldman, C., Munro, N. C., Jeffrey, P. K., Mitchell, T.]., Andrew, P. W, Boulnois, G.].,
Guerreiro, n, Rohde,]. A., Todd, H. C., Cole, P.]., et al., 1991, Pneumolysin induces the
salient histologic features of pneumococcal infection in the rat lung in vivo, Am. J Respir.
Cello Mol. Biol. 5:416-423.
43. Lock, R. A., Paton, J. C., and Hansman, D., 1988, Comparative efficacy of pneumococcal
neuraminidase and pneumolysin as immunogens protective against Streptococcus pneu-
moniae, Microbial Pathogen. 5:461-467.
43a.Rubins,J. B., Duane, P. G., Charboneau, D., andJanoff, E. N., 1992, Toxicity ofpneumo-
lysin to pulmonary endothelial cells in vitro, Infect. Immun. 60:1740-1746.
44. Coonrod, J. D., Varble, R., and Yoneda, K., 1991, Mechanism of killing of pneumococci by
lysozyme, J Infect. Dis. 164:527-532.
45. Andersson, B., Beachey, E. H., Tomasz, A., Tuomanen, E., and Svanborg-Eden, C., 1988, A
sandwich adhesin on Streptococcus pneumoniae attaching to human oropharyngeal epithelial
cells in vitro, Microbial Pathogen 4:267-278.
46. Kurono, Y., Shimamura, K., Shigami, H., and Mogi, G., 1991, Inhibition of bacterial
adherence by nasopharyngeal secretions, Ann. Otol. Rhinol. Laryngol. 100:455-458.
47. Kilian, M., Mestecky, J., Kulhavy, R., Tomana, M., and Butler, W. T., 1980, IgA. proteases
from Hemophilus injluenzae, Streptococcus pneumoniae, Neisseria meningitidis, and Streptococcus
sanguis: Comparative immunochemical studies, J Immunol. 124:2596-2600.
48. Kilian, M. J., Mestecky, J., and Schrohenloher, R. E., 1979, Pathogenic species of the genus
Hemophilus and Streptococcus pneumoniae produce IgA. protease, Infect. Immun. 26:143-149.
49. Male, C. J., 1979, Immunoglobulin A. protease production by Haemophilus injluenzae and
Streptococcus pneumoniae, Inftct. Immun. 26:254-261.
50. Kilian, M. and Reinhold, J., 1986, Interference with IgA defense mechanisms by extra-
cellular bacteria enzymes, Med. Microbiol. 5:173-208.
51. Winkelstein, J. A. and Swift, A. J., 1975, Host defense against the pneumococcus in
T-lymphocyte deficient, nude mice, Inftct. Immun. 12:1222-1223.
52. Szu, S. C., Clarke, S., and Robbins,J. B., 1983, Protection against pneumococcal infection in
mice conferred by phosphocholine-binding and relation to several types, Inftct. Immun.
39:993-999.
53. Talkington, D. F., Crimmins, D. L., Voellinger, D. C., Yother,J., and Briles, D. E., 1991, A 43-
kilodalton pneumococcal surface protein, PspA: Isolation, protective abilities, and struc-
tural analysis of the amino-terminal sequence, Infect. Immun. 59:1285-1289.
54. Au, C. C. and Eisenstein, T. K., 1981, Evaluation ofthe pneumococcal Forssman antigen (F-
polysaccharide) in the cross-serotype protection induced by pneumococcal subcellular
preparations, Infect. Immun. 31:169-173.
55. Griffoen, A. w., Rykers, G. T., Toebes, E. A. H., and Egers, B. J. M., The human in vitra
anti-type 4 pneumococcal polysaccharide antibody response is regulated by suppressor T
cells, Scand. J Immunol. 34:229-234.
56. Tvede, N., Heilmann, C., and Christensen, L. D., 1989, Interleukin 2 receptor expression
by human blood lymphocytes after vaccination with pneumococcal polysaccharides, Clin.
Exp. Immunol. 76:404-411.
57. Winkelstein, J. A., 1984, Complement and the host's defense against the pneumococcus,
Crit. Rev. Microbiol. 11(3):187-208.
58. Guckian, J. C., Christensen, G. D., and Fine, D. P., 1980, The role of opsonins in recovery
from experimental pneumococcal pneumonia,J Infect. Dis. 142:175-190.
59. Coonrod, J. D. and Rylkp-Bauer, B., 1977, Complement-fixing antibody response in
pneumococcal pneumonia, Infect. Immun. 18:617-623.
60. Mold, C., Nakayama, S., Holzer, T. J., Gewurz, H., and DuClos, T. w., 1981, C-reactive
protein is protective against Streptococcus pneumoniae infection in mice, J. Exp. Med.
154:1703-1708.
61. Kaplan, M. H. and Volnakis, J. E., 1974, Interaction of C-reactive protein complexes with
the complement system. 1. Consumption of human complement associated with the
reaction of C-reactive protein with pneumococcal C-polysaccharide and with the choline
phosphatides, lecithin and sphingomyelin,] Immunol. 112:2135-2147.
62. Osmond, A. P., Mortensen, R. F., Siegel,]., and Gewurz, H., 1975, Interactions ofC-reactive
protein with the complement system. 111. Complement-dependent passive hemolysis initi-
ated by CRP,] Exp. Med. 142:1065-1077.
63. Ziegler-Heitbrock, H. W. L., Passlick, B., KafferIein, E., Coulie, P. G., and Izbicki,]. R.,
1992, Protection against lethai pneumococcal septicemia in pigs is associated with de-
creased levels of interIeukin-6 in blood, Infect. Immun. 60:1692-1694.
64. MitcheII, T. ]., Andrew, P. w., Saunders, F. K., Smith, A. N., and Boulnois, G.]., 1991,
Complement activation and antibody binding by pneumolysin via a region of the toxin
homologous to a human acute-phase protein, Mol. Microbiol. 5(8):1883-1888.
65. Winkelstein, ]. A., Shin, H. S., and Wood, W. B., Jr., 1972, Heat labile opsonins to
pneumococcus. 111. The participation of immunoglobulin and of the alterative pathway of
C3 activation,] Immunol. 108:1681-1689.
66. Brown, E. ]., Joiner, K. A., Gaither, T. A.,Hammer, C. A., and Frank, M. M., 1983, The
interaction of C3b bound to pneumococci with factor H (ß-IH globulin), factor I (C3b/C4b
inactivator), and properdin factor B of the human complement system, ] Immunol.
131:409-415.
67. Winkelstein,]. A. and Tomasz, A., 1978, Activation ofthe alternative pathway by pneumo-
coccal cell wall teichoic acid,] Immunol. 120:174-178.
68. Winkelstein, ]. A., Bocchini,]. A., Jr., and Schiffman, G., 1976, The role of the capsular
polysaccharide in the activation of the alterative pathway by the pneumococcus,] Immunol.
116:367-370.
69. Hand, w.1. and Cantey,]. R., 1974, Antibacterial mechanisms ofthe lower respiratory tract.
I. Immunoglobulin synthesis and secretion,] Glin. Invest. 53:354-362.
70. Robertson,]., Caldwell,]. R., Castle,]. R., and Waldman, R. H., 1976, Evidence for the
presence of components of the alternative (properdin) pathway of complement activation
in respiratory secretions,] Immunol. 117:900-903.
71. Reynolds, H. Y. and Newball, H. H., 1974, Analysis of proteins and respiratory cells
obtained from human lungs by bronchiallavage, ] Lab. Glin. Med. 84:559-573.
72. Stephens, C. G., Williams, R. C., Jr., and Reed, W. P., 1977, Classical and alternative
complement pathway activation by pneumococci, Infect. Immun. 17:296-302.
73. Holzer, T. ]., Edwards, K. M., Gewurz, H., and Mold, C., 1984, Binding of C-reactive
protein to the pneumococcal capsule or cell wall results in differential localization of C3
and stimulation of phagocytosis,] Immunol. 133:1424-1430.
74. Hosea, S. w., Brown, E.]., and Frank, M. M., 1980, The critical role of complement in
experimental pneumococcal sepsis,] Infect. Dis. 142:903-909.
75. Wood, W. B., Jr., 1941, Studies on the mechanism of recovery in pneumococcal pneumonia.
I. The action of specific antibody upon the type of pulmonary lesion of experimental
pneumonia,] Exp. Med. 73:201-222.
76. Nakajima, R., Namba, K., Ishida, Y., Une, T., and Osada, Y., 1990, Protective role of
complement in the development of experimental pneumococcal pneumonia in mice,
Ghemotherapy 36:287-293.
77. Johnston, R. B., Jr., 1981, The host response to invasion by Streptococcus pneumoniae:
Protection and the pathogenesis of tissue damage, Rev. Infect. Dis. 3:282-288.
78. Osler, w., 1978, Lobar Pneumonia, in: The Principles and Practice ofMedicine: Designed For the
Use of Practitioners and Students of Medicine (A. M. Harveyand V. A. McKusick, eds.), New
York: Appleton-Century-Crofts, pp. 75-107.
79. Newman, S. L., Vogler, L. B., Feigin, R. 0., and Johnston, R. B., Jr., 1979, Recurrent
septicemia associated with congenital deficiency of C2 and partial deficiency of factor
Band the alternative complement pathway, N. Engl.] Med. 299:290-292.
80. Sjoholm, A. G., HaUberg, T., Oxelius, V. A., Hammarstrom, L., Smith, C. I. E., and
Lindgren, F., 1987, C2 deficiency, moderately low IgG2 concentrations and lack of G2m(23)
aUotype marker in a child with repeated bacterial infections, Acta. Paediatr. Scand. 76:
533-588.
81. Rosen, F. S. andJaneway, C. A., 1966, The gamma globulins. III. The antibody deficiency
syndromes, N. Engl. J Med. 275:709-715.
82. Matter, L., Wilhe1m, J. A., Angehrn, w., Skvaril, F., and Schopfer, K., 1985, Selective
antibody deficiency and recurrent pneumococcal bacteremia in a patient with Sjogren's
syndrome, hyperimmunoglobulinemia G, and deficiencies of IgG2 and IgG4 , N. Engl. J
Med. 1039-1042.
83. Twomey, J. J., 1973, Infections complicating multiple myeloma and chronic lymphocytic
leukemia, Arch. Intern. Med. 132:562-565.
84. Rimland, D., 1983, Mechanisms of ethanol-induced defects of alveolar macrophage func-
tion, Alcolwlism 8:73-76.
85. Astry, C. L., Warr, G. A., and Jakab, G. J., 1983, Impairment of PMN immigration as a
mechanism of alcohol-induced suppression of pulmonary antibacterial defenses, Am. Rev.
Respir. Dis. 128:113-117.
86. Okeson, G. C., and Divertie, M. B., 1970, Cilia and bronchial clearance: The effects of
pharmacologic agents and disease, Mayo Clin. Proc. 45:361-373.
87. Spinozzi, F., Cimignoli, E., Gerli, R., Agea, E., Bertotto, A., Rondoni, F., and Grignani, F.,
1992, IgG subclass deficiency and sinopulmonary bacterial infections in patients with
alcoholie liver disease, Arch. Intern. Med. 152:99-104.
88. Perlino, C. A. and Rimland, D., 1985, Alcoholism, leukopenia, and pneumococcal sepsis,
89. Liu, Y. K., 1980, Effects of alcohol on granulocytes and lymphocytes, Sem. Hematol. 17:
13-136.
90. Imperia, P. S., Chikkapa, G., and Phillips, P. G., 1984, Mechanism of inhibition of
granulopoiesis by ethanol, Proc. Soc. Exp. Biol. Med. 174:219-225.
91. Davis, C. C., MeUencamp, M. A., and Preheim, L. C., 1991, A model of pneumococcal
pneumonia in chronicaUy intoxicated rats, J Infect. Dis. 163:799-805.
92. Lister, P. D., Preheim, L. C., and Gentry, M. J., 1992, Ethanol impairs neutrophil chemotaxis
but not adherence or recruitment to lungs in rats with experimental pneumococcal
pneumonia, Abstracts of the 9200 General Meeting of the American Society for Microbiology,
abstract B-235:65.
93. MacGregor, R. R., Gluckman, S. J., and Senior, J. R., 1978, Granulocyte function and levels
of immunoglobulins and complement in patients admitted for withdrawal from alcohol,
J Infect. Dis. 138:747-753.
94. Perrin, L. H., Lambert, P. H., Nydegger, U. E., and Miescher, P. A., 1973, Quantitation of
C3PA (properdin factor B) and other complement components in diseases associated with a
low C3 level, Clin. Immunol. Immunopatlwl. 2:16-27.
95. MacGregor, R. R., 1986, Alcohol and immune defense,JAMA 256:l474-1479.
96. Zetterman, R. K. and Sorrell, M. F., 1981, Immunologie aspects of liver disease, Gastro-
enterology 81:616-624.
97. MeIlencamp, M. A. and Preheim, L. C., 1991, Pneumococcal pneumonia in a rat model of
cirrhosis: Effects of cirrhosis on pulmonary defense mechanisms against Streptococcus
pneumoniae,J Infect. Dis. 163:102-108.
98. Rimola, A., Soto, R., Bory, F., Arroyo, v., Piera, C., and Rodes,J., 1984, Retieuloendothelial
system phagocytie activity in cirrhosis and its relation to bacterial infections and prognosis,
Hepatology 4:53-58.
99. Grover, G. J. and Loegering, D. j., 1984, Role of the liver in host defense to pneumococcus
following splenectomy, J Surg. Res. 37:448-452.
100. Plotkowski, M. C., Puchelle, E., Beck, G., Jacquot, J., and Hannoun, C., 1986, Adherence of
type I Streptococcus pneumoniae to tracheal epithelium of mice infected with influenza
AlPR8 virus, Am. Rev. Respir. Dis. 134:1040-1044.
10l. Tristram, D. A., Miller, R. w., McMillan, J. A., and Weiner, L. B., 1988, Simultaneous
infection with respiratory syncytial virus and other respiratory pathogens, Am. J. Dis. Child.
142:834-836.
102. Korppi, M., Leinonen, M., Koskela, M., Makela, H., and Launiala, K., 1989, Bacterial
coinfection in children hospitalized with respiratory syncytial virus infections, Pediatr. Dis. J.
8:687-692.
103. Jones, E. E., Alford, P. L., Reingold, A. L., RusselI, H., Keeling, M. E., and Broome, C. v.,
1984, Predisposition to invasive pneumococcal iIIness following parainfluenzae type 3 virus
infection in chimpanzees,J. Am. Veto Med. Ass. 185:1351-1353.
104. O'Neal, B. J., and McDonald,J. C., 1981, The risk of sepsis in the asplenic adult, Ann. Surg.
194:775-778.
105. Schwartz, P. E., Sterioff, S., Mucha, P., Melton, L. J., and Offord, K. P., 1982, Postsplenec-
tomy sepsis and mortality in adults, JAMA 248:2279-2283.
106. Selby, C., Hart, S., Ispahani, P., and Toghill, P. J., 1987, Bacteremia in adults after
splenectomy or splenic irradiation, Quarterly J. Med. 63:523-530.
107. Herbert, J. C., 1989, Pulmonary antipneumococcal defenses after hemisplenectomy, J.
Trauma 29:1217-1220.
108. Zarkowsky, H. S., Gallagher, D., GiII, E M., Wang, F. C., Falletta,J. M., Lande, W. M., Levy, P.
S., Verter, J. 1., Wethers, D., and the Cooperative Study of Sickle Cell Disease, 1986,
Bacteremia in sickle hemoglobinopathies,J. Pediatr. 109:579-585.
109. Pearson, H. A., Spencer, R. P., and Cornelius, E A., 1969, Functional asplenia in young
children with sickle-cell anemia, N. Engl. J. Med. 281:923-926.
110. Bjornson, A. B., Lobel,J. S., and Harr, K. S., 1985, Relation between serum opsonic activity
for Streptococcus pneumoniae and complement function in sickle-cell disease, J. Inftct. Dis.
152:701-708.
111. Van Wyck, D. B., Witte, M. H., and Witte, C. L., 1982, Synergism between the spleen and
serum complement in experimental pneumococcemia,J. Infect. Dis. 145:514-519.
112. Redd, S. C., Rutherford, G. w., III, Sande, M., Lifson, A. R., Hadley, W. K., Facklam, R. R.,
and Spika,J. S., 1990, The role ofhuman immunodeficiency virus infection in pneumococ-
cal bacteremia in San Francisco residents,J. Inftct. Dis. 162:1012-1017.
113. Pesola, G. R. and Charles, A., 1992, Pneumococcal bacteremia with pneumonia; Mortality
in acquired immunodeficiency syndrome, Chest 101:150-155.
114. Pahwa, S. G., Quilop, M. T.J., Lange, M., Pahwa, R. N., and Grieco, M. H., 1984, Defective
B-Iymphocyte function in homosexual men in relation to the acquired immunodeficiency
syndrome, Ann. Intern. Med. 101:757-763.
115. Lane, H. C., Masur, H., Edgar, L. C., Whalen, G., Rook, A. H., and Fauci, A. S., 1983,
Abnormalities of Beeil activation and immunoregulation in patients with the acquired
immunodeficiency syndrome, N. Engl. J. Med. 309:453-458.
116. Pinching, A. J., 1991, Antibody responses in HIV infection, Clin. Exp. Immunol. 84:
181-184.
117. Parkin, J. M., Helbert, M., Hughes, C. L., and Pinching, A. J., 1989, Immunoglobulin G
subclass deficiency and susceptibility to pyogenic infections in patients with AIDS-related
complex and AIDS, AIDS 3:37-39.
118. Glaser, J. B., Volpe, S., Aguirre, A., Simpkins, H., and Schiffman, G., 1991, Zidovudine
improves response to pneumococcal vaccine among persons with AIDS and AIDS-related
complex,J. Inftct. Dis. 164:761-764.
119. Senaldi, G., Peakman, M., McManus, T., Davies, E. T., Tee, D. E. H., and Vergani, D., 1990,
Activation of the complement system in human immunodeficiency virus infection: Rele-
vance ofthe classical pathway to pathogenesis and disease severity,J Infoct. Dis. 162:1227-
1232.
120. Chou, M. Y., Brown, A., Blevins, A., and Armstrong, 0., 1983, Severe pneumococcal
infection in patients with neoplastic disease, Cancer 51:1546-1550.
121. Zurlo,].]., Schechter, G. P., and Fries, L. F., 1989, Complement abnormalities in multiple
myeloma, Am. J Med. 87:411-420.
122. Savage, 0. G., Lindenbaum, ]., and Garrett, T. ]., 1982, Biphasic pattern of bacterial
infection in multiple myeloma, Ann. Intern. Med. 96:47-50.
123. Simons, R.]. and Reynolds, H. Y., 1990, Altered immune status in the elderly, Sem. Respir.
Dis. 5:251-259.
124. Esposito, A. L., Poirier, W]., and Clark, C. A., 1989, The cardiac glycoside digoxin disrupts
host defense in experimental pneumococcal pneumonia by impairing neutrophil mobiliza-
tion, Am. Rev. Respir. Dis. 140:1590-1594.
125. Amber, I. ]., Gilbert, E. M., Schiffman, G., and Jacobson,]. A., 1990, Increased risk of
pneumococcal infections in cardiac trans plant recipients, Transplantation 49:122-125.
126. McLean, R. H., Forsgren, A., Bjorksten, B., Kim, Y., Quie, P. G., and Michael, A. F., 1977,
Decreased serum factor B concentration associated with decreased opsonization of Esche-
richia coli in the idiopathic nephrotic syndrome, Pediatr. Res. 11:910-916.
3
Pulmonary Infections Caused

by Lancefield Group A Beta-
Hemolytic Streptococcus
RICHARD H. GOLDSTEIN and ELLIE J.C. GOLDSTEIN
1. INTRODUCfION
Streptocoeci are important pulmonary pathogens, eausing epidemie pneumonias

sinee the sixteenth eentury. For most of the twentieth eentury their clinieal
importanee as a eause of pyogenie pneumonia has been in the setting of military
populations in camps. Reeently, some beta streptocoeci have beeome assoeiated
with more virulent human disease in the general population including pneu-
monia. The mortality of untreated streptoeoeeal pneumonia is between 20 and
30%. The use of appropriate antibioties reduees this mortality to less than 2-3%.
For this reason alone, the medieal praetitioner must be aware of the nature of
streptoeoeci as a pulmonary pathogen.
2. MICROBIOLOGY/IMMUNOLOGY
Streptoeoeci are gram-positive, eatalase negative, oxidase negative eoeci that

usually grow in ehains ofvarying length. Strains grown in liquid media (eg. broth,
RICHARD H. GOLDSTEIN and ELLIE ].C. GOLDSTEIN • Department of Medicine, Pul-

monary Division and Infectious Disease Division, St. John's Hospital and Health Center, Santa
Monica, California 90404, and VCLA School of Medicine, Los Ange1es, California 90024.
PulmonaTY Infeetions and Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
51
52 RICHARD H. GOLDSTEIN and ELLIE J.C. GOLDSTEIN
blood) tend to form longer chains than those grown on agar media. If the culture
is aged, the streptococci may appear as gram negative. They are facultative
anaerobes, with some strains being microaerophilic. Streptococcaceae are differ-
entiated from Micrococcaceae (Micrococcus species and Staphylococcus species) by a
negative catalase test. The usual method for catalase testing is application of
hydrogen peroxide to a colony on a blood plate. If the mixture bubbles it is
positive. However, this test is nonspecific and occasional peroxidase-producing
strains of streptococci and nonproducing strains of staphylococci may be con-
fused.! In these infrequent instances, or where there is some need to be more
definitive, the benzadine test may be used. Streptococci are benzadine negative
and staphylococci are positive. Additionally, some streptococci may form tetrads
and be mistaken for staphylococci on Gram's stain. Streptococci mayaiso appear
as small gram positive rods under certain growth conditions.
Bergey's Manual 2 divides the Streptococcaceae family into five genera:
Aerococcus, Gamella, Leuconostoc, Pedicoccus, and Streptococcus. Microbiologists fur-
ther classify the genus Streptococci by their ability to hemolyze blood (alpha, beta,
or gamma) and by specific capsular polysaccharide antigens. Streptococci are the
predominant normal oral flora of the human mouth and oropharynx.
A serologic scheme, based on cell-wall polysaccharide antigens and pi-
oneered by Dr. Rebecca Lancefield of RockefeIler University in New York City,
enabled researchers to differentiate the beta-hemolytic streptococci.3 There are
many (> 15) serologic groups ofbeta-hemolytic streptococci designated Groups A
to V Group A beta-hemolytic streptococci, Streptococcus pyogenes, is a common
human pathogen and is associated with sore throat, wound infections, impetigo,
scarlet fever, and several nonsupperative sequelae, acute rheumatic fever, and
poststreptococcal acute glomerular nephritis. Its role as a cause of pneumonia is
less well appreciated. Rare strains of S. pyogenes may not appear beta-hemolytic. It
usually appears spherical with a diameter of 0.5-1.0 fJ.m. The term pyogenes is
from Greek and means "pus producing"; the species was named in 1884 by
Rosenbach 4 and the type strain is ATCC 12344. It derives energy from fermenta-
tion and will produce acid from glucose, maltose, sucrose, and salicin.
Selective inhibition by a bacitracin disk (0.04 units) presumptively differenti-
ates S. pyogenes from other beta-hemolytic streptococci. The Lancefield precipitin
test using specific antisera should be performed soon after isolation as the specific
pro tein differences may be lost following laboratory cultivation. Occasionally,
cross reactions may occur but these are often eliminated by repeat testing
involving further dilutions, 2-hr incubation, and overnight refrigeration. Fresh
isolates may be frozen for later testing. Other typing methods include the
T-agglutination technique, immunofluorescence and counterimmune electro-
phoresis (CIE).
The cellular structure of S. pyogenes is complex. It contains a capsule of
hyaluronic acid, a cell wall made of four components: (1) the carbohydrates
rhamnose and N-acetylglucosamine in a ratio of 2:1, (2) lipoteichoic acid, (3) M,
T, and R antigenic proteins, and (4) peptidoglycans/mucopeptides, and a phos-
pholipid cell membrane. Several of these components act as virulence factors. The
LANCEFIELD GROUP A BETA-HEMOLYTIC STREPTOCOCCUS 53
capsular hyaluronic acid impedes phagocytosis. The M protein is the major

virulence factor of S. pyogenes and allows it to persist in tissues for weeks. Whereas
more than 80 sero types have been recognized and different degrees of virulence
are associated with specific M-protein types, strains entirely lacking M protein
cannot initiate disease and are avirulent. Lancefield demonstrated that phago-
cytes will avoid S. pyogenes with M proteins and readily attack those without them.
Antigenie variation of M proteins allows the organism to avoid antibody recogni-
tion. M proteins help the organism resist phagocytosis by several mechanisms.
The mucoid and matt colonial morphologie forms contain relatively large
amounts of M proteins.
M proteins are heat stable and tripsin sensitive and appear as slender
projections from the cell wall. This projection consists of aCcterminus (carboxyl
end) with a fairly constant amino-acid composition buried within the cell and an
N-terminus (amino acid) with a very variable amino-acid composition away from
the cell wall. Type-specific antibodies must be formed by the host to give immu-
nity. Consequently, this variability of M proteins can prevent the host from rapid
antibody attachment which aids phagocytosis unless the type-specific antibody is
produced/present. Thus, natural selective pressures may favor the development
ofboth variable size and different types ofM proteins. The N-terminus uses other
sophisticated, and currently not fully explained, methods to bind/attach to cell
surfaces. It also holds a net negative charge that may electrostatically repulse
PMNs. H factor, a regulatory component of the human complement system, can
bind to M proteins and thereby limit binding by the C3b component of comple-
ment and help evade phagocytosis. All this allows the organism to persist in
tissues while the body's defenses are thwarted at every turn. Consequently,
specific M protein types, which appear, disappear, and reappear in the environ-
ment, have become associated with more virulent human disease and the reem-
ergence of serous streptococcal diseases including an increase of pneumonia. A
cogent review of streptococcal M protein by Fischetti5 is available for those
interested in more information.
Cell wall lipoteichoic acid is a virulence factor that allows streptococcal
adherence to fibronectin on the epithelial cell surface. In addition, pyogenic
exotoxin, formerly called erythrogenic toxin and scarlet fever toxin, is an extra-
cellular product responsible for rash and cutaneous manifestations of infection
and is also pyogenic and cytotoxic. Streptolysin 0, which is oxygen labile and
produced by almost all strains of S. pyogenes, is also toxic to polymorphonuclear
leukocytes, platelets and other cells. It forms the basis of the antistreptolysin 0
(ASO) titer used as an indicator of recent infection. Streptolysin S, which is
elaborated in the presence of serum, is nonantigenie but can damage PMNs,
platelets, and other cells. This is the hemolysin that produces hemolysis on the
agar plate surface. S. pyogenes also produces several types of DNA-ases, hyaluroni-
dase, and streptokinase (previously called fibrinolysin) which dissolve and digest
cellular materials and help the organism spread through tissue planes. Other
extracellular products such as dinucleotidases, proteinases, amylase, and es-
terases are also elaborated and may playa role in pathogenicity.
3. PATTERNS OF HEMOLYTIC STREPTOCOCCAL PNEUMONIA
MacCallum,6 in his classic 1919 description of streptococcal pneumonia,

attributes aseries of recurrent pneumonia epidemics in Europe and the United
States beginning in the sixteenth century to hemolytic streptococci. Delirium, a
productive cough, and a rapid demise over 6-7 days characterized this illness.
Pleural effusions were common, and observers described the affected lung as
indurated and turgid, often with abscess formation. This form of epidemie
pneumonia was originally called the malignant stitch, but later typhoid or
catarrahal pneumonia.
According to MacCallum, Woodward reviewed 435 autopsied cases of pneu-
monia in US troops during the Civil War and found 300 to be lobar in distribution
and 135 to be secondary or catarrahal. The 135 cases of catarrahal pneumonia
were of a bronchopulmonary distribution and were frequently accompanied by
pleural effusion. Other medical investigators of the same period made similar
observations. In addition, Woodward noted that of the 135 cases, 101 followed
measles infections. MacCallum believed the pneumonia affecting these 135 cases
to be similar to the sixteenth- and seventeenth-century epidemics and to have a
common etiologic agent, hemolytic streptococci.
To establish the etiology, MacCallum obtained preserved lung tissue from the
US Army Medical Museum of three Civil War Soldiers who died of catarrahal
pneumonia following measles infection. MacCallum found abundant streptococci
to be present in all three. Microscopically the lungs of the Civil War soldiers
showed an interstitial bronchopneumonia, with peribronchial consolidation, and
in some instances areas of necrosis. Hepatization was less common than in cases of
lobar pneumonia. Other early observers, describing pneumonia following infec-
tions with measles, whooping cough, and diphtheria confirmed this pattern of
interstitial and peribronchial involvement.
MacCallum goes on to report his observations of pneumonia in US Army
camps during the winter-spring of 1917-18. He identified hemolytic strep in fec-
tions by their presence in sputum, blood, or pleural fluid. He found measles to be
the most important predisposing cause. At first he believed hemolytic streptococ-
TABLE I
Patterns of Hemolytic Streptococcal Pneumonia
Series Date No. pts. Mortality Comments
Miller 1918 464 31% 164 empyemas

MacCallum 1919 53 53 autopsies/c1assie description and history
Keefer 1940 55 18.1% Epidemiologie patterns
WeIch 1959 20 0.5% Primary pneumonias/l autopsy
Basiliere 1964 95 0 Incidence of pneumonia and streptococcal
carriage rates
cal pneumonia to be a secondary infection, but later in the season an epidemic of

hemolytic streptococcal pneumonia arose independent of other infections. He
attributed this to a change in the virulence of the organism. MacCallum distin-
guished two basic forms of pulmonary involvement. The most common was that
of an interstitial bronchopneumonia. In interstitial bronchopneumonia, strepto-
cocci infected the bronchi, the lymphatics, and often the pleural space. A second-
ary consolidation occurred in the adjacent peribronchial tissues, but without
tissue invasion by the streptococci. Lobular or lobar pneumonia was less common.
In lobular pneumonia, streptococci invade lung tissue and can be found in
abundant numbers in the alveoli. Necrosis of lung tissue and abscess formation
frequently accompanied lobular or lobar pneumonia.
Areport by Miller and Lusk in 19187 described an epidemic of streptococcal
pneumonia occurring from March 20, to June 1, 1918, in US Army troops
quartered at Camp Dodge. They describe 464 cases of "lobular" pneumonia due
to streptococcal infection. Overall mortality was 31.7%, rising to 61.7% in those
with empyema. One hundred and nine cases came to autopsy. Bronchopneu-
monia was present in 78 of the 109 cases. A cough in 122 cases or a sore throat in
65 cases preceded the onset of pneumonia. Only 24 cases followed measles
infection. In many the onset of illness was abrupt and dramatic. Thirty-two
percent of the pneumonias developed empyema. During the first month of the
epidemic, empyema occurred in 42% of cases. Following the first weeks, the
incidence of empyema fell to that of previous months, and was similar to that
found in lobar or pneumococcal pneumonia (17%). They found exudative peri-
carditis in 38 cases, and peritonitis in 17 cases. Other metastatic manifestations
were uncommon.
In 1939, Chester S. Keefer, Franz J. Ingelfinger, and Wesley W. Spink8
described their experience in 246 patients with hemolytic streptococcal bacte-
remia. They observed the highest incidence of bacteremia during the months of
January to May, corresponding to the peak incidence of beta hemolytic strepto-
coccal carriage rates. The overall fatality rate was 72%. Metastatic lesions in the
form of abscesses were in frequent. Of the 246 cases reported, there were three
cases of pneumonia, five cases of empyema, and two cases of mediastinitis. Eight
patients with bacteremia had detailed microbiologic studies. In all eight, the
authors isolated beta hemolytic streptococci. All belonged to Lancefield group A,
and were M strains.
Later Keefer,9 in a 1941 review of streptococcal pneumonia, noted that
hemolytic streptococci cause 3% to 5% of all pneumonias. Keefer goes on to
describe 55 cases of hemolytic streptococcal infection involving the lungs or
pleura. Ten of these cases were primary hemolytic streptococcal pneumonias. Of
the remainder, 13 were preceded by an infection of the upper respiratory tract
(measles, influenza, viral URIs, pharyngitis, or erysipelas), and 10 had an associ-
ated pneumococcal infection. Like MacCallum,6 he found hemolytic strepto-
coccus to produce patterns of both interstitial and confluent pneumonias. Em-
pyema occurred in 20% of cases. Keefer associated streptococcal pneumonia with
chronic lung disease, especially those affecting the airways: asthma, bronchi-
ectasis, and chronic cystic disease. He found the age of affected individuals to be
evenly distributed between the second and sixth decades.
Prior to the introduction of antibiotics, fatality rates varied from 30 to 60%. A
series of studies in the early 1940s cited a significant reduction in mortality by the
use of sulfonamides. However, the use of sulfonamides did not reduce the
incidence of empyema or shorten the duration of treatment. In Keefer's9 series
some received sulfonamides, while others did not receive any antibiotics. There
were ten deaths, for an overall fatality rate of 18.1%. The fatality rate was 9% in
nonbacteremic patients, and 57% in those with bacteremia. Bacteremia occurred
in seven of the 55 patients (12%). Empyema was more common in individuals
under 40 years of age (10 cases out of a total of 16 cases with empyema). Of note,
only one patient with empyema had bacteremia. In Keefer's 10 cases of primary
streptococcal pneumonia the onset was abrupt and accompanied by symptoms of
an acute respiratory infection. Patients exhibited pulmonary signs of pneumonia
without lobar consolidation. Diagnosis was by sputum culture. Patients were ill
for seven to 21 days. Two deaths occurred in this group both in association with
bacteremia. Pneumonia preceded by prior upper respiratory infection occurred
in 13 ca ses. An increase in the severity of existent symptoms accompanied the
onset of pneumonia. In all 13 cases, patients were under the age of 40 years, and
all received sulfonamides. There were no deaths in this group.
In the 10 cases of streptococcal pneumonia associated with pneumococcus
infection, some accompanied active pneumococcal infection, whereas others
caused abscess formation or late secondary infections of the lung, pleural space,
or mediastinum.
In 1946, MacFarland lO published his experience with 4,000 cases of strepto-
coccal pneumonia. He described three patterns of involvement: a disseminated
bronchopneumonia occurring secondary to some other disease process, and most
common outside of military populations; focal pneumonia following a viral
infection or streptococcal pharyngitis, this form becoming common as the inci-
dence of streptococcal disease achieves epidemie proportions in a closed popula-
tion; and primary streptococcal pneumonia characterized by a toxic course and
tendency to abscess formation.
Welch et al. ll describe an outbreak of streptococcal pneumonia among Navy
men during the winter of 1959-60. They report 20 cases, with one fatality. The
pathologie changes of interstitial bronchopneumonia found in the one case
undergoing autopsy were similar to those reported by MacCallum. 6 Nineteen of
the 20 cases had pleural effusions. Ofthe 19 survivors, all had a rather prolonged
clinical course despite antibiotic therapy. Clinically the disease was abrupt in
onset and characterized by fever, cough, and physical findings of pneumonia.
Of note, 18 of the 20 cases appeared to be primary streptococcal pneumonias,
without evidence of antecedent illness.
Basiliere et al. 12 described streptococcal pneumonia among military recruits
in training in San Diego. They describe 95 cases of beta-hemolytic streptococcal
pneumonia, occurring between July of 1964 and February of 1966. In all 95 cases,
isolates of beta-hemolytic streptococcus belonged to Lancefield group A. There
was no anteeedent epidemie illness du ring the study period. Thirty-six pereent of
patients had a history of reeent upper-respiratory infeetion. Periods of peak
streptocoeeal infeetion ineidenee corresponded to periods of inereased erowding
of reeruits. The streptoeoeeal pneumonia rates closely followed the rates of overall
streptocoeeal infeetion. Only 19% of patients had beta-hemolytie streptocoeci
present in throat eultures. Twenty-one pereent of patients gave a his tory of
previous pneumonia. Forty-six pereent of patients were smokers. Fevers, ehills,
ehest pain, and eough were the most common symptoms. Of 74 patients with
pleuritie ehest pain, two thirds developed empyema. Overall the eourse was
prolonged despite antibiotie therapy. Fever persisted for more than a week in 61 %
of the eases. Only one of 57 blood eultures demonstrated baeteremia. Empyema
oeeurred in 57% of eases. There was only one ease of aeute glomerulonephritis.
Streptococcus pyogenes eauses between 20 and 30 million infeetions/year in the
United States. 5 Aeeording to Murray and Nadel,13 streptoeoeeal pneumonia now
aecounts for less than 1% of all pneumonias. The same patterns of pulmonary
involvement as deseribed above persist, with the addition of a pieture of aeute
alveolar eapillary leak syndrome as part of the toxie shoek syndrome.
4. EPIDEMIOLOGY
Often, beta-hemolytie streptoeoeei first inhabit the oropharynx. They then

gain aeeess to the lower airways through the proeess of mieroaspiration of
oropharyngeal contents. This proeess oeeurs in healthy individuals as well as
those with altered levels of eonsciousness or impaired swallowing. 14 Seeondary
beta-hemolytie streptocoeci pneumonia (e.g., those following a viral infeetion)
then oeeurs as a result of altered host defense meehanisms. In the epidemie form,
beta-hem01ytic streptoeoeei pneumonia oeeurs both as a primary as weH as a
seeondary pneumonia. 1O Disease oeeurs beeause of an inerease in the volume or
virulenee of mieroorganisms aspirated, as well as due to a1terations in host
defense. In Basiliere's series,12 the rate of infeetion with beta-hemolytic strepto-
eoeci onee the pathogen was aequired was 40%, and 70% were primary pneu-
monias.
Blood eultures are infrequently positive in both the seeondary and epidemie
form of beta-hemolytie streptoeoeei pneumonia, and may indicate a poor prog-
nosis. 9
Invasive beta-hemolytie streptoeoeci has been reported for several eentu-
ries.l 5 Yet in reeent years the disease eaused by beta-hemolytie streptocoeci has
seemed less virulent and somewhat attenuated during most of the twentieth
eentury. In the mid-1980s, however, reeurrenees of invasive streptoeoeeal disease
including rheumatie fever l6 reminiseent of the disease of former years has re-
emerged. Stevens et al. 17 noted 20 patients from the Roeky Mountain states
developed a toxie shoek-like syndrome assoeiated with pyogenie exotoxin A and
M types 1, 3, and 28. Nineteen (95%) of their patients developed shoek, 11
patients (55%) developed acute respiratory distress syndrome, and 10 required
intubation. The overall mortality was 30%. They suggested that this was clearly a
toxin-mediated illness. Subsequendy, Schwartz et al. 18 noted the changing epide-
miology of group A streptococcal infections in the United States. They studied
M-type and T-type data of 5,193 strains sent to Center for Disease Control
between 1972 and 1988. Analysis showed that M-types 1, 3, and 18 had "increased
significantly" and were "more likely to be invasive, to cause fatal infection ... " In
contrast, the less virulent and less invasive M-4 and M-12 types had decreased.
Similarly, Gworzewska and Colman 19 found M types 1, 3, and 28 had increased in
specimens sent to the United Kingdom Central Public Health Laboratory be-
tween 1980 and 1987 and M-l type was more often associated with severe invasive
disease. Two recent reviews, one by Bisn020 and one by Stevens 15 provide addi-
tional details about invasive beta-hemolytic streptococci infections.
Pneumonia associated with invasive beta-hemolytic streptococci is associated
more frequendy with positive blood cultures. In this setting the lungs may be the
portal of entry or actually a secondary site of infection following hematogenous
dissemination. McWhinney,21 noting the high mortality associated with beta-
hemolytic streptococci sepsis (as much as 35% in some series 22 ) and the favorable
outcome in sepsis patients with pneumonia, goes on to suggest that pneumonia
may be apart of the naturallife of invasive beta-hemolytic streptococci infection
in patients who are not initially overwhehned by sepsis. 21
Invasive beta-hemolytic streptococci pneumonia is further characterized by
features of invasive beta-hemolytic streptococci infection in general. Most cases
are community acquired, although nosocomial acquisition does occur and is
particularly important in the postoperative and postradiation therapy setting.
Invasive beta-hemolytic streptococci infections are more frequent in patients with
solid tumors. The most important portal of entry is the skin. 23
5. PATHOLOGY
MacCallum, in his series of pneumonias from the winter of 1917-1918,6

describes the pattern of pulmonary involvement as that of an interstitial broncho-
pneumonia or of lobular consolidation occurring both separately and together.
Interstitial bronchopneumonia is accompanied by patchy atelectasis caused by
obstruction of the involved bronchioles and the formulation of nodules of consol-
idation about the terminal bronchioles with surrounding hemorrhage. Pleural
effusion is often present. In instances of lobular consolidation, peribronchiole
nodules do not form. Consolidated alveoli are filled with blood, leukocytes, and
streptococci. Widespread necrosis and destruction of lung tissue may lead to the
formation of abscesses. As described by MacCallum:
The essential feature, therefore, in the gross appearance of the lung in interstitial
bronchopneumonia lies in the filling of the bronchioles with opaque exudate, the
thickening of their walls, and the consolidation of the adjacent lung tissue which
appears in cross section as a nodule, in longitudinal section as a branching mantle
of dense tissue bordered with hemorrhage accompanying the bronchiole. Atelec
tasis of the intervening lung substance, great thickening and prominence of the
interlobular septa and pleura, and usually pleurisy with abundant effusion make
up the whole picture.
MacCaHum found the affected bronchioles to be fiHed with great numbers of

Streptococci as weH as leukocytes.
It appears as a primary infection of the bronchioles. In the course of the affection
thickening and infiltration of the bronchial and alveolar walls, of the interlobular
septa, perivascular adventitial tissue, and of the pleura occur. The lymphatics are
widely infected and thrombosed. The alveoli about the infected bronchi are filled
with fluid and blood, dense cagula of fibrin appear, the epithelial cells multiply
and are shed into the lumen, a few mononuclear cells appear, and organization
takes place in this exudate until bronchi and alveoli contain strands of fibrous
tissue. In the meantime the infection, probably extending by growth along the
lymphatics, reaches the pleura and sets up a pleurisy with effusion, one effect of
which is to cause extensive collapse of the lung.... It is important to consider at
this point the distribution of the bacteria in the lung in this affection. The exudate
in the lumen of each bronchiole always contains streptococci in considerable
numbers; the lymphatics in the bronchial wall and those seen elsewhere contain
them in numbers in the thrombi; the superficial layers of the exudate on the
pleural surface contain a perfect feltwork of these organisms sometimes enelosed in
leukocytes; but the contents of the alveoli and the substance of the tissue contain
none .... All this must signify a definite power of resistance on the part of the
tissues in these cases, indicating their ability to destroy the organisms whenever
they come into sufficiently elose contact with the living cells....
In lobular or lobar involvement, MacCaHum found great numbers of streptococci

in contact with the tissues. In this setting, the primary lesion was often that of an
interstitial bronchopneumonia. However, as alveoli filled with bacteria and leuko-
cytes, there arose foci of consolidation. In other instances, invasion of the alveoli
occurred without interstitial bronchopneumonia. When hemolytic streptococci
were able to invade lung tissue, lymphatic involvement was striking, and abscess
formation was not unusual. Hemolytic streptococci and an exudate of leukocytes
and fibrin fiHed the involved alveoli. Hemorrhage often occurred into these areas
of consolidation.
In patients dying of fulminant disease associated with toxic shock syndrome
the lungs demonstrate a picture of acute alveolar-capillary damage and pulmon-
ary edema. 6
6. CLINICAL MANlFESTATIONS
An abrupt onset of symptoms characterizes hemolytic streptococcal pneu-

monia. The most common symptoms include fever, cough, dyspnea, purulent
sputum, hemoptysis, and pleuritic chest pain. Patients with chest pain more
frequently develop empyema. 12 The patient is often alert, apprehensive, sleepless,
and delirious.6 Patients are often severely ill. The examiner finds cyanosis,
tachycardia, and tachypnea as weH as signs of pulmonary consolidation or pleural
60 RICHARD H. GOLDSTEIN and ELLIE ].C. GOLDSTEIN
effusion. In more fulminant forms of the disease, patients may present with all or
some of the features of the sepsis syndrome.
Fever persisting up to seven days despite appropriate antibiotics is a hallmark
of streptococcal pneumonia. In one series of military recruits with streptococcal
pneumonia, persistent fever (>7 days) occurred in 61 %. Of interest, 7I % of those
with persistent fever also had empyema.l 2 In the same series, 92% fully recovered.
A leukocytosis is often present, and in some cases may be as high as 20-
30,000. Anemia occurs in one third of patients. 13 H yponatremia and evidence of
coagulopathies or dis semina ted intravascular coagulation (DIC) may be present
but are not characteristic. Hypoxia is common.
Chest radiograph may show either an interstitial bronchopneumonia or
lobular consolidation. Lobar consolidation is rare. Infiltrates may be patchy or
homogeneous. They are usually segmental. The radiographic appearance of
streptococcal pneumonia is very similar to staphylococcal pneumonia. The two
may be very difficult to distinguish radiographically. Lung abscess or a pleural
effusion mayaIso be seen. The infiltrates are often bilateral in occurrence and
involve dependent lung portions.
7. COMPLICATIONS
The most common complications are empyema and abscess formation.

Cavitation with abscess formation is common in patients with tissue invasion
(lobular pneumonia), who do not succumb in the first few days of illness. Direct
extension may result in purulent pericarditis and mediastinitis. Miller 7 in 1918
describes pericarditis in 35% of military recruits with streptococcal pneumonia
and empyema. Bronchiectasis and pleural thickening are late complications. Non-
suppurative complications are unusual in streptococcal pneumonia and appear to
be limited to glomerulonephritis. 24 Prior to the use of antibiories pneumothorax
and bronchopleural fistula often complicated streptococcal pneumonia.
The presence of a pleural effusion, together with its rapid accumulation,
early in the course of illness is characteristic of streptococcal pneumonia. Keefer8
reports empyema in 29% of patients with streptococcal pneumonia. However,
empyema may be even more common in younger patients. In Basiliere's12 series of
streptococcal pneumonia in military recruits, 54 of 95 patients developed em-
pyema. Chest tube treatment appears to shorten hospital stay compared to ne edle
aspiration.l 2 Residual pleural thickening may develop in 15-20% of patients.
8. TREATMENT
Because of the reemergence of severe illness caused by beta-hemolytic

streptococci, interest in the susceptibility of this organism has also been re-
kindled. The organism's susceptibility is rarely tested in the routine clinical
laboratory. To date virtually all isolates have been susceptible to penicillin and
most cephalosporins. In penicillin-allergie patients, erythromycin has tradi-
tionally been used since used S. pyogenes has also traditionally been susceptible to
erythromycin and clindamycin. In 1992, however, Seppala et al. 25 reported that
coincident with a tripling of erythromycin use in Finland, "the frequency of
resistance to erythromycin in group A streptococci from blood cultures increased
from 4% in 1988 to 24% in 1990." There was also an increase in resistance from
7 to 20% and from 11 to 31% in S. pyogenes isolated from throat swabs and pus,
respectively. In addition, they noted nine of 19 patients treated with erythromycin
and harboring erythromycin-resistant isolates failed therapy compared to 1 of
26 for those patients with susceptible isolates (p = 0.008). This is of obvious
concern and may mean that laboratories may need to monitor the susceptibility of
S. pyogenes, especially if penicillin is not used. One would also suspect that the
newer macrolides would also be inactive against these resistant beta-hemolytic
streptococci.
Therapy should consist of appropriate antibiotics (penicillin G, 12-20 mil-
lion units/day) for a duration that is commensurate with the clinical and radio-
graphie resolution of the pneumonia, often for 10-14 days. If empyema is present
it should be drained. Other supportive measures are also required as clinically
indicated.
Preantibiotic mortality was 50%. Mortality is now less than 10% with proper
antibiotic treatment.I 2 Full recovery with radiographie resolution has been re-
ported by others as well. lO
REFERENCES
1. Faclam, R. R., and Washington, J. A., 1991, Streptoeoeeal and related eatalase-negative
Gram-positive coeei, in: Manual of Clinieal Mierobiology, Fifth Edition (A. Balows, W. J.
Hausler, K. L. Jerrman, H. D. Isenberg, and H. J. Shadomy, eds.), Washington, D.C., pp.
238-257.
2. Sneath, P. H. A., Mair, N. S., Sharpe, M. E., and Holt, J. G. (eds.) 1986, Bergey's Manual of
Systematie Baeteriology, Vol 2, Williams & Wilkins, Baltimore, pp. 1043-1063.
3. Laneefield, R. C., 1933, A serologie differentiation of human and other groups of hemolytic
streptocoeci,J Exp. Med. 57:571-595.
4. Hardie, J. M., 1986, Genus Streptococcus Rosenbaeh 1884, in: Bergey's Manual of Systematie
Bacteriology, Vol2 (P. H. A. Sneath, N. S. Mair, M. E. Sharpe, andJ. G. Holt, eds.) Williams &
Wilkins, Baltimore.
5. Fischetti, V. A., 1991, Streptocoecal M proteins, Sei. Amer. 264:58-65.
6. MacCallum, W. G., 1919, The pathology of the pneumonia in the United States Army camps
during the winter of 1917-18, in Monograph No. 10, The RockefeIler Institute for Medical
Research, New York City, pp. 1-142.
7. Miller,]. L., and Lusk, F. B., 1918, Empyema at Camp Dodge, Med. Clin. North Am. 2:
537-542.
8. Keefer, C. 5., Ingelfinger, F.]., and Spink, W. w., 1939, Significance of hemolytic strepto-
coccic bacteremia, Areh. Intern. Med. 60:1084-1097.
62 RICHARD H. GOLDSTEIN and ELLIE j.C. GOLDSTEIN
9. Keefer, C. S., Rantz, L. A., and Rammelkamp, C. H., 1941, Hemolytic streptococcal
pneumonia and empyema: A study of 55 cases with special reference to treatment, Ann.
Intern. Med. 14:1533-1550.
10. MacFariand, j. E., 1946,) Iowa State Med. Soe. 36:481-484.
11. Welch, C. C., Tombridge, T. L., Baker, W. j., and Kinney, R. j., 1961, Beta-hemolytic
streptococcal pneumonia: Report of an outbreak in a military population, Am.) Med. Sei.
242:157-165.
12. Basiliere, j. L., Bistrong, H. w., and Spence, W. F., 1968, Streptococcal pneumonia, recent
outbreaks in military recruit populations, Am.) Med. 44:580-589.
13. Murray, j. F., and Nadel, j. A. (eds.), 1988, Textbook o[ Respiratory Medicine, W.B. Saunders,
New York.
14. Huxley, E. j., Viroslav, j., Gray, W. R., and Pierce, A. K., 1978, Pharyngeal aspiration in
normal adults and patients with depressed consciousness, Am. ) Med. 64:564-568.
15. Stevens, D. L., 1991, Invasive group A streptococcus infections, Clin. Infoct. Dis. 14:2-13.
16. Dajani, A. S., 1991, Current status of nonsuppurative complications of group A streptococci,
Pediatr. Infoct. Dis.) 10:S25-S27.
17. Stevens, D. L., Tanner, M. H., Winship, j., Swarts, R, Ries, K. M., Schlievert, P. M., and
Kaplan, E., 1989, Severe group A streptococcal infections associated with a toxic-shock-like
syndrome and scarlet fever toxin A, N. Engl.] Med. 321:1-7.
18. Schwartz, B., Facklam, R. R, and Breiman, R. F., 1990, Changing epidemiology of Group A
streptococcal infections in the USA, Laneet 336:1167-1170.
19. Gworzewska, E. W. A., and Colman, G., 1988, Changes in the pattern of infection caused
by Streptococcus pyogenes, Epidemiol. Infoct. 100:257-269.
20. Bisno, A. L., 1991, Group A streptococcal infections and acute rheumatic fever, N. Engl.
) Med. 325:783-791.
21. McWhinney, P. H. M., and Nathwani, D., 1991, Streptococcus group A pneumonia in an
intravenous drug misuser (IVDM), Eur. Respir.) 4:761-763.
22. Barnham, M., 1989, Invasive streptococcal infections in the era before the acquired immuno-
deficiency syndrome: A 10 years' compilation of patients with streptococcal bacteraemia in
North Yorkshire,) Infoet. 18:231-248.
23. Bibler, M. R, and Rouan, G. w., 1986, Cryptogenic group A streptococcal bacteremia:
Experience at an urban general hospital and review of the literature, Rev. Inj Dis. 8:
941-951.
24. Levinson, D. A., and Litwack, K. D., 1972, Glomerulonephritis following streptococcal
pneumonia, Chest 61:397-400.
25. Seppälä, H., Nissinen, A.,järvinen, H., Huovinen, S., Henriksson, T., Herva, E., Holm, S. E.,
jahkola, M., Katila, M.-L., Klaukka, T., Kontiainen, S., Liimatainen, 0., Oinonen, S., Passi-
Metsomaa, L., and Huovinen, P., 1992, Resistance to erythromycin in group A streptococci,
N. Engl.) Med. 326:292-297.
4
Pulmonary Infections Caused

by Haemophilus injluenzae
LORRY G. RUBIN
1. INTRODUCfION
Haemophilus irifluenzae is a gram-negative coccobacillus frequently present as

part of the normal flora of the upper respiratory tract, but it may cause a wide
variety of infections in children and adults. Under natural conditions, humans
are the only known hosts of H. injluenzae. Of fundamental importance to the
understanding of infections caused by H. injluenzae is the recognition that isolates
may be nontypable (unencapsulated) or may elaborate one of six polysaccharide
capsules designated types a through f. Encapsulated strains expressing type b
capsular polysaccharide (a polymer of ribosylribitol phosphate) cause infections
primarily in children younger than 6 years old and are responsible for most cases
of systemic infection with H. injluenzae. The most common systemic infection is
meningitis, but epiglottitis, bacteremic cellulitis, septic arthritis, and pneumonia
are also frequently seen. Nontypable H. injluenzae are commonly resident flora in
the pharynx ofboth children and adults. Although they are not usually the cause
of infection with bacteremia, they are important etiologic agents of otitis media,
sinusitis, bronchitis, pneumonia, conjunctivitis, and puerperal infection. Both
encapsulated and unencapsulated H. injluenzae may cause pulmonary infection.
H. injluenzae was first described in 1883 as the Kochs-Weeks bacillus (H.
aegyptius) responsible for purulent conjunctivitis.I Based on DNA homology
LORRY G. RUBIN' Department of Pediatrics, Schneider Children's Hospital of Long Island

Jewish Medical Center, Long Island Campus for the Albert Einstein College of Medicine, New
Hyde Park, New York 11042. .
Pulmonary Inftctions and Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
63
64 LORRY G. RUBIN
studies, H. aegyptius is no longer eonsidered to be a speeies separate from H.

inJluenzae and is more properly called H. inJluenzae biogroup aegyptius. 2 During
the 1892 influenza pandemie, Pfeiffer mistakenly thought that H. injluenzae was
the cause. This association with the clinical disease, influenza, gave the species,
injluenzae, its name. In the early 1930s, Pittman recognized the importance of H.
injluenzae as a cause of serious infection. 3 She described "rough," unencapsulated
strains and the six capsular types. She recognized the important fact that type b
strains were the most significant cause of meningitis and other invasive infections.
2. THE ORGANISM: MICROBIOLOGY, TYPING, AND

VIRULENCE DETERMINANTS4
H. injluenzae are gram-negative pleomorphic bacilli that appear as cocco-

bacillary forms in Gram's stained smears prepared from clinical specimens. H.
injluenzae are considered to be fastidious organisms because they do not grow on
sheep blood agar media, and recovery of isolates from clinical specimens is
enhanced by incubation in an environment with a 5 to 10% concentration ofC02 •
Members of the genus H aemophilus are facultative anaerobes. H. injluenzae require
the supplemental growth factors, X factor (hemin), and the heat-Iabile V factor
nicotinamide adenine dinucleotide (NAD). These faetors may be supplied by
inoculating the clinical sam pie on blood agar media and cross-inoculating with a
Staphylococcus aureus strain. The S. aureus strain provides V factor and allows H.
injluenzae to groW as satellite eolonies. X and V faetors are eonveniently supplied
by chocolate agar, a medium prepared by heating blood added to an agar base to
80°C to release the X and V factors from red blood eells and inactivate V-faetor
inhibitors. This requirement for growth faetors present in blood gives the genus
name, Haemophilus from haemo (Greek for "blood") and philos (Greek for "loving").
Growth as small colonies of approximately 1 mm on chocolate agar media is
apparent 18 to 24 hours after inoculation. To facilitate recovery of H. injluenzae
from respiratory tract specimens containing other bacteria, antibiotics sueh as
bacitraein, vancomycin, and clindamycin may be incorporated into the medium. 5
Strains of H. injluenzae may be typed using several methods. Serotyping
using eapsular-type specific antisera in a slide agglutination assay will distinguish
type b (or eapsular types a, e, d, e, and f) from nontypable, unencapsulated
strains. H. injluenzae may be divided into five biotypes, I-V, on the basis of three
biochemical reaetions: indole, urease, and ornithine decarboxylase. For epide-
miologie purposes type b or nontypable strains may be further typed by classify-
ing the pattern of eleetrophoretie mobility of two major outer membrane proteins
following sodium dodeeylsulfate polyacrylamide gel electrophoresis. 6 ,7 Evolution-
ary genetie relationships among strains, that is, their clonality, are best evaluated
by studying eleetrophoretie mobility patterns of eytoplasmie enzymes. This
teehnique, based on differenees in enzyme mobility patterns beeause of polymor-
phisms at loci encoding these enzymes, is called multilocus enzyme eleetro-
phoresis. Using this technique, it has been eoncluded that (1) uneneapsulated
INFECTIONS CAUSED BY HAEMOPHILUS INFLUENZAE 65
strains are not simply capsule-deficient mutants of encapsulated strains, (2) about
a dozen clones account for all type-b isolates worldwide, and (3) unencapsulated
strains are much more diverse than type-b isolates. 8 ,9 Finally, strains may be
compared by the pattern ofbands observed after digestion ofbacterial DNA with
restriction endonucleases and electrophoresis ("DNA fingerprints"). Groeneveld
et al. studied serial H. injluenzae isolates from patients with chronic obstructive
pulmonary disease and found pairs of isolates (recovered several months apart)
with identical DNA fingerprints but dissimilar outer-membrane protein electro-
phoretic patterns lO from seven patients. These findings suggested that the major
outer membrane proteins are subject to change in vivo. The authors speculated
that these changes may enable H. injluenzae to escape immunological defense
mechanisms.
H. injluenzae may contain three main classes of surface antigens: capsular
polysaccharide, lipooligosaccharide, and outer-membrane proteins. Capsules are
composed of polysaccharides; the type-b capsule is a phosphated polymer of
ribose and ribitol. Compelling clinical evidence and virulence data using an
infant rat model of infection and genetically related strains that elaborate or are
negative for capsular polysaccharide demonstrate that the type-b capsule is a
virulence determinant for this organism. 3 ,ll The type-b capsule is present on the
bacterial cell surface and is elaborated into growth media and body fluids.
Detection of capsular antigen in body fluids such as cerebrospinal fluid and urine
provides the basis for rapid diagnosis of infection. The capsular polysaccharide
is also important because it is the chief constituent of vaccines to prevent infec-
tions caused by H. injluenzae type b.
H. injluenzae lipooligosaccharide differs from the lipopolysaccharide of
Enterobacteriaceae such as Salmonella in that they lack the long repetitive side
chain characteristic of Enterobacteriaceae. Lipooligosaccharide molecules of H.
inJluenzae undergo phase variation, frequent and reversible spontaneous changes
in the phenotype and oligosaccharide epitopes, as a result of sequential changes
in the composition and assembly of sugars.l 2 These nontoxic oligosaccharide
moieties elicit antibodies although adefinite role for antilipooligosaccharide
antibodies in protection against infection has not been established. Like all gram-
negative bacteria, H. inJluenzae lipooligosaccharide contains lipid A and it is not
surprising that lipooligosaccharide from H. inJluenzae demonstrates biologic
activity characteristic of endotoxins.
The outer membrane contains about 20 proteins with four to six proteins
accounting for most of the protein content. 13 Each strain has two major pro-
teins in the 32-42 kDa range. The one with molecular mass of 36-42 kDa
functions as a porin protein. The functions of the other major outer membrane
proteins are unknown. There is antigenic heterogeneity of outer membrane
proteins among various strains. Certain proteins, most notably a 16-kDa
peptidoglycan-associated protein designated P6 or g, express epitopes that are
widely shared among type band nontypable strains. Pili (or fimbriae) are
filamentous surface projections assembled from 24 kDa polypeptide subunits
called pilin. Pili from H. injluenzae type-b strains mediate hemagglutination, may
66 LORRY G. RUBIN
mediate binding of the bacteria to buccal epithelial cells, and contribute to the
ability of the strain to colonize the nasopharynx of monkeys (see Section 2.2). In
contrast, these hemagglutinating fimbriae do not appear to be important in the
epithelial cell binding and hemagglutination of nontypable H. injluenzae.1 4 H.
inJluenzae elaborate two extracellular proteins that may contribute to virulence:
IgA 1 protease and haemocin. H. injluenzae is one of a group of pathogenic
bacterial species that produce an extracellular enzyme, IgA 1 protease. This
protease specifically cleaves the heavy chain of human IgAl' a component of the
local immune system of the respiratory mucosa.1 5 Haemocin is a bacteriocin, a
protein elaborated by most type-b strains of H. inJluenzae, with bactericidal
activity against other nontypable and non-type b encapsulated H. inJluenzae,
other Haemophilus spp., and some other gram-negative bacteria. 16
3. EPIDEMIOLOGY OF COLONIZATION OF H. INFLUENZAE
The initial step in the pathogenesis of pulmonary infection with H. inJluenzae

is colonization. H. inJluenzae colonizes the upper respiratory tract, specifically the
oro- and nasopharynx. H. inJluenzae is probably acquired by inhalation or inges-
tion of infected respiratory tract secretions from a colo'nized individual. The
prevalence and dynamics of colonization have been studied for both type-b and
unencapsulated H. inJluenzae. Nontypable (unencapsulated) H. inJluenzae have
been recovered from throat swabs from 20 to 80% of individuals. The prevalence
of colonization is 50 to 80% if selective media are used to retard growth of
pharyngeal flora other than H. inJluenzae. Spinoza et al. compared serial phar-
yngeal isolates of nontypable H. inJluenzae isolates from three children obtained
over a 5- to 7-year period of time by biotyping and outer-membrane protein
subtyping. 17 They found a monthly rate of isolation of 20 to 40% using nonselec-
tive medium. Other studies have found rates of 30 to greater than 80%.1 8
Children were usually colonized by astrain of nontypable H. inJluenzae for several
months. Loss and acquisition of strains were common events.
In contrast, the prevalence of pharyngeal colonization with H. inJluenzae
type-b strains is less than 2% in unselected children younger than 5 years 0ld.1 9
Carriage of type b H. inJluenzae strains is rare in adults who have not been
exposed to a colonized child. Following occurrence of a case of invasive H.
inJluenzae b disease, other young children residing in the same household or
attending the same day care classroom are at increased risk for developing
invasive disease. 20 Being a member of a household or working with children in a
child care setting where a case of invasive disease has occurred is apparently a risk
factor for colonization with H. inJluenzae b. In a study of children and adults
attending or working in day care centers without recent cases of H. inJluenzae
type b invasive disease, carriage rates varied over time and ranged from 0 to 27%
with 10% of all cultures yielding H. inJluenzae type b. 21 Colonization with type-b
strains tended to persist for several months. Day care center contacts of a child
with invasive H. injluenzae type-b disease tend to have a high but variable rate of
carriage. The colonization rate varies from 20 to 75% in young children and 0 to
25% in adult contacts. 22 Household members of an index patient with H.
injluenzae b disease are frequently colonized. Approximately 50% of young
children and 15% of adults in households of an index case have been found to be
colonized. 23
These data were obtained before the implementation of universal immuniza-
tion with H. injluenzae type-b conjugate vaccines. Although limited, available data
indicate that individuals vaccinated with a conjugate vaccine have a lower preva-
lence of colonization with H. injluenzae type b. 24,25
4. PATHOGENESIS OF COLONIZATION AND INFECTION WITH

H. INFLUENZAE
The pathogenesis of colonization has been studied using infant rat and
infant monkey animal models of invasive infection, mouse and rat models of
intratracheal installation of organisms, organ culture of human or monkey
respiratory tissues such as adenoids or nasal turbinates, and suspensions of
respiratory tract epithelial cells. The series of events that result in colonization and
subsequent pulmonary infection are incompletely understood. It is generally
accepted that bacterial attachment to the mucosal surface, that is, to epithelial
cells or mucus overlying the epithelial cells, is aprerequisite for colonization.
Using suspensions ofhuman epithelial cells, unencapsulated H. injluenzae adhere
to a greater extent than type-b encapsulated bacteria, and fimbriated (piliated)
type-b cells exhibit more binding than nonfimbriated cells. The receptor for
fimbriae on human oropharyngeal epithelial cells has been identified as a sialic
acid-containing ganglioside, sialyllactosylceramide (GM3).26 Using a human
epithelial cellline derived from human conjunctival cells, St. Gerne and Falkow
found that adherence of a nontypable H. inJluenzae strain occurred independently
of piliation and that adherence requires viable bacteria capable of de novo protein
synthesis. 27
Studies in which H. injluenzae were incubated with excised human naso-
pharyngeal tissue (turbinates or adenoids) in organ culture have yielded much
information concerning the initial events in the interaction of bacteria with
epithelial cell surfaces. 28-31 Following inoculation of the organ culture with
bacteria, subsequent events were studied by combinations oflight microscopy and
transmission and scanning electron microscopy as well as by bacterial culture and
by measurements of ciliary beat frequency.
Read et al. recently described a human nasopharyngeal organ culture in
which agarose was used to seal the nonluminal side and edges of the specimen to
study the interaction of bacteria with the naturally exposed epithelial cell sur-
face. 28 Bacteria multiplied in the absence of adherence to epithelial cells. Bacteria
caused damage to the epithelial cell surface which was evident morphologically
and by noting areduction in the ciliary beat frequency. This damage may be
mediated, at least in part, by lipooligosaccharide. 32 .33 The earliest attachment of
68 LORRY G. RUBIN
bacteria was to mucus secretions rather than epithelial cells. There was then loss
of ciliary activity and sloughing of ciliated epithelial cells, events noted more
prominently following inoculation of nontypable than type-b organisms. Nontyp-
able bacteria subsequently adhered to nonciliated and basal epithelial cells but
only in areas with epithelial cell damage. Type-b bacteria appeared to be embed-
ded in a gel-like matrix elaborated by these bacteria.
In contrast to the enhanced adherence of fimbriated organisms to epithelial
cells in suspension, fimbriated bacteria did not exhibit greater adherence to
epithelial cells in organ culture or cause more epithelial damage than genetically
related fimbriae-negative cells,3o Because fimbriated bacteria appeared to ad-
here to mucus shortly after inoculation, the authors postulated that fimbriae may
confer a colonization advantage by facilitating adherence to mucus rather than by
mediating binding to epithelial cells. These findings are supported by a study by
Weber et al. in which a fimbriated H. injluenzae type-b strain was compared with a
genetically engineered nonfimbriated variant for adherence to suspensions of
human buccal epithelial cells and colonization of yearling rhesus monkeys.31 The
fimbriae-negative mutant exhibited less binding to epithelial cells and showed a
lower density of colonization in monkeys. However, all colonies recovered from
nasal swabs were fimbriae-negative colonies regardless of the phenotype inocu-
lated. These findings imply that the role of fimbriae is limited to the initiation of
colonization.
We have studied the mechanism of colonization by inoculating infant rats
intranasally with equal numbers of two variants (differing in an antibiotic-
resistance marker) of astrain of H. injluenzae type b (with an inoculum size near
the infectious dose-50 for colonization).34 When nasopharyngeal washes were
cultured several days later, it was evident that most rats became colonized with
primarily one or the other variant. These findings were consistent with the
hypo thesis that colonization may be caused by survival and multiplication of a
single organism from the infecting inoculum.
Histamine, a low-molecular-weight substance that causes bronchoconstric-
tion and is released from mast cells, may be synthesized by H. injluenzae. 35
Histamine also stimulates mucus glands and causes increased permeability of the
bronchial epithelium. Thus, if histamine is produced by colonizing H. injluenzae,
conditions induced by histamine may allow increased bacterial multiplication by
the acquisition of growth factors from the circulation and possibly facilitate
invasion by H. injluenzae.
Viral infections of the respiratory tract may facilitate colonization and infec-
tion with H. injluenzae. Viruses were recovered from upper respiratory tract
specimens in 23% of 53 children with meningitis caused by H. injluenzae type b. 36
Michaels and Myerowitz have shown that rats experimentally infected with
influenza A or parainfluenza prior to inoculation with H. injluenzae type b
develop a markedly higher density of nasal colonization with H. injluenzae than
the control group,37 Furthermore, they found that the intranasal dose of H.
injluenzae b necessary to produce bacteremia and meningitis in 5-day-old rats was
reduced IOO-fold if the animals were previously infected with influenza virus,38
Viral potentiation probably occurs by induction of mucosal inftammation by the

viral infection. This results in a higher density of colonization and bloodstream
invasion, possibly caused by growth factors for H. injluenzae released as a result
of the inftammation. 39 Other potential mechanisms such as the inftuence of the
viral infection on immune function against H. injluenzae need to be evaluated.
Development of bacteremia requires bacterial invasion beyond the epithelial
cell surface. The two possible routes ofbacterial invasion are passage between or
into epithelial cells. Stephens and Farley observed a breakdown of epithelial cell
tight junctions following inoculation of type b or nontypable H. injluenzae into
organ cultures. 4o They interpreted these findings as indicating that H. injluenzae
may invade the bloodstream by intercellular passage of bacteria, although they
observed an occasional bacterium with acelI. St. Gerne documented entry of a
small proportion of adherent nontypable H. injluenzae cells into epithelial cells. 27
This process was inhibited by cytochalasin D and colchicine, inhibitors of epithe-
lial cell microfilament and microtubule formation respectively. Following bacterial
invasion beyond the epithelial cell barrier, bacteria might gain entry to the
bloodstream by direct invasion of submucosal blood vessels or by entry into the
lymphatic system with subsequent entry into the systemic circulation. In the
infant rat model of infection with H. injluenzae type b, bacteria can be detected
in the blood shordy after intranasal inoculation. U sing this model we cultured
blood and lymph nodes that drain the lymphatic vessels underlying the na-
sopharynx to recover H. injluenzae after intranasal inoculation. 41 Our results were
consistent with H. injluenzae type-b entry into the bloodstream by direct invasion
rather than via lymphatic vessels and regionallymph nodes.
H. injluenzae may reach the lung to cause pneumonia by either the blood-
stream or contiguous spread (or aspiration) from the upper respiratory tract.
Although the relative importance of these two mechanisms is unknown, contig-
uous spread is likely to be a more important route for nontypable H. injluenzae
pneumonia and bronchitis because such organisms are not commonly recovered
in blood cultures of infected patients. In contrast, many children with pneumonia
caused by H. injluenzae type b have positive blood cultures. Recovery of H.
injluenzae from the blood of a child with pneumonia is considered documentation
that H. injluenzae played an etiologic role in the pneumonia. It remains unclear,
however, if pneumonia is initiated by bloodstream spread of bacteria to the lung
parenchyma or if bacteremia occurs secondary from an established pulmonary
focus of infection.
Several investigators have developed murine or rat models of experimental
pulmonary infection by direct intratracheal administration of organisms in a
liquid suspension resulting in an acute infection 42 or suspended organisms in
semisolid agar to establish a chronic infection. 43 These models are useful to study
the host mechanisms important for pulmonary clearance mechanisms and the
host responses to infection. 44 For example, nontypable and type-b strains are
cleared at the same rate from the lungs of mice. Neutrophils are recruited to the
lung by approximately 6 hr after inoculation, an event critical for eradication of
bacteria. 42 Oral and intratracheal immunization of rats with killed whole bacteria
70 LORRY G. RUBIN
has shown that antigenic stimulation of gut-associated lymphoid tissue can result
in improved clearance of nontypable H. inJluenzae from the lungs of rats. 45
A paradigm for the pathogenesis of pulmonary infection caused by H.
inJluenzae has been proposed in arecent review by Moxon and Wilson 46 and is
modified as folIows. H. inJluenzae colonize the upper respiratory tract by adher-
ing to mucus. They frequently may disseminate through the nasopharynx and
lower respiratory tract, but local host defenses such as mucociliary clearance and
local immunoglobulins including IgA prevent establishment of infection and
disease. Local inflammation induced by viral infections and bacterial products
such as lipooligosaccharide facilitate bacterial multiplication. With an increased
density of bacteria at the mucosal surface and in the presence of risk factors such
as impaired mucociliary clearance or inflammation associated with viral infection
or smoking, bacterial multiplication occurs in the lower respiratory tract. Epithe-
lial cell damage and interruption of the interepithelial cell tight junctions caused
by lipooligosaccharide or other toxins allow for H. inJluenzae adherence to
nonciliated and exposed basal epithelial cells and for invasion of epithelial cells.
Host responses such as the elaboration of cytokines by local macrophages may
result in an enhanced inflammatory response (as has been noted in H. inJluenzae b
meningitis and arthritis 47 ) resulting in more pronounced local and systemic
symptoms of infection.
5.IMMUNITY
The basis for protection against invasive infections caused by type-b H.

inJluenzae has been investigated intensively. The age-related occurrence of infec-
tions caused by H. inJluenzae b (cases generally limited to ages 3 months through 5
years) correlated inversely with serum concentrations of anticapsular antibodies
to the type b capsule. Sufficient concentrations of serum anticapsular antibodies
correlate with protection against disease. 48 ,49 In the case of H. inJluenzae type b,
anticapsular antibodies function by opsonizing bacteria for phagocytosis and kill-
ing by professional phagocytes. 50 ,51 Antibody-mediated, complement-dependent
bacterial lysis appears to be a less important host defense mechanism. Most
individuals older than 5 years of age have sufficient serum concentrations of
anticapsular antibodies to afford protection against invasive disease. Immuniza-
tion of infants and young children with currently licensed vaccines consisting of
capsular polysaccharide covalently bound to a protein carrier against H. inJlu-
enzae b also induce sufficient serum anticapsular antibody concentrations to
protect against invasive disease. 49 Recent evidence suggests that immunization of
young children with these vaccines results in a diminished rate of pharyngeal
colonization with type b strains. 24 This finding combined with the serum antibody
levels achieved after vaccination makes it quite likely that universal implementa-
tion of immunization with this vaccine should result in prevention of pneumonia
caused by type-b strains of H. inJluenzae. Adults who develop bacteremic pneu-
INFECfIONS CAUSED BY HAEMOPHILUS INFLUENZAE 71
monia caused by H. injluenzae b most likely have low serum anticapsular antibody
concentrations as a result of immune deficiency or waning immunity.
Antibodies to epitopes of some cell surface-exposed outer membrane pro-
teins may also contribute to protection against type-b strains although many
epitopes are apparently strain-specific. The role of antibodies to lipooligosac-
charide is less clear.
For nontypable H. inj1uenzae, less is known about the mechanisms of
antibody-mediated bacterial clearance. Infection frequently induces serum anti-
bodies with bactericidal and/or opsonic activity against the infecting strain. 13
Surface-exposed antigens on outer-membrane proteins and lipooligosaccharides
are being evaluated for reactivity with antibodies that mediate bacteriallysis by
complement and/or phagocytosisP Such studies are complicated by antigenic
variation in the expression of oligosaccharide epitopes of lipooligosaccharides
and interstrain variation in lipooligosaccharide antigens and the antigens ex-
pressed on major outer-membrane proteins. For example, H. injluenzae otitis
media induces serum antibodies bactericidal against the infecting strain but
inactive against a heterologous strain. 52
6. CLINICAL SYNDROMES
H. injluenzae causes a number of respiratory tract syndromes including

pneumonia with or without concomitant bacteremia, bronchitis, tracheitis, epi-
glottitis, otitis media, and sinusitis. The roles of nontypable and type b strains vary
according to the syndrome and the host and is summarized in Table I.
6.1. Pneumonia
In children younger than 6 years old, H. injluenzae type b is a common cause
ofbacterial pneumonia. 53 ,54 However, it is likely that the incidence of H. injluenzae
b pneumonia has significantly declined in the United States with the implementa-
tion of universal immunization with H. injluenzae b conjugate vaccines, because
the incidence of bacteremia and meningitis due to H. injluenzae has dramatically
declined. 5f>...57 As a cause of segmental pneumonia, it is second only to Streptococcus
pneumoniae in this age group.53 The median patient age is approximately one year
with an age distribution of cases identical to cases of meningitis caused by H.
injluenzae type b. The clinical presentation of pneumonia caused by H. injluenzae
type b is indistinguishable from that of pneumonia from S. pneumoniae. Children
present with high fever often preceded by upper respiratory symptoms and
accompanied by tachypnea and systemic toxicity. The patient may have concomi-
tant meningitis (approximately 20% of cases) or epiglottitis (approximately 12%
of cases). Laboratory evaluation reveals a leukocytosis with a polymorphonuclear
predominance. Blood cultures are generally positive. Radiographie studies of the
ehest most eommonly show a solitary segmental infiltrate, but the infiltrates may
72 LORRY G. RUBIN
TABLE I
Lower Respiratory Tract Syndromes Caused by H. inJluenme
Comparative frequency
of nontypable and
Clinical syndrome type b strains Features
Adults
Pneumonia
Without bacteremia Nontypable » type b Elderly, chronic lung disease, alcoholics
With bacteremia Nontypable = type b As above
With HIV infection Nontypable = type b May present subacutely mimicking
P. carinii
Bronchitis All nontypable Underlying lung disease
Tracheobronchitis Nontypable » type b Presentation similar to pneumonia
Pediatric
Childhood pneumonia Children age <5 years
Developed countries All type b
Developing countries Nontypable = type b Some cases of other types
Neonatal pneumonia Nontypable » type b Associated sepsis, high mortality
Epiglottitis All type b Airway obstruction life-threatening,
concomitant bacteremia
Tracheitis Nontypable > type b Preceding viral infection
be multiple. In 30% of cases, pneumonia is accompanied by a pleural effusion

although empyema requiring drainage is present in only a minority of cases.
Complete radiographie clearing occurs within two months following therapy.
Studies concerning the etiology of bacterial pneumonia in children in devel-
oping countries have yielded different results than those obtained in developed
countries. Predictably, S. pneumoniae was the most common etiology. However, in
contrast to childhood H. inJluenzae pneumonia in developed countries (almost all
caused by type b strains), a significant proportion of pneumonias in young chil-
dren in developing countries are due to nontypable H. inJluenzae, a finding often
documented by recovery of the isolate from culture of blood. For example, a
study of children in Pakistan with lower respiratory tract infection and bacteremia
found that 36% of 97 blood-culture isolates of H. inJluenzae were nontypable. 58
The clinieal and demographie features of children infected with nontypable H.
inJluenzae were indistinguishable from those infected with type-b strains. In
studies of acute lower respiratory tract infections in children in Papua New
Guinea, blood cultures and lung aspirates were used to document the eti-
ologies. 59•6o Of the blood and lung-aspirate H. influenzae isolates, 19% were type
b, 25% were other serotypes, and 56% were nontypable. Cultures of lung tissue
indicated that mixed infections with nontypable and typable H. inJluenzae or H.
inJluenzae and S. pneumoniae were common. Of H. influenzae blood-culture isolates,
two-thirds were type b, one-sixth were nontypable, and one-sixth were other sero-
types of H. inJluenzae.
Another setting in which nontypable H. injiuenzae may play an important

role in pediatric pneumonia is in the newborn infant. Nontypable H. injiuenzae,
particularly strains with a type IV biotype, are an important cause of postparturn
infection in women and their newborn infants. 61-63 Affected infants present
within a few hours to several days of birth with respiratory distress, diffuse
pulmonary infiltrates, and signs of sepsis. Blood cultures are alwayspositive, but
concomitant meningitis is infrequent. The mortality rate is at least 30% and may
be higher and the infection more frequent in infants who are premature or born
to mothers with obstetric complications. The clinical features of H. injiuenzae
pneumonia in newborns are indistinguishable from early-onset infection by
Group B streptococci (Streptococcus agalactiae) which presents with respiratory
distress, diffuse pulmonary infiltrates, and symptoms and signs of sepsis.
H. injiuenzae are recognized as an important cause of pneumonia in adults,
particularly elderly adults. Patients typicaIly present with fever, cough, and
purulent sputum, features indistinguishable from those of other bacterial pneu-
monias. WaIlace et al. 64 studied patients with pneumonia accompanied by a blood
culture positive for H. injiuenzae. They noted that blood culture isolates from
adults with pneumonia were frequently incorrectly classified as type b by a clinical
laboratory; their analysis revealed an equal number of type band nontypable
strains. Furthermore, they found a polymicrobial bacteremia in 35% of patients
with nontypable H. injiuenzae in blood culture. In a study of invasive H. injiuenzae
infections in adults, of 29 patients with pneumonia and bacteremia, the blood cu 1-
ture isolates were type b in 55%, nontypable in 42%, and type f in 4% of cases. 65
In cases of H. injluenzae type-b pneumonia in adults, predisposing factors are
recent upper respiratory tract infection, male gender, chronic obstructive pul-
monary disease, and alcoholism. 66 The radiographie studies may show either lobar
involvement or bronchopneumonia. Nontypable H. injiuenzae is a more common
cause of pneumonia in adults than type b strains although pneumonia caused by
nontypable strains is infrequently accompanied by bacteremia. Because nontyp-
able H. injiuenzae are common inhabitants of the upper respiratory tract, their
role as etiologic agents of pneumonia has been difficult to establish. Nevertheless,
the role of nontypable H. injiuenzae in pneumonia has been weIl documented in
elderly adults by recovery of the organism in pure culture from transtracheal
aspirates. 67 The presentation of these patients is similar to those with type-b H.
injiuenzae pneumonia except that patients tend to be older and more debilitated.
Although in both these presentations of H. injiuenzae pneumonia the infection is
most commonly community-acquired, it may be transmitted nosocomiaIly.68
Children and adults with human immunodeficiency virus (HIV) infection
have a much higher incidence of bacterial pneumonia than uninfected individ-
uals. 69 Many cases of H. injiuenzae pneumonia present acutely as is observed in
individuals not infected with HIV. Cases of H. injluenzae pneumonia in HIV-
infected in fants have been a result of serotype b strains. Both type-b and
nontypable strains have been recovered from HIV-infected adults with pneu-
monia. Infected patients generally have an excellent clinical response to institu-
tion of an antibiotic with activity against H. injiuenzae. However, some of the
74 LORRY G. RUBIN
pneumonias in adults have had unusual dinical features. A subacute presentation

with a gradual onset of nonproductive cough over a greater than two week period
of time and a predominance of bilateral interstitial or reticulonodular infiltrates
was noted in 12 HIV-infected adults. 70 These patients lacked the more common
symptoms of an acute onset, high fever, and purulent sputum usually seen in
adults with H. injluenzae pneumonia and their presentations were similar to those
seen in patients with Pneumocystis carinii pneumonia. In another study, 6 of 34
''AIDS-risk'' patients (mostly intravenous drug abusers) with H. injluenzae pneu-
monia (and without P. carinii) had bilateral interstitial or reticulonodular infil-
trates on ehest radiograph. 71 In four cases, the patient was coinfected with P.
carinii and H. injluenzae. These patients had a higher frequency of dyspnea and
elevated serum lactate dehydrogenase levels and a lower frequency of presenta-
tion with ehest pain.
6.2. Chronic Bronchitis Exacerbation

Nontypable H. injluenzae are frequently recovered from the sputum of adults
with acute exacerbation of chronic bronchitis and are probably responsible for
some cases of acute exacerbation. Because H. injluenzae are common inhabitants
of the pharyngeal flora in these patients, however, their role is not firmly
established. As summarized by Murphy and Apicella,13 three lines of evidence
have been examined to determine the role of nontypable H. injluenzae in chronic
bronchitis: (1) association of recovery of nontypable H. injluenzae with increased
symptoms and purulence of sputum, (2) serologie response to bacterial antigens,
and (3) impact of antibiotics with activity against H. injluenzae on the dinical
course. 72 These studies have yielded conflicting results.
6.3. Tracheitis and Tracheobronchitis

Less common pulmonary infections that may be caused by nontypable H.
injluenzae are bacterial tracheitis in children and acute febrile tracheobronchitis in
adults. Bacterial tracheitis affects children 1 month to 10 years of age. The dinical
course consists of a prodromal upper respiratory tract infection followed by fever,
stridor, and a brassy cough simulating epiglottitis or a severe case of viral
croup.73,74 Respiratory embarrassment occurs because of airway obstruction by
purulent secretions. Thick purulent secretions are noted during bronchoscopy or
after tracheal intubation. Staphylococcus aureus is the most common bacterial
species causing this infection; H. injluenzae and several streptococcal species may
cause this syndrome. Management consists of providing an artificial airway by
intubation and administration of appropriate antibiotic agents. Acute febrile
tracheobronchitis in adults as a result of H. injluenzae has been described by
Musher et al. 75 These patients had illnesses dinieally indistinguishable from
pneumonia yet had negative radiographs of the ehest. Beeause 61% of these
patients had ehronie obstruetive pulmonary disease, patients with aeute febrile
INFECfIONS CAUSED BY HAEMOPHILUS INFLUENZAE 75
tracheobronchitis may represent the more severe end of a spectrum of pulmonary

exacerbations in patients with chronic lung disease.
6.4. Epiglottitis
Epiglottitis is a distinctive clinical syndrome with a peak incidence in children
2 to 5 years old although cases occur occasionally in adults. Patients present with
high fever, a toxic appearance, and inspiratory stridor caused by airway obstruc-
tion as a result of edema of the epiglottitis and supraglottic structures. The child is
apprehensive and prefers to be in a sitting position. Although bacteremia with H.
injluenzae type b is present in nearly all cases, the most critical aspect of manage-
ment is securing patency of the airway by endotracheal intubation (or tracheos-
tomy if necessary) until the inflammation has subsided. The incidence of this
infection has plummeted with the implementation of immunization with H.
injluenzae b vaccines. 76
7. DIAGNOSIS
Pulmonary infections of H. injluenzae type bare frequently diagnosed by

recovery of this organism from normally sterile body fluids such as blood or
pleural fluid. This serves to definitively identify H. injluenzae b as the cause of the
clinical syndrome. A requirement for a positive blood or pleural fluid culture,
however, likely results in an underestimate of the incidence of pulmonary infec-
tion by this pathogen. Recovery of H. injluenzae from expectorated sputum is a
much less satisfactory means of assigning an etiologic role for H. injluenzae
because H. inJluenzae, especially nontypable strains, are frequently resident
flora in the oropharynx. Percutaneous aspirates of the lungs of pediatric patients
and transtracheal aspiration in adults have been performed to obtain specimens
free of contamination by resident flora of the upper respiratory tract, thereby
circumventing this difficulty. Because of the invasive nature of these procedures,
they are not used routinely as diagnostic aids.
Antigen detection is another method that may be used to diagnose infections
caused by H. injluenzae b. Capsular polysaccharide is present on the surface of the
organism and is also elaborated into body fluids (and growth media in the
laboratory). Capsular antigens may circulate in the blood and be cleared by the
kidney and be present in the urine. Using anticapsular antibodies to detect
capsular antigen in the urine by techniques such as latex agglutination, staphylo-
coccal coagglutination, or enzyme immunoassay (and, in the past, counter-
immunoelectrophoresis) has been successfully used to assign an etiology to cases
of pneumonia. 77- 79 Because capsular antigen also appears in the urine after
vaccination with H. injluenzae b vaccines, recent immunization may confound
interpretation of a positive urinary antigen assay.80 Similar tests have not been
devised for diagnosis of infections caused by nontypable strains of H. injluenzae.
76 LORRY G. RUBIN
8. ANTIBIOTIC THERAPY
Although many currently available antibiotics have activity against strains of

H. inJluenzae, ampicillin and amoxicillin have had the longest histories as safe and
effective agents for treatment of respiratory tract infections caused by H. inJlu-
enzae. In 1972, however, ampicillin-resistant strains of H. inJluenzae b were recov-
ered from children with meningitis.8 1 The mechanism of ampicillin resistance
was inactivation of ampicillin by beta-lactam ase produced by the infecting strain.
This enzyme is encoded by a plasmid and is of the TEM type, a type which had
previously been detected and characterized in Escherichia coli. The prevalence of
beta-Iactamase-producing strains has been increasing since that time and has also
been observed among nontypable H. inJluenzae. 81 In 1981 we described an
additional beta-Iactamase elaborated by astrain of H. inJluenzae from a child with
meningitis.8 2 This novel beta-Iactamase, designated ROB-l, was not detected by
the chromogenic cephalosporin method used in the clinicallaboratory for detect-
ing beta-Iactamase; as a result the strain was initially considered to be ampicillin
susceptible. Although the prevalence varies widely by geographic area, the cur-
rent prevalence rate of all beta-Iactamase-producing H. inJluenzae is approx-
imately 30% among type-b strains and 15% among nontypable strains isolated in
the United States.81 Approximately 90% of beta-Iactamase-producing strains
produce TEM beta-Iactamase and 10% produce ROB-l.8 3
Strains of H. inJluenzae may exhibit ampicillin resistance without beta-
lactamase production. The action of ampicillin is initiated by binding of the
antibiotic to cytoplasmic membrane proteins called penicillin-binding proteins.
Beta-Iactamase negative, ampicillin-resistant strains have been isolated and have
been found to have alterations of their penicillin-binding proteins resulting in a
lower affinity for penicillin. 81 Such strains have been described more frequently in
nontypable than in type-b isolates of H. inJluenzae. The prevalence ofthis type of
ampicillin resistance is currently less than 1% of strains.
The treatment of severe or invasive pulmonary infections requires hospitaliz-
ation and treatment with a parenterally administered antibiotic with activity
against ampicillin-resistant H. inJluenzae. Effective agents include some second
and all third generation cephalosporin agents as weIl as combination agents
containing a penicillin derivative and a beta-Iactamase inhibitor (ticarcillin-
clavulanic acid; ampicillin-sulbactam). In patients not requiring parenteral anti-
biotics, one of several orally administered antibiotics with activity against both H.
inJluenzae and other bacterial species that commonly cause respiratory tract
infections may be administered. In addition to amoxicillin, these include
trimethoprim-sulfamethoxazole, tetracycline, cefaclor, amoxicillin-clavulanic
acid, and in the past, chloramphenicol. Newer orally administered antibiotics with
good activity against H. inJluenzae include the fluoroquinolone antibiotic cipro-
floxacin, cefuroxime axetil, cefixime, cefpodoxime, the new azalide azithromycin
and the new macrolide clarithromycin. The latter is active against H. inJluenzae
when combined with its 14-hydroxy clarithromycin metabolite. First-generation
INFECTIONS CAUSED BV HAEMOPHILUS INFLUENZAE 77
cephalosporin antibiotics such as cefazolin or cephalexin lack adequate activity

againstH. injluenzae. Second-generation cephalosporin antibiotics have enhanced
H. injluenzae activity compared to first generation cephalosporin agents, but may
not completely resist inactivation by H. injluenzae beta-Iactamases. Chloramphenicol-
resistance, caused by production of the chloramphenicol-inactivating enzyme
chloramphenicol acetyltransferase, has been described in H. injluenzae and is
encoded on a plasmid. Strains resistant to both ampicillin and chloramphenicol
have been reported. Resistance is occasionally encountered against trimethoprim-
sulfamethoxazole, tetracycline, and rifampin, antibiotics that are generally active
against H. injluenzae. 8 ! To date, all H. injluenzae remain susceptible to third-
generation cephalosporin antibiotics such as cefotaxime and ceftriaxone and to
fluoroquinolone antibiotics.
Rifampin is the most effective antibiotic for eradication of pharyngeal
colonization with H. injluenzae band is indicated for all members of the house-
hold of a case of invasive disease caused by H. injluenzae b in which there is a child
less than 5 years of age. 83 H. injluenzae are not readily isolated from the upper
respiratory tract during antibiotic therapy, but soon thereafter may readily be
recovered on throat or nasopharyngeal swabs. This may be partly explained by
the findings of Gilsdorf and Jesperson that exposure of astrain of H. injluenzae
b to subinhibitory concentrations of one of several antibiotics resulted in a
decrease in adherence to human buccal epithelial cells. 84
9. PREVENTION
Efforts to prevent pulmonary infections caused by H. injluenzae can be

directed at immunization with H. injluenzae antigens that induce protective
immunity, immunization with putative viral vaccines to avoid viral infections that
predispose to bacterial pneumonia, avoidance of conditions such as exposure to
tobacco smoke and viral infections that predispose to the infection, and preven-
tion of secondary infections caused by H. injluenzae b using chemoprophylaxis
with rifampin. It is quite likely that respiratory tract infections caused by H.
inJluenzae type bare preventable by immunization with current conjugate vaccines
against this pathogen. 48 These vaccines, which consist of polysaccharide or oligo-
saccharide capsular antigen covalently bound to a protein carrier, induce concen-
trations of serum anticapsular antibodies sufficient to protect individuals of all
ages against infection by this pathogen. The incidences of meningitis and epiglot-
titis have declined since the introduction of H. injluenzae b vaccines. Because H.
injluenzae b pneumonia is relatively difficult to diagnose, there are no data on the
effect of vaccination on the incidence of pneumonia. It is likely that the incidence
of pneumonia in children has declined. This is particularly likely if the prevalence
of colonization is lowered by vaccination. 24 ,25 Because only a minority of cases
of H. injluenzae otitis media and sinusitis are a result of type-b strains, this vaccine
would not be expected to have a noticeable effect on the incidence of these
78 LORRY G. RUBIN
infections. Secondary cases of H. injluenzae b invasive diseases including pneu-

monia can be prevented by administration of rifam pin to the entire household of
the index case, which effectively eliminates carriage of this pathogen.
Prevention of respiratory tract infections caused by nontypable H. injluenzae
willlikely prove to be more difficult. In a double-blind, placebo-controlled study,
an orally administered vaccine consisting of formalin-killed whole nontypable H.
injluenzae significantly reduced the incidence of acute lower respiratory tract
infections in patients with chronic obstructive lung disease. 85 Although these
results may have been a result of development of mucosal immunity, differences
in colonization with H. injluenzae or salivary antibody levels to H. injluenzae were
not detected. Several surface-exposed pro teins that induce antibodies with bacte-
ricidal activity have been identified, but there is strain-to-strain antigenic vari-
ability in these proteins making it difficult to induce antibodies likely to be
protective against most or all strains.l 3 P6, a 16-kDa peptidoglycan-associated
outer-membrane protein is the most promising candidate protein to be included
in a vaccine against infection with nontypable H. injluenzae.B6 Cell surface-
exposed antigens of P6 have been identified that are widely conserved among
strains, and antibodies against this protein have bactericidal activity.
10. SUMMARY
Nontypable and type b strains of H. injluenzae are important causes of lower

respiratory infection in children and adults. Encapsulated type b strains are
responsible for most cases of H. injluenzae pneumonia in young children. In
adults, type b strains cause some pneumonias, particularly bacteremic pneu-
monias and pneumonias in adults with Human Immunodeficiency Virus infec-
tion. Pneumonias due to type b strains are preventable by immunization with an
H. injluenzae type b vaccine now routinely given to infants in the United States.
Nontypable strains of H. injluenzae are responsible for cases of pneumonia with
sepsis in newborns, a sizable proportion of pediatric pneumonias in developing
countries, most nonbacteremic pneumonias in adults, and the vast majority of
cases of bronchitis and tracheobronchitis in adults. Etiologic diagnosis of lower
respiratory tract infection may be difficult. Culture of H. injluenzae from pleural
fluid or blood is diagnostic. Visualization of pleomorphic gram-negative rods
as the predominant bacterial forms on a gram-stained smear of sputum and
culture of H. injluenzae from the specimen is adequate for a presumptive diag-
nosis. A vaccine to prevent infections due to nontypable H. injluenzae is not
available. Many antibiotics are available for treatment of respiratory infections
due to H. injluenzae. Ampicillin and amoxicillin are the agents with the longest
track record of success. However, an increasing percentage of strains of H.
injluenzae are resistant to beta-lactam antibiotics. The most common mechanism
of resistance is beta-Iactamase production resulting in antibiotic inactivation, but
a small proportion of strains are resistant to ampicillin because of a change in the
penicillin-binding protein targets of that antibiotic.
INFEGrIONS CAUSED BY HAEMOPHlLUS INFLUENZAE 79
REFERENCES
1. Koch, R., 1883, Bericht über die Thatigkeit der deutschen Cholerakommission in Aegyp-
ten, Wien Med Wsehr. 33:1548-1551.
2. Casin, 1., Grimont, E, and Grimont, P. A. 0., 1986, Deoxyribonucleic acid relatedness
between Haemophilus aegyptius and Haemophilus influenzae, Ann. Inst. PasteurlMierobiol.
137B:155-163.
3. Pittman, M., 1931, Variation and type specificity in the bacterial species H. infiuenwe,] Exp.
Med. 53:471-492.
4. Kilian, M., 1991, Haemophilus, in: Manual of Clinieal Mierobiology. (A. Balows, ed.), Fifth ed.
American Society for Microbiology, Washington nc.
5. Chapin, K. C., and Doern, G. v., 1983, Selective media for recovery of Haemophilus infiuenwe
from specimens contaminated with upper respiratory tract microbial flora,] Clin. Mierobiol.
17:1163-1165.
6. Barenkamp, S. ]., Munson, R. S., Jr., and Granoff, 0. M., 1981, Subtyping isolates of
Haemophilus infiuenwe type b by outer-membrane protein profiles,] Inf Dis. 143:668-676.
7. Murphy, T. E, Dudas, K. C., Mylotte,]. M., and Apicella, M. A., 1983, A subtyping system for
nontypable Haemophilus infiuenzae based on outer-membrane proteins, ] In! Dis. 147:
838-846.
8. Musser,]. M., Granoff, D. M., Pattison, P. E., and Selander, R. K., 1985, A population genetic
framework for the study of invasive diseases caused by serotype b strains of Haemophilus
injluenzae, Proe. Natl. Aead. Sei. USA 82:5078-5082.
9. Musser, ]. M., Barenkamp, S.]., and Granoff, 0. M., and Selander, R. K., 1986, Genetic
relationships of serologically nontypable and serotype b str;üns of Haemophilus infiuenzae,
Infect. Immun. 52:183-191.
10. Groeneveld, K., Van Alphen, L., Eijk, P. P.,Jansen, H. M., and Zanen, H. C., 1988, Changes
in outer membrane proteins of nontypable Haemophilus infiuenwe in patients with chronic
obstructive pulmonary disease,] In! Dis. 158:360-365.
11. Zwahlen, A., Kroll, ]. S., Rubin, L. G., and Moxon, E. R., 1989, The molecular basis of
pathogenicity in Haemophilus infiuenwe: Comparative virulence of genetically-related capsu-
lar transformants and correlation with changes at the capsulation locus cap, Mierob. Pathog.
7:225-235.
12. Weiser,]. N., Love,]. M., and Moxon, E. R, 1989, The molecular mechanism of phrase-
variation of H. infiuenwe lipopolysaccharide, Cell 59:657-665.
13. Murphy, T. E, and Apicella, M. A., 1987, Nontypable Haemophilus infiuenzae: A review of
clinical aspects, surface antigens, and the human immune response to infection, Rev. In! Dis.
9:1-15.
14. Gilsdorf,]. R., Chang, H. Y., McCrea, K. M., and Bakaletz, L. 0., 1992, Comparison of
hemagglutinating pili of Haemophilus infiuenwe type b with similar structures of nontype-
able H. infiuenzae, Infect. Immun. 60:374-379.
15. Plaut, A. G., 1983, The IgA 1 proteases of pathogenic bacteria, Ann. Rev. Mierobiol. 37:
603-622.
16. LiPuma,].]., Richman, H., and StulI, T. L., 1990, Haemocin, the bacteriocin produced by
Haemophilus injluenzae: Species distribution and role in colonization, Infect. Immun. 58:1600-
1605.
17. Spinola, S. M., Peacock,j., Denny, F. W, Smith, D. L., and Cannon,j. G., 1986, Epidemiology
of colonization by nontypable Haemophilus infiuenwe in children: A longitudinal study,] In!
Dis. 154:100-109.
18. Turk, 0. C., 1967, Haemophilus infiuenwe: Its CliniealImportanee. The English Universities
Press Ltd, London.
19. Hampton, C. M., Barenkamp, S. ]., and Granoff, 0. M., 1983, Comparison of outer
80 LORRY G. RUBIN
membrane protein subtypes of Haemophilus influemae type b isolates from healthy children
in the general population and from diseased patients,] GUn. Microbiol. 18:596-600.
20. Centers for Disease Control, 1986, Update: prevention of Haemophilus influemae type b
disease, MMWR 35:170-180.
21. Murphy, T. v., Granoff, D., Chrane, D. E, Olsen, K. D., Barenkamp, S. j., Dowell, S. E, and
McCracken, G. M.,Jr., 1985, Pharyngeal colonization with Haemophilus influenuu type b in
children in a day care center without invasive disease,] Pediatr. 106:712-716.
22. Li, K. I., Dashefsky, B., and Wald, E. R., 1986, Haemophilus influenuu type b colonization in
household contacts of infected and colonized children enrolled in day care, Pediatrics 78:15-20.
23. Michaels, R. H., and Norden, C. w., 1977, Pharyngeal colonization with Haemophilus
influenuu type b: A longitudinal study of families with a child with meningitis or epiglottitis
due to H. influemae type b,] Inf Dis. 136:222-228.
24. Takala, A. K., Eskola,j., Leinonen, M., Kayhty, H., Nissinen, A., Pekkanen, E., and Makela,
P. H., 1991, Reduction of oropharyngeal carriage of Haemophilus influenuu type b (Hib) in
children immunized with an Hib conjugate vaccine,] Infect. Dis. 164:982-986.
25. Murphy, T. v., Pastor, P. N., Medley, E B., Coury, S. L., Bravo, A., Osterholm, M. T., and
Granoff, D. M., 1993, Decreased Haemophilus colonization in children vaccinated with
Haemophilus influenuu type b conjugate vaccine,] Pediatr. 122:517-523.
26. Van Alphen, L., Geelen-van den Broek, L., Blaas, L., van Harn, M., and Dankert, j., 1991,
Blocking of fimbria-mediated adherence of Haemophilus influenuu by sialyl gangliosides,
Infect. Immun. 59:4473-4477.
27. St. Gerne, j. W. I., and Falkow, S., 1990, Haemophilus influemae adheres to and enters cultured
human epithelial cells, Infect. Immun. 58:4036-4044.
28. Read, R. C., Wilson, R., Rutman, A., Lund, V. j., Moxon, E. R., and Cole, P. j., 1989,
Interaction of non-typable Haemophilus influenuu (Hi) with human ciliated respiratory
epithelium in vitro, Am. Rev. Respir. Dis. 139:A37.
29. Read, R. C., Rutman, A. A.,Jeffery, P. K., Lund, v.j., Brain, A. P. R., Moxon, E. R., Cole, P.j.,
and Wilson, R., 1992, Interaction of capsulate Haemophilus inftuemae with human airway
mucosa in vitra, Infect. Immun. 60:3244-3252.
30. Farley, M. M., Stephens, D. S., Mulks, M. H., Cooper, M. D., Bricker, j. v., Mirra, S. S., et al.
1986, Pathogenesis of IgA) protease-producing and non-producing Haemophilus inftuenuu
in human nasopharyngeal organ cultures,] Infect. Dis. 154:752-759.
31. Weber, A., Harris, K., Lohrke, S., Forney, L., and Smith, A. L., 1991, Inability to express
fimbriae results in impaired ability of Haemophilus inftuemae b to colonize the nasopharynx,
Infect. Immun. 59:4724-4728.
32. Johnson, A. P., Clark, j. B., and Osborn, M. E, 1983, Scanning electron microscopy of the
interaction between Haemophilus influenuu and organ cultures of rat trachea, ] Med.
Microbiol. 16:477-482.
33. Johnson, A. P., and Inzana, T. j., 1986, Loss of ciliary activity in organ cultures of rat trachea
treated with lipo-oligosaccharide from Haemophilus inJluenuu,] Med. Microbiol. 22:
265-268.
34. Rubin, L. G., and Moxon, E. R., 1984, Haemophilus influenuu type b colonization resulting
from survival of a single organism,] Infect. Dis. 149:278.
35. Sheinman, B. D., Devalia, J. L., Davies, R. J., Crook, S. j., and Tabaqchali, S., 1986, Synthesis
of histamine by Haemophilus inftuenuu, Br. Med.] 292:857-858.
36. Kaplan, S. L., Taber, L. H., Frank, A. L., and Feigin, R. D., 1981, Nasopharyngeal viral
isolates in children with Haemophilus influenuu type b meningitis,] Pediatr. 99:591-593.
37. Michaels, R. H., and Myerowitz, R. L., 1983, Viral enhancement of nasal colonization with
Haemophilus influenuu type b in the infant rat, Pediatr Res. 17:472-473.
38. Michaels, R. H., Myerowitz, R. L., and Klaw, R., 1977, Potentiation of experimental
meningitis due to Haemophilus influenuu by influenza A virus,] Infect. Dis. 135:641-645.
INFEGTIONS CAUSED BY HAEMOPHILUS INFLUENZAE 81
39. Myerowitz, R. L., and Michaels, R. H., 1981, Mechanisms of potentiation of experimental
Haemophilus influenwe type b disease in infant rats by influenza A virus, Lab Invest. 44:
434-441.
40. Stephens, D. S., and Farley, M. M., 1991, Pathogenic events during infection of the human
nasopharynx with Neisseria meningitidis and Haemophilus influenwe, Rev. Infeet. Dis. 13:22-33.
41. Rubin, L. G., and Moxon, E. R., 1983, Pathogenesis of bloodstream invasion with Haemo-
philus influenwe type b, Infeet. Immun. 41:280-284.
42. Hansen, E. J., and Toews, G. B., 1992, Animal models for the study of noninvasive
Haemophilus influenwe disease: Pulmonary c1earance systems,] Infect. Dis. 165:S185-S187.
43. Slater, L. N., 1990, A rat model of prolonged pulmonary infection due to nontypable
Haemophilus influenwe, Am. Rev. Respir. Dis. 142:1429-1435.
44. Maciver, 1., Silverman, S. H., Brown, M. R. w., and O'Reilly, T., 1991, Rat model of chronic
lung infections caused by nontypable Haemophilus influenwe,] Med. Mierobiol. 35:139-147.
45. Wallace, F. J., Clancy, R. L., and Cripps, A. w., 1989, An animal model demonstration of
enhanced c1earance of nontypable Haemophilus influenwe from the respiratory tract after
antigen stimulation of gut-associated lymphoid tissue, Am. Rev. Respir. Dis. 140:311-316.
46. Moxon, E. R., and Wilson, R., 1991, The role of Haemophilus influenwe in the pathogenesis of
pneumonia, Rev. Infeet. Dis. 13:S518-S527.
47. Saez-Llorens, X., Ramilo, 0., Mustafa, M. M., Mertsola,J., and McCracken, G. H.,Jr., 1990,
Molecular pathophysiology ofbacterial meningitis: Current concepts and therapeutic impli-
cations,] Pediatr. 116:671-684.
48. Schneerson, R., Rodriques, L. P., Parke, J. C., Jr., and Robbins, J. B., 1971, Immunity to
disease caused by Haemophilus influenwe type b. 11. Specificity and some biologic characteris-
tics of "natural" infection-acquired, and immunization-induced antibodies to the capsular
polysaccharide of Haemophilus influenwe type b,] Immunol. 107:1081-1089.
49. Centers for Disease Control, 1991, Haemophilus b conjugate vaccines for prevention of
Haemophilus influenwe type b disease among infants and children two months of age and
older: Recommendations of the Immunization Practices Advisory Committee, MMWR
40(RR-l):1-7.
50. Weller, P. F., Smith, A. L., Smith, D. H., and Anderson, P., 1978, Role of immunity in the
dearance of bacteremia due to Haemophilus influenwe,] Infect. Dis. 138:427-436.
51. Noel, G. j., Katz, S., and Edelson, P. j., 1988, Complement-mediated early clearance of
Haemophilus influenwe type b from blood is independent of serum lytic activity,] Infect. Dis.
157:85-90.
52. Barenkamp, S. j., and Bodor, F. F., 1990, Development of serum bactericidal activity
following nontypable Haemophilus influenwe acute otitis media, Pediatr. Infect. Dis. ] 9:
333-339.
53. Jacobs, N. M., and Harris, V. j., 1979, Acute H aemophilus pneumonia in childhood, Am.] Dis.
Child. 133:603-605.
54. Asmar, B. 1., Slovis, T. L., Reed, j. 0., and Dajani, A. S., 1978, Hemophilus influenwe type b
pneumonia in 43 children,] Pediatr. 93:389-393.
55. Schreiber, J. R., Boxerbaum, B., and Morrissey, A. B., 1992, Dec1ine of Haemophilus disease
in Cleveland. Pediatr. Infect. Dis.] 11:779-780.
56. Peltola, H., Kilpi, T., and Anttila, M., 1992, Rapid disappearance of Haemophilus influenwe
type b meningitis after routine childhood immunization with conjugate vaccines, Lancet
340:592-594.
57. Black, S. B., and Shinefield, H. R., 1992, The Kaiser Permanente Pediatric Vaccine Study
Group. Immunization with oligosaccharide conjugate Haemophilus influenwe type b (HbOC)
vaccine on a large health maintenance organization population: extended follow-up and
impact on Haemophilus influenwe disease epidemiology, Pediatr. Infect. Dis.] 11:610-613.
58. Weinberg, G. A., Ghafour, A., Ishaq, Z., Nomani, N. K., Kabeer, M., Anwar, F., Burney, M.I.,
82 LORRY G. RUBIN
Qureshi, A. w., Musser, J. M., Selander, R. K., and Granoff, D. M., 1989, Clonal analysis of
Haemophilus inJluenzae isolated from children from Pakistan with lower respiratory tract
infections, J Infect. Dis. 160:634-643.
59. Lehmann, 0., 1992, Epidemiology of acute respiratory tract infections, especially those due
to Haemophilus inJluenzae, in Papua New Guinean children,J Infect. Dis. 165:S20-S25.
60. Shann, F., Gratten, M., Germer, S., Linnemann, v., Hazlett, 0., and Payne, R., 1984,
Aetiology of pneumonia in children in Goroka Hospital, Papua New Guinea, Lancet 2:
537-541.
61. Wallace, R. J., Baker, C. J., Quinones, F. J., Hollis, 0. G., Weaver, R. E., and Wiss, K., 1983,
Nontypeable Haemophilus inJluenzae (biotype 4) as a neonatal, maternal, and genital patho-
gen, Rev. Infect. Dis. 5:123-136.
62. Friesen, C. A., and Cho, C. T., 1986, Characteristic features of neonatal sepsis due to
Haemophilus inJluenzae, Rev. Infect. Dis. 8:777-780.
63. Rusin, P., Adam, R. 0., Petersen, E. A., Ryan, K. J., Sinclair, N. A., and Weinstein, L., 1991,
Haemophilus inJluenzae: An important cause of matemal and neonatal infections, Obstet.
Gynecol. 77:92-96.
64. Wallace, R.J.,Jr., Musher, 0. M., Septimus, E.J., McGowan,J. E.,Jr., Quinones, F.J., Wiss, K.,
Vance, P. H., and Trier, P. A., 1981, Haemophilus inJluenzae infections in adults: Characteriza-
tion of strains by serotypes, biotypes, and beta-Iactamase production, J Inftct. Dis. 144:
101-106.
65. Farley, M. M., Stephens, 0. S., Brachman, P. S.,Jr., Harvey, C. R., Smith,J., David, Wenger,J.
0.,1992, Invasive Haemophilus inJluenzae disease in adults: A prospective, population-based
surveillance, Ann. Intern. Med. 116:806-812.
66. Verghese, A., and Berk, S. L., 1983, Bacterial pneumonia in the elderly, Medicine 62:
271-285.
67. Berk, S. L., Holtsclaw, S. A., Wiener, S. L., and Smith,J. K., 1982, Nontypeable Haemophilus
inJluenzae in the elderly, Arch. Intern. Med. 142:537-539.
68. Smith, P. F., Stricof, R. L., Shayegani, M., and Morse, D. L., 1988, Cluster of Haemophilus
inJluenzae type binfeetions in adults,JAMA 260:1446-1449.
69. Polsky, B., Gold, J. W. M., Whimbey, E., Dryjanski, J., Brown, A. E., Schiffman, G., and
Armstrong, D., 1986, Bacterial pneumonia in patients with the acquired immunodeficiency
70. Moreno, S., Martinez, R., Barros, C., Gonzalez-Lahoz, J., Garcia-Delgao, E., and Bouza, E.,
1991, Latent Haemophilus inJluenzae pneumonia in patients infected with HIV, AIDS 5:
967-970.
71. Schlamm, H. T., and Yancovitz, S. R., 1989, Haemophilus inJluenzae pneumonia in young
adults with AIDS, ARC, or risk of AIDS, Am. J Med. 86:11-14.
72. Anthonisen, N. R., Manfreda,J., Warren, C. P. w., Hershfield, E. S., Harding, G. K. M., and
Nelson, N. A., 1987, Antibiotic therapy in exacerbations of chronic obstructive pulmonary
disease, Ann. Intern. Med. 106:196-204.
73. Sofer, S., Duncan, P., and Chernick, v., 1983, Bacterial tracheitis-an old disease redis-
covered, Clin. Pediatr. 22:407-411.
74. Donnelly, B. w., McMillan,J. A., and Weiner, L. B., 1990, Bacterial tracheitis: Report of eight
new cases and review, Rev. Infect. Dis. 12:729-735.
75. Musher, 0. M., Kubitschek, K. R., Crennan, J., and Baughn, R. E., 1983, Pneumonia and
acute febrile tracheobronchitis due to Haemophilus injluenzae in the elderly, Ann. Intern. Med.
99:444-450.
76. Topstad, 0., Waters, J., and Sheehan, G., 1993, Epiglottitis in children and adults: effect of
Haemophilus injluenzae type b vaccine. Program and Abstracts of the 33rd Interscience
Conference on Antimicrobial Agents and Chemotherapy. p. 421, abstract no. 1655.
INFECTIONS CAUSED BY HAEMOPHlLUS INFLUENZAE 83
77. Turner, R. B., Hayden, F. G., and Hendley,]. 0., 1983, Counterimmunoelectrophoresis of
urine for diagnosis of bacterial pneumonia in pediatric outpatients, Pediatrics 71:780-784.
78. Ramsey, B. W, Marcuse, E. K., Foy, H. M., Cooney, M. K., Allan, 1., Brewer, D., 1986, Use of
bacterial antigen detection in the diagnosis of pediatric lower respiratory tract infections,
Pediatrics 78:1-10.
79. Witt, C. S., Montgomery,]. M., Pomat, w., Lehmann, D., and Alpers, M. P., 1990, Detection of
Streptococcus pneumonia and Haemophilus injluenzae type b antigens in the serum and urine of
patients with pneumonia in Papua New Guinea: Comparison of latex agglutination and
counterimmunoelectrophoresis, Rev. Infect. Dis. 12:S1001-S1005.
80. Sood, S. K., Ballanco, G. A., Mather, F., Pramberg,]., and Daum, R. S., 1990, Distribution
and excretion of capsular antigen following immunization with H. injluenzae type b
polysaccharide-N. meningitidis outer membrane pro tein conjugate vaccine,] Inftct. Dis.
161:574-577.
81. Jorgensen,]. H., 1992, Update on mechanisms and prevalence of antimicrobial resistance in
Haemophilus injluenzae, Clin. Infect. Dis. 14:1119-1123.
82. Rubin, L. G., Medeiros, A., Yolken, R., and Moxon, E. R., 1981, Ampicillin treatment failure
of apparently beta-Iactamase negative Haemophilus injluenzae type b meningitis due to a
novel beta-Iactamase, Lancet 2:1008-1010.
83. Centers for Disease Control, 1986, Update: prevention of Haemophilus injluenzae type b
disease, MMWR 35:170-180.
83. Daum, R. S., Murphey-Corb, M., Shapira, E., and Dipp, S., 1988, Epidemiology ofRob beta-
lactamase among ampicillin-resistant Haemophilus injluenzae isolates in the United States,]
Infect. Dis. 157:450-455.
84. Gilsdorf,]. R., and Jesperson,]. M., 1986, Effect of antibiotics on adherence of Haemophilus
injluenzae type b, Antimicrob. Agents Chemother. 30:370-374.
85. Clancy, R. L., Cripps, A. w., Murree-Allen, K., Yeung, S., and Engel, M., 1985, Oral
immunisation with killed Haemophilus injluenzae for protection against acute bronchitis in
chronic obstructive lung disease, Lancet 2:1395-1397.
86. Nelson, M. B., Munson, R. S., Jr., Apicella, M. A., Sikkema, D. ]., Molleston, ]. P., and
Murphy, T. F., 1991, Molecular conservation of the P6 outer membrane protein among
strains of Haemophilus injluenzae: Analysis of antigenic determinants, gene sequences, and
restrietion fragment length polymorphisms, Inftct. Immun. 59:2658-2663.
5
Klebsiella Pneumonia
S. J. CRYZ, JR.
1. ETIOLOGICAL AGENT
Members of the genus Klebsiella are gram-negative nonmotile rods. Being faculta-
tive anaerobes, they are not nutritionally fastidious, enabling them to persist in a
variety of environments. Four species of Klebsiella are pathogenic for humans;
K. pneumoniae, K. oxytoca, K. ozaenae, and K. rhinoscleromatis. The overwhelming
majority oflife-threatening infections are caused by K. pneumoniae. A prominent
feature of nearly all Klebsiella clinical isolates is their ability to produce a polysac-
charide capsule, which is believed to playa critical role in the virulence of this
organism.
The ability of Klebsiella to cause human disease has been known since the late
nineteenth century when it was isolated from clinical specimens by Edwin Klebs.
Klebsiella is often referred to as Friedländer's bacillus in honor of the physician
who identified it as a significant respiratory pathogen. However, Klebsiella did not
gain prominence as a human pathogen until the antibiotic era. While Klebsiella is
most often thought of as a nosocomial pathogen, it is becoming increasingly
apparent that it can also cause serious community-acquired infections.
2. EPIDEMIOLOGY AND CLINICAL SIGNIFICANCE
Klebsiella infections, whether hospital- or community-acquired, occur almost

exclusively in debilitated patient subpopulations (Table I), and include pneu-
monia, primary and secondary bacteremia, urinary tract infections, bronchitis,
meningitis, and surgical wound infections. According to statistics compiled by the
S. J. CRYZ, JR. • Swiss Serum and Vaccine Institute, CH 3001 Berne, Switzerland.
Pulmonary Infections and Immunity, edited by Herman Chmel ct al. Plenum Press, New York, 1994.
85
86 S. J. CRYZ, JR.
TABLE I
Conditions That Predispose for Klebsiella Infection
Hospital-acquired Community-acquired
Malignancy Alcoholism
Burns Diabetes
Low birth weight Chronic obstructive pulmonary disease
Intubation Advanced age
Catheterization
Admission to intensive care unit
Thoracidthoracicoabdominal surgery
Immunosuppressive therapy
Nosocomial Infection Surveillance System,l Klebsiella was responsible for 7.4%

of all bacterial infections. It was the third most frequent cause of all types of lower
respiratory tract infections and primary bacteremia. In other large scale surveil-
lance studies, Klebsiella was found to be the leading cause of nosocomial pneu-
monia and bacteremia secondary to pneumonia. 2 Klebsiella is a leading cause of
pneumonia among elderly nursing horne, chronic care, and community pa-
tients. 3 ,4 In one report, Klebsiella was the predominant cause of pneumonia in
nursing horne patients, responsible for fully 40% of all ca ses. 3
Humans serve as the primary reservoir for Klebsiella and are instrumental in
its spread within the hospital environment. 5,6 Intestinal carriage rates among
healthy adults range widely. 5 Most studies have found that Klebsiella is excreted
for only a short period of time, suggesting that it is not part of the normal gut
flora. Carriage rates increase dramatically among hospitalized patients, especially
those receiving antibiotics. 7 Intestinal colonization predisposes to infection. 8 •9 In
one study, the infection rate increased from 10% in noncolonized patients to
45% for those colonized. Neonates are innately more prone to become colonized
with Klebsiella, even in the absence of antibiotic usage. Intestinal carriage
rates can reach 100% during an epidemic in a neonatal intensive care unit.
Nasopharyngeal carriage in the general population has been cited at 1% to
6%.5 Hospitalized patients without respiratory disease have a somewhat higher
carriage rate (2% to 19%), whereas roughly 10% of seriously ill patients harbor
Klebsiella in their upper respiratory tracts.5 Colonization rates approach 50%
among premature infants and neonates. The presence of Klebsiella in the naso-
pharynx also tends to increase with the administration of antibiotics. After the
urinary tract, the respiratory tract was the second most frequent portal of entry
associated with Klebsiella bacteremia. lO
Common-source outbreaks as a result of Klebsiella have been traced to a
variety of materials, induding blood sampling probes, breast milk and lipid
emulsions fed to infants, hand cream, and nebulizer solutions. 6 A significant
finding is that in three of these outbreaks, contamination of the common source
KLEBSIELLA PNEUMONIA 87
was traced to the hands of medical personnel. It is well known that hospital staff
often carry Klebsiella on their hands and can transmit these strains to patients. 3
The mortality rate associated with Klebsiella pneumonia and bacteremia
ranges between 24% and 48%11,12 (Table 11). Factors adversely affecting outcome
include: (1) an underlying disease state classified as rapidly fatal; (2) inadequate
antibiotic therapy; and (3) when the pulmonary tract or abdomen harbors the
foci of infection. 2,10
2.1. Antibiotic Resistance

Antibiotic resistance has also been implicated as a poor prognostic indica-
tor,6,13,14 Until recently, most clinical isolates of Klebsiella were uniformly sensitive
to third-generation cephalosporins and most aminoglycosides. Unfortunately,
resistance to these agents is emerging at an alarmingly rapid rate.l4--16 This is of
great practical concern since most hospitals employ an aminoglycoside together
with a cephalosporin as the treatment regimen of choice. Multiple antibiotic
resistance is rapidly acquired and transmitted by Klebsiella via transmissible
R-factors.l 7 Such R-factors code for several novel [3-lactamases (SHV-2 EBS,
CTX-l, CAZ, and SHV-3). Multiple resistances mayaiso develop du ring antibiotic
therapy of Klebsiella infections. 14,18 Resistant strains are more frequently associ-
ated with hospitaloutbreaks, especially among neonatal wards. I ,19,20 The mortal-
ity rate associated with infections caused by resistant strains is often substantially
higher than with nonresistant strains. 6,13,14
2.2. Pathogenesis of Klebsiella Pneumonia

Klebsiella stands out as being one of few gram-negative organisms that can
cause primary pneumonia in immunocompetent individuals. Klebsiella pneu-
monia invariably has an acute onset. Infection of the lower respiratory tract is
often associated with abscess formation, cavitation, empyema, and plural adhe-
sions. The propensity of Klebsiella to cause necrotizing, hemorrhagic foci leads
TABLE 11
Incidence and Mortality Rates for Klebsiella
Nosocomial Infection versus Sitea
Percent of total Percent deaths attributed
Infection site Klebsiella infections to Klebsiella
Urinary tract 50 7
Primary bacteremia 8 32
Lower respiratory tract 21 39
Surgical wound infection 13 14
Other 8 18
aData derived from National Nosocomial Infection Survey, 1975-1982.
88 S·1. CRYZ,1R.
to the production of "currant jelly" sputum, which is essentially diagnostic for

Klebsiella pneumonia. The sputum is often viscous because of the ability of many
Klebsiella species to produce and release copious amounts of capsular polysac-
charide material. Bacteremia secondary to pneumonia is often observed. The
clinical signs of Klebsiella sepsis are similar to those associated with other
gram-negative bacilli. However, septic shock appears to be a more common
development.
3. KLEBSIELLA VIRULENCE FACfORS
Klebsiella produce several somatic antigens that have been implicated in the
pathogenesis of Klebsiella infections. These include capsular polysaccharide
(CPS), lipopolysaccharide (LPS), adhesins, and aerobactin. Of these, CPS has
been studied in the greatest detail.
The vast majority of Klebsiella clinical isolates produce a well-defined capsu-
lar polysaccharide. 21 Electron microscopic studies have revealed the capsule to be
approximately 160 nm thick, greater than ten times that for Escherichia coli. 22 The
chemical composition of most and the primary structure of many of the 77
Klebsiella capsular serotypes have been determined. 23 ,24 All are acidic hetero-
polymers composed of repeating oligosaccharide' units. The basic repeating unit
is composed of either galacturonic or guluronic acid linked to hexoses and
deoxyhexoses. Several serotypes possess acetyl, formyl, or keto-linked pyruvate
groups.
The virulence of Klebsiella for laboratory animals is dependent on both the
presence and size of the bacterial capsule. 25 ,26 Domenico et al. 26 have shown
that Kl- and K2-bearing strains of Klebsiella that express a well-defined capsule
were capable of establishing experimentallobar pneumonia in rats. In contrast,
Kl and K2 variants that synthesized a capsule of reduced size were unable to
establish lobar pneumonia and elicited only minor lung pathology. The CPS is
believed to play a critical role in preventing efficient phagocytic uptake of
Klebsiella at several levels. The capsule has been shown to interfere with the
deposition of complement onto the bacterial cell surface, undoubtedly contribut-
ing to the serum-resistant phenotype expressed by many clinical isolates. 27,28
Klebsiella readily sheds CPS material from the cell surface. Such circulating CPS
can act to bind anti-CPS antibody at a site distant from the bacterial cello
Furthermore, detachment of CPS from the capsule layer upon antibody binding
could function to evade opsonophagocytic killing. Klebsiella CPS has also been
reported to induce astate of specific immune paralysis in a dose-dependent
manner29,30 and to block the maturation of precursor cells to macrophages in
vitro. 31 ,32
Athamna et al. 33 have recently shown that macrophages can mediate lectino-
phagocytosis of most Klebsiella CPS serotypes expressing the Mana2/3Man or
L:Rhaa2/3-L-Rha sequence via a surface lectin specific for mannose/N-acetyl-
glucosamine. Surprisingly, the CPS capsule would, therefore, serve to promote
phagocytic uptake. Interestingly, the expression of Mana2/3Man or L-Rhaa2/3-

L-Rha occurs at a significantly lower frequency among Klebsiella serotypes com-
monly associated with bacteremia.
Evidence to suggest a critical role for CPS in the pathogenesis of human
disease comes from the finding that CPS antigenemia was associated with a
significantly higher mortality rate in patients with Klebsiella bacteremia. 34 The
presence of circulating CPS in the urine and serum of patients was a poor
prognostic indicator. In most antigenernie patients, a persistent foci of infection
could be identified. Most patients with Klebsiella bacteremia mounted an immu-
noglobulin G antibody response directed against the CPS serotype expressed by
the invading strain,35 Many also responded with cross-reactive antibodies that
recognized heterologous serotypes of CPS.
Lipopolysaccharide (LPS) derived from Klebsiella possesses toxic attributes
commonly associated with this moleeule. LPS contributes substantially to the
serum-resistant phenotype expressed by many Klebsiella isolates.28.36.37 Only LPS
bearing a high molecular weight O-polysaccharide moiety functions in this
manner, however. LPS mayaiso confer a degree of resistance to intracellular
killing following phagocytosis. 38 Anti-LPS antibody can promote phagocytosis of
certain strains of Klebsiella. 28 Tomas et al. 39 have recendy shown that the 01
antigen was expressed on the surface of Klebsiella strains bearing one of 17
different capsular serotypes. It was completely masked, however, by those strains
expressing the Kl, KlO, and K16 antigen.
Straus et al. 40 •41 have described an extracellular toxic complex (ETC) com-
posed of CPS (63%), LPS (30%), and protein (7%) that is shed from the cell
surface. ETC was lethai for mice when injected intraperitoneally. When instilled
into the lungs of rats, ETC induced lesions similar to those observed subsequent
to experimental Klebsiella lobar pneumonia. In both model systems, anti-LPS
antibody completely neutralized the toxicity of ETC.
The ability of Klebsiella to cause respiratory tract infections depends on its
ability to colonize upper and lower airways. Many strains of Klebsiella synthesize
type-l pili that mediates attachment to epithelial and mucosal cell surfaces. 42
Type-l pili-bearing strains induced more extensive renal scarring than nonpili-
ated strains when evaluated in a rat pyelitis model. 43 Klebsiella outer-membrane
proteins mediate adhesion, which can be inhibited by mannose. 44-46 Expression
of the mannose-inhibitable adherence (MIAT) phenotype has been associated
with increased resistance to phagocytosis and intracellular killing. 46
The virulence of Kl- and K2-bearing strains of Klebsiella is markedly en-
hanced by the production of aerobactin, a high-affinity iron chelator encoded on
a 180 kDa plasmid. 47 The distribution of this plasmid among non-Kl and K2
strains is unknown.
3.1. Immunity to Klebsiella

Several experimental animal models have been developed that mimic natu-
ral infection. 25 .26 Curiously, only those strains expressing the Kl or K2 capsule
90 S. J. CRYZ, JR.
were found to be highly virulent in rodent models. Anti-CPS antibody has been
shown to provide serotype-specific protection against fatal experimental Klebsiella
burn wound and respiratory tract infections. 4S--50 Instillation of 106 Klebsiella
expressing the K2 CPS into the left lobe resulted in lobar pneumonia within 120
hr of infection. 49 Secondary bacteremia was noted 96 hr postinfection. Micro-
scopic examination of lung tissue revealed dissolution of alveolar septa and
massive intrabronchial infiltrate. Areas of pronounced necrosis could be observed
by gross examination. Bacterial counts exceeded 10 10 colony-forming units per
gram of lung tissue. Lung homogenates were markedly viscous, indicating the
presence of CPS in high amounts. Vaccination with K2 CPS prior to challenge
resulted in a decline in mortality rate to 8%. Lung pathology was reduced to intra-
and peribronchial involvement without consolidation and affected fewer than 5%
of airways. The lungs of immunized animals did become infected. Maximum
bacterial counts, however, were roughly three logs lower than those for nonimmu-
nized animals. There was no evidence of bacteremia in immunized animals.
Therefore, anti-CPS antibody was able to both facilitate clearance of bacteria
from the lungs and prevent invasion of the bloodstream.
Immunization of mice with native 01 LPS was able to confer significant
protection against an intraperitoneal challenge with Klebsiella strains expressing
the 01 antigen regardless of capsular serotype,39
4. CPS VACCINE
The rationale for the development of a Klebsiella CPS vaccine is based on the
above data demonstrating the protective capacity of anti-CPS antibody. Develop-
ment of such a vaccine was hampered by the existence of 77 distinct capsular
serotypes. 51 Seroepidemiological studies showed that approximately 90% of clini-
cal isolates could be serotyped. These investigations, which in most instances
involved sequentially isolated strains submitted to hospital microbiology laborato-
ries, found no clear predominant serotypes. Unfortunately, no attempt was made
to correlate isolation of a given strain of Klebsiella with an active disease state.
Given the ability of Klebsiella to colonize a variety of anatomical sites without
causing disease, one can reasonably assume that the majority of isolates analyzed
had litde or no clinical relevance. Several studies have focused on bacteremic
isolates,52-54 Unfortunately, the combined number of strains serotyped was too
small to allow any firm conclusions to be drawn in regard to seroprevalence. In an
attempt to address this issue in a systematic fashion, 708 Klebsiella blood isolates
from 13 hospitals in North America and Europe were analyzed. 21 Nearly all of
the 77 serotypes were represented among the isolates. However, three serotypes,
2, 21, and 55, were observed at a significandy higher frequency and together
accounted for 20.5% of all isolates. For the remaining serotypes, the frequency
ofisolation ranged from 0.16% to 2.8%. More than half ofthe sero types occurred
at a frequency of less than 0.5%.
Preliminary studies demonstrated that although Klebsiella CPSs could be
purified to near homogeneity, trace quantities of LPS remained that could not be
removed by physical means. 55 Based on the seroepidemiology of Klebsiella bacte-
remic isolates, any CPS-based vaccine would have to be multivalent. The cumula-
tive effect of trace LPS contamination among the monovalent components would
lead to an unacceptably high rate of adverse reactions when formulated into a
polyvalent vaccine. Treatment of purified CPS in a solution of95% ethanol-O.l N
NaOH was found to reduce the pyrogenicity (attributed to contaminating LPS) by
at least an order of magnitude without adversely affecting either the physico-
chemical or antigenic integrity of the CPS. Prototype vaccines formulated with
NaOH-treated CPSs were able to induce a good immunoglobulin G (IgG) anti-
body response in adult volunteers with a minimum of adverse reactions. 56 ,57
Vaccine-induced anti-CPS antibody was able to promote the uptake and killing of
Klebsiella by human polymorphonuclear leukocytes and to confer protection
against fatal experimental Klebsiella K2 burn-wound sepsis. The plasma from
individuals immunized with a prototype hexavalent CPS vaccine containing K2,
3,10,21,30, and 55 antigens was processed into an intravenous immune globulin
(IVIG).58 The hyperimmune IVIG was compared to commercial IVIGs for in
vitro and in vivo activities. The commercial IVIG preparations were unable to
support the opsonophagocytic killing of six Klebsiella test strains, each expressing
one of the vaccine serotypes. In contrast, the hyperimmune IVIG promoted the
uptake and killing of all six strains. Standard IVIG conferred significant protec-
tion against fatal experimental Klebsiella K2 burn-wound sepsis at doses;;:, 500
mg/kg body weight. A comparable level of protection could be obtained with the
hyperimmune IVIG at a dose of 5 mg/kg. The passive transfer ofhyperimmune
IVIG was effective at treating an established infection whereas commercialiVIG
was not. Antibiotics and the hyperimmune IVIG acted synergistically when used
to treat an ongoing infection. A similar effect was not seen with the commercial
products.
Based on the above results, a 24-valent CPS vaccine was formulated (Table
III) and tested for safety and immunogenicity in volunteers. 59-61 The criteria
TABLEIII
Characteristics of 24-Valent Klebsiella Capsular Polysaccharide Vaccine
Vaccine serotypes K2,3,~9, 10, 1~ 16. 17, 1~21,22,25,28,30,35,43,52.53,55,
60,61,62.63, and 64
Cross-reacting serotypes K7, 11, 12, 13, 14,26,31,37,46,68, and 69
Composition (wtlwt)
Polysaccharide 90%
Protein 1%
Nucleic acids 0.84%
Residual H 20 5.8%
Molecular weight >2 x 106
Pyrogenicity >24 fl.g/kg
92 S. J. CRYZ, JR.
used to select serotypes for inclusion into the vaccine included not only their
frequency of observance among bacteremic isolates, but also their ability to
engender cross-reactive antibodies. Two doses, 25 and 50 Ilg per serotype (600
and 1200 Ilg total CPS), were evaluated in a Phase 11 study. The vaccine was very
weIl tolerated at both doses. The most common complaint was transient tender-
ness at the injection site, which was reported by roughly 40% of the vaccinees.
The magnitude of the immune response engendered varied among the
serotypes. Vaccination elicited a greater than 4-fold rise in IgG levels to 75%
(18/24) of the vaccine serotypes. Two antigens were found to be poorly immuno-
genie in humans (K35 and K53). Of considerable interest was the 4-fold or greater
rise in titer that vaccinees responded with to 11 nonvaccine serotypes. Therefore,
the immune response elicited by this vaccine should provide 70-75% coverage. A
significant rise in opsonic antibody titers was noted for 21 of the 24 vaccine
antigens. The peak geometrie mean IgG antibody levels were comparable subse-
quent to immunization with either dose. Subjects who received the higher dose,
however, were more likely to manifest a 4-fold or greater rise in IgG antibody titer.
In addition, peak antibody levels were attained much earlier with the higher dose.
Therefore, all subsequent immunization programs were carried out with the
50-llg per serotype dose.
In an effort to develop a hyperimmune IVIG with an expanded range of
activity, plasma donors were simultaneously vaccinated with the above-described
Klebsiella vaccine and a polyvalent Pseudomonas aeruginosa O-polysaccharide-toxin
A conjugate vaccine,62 each administered intramuscularly into a different limb.
Coimmunization did not suppress the immune response to either vaccine as
compared to groups of volunteers who had received only a single vaccine. To date,
more than 2000 plasma donors have been vaccinated using this regimen. Systemic
adverse reactions, such as fever, chilIs, or pronounced malaise, are rare. Approx-
imately 25% of donors reported a local reaction, most often characterized by
tenderness at the injection site.
Severallots of hyperimmune IVIG have been produced from the plasma of
immunized donors,63 This hyperimmune product has significantly increased in
vitro (opsonophagocytic) and in vivo (mouse-protective) activities againstKlebsiella
and P. aeruginosa as compared to commercial IVIG preparations,63 This product
has been administered to a small number of healthy adults and intensive care-
unit patients at a dose of 100 mg/kg body weight with no untoward effect. A
significant rise in serum IgG antibody levels to all of the vaccine antigens was seen
postinfusion. Titers remained elevated for at least 5 days after infusion.
5. CONCLUSION
The hyperimmune IVIG described above is currently being evaluated in a

multicenter, double-blind, randomized, piacebo-controlled trial. Patients, who
enter either the medical or surgical intensive care unit and who are expected to
remain on-service for more than 24 hours, are randomized to receive either drug
or placebo. It is hoped that when used in such a prophylactic manner, the

hyperimmune IVIG will effect a decrease in Klebsiella and P. aeruginosa infection
rates.
REFERENCES
1. Horan, T. C., White,]. W.,Jarvis, W. R., Emori, T. G., Culver, D. H., Munn, V. P., Thornsberry,
C., Olson, D. R., and Hughes,]. M., 1986, Nosocomial infection surveillance, 1984, CDC
Surveillance Summaries, Morb. Mortal. Wkly. Rep. 35(No. 1 SS):17SS-29SS.
2. Jay, S. S., 1983, Nosocomial pneumonia: The challenge of achanging clinieal system,
Postgraduate Med. 74:221-235.
3. Garb,]. L., Brown, R. B., Garb,]. R., and TuthilI, R. w., 1978, Differences in etiology of
pneumonias in nursing horne and community patients,I Am. Med. Assoe. 240:2169-2172.
4. Ebright,]. R., and RyteI, M. w., 1980, Bacterial pneumonia in the eIderly,I Am. Geriatr. Soe.
27:345-347.
5. Montgomerie,]. Z., 1979, Epidemiology of Klebsiella and hospital-associated infections, Rev.
Infeet. Dis. 1:736-753.
6. Jarvis, W. R., Munn, V. P., Highsmith, A. K., Culver, D. H., and Hughes,]. M., 1985, The
epidemiology ofnosocomial infections caused by Klebsiella fmeumonioR, Infeet. ControI6:68-74.
7. Rose, H. D. and Schreier,]., 1968, The effect ofhospitalization and antibiotic therapy on the
gram-negative fecal flora, Am. I Med. Lab. Sei. 255:228-235.
8. Smith, R. E, Dayton, S. L., Chipps, D. D., and Blasi, D., 1973, Intestinal carriage of Klebsiella
and Pseudomonas in burned children and their comparative role in nosocomial infection,
Healih Lab. Sei. 10:173-177.
9. Seiden, R., Lee, S., Wang, W. L. L., Bennett, ]. v., and Eickhoff, T. C., 1971, Nosocomial
klebsiella infections: Intestinal colonization as areservoir, Ann. Intern. Med. 74:657-664.
10. de la Torre, M. G., Romero-Vivas, ]., Martinez-Beltran, ]., Guerrero, A., Meseguer, M.,
Bouza, E., 1985, Klebsiella bacteremia: An analysis of 100 episodes, Rev. Infeet. Dis. 7:
143-150.
11. Haddy, R. 1., Klimberg, S., and Epting, R.]., 1987, A two center review ofbacteremia in the
community hospital, I Fam. Pract. 24:253-259.
12. Kreger, B. E., Craven, D. E., Carling, P. C., and McCabe, W. R., 1980, Gram-negative
bacteremia: III Reassessment of etiology, epidemiology and ecology in 612 patients, Am.
I Med. 68:332-343.
13. Ma, M. Y., Mulligan, M. E., Friedman, M. H., and Anderson, M. S., 1983, Prognosis of
klebsiella bacteremia: Possible association with antimicrobial usage and bacterial resistance,
in: Seeond International Symposium on Infeetions in the Immunoeompromised Host, (C. S. E Eas-
mon and H. Gaya, eds.), Academic Press, New York, pp. 211-213.
14. Brun-Buisson, C., Legrand, P., Philippon, A., Montravers, E, Ansquer, M., and Duval,].,
1987, Transferable enzymatic resistance to third-generation cephalosporins du ring noso-
comial outbreak of multiresistant Klebsiella pneumonioR, Laneet 2:302-306.
15. Sanders, C. C., and Sanders, W. E., Jr., 1985, Mierobial resistance to newer generation
ß-lactam antibioties: Clinieal and laboratory implications, I Infect. Dis. 151:399-406.
16. Bauernfeind, A., 1990, Perspectives of beta-lactamases in therapy of infections caused by
Eseheriehia eoli or Klebsiella with plasmidic resistance to third generation cephalosporins,
Infection 18:556-60.
17. Mayer, K. H., Hopkins,]. D., Gilleece, E. S., Chao, L., and O'Brien, T. E, 1986, Molecular
evolution, species distribution, and clinieal consequences of an endemie aminoglycoside
resistance plasmid, Antibiot. Agents Chemother. 29:628-633.
94 S. ]. CRYZ, JR.
18. Then, R. L., Glauser, M. P., Angehrm, P., and Arisawa, M., 1983, Cephalosporin resistance in
strains of Klebsiella oxytoca isolated during antibiotic therapy, Zentralbi. Bakteriol. Hyg.
254:469-477.
19. McKee, K. T., Cotton, R. B., Stratton, C. w., Lavely, G. B., Wright, P. F., Shenai,]. P., Evans, M.
E., Melly, M. A., Farmer,].]., Karzon, D. T., and Schaffner, w., 1982, Nursery epidemic due
top multiply-resistant Klebsiella pneumoniae: Epidemiologic setting and impact on the peri-
natal health care delivery, Inftct. ControI3:150-156.
20. Woloj, M., Tolmasky, M. E., Roberts, M. C., and Crosa, ]. H., 1986, Plasmid-encoded
amikacin resistance in multiresistant strains of Klebsiella pneumoniae isolated from neonates
with meningitis, Antimicrob. Agents Chemother. 29:315-319.
21. Cryz, S.]., Jr., Mortimer, P. M., Mansfield, v., and Germanier, R., 1986, Seroepidemiology of
Klebsiella bacteremic isolates and implications for vaccine development,] Clin. Microbiol. 23:
687-690.
22. Amako, K., Meno, Y., and Takade, A., 1988, Fine structure of the capsules of Klebsiella
pneumoniae and Escherichia coli Kl,] Bacteriol. 170:4960-4962.
23. Heidelberger, M., and Nimmich, w., 1976, Immunochemical relationships between bacteria
belonging to two separate families: Pneumococci and Klebsiella, Immunochem. 13:67-82.
24. Kenne, L., and Lindberg, B., 1983, Bacterial polysaccharides, in: The Polysaccharides, Volume
2, (G. O. AspinalI, ed.), Academic Press, New York, pp. 287-363.
25. Cryz, S.]., Jr., Fürer, E., and Germanier, R., 1984, Experimental Klebsiella pneumoniae burn
wound sepsis: Role of capsular polysaccharide, Inftct. Immun. 43:440-441.
26. Domenico, P',Johanson, W. G., and Straus, D. C., 1982, Lobar pneumonia in rats produced
by clinical isolates of Klebsiella pneumoniae, Inftct. Immun. 37:327-335.
27. Trautmann, M., Cryz, S. j., Jr., Sadoff, j. C., and Cross, A. S., 1988, A murine monoclonal
antibody against Klebsiella capsular polysaccharide is opsonic in vitro and protects against
experimental Klebsiella pneumoniae infection, Microbial Pathogenesis 5:177-187.
28. Williams, P., Lambert, P. A., Brown, M. R. w., and Jones, R.]., 1983, The role of the 0 and K
antigens in determining the resistance of Klebsiella aerogenes to serum killing and phago-
cytosis,] Gen. Microbiol. 129:2181-2191.
29. Batshon, B. A., Baer, H., and Shaffer, M. F., 1963, Immunological paralysis in mice by
Klebsiella pneumoniae type 2 polysaccharide,] Immunol. 90:121-126.
30. Baer, H., Bringaze,]. K., and McNamee, M., 1954, The effect of the doses of bacterial
polysaccharide antigen in antibody production in mice,] Bact. 67:123-130.
31. Yokochi, T., Nakashima, 1., and Kato, N., 1977, Effect of capsular polysaccharide of
Klebsiella pneumoniae on the differentiation and functional capacity of macrophages cultured
in vitro, Microbiol. Immunol. 21:601-610.
32. Yokochi, T., Nakashima, 1., and Kato, N., 1979, Further studies on generation of macro-
phages in in vitro cultures of mouse spleen cells and its inhibition by the capsular polysac-
charide of Klebsiella pneumoniae, Microbiol. Immunol. 23:487-499.
33. Athamna, A., Offek, 1., Kasari, Y., Markowitz, S., Dutton, G. G. S., and Sharon, N., 1991,
Lectinophagocytosis of encapsulated Klebsiella pneumoniae mediated by surface lectins of
guinea pig alveolar macrophages and human monocyte-derived macrophages, Infect. Im-
mun. 59:1673-1682.
34. Pollack, M., 1976, Significance of circulating capsular antigen in Klebsiella infections, Infect.
Immun. 13:1543-1548.
35. Brauner, A., Cryz, S. j., Jr., Granström, M., and Wretlind, B., 1990, Serological response to
eight Klebsiella pneumoniae capsular polysaccharide antigens in patients with bacteremia,
Infect. Dis. 4:487-493.
36. Tomas,]. M., Benedi, v.]., Ciurana, B., andJofre,]., 1986, Role of capsule and 0 antigen in
resistance of Klebsiella pneumoniae to serum bactericidal activity, Inftct. Immun. 54:85-89.
37. Ciurana, B., and Tomils, ]. M., 1987, Role of lipopolysaccharide and complement in
susceptibility of Klebsiella pneumoniae to nonimmune serum, Infoct. Immun. 55:2741-2746.
38. Allen, P. M., Fisher, 0., Saunders,]. R., and Hart, C. A., 1987, The role of capsular polysac-
charide K21b of Klebsiella and of the structurally related colanic-acid polysaccharide of
Escherichia coli in resistance to phagocytosis and serum killing, J Med. Microbiol. 24:363-370.
39. Tomils, ]. M., Camprubi, S., Merino, S., Davey, M. R., and Williams, P., 1991, Surface
exposure of 01 serotype lipopolysaccharide in Klebsiella pneumoniae strains expressing
different K antigens, Infect. Immun. 59:2006-2011.
40. Straus,o. C., Atkisson, 0. L., and Garner, C. W, 1985, Importance of a lipopolysaccharide-
containing extracellular toxic complex in infections produced by Klebsiella pneumoniae, Inftct.
Immun. 50:787-795.
41. Straus, 0. C., 1987, Production of an extracellular toxic complex by various strains of
Klebsiella pneumoniae, Infect. Immun. 55:44-48.
42. Fader, R. C., and Davis, C. P., 1980, Effect of piliation on Klebsiella pneumoniae infection in rat
bladders, Infect. Immun. 30:554-561.
43. Fader, R C. and Davis, C. P., 1982, Klebsiella pneumoniae induced experimental pyelitis: The
effect of piliation on infectivity, J Urol. 128:197-208.
44. Pruzzo, C., Valisena, S., and Satta, G., 1983, Laboratory and wild-type Klebsiella pneumoniae
strains carrying mannose-inhibitable adhesin and receptors for coliphages T3 and T7 are
more pathogenic for mice than are strains without such receptors, lrifect. Immun. 39:
520-527.
45. Pruzzo, C., Deblia, E. A., and Satta, G., 1980, Identification ofthe major adherence ligand of
Klebsiella pneumoniae in the receptor for coliphage T7 and alteration of Klebsiella adherence
properties by lysogenic conversion, Infect. Immun. 30:562-568.
46. Pruzzo, C., Deblia, E., and Satta, G., 1982, Mannose-inhibitable adhesions and T3-T7
receptors of Klebsiella pneumoniae inhibit phagocytosis and intracellular killing by human
polymorphonuciear leukocytes, Infect. Immun. 36:949-956.
47. Nassif, X., and Sansonetti, P., 1986, Correlation of the virulence of Klebsiella pneumoniae Kl
and K2 with the presence of a plasmid encoding aerobactin, Inftct. Immun. 54:603-608.
48. Cryz, S. ]., Jr., Fürer, E., and Germanier, R, 1984, Proteetion against fatal Klebsiella
pneumoniae burn wound sepsis by passive transfer of anticapsular polysaccharide, Inftct.
Immun. 45:139-142.
49. Cryz, S.]., J r., Fürer, E., and Germanier, R, 1986, Immunization against fatal experimental
Klebsiella pneumoniae pneumonia, Infect. Immun. 54:403-407.
50. Riottot, M.-M., Fournier, ].-M., and Pillot, ]., 1979, Capsular serotypic specificity of the
proteetion conferred on mice by Klebsiella pneumoniae ribosomal preparations, Infect. Immun.
24:476-482.
51. 0rskov, 1. and Fife-Asbury, M. A., 1977, New Klebsiella capsular antigen, K82, and the
deletion of five of those previously assigned, Internat. J System Bacteriol. 27:386-387.
52. Maiztegui,]. 1., Biegeleisen,]. Z.,Jr., Cherry, WB., and Kass, E. H., 1965, Bacteremia due to
gram-negative rods: A ciinical, bacteriologic, serologie and immunofluorescent study, N.
Engl. J Med. 272:222-229.
53. Montgomerie, ]. Z. and Ota, ]. K., 1980, Klebsiella bacteremia, Arch. Intern. Med. 140:
525-527.
54. Kiseleva, B. S. and Krasnogolovets, V. N., 1983, Role of Klebsiella pneumoniae in the etiology
of bacterial sepsis, Zh. Mikrobiol. Epidemiol. Immunbiol. 2:20-25.
55. Cryz, S. ]., Jr., Fürer, E., and Germanier, R., 1985, Purification and vaccine potential of
Klebsiella capsular polysaccharides, Infect. Immun. 50:225-230.
56. Cryz, S.]., Jr., Fürer, E., and Germanier, R, 1985, Safety and immunogenicity of Klebsiella
pneumoniae Kl capsular polysaccharide vaccine in humans, J Inftct. Dis. 151:665-671.
96 S. j. CRYZ, JR.
57. Cryz, S. j., Jr., Mortimer, P., Cross, A. S., Fürer, E., and Germanier, R., 1986, Safetyand
immunogenicity of a polyvalent Klebsiella capsular polysaccharide vaccine in humans,
Vaccine 4:15-20.
58. Cryz, S. j., Jr., Cross, A. S., Fürer, E., Chariatte, N., Sadoff, j. C., and Germanier, R., 1986,
Activity of intravenous immune globulins against Klebsiella,j Lab. Glin. Med. 108:182-189.
59. Cryz, S. j., Jr., 1990, Klebsiella polysaccharide vaccines, in: Bacterial Vaccines: Advances in
Biotechnical Processes, Volume 13 (A. Mizrahi, ed.), Wiley-Liss, New York, pp. 87-103.
60. Cross, A. S., 1991, Klebsiella, in: Vaccines and Immunotherapy (S. j. Cryz, Jr., ed.), Pergamon
Press, New York, pp,. 178-185.
61. Granström, M., Wretlind, B., Markham, B., and Cryz, S. j., Jr., 1988, Enzyme-linked
immunosorbent assay to evaluate the immunogenicity of a polyvalent Klebsiella capsular
polysaccharide vaccine in humans,j Glin. Microbiol. 26:2257-2261.
62. Cryz, S. j.,Jr., Sadoff,j. C., Cross, A. S., and Fürer, E., 1989, Safety and immunogenicity of a
polyvalent Pseudomonas aeruginosa O-polysaccharide-toxin A vaccine in humans, Antibiot.
Ghemother. 42:177-183.
63. Cryz, S. j., Jr., Fürer, E., Sadoff, j. C., Fredeking, T., Que, j. u., and Cross, A. S., 1991,
Production and characterization of a human hyperimmune intravenous immunoglobulin
against Pseudomonas aeruginosa and Klebsiella species, j Infoct. Dis. 163:1055-1061.
6
Legionella pneumophila
JANET E. STOUT and VICTOR L. YU
1. THE DISEASE
The syndrome ofLegionnaires' disease was first recognized at the 1976 American
Legion Convention in Philadelphia where 221 Legionnaires contracted pneu-
monia and 34 died. The causative agent was identified as an aerobic gram-
negative bacterium, subsequently named Legionella pneumophila.l Although the
family Legionellaceae comprises more than 40 species, L. pneumophila remains
the most common species implicated in human disease and therefore will be the
focus of this discussion.
2. EPIDEMIOLOGY AND TRANSMISSION
The natural habitats for L. pneumophila are aquatic bodies including rivers,
lakes, streams, and thermally polluted waters, all of which contain very small
numbers of Legionella. Aquatic environments also include man-made habitats
such as cooling towers, evaporative condensors, and most important, potable
water distribution systems. The organism can survive in a wide range of condi-
tions including temperatures ofO to 63°C, pH of 5.0, dissolved oxygen concentra-
tions of 0.2 to 9.6 ppm in water. Other commensal bacteria and blue-green algae
can promote the growth of Legionella in water. 2-4
As will be discussed, L. pneumophila is an intracellular pathogen and can
infect and multiply with amoebae and ciliated protozoa. 5 ,6
JANET E. STOUT • Special Pathogens Laboratory, Pittsburgh, Pennsylvania 15261. VICfOR

L. YU • University of Pittsburgh, Infectious Disease Section, VA Medical Center, Pittsburgh,
Pennsylvania 15240.
Pulmonary Inftctions anti Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
97
98 JANET E. STOUT and VICTOR L. YU
The incidence of Legionnaires' disease depends on the degree of coloniza-

tion of the aquatic reservoir, susceptibility of the host exposed to the water, and
the intensity of that exposure. Accepted risk factors for acquisition of Legion-
naires' disease include smoking and chronic obstructive pulmonary disease, two
clinical entities in which the mechanical clearance of pathogens by respiratory-
tract cilia is impaired. Immunosuppression is a major risk factor including
corticosteroid administration and receipt of an organ trans plant. Transplant
recipients may be at increased risk of Legionella infection because of severe
immunosuppression from antirejection medications. The antirejection drug cy-
closporin A has been shown to increase susceptibility of mice and rats to Legionella
pneumonia. 7,8 The apparent mechanism of increased susceptibility to Legionella
infection from treatment with cyclosporin A is selective depletion of functional T
cells; this results in a significant reduction in the production of the macrophage-
activating cytokine interferon-gamma. Surgical trauma associated with trans-
plantation may also contribute to impaired T-cell function. In a lung trans plant
model, impaired clearance of Legionella from infected rat lungs was attributed to
disruption of lymphatic vessels. A suboptimal immune response may result as a
result of impaired lymphocyte translocation to the tracheobronchial lymph
nodes, the site of immune cell activation and proliferation. 8
Interestingly, patients with neutropenia and AIDS do not appear to have a
notably higher attack rate than that of immunocompetent patients. Infection in
children is rare but has occurred in immunosuppressed children and children
with transplants. 9 ,10
Although cooling towers have long been touted as a major reservoir for the
infecting organism, we have challenged that hypothesis. ll ,12 We point out that
once Legionella was recognized as being part of the commensal flora within a
water distribution system,13 outbreaks implicating cooling towers have essentially
disappeared.
The mode of transmission of L. pneumophila to man from water is uncertain,
although evidence exists for aerosolization, aspiration, or direct instillation
into the lung during respiratory tract manipulation (Fig. 1).1 1
Endemie nosocomial disease has been common in hospitals with water
distribution systems colonized with the organisms. Infection has been linked to
endotracheal intubation,n,14 respiratory tract equipment,15 and aspiration.16-18
In the original Pontiac Fever outbreak, L. pneumophila was isolated from the
lungs of sentinal guinea pigs exposed to aerosols generated from an evaporative
condenser.1 9 Some retrospective studies have suggested that showering, which
would provide a convenient means of aerosolization, may be a mode of transmis-
sion but numerous prospective studies have failed to confirm this link.
Humidifiers have been shown to generate aerosols containing L. pneumophila
and inhalation of aerosolized water from such devices has been linked to
nosocomial Legionnaires' disease.1 5,20-22 It has been observed that health care
personnel frequendy use tap water to rinse respiratory apparatus and tubing. If
the tap water is contaminated with L. pneumophila, the organism can be direcdy
instilled into the lung. In two studies, patients with Legionnaires' disease under-
Aerosolization ~
es
and inhalation of ~
droplet nuclei or E
amoebae cysts / Alveolar macrophage ~
tl"l
~
Aspiration of ~
contaminated water ~
~
or following
LYSOSOmeO
colonization
Alveolus
FIGURE 1. The mode of transmission of Legionella is aerosolization or aspiration. Legionella enters the alveolar macrophage via a
"coiling phagocytosis" or by conventional FC-mediated zipper phagocytosis. Virulent Legionella strains inhibit phagosome-lysosome
fusion and intracellular multiplication proceeds in the phagosome until the cell ruptures. ~
~
went endotracheal intubation significantly more often or had significantly longer

duration of intubation than patients with other pneumonia.l 1.14
Evidence has gradually accumulated suggesting that aspiration of contami-
nated water or contaminated oropharyngeal secretions may be a major mode of
transmission.l6.17.23.24 Colonization of the oral pharynx by the organism has not
been consistently shown, however. Nasogastric tube placement has been a signifi-
cant risk factor in some endemie situations; microaspiration of contaminated
water was the presumed mode of entry.18 Legionnaires' disease was found to be
the most common cause of nosocomial pneumonia in a population of head- and
neck-cancer patients who have major propensity for aspiration as a result of their
oral surgery and extensive history of cigarette smoking. 25
3. INTRACELLULAR NICHE
Pneumonia is the presenting clinical syndrome in the overwhelming majority

of cases of Legionnaires' disease. L. pneumophila evades host defense by occupying
a niche within host cells that normally rid the body of bacterial pathogens. L.
pneumophila enters, survives, and multiplies within polymorphonuclear leuko-
cytes, blood monocytes, and alveolar macrophages.
Polymorphonuclear leukocytes and monocytes are recruited from the blood
in proportion to the inocula of Legionella. 26 Initially, there is an influx of neu-
trophils, but with the transformation of the blood monocytes into macrophages,
the inflammatory response is composed of roughly equal numbers of neutrophils
and mononuclear cells. These mobile phagocytes are likely attracted by chemotac-
tic factors derived [rom the alveolar macrophages and/or the infecting organism.
Peripheral blood monocytes can ingest L. pneumophila, although L. pneumo-
phila can multiply in these cells. 27 The presence of specific antibody as an opsonin
im proves phagocytosis. 28
Both specific antibody and complement are required for neutrophils to
efficiently ingest L. pneumophila. Although neutrophils fail to produce significant
killing of L. pneumophila, intracellular replication of the organism does not
occur. 26 .28
Alveolar macrophages readily phagocytose Legionella, although the process is
more avid in the presence of specific opsonizing antibody.29 Phagocytosis of
Legionella has been described by Horwitz as a coiling process whereby pseudo-
pods wrap around the organism and engulf it within a specialized ribosome-lined
phagosome 30 (Fig. 1). The phenomenon of coiling phagocytosis is thought to
occur when complement component C3 is deposited on the bacterial surface and
the organism then attaches to the complement receptors on the phagocyte
surface.3 1 In the presence of immune antibody, Legionella enters the cell by
conventional Fc-receptor mediated zipper phagocytosis. 30 Although Legionella
enters the macrophage or monocyte, phagosomes fail to fuse with lysosomes and
do not acidify, thus allowing the organism to escape the microbicidal mechanisms
of these organelles. 3o This mechanism of intracellular survival is also seen with
LEGIONELLA PNEUMOPHlLA 101
Mycobacteria, Toxoplasma, and Chlamydia. Legionella then multiplies until the cell
ruptures. The liberated bacteria are then phagocytosed by newly recruited cells,
and the cyde of ingestion, multiplication, and liberation with cell lysis begins
anew. Inhibitors of microfilament-dependent phagocytosis and adsorptive pino-
cytosis, cytochalasin D and methylamine, have been shown to prevent the growth
of L. pneumophila in a human monocyte-like cellline (U937).3 2
Legionella also occupies an intracellular niche in the environment. Row-
botham speculated that amoebae in soil and water could act as amplifiers of
virulent Legionella in the environment and furnish the vehide for transmission of
Legionella from the environment to man. 6 Intracellular multiplication of L.
pneumophila has since been shown to occur in Acanthamoeba,33 Naegleria,34.35
and Tetrahymena. 5
4. IMMUNE DEFENSE MECHANISMS
In vitro studies suggest that humoral immunity plays only a secondary role in
host defense against Legionella. For example, antibody does not promote killing
of L. pneumophila by complement, promotes only modest killing of phagocytes,28
and facilitates uptake of L. pneumophila by macrophages but does not inhibit
intracellular multiplication in monocytes or alveolar macrophages. 28 .29 Some
investigators have speculated that humoral immunity might even be deleterious
to the host by facilitating uptake of Legionella by growth-promoting macrophages.
A beneficial role of humoral immunity is, however, suggested by the fact that
patients develop type-specific anti-Legionella IgM and IgG antibodies within
several weeks of infection and immunized animals develop a specific antibody
response with subsequent resistance to Legionella challenge.36 •37
Depletion of complement does not alter either the recruitment of neutrophils
to the lung or the degree of bacterial growth during the course of Legionella
pneumonia in rats. 38 Phagocytosis of Legionella is mediated by complement
receptors on monocytes. Legionella fixes complement component C3b to its
surface via the alternative pathway. C3 acts as a ligand on the bacterial surface that
becomes available for binding to complement receptors CRI and CR3 on the
phagocyte surface. 31 A major Legionella outer-membrane protein appears to be
the prominent acceptor molecule for C3.39
Cell-mediated immunity is accepted as the primary host defense against
Legionella. Legionnaires' disease is more common and more severe for patients
with depressed cell-mediated immunity such as transplant recipients and patients
receiving corticosteroids. Transfer of immune spleen cells but not antibody
protects guinea pigs against lethai infection with Legionella. 26
The antimicrobial action of phagocytic cells against intracellular pathogens
indudes production of organic acids, cationic proteins, toxic oxygen metabolites,
and myeloperoxidase. The ability of L. pneumophila to survive intracellularly
is not a result of resistance to the oxygen-dependent microbicidal systems found
in phagocytes. 40.41 L. pneumophila has been shown to be susceptible to the anti-
microbial action of oxygen metabolites generated by both the myeloperoxidase-

H 20 2-halide and the xanthine oxidase systems. 41 The response of human poly-
morphonuclear leukocytes to virulent L. pneumophila has also been shown to result
in a reduced oxidative response as compared to the response elicited by E. coli or
avirulent strains of L. pneumophila. 42 The effect was partially attributed to
decreased complement binding by virulent strains. Superoxide production by
alveolar exudate macrophages elicited by L. pneumophila is significantly reduced,
suggesting that resistance of exudate macrophages to L. pneumophila does not
depend on an augmented respiratory burst. 43
Some intracellular pathogens, including L. pneumophila, are dependent on
the availability of free iron within the cello Iron is acquired by mononuclear
phagocytic cells via transferrin receptors on their surfaces. Interferon-gamma
decreases the availability of intracellular iron by decreasing the number of
transferrin receptors on the monocyte surface. 44 Interferon-gamma-activated
human monocytes inhibit intracellular multiplication of L. pneumophila. 45 ,46
Growth of L. pneumophila in the interferon-gamma-activated cells can be restored
by the addition of exogenous transferrin-iron or restricted by treatment of the
cells with the iron chelator deferoxamine. 44 Transferrins and chemical chelators
have been shown to inhibit the growth of L. pneumophila, further demonstrat-
ing the sensitivity of Legionella to iron deficient conditions. 47
The importance of the macrophage in the pathogenesis of Legionnaires'
disease is also supported by reports that susceptibility to Legionella infection
among different animal species correlates with permissiveness of macrophages
to L. pneumophila. 48 ,49 Multiplication of Legionella has been shown to occur in
guinea pig, mouse and rat macrophages, monkey alveolar macrophages, human
peripheral blood monocytes, and human alveolar macrophages. 27 ,29,50-54 Species
differences in intracellular growth may be attributable to permissiveness of
macrophages for Legionella growth and differences in production of macrophage-
activating cytokines from lymphocytes stimulated with Legionella antigen. 55
Phagocytic ceHs are impaired by corticosteroids, and corticosteroid-treated
rats have a diminished resistance to L. pneumophila. 52 Corticosteroid treatment
is an established risk factor for Legionnaires' disease.
Evidence of the development of ceH-mediated immunity includes the ap-
pearance of lymphocyte proliferation and cutaneous delayed hypersensitivity to
L. pneumophila antigens within the first two weeks of infection.3 7,56-60 In both
infected patients and animals, mononuclear cells res pond to L. pneumophila
antigens with proliferation and with the generation of a variety of cytokines,
including interferon-gamma, interleukin 1 and 2 and tumor necrosis factor
(TNF).45,61-fj4
The capacity of cytokines to induce an antimicrobial state in mononuclear
phagocytes relates to their ability to augment oxidative as weIl as nonoxidative
mechanisms of cellular killing. Human recombinant interferon-gamma has been
shown to activate both of these mechanisms against Legionella infection. Expo-
sure of pulmonary macrophages and peripheral blood monocytes to recombinant
interferon-gamma but not recombinant granulocyte-macrophage colony-stimu-
LEGIONELLA PNEUMOPHILA 103
lating factor primed cells for augmentation of H 2Ü 2 production. 46 Macrophages

can be activated in vitra by interferon-gamma treatment or in vivo with Mycobac-
terium bovis BCG to inhibit the intracellular multiplication of L. pneumophila,
demonstrating that activated macrophages may be nonspecifically armed against
Legionella. 43 ,45,60,65-67 Although the activated monocytes and alveolar macro-
phages inhibit the intracellular multiplication of L. pneumophila, killing is not
enhanced.
Recombinant human TNF was shown to activate PMN-enriched suspensions
to kill L. pneumophila in a dose-dependent manner and to reduce mortality in
TNF-treated mice,68 Legionella antigens induced both TNF and interferon-
gamma in nonimmune human leukocyte culture. These cytokines were shown to
stimulate the bactericidal activity of PMN's against L. pneumophila in vitro,69
Legionella-infected monocytes were shown to produce TNF, and the addition of
interferon-gamma resulted in augmented production of TNF in a synergistic
fashion.
The cytolytic activity of human peripheral blood leukocytes (NK-like cells)
was augmented by activation of effector cells by interleukin 2. IL-2-activated
killer cells were able to lyse Legionella-infected monocytes. 63 NK cells have also
been shown to res pond in vivo and in vitra to stimulation by L. pneumophila by
producing interferon-gamma and by increased cytolytic activity. 68 lt is likely then
that both neutrophils and NK-like cells are responsible for elimination of L.
pneumophila.
5. VIRULENCE FACfORS
Strains of L. pneumophila differ in virulence. Putative factors affecting viru-

lence include sensitivity to complement-mediated bactericidal activity,70 toxicity
for alveolar macrophages,70 susceptibility to oxygen-dependent killing,71 plasmid
content,72,73 and intracellular replication 74 (Table I). Virulent strains fail to bind
components of complement to the cell surface and are resistant to the bactericidal
activity of normal human serum, whereas avirulent strains are serum-sensitive
and bind C3b to the cell surface. Virulent strains are cytotoxic for guinea pig
alveolar macrophages whereas avirulent strains exhibit very little toxicity. Fi-
nally, a toxic factor associated with the bacterial cell of L. pneumophila has been
shown to inhibit protein synthesis of Chinese hamster ovary cells, U937 cells, and
human monocytes. 75
Differences in strain virulence have also been attributed to the ability of
pathogenic strains to survive in aerosols 76 and their ability to multiply in proto-
zoa. 77 ,78In addition, multiple strains of L. pneumophila, serogroup 1, may colonize
a water distribution system but only a select strain will cause disease in patients
exposed to the water. 79- 81 Monoclonal antibody subtyping of strains of L. pneumo-
phila, serogroup 1, has shown that a surface epitope recognized by the monoclonal
antibody Mab-2 may be associated with virulence.
Virulence of L. pneumophila can be ascertained by comparing the ability of
104 JANET E. STOUT and VIGrOR L. YU
TABLE I
Properties of Legionella That Predispose
to Pathogenicity and Virulence
General eharaeteristies
Ubiquitous in nature
Range of temperature and physicoehemical requirements
Symbiotic relationship with soillwater unieellular mieroorganisms
Deereased triggering of respiratory burst
Inhibition of phagosome-lysosome fusion
Inhibition of phagosome acidifieation
Serum resistanee
Deereased complement binding
Endotoxin (LPS)
Toxins
1.2 kDa toxin inhibits respiratory burst
Protease(s) responsible for tissue damage:
Hemolysin (Iegiolysin)
Cytotoxin
38-kDa major seeretory protein (MSP), a zine metalloprotease (elastase?)
Surfaee molecules
Maerophage infeetivity potentiator (mip gene produet)
Monoclonal antibody epitope (Mab-2)
58-kDa major outermembrane protein-Heat shoek protein
L. pneumophila strains to multiply in cultures of the protozoa Tetrahymena pyriformis

to their ability to infect and kill guinea pigs. The protozoan model was able to
assay for the ability of Legionella to multiply intracellularly. The animal model was
required, however, to assay for additional virulence properties, such as toxicity.78
Other putative virulence factors of L. pneumophila include exoproteases and a
58-60 kDa protein antigen. The extracellular proteases have been shown to
exhibit cytotoxic activity82 and to produce pulmonary lesions typical of Legion-
naires' disease after intranasal administration to guinea pigS. 83 In vivo produc-
tion of a tissue-destructive protease was demonstrated in the lungs of guinea pigs
that were challenged with alethal aerosol dose of L. pneumophila. 84 Quinn and
Tompkins cloned the genetic sequence encoding a 38-kDa protease from L.
pneumophila serogroup 1 and have shown that a single polypeptide is responsible
for its proteolytic, hemolytic and cytotoxic properties. 85 This protease is referred
to as the major secretory protein (MSP) since it is the single most abundant
polypeptide produced by L. pneumophila. It is a zinc metalloprotease with activity
on a variety of substrates including casein and collagen. 86
The cloned protease DNA sequence, amino acid sequence of the protein, and
competitive inhibition studies indicate that the L. pneumophila protease is struc-
tu rally and functionally similar to the zinc metalloprotease of Pseudomonas
aeruginosa elastase. 87 Legionella MSP has also been shown to playa role in the
development of cellular immunity to L. pneumophila. 88 Vaccination of guinea pigs
LEGIONELLA PNEUMOPHlLA 105
with MSP-induced protective cell-mediated immunity against lethai aerosol chal-

lenge with L. pneumophila. 89 In addition MSP is a dominant antigen recognized by
guinea-pig T cells following either sublethai infection with virulentL. pneumophila
or vaccination with an avirulent mutant. 37 ,88,89 The gene encoding this extracellu-
lar protease has been mutated in vitro. 9o Comparisons of intracellular growth of
the wild type and the mutant demonstrated that MSP was not required for L.
pneumophila multiplication and subsequent lysis of macrophages. 90,91
Another protease, a hemolysin, which lyses human erythrocytes, has been
named legiolysin. 91 This protease was not cytolytic for Vero or CHO cells, whereas
the MSP gene product was cytotoxic for these cell lines. Legiolysin is also
responsible for the production of the melanin-like pigment produced by Le-
gionella following growth on tyrosine-containing media. 92
Legionella has also been shown to induce procoagulant activity in mono-
nuclear cells, which may be related to its ability to cause prosthetic-valve endocar-
ditis. 93 ,94
An L. pneumophila gene has been shown to encode a 24-kDa species-specific
surface protein necessary to initiate macrophage infection. 95 The cloned L.
pneumophila gene has been referred to as mip, macrophage infectivity potentiator.
Using site-specific mutagenesis, a mutant of L. pneumophila was constructed that
lacked the 24-kDa surface protein. This mutant was 80-fold less infective for
explanted human alveolar macrophages. 96 In in vivo studies using guinea pigs,
the mip-mutant strain was also shown to be attenuated, producing fewer lethai
infections and a slower-progressing disease. 96 High-stringency hybridization
of mip probes to DNAs from other Legionella species confirmed that mip is
specific to L. pneumophila, however, an analog of mip was found in all species of
Legionella. 97 It has been speculated that the function of mip is related to the
intracellular survival of L. pneumophila, rather than to its uptake or growth. 98 A
mip-like protein has been identified in the membrane or another intracellular
microorganism, Chlamydia trachomatis, suggesting a more universal antiphago-
cytic function for mip.99
A thymidine auxotrophic mutant of L. pneumophila has been isolated and
shown to be incapable of growth within human monocytes. lOO The relevance of
this finding for Legionella pathogenesis is uncertain.
The 58-60 kDa immunodominant antigen of L. pneumophila has been shown
to be a heat-shock protein and a major humoral and cellular antigen. 99,102 The
gene encoding this antigen has been sequenced.I°1,103 The extent to which heat-
shock proteins are expressed by intracellular bacteria has been suggested as a
determining factor in the resolution of disease and development of immunity.lOl
6. VACCINE DEVELOPMENT
As discussed, MSP is an immunodominant antigen of L. pneumophila. Guinea

pigs immunized subcutaneously with MSP develop humoral and cell-mediated
immune responses to MSP that result in protective immunity against lethai
aerosol challenge with L. pneumophila, serogroup 1. 89 MSP has also been shown to
induce protective immunity across serogroups of L. pneumophila but not against
other Legionella species.l°4 For these reasons, MSP has been considered as a
candidate for a vaccine preparation. No Legionella vaccine has yet been tested in
humans.
REFERENCES
1. Fraser, D. w., Tsai, T., Orenstein, w., Parkins, W E., Beecham, H.j., Sharrar, R. G., Harris,
j., Mallison, G. F., Martin, S. M., McDade,j. E., Shepard, C. C., and Brachman, P. S., 1977,
Legionnaires' disease: Description of an epidemic of pneumonia, N. Engl. J. Med.
297:1183-1197.
2. Stout, j. E., Yu, V. L., and Best, M., 1985, Ecology of Legionell,a pneumophila within water
distribution systems, Appl. Environ. Mierobiol. 49:221-228.
3. Wadowsky, R. M., Wolford, R., McNamara, A. M., and Yee, R. B., 1985, Effect of
temperature, pH, and oxygen level on the multiplication of naturally occurring Legionella
pneumophila in potable water, Appl. Environ. Mierobiol. 49:1197-1205.
4. Tison, D. L., Pope, D. H., Cherry, WB., and Fliermans, E., 1980, Growth of LegioneUa
pneumophila in association with blue-green algae (cyanobacteria), Appl. Environ. Mierobiol.
39:456-459.
5. Fields, B. S., Shotts, E. B., Feeley, j. C., Gorman, G. W, and Martin, W T., 1984,
Proliferation of Legionella pneumophila as an intracellular parasite of the ciliated protozoan
Tetrahymena pyriformis, Appl. Environ. Mierobiol. 47:467-47l.
6. Rowbotham, T. j., 1986, Current views on the relationships between amoebae, Legionellae,
and man, Isr. j Med. Sei. 22:678-689.
7. Yamamoto, Y., Klein, T. w., Newton, C., and Friedman, H., 1992, Differing macrophage
and Iymphocyte roles in resistance to Legionella pneumophila infection,j Immunol. 148:584-
589.
8. Aeba, R., Stout, j. E., Francalancia, N. A., Keenan, R. j., Duncan, A. j., Yousem, S. A.,
Burckart, G. j., Yu, V. L., and Griffith, B. P., 1993, Aspects oflung transplantation which
contribute to increased severity of pneumonia: An experimental study,j Thoraeie Cardio-
vase. Surg. 106:449-457.
9. Brady, M., 1989, Nosocomial Legionnaires' disease in a children's hospital, J Pediatr.
115:46-50.
10. Hervas, j., Lopez, P., de la Fuente, A., and Alomar, P., 1988, Multiple organ system failure
in an infant with Legionella infection, Pediatr. Infect. Dis. j 7:671-673.
11. Muder, R. R., Yu, V. L., and Woo, A., 1986, Mode of transmission of L. pneumophila: A
critical review, Arch. Int. Med. 146:1607-1612.
12. Yu, V. L., and Stout, j. E., 1992, Sporadic cases of Legionnaires' disease, N. Engl. J Med.
326:1700-170l.
13. Stout, j. E., Yu, V. L., Vickers, R. M., Zuravleff, j., Best, M., Brown, A., Yee, R. B., and
Wadowsky, R., 1982, Ubiquitousness of Legionella pneumophila in the water supply of a
hospital with endemic Legionnaires' disease, N. Engl. j Med. 306:466-468.
14. Markowitz, L., Tompkins, L., Wilkinson, H., Barbaree, j., and Broome, C. v., 1984,
Transmission of nosocomial Legionnaires' disease in heart trans plant patients, Program
and Abstracts of the 24th Intersci. Conf. Antimicrob. Ag. Chemother., Amer. Soc. Micro-
bio!., Washington D.C. 170 (abstract).
15. Woo, A. H., Goetz, A., and Yu, V. L., 1992, Transmission of Legionella by respiratory
equipment and aerosol generating devices: A review, Chest 102:1586-1590.
LEGJONELLA PNEUMOPHILA 107
16. Muder, R. R., Stout,]. E., and Yee, ]. C., 1992, Isolation of Legionella pneumophila sero-
group 5 from empyema following esophageal perforation: source of the organism and
mode of transmission, Chest 102:1601-1603.
17. Yu, V. L., Stout,]., Zuravleff,]., and Brown, A., 1981, Aspiration of contaminated water
may be mode of transmission for Legionella pneumophila just prior to a nosocomial outbreak
of Legionnaires' disease, Intersei. Conf. Antirnierob. Ag. Chemother., #293, Chicago.
18. Marrie, T.]., Haldane, D., MacDonald, S., Clarke, K., Fanning, C., Lefort-Jost, S., Bezan-
son, G., and Joly,]., 1991, Control of endemie nosocomial Legionnaires' disease by using
sterile potable water for high risk patients, Epidemiol. Infeet. 107:591-605.
19. Glick, T. H., Gregg, M. B., Berman, B., Mallison, G., Rhodes, W. w.,Jr., and Kassanoff, 1.,
1978, Pontiac fever: An epidemie of unknown etiology in a health department: Clinical and
epidemiologie aspects, Am. j Epidemiol. 149-160.
20. Arnow, P., Chou, T., Weil, D., Shapiro, E., et al., 1982, Nosocomial Legionnaires; disease
caused by aerosolized tap water from respiratory devices, j Infeet. Dis. 146:460-67.
21. Kaan, ]., Simoons-Smith, A. M., and MacLaren, D., 1985, Another source of aerosol
causing nosocomial Legionnaires' disease,j In! 11:145-148.
22. Mastro, T. D., Fields, B. S., Breiman, R. F., Campbell,]., Plikaytis, D., and Spika,]. S., 1991,
Nosocomial Legionnaires' disease and use of medication nebulizers, j Infect. Dis. 163:
667-671.
23. Wright,]. B., Athar, M. A., van Olm, T. M., Wootliff,]. S., and Costerton, ]. S., 1989,
Atypicallegionellosis: Isolation of Legionella pneumophila serogroup 1 from a patient with
aspiration pneumonia,j Hosp. Infect. 13:187-190.
24. Blatt, S. P., Parkinson, M. D., Pace, E., et al., 1992, Nosocomial Legionellosis: A one year
case-control and environmental investigation, International Symposium on Legionella,
111-6 (abstract), Am. Soc. Microbiol., Orlando, Florida.
25. Johnson, ]. T., Yu, V. L., Best, M., Vickers, R. M., et al., 1985, Nosocomiallegionellosis
uncovered in surgical patients with head and neck cancer: Implications for epidemiologie
reservoir and mode of transmission, Laneet 2:298-300.
26. Davis, G. S., Winn, w., Gump, D. w., and Beaty, H., 1983, The kinetics of early infiamma-
tory events during experimental pneumonia due to Legionella pneumophila in guinea pigs,j
Infeet. Dis. 148:823-835.
27. Horwitz, M. A., and Silverstein, S. C., 1980, The Legionnaires' disease bacterium (Legio-
nella pneumophila) multiples intracellularly in human monocytes,j Clin. Invest. 66:441-450.
28. Horwitz, M. A., and Silverstein, S. C., 1981, Interaction of the Legionnaires' disease
bacterium (Legionella pneumophila) with human phagocytes: II Antibody prornotes binding
of L. pneumophila, j Exp. Med. 153:398-406.
29. Nash, T. w., Libby, D., and Horwitz, M. A., 1984, Interaction between the Legionnaires'
disease bacterium (L. pneumophila) and human alveolar macrophages: Infiuence of anti-
body, lymphokines, and hydrocortisone,j Clin. Invest. 74:771-782.
30. Horwitz, M. A., 1984, Phagocytosis of the Legionnaires' disease bacterium (Legionella
pneumophila) occurs by a novel mechanism: Engulfment within a pseudopod coil, Cell
36:27-33.
31. Payne, N. R., and Horwitz, M. A., 1987, Phagocytosis of Legionella pneumophila is mediated
by human monocyte complement receptors,j Exp. Med. 166:1377-1389.
32. King, C. H., Fields, B. S., Shotts, E. B.,Jr., and White, E. H., 1991, Effects of cytochalasin D
and methylamine on intracellular growth of Legionella pneumophila in amoebae and human
monocyte-like cells, Infeet. Immun. 59:758-763.
33. Holden, E. P., Winkler, H. H., Wood, D. 0., and Leinbach, E. D., 1984, Intracellular growth
of Legionella pneumophila within Aeanthamoeba eastellanii neff, In/eet. Immun. 45:18-24.
34. Newsome, A. L., Baker, R. L., Miller, R. D., and Arnold, R. R., 1985, Interactions betwten
Naelglaria fowleri and Legionella pneumophila, Infect. Immun. 50:449-542.
108 ]ANET E. STOUT and VICTOR L. YU
35. TyndalI, R. L., and Domingue, E. L., 1982, Cocultivation of Legionella pneumophila and free-
living amoebae, Appl. Environ. Mierobiol. 44:954-959.
36. Rolstad, B., and Berdal, B., 1981, Immune defenses against Legionella pneumophila in rats,
Inftct. Immun. 32:805-812.
37. Breiman, R. F., and Horwitz, M. A., 1987, Guinea pigs sublethally infected with aerosolized
Legionella pneumophila develop humoral and cell-mediated immune responses,] Exp. Med.
164:799-811.
38. Davis, G. S., Winn, W, and Christman, j. W, 1989, Role of recruited nonimmunospecific
defenses in Legionnaires' disease, Ghest 95:232S-233S.
39. Bellinger-Kawahara, C. G., and Horwitz, M. A., 1987, The major outer protein is a
prominent acceptor moleeule for complement component C3 on Legionella pneumophila,
Glin. Res. 35:468A.
40. Locksley, R. M., ]acobs, R. F., Wilson, C. B., Weaver, W M., and Klebanoff, S. j., 1982,
Susceptibility of Legionella pneumophila to oxygen-dependent microbial systems,] Immunol.
2192-2197.
41. Lochner,j. E., Friedman, R. L., Bigley, R. H., and Iglewski, B. H., 1983, Effect of oxygen-
dependent antimicrobial systems on Legionella pneumophila, Infeet. Immun. 39:487-489.
42. Summersgill, j. T., Raff, M. j., Miller, R. D., 1988, Interactions of virulent and avirulent
Legionella pneumophila with human polymorphonuclear leukocytes, Mierobiol. Pathogen.
5:41-47.
43. Skerrett, S. j., and Martin, T. R., 1991, Alveolar macrophage activation in experimental
legionellosis,] Immunol. 147:337-345.
44. Byrd, T. F., and Horwitz, M. A., 1989, Interferon gamma-activated human monocytes
downregulate transferrin receptors and inhibit the intracellular multiplication of Legionella
pneumophila by limiting the availability of iron,] Glin. Invest. 83:1457-1465.
45. Bhardwaj, N., Nash, T., and Horwitz, M. A., 1986, Interferon-gamma-activated human
monocytes inhibit the intracellular multiplication of L. pneumophila,j. Immunol. 137:2662-
2669.
46. ]ensen, W A., Rose, R. M., Wasserman, A. S., Kalb, T. H., Anton, K., and Remold, H. G.,
1987, In vitro activation of the antibacterial activity of human pulmonary macrophages by
recombinant gamma interferon,] Infeet. Dis. 155:574-577.
47. Goldoni, P., Visca, P., Pastoris, M . C., Valenti, P., and Orsi, N., 1991, Growth of Legionella
spp. under conditions ofiron restriction,] Med. Mierobiol. 34:113-118.
48. Yammamoto,Y., Klein, T. W, Newton, C. A., Widen, R., and Friedman, H., 1987, Differen-
tial growth of Legionella pneumophila in guinea pig versus mouse macrophage cultures,
Inftct. Immun. 55:1369-1374.
49. Yoshida, S. 1., and Mizuguchi, Y., 1986, Multiplication of Legionella pneumophila Phila-
delphia-I in cultured peritoneal macrophages and its correlation to susceptibility of
animals, Gan.] Mierobiol. 32:438-442.
50. Berendt, R. F., Young, H. W, Allen, R. G., and Knutsen, G. L., 1980, Dose-response of
guinea pigs experimentally infected with aerosols of Legionella pneumophila,] Inftet. Dis.
141:186-192.
51. Davis, G. S., Winn, W C., Gump, D. W, and Craighead, j. E., 1982, Legionnaires'
pneumonia after aerosol exposure in guinea pigs and rats, Am. Rev. Respir. Dis. 126: 1050-
1057.
52. Skerrett, S. j., Schmidt, R., and Martin, T. R., 1989, Impaired clearance of aerosolized L.
pneumophila in corticosteroid treated rats: A model of Legionnaire's disease in the compro-
mised host,] Inftet. Dis. 160:261-268.
53. Yamamoto, Y., Klein, T. W, Newton, C. A., Widen, R., and Friedman, H., 1988, Growth of
Legionella pneumophila in thioglycolate-elicited peritoneal macrophages from AI] mice,
LEGIONELLA PNEUMOPHILA 109
54. Kishimoto, R. A., Kastello, M. D., White,j. D., Shirey, F. G., MeGann, V. G., Larson, E. w.,
and Hedlung, K. w., 1979, In vitro interaetion between normal eynomolgus monkey alveolar
maerophages and Legionnaires' disease baeteria, Infect. Immun. 25:761-763.
55. Yamamoto, Y., Klein, T. w., Newton, C., and Freidman,H., 1992, Differing maerophage
and Iymphoeyte roles in resistanee to Legionella pneumophila infeetion, J Immunol. 148:
584-589.
56. Plouffe,j. F., and Baird,1. M., 1982, Cord blood Iymphoeyte transformation responses or L.
pneumophila,Immunol. 9:119-120.
57. Friedman, H., Widen, R., Lee, 1., and Klein, T., 1983, Cellular immunity to L. pneumophila
in guinea pigs assessed by direet and indireet migration inhibition: reaetion in vitro, Infec.
Immun. 41:1132-1137.
58. Friedman, H., Widen, R., Klein, T., Searles, L., and Cabrian, K., 1984, L. pneumophila-
indueed blastogenesis of murine lymphoid eells in vitro, Infect. Immun. 43:314-319.
59. Wong, K. H., MeMaster, P. R. B., Feeley,J. C., Arko, R.J., Sealla, W. 0., and Chandler, F. w.,
1980, Deteetion of hypersensitivity to L. pneumophila in guinea pigs by skin test, Curr.
60. Nikaido, Y., Yoshida, S., Goto, Y., Mizuguehi, Y., and Kuroiwa, A., 1989, Maerophage-
activating T-eell factor(s) produeed in an early phase of Legionella pneumophila infeetion in
guinea pigs, Infect. Immun. 57:3458-3465.
61. Horwitz, M. A., 1983, Cell-mediated immunity in Legionnaires' disease, J Clin. Invest.
71:1686-1697.
62. Blanchard, 0. K., Klein, T. w., Friedman, H., and Stewart, W. E., 1985, Kineties and
charaeterization of interferon produetion by murine spleen eells stimulated with Legionella
pneumophila antigens, Infect. Immunity. 49:719-723.
63. Blanchard,D. K., Stewart, W. E., Klein, T. w., Friedman, H., and Djeu,J. Y., 1987, Cytolytic
aetivity of human peripheral blood leukoeytes against Legionella pneumophila-infeeted
monoeytes: Charaeterization of the effeetor,J Immunol. 139:551-556.
64. Klein, T. w., Newton, C. A., Blanehard, K., Widen, R., and Friedman, H., 1987, Induetion
of interleukin 1 by Legionella pneumophila antigens in mouse and human mononuclear
leukocyte eultures, Zbl. Bakt. Hyg. A 265:462-471.
65. Nash, T. w., Libby, 0. M., and Horwitz, M. A., 1988, IFN-gamma-aetivated human alveolar
macrophages inhibit the intracellular multiplication of Legionella pneumophila,} Immunol.
140:3978-3981.
66. Gibson, 0., Baskerville, A., Ashworth, L., and Fitzgeorge, R. B., 1985, Non-specifie
protection against pulmonary L. pneumophila infeetion in guinea pigs immunized and
challenged with myeobacteria, Br. J Exp. Pathol. 66:333-344.
67. Horwitz, M. A., Lewis, w., and Cohn, Z. A., 1984, Defective produetion of monoeyte-
aetivating cytokines in lepromatous leprosy, J Exp. Med. 159:666-678,
68. Blanchard, 0. K., Djeu, J. Y., Klein, T. w., Friedman, H., and Stewart, W. E., 11., 1988,
Proteetive effeets of tumor neerosis factor in experimental Legionella pneumophila infections
of mice via activation of PMN funetion, J Leuk. Biol. 43:429-435.
69. Blanehard, 0.1., Friedman, H., Klein, T. w., and Djen,J. Y.,1989, Induetion ofinterferon-
gamma and tumor necrosis faetor by Legionella pneumophila: Augmentation of human
neutrophil bactericidal activity,J Leuk. Biol. 45:538-545.
70. Caparon, M., and Johnson, w., 1988, Maerophage toxicity and complement sensitivity of
virulent and avirulent strains of Legionella pneumophila, Rev. Infect. Dis. 10:8377-384.
71. Jepras, R. 1., and Fitzgeorge, R. B., 1986, The effect of oxygen-dependent"antimierobial
systems on strains of Legionella pneumophila of different virulence, J Hyg" (London) 97:
61-69.
72. Brown, A., Vickers, R. M., Eider, E. M., et al. , 1982, Plasmid and surface antigen markers of
endemie and epidemie Legionella pneumophila strains, J Clin. Microbiol. 16:230-235.
llO JANET E. STOUT and VICTOR L. YU
73. Maher, W E., Plouffe, J. F., and Para, M. F., 1983, Plasmid profiles of clinical and
environmental isolates of Legionella pneumophila serogroup I, I Clin. Microbiol. 18:1422-
1423.
74. Horwitz, M. A., 1987, Characterization of avirulent mutant Legionella pneumophila that
survive but do not multiply within human monocytes,l Exp. Med. 166:1310-1328.
75. McCusker, K. T., Braaten, B. A., Cho, M. W, and Low, D. A., 1991, Legionella pneumophila
inhibits protein synthesis in Chinese hamster ovary cells, Infoct. Immun. 59:240-246.
76. Dennis, P. J., and Lee, J. v., 1988, Differences in aerosol survival between pathogenic and
non-pathogenic strains of Legionella pneumophila serogroup 1,1 Appl. Bacteriol. 65:135-141.
77. Fields, B. S., Barbaree, J. M., Sanden, G. N., and Morrill, W E., 1990, Virulence of a
Legionella anisa strain associated with Pontiac fever: An evaluation using protozoan, cell
culture, and guinea pig models, Infect. Immun. 58:3139-3142.
78. Fields, B. S., Barbaree, J. M., Shotts, E., Feeley, J. C., Morill,W. E., Sanden, G. N., and
Dykstra, M. J., 1986, Comparison of guinea pig and protozoan models for determining
virulence of Legionella species, Infect. Immun. 53:553-559.
79. Dournon, E., Bibb, W. F., Rajagopalan, P., Desplaces, N., and McKinney, R. M., 1988,
Monoclonal antibody reactivity as a virulence marker for Legionella pneumophila sero-
group 1 strains,l Infect. Dis. 157:496-501.
80. Plouffe, J. F., Para, M. F., Maher, W. E., Hackman, B., and Webster, L., 1983, Subtypes of
Legionella pneumophila serogroup 1 associated with different attack rates, Lancet 2:649-650.
81. Stout, J. E., Joly, J., Para, M., Plouffe, J., Ciesielski, C., Blaser, M. J., and Yu, V. L., 1988,
Comparison of molecular methods for subtyping patients and epidemiologically-linked
environmental isolates of L. pneumophila, l Infect. Dis. 157:486-494.
82. Keen, M. G., and Hoffman, P. S., 1989, Characterization of a Legionella pneumophila extra-
cellular protease exhibiting hemolytic and cytotoxic activities, Infect. Immun. 57:732-738.
83. Baskerville, A., Conlan, J. W, Ashworth, L. A., and Dowsett, A. B., 1986, Pulmonary
damage caused by a protease from L. pneumophila, Br. I Exp. Pathol. 67:527-536.
84. Conlan, J. W, Williams, A., and Ashworth, L. A. E., In vivo production of a tissue-
destructive protease by Legionella pneumophila, I Gen. Microbiol. 134: 143-149.
85. Quinn, F. D., and Tompkins, L. S., Analysis of a cloned sequence of Legionella pneumophila
encoding a 38 kD metalloprotease possessing haemolytic and cytotoxic activities, Mol.
86. Dreyfus, L. A., and Iglewski, B. H., 1986, Purification and characterization of an extra-
cellular protease of Legionella pneumophila, Infoct. Immun. 51:736-743.
87. Black, W. J., Quinn, F. D., and Tompkins, L. S., 1990, Legionella pneumophila zinc metallo-
protease is structurally and functionally homologous to Pseudomonas aeruginosa elastase,l
Bacteriol. 172:2608-2613.
88. Blander, S. J., Breiman, R., and Horwitz, M. A., A live avirulent mutant L. pneumophila
vaccine induces protective immunity against lethai aerosol challenge, I Clin. Invest. 83:
810-815.
89. Blander, S., and Horwitz, M.A., 1989, Vaccination with the major secretory protein of L.
pneumophila induces cell-mediated immunity in a guinea pig model of Legionnaires'
disease,l Exp. Med. 100:691-705.
90. Szeto, L., and Shuman, H. A., 1990, The Legionella pneumophila major secretory protein, a
protease, is not required for intracellular growth or cell killing, Infect. Immun. 58:2585-
2592.
91. Hacker,J., Ott, M., Ludwig, B., and Rdest, u., 1991, Intracellular survival and expression of
virulence determinants of Legionella pneumophila, Infection (SuPPI 4) 19:198-201.
92. Wintermeyer, E., Rdest, u., Ludwig, B., Debes, A., and Hacker,J., 1991, Characterization of
legiolysin (lly), responsible for haemolytic activity, colour production and ßuorescence of
Legionella pneumophila, Mol. Microbiol. 5(5):1135-1143.
LEG/ONELLA PNEUMOPHILA III
93. Miragliotta, G., and Fumarola, D., 1989, Production of procoagulant activity, tissue factor-
like, by human mononuclear cells and its role in the pathogenesis of Legionella prosthetic
valve endocarditis, Medical Hypotheses 30:73-74.
94. Miragliotta, G., Semeraro, N., Marcuccio, L., and Fumarola, D., 1982, Legionella pneumo-
phila and related organisms induce the generation of procoagulant activity by peripheral
mononuclear cells in vitro, Infection 10:215.
95. Cianeiotto, N. P., Eisenstein, B. 1., Mody, C. H., Toews, G. B., and Engleberg, N. C., 1989, A
Legionella pneumophila gene encoding a species-specific surfaee protein potentiates initia-
tion of intracellular infection, Inftct. Immun. 57:1255-1262.
96. Cianciotto, N. P., Eisenstein, B. 1., Mody, C. H., and Engleberg, N. C., 1990, A mutation
in the mip gene results in an attenuation of Legionella pneumophila virulence,] Infixt. Dis.
162:121-126.
97. Cianciotto, N. P., Bangsborg,J. M., Eisenstein, B.I., and Engleberg, N. C., 1990, Identifica-
tion of mip-like genes in the genus Legionella, Inftct. Immun. 58:2912-2918.
98. Engleberg, N. C., and Eisenstein, B. 1., 1991, Progress in the pathogenesis of Legionella
pneumophila, Microbiol. Pathogenesis 10:11-13.
99. Lundemose, A. G., Birkelund, S., Fey, S. J., Mose Larsen, P., and Christiansen, G., 1991,
Ghlamydia trachomatis contains a protein similar to the Legionella pneumophila mip gene
produet, Mol. Microbiol. 5(1):109-115.
100. Mintz, C. S., Chen, J. X., and Shuman, H. A., 1988, Isolation and eharaeterization of
auxotrophic mutants of Legionella pneumophila that fail to multiply in human monocytes,
Inftct. Immun. 56:1449-1455.
101. Hoffman, P. S., Houston, L., and Butler, C. A., Legionella pneumophila htpAb heat shock
operon: Nucleotide sequence and expression of the 60-Kilodalton antigen in L. pneurnophila-
infeeted HeLa cells, Infect. Immun. 58:3380-3387.
102. Sampson,J., Plikaytis, B., and Wilkinson, H., 1986, Immunologie response ofpatients with
legionellosis against m~or protein-containing antigens of Legionella pneumophila sero-
group 1 as shown by immunoblot analysis,] Glin. Microbiol. 23:92-99.
103. Sampson,J. S., O'Connor, S. P., Holloway, B. P., Plikaytis, B. B., Carlone, G. M., and Mayer,
L. W, 1990, Nucleotide sequence of htpB, the Legionella pneumophila gene encoding the 58-
Kilodalton (kDa) common antigen, formerly designated the 60-kDa common antigen,
Infect. Immun. 58:3154-3157.
104. Blander, S. J., and Horwitz, M. A., 1991, Vaccination with the major secretory protein of
Legionella induees humoral and eell-mediated immune response and protective immunity
across different serogroups of Legionella pneumophila and different speeies of Legionella,]
Immunol. 147:285-291.
7
Anaerobic Bacterial Infections

of the Lung
JOHN G. BARTLETT
l.INTRODUCfION
Anaerobic bacteria are relatively common pulmonary pathogens and they are
especially frequent with aspiration pneumonia and its suppurative sequelae: lung
abscess and empyema. The clinical features and bacteriology of this condition
have been weIl described with extensive studies done in two periods. The first
period was the turn of the century when anaerobic bacteria were initially discov-
ered as important causes of empyema 1 through the period of 1927-30 when
David Smith at Duke completed classical studies of lung abscess. 2-4 The second
period of study extended from approximately 1968 through 1980 when there was
extensive use of transtracheal aspiration as a mechanism to obtain uncontami-
nated specimens from the lower respiratory tract; access to this culture source
combined with a renewed interest in cultivating oxygen-sensitive bacteria permit-
ted clinical and bacteriological features to be redefined to include therapeutic
implications. Despite these advances, it should be acknowledged that the role of
anaerobic bacteria as pulmonary pathogens is commonly overlooked. It is still
rare to have the bacteriologic diagnosis established at the present time and there
continues to be considerable controversy regarding antibiotic options. The pres-
ent review is based on the author's experience with 193 bacteriologically con-
firmed cases studied in the 1970s5 ,6 combined with some of the more recent work
JOHN G. BARTLETT • Department of Medicine, Johns Hopkins University School of Medi-

eine, Baltimore, Maryland 21205.
113
114 JOHN G. BARTLETT
dealing with taxonomic changes, definition of virulence factors of an aerobic

bacteria, and the potential role of newer antimicrobial agents.
2. INCIDENCE
Anaerobic bacteria pose areal challenge to the clinician because specialized

methods are necessary to obtain specimens that are free of contamination from
the upper airways. The usual specimen sources that satisfy this definition require
invasive procedures such as transtracheal aspiration, transthoracic aspiration,
bronchoscopy with quantitative cultures, open-Iung biopsy, or thoracentesis.
Anaerobes also pose achallenge to the microbiologist because they are relatively
fastidious and most of the infections involve a multitude of bacteria that are often
difficult to separate and identify using modern taxonomie schema. U nfor-
tunately, these requirements have precluded extensive study of the role of
anaerobes in many settings, although the available data from reports dealing with
specific syndromes are listed in Table 1.
Most of the published reports deal with the role of an aerobic bacteria in
aspiration pneumonia or lung abscess where the recovery rates range from 62 to
100%.6--14 The usual specimen sources in these studies are transtracheal aspira-
tion and transthoracic aspiration. Perhaps the best study is by Beerens and
Tahon-Castel9 who used transthoracic needle aspiration of lung ab sees ses and
recovered an aerobic bacteria in 85% of 26 cases. A more recent version of this
study is by Gudiol et al. ll who used similar techniques and found an aerobic
bacteria in 90% of 41 cases of lung abscess.
Empyema is obviously more easily studied because of the relative ease of
obtaining pleural fluid for anaerobic culture. Here there has been a major change
in bacteriology from the preantibiotic era when S. pneumoniae accounted for 60-
70% of all empyema cases. Studies at that time suggested anaerobes accounted for
only 5_7%.1 5 ,16 More recently, there has been a sharp decline in the frequency of
pneumococcal empyema, presumably ascribed to penicillin usage. These reports
indicate a lO-fold decline in the frequency of empyema in general and a marked
change in the bacteriology of remaining cases. The pneumococcus now accounts
for only about 5-10%, and anaerobes account for 25_40%.9,17-22 The highest
yield was our report of 83 patients seen at Cooke County Hospital in Chicago and
two VA hospitals in Los Angeles; anaerobes were recovered in 63 (76%)P This
relatively high yield may re fleet the population served by these institutions which
included a large number of patients who were aspiration prone because of
alcoholism and other debilitating diseases. Also noteworthy in these studies is the
fact that all bacteriology was done by anaerobic research laboratories with a
specific interest in detection of anaerobes.
There are relatively few studies dealing with the frequency of an aerobic
bacteria among unselected cases of community-acquired pneumonia. Perhaps the
best available information is a study by Ries et al. 23 who attempted transtracheal
aspirations for an aerobic culture in all patients admitted to a general Philadelphia
ANAEROBIC BACTERIAL INFECTIONS OF THE LUNG 115
TABLE I
Incidence of Anaerobic Bacterial Infection
of the Lung and Pleura
No. No. with
Clinical setting studied anaerobes
Pulmonary abscess
Bardett et al.6--8 57 53 (93%)
Beerens and Tahon-Castel 9 26 22 (85%)
Brook and Finegold lO 10 9 (90%)
Gudiol et al.l! 41 37 (90%)
Aspirin pneumonia
Bardett et al. 6 •8 70 61 (87%)
Gonzalez-C and Calie l2 17 17 (100%)
Lorber and Swenson 13 47 29 (62%)
Brook and Finebold l4 74 69 (93%)
Empyema
Bardett et al. 17 83 63 (76%)
Beerens and Tahon-Castel9 45 23 (51 %)
Sullivan et al. 18 482 42 (19%)
Varkey et az.t 9 72 28 (39%)
Mavroudis et al. 20 100 25 (25%)
Grant and Finley 21 90 26 (32%)
Lemmer et al. 22 70 20 (30%)
Community-acquired pneumonia
Reis et aU 3 89 29 (33%)
Pollack et al. 24 74 16 (22%)
Hospital-acquired pneumonia
Barlett et al. 25 159 56 (35%)
hospital with a diagnosis of pneumonia. Only about one-third of the patients with
this diagnosis were actually studied, but among the 89 participants, anaerobic
bacteria were recovered from 29 (33%). A possibly analogous study by Pollack
et al. in Mobile, Alabama, used the bronchoscopy brush with the protected
catheter for quantitative cultures; this showed anaerobes in 16 of 75 unselected
patients (21 %).24 These studies indicated that anaerobic bacteria were second only
to the pneumococcus among all patients with community-acquired pneumonia.
The suggestion is that anaerobic bacteria are the dominant organisms in aspira-
tion pneumonia and its sequelae, lung abscess; anaerobes play an important role
in empyema, and they probably playa far greater role than generally recognized
in community-acquired pneumonia.
With regard to nosocomial pneumonia, the single study dealing with the
possible role of anaerobic bacteria was a prospective analysis done by our group
at the Sepulveda VA Hospital in Los Angeles in the early 1970s. 25 Analysis of
results was restricted to patients who had positive blood cultures, positive em-
pyema fluid cultures, or transtracheal aspirates. Among 159 cases, there were 56
(35%) in whieh anaerobes were reeovered. This was not surprising eonsidering
the importanee of aspiration as a pathophysiologie meehanism of nosocomial
pneumonia. There is also some question about the importanee or relevanee of this
observation. Approximately 80% of the eases involving anaerobes also included
aerobic baeteria as weIl, and these other organisms were predominantly gram-
negative bacilli and S. aureus, whieh may be more important pathogens in
nosoeomial pneumonia.
3. PATHOPHYSIOLOGY
Most anaerobie infeetions refleet a breeeh in the mucoeutaneous defense

meehanisms so that baeteria of the normal flora where an aerobic baeteria of the
numerieally dominant forms reaeh adjaeent, normally sterile sites. The predomi-
nant associations are dysphagia and apredisposition to aspiration. The usual
souree ofbacteria is periodontal disease, especially the organisms found along the
gum margin in pockets teeming with an aerobic bacteria that are in coneentra-
tions that approach the geometrie limits with which bacteria may oceupy spaee:
1012/gm.15.16 The second common association is predisposition to aspiration as
a result of dysphagia or compromised consciousness. Examples include alcohol-
ism, general anesthesia, seizure disorder, sedative abuse, drug abuse, esophageal
lesions, and neurologie disorders. The segments of the lung favored by gravita-
tional flow in the upright position are the basal segments of the lower lobes. More
eommonly, aspiration oeeurs in the reeumbent position where the favored seg-
ments are the posterior segments of the upper lobes or superior segments of the
lower lobes. 6 •26
Other conditions that predispose to an aerobic lung infections include infarc-
tion, pulmonary obstruction eaused by neoplasm or foreign body and bronchiee-
tasis. Eaeh of these is assoeiated with stasis or neerosis of tissue, which is the
eommon feature. Postobstruetive pneumonia is probably eaused by anaerobic
baeteria in most cases, although there are no appropriate studies to confirm this.
Support for the postulated mechanism is based on the dog model where it has
been shown that occlusion of a bronchus results in distal infeetion involving
anaerobic bacteria from the normal flora. 27 Another mechanism is bacteremie
seeding ofthe lung as illustrated by Lemierre Syndrome described in 1936;28 this
consists of septic thrombophlebitis of the jugular vein, bacteremia with Fusobac-
terium necroplwrum, and multiple septic emboli to the lungs.
A striking feature of anaerobic pulmonary infections is the tendency for
necrosis of tissue with lung abscess or a bronchopleural fistula leading to em-
pyema. The number of microbes that cause pneumonitis is legion, but the
number that cause pulmonary necrosis with abscess formation is a short list and
anaerobic bacteria are clearly at the top. This association with abseess formation
was weIl studied by David Smith using experimental animals in the period 1927-
1930. 2-4 His work was prompted by the observation that bacteria in the walls of
abscesses studied at autopsy appeared similar to those noted in the gingival
crevice, leading hirn to postulate that aspiration was the mechanism of infection.
Smith consequently challenged experimental animals using an intratracheal

inoculation of pyorrhea pus. Some animals eliminated the inoculum, some
developed pneumonitis, and a major portion of the latter went on to pulmonary
necrosis with abscess formation. He subsequently identified 17 different microbes
in the inoculum and then challenged the animals with each individually and then
with various combinations. He eventually concluded that four different anaerobic
species were required. This represents one of the first demonstrations of micro-
bial synergy, meaning organisms producing a biologic effect in combination that
neither component could produce in pure culture.
More recent studies have focused attention on the virulence properties that
might account for the extraordinary association between anaerobic bacteria and
abscesses at virtually every anatomical site, not only the lung, but also brain
abscess, dental abscess, intraabdominal abscess, tubo-ovarian abscess, and soft-
tissue abscess. There now appear to be three interrelated factors to account for
this association: First, there needs to be an adjuvant which is most likely fibrin
deposition. Secondly, the organism needs to evade early killing, which is presum-
ably facilitated by the polysaccharide capsule found with B. fraffilis, B. melanino-
genicus, and presumably other Bacteroides species as well. 3o Finally, these organ-
isms produce short-chain volatile fatty acids in the late phases of log growth that
appears to cause a pH-dependent inhibition of phagocytic killing.3 1- 33
4. CLINICAL FEATURES
The initial phase of pneumonitis caused by anaerobic bacteria may be acute,

subacute, or chronic.6· 34 We previously reviewed the clinical features of 46
patients with pneumonitis and cultures of transtracheal aspirates that yielded a
flora exclusively comprised of anaerobic bacteria. These were compared with the
clinical features of 46 patients with a transtracheal aspiration that yielded a pure
growth of S. pneumoniae. There were actually minimal differences in these two
groups in terms of patient age, changes on chest X ray, peak fever, or peripheral
leukocyte count. 34 Significant differences observed include the fact that patients
with anaerobic infections never reported true rigors, these patients had a some-
wh at longer course of symptoms prior to presentation and, as expected, they were
more likely to report associated conditions that predisposed to aspiration.
The initial feature of aspiration pneumonia is pneumonitis that, as noted
above, may appears similar in clinical presentation to pneumococcal pneumonia.
However, in many instances, the initial phase of infection is more subtle and the
patient does not seek medical attention until later in the course, often after
suppurative complications in the form oflung abscess or empyema. Studies of the
natural history of lung abscess in the animal model and with sequential X rays in
patients with defined periods of aspiration (as with general anaesthesia or a
seizure) indicate that 7-14 days are required for the appearance of a cavitary
lesions on chest X ray. A similar duration is presumably required for a broncho-
pleural fistula leading to empyema.
Other clinical features of anaerobic pulmonary infections are summarized in
Table 11, which represents a review of our experience with 193 cases. The average
temperature at the time ofhospitalization for these patients was 39.l o C and all but
five of the patients were febrile. 6 The average peripheralleukocyte count was
15,000/mm3 • As expected, patients with suppurative complications showed a
prolonged duration of symptoms prior to presentation and approximately half of
these patients noted weight loss. Of particular interest was the observation that
putrid sputum or empyema fluid was found in 40-60% of patients with abscess
or empyema and rarely found in those with pneumonitis alone. This putrid
discharge is considered diagnostic of anaerobic infections, it presumably reflects
the production of short-chain volatile fatty acids and amines, and it appears to be
a feature of late disease.
Distinctive features of anaerobic pulmonary infections compared to other
causes ofbacterial infections ofthe lower airways include (I) the high frequency of
suppurative complications, (2) the association with aspiration, especially aspira-
tion of gingival crevice material, (3) chronic or indolent symptoms, (4) putrid
discharge, and (5), lack of a likely etiologic diagnosis using conventional micro-
biological methods, primarily Gram's stain and culture of expectorated sputum.
5. LABORATORY DIAGNOSIS
The specimens that have established merit for a meaningful anaerobic

culture include pleural fluid, trans tracheal aspirates, transthoracic aspirates,
specimens obtained at thoracotomy, and fiberoptic bronchoscopic aspirates using
the protected brush catheter combined with quantitative cultures. The indications
for these diagnostic techniques are controversial since they require technical
expertise, some are regarded as invasive, and potentially dangerous, and some
TABLE 11
Clinical Features of Anaerobic Pulmonary Infections
Pneumonitis
without abscess
Abscess or empyema Empyema Total
Feature (83 patients) (79 patients) (51 patients) (193 patients)
Peak temp (mean) 102. 1°F 102.6°F 102.4°F 102.4°F

Peripheral leukocyte 15,500 13,700 21,600 15,000
count (mm 3 , median)
Age (yrs, median) 52 60 49 51
History of wt loss 36 (43%) 3 (4%) 28 (55%) 57 (30%)
Duration of symptoms 14 3 15 7
prior to presentation
(days, median)
Putrid discharge (no.) 41 (48%) 4 (5%) 32 (64%) 62 (32%)
Death caused by 3 (4%) 3 (4%) 3 (6%) 8 (4%)
infection (no.)
argue that the ease of treating anaerobes outweighs the advantages of the
procedures necessary to prove their presence. It should be noted that trans-
tracheal aspiration, a common mechanism to establish the bacteriologic diagnosis
of pulmonary infections in general in the early 1970s, was once advocated by some
as a preferred method for microbial detection for all pulmonary infections. 35-37
Enthusiasm was dearly spotty in that some institutions did many transtracheal
aspirations and some never used the procedure because they considered it
excessively invasive. Many hospitals developed strategies in which certain physi-
cians were charged with the responsibility of supervising the procedure, but it
never had the type of restraints comparable to the bronchoscopy union, with
established training criteria, hospital credentialing and, perhaps most impor-
tant, a billing code. At present, there are only rare physicians who are skilled
in the technique of transtracheal aspiration and even fewer who do it with any
frequency.
One technique that continues to generate considerable interest in fiberoptic
bronchoscopy. Initial studies show that the usual specimens aspirated from the
inner channel are universally contaminated by the salivary flora,38 but the use of a
brush catheter combined with quantitative cultures appears to obviate these
problems. 24 ,39,4o Despite this favorable experience, many physicians will have
difficulty reproducing the experience of investigators who have extensive experi-
ence coupled with laboratories that have a strong commitment to the micro-
biological methods necessary to quantitate and identify anaerobic bacteria. It
should be noted that quantitative cultures oflower airway secretions has appeared
to notably improve the diagnostic accuracy of almost any specimen source subject
to contamination induding sputum. 41-44 Nevertheless, the utility ofthese proce-
dures for an aerobic cultures has not been well verified in a fashion analogous to
the studies of TTA,37 and practical application, even if meritorious, will be
difficult. The conclusion is that a microbiological diagnosis of anaerobic bacterial
infections of the lung in the absence of empyema is difficult, and most clinicians
will continue to rely on dinical dues.
6. BACfERIOLOGY
The bacteriology of anaerobic pulmonary infections has been reported in a

large number of small series and two relatively large studies in the antibiotic era
(Table 111). One of these studies was by our group, summarizing the bacteriology
restricted to transtracheal aspirations and pleural fluid analysis in 193 patients
with both dinical and bacteriologic evidence of anaerobic infection of the lower
airways studied from 1968 to 1975.6 This study showed that approximately one-
half of the patients had anaerobic bacteria and pure culture; the remaining half
had anaerobes combined with aerobic bacteria considered potential pathogens.
Analysis of these 'two groups showed that the patients with mixed aerobic-
anaerobic infections had similar dinical features in terms of frequency of sus-
pected aspiration as the underlying mechanism, indolence in the presentation,
and the frequency of putrid discharge. Nevertheless, caution is advised with these
120 JOHN G. BARrLETT
TABLEIII
Bacteriology of Anaerobic
Pleuropulmonary Infections
Author's Finegold
experience 6 et al. 45
Period reviewed 1968-75 Early 1980s

No. cases reviewed 193 196
Total anaerobic strains 461 656
Gram-negative bacilli
Bacteroides fragilis 38 14
B. melaninogenicus 76 46
B. intermedius 60
Bacteroides sp. (other) 37 184
Fusobacterium nucleatum 56 58
Gram-positive cocci
Peptostreptococcus 87 66
Peptococcus 39
Gram-positive bacilli
Clostridium sp. 18 20
Eubacterium sp. 18 46
Actinomyces 5 28
Other 27 52
Gram-negative cocci
Veillonella 23 54
mixed infections with nosocomial acquisition because, in this setting, the aerobic
component of the infection appears to be most important.
Among the 461 strains of anaerobic bacteria, the predominant isolates were
anaerobic streptococci (Peptostreptococcus) (87 strains), B. melaninogenicus (76
strains) and Fusobacterium nucleatum (56 strains). These three organisms consti-
tute what are often referred to as the big three pathogens of anaerobic pulmon-
ary infection. The more recent comprehensive report of bacteria isolated from
anaerobic pulmonary infections is by Finegold et al. 45 reflecting studies during
the 1980s with 196 patients and a total of 656 strains of anaerobic bacteria. Again,
the specimen source was primarily transtracheal aspiration and empyema fluid,
and this also represents the work of an anaerobic research laboratory that was
responsible for processing specimens and identification of isolates. Some of the
bacteriology reported here reflect taxonomie changes during the time interval
between the two reports. The major changes are: (1) the recognition that many
of the organisms previously classified as B. fragilis were probably other penicillin-
resistant species of Bacteroides, (2) the taxonomy of anaerobic gram-positive
cocci was changed considerably to delete many organisms previously classified as
anaerobic streptococci and the genus Peptococci was largely deleted, and (3) the
species previously referred to as B. melaninogenicus was divided into multiple
species characterized by black pigmented colonies. With regard to the big three:
1. Anaerobic gram-positive cocci. Nearly all of the important clinical organ-

isms previously classified as Peptoeoeeus were reclassified as Peptostreptoeoeeus with
the exception of Pe. niger. 46 •47 The other major taxonomie change was a re-
classification based on aerotolerance of many gram-positive cocci that produced
abundant lactic acid and were formerly classified as anaerobes into the genera
Streptoeoeeus. These organisms include relatively important bacteria responsible
for infections that often simulate those produced by anaerobes including S.
intermedius, S. parvulus, S. morbillorum and S. eonstellatus. 48 •49 These organisms
actually grow best in anaerobic conditions, but they are relatively aerotolerant and
they are resistant to metronidazole, an antibiotic that is active against virtually all
strict anaerobes.
2. B. menaninogenieus. The organisms in this category, according to prior
classification, share the property of producing brown or black pigment derived
from heme, although the appellation applied was based on the erroneous assump-
tion that it was melanin. It is now recognized that organisms previously classified
in this species actually represent a diverse group that is now divided into two
genera including the saccharolytic species referred to as Bacteroides and the
asaccharolytic species referred to as Porphyromonas. There are a total of eight
species, but the most important in lung infections are B. menaninogenieus and B.
intermedius. These organisms are relatively frequent pathogens in anaerobic
pulmonary infections and they are relatively fastidious so they pose achallenge to
laboratory recovery. Of particular interest is the observation that many produce
penicillinase, which may account for some penicillin failures among patients. This
property of penicillinase production appears to represent an evolutionary pat-
tern, possibly reftecting extensive use of this drug since most of the strains
recovered in the 1960s were penicillin susceptible. (By contrast, B. fragilis strains
isolated in the 1930s were penicillin resistant indicating this organism was always
resistant to most betalactams.)
3. Fusobaeteria. This genera includes two major species, F. nucleatum, which is
the dominant form found in lung infections, and F. neeroplwrum. F. nueleatum is a
spindle-shaped rod of 5-10 um in length with pointed ends that represents the
so-called fusiform bacillus that was subject to extensive reporting of fusospiroche-
tal infections in the early 1900s. The great attention accorded this organism at the
time largely reftected its relatively unique morphologie appearance, and the
interest in anaerobic spirochetes at that time reftected the importance of syphilis.
At present, there is virtually no recognition of anaerobic spirochetes in these
infections, and it is not clear if this simply reflects absence, lack of importance,
and/or failure to perform adequate detection methods.
7. TREATMENT
As noted, most anaerobic pulmonary infections are treated empirically

wit:hout the benefit of bacteriological studies because of the difficulty in getting
appropriate specimens for anaerobic culture and the problems associated in
laboratory detection of oxygen-sensitive forms. Even when such studies are done,
in vitro sensitivity studies are rarely performed. In fact, the National Committee
on Clinical Laboratory Standards (NCCLS) feels that such studies for isolates in
cases of aspiration pneumonia and lung abscess are unnecessary.50 There are
several reasons for this:
1. Many of these infections are polymicrobial with three to five microbial
species, making this type of work long and tedious.
2. Most bacteriology laboratories do not perform in vitro sensitivity tests on
anaerobic bacteria.
3. Therapeutic decisions are usually based on drugs selected empirically
according to results of clinical trials. Thus, most physicians have well
established habits with respect to drug selection for an aerobic bacteria in
patients with aspiration pneumonia and lung abscess.
The drug selection in these cases may be done in three ways: (1) A drug may
be selected empirically on the basis of clinical trials according to published
reports; (2) Drugs may be selected on the basis of published reports of in vitro
activity against the big three; and (3) Drugs may be given for the presumed
importance of other microbes where an aerobic bacteria are not suspected. All
three are probably successful most of the time, although there is an obvious
preference for the decision according to rank order.
Clinical trials of antibiotics in patients with an aerobic pulmonary infections
may be divided into two periods:
Studies done in the 1960s, especially the work of William Weiss at Phila-
delphia General Hospital,51-54 showed that nearly all patients with primary lung
abscess responded to penicillin. Oral penicillin in a dose of 3 gm/day was as
effective as parenteral penicillin, and the rare patient who failed to res pond to
penicillin generally did well with tetracycline. These studies were done in patients
with presumed an aerobic infections for what was often referred to as nonspecific
lung abscess characterized by cavitary lung infection, the lack of a likely pathogen
with aerobic culture of expectorated sputum, and often a putrid discharge. The
work of this investigator and others clearly established penicillin as the preferred
drug for both aspiration pneumonia and lung abscess, and presumably other
forms of an aerobic infections above the diaphragm as well.
Since 1970 there have been extensive studies of lung abscess and aspiration
pneumonia including descriptions of the bacteriology, definition of virulence
factors, and pathophysiologic mechanisms. In vitro sensitivity data indicated that
a major portion of patients had isolates that were resistant to penicillin. Neverthe-
less, the relevance of these observations was questioned because the anecdotal
experience in the 1970s suggested that penicillin continued to work weH, not only
in lung infections but in orodental infections that presumably involved the same
bacteria. 55 The observation that 15-25% of patients harbored strains that were
highly resistant to penicillin begged the issue of the relative merits of penicillin
compared to alternative options. 45 The most frequently encountered penicillin-
resistant anaerobes were B. melininogenicus, B. intermedius, B. ruminocola, B.
putridinis, B. capillosis, B. gracillis, and B. ureolyticus. 50 ,56-58
More recently, there have been two therapeutic trials in patients with lung
abscess comparing penicillin and clindamycin; both showed statistically signifi-
cant benefit with clindamycin Il ,59 (Table IV). The only other drug that has
generated substantial interest in therapeutic trials in patients with anaerobic lung
infections is metronidazole, reftecting the fact that this drug is active against
virtually all anaerobic bacteria. Potential advantages are that metronidazole is
almost completely absorbed with oral administration, the drug is extremely
inexpensive so that the wholesale price of a one-week supply is approximately
$1.00, and it shows the most impressive bactericidal activity in vitra versus nearly
all anaerobic bacteria. Despite these accolades, the collective experience with 28
patients treated with metronidazole showed 12 (43%) were considered therapeu-
tic failures. 60-62 The presumed explanation for this poor track record is the
probability that metronidazole-resistant microaerophilic and aerobic streptococci
play an important role in these infections. A practical application of this observa-
tion is that metronidazole, when used, should be combined with another drug
such as penicillin. 63
The experiences noted above account for the recommendation of the Medi-
cal Letter consultants for three regimens for anaerobic pulmonary infections:
penicillin, clindamycin, or metronidazole plus penicillin. 64 Despite these recom-
mendations, it is expected that multiple antimicrobial agents will be effective in
anaerobic pulmonary infections and these other agents suffer simply by the fact
that they have not been subject to clinical trials. This certainly includes most
penicillins other than the antistaphylococcal penicillin, and most cephalosporins,
although ceftazidime may be an exception. Drugs that are particularly attractive
because of extraordinary in vitra activity against virtually all anaerobes include
imipenem, chloramphenical, and any combination of a betalactam-betalactamase
inhibitor. 50 ,57,58 Erythromycin, and presumably azithromycin and clarithro-
TABLE IV
Antibiotic Treatment of Anaerobic Lung Abscesses: Results of
Two Randomized Trials Comparing Penicillin and Clindamycin
Source: Levison et al. 59 Guidiol et al. J J
Regimen: Penicillin Clindamycin Penicillin Clindamycin

(IV/d) 6 mil units 1.8 gm 12 mil units 2.4 gm
No patients: 21 17 18 19
Drug failure 5 (290/<) 0* 7 (390/<) 1 (50/<)*
Relapse 3 (190/<) 0 2(110/<) o
Duration fever (mean) 7.7 days 4.7 days* 7.2 days 6.4 days
Duration putrid sputum 7.8 days 4.1 days* 7.3 days 3.9 days*
Side effects 0 0 o o
*Difference in outcome is statistically significant (p < 0.05).
mycin, are commonly advocated for patients with enigmatic pneumonia that may
involve anaerobic bacteria; these agents show good activity against most anaer-
obes except for fusobacteria. Drugs that have virtually no activity against anaer-
obes in vitro and would be considered almost totally inactive are aminoglycosides,
azathreonam, trimethoprim-sulfamethoxazole, and ciprofloxacin.
The duration of treatment is variable. Pneumonitis can often be treated
with the same guidelines used for the treatment of pneumococcal pneumonia in
terms of oral versus parenteral treatment and duration of treatment.6· 29 Lung
abscess usually requires prolonged courses of therapy; the initial treatment is
often with parenteral agents such as clindamycin, and therapy is changed to an
oral agent such as oral clindamycin, (300 mg 4 x daily) amoxicillin (500 mg tid), or
amoxicillin-clavulanic acid when the patient is afebrile and clinically stable. Our
practice is to continue antibiotics until the pulmonary infiltrate has cleared or
there is a smalI, stable residual lesion. OccasionallY patients fail to respond to
medical management and require surgical intervention. 6 6-69 A review of the
medicalliterature during the past two decades suggest that approximately 10-
12% of patients with pulmonary abscesses are considered surgical candi-
dates. ll •6 8-70 For patients with empyema, the recommendation is for drainage
combined with antibiotics. Patients with an aerobic bacterial empyemas often
present relatively late in the course of the disease when there are multiple
loculations, considerable fibrous material, and thick fluid that is difficult to drain.
Many eventually require open drainage procedures with a rib resection or
decortication. Antibiotics are usually continued until drainage tubes are removed
and concurrent abscesses are healed.
8. PROGNOSIS
Anaerobic infections of the lung generally respond welt to antibiotics com-

bined with drainage of empyemas. With anaerobic pneumonitis or aspiration
pneumonia ascribed to anaerobes, the average duration of fever following institu-
tion of antibiotics is two days, and 90% of patients are afebrile in seven days.6· 29
Patients with lung abscess respond more slowly with an average duration of fever
following antibiotic treatment of 4-8 days.II·58 Empyemas are often difficult for
the reasons noted above, difficult because the initial drainage procedure is often
inadequate so that a smoldering fever may persist for several weeks. 6 .15 For X-rays,
the usual rule with this as weil as other pulmonary infections is that the infiltrates
may progress during the initial stages of successful treatment. 7I After 3-7 days
there should be the beginning of gradual clearance. The time to reach cavity
closure with lung abscess depends largely on the size of the cavity at the time
treatment is started and host factors; one large study with serial X-rays showed an
average of 65 days to closure. 71
The pro gnosis for survival is good. Our experience with 193 patients showed
that this infection was a major cause of death in eight (4%) and it was a
contributing factor in 14 (7%).6 The major poor prognostic factor with aspiration
ANAEROBIC BAGTERIAL INFECTIONS OF THE LUNG 125
pneumonia is the concurrent presence of gram-negative bacteria with nosocomial

infections. Poor prognostic factors for lung abscess are large cavity size, prolonged
symptoms prior to presentation, abscess associated with bronchial obstruction,
and serious associated disease such as neoplasm and debility.
REFERENCES
1. Rendu, M., and Rist, M., 1899, Etude clinique et bacteriologique de trois cas de pleursie
putride, Bull. Mem. Soe. Med. Hop. Paris 16:133.
2. Smith, D. T., 1927, Experimental aspiratory abscess, Areh. Surg. 14:231-239.
3. Smith, D. T., 1927, Fusospirochetal disease of the lungs: Its bacteriology, pathology and
experimental reproduction, Am. Rev. Tubere. 16:584.
4. Smith, D. T., 1930, Fusospirochetal disease of the lungs produced with cultures from
Vincent's angina,] In/eet. Dis. 46:303.
5. Bardett,]. G., and Finegold, S. M., 1972, Anaerobic pleuropulmonary infections, Medieine
51:413-450.
6. Bardett,]. G., 1987, Anaerobic bacterial infections of the lung, Chest 91:901.
7. Bardett,]. G., and Finegold, S. M.,1974, Anaerobic infections ofthe lung and pleural space,
8. Bardett,]. G., and Gorbach, S. L., 1975, Treatment of aspiration pneumonia and primary
lung abscess,JAMA 234:935-937.
9. Beerens, H., and Tahon-Castel, M., 1965, in: Infections humaines a bacteries anaerobies
nontoxigenes, Presses Academiques Europeenes, Brussels, pp. 91-114.
10. Brook,l., and Finegold, S. M., 1979, Bacteriology and therapy oflung abscess in children,J
Pediatr. 94:10-14.
11. Gudiol, F., Manresa, F., and Pallares, R., 1990, Clindamycin vs. penicillin for anaerobic lung
infections, Areh. Intern. Med. 158:2525-2530.
12. Gonzalez, C. L., and Calie, F., 1975, Bacteriologic flora of aspiration-induced pulmonary
infections, Areh. Intern. Med. 135:711-714.
13. Lorber, G., Swenson, R. M., 1974, Bacteriology of aspiration pneumonia: A prospective
study of community and hospital acquired cases, Ann. Intern. Med. 81:329-331.
14. Brook, l., and Finegold, S. M., 1980, Bacteriology and therapy of lung abscess in children,
Pediatl'ies 65: 1115-1120.
15. Hirsch, R. S., and Clarke, N. G., 1989, Infection and periodontal diseases, Rev. lrifeet. Dis.
11:707-15.
16. Neiders, M. E., Chen, P. B., and Reynolds, H., 1989, Heterogenecity of virulence among
strains of Bacteroides gingivalis,] Periodontal. Res. 24:192-198.
17. Bardett,]. G., Gorbach, S. L., Thadepalli, H., and Finegold, S. M., 1974, Bacteriology of
empyema, Lancet 1:338-340.
18. Sullivan, K., O'Toole, R., and Fisher, R., 1973, Anaerobic empyema thoracic, Al·eh. Intem.
Med. 131:521-528.
19. Varkey, B., Rose, H., and Kutty, K., 1981, Empyema thoraeie during a ten year period, Areh.
Intern. Med. 141:1771-1776.
20. Mavroudis, C., Symmonds,]., and Minagi, H., 1981, Improved survival in management of
empyema thoracic,] Thorac. Cardiovase. Surg. 82:49-57.
21. Grant, D., and Finley, R., 1985, Empyema: Analysis of treatment techniques, Can.] Surg.
28:449-451.
22. Lemmer,]., Rotham, M., and Orringer, M., 1985, Modern management of adult thoracic
empyema.] Thorac. Cardiovasc. Smog. 90:849-855.
23. Ries, K., Levison, M. E., and Kaye, D., Transtracheal aspiration in pulmonary infection, Arch.
Intern. Med. 133:453-458.
24. Pollack, H. M., Hawkins, E. L., Bonner, j. R., Sparkman, T., and Bass, j. B., Jr., 1983,
Diagnosis of bacterial pulmonary infections with quantitative protected catheter cultures
obtained during bronchoscopy, j Clin. Microbiol. 17:255-59.
25. Bartlett,j. G., O'Keere, P., Tally, F. P., Louie, T.j., and Gorbach, S. L., 1986, The bacteriology
of hospital-acquired pneumonia, Arch. Intern. Med. 146:868-871.
26. Shapiro, M. S., and Matthay, R. A., 1989, Pulmonary aspiration: Keys to effective therapy,j
Respir. Dis. 10:59-74.
27. Lansing, A. M., andJamieson, W G., 1963, Mechanisms of fever in pulmonary ate!ectasis,
Arch. Surg. 87:184-89.
28. Lemierre, A., 1936, On certain septicemias due to anaerobic organisms, Lancet 701-703.
29. Mayrand, D., and Holt, S. C., 1988, Biology of asaccharolytic black pigmented Bacteroides
species, Microbiol. Rev. 52:134-149.
30. Crabb, j. H., Finberg, R., Onderdonk, A. B., and Kasper, D. L., 1990, T-cell regulation of
Bacteroides fragilis-induced intraabdominal abscesses, Rev. Infoct. Dis. 12:S178-84.
31. Rotstein, 0. D., Pruett, T. L., Fiege!, V. D., Ne!son, R. D., and Simmons, R. L., 1985, Succinic
acid: A metabolic by-product of bacteroides species inhibits polymorphonuc\ear leukocyte
function, Infoct. Immun. 48:402-08.
32. Rotstein, O. D., Vittorini, T., Kao,j., McBurney, M.l., Nasmith, P. E., and Grinstein, S., 1989,
A soluble bacteroide's by-product impairs phagocytic killing of Escherichia coli by neu-
trophils, Infoct. Immun. 57:745-53.
33. Eftimiadi, C., Tonetti, M., Cavallero, A., Sacco, 0., and Rossi, G. A., 1990, Short-chain fatty
acids produced by anaerobic bacteria inhibit phagocytosis by human lung phagocytes, j
Infoct. Dis. 161:138-42.
34. Bartlett, j. G., 1979, Anaerobic bacterial pneumonitis, Am. Rev. Respir. Dis. 119:19-23.
35. Bartlett, j. G., 1986, The technique of transtracheal aspiration, j Critical Illness 1:43-49.
36. Bartlett, j. G., Rosenblatt, j. E., and Finegold, S. M., 1973, Percutaneous transtracheal
aspiration in the diagnosis of anaerobic pulmonary infection, Ann. Intern. Med. 79:535-540.
37. Bartlett, j. G., 1977, Diagnostic accuracy of transtracheal aspiration bacteriology, Am. Rev.
Respir. Dis. 115:777-782.
38. Bartlett,j. G., Alexander,j., Mayhew,j., Sullivan-Sigler, N., and Gorbach, S. L., 1976, Should
fiberoptic bronchoscopy aspirates be cultured? Am. Rev. Respir. Dis. 114:73-78.
39. Wimberly, N., Faling, j., and Bartlett, j. G., 1979, A fiberoptic bronchoscopy technique to
obtain uncontaminated lower airway secretions for bacterial culture, Am. Rev. Respir. Dis.
110:337-43.
40. Wimberly, N. W, Bass,j. B., Boyd, B. W, Kirkpatrick, M. B., Serio, R. A., and Pollock, H. M.,
1982, Use of a bronchoscopic protected catheter brush for the diagnosis of pulmonary
infections, Chest 81:556-62.
41. Lorch, D. G.,Jr.,John,j. F.,Jr., Tomlinson,j. R., Miller, K. S., and Sahn, S. A., 1987, Protected
transbronchial needle aspiration and protected specimen brush in the diagnosis of pneu-
monia, Am. Rev. Respir. Dis. 136:565-69.
42. Chastre, j., Fagon J-Y., Soler, P., Bornet, M., Domart, Y., Trouillet, J-L., Gibert, C., and
Hance, A. j., 1988, Diagnosis of nosocomial bacterial pneumonia in intubated patients
undergoing ventilation: Comparison of the usefulness of bronchoalveolar lavage and the
protected specimen brush, Am. j Med. 85:499-506.
43. Marquette, C-H., Ramon, P., Courcol, R., Wallert, B., Tonne!, A-B., and Voisin, C., 1988,
Bronchoscopic protected catheter brush for the diagnosis of pulmonary infections, Chest
93:746-50.
44. Torres, A., De La Bellacasa, j. P., Rodriguez-Roisin, R., Jimenez De Anta, T., and Agusti-
Vidal, A., 1988, Diagnostic value ofte!escoping plugged catheters in mechanically ventilated
patients with bacterial pneumonia using the metras catheter, Am. Rev. Respir. Dis. 138:
117-20.
45. Finegold, S. M., George, W 1., and Mulligan, M. E., 1985, Anaerobic infections, DM 31:
8-77.
46. Ezaki, T., Yamamoto, N., Ninomiya, K., Suzuki, S., and Yabuuchi, E., 1983, Transfer of
Peptococcus indolicus, Peptococcus asaccharolyticus, Peptococcus prevotii, and Peptococcus magnus
to the genus Peptostreptococeus and proposal of Peptostreptoeoeeus tetradius sp. nov., Int. J Syst.
Bacteriol. 33:683-98.
47. Huss, V. A. R., Fest!, H., and Schleifer, K. H., 1984, Nucleic acid hybridization studies and
deoxyribonucleic acid base compositions of anaerobic, gram-positive cocci, Int. J Syst.
Bacteriol. 34:95-10 1.
48. Cato, E. P., 1983, Transfer of Peptostreptoeoeeus parvulus (Weinberg, Nativelle, and Prevot
1937) Smith 1957 to the genus Streptococcus: Streptoeoeeus parvulus (Weinberg, Nativelle, and
Prevot 1937) combo nov., rev., emend., Int. J Syst. Bacteriol. 33:82-4.
49. Gossling, j., 1988, Occurrence and pathogenicity of the Streptoeoceus milleri group, Rev. Infee.
Dis. 10:257-85.
50. Finegold, S. M., 1988, Susceptibility testing of anaerobic bacteria,J Glin. Mierobiol. 26:1253-
1256.
51. Weiss, W, 1968, Delayed cavity closure in acute nonspecific primary lung abscess, Am. J Med.
Sei. 255:313-19.
52. Weiss, W, 1970, Oral antibiotic therapy of penicillin and tetracycline, Gurr. Ther. Res. 12:
154-60.
53. Weiss, W, 1973, Cavity behavior in acute, primary nonspecific lung abscess, Am. Rev. Respir.
Dis. 108:1273-75.
54. Weiss, W, 1975, Letter to the editor, Ghest 67:625.
55. Bart!ett, j. G., 1975, Treatment of anaerobic pleuropulmonary infections, Ann. Intern. Med.
83:376-377.
56. Appelbaum, P. C., Spangier, S. K., andJacobs, M. R., 1990, Beta-lactamase production and
susceptibility to amoxicillin, amoxicillin-clavulanate, ticarcillin, ticarcillin-clavulanate,
cefoxitin, imipenem and metronidazole of 320 non-Bacteroides fragilis, Baeteroides isolates and
129 Fusobacteria from 28 U.S. centers, Antimierob. Ag. Ghemother. 34:1546.
57. Finegold, S. M., and Wexler, H. M., 1988, Therapeutic implications ofbacteriologic findings
in mixed aerobic-anaerobic infections, Antimierob. Ag. Ghemother. 32:611-616.
58. Wexler, H. M., Molitoris, E., and Finegold, S. M., 1991, Effect of B-lactamase inhibitors on
the activities of various B-lactam agents against anaerobic bacteria, Antimierob. Ag. Ghemother.
35:1219-1224.
59. Levison, M. E., Mangura, C. T., Lorber, B., Abrutyn, E., Pesanti, E. L., and Levt, R. S., 1983,
Clindamycin compared with penicillin for the treatment of anaerobic lung abscess, Ann.
Intern. Med. 98:466-71.
60. Tally, F. P., Sutter, V. L., and Finegold, S. M., 1975, Treatment of anaerobic infections with
metronidazole, Antimicrob. Agents Ghemother. 7:672-75.
61. Perlino, C. A., 1981, Metronidazole vs. clindamycin treatment of anaerobic pulmonary
infection, Areh. Intern. Med. 141:1424-27.
62. Sanders, C. v., Hanna, B. j., and Lewis, A. C., 1979, Metronidazole in the treatment of
anaerobic infections, Am. Rev. Respir. Dis. 120:337-43.
63. Eykyn, S. j., 1983, The therapeutic use of metronidazole in anaerobic infection: Six year's
experience in a London hospital, Surgery 93:209-14.
64. 1984, Drugs for anaerobic infections, Med. LeU. 26:87-90.
65. Heimdahl, A., Von Konow, L., and Nord, C. E., 1980, Isolation ofbeta-lactamase-producing
Baeteroides strains associated with clinical failures with penicillin treatment of human
orofacial infections, Areh. Oral Biol. 5:689-692.
66. Sokolowski, j. w., Burgher, L. W, Jones, F. L., Harris, j. A., and Patterson, j. R., 1986,
Guidelines for fiberoptic bronchoscopy, ATS News (spring):14-5.
67. Sosenko, A., and Glassroth, j., 1985, Fiberoptic bronchoscopy in the evaluation of lung
abscesses, ehest 87:489-94.
68. Bardett, j. G., 1982, Lung abscess,Johns Hopkins Med. J 150:141-47.
69. Pohlson, E. C., McNamara, j., Char, C., and Kurata, B., 1985, Lung abscess: Achanging
pattern of the disease, Am. J Surg. 150:97-101.
70. Neild, j. E., Eykyn, S. j., and Phillips, 1., 1985, Lung abscess and empyema, Q. J Med.
224:875.
71. Landay, M.j., Christensen, E. E., Bynum, L.j., and Goodman, C., 1980, Anaerobic pleural
and pulmonary infections, Am. J Roent. 134:233-40.
8
Immunology of M. tuberculosis
and Other Mycobacteria
ROBERT S. WALLIS and JERROLD J. ELLNER
1. INTRODUCfION
More than 25% of the world's population has been infected with Mycobacterium
tuberculosis. Initial infection with M. tuberculosis is usually self-limited and inap-
parent. It is associated with the subsequent acquisition of tuberculin skin test
reactivity in vivo, blastogenic T-cell responses to mycobacterial proteins in vitro,
and the coincident development of protective immunity. Nonetheless, from 6 to 8
million new cases of tuberculosis arise worldwide each year, with 2 to 3 million
fatalities, making M. tuberculosis the most common identifiable cause of death of
any infectious agent. In the developing world, tuberculosis accounts for 6.7% of
all deaths, and an estimated 26% of avoidable adult deaths.l This occurs despite
the availability of antituberculous chemotherapy and the widespread use of
vaccination with BCG.
Ten majoT prospective trials have been undertaken during the past half
century to determine the efficacy of BCG vaccination. 2-4 The results of these
trials differ considerably. A study performed by the British Medical Research
Council, performed on a population of 5,472 skin-test negative British school
children found the degree ofprotection to be approximately 70%, with complete
success in preventing miliary and meningeal disease.5 Conversely, the largest
BCG trial, involving 140,000 tuberculin-negative residents ofChingleput, India,
found no evidence for protection after 7 years, and at best, 28% protection at 15
ROBERT S. WALLIS and JERROLD J. ELLNER • Department of Medicine, Case Western

Reserve University and University Hospitals, Cleveland, Ohio 44106-4984.
Pulmonary Inftctions and Immunity, edited by Herman Chmel el al. Plenum Press, New York, 1994.
129
130 ROBERT S. WALLIS and JERROLD J. ELLNER
years, if data analysis was restricted to those who were ages 0-14 at intake. 6 •7
These disappointing dinical results have provided the impetus for the application
of modern techniques of cellular immunology and molecular biology to advance
the understanding of the immunology of tuberculosis in both human and animal
models, and to identify potentially protective mycobacterial antigens.
Several important dinical observations have shaped our understanding of
mycobacterial immunity. First, defieieneies of cellular immunity, particularly
AIDS, convey a dramatically increased risk of tuberculosis, both progressive
primary and reactivation disease.8· 9 Second, delayed-type hypersensitivity and
protection have been linked by epidemiologic studies demonstrating resistance to
exogenous reinfection in tuberculin skin-test positive subjects. lO Finally, although
active tuberculosis may arise immediately following the initial exposure, partic-
ularly in the very young, most cases of tuberculosis represent recrudescence,
years later, at a site distant to that of the initial infection. This suggests a model of
mycobacterial infection in which, although the local immune response is success-
ful in eradicating the initial infection, small numbers of organisms nonetheless
disseminate, leading to persistence of latent foei of intracellular infection. Reac-
tivation may be assoeiated with an obvious cause for impaired immunity such as
corticosteroid therapy or HIV infection although for the majority of patients the
factors responsible for the shifting balance between host and parasite are not well
understood.
2. MONONUCLEAR PHAGOCYTE IMMUNOREGULATORY

PROPERTIES
M. tuberculosis shares with other intracellular pathogens the capaeity to

replicate within the phagocytic cells whose task is their ingestion and killing. In
addition to their microbieidal functions, monocytes and macrophages must func-
tion as immune inducers, by processing and presenting mycobacterial antigens,
and by elaborating the cytokines necessary for donal T cell expansion (Fig. 1).
Patients with active tuberculosis demonstrate impairment of these monocyte
functions. From 17 to 25% of patients with pulmonary tuberculosis have negative
intermediate-strength tuberculin skin testing on initial evaluation; this percent-
age is greater in those with disseminated or miliary disease. ll •12 Negative skin
testing is assoeiated with depressed in vitro lymphocyte blastogenesis to PPD
compared to skin-test responders.l 3 Partial depletion of monocytes by plastic
adherence resulted in a 20-fold increase in responses in this group, compared to
onlya 2.3-fold increase in tuberculin responders, as shown in Fig. 2. Patients with
newly diagnosed pulmonary tuberculosis or chronic drug-resistant disease show
selectively decreased PPD-stimulated expression of interleukin-2 and surface
expression of interleukin-2 receptors. 14 •15 Although depletion of adherent cells
increased IL-2 receptor expression in some patients, individuals with the most
depressed production of the cytokine showed an increase in PPD-stimulated
blastogenesis but not expression of IL-2 following adherent cell depletion.l 4
M. TUBERCULOSIS AND OTHER MYCOBACTERIA 131
M. tuberculosis
I
infection of
macrophages
([,,1, IL6, TNF Exp........ 01 """,b''',,,ia1 on.",..
Recruitment
and activation
of CD4, C08, &
yd lymphocyres
IL-Zl IFN-,
Granuloma
formation
/
IFN-y
\ C04, yd
vitamin 0 lymphocyres
I \
Macrophage activation Lysis of heavily
for intracellular killing infected macrophages
of mycobacteria by cytotoxic cells
FIGURE 1. A model of mycobacterial immunity.
Interpretation of these data is complicated somewhat by our recent observations

that monocytes from tuberculous patients constitutively express IL-2R and re-
lease them into the supernatants spontaneously and following treatment with
PPD16 (Fig. 3). In studies of US patients with newly diagnosed tuberculosis,
exogenous IL-2 failed to reconstitute blastogenesis, nor did indomethacin exert
an effect on expression of IL-2 in chronic disease.I 4 In some of the Japanese
patients, indomethacin increased blastogenesis and IL-2 expression. The differ-
ences between these studies presumably relate to the chronicity of the infection or
the genetic background of the patients.
Although the mechanisms and mediators of suppression by adherent mono-
nuclear cells are unknown, some progress has been made. Suppression was
mediated by OKMl-positive, peroxidase-positive, E-rosette-negative cells, and
was unaffected by indomethacin. The assumption, therefore, is that the sup-
pressor cell is a monocyte. Suppression was associated with altered expression of
PPD-stimulated
production of IL-2
160
140
120
100 _MNC
IL-2 CJ T cells
U/m1 80
60
40
20
o+---__.L......-L-_ _
TB control
FIGURE 2. IL-2 production by blood mononuclear cells (shaded bars) and monocyte-depleted
T cells (open bars) of tuberculous patients and healthy tuberculin positive controls following
stimulation with PPD. (From Tweardy et al.l 7 Reproduced with permission.)
Release of IL-2 receptors by monocytes
1000~------------------------------------~
100
Dunstim
IL-2R _ PPD
10
healthy M. kansasii M. tuberculosis

FIGURE 3. Release of soluble IL-2 receptors by monocytes of healthy controls, patients with M.
kansasii or M. scrofulaceum infections, and patients with tuberculosis, either unstimulated (open
bars), or following stimulation with PPD (shaded bars). (From Tweardy el al. 17 Reproduced with
permission.)
HLA-DR and inereased produetion of interleukin 1. 17 •18 Several issues coneerning

immunosuppression by monoeytes remain unresolved. The basis for the restrie-
tion of the suppression of T-Iymphoeyte responses to PPD may be partially
explained by the observation that PPD direetly stimulates monoeyte expression of
IL-l and TNF. It is possible that in tubereulous patients the interaetions of
monoeytes with mycobaeterial produets may release eytokines induding a puta-
tive mediator of suppression. Preliminary data indicate inereased expression of
transforming growth faetor beta (TGFß) by monoeytes from tubereulous pa-
tients. TGFß in nanogram quantities suppressed antigen-indueed blastogenesis
and expression of interleukin 2, thereby mimicking the abnormalities found in
tubereulosis. Studies in progress seek to definitively identify the moleeular basis
of suppression. Although these studies do not differentiate whether the activation
of suppressor eells is the eause or effeet of tuberculosis, they serve to underscore
the eentral role of mononudear phagoeytie eells in the immunopathology of
tubereulosis.
3. SUPPRESSOR LYMPHOCYTES IN TUBERCULOSIS
The traditional suppressor-eytotoxic T-cell population CD8+ is neither ex-

panded nor activated as suppressor cells in tubereulosis.l 9•2o Lymphoeytes bear-
ing a reeeptor for the Fe portion of IgG (FcyR) are approximately doubled in
number in tuberculous patients. They possess a curious relationship to the
suppressive networks that involve monocytes. A tubereulin low-responding pa-
tient either demonstrates suppression by monocytes or FcyR + eells, but not both.
In fact, the FeyR + lymphoeytes may exert a contrasuppressor activity in that
depletion of Fey R + eells unmasks antigen-specifie suppression by monoeytes
in all tubereulous patients.
CD16+ lymphoeytes are also expanded in tubereulosis, and partieipate in
monocyte-dependent inhibition of PPD stimulated expression of interleukin 2. 21
4. MONONUCLEAR PHAGOCYTE-MYCOBACTERIAL
INTERACTIONS
Both monoeytes and maerophages are permissive of intracellular mycobae-

terial replication, with generation times of from 16 to 26 hr in vitro. 22 The rate of
intracellular growth does not appear to be related to strain virulence. The
mechanism for aetivation of intraeellular killing mycobaeteria in vitro remains in
dispute. Studies of other intracellular pathogens such as Lelfionella, Leishmania,
and Toxoplasma have suggested a model in which antigen-specific CD4+ lympho-
eytes are aetivated by parasite antigens expressed on the surfaee of infeeted
monoeytes or macrophages in association with MHC dass 11 proteins. These T
cells then release interferon--y (and perhaps other macrophage-activating factors)

which increase the microbicidal capacity of the macrophages and ultimately cause
death of the parasite. 23- 25 While interferon--y is capable of inhibiting intracellular
growth of M. tuberculosis when applied to murine peritoneal or bone marrow
macrophages,26,27 this cytokine has at best a modest effect in human systems,
whether applied to freshly obtained monocytes or monocyte-derived macro-
phages. 22 ,28,29
Tumor necrosis factor (TNF) is a cytokine produced by mononudear phago-
cytes with multiple biologie properties induding that of an endogenous pyrogen
and cachectin,3o A role for TNF in control of mycobacterial infection has been
suggested by one study in which mice treated with neutralizing antibody to TNF
failed to form granulomas and were unable to contain a BCG infection following
intravenous challenge,31
5. CELL-MEDIATED CYTOTOXICITY
An alternative approach to the problem of mycobacterial killing would be for

the host to attack the infected phagocytic cells directly, thus exposing otherwise
sequestered intracellular organisms to alternative extracellular bactericidal
mechanisms, such as antibody and complement. Although cytotoxicity generally
is thoughtofas a functionofCD8+ cells, both CD4+ and CD8+ T celldones are
capable of MHC-restricted killing of antigen-pulsed monocytes,32,33 The evi-
dence suggesting significance in vivo of these observations is indirect. Müller et al.
selectively depleted thymectomized mice of CD4+ or CD8+ cells using mono-
donal antibodies, and then challenged the animals with M. tuberculosis. 34 They
found increased numbers of mycobacteria in the spleens of mice treated with
either antibody, although the increases were greater in those animals treated with
anti-CD4, and treatment with both antibodies was not always additive. Adoptive
transfer studies by Orme and Collins35 ,36 similarly suggest that both CD4+ and
CD8+ cells are required for acquisition of resistance to tuberculosis.
6. VITAMIN D AND MYCOBACTERIAL KILLING
Several studies have demonstrated an inhibitory effect of both 1,25(OH)2-

vitamin D3 and its derivative retinoic acid on intracellular growth of M. tuber-
culosis. 22 ,37 The extent of this inhibition is generally limited to a doubling of the
generation time. Although the effect requires relatively high concentrations of
vitamin D compared to those generally present in plasma, such levels may be
present within granulomatous lesions. 38 Tuberculous pleural fluid also contains
high levels of vitamin D,39 It is possible that the apparently beneficial effect of
phototherapy of tuberculosis in the prechemotherapy era was mediated by this
effect of vitamin D mycobacterial killing.
7. 'YB T CELLS
While most T lymphocytes express a T-cell receptor comprising 0: and ß

subunits, the receptors of a minority of cells comprise two alternate chains, "I
and 8. These cells account for from 1 to 5% of peripheral blood lymphocytes, and
are evenly distributed in lymphoid tissue and skin. They bear the surface
phenotype CD3+ CD4- CD8-. Although early studies demonstrated MHC-
unrestricted cytotoxicity and elaboration of interleukin 2, interferon-'Y, and
lymphotoxin by "18 cells, no antigen-specific "18 reactivity could be identified,
despite the structural similarities between the two classes of receptors. 40 It was
speculated that the limited diversity of "{8 gene segment rearrangement allowed
for recognition of only a narrow range of foreign antigens.
Recent studies have suggested that mycobacterial proteins, and heat-shock
proteins in particular, may be important targets for "18 T cells. "18 cells can be
rapidly expanded in vitro following stimulation of blood mononuclear cells with
crude mycobacterial antigens. 41 ,42 They are found in increased numbers in
several granulomatous skin lesions, including those of lepromin skin tests, rever-
sal reactions, and localized cutaneous leishmaniasis. 43 Footpad immunization of
mice with M. tuberculosis results in increased numbers of "18 T cells in draining
lymph nodes,44 and mice administered an aerosol challenge of PPD develop
increased numbers of "18 T cells in the lungs. 45 Indeed, mycobacterial protein
antigens seem preferentially recognized by "18 T cells, even in immunologically
naive subjects. Limiting dilution analysis of CD4- CD8- blood T cells from a
healthy tuberculin-negative donor showed nearly half were M. tuberculosis reac-
tive. 42 Similarly, 10% of the murine "{8 T-cell hybridomas obtained from neonatal
th ymuses were reactive with the M. tuberculosis 65- kDa heat-shock protein. 46 I t has
been hypothesized that the "18 system represents a phylogenetically old and
relatively primitive initial defense against an ancestral group of pathogens, or
alternatively, that the reactivity with heat-shock proteins that show broad homol-
ogies with human heat-shock proteins reftects a role for "18 T cells in the clearance
of damaged or senescent autologous cells.
8. MYCOBACfERIAL ANTIGENS
M. tuberculosis culture filtrate, old tuberculin, and PPD all contain multiple
antigens,47 including polysaccharides (principally arabinogalactan and arabino-
mannan 48 ) and proteins. Mycobacterial protein antigens, some of which are
species-specific, elicit delayed type hypersensitivity and stimulate lymphocyte
blastogenic responses in both sensitized humans and guinea pigs. 49-51 The
polysaccharide antigens share antigenic cross-reactivity with nocardia, coryne-
bacteria, and staphylococci, and generally do not elicit DTH. 52-54 In 1971, the US-
Japan Cooperative Medical Sciences Program adopted the classification system of
Janicki et al., in which hyperimmune goat serum identified 11 antigens in M.
tuberculosis culture filtrate by immunoprecipitation.55,56 Antigen 5, a 38-kDa

protein, appeared to be partially restricted to M. tuberculosis. 57-<'JO Antigen 6,
alpha antigen, and BeG 85B all refer to a major secreted 32 kDa protein, the
gene for which has recently been cloned. 6l ,62
In 1985 Young et al. reported the successful application of recombinant DNA
techniques to the identification of antigens of the related organism M. leprae. 63 In
these studies, murine monoclonal antibodies were used to identify recombinant
mycobacterial proteins expressed by an M. leprae DNA library in the expression
vector Agtll. Subsequent investigation with human M. leprae-reactive T-cell clones
obtained from immunized volunteers found that nearly half were reactive to
lysates of a single E. coli colony containing epitopes of an 18 kDa mycobacterial
protein. 64
Similar genomic libraries of M. tuberculosis have been screened using panels
of animal sera and murine monoclonal antibodies. 65 ,66 At least seven antigens
have been identified, with molecular weights of 12,14,19,32,32,65, and 71 kDa.
Two of these proteins are members of families of proteins called heat-shock
proteins (HSP): the 65 kDa protein belongs to the family of GroEI-related
proteins originally identified in E. coli, whereas the 71 kDa protein is a member of
the DnaK group.67,68 Heat-shock proteins are a highly conserved group of
proteins found in all organisms, synthesis of which is increased in response to
some environmental stress. HSPs may be induced by temperature (hence the
name), irradiation, exposure to oxidizing agents, metal ions, and anoxia, or in
lymphocytes, stimulation with mitogens or lymphokines. 69-71 The amino acid
sequence of M. tuberculosis HSP-65 has been completely determined, and has
considerable (-70%) homology with a human 65-kDa heat-shock protein. Syn-
thetic peptide fragments of M. tuberculosis HSP-65 have been used to identify
B- and T-cell epitopes in both human and murine systems. The protein appears
to be an immunodominant antigen in immunized mice. Attempts to induce
immunity to M. tuberculosis infection by immunization with HSP-65 have not,
however, been successful. 72
Heat-shock proteins may have the greatest clinical relevance not in immunity
to mycobacteria, but as targets for autoimmunity in rheumatoid arthritis and
other diseases. HSP-65-reactive antibody can be found in patients with active
RA,73 and HSP-65-reactive T cells, ofwhich some express the ,,8 receptor, can be
demonstrated in synovial fluid lymphocytes. 74-76 Although expansion of these
cells may be driven by exogenous antigen, they alternately may res pond to
autologous stress signals. Monocytes stressed by interferon-" or infected with
cytomegalovirus express human HSP-65 determinants and become susceptible to
lysis by M. tuberculosis HSP-65-reactive T cells. 77 Shared amino acid sequences
between human and mycobacterial HSP-65 appear to be responsible for this
observation. 78 Indeed, HSP-65 reactive cells can be easily detected in the neonatal
murine thymus, at a time when antigen-driven expansion has not likely oc-
curred. 46 It has been suggested that the functions of these autoreactive ,,8 cells
include immune surveillance, and regulation oflymphocyte growth and develop-
ment.
The relevance of the heat shock proteins to human tuberculosis remains

unsettled, however. Prominent species-related differences exist among laboratory
animals in innate immunity to mycobacterial infection. 79 Although human
HSP-65-reactive T cell clones can be isolated from healthy tuberculin reactors,
they do not arise with the same frequency as in immunized animals,so Even
among 'VB cells, the frequency of HSP-65 reactivity in humans is only 0.016. 42 This
variation in immune repertoire may in part explain why there is no animal model
that duplicates the pattern of early resolution and late reactivation that is charac-
teristic of human infection. These questions concerning the relevance of the
mycobacterial antigens identified as targets of the immune response in murine
models have provided the impetus for the development of novel systems to study
the interactions of the tubercle bacillus with the human immune system in an
attempt to define immunodominant and potentially protective antigens.
9. HUMAN MONOCLONAL ANTIBODIES
Pleural tuberculosis is generally a self-limited process, and may constitute a

model of local protective mycobacterial immunity. Although distant reactivation
is common, recurrent pleurisy is rare; even recurrent parenchymal pulmonary
disease is as likely to be contralateral or bilateral as ipsilateral. 81 Failure to prevent
distant relapse is not associated with an inadequate local immune response.
Although diminished blastogenesis is present in blood mononuclear cells in up to
30% of patients with tuberculous pleurisy (often in association with skin test
anergy82), pleural mononuclear cell responses are preserved. This cell population
contains increased numbers of M. tuberculosis-reactive CD4+ T lymphocytes as
compared with blood,83-86 and increased numbers ofthe helper-inducer (CD4+
CD45 + ) T-cell subset. 87 Pleural fluid cells do not contain the adherent suppressor
cells found among circulating cells that may be responsible for systemic immune
unresponsiveness. 86 ,88 These findings suggest that local protective mechanisms
are effective in containing the pleural infection despite the inadequacy of the
systemic responses, and that the antigens identified by this local response might
prove to be important in the development of protective vaccines.
To identify potentially protective antigens, we developed clones of M.
tuberculosis-reactive pleural fluid B cells by transformation with EB virus. 89 Such
transformed cells spontaneously elaborate immunoglobulin and can be main-
tained in culture indefinitely, thus circumventing the lack of a suitable nonsecret-
ing human myeloma fusion partner. The frequency of transformable cells was
1122,000, presumably reflecting both the relative infrequency ofB cells in pleural
fluid and the low number of B cells transformed in the course of EBV infection.
We found the frequency of M. tuberculosis-reactive clones to be 54% and 9% in two
subjects. Eighty-three percent of the clones secreted IgM; the remainder, IgG.
Western blot analysis of M. tuberculosis using these antibodies identified six
distinct patterns of reactivity. One monoclonal antibody identified 33-, 31.5-, and
29-kDa bands on western blot analysis ofboth M. tuberculosis filtrate and sonicate;
its reactivity by ELISA was limited to M. tubereulosis. Eight antibodies identified

a 31.5 kDa band in culture filtrate, and a 62 kDa band in M. tubereulosis sonicate;
these antibodies had ELISA reactivity that extended to M. avium. The 31.5 kDa
band was identified as antigen 6/alpha antigen. Preabsorbtion with purified alpha
antigen inhibited western blot recognition of the 33- and 29-kDa bands as weIl as
of 31.5 kDa, suggesting they were closely related antigenically. Six antibodies
identified multiple bands, and had cross-reactivity that included M. avium and M.
kansasii. Three monoclonal antibodies developed from a second subject identified
bands at 70, 37, and 28 kDa inM. tubereulosis filtrate. There was no reactivity with
recombinant M. tuberculosis 71- or 65-kDa antigens; in general, reactivity in M.
tubereulosis sonicate was less intense than that with culture filtrate, or was absent
entirely. These studies suggest that antigen 6/alpha antigen, and secreted myco-
bacterial proteins in general, may be uniquely accessible to the human immune
response du ring the course of naturally occurring infection.
10. MONONUCLEAR PHAGOCYTE-ACfIVATING

MYCOBACfERIAL PROTEINS
The activation of a complex immune network hinges on the expression of

mycobacterial antigens on the surface of infected mononuclear phagocytes, and
the appropriate release of the cytokines necessary for immune induction. The
study of interactions of tubercle bacilli with mononuclear phagocytes of the naive
host may therefore yield valuable insights into mycobacterial immunity.
Interleukin 1 (lL-I) is a monocyte product that plays an essential role in
immune induction, facilitating both interleukin 2 (IL-2) release by T lympho-
cytes90 and the expression of IL-2 receptors on their surface. 9! This is critical
in the expansion of antigen-specific T lymphocytes and in the subsequent
elaboration of monocyte-activating factors such as interferon--y. The synthesis and
release of IL-I (and perhaps other required monokines) are induced by a variety
of microbial stimuli, including polysaccharides,92 phorbols,93 inert particles,94
intact gram-positive and negative bacteria,95 spirochetes,96 and mycobacteria. 97
All commonly used adjuvants are potent inducers of IL-l and TNF; it is likely that
their immune-inducing properties are a result of their macrophage-activating
effects.
The capacity for induction of IL-l by soluble factors has been thought mainly
to reside in bacterial polysaccharides; indeed, E. eoli lipopolysaccharide (LPS),
the prototypic agent used to induce IL-l release in vitro, is active in this respect
even in nanogram concentrations. M ycobacteria are potent inducers of monocyte
activation. This is not dependent on the presence of intact particles, nor is it
mediated by lipopolysaccharide, which mycobacteria lack. We have demon-
strated induction of IL-l and TNF by the dialysed filtrate of M. tuberculosis culture
medium, and by purified protein derivative (PPD).98 PPD-induced production of
IL-I is unaffected by polymyxin, a cationic polypeptide antibiotic that binds to the
lipid A moiety of LPS and inactivates most of its biologic activities,99.100 thus also
excluding a role for LPS contamination. This capacity for induction of monocyte
activation by protein antigens is novel, and is not shared by other soluble microbial
antigens such as tetanus toxoid or streptolysin 0.
To identify the specific mycobacterial constituents capable of inducing
monocyte activation, we have adapted the technique of T-cell western blot anal-
ysis,lOl,102 to study the production of cytokines by monocytes. In these experi-
ments, culture filtrate of M. tuberculosis was subjected to SDS-PAGE, transferred
to nitrocellulose paper, cut into 2-mm horizontal strips, dissolved in DMSO and
precipitated in an aqueous buffer to produce a suspension of protein-bearing
particles. The particles were incubated with monocytes in the presence of poly-
myxin Band the supernatants assayed for TNF (Fig. 4). Two fractions stimulated
monocyte production ofTNF. These fractions also expressed IL-l activity, as did
two additional fractions. The magnitude of induced cytokine production was
comparable to that of intact mycobacteria, or E. coli LPS. Many other protein
bands identified inM. tuberculosis filtrate by gold stain failed to induce production
of TNF. Nitrocellulose paper alone also failed to induce TNF activity.
T-cell Western blot analysis suggested that concordant peaks of T-cell and
Monocyte westem blot

M. tuberculosis filtrate
60 6. LPS~ PMB
.
40
-untreated
. - - LAM depleted ,, ,,
I
I ,
,
• M. tb
...... protease treated I ,
,, ,,
I ,
,.I"', ,I,
I
20
Ä LPS+PMB
15 20 ctrl
fraction
FIGURE 4. "Monocyte Western blot" of M. tuberculosis filtrate. Filtrate fractions were either
untreated (solid line), depleted of lipoarabinomannan (LAM) by affinity chromatography
(dashed line), or protease treated (dotted line) prior to culture with monocytes. TNF was
measured by L929 bioassay. Fractions were assayed in the presence of polymyxin B to minimize
effects of contaminating lipopolysaccharide (LPS). Controls included whole M. tuberculosis strain
H 37 Rv and E. coli LPS.
TABLE I
M. tuberculosis Shows T-Cell
and Monocyte Reactivity*
T-cell Monocyte
Blastogenesis TNF IL-l
19 kDa 20.5 kDa 20.5 kDa
34 35
47 46 46
68
85 90
*Fractions of M. tuberculosis filtrate were tested for their capac-
ity to induce blastogenic responses in lymphocytes of healthy
tuberculin reactors. and for the expression of the cytokines
TNF and IL-l by monocytes of tuberculin nonreactors.
monocyte reactivity could be demonstrated in M. tuberculosis filtrate. Table I

shows blastogenic responses of two healthy tuberculin reactors, in which five
major peaks ofT-cell reactivity could be detected, corresponding to fractions that
stimulate monocytes.
Although mycobacteria do not possess bacteriallipopolysaccharide, they do
contain lipoarabinomannan (LAM), a polysaccharide with many of the same
physicochemical properties as LPS. LAM has an apparent molecular weight of
approximately 30 kDa on SDS-PAGE, but can form larger complexes with
proteins and thus have variable migration on gel electrophoresis. LAM stimulates
monocyte production of TNF, although it may be a less potent stimulus than
LPS.103 To ascertain whether LAM might be responsible for our observations, M.
tuberculosis filtrate was depleted of LAM by ammonium sulfate precipitation
followed by immunoaffinity chromatography using monoclonal anti-LAM anti-
body. Depletion of approximately 95% of the initial LAM was confirmed by
western blot. The remaining LAM was detected as a single band with an apparent
molecular weight of 53 kDa. This preparation retained its capacity to induce TNF
production, as shown by the dashed line in Figure 4.
To confirm that proteins were in fact responsible for the cytokine induction,
LAM-depleted M. tuberculosis filtrate was then subjected to protease digestion.
Protein degradation was confirmed by gold staining of the nitrocellulose transfer.
Western blot analysis of the digest with anti-LAM monoclonal antibody revealed a
single band of LAM at approximately 60 kDa. The protease digest failed to
induce monocyte production of TNF (dotted line, Fig. 4).
Further studies using 2-dimensional gel electrophoresis have identified one
of the TNF-inducing proteins as a 58-kDa protein with the N-terminal amino acid
sequence ofGly-Glu-Lys-Thr-Pro-Asp-(?)-Val-Phe-(?)-(?)-Ala. o4 Given the role of
TNF in mycobacterial immunity, and the prominence of monocyte dysfunction in
myeobaeterial immunopathology, it is likely that this novel protein plays an

important role in the pathogenesis of tubereulosis.
11. MYCOBACTERIUM AVIUM
M. avium and M. intracellulare are nontubereulous myeobaeteria, members of

Runyon's group III, the nonphotoehromogens. These organisms are ubiquitous
in the environment, and may be identified in inereased numbers in the aerosol
droplets of fresh water found over rapidly Howing streams, partieularly in the
eastern USo They are relatively avirulent.
In the pre-AIDS era, M. avium was an infrequent human pathogen, mainly
eausing ehronie loeal infeetion in patients with underlying pulmonary disease. In
AIDS patients, however, disseminated infeetion with M. avium is quite eommon.
Autopsy series suggest as mäny as 50% of patients with AIDS have M. avium
infection at the time of death. Large numbers of organisms can be found in many
organs, including blood, lymph nodes, bone marrow, liver, and spleen. There
typieally is liule or no granulomatous inHammatory response, and maerophages
filled with acid-fast bacteria may be easily identified.
Treatment of M. avium infeetion is diffieult beeause the organisms are
typically multiple-drug resistant, and there is little correlation between in vitro
suseeptibility and in vivo responses. 105 Historically, drug regimens including
several antituberculous drugs have been combined with surgical resection of
localized pulmonary lesions. In AIDS patients, uncontrolled trials of regimens
including rifabutin (a rifampin derivative) and clofazamine (a antileprosy drug)
have been disappointing,I06 although drug combinations including ethambutol,
rifabutin, clofazamine, ciproHoxacin, and arnikaein may prove to be more effec-
tive.
Strains of M. avium differ in their capacity to replicate within healthy
monocytes. Flat transparent colonial morphology (as opposed to domed and
opaque) is associated with the ability to replicate intracellularly in vitro. 107 Animal
models have confirmed this as a marker for virulence. 108 Virulent organisms are
phagoeytosed less well than nonvirulent organisms. Neither AIDS-associated
strains nor those containing plasmids appear more virulent in vitro. Repeated
passage in vitro results in transition from Hat to domed colony morphology and
loss of antimicrobial resistance; this can be reversed by passage in vivo. Most
patient isolates are initially Hat and transparent.
Data on the role of interferon-" in macrophage activation for killing of M.
avium are contradictory. We have found that interferon-" increases growth inhibi-
tion of most, but not all strains of M. avium, but the magnitude of this increase is
only 20-30%.107 Other researchers have found no inhibition,109 or even enhance-
ment of intracellular mycobacterial growth llO in the presence of interferon-".
One report suggested a role for TNF in intracellular killing of M. avium,109
although we have been unable to duplieate this observation. Monocytes and
macrophages from patients with AIDS with or without M. avium infection show
142 ROBERT S. WALLIS and JERROLD ]. ELLNER
ingestion and growth inhibition comparable to that of healthy subjects,lll and

res pond normally to interferon-)' in terms of oxidative burst,1l2.113 although they
may not res pond equaHy weH to undefined serum factors. 1l4 As lymphocytes from
such patients produce subnormal amounts of interferon-'Y foHowing stimulation
with mitogens, BeG, or soluble antigens of M. avium,1l5.116 administration ofthis
lymphokine may prove beneficial in the treatment of M. avium infections.
Several issues remain unresolved induding the basis for the increased viru-
lence of the Hat, transparent colonial forms, and whether any cytokine or
combination of cytokines exceeds the MAF activity of interferon-'Y.
REFERENCES
1. Murray, C., Styblo, K., and Rouillon, A., 1989, Submission to World Bank Health Sector
Priorities Review.
2. ten Dam, H., 1984, Research on BCG vaccination, Adv. Tuberc. Res. 21:79-106.
3. Eickhoff, T., 1977, The current status of BCG immunization against tuberculosis, Annu.
Rev. Med. 28:411-423.
4. Sencer, D.]., Dull, H. B., Barrett-Connor, E., Burnett, L. S., Eisea, W R., Langmuir, A. D.,
Prather, E. C., Schiff, G. M., Shore, E. G., Meyer, H., Katz, S. L., Comstock, G. W, Koch-
Weser, D., Oatway, WH., Reichman, L. B., Sbarbaro, ]. A., and Smith, D., 1975, BCG
vaccines for tuberculosis: official ATS-CDC statement, Am. Rev. Respir. Dis. 112:478-480.
5. Hart, P. and Sutherland, 1., 1977, BCG and vole bacillus vaccines in the prevention of
tuberculosis in adolescence and early adult life, Br. Med. J. 2:293-295.
6. 1979, Tuberculosis Prevention trial: Trial of BCG vaccines in South India for tuberculosis
prevention: first report, Bull WHo. 57:819-827.
7. Tripathy, S., 1986, Presented at the 26th World Conference of the International Union
Against Tuberculosis, Singapore, November, 1986.
8. Di Perri, G., Cruciani, M., Danzi, M. C., Luzzati, R., De Checchi, G., Malena, M.,
Pizzighella, S., Mazzi, R., Solbiati, M., and Concia, E., 1989, Nosocomial epidemie of active
tuberculosis among HIV-infected patients, Lancet 2:1502-1504.
9. Selwyn, P. A., HarteI, D., Lewis, V. A., Schoenbaum, E. E., Vermund, S. H., Klein, R. S.,
Walker, A. T., and Friedland, G. H., 1989, A prospective study of the risk of tuberculosis
among intravenous drug users with HIV infection, N. Engl. J. Med. 320:545-550.
10. Stead, W W, 1981, Tuberculosis among elderly persons: An outbreak in nursing horne,
Ann. Intern. Med. 94:606-610.
11. Daniel, T. M., Oxtoby, M.]., Pinto, E., and Moreno, S., 1981, The immune spectrum in
patients with pulmonary tuberculosis, Am. Rev. Respir. Dis. 123:556-559.
12. Nash, D. R. and Douglass,]. E., 1980, Anergy in patients with pulmonary tuberculosis,
Chest 77:32-36.
13. Ellner,]., 1978, Suppressor adherent cells in human tuberculosis,j.lmmurwl. 121:2573-2578.
14. Toosi, Z., Kleinhenz, M. E., and Ellner,].]., 1984, Defective I1-2 production and responsive-
ness in pulmonary tuberculosis,J. Exp. Med. 163:1162-1172.
15. Shiratsuchi, H., Okuda, Y., and Tsuyuguchi, 1., 1987, Recombinant human I1-2 reverses in
vitro-deficient cell mediated immune responses to PPD by Iymphocytes of tuberculous
patients, Infect. Immun. 55:2126-2131.
16. Toossi, Z., Lapurger,]. P., Ondash, R.]., Sedor,]. R., and Ellner,].]., 1990, Expression of
functional interleukin-2 receptors by peripheral blood monocytes from patients with active
pulmonary tuberculosis,J. Clin.lnvest. 85:1777-1784.
17. Tweardy, D. J., Schacter, B. Z., and Ellner, J. J., 1984, Association of altered dynamics of
monocyte surface expression of HLA DR with immunosuppression in tuberculosis,J Infoct.
Dis. 149:31-37.
18. Fujiwara, H., Kleinhenz, M. E., Wallis, R. S., and Ellner,J.J., 1986, Increased interleukin-l
production and monocyte suppressor activity associated with human tuberculosis, Am. Rev.
Respir. Dis. 133:73-77.
19. Cox, R. A., Downs, M., Neimes, R. E., Ognibene, A. J., Yamashita, T. S., and Ellner, J. J.,
1988, Immunogenetic analysis of human tuberculosis,J Infoct. Dis. 158:1302-1308.
20. Kleinhenz, M. E. and Ellner, J. J., 1987, Antigen responsiveness during tuberculosis:
Regulatory interactions of T-cell subpopulations and adherent cells, J Lab. Clin. Med.
110:31-40.
21. Toossi, Z., Edmunds, K. E., Tomford,J. w., and Ellner,J.J., 1989, Suppression ofpurified
protein derivative-induced interleukin-2 production by interaction of CD16 (Leu 11 reac-
tive) lymphocytes and monocytes,J Infect. Dis. 159:352-356.
22. Crowle, A. J., and Elkins, N., 1990, Relative permissiveness of macrophages from black and
white people for virulent tubercle bacilli, Infoct. Immun. 58:632-638.
23. Nacy, C. A., Meltzer, M. S., Leonard, E.J., and Wyler, D.J.,1981, Intracellular repliation and
lymphokine-induced destruction of Leishmania tropica in C3H/HeN mouse macrophages,
J Immunol. 127:2381-2386.
24. Nathan, C., Murray, H.W., Wiehe, M. w., and Rubin, B. Y., 1985, Identification of
interferon-gamma as the lymphokine that activates human macrophage oxidative metabo-
lism and antimicrobial activity,J Exp. Med. 158:670.
25. Bhardwaj, N., Nash, T. w., and Horwitz, M. A., 1986, IFN-'Y-activated human monocytes
inhibit the intracellular multiplication of Legionella pneumophila,J Immunol. 137:2662-
2669.
26. Rook, G. A. w., Champion, B. R., Steele, J., Varey, A. M., and Stanford, J. L., 1985, I-A
restricted activation by T-cell lines of anti-tuherculosis activity in murine macrophages,
Clin. Exp. Immunol. 59:414.
27. FIeseh, I. and Kaufmann, S. E., 1987, Mycobacterial growth inhibition by interferon-
gamma-activated bone marrow macrophages and differential susceptibility among strains
of Mycobacterium tuberculosis, J Immunol. 138:4408-4413.
28. Rook, G. A., Steele, J., Fraher, L., Barker, S., Karmali, R., O'Riordan, J., and Stanford, J.,
1986, Vitamin D3, gamma interferon, and control of proliferation of Mycobacterium
tuberculosis by human monocytes, Immunology 57:159-163.
29. Douvas, G. S., Looker, D. L., Vatter, A. E., and Crowle, A. J., 1985, Gamma interferon
activates human macrophages to become tumoricidal and leishmanicidal but enhances
replication of macrophage-associated mycobacteria, Infect. Immun. 50:1.
30. Tracey, K. J., Lowry, S. F., and Cerami, A., 1988, Cachectin: A hormone that triggers acute
shock and chronic cachexia,J Infect. Dis. 157:413-420.
31. Kindler, v., Sappino, A., Grau, G. E., Piguet, 1'., and Vassaili, 1'.,1989, The inducing role of
TNF in the development of bactericidal granulomas during BCG infection, Cell 56:
731-740.
32. De Libero, G., FIeseh, 1., and Kaufmann, S. H. E., 1988, Mycobacterial-reactive Lyt-2+
T-cell lines, Eur. J Immunol. 18:59-66.
33. Ottenhoff, T. H. M., Ab, B. K., van Embden,J. D. A., Thole,J. E. R., and Kiessling, R.,1988,
The recombinant 65 kD heat shock protein of M. bovis BCG/M. tuherculosis is a target
moleeule for CD4 cytotoxic T lymphocytes that lyse human monocytes, J Exp. Med.
168:1947-1952.
34. Muller, I., Cobbold, S. P., Waldmann, H., and Kaufmann, S. H. E., 1987, Impaired
resistance to Mycobacterium tuberculosis infection after selective in vivo depletion of
L3T4+ and Lyt-2+ T cells, Infect. Immun. 55:2037-2041.
144 ROBERT S. WALLIS and jERROLD J. ELLNER
35. Orme, I. M. and Collins, F. M., 1984, Adoptive protection of the Mycobacterium
tuberculosis-infected lung, Gell Immunol. 84:113-120.
36. Orme, I. M., 1987, The kinetics of emergence and loss mediator T lymphocytes acquired in
response to infection with M. tubereulosis,] Immunol. 138:293-298.
37. Rook, G. A. W, Steele,J., Fraher, L., Barker, S., Karmali, R., O'Riordan,J., and Stanford,J.,
1986, Vitamin D3, gamma interferon, and control of proliferation of Myeobacterium
tubereulosis by human monocytes, Immunol. 57:159-162.
38. Mason, R. S., Frankei, T., Chan, Y. L., Lissner, D., and Posen, S., 1984, Vitamin D conversion
by sarcoid lymph node homogenate, Ann. Intern. Med. 100:59-6l.
39. Barnes, P. F., Modlin, R. L., Bohle, D. D., and Adams,J. S., 1989, Transpleural gradient of
1,25 hihydroxyvitamin D in tuberculosis pleuritis,] Glin.lnvest. 83:1527-1532.
40. Patel, S., Wacholtz, M. C., Duby, A. D., Thiele, D. L., and Lipsky, P. E., 1989, Analysis ofthe
functional capabilities of CD3 + CD4 - CD8- and CD3 + CD4 + CD8+ human T cell clones,]
Immunol. 143:1108-1117.
41. Haregewoin, A., Soman, G., Horn, R. C., and Finberg, R. W, 1989, Human ",/8+ T cells
res pond to mycobacterial heat-shock protein, Nature 340:309-312.
42. Kabelitz, D., Bender, A., Schondelmaier, S., Schoel, B., and Kaufmann, S. H. E., 1990, A
large fraction of human peripheral blood ",/8+ T cells is activated by M. tubereulosis but not
by its 65 kD heat shock protein,] Exp. Med. 171:667-679.
43. Modlin, R. L., Pirmez, C., Hofman, F. M., Torigian, v., Uyemura, K., Rea, T. H., Bloom, B.
R., and Brenner, M. B., 1989, Lymphocytes bearing antigen-specific gamma delta T-cell
receptors accumulate in human infectious disease lesions, Nature 339:544-548.
44. janis, E. M., Kaufmann, S. H. E., Schwartz, R. H., and Pardoll, D. M., 1989, Activation of",/8
T cells in the primary immune response to M. tubereulosis, Seienee 244:713-716.
45. Augustin, A., Kubo, R. T., and Sim, G. K., 1989, Resident pulmonary lymphocytes
expressing the ",/8 T cell receptor, Nature 340:239-24l.
46. O'Brien, R. L., Happ, M. P., DaIlas, A., Palmer, E., Kubo, R., and Born, W, 1989,
Stimulation of a major subset of lymphocytes expressing T-cell receptor ",/8 by an antigen
derived from M. tubereulosis, Gell 57:667-674.
47. Boyden, S. and Sorkin, E., 1956, Adv. Tubere. Res. 7:17.
48. Azuma, 1., Yamamura, Y., and Fukushi, K., 1968, Fractionation of mycobacterial cell wall.
Isolation of arabinose mycolate and arabinogalactan from ceIl wall fraction of Myeobac-
terium tubereulosis strain Aoyama B,] Bacteriol. 96:1885-1887.
49. Affronti, L. F. and Seibert, F. B., 1965, Some early investigations of Myeobacterium tuber-
eulosis, Am. Rev. Respir. Dis. 92:1-8.
50. Daniel T. and Hinz, C., 1974, Reactivity of purified proteins and polysaccharides from
Myeobaeterium tubereulosis in delayed skin test and cultured Iymphocyte mitogenesis assays,
51. Chaparas, S., Thor, D., Godfrey, H., Baer, H., and Hedrick, S., 1970, Tuberculin-active
carbohydrate that induces inhibition of macrophage migration by not Iymphocyte transfor-
mation, Seienee 170:637.
52. Chaparas, S., Thor, D., and Hedrick, S., 1971, Comparison of Iymphocyte transformation
inhibition of macrophage migration and skin tests using dialyzable and nondialyzable
fractions from M. bovis BCG,] Immunol. 107:149.
53. Azurna, 1., Ajisaka, M., and Yamamura, Y., 1968, Biology ofthe mycobacterioses: Chemical
and immunological studies on peptides and polysaccharides from tubercle bacilIi, Ann. NY
Aead. Sei. 154:88-97.
54. Minden, P., McClatchy, J., and Farr, R., 1972, Shared antigens between heterologous
bacterial species, Infoct. Immun. 6:574-582.
55. janicki, B., Chaparas, S., Daniel, T., Kubica, G., Wright, G., and Yee, G., 1971, A reference
system for antigens of Mycobacterium tuberculosis, Am. Rev. Respir. Dis. 104:602-604.
56. Daniel, T. and Affronti, L., 1973, Immunoelectrophoretic analysis of antigenie constituents
of Seibert fractions of mycobacterial culture filtrates: Identification with the proposed
United States-Japan reference nomenclature, Am. Rev. Respir. Dis. 108:1244.
57. Daniel, T., Ellner, j., Todd, L., McCoy, D., Payne, v., Anderson, P., and Bhe, E, 1979,
Immunobiology and species distribution of Myeobacterium tubereulosis antigen 5, Infect.
Immun. 24:77-82.
58. Daniel, T., and Olds, G., 1985, Demonstration of a shared epitope among mycobacterial
antigens using a monoclonal antibody, Clin. Exp. Immunol. 60:249-258.
59. Daniel, T., Balestrino, E., Balestrino, 0., Davidson, P., Debanne, S., Kataria, S., Kataria, Y.,
and Scocozza, j., 1982, The tuberculin specificity in humans of Myeobaeterium tubereulosis
antigen 5, Am. Rev. Respir. Dis. 126:600-606.
60. Benjamin, R. and Daniel, T., 1982, Serodiagnosis oftuberculosis using the enzyme-linked
immunoabsorbent assay (ELISA) of antibody to Myeobacterium tubereulosis antigen 5,
Am. Rev. Respir. Dis. 126:1013-1016.
61. Matsuo, K., Yamaguchi, R., Yamazaki, A., Tasaka, H., and Yamada, T., 1988, Cloning and
expression of the BCG gene for extracellular CY. antigen, J Bacteriol. 170:3847-3854.
62. Abou-Zeid, C., Ratliff, T. L., Wiker, H. G., Harboe, M., Bennedsen"j., and Rook, G. A. w.,
1988, Characterization of fibronectin-binding antigens released by M. tubereulosis and
M. bovis BCG, Infeet. Immun. 56:3046-3051.
63. Young, R., Mehra, v., Sweetwer, D., Buchanan, T., Clark-Curtiss, j., Davis, R., and Bloom,
B., 1985, Genes for the major protein antigens of the leprosy parasite M. leprae, Nature
316:450.
64. Mustafa, A., Gil!, H., Nerland, A., Britton, w., Mehra, v., Bloom, B., Young, R., and Godal,
T., 1986, Human T-cell clones recognize a major M. leprae protein antigen expressed in
E. eoli, Nature 319:63.
65. Engers, H. D., Houba, v., Bennedsen,j., Buchanan, T. M., Chaparas, S. D., and Kadival, G.,
et al., 1983, Results of a WHO-sponsored workshop to characterize antigens recognized by
Mycobacterium-specific monoclonal antibodies, Infect. Immun. 51:718-720.
66. Young, R., Bloom, B., Grosskinsky, C., Ivanyi, j., Thomas, D., and Davis, R., 1985,
Dissection of M. tubereulosis antigens using recombinant DNA, Proe. Natl. Aead. Sei. USA
82:2583-2587.
67. Shinnick, T. M., Vodkin, M. H., and Williams,j. C., 1988, The Mycobacterium tuberculosis 65-
kilodalton antigen is a heat shock protein which corresponds to common antigen ;md to the
Escherichia coli GroEL protein, Infect. Immun. 56:446-451.
68. Britton, W. j., Hellqvist, M., Basten, A., and Inglis, A. S., 1986, Immunoreactivity of a 70-
kD protein purified from M. bovis BCG by monoclonal antibody affinity chromatography,
J Exp. Med . .164:695-708.
69. Lindquist, S. and Craig, E. A., 1988, The heat-shock proteins, Annu. Rev. Genet. 22:631-677.
70. Collins, P. L. and Hightower, L. E., 1982, Newcastle disease virus stimulates the cellular
accumulation of stress (heatshock) mRNA' a and proteins, J Virol. 44:703-707.
71. Ferris, D. K., Harel-Bellan, A., Morimoto, R. 1., Welch, W. j., and Farrar, W. L., 1988,
Mitogen and lymphokine stimulation of heat shock proteins in T lymphocytes, Proe. Natl.
Acad. Sei. USA 85:3850-3854.
72. Watson,j. D., 1989, Leprosy: Understanding protective immunity, Immunol. Today 10:218-221.
73. Bahr, G. M., Rook, G. A., al-Saffar, M., Van Embden,j., Stanford,j. L., and Behbehani, K.,
1988, Antibody levels to mycobacteria in relation to HLA types: Evidence for non-HLA-
linked high levels of antibody to the 65 kD HSP of M. bovis in rheumatoid arthritis, Clin
Exp. Immunol. 74:211-2115.
74. Res, P. C. M., Breedwell, E C., Schaar, C. G., and van Eden, w., 1988, Synovial fluid T-cell
reactivity against 65-kD heat shock protein of mycobacteria in early chronic arthritis,
Ltmeet 27:478.
146 ROBERT S. WALLIS and JERROLD]. ELLNER
75. Gaston, ]. S., Life, P. F., Bailey, L. C., and Bacon, P. A., 1989, In vitra response to a 65-kD
mycobacterial protein by synovial T cells from inflammatory arthritis patients,J Immunol.
143:2494.
76. Holoshitz,]., Koning, F., Coligan,]. E., De Bruyn,]., and Strober, S., 1989, Isolation of
CD4- CD8- mycobacteria-reactive T Iymphocyte clones from rheumatoid arthritis syno-
vial fluid, Nature 339:226.
77. Koga, T., Wand-Wurttenberger, A., DeBruyn,]., Munk, M. E., Schoel, B., and Kaufmann,
S. H. E., 1989, T cells against a bacterial heat shock protein recognize stressed macro-
phages, Seience 245:1112-1115.
78. Munk, M . E., Schoel, B., Modrow, S., Karr, R. W, Young, R. A., and Kaufmann, S. H. E.,
1989, T Iymphocytes from healthy individuals with specificity for self epitopes shared by
the mycobacterial and human 65-kD HSp, J Immunol. 144:2844-2849.
79. Smith, D., 1985, Protective effect of BCG in experimental tuberculosis, Adv. Tubere. Res.
22:1-97.
80. Husson, R. and Young, R., 1987, Genes for the major protein antigens of M. tubereulosis:
The etiologic agents of TB and leprosy share immunodominant an antigen, Proe. Natl.
Aead. Sei. USA 84:1679-1683.
81. Roper, Wand Waring,]., 1955, Primary serofibrinous pleural effusion in military person-
nei, Am. Rev. Tubere. Pulm. Dis. 71:616-634.
82. Berger, H. and Mejia, E., 1973, Tuberculous pleurisy, Ghest 63:88-92.
83. Shimokata, K., Kawachi, H., Kishimoto, H., Maeda, F., and Ito, Y., 1982, Local cellular
immunity in tuberculous pleurisy, Am. Rev. Respir. Dis. 126:822-824.
84. Rossi, G., Balbi, B., and Manca, F., 1987, Tuberculous pleural effusions, Evidence for
selective presem:e of PPD-specific T-Iymphocytes at site of inflammation in the early phase
of the infection, Am. Rev. Respir. Dis. 136:575-579.
85. Fujiwara, H. and Tsuyuguchi, 1., 1986, Frequency of tuberculin-reactive T-Iymphocytes in
pleural fluid and blood from patients with tuberculous pleurisy, Ghest 89:530-532.
86. Fujiwara, H., Okuda, Y., Fukukawa, T., and Tsuyuguchi, 1., 1982, In vitro tubereuIin
reactivity of Iymphocytes from patients with tuberculous pleurisy, Infect. Immun. 35:
402-409.
87. Barnes, P., Mistry, S., Cooper, C., Pirmez, C., Rea, T., and Modlin, R., 1989, Compartmen-
talization of a CD4 + T Iymphocyte subpopulation in tuberculous pleuritis,J Immunol. 142:
1114-1119.
88. Ellner,]., 1978, Pleural fluid and peripheral blood Iymphocyte function in tuberculosis,
Ann. Intern Med. 89:932-933.
89. Wallis, R. S., Aide, S. L., Havlir, D. v., Amir-Tahmasseb, H., Daniel, T. M., and Ellner,].].,
1989, Identification of antigens of M. tubereulosis using human monoclonal antibodies,J
Glin. Invest. 84:214-219.
90. Chu, E. T., Lareau, M., Rosenwasser, L.]., Dinarello, C. A., and Geha, R. S., 1985, Antigen
presentation by EBV-B cells to resting and activated T cells: Role of interleukin 1, J
Immunol. 134:1676-1681.
91. Kaye,]., Gillis, S., Mizel, S. B., Shevach, E. M., Malek, T. R., Dinarello, C. A., Lachman, L.
B., and Janeway, C. A. , 1984, Growth of a cloned helper T ceilline induced by a monoclonal
antibody specific for the antigen receptor: IL-l is required for the expression of receptors
for interleukin 2,J Immunol. 133:1339-1345.
92. Gery, 1., Gershon, R. K., and Waksman, B. H., 1972, Potentiation of the T Iymphocyte
response to mitogens,J Exp. Med. 136:128.
93. Mizel, S. B. and MizeI, D., 1981, Purification to apparent homogeneity of murine IL-l,J
Immunol. 126:834.
94. Gery, 1., Davies, P., Derr,]., Drett, N., and Barranges, ]. A., 1981, Relationship between
production and release of Iymphocyte activating factor IL-l, Gell Immunol. 64:293.
95. Lachman, L. B., 1983, Interleukin 1: release from LPS-stimulated mononuclear phago-
cytes, in Beneficial Effeets of Endotoxin (A. Nowotny, ed.) Plenum Press, New York, p. 283.
96. Habicht, G., Beck, G., Benach, j. L., Coleman, j. L., and Leichtling, K. D., 1985, Lyme
disease spirochetes induce human and murine interleukin 1 production,} Immunol. 134:3147.
97. Valone, S. E., Rich, E. A., Wallis, R. S., and Ellner,j.j., 1988, Expression oftumor necrosis
factor in vitro by human mononuclear phagocytes stimulated with whole M. tuberculosis
BCG and mycobacterial antigens, Infeet. Immun. 56:3313-3315.
98. Wallis, R. S., Fujiwara, H., and Ellner, j. j., 1986, Direct stimulation of monocyte release of
interleukin 1 by mycobacterial protein antigens,} Immunol. 136:193-196.
99. Morrison, D. C. andjacobs, D. M., 1976, Inhibition oflipopolysaccharide-initiated activa-
tion of serum complement by polymyxin B, Infeet. Immun. 13:298-301.
100. Duff, G. W. and Atkins, E., 1982, The inhibitory effect of polymyxin B on endotoxin-
induced endogenous pyrogen production,J Immunol. Metlwds 52:333-340.
101. Abou-Zeid, C., Filley, E., Steele,j., and Rook G. A. w., 1987, A simple new method for using
antigens separated by PAGE to stimulate lymphocytes in vitro after converting bands cut
from western blots in antigen-bearing particles,} Immunol. Metlwds 98:5-10.
102. Young, D. B. and Lamb, j. R., 1986, T lymphocytes respond to a solid phase antigen: A
novel approach to the molecular analysis of cellular immunity, Immunol. 59:167-171.
103. Moreno, C., Taverne,j., Mehlert, A., Bate, C. A., Brealey, R.j., Meager, A., Rook, G. A., and
Playfair, j. H., 1989, Lipoarabinomannan from M. tubereulosis induces the production of
tumour necrosis factor from human and murine macrophages, Clin. Exp. Immunol. 76:
240-245.
104. Wallis, R. S., Phillips, M., and Paranjape, R., 1993, Identification by 2-D gel electrophoresis
of a 58 kD monocyte activating protein of Myeobacterium tuberculosis, Infeet. Immun. 61:
627-632.
105. Yajko, D. M., Nassos, P. S., and Hadley, W. K., 1987, Therapeutic implications ofinhibition
versus killing ofMycobacterium avium complex by antimicrobial agents, Antimicrob. Agents
Chemother. 31:117-120.
106. Masur, H., Tuazon, C., Gill, v., Grimes, G., Baird, B., Fauci, A. S., and Lane, H. C., 1987,
Effect of combined clofazimine and ansamycin therapy on Mycobacterium avium-
Mycobacterium intracellulare bacteremia in patients with AIDS,} Infeet. Dis. 155:127-129.
107. Toba, H., Crawford,j. T., and Ellner,j.j., 1989, Pathogenicity ofMycobacterium avium for
human monocytes: Absence of macrophage-activating factor activity of gamma interferon,
Infeet. Immun. 57:239-244.
108. Schaefer, W. B., Davis, C. L., and Cohn, M. L., 1970, Pathogenicity oftransparent, opaque,
and rough variants of Mycobacterium avium in chickens and mice, Am. Rev. Respir. Dis.
102:499-506.
109. Bermudez, L. E. and Young, L. S., 1988, Tumor necrosis factor, alone or in combination
with Il-2, but not IFN-gamma, is associated with macrophage killing of Myeobacterium
avium complex,} Immunol. 140:3006-3013.
110. Douvas, G. S., Looker, D. L., Vatter, A. E., and Crowle, A. j., 1985, Gamma interferon
activates human macrophages to become tumoricidal and leishmanicidal but enhances
replication of macrophage-associated mycobacteria, Infect. Immun. 50:1-8.
111. Schnittman, S., Lane, H. C., Witebsky, F. G., Gosey, L. L., Hoggan, M. D., and Fauci, A. S.,
1988, Host defense against M. avium complex,} Clin. Immunol. 8:234-243.
112. Murray, H. w., Scavuzzo, D., jacobs, j. L., Kaplan, M. H., Libby, D. M., Schindler, j., and
Roberts, R. B., 1987, In vitro and in vivo activation of human mononuclear phagocytes by
interferon-gamma. Studies with normal and AIDS monocytes,} Immunol. 138:2457-2462.
113. Murray, H. w., Gellene, W. A., Libby, D. M., Rothermel, C. D., and Rubin, B. Y., 1985,
Activation of tissue macrophages from AIDS patients: In vitro response of AIDS alveolar
macrophages to lymphokines and IFN'Y,} Immunol. 135:2374-2377.
148 ROBERT S. WALLIS and JERROLD ]. ELLNER
114. Crowle, A. ]., Cohn, n L., and Poche, P., 1989, Defects in sera from acquired immuno-
deficiency syndrome (AIDS) patients and from non-AIDS patients with Mycobacterium
avium infection which decrease macrophage resistance to M. avium, Irifect. Immun. 57:
1445-1451.
115. Murray, H. W, Rubin, B. Y., Masur, H., and Roberts, R. B., 1984, Impaired production of
lymphokines and immune (gamma) interferon in the acquired immunodeficiency syn-
drome, N. Engl. J Med. 310:883-889.
116. Murray, H. W, Scavuzzo, n A., Chaparas, S. n, and Roberts, R. B., 1988, T lymphocyte
responses to mycobacterial antigen in AIDS patients with disseminated Mycobacterium
avium-Mycobacterium intracellulare infection, ehest 93:922-925.
9
Bordetella pertussis
FREDERICK R. VOGEL
1. INTRODUCTION
Pertussis, or whooping cough, is a communicable, acute infection of the respira-

tory tract caused by Bordetella pertussis. Pertussis begins with the colonization and
unrestricted growth of the bacteria in the ciliated epithelium of the respiratory
tract during the catarrhal stage of the disease. The primary pathology of
pertussis is a tracheobronchitis; however, pneumonia develops in up to 20% of
pertussis patients. 1
Although the bacteria do not invade deeper tissues, systemic changes do
occur. The disease is characterized by paroxysmal coughing in which 5-20
forceful hacking coughs occur in succession. Inspiration may be impossible
during paroxysms, which can be so prolonged that anoxia and cyanosis may
result. 2 Complications occur frequently in pertussis. Neurological involvement
observed in 1.5-14% of hospitalized cases is manifested by convulsions, coma,
paralysis, blindness, and psychic disturbances. 2 Mortality as high as 60% has been
reported among patients with neurological sequela. 2
It has been proposed that an exotoxin, pertussis toxin (PT), released by
Bordetella pertussis growing in the respiratory epithelium causes the harmful
effects of whooping cough and therefore is of prime importance in the patho-
genesis of the disease.3 Furthermore, it has been suggested that immunity to this
virulence factor in the form of specific antitoxin antibodies is important in
resistance to subsequent disease. 3
This hypo thesis suggests that pertussis is a toxemia much like diphtheria and
FREDERICK R. VOGEL • Vaccine Research and Deve\opment Branch, Division of AIDS,

Department of Health and Human Services, National Institutes of Health, Bethesda, Maryland
20892.
Pulmonary Infoctions and Immunity, edited by Herman Chme\ et al. Plenum Press, New York, 1994.
149
150 FREDERICK R. VOGEL
tetanus. However, more recently, virulence factors other than PT have been
implicated in pertussis pathology and the establishment of infection by BordeteUa
pertussis.
The bacterium Bordetella pertussis is a small, nonmotile, ovoid, gram-negative
bacillus that exhibits bipolar staining. The organism was discovered by Bordet
and Gengou in 1906. 4 They also cultured the organism on a blood agar medium
(Bordet-Gengou), a medium still employed for primary isolation of the bacte-
rium. Bordetella pertussis undergoes phase variation from avirulent smooth form
(Phase I) to an avirulent rough form (Phase IV). Phase IV organisms usually grow
weIl on minimal nutrient agar, whereas Phase I organisms do not. 5 ,6
2. VIRULENCE FACfORS AND PROTECfIVE ANTIGENS
Bordetella pertussis possesses several virulence factors that may contribute to

pathogenicity of the organism either by permitting it to colonize the respiratory
tract or through direct toxic effects. Immune responses to several pertussis
virulence factors have been shown to be individually protective in animal models.
Whole-cell pertussis vaccines and most acellular pertussis vaccines are composed
of antigens representing more than one of these virulence factors. This multi-
valent approach may improve the efficacy of acellular vaccines by maximizing the
percentage of individuals in the vaccinated population who can res pond to one or
more potentially protective antigens. Certain virulence factors regulated by a
bacterial virulence gene termed the bvg locus include pertussis toxin (PT),
filamentous hemagglutinin (FHA), adenylate cydase toxin (AC), and heat-Iabile
toxin (HLT).
2.1. Pertussis Toxin

Pertussis toxin is a hexameric protein exotoxin produced by Phase I Borde-
teUa pertussis organisms. 7 PT is responsible for many of the biologic activities of
pertussis organisms, induding lymphocytosis promotion,8 induction of hypo-
glycemia through increased insulin production,9 and sensitization of mice to the
lethai effects of histamine. 10 Physicochemical purification of pertussis toxin 11 has
revealed that PT is composed of five dissimilar subunits, S-1 through S-5, that
occur in a molar ratio of 1:1:1:2:1. The molecular weight of PT is approximately
117,000. Pertussis toxin is an A-B type toxin, with enzymatic activity associated
with the A protomer and cell-binding activity associated with the B protomer. The
A protomer catalyzes the transfer of the ADP-ribose moiety of NAD to the GTP-
binding protein Gi where "i" stands for inhibitory. Ribosylation of this G protein
derepresses adenylate cydase and results in an increase of cydic AMP (cAMP).l2
2.2. Filamentous Hemagglutinin

Virulent Bordetella pertussis organisms possess two dearly identifiable hemag-
glutinins: pertussis toxin and filamentous hemagglutinin. Both contribute to the
BORDETELLA PERTUSSIS 151
virulence of Bordetella pertussis at least in part through their action as adhesins.

Filamentous hemagglutinin as observed by electron microscopy appears as fine
filaments approximately 2 nm in diameter and 40 to 100 nm in length. 13 Antibody
to purified FHA has been demonstrated to protect mice from respiratory infec-
tion but not intracerebral infection with Bordetella pertussis.14 ,15 Purified FHA is
heterogeneous in gel electrophoresis, which shows bands of Mr ranging from
90,000 to 220,000.16 Filamentous hemagglutinin helps Bordetella pertussis at-
tach to human cilia, which is essential for the establishment and maintenance of
infection. 17
2.3. Adenylate Cyclase Toxin

Adenylate cyclase (AC) is a novel toxin found in the Bordetella species pertussis,
parapertussis, and bronchisepticum. Adenylate cyclase toxin (ACT) is a cell-
associated toxin found on the external side of the cytoplasmic membrane.1 8
Bordetella AC enters mammalian target cells, where it catalyses the production of
cAMP from ATP. ACT is activated by calmodulin, catalase, and creatine phospho-
kinase.I 9 The toxin has been shown to suppress neutrophil and monocyte activ-
ities, which may contribute to the persistence of pertussis infection by inhibiting
the clearance of the organism. Adenylate cyclase toxin also may contribute to
pertussis pathogenesis by causing accumulation of cAMP in the mucosa, leading
to increased fluid secretion. Increased permeability of the mucosa through
increased cAMP levels also may enhance the entry of PT.
2.4. Beat-Labile Toxin

Heat-Iabile toxin (HLT) is a proteinaceous material found in the protoplasm
of young cells and in the culture supernatant fluid of older cultures. The toxin is
inactivated by heating at 56°C for 30 min. Intact toxin isolated from disrupted
cells is dermonecrotic and is lethai to experimental animals when injected by the
intraperitoneal or intravenous route of injection. Heat-Iabile toxin has been
implicated in the pathogenesis ofwhooping cough,18,20 but no neutralizing anti-
HLT antibody is found in convalescent serum, indicating that anti-HLT antibody
is not protective against the disease. Antibody against HLT does not protect mice
from intracerebral challenge with virulent Bordetella pertussis organisms. 3
2.5. Tracheal Cytotoxin

Tracheal cytotoxin (TCT) is a low-molecular-weight toxin (921 Da) produced
by all Bordetella species. 21 TCT is a peptidoglycan-like toxin consisting of two
residues of alanine and one residue each of glutamic acid, diaminopimelic acid,
muramic acid, and glucosamine. 21 In animal organ cell culture, TCT can cause
ciliostasis and destruction of ciliated cells. 22 ,23 In humans, destruction of these
cells leads to an accumulation of mucus, bacteria, and products of inflammation
that the infected individual attempts to clear through coughing. 21 Tracheal
cytotoxin also may contribute to the pathogenesis of pertussis by enhancing the
delivery of other pertussis virulence factors, such as PT, through injury to the
ciliated epithelium of the trachea.
2.6. Pertactin
Pertactin is an outer-membrane protein (OMP) of BordeteUa pertussis with an
apparent molecular weight of 62-69 kDa. Pertactin is a nonfimbrial agglutinogen
that can enhance binding of bacteria to cells. 1 Active immunization with pertactin
is protective in a mouse aerosol challenge model for pertussis. 24 The protein is a
component of some acellular pertussis vaccines, and antibodies to pertactin are
found in children immunized with these vaccines. 24
2.7. Agglutinogens
BordeteUa pertussis strains produce three main agglutinogens (ACCs 1, 2, and
3). Combinations of these agglutinogens make up three serotypes that infect
humans 1,2,3; 1,2; 1,3). Agglutinogens 1 and 2 are fimbriae while agglutinogen 3
appears to be a nonfimbrial surface protein.l. 25 The agglutinogens may contrib-
ute to the establishment of pertussis infection through attachment to eukaryotic
cells but not through direct attachment to cilia.l.l 7,26
2.8. Pertussis Endotoxin

Endotoxin, or bacteriallipopolysaccharide (LPS), is a cell-wall component of
gram-negative bacteria. Endotoxins are quite stable and retain their biological
activities even after heating to 100°C for 1 hr. Endotoxin from BordeteUa pertussis,
although similar to enterobacteriaceae endotoxins in its activities, is grossly
different in structure. 27 ,28 Unlike enterobacteriaceae LPS, pertussis endotoxin
possesses two LPS groups (PSI and PS2) and two lipid groups (lipid A and lipid
X). Analysis of the various subcomponents of pertussis LPS reveals that several
of the biological activities attributed to lipid A in enterobacteriaceae LPS are
associated with lipid X in pertussis. These include pyrogenicity and the ability to
induce the Shwartzmann reaction. Pertussis lipid A is weakly pyrogenic but
retains adjuvant activity.28
3. CLINICAL MANIFESTATIONS OF PERTUSSIS
3.1. Transmission
Pertussis infection spreads from person to person through inhaled respira-
tory droplets or by hand-to-nose transmission from contaminated surfaces. 29
After an incubation period of seven to ten days, the catarrhal stage of pertussis
begins as an upper respiratory infection with coryza. A dry cough develops in one
week followed by the violent coughing characteristic of the paroxysmal stage of
the disease. Vomiting often follows the bouts of coughing. Coughing usually lasts
for Ion ger than 14 days.I
3.2. Complications
Damage to the CNS is secondary to hypoxia or to hemorrhage that results
from elevated venous press ure during the paroxysms. 1 Development of secondary
bacterial bronchopneumonia through inspiration of respiratory secretions dur-
ing the whoop is a common complication. 29
3.3. Antibiotic Therapy

Erythromycin, given at high doses for two weeks, is the drug of choice for
pertussis.3° Alternatively, cotrimoxazole or amoxycillin may be used.3° Erythro-
mycin may prevent or reduce the severity of the disease if the antibiotic is used
during the incubation stage of disease, as can be accomplished in household
contacts of apertussis case. 31 Erythromycin also can alter the catarrhal stage and
reduce the length of infectivity when given at any stage of the disease. 1 Bordetella
pertussis is resistant to most cephalosporins. 1
4. PERTUSSIS VACCINES
4.1. Whole-Cell Vaccines

Effective killed whole-cell pertussis vaccines were developed by Sauer32 and
Kendrick and Elderling33 in the 1930s. Widespread immunization with pertussis
vaccines began in the United States between 1940 and 1945. Mortality caused by
pertussis declined 82% between 1900 to 1939, prior to the general use of pertussis
vaccine. However, efficacy of pertussis vaccine can be inferred by statis~ics which
indicated that for the years 1940 to 1974, 87 to 90% fewer deaths from pertussis
were observed than would have been expected by extrapolation of the 1900 to
1939 declines. 34
Although killed whole-cell vaccines provide immunity to pertussis, these
vaccines have been associated with several untoward side effects. These include
fever, anorexia, and malaise. Rarely, more serious reactions, including hypotonic
shock-like episodes and convulsions 35 have been temporally related to immuniza-
tion with whole-cell pertussis vaccines. Publicity concerning pertussis vac-
cine-associated morbidity and mortality has led to fear and distrust of the vaccine
by a segment of the public. Indeed, several countries have for a time stopped
pertussis immunizations following reports of vaccine-associated deaths. Cessation
of pertussis immunization has consistently resulted in increased numbers of
outbreaks of pertussis following shortly after the immunization is stopped. In
Japan, for example, 152,000 cases of pertussis and 17,000 deaths were reported in
1947 before pertussis immunization was established. After nationwide vaccina-
tion was begun in 1950, reported cases of pertussis declined steadily, reaching a
nadir of 206 in 1971. Vaccine administration was halted in 1975 following the
report of two pertussis vaccine-associated deaths. Pertussis subsequendy became
more prevalent, with 31,700 cases and 113 deaths reported between 1975 and
1979.36 Nationwide immunization against pertussis has since been resumed in
Japan.
4.2. Acellular Pertussis Vaccines

The necessity of providing active immunization against pertussis and the
frequency of side effects associated with whole-cell pertussis vaccines has led to
the development of acellular pertussis vaccines. These vaccines are designed to
provide potent active immunity to pertussis with decreased frequencies of ad-
verse effects including fever, pain, and malaise.
Active immunization against pertussis on aglobai scale requires the vaccine
to induce protection against pertussis disease in genetically diverse populations.
Differences in the ability of individuals to res pond to pertussis antigens (for
example, PT) indicate that greater vaccine efficacy would be obtained with
acellular vaccines containing more than one antigen capable of providing some
protection against pertussis. Therefore, while most acellular vaccines contain
detoxified PT and usually FHA, some acellular pertussis vaccines also contain
pertactin and one or more of the agglutinogens. In fact whole-cell pertussis
vaccines may succeed because many pertussis antigens are made available to the
vaccinee through these vaccines, providing broad-based protection at the popula-
tion level. Acellular vaccines have been employed in Japan since 1981 and have
recendy been recommended in the United States for the fourth and fifth diph-
theria, pertussis, tetanus (DPT) immunizations. 37 Recent evidence comparing the
immunogenicity and reactogenicity of an acellular pertussis vaccine given at 2
months of age to the same vaccine given at 3 months suggests that the current
immunization schedule for whole-cell pertussis vaccine would be acceptable for
the acellular products.3 8
5. PERTUSSIS AND AIDS
Individuals with acquired immunodeficiency syndrome are susceptible to a

variety of respiratory infections. These include tuberculosis, Pneumocystis carinii,
and cytomegalovirus and bacterial pneumonia. Bacterial pneumonia may be
caused by Streptococcus pneumoniae, or gram-negative organisms such as Haemo-
philus inJluenzae or Legionella pneumophila. 39 Bordetella pertussis also has been
isolated in cases of gram-negative pneumonia in adults and children with
AIDS.40,41 Waning immunity to pertussis vaccine and B- and T-cell deficiencies
may increase susceptibility to pertussis and increase the risk of pulmonary
infiltration by Bordetella pertussis in people with AIDS.40
Pertussis should be considered in the differential dia gnosis of Legionella
pneumophila because ßuorescent antibody cross-reactivity between Bordetella per-

tussis and Legionella pneumophila can occur. 40,42
A prospective study of 60 HIV-infected adults with respiratory symptoms
(72% cough [33% > 14 days], 40% fever) was conducted to estimate the rate of
pertussis infection in HIV patients at a tertiary medical center. Bordetella pertussis
was not isolated from any of these patients. These investigators concluded that
HIV-infected individuals are not a significant reservoir for Bordetella pertussis
in the community.43 However, the majority of the individuals in this study were
undergoing treatment far HIV and opportunistic infections with antibiotics,
including erythromycin, to which the pertussis organism is susceptible. The
effects of these drugs on the isolation of Bordetella pertussis from AIDS patients
was not evaluated in these studies.
REFERENCES
1. Woods, M. L., 1991, Pertussis: An old disease in a new era, Semin. Resp. Infeet. 6:32-43.
2. Munoz, J. and Bergman, R. K., 1977, Bordetella pertussis-Immunologie and Other Biologieal
Activities, Marcel Dekker, Inc., New York, pp. 6-8.
3. Pittman, M., 1979, Pertussis toxin: Cause of the harmful effects and prolonged immunity of
whooping cough. A hypothesis, Rev. Infoct. Dis. 1:401-412.
4. Bordet, J. and Gengou, 0., 1909, L:endotoxine coquelucheuse, Ann. Inst. Pasteur. 23:
415-419.
5. Leslie, P. H. and Gardner, A. D., 1931, The phases of Haemophilus pertussis,] Hyg. 31:
423-434.
6. Weiss, A. A. and Falkow, S., 1984, Genetic analysis of phase change in Bordetella pertussis,
Infoct. Immun. 43:263-269.
7. Vi, M., 1988, Multiple biological activities of pertussis toxin, in: Pathogenesis and Immunity in
Pertussis, (A. C. Wardiawand R. Parton, eds.), John Wiley and Sons, Chichester, VK, pp.
122-123.
8. Morse, S.l. and Morse,J. H.,1976, Isolation and properties ofthe Iymphocytosis-promoting
factor of Bordetella pertussis,] Exp. Med. 143:1483-1502.
9. Yajima, M., Hosada, K., Kanbayashi, Y., Nakamura, T., Nogimori, K., Mizushima, Y.,
Nakase, Y., and Vi, M., 1978, Islet-activating protein (IAP) in Bordetella pertussis that
potentiates insulin secretory responses in rats,] Bioehem. 83:295-303.
10. Lehrer, S. B., Vaughn,J. H., and Tan, E. M.,1975, Immunologie and biochemical properties
of histamine-sensitizing factor of Bordetella pertussis,] Immunol. 114:34-39.
11. Tamura, M., Nogimori, K., Murai, S., Yajima, M., Ito, K., Katada, T., Vi, M., and Ishii, 1982,
Subunit structure of islet-activating protein, pertussis toxin, in conformity with the A-B
model, Bioehemistry 21:5516-5522.
12. Wilson, A. D., Robinson, A., Irons, L., and Stokes, C. R., 1993, Adjuvant action of cholera
toxin and pertussis toxin in the induction of IgA antibody response to orally administered
antigen, Vaccine 11:113-118.
13. Blom,J., Hansen, G. A., and Pulsen, F. M., 1983, Morphology of cells and hemagglutinins of
Bordetella species: Resolution of substructural units in fimbriae of Bordetella pertussis, Infoet.
Immun. 42:308-317.
14. Shahin, R. D., Amsbaugh, D. F., and Leef, M. F., 1992, Mucosal immunization with
filamentous hemagglutinin protects against Bordetella pertussis infection, Infoct. Immun.
60:1482-1488.
156 FREDERICK R VOGEL
15. Sato, Y., Cowell,]. L., Sato, H., Burstyn, D. G., and Manclark, C. R, 1983, Separation and
purification of the hemagglutinins from Bordetella pertussis, Inftct. Immun. 41:313-320.
16. Irons, L. 1., Ashworth, L. A. E., and Wilton-Smith, P., 1983, Heterogeneity of the filamentous
hemagglutinin of Bordetella pertussis studied with monoclonal antibodies,] Gen. Microbiol.
13:1-8.
17. Tuomanen, E., 1988, Bordetella pertussis adhesins, in: Pathogenesis and Immunity in Pertussis,
(A. C. Wardiawand R. Parton, eds.), John Wiley and Sons, Chichester, UK, pp. 75-94.
18. Cowell,]. L., Hewlett, E. L., and Manciark, C. R, 1979, Intracellular localization of the
dermonecrotic toxin of Bordetella pertussis, Infect. Immun. 25:896-901.
19. Hewlett, E. L., and Gordon, V. M., 1988, Adenylate cyclase toxin of Bordetella pertussis, in:
Pathogenesis and Immunity in Pertussis, (A. C. Wardiawand R. Parton, eds.), John Wiley and
Sons, Chichester, UK, pp. 193-209.
20. Munoz,]., 1971, Protein toxins from Bordetella pertussis, in: Microbial Toxins, Volume 2 (T. C.
Montie and S.]. Ajl, eds.), Academic Press, New York, pp. 271-300.
21. Goldman, W E., 1988, Tracheal cytotoxin of Bordetella pertussis, in: Pathogenesis and Immunity
in Pertussis, (A. C. Wardiawand R. Parton, eds.), John Wiley and Sons, Chichester, UK, pp.
231-243.
22. Opremcak, L. B., and Rheins, M., 1983, Scanning electron microscopy of mouse ciliated
oviduct and tracheal epithelium infected in vitro with Bordetella pertussis, Can.] Microbiol.
29:415-420.
23. Collier, A. M., 1977, Pathogenesis of infection with Bordetella pertussis in hamster tracheal
organ culture,] Inftct. Dis. 136:S196-S203.
24. Shahin, R. D., Brennan, M. ]., Li, Z. M., Meade, B. D., and Manciark, C. R., 1990,
Characterization of the protective capacity and immunogenicity of the 69-kD outer mem-
brane protein of Bordetella pertussis,] Exp. Med. 171:36-73.
25. Preston, N. W, Zorgani, A. A., and Carter, E. ]., 1990, Location of the three major
agglutinogens of Bordetella pertussis by immunoelectronmicroscopy,] Med. Microbiol. 32:
63-68.
26. Tuomanen, E., and Weiss, A., 1985, Characterization of two adhesins of Bordetella pertussis
for human ciliated respiratory-epithelial cells,] Inftct. Dis. 152:118-125.
27. Le Dur, A., Caroff, M., Chaby, R., and Szabo, L., 1978, A novel type of endotoxin structure
present in Bordetella pertussis. Isolation of two different polysaccharides bound to lipid A,
Eur. ] Biochem. 84:579-589.
28. Ayme, G., Caroff, M., Chaby, R., Haeffner-CavaiIIon, N., Le Dur, A., Moreau, M., Muset, M.,
Mynard, M., Roumiantzeff, M., SchuItz, D., and Szabo, L., 1980, Biological activities of
fragments derived from Bordetella pertussis endotoxin: Isolation of a nontoxic, Shwartzmann-
negative lipid A possessing high adjuvant properties, Inftct. Immun. 27:739-745.
29. Walker, E., 1988, Clinical aspects of pertussis, in: Pathogenesis and Immunity in Pertussis, (A. C.
Wardiawand R Parton, eds.), John Wiley and Sons, Chichester, UK, pp. 273-282.
30. Hoppe, j. E., 1990, Pertussis: Diagnosis, cIinical aspects and therapy, Monatssehr Kinderheikd
138:244-248.
31. Sprauer, M. A., Cochi, S. L., ZeU, E. R., Sutter, R. W, MuUen, j. R., Euglender, S. j., and
Patriarca, P. A., 1992, Prevention of secondary transmission of pertussis in households with
early use of erythromycin, Am.] Dis. Child 146:177-181.
32. Sauer, L., 1933, Whooping cough: A study in immunizations, ] Am. Med. Assoe. 100:
239-241.
33. Kendrick, P. and Eldering, G., 1939, A study in active immunization against pertussis, Am.
] Hyg. 29:133-153.
34. Mortimer, E. A.,Jr., andJones, P. K., 1979, Pertussis vaccines in the United States: the benefit-
risk ratio, in: International Symposium on Pertussis, (C. Manciark and]. HilI, eds.), DHEW
Publication No. (NIH) 97-1830, pp. 250-255.
35. Cody, C. L., Baraff, L. j., Cherry, j. D., Marcy, S. M., and Manclark, C. R., 1981, Nature and
rates of adverse reactions associated with DPT and DT immunizations in infants and
children, Pediatrics 68:650-660.
36. Fulginiti, V. A., 1983, Pertussis vaccine: Hope for the future?,] Infect. Dis. 148:146-147.
37. CDC, 1992, Pertussis vaccination: Acellular pertussis vaccine for the fourth and fifth doses
of the DTP series update to supplementary ACIP statement. Recommendations of the
advisory committee on immunization practices (ACIP), MMWR 41(No. RR-15):1-15.
38. Kamiya, H., Ritsue, N., Matsuda, T., Yasuda, N., Christenson, P. D., and Cherry,j. D., 1992,
Immunogenicity and reactogenicity of Takeda acellular pertussis-component diphtheria-
tetanus-pertussis vaccine in 2- and 3-month-old children in Japan, Am. ] Dis. Child.
146:1141-1147.
39. Stover, D. E., White, D. A., Romano, P. A., Gellene, R. A., and Robeson, W. A., 1985,
Spectrum of pulmonary diseases associated with the acquired immunodeficiency syndrome,
Am.] Med. 78:429-437.
40. Adamson, P. C., Wu, T. C., Meade, B. D., Rubin, M. D., Manclark, C. R., and Pizza, P. A.,
1989, Pertussis in a previous\y immunized child with human immunodeficiency virus
infection,] Pediatr. 115:589-592.
41. Ng, V. L., York, M., and Hadley, W. K., 1989, Unexpected isolation of Bordetella pertussis from
patients with acquired immunodeficiency syndrome,] Clin. Microbiol. 27:337-338.
42. Benson, R. F., Thacker, W. 1., Plikaytis, B. B., and Wilkinson, W. H., 1987, Cross reactions in
Legionella antisera with Bordetella pertussis strains,] Clin. Microbiol. 25:594-596.
43. Cohn, S. E., Knorr, K. L., Gilligan, P. H., Smiley, M. L., and Weber, D.j., 1993, Pertussis is rare
in human immunodeficiency virus disease, Am. Rev. Respir. Dis. 147:411-413.
10
Coxiella burnetii and Q Fever

ANN W FUNKHOUSER and LOUIS F. FRIES
1. INTRODucnON
"Q" or Query Fever was first described in 1935 by E. H. Derrick, who investigated
an outbreak of afebrile illness in abattoir workers in Brisbane, Australia. Subse-
quently, Burnet and Freeman used sam pies of liver from guinea pigs inoculated
with the blood or urine of these infected humans to infect mice. The spleens of
the infected mice were found to harbor minute pleomorphic organisms that
Burnet classified as rickettsia. Several years later an organism pathogenic for
guinea pigs was isolated from Dermacentor andersoni ticks recovered near Nine
Mile Creek in Montana. It stained deeply with the Macchiavello and Giemsa
techniques, was pleomorphic in shape, grew in intracellular vacuoles, and infected
a laboratory worker who had symptoms very similar to those seen in the initial
Australian outbreak. Cox concluded that this agent was rickettsia-like, and Dyer
proposed that the organisms isolated by the two groups were identical. The
organism was later named Coxiella burnetii. 1
C. burnetii is currently classified in the family Rickettsiaceae (though there is
debate regarding the appropriateness of this classification), in its own genus:
Coxiella. It differs from other rickettsiae on the basis of genomic characteristics,
site of intracellular proliferation, antibiotic sensitivities, and greater resistance to
high temperature and low pH. The guanine-plus-cytosine content of C. burnetii
DNA is 42%, compared with that of other rickettsiae, whose guanine-plus-
cytosine contents are approximately 30%. C. burnetii proliferates inside phago-
lysosomes of mononuclear cells and in intracytoplasmic vacuoles in other cells,
ANN W. FUNKHOUSER • Laboratory of Infeetious Disease, National Institute of Allergy and

Infeetious Disease, Bethesda, Maryland 20892. LOUIS F. FRIES· Univax Biologics, Ine.,
Roekville, Maryland 20852.
Pulmonary lnfections and lmmunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
159
160 ANN W FUNKHOUSER and LOUIS F. FRIES
whereas other rickettsiae proliferate within the cytoplasm. Figure 1 shows vacuo-
lar proliferation of C. burnetii pushing organelles to the edge of the host cello
Whereas other rickettsiae actively promote cell entry, C. burnetii is passively
phagocytosed.l C. burnetii also differs from other rickettsiae in that it is usually
insensitive to erythromyein, ehloramphenieol, and thiomycin (see Section 7).
Further, whereas other riekettsiae generally survive only briefly outside of a host
and are sensitive to pH changes, C. burnetii resists desiceation, heat, and sunlight
and replieates at pH 4.5.
C. burnetii is a gram-negative, nonmotile, pleomorphie, oval to rod-like
organism, 0.4 to 1.0 J.Lm long and 0.2 to 0.4 J.Lm wide. The organism is not mor-
phologieally uniform. The small eell variant (SCV) is compact, with an electron-
dense nudeoid. Large eell variants (LCV s) are larger, more pleomorphie, and less
eleetron-dense. Some LCVs contain eleetron-dense bodies in the periplasmic
space, termed endospore-like [orms. In one proposed life eyde, the SCv, LCV,
FIGURE 1. C. burnetii is shown proliferating in a cytoplasmic vacuole in an L-929 cell, pushing

the nuclei to the edge of the cell. (From Baca and Paretsky,l with permission.)
COXIELLA BURNETII AND Q FEVER 161
and spores enter the phagolysosome, wherein acid pR triggers the spores to
evolve into SCVs, and then to LCVs which are in turn triggered by an unknown
mechanism to create spores. SCVs, LCVs, and spores are released when the host
cell is lysed. 2
C. burnetii also demonstrates phase variation, which is indistinguishable
morphologically but detected by immunological and cytochemical methods. The
organism is found in phase I in animals and humans. After propagation in tissue
culture it converts to phase II, and by repassage in animals reverts to phase I (see
Section 8). Surface anionic binding sites are present on phase II and not on phase
I organisms, and carbohydrate composition differs between phase I and phase II
envelopes. 3 Endotoxic lipopolysaccharide (LPS) is present in both phases, but in
phase II is present in one-tenth the amount found in phase 1. It appears that
phase I LPS may be regarded as analogous to the smooth LPS of pathogenic
enteric bacteria, whereas phase II LPS is a truncated molecule analogous to the
LPS of rough strains. Despite this difference, membrane protein antigens ap-
pear to be common to both phases. 4 Phase I LPS induces pathophysiologic
changes in animals similar to those associated with endotoxins of conventional
gram-negative bacteria including hepatic enlargement in guinea pigs, hyperther-
mia, leukocytosis, increased levels of hepatic and serum cortisol, and hypother-
mia in rats. l
Virtually all C. burnetii strains thus far examined, regardless of origin,
contain a partially conserved 36- to 45-kb DNA sequence as a plasmid or, less
commonly, as an integral part of their genome. 5 The presence of the plasmid is
unrelated to phase. Acute disease-associated strains bears a 36-kb form of the
plasmid termed QpHl, whereas strains associated with chronic disease have a
slightly larger 39-kb plasmid termed QpRS or analogous sequences integrated
into the chromosomal DNA.6,6a The products encoded by these plasmids are
under active study, and may include both cytoplasmic and outer-membrane
proteins. Restriction endonuclease digestion analysis of chromosomal DNA segre-
gates C. burnetii isolates into six groups.6b Three groups are associated with QpRI
plasmids and aC\lte disease; two groups bear QpRS sequences and are associated
with chronic illness. A final small group has a unique plasmid type and thus far
has not been associated with clinical disease. It is unclear at this point whether
these findings reßect specific pathogenic functions of plasmid or chromosome-
encoded protein variants, or a process of variant selection in chronic infection.
This is an area of active investigation. It has also been suggested on the basis of
immunoblot studies that antigenic variations in phase I LPS may be predictive of
the chronic disease potential of a given strain. 7
2. EPIDEMIOLOGY AND TRANSMISSION
Q fever has been reported from almost every region of the world. 8 The main
reservoirs for C. burnetii are dairy cows, sheep, and goats, although the organism
has been isolated from other domesticated and wild mammals, marsupials,
rodents, birds, fish, and multiple types of arthropod. 9 The role that each of these
animals play in the overall maintenance and transmission of C. burnetii is unclear.
Animals rarely become ill as a result of C. burnetii infection, but the infection is
associated with abortions and metritis with subsequent infertility in cattle.1 0
2.1. Animal Infection

Animal-to-tick-to-animal transmission has been documented, but the rela-
tive importance of tick vector versus particle inhalation versus sexual transmis-
sion in the spread of animal infection is not known. It has been noted by several
investigators that there has been a dramatic rise in C. burnetii seroprevalence in
animals over the last 10 to 20 years. 1,1l,12 Recently, the suggestion was made that
this dramatic rise may be a result of sexual transmission. The hypothesis was
tested by intraperitoneally infecting a group of male mice, then caging them with
either healthy male or female mice. None of the male mouse contacts became
infected, whereas all the female mice became infected. ll
The seroprevalence of C. burnetii in cattle varies from 1% in Uruguay13
to 44.6% in Tibet14 to 82% in one herd in California. 8 Sheep and goats also have a
variable rate of carriage of C. burnetii from 1% combined seropositivity of sheep
and goats in Tibet to 72% combined positivity of sheep and goats in Bachu,
China. 14 Goats alone vary in seropositivity to C. burnetii from 57% in California 15
to 76.8% outside Madrid. 16 Sheep, goats, and cattle shed organisms at high
concentrations in their milk and placentae. As manras 51 % of cattle in California
have been found to shed C. burnetii organisms in their milk. 1 Interestingly, the
seroprevalence of C. burnetii antibody in ewes has been found to be much higher
during lambing season than just before breeding season (55% vs. 18%).15
Bandicoots, camels, kangaroos, horses, yaks, pigs, water buffalo, desert rats,
wild rabbits, dogs, cats and mice are some of the other animals from which C.
burnetii has been isolated. 9,14 Pigeons, sparrows and other birds closely associated
with humans and domestic animals have also been found to be seropositive for C.
burnetii. The organism is believed to spread to them via ingestion or inhalation of
infected particles. Ticks are highly susceptible to C. burnetii, which naturally
infects over 40 species of ticks found on five different continents. 9 Dermacentor
marginatus tIcks seem to have the highest vector potential in at least one study.
However, C. burnetii has also been isolated from fleas, lice, and cockroaches, and
their roles as animal reservoirs are unknown. 17
2.2. Human Infection

The principal mode of transmission to humans is inhalation of infected
aerosolized particles. There are isolated case reports of adult human-to-human
transmission with questionable documentation. 18 Tick bites are also a source of
transmission to humans, though the importance ofthis source is uncertain. 2,10,19
Breast feeding has been reported as a means of transmission in endemie areas 20
and in at least one sporadic case. 21 In Hissar, India 2.6% of breast-milk sampies
were found to be positive for C. burnetii antigen, with up to 18% of breast-milk

sampies positive for C. burnetii antibody.22 At least one case of documented
perinatal transmission of Q fever has been reported with premature delivery at
29 weeks of gestation, and subsequent death of the neonate at 6 days of age. 23
Acute Q fever has been associated with abortion both early and late in gestation.
Prior to pasteurization of milk, ingestion of infected milk was a source for largely
asymptomatic infection. 15
Random serologic screens of healthy adults have shown a CoxieHa antibody
prevalence ranging from less than 1% to 5% in Canada, France, Switzerland, and
the United Kingdom. 12.24-27 In the Netherlands and West Africa, seroprevalence
was found to be as high as 40% to 60% in the healthy adult population. 12.28 In
some studies rural residents have had higher seroprevalence,24 whereas in other
studies rates are higher in urban residents. 29
Occupational exposure to the blood and/or body fluids and products oflarge
domestic animals has been weH documented as a risk factor in veterinarians,
residents of dairy farms, taxidermists, amateur wool spinners,12 slaughterhouse
workers,3()"'32 and animal laboratory researchers.33-39 Although males are re-
ported in some studies as having higher seroprevalence than females,40 this
finding has not been consistent and the observation probably reflects the greater
proportion of males in high-risk occupations. Seroprevalence in healthy humans
has been found to be as high as 83.7% in veterinarians, 68% in dairy farmers,12
and 55% in meat industry workers. 27 The incidence of C. burnetii infections in
humans has been found to be higher during lambing season in some studies. 29
Other less common sources of Q fever have included parturient cats,41-44 a
school goat that infected school children,45 a mysterious source in postal workers
whose place of work was near an old cattle market,46 the clothing of someone who
had had direct contact with an infected animal,I8 and a necropsy on a fatal case of
Q fever. 47 At least one case has been reported of suspected adult person-to-person
transmission with the index case being an individual infected during lambing
season. This man is said to have infected two other relatives whom he visited after
removing his work clothes. This report may represent person-to-person transmis-
sion or may reflect mutual exposure to an unknown variable. 18
The majority ofhumans infected with C. burnetii probably are either asymp-
tomatic or have mild flu-like symptoms. Among individuals who come to medical
attention for Q fever, there is a wide range of acuity of disease and an equaHy wide
range of clinical presentations. The classical terminology in grouping these
syndromes is to label those Q fever syndromes that are self-limited in clinical
course and accompanied by more phase II C. burnetii antibody production acute,
and the Q fever syndromes that have a protracted course either with or without
appropriate therapy and are accompanied by more phase I C. burnetii antibody
production chronic (see Section 6).
164 ANN W. FUNKHOUSER and LOUIS F. FRIES
4. ACUTE Q FEVER
The ineubatiqn period for Q fever ranges from 14 to 39 days, with the
average being 20 days. The patient may present with pneumonia, hepatitis, and/
or nonloealizing symptoms such as fever and chilis, severe headaehe (often with a
retroorbitalloealization), general malaise, and myalgia. Severe neurologie mani-
festations ofthe disease may oeeur, and ean be highly variable (see Seetion 4.3). In
all series, 90-100% of diagnosed patients experienee fever. Headaehe is also
almost universal, and is reported in 65-90% of patients in different series.
Radiographie evidenee of pneumonia is reported in 4-97% of eases of acute Q
fever, and hepatosplenomegaly is present in anywhere from 4% to 65% of
patients. 48
4.1. Pneumonia
Arecent study deseribing 164 eases of radiologieally and serologieally proven
Q fever pneumonia from the Basque eountry reviews its clinieal symptoms and
signs. 49 Fifty-three percent of eases were reported to have eough. Produetion of
purulent sputum and hemoptysis were not eommonly found, being noted in only
9% and 10.5% of patients respeetively. One-third ofthe group reported pleuritie
pain. Fever of 39°C or greater was observed in 21.5% patients, with the mean
duration being 8.3 days. Findings on physieal examination were often modest,
although signs of frank eonsolidation were noted in severe eases. No deaths
oeeurred in the series. Another reeent series eolleeted in England generated
similar findings. 50
On laboratory evaluation, the vast majority of patients with pneumonia were
found to have white blood eell counts between 5 and 10,000 per mm 3 • Elevated
erythroeyte sedimentation rate was found in approximately half of the patients.
Deereased arterial O 2 eoneentration was seen in half of the patients, with severe
respiratory insufficieney in approximately 10%.
A number of different radiographie pietures are seen with Q fever pneu-
monia. In the above-mentioned series of 164 pneumonia patients, 64.5% had
loealized patehy alveolar involvement (see Fig. 2), including part, but not all, of a
lobe, with a slight predominanee of lower-Iobe involvement. Air bronehograms
were seen in 40% of patients. Pleural effusion was seen in only 12% of patients,
and progression of the radiologie pieture from admission X-ray was rare. 49 In
other smaller series, no zonal predileetion was found, and the shape of loeal
eonsolidation tended to be rounded.5 1- 54 One study noted a differenee in the
radiologie presentations of sporadie versus epidemie eases, with sporadie eases
radiographically similar to other community-aequired pneumonias and epidemie
eases presenting more atypieally, with diffuse retieular markings and many
nodular lesions. 52 Cavitation within nodules has been reported,54 as has the
unusual presentation of adult respiratory distress syndrome with serologie doeu-
mentation of Q fever in an immunoeompromised patient. 55
who presented with pneumonia had elevated serum transaminases, the patients
with hepatitis alone were found to have signifieandy higher transaminase levels
than those with a primary pulmonary presentation. 56 In general, Q fever hepatitis
is a disease oflimited morbidity and duration; however, there have been reports of
fatalities 57 and severe disease presenting with an aeute abdomen and severe
jaundiee.58 Guinea pigs experimentally infeeted with the Nine Mile strain of C.
burnetii showed progressive depletion of hepatic glycogen, a deerease in glycogen
synthetase, an inerease in glycogen phosphorylase, and hyperglyeemia during
the aeute phase of infeetion. 59 These findings may or may not have a clinieal
eorrelate in humans.
4.3. Neurologie Manifestations

There are a substantial number of ease reports of eneephalitis, or meningo-
eneephalitis,60-64 and several ease reports of loeal neurologie deficits associated
with Q fever. 65 ,66 A reeent review of 103 eases of aeute Q fever found a surprising
22% incidenee of neurologie involvement, including confusional states, men-
ingeal signs, paresthesias, foeal motor and/or sensory defieits, and frank eneepha-
litis. 5o In a review of 18 eases of Q fever in infaney, an equal number of patients
presented with neurologie and with pulmonary manifestations, although most of
the ehildren in the study had nonspeeifie febrile illnesses. 21 White blood eell
eounts in the CSF have been doeumented as high as 667 eells/mm3 , and mono-
nuclear eells and lymphoeytes have been predominant in studies that report
results of lumbar puneture. Frequendy, however, the CSF is normal, or demon-
strates only slight inereases in protein. 50 In two of the reports of self-limited
eneephalitis, phase I-specifie antibody (see Seetion 8) was deteeted, either in
addition to or instead of phase II -specifie antibody to C. burnetii, raising questions
about the classifications of these cases as acute or ehronie Q fever (see Seetion
6).6°,61 Isolated optie neuritis was associated with C. burnetii infeetion in one ease
in whieh a rise in IgM titer oeeurred simultaneously with loss of vision in one eye
and development of bilateral thickened optie nerves on CT sean several weeks
after an episode of fever and malaise.66 Self-limited lateral reetus muscle paralysis
and papilledema in both eyes associated with an inerease in IgM and IgG anti-
bodies to C. burnetii in the serum have also been reported. 64
4.4. Other Uneommon Clinieal Correlates

Other uneommon manifestations of Q fever have been deseribed in anee-
dotal ease reports. Erythema nodosum was reported two weeks after the onset of
Q fever pneumonia in one ease, with resolution oeeurring within two weeks of
onset. 67 Coombs-negative hemolytie anemia and transient severe bone marrow
hypoplasia assoeiated with anemia and thromboeytopenia have both been de-
seribed.68 ,69 Several eases of doeumented C. burnetii septie arthritis of the hip
have also been reported. 70
COXIELLA BURNETll AND Q FEVER 165
FIGURE 2. This is a ehest radiograph of a 40-year old male with acute Q tever showing patehy,
inhomogeneous, alveolar densities in a diffuse distribution. (From Gordon et al. ,52 with per-
mission.)
4.2. Hepatitis
In studies in which liver function tests have been performed in patients with
acute Q fever, approximateiy 10% of patients have elevated bilirubins, one third
of patients have elevated alkaline phosphatase, and up to 70% have elevated
aspartate aminotransferase. 48 In one series, as many as 42.1 % of Q fever patients
presented with hepatitis alone. Although approximately one-third of patients
4.5. Pathology
There are a number of reports of Q fever hepatie histology. In one series of
18 eases, 16 specimens were found to have granulomas. These granulomas were
nonspecifie, with histioeytes, neutrophils, and mononuclear eells, sometimes
surrounding a eentral area of fat neerosis. Some granulomas had a fibrin ring
around the lipid eore (as in Fig. 3).71 Sueh so-ealled "doughnut granulomas" are
not pathognomonie of Q fever, but are highly suggestive in the appropriate
clinieal setting. When serial biopsies have been taken in patients who eontinue
to be symptomatie, granulomas have been found to be persistent, with only slight
histologieal ehange in some instanees,71,72 and replaeed by fibrosis in others. 73
Bone marrow biopsies show a similar histologie pieture; granulomas with
or without eentral fat neerosis, surrounded by fibrin and/or inflammatory
eells. 71 ,74--76 Bone marrow histioeytie proliferation and hemophagoeytosis have
been noted in one ease, a finding previously reported with other infeetions sueh as
tuberculosis, brueellosis, and typhoid fever. 77 Histologie examination of mouse
lung after C. burnetii ehallenge shows invasion of type I and type 11 pneumoeytes
and pulmonary fibroblasts, with foeal intraalveolar infiltration with maerophages.
Uninfeeted pneumoeytes and vaseular endothelial eells were also damaged. 78
Histologie studies of the pulmonary lesion of Q fever in man are limited. In very
severe eases, neerotizing pneumonia and bronehitis have been seen, with alveolar
and interstitial infiltration of maerophages and lymphoid eells. 79
5. CHRONIC Q FEVER
Infeetions with C. burnetii that reeur or that linger over months to years have
eome to be called chronic Q fever. This entity has most often been seen in the
clinieal context of endoearditis, but has also been noted in patients with persistent
hepatitis, or nonfoeal infeetion. The norms for serologie doeumentation of
chronie Q fever were developed in the evaluation of endoearditis. It was noted in
that eontext that Q fever endoearditis was associated with markedly elevated
antibody titers specifie for phase I, exeeeding titers against phase 11 C. burnetii.
As a result, these serologie patterns have become part of the definition of ehronie
disease, though the eorrelation between aetual ehronicity of symptoms and
presenee and magnitude of phase 1- or phase I1-specifie antibody titers is not
perfeet (see Seetion 6). In this seetion, we diseuss endoearditis as the prototype for
ehronie Q fever, and follow that diseussion with aeeounts ofless frequent presenta-
tions of ehronie Q fever.
5.1. Endocarditis
C. burnetii is a classie eausative agent of the syndrome of eulture-negative
endoearditis. In one epidemiologie study,80 14.5% of all eases of Q fever repre-
FIGURE 3. This photomicrograph of a liver biopsy specimen from a patient with acute Q fever
shows a lipid granuloma with a fibrinoid ring (arrow) within an inflammatory foeus. The
granuloma is surrounded by normalliver tissue. (From Srigley et al .•7l with permission.)
sented chronic Q fever, and 75% of chronic cases were cases of endocarditis. In
this study, composed entirely of adults, only 33% of patients with endocarditis
were reported to have underlying valvular heart disease. In other series, how-
ever, up to 100% of patients had underlying valvular heart disease, either
congenital or acquired, with or without a history of surgical intervention. 81 At
least one case of Q fever endocarditis in a child has been reported in a patient (7
years of age) clearly witlwut underlying valvular heart disease. 82 In a meta-
analysis of reports of Q fever endocarditis,81 the majority of patients presented
with fever, cardiac failure, and hepatosplenomegaly. Most patients had increased
erythrocyte sedimentation rates, thrombocytopenia and anemia, and elevated
transaminases. The mortality rate, reported by some to be as high as 50%, was
37% in the meta-analysis, and the me an time from symptom onset to diagnosis
was found to be 12 months. Additional reported complications of Q fever endocar-
ditis include arterial embolic phenomena, aortic valve ring abscess formation,83
persistent hypersplenism, 84 leukocytoclastic vasculitis, and glomerular nephrop-
athy with variable outcomes,85
5.2. Other Chronic Q Fever Syndromes

As mentioned above, most cases of Q fever endocarditis have evidence of
hepatic involvement to one degree or another. Occasionally chronic Q fever has
been associated with hepatitis alone,86 with sequential liver biopsies showing
persistent fibrin ring granulomas as described above. Other manifestations of
chronic Q fever are uncommon and constitute anecdotes. One case of chronic Q
fever presented with thrombocytopenia in the third trimester of pregnancy. The
symptoms resolved at delivery and the placenta was found to be infected with C.
burnetii, though the fetus was uninfected. 87 Another unusual case was reported as
Sudden Infant Death Syndrome in a nine-month-old infant who was found on
postmortem examination to have C. burnetii infection and both a phase land
phase II-specific antibody response. 23 There have been instances of chronic Q
fever, as defined by a rise in phase I-specific antibody titer, that have been
asymptomatic and were picked up on routine screening evaluations,88 or that
have had prolonged febrile illness without an obvious source,89 One such case
featured inappropriate secretion of antidiuretic hormone,89
6. DIAGNOSIS
Isolation of C. burnetii is both difficult and hazardous as a means of routine

diagnosis of Q fever, although the organism can be recovered in tissue culture
from specimens of chronically infected cardiac valve tissue. 90 Polymerase chain-
reaction technology may be of utility in the future, but a standardized assay is
not yet available as of this writing. Serologic testing is normally used for diagnosis,
but little standardization has been achieved. There is lack of uniformity in the
assay techniques themselves, and also in the antigens used to set up the assays (i.e.,
some use phase I, some phase 11, and some a mixture of both). Diffieulty in data
standardization also arises from the faet that different studies measure different
antibody classes (some IgM, some IgG or IgA) and use patient populations that
may not be homogeneous or weIl defined and are essentially never eulture-proven.
Although complement fixation (CF) and indireet fluoreseent antibody (IFA)
assays are the most widely used serologie methods, there are a number of reports
of other assays that are either as sensitive or more sensitive than CF and IFA, and
that are comparable in specificity. Some of these include mieroagglutination
(MA) and enzyme-linked immunosorbent assay (ELISA) or other immuno-
enzymatie methods.
The sensitivity of serologie assays is affeeted by the antigen used, by the
proeessing of the antigen, and by the variability in assay sensitivity to partieular
antibody classes. 91 ,92 One study measured IgG to C. burnetii phase 11 antigen in
sera of 213 patients who had been ill du ring a Q fever outbreak in Switzerland
(length and type of symptoms not specified). Whereas 94.8% of sera were posi-
tive by ELISA; 90.6% were positive by IFA, and 77.8% were positive by CF. The
speeifieities of the three assays were comparable, and were demonstrated by
negative results in the sera of 36 individuals who had non-Q fever pneumonia. 93
In another study, ELISA and IFA were again demonstrated to be more sensitive
than CF; IgM antibodies to C. burnetii phase 11 antigen were deteeted for 17 weeks
from onset of illness using ELISA and IFA, and 10 weeks after onset of illness by
CF assay.94 The timing of peak antibody titers was analyzed for a group of
patients with classie acute Q fever symptomatology. Combined IgM and IgG
antibody titers to phase 11 (deteeted by both CF and IFA) peaked at 12 to 13 weeks
after the onset of symptoms. Combined IgG and IgM antibody to phase I were
virtually undeteetable du ring the aeute illness, and slowly inereased to a low level
at 1 year from the onset of symptoms. Anti-phase 11 IgM deteeted by IFA was
very high and peaked in week 3, dropping to undeteetable levels by week 26.
Anti-phase I IgM had a similar pattern, but had a peak one-third the magni-
tude of the peak for anti-phase 11. 95 A eareful study of 16 aeute Q fever patients
in a point-souree outbreak has confirmed that phase II-specifie IgG and IgM rise
mueh more rapidly and to higher titers in acute disease. Phase I-specifie IgM and
IgG responses did oeeur later in the course, but attained lower peak levels. 96
A number of investigators have reeommended approaehes to serologie
differentiation of acute Q fever from Q fever endoearditis and/or ehronie Q fever
hepatitis. Elevated IgG, IgM, and especially IgA titers speeifie for phase I antigen
have been noted to be assoeiated with ehronie forms of Q fever. 97- 99 Presenee
of IgA antibody to phase I determinants is strongly suggestive of ehronie disease,
but a reeent study indieates a need for eaution, beeause a few patients with aeute
disease and no clinieal evidenee of ehronicity have been found to develop phase
I-specifie IgA on long-term serologie foIlow-up.96 The ratio of antibody (usually
IgG) against phase 11 antigen to antibody to phase I antigen has been shown to be
helpful, with a ratio greater (usually much greater) than 1 in acute Q fever, equal
to or greater than 1 in granulomatous hepatitis, and less than or equal to 1 in Q
fever endoearditis. 97 Given the laek of standardization, it is important for the
clinician to be familiar with the particular assays performed by his or her local
laboratory.
7. THERAPY
Because acute Q fever is generally self-limited with no therapy, it is difficult to

interpret clinical data showing improvement with antibiotic treatment unless a
control group is also studied. Rifampin, doxycycline, and tetracycline have been
shown to increase the survival of C. burnetii-infected chicken embryos, whereas
clindamycin, erythromycin, and cephalothin showed no rickettsiostatic proper-
ties.l°o Although there has been no conclusive demonstration that any antibiotic is
rickettsiocidal for C. burnetii in vitro, some investigators feel that rifampin and
ftuoroquinolones may be rickettsiocidal for some strains. 101 ,102 In a controlled
clinical trial in patients with acute Q fever, the duration of fever was decreased
from 3.3 days in a group receiving no therapy to 1. 7 and 2 days by the administra-
tion of either doxycycline or tetracycline, respectively.32 Despite inconsistent in
vitro activity, some authors continue to suggest that erythromycin is useful
clinically, although the data represent small numbers of patients studied retro-
spectively.102a
Based on the currently available data, it is unclear whether treatment of acute
Q fever has any effect on future probability of chronic infection, but treatment of
diagnosed cases appears prudent. In vitro, data comparing acutely and chron-
ically infected cells suggest that organisms inhabiting recently infected cells
(22 days) have different susceptibilities than those in chronically infected cells
(>400 days). This variability in susceptibility is strain dependent, with the Nine
Mile strain being more sensitive to oftoxacin, rifampin, and doxycycline than the
Priscilla strain in both acutely and chronically infected cells. The Nine Mile strain
is equally sensitive to oftoxacin and rifampin in persistently and recently infected
cells, but more sensitive to doxycycline in recently infected than in persistently
infected cells. 103 It is therefore postulated that along with potential host factors
that may contribute to the chronicity of Q fever (see Section 8), both the strain
of C. burnetii and the antibiotic used may conceivably modulate the potential
development of the chronic disease state. Unfortunately, because the organism is
dangerous and difficult to isolate and cultivate, antibiotic sensitivities in individ-
ual clinical cases do not seem feasible in the foreseeable future.
The currently recommended therapy for acute Q fever is a 2-week course of
tetracycline 25 mg/kg/day up to I to 2 g/day in four divided dos es in adults and
children over 8 years of age. An alternative would be doxycycline 100 mg twice
daily for adults, or 0.45 mg/kg/day twice daily for children over 8 years of age.
Quinolones and trimethoprim/sulfamethoxazole have impressive in vitro activity
against C. burnetii, but clinical experience is largely lacking for these agents.l° 3a
There have been no randomized clinical trials comparing the effectiveness of
different therapies on chronic Q fever. Therapy should probably include tetra-
cycline or one of its derivatives for at least one year, if not indefinitely. Some
investigators suggest adding a quinolone or rifampin for synergistic effect. 104

Since drugs in the tetracycline family are contraindicated in children 8 years old
and under, quinolones have not yet been approved in children, and rifampin is
probably not an acceptable single agent, the choice of therapy in children 8 years
and under is severely limited. There are no currently available data on this
subject. Possible modes of therapy in young children might be trimethopriml
sulfamethoxazole or chloramphenicol, which have activity in vitro. As stated
above, neither of these drugs has been shown to be rickettsiacidal, their rickettsio-
static activity is modest, and there are no clinical data proving the efficacy of
either drug in children or adults. Given the serious long-term consequences of
inadequately treated Q fever endocarditis, the known risks and adverse effects
of tetracyclines in children may be acceptable in this setting. Fortunately, Q fever
is a predominantly adult infection.
Q fever endocarditis is difficult to eradicate and may relapse after protracted
therapy. Serology is, unfortunately, not a useful tool for documentation of
recovery from chronic Q fever since serologie positivity remains for years, in many
cases, with very little apparent correlation with clinical disease.l°4 Intractable
infection or hemodynamic compromise require surgical intervention, under the
coverage of medical therapy.
8. IMMUNOLOGY
The bulk of evidence from animal models suggests that definitive protective
immunity to C. burnetii is primarily mediated by cellular mechanisms, although
antibody may have a role. In inbred strains of mice with varying degrees of
susceptibility to C. burnetii, resistance was found to correlate with the capacity to
mount and maintain a vigorous lymphocyte proliferative response to whole-cell
antigen preparations, whereas all strains generated roughly similar antibody
titers. I05 Interestingly, these experiments showed resistance to be determined by
unidentified genes outside the H-2 complex. Further, an active role for the
organism in modulating T-cell responses was indicated by the finding that even
the more susceptible mouse strains could be successfully immunized with dead
C. burnetii.
Resistance to C. burnetii can be passively transferred by lymphoid cells of
immune animals, and guinea pigs immunized with whole-cell antigens on a
schedule that generated little or no antibody response were nonetheless solidly
resistant to challenge .106,107 These animals demonstrated significantly enhanced
peripheral blood lymphocyte proliferation in response to C. burnetii antigens.
Finally, athymic nude mice fail to control the organism, even in the presence of
significant amounts of antibody.108 The precise mechanism of action of immune
lymphocytes, presumptively T cells, has not been fully elucidated. Cocultivation
of infected normal guinea pig macrophages with either immune lymphocytes or
their supernatants (produced after antigen stimulation) suppressed the replica-
tion of C. burnetii 109 within the phagocytic cells.
Turco and colleagues showed that recombinant interferon-J' or crude

lymphokine-containing supernatants of stimulated murine cells could inhibit
growth of C. burnetii in L929 murine fibroblasts by a cycloheximide-sensitive
mechanism. 110 Although interferon-J' was clearly present in the crude super-
natants, its concentration appeared insufficient to account for the activity of these
preparations relative to the recombinant product, and other unidentified soluble
T-cell products were also implicated.
The presence of passively transferred antibody to C. burnetii given before the
time of challenge is associated with improved clearance of the organism at 1 week
in mice. lll Administration of antibody simultaneous with, or after, challenge
has no effect. This, and data of Kazar that indicate that proteetion correlates with
the opsonizing activity of antibody preparations,1I2 suggests that the role of
antibody consists of enhanced initial delivery of the pathogen to professional
phagocytes. Whether such an effect might in some instances be detrimental by
swiftly delivering the organism to its preferred phagolysosomal site remains
controversial. 1I3 In any event, an active T-cell response appears necessary for
definitive elimination of C. burnetii, as indicated by the previously cited nude
mouse model.
Several investigators have demonstrated that the T-cell response to C. burnetii
in humans is more persistent and more vigorous on initial infection than the
antibody response. On serial examinations of patients with Q fever infection,
90% had both positive intradermal skin tests with C. burnetii antigens and
antibody responses by microhemagglutination (MA) at 1 to 3 months postinfec-
tion. At 1 year, 90% had a positive response by skin testing, while 73% had positive
serology to phase I antigen, and 77% to phase 11 antigen. At 11 years, 90% still
had a positive skin test, whereas only 55% had positive serology with phase I,
and a similar proportion with phase 11. The clinical course of the subjects studied
was not reported. 1l3
Another study compared the T-cell response to C. burnetii in persons previ-
ously occupationally exposed to Q fever (documentation of previous infection not
given) with that of individuals vaccinated with a phase I trichloroacetic acid
extracted vaccine. Both skin testing (using intradermal injection of the vaccine)
and lymphocyte proliferation assays were used. The assays and skin tests were
performed at variable intervals after vaccination, and the timing of natural
exposure was variable. In the group of vaccinees, 77% demonstrated lymphocyte
proliferation in response to phase I antigen and 36% had a positive skin test. None
had lymphocyte proliferation in response to phase 11 antigen. Among the natu-
rally exposed individuals, 100% had lymphocyte proliferation in response to
phase I, 44% in response to phase 11, and 89% had a positive skin test. In ten
individuals with no previous exposure to Q fever, the skin test and the lymphocyte
proliferative assays were negative. 1I4
Host immunologie parameters have been postulated to contribute to the
predisposition to, and potential for evolution of, chronic Q fever. One study
comparing peripheral blood mononuclear cells from patients with Q fever hepa-
titis, endocarditis, acute infection, and controls demonstrated a dramatic lympho-
cyte proliferative response in hepatitis patients, a moderate response in acute Q

fever patients, and no significant response in endocarditis or control patients.
Lack ofT-cell responsiveness was specific to the Coxiella antigen; when stimulated
with Candida antigen, there was a significant proliferative response in all four
groups tested. ll5 Koster, Williams, and Goodwin demonstrated a possible role for
both CD8 + T cells and an adherent cell secreting prostaglandins in mediating the
Coxiella antigen-specific unresponsiveness of lymphocytes in patients with Q
fever endocarditis. They postulated the existence of an antigen-specific, T cell-
monocyte suppressor circuit driven by chronic antigenemia in such patients. ll6 It
remains unelear as to whether this mechanism represents a cause or effect of
chronic infection, although it does not appear to be operative in hepatitis or acute
Q fever patients. There is anecdotal data suggesting an association between
chronic Q fever and auto immune phenomena such as presence of smooth musele
antibody or elevated cold agglutinin titers. ll7
9. VACCINES
Attempts to develop Q fever vaccines have been ongoing since the pathogen
was discovered. In the first 30 years, the vaccines were either of low immuno-
genicity or of high reactogenicity, especially in previously sensitized individuals,
where sterile abscesses were found to occur frequently. It has been found, based
on animal challenge studies, that vaccines must contain at least phase I LPS
antigens to protect against challenge with phase I organisms. Several purified
protein antigens are currently under study, but their potential as vaccines in the
absence of LPS is as yet unelear. 1l8 In recent years, the leading vaccine candidates
have been formalin-killed phase I C. burnetii whole cells (WC), chloroform-
methanol-treated phase I C. burnetii cells (CM), and trichloroacetic acid-extracted
phase I antigen (TCAE).l12 A live attenuated C. burnetii went to field trial in the
Soviet Union, but was subsequentlY discarded because of concerns regarding
safety.
The WC, CM, and TCAE vaccines have been compared in animal models. In
one study, guinea pigs and mice were given one of the three vaccines and
subsequently challenged with live C. burnetii. WC gave the highest level of
protection and the earliest onset of protection, independent of the route of
immunization or challenge (intraperitoneal vs. subcutaneous and intraperitoneal
vs. aerosolized, respectively).1l9 In pregnant ewes, the WC and CM vaccines both
dramatically decreased, but did not entirely prevent, Coxiella shedding in placen-
tae, amniotic fluid, and colostrum following C. burnetii challenge at the 100th
day of gestation. In addition, the lambs born to the vaccinated groups experi-
enced no neonatal mortality, and were larger and stronger than those born to the
control groups.l20 In a variety of animals, ineluding mice, guinea pigs, rabbits,
and cattle, WC vaccination induced higher antibody levels and better delayed-
hypersensitivity responses than CM vaccination at identical doses. The WC
preparation, however, caused local inflammation at lower doses than CM. All
three existing preparations, however, may demonstrate significant local reacto-

genicity at higher doses associated with strong immune responses.l21-123
The vaccine that has been used most widely in humans, and the only Q Fever
vaccine that has been tested in a human efficacy trial, is a partially purified
formalin-inactivated phase I whole cell vaccine. Descriptive data has been col-
lected in 4,000 abattoir workers in Australia who have received vaccine over a
7-year period. Individuals were prescreened using both skin testing and antibody
measurement, so that those previously exposed to C. burnetii were presumably
exeluded. In this group, two vaccine es developed sterile abscesses after vaccl.na-
tion, and two had small persistent subcutaneous lumps. Eight Q fever cases were
observed in vaccinees, and 97 cases were detected in unvaccinated, exposed
persons (with unknown previous immunity). There was an 80% seroconversion
rate immediately after vaccination, falling to 60% at 20 months, and approx-
imately 90% of vaccine es developed lymphoproliferative responses after vaccina-
tion that remained detectable for 5 years. 124 In the only randomized, blinded,
placebo-controlled trial in seronegatives, no cases of Q fever were observed in 98
vaccinees, and 7 were observed in 102 controls.l 25 Thus, the extant whole-cell
vaccines appear to have good efficacy, purchased at the price of some local
reactogenicity. This can be minimized, but not eliminated, by cumbersome
prescreening with skin tests, serology, and possibly in vitro proliferative assays. A
better vaccine is needed.
10. SUMMARY
Q fever is a relatively recently recognized disease syndrome with serious

consequences. Much work has been done in defining the organism and its elinical
manifestations, but many important questions regarding the microbiology,
pathogenesis, optimal therapeutic regimen(s), and optimal preventive measures
have yet to be answered.
This review has attempted to summarize much of what is currently known
about C. burnetii infection. C. burnetii is different in structure and function than
other members of the Rickettsiaceae family and, importantly, is more resistant to
desiccation. The organism is ubiquitous, and has been shown to colonize, if not
infect, most of the animal species in which evidence of infection has been sought,
ineluding many mammals, arthropods, and a number of birds. In humans,
seropositivity to C. burnetii is especially high in persons who are regularly in elose
contact with farm animals. Transmission to humans is most frequently via
inhalation of aerosolized partieles, with other documented but less common
modes of transmission being tick bites, breast feeding, and perinatal transmis-
sion. The major mode of transmission among animals is less elear, though
transmission has been documented by partiele inhalation, sexual activity, and
tick bites.
Clinical Q fever has significant mortality in its chronic form, especially in the
setting of endocarditis. In endemic areas, Q fever pneumonia can be a significant
176 ANN W FUNKHOUSER and LOUIS E FRIES
contributor to respiratory morbidity, though the course is usually self-limited.

Though serologic dia gnosis has not been standardized, the sensitivity of serologic
detection is reasonable for acute disease. There is good evidence that the T-cell
response is Ion ger lasting than the B-cell response to C. burnetii. Cellular immu-
nity is primarily responsible for eradicating infection, but humoral immunity may
contribute to resistance. In addition, a defect in specific T-cell responses seems
to be associated with the development of chronic Q fever, either as a cause for
prolonged infection, or as a result of it. Among the current candidate vaccines, the
formalin inactivated whole-cell vaccine seems to be more immunogenic but more
reactogenic in animals than either the chloroform-methanol-treated or trichloro-
acetic acid extracted vaccines. The whole-cell vaccine is quite efficacious in man,
but has troublesome local reactogenicity.
There are many unresolved questions surrounding C. burnetii and Q fever.
Some of the important questions to be addressed in future research might
indude the following:
1. What is the life cyde of C. burnetii? What are the determinants of the
resistance of this organism to environmental changes that normally destroy
rickettsiae and other organisms? Although spore-like partides have been ob-
served microscopically, further study is required to understand the determinants
of spore formation, antigenie characterization of spores, and the role of spores in
propagation of further generations.
2. Do specific plasmid variants, restrietion endonudease analysis of chromo-
somal DNA, or LPS serotypic markers actually predict the potential of a given
strain to produce chronic disease? If so, these data might indicate important
targets for immunization, identify "safe" strains that could be used to study
prophylactic and therapeutic approaches in man, or provide useful markers for
evaluation and therapeutic planning in individual dinical cases.
3. How frequent is human-to-human transmission? Although there is anec-
dotal evidence for casual human-to-human transmission, this subject has not been
formally studied. Although perinatal and breast milk transmission have been docu-
mented, the importance of these modes of transmission in humans is unknown.
4. How can C. burnetii be safely and cheaply isolated or identified and
studied in a laboratory setting? Convenient and rapid diagnostic methodology is
crucial to further advances in therapy for both acute and chronic Q fever. The
availability of antibiotic sensitivities in patients with Q fever might prevent some
portion of chronic infection, and might shorten the course of acute disease.
5. What is/are the most effective therapeutic regimen(s) in adults and in
children? Are different drugs optimal in the acute and chronic settings?
6. Is animal or human immunization more cost-effective for controlling in-
feetion in the human population? Ifhuman immunization is more cost-effective,
which populations ofhumans should be vaccinated to optimally prevent disease?
7. What is/are the safest and most efficacious vaccine(s) for use in humans?
What is the duration of immunity, and optimal dosing regimen for the candidate
vaccine(s)?
REFERENCES
1. Baca, O. G. and Paretsky, D., 1983, Q fever and Coxiella burnetii: A model for host-parasite
interactions, Mierobiol. Rev. 47:127-149.
2. Aitken,1. D., Bogei, K., Cracea, E., Edlinger, E., Houwers, D., Krauss, H., Rady, M.,
Rehacek,J., Schiefer, H. G., Schmeer, N., Tarasevich, I. v., and Tringali, G., 1987, Q fever in
Europe: Current aspects of aetiology, epidemiology, human infection, diagnosis and
therapy, Infection 15:323-327.
3. Krauss, H., Schiefer, H. G., and Schmatz, H. D., 1977, Ultrastructural investigations on
surface structures involved in Coxiella burnetii phase variation, Infect. Immun. 15:890-896.
4. Hackstadt, T., 1990, The role of lipopolysaccharide in the virulence of Coxiella burnetii,
Ann. N.Y. Aead. Sei. 590:27-32.
5. Savinelli, E. A. and Mallavia, L. P., 1990, Comparison of Coxiella burnetii plasmids to
homologous chromosomal sequences present in a plasmidless endocarditis-causing isolate,
Ann. N.Y. Aead. Sei. 590:523-533.
6. Samuel, J. E., Frazier, M. E., and Mollavia, L. P., 1985, Correlation of plasmid type and
disease caused by Coxiella burnetii, Infect. Immun. 49:775-779.
6a.Minnick, M. F., SmalI, C. L., Franzier, M. E., and Mallavia, L. P., 1991, Analysis ofthe cbhE'
plasmid gene from acute disease-causing isolates of Coxiella burnetii, Gene 103:113-118.
6b.Hendrix, L. R., Samuel, J. E., and Mallavia, L. P., 1991, Differentiation of Coxiella burnetii
isolates by analysis of restriction-endonudease-digested DNA separated by SDS-PAGE,J
Gen. Mierobiol. 137:269-276.
7. Hackstadt, T., 1986, Antigenic variation in the phase I lipopolysaccharide of Coxiella
burnetii isolates, Infeet. Immun. 52:337-340.
8. Kaplan, M. M. and Bertagna, P., 1955, The geographical distribution of Q fever, Bull.
WH.O. 13:829-860.
9. Babudieri, B., 1959, Q fever: A zoonosis, Adv. Veto Sei. 5:81-154.
10. Krauss, H., 1989,Clinical aspects and prevention of Q fever in animals, Eur. J Epidemiol.
5:454-455.
11. Tylewska-Wierzbanowska, S. K. and Kruszewska, D., 1990, Q fever-sexually transmitted
infection? J Infect. Dis. 161:368-369.
12. Richardus, J. H., Donkers, A., Dumas, A. M., and Schaap, G. J. P., 1987, Q fever in the
Netherlands: A sero-epidemiological survey among human population groups from
1968 to 1983, Epidemiol. Infect. 98:211-219.
13. Somma-Moreira, R. E., Caffarena, R. M., Somma, S., Perez, G., and Montiero, M., 1987,
Analysis of Q fever in Uruguay, Rev. Infect. Dis. 9:386-387.
14. Ming-yuan, F., Walker, D. H., Shu-rong, Y., and Qing-huai, L., 1987, Epidemiology and
ecology of rickettsial diseases in the People's Republic of China, Rev. Infect. Dis. 9:823-840.
15. Behymer, D. and Riemann, H. P., 1989, Coxiella burnetii infection (Q fever), JA. V.M.A.
194:764-767.
16. Tellez, A., Martin, A., Anda, P., de la Fuente, L., Benitez, P., Garcia, C., and Leon, P., 1989,
Study of C. burnetii human and animal seroprevalence in a rural population in Madrid
community, Eur. J EpidemiDI. 5:444-446.
17. Rehacek, J., 1989, Ecological relationships between ticks and rickettsiae, Eur. J EpidemiDI.
5:407-413.
18. Mann,J. S., Douglas,J. G., Inglis,J. M., and Leitch, A. G., 1986, Q fever: Person to person
transmission within a family, Tlwrax 41:974-975.
19. Beaman, M. H. and Hung, J., 1989, Pericarditis associated with tick-borne Q fever, Aust.
N.Z. J Med. 19:254-256.
20. Kumar, A., Yadav, M. P., and Kakkar, S., 1981, Human milk as a source of Q fever infection
in breast-fed babies, Indian J Med. Res. 73:510-512.
21. Richardus,]. H., Dumas, A. M., Huisman,]., and Schaap, G.]. P., 1985, Q fever in infancy:
A review of 18 cases, Pediatr. Infect. Dis. 4:369-373.
22. Prasad, B. N., Chandiramani, N. K., and Wagle, A., 1986, Isolation ofCoxiella burnetii from
human sources, Int.] Zoonoses 13:112-117.
23. Ellis, M. E., Smith, C. C., and Moffat, M. A.]., 1983, Chronic or fatal Q-fever infection: A
review of 16 patients seen in North-East Scotland (1967-80), Quart.] Med. 205:54-66.
24. Marrie, T.]., 1988, Seroepidemiology ofQ fever in New Brunswick and Manitoba, Can.]
Microbiol. 34:1043-1045.
25. Marrie, T.]., Van Buren,]., Faulkner, R. S., Haldane, E. v., Williams,]. C., and Kwan, C.,
1984, Seroepidemiology of Q fever in Nova Scotia and Prince Edward Island, Can. ]
26. Raoult, D., Toga, B., Chaudet, H., and Chiche-Portiche, C., 1987, Rickettsial antibody in
southern France: Antibodies to Rickettsia conorii and Coxiella burnetii among urban, sub-
urban and semi-rural blood donors, Trans. R. Soc. Trop. Med. Hyg. 81:80-81.
27. Lang, G. H., 1989, Q fever: An emerging public health concern in Canada, Can.] Veto Res.
53:1-6.
28. Sixl, W. and Sixl-Voigt, B., 1987, Research on a possible Q-fever infection in humans and
animals on the Cap Verde Islands (Santa CruziSantiago, West Africa),] Hyg. Epidemiol.
Microbiol. Immunol. 31(Suppl):472-474.
29. Cracea, E., Constantinescu, S., Tofan, N., Caruntu, F., and Dogaru, D. 1989, Q fever urban
cases in Romania, Arch. Roum. Path. Exp. Microbiol. 48:13-17.
30. Centers for Disease Control, 1986, Q fever among slaughterhouse workers-California,
M.M. W.R. 35:223-226.
31. Derrick, E. H., 1937, Q fever, a new fever entity: C1inical features, diagnosis and laboratory
investigations, Med.] Aust. 2:281-305.
32. Spelman, D. w., 1982, Q fever: A study of 111 consecutive cases, Med.] Aust. 1:547-553.
33. Graham, C.]., Yamauchi, T., and Rountree, P., 1989, Q fever in animallaboratory workers:
An outbreak and its investigation, Am. J Infect. Contral. 17:345-348.
34. Rauch, A. M., Tanner, M., Pacer, R. E., Barrett, M.]., Brokopp, C. D., and Schonberger,
L. B., 1987, Sheep-associated outbreak of Q fever, Idaho, Arch. Intern. Med. 147:341-344.
35. Simor, A. E., Brunton,]. L., Salit, I. E., Vellend, H., Ford-Jones, L., and Spence, L. P., 1984,
Q fever: Hazard from sheep used in research, 130:1013-1016.
36. Meiklejohn, G., Reimer, L. G., Graves, P. S., and Helmick, C., 1981, Cryptic c;pidemic of Q
fever in a medical school,] Infect. Dis. 144:107-113.
37. Hall, C.]., Richmond, S.]., Caul, E. 0., Pearce, N. H., and Silver, I. A., 1982, Laboratory
outbreak of Q fever acquired from sheep, Lancet 1:1004-1006.
38. Schachter, ]., Sung, M., and Meyer, K. F., 1971, Potential danger of Q fever in a university
hospital environment,] Infect. Dis. 123:301-304.
39. Curet, L. B. and Paust,]. C., 1972, Transmission of Q fever from experimental sheep to
laboratory personneI, Am.] Obstet. Gynecol. 114:566-568.
40. Dupuis, G., Petite,]., Peter, 0., and Vouilloz, M., 1987, An important outbreak of human Q
fever in a Swiss alpine valley, 16:282-287.
41. Langley,]. M., Marrie, T.]., Covert, A., Waag, D. M., and Williams,]. C., 1988, Poker players'
pneumonia: An urban outbreak of Q fever following exposure to a parturient cat, N.
Engi.] Med. 319:354-356.
42. Marrie, T. ]., Langille, D., Papukna, v., and Yates, L., 1989, Truckin' pneumonia-an
outbreak of Q fever in a truck repair plant probably due to aerosols from c10thing
contaminated by contact with newborn kittens, Epidemiol. Infect. 102:119-127.
43. Marrie, T.]., Durant, H., Williams,]. C., Mintz, E., and Waag, D. M., 1988, Exposure to
parturient cats: A risk factor for acquisition of Q fever in maritime Canada,] Infeel. Dis.
158:101-108.
44. Marrie, T. J., MaeDonald, A., Durant, H., Yates, L., and MeCormick, L., 1988, An outbreak
of Q fever probably due to eontaet with a parturient cat, Chest 93:98-103.
45. Jorm, L. R., Lightfoot, N. F., and Morgan, K. L., 1990, An epidemiological study of an
outbreak of Q fever in a seeondary sehool, Epidemiol. Infect. 104:467-477.
46. Winner, S. J., Eglin, R. P., Moore, V. I. M., and Mayon-White, R. T., 1987, An outbreak of Q
fever affeeting postal workers in Oxfordshire,] Infect. 14:255-261.
47. Marmion, R. A. and Stoker, M. G. P., 1950, Q fever in Britain, Epidemiology of an
outbreak, Lancet ii:611-616.
48. Sawyer, L. A., Fishbein, D. B., and MeDade, J. E., 1987, Q fever: Current eoneepts, Rev.
Infect. Dis. 9:935-946.
49. Sobradillo, v., Ansola, P., Baranda, F., and Corral, C., 1989, Q fever pneumonia: A review
of 164 eommunity-aequired eases in the Basque eountry, Eur. Respir.] 2:263-266.
50. Reilly, S., Northwood,J. L., and Caul, E. 0.,1990, Q fever in Plymouth, 1972-88: A review
with partieular referenee to neurological manifestations, Epidemiol. Infect. 105:391-408.
51. Miliar, J. K., 1978, The ehest film findings in 'Q' fever-a series of 35 eases, Clin. Radiol.
29:371-375.
52. Gordon, J. D., MaeKeen, A. 0., Marrie, T. J., and Fraser, D. B., 1984, The radio-
graphie features of epidemie and sporadie Q fever pneumonia,] Can. Assoc. Radiol. 35:
293-296.
53. Lipton, J. H., Fong, T. C., GilI, M. J., Burgess, K., and Elliott, P. 0., 1987, Q fever
inflammatory pseudotumor of the lung, Chest 92:756-757.
54. Seggev,J. S., Levin, S., and Sehey, G., 1986, Unusual radiologieal manifestations ofQ fever,
Eur.] Respir. Dis. 69:120-122.
55. Torres, A., de Celis, M. R., Roisin, R. R., Vidal,J., and Vidal, A. A., 1987, Adult respiratory
distress syndrome in Q fever, Eur.] Respir. Dis. 70:322-325.
56. Domingo, P., Orobitz, J., Colombina, J., Alvarez, E., and Cadafalch, J., 1988, Liver involve-
ment in acute Q fever, Chest 94:895-896.
57. Berkoviteh, M., Aladjem, M., Beer, S., and Cohar, K., 1985, A fatal ease of Q fever hepati-
tis in a ehild, Helv. Paediatr. Acta. 40:87-91.
58. Kelly, R. P., Byrnes, 0. J., and Turner, J., 1986, Acute, severe hepatitis due to Coxiella burnetii
infeetion, Med.] Aust. 144:151-152.
59. Joshi, M. V. and Banerjee, K., 1987, Studies on some aspeets of glucose metabolism in
guineapigs infeeted with Coxiella burnetii, Indian] Med. Res. 86:20-24.
60. Brooks, R. G., Lieitra, C. M., and Peaeoek, M. G., 1986, Eneephalitis eaused by Coxiella
burnetii, Ann. Neural. 20:91-93.
61. Cameron, 0. A., Freedman, A. R., and Wansbrough-Jones, M. H., 1990, Q fever eneepha-
litis,] Infect. 20:159-162.
62. Marrie, T. J., 1985, Pneumonia and meningo-eneephalitis due to Coxiella burnetii,] Infect.
11:59-61.
63. Gomez-Aranda, F., Diaz,J. P., Acebai, M. R., Cortes, L. L., Rodriguez, A. N. and Morena,
J. M., 1984, Computed tomographie brain sean findings in Q fever eneephalitis, Neuro-
radiology 26:329-332.
64. Smith, C. C., Smith, 0. F., and Hudson, M., 1989, Progressive demyelinating disease
following acute Q fever,] Infect. 19:153-155.
65. Shaked, Y. and Samra, Y., 1989, Q fever meningoeneephalitis associated with bilateral
abdueens nerve paralysis, bilateral optie neuritis and abnormal eerebrospinal fluid find-
ings, Infection 17:394-395.
66. Sehuil,J., Riehardus,J. H., Baarsma, G. S., and Sehaap, G.J. P., 1985, Q fever as a possible
eause of bilateral optic neuritis, Br.] Ophthalmol. 69:580-583.
67. Conget, 1., Mallolas, J., Mensa, J., and Rovira, M., 1987, Erythema nodosum and Q fever,
Arch. Dermatol. 123:867.
68. Cardellach F., Font, j., Agusti, A. G. N., Ingelmo, M., and Balcells, A., 1983, Q fever and
hemolytic anemia,] Infect. Dis. 148:769.
69. Hitchins, R., Cobcroft, R. G., and Hocker, G., 1986, Transient severe hypoplastie anemia in
Q fever, Patlwlogy 18:254-255.
70. Raoult, 0. and Espejo-Arenas, E., 1989, Osteoarticular infeetion due to Coxiella burnetii,]
Infect. Dis. 159:1159-1160.
71. Srigley, j. R., Vellend, H., Palmer, N., Phillips, M. j., Geddie, W R., Van Nostrand, A. W P.,
and Edwards, V. 0., 1985, Q fever: The liver and bone marrow pathology, Am.] Surg. Patlwt.
9:752-758.
72. Qizilbash, A. H., 1983, The pathology of Q fever as seen on liver biopsy, Arch. Patlwl. Lab.
Med. 107:364-367.
73. Atienza, P., Ramond, M.-j., Degott, C., Lebree, 0., Rueff, B., and Benhamou, j.-P., 1988,
Chronie Q fever hepatitis eomplieated by extensive fibrosis, Gastroenterology 95:478-481.
74. Travis, L. B., Travis, W 0., Li, C.-Y., and Pierre, R. v., 1986, Q fever: A clinieopathologie
study of five eases, Arch. Patlwl. Lab. Med. 1l0:1017-1O20.
75. Silver, S. S. and McLeish, W A., 1984, "Doughnut" granulomas in Q fever, Can. Med. Assoe.]
130:102-104.
76. Voight, j. j., Delsol, G., and Fabre, j., 1983, Liver and bone marrow granulomas in Q fever,
Gastroenterology 84:887-890.
77. Estrov, Z., Bruek, R., Shtalrid, M., Berrebi, A., and Resnitzky, P., 1984, Histioeytie hemo-
phagoeytosis in Q fever, Arch. Patlwl. Lab. Med. 108:7.
78. Khavkin, T. and Tabibzadeh, S. S., 1988, Histologie, immunofluoreseenee, and eleetron
mieroseopie study of infeetious proeess in mouse lung after intranasal ehallenge with
Coxiella burnetii, Infect. Immun. 56:1792-1799.
79. Lillie, R. 0., Perrin, T. L., and Armstrong, C., 1941, An institutional outbreak of pneu-
monitis. III. Histopathology in man and rhesus monkeys in the pneumonitis due to the
virus of "Q" fever, Pub. Health Rep. 56:149-155.
80. Palmer, S. R. and Young, S. E. j., 1982, Q fever endocarditis in England and Wales, 1975-
1981, Lancet ii:1448-1449.
81. Raoult, 0., Etienne, j., Maissip, P., Iaocono, S., Prinee, M. A., Beaurain, P., Beniehou, S.,
Auvergnat,j. C., Mathieu, P., Buehet, P. H., and Serradimigrii, A., 1987, Q fever endoear-
ditis in the south of Franee,] Infect. Dis. 155:570-573.
82. Beaufort-Krol, G. C. M. and Storm, C. j., 1987, Chronie Q fever endocarditis,] Pediatr.
1l0:330-331.
83. Fort, S., Fraser, A. G. and Fox, K. A. A., 1989, Extensive aortic valve ring abseess formation:
A rare eomplieation of Q fever endoearditis, Postgrad. Med.] 65:384-386.
84. Laufer, 0., Lew, P. 0., Oberhansli, 1., Cox,j. N., and Longson, M., 1986, Chronic Q fever
endoearditis with massive splenomegaly in ehildhood,] Pediatr. 108:535-539.
85. Perez-Fontan, M., Huarte, E., Tellez, A., Rodriguez-Carmona, A., Pieazo, M. L,. and
Martinez-Ara, j., 1988, Glomerular nephropathy assoeiated with ehronie Q fever, Am.]
Kidney Dis. 4:298-306.
86. Yebra, M., Marazuela, M., Albarran, F., and Morena, A., 1988, Chronie Q fever hepatitis,
Rev. Infect. Dis. 10:1229-1230.
87. Rieehman, N., Raz, R., Keysary, A., Goldwasser, R., and Flatau, E., 1988, Chronie Q fever
and severe thromboeytopenia in a pregnant woman, Am. ] Med. 85:253-254.
88. Fergusson, R.j., Shaw, T. R. 0., Kitehin, A. H., Matthews, M. B., Inglis,j. M., and Peutherer,
R., 1985, Subclinieal ehronie Q fever, Q.] Med. 57(222):669-676.
89. Biggs, B. A., Douglas, j. G., Grant, I. WB., and Crompton, G. K., 1984, Prolonged Q fever
assoeiated with inappropriate seeretion of anti-diuretie hormone,] Infect. 8:61-63.
90. Fernandez-Guerrero, M. L., Muelas,j. M., Aguado,j. M., Renedo, G., Fraile,j., Soriano, F.,
and de Villalobos, E., 1988, Q fever endoearditis on porcine bioprosthetie valves, Ann.
Intern. Med. 108:209-213.
COXIELLA BURNETII AND Q FE VER 181
91. Kovacova, E., Gallo, ]., Schramek, S., Kazar, ]. and Brezina, R., 1987, Coxiella burnetii
antigens for detection of Q fever antibodies by ELISA in human sera, Acta Viral. 31:
254-259.
92. Roges, G. and Edlinger, E., 1986, Immunoenzymatic test for Q-fever, Diagn. Microbiol.
Infect. Dis. 4:125-132.
93. Peter, 0., Dupuis, G., Peacock, M. G., and Burgdorfer, W, 1987, Comparison of enzyme-
linked immunosorbent assay and complement fixation and indirect fluorescent-antibody
tests for detection of Coxiella burnetii antibody,J Clin. Microbiol. 25:1063-1067.
94. Field, P. R., Hunt,]. G., and Murphy, A. M., 1983, Detection and persistence ofspecific IgM
antibody to Coxiella burnetii by enzyme-linked immunosorbent assay: A comparison with
immunofluorescence and complement fixation tests, J Infoct. Dis. 148:477-487.
95. Dupuis, G., Peter, 0., Peacock, M., Burgdorfer, W, and Haller, E., 1985, Immunoglobulin
responses in acute Q fever,J Clin. Microbiol. 22:484-487.
96. Embil,]., Williams,]. C., and Marrie, T.]., 1990, The immune response in a cat-related
outbreak of Q fever as measured by the indirect immunofluorescence test and the enzyme-
linked immunosorbent assay, Can. J Micrabiol. 36:292-296.
97. Peacock, M. G., Philip, R. N., Williams,]. C., and Faulkner, R. S., 1983, Serologic evaluation
ofQ fever in humans: Enhanced phase I titers ofimmunoglobulins Gand Aare diagnostic
for Q fever endocarditis, Infoct. Immun. 41:1089-1098.
98. Peter 0., Dupuis, G., Bee, D., Luthy, R., Nicolet,]., and Burgdorfer, W, 1988, Enzyme-
linked immunosorbent assay for diagnosis of chronic Q fever,J Clin. Microbiol. 26:1978-1982.
99. Raoult, D., Urvolgyi,]., Etienne,]., Roturier, M., Puel,]., and Chaudet, H., 1988, Diagnosis
of endocarditis in acute Q fever by immunofluorescence serology, Acta Viral. 32:70-74.
100. Spicer, A.]., Peacock, M. G., and Williams,]. C., 1981, Effectiveness of several antibiotics in
suppressing chick embryo lethality during experimental infections by Coxiella burnetii,
Riekettsia typhi, and R. riekettsii, in: Riekettsiae and Riekettsial Diseases, (W Burgdorfer and
R. L. Anacker, eds.) New York: Academic Press, pp. 375-383.
101. Yeaman, M. R. and Baca, 0. G., 1990, Unexpected antibiotic susceptibility of a chronic
isolate of Coxiella burnetii, Ann. N.Y. Aead. Sei. 590:297-305.
102. Yeaman, M. R., Mitscher, L. A., and Baca, 0. G., 1987, in vitra susceptibility of Coxiella
burnetii to antibiotics, including several quinolones, Antimierab. Agents Chemother. 31:1079-
1084.
102a.perez-del-Molino, A., Aguado, ]. M., Riancho, ]. A., Sampedro, 1., Matoras, P., and
Gonzalez- Macias,]., 1991, Erythromycin and the treatment of Coxiella burnetii pneumonia,
J Antimicrob. Chemother. 28:455-459.
103. Yeaman, M. R., Roman, M.]., and Baca, O. G., 1989, Antibiotic susceptibility oftwo Coxiella
burnetii isolates implicated in distinct clinical syndromes, Antimicrab. Agents Chemother.
33:1052-1057.
103a.Raoult, D., Torres, H., and Drancourt, M., 1991, Shell-vial assay: Evaluation of a new
technique for determining antibiotic susceptibility, tested in 13 isolates of Coxiella burnetii,
Antimierab. Agents Chemother. 35:2070-2077.
104. Raoult, D., 1989, Antibiotic susceptibility of Rickettsia and treatment of rickettsioses, Eur.J
Epidemiol. 5:432-435.
105. Scott, G. H., Williams,]. C., and Stephenson, E. H., 1977, Animal models of Q fever:
Pathological responses of inbred mice to phase I Coxiella burnetii, J Gen. Mierobiol. 133:
691-700.
106. Kazar,]., El-Najdawi, E., Brezina, R., and Schramek, S., 1977, Search for correlates of
resistance to virulent challenge in mice immunized with Coxiella burnetii, Acta Viral. 21:
422-430.
107. Kishimoto, R. A.,Johnson,]. W, Kenyon, R. H., Ascher, M. S., Larson, E. W, and Pederson,
C. E., 1978, Cell-mediated immune responses of guinea pigs to an activated phase I Coxiella
burnetii vaccine, Infoct. Immun. 19:194-198.
108. Kishimoto, R. A., Rozmiarek, H., and Larson, E. W, 1978, Experimental Q fever infec-
tion in congenitally athymic nude mice, Infect. Immun. 22:69-71.
109. Heinrich, 0. j. and Jerrells, T. R., 1976, In vitra evaluation of immunity to Caxiella burnetii,j
Immunal. 117:996-1003.
110. Turco, j., Thompson, H. A., and Winkler, H. H., 1984, Interferon-'Y inhibits growth of
Caxiella burnetii in mouse fibroblasts, Infect. Immun. 45:781-783.
111. Humphres, R. C. and Hinrichs, 0. j., 1988, Role for antibody in Caxiella burnetii infection,
112. Kazar, j., 1988, Immunity in Q fever, Acta Viral. 32:358-368.
113. Kazar, j., Schramek, S., and Brezina, R., 1984, The value of skin test in Q fever convales-
cents and vaccinees as indicator of antigen exposure and inducer of antibody recall, Acta
Viral. 28:134-140.
114. Gajdosova, E. and Brezina, R., 1989, Cell-mediated immune response to Caxiella burnetii
antigens in Q fever convalescents and vaccinees, Acta Viral. 33:474-481.
115. Koster, F. T., Williams, j. C., and Goodwin, j. S., 1985, Cellular immunity in Q fever:
Specific lymphocyte unresponsiveness in Q fever endocarditis,j Infect. Dis. 152:1283-1289.
116. Koster, F. T., Williams, j. C., and Goodwin, j. S., 1985, Cellular immunity in Q fever:
modulation of responsiveness by a suppressor T cell-monocyte circuit, j Immunal. 135:
1067-1072.
117. Levy, P., Raoult, 0., and Razongles, j. j., 1989, Q-fever and autoimmunity, Eur. j Immunal.
5:447-453.
118. Williams,j. C., Hoover, T. A., Waag, 0. M., Banerjee-Bhatnagar, N., Bolt, C. R., and Scott,
G. H., 1990, Antigenie structure of Caxiella burnetii-a comparison of lipopolysaccharide
and protein antigens as vaccines against Q fever, Ann. N.Y. Aead. Sei. 590:370-380.
119. Kazar, j., Votruba, 0., Propper, P., and Schramek, S., 1986, Onset and duration of immunity
in guinea pigs and mice induced with different Q fever vaccines, Acta Viral. 30:499-506.
120. Brooks, 0. L., Ermel, R. W, Franti, C. E., Ruppanner, R., Behymer, 0. E., Williams, j. C.,
and Stephenson, J. C., 1986, Q fever vaccination of sheep: Challenge of immunity in ewes,
Am. j Veto Res. 47:1235-1238.
121. Kazar, j., Schramek, S., Lisak, v., and Brezina, R., 1987, Antigenicity of chloroform-
methanol-treated Caxiella burnetii preparations, Aeta Viral. 31:158-167.
122. Resl, M., Sidorenko, S. v., and Zhuljabin, E. 1., 1989, Development oflocal morphological
changes after intradermal inoculation of Q-fever chemovaccine, Acta Viral. 33:167-171.
123. Ascher, M. S., Williams,j. C., and Berman, M. A., 1983, Dermal granulomatous hypersen-
sitivity in Q fever: Comparative studies of the granulomatous potential of whole cells of
Caxiella burnetii phase land subfractions, Infect Immun. 42:887-889.
124. Marmion, B. P., Ormsbee, R. A., Kyrkou, M., Wright, j., Worswick, 0. A., Izzo, A. A.,
Esterman, A., Feery, B., and Shapiro, R. A., 1990, Vaccine prophylaxis of abattoir-
associated Q fever: Eight years' experience in Australian abattoirs, Epidemial. Infect.
104:275-287.
125. Shapiro, R. A., Siskind, v., Schofield, F. 0., Stallman, N., Worswick, 0. A., and Marmion,
B. P., 1990, A randomized, controlled, double-blind, cross-over, clinical trial of Q fever
vaccine in selected Queensland abattoirs, Epidemial. Infect. 104:267-273.
11
The Chlamydial Pneumonias

BURKE A. CUNHA
1. INTRODUCfION
Chlamydia affeets all age groups and is ubiquitous in distribution. In the 1930s the
life eyde of Chlamydia psittaci, then known as Bedsonia, was first eharaeterized.
Psittaeosis has long been reeognized as a disease of psittacine birds, and for this
reason psittaeosis is oeeasionally termed ornitlwsis. Psittaeosis has been reeognized
as a distinet dinieal entity for deeades and the inerease of psittacosis has inereased
steadily sinee the 1970s. Infant neonatal pneumonia eaused by Chlamydia traclw-
matis was reported and deseribed in the late 1970s. Infant ehlamydia pneumonia
is not uneommonly seen in the hospital setting, but must eertainly be more
eommon and largely unreeognized in the ambulatory setting. In 1983, a distinet
Chlamydia species was isolated from a patient with an aeute respiratory illness.
This isolate, whieh was unrelated to C. traclwmatis or C. psittaci, was initially ealled
TWAR agent sinee the first isolate was from Taiwan, henee the TW, and the isolate
was associated with an aeute respiratory illness, henee the AR.
The term TWAR continued to be used until relatively reeently when the
organism was formally dassified as a new species of Chlamydia, that is Chlamydia
pneumoniae. C. pneumoniae may be as eommon or more eommon than Mycoplasma
pneumonia in the young adult population, but has been deseribed less frequently
in the elderly population as well. Chlamydiae have an interesting biologieallife
eyde, unique among the obligate intraeellular pathogens eausing human disease.
Chlamydial infeetions are eommon, in general diffieult to eradieate, tend to
persist, and are prone to relapse beeause of their eomplex life eyde and their
BURKE A. CUNHA • Infectious Disease Division, Winthrop-University Hospital, Mineola,

New York 11051; and SUNY School of Medicine, Stony Brook, New York 11790.
Pulmonary Infoctions arui Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
183
184 BURKE A. CUNHA
protected intracellular location. l -6 The chlamydial species responsible for caus-

ing pneumonias are difficult to isolate in the dinical setting and the dia gnosis is
usually confirmed retrospectively by serological methods. Fortunately, the dinical
presentation of each of the chlamydial pneumonias is distinctive, providing the
basis for an accurate, presumptive diagnosis pending serological verification. By
appreciating the key diagnostic features that distinguish the chlamydial pneu-
monias from other pulmonary pathogens, the dinician can prescribe appropriate
empirie antimicrobial therapy. Fortunately chlamydial agents are susceptible to
available and commonly prescribed nontoxic antimicrobial agents which, when
given in the proper dose for adequate duration, are able to effectively treat all
of the chlamydial pneumonias. 7- 1O
2. MICROBIOLOGY
Chlamydia are obligate, intracellular microorganisms that are trophic for

squamocolumnar epithelial cells lining mucous membranes. Chlamydia differ
from viruses in that they contain RNA and DNA, synthesize protein indepen-
dendy, contain ribosomes, and are susceptible to antibiotics. The trilaminar cell
walls of Chlamydia, like gram-negative bacilli, are responsible for group specific
reactions. Chlamydial cell walls lack muramic acid and are intermediate in size
between virus es and bacteria.
Chlamydia exist intracellularly in two morphologically unique forms. The
elementary body (EB) is approximately 0.4 /-Lm in diameter and is the infectious
unit in the chlamydiallife cyde. The outer membrane proteins of the EBs are
extensively cross-linked in a peptidoglycan matrix utilizing extensive disulfide
bond cross-linking. EBs are the infectious partides, are responsible for chla-
mydial attachment to the host cell, induce phagocytosis by the host cell, and
inhibit phagolysosome formation within the host cello EBs are capable of extra-
cellular existence but require an intracellular location in the appropriate host cell
for metabolism and reproduction.3· 4
Reticulate bodies (RB) are incapable of existence outside the host cell but are
the metabolically active particles in chlamydial infections. RBs are approximately
0.8 /-Lm in diameter with an osmotically sensitive outer membrane. RBs are
responsible for chlamydial replication by binary fission and synthesize their own
proteins and nudeic acids.
The Chlamydia life cyde is unique among microorganisms and complex. EBs
induce phagocytosis after attachment to host epithelial cells. After gaining en-
trance into the host cell by phagocytosis, the EBs inhibit phagolysosome forma-
tion thus assuring their intracellular survival. After several hours, the ingested
EBs migrate to the perinudear area surrounded by the host's endosomal cellular
membrane. The EBs undergo a morphological transformation and are reorga-
nized into RBs after going through a spheroplast-like phase. The RBs then utilize
host ATP to generate high-energy phosphate bonds permitting further growth of
the RBs and providing the energy for binary fission. The RBs contain several
THE CHLAMYDIAL PNEUMONlAS 185
cylindrical surface projections, usually 18, arranged in a hexagonal configura-

tion, which penetrate the endosomal membrane and penetrate into the host
cytoplasm. These tubular structures may be responsible for nutrient or energy
transport between the RB and the host cell cytoplasm. After several hours,
following replication by binary fission, the RBs condense into wh at appear to be
EBs in the inclusion body in the host cell cytoplasm. Cytoplasm inclusion bodies
vary on the basis of species and have been used to identify Chlamydia. For
example, multiple inclusion bodies are characteristic of C. psittaci, whereas C.
traclwmatis usually are associated with single inclusion bodies.
Chlamydia-specific lipopolysaccharide is transported to the surface of the
host cell and is thought to decrease chlamydial plasma protein fluidity, thereby
protecting Chlamydia from attack by cytotoxic T-lymphocytes. It is not certain how
Chlamydia cause death of the host cell, with resultant lysis of the host cell
membrane, but this clearly occurs as the end result of chlamydial cell invasion
following cell death and lysis by internal osmotic pressure, which causes bursting
of the weakened host cell membrane and the release of a variable number of
infected cells. 2-4
Three species of Chlamydia comprise the genus. C. traclwmatis has three
biovars. The lymphogranuloma venereulis (LGV) biovar and the trachoma biovar
affect human squamocolumnar cells and the third biovar is responsible for mouse
pneumonitis. The number of serovars for C. psittaci is unknown, but many
unidentified serotypes exist. C. psittaci causes disease in many animals and is the
cause of psittacosis in humans. C. pneumoniae was initially thought to represent
a C. psittaci variant, but on the basis of DNA hemology studies and a specific
antibody response, it has now been given species status. C. pneumoniae is thought
to infect only humans. All three chlamydial species affect primarily respiratory
epithelium, but cause a variety of other infectious diseases in addition (Table I).
Chlamydia are difficult to grow and isolate in the laboratory. The yolk sac of
embryonated chick eggs and HeLa cell cultures have been used to grow these
organisms. C. pneumoniae is the most fastidious of the chlamydial species and is
the most difficult to pro pagate in self culture. Growth in cell culture may be
enhanced by centrifugation, cycloheximide, or DEAE-dextran. 2 ,3
3. PATHOPHYSIOLOGY
Chlamydia, like viruses, are obligate, intracellular pathogens, and gain en-
trance into the host cell via endocytosis. It is thought that there is a specific
receptor on the host cell surface. After receptor mediator endocytosis, the EBs
localize near the Golgi region of the eukaryotic host cell. In the process, there is
thought to be a restructuring or redistribution of EB coating material which is
believed to thwart endolysosome-lysosome fusion by sending confusing chemical
signals as a result of modifying the coat configurations of the EB.
Chlamydia localize at the bases of microvilli in coated pits. After gaining
entrance into the host cell, as described above, chlamydial organisms are success-
186 BURKE A. CUNHA
TABLE I
Human Chlamydia Species
Iodine
Elementary containing Sulfa
Species Diseases Serotypes bodies inclusions sensitivity
Chlamydia Trachoma 15 Multiple, + +
tracJwmitis LCV, STD's coccoid
Urethritis
Conjunctivitis
Infant pneumonia
Chlamydia Psittacosis Unknown Single,
psittaci Culture negative coccoid
Endocarditis
FUO
Chlamydia Sinusitis Multiple, pear
pneumoniae Pharyngitis shaped
Laryngitis
Bronchitis
Pneumonia
ful obligate intraeellular parasites beeause of their ability to evade intraeellular

host defenses. There is virtually no information on T-eell abnormalities in
ehlamydial infeetions. It may be postulated, however, that beeause relatively few
eells are infeeted by hundreds of EBs liberated from lysed host eells, intraeellular
host defenses limit attaehment and/or infeetion in some way. Lymphokines elabo-
rated from T eells inhibit Chlamydia. Alpha, beta and especially gamma interferon
inhibits Chlamydia intraeellularly. Gamma interferon primarily delays the devel-
opmental eycle at the reorganization and replieation stages within the host eell.
The immunology of these very complex intraeellular pathogens is not weIl
understood and a single infeetion does not confer long-term proteetive immunity.
Beeause of the ineomplete immune response and their intraeellular loeation,
ehlamydial infeetions are, in general, prone to persistenee and relapse. 2-4
Pathologieally, ehlamydial organisms gain entranee to the respiratory traet
presumably via airborne transmission, and initially attaeh and replieate in the
nasopharyngeal mucosa. Some Chlamydia, for example, C. psittaci, may spread via
the bloodstream to loealize and persist in the reticuloendothelial system of the
lungs, liver, and spleen. Chlamydia elieit a moderate, inftammatory response of
mixed eells with a mononuclear eell predominanee. Infeeted respiratory epithe-
lium usually remains intaet, but alveolar eells and interstitial eells of the lung may
show loeal edema, neerosis, and infrequendy may hemorrhage. Relatively few
hUI]1an specimens from patients with ehlamydial pneumonias have been available
for pathologie examination. However, all the Chlamydia that eause pneumonia
have a lymphoeytie interstitial pneumonitis pathologieally.8-1O Chlamydia infeetion
elicits a rise in IgM antibody levels that peak acutely and decrease after approx-
imately one month. IgG antibody responses follow a decrease in IgM levels and
may persist for years following infection. There is a high prevalence of low titer
IgG antibody levels in the general population, and a fourfold increase in serum
titer is necessary for dia gnosis. Tests used to identify Chlamydia indude the
complement fixation test (CF) which uses group specific antigens and is used only
for the dia gnosis of psittacosis. Microimmunofluorescent techniques (IF) measur-
ing IgM responses are specific and available for C. tracJwmatis and C. pneu-
moniae. I1 •12
Chlamydia indusion staining has been used to identify the presence of
Chlamydia from pathological specimens. Initially inclusions were stained by io-
dine, since the inclusions also contain glycogen. However, iodine staining is less
sensitive than Giemsa staining and is primarily useful for C. tracJwmatis infections.
Although squamous cells frequendy have glycogen indusions, glycogen is not
present during the entire life cyde, making inclusion staining techniques useful
during only part of the chlamydiallife cycle. Immunofluorescent techniques are
now used to stain for EBs using fluorescein co~ugated antibody. By electron
microscopy, specific dia gnosis may be made by the characteristic number or shape
of the inclusion bodies. 2-4
4.1. Infant Chlamydial Pneumonia

4.1.1. Clinical Presentation
Infant chlamydial pneumonia is caused by the D-K serotypes of C. tracJw-
matis. This entity was initially described as the eosinophilic pertussoid neonatal
pneumonia. Typically, infant chlamydial pneumonia presents between the 3rd
and 11th week of life. The mother may be symptomatic or asymptomatic and
transmits the chlamydial infection to the child, usually at birth. Coexisting
conjunctivitis may be a clue to the diagnosis. Chlamydial conjunctivitis is usually
unilateral and presents with lid edema and a watery-to-purulent discharge. The
organisms are readily cultured from conjunctival scrapings or by direct fluores-
cent antibody techniques of infected conjunctival secretions. Antecedent con-
junctival infection is not always present but is a helpful clue to the diagnosis when
present. Infants usually become symptomatic by the 8th week of life, and usually
present with litde or no fever. Upper respiratory tract congestion without a nasal
discharge is a frequent finding, as is concomitant otitis media in some series.
Patients with pneumonitis are tachypneic and have the characteristic pertussoid
cough. Bilateral rales may be heard but auscultatory findings of the lung are
usually not remarkable. Expiratory wheezes are not characteristic of infant
chlamydial pneumonia. Chest X ray shows bilateral interstitial infiltrates with
variable degrees of hyperinflation of the lungs. The infiltrates are usually bilat-
eral, symmetrieal, and perihilar. Occasionally focal consolidations are seen and
188 BURKE A. CUNHA
when present, these wedge-shaped infiltrates are characteristic of infant chla-

mydial pneumonia. 13- 18
4.1.2. Laboratory Tests

Nonspecific laboratory findings indude polydonal hypergammaglobulin-
emia found on serum protein electrophoresis. Low-grade peripheral eosinophilia
is an important diagnostic due to the presence of chlamydial pneumonia in
infants. Tracheal secretions stained for eosinophils show a characteristic eosino-
philic exudate, which provides a rapid technique to arrive at a presumptive
diagnosis. The peripheral white blood cell count is usually not elevated. Hypox-
emia by arterial blood gas determination is variable but persists even after the
acute phase of the infection. 7 ,8
4.1.3. Diagnosis
The organism may be cultured from tracheal aspirates or nasopharyngeal
aspirates and these secretions may be stained by DFA techniques for a rapid
diagnosis. Serological diagnosis is by demonstrating a single mlcrOlmmuno-
fluorescent titer ;;:'1:32 of IgM antibody to C. traclwmatis. 2 ,3
4.1.4. Differential Diagnosis

Differential diagnosis of infant chlamydial pneumonia indudes other dis-
eases that present with proximal cough, for example, adenovirus, respiratory
syncytial virus, or parainfluenza virus. Usually the main differential diagnosis in
infants is between respiratory syncytial virus, which is common in the winter
months and is frequently associated with expiratory wheezing. In contrast, infant
chlamydial pneumonia has no seasonal distribution and is associated with periph-
eral eosinophilia and is not characterized by expiratory wheezing. Usually a
dinical diagnosis is possible because of the characteristic pertussoid cough in a
neonate who is afebrile and has low grade peripheral eosinophilia. In compro-
mised hosts, the chest X ray may resemble Pneumocystis carinii pneumonia or
cytomegalovirus pneumonitis. 7 ,8
4.1.5. Treatment
Infant chlamydial pneumonia is usually self limiting and nonfatal. Some
infants, however, require ventilatory support during the acute phase of severe
illness. Therapy is with erythromycin or sulfonamides for 2-3 weeks. Patients
treated do better and their symptoms res pond more quickly than those that are
untreated. Coexisting chlamydial co~unctivitis should be treated topically as weil
as systemically. Treatment of antecedent chlamydial conjunctivitis has been
shown not to prevent chlamydial pneumonitis. Since C. traclwmatis has been
isolated from gastric aspirates of newborns, it is thought that most infections are
acquired du ring the birthing process and are not secondary to conjunctivitis. 10
4.1.6. Complications
Untreated infant chlamydial pneumonia persists for weeks or months until

the infection clears. The range of illness is 24-61 days with an average of 43 days.
There is a higher than expected incidence of asthma in childhood in patients
who had neonatal chlamydial pneumonia. 3- 5
4.2. Psittacosis
Psittacosis caused by C. psittaci is a systemic illness characterized by promi-

nent pulmonary manifestations. Psittacosis affects psittacine birds and humans,
but human infection results from a wide variety of non-psittacine birds. Psitta-
cosis frequently occurs in those working with or handling birds or bird products.
Chlamydia is present in bird nasal secretions and excreta as weIl as on the feathers
of infected birds. Siek birds are more likely than weIl-appearing birds to transmit
the illness. 9
After a 1-2-week incubation period, psittacosis presents as an acute febrile
pneumonitis with temperatures increasing over several days. A nonproductive
dry cough is characteristic and is usually associated with a severe headache
and prominent myalgias. Epistaxis and photophobia are frequent early com-
plaints and provide an important clue to the diagnosis of psittacosis. Patients may
also complain of a mild sore throat or nausea, vomiting, or diarrhea in some
cases. 3
Changes in mental status such as mild encephalopathy, are present in some
cases. Back and neck muscle stiffness frequently accompany myalgias. On physi-
cal examination the patient may have a facial rash resembling the rose spots of
typhoid fever, in effect, Horder's spots. An important clue in the diagnosis is
relative bradycardia and splenomegaly. Splenomegaly occurs in 10-70% of cases,
and in a patient with pneumonia, should suggest psittacosis. Nontender hepato-
megaly also may be present. Examination of the lungs is usually unremarkable
but signs of consolidation may occasionally be present mimicking a bacterial
pneumonia. 9
Psittacosis may present as an upper respiratory illness without pneumonitis,
as culture-negative endocarditis, or as a fever of unknown origin.I°
The peripheral white blood cell count is normal with an unremarkable

differential cell count. Erythrocyte sedimentation rate (ESR) is usually normal,
liver function tests are frequently slightly abnormal, the ehest X ray is not
characteristic and may show little in the way of infiltrates or bilateral interstitial
infiltrates which are the usual pattern, or there may be signs of lobar consolida-
tion. 9 ,10
190 BURKE A. CUNHA
4.2.3. Diagnosis
Diagnosis is by isolation or by complement fixation (CF) antibody titers.

Acute and convalescent specimens should be obtained to document a fourfold
increase in CF titers. There is group specific cross-reactivity between C. psittaci
and C. pneumoniae as well as C. traclwmatis, however, cross-reactions to the micro-
immunoftuorescent test are less common. In any event, the titer rise to C. psittaci
should be of a much greater magnitude as to permit differentiation from its
serologically related elose relatives C. traclwmatis and C. pneumoniae. Cytoplasmic
glycogen inelusion staining is negative because C. pneumoniae infections are not
characterized by glycogen inelusions. 2,3
4.2.4. Differential Diagnosis
Psittacosis needs to be differentiated from bacterial pneumonia and from

other atypical pneumonias. The patient with psittacosis presenting with lobar
consolidation may elosely mimic a bacterial pneumonia. The best way to differen-
tiate these two entities is to utilize the extrapulmonary manifestations of psitta-
cosis to differentiate it from the common bacterially caused community-acquired
pneumonias. Community-acquired bacterial pneumonias are not associated with
pulse temperature deficits or splenomegaly. These two findings alone would be
adequate to rule out common community-acquired bacterial pathogens as a cause
of the patient's X ray findings. Furthermore, patients with bacterial pneumonias
do not usually have the prominent headache, myalgias and photophobia, or
epistaxis which are common early features of psittacosis. Finally, bacterial patho-
gens can be isolated from sputum or blood cultures, which is not the case and
should not be attempted with C. psittaci. 1O ,19
It is more difficult to tell psittacosis from the other causes of atypical
pneumonia, for example, Mycoplasma pneumonia, Legionnaire's disease, Q fever,
and tularemia. The most frequent differential diagnosis will be between Legion-
naires' disease, Q fever, and occasionally tularemic pneumonia. Legionnaires'
disease and Q fever are also associated with pulse temperature deficits, such as
relative bradycardia, whereas Mycoplasma and tularemic pneumonia are not.
Abnormalliver function tests are more frequent in Legionnaires' disease but also
occur in psittacosis as well as Q fever. Splenomegaly, if present, strongly suggests
the diagnosis of psittacosis, but may occur rarely in patients with Q fever.
Diarrhea is more prominent in Mycoplasma and Legionnaires' disease, but does
occur with psittacosis. The peripheral white blood cell count is normal in the
other causes of atypical pneumonia. H ypophosphatemia, if present, immediately
points to the dia gnosis of Legionnaires' disease in the absence of other causes of
hypophosphatemia in patients with pneumonia. The presence of pharyngitis also
limits the differential diagnosis to psittacosis and Mycoplasma pneumonia because
sore throats are not common with the other causes of nonchlamydial pneumonias.
In terms of differentiated psittacosis from C. pneumoniae, the presence of sinusitis,
bronchitis, or laryngitis with an atypical pneumonia strongly favor diagnosis of C.
pneumoniae. These features are not present in patients with psittacosis. Also, a
pulse temperature deficit occurs with psittacosis and is not a feature of C.
pneumoniae pneumonia.
Therefore, in summary, a presumptive diagnosis of psittacosis may confi-
dently be made on the basis of occupational exposure to birds as weIl as the
characteristic clinical picture. It may be differentiated easily from bacterial
pneumonias and less easily with respect to the other causes of atypical pneu-
monias such as Legionnaires' disease, Q fever, and so forth. IO ,19
4.2.5. Treatment
The treatment of psittacosis is usually with a tetracycline for 1-4 weeks. In
patients unable to tolerate tetracycline, or in children as weIl as pregnant mothers,
erythromycin provides alterative therapy. Since patients with psittacosis res pond
better to tetracycline than erythromycin, a tetracycline analog such as doxy-
cycline, is usually preferred over erythromycin for treatment. 2•6
Psittacosis usually resolves without residual lung dysfunction. During the
convalescent phase of psittacosis, the patient's course may be complicated by
pulmonary infarction or obscure thrombophlebitis. 9
4.3. Chlamydia pneumoniae Pneumonia

C. pneumoniae was formerly known as TWAR strain of Chlamydia because it
could not be grouped with either C. trachomatis or C. psittaci. C. pneumoniae
appears to occur only in humans and is biologically and clinically different than
C. psittaci. C. pneumoniae is a disease of adolescence and young adults but may be
found in aB age groups. Transmission presumably occurs via droplet secretion
wh ich results in pneumonia in a minority of cases. Most infections are clinicaBy
inapparent and occur in childhood. Reinfection is not uncommon in later
life. 20-23

Patients with C. pneumoniae may present with an upper respiratory infection
that precedes pneumonia, or pneumonia may be the sole manifestation of
infection. Patients presenting with upper respiratory tract disease usually present
with sinusitis or bronchitis. A minority of patients present with primary phar-
yngitis. In the patients presenting with upper respiratory symptoms and subse-
quent pneumonitis, the upper respiratory tract involvement provides an impor-
tant clue as to the etiology of the patient's lower respiratory tract infection. For
example, in patients with C. pneumoniae pneumonia, there is frequently anteced-
ent or coexisting sinusitis, pharyngitis, or laryngitis. A sore throat is unimpressive
192 BURKE A. CUNHA
on physical examination but prominent in terms of the patient's sym ptomatology,

and is an important clue to C. pneumoniae infection. There is a paucity of physical
findings in the pulmonary examination in patients with C. pneumoniae. Extra-
pulmonary findings such as rash, are not part of the illness. lO Typically, the patient
presents with a Mycoplasma-like illness, in effect, nonproductive, dry hacking
cough which has persisted for weeks in spite of apparently adequate erythro-
mycin therapy. Diarrhea and ear findings are less common in patients with C.
pneumoniae pneumonia than in patients with Mycoplasma pneumonia. Tempera-
ture tends to be lower than with Mycoplasma pneumonia and there is no pulse
temperature deficit as is the case with Legionnaires' disease. C. pneumoniae occurs
in older adults with chronic obstructive pulmonary disease (COPD) or heart
failure and may be associated with coronary artery disease. 24-27

Laboratory findings in C. pneumoniae infection are unremarkable. The sedi-
mentation rate is minimally elevated and the white blood cell count is usually
unhelpful. lO Chest X ray frequently shows single subsegmentallesions. Consolida-
tion and pleural effusion are rare, as are bilateral infiltrates. Elevated cold
agglutinin titers and elevated liver function tests are not features of C. pneumoniae
pneumonia. 19 ,28
4.3 .3. Diagnosis

Diagnosis is by single elevated C. pneumoniae specific microimmunofluores-
cent (MIF) IgM titer ~1:32 or an IgG titerof~512. Serological responses may not
be evident for at least four weeks and may be delayed, blunted, or eliminated
entirely by early initiation of appropriate antimicrobial therapy. Nasopharyngeal
cultures are preferred over throat cultures for growing the organism. C. pneu-
moniae grows readily on HL and HEp-2 celliines in tissue culture. The diagnosis is
usually made serologically.29,30
4.3 .4. Differential Diagnosis
C. pneumoniae pneumonia should not be confused with bacterial pneu-

monias, and other atypical pathogens increase the possibility of diagnostic confu-
sion. Of the atypical pneumonias, C. pneumoniae most closely resembles Myco-
plasma pneumonia. There are enough distinguishing features, however, to permit
an accurate presumptive diagnosis on clinical grounds alone. 19 Because many
patients will be misdiagnosed as Mycoplasma pneumonia, the clinician should
suspect the diagnosis in such patients who have failed to res pond to an adequate
trial of erythromycin therapy.25 Although the age group of both patients with
Mycoplasma and C. pneumoniae are the same, there are important clues that may be
discerned from the pattern of organ involvement. For example, hoarseness is only
rarelya feature of Mycoplasma pharyngitis but is a common accompaniment of
C. pneumoniae infections of the pharynx. This is an extremely important find-

ing. 31 Ear involvement characterizes Mycoplasma pneumonia but it rarely, if ever,
occurs with C. pneumoniae infection. Sinusitis or bronchitis favor the diagnosis of
C. pneumoniae over a Mycoplasma pneumonia. Diarrhea and gastrointestinal com-
plaints are distinctly more common with Mycoplasma than with C. pneumoniae. 28
Mycoplasma may be associated with a variety of extrapulmonary manifestations,
for example, meningoencephalitis, cardiac or neural involvement, skin involve-
ment, and so forth, which are not features of C. pneumoniae. C. pneumoniae has
been associated recently with myocarditis in some cases, however. On chest X ray
the two diseases may be indistinguishable and therefore X rays should not be
relied on to differentiate between these two disease processes. Elevation of cold
agglutinins does not occur in all cases of Mycoplasma pneumonia but is helpful
especially when the cold agglutinin titers are highly elevated, for instance ~1:64.
Cold. agglutinins are not a feature of C. pneumoniae pneumonia, therefore the
clinician should be able to readily differentiate these two entities on clinical
grounds as weIl as the lack of responsiveness of patients with Mycoplasma pneu-
monia to erythromycin therapy28,30 (Table 11).
4.3.5. Therapy
Optimal treatment for C. pneumoniae pneumonia has not yet been deter-
mined, however, most experts suggest tetracycline therapy for a 2-4 week period.
Tetracyclines, such as doxycycline or minocycline, are preferred over erythro-
mycin derivatives because erythromycins are clinically clearly less effective than
tetracyclines. Macrolides would be preferred in children or pregnant mothers
who need to be treated for C. pneumoniae infection. 30
Recently, it has been suggested that C. pneumoniae may result in chronic

sinusitis or bronchitis following initial infection. Furthermore, C. pneumoniae has
been implicated in the etiology of coronary artery disease and possibly sarcoid-
osis. Recurrent pneumonia with a different pathogen or relapse with the same
organism is a common complication of C. pneumoniae pneumonia. 32- 36
5. SUMMARY
The chlamydial pneumonias caused by C. traclwmatis, C. psittaci, and C.

pneumoniae are distinct clinical entities with different patterns of organ involve-
ment, which permit an accurate presumptive clinical diagnosis. This is fortunate
because all of these pathogens are difficult to grow in tissue culture and do not
grow on ordinary media. The diagnosis is usually by direct immunoßuorescence
technique or by serological methods. Fortunately, macrolides or tetracyclines
provide adequate therapy for all of the chlamydial pneumonias. The ultrastruc-
194 BURKE A. CUNHA
TABLE 11
Differential Diagnostic Features of Chlamydial Pneumonias
c. pneumoniae
Key characteristics C. psittaci C. trachomatis (TWAR strain)
Symptoms
Mental status changes +
Prominent headache +
Meningismus ±
Myalgias +
Nonproductive cough +a ± +
Photophobia +
Nasal congestion +
Nausea/vomiting/diarrhea ±
High fever + ±
Signs
Relative bradycardia +
Rash ±
(Horder's spots)
Conjunctivitis +
Epistaxis +
Otitis +
Sinusitis +
Nonexudative pharyngitis ± +
Cervical adenopathy ±
Laryngitis +
Lobar consolidation ±
Cardiac involvement ±b
Pericarditis
Myocarditis
Splenomegaly + ,
Chest film infiltrate No infiltrate or Perihilarlinter- Small single
patchy infiltratel stitial infiltrates segmental
consolidation infiltrate'
Pleural effusion ±
Laboratory abnormalities
WBC count N N N/j
Eosinophilia +
Increase in SGOT/SGPT ±
Therapy: Response to
Erythromycin ± +
Sulfonamides +
Tetracycline/doxycycline + + +
Complications Thrombophlebitis Asthma Chronic sinusitis
Pulmonary Chronic bronchitis
infarction
aMay have bloody sputum.
hMay present as culture negative endocarditis.
'M uhiple infiltrates may occur in the e1derly.
ture and reproductive cyde of Chlamydia are unique in the biological world.
Advances in the basic sciences will provide further understanding into the basis
for the dinical manifestations that are appreciated by physicians caring for
patients infected with these fascinating obligate intracellular organisms.
REFERENCES
1. Schachter,j., Lum, L., Gooding, C. A., and Ostler, B., 1975, Pneumonitis following inclusion
blennorrhea,j Pediatr. 87:779-780.
2. Schachter,j., and Dawson, C. R. (eds.), 1978, Human chlamydial infections, PSG Publishing
Company, Littleton, Massachusetts.
3. Schachter, j. and Grossman, M., 1981, Chlamydial infections, Annu. Rev. Med. 32:45-61.
4. Fraiz, j. and Jones, R. B., 1988, Chlamydial infections, Annu. Rev. Med. 39:357-370.
5. Ridgeway, G. L., 1986, Chlamydial infection in man, Postgrad. Med. 62:249-253.
6. Cunha, B. A., 1987, Chlamydial infections in man, Infect. Dis. Prac. 10:1-12.
7. Remington, j. S. and Klein, j. 0. (eds.), 1990, Infectious Diseases of the Fetus and Newborn Infant,
3rd ed., W B. Saunders Company, Philadelphia.
8. Feigin, R. 0. and Cherry, j. D. (eds.), 1987, Textbook of Pediatric Infectious Diseases, Volumes I
and 11, 2nd ed., W B. Saunders Company, Philadelphia.
9. Wilson, j. 0., Braunwald, E., Isselbacher, K. j., Petersdorf, R. G., Wilson, j. P., Martin, j. B.,
and Fauci, A. S. (eds.), 1991, Harrison's Principles of Internal Medicine, 12th ed., McGraw-Hill,
New York.
10. Gorbach, S. L., Bartlett, j. G., and Blacklow, N. R. (eds.), 1992, Infectious Diseases, W B.
Saunders Company, Philadelphia.
11. Wang, S-P., Grayston, j. T., Alexander, E. R., and Holmes, K. K., 1975, Simplified micro-
immunofluorescence tests with trachoma-Iymphogranuloma venereum (Chlamydia tra-
chomatis) antigens for use as a screening test for antibody,j Clin. Microbiol. 1:250,255.
12. Paisley, j.W, Lauer, B. A., Melinkovich, P., Gitterman, B. A., Feiten, 0. j., and Berman, S.,
1986, Rapid diagnosis of Chlamydia trachomatis pneumonia in infants by direct immuno-
ftuorescence microscopy of nasopharyngeal secretions, j Pediatr. 109:653-655.
13. Tipple, M., Beem, M. 0., and Saxon, E., 1979, Clinical characteristics of the afebrile
pneumonia associated with Chlamydia trachomatis infection in infants less than 6 months of
age, Pediatrics 63:192-197.
14. Arth, C., Von Schmidt, B., Grossman, M., and Schachter,j., 1978, Chlamydial pneumonitis,
j Pediatr. 93:447-449.
15. Beem, M. O. and Saxon, E. M., 1977, Respiratory tract colonization and a distinctive
pneumonia syndrome in infants infected with Chlamydia trachomatis, N. Engl. j Med.
296:306-310.
16. Brasfield, 0. M., Stagno, S., Whitley, R. j., Clouda, G., Cassella, G., and Tiller, R. E., 1987,
Infant pneumonitis associated with cytomegalovirus, Chlamydia, Pneumocystis, and Urea-
plasma: follow-up, Pediatrics 79:76-83.
17. FrommelI, G. T., Bruhn, F. W, and Schwartzman, j. 0., 1977, Isolation of Chlamydia
trachomatis from infant lung tissue, N. Engl. j Med. 296:1150-1152.
18. Harrison, H. R., English, M. G., and Lee, C. K., 1978, Chlamydia trachomatis infant pneu-
monitis: Comparison with matched controls and other infant pneumonitis, N. Engl. j Med.
298:702-708.
19. Johnson, 0. H., and Cunha, B. A., 1993, Atypical pneumonias, Postgrad Med. 93:69-82.
20. Grayston,j. T., Kuo, C-C., Wang, S. P., and Altman, T., 1986, A new Chlamydia psittaci strain,
TWAR, isolated in acute respiratory tract infections, N. Engl. j Med. 315:161.
196 BURKE A. CUNHA
21. Grayston, J. T., 1989, Chlamydia pneumoniae, strain TWAR, Chest 95:664-669.
22. Grayston,J. T., Campbell, L. A., Kuo, C-C., Mordhorst, C. H., and Saikku, P., 1990, A new
respiratory tract pathogen: Chlamydia pneumoniae strain TWAR,j Infect. Dis. 161:618-625.
23. Aldous, M. B., Grayston, J. T., Wang, S-P., and Foy, H. M., 1992, Seroepidemiology of
Chlamydia pneumoniae TWAR infection in Seattle families, 1966-1979, j Infect. Dis. 166:
646-649.
24. Thorn, D. H., Grayston,J. T., Siscovick, 0. S., Wang, S-P., Weiss, N. S., and Daling,J. R., 1992,
Association of prior infection with Chlamydia pneumoniae and angiographically demon-
strated coronary artery disease,JAMA 268:68-72.
25. Saikku, P., Leinonen, M., Tenkanen, L., Linnanmaki, E., Ekman, M-R., Manninen, v.,
Manttari, M., Frick, M. H., and Huttunen, J. K., 1992, Chronic Chlamydia pneumoniae
infection as a risk factor for coronary heart disease in the Helsinki heart study, Ann. Intern.
Med. 116:273-278.
26. Kuo, C. C., Shor, A., Campbell, L. A., et al., 1993, Demonstration of Chlamydia pneumoniae in
atherosclerotic lesions of coronary arteries. j Infect. Dis. 167:841-849.
27. Beaty, C. 0., Grayston, J. T., Wang, S-P., Kuo, C-C., Reto, C. S., and Martin, T. R., 1991,
Chlamydia pneumoniae, strain TWAR, infection in patients with chronic obstructive pulmon-
ary disease, Am. Rev. Respir. Dis. 144:1408-1410.
28. Atmar, R. L. and Greenberg, S. B., 1989, Pneumonia caused by Mycoplasma pneumoniae and
the TWAR agent, Semin. Respir. Ther. 4:19-31.
29. Marrie, T. J., Grayston, J. T., Wang, S-P., and Kuo, C. C., 1987, Pneumonia associated with
the TWAR strain of Chlamydia, Ann. Intern. Med. 106:507-511.
30. Grayston, J. T., 1992, Chlamydia pneumoniae, strain TWAR pneumonia, Annu. Rev. Med.
43:317 -323.
31. Kazuhiro, H., Ogawa, H., and Kazuyama, Y., 1992, Seroprevalence of Chlamydia pneumoniae
infections in otolaryngeal diseases, j Laryngol. Otol. 106:208-210.
32. Jones, R. B., Priest, J. B., and Kuo, C. C., 1982, Subacute chlamydia endocarditis, JAMA
247:655-658,.
33. Marrie, T. J., Harczy, M., Mann, O. E., Landymore, R. W, Raza, A., Wang, S. P., and
Grayston,J. T., 1990, Culture-negative endocarditis probably due to Chlamydia pneumoniae,j
Infect. Dis. 161:127-129.
34. Gronhagen-Riska, C., Saikku, P., Riska, H., Froseth, B., and Grayston,J. T., 1988, Antibodies
to TWAR-a novel type of chlamydia-in sarcoidosis, in: Sarcoidosis and Other Granu-
lomatous Disorders, (C. Grassi, C. Rizzato, and G. E. Pozzi, eds.) Elsevier Science Publishers,
Amsterdam, pp. 297-301.
35. Hammerschlag, M. R., Chirgwin, K., Roblin, P. M., GeIling, M., Dumornay, W, Mandel, L.,
Smith, P., and Schachter, J., 1992, Persistent infection with Chlamydia pneumoniae following
acute respiratory iIIness, Clin. Infect. Dis. 14:178-182.
36. Hammerschlag, M. R., 1994, Chlamydia pneumoniae infections, Infections in Medicine 11:
64-70.
12
H istoplasma capsulatum
STANLEY W CHAPMAN and
HAROLD M. HENDERSON
1. INTRODUCfION
Histoplasmosis, the disease eaused by inhalation of the dimorphie fungus Histo-

plasma capsulatum, is the most eommon systemie fungal infeetion in the United
States. Most infeetions with H. capsulatum are asymptomatic. In the normal host,
symptomatie acute pulmonary histoplasmosis may occur after the inhalation of a
large inoculum of infeetious mieroconidia. In the abnormal host, clinieal in fee-
tion may take the form of ehronie pulmonary diseases or progressive dissemi-
nated illness.
2. HISTORY
H. capsulatum was first diseovered by Samuel Darling in 1905. 1 While examin-

ing autopsy speeimens from a patient with disseminated disease, he identified
what he presumed to be a new eneapsulated protozoan inside histioeytes, henee
the name Histoplasma capsulatum. In 1934 DeMonbreun isolated the organism
from an infeeted infant and demonstrated it to be a dimorphie fungus. 2 H.
capsulatum was originally thought to eause only disseminated disease but autopsy
and clinieal studies subsequently doeumented a ehronie pulmonary form of
disease resembling tubereulosis. 3 ,4 In 1945 and 1946, epidemiologie studies of
histoplasmin skin-test reaetivity defined the endemie areas of infection in the
United States. 5 ,6 Contrary to previous reports, histoplasmosis proved to be a
STANLEY W. CHAPMAN and HAROLD M. HENDERSON • Division of Infectious Diseases,

University of Mississippi Medical Center, Jackson, Mississippi 39216-4505.
Pulmonary Inftctions and Immunily, edited by Herman Chmel el al. Plenum Press, New York, 1994.
197
198 STANLEY W CHAPMAN and HAROLD M. HENDERSON
common, usually asymptomatic infection. Aseries of subsequent environmental

isolations identified soil as the natural reservoir of H. capsulatum. 7,8
3. MYCOLOGY
Histoplasma capsulatum is the imperfect state of Ajellomyces capsulatus. 9 ,10 It is

a dimorphie fungus that grows in mycelial form at room temperature and as a
yeast at 37°C. The mycelial form grows on agar as ftuffy white or buff-colored
colonies. Hyphae are septate and branching, producing both microconidia and
tuberculate macroconidia. The conidia are easily airborne when contaminated
soil is disturbed.
Conversion to the yeast form at 37°C in culture requires 7 to 14 days, but
transformation in tissue is more rapid. Yeast colonies are usually beige or cream-
colored and may be smooth or wrinkled. The uninucleate yeast cells are oval, have
thin walls and measure 2 X 2 to 3 X 3 jl.m. Reproduction is by thin-based, polar
budding. When seen in infected tissue, the characteristic yeast are found almost
exclusively intracellularly within macrophages.
A stable variant, H. capsulatum var. dubosii, is found in central Africa. 9 It
differs from classical H. capsulatum by producing larger yeast forms (i.e., 7-15
jl.m) in culture and in tissue. Clinical disease, commonly referred to as African
histoplasmosis, is more likely to involve skin and bone.
4. HOST DEFENSE
The primary host defense against H. capsulatum is cell-mediated immunity,

and is dependent on a cooperative interaction of T lymphocytes and macro-
phages. 9 ,1l The importance of cellular immunity is supported by severallines of
evidence in murine models of infection. First, nude mice have an enhanced
susceptibility to experimental infection with Histoplasma.l 2 Second, T lympho-
cytes from mice injected with a sublethai inoculum of H. capsulatum secrete a
factor that enables macrophages to inhibit intracellular growth of the organism. 13
Third, CD4 + T cells from mice immunized with H. capsulatum transfer protective
immunity to naive mice.l 4 Fourth, 100% of mice depleted of CD4 + T cells die
after sublethai challenge with H. capsulatum, while infected controls survive. 15 In
addition, cloned populations of CD4 + T cells that recognize Histoplasma antigens
have been isolated from humans with prior infection by H. capsulatum. 16
Once inhaled into the lungs, the conidia of H. capsulatum convert into yeast
forms after 2 or 3 days and are rapidly phagocytized by alveolar macrophages. 9 ,1l
In the nonimmune host, the yeast forms continue to proliferate intracellularly,
with more macrophages and lymphocytes being recruited to the site of infection.
The characteristic epithelioid granuloma, often with multinucleated giant cells,
eventually develops. During these early stages infected macrophages migrate to
the mediastinallymph nodes and may disseminate to organs of the reticuloendo-
HISTOPLASMA CAPSULATUM 199
thelial system (i.e., liver, spleen, and bone marrow). Cellular immunity develops
one to three weeks after infection, activating macrophages to kill intracellular
organisms. Central caseous necrosis may then occur in the granuloma, along with
surrounding fibrosis and calcification.
In the immune host the responses noted above limit the replication of
intracellular organisms within the first few days and control the infection. A very
heavy inoculum, however, may result in symptomatic illness in immune individ-
uals. In comparison to infection of nonimmune patients, the incubation period is
shorter (less than 5 days), and the illness is milder and lasts a shorter period of
time. Acquired immunity and skin test reactivity wane over time unless reinfee-
tion occurs.
In some patients cellular immunity never develops and the organism con-
tinues to proliferate inside macrophages. There is a progressive parasitization of
the reticuloendothelial system which results in the clinical syndrome of dis sem i-
nated histoplasmosisP Persons at greater risk of disseminated disease include
those at the extremes of age, and immunosuppressed patients such as those with
hematologic malignancies, trans plant recipients, and individuals treated with
corticosteroids or cytotoxic agents.l8-20 Patients with AIDS have a severe defi-
ciency of cell-mediated immunity and are especially prone to develop dissemi-
nated infection with H. capsulatum. 21- 23 In 10-20% of patients with disseminated
infection, no underlying disease or defect in cellular immunity can be identified.
It is speculated that in such cases the infection itself may cause a transient
immunosuppression.1 7
5. EPIDEMIOLOGY
Histoplasma capsulatum is distributed worldwide. Evidence for its existence

has been found in nearly all river valleys in temperate and tropical zones of the
world, including Europe, Africa, Asia, Central America, and South America. In
the United States most cases of histoplasmosis occur in the Mississippi and Ohio
River valleys. These areas of the country have been shown by skin-test surveys
to be highly endemie for infection with the fungus (Fig. 1).
H. capsulatum exists in nature as a saprophytic mold, and is found in soil
under certain conditions that promote growth of the fungus. These conditions are
not completely understood but are associated with the low altitude and high
humidity found in river valleys.24 Soil enriched by bird droppings has long been
associated with the growth of H. capsulatum. This was first demonstrated for
chickens,8 and has since been well-documented for starlings, blackbirds, grackles,
pigeons, and other birds as well. The organism has also been isolated from bats
and bat guano. 25 The requirement of the fungus for specific conditions that
promote its growth leads to a patchy distribution in the environment, even in
highly endemie areas. This has given rise to the concept of microfoci of infection.
Caves, attics and old buildings frequented by bats, starling and blackbird roosts,
chicken houses, old equipment contaminated by bird or bat droppings, and
Nl
o
o
HISTOPLASMIN
[Jl
>i
zt"'
t"'I
><
~
C"l
::r::
~
"C
s::~
Z
I'l
::l
0-
PERCENTAGE ::r::
• REACTlON5 4 + MM ~
:-- c 5 _40 - 59 ~
::= 5 - 9 _60 - 79 275,558 NAVY RECRUIT5 t"'
t:)
GEl0 - 19 _80+ 1958 -1965
_20 -39 ~
whit. males 17-21 y.ars ::r::
~"..,.. U.S,N:l11C HeGI'" S~ t"'I
lIfiTIMI ONI COUNn IlSIOINTS Z
FIGURE 1. Endemie areas for infeetion with H. capsulatum in the United States as shown by skin testing of naval ~
reeruits. (Reprodueed by permission from Edwards et al. 23a ) ~
~
canebreaks that serve as bird roosts are aB examples of microfoci. Although

originaBy detected in rural settings, microfoci have been found in urban locales
as well. Epidemics of acute pulmonary histoplasmosis occur if susceptible individ-
uals are subjected to heavy exposures of aerosolized spores when a microfocus is
disturbed. 26-30
The clinical manifestations of infection with Histoplasma capsulatum are

dependent on the complex interaction between the organism and the host's
immune system. Inhalational exposure to a very large inoculum frequently
causes symptomatic iBness even in persons with prior immunity. Conversely, a
light exposure is asymptomatic in aB but the most immunocompromised indi-
viduals.
6.1. Acute Pulmonary Histoplasmosis

The sudden onset of a ftu-like illness an average of 10-16 days foBowing an
exposure is characteristic of acute pulmonary histoplasmosis. Incubation periods
ranging from 3 to 24 days have been reported. 31 Fever and weakness are present
in more than 90% of patients, whereas headache, myalgias, cough, and ehest pain
occur in 60_80%.26.27.29,30,32-34 Cough usuaBy appears 2-3 days after the onset
of other symptoms, and is mild and nonproductive. Chest pain is most often
described as a retrosternal ache or discomfort, but on occasion may be pleuritic.
Weight loss may occur with more prolonged illness. Erythema nodosum, arthral-
gias, or arthritis may be seen in 5-10% of patients. 35 ,36 U ncommon complications
include tracheal, bronchial, or esophageal compression by enlarged lymph nodes,
broncholithiasis, and the formation of fistulas within the mediastinum.
Roentgenograms in patients who are asymptomatic foBowing exposure to H.
capsulatum usuaBy show no abnormalities. A patchy pneumonitis with hilar or
mediastinal lymph node enlargement not uncommonly occurs after mild-to-
moderate exposures in hoth symptomatic and asymptomatic individuals (Fig. 2).
Patients with heavy inhalational exposures and symptomatic illness generaBy
have bilateral, patchy, nodular infiltrates, and mediastinal or hilar lymphadenop-
athy (Fig. 3). Pleural effusion is uncommon. Routine laboratory studies are
usually unremarkable, with mild anemia or leukocytosis occurring in less than
25% of cases. Physical findings are generally minimal, with hepatic or splenie
enlargement present on occasion. Most cases resolve in 7-10 days without specific
antifungal therapy.
6.2. Chronic Pulmonary Histoplasmosis

From 2-10% of patients with symptomatic pulmonary histoplasmosis pres-
ent with chronic symptoms of fever, fatigue, productive cough, weight loss of
~
...,'"
S5
~
~
Q
~
~
...,
~
FIGURE 3. Acute pulmonary histoplasmosis with hypoxemia caused by heavy exposure after clearing a canebreak
used as a roost by blackbirds. (Courtesy of Dr. James E. Griffith.) Nl
o
""
several weeks duration, and apical pulmonary infiltrates (Fig. 4). In some affected
persons these signs and symptoms resolve spontaneously, but in many, symptoms
persist and parenchymal disease proceeds over the ensuing months to form single
or multiple thick-walled cavities (Fig. 5).37,38 In the most severe cases these
changes are progressive and pulmonary insufficiency is the ultimate outcome.
Reinfection in an immune subject with structural defects of the lung (i.e., bullous
and centrilobular emphysema) allowing colonization by H. capsulatum has been
implicated in the pathogenesis of chronic pulmonary histoplasmosis. Male sex,
age greater than 40 years, and preexisting chronic lung disease are risk factors
associated with this syndrome.38-40
6.3. Mediastinal Granuloma

In some patients mediastinallymph nodes containing caseous material may
coalesce with the formation of a single surrounding capsule. Mediastinal granu-
lomas are often asymptomatic, particularly if the capsule is no more than 2-5
mm in thickness, and appear as mediastinal masses on routine chest roentgeno-
gram. Granulomas with thicker capsules (6-9 mm) more commonly cause symp-
toms from esophageal compression, bronchial obstruction, or the superior vena
cava syndrome. 41 In selected symptomatic individuals surgical excision of the
mass may be of benefit.
6.4. Mediastinal Fibrosis

On rare occasions, an exhuberant, very thick (1 cm or more) fibrotic capsule
will form in the mediastinal perinodal region, with actual invasion or com pression
of adjacent structures in the mediastinum. Mediastinal fibrosis is a late sequela,
typically occurring years after the initial infection, and tends to be slowly but
relentlessly progressive. Patients usually present with cough, dyspnea, hemopty-
sis, and less commonly pleuritic chest pain. 42 Chest roentgenograms and CT seans
reveal mediastinal widening or mass lesions. It has been proposed that mediasti-
nal fibrosis represents an idiosyneratie, exeessive fibrogenie response to persisting
antigenie material within easeous lymph nodes. 41 ,42 The literature does not
support the hypothesis that mediastinal granuloma represents an earlier stage on
a continuum to mediastinal fibrosis. Thus, reseetion of mediastinal granuloma to
prevent the development of the more serious eondition is not indieated. 42 ,43
Surgieal treatment of mediastinal fibrosis is associated with high complieation
rates and has been disappointing.
6.5. Pericarditis
Pericarditis has been documented in up to 6% of patients following an
outbreak of aeute pulmonary disease. 44 Fever, ehest pain; and a history of an
upper respiratory traet infeetion 2-6 weeks prior to presentation are typical
findings. A pericardial frietion rub is heard in 75-90% of eases, and ehest
FIGURE 5. Single right upper lobe cavity of progressive chronic pulmonary histoplasmosis.
roentgenograms usually demonstrate an enlarged cardiac shadow, mediastinal

adenopathy, or pulmonary infiltrates. In contrast to acute pulmonary infection,
pleural effusion is present in 50% of patients. 44 .45 Cultures and histopathologie
examination of the pericardium and pericardial fluid almost always are negative
for H. capsulatum. These findings lend support to the concept that pericarditis
is an inflammatory complication of histoplasmosis, rather than a true infective
pericarditis.31.44.45
6.6. Disseminated Histoplasmosis

The true incidence of dissemination following infection with Histoplasma
capsulatum is unknown. In the largest outbreak studied to date, consisting of a
presumed 120,000 new infections, 61 patients with disseminated histoplasmosis
were identified for a rate of 0.5 per 1000 infected persons.33.46 As noted earlier,
immunosuppressed persons are at greatest risk for disseminated disease. Dissem-
ination has been thought for many years to occur primarily in the setting of acute
primary infection or acute reinfection. 17.46 In recent years recrudescence of a
previously dormant focus of infection in an immunosuppressed patient, partic-
ularly someone with AIDS, has been increasingly recognized as an important
pathogenic mechanism.18.22.47
The onset of symptoms in patients with disseminated histoplasmosis is
generally subacute to chronic,l7·46 Fever of 101°F or higher occurs in virtually all
patients. Malaise and weight loss are present in more than 50%, and respiratory
complaints such as cough or dyspnea are common. Hepatomegaly and/or spleno-
megaly are present in a third or more of patients. Ulcerations of the oropharynx
and larynx are seen frequently and are important diagnostic clues. Anemia and
leukopenia are common hematologic abnormalities. Chest roentgenograms may
show diffuse interstitial or nodular infiltrates, but are often normal. Patients with
AIDS and disseminated histoplasmosis have more severe manifestations and may
present with an acute sepsis-like syndrome, coagulopathy, respiratory failure, or
central nervous system involvement. 23
7. DIAGNOSIS
The diagnosis of active infection with Histoplasma capsulatum may be difficult.

Clinical findings are frequently nonspecific, isolation of the organism from
culture is time consuming, and the specificity of the available serologie tests is
limited. A thorough understanding of the various strengths and limitations of
each diagnostic test and modality is critical to the optimal management of
patients with histoplasmosis.
7.1. Skin Tests

Skin tests with the mycelial-phase antigen histoplasmin are positive in nearly
all normal hosts within four weeks of infection by H. capsulatum. Skin-test
208 STANLEY W. CHAPMAN and HAROLD M. HENDERSON
reaetivity persists for many years, and is an exeellent epidemiologie tool for
measuring the background incidenee of infeetion in a partieular geographie
region. Skin testing eannot distinguish between new and past infeetion and thus
is not a useful diagnostie tool. Skin tests are negative in as many as two-thirds of
patients with disseminated disease,17 and may be falsely positive in persons with
other fungal infeetions.
7.2. Tests for Antibodies

Antibodies to Histoplasma antigens, deteetable by serologie tests, are present
within 3 to 5 weeks in most persons with aeute pulmonary histoplasmosis.
Antibody levels rise to a peak over the next few months before declining to
undeteetable levels after two or more years. 29 Com plement fixation titers of 1: 8 or
higher to either the yeast or myeelial antigens or both are seen in more than
90% of symptomatie patients following exposure, and titers of 1:32 or higher are
seen in 70%.48
The immunodiffusion test for precipitin bands to the Hand M antigens may
also be useful. Immunodiffusion is not as sensitive as eomplement fixation, and
the deteetion of an H or M band generally lags behind the appearanee of a
eomplement fixation titer by 2-4 weeks. However, it is simple to perform and is
more specifie than complement fixation. 49 The M band appears after aeute
infeetion and may persist for years, and thus eannot differentiate between new
and remote infeetion. 50 The presenee of an H band does correlate with aetive
infection, but is seen in only 10-15% of patients with aeute pulmonary histo-
plasmosis. 48
There are severallimitations to serologie tests. As mentioned above, a delay
in the appearanee of deteetable antibody of three or more weeks is routine, and
the tests will therefore be negative in some patients if performed very early in the
course of symptomatie illness. Serologie tests are negative in 30-50% of individ-
uals with disseminated histoplasmosis.1 9.23 In endemie areas where infeetion is
eommon and reinfeetion may oeeur frequently, false-positive eomplement fixation
titers of 1:8 or 1:16 oceur in 5-15% of healthy persons. 51 Finally, false-positive
results oeeur in 20-40% of patients with other fungal and granulomatous
diseases, most commonly as eomplement fixation titers of 1: 8 or 1: 16.49.52.53 Newer
tests for antibody deteetion such as enzyme immunoassay (EIA) and radio-
immunoassay (RIA) are highly sensitive, but are limited by frequent false-positive
results and lack of eommercial availability.54
The serologie tests are most helpful when used in eonjunetion with a elinieal
syndrome compatible with aetive histoplasmosis. A eomplement fixation titer of
1:32 or more in a patient with pulmonary infiltrates and a typieal exposure history
is strongly suggestive of aeute pulmonary histoplasmosis, and further diagnostie
tests generally are not required. A fourfold rise in eomplement fixation titers is
also strong presumptive evidenee of aeute histoplasmosis. Immunodiffusion may
be used to confirm active histoplasmosis in the presence of complement fixation
titers of 1:8 and 1:16. Beeause they may remain positive for years, serologie tests
are not reliable for identifying the etiology of a pulmonary nodule or mass. In
patients suspected of having disseminated histoplasmosis, positive serologie tests
results support the institution of empirie antifungal therapy pending histo-
pathology or culture results (see below). Negative serologies do not exclude the
diagnosis of disseminated histoplasmosis.
7.3. Tests for Antigen

Detection by radioimmunoassay of a polysaccharide antigen of Histoplasma
capsulatum has recently been described. 55 The antigen may be found in the urine
or serum of the majority of patients with disseminated disease. Antigen is
detected infrequently in patients with other forms of histoplasmosis. Unlike
anti-Histoplasma antibodies, antigen levels quickly become undetectable following
resolution of infection, but remain elevated in patients with persistent disease.
Detection of antigen will likely prove most useful for the diagnosis of dissemi-
nated histoplasmosis, and in monitoring the response to therapy in AIDS patients
with disseminated illness. 56 At present antigen testing is not commercially avail-
able.
7.4. Culture and Histopathology

Cultures are positive in 75-90% of patients with disseminated histoplas-
mosis. Blood cultures will be positive in 40-70% of cases. The yield from blood
cultures is improved with the use of the lysis-centrifugation method.5' Bone
marrow cultures are positive in 64-90% of patients with dis semina ted disease
and are particularly useful in patients with AIDS. The organism may be cultured
from urine or sputum in more than half of patients with disseminated histoplas-
mosis. Other specimens from which the organism may be isolated include
cerebrospinal fluid, lung, lymph node, liver, and mucous membrane ulcers. The
chieflimitation of culture is the length of time required for growth and identifica-
tion of the organism, often 4-6 weeks. A more rapid means of establishing the
diagnosis if disseminated histoplasmosis, often within 24 to 48 hr, is the demon-
stration of typical organisms by methenamine silver stain in biopsy of oral
ulceration, bone marrow, liver, or lymph node specimens. A silver or Wright's stain
of blood or respiratory secretions will be positive for intracellular organisms
in more than 50% of patients with dissemination (Fig. 6).
In patients with a nondisseminated form of histoplasmosis, the ability to
isolate the fungus in culture is decreased compared with disseminated illness.
Sputum cultures are positive in 35-60% of patients with chronic cavitary histo-
plasmosis. 38 ,40 Cultures ofbronchial washings in patients with cavitary disease are
positive in about 40% of cases. 40 In acute pulmonary histoplasmosis, sputum
cultures are positive in no more than 10% of cases,34,48 although patients with
heavy exposures and more severe symptomatic illness are more likely to yield H.
capsulatum from their sputum. 31 ,58 Bronchial washings are seldom culture positive
in self-limited acute histoplasmosis. 48 In acute pulmonary disease the diagnosis
~
:::;
r.n
~
Z
t""
i:'rj
-<
~
C"l
::r:
;.-
~
s:::
;.-
Z
III
::l
0..
::r:
;.-
~
t""
t:I
~
::r:
i:'rj
Z
t:I
FIGURE 6. Wright's stain of peripheral blood smear from a patient with disseminated histoplasmosis. Note the i:'rj
~
intracellular Histoplasma yeasts. r.n
0
Z
may be established by open lung biopsy, but this should be reserved for immuno-
suppressed patients or those with severe pulmonary compromise for whom
diagnosis and therapeutic decisions must be made quickly. Cultures are almost
always negative in patients with mediastinal fibrosis, but methenamine silver stain
of histologie material may be positive in many cases. 42
8. TREATMENT
8.1. Disseminated Histoplasmosis

Untreated, the mortality rate of disseminated histoplasmosis is approx-
imately 90%.45,58 In patients with disseminated disease not infected with HIV,
treatment with amphotericin B is usually associated with rapid clinical improve-
ment and areduction in the mortality rate to 7-15%. Most deaths from dissemi-
nated histoplasmosis occur within 4 to 6 weeks of the diagnosis in patients who
receive less than 500 mg of amphotericin. 19 ,59,60 Relapse after clinical improve-
ment may occur in from 10 to 20% of persons treated with amphotericin B.
Relapse is most common in the immunosuppressed, and in patients with endo-
vascular infection, endocarditis, or meningitis, or those who receive less than
30 mg/kg total dose of amphotericin. 46 ,59,61 An amphotericin B total dose of
30 mg/kg or greater should thus be considered the standard therapeutic dose for
non-AIDS patients with disseminated histoplasmosis.
Ketoconazole is an oral antifungal agent active against H. capsulatum and has
been effective in the treatment of disseminated histoplasmosis. Response rates of
70-100% have been seen in two small series ofpatients. 62 ,63 A regimen of 400 mg
daily for at least 6 months is used most commonly. However, clinical improvement
is not as rapid as that seen with amphotericin, and treatment failures occur
routinely when ketoconazole is used in immunocompromised patients. 62 ,63 Keto-
conazole should therefore be reserved for patients with disseminated disease who
do not have meningeal involvement or other life-threatening illness, and who are
immunocompetent. Itraconazole, a new triazole, is also an effective oral drug for
dis semina ted histoplasmosis. 64 Response rates at a dose of 200 mg daily for a
minimum of6 months are similar to those seen with ketoconazole. Itraconazole is
also effective in immunocompromised patients,64 in particular patients with
AIDS, as discussed below. Itraconazole seems to be better tolerated than ketocon-
azole and is the oral agent of choice for disseminated disease at this institution.
AIDS patients with severe disseminated histoplasmosis likely to die within 7
days or with CNS involvement are best managed with an initial induction phase of
15-30 mg/kg amphotericin B over 4-8 weeks. 22 ,23 Patients with mild or moder-
ately severe disease may be treated with itraconazole 300 mg twice daily for 3 days
then 200 mg twice daily for 12 weeks as induction therapy.65 Following treatment
of the acute illness AIDS patients must remain on lifelong maintenance therapy to
prevent relapse. Itraconazole 200-400 mg daily is the maintenance regimen of
choice due to its efficacy (>90%) and tolerability.65,66 Amphotericin B 50 mg
given every 1-2 weeks is an alternative, with relapse rates of 9_19%.23,67 Keto-
conazole is ineffective for acute therapy or when used as maintenance for
disseminated disease in AIDS patients.
8.2. Nondisseminated Forms of Histoplasmosis

The large majority of cases of acute pulmonary histoplasmosis are self-
limited and require no therapy. Patients with heavy exposures and more symp-
tomatic disease may benefit from a brief course of amphotericin B. Doses of 150-
500 mg over 7-14 days have been used with success,6S,69 Anecdotal reports of
rapid responses following the administration of corticosteroids have been pub-
lished.5 s,7o
Decisions regarding the treatment of chronic cavitary histoplasmosis may
be difficult. As many as 30% of cases of cavitary disease spontaneously regress,
but prognostic predictions based on clinical symptoms and roentgenographic
findings are not always reliable. 4o For patients who are not severely ill a 2-3 month
period of observation is recommended,3s,71 If cavitation persists and clinical
improvement does not occur, antifungal therapy is indicated. Treatment with
ketoconazole 400 mg per day or itraconazole 200-400 mg per day has yielded
response rates of65-80% when administered for six months or longer. 62-64 This
response rate compares favorably with that of amphotericin B in several se-
ries,3S,40,72-74 Initial therapy with an oral agent is thus a reasonable approach.
Treatment failures with ketoconazole are well-described, however,40,75 and pa-
tients should be followed closely both during and after completion of therapy.
If clinical and roentgenographic improvement do not occur, or sputum cultures
remain positive, therapy with amphotericin B should be initiated. Total doses
of 35 mg/kg have given the best outcomes. 72,74 Patients who relapse after a
course of an oral azole may be treated with amphotericin or retreated with an oral
drug.
Antifungal therapy for the inftammatory complications ofhistoplasmosis has
not been evaluated in any controlled, meaningful fashion. Patients with pericar-
ditis generally recover with bed rest and antiinflammatory agents such as
ibuprofen, although cardiac tamponade (up to 25% of cases. 44 ) may require
surgical intervention. Some persons with acute massive lymphadenitis or medi-
astinal granuloma with symptomatic compression of adjacent structures have
been treated with ketoconazole or amphotericin. No data exist to form the basis
for a firm recommendation on the use of these drugs, and surgical decompression
may be required. 43 Antifungal agents are of no proven benefit in patients with
mediastinal fibrosis. 31
REFERENCES
1. Darling, S. T., 1906, A protozoan general infeetion produeing pseudotubercles in the lungs
and foeal neeroses in the liver, spleen and Iymph nodes,JAMA 46:1283-1285.
2. DeMonbreun, W. A., 1934, Cultivation and eultural eharaeteristies of Darling's Histoplasma

capsulatum, Am. J Trop. Med. 14:93-125.
3. Meleney, H. E., 1934, Pulmonary histoplasmosis: Report of 2 eases, Am. Rev. Tuberculosis
44:240-247.
4. Johnson, H E. and Batson, R., 1948, Benign pulmonary histoplasmosis: Case report with
brief review of literature, Dis. Chest 14:517-524.
5. Palmer, C. E., 1945, Non-tuberculous pulmonary ealcifieation and sensitivity to histo-
plasmin, Public Health Rep. 60:513-520.
6. Palmer, C. E., 1946, Geographie differenees in sensitivity to histoplasmin among student
nurses, Public Health Rep. 61:475-487.
7. Emmons, C. w., 1949, Isolation of Histoplasma capsulatum from soil, Public Health Rep.
64:892-896.
8. Zeidberg, L. D., Ajello, L., Dillon, A., and Runyon, L. C., 1952, Isolation of Histoplasma
capsulatum from soil, Am. J Public Health 42:930-935.
9. Kwon-Chung, K.]. and Bennett, ]. E., 1992, Histoplasmosis, in: Medical Mycology, Lea and
Fibiger, Malvern, Pennsylvania, pp. 464-513.
10. Walsh, T.]. and MitehelI, T. G., 1991, Dimorphie fungi eausing systemie myeoses, in: Manual
of Clinical Microbiology, 5th.ed. (A. Balows, W. ]., Hausler, Jr., K. L. Herrmann, H. D.
Isenberg, and H.]. Shadomy, eds.), Ameriean Soeiety for Mierobiology, Washington D.C.,
pp. 630-643.
11. Loyd, ]. E., DesPrez, R M., and Goodwin, RA., Jr., 1990, Histoplasma capsulatum in:
Principles and Practice ofInfectious Diseases, 3rd ed. (G. L. MandelI, R G. Douglas,Jr., and]. E.
Bennett, eds.), Churehill Livingstone Inc., New York, pp. 1989-1999.
12. Williams, D. M., Graybill, ]. R, and Drutz, D.]., 1978, Histoplasma capsulatum infeetion in
nude mice, Infect. Immun. 21:973-977.
13. Wu-Hsieh, B. and Howard, D. M., 1984, Inhibition of growth of Histoplasma capsulatum by
lymphokine-stimulated maerophages, J Immunol. 132:2593-2597.
14. Deepe, G. S. Jr., 1988, Proteetive immunity in murine histoplasmosis: Funetional comparison
of adoptively transferred T eell clones and splenie T eells, Infect. Immun. 56:2350-2355.
15. Gomez, A. M., Bulloek, W. E., Taylor, C. L., and Deepe, G. S.,Jr., 1988, Role ofL3T4+ T eells
in host defense against Histoplasma capsulatum, Infect. Immun. 56:1685-1691.
16. Henderson, H. M. and Deepe, G. S., Jr., 1992, Reeognition of Histoplasma capsulatum
yeast-eell antigens by human Iymphoeytes and human T-eell clones, J Leuk. Biol. 51:
432-436.
17. Goodwin, R A.,Jr., Shapiro,]. L., Thurman, G. H., Thurman, S. S., and Des Prez, R M.,
1980, Disseminated histoplasmosis: Clinieal and pathologie correlations, Medicine 59:1-33.
18. Davies, S. T., Kahn, M., and Sarosi, G. A., 1978, Disseminated histoplasmosis in immuno-
logieally suppressed patients: Oeeurrenee in a nonendemie area, Am. J Med. 64:94-99.
19. Kauffman, C. A., Israel, K. S., Smith,]. S., White, A. C., Sehwar,]., and Brooks, G. F., 1978,
Histoplasmosis in immunosuppressed patients, Am. J Med. 64:923-932.
20. Wheat, L.]., Slama, T. G., Norton,]. A., Kohler, R. B., Eitzen, H. E., Freneh, M. L. v., and
Sathapatayavongs, B., 1982, Risk faetors for disseminated or fatal histoplasmosis, Ann.
Int. Med. 96:159-163.
21. Bonner,]. R., Alexander,]. w., Dismukes, W. E., App, w., Griffin, F. M., Little, R., and Shin,
M. S., 1984, Disseminated histoplasmosis in patients with the aequired immunodeficieney
syndrome, Arch. Int. Med. 144:2178-2181.
22. Johnson, P. C., Khardori, N., Najjar, A. F., Butt, F., ManseIl, P. W. A., and Sarosi, G. A., 1988,
Progressive disseminated histoplasmosis in patients with aequired immunodeficieney syn-
drome, Am. J Med. 85:152-158.
23. Wheat, L.]., Connolly-Stringfield, P. A., Baker, R. L., Curfman, M. F., Eads, M. E., Israel, K.
S., Norris, S. A. Webb, D. H., and ZeckeI, M. L., 1990, Disseminated histoplasmosis in the
214 STANLEY W. CHAPMAN and HAROLD M. HENDERSON
aequired immunodefieieney syndrome: Clinieal findings, diagnosis, treatment and review

of the literature, Medicine 69:361-374.
23a.Edwards, L. B., Aequaviva, F. A., Livesay, V. T., Cross, F. w., and Palmer, C. E., 1969, An atlas
of sensitivity to tubereulin, PPD-B, and histoplasmin in the United States, Am. Rev. Resp. Dis.
99(supplement):1-132.
24. Furcolow, M. L., 1965, Environmental aspeets of histoplasmosis, Arch. Env. Health 10:4-10.
25. Emmons, C. w., Klite, P. D., Baer, G. M., and Hili, W. B., ]r., 1966, Isolation of Histoplasma
capsulatum from bats in the United States, Am.] Epidemiol. 84:103-108.
26. Waldman, R. W ., England, A. C., Tauxe, R., Kline, T., Weeks, R.]., Ajello, L., Kaufman, L.,
Wentworth, B., and Fraser, D. w., 1983, A winter outbreak of acute histoplasmosis in
northern Miehigan, Am.] Epidemiol. 117:68-75.
27. Furcolow, M. L., Tosh, F. E., Larsh, H. w., Lynch, H.].,]r., and Shaw, G. 1961, The emerging
pattern of urban histoplasmosis, N. Engi.] Med. 264:1226-1230.
28. Seward, C. w., Mohr, ]. A., and Rhoades, E. R., 1970, An outbreak of histoplasmosis in
Oklahoma, Am. Rev. Respir. Dis. 102:950-958.
29. DiSalvo, A. F. and]ohnson, W. M., 1979, Histoplasmosis in South Carolina: Support for the
mierofoeus concept, Am.] Epidemiol. 109:480-492.
30. Brodsky, A. L., Gregg, M. B., and Lowenstein, M. S., 1973, Outbreak of histoplasmosis
assoeiated with the 1970 Earth Day aetivities, Am.] Med. 54:333-342.
31. Goodwin, R. A., Loyd,]. E., and Des Prez, R. M., 1981, Histoplasmosis in normal hosts,
Medicine 60:231- 266.
32. Chin, T. D. Y., Ney, P. E., Saltzman, B. N., Paxton, G. B., Rakieh, J. H., Ware, M., Whit-
more, M., and Furcolow, M. L., 1956, An epidemie ofhistoplasmosis among sehool ehildren
in Arkansas, South Med.] 49:785-791.
33. Wheat, L.]., Slama, T. G., Eitzen, H. E., Kohler, R. B., Freneh, M. L. V. and Bieseeker,]. L.,
1981, A large urban outbreak of histoplasmosis: Clinical features, Ann. Intern. Med. 94:
331-337.
34. Ward, ]. 1., Weeks, M., Allen, D., Hutcheson, R. H., ]r., Anderson, R., Fraser, D. w.,
Kaufman, L., Ajello, L., and Spiekard, A.,1979, Acute histoplasmosis: Clinieal, epidemio-
logie, and serologie findings of an outbreak associated with exposure to a fallen tree, Am.
] Med. 66:587-595.
35. Medeiros, A. A., Marty, S. D., Tosh, F. E., and Chin, T. D. Y., 1966, Erythema nodosum and
erythema multiforme as clinieal manifestations of histoplasmosis in a community outbreak,
N. Engi.] Med. 274:418-420.
36. Rosenthai,]., Brandt, K. D., Wheat, L.]., and Slama, T. G., 1983, Rheumatologie manifesta-
tions of histoplasmosis in the reeent Indianapolis epidemie, Arth. Rheum. 26:1065-1070.
37. Lehan, P. L., Brasher, C. A., Larsh, H. w., Furcolow, M. L., 1957, Evaluation of clinical aids to
the diagnosis of ehronic progressive eavitary histoplasmosis, Am. Rev. Tuberculosis 75:
938-948.
38. Goodwin, R. A., Owens, F. T., Snell,]. D., Hubbard, W. w., Buehanan, R. D., Terry, R. T., and
Des Prez, R. M., 1976, Chronie pulmonary histoplasmosis, Medicine 55:413-452.
39. Latham, R. H. Kaiser, A. B., Dupont, W. D., and Dan, B. D., 1980, Chronie pulmonary
histoplasmosis following the exeavation of a bird roost, Am.] Med. 68:504-508.
40. Wheat, L. ]., Wass, ]., Norton, ]., Kohler, R. B., and Freneh, M. L. V., 1984, Cavitary
histoplasmosis oeeurring during two large urban outbreaks, Medicine 63:201-209.
41. Goodwin, R. A., NickelI,]. A., and Des Prez, R. M., 1972, Mediastinal fibrosis complieating
healed primary histoplasmosis and tuberculosis, Medicine 51:227-246.
42. Loyd,]. L., Tiliman, B. F., Atkinson,]. B., and Des Prez, R. M., 1988, Mediastinal fibrosis
eomplieating histoplasmosis, Medicine 67:295-310.
43. Garrett, H. E. and Roper, C. L., 1986, Surgieal intervention in histoplasmosis, Ann. Thor.
Surg. 42:711-722.
44. Wheat, L.]., Stein, L., Corya, B. C., Wass,]. W, Nonon,]. A., Grider, K., Slama, T. G., French,
M. L., and Kohler, R. B., 1983, Pericarditis as a manifestation of histoplasmosis during two
large urban outbreaks, Medicine 62:110-119.
45. Picardi,]. L., Kauffman, C. A., Schwarz,]., Holmes,]. C., Phair,]. P., and Fowler, N. 0., 1976,
Pericarditis caused by Histoplasma capsulatum, Am. J Cardiol. 37:82-88.
46. Sathapatayavongs, B., Batteiger, B. E., Wheat,]., Slama, T. G., and Wass,]. L., 1983, Clinical
and laboratory features of disseminated histoplasmosis du ring two large urban outbreaks,
Medicine 62:263-270.
47. Mandell, W, Goldberg, D. M., Neu, H. C., 1986, Histoplasmosis in patients with acquired
immunodeficiency syndrome, Am. J Med. 81:974-978.
48. Wheat,]., French, M. L. v., Kohler, R. B., Zimmerman, S. E., Smith, W R., Nonon, ]. A.,
Eitzen, H. E., Smith, C. D., and Slama, T. G., 1982, The diagnostic laboratory tests for
histoplasmosis: Analysis of experience in a large urban outbreak, Ann. Intern. Med. 97:
680-685.
49. Wheat,]., French, M. L. v., Kamel, S., and Tewari, R. P., 1986, Evaluations of cross-reactions
in Histoplasma capsulatum serologie tests, J Clin. Micro. 23:493-499.
50. Busey,]. F. and Hinton, P. F., 1965, Precipitins in histoplasmosis, Am. Rev. Resp. Dis. 92:
637-639.
51. George, R. B. and Lambert, R. S., 1984, Significance of serum antibodies to Histoplasma
capsulatum in endemie areas, South. Med. J 77:161-163.
52. Terry, P. B., Rosenow, E. C., and Roberts, G. D., 1978, False-positive complement fixation
serology in histoplasmosis. A retrospective study, JAMA 239:2453-2456.
53. ]ohnson,]. E., DeRemee, R. A., Kueppers, F., and Roberts, G. D., 1977, Prevalence offungal
complement-fixing antibodies in sarcoidosis, Am. Rev. Resp. Dis. 116:145-147.
54. American Thoraeie Society Statement: Laboratory diagnosis of mycotic and specific fungal
infections, 1985, Am. Rev. Resp. Dis. 132:1373-1379.
55. Wheat, L.]., Kohler, R. B., and Tewari, R. P., 1986, Diagnosis of disseminated histoplasmosis
by detection of Histoplasma capsulatum antigen in serum and urine specimens, N. Eng/.
J Med. 314:83-88.
56. Wheat, L. ]., Connolly-Stringfield, P., Blair, R., Connolly, K., Garringer, T., and Katz, B.,
1991, Histoplasmosis relapse in patients with AIDS: Detection using Histoplasma capsulatum
variety capsulatum antigen levels, Ann. Intern. Med. 115:936-941.
57. Bille,]., Stockman, L., Roberts, G. D., Horstmeier, C. D., and Ilstrup, D. M., 1983, Evaluation
of a Iysis-centrifugation system for recovery of yeasts and filamentous fungi from blood, J
Clin. Micro. 18:469-471.
58. Kataria, Y. P., Campbell, P. B., and Burlingham, B. T., 1981, Acute pulmonary histo-
plasmosis presenting as adult respiratory distress syndrome: Effect of therapy on clinical and
laboratory features, South. Med. J 74:534-537.
59. Sarosi, G. A., Voth, D. v., Bernhoff, A. D., Doto, I. L., and Tosh, F. E., 1971, Disseminated
histoplasmosis: Results of long-term follow-up, Ann. Intern. Med. 75:511-516.
60. Reddy, P., Goreliek, D. F., Brasher, C. A., and Larsh, H., 1970, Progressive disseminated
histoplasmosis as seen in adults, Am. J Med. 48:629-636.
61. Bradsher, R. W, Alford, R. H., Hawkins, S. 5., and Spickard, W A., 1982, Conditions
associated with relapse of amphotericin B-treated disseminated histoplasmosis, Johns Hop-
kins Med. J 150:127-131.
62. National Institute of Allergy and Infectious Diseases Mycoses Study Group: Treatment of
blastomycosis and histoplasmosis with ketoconazole, 1985, Ann. Intern. Med. 103:861-872.
63. Slama, T. G., 1983, Treatment of disseminated and progressive cavitary histoplasmosis with
ketoconazole, Am. J Med. 74(IB):70-73.
64. Dismukes, W E., Bradsher, R. W, Cloud, G. C., Kauffman, C. A., Chapman, S. W, George,
R. B., Stevens, D. A., Girard, W M., Saag, M. S., Bowles-Patton C., and the NIAID Mycoses
Study Group, 1992, Itraconazole therapy of blastomycosis and histoplasmosis, Am. J Med.
93:489-497.
65. Wheat, L. J., Hafner, R. E., Ritchie, M., and Schneider, D., 1992, Itraconazole is effective
treatment for histoplasmosis in AIDS: prospective multicenter non-comparative trial. 32nd
Interscience Conference on Antimicrobial Agents and Chemotherapy, abstract 1206.
66. Wheat, J., Hafner, R., Wulfsohn, M., Spencer, P., Squires, K., Powderly, W, Wong, B.,
Rinaldi, M., Saag, M., HamiII, R., Murphy, R., Connolly-Stringfield, P., Briggs, N., and
Owens, S., 1993, Prevention of relapse of histoplasmosis with itraconazole in patients with
the acquired immunodeficiency syndrome, Ann. Intern. Med. 118:610-616.
67. McKinsey, D. S., Gupta, M. R., Riddler, S. A., Driks, M. R., Smith, D. L., and Kurtin, P. J.,
1989, Long-term amphotericin B therapy for disseminated histoplasmosis in patients with
the acquired immunodeficiency syndrome (AIDS), Ann. Intern. Med. 111:655-659.
68. Naylor, B. A., 1977, Low-dose amphotericin B therapy for acute pulmonary histoplasmosis,
Chest 71:404-406.
69. Wynne, J. W, Olsen, G. N., 1974, Acute histoplasmosis presenting as the adult respiratory
distress syndrome, Chest 66:158-161.
70. Tegeris, A. S., Smith, D. T., 1958, Acute disseminated pulmonary histoplasmosis treated
with cortisone and MRD-1l2, Ann. Intern. Med. 48:1414-1419.
71. Goodwin, R. A., Snell, J. D., Hubbard, W W, and Terry, R. T., 1966, Early chronic
pulmonary histoplasmosis, Am. Rev. Resp. Dis. 93:47-61.
72. Parker, J. D., Sarosi, G. A., Doto, I. L., Bailey, R. E., Tosh, F. E., 1970, Treatment of chronic
pulmonary histoplasmosis, N. Engl. J Med. 283:225-229.
73. Sutliff, W D., Andrews, C. E., Jones, E., and Terry, R. T., 1964, Histoplasmosis cooperative
study 11. Chronic pulmonary histoplasmosis treated with and without amphotericin B,
Am. Rev. Resp. Dis. 89:641-650.
74. Sutliff, W D., 1972, Histoplasmosis cooperative study V. Amphotericin B dosage for chronic
pulmonary histoplasmosis, Am. Rev. Resp. Dis. 105:60-67.
75. Quinones, C. A., Reuben, A. G., Hamill, R. j., Musher, D. M., Gorin, A. B., and Sarosi, G. A.,
1989, Chronic cavitary histoplasmosis: Failure of oral treatment with ketoconazole, Chest
95:914-916.
13
Blastomyces dermatitidis and

Paracoccidioides brasiliensis
ROBERT W. BRADSHER and RICHARD W. MCDONNELL
1. INTRODUCnON
Blastomyeosis is a systemie fungal infeetion eaused by Blastomyces dermatitidis. The

manifestations range from self-limited infeetions, foeal eutaneous or pulmonary
lesions, destruetive foei in various organs to rapidly progressive and fatal in fee-
tion.l The vast majority ofblastomyeosis eases begin from a pulmonary portal of
entry 2,3 although a few eases of eutaneous inoeulation of the fungus in laboratory
workers have been reported. 4
Paracoeeidioidomycosis is a systemie fungal infeetion eaused by Paracoccidi-
oides brasiliensis. Manifestations include self-limited pulmonary infeetion, foeal
mueosal or eutaneous lesions, progressive pulmonary lesions, foeal lesions in
other organs, and widespread disseminated infeetion. Primary infeetion oeeurs
after inhalation of conidia in nearly all eases, with most infeetions being benign
and self-limited. 5
2. MYCOLOGY
B. dermatitidis and P. brasiliensis are both round, budding, thiek-walled yeast

eells in tissue infeetion. Like Histoplasma capsulatum, and Coccidioides immitis both
P. brasiliensis and B. dermatitidis are thermally dimorphie yeasts whieh relates to the
ROBERT W. BRADSHER and RICHARD W. MCDONNELL • Department of Medicine,

University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205-7199.
217
218 ROBERT W. BRADSHER and RICHARD W MCDONNELL
transition from myeelial forms in nature to yeast in tissue. This also is seen in the
properties of growth in eultures as myeelial at 25°C and as yeast at 37°C. The
physiologie change from myeelial to yeast oecurs beeause of a heat-shoek-related
insult with uneoupling of oxidative phosphorylation. 6
B. dermatitidis in clinieal specimens varies in size from 5 to 15 jJ.m and the
number of organisms in tissue ean differ greatly. A single broad-based bud forms
from the internal surfaee of the yeast eell in a eharaeteristie fashion. P. brasiliensis,
on the other hand, frequently has multiple buds on a single mother eell in a
pathognomonie pilot wheel configuration. In tissue, the budding eells vary from
12 to 40 jJ.m in diameter, with multiple small buds around the periphery.
3. PATHOPHYSIOLOGY
Infeetion with either of these organisms begins with inhalation into the lung
of eonidia of the myeelial phase of the fungus, whieh is followed by clearing of the
organism by bronchial pulmonary phagocytes. 3 As the fungus undergoes transition
to yeast eells, growth ean oeeur in the lung itself or spread via the bloodstream
and lymphaties to distant sites. With development of immunity, inftammatory reae-
tions oeeur at the initial infeetion site and at these metastatie foci. Initially the
pathologie response is suppurative with polymorphänuclear leukoeyte (PMN) infil-
tration and is followed by a granulomatous formation with lymphoeyte and mono-
eyte-derived maerophages. 7 This pyogranulomatous response is typieal ofblasto-
mycosis and paracoccidioidomycosis although necrosis or fibrosis can also be found.
The first leukoeyte response is thought to be a nonspecifie reaetion whereas
the formation of a granuloma is the result of development of specifie eellular
immunity. In human infeetions with B. dermatitidis or P. brasiliensis, pathologieal
seetions frequently demonstrate the fungus either inside or attaehed to mono-
eytes, maerophages, or giant eells. In one series of pathologie findings in blasto-
myeosis,7 from one to several yeast eells were loeated in the eytoplasm of giant
eells as a routine finding. The pathology in paraeoecidioidomycosis is remarkably
similar to fi'ndings in blastomycosis. 8 Differenees include greater lymphoid tissue
involvement and oeeasional instanees of widespread intestinal involvement. Dif-
ferentiation of these diseases ean be made when the pathognomonie "pilot wheel"
peripheral budding is observed.
A unique feature of paracoeeidioidomyeosis is the apparent inftuenee of
estrogens on the myeelia-to-yeast transition. Restrepo and colleagues have dem-
onstrated that a physiologie level of 10- 10 M 17 ß-estradiol signifieantly inhibited
the transition of myeelia to yeast in vitro, while other steroid hormones had no
effeet. 9 This may be a faetor in the large preponderanee of males with disease.
4 .. EPIDEMIOLOGY
Generally, blastomycosis is found in the states surrounding the Mississippi

and Ohio river valleys. 3 Adult men are more likely than women or ehildren to
BLASTOMYCES DERMATITIDIS AND PARACOCCIDIOIDES BRASILIENSIS 219
present with this infection, which may relate to greater potential for exposure to
the organism in nature from occupational or recreational sources.l° The organ-
ism was once thought to be only in the North American continent, prompting the
tide North American Blastomycosis in contradistinction to another disease termed
South American Blastomycosis. This terminology has been abandoned because
blastomycosis cases have been reported with increased frequency on other conti-
nents.l1.l 2 Also, the appropriate name for what was formerly called South Ameri-
can Blastomycosis is paracoccidioidomycosis and it is caused by P. brasiliensis. This
infection is geographically limited to Latin America, occurring from southern
Mexico to Argentina.8 Most cases occur in adults with a male:female ratio of
approximately 12:1. Skin test surveys have demonstrated equal reactivity to P.
brasiliensis antigens in men and women, 13 suggesting that hormonal factors may be
responsible for development of disease.
5. MANIFESTATIONS
Blastomycosis and paracoccidioidomycosis cause constitutional symptoms

including weight loss, fever, malaise, fatigue, and other nonspecific symptoms.
Most patients with blastomycosis present with either an acute pulmonary infec-
tion mimicking acute bacterial pneumonia, or a more chronic, indolent pneu-
monia that resembles tuberculosis or cancer. 2,3 Oral mucosal and cutaneous
involvement is the most common presentation of paracoccidioidomycosis, 8 and a
skin lesion is the most common extrapulmonary manifestation in blastomycosis,
appearing either as a verrucous lesion or an ulcer. Involvement of the genitourin-
ary tract, bones, nervous system, or other organs mayaiso be found in humans
with either of these organisms. 2,3,8
6. CLINICAL MARKERS
6.1. Identification of the Organism

The diagnosis ofblastomycosis or paracoccidioidomycosis is made by visual-
ization of the fungus in tissue or exudate followed by culture. Both organisms are
easy to detect in smears and cultures and that detection is enough to confirm the
diagnosis. For example, cultures were positive in each case ofblastomycosis in one
series of patients 14 and in every patient who had exudate, sputum, or tissue
examined, the fungus was identified on microscopy. In addition to examinations
of sputum by wet preparations with potassium hydroxide, cytologic preparations
can be used for a reliable diagnosis of infection with these two fungi. Pathologists
can identify the appearance of the fungus after Papanicolaou staining of clinical
materia}l5; because the clinical picture of chronic pneumonia caused by blasto-
mycosis may resemble that of carcinoma of the lung, cytologic examination of
sputum is not an infrequent means of diagnosis. In areas with a low frequency of
this infection, many cases will be diagnosed only after invasive procedures.
220 ROBERT W. BRADSHER and RICHARD W. MCDONNELL
Specific fungal staining of tissue with Gomori methenamine silver staining allows
the diagnosis to be made.
6.2. Serology
It is fortunate that diagnosis by smear or culture is relatively easy because
serodiagnostic techniques used for infections are not reliable in blastomycosis.
Tests include complement fixation (CF) antibodies, immunodiffusion (ID) pre-
cipitin bands, and antibodies by enzyme immunoassay (EIA).2,16 These tests have
been useful as tools for epidemiologie assessments in blastomycosis, but not for
clinical diagnosis. Reactivity to antigens of other fungi, particularly H. capsu-
latum, is severely limiting to specificity of the assays.16 Since the geographie
regions for these two organisms overlap, clinical confusion can result from the
cross-reactions. For example, persons with blastomycosis are just as apt to demon-
strate CF antibodies against histoplasmin as against blastomycin. 3 This test's poor
sensitivity prompted the development of immunodiffusion testing, which re-
sulted in sensitivity rates of up to 80% in the initial reports.l 6 Klein et alP
reported even better results with EIA techniques using a yeast-phase antigen (A-
antigen) than immunodiffusion tests in sera of patients with localized blasto-
mycosis. Similar results with ID and EIA were obtained when sera from patients
with disseminated disease were testedP In the largest outbreak ofblastomycosis
reported, Klein et al. described antibody detection by EIA, ID, and CF techniques
in 77%, 28%, and 9%, respectively.l8 Although documented cases had positive
reactions, false-positive results were also detected by Lambert and George with a
similar EIA technique on specimens from persons in endemie areas for histoplas-
mosis and blastomycosis.l 9 Therefore, as noted by Sarosi et al.,20 serodiagnosis of
blastomycosis is a problem because of potentiallow sensitivity and low specificity
rates.
Newer antigens may change this observation in blastomycosis. Klein and
Jones have isolated a 120-kDa yeast-cell-surface protein of B. dermatitidis called
WI-l which was useful in detection of antibody in patients with this infection. 21
An antibody to WI-l was detected by radioimmunoassay (RIA) in 85% of sera
from blastomycosis patients but only in 2 of 73 patients with histoplasmosis,
coccidioidomycosis, or sporotrichosis and in none of the control donors with no
his tory of fungal infection. In a comparative study, RIA to WI-l detected
antibodies specific for B. dermatitidis in 18 sera that did not have detectable
antibody to A-antigen by EIA; in the 44 sera with identification of antibodies by
both methods, RIA had substantially higher titers detected. The same antigen
has more recently been utilized in cellular immunity studies to be detailed later
in the chapter. 22 Following modification of A-antigen, 83% of 125 patients with
culture-proven blastomycosis had detection of antibody with a commercially
available EIA technique. 23 Cross-reactions did occur in patients with histoplas-
mosis but the quantitative amount of antibody was much lower than in the
patients with blastomycosis; mean index values from those with positive results
were 24.3 for blastomycosis patients and 3.8 for histoplasmosis patients. 23
Serologic testing is historically more reliable in paracoccidioidomycosis than

in blastomycosis, but caution is still required in interpretation. 24 Tests include
CF, ID, and EIA. CF antibodies correlate with disease activity, but cross-react with
other mycoses, including histoplasmosis. Precipitin bands seen with ID are highly
specific and sensitive, but do not correlate weIl with disease activity and may
persist for prolonged periods of time. Band 1 is most commonly detected and is
identical to Band B described by Yarzabal et al. 25 In double ID testing with a
positive control serum, a line of identity for Band 1 is 100% specific for P.
brasiliensis and can approach 100% sensitivity.26 Band 1 shows total identity with a
soluble 43-kDa glucoprotein described by Puccia and colleagues as an immuno-
dominant antigen. 27 This antigen has been detected in sera of patients with
paracoccidioidomycosis by the immunoblot technique using rabbit-monospecific
antise~a as the probe. 28 Sera from patients with other systemic mycoses cross-
reacted significantly with the 43-kDa glycoprotein in an EIA assay, whereas a
tube-precipitant assay was 100% specific. 29 CF and ID are the most commonly
available tests. With the progress seen in other mycotic diseases in recent years, it
appears likely that further improvements in serologic testing will occur for para-
coccidioidomycosis.
6.3. Skin testing

Skin testing with blastomycin is, unfortunately, no better than serology as a
diagnostic procedure. With this crude mycelial-phase filtrate of B. dermatitidis in
two large series30 ,31 59% and lOO% of patients with positive cultures had negative
reactions to blastomycin skin tests. Blastomycin does not provide sufficient
specificity or sensitivity for reliable patient assessment and is no longer available
clinically. It may be useful epidemiologically; in the largest blastomycosis out-
break, 15 of 48 tested had a positive skin reaction. 32 Unlike the detection oflong
lasting immunity with in vitro antigen-induced lymphocyte reactivity in blasto-
mycosis,33 blastomycin skin-test reactions rapidly diminish and become nega-
tive on repeated testing over time,34
Patients rarely have skin-test reactivity to paracoccidioidin or heterologous
antigens with active, progressive paracoccidioidomycosis but do with benign, self-
limited infection. 35 This makes skin testing useless in the dia gnosis of active
disease. Paracoccidioidin can be useful epidemiologically, as cross-reactivity with
histoplasmosis and other mycoses is relatively low. 13
7. IMMUNOLOGY
Cellular immunity is considered to be the major protection factor in prevent-

ing progressive disease caused by slow-growing but pathogenic fungi like B.
dermatitidis,33 P. brasiliensis,36 H. capsulatum,37 and C. immitis. 38 Although anti-
bodies may be helpful for diagnosis, humoral immunity is not critical for protec-
tion because patients with hypogammaglobulinemia handle fungal infections
222 ROBERT W BRADSHER and RICHARD W MCDONNELL
adequately.39 Mice with congenital absence of the thymus are more susceptible to
these fungi because they lack cellular immunity.4o With restoration of thymus
function, protection from further infection is demonstrated. 41 Cellular immuno-
suppression with transplantation or with HIV infection leads to infection with B.
dermatitidis 42 •43 and P. brasiliensis. 44•45
7.1. In Vivo Markers

Experimental animal models of blastomycosis have suggested that cellular
immunity is the essential arm of the immune system for preventing progressive
disease. Spencer and Cozad 46 immunized mice with either live or killed yeast
cells of B. dermatitidis and documented delayed hypersensitivity by antigen-
induced foot pad swelling; increased thickness as a marker of cellular immunity
was noted on days 12 and 20 after immunization. More important, levels of
delayed hypersensitivity correlated with resistance of immunized mice to lethai
challenge with live B. dermatitidis organisms. 47
Deepe and colleagues48 described a reproducible model of disseminated
blastomycosis in mice. Tail-vein injection of avirulent B. dermatitidis strain
resulted in a 200-fold increase in number of organisms in the lung and brain by
the fifth week of infection. Initially, acute inftammatory cells were found at sites of
infection but by three weeks, multiple granulomas were present. Foot pad re-
sponse to a specific Blastomyces antigen was not found, presumably secondary to
anergy. With subcutaneous inoculation and a lower yeast inoculum, delayed
hypersensitivity was measured by spleen-cell proliferation in response to Blasto-
myces antigen. The suppression of immune response with a large antigenic
challenge was considered to mimic human disease with widespread dissemina-
tion.
An animal model fo dis semina ted P. brasiliensis has been described using
testicular inoculation of hamsters. 49 An intense cellular response resulted in
compact granulomas with few organisms; after 23 weeks, the granulomas had
become less compact and profound fungal proliferation was noted in association
with depressed cellular immunity and progressive disease. Administration of
levamisole, an antihelminthic agent known to augment T-cell function, to the
animals increased lymphokine production in response to P. brasiliensis antigens,
and allowed continuation of the compact granuloma forms in the animals. 5o
The importance of thymus function in P. brasiliensis infection has been
demonstrated in studies using nude mice, which succumbed slowly without
granuloma formation. 51 Thymus transplantation resulted in resistance to P.
brasiliensis challenge to these animals.
Defavere et al. used a P. brasiliensis-immunized mouse model to show a
granulomatous response to P. brasiliensis antigen-coated bentonite particles in the
lung. 52 With intratracheal challenge with live yeast-phase organisms, an intense
pulmonary hypersensitivity pattern with few organisms was noted in immune
animals whereas control animals developed granulomas but also greater numbers
of organisms.5 3 Immunization resulted in antigen-induced foot pad swelling
which decreased after pulmonary challenge with the fungus; this related to
trapping of sensitized lymphocytes in the lung. Cell-mediated immunity was
responsible for resistance to infection in this model.
After Restrepo and colleagues54 developed a method to isolate viable P.
brasiliensis conidia, McEwen and colleagues emulated natural infection with
intranasal challenge in Balblc mice.5 5 Conidia reached the alveoli and converted
to yeast form within 12 hr. Initial cellular response was composed of PMN, but
by 6 days, lymphocyte and macrophages predominated. Multinucleated giant
cells appeared only after 6 weeks, and progressive increases in the number of
viable fungi were observed over time. This model was used to study development
of pulmonary fibrosis with progressive collagenization over 16 weeks and alter-
ations in the proportion of collagen fibers land 11. 56 This model may facilitate
understanding of the pathogenesis of fibrotic changes seen in the lungs of some
patients with this disease.
7.2. In Vitra Lymphocyte Studies

Markers of cellular immunity have correlated with protection from progres-
sion of infection with H. capsulatum 37 •4o and C. immitis. 38•57 Because of inadequate
antigen, study of in vitro markers as correlates of cellular immunity in blasto-
mycosis has lagged behind these other fungal infections. A more active and
specific Blastomyces antigen was developed and reported by Cox and Larsh 58 .59
and Deighton et al. 6o The antigen called B-ASWS is an alkali- and water-soluble
preparation of yeast cells of B. dermatitidis. Sensitized animals responded in a
specific fashion by foot pad thickness assays following injection of the antigen.
Cox, using B-ASWS antigen in human lymphocyte proliferative assays with cells
from healthy donors with no history of blastomycosis demonstrated some cross-
reactivity in histoplasmin skin-test positive donors. 57 Less cross-reaction occurs
than with blastomycin as antigen, however. With the same techniques and antigen,
distinction was possible between patients with previously treated blastomycosis
and healthy donors with no his tory of infection or those with previous histoplas-
mosis.3 3 In subsequent studies with B-ASWS-induced lymphocyte reactivity
assays, patients with acute pulmonary blastomycosis did not have in vitro markers
of specific Blastomyces immunity but did upon retesting within 4 weeks of antifun-
gal therapy.6 1 In contrast, patients with extrapulmonary blastomycosis who had
experienced symptoms for months prior to diagnosis, which allowed develop-
ment ofimmunity, had signifieant lymphoeyte responses to B-ASWS at the onset
of therapy.61 The time to develop specific eellular immunity in these patients with
pulmonary blastomycosis was similar to development of immunity in Deepe's
disseminated blastomycosis murine model. 48
Klein and colleagues 22 examined human lymphocyte response to WI-l, the
120-kDa B. dermatitidis yeast-cell-surface protein previously described for sero-
logie assessment in blastomycosis. 21 Cells from patients with treated blastomyeosis
responded to WI-l in a similar fashion as to stimulation with B_ASWS.32.33.61
T-cell clones were developed that proliferated in response to stimulation with
WI-l and B-ASWS but not to antigens of H. capsulatum or C. albicans. 22 This led to
the conclusion that WI-l is the immunodominant antigen for the ceIl-mediated
response as it is for the humoral immune response,.21
Lymphocyte function in paracoccidioidomycosis has been investigated in
both animal models and in humans with infection. 36 Pera\oli et al. 62 studied the
transfer of ceIl-mediated immunity to P. brasiliensis in hamsters with dialyzable
leukocyte extracts (DLE). DLE from immunized hamster lymph nodes and spleen
cells resulted in positive macrophage migration inhibition as a measure of
lymphokine production as weil as positive skin tests with P. brasiliensis antigen
compared to control animals. DLE from both control and immune hamsters
increased recipient response to Candida albicans antigen and BCG, which indi-
cated a nonspecific augmentation of cellular immunity as weil as the specific
enhancement. Further studies may better define the factors for specific and
nonspecific amplification of ceIl-mediated immunity.
Castaneda et al. 63 studied the regulation of cellular responses in chronic
murine P. brasiliensis infection. At 18 weeks of infection, peripheral blood lympho-
cytes had depressed concanavalin A responsiveness in vitro. When those cells were
mixed with peripheral blood lymphocytes from noninfected animals in a 1:1 ratio,
response to concanavalin A was reduced by 95% compared to responses of
normal cells without mixture of cells from chronically infected animals. Depletion
of suppressor cells in the mixed cultures also returned responsiveness to normal.
Injection of immune mouse sera with high antibody titers against P. brasiliensis
significantly reduced in vivo delayed hypersensitivity responses, demonstrating
the possible regulation of cellular response by humoral immunity.
Jimenez-Finkel and Murphy64 studied the induction of antigen-specific
T-suppressor cells by a soluble P. brasiliensis antigen in a murine model. Subcuta-
neous injection of the antigen in Freund's adjuvant induced delayed-type hyper-
sensitivity whereas intravenous injection induced a population oflymph node and
spleen cells that suppressed the development of cellular immunity to specific
antigen on adoptive transfer. These cells were present at 7 days and absent by 14
days and were called afferent suppressor ceIls. 64 In additional studies, a second
population of cells were described which had other suppressor functions that
inhibited delayed-type hypersensitivity in a foot pad thickness assay in previously
immunized mice;65 these cells were called efferent suppressor cells. A soluble
factor from the afferent suppressor cell was shown to induce the efferent sup-
pressor T cells. This model of suppressor-cell induction by intravenous antigen
may mimic the depressed cell-mediated responses observed in paracoccidioido-
mycosis patients who have been shown to have P. brasiliensis antigens and immune
complexes in the serum. 66-68 Chequer-Bon-Haviv et al. 68 extended work on the
immunosuppressive effect of sera from paracoccidioidomycosis patients on the
proliferative response of normal mononuclear cells by studying the inftuence of
immune complexes. Treatment of sera from patients with the chronic moderate
form of the disease with 2.5% polyethyleneglycol to precipitate immune com-
plexes significantly reduced the inhibitory activity of the sera upon normal
mononuclear cell mitogen-induced proliferation. The precipitates were shown to
BLASTOMYCES DERMATITIDIS AND PARACOCCIDIOIDES BRASIUENSIS 225
contain a 34-kDa polypeptide that reacted with rabbit anti-Po brasiliensis IgG. This
study supports the concept that circulating P. brasiliensis antigens may have a
negative immunoregulatory effect on lymphocyte transformation.
Perac;oli et al. 69 found increased numbers of natural killer cells in patients
with acute and chronic paracoccidioidomycosis with the cytotoxic activity of the
NK cells being significantly lower than in a control group. Mota et al. 7o studied
mononuclear cell subsets in 70 patients with paracoccidioidomycosis and demon-
strated low helperlsuppressor cell ratios and increased monocyte/null cell popula-
tions compared to control groups. Silva and Fuguieiredo 71 found elevated levels of
tumor necrosis factor in 30 patients with paracoccidioidomycosis. Further studies
may delineate the role of lymphokines, cytokines, and immune complexes in
changing immune function.
7.3. In Vitro Polymorphonuclear Studies

The role of PMN in defense against blastomycosis has been controversial in
the literature. Despite the almost universal presence of these acute inflammatory
cells in histologie sections from patients, whether the cell is a defense or actually
an aid to the infection has been debated.
The interaction of one clinical isolate of B. dermatitidis and PMN chemotaxis,
chemiluminescence, phagocytosis, and killing was reported by Sixbey and co-
workers. 72 High rates of phagocytosis were recorded by microscopy and con-
firmed by electron microscopy. With other strains of B. dermatitidis, investigators
have not been able to reproduce phagocytosis of yeast cells by PMN; it is
considered that Sixbey and colleagues studied an unusually small yeast cell strain
of this fungus. In addition to phagocytosis, Sixbey et al. described a 29%
reduction in colony-forming units of yeast counts in coculture experiments with
PMN.72 Chemiluminescence of PMN was noted in response to the yeast, and broth
culture supernatants of the organism contained PMN chemotactic activity.72
In contrast, Brummer and Stevens reported up to 45% enhancement of
growth of B. dermatitidis with human PMN cocultures at 24 hr and 68% enhance-
ment at 72 hr. 73 Even after two cycles of freezing and thawing the PMN to ensure
nonviability, enhancement of growth was 40% compared to a 43% enhancement
with viable PMN.73 Subsequently, Brummer and Stevens 74 reported that cocul-
tures of B. dermatitidis and murine peritoneal PMN resulted in enhancement of
fungal replication. Injection of B. dermatitidis along with autologous PMN caused
90% greater growth of the subcutaneous organisms than when the yeast were
injected alone. 74 These studies suggested that PMN might actually accelerate or
exacerbate the infection rather than help control the infection. Other investiga-
tions of the interactions of PMN and B. dermatitidis have not reproduced this
enhancement of fungal growth by PMN but differences in methods, such as the
PMN-to-fungus ratios, have been utilized.
Brummer and colleagues, in aseries of subsequent reports, related immune
activation of murine PMN with killing of B. dermatitidis. Cells obtained from
injection of either nonviable B. dermatitidis or sodium caseinate into the peritoneal
cavity of immunized mice resulted in reduction of yeast in coculture compared

with growth in media alone 75 or with murine peripheral blood PMN or
thioglycollate-induced peritoneal PMN.76 More effective killing of B. dermatitidis
in coculture was also found using Blastomyces antigen-induced culture super-
natants of spleen cells from immunized mice, presumably from Iymphokine
activation of PMN. 77 Further support to PMN activation and killing was added
with the report that immunologically activated PMN had a lO-fold higher peak of
oxidative burst as measured by chemiluminescence 78 and produced greater
amounts of superoxide anion in response to B. dermatitidis than did thioglycollate-
induced peritoneal murine PMN.79
In contrast to PMN enhancement previously reported, a reduction in yeast in
cocultures with peripheral blood murine PMN ranged from 33% with normal,
nonimmune mice to 54% with immunized mice. 80 Killing was significantly
greater than with peripheral blood PMN, either from nonimmune animals given
intraperitoneal yeast or from immunized animals without intraperitoneal anti-
gens. Curiously, peripheral blood PMN from nonimmune mice had greater
fungicidal activity than similar cells from immunized mice without intra-
peritoneal challenge. 80 The explanation for this apparent contradiction was
postulated to be that different populations of peripheral blood PMN were
studied because of localization of some PMN at the Blastomyces immunization site.
Greater killing of the yeast by PMN from stimulated immune animals was thought
to be a result of soluble mediators from the local immune re action that activated
the PMN in the peripheral blood. 80 Similarly, Iymphokine activation of murine
PMN from peripheral blood81 with in vitra addition of recombinant INF--y to
PMN was noted to cause greater killing of B. dermatitidis in cocultures.
Drutz and Frey82 reported that approximately 10-18% of B. dermatitidis were
killed by PMN from human volunteers who had no evidence of prior blastomyco-
sis. Killing of yeast was assessed by methylene blue dye exclusion and confirmed
by Chromium release of B. dermatitidis previously grown in the presence of the
isotope. 51 Greater than 97% of the yeast were surrounded by PMN after 2 hr of
coculture incubation but the yeast were extracellular. Electron-dense material,
presumed to be from PMN granules, encircled the organisms by electron micro-
seopie examination.
Experiments were also performed to examine the effect of PMN on B.
dermatitidis conidia. 82 Both phagocytosis and killing of the conidia occurred.
Reduction of nitroblue tetrazolium dye by PMN and either yeast or conidia
cultures suggested that superoxide anion generation took place; conidia caused a
greater degree of oxidative burst than did yeast cells. These results supported the
concept that human PMN were effective in killing conidia, which is the infectious
form of this fungus in nature. If a small enough number of conidia were inhaled,
the PMN response might be adequate to eliminate the fungus before transforma-
tion to the yeast form and subsequent progressive infection. 82
Schaffner and colleagues compared the susceptibility of dimorphie fungi
and opportunistic fungi to killing by human PMN.83 The major conclusions to
their studies were that virulent dimorphie fungi (B. dermatitidis, H. capsulatum, P.
brasiliensis, and Sporothrix schenkii} were more resistant to killing by PMN than
were opportunistic organisms (Aspergillus, Mucorales, and Petriellidium boydii).
Specifically with B. dermatitidis, approximately 45-55% of conidia were killed by
human PMN but the yeast phase organisms were resistant to killing. In these
experiments, even with low numbers of yeast and a high concentration of PMN,
killing was not observed.83 In chemiluminescence assays, it was noted that B.
dermatitidis yeast cells caused an oxidative burst even in the absence of PMN, and
production of hydrogen peroxide by the yeast itself was measured calling into
question previous works of chemiluminescence with luminol-amplification with B.
dermatitidis. 72 ,78,82 In experiments using other techniques, however, unequivocal
evidence was obtained of PMN oxidative burst stimulation by B. dermatitidis as
with the other dimorphic fungi that were studied. 83
The role that PMN play in paracoccidioidomycosis is not dear. PMN are
usually present in infected tissues, and in animal models, PMN are the initial
cellular response to fungal challenge and are the only inftammatory cell for the
first several days. Their role in prevention of progressive infection or on the
course of the disease remains uncertain.
Goihman-Yahr and colleagues demonstrated that yeast-phase organisms of
B. brasiliensis are phagocytosed and digested by PMN from normal hosts or
patients with other granulomatous diseases, whereas PMN from patients with
paracoccidioidomycosis do accomplish phagocytosis but are comparatively defi-
cient in digestion. 84 Similar results were also demonstrated with organisms that
were killed by autodaving prior to phagocytosis.85 Yeast exposed to 500 ILg/ml of
amphotericin B for 18 hr in vitro were digested by PMN from paracoccidioido-
mycosis patients in a normal fashion.
Calich et al. 86 studied the rapid inftux of PMN to P. brasiliensis yeast cells at
subcutaneous inoculation sites in mice and found that neither depletion of
complement with cobra venom factor nor testing in C5-deficient mice altered the
inftux of PMN to the infection site. This suggests that complement does not
control chemotaxis in paracoccidioidomycosis. These investigators further dem-
onstrated that murine peritoneal macrophages incubated for 6 hr with yeast cells
of P. brasiliensis release a soluble factor that induced inftux of PMN in vivo. 87 The
factor was shown to be a protein produced by glass-adherent cells, with a
molecular weight of less than 15 kDa. Puromycin, a pro tein inhibitor, suppressed
production of the factor.
7.4. In Vitro Monocyte and Macrophage Studies

A number of experiments have been reported pointing to the importance of
cell-mediated immunity in blastomycosis. Delayed hypersensitivity to Blastomyces
antigen was demonstrated by transfer of lymphocytes from immunized mice but
not by serum,88 and the adoptive transfer of immunity protected animals from
challenge with live organisms. 47 One of the effector mechanisms of this protec-
tion was studied with a coculture of yeast and monolayers of peritoneal macro-
phages from mice immunized with killed B. dermatitidis,s9 Scanning electron
microscopy confirmed the phagocytosis and intracellular location of the yeast

seen with light microscopy of these activated macrophages. In other experiments,
extracellular yeast were removed from the macrophage monolayers after 3 hr of
incubation to determine the intracellular fate of the fungus. Macrophages from
nonsensitized animals had a 1.5 log increase in yeast numbers in cell cultures over
96 hr whereas macrophages from immune mice inhibited the intracellular
growth of the yeast. 89 This inhibition of growth of the fungus by sensitized
macrophages supported the hypothesis that macrophage activity correlated with
specific cellular immunity to B. dermatitidis.
Interactions of B. dermatitidis with murine peritoneal macrophages have been
reported by Brummer and colleagues90 in coculture experiments. Unstimulated
mice cells caused a 24% reduction, and concanavalin A-treated mice macro-
phages had a 64% reduction of colony forming units of yeast compared to the
growth of the fungus in medium alone. Peritoneal macrophages obtained 4 weeks
after fungal challenge from Blastomyces-immunized mice showed a 41 % reduc-
tion in yeast cells. Brummer and others91 next reported that avirulent strains
of the fungus were inhibited from growth but virulent strains of B. dermatitidis
could escape inhibition of growth in coculture with peritoneal murine macro-
phages under similar conditions as above. 90 Similar differences between growth
of avirulent and virulent strains were obtained with macrophages activated by
prior concanavalin A treatment of the mice. Complement and antibodies were not
necessary for inhibition of growth of the fungus but protein synthesis inhibitors
added prior to macrophage activation could block killing. 92
Brummer and colleagues93 examined activation of peritoneal murine macro-
phages by INF--y in vitro and the mechanisms for the 37% reduction in yeast
numbers that was observed in coculture experiments. A 100-fold higher concen-
tration of INF--y was required for macrophages to kill B. dermatitidis than required
for Candida albicans. The investigators were not able to demonstrate that the
macrophages could execute phagocytosis of the B. dermatitidis strain studied.
Inhibitors of oxidative metabolism in similar cocultures did not inhibit killing of
the yeast,94 which suggested that the fungicidal activity of the macrophages
occurred independendy of an oxidative burst. 94 Murine alveolar macrophages
were found to have similar killing activity for B. dermatitidis as peritoneal macro-
phages. 95 Cells from immunized animals killed 15-24% of the yeast compared to
cultures in medium alone, whereas cells from unstimulated animals killed 21-
33%. These observations were followed by studies of INF-'Y or lymphokine-
activated murine alveolar macrophages with resultant increased killing of B.
dermatitidis. 96 •97 Without immune activation, this study did not demonstrate
alveolar macrophage killing of the fungus.
Sugar and Picard98 examined growth of avirulent and virulent strains of B.
dermatitidis conidia on coculture with murine alveolar macrophages. With these
cells that had not undergone immune stimulation, conidia of the fungus were
blocked from transition from the mycelial phase to the yeast phase. Hydrogen
peroxide was also able to block this phase transition. This macrophage-induced
inability of the fungus to convert to the tissue phase might be another nonspecific
host defense like PMN,82 which could account for human cases of subclinical
blastomycosis.
In addition to the experiments with human PMN and B. dermatitidis de-
scribed above, Drutz and Frey also examined human monocyte and macrophage
interactions with this fungus.8 2 Monocytes phagocytized 40% and killed 35% of
the mycelial phase of B. dermatitidis. Monocyte-derived macrophages from sub-
jects with no history of blastomycosis were shown to ingest and kill more than
80% of conidia. Monocytes were unable to phagocytize yeast but macrophages
ingested approximately 20% of the yeast. Of interest, 40% of the yeast were killed
by the macrophages, apparently by extracellular mechanisms.8 2 Brummer and
Stevens73 reported interactions of B. dermatitidis and monocytes and macrophages
from peripheral blood of healthy human volunteers with no history of blasto-
mycosis. Colony-forming units of fungus were reduced by monocyte coculture by
65% and 45% of avirulent and virulent strains, respectively, compared to growth
of fungus in medium alone. After maturation to macrophages, the reduction of
growth of yeast was 85%.73
Bradsher and colleagues performed three sets of experiments with periph-
eral blood mononuclear cells,99 monocyte-derived macrophages,IOO and alveolar
macrophages 101 respectively from persons with culture-proven blastomycosis and
from uninfected controls. A clinical isolate of B. dermatitidis was added at ratios of
yeast to peripheral blood mononuclear cells from 0.001 to 10 in tumbled suspen-
sion cocultures with tritiated thymidine uptake by lymphocytes measured after 5
days of incubation. 99 Inocula from 103 to 107 yeast did not cause lymphocytic
uptake of tritiated thymidine as a measure of immune stimulation from the
control group but the yeast did stimulate uptake in a dose-dependent fashion with
cultures from the blastomycosis patient group. Therefore, live yeast caused
specific lymphocyte stimulation from blastomycosis donors like the soluble anti-
gen, B-ASWS.33.61 In addition, counts ofyeast increased from the 5 x 104 initial
inoculum of B. ~rmatitidis to 3.2 X 105 over the 5 days of coculture with
nonimmune cells and 3.6 X 105 with immune donor cells. The absence of
inhibition of replication of yeast by the immune cells was in contradistinction to
the intracellular growth inhibition by murine macrophages reported by McDaniel
and Cozad;88 this difference was considered to be a result of extracellular growth
of the fungus because removal was not possible with these tumbled suspension
cultures.
Macrophage monolayers of cells from patients with culture-proven blasto-
mycosis or normal controls were challenged with B. dermatitidis at a ratio of 1
yeast to 5 macrophages for a 2 hr incubation with subsequent removal of
extracellular organisms. 1OO Previously infected donors had 35-44% of macro-
phages with ingested yeast whereas phagocytosis was observed with macrophages
from nonimmune donors in only 12-16% of cells. Electron microscopy confirmed
the intracellular location of the yeast. The numbers of yeast increased over 72 hr
in coculture with nonimmune cells but growth of B. dermatitidis was inhibited in
cocultures with immune macrophages, paralleling the experiments with immune
murine macrophages. 88 The mean increase for growth of yeast inside the non-
immune cells was 78.8 ± 4%; a 4 ± 8% decrease in yeast numbers was found in
cultures of cells from infected patients. IOO Growth of the organism in media alone
was similar to that inside normal cells (85% ± 3%).
Because alveolar macrophages are one of the first lines of defense against
pulmonary fungal pathogens,102 these macrophages were examined after being
obtained by bronchoalveolar lavage performed on patients with recently treated
blastomycosis and on healthy control subjects. 101 In addition, comparisons were
made to peripheral blood-monocyte-derived macrophages from the same donors.
U sing methods as described for the previous study,100 a greater number of
macrophages from blastomycosis patients had ingested B. dermatitidis than the
cells from normal controls.I Ol The range for blastomycosis patients was 25-38%
of macrophages with intracellular yeast after challenge, compared to 12-18.8%
for nonimmune donor cells. No significant differences were found in phagocytosis
rates between peripheral macrophages and alveolar macrophages from the same
donor. 101 Both alveolar and peripheral macrophage cultures followed the same
pattern of growth inside nonimmune donor cells and inhibition of intracellular
growth with blastomycosis patients (i.e., immune) cells. Again, no differences
between alveolar and peripheral macrophages were detected for either group.
In these studies,IOO,101 there were higher numbers of B. dermatitidis at the
initial 2-hr time point in the immune macrophage cocultures than with the non-
immune cells. This was a result of the increased rate of phagocytosis by the
immune cells because extracellular yeast were removed before counts were done.
The 12-18% rate ofphagocytosis of B. dermatitidis by nonimmune cells compares
to the rate of 17.6% of macrophages from uninfected normal persons reported by
Drutz and Frey.B 2 Murine peritoneal macrophages were reported by Brummer et
al. 89 ,93 to be unable to ingest B. dermatitidis because of the size of the yeast, but
McDaniel and Cozad88 were able to demonstrate murine macrophage phagocyto-
sis of this yeast.
One reason for increased phagocytosis and intracellular growth inhibition by
alveolar lOl and peripheraPOO,lOI macrophages was thought to be immune activa-
tion of the cells in the coculture experiments as occurred in murine models with
INF--y enhancement. 96,97 Lymphocytes, accounting for 2-5% of the cells of the
human macrophage monolayers, were stimulated by the live B. dermatitidis organ-
isms to secrete cytokines. Lymphocyte activation with immune cells was clearly
shown with the tumbled, suspension culture technique in response to live yeast
challenge. 99 This hypothesis of cytokine activation was examined by treating
macrophages from nonimmune donors with supernatants from cultures ofB-ASWS
stimulated lymphocyte obtained from an immune donor.I° 1 Macrophages in-
duced by active supernatant inhibited intracellular B. dermatitidis growth and had
greater phagocytosis than macrophages treated with supernatants from non-
stimulated cultures of lymphocytes. 101
Paracoccidioidomyeosis has not been studied in the same depth with mono-
eyte and maerophage interaetions as in blastomycosis. Cano et al. 103 studied the
interaetion of P. brasiliensis eonidia with mouse peritoneal macrophages. Conidia
readily transformed to yeast eells after ingestion, and began budding in compari-
son to cell-free culture conditions. When the macrophages were exposed to

cytokines from the spleen cells of immunized mice, there was significant inhibi-
tion of transformation to yeast forms, although some macrophages continued to
support transformation of ingested conidia. This study demonstrates the ability
of conidia to transform within macrophages, and how macrophages activated by
cytokines can inhibit the process. These experiments were also performed with
alveolar macrophages from nonimmunized mice, with killing of 28% of ingested
conidia. I04 Exposure of the macrophages to supernatants of mononuclear cells
from immunized mice increased killing to 73%, demonstrating that activated
alveolar macrophages can effectively kill conidia of P. brasiliensis.
Brummer et al. 97 studied the ability of murine alveolar macrophages to kill
ingested P. brasiliensis yeast cells after in vitro and in vivo exposure to INF-'Y.
Control alveolar macrophages killed 15-25% of the organisms, whereas 45%
killing resulted from interferon exposure. Supernatants of cultures of concana-
valin A lymph node cells or spleen cells also stimulated macrophage killing of this
fungus. In contrast, peritoneal macrophages ingested but did not kill the yeast
ceIls.I 05 Overnight exposure to INF-'Y or lymphokines resulted in 35-55% killing
of the ingested organisms.
8. CONCLUSIONS
B. dermatitidis and P. brasiliensis infections begin with inhalation of conidia

into the lung. Both PMN and alveolar macrophages have been shown to ingest
and kill these conidia. This defense could then keep the human host from
developing subsequent clinical infection with this organism. The numbers of
persons with subclinical or asymptomatic blastomycosis are not as weIl known
as in paracoccidioidomycosis, histoplasmosis, or coccidioidomycosis since skin
tests are not considered to provide as reliable results as other fungal skin tests.
Epidemies ofblastomycosis include persons who are infected but recover without
therapy.I06,107 Lymphocyte reactivity studies of persons from the endemie area of
blastomycosis also confirm that some individuals become infected with this
organism but recover spontaneously.108
In paracoccidioidomycosis, the large preponderance of males with the dis-
ease, despite equal skin-test reactivity for both sexes, suggests a hormonal inftu-
ence. 9 Skin-test surveys suggest that subclinical disease is common in endemic
areas. 13 The elimination of conidia by macrophages and PMN is a likely mecha-
nism for these infections to be halted before clinical disease develops.
Macrophages from humans are able to ingest and inhibit the growth of
intracellular B. dermatitidis. Macrophages from immune miee have likewise been
reported to ingest yeast and eliminate fungal replication. This supports the
concept that cellular immunity is the critical host defense in blastomycosis. The
mechanisms of the killing of B. dermatitidis by macrophages are not fully under-
stood and should be an area of further research.
Macrophages are able to phagocytose and inhibit growth of intracellular P.
brasiliensis. Activation of macrophages increases inhibition of growth, whereas

depression of cellular hypersensitivity correlates with progressive infection in
animal models. Cellular immunity appears to be the central host defense in
paracoccidioidomycosis. Further work may better define the pathogenesis of
depressed cellular immunity seen with the disease.
REFERENCES
1. Chapman, S., 1990, Blastomyces dermatitidis, in: Principles and Practice of Infectious Diseases,
3rd ed., (G. L. MandelJ, R. G. Douglas, and]. E. Bennett, eds.), Churehill Livingstone, New
York.
2. Bradsher, R. w., 1988, Blastomycosis, Infect. Dis. Clin. North Am. 2:877-898.
3. Sarosi, G. A. and Davies, S. F., 1979, Biastomycosis. State of the art, Am. Rev. Respir. Dis.
120:911-938.
4. Larson, D. M., Eckman, M. R., Alber, R. L., and Goldschmidt, V. F., 1983, Primary
cutaneous (inoculation) blastomycosis: An occupational hazard to pathologists, Am. j Clin.
Pathol. 79:253-255.
5. Restrepo, A., 1990, Paracoccidioides brasiliensis, in: Principles and Practice of Infectious Diseases,
3rd ed., (G. L. MandelJ, R. G. Douglas, and]. E. Bennett, eds.), Churchill Livingstone, New
York.
6. Medoff, G., Painter, A., and Kobayashi, G. S., 1987, Mycelial to yeast-phase transition of
dimorphie fungi Blastomyces dermatitidis and Paracoccidioides brasiliensis, j Bacteriol.
169:4055-4060.
7. Schwarz,]. and Salfelder, K., 1977, Blastomycosis: a review of 152 cases, Curr. Top. Pathol.
65:165-200.
8. Rippon,]. w., 1988, Medical Mycology, 3rd ed., W.B. Saunders Co., Philadelphia.
9. Restrepo, A., Salazar, M. E., Cano, L. E., Stover, E. P., Fredman, D., and Stevens, D. A., 1984,
Estrogens inhibit mycelium-to-yeast transformation in the fungus Paracoccidioides brasili-
ensis: Implication for resistance of females to paraeoccidioidomyeosis, Infect. Immun.
46:346-353.
10. Bradsher, R. w., 1987, Water and blastomycosis: Don't blame beaver, Am. Rev. Respir. Dis.
136:1324-1326.
11. Fragoyannis, S., VanWyk, G., and DeBeer, M., 1977, North Ameriean blastomycosis in
South Africa, S. Afr. Med.j 51:169-171.
12. Baily, G. G., Robertson, v.]., Neill, P., Garrido, P., and Levy, L. F. 1991, Blastomycosis in
Africa: Clinieal features, diagnosis and treatment, Rev. Infect. Dis. 13:1005-1008.
13. Conti-Diaz, 1. A., 1972, Skin tests with paracoecidioidin and their importance in Paracoc-
cidioides brasiliensis, Proeeedings of the First Pan Ameriean Symposium, Washington D.C.,
Pan Ameriean Health Organization, p. 197.
14. Bradsher, R. w., Riee, D. C., and Abernathy, R. S., 1985, Ketoconazole therapy of endemie
blastomycosis, Ann. Intern. Med. 103:872-879.
15. Sutliff, W. D. and Cruthirds, T. P., 1973, Blastomyces dermatitidis in eytologie preparations,
16. Davies, S. F. and Sarosi, G. A., 1987, Serodiagnosis of histoplasmosis and blastomyeosis,
17. Klein, B. S., Kuritsky,]. N., Chappell, W. A., Kaufman, L., Green,]., Davies, S. F., Williams,
]. E., and Sarosi, G. A., 1986, Comparison ofthe enzyme immunoassay, immunodiffusion,
and complement fixation tests in detecting antibody in human serum to the A antigen of
Blastomyces dermatitidis, Am. Rev. Respir. Dis. 133:144-148.
BLASTOMYCES DERMATITIDIS AND PARACOCCIDJOIDES BRASILIENSIS 233
18. Klein, B. S., Vergeront,]. M., Kaufman, L., Turner, S., Kumar, U. N., Mathai, G., Bradsher,
R. w., Varkey, B., and Davis,]. P., 1987, Serologie tests for blastomycosis: Assessment dur-
ing a large point source outbreak in Wisconsin,] Infect. Dis. 155:262-268.
19. Lambert, R. S. and George, R. B., 1987, Evaluation of enzyme immunoassay as a rapid
screening test for histoplasmosis and blastomycosis, Am. Rev. Respir. Dis. 136:316-319.
20. Sarosi, G. A., Davies, S. F., Klein, B., Dismukes, w., and Bradsher, R. w., 1986, Recent
deve!opments in blastomycosis, Am. Rev. Respir. Dis. 134:817-818.
21. Klein, B. S. and Jones,]. M., 1990, Isolation, purification, and radiolabe!ing of a nove!
120KD surface protein on Blastomyces dermatitidis yeasts to detect antibody in infected
patients,] Clin. Invest. 85:152-161.
22. Klein, B. S., Sonde!, P. M., and Jones,]. M., 1992, WI-l, a novel 120-kilodalton surface
protein on Blastomyces dermatitidis yeast cells, is a target antigen of cell-mediated immunity
in human blastomycosis, Infect. Immun. 60:4291-4300.
23. Bradsher, R. w., Papay, ]., and Pappas, P. G., 1992, Detection of antibody in human
blastomycosis with a commercially available enzyme immunoassay, Interscience Confer-
ence on Antimicrobial Agents and Chemotherapy, Abstract 1066, Anaheim, California.
24. Restrepo, A., 1987, Immune response to P. brasiliensis in human and animal hosts, in
Current Topics in Medical Mycology, Volume 2 (McGinnis, M. K., ed.) Springer-Verlag, New
York, pp. 239-277.
25. Yarzabal, L. A., Bout, D., Naquira, F., Fruit,]., and Andrieu, S., 1977, Identification and
purification of the specific antigen of Paracoccidioides brasiliensis responsible for immuno-
e!ectrophoresis band E, Sabouraudia 15:79-85.
26. Camagro, Z. P., Unterkireher, C., Campoy, S. P., and Travassos, L. R., 1988, Production of
Paracoccidioides brasiliensis exoantigens for immunodiffusion tests,] Clin. Micro. 26:2147-
2151.
27. Puccia, R., Schenkman, S., Gorin, P. A., and Travassos, L. R., 1986, Exocellularcomponents
of Paracoccidioides brasiliensis: Identification of a specific antigen, Infect. Immun. 53:199-
206.
28. Mendes-Giannini, M. ]. S., Bueno, ]. B., Shikanai-Yasuda, M. A., Ferreira, A. w., and
Masuda, A., 1989, Detection of the 43,000-molecular-weight glycoprotein in sera of
patients with paracoccidioidomycosis,] Clin. Micro. 27:2842-2845.
29. Puccia, R. and Travassos, L. R., 1991, 43-Kilodalton glycoprotein from Paracoccidioides
brasiliensis: Immunochemical reactions with sera from patients with paracoccidioido-
mycosis, histoplasmosis, or Jorge Lobo's disease,] Clin. Micro. 29:1610-1615.
30. Witorsch, P. and Utz,]. P., 1968, North American blastomycosis: A study of 40 patients,
Medicine 47:169-200.
31. Busey,]. F., 1964, Blastomycosis: A review of 198 collected cases in VA hospitals, Am. Rev.
Respir. Dis. 89:659-672.
32. Klein, B. S., Vergeront, ]. M., Weeks, R.]., Kumar, U. N., Mathai, G., Varkey, B., Kauf-
man, L., Bradsher, R. w., Stoebig, ]. F., and Davis, ]. P., 1986, Isolation of Blastomyces
dermatitidis in soil associated with a large outbreak ofblastomycosis in Wisconsin, N. Engl.
] Med. 314:529-'534.
33. Bradsher, R. W. and Alford, R. H., 1981, Blastomyces dermatitidis antigen-induced Iympho-
cyte reactivity in human blastomycosis, Infect. Immun. 33:485-490.
34. Sarosi, G. A. and King, R. A., 1977, Apparent diminution of the blastomycin skin test,
35. Mendes, E. and Rafae!, A., 1971, Impaired de!ayed hypersensitivity in patients with South
American blastomycosis,] Allergy. Clin. Immunol. 47:17-22.
36. Musatti, C. C., Rezkallah, M. T., Mendes, E., and Mendes, N. F., 1976, In vivo and in vitro
evaluation of cell-mediated immunity in patients with paracoccidioidomycosis, Cell Immu-
nol. 24:365-378.
37. Alford, R. H. ancl Goodwin, R. A., 1973, Variations in lymphocyte reactivity to histo-
plasmin during chronic pulmonary histoplasmosis, Am. Rev. Respir. Dis. 108:85-92.
38. Cox, R. A., Vivas,J. R., Gross, A., Lecara, G., Miller, E., and Brummer, E., 1976, In vivo and
in vitro cell-mediated responses in coccidioidomycosis, Am. Rev. Respir. Dis. 114:937-943.
39. Biggar, W D., Menwissen, H. J., and Good, R.A., 1971, Successful defense against Histo-
plasma capsulatum, Arch. Intern. Med. 128:585-587.
40. Williams, D. M., Graybill, J. R., and Drutz, D. J., 1978, Histoplasma capsulatum infection
in nude mice, Infect. Immun. 21:973-977.
41. Longley, R. E. and Cozad, G. C., 1983, Thymosin restoration of cellular immunity to
Blastomyces dermatitidis in T-cell-depleted mice, Infect. Immun. 40:733-740.
42. Pappas, P. G., Pottage, J. C., Powderly, W G., Fraser, V. J., Stratton, C. W, McKenzie, S.,
Tapper, M. L., Chmel, H., Bonebrake, F. C., Blum, R., Shafer, R. W, King, C., and
Dismukes, W E., 1991, Blastomycosis in patients with the acquired immunodeficiency
43. Fraser, V. J., Keath, E. J., and Powderly, W G., 1991, Two cases of blastomycosis from a
common source; DNA restrietion analysis to identify strains,J. Infect. Dis. 163:1378-1380.
44. Benard, G., Bueno,J. P., Yamashiro-Kanashiro, E. H., Shikanai-Yasuda, M. A., Del Negro,
G. M. B., Melo, N. T., Sato, M. N., Neto, V. A., Shiroma, M., and Duarte, A. J. S., 1990,
Paracoccidioidomycosis in a patient with HIV infection: Immunological study, Trans.
R. Soc. Trop. Med. Hygiene 84:151-152.
45. Goldani, L. Z., Coelho, I. C. B., Machado, A. A., and Martinez, R., Paracoccidioidomycosis
and AIDS, Scand. J. Infect. Dis. 23:393.
46. Spencer, H. D. and Cozad, G. C., 1973, Role of delayed hypersensitivity in blastomycosis of
mice, Infect. Immun. 7:329-334.
47. Cozad, G. C. and Chang, C.T., 1980, Cell-mediated immunoprotection in blastomycosis,
48. Deepe, G. S., Jr., Taylor, C. L., and Bullock, W E., 1985, Evaluation of inflammatory
response and cellular immune response in a murine model of disseminated blastomycosis,
49. Iabuki, K. and Montenegro, M. K., 1979, Experimental paracoccidioidomycosis in the
Syrian hamster: Morphology, ultrastructure and correlation of lesions with presence of
specific antigens and serum levels of antibodies, Mycopathologia 67:131-141.
50. Rezkallah-Iwasso, M. T., Mota, N. G. S., Gomes, M. C. Go, and Montenegro, M. R., 1983,
Effect of levamisole on experimental paracoccidioidomycosis in the Syrian hamster:
Immunologie and histopathologie correlation, Mycopathologia 84:171-180.
51. Robledo, M. A., Graybill,J. R., Ahrens,J. Restrepo, A., Drutz, D.J., and Robledo, M.,1982,
Host defense against experimental paracoccidioidomycosis, Am. Rev. Respir. Dis. 125:563-567.
52. Defaveri, J., Martin, L. C., and Franeo, M., 1989, Histological and ultrastructural study of
the inflammation evoked by Paracoccidioides brasiliensis antigen in previously immunized
mice, Mycopathologia 105:53-58.
53. Defaveri,J., Rezkallah-Iwasso, M.T., and Franeo, M. F., 1992, Pulmonary paracoccidioido-
mycosis in immunized mice, Mycopathologia 119:1-9.
54. Restrepo, A., Salazar, M. E., Cano, L. E., and Patino, M. M., 1986, A technique to collect and
dislodge conidia produced by Paracoccidioides brasiliensis mycelial form, J. Med. Veto Mycol.
24:247-250.
55. McEwen,J. G., Bedoya, v., Patino, M. M., Salazar, M. E., and Restrepo, A., 1987, Experi-
mental murine paracoccidioidomycosis induced by the inhalation of conidia, J. Med. Veto
Mycol. 25:165-175.
56. Restrepo, S., Tobon, A., Trujillo, J., and Restrepo, A., 1992, Development of pulmonary
fibrosis in mice during infection with Paracoccidioides brasiliensis conidia, J. Med. Veto Mycol.
30:173-184.
BLASTOMYGES DERMATITIDIS AND PARAGOGGIDIOIDES BRASILIENSIS 235
57. Cox, R. A., 1979, Cross-reactivity between antigens of C. immitis, H. capsulatum, and B.
dermatitidis in lymphocyte transformation assays, Infeet. Immun. 25:932-938.
58. Cox, R. A. and Larsh, H. W, 1974, Isolation of skin test-active preparations from yeast-
phase cells of Blastomyees dermatitidis, Infect. Immun. 10:42-47.
59. Cox, R. A. and Larsh, H. W, 1974, Yeast and mycelial-phase antigens of Blastomyees
dermatitidis: Comparison using disc gel electrophoresis, Infoet. Immun. 10:48-53.
60. Deighton, F., Cox, RA., Hall, N. ]., and Larsh, H. W, 1977, In vivo and in vitro cell-
mediated immune responses to a cell wall antigen of Blastomyees dermatitidis, Infeet. Immun.
15:429-435.
61. Bradsher, R. W, 1984, Development of specific immunity in patients with pulmonary and
extrapulmonary blastomycosis, Am. Rev. Respir. Dis. 129:430-434.
62. Perac;:oli, M. T S., Montenegro, M. R., Soares, A. M. V. C., and Mota, N. G. S., 1990,
Transfer of cell-mediated immunity to Paracoeeidioides brasiliensis in hamsters with dialys-
able leukocyte extracts,] Med. Veto Myeol. 28:35-46.
63. Castaneda, E., Brummer, E., Pappagianis, 0., and Stevens, 0. A., 1988, Regulation of
immune responses by T suppressor cells and by serum in chronic paracoccidioidomycosis,
Gell Immunol. 117:1-11.
64. Jimenez-Finkel, B. E. and Murphy,]. W, 1988, Characterization of efferent T suppressor
cells induced by Paracoeeidioides brasiliensis-specific afferent T suppressor cells, Infoct.
Immun. 56:744-750.
65. Jimenez-Finkel, B. E. and Murphy,]. W, 1988, Induction of antigen-specific T suppressor
cells by soluble Paraeoeeidioides brasiliensis antigen, Infect. Immun. 56:734-743.
66. Ferreira-Da-Cruz, M. F., Galvao-Castro, B., and Daniel-Ribeiro, C. T., 1991, Sensitive
immunoradiometric assay for the detection of Paraeoeeidioides brasiliensis antigens in human
sera,] Glin. Miero. 29:1202-1205.
67. Freitas-Da-Silva, G. and Roque-Barreira, M. C., 1992, Antigenemia in paracoccidioido-
mycosis,] Glin. Miero. 30:381-385.
68. Chequer-Bou-Habib, 0., Ferreira-Da-Cruz, M. F., and Galvao-Castro, B., 1992, Immuno-
suppressive effect of paracoccidioidomycosis sera on the proliferative response of normal
mononuclear cells, Myeopatlwlogia 119:65-71.
69. Perac;:oli, M. T S., Soares, A. M. V. C., Mendes, R. P., Marques, S. A., Pereira, P. C. M., and
Rezkallah-Iwasso, M. T., 1991, Studies of natural killer cells in patients with paracoccidi-
oidomycosis,] Med. Veto Mycol. 29:373-380.
70. Mota, N. G. S., Perac;:oli, M. T S., Mendes, R. P., Gattass, C. R, Marques, S. A., Soares,
A. M. V. C., Izatto, L C., and Rezkallah-Iwasso, M. T, 1988, Mononuclear cell subsets in
patients with different dinical forms of paracoccidioidomycosis,f Med. Veto Myeol. 26:105-111.
71. Silva, C. L. and Figueiredo, F., 1991, Tumor necrosis factor in paracoccidioidomycosis
patients,] Infect. Dis. 164:1033-1034.
72. Sixby,]. w,. Fields, B. T, Sun, C. N., Clark, RA., and Nolan, C. M., 1979, Interactions
between human granulocytes and Blastomyees dermatitidis, Infoct. Immun. 23:41-44.
73. Brummer, E. and Stevens, 0. A., 1982, Opposite effects of human monocytes, macro-
phages, and polymorphonuclear neutrophils on replication of Blastomyees dermatitidis in
vitro, Infoct. Immun. 36:297-303.
74. Brummer, E. and Stevens, 0. A., 1983, Enhancing effect of murine PMN on the multiplica-
tion of B. dermatitidis in vitro and in vivo, Glin. Exp. Immun. 54:587-594.
75. Brummer, E., Sugar, A. M., and Stevens, 0. A., 1984, Immunological activation of poly-
morphonuclear neutrophils for fungal killing: Studies with murine cells and Blastomyees
dermatitidis in vitro,] Leuk. Biol. 36:505-520.
76. Brummer, E., McEwen, ]. G., and Stevens, 0. A., 1986, Fungicidal activity of murine
infiammatory polymorphonuclear neutrophils: Comparison with murine peripheral blood
PMN, Clin. Exp. Immunol. 66:681-690.
77. Brummer, E. and Stevens, D. A., 1984, Activation of murine poly-morphonucIear neutro-
phils for fungicidal activity with supernatants from antigen-stimulated immune spleen cell
cultures, Infect. Immun. 45:447-452.
78. Brummer, E., Sugar, A. M., and Stevens, D. A., 1985, Enhanced oxidative burst in
immunologically activated but not elicited PMN leukocytes correlates with fungicidal
activity, Infeet. Immun. 49:396-401.
79. Morrison, C.]., Isenberg, R. A., and Stevens, D. A., 1988, Enhanced oxidative mechanisms
in immunologically activated versus elicited polymorphonucIear neutrophils: Correlations
with fungicidal activitY,J Med. Mierobiol. 25:115-121.
80. Morrison, C.]., Brummer, E., and Stevens, D. A., 1987, Effects of a local immune reaction
on peripheral blood polymorphonucIear neutrophil microbicidal function: Studies with
fungal targets, Cell Immunol. 110:176-182.
81. Morrison, C. ]. and Stevens, D. A., 1989, Enhanced killing of Blastomyees dermatitidis by
gamma interferon-activated murine peripheral blood polymorphonucIear neutrophils,
Int. J Immunopharmac. 11:855-862.
82. Drutz, D. ]. and Frey, C. L., 1985, Intracellular and extracellular defenses of human
phagocytes against B. dermatitidis conidia and yeasts,J Lab. Clin. Med. 105:737-750.
83. Schaffner, A., Davis, C. E., Schaffner, T., Market, M., Douglas, H., and Braude, A., 1986, In
vitro susceptibility of fungi to killing by PMN discriminates between primary pathogenicity
and opportunism,J Clin. Invest. 78:511-524. .
84. Goihman-Yahr, M., Essenfeld-Yahr, E., Albornoz, M. C., Yarzabal, L., G6mez, M. H., San
Martin, B., Ocanto, A., Gil, F., and Convit,]., 1980, Defect of in vitro digestive ability of
polymorphonucIear leukocytes in paracoccidioidomycosis, lrifeet. Immun. 28:557-566.
85. Goihman-Yahr, M., Rothenberg, A., Bretafta, A., Isturiz, G., Rosquete, R., Avila-Millan,
Viloria, N., Borges, N. S., Carrasquero, M., Fernandez, B. P., Maring, B. S., Roman, A.,
G6mez, M. H., Pereira, ]., and Molina, T., 1989, Digestion of killed Paracoeeidioides
brasiliensis by neutrophils, Myeopatlwlogia 106:53-58.
86. Calich, V. L. G., Kipnis, T. L., Mariano, M., Neto, C. F., and Silva, W. 0., 1979, The activa-
tion of the complement system by Paracoeeidioides brasiliensis in vitra: Its opsonic effect and
possible significance for an in vivo model of infection, Clin. Immunol. Immunopath. 12:
20-30.
87. Calich, V. L. G., Vaz, C. A. C., and Burger, E., 1985, PMN chemotactic factor produced by
glass-adherent cells in the acute inßammation caused by Paracoeeidioides brasiliensis, Br.
J Exp. Path. 66:57-65.
88. Scillian,].]., Cozad, G. C., and Spencer, H. D., 1974, Passive transfer of delayed hypersen-
sitivity to Blastomyees dermatitidis between mice, Infoct. Immun. 10:705-711.
89. McDaniel, L. S. and Cozad, G. C., Immunomodulation by Blastomyees dermatitidis: Func-
tional activity of murine peritoneal macrophages, lrifeet. Immun. 40:733-740.
90. Brummer, E., Morozumi, P. A., and Stevens, D. A., 1980, Macrophages and fungi: In vitro
effects of method of macrophage induction, activation by different stimuli, and soluble
factors on Blastomyees, J Retieuloendothelial Soe. 28:507-518.
91. Brummer, E., Morozumi, P. A. Philpott, D. E., and Stevens, D. A., 1981, Virulence of fungi:
Correlation of virulence of B. dermatitidis in vitro with escape from macrophage inhibition
of replication in vitro, Infoct. Immun. 32:864-871.
92. Brummer, E., Hanson, L. H., and Stevens, D. A., 1991, Kinetics and requirements for
activation of macrophages for fungicidal activity: Effect of protein synthesis inhibitors and
immunosuppressants on activation and fungicidal mechanism, Cell Immunol. 132:236-245.
93. Brummer, E., Morrison, C.]., and Stevens, D. A., 1985, Recombinant and natural gamma-
interferon activation of macrophages in vitro: Different dose requirements for induction of
killing activity against phagocytizable and nonphagocytizable fungi, Inftet. Immun. 49:
724-730.
94. Brummer, E. and Stevens, D. A., 1987, Fungicidal mechanisms of activated macrophages:
Evidence for nonoxidative mechanisms for killing of B. dermatitidis, Infect. Immun. 55:3221-
3224.
95. Sugar, A. M., Brummer, E., and Stevens, D. A., 1986, Fungicidal activity of murine broncho-
alveolar macrophages against Blastomyces dermatitidis,J Med. Microbiol. 21:7-11.
96. Brummer, E. and Stevens, D. A., 1987, Activation of pulmonary macrophages for fungici-
dal activity by gamma-interferon or Iymphokines, Clin. Exp. Immunol. 70:520-528.
97. Brummer, E., Hanson, L. H., Restrepo, A., and Stevens, D. A., 1988, In vivo and in vitro
activation of pulmonary macrophages by IFN--y for enhanced killing of Paracoccidioides
brasiliensis or Blastomyces dermatitidis, J Immunol. 140:2786-2789.
98. Sugar, A. M. and Pieard, M., 1991, Macrophage and oxidant-mediated inhibition of the
ability of live B. dermatitidis conidia to transform to the pathogenie yeast phase: Implica-
tions for the pathogenesis of dimorphie fungal infections,J Infect. Dis. 163:371-375.
99. Bradsher, R. w., 1984, Live Blastomyces dermatitidis yeast-induced responses of immune and
nonimmune human mononuc1ear cells, Mycopatlwlogia 87:159-166.
100. Bradsher, R. w., Ulmer, C., Marmer, D. J., Townsend, J. w., and Jacobs, R. F., 1985,
Intracellular growth and phagocytosis of Blastomyces dermatitidis by monocyte-derived
macrophages from previously infected and normal subjects,J Infect. Dis. 151:57-64.
101. Bradsher, R. w., Balk, R. A., and Jacobs, R. F., 1987, Growth inhibition of Blastomyces
dermatitidis in alveolar and peripheral macrophages from patients with blastomycosis,
102. Green, G. M.,Jakab, G. H., Low, R. B., and Davis, G. S., 1977, Defense mechanisms ofthe
respiratory membrane, Am. Rev. Respir. Dis. 115:479-514.
103. Cano, L. E., Brummer, E., Stevens, D. A., and Restrepo, A., 1992, Fate of conidia of
Paracoccidioides brasiliensis after ingestion by resident macrophages or cytokine-treated
macrophages, Infect. Immun. 60:2096-2100.
104. Cano, L. E., Arango, R., Salazar, M. E., Brummer, E., Stevens, D. A., and Restrepo, A.,1992,
Killing of Paracoccidioides brasiliensis conidia by pulmonary macrophages and the effect of
cytokines,J Med. Veto Mycol. 30:161-168.
105. Brummer, E., Hanson, L. H., and Stevens, D. A., 1988, Gamma-interferon activation of
macrophages for killing of Paracoccidioides brasiliensis and evidence for nonoxidative mecha-
nisms, Int. J Immunopharmacol. 10:945-952.
106. Klein, B. S., Vergeront, J. M., DiSalvo, A. F., Kaufman, L., and Davis, J. P., 1987, Two
outbreaks of blastomycosis along rivers in Wisconsin: Isolation of Blastomyces dermatitidis
from riverbank soi! and evidence of its transmission along waterways, Am. Rev. Respir. Dis.
136:1333-1338.
107. Tosh, F. E., Hammerman, K. J., Weeks, R. J., and Sarosi, G. A., 1974, A common source
epidemie of North American blastomycosis, Am. Rev. Respir. Dis. 109:525-529.
108. Vaaler, A. K., Bradsher, R. w., and Davies, S. F., 1990, Evidence of subc1inical blastomycosis
in forestry workers in northern Minnesota and northern Wisconsin, Am. J Med. 89:
470-475.
14
The Immunology of
Coccidioidomycosis
STANLEY C. DERESINSKI
1. INTRODUCTION
Coccidioidomycosis is a disease of protean manifestations. Of those infected with

Coccidioides immitis (CI), only approximately 40% become ill and, in most of these,
the illness consists of prolonged and often severe, but self-limited, influenza-like
symptoms, commonly with pneumonitis. Approximately 5% are left with pulmo-
nary residuals such as nodules and cavities which may be variably symptomatic.l
Fewer than 1% of those infected develop clinically apparent extrapulmonary
dissemination.
Some evidence suggests that subclinical dissemination may, however, be
common, as it is with infection with Mycobacterium tuberculosis. Thus, chorioretinal
lesions can be found in as many as 40% of patients without other clinical evidence
of dissemination. 2•3 CI can frequently, in addition, be recovered from urine in
patients thought clinically to have only pulmonary infection. 4 •5
Pneumonitis caused by infection with CI is usually self-limited. In some
cases, however, it may be acutely progressive, rapidly leading to respiratory
failure.6-8 A more chronically progressive form of pneumonitis also occurs. 9
Infection mayaiso result in pulmonary parenchymal cavitation; cavitary disease
may then take a highly variable course.lO-12
STANLEY C. DERESINSKI • Department of Medieine. Stanford University School of Medi-

eine, Stanford, California 94305; Department of Medieine, Santa Clara Valley Medical Center,
San Jose, California 95128; AIDS Community Research Consortium, Redwood City, California
94062.
Pulmonary Infoctions and Immunity, edited by Herman ChmeI et al. Plenum Press, New York, 1994.
239
240 STANLEY C. DERESINSKI
The risk of clinieally important dissemination of eoecidioidal infeetion varies

among different groups of individuals. Filipinos may have a risk of dissemination
that is 10-175 times that of non-Hispanie Caueasians. Blaeks and Hispanies also
have an inereased risk of dissemination, although it is of lesser magnitude.I 3 ,14
Infeeted patients with blood group Band those who are HLA-A9 positive are at
inereased risk of dissemination, but these phenotypes are overrepresented in
blaeks and Filipinos, making interpretation diffieult.I 5,16 Pregnaney abrogates the
apparent relative proteetion of white females from dissemination,I',18 possibly
beeause of the trophie effeets of estrogen and progesterone on CI.19,20 The rate of
growth and release of endospores of CI in vitra is stimulated, via binding to
eytosol proteins, by testosterone, 17 ß-estradiol and progesterone. In the ease of
the lauer two hormones, this is aeeomplished by eoneentrations that are physio-
logie during pregnaney,19,20 and may thus aeeount for the inereased risk of
dissemination du ring late gestation.
Immunosuppressive disease or therapies that affeet eellular immune re-
sponse also inerease the risk of development of severe disease. 21 ,22 Thus, patients
with aequired immunodeficieney syndrome exposed to CI appear to be at
inereased risk of adverse outeome. 22 Although patients with diabetes mellitus
do not appear to be at inereased risk of disseminated disease, aneedotal evidenee
suggests that they are at inereased risk of pulmonary eavitation, whieh may
beeome ehronic and progressive.
2. COCCIDIOIDESIMMITIS
CI is a dimorphie fungus whose taxonomy remains unclear. The fungus

exists as a myeelium in soil and in routine eulture on solid agar, but as an
endosporulating spherule in the infeeted host. This dimorphism is not, as with
Histoplasma capsulatum and Blastomyces dermatitidis, temperature dependent. An
important faetor in the eonversion from the hyphae to the endosporulating
spherules is, instead, an ambient CO 2 tension of 20-60 torr. 23 Inhaled arthro-
eonidia, whieh have become disartieulated from myeelial colonies of CI in the soil
and become airborne, are inhaled by the unlueky mammalian host. These barrel-
shaped fragments are usually approximately 2-3 fJ.m by 4-5 fJ.m in size and are
thus, although not optimally sized, eapable of reaehing the pulmonary alveoli.
Primate studies indieate that infeetion may result from inhalation of as few as
10 arthroconidia. 24
The virulenee meehanisms of CI are poorly understood. An enzyme with
elastase (and to a lesser degree, collagenase) aetivity has been isolated from eul-
ture filtrates of the parasitic phase of CI. This enzyme is possibly identieal to, or a
subunit of, the serine proteinase thought to be important in the dissolution of the
segmentation apparatus during spherule-endospore reproduetion, and its aetiv-
ity peaks at the time of endospore release. Its elastase aetivity allows it to
enzymatieally degrade eonneetive-tissue matrix maeromoleeules and it may thus
have an important role in the pathogenesis of pulmonary eoecidioidomycosis.
IMMUNOLOGY OF COCCIDIOIDOMYCOSIS 241
Breakdown of pulmonary connective-tissue matrix at the time of endospore

release may allow intra- and extrapulmonary spread of the organism and may
contribute to progressive loss of pulmonary parenchyma. 25 Furthermore, the
enzymatic breakdown products of elastin possess chemotactic activity and may
thus further contribute to the inflammatory response. 26 The influx of phagocytic
cells may lead to the release of inflammatory mediators, oxidants, and additional
enzymes that may contribute further to parenchymal damage.
3. IMMUNE RESPONSE
The initial inflammatory response to arthroconidia consists of an influx of

polymorphonuclear leukocytes (PMNLs)27 that appear in response to an organism-
derived chemotaxinogen. 28 The arthroconidia, however, may resist ingestion by
these cells because of the presence of an outer cell-walllayer with antiphagocytic
properties. 29
Within 72 hr of reaching the alveoli, the surviving arthroconidia, influenced
by the local CO2 tension, convert to the spherule-endospore phase. With this
morphological conversion, the inflammatory infiltrate converts to one consisting
predominantly of mononuclear cells and, eventually, well-formed granulomas may
result. The spherules reproduce by endosporulation. Each spherule may produce
hundreds of endospores that are released on rupture of the spherule wall. As
spherule rupture and endospore release occurs, PMNLs appear in the infiltrate
once again.3° As the endospores enlarge and become thick-walled spherules, the
PMNLs once again disappear and the infiltrate resumes its mononuclear nature.
The transient appearance of neutrophils at the time of spherule rupture is
consistent with the observation that spherule lysates are chemotaxigenic for these
cells in a complement-dependent manner.3 1•32 The ability of PMNLs to damage
the fungus significantly seems limited, but is not nonexistent. Thus, while PMNLs
of patients with chronic granulomatous disease are ineffective, those of normal
humans inhibit the incorporation of the chitin precursor, N-acetylglucosamine,
into cell-wall chitin of arthroconidia. 33 Lysozyme, which may possibly damage the
surface of the spherule wall, is released by PMNLs.29.34 The effect of PMNLs on
spherules, however, appears to be even more limited, possibly as a result of the
size of these fungal elements and the presence of an extracellular glucoprotein
fibrillar matrix that has been proposed to be antiphagocytic. 35 In fact, the influx
of PMNLs may be counter-productive because these cells may play a role in
maintaining the organism in the spherule-endospore phase in vivo,36,37 although
this has been disputed.
Murine experiments have confirmed clinical observations indicating the
primacy of the cellular immune response in controlling infection with CI. Adop-
tive transfer of immunity to CI in mice is T-Iymphocyte dependent. 38 The
administration of recombinant human interleukin 2 (rh IL-2), however, had no
effect in a murine model of coccidioidomycosis. 39 The susceptibility of inbred
strains of mice to infection with CI appears to be under the control of a single
gene that is expressed by spleen eells and is assoeiated with an aequired suppres-
sion of eeIl-mediated immunity.40-42
As stated above, mature spherules eannot be ingested by professional phago-
eytes, possibly beeause of their size or their extraeellular fibrillar matrix. Arthro-
eonidia and endospores, however, are readily phagoeytized by murine as weIl as
by primate (including alveolar) maerophages, as weIl as by human polymorpho-
nuclear leukoeytes. 43 .35 Possibly as the result of failure of phagosome-Iysosome
fusion, however, killing of the ingested organisms fails to result. 44 ,45 When
aetivated prior to infection with CI by ineubation with erude supernatants
obtained from sensitized lymphocytes exposed to spherule-derived antigen,
however, phagosome-Iysosome fusion may be enhaneed and killing of CI results. 46
Cultured murine alveolar and peritoneal maerophages aetiva.ted by reeombinant
INF-'Y are also eapable of restricting the growth of CI.47 Studies involving
experimental CI infeetion of eongenitally athymie nude, as weIl as beige
(C57B1I6J/bgj/bgj) mice, provide evidenee that the primary effector cells in
resistance to this fungus are macrophages and that PMNLs mayaiso playa role. 48
Peripheral blood lymphocytes from patients with controlled CI infeetion
res pond in vitro to soluble antigens of CI as weIl as to intact spherules, arthro-
spores, and endospores. 49 Human natural killer eells inhibit the growth of
endospores and "young" spherules. 5o Human glass-adherent peripheral mono-
nuclear cells ingest endospores51,52 and arthrospores and reduce in vitro uptake
of N-aeetylglucosamine into the ceIl-waIl chitin of the lauer while inhibiting their
growth. 43 In one study, 25% of arthroconidia were killed after ingestion by
human peripheral blood mononuclear eells whereas PMNLs were largely ineffec-
tive (5% killing) in this regard. Killing of phagocytized mature spherules by
mononuclear cells was only one-third of that of arthroeonidia. Preincubation with
either INF--y or TNF-o: failed to enhanee killing of ingested arthroconidia by
these eells. 53 The lauer observation is, however, diffieult to reeoncile with studies
involving murine ceIls47 and the observation that, when aetivated by INF-'Y or
TNF-o:, either alone or in combination, human peripheral blood mononuclear
cells are able to inhibit the intracellular growth of endospores. 52
Patients with disseminated coccidioidomyeosis often have, in addition to
defeetive eutaneous delayed hypersensitivity, impaired lymphoeyte response to
coceidioidal antigens. 54-56 Loss of immunologie responsiveness in CI-infeeted
mice appears to be a result of the aetivation of a splenie suppressor cell population
induced by circulating coccidioidal antigen.57 Consistent with the observation
that high levels of serum antibody to CI correlates with disease dissemination,
suppression of the invitro response of lymphocytes obtained from patients with
coccidioidomycosis may be media ted by IgG, either alone or in immune com-
plexes. 58
Humans with CI infection who fail to develoP delayed hypersensitivity to
coccidioidal antigens have a poor prognosis. These same patients often have high
titers of complement-fixing antibody to CI chitinase,59 indicating that antibody
is not protective. In fact, the height of the complement-fixing antibody titer tends
to correlate with the extent of infection. Furthermore, patients with dissemination
and high antibody titers often have absent delayed hypersensitivity. Such absent
response is often selective, in that delayed hypersensitivity responses to control
antigens is preserved. 54 ,60
4. DETERMINANTS OF OUTCOME OF COCCIDIOIDAL

INFECfION-HYPOTHESIS
These observations indicate that the outcome of infection with CI appears to

depend on the development of an effective cellular immune response. Thus, the
evidence reviewed here confirms that the T lymphocyte and macrophage are key
to the immune response to CI. CD4 + lymphocytes (T-helper cells) function
both in the system of cellular immunity and as helper cells in the production of
antibody by B lymphocytes. Evidence now indicates that T-helper cells are
composed of at least two distinct subsets called T-helper 1 (Thl) and T-helper 2
(Th2). The former cells produce IL-2 and INF--y and are thus critically important
in cellular immune responses, including macrophage activation and delayed
hypersensitivity reactions. Th2 cells, on the other hand, which produce IL-4,
IL-5, and IL-lO, serve a hel per function for antibody production by B lympho-
cytes. IL-lO, in particular, is an immunosuppressive cytokine which, among other
things, interferes with antigen presentation by macrophages. 61
Current evidence suggests that Thl cells arise from undifferentiated (naive)
T cells under the influence of IL-l2, a cytokine derived from B lymphocytes and
macrophages, which is a stimulator of INF--y production. IL-l2 (also called NK
cell stimulating factor), by virtue of its ability to drive naive T lymphocytes to
become Thl cells, at least in the absence of significant countervailing press ure
from IL-4, is therefore critical to the development of an effective cellular immune
response.
Although supporting data are not yet available, it is likely that the determina-
tion of the ultimate outcome of coccidioidal infection depends on the type of
T-helper cell activity that predominates initially. An analogy with studies of
murine infection with Leishmania major may be appropriate here. Leishmaniasis is
a protozoan disease which, like coccidioidomycosis, has protean manifestations,
ranging from a spontaneously healing single cutaneous ulceration to widespread
viscerally disseminated infection (kala azar). Although some of this clinical
variation is the result of varying degrees and types of virulence of the individual
species causing this infection, the outcome of infection with L. donovani (the
etiologic agent of kala azar and an obligate intracellular parasite) is also highly
variable, and appears to be determined by the nature ofthe lymphocyte response
to the infection. Mice that respond with a Thl response to infection with L. major
survive while those that develop a dominant Th2 response die. It is likely that a
similar result is applicable to coccidioidomycosis. In this case a Th2 response,
with its attendant predominant B-cell response and lack of cellular immune
response, would dominate in the patient destined to suffer from dissemination of
CI infection. Such a patient would have, as a consequence, a high complement-
fixing antibody titer to CI antigen and absent delayed hypersensitivity with

specific anergy to CI. The patient destined to control his or her infection would,
on the other hand, develop a dominant Thl response with its attendant effective
cellular immunity.
This hypothesis, while attractive, requires experimental confirmation. If
proven true, however, new avenues of therapeutic immunomodulatory interven-
tion in patients with coccidioidomycosis will become available.
AcKNOWLEDGMENT. Barbara Houghton, M.L.S., Medical Staff Librarian, Se-

quoia District Hospital, Redwood City, CA 94062.
REFERENCES
1. Smith, C. E., Beard, R. R., Whiting, E. G., and Rosenberger, H. G., 1946, Varieties of
coccidioidal infection in relation to the epidemiology and control of the disease, Am. J Publ.
Health 36:1394-1402.
2. Blumenkranz, M. S. and Stevens, D. A., 1980, Endogenous coccioiodal endophthalmitis,J
Ophthalmol. 87:974-984.
3. Rodenbiker, H. T., Ganley, J. P., Galgiani, J. N., and Axline, S. G., 1981, Prevalence of
chorioretinal scars associated with coccidioidomycosis, Arch. Ophthalmol. 99:71-75.
4. Petersen, E. A., Friedman, B. A., and Crowder, E. D., 1976, Coccidioiduria: Clinical
significance, Ann. Intern. Med. 85:34-38.
5. DeFelice, R., Wieden, M. A., and Galgiani, J. N., 1982, The incidence and implications of
coccidioiduria, Am. Rev. Respir. Dis. 125:'49-52.
6. Bayer, A. S., 1981, Fungal pneumonias: Pulmonary coccidioidal syndromes (Part I). Primary
and progressive primary coccidioidal pneumonias-diagnostic, therapeutic, and prognostic
considerations, Chest 79:575-583.
7. Bayer, A. S., 1981, Fungal pneumonias: pulmonary coccidioidal syndromes (Part 2). Miliary,
nodular, and cavitary pulmonary coccidioidomycosis: Chemotherapeutic and surgical con-
siderations, Chest 79:686-691.
8. Larsen, R. A., Jacobson, J. A., Morris, A. H., and Benowitz, B. A., 1985, Acute respiratory
failure caused by primary pulmonary coccidioidomycosis: Two case reports and a review
of the literature, Am. Rev. Respir. Dis. 131:797-799.
9. Bayer, A. S., Yoshikawa, T. T., and Guze, L. B., 1979, Chronic progressive coccidioidal
pneumonitis: Report of six cases with clinical, roentgenographic, serologie and therapeutic
features, Arch. Intern. Med. 139:536-540.
10. Cunningham, R. T. and Einstein H., 1982, Coccidioidal pulmonary cavities with rupture,]
Thorac. Cardiovasc. Surg. 84:172-177.
1l. Winn, W. A., 1968, A long term study of 300 patients with cavitary-abscess lesions ofthe lung
of coccidioidal origin: An analytical study with special reference to treatment, Dis. Chest
54(Suppl 00. 1):12-16.
12. Putnam, J. S., Harper, W. K., Greene, J. F., Jr., Nelson, K. G., and Zurek, R. C., 1975,
Coccidioides immitis: A rare case of pulmonary mycetoma, Am. Rev. Respir. Dis. 112:733-738.
13. Smith, C. E., Beard, R. R., Whiting, E. G., and Rosenberger, H. G., 1946, Varieties of
coccidioidal infection in relation to the epidemiology and control of the disease, Am. J Publ.
Health. 36:1394-1402.
14. Hugenholtz, P., 1956, Climate and coccidioidomycosis, in: Proceedings of the Symposium
on Coccidioidomycosis. U.S. Public Health Service Publication No. 575, pp. 136-143.
15. Deresinski, S. C., Pappagianis, D., and Stevens, D. A., 1979, Association of ABO blood group
and outcome of coccidioidal infection, Sabouraudia 17:261-264.
16. Scheer, M., Opelz, G., Terasaki, P., and Hewitt, w., 1973, The association of disseminated
coccidioidomycosis and histocompatibility type. 13th Interscience Conference on Anti-
microbial Agents and Chemotherapy, American Society for Microbiology, Abstract No. 157.
17. Smale, L. E. and Birsner,j. w., 1949, Maternal deaths from coccidioidomycosis,J Am. Med.
Assoc. 140:1152-1154.
18. Wack, E. E., Ampel, N. M., Galgani,j. N., and Bronmann, 0. A., 1988, Coccidioidomycosis
during pregnancy: An analysis often cases among 47,120 pregnancies, Chest 94:376-379.
19. Powell, B. L., Drutz, D.j., Huppert, M., and Sun, S. H., 1983, Relationship ofprogesterone-
and estradiol-binding proteins in Coccidioides immitis to coccidioidal dissemination in preg-
nancy, Infect. Immun. 40:478-485.
20. Powell, B. L. and Drutz, 0. A., 1984, Identification of a high-affinity binder for estradiol and
a low-affinity binder for testosterone in Coccidioides immitis, Infect. Immun. 45:784-786.
21. Cohen, I. M., Galgiani,j. N., Potter, D., and Ogden, D. A., 1982, Coccidioidomycosis in renal
replacement therapy, Arch. Intern. Med. 142:489-494.
22. Bronnimann,D. A., Adam, R. 0., Galgiani,j. N., Habib, M. P., Petersen, E. A., Porter, B., and
Bloom, j. w., 1987, Coccidioidomycosis in the acquired immunodeficiency syndrome, Ann.
Intern. Med. 106:372-379.
23. Klotz, S. A., Drutz, 0. j., Huppert, M., Sun, S. H., and DeMarsh, P. L., 1984, Thi: critical
role of CO2 in the morphogenesis of Coccidioides immitis in cell-free subcutaneous chambers,
J Infect. Dis. 150:127-134.
24. Converse, j. L. and Reed, R. E., 1966, Experimental epidemiology of coccidioidomycosis,
Bacteriol. Rev. 30:678-695.
25. Resnick, S., Pappagianis, D., and McKerrow,j. H., 1987, Proteinase production by the para-
sitic cyde of the pathogenic fungus Coccidioides immitis, Infect. Immun. 55:2807-2815.
26. Hunninghake, G. w., Davidson, j. M., Rennard, S., Szapiel, S., Gadek, j. E., and Crystal,
R. G., 1981, Elastin fragments attract macrophage precursors to diseased sites in pulmonary
emphysema, Science 212:925-927.
27. Savage,D. C. and Madin, S. H., 1968, Cellular responses in lungs of immunized mice to
intranasal infection with Coccidioides immitis, Sabouraudia 6:94-102.
28. Galgiani,J. N., Isenberg, R.A., and Stevens, D. A., 1978, Chemotaxigenie activity of extracts
from the mycelial and spherule phases of Coccidioides immitis for human polymorphonudear
leukocytes, Infect. Immun. 21:862-865.
29. Drutz, D. j. and Huppert, M., 1983, Coccidioidomycosis: Factors affecting the host-parasite
interaction,J Infect. Dis. 147:372-390.
30. Ophüls, W. and Moffitt, H. C., 1900, A new pathogenie mould (formerly described as a
protozoan (Coccidioides immitis pyogenes): Preliminary report, Philadelphia Med. J 5:1471-
1472.
31. Galgiani, j., Isenberg, R.A., and Stevens, 0. A., 1978, Chemotaxigenic activity of extracts
from the mycelial and spherule phases of Coccidioides immitis for human polymorphonudear
leukocytes, Infect. Immun. 21:862-865.
32. Galgiani, j. N., Yam, P., Petz, L. D., Williams, P. L., and Stevens, 0. A., 1980, Complement
activation by Coccidioides immitis: In vitro and dinical studies, Infect. Immun. 28:944-949.
33. Galgiani, j. N., Payne, C. M., and jones, j. E, 1984, Human polymorphonudear-Ieukocyte
inhibition of chitin precursors into mycelia of Coccidioides immitis, J Infect. Dis. 149:
404-411.
34. Collins, M. S. and Pappagianis, D., 1973, Effects oflysozyme and chitinase on the spherules
of Coccidioides immitis and Histoplasma capsulatum, Contrib. Microbiol.Immunol. 3:106-125.
35. Frey, C. L. and Drutz, 0. j., 1986, Influence of fungal surface components on the interaction
of Coccidioides immitis with polymorphonudear neutrophils, J Infect. Dis. 153:933-943.
36. Baker, O. and Braude, A. 1., 1956, A study of stimuli leading to the production of spherules
in coccidioidomycosis,] Lab. Clin. Med. 47:169-181.
37. Galgiani,J. N., Hayden, R, and Payne, C. M., 1982, Leukocyte effects on the dimorphism of
Coccidioides immitis,] Infeel. Dis. 146:56-63.
38. Beaman, L., Pappagianis, D., and Benjamini, E., 1977, Significance ofT cells in resistance to
experimental murine coccidioidomycosis, Infeel. Immun. 17:580-585.
39. Hoeprich, P. D. and Merry, j. M., 1988, Effect of recombinant human interleukin-2 in
experimental murine coccidioidomycosis, Diagn. Microbiol. Infeel. Dis. 9:115-118.
40. Kirkland, T. N. and Fierer, j., 1983, Inbred mouse strains differ in resistance to lethai
Coccidioides immitis infection, Inftel. Immun. 40:912-916.
41. Kirkland, T. N. and Fierer, j., 1985, Genetic control of resistance to Coccidioides immitis: A
single gene that is expressed in spleen cells determines resistance,j. Immunol. 135:548-552.
42. Cox, R. A., KennelI, W, Boncyk, L., and Murphy, j. W, 1988, Induction and expression of
cell-mediated immune responses in inbred mice infected with Coccidioides immitis, Infeel.
Immun. 56:13-17.
43. Ampel, N. M. and Galgiani,j. N., 1991, Interaction ofhuman peripheral blood mononuclear
cells with Coccidioides immitis arthroconidia, Cell Immunol. 133:253-262.
44. Beaman, L., Be~amini, E., and Pappagianis, D., 1981, Role oflymphocytes in macrophage-
induced killing of Coccidioides immitis in vitro, Infeel. Immun. 34:347-353.
45. Beaman, L. and Holmberg, C. A., 1980, In vitro response of alveolar macrophages to
infection with Coccidioides immitis, Inftel. Immun. 28:594-600.
46. Beaman, L., Benjamini, E., and Pappagianis, D., 1983, Activation of macrophages by
lymphokines: Enhancement of phagosome-lysosome fusion and killing of Coccidioides im-
mitis, Infect. Immun. 39:1201-1207.
47. Beaman, L., 1987, Fungicidal activation of murine macrophages by recombinant gamma
interferon, Inftct. Immun. 55:2951-2955.
48. Clemons, K.Y., Leathers, C. R., and Lee, K. W, 1985, Systemic Coccidioides immitis infection
in nude and beige mice, Infect. Immun. 47:814-821.
49. Deresinski, S. C., Stevens, D. A., Applegate, R. j., Levine, H. B., and Stevens, D. A., 1977,
Cellular immunity to Coccidioides immitis: In vitro lymphocyte response to spherules, arthro-
spores, and endospores, Cell Immunol. 32:110-119.
50. Petkus, A. F. and Baum, L. L., 1987, Natural killer cell inhibition of young spherules and
endospores of Coccidioides immitis, J Immunol. 139:3107-3111.
51. Deresinski, S. C., Levine, H. B., and Stevens, D.A., 1978, Coccidioides immitis endospores:
Phagocytosis by human cells, Mycopatlwlogia 3:179-181.
52. Beaman, L., 1991, Effects of recombinant gamma interferon and tumor necrosis factor on in
vitro interactions of human mononuclear phagocytes with Coccidioides immitis, Infeet. Immun.
59:4427 -4429.
53. Ampel, N. M., Bejarano, G. C., and Galgiani, j. N., 1992, Killing of Coecidioides immitis by
human peripheral blood mononuc\ear cells, Infoct. Immun. 60:4200-4204.
54. Smith, C. E., Whiting, E. G., Baker, E. E., Rosenberger, H. G., Beard, R R., and Saito, M. T.,
1948, The use of coccidioidin, Am. Rev. Tuberc. 57:330-360.
55. Cox, R. A. and Vivas,j. R., Spectrum of in vivo and in vitro cell-mediated immune responses
in coccidioidomycosis, Cell Immunol. 31:130-141.
56. Cox, R. A., Vivas, J. R., Gross, A., Lecara, G., and Brummer, E., 1976, In vivo and in vitro cell-
mediated immune responses in coccidioidomycosis. I. Immunologic responses of persons
with primary, asymptomatic infection, Am. Rev. Resp. Dis. 114:937-942.
57. Cox, R A. and KennelI, W, 1988, Suppression of T-lymphocyte response by Coeeidioides
immitis antigen, Infeel. Immun. 56:1424-1429.
58. Cox, R. A. and Pope, R. M., 1987, Serum-mediated suppression oflymphocyte transforma-
tion responses in coccidioidomycosis, Infeel. Immun. 55:1058-1062.
59. Johnson, S. M. and Pappagianis, D., 1992, The coccidioidal complement fixation and
immunodiffusion-complement fixation antigen is a chitinase, Infect. Immun. 60:2588-
2592.
60. Smith, C. E., Saito, M. T., and Simmons, S. A., 1956, Pattern of 39,500 serologie tests in
coccidioidomycosis, J Am. Med. Assoe. 160:546-552.
61. Scott, P., 1993, IL-12: Initiation cytokine for cell-mediated immunity, Science 260:
496-497.
15
Pulmonary Cryptococcosis
Pathophysiological and Clinical
Characteristics
MIRIAM L. CAMERON and JOHN R. PERFECT
1. INTRODUCfION
Cryptococcus neoformans is an encapsulated yeast that has become a significant

human pathogen in our enlarging population of immunocompromised hosts.
The yeast was originally isolated by Sanfelice in 1894 from peachjuice,l but its
role as a human pathogen was described soon after in 1894-1895 (Buschke and
Busse) from a bone lesion. 2,3 C. neoformans mayaiso be a pulmonary pathogen in
other mammals and even the bottle-nosed dolphin is not safe from a pulmonary
infection. 4 Since its discovery, C. neoformans has been a consistent cause of invasive
mycoses, but with the advent of iatrogenic immunosuppression (glucocorticoids,
cancer chemotherapeutic agents, anti-trans plant rejection therapeutics),5,6 and
with the appearance of the acquired immunodeficiency syndrome (AIDS) during
the past 10 years,7-1O this particular fungal infection has mushroomed into a
common clinical problem.
C. neoformans is ubiquitous in nature worldwide ll- 15 hence, compared to
certain endemie mycoses, such as Histoplasma, Blastomyces, and Coccidioides,
epidemiologie and travel histories are less relevant to diagnosis, although expo-
sure to bird (pigeon) excreta, which carries a high burden of organisms, has been
implicated in some patients such as pigeon breeders.I6-18 In addition, pulmonary
MIRIAM L. CAMERON and lOHN R. PERFECT • Division of Infectious Diseases, Duke

University Medical Center, Durham, North Carolina 22710.
249
250 MIRIAM L. CAMERON and lOHN R. PERFECT
cryptococcosis has been described in a bagpipe player with yeast contamination

of his instrument. 19 Experimentally, pigeon excreta and the surrounding dust
have been aerosolized and found to contain yeast forms of C. neoformans in small
enough partide size of 1-5.5 11M to allow deposition in alveoli. 2o- 22 Various soil
sites in nature have been found to contain the yeasts. 23 .24 Recently, Ellis et al. have
found that Cryptococcus neoformans var gatti strains (serotype B/C) have been found
associated with Eucalyptus camaldulensis trees (Red River gum) in Australia. 25 .26
These investigators have even identified structures that may be basidiospores in
the flower of these trees and postulate that the shedding of this fungus may be
interrelated to the flowering mechanism ofthe tree. Although the more common
C. neoformans var neoformans (serotype AID) yeasts have been isolated from soil
specimens and animals, and may even asymptomatically colonize the human
respiratory tree,27.28 it is speculated that the natural reservoir for this variety of
yeasts might reside in some type of flowering grasses. The yeasts could then be
picked up by animals or birds feeding on these plants and spread into the
environment through their feces or guano. The human host generally acquires C.
neoformans infection via inhalation of aerosolized yeasts from these excreta or the
surrounding soiI.21.22.29 This inhalation leads typically to an inapparent pulmon-
ary infection in the immune competent hosts. On the other hand, this infection
can lead to widespread dissemination, particularly into the subarachnoid space,
even in apparently normal hosts. Another route of infection indudes a few well-
documented case reports of accidental percutaneous inoculation.3°.31 In most
cases, however, skin lesions containing C. neoformans represent dissemination
to the skin from a pulmonary infection, or an established and active subarachnoid
infection. As the usual site of entry for C. neoformans, the lungs play a major role
in the pathogenesis and treatment outcome of C. neoformans infection. This
chapter focuses specifically on the various aspects related to pathogenesis and
treatment of cryptococcosis at this site.
2.0RGANISM
C. neoformans is a saprobic fungus found in nature in an asexual, haploid
yeast phase generally considered to be the infectious form. A sexual stage, Filo-
basidiella neoformans, indudes alpha- and a mating-type yeasts which, under
certain environmental conditions and physical proximity, can mate, produce
damp connections and true hyphae, and then form basidiospores on the ends of
these hyphae. 32 Once detached, basidiospores will morphologically change to
sm all yeasts within hours, while growing on routine culture media. This life cyde
has been reproduced in vitro but has not been convincingly found in vivo,
although recent work with the eucalyptus tree encouraged researchers that the
perfect state may have been seen in nature. 26
The value of this life cyde and the ability to undergo meiosis has two
potentially important aspects. First, this pathogenic yeast can be studied both on a
molecular and genetic basis. The use of meiosis is an extremely valuable genetic
tool,33 and potentially makes it a beuer fungal pathogen model to study molecu-
PULMONARY CRYPTOCOCCOSIS 251
lar mechanisms than the common diploid, asexual pathogen, Candida albicans.
Second, the basidiospores produced by these sexual structures appear ideal as
aerosol infectious agents. 34 The basidiospore is a I-211m ovoid shaped spore that
could easily be inhaled as an infectious agent. Serious challenges to the basidio-
spore as the infectious agent are (I) more than 90% of the C. neoformans isolates
from elinical pulmonary infections are mating-type alpha,35 and (2) these basidio-
spores have not been readily found in nature. If the basidiospore was an infectious
agent by simple Mendelian characteristics, the number of mating-type alpha and
mating-type a infections should be approximately 50% each. It is possible,
however, that the mating-type alpha locus contains some virulence factors that
allow it to express infection better than the mating-type a locus, and there is some
recent genetic support that mating-type alpha locus and/or genes in elose prox-
imity may be associated with virulence,36 On the other hand, even in nature, the
vast majority of isolates are mating-type alpha. 35 Therefore, at this time most
investigators believe that the infectious form of C. neoformans is a poorly encapsu-
lated yeast measuring 1-5 11m in diameter, whieh is inhaled and deposited in the
alveolus.
The yeast body itself generally measures 4-6 11m in diameter in vitro, but in
host tissue and in vitro cultures under certain conditions, the yeast will form a
polysaccharide capsule that measures from 1-30 11m in width. There is strain
variation in size of the capsule and although its presence is a major virulence
factor, the direct measurement of capsule size is not. 37 For example, acapsular or
hypocapsular mutants have been isolated from patients with human immuno-
deficiency virus-I (HIV-I) infection du ring cryptococcosis. 38,39 These findings
suggest that during a severe, persistent, immune depressive state, capsule regula-
tion may be less important than certain host factors. However, the polysaccharide
capsule remains an extremely important virulence factor, is a taxonomic feature,
and also aids in the diagnosis of infection by its detection in various body fluids.
The biochemistry of the glucuronoxylomannan polysaccharide capsule structure
has been weIl studied and its variable characteristics used for serotyping. 4O-43
There are two varieties of the yeast, C. neoformans var. neoformans (serotype
A or D) and C. neoformans var. gattii (serotype B or C). These two varieties are
distinct by both DNA homology studies 44 and biochemieal assays and karyotyp-
ing. 45 C. neoformans var. neoformans is the most common elinical variety worldwide,
and is recovered often in both immunocompromised and immunocompetent
hosts. The vast majority of AIDS patients (> 95%) are infected with this vari-
ety.46-48 On the other band, C. neoformans var. gattii is more geographically bound
and is found in infections from Australia, Southern California, Southeast Asia,
and Central Aftica. 49 ,50 C. neoformans var gattii infection may occur years after
living or traveling in an endemie area. 5l ,52 It appears to cause disease in immuno-
competent hosts over immune deficient ones,5l,53 and may have a higher propen-
sity for invading the brain parenchyma rather than the subarachnoid space and
may be more difficult to treat,53 but comparative studies are needed to ensure the
validity of these pathogenie concepts. Interestingly, even in areas of high AIDS
prevalence, such as Southern California, serotypes A and D are almost exelusively
252 MIRIAM L. CAMERON and JOHN R. PERFECf
isolated from patients even though Band C serotypes exist in the· environ-
ment. 47 .48 The differences and similarities in pathobiology and molecular under-
standing of these two varieties of cryptococcus remain interesting investigational
subjects.
Studies have shown that virulence factors can be identified that allow various
strains of C. neoformans to vary in their ability to cause infection. Some factors have
been identified and others remain to be discovered. The polysaccharide capsule is
known to inhibit phagocytosis of the yeast,54 and generally its presence is needed
to cause disease. Even this primary virulence factor cannot, however, explain the
reports that nonpathogenic, acapsular strains have apparently been isolated from
immunocompetent hosts and immunocompromised hosts, induding AIDS pa-
tients with cryptococcosis. 38 .55-58 In addition, there are species of cryptococci
other than C. neoformans with capsules that are not pathogenic. Therefore, other
factors are likely essential to produce disease through a dynamic relationship with
host responses. Several yeast enzymes induding proteases59 and 3,4-dihydroxy-
phenylalanine-phenoloxidase60.61 have been suggested as virulence factors. The
importance of the phenoloxidase system for virulence has been the best stud-
ied. 60 .61 Strains producing increased amounts of phenoloxidase activity are more
infectious to mice than those producing lesser amounts; mutants that lack
phenoloxidase activity are avirulent.
The genetics of this organism have been successfully used to study certain
characteristics, induding gene analysis and linkages with the phenoloxidase
phenotype and capsular genes, respectively.33.60.61 With present molecular biolog-
ical approaches and a proven genetic system, the potential exists for determining
specific antifungal targets for directed chemotherapies. A basic molecular foun-
dation for C. neoformans is being developed. Ribosomal DNA genes have been
cloned62 and restriction fragment polymorphisms between different species in
their rDNA have been used to determine the relationship between various
cryptococcal species. 63 Molecular epidemiological studies can be performed
using either mitochondrial DNA polymorphism64 or differences in strain karyo-
types using pulsed-field electrophoresis. 45 Two transformation schemes have now
been designed to allow transfer of DNA into C. neoformans65 •66 and a variety of
genes have been doned and sequenced.65.67.68 Therefore, this organism has the
potential to be the prototype fungus to study for the molecular determinants of
the host-fungus interactions.
C. neoformans is easy to isolate in culture on routine bacteriologic and
mycologic media, usually within a week of inoculation of body fluids (blood,
cerebrospinal fluid, urine, etc.) or tissue onto the media. The yeast grows at
37°C in air. The phenoloxidase enzyme that converts phenolic compounds to
melanin can be used to help identify the organism. On media containing 3,4-
dihydroxycinnamic acid (caffeic acid plates, bird-seed agar), the phenoloxidase
oxidizes the O-diphenol to produce dark or black colonies.69 This is particularly
useful in mixed flora cultures such as sputum or environmental sampies. It is also
helpful to have the laboratory save sputum culture plates for several extra days in
high-risk patients to allow time for identification of the C. neoformans isolates,
whieh may be missed if plates are routinely discarded after 48 hr. Identification is
also aided by India ink examination showing typical encapsulated budding yeast
from direct specimens, and by a simple measuring of urease production by the
isolated yeasts.
Although the focus of this chapter is Cryptococcus neoformans, other crypto-
coccal species have been known to produce pulmonary infections. Both Cryptococ-
cus albidus70 and Cryptococcus laurenti71 ,72 have been reported to cause invasive
infection in the lung. These cases are rare reports and therefore, further consid-
eration of these cryptococcal species are not made in this discussion.
3. HOST DEFENSES
The pathological description of human pulmonary cryptococcosis has been

weIl described in the textbook by Littman and Zimmerman. 29 Gross examination
of lung can vary from a diffuse pneumonia to a smaIllocalized nodule only found
at autopsy. There is generally an inverse relationship between numbers of host
cells and yeast seen within tissue. This is also typical of other areas of the body
such as the subarachnoid space. The smaller granulomas of the lung are seen in
the periphery or subpleurallocation with an abundant and typical granuloma-
tous reaction. Larger lesions generally do not form a fibrous encapsulation like a
tuberculoma and there is little tissue necrosis or microabscess formation as seen
with the endemie mycoses. Exceptions will occur, however, and fibrocaseous
lesions have been observed. The organisms can be seen with periodie acid-schiff
(PAS), methenamine silver, alcian blue, or mucicarmine stains, and polysac-
charide material may be identified scattered throughout the cytoplasm of macro-
phages with or without ingested yeasts. 29,73 Acapsular strains of C. neoformans may
be visualized using a Fontana-Masson stain.56 In an individual with a severelY
depressed immune system, mieroscopie examination of the lung tissue may reveal
clusters of yeasts, packing the alveolar spaces with only a few host cells pres-
ent. 74 ,75
It has previously been suggested that a poorly encapsulated yeast form
rather than a basidiospore is the first infectious form the pulmonary immune
system must attack. Although there are no studies in humans, in mice, intranasal
colonization can occur with C. neoformans weeks or months prior to the develop-
ment of infection of the lung. 76 The bronchoalveolar fluid would be the next
site of in vivo exposure for these yeasts; in our experience with bronchoalveolar
lavage fluids from rabbits, this fluid does not normally contain potent anticrypto-
coccal activity (unpublished data). Therefore, except for the physical removal of
yeast particles by the ciliated epithelial cells in the airway and thus limitation of
particle or propagule size to below 10 ,...m, the first line of defense against the
inhaled yeast is the alveolar macrophages. There have been aseries of studies on
the interaction of C. neoformans and alveolar macrophages. Rat alveolar macro-
phages have been shown to ingest and kill C. neoformans in the absence of serum. 77
This is likely an important feature of these cells, whieh are exposed to limited
254 MIRIAM L. CAMERON and JOHN R. PERFECT
complement and immunoglobulins in the alveolar fluid. However, C. neoformans

has developed important survival adaptations as it reaches the lung tissue. It has
been shown that capsule size of C. neoformans is regulated by ambient pC02
concentrations. 78 At environmental CO 2 concentrations, most C. neoformans pos-
sess small capsules, however, as the pC02 tension increases to concentrations
found in lung tissue, the capsule size is dramatically increased. Since the capsule
is a protective armor against phagocytosis and intracellular killing by host cells,
this regulation of capsular synthesis appears to be a survival mechanism for the
yeast. On the other hand, recent studies suggest that murine alveolar macro-
phages may be more efficient at killing the more heavily encapsulated yeasts,
compared to the hypocapsular strains via nonoxidative mechanisms. 79 This
appears to be a paradoxical situation compared to other areas of the body where
the capsule is protective. This selective killing by the normal alveolar macro-
phages may result in elimination of the majority of the infectious inoculum,
particularly the large extracellular yeasts, but a few of the poorly encapsulated
yeasts within the inoculum may be easily ingested by the phagocytes, and not
killed. These yeasts would then be part of a pulmonary lymph node complex and
remain dormant in a so-called persistent state until reactivation and dissemina-
tion of infection occur when the protective immune system is abrogated. It is
likely from both pathological and clinical descriptions that C. neoformans, like H.
capsulatum and M. tuberculosis, forms lymph node complexes where organisms
remain dormant without producing immediate disease until reactivation oc-
curs. 80 ,8l
The specific macrophage-Cryptococcus neoformans interaction has been weH
studied in vitro. It is likely that there is both an extracellular fungistasis mecha-
nism, which requires both macrophage mobility and direct contact with the yeast,
and a second intracellular killing mechanism,s2 Tissue macrophages and mono-
cytes from rodents generally require astate of activation by the addition of
biological modifiers such as INF-')' and endotoxin for fungistasis and fungicidal
activities. 83 ,84 These biological activators have also been shown to improve the
anticryptococcal activity of rat alveolar macrophages,77 but these results appear
species specific. In rabbits and humans, resident alveolar macrophages do not
require activation for their anticryptococcal activity85,86 (Lee-See, personal com-
munication). This may partially explain the relative resistance to cryptococcal
infections in these two species where the resident alveolar cells appear primed to
inhibit growth of C. neoformans. Rabbit alveolar macrophages exhibit exquisite
fungistasis against C. neoformans in vitro. Macrophage mobility stimulated by a
serum component appears to be required for fungistasis to occur (Lee-See,
personal communication). Human alveolar macrophages mediate fungistasis in
vitro independent of endotoxin and INF-')',s5,86 In fact, the combination of
endotoxin and INF-')' actually inhibits human alveolar macrophage-mediated
fungistasis, though endotoxin or INF-')' alone have no effect on human alveolar
macrophage-mediated fungistasis. 86 Yet, in a similar in vitro system, INF-')' ap-
pears to have a dual effect on human alveolar macrophage-mediated fungistasis,
by decreasing the rate of phagocytosis, while inducing cytolytic activity against
intracellular yeasts. 87 Overall, these two effects had no impact on human alveolar
macrophage-mediated fungistasis as measured in vitro. The mechanism of these
effects is unknown, though INF-'Y may inhibit complement receptor 1 (CRl) yet
appears to have no effect on complement receptors 3 (CR3) or 4.8 7 C. neoformans is
thought to bind to mononuclear phagocytes via CR3. 88 The polysaccharide
capsule of the yeasts stimulates the alternate complement pathway, leading to
iC3b deposition on the yeast surface and then to subsequent binding and inges-
tion of the yeast via iC3b-CR3 interactions.89,90 Hence, the mechanism by which
endotoxin plus INF-'Y inhibit human alveolar macrophage-mediated fungistasis
remains unknown; the specific pathways for fungistasis and fungicidal activity of
alveolar macrophages also remain unclear. It is likely that there is more than one
pathway. The recently described L-arginine-dependent nitrogen oxidation sys-
tem required for murine macrophage-mediated fungistasis 84 has not been found
in human alveolar macrophages.8 6 In fact, human alveolar macrophages, peri-
toneal macrophages, and blood monocytes operate their fungistasis via a mecha-
nism independent of L-arginine nitrogen oxidation. 86 Further studies in this
area of human alveolar macrophage biology with cell components such as
defensins 91 are needed, as this cell has demonstrated potent intracellular fungi-
stasis and probably some intracellular killing as measured in vitro. 86
Some in vitro studies have been done examining the effect of HIV infection
on human mononuclear phagocyte (alveolar macrophage, peritoneal macro-
phage, blood monocyte) mediated anticryptococcal activity. In vitro inoculation of
human mononuclear phagocytes with monocytotropic HIV-I strain BaL dirn in-
ishes the anticryptococcal activity of peritoneal macrophages and monocytes. 92 ,93
In contrast, alveolar macrophages retain anticryptococcal activity, and appear less
sensitive to HIV infection. 94 Mononuclear phagocytes may be a reservoir of HIV
in vivo,95-99 hence, alteration of mononuclear phagocyte function including
anticryptococcal activity conceivably could occur in vivo. Because alveolar macro-
phages inoculated in vitro with HIV retain normal anticryptococcal activity as
measured in vitro, the question of the roles of other immune effector cells in
mediating acquired immunity for antifungal function in the lung is again raised.
Two possibilities exist. First, other cell types, including natural killer (NK) cells,lOo
cytotoxic T cells, CD4lymphocytes,IOI and neutrophils lO2 may be required in vivo.
Clearly, CD4lymphocytes are affected by in vivo HIV infection. 103 Other immune
effector cells such as cytotoxic T cells can also be affected by HIV infection. 104,105
Second, mononuclear phagocytes from different tissue sites may have different
susceptibilities to HIV infection.I°6 In vitro HIV inoculation of monocytes and
peritoneal macrophages has shown a difference in susceptibility in viral expres-
sion.I°6 Thus, it is possible that alveolar macrophages in patients seropositive for
HIV may function normally against cryptococci, but the presence of dysfunc-
tional CD4 lymphocytes could lead to cryptococci escape from the lungs into the
blood and central nervous system where blood monocytes, other forms of tissue
macrophages, or other types of inflammatory cells are inadequate in controlling
cryptococcal infection. Supporting data for this theory is that patients without
AIDS who have cryptococcal meningitis but minimal to no inflammatory re-
sponse as measured by cerebrospinal fluid studies have an extremely poor prog-

nosis because of cryptococcal disease,l°7,108
Despite the consistent anticryptococcal activity measured in vitro, studies
show that alveolar macrophages do not provide efficient killing of C. neoformans
and that other lung celIs are also essential for eradication or suppression of
infection in the lung. It is also important to note that in the host there is adynamie
turnover of cells with freedom of mobility, and interaction with other cell types,
including cytotoxic T ceIls, NK ceIls, and polymorphonuclear cells in the inflam-
matory milieu, which cannot be duplicated by in vitro studies. The quantitative
inflammatory response is difficult to reproduce in vitro, but may be an essential
component of the successful immune response in the lung. For example, the
initial natural immunity to a large inoculum of yeasts in the lung may even
require polymorphonuclear ceUs in the response. Although not primarily consid-
ered a major host ceU in this infection, the neutrophil has been shown to have
potent anticryptococcal activity in vitro by both oxidative and nonoxidative mech-
anisms. 91 ,102 In C5 deficient mice with increased susceptibility to C. neoformans,109
the inability to make the powerful chemoattractant C5a for polymorphonuclear
ceHs and macrophages may illustrate that a large quantity of these ceUs may be an
important part of the successful immune reaction. Murine NK ceUs have been
shown to possess anticryptococcal activitylOO but when relative importance for the
immune response within the lung was examined, there was litde effect of murine
NK ceUs on the pulmonary infection. 110 Human NK ceUs have been shown to
inhibit C. neoformans growth in vitro in the presence of specific antieapsular
antibody but have minimal inhibitory effect in the absence of antibody.lll Local
antibody production has not been detected in pulmonary cryptococcosis, hence
the role of NK ceUs in pulmonary infection is probably minimal. It has also been
shown that CD8 (cytotoxic T ceHs) are important in the clearance of C. neoformans
from murine lungs. 1l2,l13 CD8 ceHs may be crucial in recruiting effector ceUs
or may help in lysing unactivated macrophages to release yeasts so they can be
killed by activated ceUs.
Clinieal and pathological studies suggest that the majority of pulmonary
infections are subclinieal and that most disseminated disease occurs from reacti-
vated infections. The biology of the yeast that remains in this persistent state
within the lung remains a fascinating and poorly understood phenomenon. On
the other hand, we have begun to identify important immune ceUs, in our
acquired immunity, whieh apparently keep these yeasts in their dormant or
persistent state, and thus, likely localized within the lung. During the HIV
epidemie, it has been observed that as the CD4 count of the host drops during
infection below 200 ceUs/I.l.I, the risk of a disseminated cryptococcal infection rises
significantly.103 Alveolar macrophages from HIV-seropositive patents exhibit
enhanced accessory ceH function, which may lead to depletion of CD4 cells from
the lung,l14 These data suggest that the CD4 ceU is a crucial ceU in the mainte-
nance mechanism(s) whieh limits C. neoformans infection to the lung. Well-
controlled animal studies have supported this concept. The CD4 subset of lym-
phocytes was depleted from mice by treating them with a monoclonal antibody to
these cells. After intratracheal inoculation of C. neoformans, dissemination from

the lung to other organs occurred earlier and the burden of organisms in extra-
pulmonary tissue was greater in the CD4 depleted mice compared to controls,
despite similar levels of organisms in the lung tissue in both groups.l0l,1l2,1l3 Both
the animal and human findings make it clear that the CD4 cell is an essential host
cell in preventing C. neoformans from escaping to extrapulmonary sites.
Although humoral immunity, including both certain complement compo-
nents ll5 and antibodies, 116,l17 can influence phagocytosis of C. neoformans and may
be helpful in eradicating infection, there remains controversial proof of their
worth in the inflammatory reaction to the primary or reactivation pulmonary C.
neoformans infection. Murine macrophages require a macromolecular component
in serum in addition to L-arginine to mediate fungistasis.l 18 Human mono-
nuclear phagocytes also require serum to mediate fungistasis in vitro.8 5,86,1l9 Yet,
human serum depleted of complement by heating to 56°C for 30 min is still able
to support human alveolar mediated fungistasis,86 hence the role of complement
is unclear. Nevertheless, the polysaccharide capsule does activate the alternate
complement pathway leading to iC3b deposition on the yeast surface. 90 This
allows binding of the yeasts to the CR3 receptor on mononuclear phagocytes. The
role of antibodies in the inflammatory response to C. neoformans is even more
murky. We have shown no improvement in human macrophage mediated fungi-
stasis,85 yet other groups have shown that anticryptococcal capsule antibody may
enhance fungistasis or killing of certain strains of C. neoformans .78,82,120 Clinically,
the role of complement components and antibodies is also unclear. Patients with
known complement deficiencies have not been reported to develop cryptococcal
infection, although C5-deficient mice are more susceptible.l°9 Patients withJob's
syndrome, hyperimmunoglobulin E, may have an increased incidence of cryp-
tococcal infection,121,122 and hypogammaglobulinemia has been seen in a patient
with cryptococcosis.1 23
The animal models in cryptococcosis have been helpful in defining the
importance of certain host factors. There are several murine genetic immune
defect models, .including C5 deficient, nude, and beige mice,109,124 which are
more susceptible to overwhelming infection with C. neoformans. These results
suggest that certain complement components, NK cells and an intact T-cell
immune system are important to successful immunity from this infection, both
at the pulmonary level and in other body sites. An excellent animal model of
pulmonary cryptococcosis in rats and guinea pigs examined the effect of steroids
on this infection.l25 These experiments showed that sublethai doses of C. neofor-
mans could be harbored asymptomatically in animals infected in the pulmonary
tree; these latently infected animals are administered cortisone. Cortisone-treated
animals developed extensive extrapulmonary cryptococcosis and died. Clinical
experience in humans suggests that pharmacologic doses of corticosteroids can
produce the same effects and these agents have become a second major factor in
the presentation of cryptococcosis along with HIV infection. 27 ,126-131 In addition,
patients with endogenous glucocorticoid excess, Cushing's syndrome, also have a
predilection for pulmonary cryptococcosis, with or without dissemination. 132
There are several other important concepts regarding the yeast when consid-
ering host immune defenses. First, it should be emphasized that in most of the
immunological studies on the pulmonary defenses, there is generally a difference
between various strains in their ability to be killed by host cells. The phenotypes of
these strains is not explained by capsule size, and further molecular and bio-
chemical work with the yeast is needed to determine characteristics that may
make some C. neoformans strains intrinsically more invasive into lung tissue.
Second, it is now also possible to determine if a patient can be coinfected with two
different strains at the same time but at different locations, and whether reinfee-
tion of the lung with a second strain can occur. Molecular karyotyping or various
methods of DNA restriction fragment polymorphisms of clinical isolates generally
gives unique strain differences, and can potentially be used to answer these
epidemiological questions.
4. CLINICAL PRESENTATIONS
C. neoformans infection in the normal immunocompetent host tends to be 10-

calized to the lung,133 but in the immunocompromised host dissemination to
other sites frequently occurs without clinical evidence of the initial pulmonary
site, for instance, meningitis, cryptococcemia, and multiorgan system involvement,
including the eyes, heart, liver, spleen, bone, urinary tract and skin.l 6,29,51,133-144
Cryptococcal disease occurs more frequently in Caucasian adult males than in
other population groups.133 This is consistent with animal studies where males are
more predisposed to infection, but there have been no known hormonal receptors
on this yeast found to explain the predilection. Occupational and recreational
exposures have been postulated as the reason for this epidemiology, but no
definitive proof exists. With the present HIV epidemie in Africa and the Amer-
icas, the increased numbers ofinfection in blacks has risen dramatically. Unques-
tionably, HIV infection has had the greatest impact on clinical epidemiology of
cryptococcosis since its discovery a century ago. Another interesting and unex-
plained observation is the infrequent reporting of cryptococcosis in children. 145 A
recent report, however, of pulmonary cryptococcal nodules in children with
sarcomas suggests that certain underlying diseases in this age group may predis-
pose to this infection. 146
Pulmonary cryptococcosis has different clinical presentations, treatments,
and outcomes depending on whether the host is immunocompetent or immuno-
compromised and therefore each group is discussed separately.
4.1. Immunocompetent Host

Hundreds of cases of pulmonary cryptococcosis have been described in the
medicalliterature,12,27,126,145,147-161 but Campbell's comprehensive review of the
English literature in 1965 remains the most complete description of primary
pulmonary cryptococcosis.l49 In this review, only 11 ofthe 101 patients described
had underlying diseases such as malignancy, diabetes, tuberculosis, alveolar pro-

teinosis, and pemphigus vulgaris. Campbell's review would suggest that primary
pulmonary cryptococcosis in the immunocompetent host may present asympto-
matically in 32%. For example, it may be discovered through an abnormal finding
on a ehest radiograph taken for other reasons. The majority of patients described,
however, do present with symptoms including cough 54%, ehest pain 46%, spu-
tum production 32%, weight loss and fever 26% each, and hemoptysis 18%.
Other presentations including dyspnea, night sweats, and superior vena cava
obstruction are rarer. 51 ,162 Several unusual presentations have been reported. A
case report of "allergie" cryptococcal pneumonia characterized by urticaria,
hypotension, and dyspnea has also been described. 163 Cryptococcal pneumonia
has presented with similar findings to a Pancoast's tumor l64 and even develop-
ment ofbronchiolitis obliterans-organizing pneumonia has been implicated with
a cryptococcal infection.I65 Pulmonary cryptococcosis in the immunocompetent
host has also been found in conjunction with Mycobacterium tuberculosisl 66-168 and
pulmonary echinococcosis.169 In one case, pulmonary tuberculosis occurred
years after lung resection for cryptococcosis.1 68 Cavitary pulmonary cryptococ-
cosis has also been complicated by aspergilloma. 150 Of course, the nature of these
reported cases suggest a bias toward clinical manifestations. It remains likely,
however, that the vast majority ofhuman pulmonary infections are asymptomatic
or do not reach the clinician's attention. These silent clinical presentations are
similar to cases of asymptomatic tuberculosis or histoplasmosis, but in many
respects less observable. C. neoformans of the lung simply doesn't produce as much
scar tissue or encapsulation as tuberculosis or add as much calcium to infected
necrotic foci as histoplasmosis, and therefore it is radiographically silent. Also, the
skin test has not been used as extensively as the PPD or histoplasma skin test to
demonstrate the numbers of infected individuals.170-174
Diagnosis of infection in immunocompetent hosts has been made ante
mortem by lung biopsy, either excisional or incisional, for histopathology amll
or culture, cytopathology, sputum culture, antigen testing, and by ehest X
ray.73,74,170,175-185 Several series of cryptococcal pneumonia have shown asymp-
tomatic colonization of the respiratory tree by C. neoformans.I48,158,160,186 This
generally occurs in patients with an underlying lung disease such as chronic
obstructive pulmonary disease and therefore a sputum culture positive for C.
neoformans with a lack of clinical symptoms or consistent ehest radiograph find-
ings needs to be interpreted carefully and may not always require therapy. In the
normal host, C. neoformans generally does not disseminate, therefore blood
cultures, urine cultures, cerebrospinal fluid (CSF) cultures, and serum or CSF
cryptococcal polysaccharide antigens are usually negative. Nevertheless, it is our
opinion that all patients with cryptococcal pneumonia and/or pulmonary coloni-
zation should be evaluated for cryptococcal meningitis with a lumbar puncture
because of the disease's propensity to invade the central nervous system, even in
the absence of neurologie signs or symptoms. 51 ,133 Other sites of dissemination
found in patients with what was originally classified as pulmonary cryptococcosis
include blood, bone, skin, and eyes. 5 ,6,16,187,188
Radiographieally, C. neoformans pneumonia in the normal host may present

with well-defined, nonealcified single or multiple lung nodules, indistinet to mass-
like infiltrates, hilar and mediastinal lymphadenopathy, oeeasionally pleural
effusions, and more rarely, eavitation.I47 ,151,155,17B,lBl,lB9-191 Single or multiple
peripheral nodules are the most eommon radiographie findings and many of
these infeetions are diseovered when the nodules are aspirated or removed to rule
out malignaney.I B9 Aseries of ehest X rays are displayed in Figs. 1-3 to demon-
strate the range of radiologie manifestations of pulmonary eryptocoeeosis.
Other diagnostie tests are rarely useful in primary pulmonary eryptoeoecosis
unless there is evidenee of subsequent dissemination. Oeeasionally, the serum
eryptoeoeeal antigen is positive and it should always be assessed.I 54,170,175,lB4,192
There remains the importanee of an elevated serum eryptoeoeeal antigen in
patients with apparent pulmonary disease alone. Aserum eryptocoeeal antigen
of 1:8 or more with doeumented pulmonary involvement suggests to us a high
burden of organisms in the lung and/or disseminated infeetion and would make
us eonsider speeifie antifungal treatment.
Whether treatment is indieated, and the most appropriate therapy to be used,
are still subjeets of controversy for pulmonary eryptoeoecosis in the immunoeom-
petent individual. Normal immunoeompetent hosts with single asymptomatie
pulmonary nodules do not neeessarily need treatment. If the organism is viable in
the lesion (i.e., eultures are positive), however, there is a eogent argument for
relatively nontoxie oral therapy to potentially eliminate the yeast from the host as
there is no assuranee that the patient will not become immunosuppressed in the
future and thus be at risk for reaetivation and dissemination of infeetion. If the
organism is seen on histopathology but eultures are negative, then it may be
reasonable to withhold treatment beeause there is no proof that the organism is
growing sufficiently to respond to antifungal ehemotherapy. Those patients with
symptoms, multiple nodules, mass- or pseudotumor-like lesions, pleural effu-
sions, or more extensive interstitial lung disease, probably deserve antifungal
therapy. There have been multiple eases of pleural effusions eaused by C.
neoformans.I91 ,193,194 Most ean be treated with antifungal agents without surgieal
drainage, out there is always the eomplex ease where tube drainage has been
eonsidered beneficiaI.l95 Prior to the development of amphotericin B, surgieal
reseetion of pulmonary lesions was a main therapeutie modality.I 96 It is rarely
needed at the present, unless to aid in diagnosis by excisional biopsy, or in eases of
large pseudotumor-like masses unresponsive to ehemotherapy;15B,lBB,197 however,
sinee the pathophysiology suggests that a hilar lymph node eomplex exists,BO,Bl it
is also unlikely that surgery alone on a mass or nodule ean completely eliminate
the organism from the lung. We also remain uneonvineed that surgery, by itself,
on the lung of a patient with pulmonary eryptoeoeeosis, predisposes to dissemi-
nated disease.I 9B
Most studies of antieryptoeoeeal therapy have foeused on treatment of
eryptoeoeeal meningitis, therefore, litde or no data is available coneerning the best
treatment for primary pulmonary eryptoeoeeosis without evidenee of dissemina-
tion. It is reasonable, however, to use the same drugs and doses as used for
~
r
~~
Cj
~
~
Cj
oCj
Cj
~
Vl
-
FIGURE 1. ehest roentgenogram demonstrating a right lower lobe nodule in anormal host.
~
:::
~
O'l
~
~
;;;
:;
~
t'
n
>-
~
i:"1
B
z
tlJ
8-
'g
z
r=
;:g
FIGURE 2. ehest roentgenogram showing left upper lobe and right lower lobe mass-like infiltrates in anormal ::0
"I'j
hast. i:"1
q
Azole therapy may also be appropriate for cryptococcal pneumonia, based on

therapy results for meningitis. Ketoconazole has been reported to successfully treat
cryptococcal pneumonia in a few patients205 but is not successful in meningitis. 206
If ketoconazole is used, a reasonable dose is 400 mg orally per day for 3-6 months.
Itraconazole is not available in all countries, but has been shown to have efficacy in
cryptococcal meningitis and is likely to be successful in pneumonia.207-209 Flucon-
azole has been used with increasing success to treat cryptococcal meningitis in
immunocompromised patients, including those with AIDS.210-212 Although no
substantial reports have been made concerning its use in the normal immuno-
competent host with C. neoformans infection, it is reasonable to suggest efficacy.
Recent data show ßuconazole is as efficacious as amphotericin B in the treatment
of cryptococcal meningitis in patients with AIDS when used at a dose of at least
400 mg orally per day.2I1 As with amphotericin B, the exact dose and duration of
therapy are unknown, but it would be reasonable to use 200-400 mg ßuconazole
orally per day for 3-6 months.210.213 Toxicity associated with ßuconazole is
minimal and the value of oral therapy in patients with minimal symptoms cannot
be underestimated.
The prognosis for cryptococcal pneumonia in the normal host is excellent
with or without therapy. Morbidity has occurred predominantly in patients with
extensive disease, and after dissemination in patients with a relapse. 29.196 Thus,
some patients with primary pulmonary cryptococcosis can be observed with the
expectation of a good outcome with or without therapy. The more aggressive
clinician, however, will justify treating most patients with viable C. neoformans in
pulmonary tissue. 214
4.2. Immunocompromised Host without HIV Infection

Immunocompromised patients with cryptococcal pneumonia can have a
completely different and more rapid clinical course than that seen in immuno-
competent hosts. Though C. neoformans typically enters through the lungs, in the
immunocompromised host it tends to disseminate to the central nervous system
(CNS), hence patients normally present with a meningitis syndrome rather than a
pulmonary syndrome. However, an overwhelming pneumonia with adult respira-
tory distress syndrome126.159.215 can occur on initial presentation. High-risk im-
munocompromised hosts for disseminated cryptococcosis include patients with
human immunodeficiency virus (HIV) infection, cirrhosis, diabetes, Cushing's
syndrome, sarcoidosis, leukemia, lymphoma, sickle cell disease, and treatment
with glucocorticoids, as weIl as patients after organ transplantation.132-134.216-220
Kerkering et al., in a classic article in 1981 133 described pulmonary cryptococcosis
in 41 patients, 34 of whom had an underlying immunocompromising condition
other than HIV. Twenty-nine of these patients developed disseminated disease,
but only one of the patients with disseminated disease was identified as immuno-
competent. Unlike immunocompetent hosts who generally have an inapparent
pneumonitis, 83% of the patients in Kerkering's series had constitutional symp-
toms. The most common presenting symptoms were fever in 63%, malaise 61%,
HGURE 3. Chest roentgenogram exhibiting interstitial infiltrates in a patient with AIDS and
cryptococcal pneumonia.
cryptococcal meningitis. Amphotericin B 0.3-0.5 mg/kg intravenously per day

has been a mainstay of therapy for cryptococcosis.108.199 The addition of flu-
cytosine at 75-100 mg/kg orally per day in four divided dos es has allowed for use
of a reduced amphotericin B dosage and shorter duration of therapy.108.200 If
amphotericin B is used alone, probably 6-8 weeks of therapy are indicated. When
flucytosine is used in conjunction with amphotericin B, 4-6 weeks of therapy
should be adequate. 108 Therapy also needs to be tailored to clinical response and
drug toxicity, which includes renal insufficiency, potassium and magnesium
wasting from amphotericin B, and bone marrow suppression from flucyto-
sine.201.202 Drug levels of ftucytosine should be maintained at 100 mg/dl or less,
checked 2 hr after dosing. 202 Flucytosine can also he used alone in cryptococcal
pneumonia in the nonimmunocompromised host with a reasonable success
rate,196.203.204 hut there remains concern ahout development of drug resistance
if the hurden of organisms is high.
chest pain 44%, weight loss 37%, dyspnea 27%, night sweats 24%, cough 17%,
hemoptysis and headache 7% each. The three patients in this series with head-
ache at presentation and one patient without headache had concurrent crypto-
coccal meningitis. 133
A definitive diagnosis was made in Kerkering's series by sputum culture,
bronchoscopy, thoracentesis, open lung biopsy, and needle aspiration. Subsequent
to diagnosis of pulmonary cryptococcosis, dissemination to the meninges oc-
curred in 25 patients within 2-20 weeks after presentation. All of the patients
had abnormal chest roentgenograms. The most common finding, in order of
frequency, was alveolar or interstitial infiltrates followed by single- or multiple-
coin lesions, masses, cavitary lesions and pleural effusions. In this artide it
became dear that immunodeficient hosts require antifungal therapy.l33 Confu-
sion may arise about who is induded as immunodeficient, but in general, it is
probably safer to err on the side of overtreatment.
The more controversial issue in this area is the best regimen for treatment of
pneumonia in the immunocompromised host. First, acheck must be made for
extrapulmonary sites of infection: CSF, blood, and urine cultures after prostatic
massage, and a good skin exam. Although formal studies have never been
performed for pulmonary infection in the immunocompromised patient, crypto-
coccal meningitis has been extensively studied and retrospective reports suggest
certain concepts. First, flucytosine should not be used alone because of the risk for
development of drug resistance and thus failure or relapse. 133 Amphotericin B
remains the best characterized treatment and the addition of flucytosine shortens
the course of treatment. It seems reasonable to use a 6-week course with doses
previously mentioned in the section on the immunocompetent host. Judgment on
longer primary courses will need to be made on an individual basis. Primary
therapy with azole compounds in this group remains less characterized, and
because of the azoles' fungistatic characteristics, it may require longer courses and
theoretically could have higher relapse rates, although this possibility remains
unproven. If azoles, ketoconazole or fluconazole, are used, a reasonable regimen
should be 400 mg orally per day for at least 3-6 months.
4.3. Immunocompromised Host as a Result of HIV Infection

Cryptococcosis is an important opportunistic infection in patients with
AIDS, occurring in 8_10%.7.8.221 Pulmonary cryptococcosis is less common than
meningitis but weIl described. Pulmonary cryptococcosis in patients with HIV
infection has slighter different manifestations than in other types of immuno-
compromised hosts. We and others have described small series of pulmonary
cryptococcosis in patients with AIDS.219.222-225 In one series of pulmonary
manifestations of AIDS, C. neoformans was the implicated pathogen in approx-
imately 10% ofthe cases. 226 Combining three similar small series of cryptococcal
pneumonia in patients with AIDS leads to some observations.219.222.223 Almost
all of the patients presented with symptoms induding fever 81%, cough 63%,
dyspnea 50%, weight loss 47%, and headache 41%. Pleuritic chest pain was
mentioned in one series in three out of five patients. 222 Hemoptysis occurred in
another series in two out of 12 patients. 219 Dissemination of cryptococci, partic-
ularly to the meninges or blood, occurred in 94% based on the results from two of
the series.219.222 The third series 223 also showed evidence of dissemination though
the exact level was not stated. Physical exam findings were not stated in most
reports, but may include lymphadenopathy, rales, tachypnea, and splenomegaly.219
Concurrent oral candidiasis was found frequently in one series, and this finding
should alert the clinician to an underlying HIV infection if this information is not
already known, and again emphasizes the importance of a CD4 count under 200
celllf,d to clinical disease. Concomitant second infections with other opportunistic
pathogens including Pneumocystis carinii, Mycobacterium avium, eytomegalovirus,
and Histoplasma capsulatum, have oecurred in conjunetion with pulmonary erypto-
coecosis.219.223.224 In addition, C. neoformans pneumonia may occur as a conse-
quenee of steroid therapy for Pneumocystis carinii.I 27
Diagnostic studies should include arterial P02' ehest radiography, cultures,
and cryptococcal antigen detection. Mild to moderate hypoxemia can be found
though both normal P02 and profound hypoxemia have occurred with crypto-
coccal pneumonia.219.223 Adult respiratory distress syndrome has also oceurred in
this population.215.227.228
Chest radiographs most often reveal interstitial infiltrates (Fig. 3), either focal
or diffuse, and lymphadenopathy, unlike immunoeompetent and other types of
immunoeompromised hosts, nodular and alveolar infiltrates are quite rare.219.223.229
Large masses and pleural effusions are also unusual. 219 Beeause the most com-
mon radiographie picture is interstitial infiltrates, C. neoformans pneumonia can
easily be confused with Pneumocystis carinii pneumonia, the most eommon eause
of interstitial infiltrates in patients with AIDS.230
Cryptoeoceal antigen detection using latex agglutination ean be extremely
helpful while awaiting eulture results. Sinee patients frequently have dis sem i-
nated eryptoeoeeosis, serum, CSF, and urine cryptococcal antigens may be
positive.219.223 It is important that a prozone effeet be ruled out, sinee serum
antigen titers greater than 1:1 million can oecur,231 and that in addition, the
antigen kit used contains pronase to remove inhibitors in the serum. It is
imperative to assess evidenee of dissemination, particularly to the meninges, for
patients with AIDS. Therapy for pulmonary cryptococcosis should, however,
probably be the same as for meningitis. Pulmonary and extrapulmonary cultures
are pivotal in making the diagnosis. Regular sputum, bronehoalveolar lavage,
transbronchiallung biopsy or needle aspiration, and pleural fluid cultures are all
suitable for making the diagnosis.219.224.232.233 Blood and CSF are high-yield
cultures in disseminated disease in this patient population. Occasionally, urine
culture, preferably after prostatic massage, bone marrow eultures or eultures of
skin lesions may lead to the diagnosis.l36.231.234
Treatment of pulmonary eryptococeosis in AIDS patients is similar to that
used for cryptococeal pneumonia or meningitis in other populations of immuno-
eompromised hosts, with the eaveat that relapse and dissemination occur fre-
quently in this group, particularly if not given maintenance therapy.7-9.235 Recent
data have shown that ftuconazole and amphotericin Bare comparable in the
treatment of cryptococcal meningitis in patients with AIDS.210-212 Fluconazole
may sterilize the CSF at a slower rate than amphotericin B, hence if it is used in a
patient initially with only primary pulmonary cryptococcosis, it is possible that
there would be a higher risk for dissemination while the patient is on induction
therapy. Nevertheless, it is a reasonable first choice drug in a patient without
factors that could possibly predict poor outcome, including hypoxemia, ARDS,
severe meningitis with increased intracranial pressure, or obtundation. After
acute therapy for cryptococcal pneumonia with or without dissemination, with
either ftuconazole or amphotericin B with or without ftucytosine maintenance
therapy with ftuconazole should be continued indefinitely, since relapse is com-
mon in this patient group.236.237 The prostate and the basal eisterns may serve as
reservoirs of C. neoformans, despite apparently adequate therapy to sterilize
the CSF or pulmonary speeimens.238.239 We use a maintenance dose of 200-400
mg/day of ftuconazole, based on whether or not the patient has had a relapse
previously, and based on renal function. This concept of maintenance therapy
may be extended to other severely and persistently immunocompromised hosts
although supportive data are unfortunately not available. Acute mortality from
disseminated cryptococcosis has been quite high in certain risk groups with AIDS
though the prognosis of primary pulmonary cryptococcosis in this patient popu-
lation is unknown, but is likely to be beUer.
5. CONCLUSION
Pulmonary cryptococcosis may occur in a number of presentations, based on

the immune competency of the host. We antieipate that an increasing number of
cases will be seen in the future because of an increase in the number of solid
organ and bone marrow trans plant reeipients, more aggressive use of chemo-
therapy agents in cancer patients, espeeially those with hematologic malignan-
eies, and the increasing ineidence of HIV infection. Ongoing trials in patients
with HIV infection using ftuconazole or clotrimazole versus no antifungal agent
may allow future recommendations to he made for prophylaxis in these popula-
tions at risk for profound immunodefieiency against cryptococcosis. For the
present, a clinical suspieion of cryptococcal pneumonia, rapid diagnosis, and
appropriate therapy are key to a successful outcome of this increasing clinical
infection. In the future, discovery of new fungieidal agents and hetter under-
standing and manipulation of host immunity for treatments will be welcomed
additions to our armentarium in the fight against this encapsulated yeast.
REFERENCES
1. Sanfeliee, E, 1894, Contributo alla morfologia e biologia dei blastomiceti ehe si sviluppano
nei sueehi di alcuni frutti, Ann. igiene 4:463.
268 MIRIAM L. CAMERON and lOHN R. PERFECT
2. Buschke, A., 1895, Ueber, eine durch coccidien hemorgerufene krankheit des menschen,
Dtsch. Med. Wochenschr. 21:14.
3. Busse, 0., 1894, Ueber parasitare zelleinschulusse and ihre zuchtung, Centralbl f Bakt
16:175.
4. Migaki, G., Gunnels, R. D., and Casey, H. W, 1978, Pulmonary cryptococcosis in an
Atlantic bottlenosed dolphin (TURSIOPS TRUNCATUS), Lab. Anim. Sei. 28:603-606.
5. Perfeet, ]. R., Durack, D. T., and Gallis, H. A., 1983, Cryptococcemia, Medicine 62:
98-109.
6. Perfeet, ]. R., 1989, Cryptococcosis, Infect. Dis. Clin. North Am. 3:77-102.
7. Zuger, A., Louie, E., Holzman, R. S., Simberkoff, M. S., and Rahal,].]., 1986, Cryptococcal
disease in patients with acquired immunodeficiency syndrome: Diagnostic features and
outcome of treatment, Ann. Intern. Med. 104:234-240.
8. Kovacs,]. A., Kovacs, A. A., Polis, M., Wright, W C., Gill, v.]., Tuazon, C. 0., Gelmann, E. P.,
Lane, H. C., Longfield, R., Overturf, G., Macher, A. M., Fauci, A. S., Parrillo,]. E., Bennett,
]. E., and Masur, H., 1985, Cryptococcosis in the Acquired Immunodeficiency Syndrome,
9. Chuck, S. L. and Sande, M. A., 1989, Infections with Cryptococcus neoformans in the acquired
immunodeficiency syndrome, N. Engi.] Med. 321:794-799.
10. Clark, R. A., Greer, D., Atkinson, W, Valainis, G. T., and Hyslop, N., 1990, Spectrum of
Cryptococcus neoformans infection in 68 patients infected with acquired immunodeficiency
virus, Rev. Infect Dis. 12:768-777.
11. Edwards, V. E., Sutherland,]. M., and Tyrer,]. H., 1970, Cryptococcosis of the central
nervous system,] Neural. Neurosurg. Psychiatry 33:415-425.
12. Newberry, W M.,lr., Walter,]. E., Chandler,]. W,lr., and Tosh, F. E., 1967, Epidemiologie
study of Cryptococcus neoformans, Ann. Intern. Med. 67:724-732.
13. Falconer, H., Terry, S. 1., and Spencer, H., 1980, Cryptococcosis in the West Indies, West
Indian Med.] 29:142.
14. Slobodniuk, R. and Naraqi, S., 1980, Cryptococcal meningitis in the central province of
Papua New Guinea, New Guinea Med.] 23:111.
15. Misha, S. K., Staib, F., Folkens, v., and Fromtling, R. A. 1981, Serotypes of Cryptococcus
neoformans strains isolated in Germany,] Clin. Microbiol. 14:106.
16. Bisseru, B., Baj'Ü, A., Carruthers, R. H., and Chhabra, H. N., 1983, Pulmonary and
bilateral retinochoroidal cryptococcosis, Br.] Ophthalmol. 67:157-161.
17. Randhawa, H. S. and Paliwal, D. K., 1979, Survey of Cryptococcus neoformans in the
respiratory tract of patients with bronchopulmonary disorders and in the air, Sabouraudia
17:399-404.
18. Fink,]. N., Barboriak,]. ]., and Kaufman, L., 1968, Cryptococcal antibodies in pigeon
Breeder's disease,] Allergy Clin. Immunol. 41:297-301.
19. Cobcroft, R., Kronenberg, H., and Wilkinson, 1., 1978, Cryptococcus in bagpipes, Lancet
1:1368-1369.
20. Powell, K. E., Dahl, B. A., Weeks, R. ]., and Tosh, F. E., 1972, Airborne Cryptococcus
neoformans: Particles from pigeon excreta compatible with alveolar deposition,] Infect. Dis.
126:412-415.
21. Smith, C. D., Ritter, R., Larsh, H. W, and Furculow, M. L., 1964, Infections of white mice
with airborne Cryptococcus neoformans, ] Bacteriol. 87:1363-1368.
22. Neilson,]. B., Fromtling, R. A., and Bulmer, G. S., 1977, Cryptococcus neoformans: Size range
of infectious particles from aerosolized soil, Infect. Immun. 17:634-638.
23. Emmons, C. W, 1951, Isolation of Cryptococcus neoformans from soil,] Bacteriol. 62:
685-690.
24: Ruiz, A., Fromtling, R. A., and Bulmur, G. S., 1981, Distribution of Cryptococcus neoformans
in a natural site, Infect. Immun. 31:560-563.
25. Ellis, D. H. and Pfeiffer, T.]., 1990, Ecology, life cycle, and infectious propagule of Cryptoeoeeus
neoformans, Laneet 336:923-925.
26. Ellis, D. H. and Pfeiffer, T.]., 1990, Natural habitat of Cryptoeoeeus neoformans var gattii,
] Clin. Mierobiol. 28:1642-1644.
27. Duperval, R., Hermans, P. E., Brewer, N. S., and Roberts, G. D., 1977, Cryptococcosis, with
emphasis on the significance of isolation of Cryptoeoeeus neoformans from the respiratory
tract, Chest 72:13-19.
28. Levitz, S. M., 1991, The etiology of Cryptoeoeeus neoformans and the epidemiology of
cryptococcosis, Rev. Infect. Dis. 13:1163-1169.
29. Littman, M. L. and Zimmerman, L. E., 1956, Cryptoeoeeosis, torulosis or European blasto-
myeosis, Grune and Stratton, New York.
30. Glaser,]. B. and Garden, A., 1985, Inoculation of cryptococcosis without transmission
of the acquired immunodeficiency syndrome, N. Engi.] Med. 313:264.
31. Shuttleworth, D., Philpot, C. M., and Knight, A. G., 1989, Cutaneous cryptococcosis
treatment with oral fluconazole, Br.] Dermatol. 120:683-687.
32. Kwon-Chung, K. ]., 1975, A new genus, Filobasidiella, the perfect state of Cryptococcus
neoformans, Mycologia 67:1197.
33. Still, C. N. and jacobson, E. S., 1983, Recombinational mapping of capsule mutations in
Cryptococeus neoformans, ] Bacteriol. 156:460-462.
34. Cohen,]., Perfect,]. R., and Durack, D. T., 1982, Cryptococcosis and basidiospores, Laneet
1:1301.
35. Kwon-Chung, K.]. and Bennett,]. E., 1978, Distribution of"a" and "alpha" mating types of
Cryptoeoeeus neoformans among natural and clinical isolates, Am.] Epidemiol. 108:337.
36. Kwon-Chung, K.]., Edman,]. C., and Wickes, B. L., 1992, Genetic association of mating
types and virulence in Cryptococcus neoformans, Infeet. Immun. 60:602-605.
37. Dykstra, M. A., Friedman, L., and Murphy, ]. W, 1977, Capsule size of Cryptocoeeus
neoformans: Control and relationship to virulence, Infeet. Immun. 16:129.
38. Gal, A. A., Evans, S., and Meyer, P. R., 1987, The clinical laboratory evaluation of
cryptococcal infections in the Acquired Immunodeficiency Syndrome, Diagn. Microbiol.
Infect. Dis. 7:249-254.
39. Bottone, E. ]. and Wormser, G. P., 1985, Capsule-deficient cryptococci in AIDS, Laneet
2:553.
40. Kozel, T. R., 1989, Antigenic structure of Cryptoeoeeus neoformans capsular polysaccharides,
in: Immunology of Fungal Diseases, (E. Kurstak, ed.), Marcel Dekker, Inc., New York, pp.
63-86.
41. Cherniak, R., Reiss, E., Slodki, M. E., Plattner, R. D., and Blumer, S.O., 1980, Structure and
antigenie activity of the capsular polysaccharide of Cryptoeoccus neoformans serotype A, Mol.
Immunol. 17:1025-1032.
42. Cherniak, R., Reiss, E., and Turner, S. H., 1982, A galactoxylomannan antigen of Crypto-
eoeeus neoformans serotype A, Carbohydr. Res. 103:239-250.
43. Bhattacharjee, A. K., Bennett, ]. E., and Glaudemans, C. P. ]., 1984, Capsular polysac-
charides of Cryptoeoeeus neoformans, Rev. Infect. Dis. 6:619-624.
44. Erke, K. H. and Schneidau,]. D., 1973, Relationship of some Cryptoeoeeus neoformans hypha-
forming strains to standard strains and to other species of yeasts as determined by
deoxyribonucleic acid base ratios and homologies, Infect. Immun. 7:941-948.
45. Perfect, ]. R., Magee, B. B., and Magee, P. T. 1989, Separation of chromosomes of
Cryptoeoeeus neoformans by pulsed fieid gei electrophoresis, Infect. Immun. 57:2624-2627.
46. Bottone, E.]., Salkin, 1. E, Hurd, N.]., and Wormser, G. P., 1987, Serogroup distribution of
Cryptoeoceus neoformans in patients with AIDS,] Infect. Dis. 156:242.
47. Shimizu, R. Y., Howard, D. H., and Claney, M. N., 1986, The variety of Cryptocoeeus
neoformans in patients with AIDS,] Infect. Dis. 159:1042.
48. Rinaldi, M. G., Drutz, D.]., Howell, A., Sande, M. A., Wofsky, C.B., and Hadley, W K., 1986,
Serotypes of Cryptococcus neoformans in patients with AIDS,] Infect. Dis. 153:642.
49. Bennett, ]. E., Kwon-Chung, K.]., and Howard, D. H., 1977, Epidemiology differences
among serotypes of Cryptococcus neoformans, Am. ] Epidemiol. 105:582.
50. Kwon-Chung, K.]., and Bennett, ]. E., 1984, Epidemiologie differences between the two
varieties of Cryptococcus neoformans, Am.] Epidemiol. 120:123-140.
51. Lehmann, P. F., Morgan, R.]., and Freimer, E. H., 1984, Infection with Cryptococcus neo-
formans var. gattii leading to a pulmonary cryptococcoma and meningitis,] Infect. 9:301-306.
52. Bottone, E. ]., Kirschner, P. A., and Salkin, I. F., 1986, Isolation of highly encapsulated
Cryptococcus neoformans serotype B from a patient in New York City,] Clin. Microbiol.
23:186-188.
53. Henderson, D. K., Edwards,]. E., Dismukes, W E., and Bennett, ]. E., 1981, Meningitis
produced by different serotypes of Cryptococcus neoformans, in: Proceedings of the 81st
annual meeting of the American Society for Microbiology, American Society of Microbiology
Proceedings (Abstract).
54. Bulmer, G. S. and Tacker,]. R., 1975, Phagocytosis of Cryptococcus neoformans by alveolar
macrophages, Infect. Immun. 11:73-79.
55. Bottone, E.]., Johnasson, B. E., Toma, M., and Wormser, G. P., 1986, Poorly encapsulated
Cryptococcus neoformans from patients with AIDS. I: preliminary observations, AIDS Res.
Hum. Retroviruses 2:221.
56. Ro,]. Y., Lee, S. S., and Ayala, A. G., 1987, Advantage ofFontana-Masson stain in capsule-
deficient cryptococcal infection, Arch. Pathol. Lab. Med. 111:53-57.
57. Moser, S. A., Friedman, L., and Varraux, A. R., 1978, Atypical isolate of Cryptococcus
neoformans cultured from sputum of a patient with pulmonary cancer and blastomycosis,]
Clin. Microbiol. 7:316-318.
58. Harding, S. A., Scheld, W M., Feldman, P. S., and Sande, M. A., 1979, Pulmonary infection
with capsule-deficient Cryptococcus neoformans, Virchows Arch {A]182:113-118.
59. Brueske, C. H., 1986, Proteolytic activity of a clinical isolate of Cryptococcus neo[ormans,]
Clin. Microbiol. 23:631.
60. Rhodes, ]. C., Polacheck, 1., and Kwon-Chung, K. ]., 1982, Phenoloxidase activity and
virulence in isogenie strains of Cryptococcus neoformans, Infect. Imm'un. 36:1175-1184.
61. Kwon-Chung, K. ]. and Rhodes, ]. C., 1986, Encapsulation and melanin formation as
indicators of virulence in Cryptococcus neoformans, Infect. Immun. 51:218-223.
62. Restrepo, B. I. and Barbour, A. G., 1989, Cloning of the 18s and 25s ribosomal DNA from
the pathogenic fungus, Cryptococcus neoformans,] Bacteriol. 171:5596-5600.
63. Vilgalys, R. and Hester, M., 1990, Rapid genetic identification and mapping of en-
zymatically amplified ribosom al DNA from several Cryptococcus species, ] Bacteriol.
172:4238-4246.
64. Varma, A. and Kwon-Chung, K.]., 1989, Restriction fragment polymorphism in mito-
chondrial DNA of Cryptococcus neoformans,] Gen. Microb. 135:3353-3362.
65. Edman,]. C. and Kwon-Chung, K.]., 1990, Isolation of the URA5 gene from Cryptococcus
neoformans var. neoformans and its use as a selective marker for transformation, Mol. Cell.
Biol. 10:4538-4544.
66. Toffaletti, D. L., Rude, T. H., Johnston, S. A., and Durack, D. T., 1993, Gene transfer in
Cryptococcus neoformans using biolistic delivery of DNA.] Bacteriol. 175:1405-1411.
67. Perfect,]. R., Rude, T. H., Penning, L. M., and Johnston, S. A., 1992, Cloning of the
Cryptococcus neoformans TRP 1 gene by complementation in Saccharomyces cerevisiae, Gene
122:213-217.
68. Perfect, ]. R., Rude, T., Penning, L., and Wong, B., 1991, Cloning of a Cryptococcus
neoformans mannitol dehydrogenase gene, Internat. Soc. Human Animal Mycol. PS1.45.
(Abstract).
69. Chaskes. S. and TyndalI. R. L., 1975. Pigment production by Cryptococcus neoformans from
para- and ortho-diphenols: Effect of the nitrogen source,) Clin. Microbiol. 1:509-514.
70. Krumholz, R. A., 1972, Pulmonary cryptococcosis: A case due to Cryptococcus albidus,
71. Lynch, j. P., III, Schaberg, 0. R., Kissner, D. G., and Kauffman, C. A., 1981, Cryptococcus
laurmtii lung abscess, Am. Rev. Respir. Dis. 123:135-138.
72. Krajden, S., Summerbell, R.C., Kane,j., Salkin, I. F., Kemna, M. E., Rinaldi, M. G., Fuksa,
M., Spratt, E., Rodriques, C., and Choe, j., 1991, Normally saprobie cryptococci isolated
from Cryptococcus neoformans infections,) Clin. Microbiol. 29:1883-1887.
73. Gupta, R. K., 1985, Diagnosis of unsuspected pulmonary cryptococcosis with sputum
cytology, Acta Cytol. 29:154-156.
74. Gal, A. A., Koss, M. N., Hawkins,j., Evans, S., and Einstein, H., 1986, The pathology of
pulmonary cryptococcal infections in the acquired immunodeficiency syndrome, Arch.
Pathol. Lab. Med. 110:502-507.
75. Gal, A. A., Kos, M. N., and Hawkins, j., 1986, The pathology of pulmonary cryptococcal
infections in the acquired immunodeficiency syndrome, Arch. Pathol. Lab. Med. 110:502.
76. Anderson, D. A. and Sagha, H. M., 1988, Persistence of infection in miee inoculated
intranasally with Cryptococcus neoformans, Mycopathologia 104:161-169.
77. Mody, C. H., Tyler, C. L., Sitrin, R. G., Jackson, C., and Toews, G. B., 1991, Interferon-
gamma activates rat alveolar macrophages for anticryptococcal activity, Am.) Resp. Cello
Mol. Biol. 5:19-26.
78. Granger, D. L., Perfeet, j. R., and Durack, 0. T., 1985, Virulence of Cryptococcus
neoformans: Regulation of capsule synthesis by carbon dioxide,) Clin.lnvest. 76:508-516.
79. Levitz, S. M. and DiBenedetto, 0. j., 1989, Paradoxieal role of capsule in murine broncho-
alveolar macrophage-mediated kiIIing of Cryptococcus neoformans,) Immunol. 142:659-665.
80. Salyer, W. R., Salyer, D. C., and Baker, R. D., 1974, Primary complex of Cryptococcus and
pulmonary Iymph nodes,) Inftct. Dis. 130:74-77.
81. Baker, R. 0., 1976, The primary pulmonary Iymph node complex of cryptococcosis, Am.)
Clin. Pathol. 65:83-92.
82. Granger, 0. L., Perfect,j. R., and Durack, D. T., 1986, Macrophage-mediated fungistasis in
vitro: Requirements for intracellular and extracellular cytotoxicity,) Immunol. 136:672-
680.
83. FIeseh, I. E. A., Amberger, G., and Kaufmann, S. H. E., 1989, Fungicidal activity of IFN-
gamma activated macrophages. Extracellular kiIIing of Cryptococcus neoformans,) Immunol.
142:3219-3224.
84. Granger, D. L., Hibbs, j. B., Perfect, j. R., and Durack, D. T., 1988, Specific amino acid
(L-Arginine) requirement for the mierobiostatic activity of murine macrophages,) Clin.
Invest. 81:1129-1136.
85. Weinberg, P. B., Becker, S., Granger, D. L., and Koren, H. S., 1987, Growth inhibition of
Cryptococcus neoformans by human alveolar macrophages, Am. Rev. Respir. Dis. 136:1242-
1247.
86. Cameron, M. L., Granger, D. L., Weinberg,j. B., Kozumbo, w.j., and Koren, H. S., 1990,
Human alveolar and peritoneal macrophages mediate fungistasis independently of L-ar-
ginine oxidation to nitrite or nitrate, Am. Rev. Respir. Dis. 142:1313-1319.
87. Becker, S. and Koren, H. S., 1989, Dual effects of interferon-gamma on the interaction
between human alveolar macrophages and Cryptococcus neoformans, Chest 95:S172-S174.
88. Newman, S. L. and Mikus, L. K., 1985, Deposition of C3b and iC3b onto particulate
activators of the human complement system. Quantitation with monoclonal antibodies to
human C3,) Exp. Med. 161:1414-1431.
89. Kozel, T. R., Wilson, M. A. Pfrommer, G. S. T., and Schlageter, A. M., 1989, Activation and
binding of opsonie fragments of C3 on encapsulated Cryptococcus neoformans by using an
alternative complement pathway reconstituted from six isolated proteins, Infect. Immun.
57:1922-1927.
90. Kozel, T. R. and Pfrommer, G. S. T., 1986, Activation of the complement system by
Cryptococcus neoformans leads to binding of iC3b to the yeast, Infect. Immun. 52:1-5.
91. Lehrer, R. I. and Ladra, K. M., 1972, Fungicidal components of mammalian granulocytes
active against Cryptocoeeus neoformans, ] Infect. Dis. 136:96.
92. Cameron, M. L., Granger, D. L., Matthews, T. J., Haney, A. F., and Weinberg, J. B., 1991,
Peritoneal macrophages infected in vitro with HIV-l have decreased anticryptococcal
activity, ICAAC (Abstract).
93. Cameron, M. L., Granger, D. L., Weinberg, J. B., and Matthews, T. J., 1990, Human
monocytes infected with HIV-l have reduced anti-cryptococcal activity (abstract 362A),
Clin. Res. 38:(Abstract).
94. Cameron, M. L., Granger, D. L., Matthews, T. J., and Weinberg, J. B., 1991, Alveolar
macrophages infected in vitro with HIV-l retain normal cryptococcal activity, VII Interna-
tional AIDS Conferenee, Florenee, Italy (Abstract).
95. Meltzer, M. S., Skillman, D. R., Gomatos, P. J., Kalter, D. C., and Gendelman, H. E., 1990,
Role of mononuclear phagocytes in the pathogenesis of Human Immunodeficiency Virus
infection, Annu. Rev. Immunol. 8:169-194.
96. Mann, D. L., Gartner, S., Le Sane, F., Buchow, H., and Popovic, M., 1990, HIV-l transmis-
sion and function of virus-infected monocytes/macrophages,] Immunol. 144:2152-2158.
97. Pauza, C. D., 1988, HIV persistence in monocytes leads to pathogenesis and AIDS, Cello
Immunol. 112:414-424.
98. Landay, A., Kessler, H. A., Benson, C. A., Pottage,J. C.,Jr., Murphy, R., Urbanski, P., Kucik,
S., and Phair, J., 1990, Isolation of HIV-l from monocytes of individuals negative by
conventional culture,] Infect. Dis. 161:706-710.
99. Gartner, S., Markovits, P., Markovitz, D. M., Kaplan, M. H., Gallo, R. C., and Popovic, M.,
1986, The role of mononuclear phagocytes in HTLV-IIIILAV infection, Science 233:
215-219.
100. Murphy, J. W. and McDaniel, D. D., 1982, In vitro reactivity of natural killer (NK) cells
against Cryptoeoeeus neoformans, ] Immunol. 128:1577.
101. Mody, C. H., Lipscomb, M. F., Street, N. E., and Toews, G. B., 1990, Depletion of CD4+
(L3T4 +) Iymphocytes in vivo im pairs murine host defenses to Cryptoeoeeus neoformans, ]
Immunol. 144:1472-1477.
102. Diamond, R. D., Root, R. K., and Bennett, J. E., 1972, Factors inßuencing killing of
Cryptoeoeeus neoformans by human leukocytes in vitro,] Infeet. Dis. 125:267-276.
103. Masur, H., Ognibene, F. P., Yarchoan, R., Shelhamer, J. H., Baird, B. F., Travis, w.,
Suffredini, A. F., Deyton, L., Kovacs, J. A., Falloon, J., Davey, R., Polis, M., Metcalf, J.,
Baseler, M., Wesley, R., Gill, V. J., Fauci, A. S. and Lane, H. C., 1989, CD4 counts as
predictors of opportunistic pneumonias in human immunodeficiency virus (HIV) infec-
tion, Ann. Intern. Med. 111:223-231.
104. Plata, F., Autran, B. Martins, L. P., Wain-Hobson, S., Raphael, M., Mayaud, C., Denis, M.,
Guillon, J., and Debre, P., 1987, AIDS virus-specific cytotoxic T Iymphocytes in lung
disorders, Nature 328:348-351.
105. Mackewicz, C. E., Ortega, H. w., and Levy, J. A., 1991, CD8+ cell anti-HIV activity
correlates with the clinical state of the infected individual,] Clin. Invest. 87:1462-1466.
106. Olafsson, K., Smith, M. S., Marshburn, P., Carter, S. G., and Haskill, S., 1991, Variation of
HIV infectibility of macrophages as a function of donor, stage of differentiation, and site of
origin,] Aequir. Immune Defie. Syndr. 4:154-164.
107. Diamond, R. D. and Bennett, J. E. 1974, Prognostic factors in cryptococcal meningitis: A
study of III cases, Ann. Intern. Med. 80:176-18l.
108. Dismukes, W. E., Cloud, G., Gallis, H. A., Kerkering, T. M., Medoff, G., Craven, P. L.,
Kaplowitz, L. G., Fisher, J.F., Gregg, C. R., and Bowles, C. A., 1987, Treatment of
Cryptococcal meningitis with combination amphotericin Band ftucytosine for four as

compared with six weeks, N. EngI.] Med. 317:334-341.
109. Rhodes, j. C., Wicker, L. S., and Urba, W. j., 1980, Genetic control of susceptibility to
Gryptococcus neoformans in mice, Infect. Immun. 29:494.
110. Lipscomb, M. E, Alvarellos, T., Toews, G. B., Tompkins, R., Evans, Z., Koo, G., and Kumar,
v., 1987, Role of natural killer cells in resistance to Gryptococcus neoformans infections
in mice, Am.] Patlwl. 128:354-361.
lU. MilIer, M. E, MitchelI, T. G., Storkus, W. j., and Dawson,]. R., 1990, Human natural killer
. cells do not inhibit growth of Gryptococcus neoformans in the absence of antibody, Infect.
Immun. 58:639-645.
112. Hili,]. 0. and Harmsen, A. G., 1991, Intrapulmonary growth and dissemination of an
avirulent strain of Gryptococcus neoformans in mice depleted of CD4 + or CD8 + T cells,]
Exp. Med. 173:755-758.
U3. Huffnagle, G. B., Yates, ]. L., and Lipscomb, M. F., 1991, Immunity to a pulmonary
Gryptococcus neoformans infection requires both CD4+ and CD8+ T cells, ] Exp. Med.
173:793-800.
114. Twigg, H. L., In, Lipscomb, M. E, Yoffe, B., Barbaro, D. ]., and Weissler, j. C., 1989,
Enhanced accessory cell function by alveolar macrophages from patients infected with the
human immunodeficiency virus: Potential role for depletion of CD4 + cells in the lungs,
Am.] Resp. Gell Mol. Biol. 1:391-400.
115. Calderone, R. A., 1989, The role of complement in host resistance to systemic fungal
infection, in: Immunology of Fungal Diseases, (E. Kurstak, ed.), Marcel Dekker, New York,
pp. 225-242.
116. Waldorf, A. R., 1989, Pulmonary defense mechanisms against opportunistic fungal patho-
gens, in: Immunology of Fungal Diseases, (E. Kurstak, ed.), Marcel Dekker, New York, pp.
243-271.
117. Murphy, ]. w., 1989, Immunoregulation in cryptococcosis, in: Immunology of Fungal
Diseases, (E. Kurstak, ed.), Marcel Dekker, New York, pp. 319-345.
U8. Granger, D. L., Perfect,]. R., and Durack, D. T., 1986, Macrophage-mediated fungistasis:
Requirement for a macromolecular component in serum,] Immunol. 137:693-701.
119. Davies, S. E, Clifford, D. P., Hoidal,]. R., and Repine,]. E., 1982, Opsonic requirements for
the uptake of Gryptococcus neoformans by human polymorphonuclear leukocytes and mono-
cytes,] Infect. Dis. 145:870-874.
120. MilIer, G. P. G. and Kohl, S., 1983, Antibody-dependent leukocyte killing of Gryptococcus
neoformans,] Immunol. 131:1455-1459.
121. Hutto,j. 0., Bryan, C. S., Greene, E L., White, C. j., and Gallin,j. 1., 1988, Cryptococcosis
of the colon resembling Crohn's disease in a patient with hyperimmunoglobulin E-recur-
rent infection (Job's) syndrome, Gastroenterology 94:808-812.
122. Jacobs, D. H., Macher, A. M., Handler, R., Bennett, ]. E., Collern, M. j., and Gallin,]. 1.,
1984, Esophageal cryptococcosis in a patient with the hyperimmunoglobulin E-recurrent
infection (Job's) syndrome, Gastroenterology 87:201-203.
123. Gupton, S., Ellis, M., Cesario, T., Ruhling, M., and Vayuvegula, B., 1987, Disseminated
cryptococcal infection in a patient with hypogamma-globulinemia and normal T cell
functions, Am.] Med. 82:129-131.
124. Marquis, G., Montplaisir, S., and Pelletier, M., 1985, Genetic resistance to murine crypto-
coccosis: The beige mutation (Hediak-Higashi syndrome) in mice, Infect. Immun. 47:
288-293.
125. Gadebusch, H. H. and Gikas, P. w., 1965, The effect of cortisone upon experimental
pulmonary cryptococcosis, Am. Rev. Respir. Dis. 92:64.
126. Kent, T. H. and Layton,]. M., 1962, Massive pulmonary cryptococcosis, Am.] Glin. Patlwl.
38:596-604.
127. Lambertus, M. W, Goetz, M. B., Murthy, A. R., and Mathisen, G. E., 1990, Complications
of corticosteroid therapy in patients with the acquired immunodeficiency syndrome and
Pneumocystis carinii pneumonia, Chest 98:38-43.
128. Wiest, P. M., Flanigan, T., Salata, R. A., Shlaes, D. M., Katzman, M., and Lederman, M. M.,
1989, Serious infectious complications of corticosteroid therapy for COPD, Chest 95:1180-
1183.
129. Goldstein, E. and Rombo, o. N., 1962, Cryptococcal infection following steroid therapy,
Ann. Intern. Med. 56:114.
130. Bennington,J. L., Haber, S. L., and Morgenstern, N. L., 1964, Increased susceptibility to
cryptococcosis following steroid therapy, Dis. Chest 45:262.
131. Zee, C. S., Segall, H. 0., Boswell, W, Ahmadi, J., Nelson, M., and Colletti, P., 1988, MR
imaging of neurocysticercosis,J Comput. Assist. Tomogr. 12:927-934.
132. Kramer, M., Corrado, M. L., Bacci, \1., Carter, A. C., and Landesman, S. H., 1983,
Pulmonary cryptococcosis and Cushing's syndrome, Ann. Intern. Med. 143:2179-2180.
133. Kerkering, T. M., Duma, R. J., and Shadomy, S., 1981, The evolution of pulmonary
cryptococcosis: Clinical implications from a study of 41 patients with and without compro-
mising host factors, Ann. Intern. Med. 94:611-616.
134. Finke, R., Strobel, E. S., Kroepelin, T., Guzman,J., Klein, P.J., Schaefer, H. E., Kappe, R.,
and Muller, J., 1988, Disseminated cryptococcosis in a patient with malignant lymphoma,
Mycoses 31:102-108.
135. Okun, E. and Butler, W T., 1964, Ophthalmologic complications of cryptococcal men-
ingitis, Arch. Ophthalmol. 71:52-57.
136. Borton, L. K. and Wintroub, B. u., 1984, Disseminated cryptococcosis presenting as
herpetiform lesions in a homosexual man with acquired immunodeficiency syndrome,
J Am. Acad. Dermatol. 10:387-390.
137. Mayers,D. L., Martone, WJ., and MandelI, G. L., 1981, Cutaneous cryptococcosis mimick-
ing gram-positive cellulitis in a renal transplant patient, South. Med. J 74:1032.
138. Schupbach, C. W, Wheeler, C. E., and Briggaman, R. A., 1976, Cutaneous manifestations
of disseminated cryptococcosis, Arch. Dermatol. 112:1734.
139. Rico, M. J. and Penneys, N. S., 1985, Cutaneous cryptococcosis resembling molluscum
contagiosum in a patient with AIDS, Arch. Dermatol. 121:901.
140. Gosling, H. R. and Gilmer, WS., 1956, Skeletal cryptococcosis: Reportofa case and review
of the literature, J Bone Joint Surg. 38A:660.
141. Lombardo, T. A., Rabson, A. S., and Dodge, H. T., 1957, Mycotic endocarditis, Am.J Med.
22:664.
142. Colmers, R. A., Irniger, W, and Steinberg, 0. H., 1967, Cryptococcus neoformans endocarditis
cured by amphotericin B, JAMA 199:762.
143. Proknow, J. J., Benfield, J. R., and Rippon, J. W, 1965, Cryptococcal hepatitis presenting
as a surgical emergency, JAMA 191:269.
144. RandalI, R. E., Jr., Stacy, W. K., and Toone, E. C., 1968, Cryptococcal pyelonephritis,
N. Engl. J Med. 279:60.
145. Cohen, 1., 1985, Isolated pulmonary cryptococcosis in a young adolescent, Pediatr. Infect.
Dis. J 4:416-419.
146. Allende, M., Pizzo, P. A., and Walsh, T. J., 1992, Pulmonary cryptococcosis presenting as
metastatic tumor in children with soft tissue sarcomas, 92nd American Society ofMicrobiology,
New Orleans (Abstract).
147. Paillas, J., Sinico, M., Viellefond, A., and Blaye, N., 1982, Un cas de cryptococcose
pulmonaire pseudo-tumorale, Sem. Hop. Paris 58:2402-2404.
148. Warr, w., Bates, J. H., and Stone, A., 1968, The spectrum of pulmonary cryptococcosis,
Ann. Intern. Med. 69:1109-1116.
149. Campbell, G. 0.,1966, Primary pulmonary cryptococcosis, Am. Rev. Respir. Dis. 94:236-243.
150. Rosenheim, S. H. and Schwarz,J. 1975, Cavitary pulmonary cryptococcosis complicated by

aspergilloma, Am. Rev. Respir. Dis. 111:549-553.
151. Zlupko, G. M., Fochler, E j., and Goldschmidt, Z. H., 1980, Pulmonary cryptococcosis
presenting with multiple pulmonary nodules, Ghest 77:575.
152. McDonnell, j. M. and Hutchins, G. M., 1985, Pulmonary cryptococcosis, Rum. Pat/wl.
16:120-128.
153. Middleton, E G. and Prioleau, W H. 1982, Pulmonary cryptococcosis: Review of eleven
cases,J S.G. Med. Assoc. 78:419-423.
154. Town, G. 1., 1985, Pulmonary cryptococcosis: Areport of two cases and review of the
literature, N.Z. Med. J 894-895.
155. Hunt, K. K.,Jr., Enquist, R. W, and Bowen, T. E., 1976, Multiple pulmonary nodules with
central cavitation, Ghest 69:529-530.
156. Diamond, R. D. and Levitz, S. M., 1988, Gryptococcus neojOT1fllJns pneumonia, in: Respiratory
Infections: Diagnosis and Management, 2nd ed., (J. E. Pennington, ed.), Raven Press, New
York, pp. 457-471.
157. McDonnell, j. M. and Hutchins, G. M., 1985, Pulmonary cryptococcosis, Rum. Pat/wl.
16:121-128.
158. Hammerman, K. j., Powell, K. E., Christianson, C. S., Huggin, P. M., Larsh, H. W, Vivas,
j. R., and Tosh, E E., 1973, Pulmonary cryptococcosis: Clinical forms and treatment. A
Center for Disease Control Cooperative Mycoses Study, Am. Rev. Respir. Dis. 108:1116-1123.
159. Henson, D. J. and Hili, A. R., 1984, Cryptococcal pneumonia: A fulminant presentation,
Am. J Med. 228:221.
160. Tynes, B., Mason, K. N., Jennings, A. E., and Bennett, j. E., 1968, Variant forms of
pulmonary cryptococcosis, Ann. Intern. Med. 69:1117-1125.
161. Warr, W, Bates, j. H., and Stove, A., 1968, The spectrum of pulmonary cryptococcosis,
Ann. Intern. Med. 69:1109-1116.,
162. Menon, A., and Rajamani, R., 1976, Giant 'cryptococcoma' of the lung, Br. J Dis. Ghest
70:269-272.
163. Gerstenhaber, B. j., Weiner, B., Morecki, R, Bernstein, R., and Luftschein, S., 1977,
''Allergic'' cryptococcal pneumonia, Lung 154:195-199.
164. MitchelI, D. H. and Sorrell, T. C., 1992, Pancoast's syndrome due to pulmonary infection
with Gryptococcus neoformans variety gatti, Glin. Infect. Dis. 14:1142-1144.
165. Carey, C. E, Mueller, L., Fotopoulos, C. L., and Dall, L., 1991, Bronchitis obliterans-
organizing pneumonia associated with Gryptococcus neofOT1fllJns infection, Rev. Infect. Dis.
13: 1253~ 1254.
166. Kahn, E W, England, D. M., and Jones, j. M., 1985, Solitary pulmonary nodule due to
Gryptococcus neojOT1fllJns and Mycobacterium tuberculosis, Am. J Med. 78:677-681.
167. Corpe, R. E and Parr, L. H., 1953, Pulmonary torulosis complicating pulmonary tuber-
culosis treated by resection,J Thorac. Gardiovasc. Surg. 27:392-398.
168. Riley, E. and Cahan, W G., 1972, Pulmonary cryptococcosis followed by pulmonary
tuberculosis: Acase report, Am. Rev. Respir. Dis. 106:594-599.
169. Dalgleish, A. G., 1981, Concurrent hydatid disease and cryptococcosis in a 16-year-old girl,
Med. JAust. 2:144-145.
170. Davies, S. E and Sarosi, G. A., 1987, Role of serodiagnostic tests and skin tests in the
diagnosis of fungal disease, Gun. Ghest Med. 8:135-146.
171. Bennett, j. E., Hasenclever, H. E, and Baum, G. L., 1965, Evaluation of skin test for
cryptococcosis, Am. Rev. Respir. Dis. 91:616.
172. Atkinson, A.j. and Bennett,j. E., 1968, Experience with a new skin test antigen prepared
from Gryptococcus neojOT1fllJns, Am. Rev. Respir. Dis. 97:637-643.
173. Muchmore, H. G., Feiton, E G., Salvin, S. B., and Rhoades, E. R, 1969, Delayed hypersen-
sitivity to cryptococcin in man, Sabouraudia 6:285-288.
174. Sehimpff, S. C. and Bennett, J. E., 1975, Abnormalities in eell-mediated immunity in

patients with Gryptococcus neoformans infeetion,j. Allergy Glin. Immunol. 55:430-441.
175. Kauffman, C. A., Bergman, A. G., Severance, P. j., and McClatchey, K. D., 1981, Detec-
tion of cryptoeoccal antigen: Com parison of two latex agglutination tests, Am. j. Glin. Patlwl.
75:106-109.
176. Klotz, S. A., Penn, R. L,. and George, R. B., 1986, Antigen detection in the diagnosis of
fungal respiratory infections, Semin. Respir. In/ect. 1:16-2l.
177. Davies, S. E, 1988, Diagnosis of pulmonary fungal infections, Semin. Respir. Infect. 3:
162-171.
178. Murata, K., Khan, A., and Herman, P. G., 1989, Pulmonary parenchymal disease: Evalua-
tion with high-resolution CT, Radiology 170:629-635.
179. Silverman, J. E and Johnsrude, I. S., 1985, Fine needle aspiration cytology of granulo-
matous cryptococcosis of the lung, Acta Gytol. 29:157-161.
180. Fouret, P., Roux, P., Touboul, J. L., Lancastre, E, and Roland, J., 1985, Diagnostic rapide
d'une cryptococcose pulmonaire par examen eytologique du liquide de lavage broncho-
alveolaire (LBA), Arch. Anat. Gytol. Patlwl. 33:90-92.
18l. Feigin, D. S., 1983, Pulmonary cryptococcosis: Radiologie-pathologie correlates ofits three
forms, AJR 141:1263-1272.
182. Gleason, T. H., Hammar, S. P., Barthas, M., Kasprisin, M., and Bockus, D., 1980, Cytologi-
cal diagnosis of pulmonary cryptococcosis, Arch. Patlwl. Lab. Med. 104:384-387.
183. Smith, ES., Gibson, P., Nicholls, T. T., and Simpson, J. A., 1976, Pulmonary resection for
loealized lesions of eryptococcosis (torulosis): A review of eight cases, Tlwrax 31:121-126.
184. Jensen, W A., Rose, R. M., Hammer, S. M., and Karchmer, A. W, 1985, Serologie diagnosis
of focal pneumonia caused by Gryptococcus neoformans; Am. Rev. Respir. Dis. 132:189-191.
185. George, R. B., Jenkinson, S. G., and Light, R. W, 1978, Fiberoptic bronchoscopy in the
diagnosis of pulmonary fungal and nocardial infections, Ghest 73:33-36.
186. Subramanian, S., Kherdekar, S. S., Babu, P. G. v., and Christianson, C. S., 1982, Lipoid
pneumonia with Gryptococcus neoformans eolonisation, Tlwrax 37:319-320.
187. Lewis, j. L. and Rabinovich, S., 1972, The wide spectrum of cryptocoecal infeetions, Am.
j. Med. 53:315-322.
188. Spiekard, A., 1973, Diagnosis and treatment of cryptococcal disease, South. Med. j. 66:
26-31.
189. Khoury, M. B., Godwin, J. D., Ravin, C. E., Gallis, H. A., Halvorsen, R. A., and Putnam,
C. E., 1984, Thoracie cryptococcosis: Immunologie competenee and radiologie appear-
anee, AJR 141:893-896.
190. McAllister, K., Ognibene, A. J., Davis, C.D., Jr., and Carpenter, J. L., 1984, Cryptococcal
pleuro-pulmonary disease: Infection of the pleural fluid in the absence of disseminated
cryptococcosis. Case report, Milit. Med. 149:684-686.
191. Young, E. J., Hirsh, D. D., Fainstein, v., and Williams, T. W, 1980, Pleural effusions due to
Cryptococcus neoformans: A review of the literature and report of two cases with eryptoeoeeal
antigen determinations Am. Rev. Respir. Dis. 121:743-746.
192. Duvie, M., Lowe, L., Rios, A., MaeDonald, E., and Vanee, P., 1987, Superficial phaeohypho-
mycosis of the seroturn in a patient with the acquired immunodeficiency syndrome, Arch.
Dermatol. 123:1597-1599.
193. Epstein, R., Cole, R., and Hunt, K. K., 1972, Pleural effusion secondary to pulmonary
cryptoeoeeosis, Chest 61:296.
194. Salyer, W R. and Salyer, D. C., 1974, Pleural involvement in eryptococcosis, Chest 66:139.
195. Tenholder, M. E, Ewald, E W,Jr., Khankhanian, N. K., and Crosby,J. H., 1992, Complex
cryptoeoceal empyema, Ghest 101:586-588.
196. Perkins, W, 1969, Pulmonary cryptoeoceosis: Report on the treatment of nine cases, Dis.
Chest 56:389-394.
197. Inoue, N., Uozumi, T., Yamamoto, T., Ohishi, T., Murai, Y., Murakami, M., and
Yoshimatsu, H., 1984, Treatment of cryptococcal meningitis with pulmonary granuloma,J
Neurol. 231:109-11l.
198. Majid, A. A., 1989, Clinical notes: Surgical resection of pulmonary cryptococcomas in the
presence of cryptococcal meningitis, J R. Colt. Surg. Edinb. 34:332-333.
199. Chechani, V. and Kamholz, S. L., 1989, Recommended dosage change for amphotericin
B, J Infect. Dis. 160:555.
200. Bennett,J. E., Dismukes, w., Duma, R.J., Medoff, G., Sande, M. A., Gallis, H., Leonard,J.,
Fields, B. T., Bradshaw, M., Haywood, H., McGee, Z. A., Cate, T. R., Cobbs, C. G., Warner,
J. F., and Alling, D. w., 1979, A comparison of amphotericin B alone and combined with
ßucytosine in the treatment of cryptococcal meningitis, N. Engl. J Med. 301:126-13l.
201. Schmutzhard, E. and Vejajjiva, A., 1988, Treatment of cryptococcal meningitis with high-
dose, long-term combination amphotericin Band ßucytosine, Am. J Med. 85:737-738.
202. Stamm, A. M., Diasio, R. B., Dismukes, W. E., Shadomy, S., Cloud, G. A., Bowles, C. A.,
Karam, G. H., and Espinel-Ingroff, A., 1987, Toxicity of amphotericin B plus ßucytosine
in 194 patients with cryptococcal meningitis, Am. J Med. 83:236-242.
203. Utz,J. 1'., Shadomy, S., and McGehee, K. F., 1969, 5-ßucytosine: Experience in patients with
pulmonary and other forms of cryptococcosis, Am. Rev. Respir. Dis. 99:975.
204. Fass, R. J. and Perkins, R. L., 1971, 5-ßuorocytosine in the treatment of cryptococcal and
candida mycoses, Ann. Intern. Med. 74:535.
205. Dismukes, W. E., Cloud, G., and Gallis, H. A., 1983, Treatment of systemic mycoses with
ketoconazole: Emphasis on toxicity and clinical response in 52 patients, Ann. Intern. Med.
98:13.
206. Perfect,J. R., Durack, D. T., Hamilton,J. D., and Gallis, H. A., 1982, Failure ofketoconazole
in cryptococcal meningitis, JAMA 247:3349-3351.
207. Denning, D. w., Tucker, R. M., Hanson, L. H., Hamilton, J. R., and Stevens, D. A., 1989,
Itraconazole therapy for cryptococcal meningitis and cryptococcosis, Arch. Intern. Med.
149:2301-2308.
208. Viviani, M. A., Tortorano, A. M., Giani, I'. C., Arici, C., Goglio, A., Crocchiolo, 1'., and
Alma-viva, M., 1987, Itraconazole for cryptococcal infection in the acquired immunodefi-
ciency syndrome (letter), Ann. Intern. Med. 106:166.
209. Viviani, M. A., Tortorano, A. M., Langer, M., Alma-viva, M., Negri, E., Christina, S., Soccia,
S., De Maria, R., Fiocchi, R., and Ferrazzi, 1'., 1989, Experience with itraconazole in
cryptococcosis and aspergillosis,J Inftct. 18:151-165.
210. Saag, M. S., Powderly, W. G., Cloud, G. A., Robinson, P., Grieco, M. H., Sharkey, I'. K.,
Thompson, S. E., Sugar, A., Tuazon, C. u., Fisher, J. F., H yslop, N., Jacobson, J. M., Hafner,
R., and Dismukes, W. F., 1992, Comparison of amphotericin B with ßuconazole in the
treatment of acute AIDS-associated cryptococcal meningitis, N. Engl. J Med. 326:83-89.
211. Larsen, R. A., Leal, M. A. E., and Chan, L. S., 1990, Fluconazole compared with
amphotericin B plus ßucytosine for cryptococcal meningitis in AIDS, Ann. Intern. Med.
113:183-187.
212. Stern, J. J., Hartman, B. J., Sharkey, 1'., Rowland, v., Squires, K. E., Murray, H.W., and
Graybill, J. R., 1988, Oral ßuconazole therapy for patients with the acquired immunodefi-
ciency syndrome and cryptococcosis: Experience with 22 patients, Am. J Med. 85:477-480.
213. Bozette, S. A., Larsen, R. A., Chiu,J., Leal, M. A. E.,Jacobsen,J., Rothman, 1'., Robinson, P.,
Gilbert, G., McCutchan, J. A., Tilles, J., Leedom, J. M., and Richman, D. D., 1991, A
placebo-controlled trial of maintenance therapy with ßuconazole after treatment of crypto-
coccal meningitis in the acquired immunodeficiency syndrome, N. Engl. J Med. 324:
580-584.
214. Perfect, J. R., 1990, Cryptococcal pulmonary infections, in: Internal Medicine (J. Kassier,
ed.), B.C. Decker, Philadelphia, pp. 713-175.
278 MIRIAM L. CAMERON and JOHN R PERFECT
215. Murray, R. j., Becker, P., Furth, P., and Criner, G. j., 1988, Recovery from cryptococcemia
and the adult respiratory distress syndrome in the acquired immunodeficiency syndrome,
Chest 93:1304-1307.
216. Whitley, T. H., Graybill,j. R., and Alford, R. H., 1976, Pulmonary cryptococcosis in chronic
Iymphocytic leukemia, South. Med. I 69:33-37.
217. Hardy, R. E., Cummings, C., Thomas, F., and Harrison, D., 1986, Cryptococcal pneumonia
in a patient with Siekle Cell disease, Chest 89:892-894.
218. Christoph, 1., 1990, Pulmonary Cryptococcus neoformans and disseminated Nocardia
brasiliensis in an immunocompromised host, North Carolina Med. I 51:219-220.
219. Cameron, M. L., Bartlett,j. A., Gallis, H. A., and Waskin, H. A., 1991, Manifestations of
pulmonary cryptococcosis in patients with acquired immunodeficiency syndrome, Rev.
lrifect. Dis. 13:64-67.
220. Blanc, M., Leuenberger, P. H., and Favez, G., 1982,Cryptococcose pulmonaire invasive
isolee. Presentation d'un cas, Schweiz. Med. Wochenschr 112:421-424.
221. Eng, R. H., Bishburg, E., and Smith, S. M., 1986, Cryptococcal infections in patients with
acquired immune deficiency syndrome, Am. I Med. 81:19-23.
222. Wasser, L. and Talavera, w., 1987, Pulmonary cryptococcosis in AIDS, Chest 92:692-695.
223. Clark, R. A., Greer, 0. L., Valainis, G. T., and Hyslop, N. E., 1990, Cryptococcus neoformans
pulmonary infection in HIV-l-infected patients,I Acquir. Immune Defic. Syndr. 3:480-485.
224. Chechani, V. and Kamholz, S. L., 1990, Pulmonary manifestations of disseminated crypto-
coccosis in patients with AIDS, Chest 98:1060-1066.
225. Khardori, N., Butt, F., and Roiston, K. V. 1., 1988, Pulmonary cryptococcosis in AIDS, Chest
93:1319-1320.
226. Suster, B., Akerman, M., Orenstein, M., and Wax, M. R., 1986, Pulmonary manifesta-
tions of AIDS: Review of 106 episodes, Radiolog) 161:86-93.
227. Perla, E. N., Maayan, S., Miller, S. N., Ramaswamy, G., and Eisenberg, H., 1985, Dissemi-
nated cryptococeosis presenting as the adult respiratory distress syndrome, N.Y. State
I Med. 85:704-706.
228. Similowski, T., Datry, A., jais, P., Katlama, C., Rosenheim, M., and Gentilini, M., 1989,
AIDS-associated cryptoeoeeosis eausing adult respiratory distress syndrome, Respir. Med.
83:513-515.
229. Miller, W. T. and Edelman, j. M., 1990, Cryptocoeeal pulmonary infection in patients with
AIDS: Radiographie appearanee, Radiology 175:725-728.
230. Loerine, A. M., Bottone, E. j., Finkei, L. j., and Teirstein, A. S., 1988, Primary eryptococeal
pneumonia mimieking Pneumocystis carinii pneumonia in a patient with AIDS, Mt. Sinai
I Med. 55:181-186.
231. Eng, RH. K., Bishburg, E., Smith, S. M., and Kapila, R, 1986, Cryptocoeeal infeetions in
patients with acquired immune deficieney syndrome, Am. I Med. 81:19-23.
232. Sobonya, R. E., Barbee, R. A., Wiens,j., and Trego, 0.,1990, Deteetion of fungi and other
pathogens in immunocompromised patients by bronehoalveolar lavage in an area endemie
for coccidioidomycosis, Chest 97:1349-1355.
233. Malabonga, V. M. and Kamholz, S. L., 1991, Utility ofbronehoscopic sampling techniques
for cryptococcal disease in AIDS, Chest 99:370-373.
234. Witt, 0., McKay, 0., Schwam, L., Goldstein, 0., and Gold, j., 1987, Acquired immune
deficieney syndrome presenting as bone marrow and mediastinal cryptococcosis, Am.
I Med. 82:149-150.
235. Dismukes, W. E., 1988, Cryptococcal meningitis in patients with AIDS, I Inftct. Dis.
157:624-628.
236. Sugar, A. M. and Saunders, C., 1988, Oral flueonazole as suppressive therapy of dissemi-
nated cryptococcosis in patients with acquired immunodeficiency syndrome, Am. I Med.
85:481-489.
237. Zuger, A., Schuster, M., Sjmberkoff, M. S., Rahal, ]. ]., and Holzman, R. S., 1988,
Maintenance amphotericin B for cryptococcal meningitis in the acquired immunodefi-
ciency syndrome (AIDS), Ann. Intern. Med. 104:592-593.
238. King, C., Finley, R., and Chapman, S. w., 1990, Prostatic cryptococcal infection, Ann. Intern.
Med. 113:720.
239. Larsen, R. A., Bozzette, S., McCutchan, A., Chiu,]., Leal, M. A., and Richman, D. D.,1989,
Persistent Cryptococcus neoformans infection of the prostate after successful treatment of
meningitis, Ann. Intern. Med. 111:125-128.
16
Influenza Viruses
RICHARD V. SPERA, JR., and DAVID H. SHEPP
1. INTRODUCTION
Influenza virus es are orthomyxoviruses that infect a wide range of mammalian

hosts, including humans. Immunity develops after infection, but is strain specific.
Influenza has a remarkable ability to alter its genetic and antigenie makeup such
that new strains or subtypes emerge regularly. As a result, yearly epidemie or
pandemie disease occurs, affecting millions of individuals. The vast majority of
influenza infections are brief, self-limited, and cause only minor morbidity.
Nevertheless, in developed countries, uncomplicated cases cause large losses in
economic productivity as a result of absenteeism in the workplace and significant
consumption of health-care resources for visits to health-care providers and
prescriptions of symptomatic remedies. Although complications of influenza
occur in only a sm all fraction of cases, complicated influenza is associated with
major morbidity and mortality. Because of the enormous number of infections
that occur every year during seasonal epidemics, the aggregate impact of influ-
enza virus infection on public health is very great. Much of the serious morbidity
and mortality in influenza is a consequence of superimposed bacterial respiratory
infection. This predisposition to serious bacterial infection results from the
deleterious effects of influenza on host immune mechanisms essential for resis-
tance to common bacterial respiratory pathogens.
RICHARD V. SPERA, JR., and DAVID H. SHEPP • Department of Medicine, North Shore
University Hospital, Cornell University Medical College, Manhasset, New York 11030. Present
address for R. V.S., Jr.: Department of Internal Medicine, The Brooklyn Hospital Center,
Brooklyn, New York, 1120l.
281
282 RICHARD V. SPERA, JR., and DAVID H. SHEPP
2. EPIDEMIOLOGY
In temperate cIimates, outbreaks of influenza are seasonal, with the greatest

ineidenee in the late autumn and winter. The yearly oeeurrenee of influenza virus
epidemies aeeounts in part for the inereased population death rate seen during
winter months. l -4 Exeess mortality above levels usually seen in winter months is
associated with the severity of the influenza epidemie in a partieular year.
Complieations of influenza virus infeetion oeeur with greater frequeney in in-
fants, the elderly, those with ehronie eardiovaseular or pulmonary disease, and
immunocompromised hosts. Infeetion in these groups aeeounts for the vast
majority of deaths.I In tropical cIimates, outbreaks are not seasonal, but tend
to follow major weather ehanges. 5
An influenza epidemie is an outbreak of influenza oecurring within a
geographieally defined population. The natural history of an influenza epidemie
has been deseribed in detail. 6 An epidemie usually peaks within 2-3 weeks of
onset, and lasts approximately 5-6 weeks. Initially, there is an inerease in febrile
respiratory illnesses in ehildren followed by a similar inerease in adults. The
number of hospital admission for pneumonia, eroup, and exaeerbations of
eongestive heart failure and ehronie obstruetive pulmonary disease inerease. 6 ,7
During major epidemie years, these events are associated with an overall in-
ereased death rate.I-7 The overall attaek rate has been reported as between 10
and 20% in most outbreaks. Attaek rates of 50% have been reported in seleeted
subpopulations, however,s
An influenza pandemie is a worldwide epidemie. There have been five such
pandemies in the past 100 years-1890, 1900, 1918, 1957, and 1968. Eaeh has
been associated with the appearanee of a new viral subtype to whieh few persons
had preexisting antibodies. The potential effeets of this dramatie change in viral
strueture are demonstrated by the 1918 pandemie, whieh was responsible for
21,000,000 deaths. 5
3. CLINICAL OVERVIEW
3.1. Uncomplicated Influenza

Although mild symptoms and partial clinieal syndromes are eommon and
underdiagnosed, classic influenza is eharaeterized by a sudden onset of symp-
toms 1-3 days after exposure. Fever, chilIs, headaehe, myalgia, arthralgia, and
malaise are the predominant symptoms. Nonproduetive eough is a prominent.
feature but is sometimes absent at initial presentation. Fever, whieh ean reaeh
40°C, usually lasts 3-4 days. Gastrointestinal symptoms such as nausea and
vomiting may oeeur but suggest other diagnoses when they predominate. The
respiratory symptoms, whieh in addition to eough include coryza and sore throat,
appear later in the illness but persist longer than do the constitutional symptoms.
Complete recovery may require up to 2 weeks.
Disease manifestations are generally most severe with influenza A. Influenza
INFLUENZA VIRUSES 283
B produees a similar but less severe syndrome. 9 Influenza C generally produees a

mild, nonspecifie respiratory infeetion and epidemies do not oeeur.l°
3.2. Complications of Influenza

Among the potential eomplieations of influenza, baeterial superinfeetion of
the respiratory tree oeeurs most eommonly. Sinusitis, otitis media, bronehitis, and
pneumonia all oeeur with inereased frequeney following acute influenza virus
infeetion. Seeondary baeterial pneumonia is the most eommon pulmonary eom-
plieation of influenza virus infeetion and is the major eause of influenza A-
related morbidity and mortality.3.4.11-13 Baeterial pneumonia oecurs most fre-
quently in the elderly and those with signifieant eomorbid illness. Signs and
symptoms of baeterial pneumonia usually appear several days to 2 weeks after
eonvaleseenee from the typical symptoms of influenza has begun, making influ-
enza eomplieated by baeterial pneumonia a biphasie illness. The most common
eausative organisms are Streptococcus pneumoniae, Staphylococcus aureus, and Hae-
mophilus inJiuenzae.5
An assoeiation between influenza and other serious baeterial infeetions has
also been suggested. An inereased incidenee of invasive Neisseria meningitidis
infeetions during influenza epidemies has been deseribed.l 4,15 The onset of N.
meningitidis infeetion in relationship to influenza is similar to that for baeterial
pneumonia. Toxie shoek syndrome eaused by eolonization with TSS-I producing
strains of S. aureus associated with influenza also has been reported.l 6 •17
Primary influenza viral pneumonia is a potentially devastating eomplieation.
In influenza pneumonia, the usual signs and symptoms of uneomplieated illness
progress within a few days to respiratory distress and respiratory failure. Viral
eulture of traeheobronehial seeretions yields high titers of influenza virus. Mortal-
ity was reported to be as high as 83% during the 1957-1958 pandemie.l 3
Although this eomplieation has been reported to oeeur in healthy young adults, it
is seen more eommonly in pregnant women and patients with eardiovaseular
disease.l· 3.4 •8 An overlap syndrome having features of both primary viral and
seeondary baeterial pneumonia also may oeeur. It has a mueh more favorable
pro gnosis than does primary viral pneumonia and usually responds favorably to
antibaeterial therapy.13
Decompensated congestive heart failure is the most eommon eardiovascular
eomplieation of influenza virus infeetion. Viral myoearditis and pericarditis have
both been reported, but are rare.l 8 •19 Myositis oeeurs most eommonly in ehildren
infeeted with influenza B.20.21
The most eommon neurologie eomplieation of influenza virus infeetion is
Reye's syndrome. This hepatie and neurologie syndrome has been reported in
association with influenza B more often than influenza A,22 as weIl as other acute
viral infeetions. With rare exeeptions, Reye's syndrome is seen only in ehildren. 23
The illness is eharaeterized by alteration in level of eonseiousness, anormal
eerebrospinal fluid profile, nausea, vomiting and elevated serum ammonia and
hepatic transaminase levels. The mortality rate is 10-40% and severe neurologie
sequelae in survivors is frequent. 22 Use of aspirin to ameliorate the symptoms of
influenza or other viral illnesses greatly inereases the risk of Reye's syndrome. 24
Beeause of this assoeiation, aspirin use should be avoided in ehildren and adoles-
cents with febrile illnesses 25 and ehildren who require ehronie aspirin therapy
should be immunized against influenza. 26 Since the 1980s it has become standard
pediatrie praetiee to use antipyreties other than aspirin, and the incidence of
Reye's syndrome has declined substantiaIly. The pathogenetie meehanism by
whieh eoncurrent influenza Binfeetion and aspirin administration lead to Reye's
syndrome remains unknown.
Eneephalitis, transverse myelitis and Guillain-Barre syndrome have been
reported following influenza. U suaIly, the onset of illness has oeeurred after
recovery from an aeute influenza syndrome and an immunologie meehanism has
been postulated. 27 .28
4. VIRAL STRUCTURE AND FUNCfION
Influenza virus is a spherieal, enveloped, single-stranded, negative sense

RNA virus approximately 80-120 nm in diameter. The influenza virus genome
eneodes ten proteins: hemagglutinin, neuraminidase, nucleoprotein, two matrix
proteins, three RNA polymerases, and two nonstruetural proteins. Influenza
viruses are typed based on antigenie differenees in nucleoprotein and subtyped
based upon antigenie differenees in hemagglutinin and neuraminidase. Major
and minor ehanges in the strueture of hemagglutinin and neuraminidase are
associated with pandemie and epidemie influenza infection. Strains of influenza
virus are distinguished using a nomenclature that includes the virus type, the
geographie area of initial isolation, the strain number, the year of isolation, and
the hemagglutinin and neuraminidase subtype, for example Atrexas/36/91(HIN1).
For nonhuman strains, the species of origin is also indieated.
The viral envelope consists of phospholipid bilayer with four membrane-
assoeiated proteins. Hemagglutinin and neuraminidase are the major surfaee
glycoproteins present on virions and infeeted eells. The matrix protein, M2, is a
membrane protein with a small domain expressed on the surfaee of infeeted
eeIls. 29 The fourth protein, MI matrix protein, is associated with the inner surfaee
of the bilayer. 3o The nucleoeapsid of influenza virus consists of eight fragments of
RNA individually complexed with nucleoprotein and RNA polymerases. This
unique genomie arrangement makes possible the reassortment of genes of influenza
virus and provides a powerful meehanism for antigenie variability. Nucleoprotein
is highly antigenieally eonserved and is the basis for typing of influenza isolates.
Hemagglutinin is one of two major surfaee antigens loeated on the influenza
virus envelope. The strueture of hemagglutinin has been reviewed in detail
previously.31 I t is a trans membrane glyeoprotein eom posed of two subunits of 36
and 26 kDa moleeular weight, is synthesized as a single unit and undergoes
postranslational cleavage into two distinet subunits joined by disulfide bonds.
Cleavage of this molecule at three sites is neeessary for virion infeetivity and is
aecomplished by arginine-speeific proteases ofhost-eell origin. 32 Infeetion of eell
types lacking such proteolytic enzymes results in noninfeetious viral particles, but
infectivity can be enhanced by addition of trypsin to the culture medium. Three

hemagglutinin molecules aggregate on the outer surface of the envelope to form
spike-shaped structures which may be seen in electron micrographs. 33
Hemagglutinin mediates viral attachment and penetration into ciliated col-
umnar epithelial cells. It binds to receptor molecules on the cell surface that are
rich in sialic acid. 34 It also permits viral binding to the erythrocytes of several
mammalian species, causing agglutinination. The ability of infected cell lines
expressing hemagglutinin on the cell surface to bind erythrocytes has been used
in diagnostic virology laboratories as an aid in the identification of influenza
isolates. The active site of hemagglutinin is a pocket formed in the center of each
molecule. Although the active site is not immunogenic, antibody binding to any of
five other epitopes on the molecule inhibits hemagglutinin activity and neutral-
izes viral infectivity.31
Neuraminidase is also a transmembrane glycoprotein. It is composed of a
single chain that does not undergo posttranslational cleavage. 35 Four molecules
aggregate to form a mushroom-shaped structure which, like hemagglutinin
spikes, may be seen on electron micrographs.3 3 Neuraminidase cleaves sialic acid
residues from glycopeptides and glycolipids. 35 The active site ofthe molecule is a
pocket on the distal surface. There are four antigenic sites distinct from the active
site. Antibodies to two of these sites inhibit enzyme activity while antibodies to a
third site have neutralizing activity.6 Much higher titers of antibody to neura-
minidase are required to neutralize viral infectivity in tissue culture than are
required for hemagglutinin, however. 37 ,38 Subneutralizing titers of antibody to
neuraminidase have been shown to decrease the number of new virions liberated
from infected ceIls.38 The function of neuraminidase in viral pathogenesis is
uncertain. Postulated roles include removal of receptor from hemagglutinin,
receptor degradation, promotion of virion release from infected ceIls, and pre-
vention of virion self-aggregation. 35
Nucleoprotein is a structural element of the nucleocapsid and also functions
in the regulation ofviral RNA transcription.3 9 The MI matrix protein is thought
to inhibit transcription. 40 In infected ceIls, the M2 matrix protein appears to
function in assembly and budding of whole virions and is also incorporated into
virions in small amounts. 41 Because resistance to the tricyclic amine antiviral
agents is associated with mutation in M2 protein,42 it probably plays a role in viral
penetration as weIl. Two ofthe influenza virus polymerases, PBl and PB2, appear
to function in the synthesis of complementary RNA from the negative sense virion
RNA. The third polymerase, BPA, functions in synthesis of virion RNA.43 The
function of the two nonstructural proteins is unknown.
5. PATHOGENIC MECHANISMS
5.1. Transmission
Human-to-human spread of influenza virus occurs when infectious virions

reach the respiratory epithelium either by inhalation of respiratory droplets or
direct contact with fomites contaminated by these droplets. The primary site of
replication of influenza virus is the ciliated columnar epithelial cell of the
respiratory tree. In adults, viral replication usually peaks 3 days after exposure,
declines thereafter and is no longer detected about 1 week after onset of illness. 44
In children, viral shedding can last up to 2 weeks. 45
5.2. Histopathology
Bronchial biopsies from patients with clinically uncomplicated influenza
show desquamation of the respiratory epithelium and loss of cilia, edema,
hyperemia, and a mononuclear cell infiltrate in the submucosa. 46 Studies of lung
tissue taken from fatal cases of influenza pneumonia show hemorrhagic, necrotiz-
ing tracheobronchitis, hemorrhage and hyaline membrane formation within
alveoli, and an interstitial mononuclear inflammatory infiltrate. 13 ,47 Viral anti-
gen is found in type 1 and 2 pneumatocytes and in alveolar macrophages. 48
5.3. Antigenie Variation

The success of influenza virus as a human pathogen can be attributed in
large part to its ability to evade or inhibit host immunologie mechanisms. Both
hemagglutinin and neuraminidase undergo two forms of antigenie variation.
The first, called anti genie drift, is a minor structural change in the antigenic sites
of either hemagglutinin or neuraminidase. Antigenie drift arises from point
mutations in either the hemagglutinin or neuraminidase genes. Accumulation of
point mutations leads to changes in the amino acid sequences in one or both of
these proteins, which eventually are sufficient to change the three-dimensional
structure such that the antigenie determinants are no Ion ger recognized by
strain-specific antibody.49-52 Antigenie drift has been shown to occur with in-
creasing frequency after the appearance of a new viral subtype and is associated
with epidemics of disease. 51
The second form of variation, called antigenie shift, produces a major
change in the amino-acid sequence of either hemagglutinin or neuraminidase
such that either an entirely new subtype of virus appears, or an old subtype
reappears after many years. Each of five influenza pandemics from 1890 to 1968
resulted from antigenic shift. Only the 1977 shift did not result in a pandemic. 5
There are at least three proposed mechanisms whereby antigenic shift may
occur. 32 A new viral subtype may arise from the reassortment of a mammalian or
avian hemagglutinin gene into a human strain. This has been shown to occur
both in vitro and in vivo. 53 ,54 This mechanism is postulated to have occurred with
the AlHong Kong/68(H3N2) strain, since there is 98% homology with the
hemagglutinin gene from a duck virus. 55 Alternately, astrain may persist in an
animal reservoir and reappear years later. This mechanism is postulated for the
appearance of the AlUSSR/77(HIN1) strain. DNA-homology studies demon-
strated only eight base-pair changes between this isolate and AlFW/50(HIN1)
which circulated 27 years earlier. 56 Also, although the AlHong Kong/68(H3N2)
appears to have arisen from a human-avian gene reassortment, its persistence in

swine was demonstrated nine years after the initial human epidemic. 57 A third
possible mechanism involves a genetic recombination or accumulation of muta-
tions in an animal strain resulting in a new virus with tropism for humans. This is
postulated to have occurred in 1918 when an HINI strain apparently caused
simultaneous pandemic disease in both humans and in swine. 58
These two mechanisms of antigenic variation, shift and drift, allow influenza
A to escape from host immunologic memory induced by prior exposures to
influenza, and endow the virus with a remarkable ability to produce epidemics of
illness on a yearly basis and pandemic disease approximately every 10-20 years.
Influenza Band C are more genetically stable than influenza A and exhibit
considerably less tendency toward antigenic variation.
5.4. Effeet of Infeetion on Immunologie Defenses

The immunopathogenesis of influenza-related bacterial infections has been
the subject of numerous investigations. Experimental evidence suggests that
influenza virus infection induces multiple abnormalities in host defenses leading
to the increased susceptibility to infection by bacterial pathogens.
Adherence of Staphylococcus aureus to epithelial cells from the upper respira-
tory tract is increased in ferrets infected with influenza virus. 59 Increased adher-
ence of S. aureus, H. inJiuenzae, and S. pneumoniae to oropharyngeal epithelium in
human volunteers experimentally infected with influenza virus also has been
demonstrated. 60 ,61 This increase in colonization ofthe upper respiratory tract by
common respiratory pathogens is thought to be the initiating event leading to
bacterial superinfection of the sinuses, bronchi, or pulmonary parenchyma.
Normal mucociliary clearance of small numbers of inhaled or aspirated
bacteria in the trachea or bronchi is an important mechanism for maintenance of
sterility in the lower respiratory tree. During influenza virus infection, mucocili-
ary function is abnormal. 62-64 Thus, following colonization of the upper respira-
tory tree because of alterations in epithelial adherence, mucociliary dysfunction
allows bacteria to invade the lower respiratory tree and predisposes to the
development of secondary bacterial pneumonia.
Ifbacteria gain entry to the normally sterile portions of the respiratory tree, a
variety of resident or migratory leukocytes may still function to prevent the
development of infection. Influenza virus alters the function of immune effector
cells, however. Influenza virus infection of neutrophils is abortive; viral proteins
but not infectious virus are produced. 65 Infection of neutrophils with influenza
virus in vitro depresses chemotaxis, oxidative burst, secretory function, and
bactericidal activity.66-73 Phagocytic function has been found to be normal or
depressed. 71 ,74 Similar functional abnormalities also have been seen in neutro-
phils taken from humans or experimental animals with influenza virus infec-
tion. 75- 80 Stimulation of an oxidative burst in monocytes and neutrophils by
influenza virus results in a decreased oxidative burst when these cells are subse-
quently challenged with bacteria. 71 This reduction in oxidative burst does not
appear to result merely from depletion of oxidative reserve capacity caused by

repetitive stimulation, but rather from metabolie dysfunction induced by viral
infection.B 1The decrease in oxidative burst particularly may impair intracellular
killing of catalase-positive bacteria such as Staphylococcus aureus, one ofthe classic
pathogens in postinfluenza bacterial pneumonia.
Influenza virus readily infects macrophages. Hemagglutinin, neuraminidase
and matrix proteins are expressed on infected ceIls,82 but infectious virus is
produced ineffectively or not at aIl.82-84 Human peripheral blood macrophages
infected with influenza virus provide a depressed level of accessory cell function
to mitogen-driven peripheral blood lymphocyte proliferation, whereas accessory
ceU function of alveolar macrophage appears to be unaltered.B5,86 Phagocytic
function of murine alveolar macrophages is depressed 3 days after experimental
influenza, then returns toward baseline.B7 Significant dysfunction of alveolar
macrophage phagocytic and bactericidal activity was not, however, evident in
another murine model 1 week after infection, despite persistently abnormal
whole-Iung clearance of Staphylococcus aureus, 88 suggesting that defects other than
alveolar macrophage abnormalities contribute to the pathogenesis of staphylo-
coccal pneumonia following influenza.
Human macrophages exposed to influenza virus produce large amounts of
INF-a and INF_'Y85 ,86,89,90 and elevated levels of interferon appear in the serum
and respiratory secretions of patients with influenza. 91 ,92 Influenza A virus
infection of human peripheral blood and rat alveolar macrophages augments
production ofboth tumor necrosis factor alpha messenger RNA and tumor necrosis
factor alpha and greatly potentiates release of tumor necrosis factor in response to
bacteriallipopolysaccharide. 84 Acute influenza has been associated with transient
anergy to common skin-test antigens. 93 Depression of delayed-type hypersensitiv-
ity reactions probably results from the effects of influenza on macrophage function.
Influenza virus also abortively infects lymphocytes. After exposure to influ-
enza, expression of viral protein can be found in 70% or more of resting lympho-
cytes. 94 Effects of clinical influenza on lymphocyte number and function have
been observed. Lymphopenia occurs du ring uncomplicated influenza in young
adults and is caused by proportional decreases in both T and B ceIls.95 Among T
ceIls, both the CD4 and CD8 subsets are decreased. In mice with experimental
influenza, there is a marked influx ofboth CD4 and CD8lymphocytes into both
lung exudate and mediastinallymph nodes, with a predominance of activated
CD8 lymphocytes, suggesting that the peripherallymphopenia noted in humans
may be a result of trafficking of lymphocytes into the infected tissues. 96 Reduced
lymphocyte proliferative responses to mitogen have been described,95 but other
investigators have found that mitogen-driven proliferation is unaltered if acces-
sory cell function from normal uninfected macrophages is provided. 85 Like
macrophages, lymphocytes also respond to influenza virus infection with high
levels of interferon production. 85 ,90 Poke weed mitogen-driven synthesis of IgG
by B-Iymphocytes is depressed following in vitro infection with influenza viruS. 94
Natural killer-cell activity is increased during clinical influenza, probably as a
consequence of the high levels of interferon produced. 91 ,95
5.5. Effect of Bacterial Infection on Influenza Virus

Although influenza virus facilitates bacterial superinfection, bacterial infec-
tion also may enhance influenza virus infection. The formation of fully infectious
influenza virions requires trypsin-like protease activity in the host cell to cleave the
hemagglutinin precursor polyprotein into its mature form. Not all host cells
perform this function with full efficiency. Several bacterial proteases, especially
those of staphylococci and streptococci, are capable of providing this trypsin-like
activity and have been shown to augment influenza virus infectivity in mice. 97 ,98
Thus, influenza virus infection with concurrent bacterial infection may be more
virulent than uncomplicated influenza.
5.6. Effect of Infection on Respiratory Physiology

Influenza virus infection of the lower respiratory tree can alter airway
physiology. In many patients, pulmonary function tests show mild, reversible
obstructive airway changes and decreased carbon monoxide diffusing capac-
ity.99-101 Influenza virus induces production of a histamine-releasing factor by
monocytes that may be the mechanism by which airway reactivity is increased. 102 ,103
Such alterations may contribute to the pathogenesis of croup in young children 104
and exacerbations of asthma or chronic obstructive pulmonary disease in chil-
dren or adults.1° 5 ,106
6. HOST IMMUNE RESPONSE TO INFLUENZA VIRUS
6.1. Protective Immunity to Influenza

Because hoth influenza virions and influenza virus-infected cells express
foreign antigen, they are targets for recognition by the immune system. Hemag-
glutinin, neuraminidase, matrix proteins 1 and 2, nucleoprotein and the three
RNA polymerases are all potent antigens and stimulate both humoral and cellular
responses. 107 Certain antigenic determinants of influenza, such as nucleoprotein,
matrix proteins 1 and 2, and the RNA polymerases, are highly conserved among
virus isolates within a specific virus type. Antibody to these antigens is not
protective against infection, however. 108 Whereas hemagglutinin exhibits greater
antigenic variability, antibody to it is associated with resistance to infection as a
result of neutralization of infectivity. Several lines of evidence suggest that
antibody to hemagglutinin is protective against infection. Passive transfer of
antihemagglutinin antibodies protects animals from experimental influenza,109
and protection against influenza in the neonate correlates with passive acquisi-
tion of maternal neutralizing or antihemagglutinin antibodies. l1O A recombinant
vaccinia virus expressing the influenza hemagglutinin gene product but not other
influenza antigens can induce complete protective immunity in experimental
animals.I l1 In humans, the presence of antihemagglutinin antibodies in nasal
secretions is associated with protection against experimental challenge.I 12 ,113
Protective anti-hemagglutinin antibodies are only partially cross-reactive among

strains belonging to a given viral subtype, however. Therefore, antigenic drift can
result in outbreaks of both infection and clinical illness as a result of subtle
changes in the stereochemistry of the hemagglutinin molecule that cause it to
bind preexisting neutralizing antibody less avidly. Further evidence for the role of
antihemagglutinin neutralizing antibodies in protective immunity is provided by
the genetic variability of the hemagglutinin molecule. The amino-acid sequences
of the neutralizing antibody-binding epitopes change with the greatest frequency
during antigenie drift.3 I,I08
Antibody to neuraminidase provides less potent viral neutralization and is
not protective against infection except at titers that are probably not physiologi-
cally relevant. 37 ,I09 Antineuraminidase antibodies can decrease the number of
intact virions released from infected cells38 and are associated with amelioration
of symptoms and more rapid resolution of clinical influenza1l4 Antibodies to
nucleoprotein or matrix proteins have no protective effect. 109
6.2. Mucosal Immunity

Because the respiratory tract is the portal of entry of influenza virus, mucosal
immunity, especially local secretion of IgA, plays a critical role in resistance to
infection. In mice, passive transfer of polymerie antihemagglutinin IgA results in
appearance of antibody in nasal secretions and confers protection against experi-
mental challenge. Proteetion is not achieved with administration of monomeric
antihemagglutinin IgA or IgG.l15 Adults or children with any detectable anti-
hemagglutinin IgA antibody in nasal secretions are resistant to challenge with
homologous virus. ll2•113 Antihemagglutinin IgG antibody in nasal secretions also
correlates with proteetion, although at higher titers than are required for IgA.1l3
Individuals who are seronegative for a given hemagglutinin subtype produce
similar nasal-wash antibody responses to naturally occurring infection and intra-
nasal vaccination with live, attenuated virus. Within 14 days of intranasal expo-
sure, IgA, IgG, and IgM antihemagglutinin antibodies can be detected, with IgA
titers exceeding those for IgG or IgM.I08.116 Both IgA and IgM are locally
produced, whereas most IgG in mucosal secretions derives by passive diffusion
from serum. 1l6.117 Following reexposure to influenza in seropositive individuals,
the mucosal antibody response is predominantly IgA.1l8
Induction of mucosal immunity by live virus infection and parenteral immu-
nization with inactivated vaccine differ. Nasal-wash IgA was detectable in 83% of
subjects given live attenuated virus vaccine as compared to only 38% of those
given inactivated vaccine, and titers of IgA were 5- to 8-fold greater in the live
vaccine recipients. 1l9 Subjects previously immunized with inactivated vaccine
respond to homologous live virus infection with an anamnestic mucosal IgA and
IgG response, however. ll3 Although intranasal administration of inactivated virus
is poody immunogenic in seronegative subjects, it also elicits an anamnestic
mucosal IgA response in seropositive subjects.I20 In contrast to the effect on IgA
induction, 94% of subjects immunized with inactivated vaccine generated nasal-
wash IgG, as compared to only 59% given live, attenuated virus. H9 Among
responders, the nasal-wash IgG titers were similar in both groups.H9
6.3. Humoral Immunity

Although perhaps less independently related to immunity than mucosal
antibody, serum antibody also correlates with protection from influenza. Infants
with higher levels of passively acquired maternal neutralizing or antihemagglu-
tinin antibodies experienced more prolonged protection against influenza than
those with low or absent levels. HO High levels of serum antihemagglutinin IgG are
associated with decreased viral shedding after experimental challenge. ll3 Higher
levels of serum antineuraminidase antibody also have been correlated with less
severe clinical disease manifestations. H4 Following primary influenza infection in
adults seronegative for subtype-specific antibody, IgG or IgA, but IgM responses
are detected in 86-100% of sera. H8 In sero positive individuals, two-thirds to
three-quarters have a significant rise in serum IgG or IgM but IgM responses are
rare. Serum IgA responses are most likely to occur in those with concurrent local
IgA production in nasal secretions. H6.118
Induction of antihemagglutinin antibodies in the serum of seronegative
children by live, attenuated virus vaccine and inactivated virus vaccines is simi-
lar. 121 A single dose of some inactivated vaccine preparations may, however, elicit
protective levels of serum antibody in fewer than one-half of immunologically
naive children.122.123 A second dose has been found to result in protective
antibody levels in 74-93% of vaccines and therefore a two-dose immunization
schedule is recommended in children.26.122.123 In adults who are seronegative for
subtype-speeific antihemagglutinin antibody, inactivated virus produces serum
IgG titers 2-4 times in excess of those produced by live, attenuated virus
vaccine. Hg The reduced immunogenicity of live, attenuated influenza virus vac-
eines in adults is thought to be caused by partial suppression of viral replication by
the host's anamnestic immunologie response to antigens other than hemagglu-
tinin or neuraminidase which are more broadly shared among different influenza
A subtypes. In particular, type-speeific cytotoxic T lymphocytes can limit viral
replication and so reduce expression ofnew hemagglutinin antigens to which the
individual has not previously been exposed. Titers of antibody to shared anti-
genie determinants to which immunologie memory exists may be boosted but
lower titer antibody responses to new antigenie determinants are produced. 108
6.4. Durability of Antibody Responses

The durability of both the local and serum antibody response depends on
age and prior experience with natural influenza or vaccine. Antibody titers
induced by natural infection persist for years. When studied during influenza
vaccination programs in 1978, most individuals 25 years of age or older were
found to remain seropositive for the Hl subtype of hemagglutinin despite the
absence of this antigen in eirculating influenza strains for more than twenty
years.124.125 Antibody induced by inactivated vaccine declines moderately over

time. In adults seronegative for the immunizing hemagglutinin subtype, mean
titers decline about twofold 6 months after vaccination. 125,126 NasaligA and IgG
and serum IgG are, however, still detected for at least 7 months after immuniza-
tion at levels associated with protection. 1l9 ,127 In children, antibody declines may
be fourfold or greater after inactivated virus vaccine,108 Responses to live, attenu-
ated virus vaccine are more durable,l°8 Both IgA and IgG are detectable in nasal
washings for up to 12 months in approximately 50% of children following
intranasal exposure to either wild-type virus or live, attenuated vaccine virus
strains. 108
6.5. Cell-Mediated Immunity

Cell-mediated immunity to influenza plays a vital role in limiting influenza
virus infection. The cytotoxic T-Iymphocyte response is predominantly mediated
by cells bearing the CD8 marker and restricted by the class I major histocom-
patibility (MHC) antigen. 128 CD4+ cytotoxic T lymphocytes restricted by the class
Il MHC that recognize influenza virus antigens also have been identified,
however.l 29 Unlike the mucosal and serum antibody response, the predominant
cytotoxic T-Iymphocyte response to influenza is type specific, a finding that
suggested that conserved antigens are recognized by cytotoxic T lympho-
cytes.l 30 ,131 Experiments employing either vaccinia recombinants or synthetic
peptides have confirmed that nucleoprotein is a major target for influenza-
specific cytotoxic T lymphocyte.l 32 ,133 Other viral antigens, however, including
hemagglutinin, polymerase, and matrix protein, are recognized by cytotoxic T
lym phocytes. 107 ,134-137
Influenza-specific cytotoxic T-Iymphocyte activity does not prevent infection
but is an important defense leading to resolution of infection. In mice, adoptive
transfer of polyclonal influenza virus-specific class I MHC-restricted cytotoxic
T-Iymphocytes or cytotoxic lymphocyte clones specific for nucleoprotein or hem-
agglutinin results in more rapid recovery from experimental influenza pneu-
monia and decreased viral titers in lung tissue. 13 8-140 Type-specific cytotoxic
T-Iymphocyte activity is associated with improved clearance of virus after experi-
mental influenza in humans, even in the absence of subtype-specific antibody.l3l
Class I MHC-restricted cytotoxic T-Iymphocytes are not an absolute requirement
for recovery, however. Transgenic or CD8 lymphocyte-depleted mice clear influ-
enza virus in the absence of a class I MHC-restricted cytotoxic T-Iymphocyte
response, although they do so less efficiently than normal mice. 141 ,142 Adoptive
transfer of class Il MHC-restricted cytotoxic T-Iymphocyte clones to congenitally
athymic mice is associated with resolution of infection either because of cytotoxic
activity or enhanced antibody formation.l 43 Cytotoxic T-Iymphocyte responses
can be induced in humans by both inactivated and live, attenuated virus vaccines,
but are transient,l44 Induction of a somewhat broader range of responses by live,
attenuated virus vaccine has been suggested. 145
Natural killer cells are cytotoxic for influenza virus-infected cells in a
manner that is neither antigen-specific nor restricted by recognition of the major

histocompatibility complex. Natural killer-cell activity is augmented during influ-
enza virus infection as a consequence of the high levels of interferon pro-
duced. 91 ,95 Antibody-dependent cell-mediated cytotoxicity has also been demon-
strated.l 46 The triggering antibodies are antihemagglutinins, and lysis of the
infected cell is mediated by non-T, non-B lymphocytes. 147,148 Cytotoxicity to
influenza virus-infected cells mediated by antibody and complement also oc-
curS. 149 The antibodies participating in this reaction are antihemagglutinins,
although activity of more broadly reactive antibodies has been suggested. 150
Delayed-type hypersensitivity response to influenza A antigens has been demon-
strated in humans. This response is semiquantitatively correlated with in vitro
antigen-induced T-cell proliferation, but does not correlate well with resistance to
infection. 151
6.6. Immunopathogenesis
Although the immune response to influenza benefits the host by preventing

or limiting infection, there is some evidence that it also injures tissue or produces
other deleterious effects on the host. Mice experimentally infected with influenza
had less lung consolidation when treated with antilymphocyte serum. 152 Congeni-
tally athymic mice excrete virus longer yet have improved survival when com-
pared to immunologically normal mice.l 53 Neutrophil-derived oxygen free radi-
cals playa role in the mediation of pulmonary pathology in a murine model of
influenza virus infection.l 54 These data suggest that T-Iymphocyte and neutrophil-
mediated responses participate in pathogenesis of influenza virus infection. The
relative contribution of immunopathogenic mechanisms and direct viral effects
in causing human influenza is not fully known, however.
7. PREVENTION AND THERAPY
7.1. Vaccination
Influenza virus vaccine provides safe and effective prophylaxis against the
strains incorporated into the formulation. Influenza vaccines presently in use in
the United States are trivalent inactivated-virus vaccines. Vaccine for clinical use is
reformulated annually using influenza strains bearing the hemagglutinin and
neuraminidase antigens seen with greatest frequency in the previous year and
genes governing the ability to replicate efficiently in eggs. At present, two
influenza Astrains and one influenza B strain are included. Vaccines prepared
from intact, formalin-inactivated viruses are termed whole-virus preparations.
Vaccines prepared from disrupted, partially purified hemagglutinin and neur-
aminidase antigens are termed split-product vaccines.3 2 Each does of vaccine
should contain 7-21 I1g of hemagglutinin for each component strain.l 24 Neur-
aminidase is too labile to permit antigenic mass standardization.
Following immunization, neutralizing antibody at a titer of 1:8 or greater or

hemagglutinin-inhibiting antibody at a titer of 1:40 or greater are associated with
protection against clinical illness. 5 Individuals with prior exposure to influenza
have a satisfactory response to vaccine containing standard doses of hemagglu-
tinin, whereas seronegative individuals require a single dose of least 60 ~g of
hemagglutinin or a second dose to produce equivalent geometrie mean titers of
neutralizing antibody.12!l-126 Therefore, two doses given at least 1 month apart are
recommended for children less than 9 years of age who are more likely to be
seronegative, and for seronegative adults who are receiving influenza vaccine for
the first time. 26,122,123
The immunogenicity of influenza-virus vaccine in high-risk and immuno-
compromised patients has been studied. Elderly vaccinees generally produce
titers equivalent to those of younger individuals. 155,156 Children less than 6
months of age with chronic cardiac or pulmonary conditions have lower responses
to standard immunization than do children who are 6 months of age or 01der. 157
In the younger age group, protective antibody levels to HINI strains and influ-
enza B were achieved in less than half of vaccinees. Cancer patients receiving
systemie chemotherapy produce lower titers than do controls. 158 The serum
antibody response in patients with renal failure undergoing dialysis is better than
that seen in similar patients not receiving dialysis. 159 After renal transplantation,
patients receiving immunosuppression with cyclosporine generate lower geomet-
ric mean titers than those treated with azathioprine.I 60
Influenza vaccine has been shown to be somewhat less immunogenic in
patients with systemic lupus erythematosus than in normal controls, but antibody
responses were least affected in patients treated only with low-dose or alternate-
day corticosteroids. 161 No increase in autoantibody production or flare of clinical
illness was seen after immunization. Antibody responses to influenza vaccine are
significantly lower in HIV-infected individuals than in HIV-negative controls. In a
group of subjects with early HIV infection and mean CD4lymphocyte counts still
within the normal range, antihemagglutinin antibody responses to some vaccine
components were near normal but only one-third developed protective responses
to certain other vaccine strains.I62 In advanced HIV disease, protective antibody
responses to component vaccine strains ranged from 13 to 40%. Use of
zidovudine was associated with moderate improvement in responsiveness to
vaccine.1 62
The standard by which the clinical utility of influenza virus vaccines is
measured is its efficacy in reducing influenza morbidity and mortality. Vaccina-
tion provides a reduction in both the frequency and severity of symptomatic
illness in vaccinees compared with matched controls. The protective efficacy of
inactivated virus vaccine against clinieal illness has been shown to be in the range
of 60-80% depending on the degree ofhomology of the strains in the community
with those in the vaccine.1 24 Although vaccination decreases the frequency of
clinical illness, a reduction in the incidence of infection is less frequently seen.
Vaccination can reduce virus spread in populations in whieh it is widely em-
ployed, however.1 63 Although influenza vaccine is immunogenic in the elderly,
protection against clinical illness appears to be less than the 60-80% reduction
reported in younger adults. Protection against the complications of influenza,
including death, however, is significantly associated with vaccine administration
and is much greater than protection against uncomplicated illness.l64.165
Administration of inactivated influenza virus vaccine has a remarkable safety
record. In adults, the incidence of systemic reactions in subjects given clinically
efficacious doses of vaccine is not greater than that in placebo recipients. 124 When
systemic reactions are seen, they are most likely to occur in seronegative vaccinees.
In children, whole-virus preparations cause more systemic reactions than do split-
virus vaccines.l 22 Reactions in children given two doses of split-product vaccine
with well-standardized antigenic mass are minimal,l23 Although there was a six-
fold increase in Guillian-Barre syndrome in the 1976 swine influenza immuniza-
tion program, this complication has not been associated with other vaccine
preparations.166.167 Although cases of meningoencephalitis following influenza
vaccination have been reported, adefinite causal relationship has not been
established. 168
The frequency of immediate hypersensitivity reactions to inactivated virus
vaccine is thought to be 113900,32 Allergy to egg protein contaminating the
vaccine preparation is the probable cause of such reactions and a history of egg
allergy has been considered a contraindication to vaccination. Patients with a
history of egg allergy and a strong indication for influenza vaccination may,
however, be successfully immunized after desensitization. 169
Influenza vaccination is recommended for individuals at increased risk for
influenza-related morbidity and mortality or those who may readily transmit
infection to such individuals. 26 At highest risk are children and adults with
chronic cardiopulmonary conditions, including asthma, and residents of nursing
hornes or other chronic-care facilities. Others at risk for increased influenza-
related morbidity and mortality are healthy patients over age 65, patients with
noncardiopulmonary chronic illness such as chronic renal failure, diabetes
mellitus, and hemoglobinopathies, and children and adolescents receiving
chronic aspirin therapy. Health care workers and household contacts of high-risk
patients also should be immunized routinely. The inactivated virus vaccine has
never been shown to be harmful to the fetus and vaccination is recommended for
pregnant women with antecedent chronic illnesses. Immunocompromised pa-
tients, including transplant recipients, those receiving cytotoxic chemotherapy
for cancer, those on chronic corticosteroids, or those with AIDS should also be
vaccinated, but the expected efficacy of immunization is lower than that of
immunocompetent individuals. Vaccine mayaiso be administered to any other
individuals not covered by the above guidelines who wish to avoid illness from
influenza. 26
Although clearly valuable in the prevention of influenza, inactivated, paren-
terally administered influenza vaccine has limitations. It has limited ability to
induce mucosal immunity, has relatively poor immunogenicity in immunologi-
cally naive children and adults, does not prevent infection in the majority of cases,
and provides less than complete protection against clinical disease expression. For
these reasons, there is a continuing research effort to develop live attenuated

influenza vaccines for clinical use. Two strategies to achieve attenuation have been
successful. Both cold-adapted mutant and avian-human recombinant viruses
have been developed. 170,l7t These vaccines are administered intranasally, Adverse
effects are minimal. Mild influenza-like symptoms are seen only when high doses
are used. H2 Genetic reversion to wild strains has not been demonstrated and there
is little evidence of transmission of the vaccine strain.l 12 ,172,173 Live, attenuated
vaccine is also immunogenic in immunologically naive individuals and induces
higher titers of nasal-wash antibody than does inactivated vaccine. H9 Experimen-
tal challenge with homologous, wild-type virus demonstrated protection against
both clinical illness and infection that is comparable to or better than that seen
with inactivated vaccine,l74-176 High levels of protection against naturally occur-
ring influenza has also been demonstrated.l 21 ,177 Although most studies of the
protective efficacy of live, attenuated-virus vaccine have been carried out in
seronegative subjects, protection in seropositive subjects has also been re-
ported,l78
Although live, attenuated vaccines are promising, practical obstacles to wide-
spread use remain. Reduced or transient immune responses have been seen in the
elderly when live, attenuated-virus vaccine is used alone but improved protection
has been reported when live and inactivated vaccines have been combined.l 79 ,180
The optimal strategy for use of live attenuated-virus vaccines relative to inacti-
vated vaccine remains to be defined. Methods to produce and distribute rapidly
large pools of stable, viable, attenuated viruses annually, reflecting the immuno-
logie composition of several appropriate epidemie strains, need to be developed.
Encouragingly, preliminary trials of a trivalent, live, attenuated vaccine recently
have been reported. 181
7.2. Antiviral Agents

Antiviral agents offer both an alternative approach in the prevention of
influenza and treatment of established infection. Amantadine and rimantadine
are tricyclic amines with antiviral activity specific for influenza A virus. Other
influenza virus types are not affected. Although the mechanism of action is not
fully understood, these compounds exert their effect after virus adsorption to the
target cell but before RNA transcription. They interact with the M2 matrix
protein and are thought to interfere with influenza-virus uncoating. 182 Mutations
in M2 confer amantadine and rimantidine resistance. 42 ,183
Amantadine and rimantadine are effective in the prevention of influenza A
virus infection. The efficacy of chemoprophylaxis is similar to that of inactivated
virus vaccine.l 84 A randomized, controlled trial of amantadine versus riman-
tadine demonstrated that the protective efficacy of these two agents is compa-
rable. 185
Amantadine and rimantadine are also active as therapy for acute influenza.
Both agents shorten the duration of symptoms when used as treatment for
uncomplicated influenza in otherwise healthy children and young adults, but the
period of symptoms usually is abbreviated by only 1-2 days.186-188 Viral shedding

is only transiently effected. Rimantadine has been reported to be effective in
treatment of uncomplicated influenza in the elderly,189 but little information is
available about use of these agents in severe or complicated influenza, or their
ability to reduce complications in high-risk patients.
The most frequent adverse effects of amantadine are related to the central
nervous system and gastrointestinal tract and include insomnia, dizziness, diffi-
culty concentrating, and nausea. Amantadine is thought to lower the seizure
threshold, and an increased frequency of seizures has been observed in patients
with therapeutic anticonvulsant levels. Rimantadine appears to have a signifi-
cantly lower incidence of adverse effects than amantadine,185,190
Influenza virus resistant to amantadine and rimantadine has been found to
emerge in approximately 30% of patients treated for acute influenza. 187 ,188,191
Resistant virus appears between 2 and 5 days after the start of treatment. In
uncomplicated influenza, clinical benefit from treatment has still been seen
despite emergence of resistance. Symptoms tend to be more prolonged in patients
shedding resistant virus, however,188 and it is possible resistance will result in loss
of therapeutic benefit or relapsing illness in high-risk or immunodeficient pa-
tients with more severe manifestations of influenza, in whom either the viral
burden is greater, the immune response weaker, or both. Resistant influenza
viruses retain full virulence in an animal model of infection. 192 Failure of
chemoprophylaxis as a result of transmission of resistant influenza virus from an
index case to household or institutional contacts has been documented and
clinical illness caused by resistant influenza is similar to that produced by sensitive
virus,191,193
Because of its cost and side-effect profile, amantadine should not replace
vaccination as the primary modality of influenza A virus prevention. Chemo-
prophylaxis is recommended for individuals who would otherwise be candidates
for vaccination but have a his tory of egg allergy and a reactive-vaccine skin test or
otherwise cannot be desensitized. Such patients should receive amantadine for
the duration of the influenza A outbreak in the community, usually 5 to 7 weeks.
During an established influenza outbreak, unvaccinated high-risk individuals
without a contraindication to vaccination should be given amantadine in conjunc-
tion with vaccination. Chemoprophylaxis can be discontinued after 2 weeks, at
which time protective antibody titers should have developed. Amantadine can
also be used to supplement the protective effect of vaccination in high-risk
individuals who are likely to have a suboptimal response to vaccination, for
example, immunodeficient patients. Finally, amantadine may be used during an
outbreak with an influenza strain not related to any contained in an available
vaccme.
Indications for antiviral therapy for active influenza are less clear than those
for chemoprophylaxis. Although treatment does shorten the average duration of
symptoms in uncomplicated influenza, routine treatment in low-risk cases is not
necessary because the vast majority of cases are self-limited and acceleration of
recovery associated with treatment is modest. Despite the lack of controlled trials,
it may be reasonable to consider amantadine treatment in patients at high risk of

influenza-related morbidity and mortality who have a characteristic illness du ring
a confirmed outbreak of influenza. Because of the problem with transmission of
resistant viruses, treatment oflow-risk household contacts ofhigh-risk individuals
receiving amantadine prophylaxis should be avoided. Similarly, in an institutional
setting such as a chronic care facility, concurrent treatment of an index case and
prophylaxis of other residents should be avoided unless the index case can be
isolated effectively.
Ribavirin is another anti viral agent with activity in vitro against influenza
viruses. Both influenza A and Bare susceptible. Limited data suggest that
aerosolized ribavirin can shorten the duration of symptoms and viral shedding in
uncomplicated influenza.I 94 ,195 Orally administered ribavirin has also been stud-
ied and found to reduce clinical symptomatology, although a significant antiviral
effect was not documented. 196 Intravenous ribavirin therapy of severe influenza
myocarditis has not been successful, despite an apparent antiviral effectJ97
REFERENCES
1. Tillett, HE., Smith,]. W. G., and Clifford, R. E., 1980, Excess morbidity and mortality
associated with influenza in England and Wales, Lancet 1:793-795.
2. Glezen, W. P., Payne, A. A., Snyder, 0. N., and Downs, T. 0.,1982, Mortality and influenza,
J Inftct. Dis. 146:313-321.
3. Alling, D. w., Blaek, W. C., and Stuart-Harris, C. H., 1981, A study of excess mortality
du ring influenza epidemics in the United States, 1968-1976, Am.] Epidemiol. 113:30-43.
4. Barker, W. H. and Mullooly,]. P., 1982, Pneumonia and influenza deaths during epidemics:
Implieations for prevention, JAMA 142:85-89.
5. Betts, R. F. and Douglas, R. G., Jr., 1990, Influenza virus, in: Principies and Practice 0/
InJectious Diseases (G. E. MandelI, R. G. Douglas, Jr., and]. E. Bennett, eds.), Churehill
Livingstone, New York, pp. 1306-1325.
6. Glezen, W. P. and Couch, R. B., 1978, Imerpandemie influenza in the Houston area, 1974-
76, N. Engl. J Med. 298:587-592.
7. Sugaya, N., Nerome, K., Ishida, M., Nerome, R., Nagae, M., Takeuchi, Y., and Osano, M.,
1992, Impact of influenza virus infection as a cause of pediatric hospitalization,J InJect. Dis.
165:373-375.
8. Monto, A. S. and Kioumehr, F., 1975, The tecumseh study of respiratory ilIness. IX.
Occurrence of influenza in the community, 1966-71, Am. J Epidemiol. 102:553-563.
9. Nigg, C., Eklund, C. M., Wilson, 0. E., and Crowley,]. H., 1942, Study of an epidemie of
influenza B, Am. J Hyg. 35:265-284.
10. Moriuehi, H., Katsushima, N., Nishimura, H., Nakamura, K., and Numazaki, Y., 1991,
Community-acquired influenza C virus infection in children, J Pediatr. 118:235-238.
11. Martin, C. M., Kunin, C. M., Gottiieb, L. S., and Finland, M., 1959, Asian influenza A in
Boston, 11. Severe staphylococcal pneumonia complieating influenza, Ann. Intern. Med.
103:532-542.
12. Schwarzmann, S. w., Adler, ]. L., Sullivan, R. ]., and Marine, W. M., 1971, Bacterial
pneumonia during the Hong Kong influenza epidemie of 1968-69, Arch. Intern. Med.
127:1037-1041.
13. Louria, D.B., Blumenfeid, H. L., Ellis,]. T., Kilbourne, E. D., and Rogers, D. E., 1959,
Studies of influenza in the pandemie of 1957-58. 11. Pulmonary complications of influ-

enza,} Glin. Invest. 38:213-265.
14. Young, L. S., La Force, F. M., Head, j. j., Feeley, j. C., and Bennett, j. v., 1972, A simulta-
neous outbreak of meningococcal and influenza infections, N. Eng!.} Med. 287:5-9.
15. Cartwright, K. A. v.,Jones, D. M., Smith, A. S., Stuart,j. M., Kaczmarski, E. B., and Palmer,
j. R., 1991, Influenza A and meningococcal disease, Lancet 338:554-557.
16. Center for Diseases Control, 1986, Toxie shock syndrome associated with influenza-
Minnesota, MMWR 35:143-144.
17. Sperber, S. j. and Francis,j. B., 1987, Toxie shock during an influenza outbreak,JAMA 257:
1086-1087.
18. Finland, M., Parker, F., Jr., Barnes, M. w., and Joliffe, L. S., 1945, Acute myocarditis in
influenza A infections: Two cases of nonbacterial myocarditis, with isolation of virus
from the lungs, Am.} Med. Sei. 207:455-468.
19. Hildebrandt, H. M., Maassab, H. F., and Willis, P. w., 1962, Influenza virus pericarditis,
Am.} Dis. Ghild. 104:179-182.
20. FarreH, M. K., Partin, j. C., Bove, K. E., Jacobs, R., and Hilton, P. K., 1980, Epidemie
influenza myopathy in Cincinnati, 1977,} Pediatr. 96:545-551.
21. Greco, T. P., Askenase, P. w., and Kashgarian, M., 1977, Postviral myositis: Myxovirus-like
structures in affected muscle, Ann. Intern. Med. 86:193-194.
22. Corey, L., Rubin, R. j., Hattwiek, M. A., Noble, G. R., and Cassidy, E., 1976, A nationwide
outbreak ofReye's syndrome: Its epidemiologie relationship to influenza B, Am.} Med. 61:
615-625.
23. Varma, R. R., Riedel, D. R., Komorouski, R. A., Harrington, G. j., and Nowak, T. v., 1979,
Reye's syndrome in nonpediatric age groups,JAMA 242:1373-1375.
24. Starks, K. M., Ray, G., and Dominguez, L. B., 1980, Reye's syndrome and salicylate use,
Pediatries 66:859-864.
25. Fulginiti, V. A. and the Committee on Infectious Diseases, 1982, Special report: Aspirin
and Reye's syndrome, Pediatries 69:810-812.
26. Centers for Disease Control, 1992, Prevention and control of influenza, MMWR 41(no.
RR-9):1-17.
27. WeHs, C. E. C., 1971, Neurologie complications of so-caHed influenza: A winter study in
Southeast Wales, Br. Med.} 1:369-373.
28. Baine W. B. and Luby,j. P., 1980, Severe illness with influenza B, Am.} Med. 68:181-189.
29. Lamb, R. A., Zebedee, S. L., Richardson, C. D., 1985, Influenza virus M2 protein is an
integral membrane protein expressed on the infected cell surface, GcIl40:627-633.
30. Schulze, I. T., 1973, Structure of the influenza virion, Adv. Virus Res. 18:1-55.
31. Wiley, D. C. and Skehel, j. j., 1987, The structure and function of the hemagglutinin
membrane glycoprotein of influenza virus, Annu. Rev. Bioehern. 56:365-394.
32. Murphy, B. R. and Webster, R. G., 1990, Orthomyxoviruses, in: ViTOlogy (B. N. Fields, ed.),
Raven Press, New York, pp. 1091-1152.
33. Murti, K. G. and Webster, R. G., 1986, Distribution of hemagglutinin and neuraminidase
on influenza virions as revealed by immunoelectron mieroscopy, Virology 149:36-43.
34. Suzuki, Y., Nagao, Y., Kato, H., Matsumoto, M., Nerume, K., Nakajima, K., and Nobusawi,
E., 1986, Human influenza A virus hemagglutinin distinguishes sialoligosaccharides in
membrane-associated gangliosides as its receptor whieh mediates adsorption and fusion
processes ofvirus infection,} Biol. Ghern. 261:17057-17061.
35. Colman, P. M. and Ward, C. w., 1985, Structure and diversity of influenza virus neuramini-
dase, Gurr. Top. Microbio!. Immunol. 11:177-255.
36. Webster, R. G., Brown, L. E., and Laver, W. G., 1984, Antigenic and biologieal characteriza-
tion of influenza virus neuraminidase (N2) with monoclonal antibodies, Virology 135:
30-42.
300 RICHARD V. SPERA, JR, and DAVID H. SHEPP
37. Webster, R. G., Reay, P. A., and Laver, W. G., 1988, Protection against lethai influenza with
neuraminidase, Virolog) 164:230-237.
38. Schulman, j. L., 1975, Immunology of influenza, in: The Influenw Viruses and Influenza,
(E. D. Kilbourne, ed.), Academic Press, Orlando, pp. 373-393.
39. Van Wyke, K. L., Bean W. j., Jr., and Webster, R G. 1981, Monoclonal antibodies to
influenza A virus nucleoprotein affecting RNA transcription,J Virol. 39:313-317.
40. Ye, Z., Pal, R., Fox,]. w., and Wagner, R. R, 1987, Functional and antigenic domains ofthe
matrix (M) protein of influenza virus, J Virol. 61:239-246.
41. Zebedee, S. L. and Lamb, R. A., 1988, Influenza A virus M2 protein. Monoclonal antibody
restriction of virus growth and detection of M2 in virions, J Virol. 62:2762-2772.
42. Hay, A. j., Zambon, M. C., Wolstenholme, A. j., SkeheJ, j. j., and Smith, M. H., 1986, The
molecular basis of resistance to influenza A virus to amantadine, J Antimierob. Chemother.
18(Suppl B):19-29.
43. Palese, P., Ritchey, M. B., and Schulman, j. L., 1977, PBl and PB3 proteins of influenza
virus are required for complementary RNA synthesis,J Virol. 21:1187-1195.
44. Murphy, B. R, Chalhub, E. G., Nusinoff, S.R., Kasel, j., and Channock, R. M., 1973,
Temperature-sensitive mutants of influenza virus. III. Further characterization of the
ts-[E] influenza recombinant (H3N2) virus in man, J Infect. Dis. 128:479-487.
45. Hall, C. B. and Douglas, R. G., Jr., 1975, Nosocomial influenza infection as a cause of
intercurrent fevers in infants, Pediatries 55:673-677.
46. Walsh, j. j., Dietlein, L. F., Low, F. N., Burch, G. E., and Mogabgab, w.]., 1961, Broncho-
tracheal response in human influenza, Areh. Intern. Med. 108:98-109.
47. Oseasohn, R., AdeJson, L., and Kaji, M., 1959, Clinicopathologic study ofthirty-three fatal
cases of Asian influenza, N. Engl. J Med. 260:509-518.
48. Mulders, j. and Hers, j. F., 1979, Influenw, Walter-Noordhoff, Groniger, Netherlands.
49. Webster, R. G., Hinshaw, V. S., Berton, M. T., Laver, W. G., and Air, G., 1981, Antigenic drift
in influenza viruses and association of biological activity with the topography of the
hemagglutinin molecule, in: Genetie Variation Among Influenw Viruses. ICN-UCLA sympo-
sia on molecular and cellular biology, Vol XXI (Nayak, D. P., Fox, C. F., eds.), Academic
Press, Orlando, pp. 309-322.
50. Velazzo,j. G., Couch, R. B., Six, H. R., and Glezen, W. P., 1983, Surveillance ofinfluenza in
Houston, Texas, USA: Gradual transition from AlVictorial75(H 3 N 2 ) to AlTexasI77(H 3 N2 ),
predominance and characterization of intermediate strains, Bull. WHO 61:345-352.
51. Both, G. w., Sleigh, M.J., Cox, N. J., and Kendal, A. P., 1983, Antigenic drift in influenza
virus H3 hemagglutinin from 1968 to 1980: Multiple evolutionary pathways and sequential
amino acid changes of key antigenic sites, J Virol. 48:52-60.
52. Webster, R. G., Laver, W. G., Air, G. M., Ward, C., Gerhard, w., and Van Wyke, K. L., 1980,
The mechanism of antigenic drift in influenza viruses: Analysis of Hong Kong (H 3N 2)
variants with monoclonal antibodies to the hemagglutinin molecule, Ann. N.Y. Aead. Sei.
254:142-16l.
53. Webster, R. G., Campbell, C. H., and Granoff, A., 1972, The in vivo production of new
influenza A viruses. I. Genetic recombination between avian and mammalian viruses,
Virolog) 44:317~328.
54. Webster, R. G., Campbell, C. H., and Granoff, A., 1973, The in vivo production of new
influenza viruses. III. Isolation of recombinant influenza viruses under simulated condi-
tions of natural transmission, Virolog) 51:149-)62.
55. Kida, H., Kawaoka, Y., Naeve, C. w., and Webster, R. G., 1987, Antigenic and genetic
conservation of H3 influenza in wild ducks, Virolog) 159:109-119.
56. Nakajima, K., Desselberger, u., and Palese, P., 1978, Recent human influenza A viruses are
closeJy reJated genetically to strains isolated in 1950, Nature 274:334-339.
57. Shortridge, K. F., Webster, R G., Butterfield, W K., and Campbell, C. H., 1977, Persistence
of Hong Kong influenza virus variants in pigs, Science 196:1454-1455.
58. Dowdle, W R. and Hattwick, M. A., 1977, Swine influenza infections in humans,] Infoct.
Dis. 46:S386-389
59. Sanford, B. A. and Ramsay, M. A., 1987, Bacterial adherence to the upper respiratory tract
of ferrets infected with influenza A virus, Proc. Soc. Exp. Biol. Med. 185:120-128.
60. Fainstein, v., Musher, D. M., and Cate, T. R, 1980, Bacterial adherence to pharyngeal cells
during viral infection,] Infect. Dis. 141:174-176.
6l. Smith, C. B., Colden, C., Klauber, M. R., Kanner, R., and Renzetti, A., 1976, Interactions
between viruses and bacteria in patients with chronic bronchitis,] Infoct. Dis. 134:552-561.
62. Camner, P.,Jarstrand, c., and Philipson, K., 1973, Tracheo-bronchial c1earance in patients
with influenza, Am. Rev. Respir. Dis. 108:131-135.
63. Westerberg, S. C., Smith, C. B., and Wiley, B. B., 1972, Pathogenesis of influenza virus
infection in mouse tracheal organ cultures, Proc. Soc. Exp. Biol. Med. 140:846-850.
64. Wilson, R, Alton, E., and Rutman, A., 1987, Upper respiratory tract viral infection and
mucociliary c1earance, Eur.] Respir. Dis. 70:272-280.
65. Cassidy, L. F., Lyles, D. S., and Abramson, J. S., 1988, Synthesis of viral proteins in
polymorphonuclear leukocytes infected with influenza A virus,] Clin. Microbiol. 26:1267-
1270.
66. Abramson, J. S. and MiIIs, E. L., 1988, Depression of neutrophil function induced by
viruses and its role in secondary microbial infections, Rev. Infect. Dis. 10:326-341.
67. Abramson, J. S., Parce, J. W, Lewis, J. C., Lyles, D. S., MiIIs, E. L., Nelson, R D., and Bass,
D. A., 1984, Characterization of the effect of influenza virus on polymorphonuclear
leukocyte membrane responses, Blood 64:131-138.
68. Abramson, J. S., Wiegand, G. L., and Lyles, D. S., 1982, Neuraminidase activity is not the
cause of influenza virus-induced neutrophil dysfunction,] Clin. Microbiol. 22:129-131.
69. Larson, H. E., Parry, R P., GiIchrist, C., Luquetti, A., and Tyrrell, D. A. J., 1977, Influenza
viruses and staphylococci in vitro: Some interactions with polymorphonuclear leucocytes
and epithelial cells, Br.] Exp. Patlwl. 58:281-288.
70. Ruutu, P., Vaheri, A., and Kosunen, T. U, 1977, Depression of human neutrophil motility
by influenza virus in vitro, Scand. ] Immunol. 6:897-906.
71. Abramson,j. S., MiIIs, E. L., Giebink, G. S., and Quie, P. G., 1982, Depression ofmonocyte
and polymorphonuclear leukocyte oxidative metabolism and bactericidal activity by influ-
enza A virus, Infect. Immun. 35:350--355.
72. Shult, A., Dick, E. C.,Joiner, K. A., and Busse, W W, 1985, Role of serum in stimulation of
polymorphonuclear leukocyte luminol-dependant chemiluminescence by influenza A,
73. Abramson, J. S., Lewis, J. C., Lyles, D. S., Heller, K. A., MiIIs, E. L., and Bass, D. A., 1982,
Inhibition of neutrophillysosome-phagosome fusion associated with influenza virus infec-
tion in vitro: Role in depressed bactericidal activity,] Clin.Invest. 69:1393-1397.
74. Larson, H. E. and Blades, R, 1976, Impairment of human polymorphonuclear leukocyte
function by influenza virus, Lancet 1:283.
75. Abramson, J. S., Giebink, G. S., Mills, E. L., and Quie, P. G., 1981, Polymorphonuclear
leukocyte dysfunction during influenza virus infection in chinchillas,] Infect. Dis. 143:
836-845.
76. Abramson, J. S., Giebink G. S., and Quie, P. G., 1982, Influenza A virus-induced poly-
morphonuclear leukocyte dysfunction in the pathogenesis of experimental otitis media,
77. Ruutu, P., 1977, Depression of rat neutrophil exudation and motility by influenza virus,
Scand. ] Immunol. 6:1113-1120.
78. Larson, H. E., Parry, R. P., and Tyrrell, D. A. ]., 1980, Impaired polymorphonucIear
leukocyte chemotaxis after influenza virus infection, Br.] Dis. ehest 74:56-62.
79. Martin, R. R., Couch, R. B., Greenberg, S. B., Cate, T. R., and Warr, G. A., 1981, Effects of
infection with influenza virus with leukocytes and its effects on phagocytosis,] Infect. Dis.
144:279-286.
80. Sawyer, W. D., 1969, Interaction of influenza virus with leukocytes and its effect on
phagocytosis,] Infeet. Dis. 119:541-556.
81. Abramson,]. S., Lyles, D. S., Heller, K. A., and Bass, D. A., 1982, Influenza A virus-induced
polymorphonucIear leukocyte dysfunction, Infeet. Immun. 37:794-799.
82. Roberts, N.]. and Horan, P. K., 1985, Expression of viral antigens after infection of human
Iymphocytes, monocytes and macrophages with influenza virus,] Infeet. Dis. 151:308-313.
83. Brownson,]. M., Mahy, B. w., and Hazelman, B. L., 1979, Interaction ofinfluenza A virus
with peripheral blood Iymphocytes, Infeet. Immun. 25:749-756.
84. Nain, M., Hinder, F., Gong,]. H., Schmidt, A., Bender, A., Sprenger, H., and Gemsa, D.,
1990, Tumor necrosis factor alpha production of influenza A virus-infected macrophages
and potentiating effect of Iipopolysaccharide,] Immunol. 145:1921-1928.
85. Roberts, R.]., 1982, Different effects of influenza virus, respiratory syncytial virus and
sendai virus on human Iymphocytes and macrophages, Infect. Immun. 35:1142-1146.
86. Ettensohn, D. B. and Roberts, N.]., Jr., 1984, Influenza virus infection of human alveolar
and blood derived macrophages: Differences in accessory cell function and interferon
production,] Infeet. Dis. 149:942-949.
87. Nickerson, C. L., and Jakab, G.]. 1990, Pulmonary antibacterial defenses during mild and
severe influenza virus infection, Infect. Immun. 58:2809-2814.
88. Nugent, K. M. and Pesanti, E. L., 1982, Staphylococcal cIearance and pulmonary macro-
phage function during influenza infection, Infeet. Immun. 38:1256-1267.
89. Roberts, N.]., Douglas, R. G., Simons, R. M., and Diamond, M. E., 1979, Virus-induced
interferon production by human macrophages,] Immunol. 123:365-369.
90. Ennis, F. A. and Meager, A., 1981, Immune interferon produced to high levels by antigenic
stimulation of human Iymphocytes with influenza virus,] Exp. Med. 154:1279-1289.
91. Ennis, F. A., Beare, A. S., Riley, D., Schild, G. L., Meager, A., Qi, Y-H., Schwartz, G., and
Rook, A. H., 1981, Interferon induction and increased natural killer cell activity in
influenza infections in man, Laneet 2:891-893.
92. Green,]. A., Charette, R. P., Yeh, T-]., and Smith, C. B., 1982, Presence of interferon in
acute and convalescent phase sera of humans with influenza or an influenza-like iIlness of
undetermined etiology,] Infeet. Dis. 145:837-841.
93. Reed, W. P., Old,]. w., and Kisch, A. L., 1972, Decreased skin test reactivity associated with
influenza,] Infect. Dis. 125:398-402.
94. Casali, P., Rice, P. A., and Oldstone, M. A., 1984, Viruses disrupt functions of human
Iymphocytes: Effects of measles virus and influenza virus on Iymphocyte-mediated killing
and antibody production,] Exp. Med. 159:1322-1337.
95. Lewis, D. E., Gilbert, B. E., and Knight, v., 1986, Influenza virus infection induces
functional alterations in peripheral blood Iymphocytes,] Immunol. 137:3777-3781.
96. Allan, w., Tabi, Z., Cleary, A., and Doherty, P. C., 1990, Cellular events in the Iymph node
and lung of mice with influenza: Consequences of depletion CD4 + T cells,] Immunol.
144:3980-3986.
97. Tashiro, M., Cibrowski, P., Reinacher, M., Klenk, H. D., Pulverer, G., and Rott, R., 1987,
Synergistic role of staphylococcal proteases in the induction of influenza virus patho-
genicity, Virology 157:421-430.
98. Scheiblauer, H., Reinacher, M., Tashiro, M., and Rott, R., 1992, Interactions between
bacteria and influenza A virus in the development of influenza pneumonia,] Infect. Dis.
166:783-791.
99. Little, J. w., Hall, W. J., Douglas, R. G., Jr., Mudholkar, G. S., Speers,D. M. and PateI, G.,
1987, Airway hyperreactivity and peripheral airway dysfunction in influenza A infection,
100. Horner, G. J. and Gray, F. 0., Jr., 1973, Effect of uncomplicated influenza on the diffusing
capacity of the lung, Am. Rev. Respir. Dis. 108:866-869.
101. Hall, w.J., Douglas, R. G.,Jr., Hyde, R. w., Roth, F. K., Cross, A. S., and Speers, 0. M., 1976,
Pulmonary mechanics following uncomplicated influenza A infection, Am. Rev. Respir. Dis.
113:141-147.
102. Clementsen, P.,Jensen, C. B., Hannoun, C., Soborg, M., and Norn, S., 1988, Influenza A
virus potentiates basophil histamine release caused by endotoxin induced complement
activation, Allergy 43:93-99.
103. Chonmaitree, T., Lett-Brown, M. A., and Grant, J. A., 1991, Respiratory viruses induce
production of histamine releasing factor by mononucIear leukocytes a possible role in the
mechanism of virus-induced asthma.] Infoct. Dis. 164:592-594.
104. Howard,J. B., McCracken, G. H.,Jr., and Luby,J. P., 1972, Influenza A virus as a cause of
croup requiring tracheostomy,] Pediatr. 81:1148-1150.
105. Lin, C. Y., Kuo, Y. C., Liu, W. T., and Lin, C. C., 1988, Immunomodulation of influenza
virus infection in the precipitating asthma attack, Chest 93:1234-1238.
106. Stark, J. E., Heath, R. B., and Curwen, M. P., 1965, Infection with influenza and para-
influenza viruses in chronic bronchitis, Tlwrax 20:124-127.
107. Lamb, J. R., Mc Michael, A. J., and Rothbard, J. B., 1987, T-cell recognition of influenza
viral antigens, Human. Immunol. 19:79-89.
108. Murphy, B. R. and Clements, M. L., 1989, The systemic and mucosal immune responses of
humans to influenza A virus, Curr. Top. Mierobiol. Immunol. 146:107-116.
109. Potter, C. W. and Oxford, J. S., 1979, Determinants of immunity to influenza infection in
man, Br. Med. Bull. 53:69-80.
110. Puck, J. M., Glezen, W. P., Frank, A. L., and Six, H. R., 1980, Protection of infants from
infection with influenza A virus by transplacentally acquired antibody, ] Infoct. Dis.
142:844-849.
111. Smith, S. L., Murphy, B. R., Moss, B., 1983, Construction and characterization of an
infectious vaccinia virus recombinant that expresses the influenza hemagglutinin gene and
induces resistance to influenza virus infection in hamsters, Proc. Nati. Aead. Sei. 80:7155-
7159.
112. Clements, M. L., O'Donnell, S., Levin, M. M., Chanock, R. M., Murphy, B. R., 1983, Dose
response of AlAlaska/6177(H3N2) cold-adapted reassortant vaccine virus in adult volun-
teers: Role of local antibody in resistance to infection with vaccine virus, Infoet. Immun. 40:
1044-1O5l.
113. Johnson, P. R., Feldman, S., Thompson, J. M., Mahoney, J. 0., and Wright, P. F., 1986,
Immunity to influenza A virus infection in young children: A comparison of natural
infection, live cold adapted vaccine and inactive vaccine,] Infoet. Dis. 154:121-127.
114. Murphy, B. R., Kasel, J. A., and Chanock, R. M., 1972, Association of serum anti-neur-
aminidase antibody with resistance to influenza in man, N. Engl.] Med. 286:1329-1332.
115. Renegar, K. B. and SmalI, P. A., Jr., 1991, Passive transfer of local immunity to influenza
virus infection by IgA antibody,] Immunol. 146:1972-1978.
116. Murphy, B. R., Nelson, 0. L., Wright, P. F., Tierney, E. L., Phelan, M. A., and Chanock, R.
M., 1982, Secretory and systemic immunological responses in children infected with live
attenuated influenza A virus vaccines, Infoct. Immun. 36:1102-1108.
117. Wagner, 0. K., Clements, M. L., Reiner, C. B., Snyder, M., Nelson, D. L., and Murphy, B. R.,
1987, Analysis of immunoglobulin antibody responses after administration of live and
inactivated influenza A vaccine indicated that nasal wash immunoglobulin G is a trans-
udate from serum,] Clin. Microbiol. 25:559-562.
304 RICHARD V. SPERA, jR., and DAVID H. SHEPP
118. Burlington, D. B., Clements, M. L., Meiklejohn, G., Phelan, M., and Murphy, B. R., 1983,
Hemagglutinin specific antibody responses in immunoglobulin A, G and M isotypes as
measured by enzyme-linked immunosorbent assay after primary or experimental second-
ary infection with influenza A virus in humans, Infect. Immun. 41:540-545.
119. Clements, M., L. and Murphy, B. R., 1986, Development and persistence of local and
systemic antibody responses in adults given live attenuated or inactivated influenza A
vaccine,] Clin. Mierobiol. 23:66-72.
120. Wright, P. F., Murphy, B. R., Keruna, M., Lawrence, L. M., Phelan, M. A., and Karzon, D. T.,
1983, Secretory immunological response after intranasal inactivated influenza A virus
vaccinations: Evidence for immunoglobulin a memory, Infect. Immun. 40:1092-1099.
121. Feldman, S., Wright, P. F., Webster, R. G., Roberson, P. K., Mahoney, j., Thompson, j.,
Doolittle, M., Lott, L., johnson, P., and Christoph, R. C., 1985, Use of influenza A virus
vaccines in seronegative children: Live cold-adapted versus inactivated whole virus,] Inftct.
Dis. 153:1212-1218.
122. Wright, P. F., Thompson, j., Vaughn, W K., FolIand, D. S., Seil, S. H., and Karzon, D. T.,
1977, Trials of influenza NNew jersey176 virus vaccine in normal children: An overview of
age-related antigenicity and reactogenicity,] Inftct. Dis. 136:731-741.
123. Wright, P. F., Cherry,j. D., Foy, H. M., Glezen, W P., Hall, C. B., McIntosh, K., Monto, A. S.,
Parrott, R. H., Portnoy, B., and Taber, L. H., 1983, Antigenicity and reactogenicity of
influenza NUSSRl77 virus vaccine in children-A multicentered evaluation of dosage and
safety, Rev. Infeet. Dis. 5:758-764.
124. LaMontagne,j. R., Noble, G. R., Quinnan, G. v., Curlin, G. T., Blackwelder, W C., Smith,
j. 1., Ennis, F. A., and Bozeman, F. M., 1983, Summary of clinical trials of inactivated
influenza vaccine, Rev. Infect. Dis. 5:723-736.
125. Cate, T. R., Couch, R. B., Parker, D., and Baxter, B., 1983, Reactogenicity, immunogenicity,
and antibody persistence in adults given inactivated influenza vaccines-1978, Rev. Inftet.
Dis. 5:737-747.
126. Cate, T. R., Couch, R. B., Kasel, j. A., and Six, H. R., 1977, Clinical trials of monovalent
influenza NNew jersey176 virus vaccine in adults: reactogenicity, antibody response and
antibody persistence,] Infeet. Dis. 136:S450-S455.
127. Clements, M. L., Betts, R. F., Tierney, E. L., and Murphy, B. R., 1986, Serum and nasal wash
antibodies associated with resistance to experimental challenge with influenza A wild-type
virus,] Clin. Mierobiol. 24:157-160.
128. McMichael, A. j., Parham, P., Bordsky, F. M., and Pilch, j. R., 1980, Influenza virus specific
cytotoxic T-lymphocytes recognize HLA molecules; blocking by monoclonal anti-HLA
antibodies,] Exp. Med. 152:195S-203S.
129. Kaplan, D. R., Griffith, R., Braciale, V. L., and Braciale, T. j., 1984, Influenza virus-specific
human cytotoxic T cell clones: Heterogeneity in antigen specificity and restriction by class
II MHC products, Cello Immunol. 88:193-206.
130. Biddison, W E., Shaw, S., and Nelson, D. L., 1979, Virus specificity of human influenza
virus immune cytotoxic T-cells,] Immunol. 122:660-668.
131. McMichael, A. j., Fotch, F. M., Nobel, G. R., and Beare, P. A. S., 1983, Cytotoxic T-cell
immunity to influenza, N. Engi.] Med. 309:13-17.
132. McMichael, A.j., Michie, C. A., Gotch, F. M., Smith, G. L., and Moss, B., 1986, Recognition
of influenza A virus nucleoprotein by human cytotoxic T lymphocytes,] Gen. Virol.
67:719-726.
133. Townsend, A. R. M., Rothbard,j., Gotch, F. M., Bahadur, G., Wraith, D., and McMichael, A.
j., 1986, The epitopes of influenza nucleoprotein recognized by cytotoxic T Iymphocytes
can be defined with short synthetic peptides, Cell 44:959-965.
134. Ennis, R. A., Martin, W j., and Verbonitz, M. W, 1977, Haemagglutinin specific cytotoxic
T cell responses during influenza infection,] Exp. Med. 146:843-898.
135. Bennink, j. R, Yewdell, j. w., Smith, G. L., Moller, C., and Moss, B., 1984, Recombinant
vaccinia virus primes and stimulates influenza haemagglutinin-specific cytotoxic T-cells,
Nature 311:578-579.
136. Bennink, j. R., Yewdell, j. w., and Gerhard, w., 1982, A viral polymerase gene involved in
recognition of influenza-infected cells by a cytotoxic T cell clone, Nature 296:75-76.
137. Gotch, F., McMichael, A., Smith, G., and Moss, B., 1987, Identification ofviral molecules
recognized by influenza-specific human cytotoxic T lymphocytes,] Exp. Med. 165:408-416.
138. Wells, M. A., Ennis, F. A., and Albrecht, P., 1981, Recovery from a viral respiratory infection.
II. Passive transfer of immune spleen cells to mice with influenza pneumonia,] Immunol.
126:1042-1046.
139. Taylor, P. M. and Askonas, B. A., 1986, Influenza nucleoprotein-specific cytotoxic T-cell
clones are protective in vivo, Immunology 58:417-420.
140. Kuwano, K., Scott, M., Young,j. F., and Ennis, F. A., 1988, HA2 subunit ofinfluenza A Hl
and H2 subtype viruses induces a protective cross-reactive cytotoxic T Iymphocyte re-
sponse,] Immunol. 140:1264-1268.
141. Eichelberger, M., Allan, w., Zijlstra, M.,Jaenisch, R, and Doherty, P. C., 1991, Clearance of
influenza virus respiratory infection in mice lacking class I major histocompatibility
complex-restricted CD8+ T cells,] Expl. Med. 174:875-880.
142. Bender, B. S., Croghan, T., Zhang, L., and Small, P. A., 1992, Transgenic mice lacking class
I major histocompatibility complex-restricted T cells have delayed viral clearance and
increased mortality after influenza virus challenge,] Exp. Med. 175:1143-1145.
143. Scherle, P. A., Palladino, G., and Gerhard, w., 1992, Mice can recover from pulmonary
influenza virus infection in the absence of class I-restricted cytotoxic T cells,] Immunol.
148:212-217.
144. Ennis, F. A., Qi, Y-H., Schild, G. C., Pratt, R, and Poiter, C. w., 1981, HLA-restricted virus-
specific cytotoxic T-Iymphocyte response to live and inactivated influenza vaccines, Lancet
2:887-891.
145. Gorse, G. j. and Belshe, RB., 1990, Enhancement of anti-influenza A virus cytotoxicity
following influenza A virus vaccination in older, chronically iII adults,] GUn. Microbiol.
28:2539-2550.
146. Greenberg, S. B., Criswell, B. S., and Couch, RB., 1975, Lymphocyte mediated cytotoxicity
against influenza virus-infected cells: An in vitro method,] Immunol. 115:601-603.
147. Greenberg, S. B., Criswell, B. S., Six, H. R., and Couch, RB., 1977, Lymphocyte
cytotoxicity to influenza virus infected cells. 11. Requirement for antibody and non-T
Iymphocytes,] Immunol. 119:2100-2106.
148. Justewicz, D. M., Lecomte,j., and Mandeville, R, 1984, Blockingofinfluenza virus-infected
cell-mediated cytotoxicity by hemagglutinin-specific monoclonal antibody,] Infoct. Dis.
150:348-357.
149. Verbonitz, M. w., Ennis, F. A., Hicks, j. T., and Albrecht, P., 1978, Hemagglutinin-specific
complement-dependent cytolytic antibody response to influenza infection,] Exp. Med.
147:265-270.
150. Quinnan, G. v., Ennis, F.A., Tuazon, C. u., Wells, M. A., Butchko, G. M., Armstrong, R.,
McClaren, C., Manischewitz, j. F., and Kiley, S., 1980, Cytotoxic Iymphocytes and antibody-
dependent complement-mediated cytotoxicity induced by administration of influenza
vaccine, Infoct. Immun. 30:362-369.
151. Habershon, RB., Molyneux, M. E., Slavin, G., Loewi, B., and Tyrell, D. A.j., 1973, Skin test
with influenza virus,] Hyg. 71:755-761.
152. Suzuki, F., Ohya, j., and Ishida, N., 1974, Effect of antilymphocyte serum on influenza
virus infection in mice, Proc. Soc. Exp. Biol. Med. 146:78-84.
153. Sullivan,j. L., Mayner, R. E., Barry, D. w., and Ennis, F. A., 1976, Influenza virus infection
in nude mice,] Infoct. Dis. 133:91-94.
154. Oda, T., Akaike, T., Hammamoto, T., Sukuki, F., Hirano, T., and Maeda, H.,1989, Oxygen
radicals in influenza-induced pathogenesis and treatment with pyran polymer-conjugated
SOD, Science 244:974-976.
155. Brandriss, M. w., Betts, R. F., Mathur, u., and Douglas, R. G.,Jr., 1981, Responses of elderly
subjects to monovalent AlUSSR/77(HINI) and trivalent AlUSSR/77(HINI)-AlTexasI77
(H3N2)-B/HongKongI72 vaccines, Am. Rev. Respir. Dis. 124:681-684.
156. Gross, P. A., Weksler, M. E,. Quinnan, G. v., Douglas, R. G.Jr., Gaerlan, P. F., and Denning,
C. R., 1987, Immunization of elderly people with two doses of influenza vaccine,] Clin.
Microbiol. 25:1763-1765.
157. Groothuis, j. R., Levin, M. j., Rabalais, G. P., Meiklejohn, G., and Lauer, B. A., 1991,
Immunization of high-risk infants younger than 18 months of age with split-product
influenza vaccine, Pediatrics 87:823-828.
158. Gross, P. A., Gould, L., and Brown, A. E., 1985, Effect of cancer chemotherapy on the
immune response to influenza virus vaccine: Review of the published studies, Rev. Infoct.
Dis. 7:613-618.
159. Beyer, W. E. P., Versluis, 0. j., Kramer, P., Diedrich, P. M. N., Weimer, w., and Masurel, v.,
1987, Trivalent influenza vaccine in patients on hemodialysis: Impaired seroresponse with
differences for A-H3N2 and A-HINI vaccine components, Vaccine 5:43-48.
160. Versluis, D.j., Beyer, W. E. P., Masurel, v., Wenting, G.j., and Weimar, w., 1986, Impairment
of the immune response to influenza vaccination in renal transplant recipients by cyclo-
sporine, but not azathioprine, Transplantation 42:376-379.
161. Williams, G.w., Steinberg, A. 0., Reinertsen,j. L., Klassen, L. w., Decker,j. L., and Dolin,
R., 1978, Influenza immunization in systemic lupus erythematosus. A double-blind trial,
162. Nelson, K. E., Clements, M. L., Miotti, P., Cohn, S., and Polk, B. F., 1988, The influence of
human immunodeficiency virus (HIV) infection on antibody responses to influenza
vaccines, Ann. Intern. Med. 109:383-388.
163. Warburton, M. F.,Jacobs, 0. S., Langsford, W. A., and White, G. E., 1972, Herd immunity
following subunit influenza vaccine administration, Med.] Aust. 2:67-70.
164. Patriarca, P. A., Weber,j. A., Parker, R. A., Hall, W. N., Kendal, A. P., Bregman, D.j., and
Schonberger, L. B., 1985, Efficacy of influenza vaccine in nursing hornes. Reduction in
illness and complications during an influenza A (H3N2) epidemic,JAMA 253:1136-1139.
165. Gross, P. A., Quinnan, G. v.,Jr., Rodstein, M., LaMontagne,j. R., Kaslow, R. A., Saah, A.j.,
Wallenstein, S., Neufeld, R., Denning, C., and Gaerlan, P., 1988, Association of influenza
immunization with reduction in mortality in and elderly population: A prospective study,
Arch. Intern. Med. 148:562-565.
166. Hurwitz, E. S., Schonberger, L. B., Nelson, 0. B., and Holman, R. C., 1981, Guillain-Barre
syndrome and the 1978-1979 influenza vaccine, N. Engl.] Med. 304:1557-1561.
167. Kaplan, j. E., Katara, P., Hurwitz, E. S., and Schonberger, L. B., 1982, Guillain-Barre
syndrome in the United States, 1979-1980 and 1980-1981: Lack of an association with
influenza virus vaccination, JAMA 248:698-700.
168. Guerrero, I. C. and Retailleau, H. F., 1976, Increased meningoencephalitis after influenza
vaccine, N. Engl.] Med. 300:565.
169. Murphy, K. R. and Strunk, R. C., 1985, Safe administration of influenza vaccine in
asthmatic children hypersensitive to egg proteins,] Pediatr. 106:931-933.
170. Maasaab, H. F., LaMontagne, j. R., and DeBorde, D. C., 1988, Live influenza virus vaccines,
in: Vaccines (S. A. Plotkin and E. A. Mortimer, eds.). Saunders, Philadelphia, pp. 435-457.
171. Murphy, B. R., Sly, 0. L., Tierney, E. L., Hosier, N. T., Massicot j. G., London, W. T.,
Chanock, R. M., Webster, R. G., and Hinshaw, V. S., 1982, Reassortant virus derived from
avian and human influenza viruses is attenuated and immunogenic in monkeys, Science
218:1330-1332.
172. Wright, P. F. and Karzon, D. T., 1987, Live attenuated influenza vaccine, Prog. Med. Virol.
34:70-88.
173. Belshe, R. B. and Van Voris, L. P., 1984, Cold-recombinant influenza AlCalifornia:10178
(HINl) virus vaccine (CR-37) in seronegative children: Infectivity and efficacy against
investigational challenge, ] Infect. Dis. 149:735-740.
174. Clements, M. L., Betts, R. F., and Murphy, B. R., 1984, Advantage of live attenuated cold-
adapted influenza A virus over inactivated vaccine for AlWashington/80 (H3N2) wild-type
virus infection, Lancet 1:705-708.
175. Clements, M. L., Betts, R F., Tierney, E. L., and Murphy, B. R., 1986, Resistance of adults to
challenge with influenza A wild-type virus after receiving live or inactivated virus vaccine,]
Glin. Microbiol. 23:73-76.
176. Sears, S. D., Clements, M. L., Betts, R. F., Maassab, H. F., Murphy, B. R., and Snyder, M. H.,
1988, Comparison of live, attenuated HINI and H3N2 cold-adapted and avian-human
influenza A reassortant viruses and inactivated virus vaccine in adults,] Infect. Dis. 158:
1209-1218.
177. Belshe, RB., Van Voris, L. P., Bartram, j., and Crookshanks, F. K., 1984, Live attenuated
influenza A virus vaccines in children: Results of a field trial,] Infect. Dis. 150:834-840.
178. Treanor, j. j., Roth, F. K., and Betts, R. F., 1990, Use oflive cold-adapted influenza A HINI
and H3N2 virus vaccines in seropositive adults,] Glin. Microbiol. 28:596-599.
179. Powers, D. C., Sears, S. D., Murphy, B. R., Thumar, B., and Clements, M. L., 1989, Systemic
and local antibody responses in e1derly subjects given live or inactivated influenza A virus
vaccines,] Glin. Microbiol. 27:2666-2671.
180. Treanor,j.j., Mattison, H. R., Dumyati, G., Yinnon, A., Erb, S., O'Brien, D., Dolin, R., and
Betts, R F., 1992, Protective efficacy of combined live intranasal and inactivated influenza
A virus vaccines in the e1derly, Ann. Intern. Med. 177:625-633.
181. Belshe, RB., Swierkosz, E. M., Anderson, E. L., Newman, F. K., Nugent, S. L., and
Maassab, H. F., 1992, Immunization of infants and young children with live attenuated tri-
valent cold-recombinant influenza A HINl, H3N2 and B vaccine,] Infect. Dis. 165:727-732.
182. Richman, D. D., Hostetler, K. Y., Yazki, P. j., and Clark, S., 1986, Fate of influenza A virion
proteins after entry into subcellular fractions of LLC cells and the effect of amantadine,
Virology 151:200-210.
183. Belshe, R. B., Smith, M. H., Hall, C. B., Betts, R., and Hay, A. j., 1988, Genetic basis of
resistance to rimatindine emerging during treatment of influenza virus infection,] Virol.
62:1508-1517.
184. Atkinson, W. L., Arden, N. H., Patriarca, P. A., Leslie, N., Lui, K-j., and Gohd, R., 1987,
Amantadine prophylaxis during an institutional outbreak of type A (HINl) influenza,
Arch. Intern. Med. 146:1751-1756.
185. Dolin, R., Reichman, R L., Madore, H. P., Maynard, R, Linton, P. N., and Webber-jones,j.,
1982, A controlled trial of amantadine and rimantidine in the prophylaxis of influenza A
infection, N. Engi.] Med. 307:580-584.
186. Younkin, S. w., Beils, R F., Roth, F. K., and Douglas, R G.,jr., 1983, Reduction in fever and
symptoms in young adults with influenza AlBraziIl78 (HINl) infection after treatment with
aspirin or amantadine, Antimicrob. Agents Ghemother. 23:577-582.
187. Hall, C. B., Dolin, R, Gala, C. L., Markovitz, D. M., Zhang, Y. Q., Madore, P. H., Disney,
RA., Talpey, W. B., Green,j. L., Francis, A. B., and Pichichero, M. E., 1987, Children with
influenza A infection: Treatment with rimantidine, Pediatrics 80:275-282.
188. Hayden, F. G., Sperber, S. j., Belshe, R. B., Clover, R. D., Hay, A. j., and Pyke, S., 1991,
Recovery of drug-resistant influenza Avirus during therapeutic use of rimantidine, Anti-
microb. Agents Ghemother. 35:1741-1747.
189. Betts, R. F., Treanor,j.j., Graman, P. S., Bentley, D. w., and Dolin, R., 1987, Antiviral agents
to prevent or treat influenza in the elderly,] Respir. Dis. 8(suppl. llA):S56-S59.
190. Brady, M. T., Sears, S. D., Pacini, D. L., Samorodin, R., DePamphilis,]., Oakes, M., Soo, W,
and Clements, M. L., 1990, Safety and prophylactic efficacy of low-dose rimantadine in
adults during an influenza A epidemie, Antimicrob. Agents Chemother. 34:1633-1636.
191. Hayden, F. G., Belshe, R. B., Clover, R. D., Hay, A.]., Oakes, M. G., and Soo, W, 1989,
Emergence and apparent transmission of rimantadine-resistant influenza A virus in
families, N. Engl.J Med. 321:1696-1702.
192. Sweet, C., Hayden, F. G., Jakeman, K.]., Grambas, S., and Hay, A.]., 1991, Virulence of
rimantadine-resistant human influenza A (H3N2) viruses in ferrets, J Inftct. Dis. 164:
969-972.
193. Degelau, ]., Somani, S. K., Copper, S. L., Guay, D. R. P., and Crossley, K. B., 1992,
Amantadine-resistant influenza A in nursing facility, Arch. Intern. Med. 152:390-392.
194. Knight, v., McClung, H. W, Wilson, S. Z., Quaries, ]. M., Waters, B. K., Cameron, R. W,
Greggs, S. E., Zewas, J. M., and Couch, R. B., 1981, Ribavirin small particle aerosol
treatment of influenza, Lancet 2:945-949.
195. Bernstein, D. 1., Reuman, P. D., Sherwood, j. R., Young, E. C., and Schiff, G. M., 1988,
Ribavirin small-particle aerosol treatment of influenza B virus infection, Antimicrob. Agents
Chemother. 32:761-764.
196. Stein, D. S., Creticos, C. M.,Jackson, G. G., Bernstein,]. M., Hayden, F. G., Schiff, G. M.,
and Bernstein, D. 1., 1987, Oral ribavirin treatment of influenza A and B, Antimicrob. Agents
Chemother. 31:1285-1287.
197. Ray, C. G., Icenogle, T. B., Minnich, L. L., Copeland,j. G., and Grogan, T. M., 1989, The
use of intravenous ribavirin to treat influenza virus-associated acute myocarditis, J Inflct.
Dis. 159:829-836.
17
Parainfluenza Viruses
RICHARD D. CLOVER
1. INTRODUCfION
The parainfluenza viruses are members of paramyxoviridiae family and the Para-
myxovirus genus. The first parainfluenza virus was isolated from a child with croup
and was initially referred to as CA (croup-associated) virus.I· 2 These viruses are
now recognized as important causes of respiratory disease in infants and young
children, producing a spectrum of disease ranging from a mild upper-respiratory-
tract infection to croup, bronchiolitis, and pneumonia. 3
2. VIROLOGY
Parainfluenza viruses are spherical, single-stranded RNA viruses, averaging

between 150 and 300 nm in diameter. 4 The RNA has a molecular weight of 5
x 10 3 kDa and is of negative polarity. Replication of the nucleocapsid occurs
within the cytoplasm of an infected cello The nucleocapsid is encased within an
envelope that is comprised of proteins specified by the virus and lipids derived
from the host cell membrane. The two surface glycoproteins that protrude from
the envelope as lO-nm spikes are the HN protein and the F protein.5 The inner
layer of the envelope consists of the matrix (or M) protein that maintains the
structural integrity of the virus.
The surface glycoproteins have been extensively studied. 6.7 The hemagglu-
RICHARD D. CLOVER • Department of Family Medicine, The University of Texas Medical

Branch. Galveston, Texas 77555.
Pulmonary Infoctions and Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
309
310 RICHARD D. CLOVER
tinating and neuraminidase aetivities of the virus are found in a single glyco-
protein, the HN protein, whieh is different from influenza viruses in whieh these
aetivities are found on two different glycoproteins. 8 The HN moleeule has six
topographieally nonoverlapping antigenie sites, eaeh eontaining multiple epi-
top es. 9 The F protein is involved with eell fusion and hemolytie aetivities of
the virus. The F pro tein preeursor, FO, is cleaved by host-eell proteinases into the
aetivated fusion protein. This pro tein consists of two subunit polypeptides, Fl
and F2. The amino terminus of the Fl is hydrophilie and highly conserved among
the parainfluenza viruses for the first 25 amino acids.l°,ll
Parainfluenza virus ean be separated into four antigenie types: Types 1, 2, 3,
and 4a and 4b. Similar viruses to types 1, 2, and 3 have been isolated from
animals, but infeetion in humans of these animal strains does not oeeur. The
human parainfluenza viruses have been antigenieally stable for more than 30
years. 12
3. EPIDEMIOLOGY
Parainfluenza viruses have a worldwide distribution. 13 ,14 The epidemie be-

havior is eharaeterized for types 1, 2, and 3, but the behavior oftype 4 is not weIl
eharaeterized beeause the disease resulting from this strain is so mild that it
usually does not require medieal attention. Parainfluenza viruses types 1 and 2 are
eommonly recovered in epidemies of respiratory illness in young presehool
children with lower laryngotracheobronchitis or eroupJ5-18 Passive maternal
antibody may be proteetive in infants less than 4 months of age and is a faetor
in the laek of severe disease in young infants produeed by these two viruses.l 9 In
eontrast, immunologie surveys indieate that parainfluenza type 3 is often experi-
eneed in the first year oflife, with 50% of the ehildren being serum positive by 12
months of age. 20-22 These infeetions in infants ean be severe, resembling the
bronehopneumonia eaused by respiratory syneytial viruS. 23
Early studies suggested that parainfluenza types 1 and 2 alternated years in
producing epidemies. 24 ,25 In reeent years this trend has not been seen, however.
Parainfluenza types 1 and 2 commonly produee illness in the fall prior to the
influenza season,15,17,26 Parainfluenza type 3 is more endemie in the eommunity
than are types 1 and 2, although type may be episodie with peak infeetions after
the influenza season. 21 Reinfeetions with all parainfluenza viruses are possible,
although reinfeetions are usually milder illnesses. 27 Infeetion with one type of
parainfluenza does not proteet that individual from the other types, however.
In humans, parainfluenza infeetions are assumed to be transmitted by the
respiratory route. 28 ,29 Transmission by hand contamination with nasal seere-
tions may be an additional mode of transmission of parainfluenza viruses. Noso-
eomial spread of parainfluenza and infeetion in hospitalized pediatrie patients
has been weIl doeumented, with one study reporting that approximately one-fifth
of hospitalized ehildren under 18 months aequire infeetion with parainfluenza
type 3.3°
PARAINFLUENZA VIRUSES 311
4. CLINICAL MANIFESTATIONS
The type of respiratory illness caused by parainfluenza virus is diverse. In

adults, the illness is predominately confined to upper respiratory tract,31 although
lower tract disease has been described,32 In children, the severity of the illness is
influenced by their age, the type ofvirus, and whether it's a primary infection or a
reinfection.16.19.33-35 By far the predominant form of the illness that leads to
hospitalization for parainfluenza virus is laryngotracheobronchitis or croup.15
Croup characteristically begins with upper respiratory symptoms of rhinorrhea
and sore throat with low grade fever. Over the next several days, the child
develops hoarseness and increasing respiratory compromise. The child then
presents to the physician with tachypnea, stridor, and intercostal retractions
secondary to subglottic edema. The cause of the respiratorycompromise is
demonstrated radiographically by a "steeple sign"-a subglottic narrowing pro-
duced by edema.
A second common cause of hospitalization for a parainfluenza infection is
bronchiolitis/bronchopneumonia, although this illness is most frequently caused
by RSv. Like croup, children with bronchiolitis usually have a several-day history
of an upper-respiratory infection manifested by rhinorrhea, sore throat, cough,
and low-grade fever. The children then develop lower-respiratory disease mani-
fes ted by wheezing, rhonchi, and/or rales. The cardinal signs of bronchiolitis are
wheezing and hyperaeration of the lungs. These children mayaiso have intercos-
tal retractions and nasal flaring. The children usually clinically improve over the
next 3 or 4 days.
Parainfluenza viral infections have also been associated with more chronic
conditions. Parainfluenza viruses may trigger episodes of wheezing in asthmatic
children. 36.37 Adults with exercise-induced bronchial constriction frequently
have a history of croup as a child. 38 In immunocompromised patients, para-
influenza virus infections can be quite severe. 39-43 Parainfluenza virus has been
shown to complicate bone-marrow transplantations by producing lower-respiratory-
tract disease that occasionally results in the death ofthe patient. 44 Parainfluenza
virus type 3 appears to be particularly frequent and severe in patients with severe
combined immunodeficiency.45.46 In addition, parainfluenza viruses have been
implicated in several central nervous system disorders including meningitis,47
Reye's syndrotne,48 and Guillain-Barre's syndrome. 49
5. DIAGNOSIS
Because the virus is shed in the respiratory tract, it is readily recovered by

nasal washing or suctioning. 50 Parainfluenza virus can be best isolated in primary
rhesus or continuous monkey kidney cells. 51 .52 Other cell lines like HeLa,
HEp-2, or human embryonic lung fibroblasts (WI-38 strain) are not as sensitive
for primary isolation of parainfluenza. Prompt inoculation provides best results,
because freezing and thawing or pH changes can produce significant decreases in
titers of virus. The eultures are subsequently observed for eytopathie effeet,
although this effeet is usually only seen with type 2. Infeeted eultures ean also be
deteeted by hemadsorption of guinea pig red eells to the eell monolayer. 53 The
eultures are read for hemadsorption at 4°C and again after elution at 23°C.54
Further identifieation of the virus may be aeeomplished by hemagglutination
inhibition, hemadsorption inhibition, or eomplement fixation using commercially
available antisera. 51 Immunofluoreseent staining of infeeted eells offers another
way of diagnosing parainfluenza.l 9,53,55
Serological diagnosis of infeetion may be aeeomplished by demonstrating a
fourfold antibody titer rise to the infeeting virus by utilizing one of several
teehniques including neutralization, hemagglutination inhibition, or eomplement
fixation.5 1,56 Serologieal dia gnosis of a specifie serotype is limited by heterotypie
antibody responses, especially during reinfeetions.57
6.IMMUNITY
After the parainfluenza virus is exposed to the respiratory mueosa, virus

replieation oeeurs. The incubation period of parainfluenza varies from 3 to 6
days.16,58 Although the virus ean be demonstrated in eiliated, eolumnar epithelial
eells,55 the meehanism for its subglottie involvement is not weIl understood. 59
Histopathologie ehanges in the lung produeed by parainfluenza are indis-
tinguishable from those produeed by other respiratory viruses.
After a parainfluenza virus infeetion, both serum and local antibodies are
found. 26 Following primary infeetion, antibody produetion to the HN protein is
greater than that for the F protein, and the HN-speeifie protein appears to
eontinue to rise for several months. 9 Antibody to the HN protein inhibits hem-
agglutinating and neuraminidase aetivity and inhibits exogenous virus from
infeeting a eell, although it does not inhibit virus spread by fusion. ll ,60 Antibody
to the F pro tein inhibits eell fusion and hemolyzing aetivity, and inhibits eell-to-
eell spread of parainfluenza viruS. 61 Reeent studies suggest the antibody response
to the fusion protein may be important in determining proteetion, and response
was only seen after several infeetions with parainfluenza viruses. 61,62
IgA antibody is deteetible in nasal seeretions after parainfluenza infeetion
within a week of onset of illness and is probably important in clearing the
infeetion. 63 ,64 A strong eorrelation with the level of seeretory IgA antibody and
proteetion from experimental infeetion has been demonstrated in adults. 65 In
addition, interferon has also been deteeted in nasal seeretions and may aid in
clearing the infeetion. 66
7. TREATMENT AND PREVENTION
Treatment for the parainfluenza infeetions is primarily symptomatic. The

need for hospitalization is based on the clinieal evaluation. 68 The degree of
tachypnea, dyspnea, and hypoxemia, the ability of the child to maintain adequate
hydration, and the comfortableness ofthe parent(s) or guardian(s) caring for the
child are factors that should be evaluated when considering hospitalization. The
administration of cool humidified air to prevent drying of secretions and to
soothe the inftamed glottis and airways is important in the management of croup.
The child should be closely observed for respiratory failure and potential need for
intubation and mechanical ventilation. For hospitalized children with croup,
nebulized racemic epinephrine appears to offer temporary relief of symptoms
and is supported by some authors. 69 The benefit of corticosteriods is unclear70
and is therefore not generally recommended. The therapeutic use of interferon
which is produced in respiratory secretions during parainftuenza infection67 is
not weIl established.
In vitra and experimental clinical studies have shown amantadine and riman-
tadine are not effective against parainftuenza virus. 71 Ribavirin that has antiviral
activity against a wide variety of DNA and RNA viruses inhibits parainftuenza
viruses at a minimal inhibitory concentration of 0.01 to 0.1 mg/m1. 55 ,64 Clinical
trials assessing the efficacy of ribavirin in children with parainftuenza infections
are lacking. Newer nucleoside analogs have been developed and are currently
under investigation. 64 ,72
Inactivated and live attenuated parainftuenza vaccines have been developed
and studied, although initial human trials did not demonstrate acceptable effi-
cacy.73-83 Because of ability of the parainftuenza viruses to reinfect, it is unlikely
that any vaccine can prevent total disease. A benefit would, however, exist if a
parainftuenza vaccine could be developed that would convert a severe primary
infection in an infant or young child to a mild illness of "reinfection". Intranasally
administered live attenuated vaccines are logical candidates because they can
produce local nasal antibodies that have been correlated with immunity to
parainftuenza. Since the human and bovine parainftuenza type 3 viruses share
several neutralization epitopes on the Fand HN glycoproteins, a bovine vaccine
has been developed and evaluated.B 2,84 In addition, several cold-adapted para-
inftuenza type 3 vaccines have been studied. 85 ,86 Although these studies show
some promise, further studies are needed to ensure these vaccines are adequately
attenuated, immunogenic, and efficacious in young in fants.
Another approach has also been taken in the development of a parainftuenza
vaccine. The recent demonstration of the importance of fusion protein in the cell-
to-cell spread of parainftuenza virus has led to efforts to develop a subunit vaccine
containing this glycoprotein and the HN glycoprotein. 87 The initial failure of
inactivated parainftuenza vaccines may have been a result of inadequate stimula-
tion of an antiviral response to the F protein. ll ,60,88
Prevention of nosocomial spread of parainftuenza viruses in hospitals and
other institutions has proved to be quite difficult. Infection control measures
include handwashing, respiratory isolation, and cohorting.B9 Although placing
these patients in respiratory isolation and having hospital personnel adhere to
good handwashing techniques would theoretically decrease transmission, nosoco-
mial transmission still occurs. This problem may be partially explained by the
314 RICHARD 0. CLOVER
prolonged shedding of parainftuenza (especially type 3) even in asymptomatic

patients. 41 •90,91
REFERENCES
l. Chanock, R M., 1956, Association of a new type of cytopathogenic myxovirus with infantile
croup, J Exp. Med. 104:555-576.
2. Chanock, R. M., Parrott, R. H., Cook, K., et al. , 1958, Newly recognized myxoviruses from
children with respiratory diseases, N. Engl. J Med. 258:207-213.
3. Monto, A. S., 1973, The Tecumseh study of respiratory illness. V. Patterns of infections with
the parainftuenza viruses, Am. J Epidemiol. 97:338-348.
4. Waterson, A. P. and Hurrell,]. M. W, 1962, The fine structure ofthe parainftuenza viruses,
Areh. Virusforseh.12:138-142.
5. Choppin, P. Wand Scheid, A., 1981, The functions and inhibition of the membrane
glycoproteins of paramyxoviruses and myxoviruses and the role of the measles virus M
protein in subacute sclerosing panencephalitis, J Infeet. Dis. 143:352-363.
6. Scheid, A., Caliguiri, L. A., Compans, R. W, et al., 1972, Isolation of paramyxovirus
glycoproteins. Association of both hemagglutinating and neuraminidase activities with the
larger SV5 glycoprotein, Virology 50:640-652.
7. Shioda, T., Iwasaki, K., and Shibuta, H., 1986, Determination of the complete nucleotide
sequence of the Sendai virus genome RNA and the predicted amino acid sequence of the
F, HN and L proteins, Nucleie Acids Res. 14:1545-1563.
8. Choppin, P. W, and Scheid, A., 1980, The role of viral glycoproteins in adsorption,
penetrations and pathogenicity of viruses, Rev. Infeet. 2:40-61.
9. Heilman, C. A., 1990, From the National Institute of Allergy and Infectious Diseases and the
World Health Organization: Respiratory syncytial and parainftuenza viruses, J Infeet. Dis.
161:402-406
10. Hsu, M-C., Scheid, A., and Choppin, P. W, 1981, Activation of the Sendai virus fusion
protein (F) involves a conformational change with exposure of a new hydrophobie region,
J Biol. Chem. 256:3557-3563.
11. Merz, D. C., Scheid, A., and Choppin, P. W, 1981, Immunological studies ofthe functions of
paramyxovirus glycoproteins, Virology, 109:94-105.
12. van Wyke Coelingh, K. L., Winter, C. C., and Murphy, B. R, 1988, Nucleotide and deduced
amino acid sequence of the hemagglutinin-neuraminidase genes of the human type 3
parainftuenza viruses isolated from 1957 to 1983, Virology, 162:137-143.
13. Chanock, R M., and Parrott, R. H., 1965, Para-inftuenza viruses, in: Viral and Riekettsial
Infections of Man. (4th ed. F. L. Horsfall, and I. Tamm, eds.), JB Lippincott, Philadelphia,
741-754.
14. Brandt, C. 0., Kim, H. W, Chanock, R M., et al., 1974, Parainftuenza virus epidemiology,
Pediatr. Res. 8:422.
15. Kilgore, G. E. and Dowdle, W R., 1970, Antigenie characterization of parainftuenza 4A and
4B by the hemagglutination-inhibition test and distribution of HI antibody in human sera,
Am. J Epidemiol. 91:306-316.
16. Downham, M. A. P. S., McQuillin, ]. and Gardner, P. S., 1974, Diagnosis and c1inieal
significance of parainftuenza virus infection in children, Areh. Dis. Child. 49:8-15.
17. Clarke, S. K. R, 1973, Parainftuenza virus infections, Postgrad. Med.J 49:792-797.
18. Denny, F. W, Murphy, T. F., Wallace, A. C., Jr., el al., 1983, Croup: An ll-year study in a
pediatrie practice, Pediatries 71:871-880.
19. Glezen, W P., Loda, F. A. and Denny, F. W, 1982, The parainftuenza viruses, in: Viral
Infeetions of Humans, (A. S. Evans, ed.), Plenum, New York, pp. 441-454.
20. Glezen, W P., Denny, E W, 1973, Epidemiology of acute lower respiratory disease in
children, N. Engl. J Med. 288:498-505.
21. Glezen, W P., Frank, A. L., Taber, L. H., et al. , 1984, Parainfluenza virus type 3: Seasonality
and risk ofinfection and reinfeetion in young children,J Infect. Dis. 150:851-857.
22. Denny, E Wand Clyde, W A., Jr., 1986, Acute lower respiratory tract infection in non-
hospitalized children, J Pediatr. 108:635-646.
23. Mufson, M. A., Krause, H. E., Mocega, H. E., et al. , 1970, Viruses, Mycoplasma pneumoniae
and bacteria associated with lower respiratory tract disease among infants, Am. J Epidemiol.
91:192-202.
24. Hall, C. B., 1981, Nosocomial viral respiratory infections: Perennial weeds on pediatric
wards, Am. J Med. 70:670-676.
25. Canchola, j., Vargosko, A. j., Kim, H. W, et al., 1964, Antigenic variation among newly
isolated strains of parainfluenza type 4 virus, Am. J Hyg. 79:357-364.
26. Chanock, R. M., Wong, D. C., Heubner, R. j., et al., 1970, Serologie response of individuals
infected with parainfluenza viruses, Am. J Public Health 50:1858-1865.
27. Loda, F. A., Glezen, W P., and Clyde, W A.,Jr., 1972, Respiratory disease in group day care,
Pediatrics 49:428-437.
28. Miller, W S. and Artenstein, M. S., 1967, Aerosol stabiJity of three acute respiratory disease
viruses, Proc. 80c. Exp. Biol. Med. 125:222-227.
29. McLean, D. M., Bannatyne, R. M., and Giban, K., 1967, Myxovirus dissemination by air, Can.
Med. Assoc. J 96:1449-1453.
30. Mufson, M. A., Mocega, H. E., and Krause, H. E., 1973, Acquisition ofparainfluenza 3 virus
infection by hospitalized children. I. Frequencies, rates and temporal data, J Infect. Dis.
128:141-147.
31. Bloom, H. H., Johnson, K. M., Jacobsen, R., et al., 1961, Recovery of parainfluenza viruses
from adults with upper respiratory illness, Am. J Hyg. 74:50-59.
32. Wenzel, R. P., McCormick, D. P., and Beam, W E., Jr., 1972, Parainfluenza pneumonia in
adults, JAMA 221:294-295.
33. Centers for Disease Control, 1978, Parainfluenza outbreaks in extended-care faciJities-
United States, MMWR 27:475-476.
34. Gardner, S. D., 1969, The isolation of parainfluenza 4 sub-types A and B in England and
serological studies of their prevalence, J Hyg. Camb. 67:540-545.
35. Chanock, R. M., Bell, j. A., and Parrott, R. H., 1960, Natural history of parainfluenza
infection, in: Perspectives of Virology II, (Pollard, ed.), Harper and Row, New York, pp.
126-139.
36. Mdntosh, K., Ellis, E. F., Hoffman, L. S., et al.,1973, The association of viral and bacterial
respiratory infections with exacerbations of wheezing in young asthmatic children,J Pediatr.
82:578-590.
37. Minor, T. E., Baker, j. W, Dick, E. C., et al., 1974, Greater frequency of viral respiratory
infections in asthmatic children as compared with their nonasthmatic siblings, J Pediatr.
85:472-477.
38. Loughlin, G. M. and Taussig, L. M., 1979, Pulmonary function in children with a history of
laryngotracheobronchitis, J Pediatr. 94:365-369.
39. Craft, A. W, Reid, M. M., Gardner, P. S., et al., 1979, Virus infections in children with acute
lymphoblastic leukemia, Arch. Dis. Child. 54:755-759.
40. Fishaut, M., Tubergen, D., and Mdntosh, K., 1980, Cellular response to respiratory viruses
with particular reference to children with disorders of cell-mediated immunity, J Pediatr.
96:179-180.
41. Gross, P. A., Green, R. H., and Curnen, M. G. M., 1973, Persistent infection with para-
influenza type 3 virus in man, Am. Rev. Respir. Dis. 108:891-898.
42. Karp, D., Willis, j., and Wilfert, C., 1974, Parainfluenza virus II and the immuno-
compromised host, Am. J Dis. Child. 127:592-593.
316 RICHARD 0. CLOVER
43. DeFabritus, A. M., Riggio, R R., David, S. 0., et al., 1979, Parainfluenza type 3 in a
trans plant unit, JAMA 241:384-386.
44. Wendt, C. H., Weisdorf, 0. j., Jordan, M. C., Balfour, H. H., Jr., and Hertz, M. 1., 1992,
Parainfluenza virus respiratory infection after bone marrow transplantation, N. Engl.] Med.
326(14):921-926.
45. Delage, G., Brochu, P., Pelletier, M., et al., 1979, Giant-cell pneumonia caused by para-
influenza virus,] Pediatr. 94:426-429.
46. Jarvis, W R., Middleton, P. j., and Gelfand, E. W, 1979, Parainfluenza pneumonia in severe
combined immunodeficiency disease,] Pediatr. 94:423-425.
47. Arguedas, A., Stutman, H. R., and Blanding, j. G., 1990, Parainfluenza type 3 meningitis,
Glin. Pediatrics 29(3):175-178.
48. Powell, H. C., Rosenberg, R. N., and McKeller, B., 1973, Reye's syndrome: Isolation of
parainfluenza virus, Arch. Neurol. 29:135-139.
49. PhiIlips, C. A. and Roman, G. 1978, Parainfluenza virus type 3. Isolation from CSF of a
patient with GuiIlain-Barre syndrome, JAMA 240:1613-1615.
50. Hall, C. B. and Douglas, R. G., 1975, Clinical useful method for the isolation of respiratory
syncytial virus,] Inftct. Dis. 131: 1-5.
51. Chanock, R. M., 1967, Parainfluenza viruses, in: Diagnostic Procedures for Viral and Rickettsial
and Ghlamydial Infections, 4th ed. (E. H. Lennette and N. j. Schmidt, eds.), American Public
Health Assoc., New York. pp. 611-632.
52. Frank, A. L., Couch, R. B., Griffis, C. A., et al., 1979, Comparison of different tissue cultures
for isolation and quantitation of influenza and parainfluenza viruses, ] Glin. Microbiol.
10:32-36.
53. Wong, 0. T., Welliver, R. C., Riddlesberger, K. R, et al., 1982, Rapid diagnosis of para-
influenza virus infection in children,] Glin. Microbiol. 16: 164-167.
54. Herrmann, E. C., Jr., and Hable, K. A., 1970, Experiences in laboratory diagnosis of
parainfluenza viruses in routine medical practice, Mayo Glin. Proc. 45:177-188.
55. Gardner, P. S., McQuillin, V j., McGuckin, R., et al., 1971, Observations on clinical and
immunofluorescent diagnosis of parainfluenza virus infections, Br. Med.] 2:7-12.
56. Frank A. L., Puck,j., Hughes, B. j., et al., 1980, Microneutralization test for influenza A and B
and parainfluenza 1 and 2 viruses that uses continuous celilines and fresh serum enhance-
ment,] Glin. Microbiol. 12:426-432.
57. Lennette, E. H., Jensen, F. W, Guenther, R. W, et al.,1963, Serologic responses to para-
influenza viruses in patients with mumps virus infection,] Lab. Glin. Med. 61:780-788.
58. Rendtorff, R. C., Walker, L. C., and Roberts, A. N., 1963, A parainfluenza 3 virus outbreak in
an orphanage nursery, Am.] Hyg. 77:82-97.
59. Zinserling, A., 1972, Peculiarities of lesion in viral and mycoplasma infections of the
respiratory tract, Virchows. Arch. 356:259.
60. Merz, 0. C., Scheid, A., and Choppin, P. W, 1980, Importance of antibodies to the fusion
glycoprotein of paramyxoviruses in the prevention of spread of infection,] Exp. Med.
151:275-288.
61. Kasel,]. A., Frank, A. L., Keitel, W A., etat., 1984, Acquisition ofserum antibodies to specific
viral glycoproteins of parainfluenza virus 3 in children,] Virol. 52:828-832.
62. Tyeryar, F. j., 1983, Report of a workshop on respiratory syncytial virus and parainfluenza
viruses, ] Infect. Dis. 184:588-598.
63. Yanagihara, R, and McIntosh, K., 1980. Secretory immunological response in infants and
children to parainfluenza virus types 1 and 2, Infect. Immun. 30:23-28.
64. Smith, C. B., Bellanti,j. A., and Chanock, R. M., 1967, Immunoglobulins in serum and nasal
secretion following infection with type I parainfluenza virus and injection of inactivated
vaccines,] Immunol. 99:133-141.
65. Smith, C. B., Purcell, R. H., Bellanti, j. A., et al., 1966, Protective effect of antibody to
parainfluenza type 1 virus, N. Engi.] Med. 275:1145-1152.
66. Shigeta, S., Mori, S., Baba, M., Ito, M., et al., 1992, Antiviral actIvlties of ribavirin,
5-ethynyl-l-beta-D-ribofuranosylimidazole-4-carboxamide, and 6' -(R)-6' -C-methylne-
planocin A against several ortho- and paramyxoviruses, Antimicrob. Agents Chemother.
36(2):435-439.
67. McKintosh, K., 1978, Interferon in nasal secretions from infants with viral respiratory
infections,] Pediatr. 93:33-36.
68. Newth, C. J., Levinson, H., and Byron, A. C., 1972, The respiratory status of children with
croup,] Pediatr. 81: 1068-1073.
69. Gardner, H. G., Powell, K. R., Roden, v., et al. , 1973, The evaluation of racemie epinephrine
in the treatment of infectious croup, Pediatrics 52:52-55.
70. Cherry, J. D., 1979, The treatment of croup: Continued controversy due to failure of
recognition of historie, ecologie, etiologic and c1inical perspectives,] Pediatr. 94:352-354.
71. Smith, C. B., Purcell, R. H., and Chanock, R. M.,1967, Effect of amantadine hydrochloride
on parainßuenza type 1 virus infectious in adult volunteers, Am. Rev. Respir. Dis. 95:689-690.
72. De Clercq, E., Cools, M., Balzarini, J., Snoeck, R., et al., 1991, Antiviral activities of
5-ethynyl-l-beta-D-ribofuranosylimidazole-4-carboxamide and related compounds, Anti-
microb. Agents. Chemother. 35(4):679-684.
73. Crookshanks, R. F. and Belshe, R. B., 1984, Evaluation of cold adapted and temperature
sensitive mutants of parainßuenza virus type 3 (para 3) in weanling hamsters,] Med. Virol.
13:243-249.
74. Belshe, R. B. and Hissom, R. K., 1982, Cold adaptation of parainßuenza virus type 3:
Induction of three phenotypic markers,] Med. Virol. 10:235-242.
75. Kim, H. w., Canchola, J. G., Vargosko, A. J., et al., 1966, Immunogenicity of inactivated
parainßuenza type I, type 2, and type 3 vaccines in infants, JAMA 196:819-824.
76. Sweet, B. H., Tyrell, A. A., Potash, L., et al., 1966, Respiratory virus vaccine III. Pentavalent
respiratory syncytial parainßuenza Mycoplasma Pneumonia vaccine, Am. Rev. Respir. Dis.
94:340-349.
77. Chin, J., Magoffin, R. L., Shearer, L. A., et al., 1969, Field evaluation of arespiratory syncytial
virus vaccine and a trivalent parainßuenza virus vaccine in a pediatric population, Am. ]
Epidemiol. 89:449-463.
78. Fulginiti, V. A., Amer, J., Eller, J. J., et al., 1967, Parainßuenza virus immunization. IV.
Simultaneous immunization with parainßuenza types I, 2, and 3 aqueous vaccines, Am.] Dis.
Child. 114:26-28.
79. Fulginiti, V. A., Eller,J. J., Sieber, 0. F., el al., 1969, Respiratory virus immunization. I. A field
trial oftwo inactivated respiratory vaccines; an aqueous trivalent parainßuenza virus vaccine
and an alum~precipitated respiratory syncytial virus vaccine, Am.] Epidemiol. 89:435-448.
80. Vella, P. P., Weibel, R. E., Woodhour, A. F., el al., 1969, Respiratory virus vaccine. VIII. Field
evaluation of trivalent parainßuenza virus vaccine among preschool children in families,
1967-1968, Am. Rev. Respir. Dis. 99:526-541.
81. Wigley, F. M., Fruchtman, M. H., and Waldman, R. H., 1970, Aerosol immunization of
humans with inactivated parainßuenza type 2 vaccine, N. Engi.] Med. 283:1250-1253.
82. van Wyke Coelingh, K. L., Winter, C. C., Tierney, E. L., et al., 1988, Attenuation of bovine
parainßuenza virus type 3 in nonhuman primates and its ability to confer immunity to
human parainßuenza virus type 3 challenge, ] Inftct. Dis. 157:655-662.
83. Potash, L., Lees, R. S., Greenberge,J. L., et al., 1970, A mutant ofparainßuenza type 1 virus
with decreased capacity for growth at 38C and 39C,] Infect. Dis. 121:640-647.
84. Clements, M. L., Belshe, R. B., King, J., Newman, F., et al., 1991, Evaluation of Bovine, cold-
adapted human, and wild-type human parainßuenza type 3 viruses in adult volunteers and
in.chimpanzees,] Clin. Microbiol. 29(6):1175-1182.
85. Hall, S. L., Stokes, A., Tierney, E. L., London, W. T., el al., 1992, Cold-passaged human
parainßuenza type 3 viruses contain ts and non-ts mutations leading to attenuation in rhesus
monkeys, Virus Res. 22(3):173-184.
86. Belshe, R. B., Karron, R. A., Newman, F. K., Anderson, E. L., et al. , 1992, Evaluation of a live
attenuated, cold-adapted parainfluenza virus type 3 vaccine in children.} Clin. Microbiol.
30(8):2064-2070.
87. Ray, R., Brown, V. E., and Compans, R. w., 1985, Glycoproteins of human parainfluenza
virus type 3: Characterization and evaluation as a subunit vaccine,} Infect. Dis. 152:1219-
1230.
88. Ray, R., Glaze, B. j., and Compans, R. w., 1987, Role of the individual glycoproteins of
human parainfluenza type 3 in the induction of a protective immune response, Virology
61:3416-3423.
89. Graman, P. S. and Hall, C. B., 1989, Epidemiology and control ofNosomial viral infections,
Infectious Disease Clinics of N(ffth America 3(4):815-841.
90. Frank, A. L., Taber, L. H., Wells, C. R., et al., 1981, Patterns of shedding of myxovirus and
paramyxoviruses in children,} Infect. Dis. 144:433-441.
91. Muchmore, H. G., Parkinson, A. j., Humphries, j. E., et al.,1981, Persistent parainfluenza
virus shedding during isolation at the South Pole, Nature 289:187-189.
18
Varicella-Zoster Virus
GERALD LANCZ and STEVEN SPECTER
1. INTRODucnON
Descriptions of the clinical entity varicella, also known as chickenpox, and of

zoster, also known as shingles, were found in the writings of the ancient Greek
physicians. The name varicella is a word that is a diminutive form of the word for
variola, which refers to the viral disease smallpox. We recognize today that the
relationship between these two diseases is more literary than medical, although
both viral agents produce vesicular eruptions that present as exanthema and
these lesions are somewhat similar in appearance. That the lesions of smallpox
are often larger and more numerous than those of varicella no doubt contributed
to the mistaken notion that the latter disease was a diminutive form of the former.
There is, however, no significant physical, chemieal, or antigenie relationship
between the viral agents that produce these two diseases. How varicella also
became known as chickenpox is unclear. This popular name for this disease
unfortunately also suggests a relationship to smallpox. By contrast, zoster is
derived from a Greek word which means belt or girdle, an obvious reference to
the localized type of infection associated with the intercostal clinical manifesta-
tions of this disease.
The epidemiological features of the two clinical syndromes, varicella and
zoster, are very distinct. Varicella displays seasonality whereas zoster does not.
Varicella infection occurs in epidemics and zoster appears as isolated cases in
individuals. Perhaps because of the similarity in the physical appearance of the
vesicles produced, a relationship between the etiologic agents causing varicella
and zoster was hypothesized in the early twentieth century and then demon-
GERALD LANCZ and STEVEN SPECTER • Department of Medical Microbiology and Immu-
nology, University of South Florida College of Medicine, Tampa, Florida 33612.
Pulmonary Infections arul Immunity, edited by Herman Chmel et al. Plenum Press, New York, 1994.
319
320 GERALD LANCZ and STEVEN SPECTER
strated when it was shown that individuals who received an inoculation of fluid
taken from a zoster patient developed chickenpox. 1.2 The initial cultivation of
varicella by Weller in 19533 enabled the direct comparison of the antigenic
properties of the agents isolated from individuals with chickenpox and shingles.
These investigations indicated that the viruses were either identical, antigenically
speaking, or were very closely related infectious agents. 4 •5 With the develop-
ment of the concepts of virallatency and the reactivation of latent viral agents,
with herpes simplex virus as the model, it followed quickly that zoster may
represent the reactivated form of a primary varicella infection. Verification at the
genetic level that a single viral agent is responsible for the production of chicken-
pox and shingles was provided by Straus et al. 6 who isolated a viral agent from an
individual who had experienced varicella and subsequently zoster. Using restric-
tion enzyme analysis, this study showed that the DNA recovered from the two viral
isolates was identical.
Chickenpox tends to be regarded as a relatively benign disease that is seen
principally in children, whereas zoster is seen principally in an elderly popula-
tion. Both of these infections are associated with a significant number of physi-
cian visits annually. For example, varicella epidemics are responsible for signifi-
cant absenteeism from schools. In immunocompromised individuals, varicella
infection and/or exposure to the varicella-zoster virus is considered a life-
threatening event. This is particularly true for children and adults who are
immunosuppressed as a result of cancer or who are infected with the human
immunodeficiency virus (HIV). The potential medical problems that arise in this
population stern from the primary infection in an individual whose immune
system is depressed, as weIl as the possibility of reactivation of endogenous latent
virus. Thus, varicella-zoster virus (VZV) infections, either primary or a reacti-
vated latent infection, represent medically important viral disease states that
require continuing surveillance.
2. PROPERTIES OF THE VIRUS
Analysis of the physical-chemical properties of VZV indicated that it was

morphologically different compared to the smallpox virus, but was related to the
herpesviruses. The virus nucleocapsid is approximately 100 nm in diameter and
is surrounded by a lipid-containing envelope. The nucleocapsid surrounds the
viral core, which contains the double stranded DNA genome, and is composed of
162 hexagonal capsomeres that bear a central hole measuring 4 nm in diameter.
The nucleoca psid possesses a 5: 3: 2 axial symmetry typical of its icosadeltahedral
shape. The DNA is in the form of a toroid and forms the core of the virus. The
tegument is the region that lies between the external reaches of the nucleocapsid
and the inner surface of the envelope. It contains viral proteins that function as
regulators of viral replication. The envelope surrounds the tegument region and
the nucleocapsid and contains a number of virion-directed glycoproteins that
serve to promote and enable the attachment of the virus to a susceptible cello The
VARICELLA-ZOSTER VIRUS 321
enveloped virion ranges from 180 to 200 nm in overall diameter. Since the
envelope contains the viral proteins employed to initiate infection, the loss of the
virion envelope renders the virus noninfectious. The presence of an envelope is
also referred to as ether sensitivity. Thus, physical-chemical properties and
morphological characteristics place VZV in the herpesvirus group.
The virus contains a double-stranded DNA genome with a molecular weight
previously estimated at 100 x 10 3 kDa. 7 More precise techniques employing
restrietion enzyme analysis of virion DNA, however, suggest that the DNA has a
molecular weight closer to 80 x 10 3 kDa. 8 ,9 The genome recently has been
sequenced by Davidson and ScottlO and found to contain approximately 125,000
base pairs. The virion DNA is a linear molecule characterized by having a long
unique segment (approximately 83% of the genome) and a shorter unique
segment that can invert relative to the long segment. Thus, the VZV DNA exists
in two isomerie forms based on the physical orientation of the short unique
sequence. ll Viruses possessing either physical form of DNA are infectious. It has
been shown also that up to 5% of the time, the long unique sequence inverts as
weIl. 12 The isomerie form of the virion nucleic acid appears to be inconsequential
to the virion as a comparison of the genetic maps of different VZV isolates
demonstrates the genetic relatedness of all virus isolates regardless of whether
it was obtained from an individual with chickenpox or shingles. Some structural
variation of the genome is present, however, in the OKA strain of VZv, astrain
that is being evaluated as a candidate vaccine in humans.I 3 ,14
The physical maps of the different members of the human herpesvirus
group show some variation, particularly with regard to repeat sequences found
internally as weIl as at the termini of the genome. As a result of repeat nucleotide
sequences and a recombination-like mechanism, however, the DNA present in
different virus particles within a single virus population may be in different
isomerie forms. These DNA moleeules differ as a result of the ability of segments
of the genome to invert relative to one another. 15 Thus, the genome of the
different human herpesviruses exists in multiple isomerie forms that can vary
with the particular herpesvirus. Nonetheless, the different physical forms of DNA
have no effect on the ability of the virus to replicate in susceptible cells. Compara-
tive analysis of the physical maps of the human herpesvirus DNAs indicates
homology between major segments and critical regions associated with replica-
tion of the viruses. Thus, it is held that the general and overall processes
associated with viral replication are similar for all the human herpesviruses.
VZV is highly susceptible to inactivation and is one of the more labile
members of the herpesvirus group. In addition to susceptibility to trypsin and
ether treatment, VZV readily inactivates when it is extracellular. This extreme
lability ofVZV assurnes great significance when considering storage and distribu-
tion of candidate vaccines. Cell-free preparations of virus used to study the
regulation of virus replication are prepared by sonication of infected cells. It has
been shown, however, that liberation of the virus from its cell association reduces
the virus titer by approximately 210g10 within 2 min of sonication. 16 ,17 By contrast,
virus infectivity is enhanced when virus-infected cells are frozen under conditions
322 GERALD LANCZ and STEVEN SPECfER
that favor survivability of the infected cells. This indudes freezing cells in the
presence of glycerol,18 as weIl as frozen preservation of infected cells as a
monolayer. 19 Grose et al. 20 showed that virus and virus infected cells would remain
viable iflyophilized and stored at -20 or -70°C. Further evaluation of conditions
to stabilize the virus for storage and transport, preferably in a lyophilized state,
assurnes medical significance for distribution of suitable vaccine to remote geo-
graphie areas.
3. VIRUS REPLICATION
VZV most likely infects a cell as a result of the fusion of the virion envelope
with the plasma membrane of the cello This act of membrane fusion provides the
nudeocapsid and tegument proteins entry to the cytosol of the cello The tegument
proteins are liberated, and some then function within the cytosol by exerting
regulatory control of cell protein synthesis, while the nudeocapsid traverses the
cytosol and provides for the deposition of virion DNA into the nudeus. Additional
tegument proteins that are a constitutive part of the virion function as transcrip-
tional regulators within the cell nudeus. The synthesis of virion transeripts and
their translation is sequentially ordered and temporally regulated. By 8 to 12 hr
after infection, virus-specific proteins can be detected by immunofluorescence
that corresponds to the observed dissemination of virus-associated cytopathol-
ogy.21,22 With the synthesis of the late viral structural proteins, the virion nudeo-
capsid is assembled in the nudeus of the cell where progeny VZV DNA already
has been synthesized. Self-assembly of virion proteins results in a procapsid
structure. This structure surrounds or incorporates the virion DNA by mecha-
nisms unknown, resulting in the encapsidation of the genome and the formation
of the VZV nudeocapsid. The critical and regulatory tegument proteins either
become associated with the nudeocapsid by some specific selection process or
become physically trapped by a more random process. This structure then buds
from the nudear membrane and in so doing picks up its envelope, which already
contains the virion-directed glycoproteins involved in virion attachment to sus-
ceptible cells. Virion egress from the external recesses of the plasma membrane
results in the fusion of the virus partide with adjacent cells, initiating a new round
of infection. This ensures VZV's cell association and helps overcome its suscep-
tibility when cell-free. The extreme lability of the extracellular virus has ham-
pered the ability to perform genetic analysis and to more thoroughly dissect the
replicative cyde of the virus because it is difficult to obtain plaque-purified virus
populations. Virus infection is most efficiently initiated by infecting cell cultures
with other infected cells.
Regarding immunity to virus infection, the virion-directed glycoproteins
represent a major target for investigation. These viral-directed proteins, ofwhich
there are six to eight, are found in the most external portion of the virus,
embedded in the virion envelope. Thus, these proteins are the first that are
encountered by the host immune system when it encounters an infecting virus, as
weIl as being present in the infected cell membranes. gpI, located at approx-
imately 0.94 map units on the genome,23 gpIl, located at 0.47 map units,24 and
gpIII, located at map unit 0.53,25 are the three most abundant glycoproteins
produced by the virus during the cyde of replication. Each of these glycoproteins
will elicit antibodies that neutralize the virus. 26-28 Two remaining glycoproteins,
gpIVand gpv, are respectively located at map units 0.92 and 0.17.1°,29 These two
glycoproteins are produced in lesser abundance in virus-infected cells. The
source and status of other glycoproteins detected in VZV and infected cells is
under investigation.
4. EPIDEMIOLOGY OF VARICELLA-ZOSTER VIRUS INFECTIONS
Man is the natural reservoir for VZv. Thus, the development of an annual
outbreak of virus infections, generally seen in the late winter and early spring of
the year, results from the exposure of susceptible individuals, defined as being
seronegative, to virus shed from other humans. elose contact is required for
successful spread of the virus, which is believed to enter the susceptible individual
via the respiratory route. Varicella is relatively contagious, with a 70-90% attack
rate observed in susceptible household contacts living with an infected individ-
ual. 30 There are an estimated 3 million infections annually.31,32 The disease is
often seen in children 3 years of age or less, although school-age children
represent a significant segment of susceptibles.
By contrast, individuals susceptible to developing zoster, or shingles, are
those who previously have had dinical or subdinical chickenpox. Zoster, a
sporadic disease, is the reactivated form of latent VZv. All age groups are
susceptible to developing zoster, but the disease is more commonly seen in the
elderly.33 Thcrc is no scasonal variation associatcd with thc dcvelopmcnt of zostcr,
but the incidence of zoster has been found to increase in individuals whose
cellular immune system is deficient or impaired, for example, in those with
malignancies, transplant recipients, AIDS patients, and so forth. 34,35
5. PATHOGENESIS
5.1. Varicella
A susceptible individual becomes exposed to the virus as a result of relatively

dose contact with another infected human. The incubation period is generally 14
days with a range of 10-20 days.32,36 Virus is believed to gain entry via the
respiratory route and/or possibly via the conjunctiva. Local replication takes place
at the site of entry, eventually resulting in a mild viremia that disseminates the
virus throughout the body. Virus is picked up by cells of the reticuloendothelial
system, within which viral replication takes place. VZV then establishes a second-
ary viremia that coincides with the prodromal sym ptoms associated with the onset
of the disease state.3 7- 39 Virus replication in the endothelial cells of the capillary
beds enables rapid spread of virus to the epithelial cells in adjacent tissue, such as
epidermis. Replication of the virus in the basal levels of the epidermis results in a
separation of the dermal and epidermal cell layers and the development of a
fluid-containing vesicle. The fluid is initially clear and the vesicle takes on a
dewdrop-like appearance. Simultaneously, the area is infiltrated with leukocytes,
initially into the dermallayers and subsequently into the vesicle itself, turning the
fluid cloudy. These cells ingest and destroy free VZV present in the fluid, as weIl
as being the likely source of interferon found in vesicular fluid. 4o The vesicular
fluid is resorbed over the next day or so, or the vesicular roof may dry and the
crusted vesicle is sloughed.
The pathologie changes observed in cells surrounding the vesicle mimic
those observed in tissue culture cells. In these VZV-infected ceIls, ballooning
degeneration is readily apparent, accompanied by margination of nuclear chro-
matin and development of both intranuclear inclusions and multinucleate giant
cells, such as syncytia. 41 ,42
Although a relatively benign disease of normal children, varicella infection
can be severe in neonates, the immunocompromised and, for reasons unexplained,
is more severe in normal adult populations. 43 ,44 The development of generalized
varicella in the neonate and the development of varicella pneumonia in the adult
and immunocompromised populations represent significant medical problems.
Varicella pneumonia often results in focal areas of infection, resulting in a
hemorrhagic necrosis. Studies of the immune response associated with VZV
infection suggest that cell-mediated immune responses are of paramount impor-
tance in recovering from infection. Individuals with defects in cell-mediated
immune (CMI) systems have more problems in general with viral infections than
individuals who have normal CMI responses but who may have a defect in their
humoral immune system. (Enterovirus infections are one notable exception).
Thus, immunocompromised individuals and neonates in whom cellular immune
responses are deficient are more prone to severe, life-threatening VZV infections.
5.2. Herpes Zoster

Herpes zoster represents the reactivation of latent VZv. One current notion
of the pathogenic process was put forth by Hope-Simpson. 33 This notion suggests
that dermal lesions associated with varicella present virus to the sensory nerve
endings innervating the skin that be ars the vesicular eruptions. The virus makes
its way in aretrograde manner along nerve fibers to the sensory ganglia wherein
the virus establishes a latent infection in nerve celis. Presence of virus DNA and
some viral RNA transcripts in sensory ganglia have been demonstrated. 45 ,46 The
specific triggering event(s) associated with reactivation of the latent virus are not
known with any degree of certainty but, as suggested by Hope-Simpson,33 the
immune system seems to playa major role in keeping the latent virus in a latent
state. According to the notion, when specific anti-VZV immunity declines below a
threshold level, reactivated virus is no longer held in check. It begins to repli-
cate in and establishes a spreading infection of epithelial cells, having passed

down the sensory nerve cells from the ganglion. The striated demarkation of the
resulting vesicular eruption, whose histologic morphology is identical to that of
varicella, is a hallmark characteristic of zoster and reßects the dermatome or
innervated locus that bore most heavily the primary varicella vesicular lesions.
Cell-mediated immune responses seem to be more specifically associated with
containment ofVZV in a latent state. Studies indicate that a transient decrease in
CMI responsiveness to VZV is present during the early and acute phases of
zoster. 47 ,48 Hence, children with lymphoma as well as AIDS patients who have
decreased CMI are more susceptible to developing disseminated zoster infections
and the increased morbidity and mortality observed. An overall decrease in VZV
CMI does not explain why highly localized zoster lesions are observed instead of
generalized virus infection, however.
5.3. Clinical Course

5.3.1. Varicella
In the normal child, the prodromal period prior to the onset of the disease
state is usually unrecognizable. The typical incubation period is 14 days and the
disease state is initiated with the onset of the exanthem, or some generalized and
nondescript prodromal symptoms immediately before the appearance of the
exanthem. The development of a low-grade fever accompanied by chills and
malaise, probably reßects the release ofbiologically active cytokines that mediate
these host responses. The disease state in children is normally benign and the
mortality rate is less than 2 per 100,000. In the normal adult population, the
morbidity and mortality rise 10- to 20-fold over those seen in normal children, the
fever response is higher, and the overall symptoms observed are more severe.
Illness is associated with fever and chills of several days duration, headache,
backache, and other constitutional symptoms.
The exanthem is typically seen initially on the body and head and, over time,
progresses to the extremities (Fig. I). The lesions rapidly develop from a mac-
ule to a papule with an erythematous base (Fig. 2) and then to a vesicle with a clear
dewdrop appearance. Multiple crops of vesicles develop over a successive 3- to
4-day period. Typically, 100 or more lesions will develop. The lesions undergo a
rapid evolution and within a day begin to dry and form a crust that drops off over
the next few days. Crusts are typically gone over a 10- to 12-day period of time.
Altered pigmentation of the skin results when the crusted lesion falls off (Fig. 3).
These spots of altered pigmentation and the shallow ulcers that may remain when
crusts drop off (Fig. 4) are indicators of past VZV infection.
The lesions are pruritic, which leads to the most common complication of
chickenpox, bacterial infection of the skin bearing the vesicular exanthem.
Bacterial infection of the vesicular lesions is treated with antibiotics. In the normal
adult, the most common serious complication seen is varicella pneumonia, which
has an incidence of 10_15%.44,49 X rays indicate that varicella pneumonia pro-
liGU!lE 1. lnfected 7-year-old male with numerous pruritic chickenpox lesions on head and
neck. VesicIes are present on lower lip, left eyeIid and neck. Oral lesions also deveIoped (not
shown). Note: Lesions are more numerous on the head, neck and torso than on the extremities..
FIGURE 2. Typical varicella lesion associated with primary infection producing chickenpox.
Note the erythematous base (arrows).
duces diffuse densities in both lungs 49 which leave nodular calcifications upon
resolution. The mortality is approximately 10% in the normal adult population
and is higher in individuals who are immunoincompetent. Pregnant women have
a higher incidence of varicella pneumonia than other adults because of natural
immunosuppression associated with pregnancy.
Congenital infection with VZV has been associated with anomalies, espe-
cially if infection takes place in the first trimester. In utero infection during the
second half of gestation has a lower incidence of severe congenital malformations.
Neonatal VZV infection may result when the mother develops primary infection
within a few days of delivery, however. In this scenario, the mother has not had
sufficient time to develop significant humoral or cellular immune responses. As a
consequence, protective antibodies are not transferred to the neonate and the
neonate may develop a severe generalized VZV infection. Neonatal varicella
results in a mortality rate which can approach 30%,50 The illness initiates with
infection shortly after birth in the infant whose passively transferred maternal
protection is not sufficient to fully protect the infant from disease. In these
instances, a generalized type of infection may occur, resulting in the higher
mortality observed. Maternal transfer of lower levels of protective VZV antibody
may result in a modified form of VZV with a correspondingly lower morbidity
and mortality. In children who are immunoincompetent such as children with
leukemia, and in the AIDS patient, the incidence and severity of varicella are
significantly heightened. Children are at a high risk for infection of visceral
organs and in these cases, approach a fatality rate of 10-20%. In addition,
secondary bacterial infections that may become generalized are an additional
complication specifically in children with malignancies of the neutrophil cell
lineage. Similar types of generalized bacterial infections are observed in AIDS
patients, resulting in a much higher level of pneumonitis in both children and
adults.
Another severe postvaricella complication is encephalitis. Fortunately, this
complication is exceedingly rare, although the case incidence increases with the
increasing age of the patient. In the adult population, encephalitis can be life-
threatening and has been reported to occur in 0.1 % of adults. 51 Progressive
disease in this population results in altered levels of consciousness and a mortality
rate of approximately 15%. Neurologie sequelae are present and occur in as
many as 15% of survivors.
5.3.2. Zoster
The onset of zoster is typically preceded by pain in the affected dermatome

for a 2- to 3-day period. Constitutional symptoms induding fever and so forth are
generally not observed. The vesicular eruption continues to form over a 3- to 5-day
period. The lesions go through the same progression as with varicella but are
localized. They delineate the infected dermatome, and do not cross the midline of
the body, for example, intercostal zoster. The most common manifestations of
zoster occur in the thoraeie or lumbar regions. Less frequently, ophthalmie zoster
FIGURE 3. Altered (lighter) skin pigmentation as a result of sloughing of crusted chickenpox

vesicIes. Lesions present in an 8-year-old male.
FIGURE 4. Shallow ulcer that remains following the loss of crusted, dried chickenpox in a 15-
year-old male. These permanent ulcers have been used as evidence of prior exposure to VZv.
is observed as a result of reactivation of latent virus in the trigeminal ganglion.

The lesions heal in approximately 2 weeks but the associated neuralgia may
persist for a month or longer. The postherpetic neuralgia, less common in the
young, may occur in as many as half of the patients over 50 years of age, and
may be quite debilitating. 52 •53
In the immunocompromised host, herpes zoster is a more severe disease, as
lesions may continue to develop for up to a 2- to 3-week period of time. Scab-
bing is not complete until 1 to 2 months later.54 In HIV-infected individuals,
there is also an increased incidence of zoster. 55 Fortunately, even in the immuno-
compromised patient, dissemination of the zoster infection is relatively rare and is
not usually fatal.
6. DIAGNOSIS
The diagnosis ofboth chickenpox and shingles can be made by clinical signs
and sym ptoms with a high degree of accuracy because of the characteristic lesions
produced. Varicella is generally seen in children in epidemie form in the winter or
spring of the year. The development of typieallesions and the presence of lesions
in all stages of development, from papular to crust, in a single anatomieal area is
highly indicative of varicella infection. Similarly, the zoster lesions charac-
teristically develop along a specific dermatome and do not cross the midline of the
body. In cases of atypical or modified varicella, in individuals who have some
degree but not total immunity, the clinical presentation can be confused with
other vesieulogenic disease states. Thus, a vesicular eruption on an erythematous
base can be caused by bacterial or other viral infections, as well as allergie
reactions. A laboratory-based differential diagnosis with data that support the
clinical evaluation may be criticaHy important, particularly in individuals where
enhanced morbidity and mortality is likely to ensue in the infected individual.
Serologie diagnosis can be performed by monitoring a minimal fourfold rise
in antibody titer to VZv. A complement-fixation test was the first developed to
monitor VZV antibody56 but has been replaced by other, more specific tests,
because of its lower sensitivity. Alternative tests include ELISA, as weH as various
immunofluorescent assays, to detect virus-induced membrane antigens or other
viral products.57•58 A single antibody can be tested for the presence of relatively
high levels of IgM directed against VZV which would be indieative of a primary
VZV infection, although this test is not routinely performed.
Alternatively, viral diagnosis is accomplished by virus isolation and confirma-
tion of the virus isolate as VZV. This is best accomplished by obtaining fluid from
clear vesicles with the aid of a small-bore syringe and then immediately inoculat-
ing appropriate cell cultures. Recall that the virus is quite labile, thus, an
immediate inoculation of cell cultures enhances virus recovery. When this is not
possible, store the viral aspirate in tissue culture medium or skim milk containing
transport medium for 24 hr at 4°C or for longer periods at -70°C. The viral
specimen is inoculated onto susceptible tissue culture cells, and human embry-
onic or diploid fibroblasts are commonly em ployed. The cell cultures are observed
for the development of cytopathology typical of virus isolation. The time in which
a virus can be detected can vary from 2 to 14 days and may be enhanced if the
inoculated cell cultures are centrifuged immediately after inoculation. The virus
can be identified by a variety of antibody-based techniques including immuno-
fluorescence, immunoperoxidase, and so forth and the presence ofVZV DNA has
recently been detected in CSF using PCR as a means of diagnosing infections. 59
These latter serologic and molecular probe tests are highly specific for VZv,
offering absolute confirrnation of infection, and have replaced the use of non-
specific tests such as the Tzanck smear, which sought the presence of viral
intranuclear inclusion bodies as a means of diagnosing VZV infection.
7. THERAPY
The administration of antiviral chemotherapeutic agents for chickenpox is

not indicated for infections of normal populations. In the case of children with
chickenpox, symptomatic treatment to restrict complications associated with the
disease state is important, that is, because the lesions are pruritic, decreasing the
itch decreases the amount of scratching which decreases the incidence of second-
ary bacterial infection of the skin. Acetaminophen is recommended as an anti-
pyretic. Note: The use of aspirin for children infected with VZV (and influenza)
as an antipyretic is contraindicated because of the association between aspirin and
Reye's syndrome. In the normal adult population, although the risk of viral
pneumonitis is greater than in children, generally speaking administration of
anti-VZV drugs is not indicated unless evidence of pneumonitis or pneumonia is
present.
Therapy for the immunocompromised host is another story. Individuals
immunosuppressed as a result of cancer, pregnancy, AIDS, organ transplantation,
and so forth, have been shown to benefit from receiving anti-VZV chemothera-
peutic agents such as adenine arabinoside (Vidarabine), or acyclovir (Zovirax).
For example, Landsberger, Hager, and Grossman report successful management
of 3 pregnant females who had been exposed to or developed VZV pneu-
monitis. 60 They employed a combination of aggressive supportive therapy, ventila-
tion, and respiratory support in the setting of an intensive care facility, along with
the use of these antiviral agents. In each instance, the combination of aggressive
support and antiviral therapy resulted in clearing of the lungs as evidenced by
X ray and resulted in the prevention of virus dissemination. Similarly, treatment
of immunosuppressed children with acyclovir or Vidarabine results in a de-
creased number of lesions, a lessening of the severity of disease, a decreased time
ofhealing, and a decrease in the number of serious complications associated with
VZV infection. 61 •62 Leukocyte interferon is also under evaluation for use in
humans and has been found to provide positive results in children with cancer. 63
Zoster often appears with a greater frequency in immunosuppressed individ-
uals, including the elderly who are immunosuppressed as a result of increasing
age. The use of antiviral chemotherapeutics such as acyclovir and adenine

arabinoside has been successful in decreasing the severity of the lesions and the
time of healing. As a result, individuals treated with these nucleoside analogs
have a decreased incidence of disseminated lesions and frequency of visceral
disease. 64-66
The search for new, effective anti-VZV chemotherapeutic agents is being
actively pursued. Modified nucleoside analogs that inhibit and/or are specific for
viral targets such as virion DNA polymerase, virion thymidine kinase, and so
forth, have been synthesized and tested in vitro. 67- 7o These compounds have
therapeutic indices that portend their in vivo effectiveness subject to a lack of
pharmacologic toxicity.
Children born with congenitally acquired infections should receive varicella
zoster immune globulin (VZIG) or zoster immune plasma (ZIP) therapy within 72
hr of birth so that the passively transferred immune globulin will significantly
retard or preclude the establishment of a generalized VZV infection in the neo-
nate. Similarly, VZIG is employed in children and other nonimmune and immuno-
compromised individuals who receive a primary exposure to VZV. Passive admin-
istration of the immune globulin should follow the established guidelines of the
agency providing the material. Specifics regarding the availability and adminis-
tration of these antibody preparations can be obtained from the American Red
Cross, which distributes these materials through regional blood centers. 7I
The treatment of postherpetic neuralgia is particularly problematic, as a vast
array of agents and modalities have been tested with mixed results. Overall the
advent of specific anti-VZV chemotherapeutic agents has been beneficial for the
treatment of both primary varicella and recurrent zoster infections as used in
specialized patient populations. Their prompt administration, along with the
prompt administration of passive immune globulin in the form of VZIG, yield
maximum beneficial results.
8. PREVENTION
With the exception of developing viral encephalitis, the incidence of signifi-

cant and severe complications as a result of a viral infection in normal children in
the age range of 3 years to teenager is low. This is not the case for the neonate,
adult, or immunosuppressed individual. In these populations, there is a specific
need to prevent VZV infections, as the risk of severe disease is heightened,
thereby presenting the physician with an increasing number of life-threatening
medical situations.
Disease prevention is most effectively accomplished by the prompt use of
VZIG or plasma obtained from immune individuals, such as ZIP. The passive
administration of these immune globulin preparations must be initiated within
72 hr of the exposure of the susceptible individual to a source known to be
actively infected with VZV. In addition, the individual must have a suppressed or
immature immune status as a result of either being a neonate, harboring a cancer
(e.g., child with leukemia), be HIV infected, and so forth. The VZIG and ZIP
products have been shown to modify or prevent the establishment of primary
varicella in susceptible individuals. 72- 74 With prompt administration, these anti-
body preparations will neutralize the VZV prior to its ability to disseminate
throughout the body, thereby preduding a major step in virion pathogenesis
associated with the establishment of a generalized viral infection.
A live virus vaccine has been developed by Takahashi and co-workers. 75 ,76
The specific nature of the attenuation obtained has not been defined at the
genetic level as yet, although the vaccine strain (OKA) has been found to contain
altered nudeotide sequences relative to wild type VZVJ3,77 The OKA strain
vaccine is highly immunogenic, resulting in seroconversion rates of 85-90% of
children receiving a single dose of vaccine. The vaccine seems to be less immuno-
genic in adults, resulting in a protective rate of immunity that is approximately
50%, although, in those not fully protected, the disease was modified. 7B One of
the highly beneficial effects of the OKA vaccine is that it is effective in providing
protection to children with leukemia. In these individuals, because of their
underlying malignancy, they are susceptible to establishing a generalized varicella
infection with an accompanying high mortality rate. The OKA strain, with its
reduced virulence and its inability to establish a massive generalized infection,
stimulates the immune response of these children such that they develop a
protective immunity that either predudes the establishment of infection when
exposed to wild type virus or that results in modified varicella. The net result is
that the mortality rate in OKA-vaccinated leukemic children caused by VZV
infection is dramatically reduced. The immunity that develops has been docu-
mented to last 7-10 years, resulting in either proteetion or modification of disease
over this period. 79 Studies by Gershon et al. BO indicate the immunity that develops
in leukemic children and normal adults was similar, as monitored by tests of
humoral and ceIl-mediated immunity to VZv, but was lower than that which
resulted from a natural infection. Thus, it should be possible to use this vaccine,
once it is licensed in the United States, to immune not only susceptible immuno-
incompetent populations, but also the susceptible normal adult population as
weIl.
Regarding the universal vaccination of children with the OKA vaccine strain,
tests have shown that children as young as 1 to 2 years of age will seroconvert in
response to receiving the vaccine alone Bl whereas other investigators have tested
the varicella vaccine in combination with measles, mumps, and rubella vaccina-
tion. B2 The efficacy of the VZV vaccine has also been tested in normal children
and, in one study, found to be 100% effective in preventing varicella in normal
healthy children who were vaccinated in the age range of 1-14 years. B3 The status
of recommending universal vaccination of a varicella vaccine is controversial at
this point, considering the prospect that the virus may subsequently reactivate,
producing zoster in the vaccinees, as weIl as lingering questions regarding the
total safety of administering a virus that has the ability to establish a latent
infection in the human population. 84 Thus, additional tests need to be performed
to assess the safety of this vaccine candidate prior to its adoption for universal
administration. The usefulness and efficacy of this vaccine strain for use in
specialized immunosuppressed populations, especially leukemic children, is
clear.
In summary, VZV is a DNA virus that has the capability of infecting the
human population, its natural host, and establishing a latent infection that
persists throughout the life of the individual. The primary infection is charac-
terized by a rapidly developing vesicular exanthem as a result of the establish-
ment of primary and secondary viremias during the prodromal incubation and
early phases of the infection respectively. The vesicular exanthem develops
rapidly from a maculopapular state and proceeds to a dried crust, at which time
the lesion is no longer infectious. The virus fixes in sensory nerve endings
innervating the anatomical site bearing the vesicular exanthem. The virus moves
in aretrograde manner to establish latent infection in the sensory ganglion where
it remains dormant. The specific stimuli that reactivate latent VZV are unknown
but seem to be associated with a decrease in specific immunity to the virus or a
generalized decreased immunity in the individual associated with increasing age.
Other nonimmunologic mechanisms are also possible. The reactivated virion
establishes recrudescent disease in the form of a localized infection restricted to
an infected and affected dermatome. Although varicella is a relatively benign
disease in the young child, it can be quite serious in normal adult populations,
causing a severe pneumonia, as weIl as in immunocompromised populations in
which it can also establish generalized infection. In these populations, studies
have shown that aggressive administration of supportive therapy as weIl as the
administration of antivirals including Zovirax and Vidarabine, are effective in
halting the spread of the virus infection in the human host. Similar results have
been found in studies on zoster patients. The advent of newer antiviral chemo-
therapeutic agents is a matter of further testing of promising candidates. Alter-
natively, the rapid administration of VZIG or ZIP to susceptible immuno-
compromised populations can effectively either preclude or dampen a primary
VZV infection, resulting in decreased morbidity and mortality caused by this
virus. The development of the OKA strain of VZV as a vaccine, initiated in Japan
in the mid-1970s and still under test in various populations in the United States
and throughout the world, represents a major breakthrough in further reducing
the morbidity and mortality associated with VZV infections.
REFERENCES
l. Kundratitz, K., 1925, Experimentelle ubertragungen von herpes zoster auf menschen und
die beziehungen von herpes zoster zu varicellen, Z. Kinderheilkd. 39:379-387.
2. Brunsgaard, E.,1932, The mutual relation between zoster and varicella, Br.J Dermatol. Syph.
44:1-21.
3. Weller, T. H., 1953, Serial propagation in vitro of agents producing inclusion bodies derived
from varicella and herpes zoster, Proc. Soc. Sys. Br. Med. 83:340-346.
4. Weller, T. H. and Coons, A. H., 1954, Fluorescent antibody studies with agents of varicella
and herpes zoster propagated in vitro, Proc. Soc. Exp. Biol. Med. 86:789-794.
5. Weller, T. H., Witton, H. M., and Bell, E. J., 1958, The etiologic agents of varicella and herpes
zoster: Isolation, propagation and cultural characteristics in vitro,] Exp. Med. 108:843-868.
6. Straus, S. E., Reinhold, w., Smith, H. A., Ruyechan, W. T., Henderson, D. K. Blaese, R. M.,
and Hay,J., 1984, Endonuclease analysis ofviral DNA from varicella and subsequent zoster
virus infections in the same patient, N. Engi.] Med. 311:1362-1364.
7. Iltis, J. B., Oakes, J. E., Hyman, R. w., and Rapp, F., 1977, Comparison of the DNAs of
varicella-zoster viruses isolated from clinical cases of varicella and herpes zoster, Virology
82:345-352.
8. Dumas, A. M., Geelen, J. L. M. C., Westrate, M. w., Werthein, P., and van der Noordaa, J.,
1981, XbaI, PstI, and BgllI restriction enzyme maps for the two orientation of the varicella-
zoster virus genome,] Virol. 39:390-400.
9. Straus, S. E., Aulakh, H. S., Ruyechan, W. T., Hay, J., Casey, T. A., Vande Woude, G. F.,
Owens, J., and Smith, H. A., 1981, Structure of varicella-zoster virus DNA,] Virol. 40:
516-525.
10. Davison, A. J. and Scott, J. E., 1986, The complete DNA sequence of varicella-zoster virus,
] Gen. Virol. 67:1759-1816.
11. Gelb, L. D., 1990, Varicella-Zoster virus, in: Virology, 2nd ed. (B. Fields, D. Knipe, R.
Chanock, M. Hirsch,J. Melnick, T. Monath, and B. Roizman, eds.) Raven Press, New York,
pp. 2011-2054.
12. Kinchington, P. R., Reinhold, W. C., Casey, T. A., Straus, S. E., Hay,J., and Ruyechan, W. T.,
1985, Inversion and circularization of the varicella-zoster virus genome,] Virol. 56:194-200.
13. Ecker, J. R. and Hyman, R. w., 1981, Varicella-zoster virus vaccine DNA differs from the
parental virus DNA,] Virol. 40:314-318.
14. Vlazny, D. A. and Hyman, R. w., 1985, Errant processing and structural alterations of
genomes present in a varicella-zoster virus vaccine,] Virol. 56:92-10l.
15. Roizman, B., 1979, The structure and isomerization of herpes simplex virus genomes, Cell
16:481-494.
16. Caunt, A. E. and Taylor-Robinson, 0.,1964, Cell-free varicella-zoster virus in tissue culture,
] Hyg. (Lond). 62:413-424.
17. Shaw, D. G., 1968, Laboratory studies on varicella-zoster virus, Ph.D. Thesis, Liverpool
University.
18. Rapp, F. and Benyesh-Melnick, M., 1963, Plaque assay for measurement of cells infected
with zoster virus, Seienee 141:433-434.
19. Meurisse, E. v., 1969, Laboratory studies on the varicella-zoster virus,] Med. Mierobiol.
2:317-325.
20. Grose, C., Friedrichs, W. E., and Smith, K. 0., 1981, Cryoprecipitation of varicella-zoster
virus without loss of structural integrity or infectivity, Intervirology 15:154-160.
21. Rapp, F. and Vanderslice, D., 1964, Spread of zoster virus in human embryonic lung cells
and the inhibitory effect of idoxyuridine, Virology 22:321-330.
22. Vaczi, L., Geder, L., Koller, M., and Veney, E., 1963, Influence of temperature on the
multiplication of varicella virus, Acta Mierobiol. Aead. Sei. Hung. 10:109-115.
23. Ellis, R. w., Keller, P. M., Lowe, R. S., and Zivin, R. A., 1985, Use of a bacterial expression
vector to map the varicella-zoster virus major glycoprotein gene, gC,] Virol. 53:81-88.
24. Keller, P. M., Davison, A. J., Lowe, R. S., Bennett, C. D., and Ellis, R. w., 1986, Identification
and structure of the gene encoding gpII, a major glycoprotein of varicella-zoster virus,
Virology 152:181-19l.
25. Keller, P. M., Davison, A.,J., Lowe, R. S., Riemen, M. w., and Ellis, R. w., 1987, Identification
and sequence of the gene encoding gpIII, a major glycoprotein of varicella-zoster virus,
Virology 157:526-533.
26. Forghani, B., DuPuis, B. w., and Schmidt, N. J., 1984, Varicella-zoster viral glycoproteins
analyzed with monoclonal antibodies,] Virol. 52:55-62.
27. Keller, P. M., Neff, B.]., and ElIis, R. w., 1984, Three major glycoprotein genes ofvaricella-
zaster virus whose products have neutralization epitopes, J Virol. 52:293-297.
28. Vafai, A., Wroblewska, Z., Wellish, M., Green, M., and Gilden, D., 1984, Analysis ofthree late
varicella-zaster virus proteins, a 125,000 molecular-weight protein and small gp 1 and gp 3,J
Virol. 52:953-959.
29. Davison, A. ]., Waters, D. ]., and Edson, C. M., 1985, Identification of the products of a
varicella-zaster virus glycoprotein gene,J Gen. Virol. 66:2237-2242.
30. Weller, T. H., 1979, Varicella-herpes zaster virus, in: Viral Infection of Humans. Epidemiology
and Control. (A. S. Evans, ed.), Plenum Press, New York. pp. 457-480.
31. Preboud, S. R., 1968, Varicella: Complications and costs, Pediatrics 78:728-735.
32. Preboud, S. R., Orestein, W. A., and Bart, K.]., 1984, Varicella: Clinical manifestations,
epidemiology and health impact in children, Pediatr. Infect. Dis. 3:505-509.
33. Hope-Simpson, R. E., 1965, The nature of herpes zaster: A long-term study and a new
hypothesis, Proc. R. Soc. Med. 58:9-20.
34. Jacobson, M. A., Berger, T. G., Fikrig, S., Becherer, P., Moohr,]. w., Stanat, S. C., and Biron,
K. K., 1990, Acyclovir-resistant varicella-zaster virus infection after chronic oral acyclovir
therapy in patients with the acquired immunodeficiency syndrome (AIDS), Ann. Int. Med.
112:187-191.
35. Oxman, M. N., 1981, Varicella and herpes zaster, in: Medical Microbiology and Infectious
Diseases, (A. Braude, C. E. Davis, and]. Fierer, eds.) W. B. Saunders, Philadelphia, pp. 1652-
1671.
36. Hope-Simpson, R. E., 1952, Infectiousness of communicable diseases in the household
(measles, chickenpox and mumps), Lancet 2:549-554.
37. Arvin, A. M., Koropchak, C. M., Williams, B. R. G., Grumet, F. C., and Foung, S. K. H., 1986,
Early immune response in healthy and immunocompromised subjects with primary
varicella-zaster virus infection,J Infect. Dis. 154:422-429.
38. Asano, Y., Itakura, N., Hiroishi, Y., Hirose, S., Nagai, T., Ozaki, T., Yazaki, T., Yamanishi, K.,
and Takahashi, M., 1985, Viremia is present in incubation period in nonimmunocompro-
mised children with varicella,J Pediatr. 106:69-71.
39. Gold, E., 1966, Serologic and virus-isolation studies of patients with varicella or herpes-
zaster infection, N. Engl. J Med. 274:181-185.
40. Armstrong, R. w., Gurwith, M. ]., WaddelI, D., and Merigan, T. C., 1970, Cutaneous
interferon production in patients with Hodgkins disease and other cancers infected with
varicella or vaccinia, N. Engl. J Med. 283:1182-1187.,
41. Cowdry, E. v., 1934, The problem of intranuclear inclusions in virus diseases, Arch. Pathol.
18:527-542.
42. Taylor-Robinson, D. and Caunt, A. E., 1972, Varicella Virus. Springer-Verlag, Berlin.
43. Fitz, R. H. and Meiklejohn, G., 1956, Varicella pneumonia in adults, Am. J Med. Sei. 232:
489-499.
44. Guess, H. A., 1986, Population-based studies ofvaricella complications, Pediatrics 78:723-727.
45. Gilden, D. H., Vafai, A., Shtram, Y., Becker, Y., Devlin, M., and Wellish, M., 1983, Varicella-
zaster virus DNA in human sensory gang1la, Nature 306:478-480.
46. Tenser, R. D. and Hyman, R. w., 1987, Latent herpes virus infections ofneurons in guinea
pigs and humans, YaleJ Biol. Med. 60:159-167.
47. Baadsgaard, 0., Lindskov, R., and Geisler, C., 1987, Reduction ofthe number ofimmuno-
competent cells in the acute stage of herpes zaster, Arch. Dermatol. Res. 279:374-378.
48. RusselI, A. S., Maini, R. A., Baily, M., and Dumonde, D. C., 1973, Cell mediated immunity to
varicella zaster antigen in acute herpes zaster (shingles), Clin. Exp.lmmunol. 14:181-185.
49. Triebwasser, ]. H., Harris, R. E., Bryant, R. E., and Rhoades, E. R., 1967, Varicella
pneumonia in adults: Areport of seven cases and a review of the literature, Medicine 46:
409-423.
50. BrunnelI, P. A., 1983, Fetal and neonatal varicella zoster infections, Semin. Perinatol. 7:47-56.
51. Johnson, R. and Milbourn, P. E., 1970, Central nervous system manifestations of chickenpox,
Gan. Med. Assoc. j 102:831-834.
52. de Moragas, j. M. and Kierland, R. R., 1957, The outcome of patients with herpes zoster,
Arch. Dermatol. 73:193-196.
53. Esmann, v., Geil, j. P., Kroon, S., Fogh, H., Peterslund, N. A., Petersen, C. S., Ronne-
Rasmussen, j. 0., and Danielsen, L., 1987, Prednisolone does not prevent post-herpetic
neuralgia, Lancet 2:126-129.
54. Whitley, R. j., 1984, Varicella-zoster infections, in: Antiviral Agents and Viral Infections olMan,
(Galasso, G., Merigan, T., Buchanan, R., eds.), Rayburn Press, New York, pp. 517-541.
55. Friedman-Kien, A. A., Lafleur, F., Gendler, A. E., Hennessey, N., Montagna, R., Halbert, S.,
Rubenstein, P., Krasinski, K., Zang, E., and Poisey, B., 1986, Herpes zoster: A possible early
clinical sign for development of acquired immunodeficiency syndrome in high risk individ-
uals,j Am. Acad. Dermatol. 14:1023-1028.
56. Weller, T. H., Witton, H. M., and Bell, E.j., 1958, The etiologic agents ofvaricella and herpes
zoster: Isolation, Propagation and cultural characteristics in vitro,] Exp. Med. 108:843-868.
57. Forghani, B., Schmidt, N., and Denis, j., 1978, Antibody assays for varicella zoster virus:
Comparison of enzyme immunoassay with neutralization, immune adherence hemagglu-
tination and complement fixation, j GUn. Microbiol. 8:545-552.
58. Schmidt, N. j. and Emmons, R., 1989, Diagnostic procedures for viral, rickettsial and
chlamydial infections. 6th ed. American Public Health Association, Washington, D.C.,
347-367.
59. Puchhammer-Stockl, E., Popow-Kraupp, T., Heinz, F. X., Mandl, C. w., and Kunz, C., 1991,
Detection of varicella-zoster virus DNA by polymerase chain reaction in the cerebrospinal
fluid of patientssuffering from neurological complications associated with chickenpox or
herpes zoster,j Glin. Microbiol. 29:1513-1516.
60. Landsberger, E., Hager, w., and Grossman, j., 1986, Successful management of varicella
pneumonia complicating pregnancy,j Reprod. Med. 31:311-314.
61. Whitley, R.j., Hilty, M., Haynes, R., Bryson, Y., Connor,]. D., Soong, S.]., and Alford, C. A.,
1982, Vidarabine therapy of varicella in immunosuppressed patients,j Pediatr. 101: 125-131.
62. Prober, C. G., Kirk, L. E., and Keeney, R. E., 1982, Acyclovir therapy of chickenpox in
immunosuppressed children: A collaborative study, j Pediatr. 101:622-625.
63. Arvin, A. M., Kushner,j. H., Feldman, S., Baehner, R. L., Hammond, 0., and Merigan, T. C.,
1982, Human leukocyte interferon for the treatment of varicella in children with cancer, N.
Engl. j Med. 306:761-765.
64. Shepp, 0. H., Dandliker, P. S., and Myers,]. 0., 1986, Treatment of varicella zoster virus
infection in severely immunocompromised patients: A randomized comparison of acyclovir
and vidarabine, N. Engl. j Med. 314:208-212.
65. Huff,j. C., Bean, B., Balfour, H. H.Jr., Laskin, 0. L., Connor,j. 0., Corey, L., Bryson, Y.j.,
and McGuirt, P., 1988, Therapy of herpes zoster with oral acyclovir, Am. j Med. 85:85-89.
66. Morton, P. and Thompson, A. N., 1989, Oral acyclovir in the treatment of herpes zoster in
general practice, N. Z. Med. j 102:93-95.
67. Sakuma, T., Saijo, M., Suzutani, T., Yoshida, 1., Saito, S., Kitagawa, M., Hasegawa, S., and
Azuma, T., 1991, Antiviral activity of oxetanocins against varicella-zoster virus, Antimicrob.
Agents Ghemother. 35:1512-1514.
68. Koszalka, G., Averett, 0., Fyfe,j., Roberts, G. Spector, T., Biron, K., and Krenitsky, T., 1991,
6-N substituted derivatives of adenine arabinoside as selective inhibition of varicella-zoster
virus, Antimicrob. Agents Ghemothl!1: 35:1437-1443.
69. Machida, H. and Watauabe, Y., 1991, Inhibition of DNA synthesis in varicella-zoster virus-
infected cells by BV-araU, Microbiol. Immunol. 35:139-145.
70. Averett, 0. R., Koszalka, G. w., Fyfe, j. A., Purifoy, 0. j., and Krenitsky, T. A., 1991,
6-Methoxypurine arabinoside as a selective and potent inhibitor of varicella-zoster virus,

Antimicrob. Agents Chemother. 35:851-857.
71. Centers for Disease Control, 1984, Varicella zoster immune globulin for the prevention of
chickenpox: Recommendations of the immunizations practices advisory committee, Ann.
Intern. Med. 100:859-865.
72. BrunelI, P. A., Ross, A., Miller, L. H., and Kuo, B., 1969, Prevention of varicella by zoster
immune globulin, N. Engl.] Med. 280:1191-1194.
73. Balfour, H. H., Jr., Groth, K. E., McCullough, j., Kalis, j. M., Marker, S. C., Nesbit, M. E.,
Simmons, R. L., and Najarian,j. S., 1977, Prevention or modification of varicella using zoster
immune plasma, Am. ] Dis. Child 131:693-696.
74. Zaia,j. A., Levin, M. j., Preblud, S. R., Leszczynski,j., Wright, G. G., Ellis, R. j., Curtis, A. C.,
Valerio, M. A., LeGore, j., 1983, Evaluation of varicella-zoster immune globulin: Protection
ofimmunosuppressed children after household exposure to varicella,] Infect. Dis. 147:737-
743.
75. Takahashi, M., Otsuka, T., Okuno, Y., Asano, Y., Yazaki, T., and Isomura, S., 1974, Live
vaccine used to prevent the spread of varicella in children in hospital, Lancet 2:1288-1290.
76. Takahashi, M., Okuno, Y., Otsuka, T., Osame, j., and Takamizawa, A., 1975, Development
of a live attenuated varicella vaccine, Billen] 18:25-33.
77. Ecker,j. R., Kudler, L., and Hyman, R. w., 1984, Variation in the structure ofvaricella zoster
virus DNA, Intervirology 21:25-37.
78. Gershon, A., Steinberg, S., LaRussa, P., Ferrara, A., Hammerschlag, M., Gelb, L., and the
NIAID Varicella Vaccine Collaborative Study Group, 1988, Immunization of healthy adults
with live attenuated varicella vaccine,] Infect. Dis. 158:132-136.
79. Asano, Y., Nagai, T., Miyata, T., Yazaki, T., Ito, S., Yamamishi, K., and Takahashi, M., 1985,
Long-term protective immunity of recipients of the OKA strain of live varicella vaccine,
Pediatrics 75:667-670.
80. Gershon, A., Steinberg, S., and the National Institute of Allergy and Infectious Diseases
Varicella Vaccine Collaborative Study Group, 1990, Live attenuated varicella vaccine: Protec-
tion in healthy adults compared with leukemic children,] Infect. Dis. 161:661-666.
81. Johnson, C., Shurin, P., Fattlir, 0., Rome, L., and Kumar, M., 1988, Live attenuated varicella
vaccine in healthy 12-24 month old children. Pediatrics 81:512-518.
82. Arbeter, A. M., Baker, S., Starr, S., Levine, B., Brooks, E., and Plotkin, S., 1986, Combination
measles, mumps, rubella, and varicella vaccine, Pediatrics 78:742-747.
83. Weibel, R., Neff, B., Kuter, B., Guess, H., Rothenberger, C., Fitzgerald, A., Connor, K.,
McLean, A., Hilleman, M., Buynak, E., and Scolnick, E., 1984, Live attenuated varicella
virus vaccine: Efficacy trial in healthy children, N. Engl.] Med. 310:1409-1415.
84. Starr, S. E., 1989, Status of varicella vaccine for healthy children: Commentary, Pediatrics
84:1097-1099.
Index
Abortion, Coxiella burnetii-related, 163 Acyclovir, as varicella-zoster therapy, 330,

Abscess, anaerobic bacterial 331
cerebral, 117 Adenine arabinoside, as varicella-zoster
dental,117 therapy, 330, 331
intraabdominal, 117 Adenylate cyclase toxin, 150, 151
pulmonary, 113, 114, 115, 116-117, 118, Adult respiratory distress syndrome
122-123, 124, 125 cryptococcal pneumonia-related, 264, 266,
soft-tissue, 117 267
tubo-ovarian, 117 macrophages in, 15
N-Acetylglucosamine, 52 Q fever-related, 164
in coccidioidomycosis, 241 Aerobactin, 88
N-Acetylmuramyl-L-alanine amidase, 30-31, Aerococcus, 52
32 Agglutination, by influenza viruses, 285
Acquired immune deficiency syndrome Agglutinogens, of Bordetella pertussis, 152
(AIDS) Airway
CD4+ T cell deficiency and, 10 conducting, 2
coccidioidomycosis and, 240 host defense mechanisms of
cryptococcal infections and, 249, 251-252, bacterial mucosal adherence, 6
264, 265-267 immunologie defenses, 6
cryptococcal meningitis and, 264, 265 muciliary escalator, 3-6
cytomegalovirus infections and, 154, 266 mucus-secretory epithelium of, 4
histoplasmosis and, 199, 207, 209 Ajellomyces capsulatus, 198
treatment, 211-212 Albumin, of alveolar lining fluid, 13
Legionnaires' disease and, 98 Alcoholism
macrophage-lymphocyte interactions in, 22 as aspiration pneumonia risk factor, 114,
Mycobacterium avium infections and, 141- 116
142 as pneumococcal pneumonia risk factor,
pertussis and, 154-155 38-39
pneumococcal infection and, 40-41 Alpha-l anti trypsin, of alveolar lining fluid,
pneumococcal pneumonia and, 40-41 13
respiratory infections and, 154 Alveolar milieu, host defenses of, 12-20
T cell ratio in, 18 of alveolar lining fluid, 12-13
tuberculosis and, 130 alveolar space cells, 14-20
zoster infections and, 325, 328 complement system components, 14
339
340 INDEX
Alveolar milieu, host defenses of (cont.) Antibiotic therapy (cont.)

immune opsonins, 14 for pertussis, 153
nonimmune opsonins, 13 for psittacosis, 191
Alveoli See also specific antibiotics
bacterial dearance mechanisms of, 20-22 Antibody tests
gas exchange function of, 2 for Histoplasma capsulatum antibody
Amantadine detection, 208-209
as influenza therapy/prophylaxis, 296-298 See also specific antibody tests
as parainfluenza virus infection therapy, Antigenic drift, 286, 287
313 Antigenic shift, 286-287
American Red Cross, 331 Antigen-presenting cell, 9-10, 11-12
Aminoglycosides, of Klebsiella, 87 Antigen tests, for Histoplasma capsulatum, 209
Amoebae, Legionella pneumophila infection Antiprotease, of alveolar maerophages, 16, 17
of, 97, 101 Apical junction, 5
Amoxicillin Aquatic environment, as Legionella
as anaerobic bacterial infection therapy, pneumophila habitat, 97, 98
124 Arthritis, septic
as Haemophilus injluenzae infection therapy, Haemophilus injluenzae-related, 63, 70
76, 78 Q fever-related, 166
as pertussis therapy, 153 Arthroconidia, 240
Amphotericin B immune response to, 241-243
as cryptococcal infection therapy, 263, Aspergillosis, allergie bronchopulmonary, 20
264, 265, 266-267 Aspiration
as histoplasmosis therapy, 211-212 for anaerobic bacteria recovery
effect on polymorphonudear neutrophil transthoracic, 114, 118
yeast phagocytosis, 227 transtracheal, 113, 114-115, 118-119
Ampicillin, as HaemofJhilus injluenzae See also Pneumonia, aspiration
infection therapy, 76, 78 Aspirin, as Reye's syndrome risk factor, 283-
Ampicillin resistance, of Haemophilus 284, 330
injluenzae, 76, 77, 78 Asthma
Anaerobic bacteria histamine-releasing factor in, 289
nasooropharnygeal, 6 parainfluenza virus and, 311
as pulmonary infection causal agents, 113- L-Arginine-dependent nitrogen oxidation
128 system, 255
bacteriology, 119-121 Autolysin, cell wall-associated: see
dinical features, 117-118 N- Aeetylmuramyl-L-amidase
incidence of infections, 114-116 Azithromycin, as anaerobic baeterial
laboratory diagnosis, 118-119 infeetion therapy, 123-124
pathophysiology, 116-117 Azoles, as cryptococeal infeetion therapy,
prognosis, 124-125 264, 265
treatment, 121-124
Anemia, Q fever-related, 166, 169 Bacteremia
Angiotensin-converting enzyme, alveolar Haemophilus injluenzae-related, 71, 75
macrophage-produced, 16, 17 hemolytic streptococeus-related, 55, 56
Animals, as Coxiella burnetii vectors, 161-162, Klebsiella-related, 85, 86, 87, 89, 90
163 Bacteria, nasopharyngeal mucosal adherence
Antibiotic therapy by,6
for anaerobic bacterial pulmonary See also specific bacteria
infections, 121-124 Bacterial superinfection, in influenza, 287-
for chlamydial pneumonia, 193 289
for infant chlamydial pneumonia, 188 Bacteroides, polysaccharide capsule, 117
INDEX 341
Bacteroides fragilis, misclassification, 120 Blastomycosis (cont.)

Bacteroides intermedius, 121, 122 clinical features, 219
Bacteroides melanogenicus, 117, 120, 121, 122 diagnosis, 219-221
Bacteroides spp., penicillin-resistant, 122 epidemiology, 218-219
Bagpipes, cryptococcal contamination of, South American: see
249-250 Paracoccidioidomycosis
Basement membrane, 5 Bordetella bronchisepticum, 151
Basidiospore, of Cryptococcus neofOT11Ulns, Bordetella parapertussis, 151
250, 251 Bordetella pertussis, 149-150
Bat, as Histoplasma capsulatum carrier, 199 cilia binding by, 20
B cell, 7 phase variation, 150
of alveolar lining fluid, 18 virulence factors, 149-152
in bronchus-associated lymphoid tissue, 11 Bordet-Gengou blood agar medium, 150
HIV-related dysfunction, 40 Breast feeding
BCG vaccination, efficacy, 129-130 as Coxiella burnetii transmission method,
Bedsonia, 183 162-163
Benzadine test, 52 as Klebsiella transmission method, 86
Beta-lactam antibiotics Bronchiectasis, 4
Haemophilus injiuenzae resistance to, 76, as anaerobic lung infection risk factor, 116
77, 78 Bronchiole, respiratory, 2
Klebsiella resistance to, 87 Bronchiolitis, parainfluenza virus-related, 311
Beta-lactamase Bronchitis
antibiotic-inactivating activity of, 76, 77 Haemophilus injiuenzae-related, 74
of Haemophilus injiuenzae, 76, 77 as influenza complication, 283
Biopsy Klebsiella-related, 85
bone marrow Bronchoalveolar fluid, in cryptococcal
for histoplasmosis diagnosis, 209 infections, 253
for Q fever diagnosis, 167 Bronchoconstriction, reflex, 3-4
open-lung Bronchopneumonia, interstitial, hemolytic
for anaerobic bacterial infection streptococcal, 55, 58, 59
diagnosis, 114 Bronchoscopy, with quantitative cultures,
for histoplasmosis diagnosis, 209, 211 114, 115, 118, 119
Birds
as Chlamydia carriers, 183, 189 Cancer, as pneumococcal infection risk
as Cryptococcus neoformans carriers, 249, factor, 41
250 Cancer patients, varicella-zoster therapy in,
as Histoplasma capsulatum carriers, 199, 330
201,203 Candida albicans, diploidy, 250-251
Blastomyces antigen, 223-224 Candidiasis, oral, as HIV infection indicator,
delayed hypersensitivity to, 227-228 266
Blastomyces dermatitidis Capillary leak syndrome, 57
mycology of, 217-218 Catalase test, 52
phase transition blockage in, 228-229 CD4+ T cell, 9, 10
polymorphonuclear neutrophil of alveolar lining fluid, 18
interactions, 225-227 in coccidioidomycosis, 243
temperature-dependent dimorphism, 240 in cryptococcal infections, 255-257
Blastomycin, skin testing with, 221 in histoplasmosis, 198
Blastomycosis, 217 in influenza, 288, 292
animal models, 222 in macrophage cytotoxic activity, 133-134
cell-mediated immunity in, 221-222, in paracoccidioidomycosis, 224-225
223-224, 225-230, 231 in pneumococcal infections, 35
342 INDEX
CD8+ T cell Chlamydia

in cryptococcal infections, 256 as intracellular parasite, 100-101, 185-186
in influenza, 288, 292 life cyde, 183, 184-185
in tuberculosis, 133 microbiology, 184-185
CDl6+ T cell, 133 morphological variant forms, 184, 185
Cefazolin, inefficacy as Haemophilus Chlamydia pneumoniae, 183, 185, 186, 191-
injluenwe infection therapy, 76-77 193; see also Pneumonia, chlamydial
Cefixime, as Haemophilus injluenzae infection Chlamydia psittaci, 186
therapy,76 life cyde, 183
Cefpodoxime, as Haemophilus injluenzae serovars, 185
infection therapy, 76 See also Psittacosis
Ceftaxime, as Haemophilus injluenzae Chlamydia trachomatis
infection therapy, 77 biovars, 185
Ceftriaxone, as Haemophilus injluenzae as interstitial pneumonitis causal agent,
infection therapy, 77 186, 187, 188-189
Cefuroxime axefil, as Haemophilus injluenzae mip protein, 105
infection therapy, 76 as neonatal pneumonia causal agent, 183
Cell-mediated immunity Chloramphenicol resistance
in blastomycosis, 221-222, 223-224, 225- of Coxiella bumetii, 160
230, 231 of Haemophilus injluenwe, 77
in coccidioidomycosis, 241-244 Chronic obstructive pulmonary disease
in histoplasmosis, 198-199 chlamydial pneumonia and, 192
in influenza, 292-293 Haemophilus injluenwe infections and, 65,
in Legionella infection, 101-103, 104-105 74-75
in mycobacterial infections, 134 histamine-reIeasing factor in, 289
in paracoccidioidomycosis, 221-223, 224- Legionnaires' disease and, 98
225, 226-227, 230-232, 232 Churg-Strauss syndrome, 20
in varicella-zoster virus infection, 324, 325 Cilia, 4, 5
Cellulitis, Haemophilus injluenwe-related, 63 microoganism binding by, 20
Cephalexdin, inefficacy as Haemophilus Ciliary activity
injluenzae infection therapy, 76-77 in Haemophilus injluenwe infections, 68
Cephalosporin-resistance, of Klebsiella, 87 pneumolysis-related inhibition of, 34, 35
Cephalosporins, as Haemophilus irifluenwe Ciliary dyskinetic syndromes, 4
infection therapy, 76-77 Ciprofloxacin, as Haemophilus injluenwe
Chemiluminescence, of polymorphonudear infection therapy, 76
neutrophils, 225, 226, 227 Cinhosis
Chemotactic factors as cryptococcosis risk factor, 264
of alveolar macrophages, 22 as pneumococcal pneumonia risk factor,
of neutrophils, 16, 18 38, 39
Chickenpox: see Varicella Civil War soldiers, hemolytic pneumonia in,
Children 54
cryptococcosis in, 258 Clara cell, 5
influenza vaccination of, 294, 295 Clarithromycin
Legionnaires' disease in, 98 as anaerobic bacterial pulmonary infection
parainfluenza virus infections in, 310, 311, therapy, 123-124
312-313 as Haemophilus injluenwe pulmonary
Reye's syndrome in, 283-284, 330 infection therapy, 76
varicella-zoster virus infections in, 320, Clindamycin, as anaerobic bacterial lung
323, 324, 325-328 infection therapy, 123, 124
prevention of complications, 331 Clofazamine, 141
vaccination against, 332-333 Coccidioides immitis, 239, 240-241
INDEX 343
Coccidioides immitis (cont.) Corticosteroids

cellular immunity markers, 223 as eryptoeoceosis risk faetor, 257
See also Coccidioidomycosis as Legionnaires' disease risk factor, 102
Coccidioidomycosis Corynebacteria, nasooropharyngeal
asymptomatic, 239 colonization by, 6
disseminated, 239-240, 242-243 Cotrimoxazole, as pertussis therapy, 153
determinants of outcome of, 243-244 Cough
immunology of, 239-247 as airway dearanee meehanism, 3, 4
race factors in, 240 chlamydial penumonia-related, 192
Colchicine, as Haemophilus influenwe histoplasmosis-related, 201
bacteremia inhibitor, 69 influenza-related, 282
Collagenase, alveolar macrophage-produced, pertussis-related, 149, 151, 152-153
16,17 Q fever-related, 164
Complement Coxiella burnetii, 159-163
of alveolar lining fluid, 13 antibiotic resistance, 160
pneumococcal activation of, 32 dassification, 159-160
Complement cascade, pneumococcal intracellular vacuolar replication, 159-160
activation of, 34, 35-37 life cyde, 160-161, 176
Complement component(s), of alveolar morphological variants, 160-161
lining fluid, 14 phase variation, 161
Complement component C2 deficiency, as plasmid, 161
pneumococcal infection risk factor, 38 transmission, 161-163
Complement component C3 virulence factors, 161
in Legionella infection, 101, 103 See also Q fever
in pneumococcal pneumonia, 36, 38 C-polysaccharide: see Polysaccharide,
Complement component C3 deficiency, in capsular
pneumococcal infection, 38, 41 C-reactive protein, 3i, 32, 36
Complement component C3 receptor, Croup, 310, 3Il, 313
cryptococcal binding by, 255, 257 Cryptococcosis, pulmonary, 249-279
Complement component C5, in dinical presentation, 258-267
pneumococcal pneumonia, 36-37 in HIV-infected patients, 265-267
Complement eomponent e5 deficiency, as in immunoeompetent patients, 258-264
cryptococcal infeetion risk faetor, 257 in immunocompromised patients, 264-
Complement fixation test 267
for Blastomyces dermatitidis detection, 220 host defenses in, 253-258
for Chlamydia deteetion, 187, 190 Cryptococcus albidus, 253
for Coxiella burnetii deteetion, 170 Cryptococcus laurenti, 253
for Histoplasma eapsulatum deteetion, Cryptococcus neoformans
208 alveolar maerophage activity against, 16
for Paracoccidioides brasiliensis detection, distribution, 249-250
221 genetics, 252
for parainfluenza viruses detection, 312 host defenses against, 16, 253-258
for varicella-zoster virus detection, 329 isolation, 252-253
Congestive heart disease life eyde, 250-251
as influenza complication, 283 lymph node complex, 254, 260
as pneumococcal pneumonia risk factor, microbiology, 250-253
41 natural reservoir, 250
Conidia, polymorphonudear neutrophil virulenee factors, 252
killing of, 226-227 Cryptococcus neoformans var. gattii, 250, 251-252
Conjunctivitis, chlamydial, 186, 187, 188 Cryptococcus neoformans var. neoformans, 250,
Cooling towers, as Legionella reservoir, 98 251-252
344 INDEX
Cushing's syndrome, as cryptococcosis risk Empyema (cont.)

factor, 257, 264 beta-hemolytic streptococcal pneumonia-
Cyclosporin A, as Legionnaires' disease risk related, 55, 56, 57, 58, 59, 60
factor,98 Haemophilus injluenzae pneumonia-related,
Cystic fibrosis, 4 72
Cytochalsin D, as Haemophilus influenzae pneumococcal, 114, 115
bacteremia inhibitor, 69 Encephalitis
Cytokines influenza-related, 284
as alveolar macrophage enhancers, 22-23 Q fever-related, 166
alveolar macrophage-produced, 16, 17 varicella-related, 328
antimicrobial state induction by, 102-103 Endocarditis
Cytomegalovirus histoplasmosis-related, 211, 212
in AIDS/HIV-infected patients, 154, 266 Q fever-related, 167, 169, 170, 172, 173-
alveolar macrophage activity against, 16 174
Endospore, of Coccidioides immitis, 240-241,
Delayed hypersensitivity, in 242
paracoccidioidomycosis, 224 Endospore-like form, of Coxiella burnetii,
Dendritic cell, 12 160-161
Dental infections, anaerobic bacterial, 116- Endotoxin
117,122 in macrophage-mediated fungistasis, 254,
Diabetes 255
coccidioidomycosis and, 240 See also Lipopolysaccharide
cryptococcosis and, 264 Enterovirus infections, 324
Dialyzable leukocyte extract, 224 Enzyme e1ectrophoresis, multilocus, 64-65
Digoxin,41 Enzyme-Iinked immunosorbent assay
Disseminated intravascular coagulation, (ELISA)
pneumococcal cell wall in, 32 for Blastomyces dermatitidis detection, 220
DNA fingerprinting, of Haemophilus for Coxiella burnetti detection, 170
injluenzae strains, 65 for Histoplasma capsulatum detection, 208
Doxycycline for Paracoccidioides brasiliensis detection,
as chlamydial pneumonia therapy, 193 221
as Q fever therapy, 171 for varicella-zoster virus detection, 329
Dysphagia, as aspiration pneumonia risk Eosinophil, of alveolar Iining fluid, 19-20
factor, 116 Eosinophilia, 19, 20
Eosinophil peroxidase, 19-20
Eicosanoids, 16 Epiglottitis, Haemophilus injluenzae-related,
Elastase 63, 71, 75, 77
alveolar macrophage-produced, 16, 17 Epinephrine, as croup therapy, 313
of Coccidioides immitis, 240-241 Epithelium, ciliated respiratory, 4
polymorphonuclear neutrophil-produced, 19 Erythema nodosum, Q fever-related, 166
Elderly persons Erythromycin
Haemophilus injluenzae pneumonia in, 73 as anaerobic bacterial infection therapy,
influenza complications in, 282 123-124
influenza vaccination of, 294-295 as infant chlamydial pneumonia therapy,
Klebsiella pneumonia in, 86 188
pneumococcal pneumonia in, 41 as pertussis therapy, 153, 155
Elementary body, 184, 185, 186 as psiuacosis therapy, 191
immunofluorescent staining, 187 Erythromycin-resistance, of Coxiella burnetii,
Empyema 160
anaerobic bacterial, 113, 114, 116, 117, 118, Estrogens
124 in coccidioidomycosis, 240
INDEX 345
Estrogens (cont.) Guillain-Barre syndrome (cont.)

in paracoccidioidomycosis, 218 swine influenza vaccine-related, 295
Eucalyptus trees, Cryptococcus neoformans
association with, 250 Haemocin, 66
Extracellular toxic complex, of Klebsiella, 89 Haemophilus, nasooropharyngeal colonization
by,6
Factor B, 14 Haemophilus aegyptius, 63
FcyR+ cell, 133 Haemophilus injluenzae, 63-83
Fibrosis adherence in influenza, 287
histoplasmosis-related, 205 c1inical syndromes, 63, 71-75
paracoccodioidomycosis-related, 223 antibiotic therapy, 76-77
Filobasidiella neoformans, 250 diagnosis, 75, 78
Fimbriae: see Pili immunization against, 70, 71, 75, 77, 78
Fistula, bronchopleural, 117 prevention, 77-78
Fluconazole, as cryptococcal infection colonization by
therapy, 254, 265, 266-267 epidemiology of, 66-67
Flucytosine, as cryptococcal infection pathogenesis of, 67-70
therapy, 263, 265, 267 growth factor regulation, 64
Fluoroquinolones, as Q fever therapy, 171 host immune responses to, 70-71
Forssman antigen, 32 immunoglobulin A of, 20
F protein, 309, 310, 312, 313 lung c1earance of, 69-70
Friedlander's bacillus, 85 microbiology, 64
Fungal infections non-pulmonary infections caused by, 63
hypogammaglobulinemia in, 221-222 phagocytosis of, 20, 22
See also specific types of fungal infections as pneumonia causal agent, 283
Fungi, susceptibility to polymorphonuclear relationship to influenza, 64
neutrophils, 226-227 surface antigens, 65-66
Fungistasis, macrophage-mediated, 254- typing of, 63, 64-66
255, 257 in pneumonia, 72
Fusobacteria, 121 virulence factors, 65-66
Fusobacterium necrophorum, 116, 121 Haemophilus injluenzae biogroup aegyptius, 63
Fusobacterium nucleatum, 120, 121 Heart transplant patients, pneumococcal
Fusospirochetal infection, 121 infection in, 41
Heat-Iabile toxin, 150, 151
Gamella,52 Heat-shock proteins, 135, 136-137
Glucocorticoids, as cryptococcosis risk factor, Hemadsorption inhibition, of parainfluenza
264 viruses, 312
Glycoproteins, of varicella-zoster virus, 322- Hemagglutination inhibition, of
323 parainfluenza viruses, 312
Goblet cell, 4-5 Hemagglutinin
Granuloma of Bordetella pertussis, 149, 150-151, 152
blastomyosis-related,217-218 filamentous, 150-151
coccidioidomycosis-related, 241 of influenza viruses, 284-285, 286, 288, 289
histoplasmosis-related, 198, 205 antibodies, 289-290, 291
paracoccidioidomycosis-related, 218, 222- vaccine and, 293, 294
223 of parainfluenza viruses, 309-310
Q fever-related, 167, 168 Hemophilus injluenzae: see Haemophilus
Granulomatosis, allergie, 20 injluenzae
Guillain-Barre syndrome Hepatitis, Q fever-related, 165-166, 170,
influenza-related, 284 173-174
parainfluenza virus-related, 311 Herpesviruses, varicella-zoster virus as, 321
346 INDEX
Herpes zoster, 324-325 Hyperimmunoglobulin E, 257

in immunocompromised patients, 329 H ypogammaglobulinemia
See also Zoster in fungal infections, 221-222
Histamine, Haemophilus influenzae synthesis in HIV infection, 40
of,68 as pneumococcal infection risk factor, 38, 41
Histoplasma capsulatum, 197-216
cellular immunity markers, 223 Immunization: see Vaccines
culture, 209-211 Immunocompromised patients
epidemiology, 199-201 cryptococcal infections in, 253, 264-265
in HIV infection, 266 herpes zoster in, 329
host defenses against, 198-199 histoplasmosis in, 199, 207, 209, 211
temperature-dependent dimorphism, 240 parainfluenza virus infections in, 311
Histoplasma capsulatum var. dubosii, 198 Q fever in, 164
Histoplasmosis, 197-198 varicella-zoster vaccination of, 332
acute pulmonary, 201, 202-203 varicella-zoster virus infections in, 320,
asymptomatic, 197-198 324, 328
chronic pulmonary, 201, 204-205, 206 therapy, 330-331, 333
clinical syndromes, 201-202 See also Acquired immune deficiency
diagnosis, 207-211, 220 syndrome (AIDS); Human immune
disseminated, 197, 199,207,209 deficiency virus (HIV)-infected
diagnosis, 209, 210 patients
treatment, 211-212 Immunodiffusion precipitin test
epidemiology, 199-201 for blastomycosis diagnosis, 220
treatment, 211-212 for histoplasmosis diagnosis, 208
HLA-DR, in tuberculosis, 131, 133 for paracoccidioidomycosis diagnosis, 221
HN protein, 309-310, 312, 313 Immunofluorescent assays, for varicella-
Horder's spots, 189, 194 zoster virus detection, 329, 330
Hospital staff, Klebsiella transmission by, 86- Immunoglobulin(s), of respiratory tract, 7-9
87 of alveolar lining fluid, 13
Human immune deficiency virus (HIV)- Immunoglobulin A
infected patients of alveolar lining fluid, 14
blastomycosis in, 222 in influenza, 290-291, 292
cryptococcal infections in, 257 in parainfluenza infections, 312
Haemophilus influenzae pneumonia in, 73- in pneumococcal-related complement
74, 78 activation, 36
herpes zoster in, 329 as pneumococci adherence inhibitor, 34-35
influenza vaccination of, 294, 295 of respiratory tract, 7, 9
mononuclear phagocyte anticryptococcal Immunoglobulin A protease, 20, 66
activity in, 255-256 Immunoglobulin E
paracoccidioidomycosis in, 222 mast cell affinity, 7
pneumococcal infection in, 40-41 of respiratory tract, 7, 9
varicella-zoster virus infections in, 320 Immunoglobulin G
Humidifiers, Legionnaires' disease in alcoholic liver disease, 38
transmissiom via, 98 of alveolar lining fluid, 14
Humoral immunity in chlamydial infections, 187
in influenza, 291 in influenza, 290-291, 292
in Legionella infection, 101 in Legionella infections, 10 1
Hydrogen peroxide in phagocytosis, 20, 22
alveolar macrophage-produced, 18 in pneumococcal-activated complement
polymorphonuclear neutrophil-produced, cascade,35
19 of res pi ra tory tract, 7, 9
INDEX 347
Immunoglobulin G (cont.) Influenza C virus, genetic stability, 287

structure, 7, 8 Influenza C virus infections, symptoms of,
Immunoglobulin M 283
in chlamydial infections, 186-187 Influenza epidemics, 281, 282
in influenza, 290 natural history, 282
in Legionella infections, 10 1 seasonality, 282
in pneumococcal-activated complement Influenza pandemics, 281, 286-287
cascade,35 Influenza viruses, 281-308
structure, 7 antigene variation, 286-287
in varicella-zoster virus infections, 329 host immune responses, 289-293
Immunosuppression pathogenic mechanisms, 285-289
as Legionnaires' disease risk factor, 98 prevention, 293-296
See also Acquired immune deficiency structure and function, 284-285
syndrome (AIDS); Human immune implications for pandemics, 282
deficiency virus (HIV)-infected therapy, 296-298
patients See also Influenza A virus; Influenza B
Inappropriate secretion of antidiuretic virus; Influenza C virus
hormone, 169 Interferon-IX
IncJusion staining, of Chlamydia, 187 as Chlamydia inhibitor, 186
Indirect fluorescent antibody assay, of in influenza, 288
Coxiella burnetii, 170 Interferon-ß, as Chlamydia inhibitor, 186
Infants Interferon-'Y
chlamydial pneumonia in, 183, 187-189 in blastomycosis, 228
Haemophilw; irifluenzae pneumonia in, 73, 78 as Chlamydia inhibitor, 186
influenza complications in, 282 as Coxiella burnetti inhibitor, 173
Klebsiella colonization in, 86 endospore-inhibiting activity, 242
parainfluenza virus infections in, 310 in influenza, 288
varicella in, 324 in Legionella infection, 102-103
Inflammatory response, to Cryptococcw; in macrophage-mediated fungistasis, 254-
neoformans, 256, 257 255, 257
Influenza in mycobacterial infections, 133-134, 141-
complications, 281, 283-284 142
vaccine-related prevention of, 295 in paracoccidioidomycosis, 231
epidemiology, 282 Interleukin-l, 102
Haemophilw; injluenzae relationship, 64, 68 bacterial polysaccharide-induced, 138
histopathology, 286 in mycobacterial infections, 138-139, 140
host immune responses, 287-293 pneumococcal, 31
as pneumococcal pneumonia risk factor, 39 in tuberculosis, 131, 133
transmission, 285-286 Interleukin-2, 102, 103
uncomplicated, 282-283 induction of, 138
Influenza A virus in tuberculosis, 130-131, 132
antigenie variation, 287 Intravenous immune globulin, Klebsiella, 91-
antiviral therapy, 296, 297, 298 93
delayed hypersensitivity response, 293 Itraconazole
vaccine, 293, 293, 294 as cryptococcal infection therapy, 264
Influenza A virus infections, symptoms, 282 as histoplasmosis therapy, 211
Influenza B virus, 283-284
antiviral therapy, 298 Job's syndrome, 257
genetic stability, 287
Influenza B virus infections, symptoms of, Kala azar, 243
282-283 Kartagener's syndrome, 4
348 INDEX
Ketoconazole Lipopolysaccharide
as cryptococcal infection therapy, 264, 265 of Chlamydial, 185
as histoplasmosis therapy, 211, 212 of Coxiella bumetii, 161
Klebs, Edwin, 85 as Q fever vaccine, 174
Klebsiella, 85, 96 of Haemophilus injluenzae, 65, 67, 70
carriage rates, 86 antibodies to, 71
dinical significance, 85-87 as interleukin-l inducing agents, 138-139
host immunity to, 89-90 of Klebsiella, 88, 89
virulence factors, 88-90 as vaccine contaminant, 90-91
Klebsiella infections pertussis: see Pertussis toxin
nosocomial, 86 Lipoteichoic acid, 52, 53
predisposing factors, 86 Listeria monocytogenes, alveolar macrophage
Klebsiella oxytoca, 85 activity against, 16
Klebsiella ouumae, 85 Liver cirrhosis, as pneumococcal pneumonia
Klebsiella pneumoniae, 85 risk factor, 39
phagocytosis of, 20, 22 Liver disease, as pneumococcal pneumonia
Klebsiella rhinoscleromatis, 85 risk factor, 38-39
Koch-Weeks bacillus, 63 Livestock, as Coxiella bumetii vector, 161-162,
Kupffer cell, 15 163
Lymph node complex, cryptococcal, 254,
Lamina propria, 5 260
Lancefield precipitin test, 52 Lymphocyte
Langerhans cell, 15 of alveolar lining fluid, 18
Laryngotracheobronchitis: see Croup in blastomycosis, 223-224
Larynx, as upper airway structure, 1 in influenza, 288
Legiolysin, 105 in paracoccidioidomycosis, 224
LegioneUaceae, 97 of respiratory tree, 7-12
Legionella pneumophila, 97-111 Lymphogranuloma venereulis biovar, of
alveolar macrophage activity against, 16 Chlamydia, 185
aquatic reservoir, 97, 98 Lymphoid tissue
differential diagnosis, 155 bronchus-associated, 10-11
host immune defenses against, 22, 100- in respiratory tree, 10-11
103 Lymphoma, as cryptococcosis risk factor, 264
as intracellular pathogen, 97, 100-101 Lymphoma patients, varicella-zoster virus
macrophage-lymphocyte response to, 22 infections in, 325
transmission, 98-100 Lysozyme, 16, 17
vaccine, 105-106 anti-pneumococcal activity, 34
virulence factors, 103-105
Legionnaires' disease, 97 Macrophage
differentiated from psittacosis, 190, 191 of airway, 11-12
in immunocompromised patients, 101 alveolar, 14-18
risk factors, 98 Blastomyces dermatitidis interactions, 228-
Leishmania major, 243 230,231
Leishmaniasis, 243 in cryptococcal infections, 253-256
Lemierre syndrome, 116 cytokine-enhanced activity 0[, 22-23
Leuconostoc, 52 Histoplasma capsulatum phagocytosis by,
Leukemia, as cryptococcosis risk factor, 264 198
Leukemia patients, varicella-zoster virus in HIV-seropositive patients, 256
vaccination of, 332 Legionella replication in, 102
Leukotrienes, in pneumococcai pneumonia, 37 Legionella phagocytosis of, 99, 100-101,
Lipoarabinomannan, 138, 140 103
INDEX 349
Macrophage (cont.) Mip protein, 105

alveolar (cont.) Monocyte
Paracoccidioides brasiliensis interaction, Blastomyces dermatitidis interactions, 229
231 Legionella infection of, 100-101
secretory products 0[, 16, 17 in tuberculosis, 130, 131-133
arthocondia phagocytosis by, 242 Monocyte-activating substances, in
of bronchus-associated lymphoid tissue, 11 mycobacterial infections, 138-141
in influenza, 288 Moraxella catarrhalis, nasooropharyngeal
Klebsiella lectinophagocytosis, mediation colonization by, 6
by,88-89 Mucociliary cIearance, 3-6
mycobacterial replication in, 133 in influenza, 287
Paracoccidioides brasiliensis interaction, 231- Mucosal immunity, in influenza, 290-291,
232 295
peritoneal Mucous gland, bronchial, 5-6
Blastomyces dermatitidis interaction, 227- Mycobacteria
228 intracellular survival mechanisms, 100-101
Paracoccidioides brasiliensis interaction , mononucIear phagocyte interactions of,
230-231 133-134
pulmonary intravascular, 15 Mycobacterial infection
Major histocompatibility complex, 9-10 'V8 T-cell system in, 135, 137
Major secretory pro tein, 104-106 See also Tuberculosis
Malignant stitch, 54 Mycobacterial proteins, 135, 138
Mannose-inhibitable adherence phenotype, mononucIear phagocyte-activating, 138-
89 141
Matrix (M) proteins Mycobacterium avium, 141-142
antiviral agent interactions, 296 in HIV-infected patients, 266
of influenza viruses, 284, 285, 288, 289 monocIonal antibodies, 138
antibodies, 290 Mycobacterium intracellulare, 141
of parainfluenza viruses, 309 Mycobacterium leprae, antigens, 136
as streptococcal cell wall component, 52, Mycobacterium tuberculosis
53 alveolar macrophage activity against, 16
Meiosis, by Cryptococcus neoformans, 250-251 antigens, 135-137
Meningitis filtrate, blastogenic response induction by,
cryptococcal, 255-256, 259, 264, 265 139-141
therapy, 260, 263-264, 266 intracellular replication, 130
Haemophilus injluenzae-related, 63, 70, 71, macrophage-lymphocyte response to, 22
77 monocIonal antibodies, 137-138
Klebsiella-related, 85 mononucIear phagocyte
parainfluenza virus-related, 311 immunoregulatory properties, 130-
pneumococcal cell-wall components in, 31 133
Meningoencephalitis prevalence of infection by, 129
influenza vaccine-related, 295 with pulmonary cryptococcosis, 259
Q fever-related, 166 reactivation of infection with, 130
Microagglutination assay, of Coxiella burnetii, See also Tuberculosis
170 Mycoplasma, cilia interactions, 4, 20
Microglial cell, 15 Myelitis, influenza-related, 284
Microimmunofluorescence, for Chlamydia Myocarditis, as influenza complication, 283
detection, 187, 188 Myositis, as influenza complication, 283
Migration inhibitory factor, 22, 23
MinocycIine, as chlamydial pneumonia Nasopharynx
therapy, 193 anatomy,1
350 INDEX
Nasopharynx (cont.) Organ transplantation

Streptococcus pneumoniae adherence in, 34- as blastomycosis risk factor, 222
35 as cryptococcosis risk factor, 264
Natural killer cell as Legionnaires' disease risk factor, 98
anticryptococcal activity, 256, 257 as paracoccidioidomycosis risk factor, 222
in coccidioidomycosis, 242 Ornithosis, 183
in influenza, 288, 292-293 Orthomyxoviruses, 281
in paracoccidioidomycosis, 225 Osteolcast, 15
Natural killer-like cell, in Legionnaire's Otitis media, Haemophilus injluenzae-related,
disease, 103 63,71,77
Necrosis, pulmonary, anaerobic bacterial, Outer-membrane proteins
116-117 of Bordetella pertussis, 152
Neisseria meningitides, immunoglobulin A 0[, 20 of Haemophilus injluenzae, 65-66
Neisseria meningitides infection, influenza- of Klebsiella, 89
related, 283 of Legionella, 101
Nephrotic syndrome, idiopathic, 41
Neuralgia, postherpetic, 331 P6 protein, 65, 78
Neuraminidase Paracoccidioides brasiliensis, 221
of influenza viruses, 285, 286, 288, 289 macrophage interactions, 230-232
antibodies, 290 mycology 0[,217-218
of parainfluenza viruses, 309-310 Paracoccidioidin, skin testing with, 221
pneumococcal, 34 Paracoccidioidomycosis, 217
Neurologie disorders, Q fever-related, 166 animal m~dels, 222-223
Neutropenia, as Legionnaires' disease risk cell-mediated immunity in, 221, 223, 224-
factor,98 225, 226-227, 230-232
Neutrophil: see Polymorphonuclear clinical features, 219
neutrophil diagnosis, 219-221
Nitrogen oxidation, L-arginine-dependent, geographicallocation, 219
255 hormonal factors in, 231
Nose, particulate filtration by, 3 subclinical, 231
Nosocomial transmission Parainfluenza viruses, 309-318
of aspiration pneumonia, 115-116 clinical manifestations, 311
of Klebsiella, 86 diagnosis, 311-312
of parainfluenza viruses, 310 epidemics,310
prevention of, 313-314 epidemiology, 310
Nucleoprotein, of influenza viruses, 285 immune responses to, 312
antibodies to, 290 prevention, 313-314
Nucleoside analogs, as varicella-zoster treatment, 312-313
therapy, 330, 331 vaccine, 313
Nursing horne patients, Klebsiella pneumonia virology, 309-310
in, 86 Paramyxoviridiae, 309
Paramyxovirus, 309
Obstruction, pulmonary, as anaerobic lung Particulate matter, airway filtration of, 3
infection risk factor, 116 Pedicoccus, 52
Ophthalmologie disorders, Q fever-related, Penicillin, as anaerobic bacterial infection
166 therapy, 122, 123
Opsonin Penicillin resistance, of Bacteroides species, 122
immune, 14 Peptostreptococcus, 120, 121
nonimmune, 13 Pericarditis
Opsonization, of Streptococcus pneumoniae, histoplasmosis-related, 205, 207, 212
35-37 influenza-related, 283
INDEX 351
Periodontal disease, as anaerobie baeteria Pneumonia (cont.)

souree, 116-117 beta-hemolytic streptococcal (cont. )
Pertaetin, 152 clinical manifestations, 59-60
Pertussis, 149-150 complications, 60
AIDS and, 154-155 epidemiology, 57-58
clinieal manifestations, 152-153 pathology, 58-59
mortality rate, l49, 153-154 primary,57
vaeeines, 153-154 secondary, 57
Pertussis toxin, 149, 150-151, 152 treatment, 60-61
Phagoeyte, mononuclear, in HIV infeetion, blastomycosis-related, 219
255-256 chlamydial, 183-196
Phagoeytosis clinical syndromes, 187-193, 194
by alveolar maerophages, 16 infant, 183, 187-189
Chlamydia-indueed, 184 interstitial pneumonitis and, 186
"eoiling," of Legionella, 99, 100 pathophysiology, 185-187
matrix protein-related resistanee to, 53 cryptococcal, 253
by polymorphonuclear neutrophils, 19 in AIDS patients, 265-267
Phenoloxidase, as eryptocoeeal virulenee allergie, 259
faetor, 252 clinical features, 259, 260, 265-266
Pili in immunocompromised patients, 264-
of Haemophilus injluenzae, 65-66 267
Klebsiella type-I, 89 radiographie evaluation, 260, 261-263,
in mueus adherenee, 68 265, 266
Plasma membrane, pneumoeoeeal, 32 therapy, 260, 263-264, 265, 266-267
Platelet-aetivating faetor, pneumoeoeeal eell eosinophilic, 20
wall-indueed, 32 Haemophilus injluenzae-related, 63, 70-74
Pleural effusion, cryptocoeeal infeetion- bacteremic, 69,72, 73
related, 260 in children, 69, 70, 71-74, 77
Pneumococei, metabolie requirements, 30 in developing countries, 72
Pneumocystis carinii in HIV-infected individuals, 73-74, 78
in AIDS/HIV-infeeted patients, 154, 266 immunity against, 77
alveolar maerophage aetivity against, 16 pathogenesis, 70
maerophage-lymphoeyte response to, 22 T cell ratio in, 18
Pneumolysin, 33-34, 36 influenza-related, 283, 286
Pneumonia Klebsiella-related, 85-96
aspiration, anaerobie baeteria-related, 113 clinical signifieance, 85-87
baeteriology, 119-121 epidemiology, 85-87
clinieal features, 117-118 mortality rate, 87
ineidenee, 114-116 pathogenesis, 87-88
laboratory diagnosis, 118-119 Legionella-related, 100
nosocomial, 115-116 mycoplasmal, differentiated from
pathophysiology, 116-117 chlamydial pneumonia, 192-193
prognosis, 124-125 nosocomial, 86, 115-116
treatment, 121-124 pertussis-related, 149, 153
baeterial pneumococcal, 29-49
in AIDS patients, 154 causal organism, 29-31
differentiated from psittaeosis, 190 epidemiology, 29-30
as influenza complieation, 283 immune defense, 34-37
See also speeific types of bacterial inflammatory response 0[, 31-32, 33
pneumonia predisposing disease states, 37-41
beta-hemolytic streptococcal, 54-61 Pneumocystis carinii-related, 22, 154, 266
352 INDEX
Pneumonia (cont.) Protease

postobstructive, 116 of Legionella, 104-lO5
Q fever-related, 164-166, 167 polymorphonuclear neutrophil-produced, 19
streptococcal, with Haemophilus inftuenwe Protozoa, Legionella pneumophila infection of,
pneumonia, 72 97,101, lO3-104
varicella-related, 324, 325, 328, 330, 333 Pseudomonas aeruginosa
Pneumonitis ciliary-inhibiting activity of, 4
anaerobic bacterial, 116-117 immunoglobulin A of, 20
aspiration pneumonia-related, 117, 118 impaired pulmonary clearance of, 14
chlamydial, 186, 187, 188, 191 phagocytosis of, 20, 22
coccidioidomycosis-related, 239 Psittacosis, 183, 189-191
histoplasmosis-related, 201 Purified protein derivatives (PPD), as
therapy, 124 interleukin inducing agent, 130, 131,
varicella, 330 132, 138-139
Pneumonitis biovar, of Chlamydia, 185 Puromycin, 227
Polymorphonuclear neutrophil
of alveolar lining fluid, 18-19 Q fever, 159-182
alveolar macrophage interactions, 16, 18 acute, 164-167, 168
arthroconidia phagocytosis by, 242 causal agent, 159-163
in blastomycosis, 218, 231 chronic, 167, 169, 175-176
capsular polysaccharide-related inhibition clinical syndromes, 163
of,33 diagnosis, 169-171
chemiluminescence, 225, 226, 227 differentiated from psittacosis, 190, 191
chemotactic factors of, 16, 18 immunology, 172-174
in coccidioidomycosis, 241 therapy, 171-172
in cryptococcosis, 256 transmission, 161-163, 176
endospore phagocytosis by, 242 vaccines, 174-175, 176
immune response of, 19 Query fever: see Q fever
influenza virus infection of, 287-288
macrophage interactions, 22 Radioimmunoassay
in meningitis, 31 for blastomycosis diagnosis, 220
in paracoccidioidomycosis, 218, 225-227, for histoplasmosis diagnosis, 208
231 Renal failure patients, influenza vaccination
Polysaccharide, capsular of,294
anaerobic bacterial, 117 Respiration, process of, 1-3
cryptococcal, 251, 252 Respiratory tract
carbon dioxide response, 254 host immune defenses, 1-27
as complement activator, 257 airway defenses, 3-6
of HMmophilus inftuenwe, 65, 75 in alveolar milieu, 12-20
of Klebsiella, 85, 88-90 integrated response of, 20-22
as Klebsiella vaccine, 90-93 lymphocytes, 2, 7-12
pneumococcal, 31, 32, 33, 35, 36 primary function, 1
Pregnancy Respiratory tract infections
chlamydial pneumonia therapy during, in AIDS patients, 154
193 as HMmophilus inftuenwe potentiators, 68-
coccidioidomycosis during, 240 69, 70, 73
psittacosis therapy during, 191 Klebsiella-related, 86, 89
Q fever during, 163, 169 Respiratory tree, anatomy, 1
varicella pneumonia during, 328 Reticulate body, 184-185
varicella-zoster therapy during, 330 Reticuloendothelial system
Progesterone, in coccidioidomycosis, 240 in histoplasmosis, 198-199
INDEX 353
Reticuloendothelial system (cont.) Smoking (cont.)

in liver cirrhosis, 40 as Legionnaires' disease risk factor, 98
Reye's syndrome Sneezing, as nasal clearance mechanism, 3-4
aspirin-related, 283-284, 330 Spherule, of Coccioides immitis, 240, 241, 242
parainfluenza virus-related, 311 Spirochete, anaerobic, 121
R-factors, 87 Splenectomy, as pneumococcal infection risk
Rhamnose, as streptococcal cell wall factor, 39-40
component, 52 Sputum
Rheumatic fever, beta-hemolytic "currantjelly," 87-88
streptococcal, 57 putrid, in anaerobic aspiration
Ribavirin pneumonia, 118, 122
as influenza therapy, 298 Sputum culture, of Histoplasma capsulatum,
as parainfluenza virus infection therapy, 313 209,212
Rickettsiaceae, Coxiella burnetii classification Staphylococcus, nasooropharyngeal
as, 159-160 colonization by, 6
Rifabutin, 141 Staphylococcus aureus
Rifampin adherence in influenza, 287
as Haemophilus influenzae infection therapy, phagocytosis of, 20, 22
77,78 as pneumonia causa! agent, 283
as Q fever therapy, 171-172 "Steeple sign," 311
Rimantadine Streptococcaceae, 51-52
as influenza therapy/prophylaxis, 296-297 Streptococcus
as parainfluenza virus infection therapy, anaerobic, reclassification, 120, 121
313 beta-hemolytic; see also Pneumonia, beta-
RNA polymerase, in influenza, 284, 285, 289 hemolytic streptococcal
R proteins, 52 differentation, 52
Lancefield group A, 56-57
Sarcoidosis classification, 52
angiotensin-converting enzyme in, 16 nasooropharyngeal colonization by, 6
as cryptococcosis risk factor, 264 Streptococcus constellatus, 121
T cell ratio in, 18 Streptococcus intermedius, 121
Sarcoma, cryptococcosis and, 258 Streptococcus morbillorum, 121
Serous cell, 5 Streptococcus parvulus, 121
Shingles: see Zoster Streptococcus pneumoniae
Shock, septic capsular polysaccharide, 33
Klebsiella-related, 88 cell envelope, 30-31
See also Toxic shock syndrome cell wall, 31-32,42
Showering, as Legionnaires' disease as complement activator, 36
transmissiom method, 98 cilia inhibitory activity, 20
Shwartzmann reaction, 152 cytoplasmic factors, 33-34
Siayllactosykeramide, 67 host immune responses, 34-37,42
Sickle cell disease impaired pulmonary clearance of, 14
as cryptococcosis risk factor, 264 inflammatory response, 42
as pneumococcal infection risk factor, 40 in influenza, 287
Skin tests, for Histoplasma capsulatum nasopharygneal adherence, 34-35
detection, 207-208 oropharyngeal colonization by, 29-30
Smallpox, 319 phagocytosis 0[, 20, 22
Smokers, angiotensin-converting enzyme in, plasma membrane, 32
16 as pneumococcal pneumonia causal
Smoking organism, 29-31
effect on dendritic cell activity, 12 with influenza pneumonia, 283
354 INDEX
Streptococcus pyogenes T proteins, 52

antibiotic susceptibility, 61 Trachea, 2
cellular structure, 52-53 Tracheal cytotoxin, 151-152
See also Pneumonia, beta-hemolytic Tracheitis, Haemophilus injluenzae-related, 74
streptococcal Tracheobronchitis
Streptolysin, 33, 36 Haemophilus injluenzae-related, 74-75
Streptolysin 0, 53 pertussis-related, 149
Streptolysin S, 53 Trachoma biovar, of Chlamyditl, 185
Subarachnoid space, cryptococcal infections Transferrin
of, 250, 253 of alveolar lining fluid, 13
Sudden infant death syndrome, 169 as Legionella growth inhibitor, 102
Sulfonamides Transforming growth factor-b, in
as hemolytie streptococcal pneumonia tuberculosis, 133
therapy,56 Tuberculosis
as infant chlamydial pneumonia therapy, in AIDS patients, 154
188 BCG vaccination, 129-130
Superoxide, 19 host immune response, 130-133
Surface protein A, pneumococcal, 35 in immunocompromised individuals, 130,
Surfactant, of alveolar lining fluid, 13 154
Syncytial virus, respiratory, dfferential monoclonal antibodies in, 137-138
diagnosis, 188 mononuclear phagocyte-mycobacterial
Synergy, microbial, 117 interactions in, 133-134
as mortality cause, 129
T cell, 9-10 negative tuberculin skin-tests in, 130
alveolar, 7, 18 reactivation, 130, 137
antigen presentation to, 11-13 suppressor lymphocytes in, 133
antigen recognition by, 9-10 vitamin D synthesis in, 134
in bronchus-associated lymphoid tissue, 11 Tumor necrosis factor
helper: see CD4+ T cell induction of, 138
in Q fever, 173, 176 in influenza, 288
suppressor: see CD8+ T cell in Legionella infections, 102, 103
'Yll, 135, 137 in mycobacterial infections, 134, 139, 140-
Teichoic acid, 31, 32 141
Tetracycline in paracoccidioidomycosis, 225
as chlamydial pneumonia therapy, 193 in tuberculosis, 133, 140-141
as psittacosis therapy, 191 Tumor necrosis factor-a, endospore-
as Q fever therapy, 171-172 inhibiting activity, 242
ThlITh2 cells, in coccidioidomycosis, 243- TWAR strain, of Chlamydia, 183, 191, 194
244 Tzanck smear, 330
Thiomycin-resistance, of Coxiella burnetii,
160 Upper airway, function, 1-2
Thoracentesis, 114 Urinary antigen assay, for Haemophilus
Thoracotomy, for anaerobie bacterial species injluenzae detection, 75
recovery, 118 Urinary tract infection, Klebsiella-related, 85,
Thymus, in paracoccidioidomycosis, 222 86
Tick, as Coxiella burnetii vector, 159, 162
Toxie shock syndrome, 57-58, 59 Vaccines
Toxoplasma, intracellular survival diphtheria, pertussis, tetanus (DPT), 154
mechanisms,100-101 Haemophilus injluenzae, 70, 71, 75
Toxoplasma gondii, alveolar macrophage influenza, 291-292, 293-296
activity against, 16 Klebisiella capsular polysaccharide, 90-93
INDEX 355
Legionella pneumoniae, 105-106 Varieella-zoster virus infections (cont. )

parainfiuenza virus, 313 therapy, 330-331
pertussis, 153-154 Variola, 319
Pseudomonas aeruginosa, 92, 93 Virulence factors
Q fever, 174-175, 176 of Bordetella pertussis, 149-152
varicella-zoster virus, 332-333 of Coccidioides immitis, 240-241
Varicella of Cryptococcus neoformans, 252
in children, 323 ofHaemophilus influenwe, 65-66
clinieal course, 325-328 of Klebsiella, 88-90
diagnosis, 329 of Legionella pneumophila, 103-105
epidemies, 320 of Streptococcus pyogenes, 52-53
epidemiology, 319 Vitamin D, as mycobacterial inhibitor, 134
pathogenesis, 323-324
relationship to zoster, 319-320, 323 Whooping cough: see Pertussis
Varicella-zoster immune globulin, 331-332, WI-l protein, 220, 223-224
333 Wound infections, Klebsiella-related, 85
Varicella-zoster virus, 319-337
host immune responses to, 322-323 Yeast, thermally dimorphie, 217-218
physical-chemical properties, 320-322
reactivation, 333 Zidovudine, 40
replieation, 322-323, 324 Zoster
vaccine, 332-333 clinical course, 328-329
Varicella-zoster virus infections epidemiology, 323
congenital, 328 in immunocompromised individuals, 330-
diagnosis, 329-330 331
epidemiology, 323 pathogenesis, 324-325
neonatal, 328 relationship to varicella, 319-320, 323
pathogenesis, 323-329 Zoster immune plasma, 331-332, 333
prevention, 331-333

(Infectious Agents and Pathogenesis) Ramon G. Canto, George R. Robinson II (Auth.), Herman Chmel, Mauro Bendinelli, Herman Friedman (Eds.) - Pulmonary Infections and Immunity-Springer US (1994)

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

(Infectious Agents and Pathogenesis) Ramon G. Canto, George R. Robinson II (Auth.), Herman Chmel, Mauro Bendinelli, Herman Friedman (Eds.) - Pulmonary Infections and Immunity-Springer US (1994)

Uploaded by

Copyright:

Available Formats

Pulmonary Infections

FUNGAL INFECTIONS AND IMMUNE RESPONSES

NEUROPATHOGENIC VIRUSES AND IMMUNITY

PSEUDOMONAS AERUGINOSA AS AN OPPORTUNISTIC PATHOGEN

PULMONARY INFECTIONS AND IMMUNITY

Springer Science+Business Media, LLC

Pulmonary Infections and Immunity I edited by Herman Chmel, Mauro

1. Respiratory Infectlons--Immunologlcal aspects. I. Chmel,

ISBN 978-1-4899-1065-3 ISBN 978-1-4899-1063-9 (eBook)

© 1994 Springer Science+Business Media New York

JOHN G. BARTLETT • Department of Medicine, Johns Hopkins University

ANN W. FUNKHOUSER • Laboratory of Infectious Disease, National Insti-

JANET E. STOUT • Special Pathogens Laboratory, Pittsburgh, Pennsylvania

The mechanisms of disease production by infectious agents are presently the

It is widely accepted that most microorganisms do not cause infection of the

appropriate control measures for respiratory infections. In this regard, exciting

1. Defense Meehanisms of the Respiratory Traet

2. The Immunology of Pneumococcal Pneumonia

3. Pulmonary Infections Caused by Lancefield Group A Beta-

4. Pulmonary Infections Caused by Haemophilus inftuenzae

1. Etiological Agent .............................................. 85

1. The Disease .................................................. 97

6. Vaccine Deve10pment .......................................... 105

7. Anaerobic Bacterial Infections of the Lung

1. Introduction .................................................. 113

8. Immunology of M. tuberculosis and Other Mycobacteria

2.2. Filamentous Hemagglutinin ............................... 150

10. Coxiella burnetii and Q Fever

11. The Chlamydial Pneumonias

12. Histoplasma capsulatum

1. Introduction ................................................. 197

13. Blastomyces dermatitidis and Paracoccidioides brasiliensis

1. Introduction ................................................. 217

14. The Immunology of Coccidioidomycosis

1. Introduction ................................................. 239

15. Pulmonary Cryptococcosis: Pathophysiological and Clinical

1. Introduction ................................................. 249

5. Conclusion .................................................. 267

16. Influenza Viruses

17. Parainfluenza Viruses

18. Varicella-Zoster Virus

1. Introduction ................................................. 319

RAMON G. CANTO and GEORGE R. ROBINSON II • Pulmonary/Critical Care Division; and

The respiratory tract contains a complex series of mechanisms that defend

2.1. Aerodynamic Defenses and the Mucociliary Escalator

the nasal passageways of secretions and trapped matter. Reflex bronchoconstric-

' - - - r , f - - - - - inner dynein arm

2.2. Bacterial Mueosal Adherenee

2.3. Immunologie Defense of the Airways

Complement Binding Region

3.3. Lymphoid Tissue and Cell-Mediated Defense

The overlying epithelium is histologically different from nearby tissue as it tends

3.4. Antigen-Presenting Cells

4. DEFENSES IN THE ALVEOLAR MILIEU

The conducting airways terminate at the level of the respiratory bronchioles,

4.1. Noncellular Components of the Alveolar Lining Fluid

4.2. Nonimmune Opsonins

4.3. Immune Opsonins

4.4. Complement System Components

4.5. Cells in the Alveolar Space

4.5.1. Alveolar Macrophages

The alveolar macrophage is one of many mononuclear phagocytes that

such as Staphylococcus aureus and by the formation of immune complexes. This

classic and alternative complement pathways can also generate complement